Figure 1.
List of 16 representative JH agonists and their biological activities in Drosophila morphogenetic assay.
The biological activity (ED50) is expressed in µg of the compound per animal, then it is converted to nmol per animal, and finally to -log of nmol values that are used for CoMFA and CoMSIA computations. Compounds 1–28 represent Class I agonists, whereas compounds 35–86 within this Table represent Class II agonists. Complete list of tested JH agonists is provided under Supporting Table S1. 1 = (2E,6E)-9-((2R)3,3-Dimethyl-oxiranyl)-3,7-dimethyl-nona-2,6-dienoic acid methyl ester (JH-III, also known as methylepoxyfarnesoate). 6 = (2E,6E)-(S)-10,11-Dihydroxy-3,7,11-trimethyl-dodeca-2,6-dienoic acid (JH-III acid diol). 12 = (E)-(R)-11-Chloro-3,7,11-trimethyl-dodec-2-enoic acid methyl ester. 16 = Tioethyl-(2E,4E)-(R)-11-methoxy-3,7,11-trimethyl-2,4-dodecadienoate (triprene; ZR-619). 23 = (2E,4E)-(R)-11-Methoxy-3,7,11-trimethyl-dodeca-2,4-dienoic acid diethylamide (ZR-618). 27 = (2E,4E)-(R)-3,7,11-Trimethyl-dodeca-2,4-dienoic acid prop-2-ynyl ester (kinoprene; ZR-777). 28 = 2-((E)-(8R,9S)-9-Ethoxy-4,8-dimethyl-dec-3-enyl)-2-methyl-(2R,3S)-cyclopropanecarboxylic acid isopropyl ester (ZR-4429). 35 = (R)-3-[5-(3-Ethyl-phenoxy)-3-methyl-pentyl]-2,2-dimethyl-oxirane. 39 = (R)-3-[5-(3-Ethoxymethoxy-phenoxy)-3-methyl-pentyl]-2,2-dimethyl-oxirane. 41 = (R)-3-{[(E)-4]-4-Chloro-phenoxy)}-3-methyl-but-3-enyl]-2,2-dimethyl oxirane. 48 = 5-[(E)-4-((2R,3S)-3-Ethyl-3-methyl-oxiranyl)-2-methyl-but-1-enyloxy]-benzo-[1], [3]-dioxole. 53 = (S)-4-{2-[3-(2,2,2-Trichloro-acetyl)-ureido]-propionylamino}-benzoic acid ethyl ester. 57 = [2-(4-Phenoxy-phenoxy)-ethyl]-carbamic acid ethyl ester (fenoxycarb). 62 = (R)-2-[1-Methyl-2-(4-phenoxy-phenoxy)-ethoxy]-thiazole. 80 = (S)-{2-[4-(1,4-Dioxa-spiro-[4], [6]-undec-6-ylmethyl)-phenoxy]-ethyl}-carbamic acid ethyl ester. 86 = 2-[1-Methyl-2-(4-phenoxy-phenoxy)-ethoxy]-pyridine (pyriproxyfen; Sumitomo 31183)
Figure 2.
The superpositional alignment of congeners of JH agonists analyzed in this study.
Complete set of all 86 JH compounds is shown in A, whereas B shows alignment of two selected agonists, natural JH-III (1) as representative of Class I and the most rigid structure of ZR-10852 (82) as representative of Class II.
Table 1.
Summary of results from CoMFA analyses for the common, unseparated training set of 76 JH agonists as well as for this set split into Class I and Class II compounds.
Table 2.
Summary of results from CoMSIA analyses for a training set of 76 JH agonists.
Figure 3.
CoMFA steric and electrostatic fields contour plot for Class I and Class II JH agonists.
Green polyhedra represent sterically favored regions where more bulky substituents increase biological activity, while yellow polyhedra surrounded regions indicate sites where less bulky substituents are appreciated for increasing biological activity. Blue polyhedra represent electrostatic regions where positively charged groups will be favorable and will enhance biological activity, whereas the red contours represent regions where negative charge is favorable. The importance of the electronegative oxygens or nitrogens on the ends of the aligned structures is indicated by red polyhedra near the positions of these atoms. The presence of smaller red polyhedra in the middle of the JH agonist structure indicates an additional site where an electronegative atom or group can enhance biological activity. The contour map of Class II (B) differs from that of Class I (A) mainly in the steric fields. The large green polyhedra of Class II (B) near the part of the phenoxyphenyl group (left side) indicates that the presence of steric bulk substituents in this part of the molecule enhances biological activity.
Figure 4.
Contour maps of Class I and Class II JH agonists CoMSIA analyses.
Green polyhedra represent sterically favored regions in which more bulky substituents will increase biological activity, while yellow polyhedra represent sterically disfavored regions where less bulky substituents are appreciated for increasing the activity of both Class I (A) and Class II (B) JH agonists, respectively. In the electrostatic contour plot, the red polyhedra represent favorable regions where negatively charged groups will enhance activity and the blue polyhedra represent disfavored regions where positively charged groups will enhance activity. Contour maps for the hydrogen bond acceptor and donor fields are illustrated in C (for Class I agonists) and D (for Class II agonists). Magenta areas indicate regions where hydrogen bond acceptors are favorable for increasing biological activity (oxygens and nitrogens in the ligand), cyan areas indicate fields where hydrogen bond donors are favorable (NH and OH groups in ligand). Orange polyhedra surround area where H-bond acceptors are unfavorable and white polyhedra areas where H-bond donors are unfavorable.
Figure 5.
Pharmacophore models of Class I and Class II JH agonists based on CoMFA and CoMSIA result.
Models show key structural elements responsible for the hormonal activity of JH agonists. Both classes of JH compounds share regions that favor negative charge and a field that requires hydrogen bond acceptor. In both cases these elements are located at the sides of JH agonists. However, Class I compounds (A) have also additional hydrogen bond acceptor element located on the esteratic side, and a region which favors positive charges centrally. A specific feature of Class II compounds (B) is a hydrophobic region that is favored around the outside edge of a phenoxyphenol moiety. Due to the high flexibility of Class I compounds the distance between hydrogen bond acceptor atoms (e.g. esteratic and epoxy oxygens at two sides) must be between 11.5 and 13.5 Å to possess agonist biological activity in Drosophila. Highly rigid Class II compounds that show JH biological activity in Drosophila automatically fit to this requirement.
Figure 6.
Three-dimensional structure of the pseudoreceptor models for Class I (A) and Class II (B) JH agonists.
Selected compounds are aligned in the middle with the surrogate of eight amino acids surrounding them. Both models are composed of eight amino acid residues reflecting all interactions that are predicted to be required from pharmacophore analysis. For the sake of clarity, bonding interactions and vectors have not been displayed.
Table 3.
ΔG Experimental versus ΔG PrGen predicted activities for JH agonists of Class I.
Table 4.
ΔG Experimental versus ΔG PrGen predicted activities for JH agonists of Class II.