The first member of the CYP15 family, CYP15A1, was cloned from the corpora allata (endocrine glands producing juvenile hormones) of the cockroach Diploptera punctata (Helvig et al., 2004).

Purified Diploptera punctata CYP15A1 has a high affinity for methyl farnesoate, showing a type I spectrum with a Ks of 6 μM. The enzyme, when reconstituted with fly P450 reductase, catalyzes the NADPH-dependent epoxidation of 2E, 6E-methyl farnesoate to JH III. The epoxidation is highly stereoselective (98:2) to the natural 10R enantiomer over its diastereomer. The enzyme also has a high substrate specificity, epoxidizing the 2E,6E isomer preferentially over the 2Z,6E isomer, and accepting no other substrate tested including farnesoic acid. The rank order of inhibition of CYP15A1 activity by substituted imidazoles is identical to that of JH biosynthesis by isolated corpora allata (Helvig et al., 2004).

CYP15A1 is expressed selectively in the CA of adult Diploptera punctata, Schistocerca gregaria and Blattella germanica (Helvig et al., 2004; Marchal et al., 2011; Maestro et al., 2010) as well as in embryos of the latter species. RNAi silencing of the desert locust S. gregaria ortholog of CYP15A1 causes a decrease of spontaneous and farnesoic acid-stimulated JH III synthesis, as well as methyl farnesoate accumulation in the glands (Marchal et al., 2011).

RNAi of CYP15A1 in Tribolium castaneum does not cause precocious metamorphosis (Minakuchi et al., 2015), suggesting that the physiological function of MF was not abolished by the emergence of JH III, although other interpretations are possible, such as an extra-allatal epoxidation by other P450 enzymes.

Recently, Nouzova et al. (2021) showed that genetic knockout of CYP15B1 in Aedes aegypti was not lethal but impaired reproductive success, suggesting that epoxidation of the “ancestral” hormone MF was a key innovation of insects.

The CYP15 sequences do not form a monophyletic clade. The founder CYP15A1 from Diploptera punctata is the highly specific methyl farnesoate (MF) epoxidase that makes juvenile hormone III (methyl farnesoate 10R,11-epoxide) in the corpora allata of this cockroach (Helvig et al., 2004). It is part of a strongly supported clade (93/93.8/72) of CYP15A sequences, related to CYP15B1 of mosquitoes and to the CYP15C clade of Lepidoptera. CYP15C1 of Bombyx mori is the allatal farnesoic acid epoxidase (Daimon et al., 2012),

Bombyx mori CYP15B1 catalyses (homo and bishomo) farnesoic acid (i.e. JH, II, JH I and JH III -acid) epoxidation (T. Shinoda, pers. comm.) as expected from Lepidoptera where epoxidation takes place before esterification by JHAMT (juvenile hormone acid methyl transferase). The CYP15 enzymes have therefore evolved a lineage-specific substrate selectivity. Farnesoic acid is a better substrate for CYP15B1 than methyl farnesoate (Daimon et al., 2012), while CYP15A1 of Tribolium castaneum is a mixed farnesoic acid/MF epoxidase (Minakuchi et al., 2015).

The CYP15A/C clade therefore represents the bona fide enzymes producing the epoxide characteristic of juvenile hormones (JH). The CYP nomenclature places several CYP2 clan sequences in the CYP15 family, including two CYP15H from locusts, six from Machilis hrabei, three from Calopterx splendens, eight from Catajapyx aquilonaris and three from Sinella curviseta, but these do not form a single well supported clade and their function is unknown.

Phylogeny of CYP15 and related P450s Biochemically charaterized CYP15 are marked with *. (Fig. S15 from Nouzova et al., 2021)

In the termite Reticulotermes flavipes, CYP15F1 is induced by JH and facilitates JH-dependent soldier caste differentiation but its precise biochemical function is unknown (Tarver et al., 2012). The three Calopteryx splendens CYP15 sequences are also outside the 15A/C clade. This raises the question of the precise origin of the CYP15A/C clade and when during evolution of insects the JH function switched from the “ancestral” MF to its epoxidized congener, JH III.

The deepest branch within the class Insecta in which a CYP15 epoxidase was found with high degree of confidence is therefore the order Zygentoma (Thermobia domestica, genomic sequence, and Tricholepidion gertschi, transcript GASO02035928.1). Interestingly, this distribution matches the distribution of the closely related CYP305 gene, also a highly conserved and generally single-copy gene restricted to insects, but of unknown function. While Crustacea and “primitive” hexapods have large numbers of CYP2 clan P450s, this is not the case in Condylognatha, Psocodea, and Holometabola. This suggests on one hand that the epoxidase emerged from a pool of diverse CYP2 clan genes, probably including P450s with a detoxification function. On the other hand, gene loss during insect evolution spared the epoxidase, CYP303 and CYP305 lineages.

Two lineages in insects have a different type of JH, the higher Diptera which make a JH “bisepoxide” (methyl farnesoate 6S,10R-diepoxide) and bugs, such as Rhodnius prolixus (in which JH was first discovered) and several other Hemiptera which make a JH “skipped bisepoxide” (methyl farnesoate 2R,10R-diepoxide) (Kotaki et al., 2009; Villalobos-Sambucaro et al., 2020). R. prolixus has a single CYP15A1 gene, suggesting that this P450 can epoxidize at one if not at both sites. Aphids have duplicated the CYP15 gene with three copies in the peach aphid.

CYP15 is missing in higher Diptera, specifically Cyclorrhapha. but is found in other Diptera such as Empidoidea (partial TSA sequence in Heteropsilopus ingenuus), Asiloidea (Proctacanthus coquilletti), Bibionomorpha (Mayetiola destructor, Bradysia coprophila), Tipulomorpha, Psychodomorpha and Culicomorpha. The identity of the epoxidase in Cyclorrhapha is therefore unknown. Epoxidation of a farnesoid molecule can be achieved by other enzymes (e.g. CYP6A1), so it is conceivable that CYP15 was lost in higher Diptera and this loss compensated by either the utilization of non epoxidized farnesoids or the recruitment within or outside the CA of other P450 enzymes, or both. One candidate may be CYP6G2.

CYP6G2 is highly expressed in the allatal portion of the Drosophila ring gland, and its RNAi knockdown is lethal (Chung et al., 2009; Ou et al., 2016; Christesen et al., 2017). The biochemical function of CYP6G2 is unknown, but many CYP6 enzymes have a broad substrate specificity, and house fly CYP6A1 is highly active as 6- and 10-epoxidase of MF (Andersen et al., 1997). Similarly, CYP6G2 does not appear to be a highly specific enzyme, because its transgenic, ectopic overexpression results in nitenpyram resistance (Daborn et al., 2007).