Detection of juvenile hormone agonists by a new reporter gene assay using yeast expressing Drosophila methoprene‐tolerant

Juvenile hormones (JHs) are sesquiterpenoids that play important roles in the regulation of growth, metamorphosis, and reproduction in insects. Synthetic JH agonists (JHAs) have been used as insecticides and are categorized as a class of insect growth regulators (IGRs). Natural JHs and synthetic JHAs bind to the JH receptor methoprene‐tolerant (Met), which forms a functional JH‐receptor complex with steroid receptor coactivators, such as Drosophila melanogaster Taiman (Tai). The ligand‐bound Met–Tai complex induces the transcription of JH response genes by binding to specific DNA elements referred to as JH response elements (JHREs). In the present study, we established a reporter gene assay (RGA) for detecting natural JHs and synthetic JHAs in a yeast strain expressing D. melanogaster Met and Tai. The yeast RGA system detected various juvenoid ligands in a dose‐dependent manner. The rank order of the ligand potencies of the juvenoids examined in the yeast RGA linearly correlated with those of RGAs for Met–Tai established in mammalian and insect cells. Our new yeast RGA is rapid, easy to handle, cost‐effective, and valuable for screening novel JHAs.

Juvenile hormones (JHs) are sesquiterpenoids that play important roles in the regulation of growth, metamorphosis, and reproduction in insects. Synthetic JH agonists (JHAs) have been used as insecticides and are categorized as a class of insect growth regulators (IGRs). Natural JHs and synthetic JHAs bind to the JH receptor methoprene-tolerant (Met), which forms a functional JH-receptor complex with steroid receptor coactivators, such as Drosophila melanogaster Taiman (Tai). The ligand-bound Met-Tai complex induces the transcription of JH response genes by binding to specific DNA elements referred to as JH response elements (JHREs). In the present study, we established a reporter gene assay (RGA) for detecting natural JHs and synthetic JHAs in a yeast strain expressing D. melanogaster Met and Tai. The yeast RGA system detected various juvenoid ligands in a dose-dependent manner. The rank order of the ligand potencies of the juvenoids examined in the yeast RGA linearly correlated with those of RGAs for Met-Tai established in mammalian and insect cells. Our new yeast RGA is rapid, easy to handle, cost-effective, and valuable for screening novel JHAs.
Drosophila species [18,19]. It was shown that Gce had a partly conserved function with Met in transducing JH action [20,21]. Further study demonstrated that JH-binding capacity of Met and Gce were required to mediate JH action during normal development of D. melanogaster, establishing Met/Gce as JH receptors in vivo [22]. Ligand-bound Met/Gce proteins heterodimerize with other bHLH-PAS family proteins, such as D. melanogaster Taiman (Tai) [23,24]. The complex of JH-bound Met and Tai binds to specific cis-acting DNA sequences referred to as JH response elements (JHREs) in order to induce the transcription of early JH response genes [23][24][25]. The antimetamorphic gene Kr€ uppel homolog 1 (Kr-h1) was identified as a target of Met, and a 13-nucleotide motif containing an E-box (CACGTG) as JHRE. The DNA motif was essential for the binding of Met-Tai to induce the transcriptional activation of Kr-h1 [26][27][28].
The activities of natural JHs and synthetic JHAs have traditionally been examined using whole body [29][30][31][32][33] or cultured insect cell lines to detect JHdependent morphological changes [34][35][36]. Since the cloning of Met/Gce genes, in vitro ligand-receptor binding assay procedures have been established [17,22,37,38]. A reporter gene assay (RGA) is an alternative procedure for examining the JH-like activity of synthetic chemicals. In RGA systems, the JHdependent activation of Met-mediated gene expression is monitored using reporter gene constructs under the control of JHRE. The two-hybrid-based method to detect JH-dependent interactions between Met and Tai is also employed. To date, a number of RGAs for Met have been established in insect and mammalian cells [39]. We previously established yeast RGAs for the ligand-dependent transcription factors of animals, such as mammalian aryl hydrocarbon receptors (AhRs)-AhR nuclear translocators (Arnts) in bHLH-PAS family proteins [40,41] and human steroid hormone receptors and insect ecdysone receptors (EcRs)ultraspiracles (USPs) in the nuclear receptor superfamily [42][43][44]). These yeast RGAs are simpler, easier to handle, and more cost-effective than mammalian or insect cell-based bioassays and instrumental analyses.
In the present study, we established a novel yeast RGA to quantitatively measure the activities of JHs and JHAs in a recombinant yeast strain. Met and Tai from D. melanogaster were expressed in yeast cells carrying the lacZ reporter plasmid, with JHRE being identified in the Bombyx mori Kr-h1 gene [26] (Fig. 1). The ligand responses of the yeast RGA were examined using various juvenoids, and its sensitivity was compared with those of previously established mammalian or insect cell RGAs for Met-Tai.

Plasmid construction
The D. melanogaster Met expression plasmid and reporter plasmid carrying the JHRE identified in the B. mori Kr-h1 gene promoter [26] were constructed to develop the Met ligand assay yeast strain. The primers used in the present study were synthesized by Sigma-Aldrich Japan (Tokyo, Japan) and are listed in Table S1.
DNA fragments containing the Met (DDBJ/EMBL/Gen-Bank accession number NM_078571) open reading frame (ORF) were obtained from D. melanogaster larva poly A + RNA (Clontech, Palo Alto, CA, USA) using a reversetranscription polymerase chain reaction (RT-PCR). First, cDNA was obtained using SuperScript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA). Met cDNA was amplified by PCR with the primer pair DmMet Fwd and DmMet Rev, which contained a restriction site and/or yeast ribosomal binding consensus sequence near the initiation codon. PCR was performed using high-fidelity PCR polymerase KOD-plus-ver. 2 (TOYOBO Co., Ltd.) according to the manufacturer's instructions. Amplified fragments were digested with SmaI and EcoRI, and cloned into the corresponding sites of multicloning site (MCS) 2 of the expression vector pUdp6 [47]. Plasmids were isolated and purified using a QIAGEN Mini Prep Kit (Valencia, CA, USA), and the nucleotide sequences of Met ORF were confirmed using the ABI DNA sequencer. A sequence analysis was performed using the Genetyx-Mac ver. 13.0.3 and Genetyx-Mac/ATSQ ver. 4.2.4 programs (Genetyx Corporation, Tokyo, Japan).
We found a sequence alteration in the cloned Met ORF, an A to C alteration at nucleotide position 307, which was associated with an amino acid substitution, threonine (ACC) to proline (CCC) at codon 103. Therefore, Met ORF was amplified again from cDNA by PCR using KOD-plus-ver. 2 in three independent experiments and then purified and sequenced. Codon 103 of Met was reconfirmed as the CCC encoding proline in all three independently amplified DNA fragments. The T103P alteration in our Met clone appeared to be polymorphism [48]. Since the plasmid was functional in terms of the JH/JHA-dependent induction of reporter genes in yeast cells, it was designated as pUdp6-DmMet and used to establish the yeast strain for the Met-Tai assay in the present study.
To construct the reporter plasmids carrying JHRE of B. mori Kr-h1 (BmkJHRE), one or two copies of oligonucleotides containing these sequences were inserted into the MCS of the pRW95-3 vector [49], upstream of the CYC minimal promoter. The oligonucleotides Bm-kJHRE Fwd and Bm-kJHRE Rev were annealed, phosphorylated with T4 polynucleotide kinase, ligated, and inserted into the Spe I site of pRW95-3. The copy numbers and orientation of the oligonucleotides on each plasmid were confirmed by sequencing. The resultant plasmids were designated as pYTb-BmkJHRE 9 1 and pYTb-BmkJHRE 9 2.
The plasmid pESC-Leu-DmTai for the expression of Tai was constructed in our previous study [44]. Briefly, Tai ORF amplified from D. melanogaster larva poly-A + RNA by RT-PCR was cloned into MCS 2 on the yeast expression vector pESC-Leu (Agilent Technologies, Inc., Santa Clara, CA, USA).

Establishment of the yeast RGA
Yeast transformation was performed using the lithium acetate procedure as previously described [50]. To develop a yeast RGA for Met-Tai, one of the reporter plasmids and the expression plasmid for Tai, pESC-Leu-DmTai, were introduced into the wild-type yeast strain W303a. A transformant grown on SDC-TRP/LEU agar medium was isolated and used as a host for subsequent transformation. The Met expression plasmid pUdp6-DmMet was linearized by BstBI digestion and integrated into the ura3 locus in the yeast genome by homologous recombination (Fig. 1). Transformants were selected on SCD-TRP/LEU/URA agar plates. In order to establish reference strains without Met expression, the empty plasmid pUdp6 was linearized by EcoRV digestion and integrated into W303a, as described above. To obtain another reference strain that does not express Tai, the empty plasmid pESC-Leu was used in the first transformation experiment instead of pESC-Leu-DmTai.

Measurement of JH/JHA activity using yeast RGA
The yeast RGA was performed as previously described [42,44]. Single colonies of yeast strains were grown in SDC-TRP/LEU/URA medium at 30°C overnight, and the optical density (OD) at 595 nm of each culture was adjusted to 1.0 with the same medium.  [51]. Data were analyzed using Student's t-test to assess the significance of differences between two sets of values. Probability (P) values < 0.05 were considered to be significant.

Optimization of the yeast RGA for Met
To optimize the yeast RGA for Met, we constructed new reporter plasmids carrying one or two copies of the E-box (CACGTG)-containing JHRE from the B. mori Kr-h1 gene (BmkJHRE) [26]. These reporter plasmids were introduced into the wild-type yeast strain W303a along with the Met and Tai expression plasmids. A yeast strain carrying two copies of BmkJHRE exhibited reporter activity in response to JH III (0.1 lM) (Fig. 2A). Therefore, we selected pYTb-BmkJHRE 9 2 containing two copies of Bm-kJHRE as a reporter plasmid for the Met-Tai RGA in the present study.
We then compared the ligand responses of yeast strains expressing Met-Tai, Met, and Tai, respectively. Only the yeast strain co-expressing Met and Tai responded to JH III and induced lacZ reporter gene expression. Neither of the strains expressing Met or Tai alone responded to JH III or a reference strain carrying empty plasmids instead of the Met and Tai expression plasmids (Fig. 2B).

Responses of Met-Tai assay yeast to natural JHs
We initially investigated the responses of the Met-Tai assay yeast strain to the commercially available natural JHs, JH III, and methyl farnesoate. As shown in Fig. 3A, Met-Tai assay yeast responded to these JHs in a dose-dependent manner. Both compounds induced reporter gene expression at 0.01 lM ( Table 1). We also used in-house synthesized natural JHs [46] in the yeast RGA. JH I and II activated Met at 0.01 lM, and the b-gal level increased in a dosedependent manner, as well as JH III (Fig. 3B, Table 2). Met-Tai assay yeast did not exhibit a dosedependent response up to 0.01 lM in the assay of JH 0; however, b-gal activity was higher than that of the solvent control. JH 0 slightly potentiated reporter gene expression at high doses (Fig. 3B). The EC 50 values of natural JHs calculated from dose-response curves were summarized in Tables 1 and 2. The EC 50 value of JH III was approximately 2.5-fold lower than that of methyl farnesoate, and the difference was significant (P < 0.001) ( Table 1). Regarding inhouse synthesized natural JHs, JH 0 was a less potent ligand than other JHs (3.3-to 4.3-fold difference, P < 0.001 for JH 0 vs. JH I or III, P < 0.005 for JH 0 vs. JH II, Table 2). No significant differences were observed in EC 50 values among JH I, II, and III (P = 0.27~0.52). The rank order of the ligand potencies of natural JHs in yeast RGA was JH II ≥ JH III ≥ JH I > methyl farnesoate > JH 0.

Responses of Met-Tai assay yeast to synthetic JHAs
We performed the yeast RGA using the synthetic JHAs, methoprene, (S)-hydroprene, pyriproxyfen, and fenoxycarb as ligands. Met-Tai assay yeast responded to these ligands in a dose-dependent manner. Fenoxycarb reduced b-gal activity at the highest concentration due to its cytotoxicity to yeast cells (Fig. 4). Relative cell growth rate decreased to approximately 40% after 18 h exposure to fenoxycarb at 100 lM, while over 80% of the cells survived in the assays with other JHAs. Significant differences were observed in the EC 50 values of these compounds (Table 3), and the rank order of synthetic JHAs in yeast RGA was fenoxycarb >> pyriproxyfen >> (S)-hydroprene > methoprene.

Relationships between ligand potencies of juvenoids among RGAs for Met-Tai established in yeast and animal cells
Three RGAs for Met-Tai were recently established in insect and mammalian cell systems. As a mammalian RGA cell system, the expression plasmids for Met and Tai and the luciferase reporter plasmid under the control of BmkJHRE were introduced into HEK293T cells derived from human embryonic kidney cells [46]. To establish a RGA in insect cells, Bittova et al. introduced a luciferase plasmid under the control of the JHRE of the A. aedes early trypsin (Aa ET) gene into D. melanogaster Schneider 2 (S2) cells expressing endogenous JH receptors [38]. A two-hybrid-based RGA in which Met and Tai were fused to the VP16 activation domain (VP16 AD) and GAL4 DNAbinding domain (GAL4 DBD), respectively, were also established in Chinese hamster ovary (CHO) cells [48]. We compared the potencies of natural JHs and synthetic JHAs between yeast and RGAs established in HEK293T, Drosophila S2, and CHO cells by plotting pEC 50 values, the negative logarithms of EC 50 (Fig. 5). The sensitivity of yeast RGA was 2-to 30-fold lower than that of the HEK293T RGA. On the other hand, yeast RGA was more sensitive than the Drosophila S2 RGA and CHO two-hybrid RGA: ligand potencies in the yeast RGA were 4-to 27-fold and 5-to 11-fold higher than those in the Drosophila S2 and CHO two-hybrid RGAs, respectively. The correlation coefficient (r) of regression lines between yeast and other RGAs were high, with r values of 0.975-0.986 (Fig. 5). This result indicates that the rank order of the ligand potencies of test compounds strongly correlated among yeast, mammalian, and insect cell RGA systems (Fig. 5).

Discussion
RGA is a widely used experimental method that has been employed to examine the activity of synthetic chemicals that mimic the MHs and JHs of insects. These insect hormones exert their effects by activating the specific receptors EcR-USP and Met for MHs and JHs, respectively. EcR-USP and Met are liganddependent transcription factors that belong to the nuclear receptor superfamily and bHLH-PAS domain protein family, respectively (reviewed by Ito-Harashima and Yagi [39]). We previously established RGA systems for insect EcR-USP in the yeast S. cerevisiae, in which EcRs and USPs from three different taxonomic orders were introduced in conjunction with the steroid receptor coactivator Tai and reporter plasmids containing appropriate response elements [44]. S. cerevisiae is the simplest eukaryote possessing highly conserved gene expression systems with those of animals, including insects and humans [52]. The mechanisms of action and ligand responses of EcR-USP from three insect species were reconstituted well in yeast RGAs [44]. In insect cells, the functional JH receptor is constituted by the heterodimer of Met and Tai proteins [23][24][25]. In the present study, we newly established a yeastbased RGA to detect JHs/JHAs by introducing cDNA fragments containing full-length D. melanogaster Met and Tai ORFs (Fig. 1). We showed that the E-boxcontaining JHRE identified in the B. mori Kr-h1 gene (BmkJHRE) [26] was effective for JH III-dependent gene expression in the yeast RGA ( Fig. 2A). This result suggests that the DNA-binding properties of Met proteins were conserved among the different insect species because of the high sequence similarity for the bHLH domain involved in DNA binding  The EC 50 values of synthetic JHAs were significantly different from each other: *methoprene vs. (S)-hydroprene (P < 0.05); **methoprene vs. pyriproxyfen or fenoxycarb (P < 0.005); # (S)-hydroprene vs. pyriproxyfen (P < 0.01); ## (S)-hydroprene vs. fenoxycarb (P < 0.005); $ pyriproxyfen vs. fenoxycarb (P < 0.001). [23,25,38]. We also demonstrated that JH IIIdependent reporter gene expression was only observed in the yeast strain co-expressing Met and Tai (Fig. 2B), suggesting that the heterodimerization of Met and Tai was required for ligand-dependent transactivation in yeast as well as insect cells [23][24][25]. We examined whether the established yeast RGA for Met-Tai was responsive to known juvenoids. As shown in Figs. 3 and 4, Met-Tai assay yeast exhibited dose-dependent responses to natural JHs and synthetic JHAs. Differences in ligand potencies among juvenoids, particularly synthetic JHAs, were clearly detectable (Tables 1-3). The sensitivity of the yeast RGA was lower than that of the HEK293T RGA, but higher than RGAs established in Drosophila S2 and CHO cells. To improve the sensitivity of the yeast RGA, increases in the copy number of BmkJHRE or the use of JHRE of the D. melanogaster Kr-h1 gene [53] may be effective. Alternatively, the deletion of the genes encoding cell wall mannoproteins (CWP1/ CWP2) and/or plasma membrane efflux pumps (PDR5/PDR10), or ERG6, which is involved in the synthesis of plant/fungal-specific ergosterol, may improve the permeability of the cell wall/plasma membrane and increase the intracellular concentrations of test compounds [42,43,[54][55][56]. We also showed that the rank order of ligand potencies in the yeast RGA strongly correlated with those obtained in the other RGA systems (Fig. 5). JH III is the most common in various insects while unique types of JHs have been found in some insect orders [9,57]. JH III bisepoxide (JHB 3 ) was identified as a JH specific to higher Dipteran insects including D. melanogaster [58,59], and its agonist potency for Drosophila JH receptors was indicated in vitro [38]. It is important to evaluate the effectiveness of our yeast RGA for Met-Tai in detecting JHB 3 compared to other JHs. It is also interesting to establish JH-type specific RGAs in the future. Twohybrid-based yeast RGAs to detect interactions between A. aedes Met and Tai orthologs were recently established and utilized for the screening of plant compounds with JH/anti-JH activity [60][61][62]. We have previously shown that the yeast RGAs for human nuclear receptors detected not only agonist but also antagonist activities of various ligands [42,43,47]. Our newly established yeast RGA for Met-Tai is useful for screening more potent JH/anti-JH-like compounds from a pool of natural and synthetic chemical compounds.
In the present study, the JH-Met signaling pathway for the transcriptional activation of target genes was reconstituted via B. mori Kr-h1 JHRE in the yeast strain co-expressing Met and Tai. This procedure is simple and may be applied to the establishment of new yeast RGAs for Met from other insect and arthropod species. Further studies are warranted to examine whether species-selective responses to JHs/JHAs exist. In contrast to synthetic MH agonists, JHAs with potent insect selectivity have not yet been discovered [63]. Extensive efforts have recently been made to obtain a novel class of JHAs using combined approaches, such as (quantitative) structure-activity relationship ((Q)SAR) studies, virtual screening methods, including the three-dimensional (3D) modeling of  [38]. (C) Comparison with the two-hybrid-based RGA established in CHO cells [48]. Natural JHs; JH I (red square), JH II (green square), and JH III (purple square), and methyl farnesoate (orange triangle). Synthetic JHAs: methoprene (blue diamond), pyriproxyfen (red diamond), and fenoxycarb (green diamond). Correlation coefficients (r) are shown in each graph.
the ligand-binding pocket and ligand docking studies, and RGA methods [46,64,65]. The yeast RGA is available as an activity-based high throughput screening method for the compounds identified in (Q)SAR and/ or virtual screening approaches. Yeast-based RGA systems for the insect hormone receptors, Met-Tai (this study) and EcR-USP of three different taxonomic orders [44], are rapid, easy to handle, and costeffective experimental methods, and valuable as primary screening tools for detecting new IGRs with MH-and JH-like activities.