Metamorphosis was a key innovation in insect evolution, wherein the individual acquires characteristic adult features and stops molting during postembryonic development. The ancestral metamorphosis mode was hemimetaboly, in which the embryogenesis gives rise to a first instar nymph with the essential adult body structure. The nymphs grow gradually and the final molt to the adult stage completes the formation of functional genitalia and wings. The metamorphosis mode known as holometaboly emerged from hemimetaboly, which is characterized by embryogenesis that produces a larva with a body form that may be substantially different from that of the adult. The larva grows through various stages until molting to the pupal stage, which bridges the gap between the morphologically divergent larva and that of the winged and reproductively competent adult. In the hemimetabolan and holometabolan modes, metamorphosis is regulated by two hormones: the juvenile hormone (JH) and the ecdysone, plus its biologically active derivative, 20-hydroxyecdysone (20E). 20E is a steroid, and its main role is to promote the successive molts, including the metamorphic one, whereas JH is a terpenoid, whose function is to repress metamorphosis [1]. The action of these hormones is underpinned by the mechanisms that transduce the hormonal signal through a pathway of gene activation. The 20E signaling pathway was first described in the 1990s [2], whereas the most important details of the JH pathway were unveiled recently. Important components of the JH signaling pathway are the JH receptor, which is the basic helix–loop–helix-Per-ARNT-Sim (bHLH-PAS) protein known as methoprene tolerant (Met), to which JH binds, as unveiled by Jindra's group in the 2010 decade. Another important component is Taiman, also a bHLH-PAS protein that plays the role of co-receptor. Finally, the transcription factor Krüppel homolog 1 (Kr-h1) is the main transducer of the antimetamorphic signal of JH [1].

Krüppel homolog 1

Kr-h1 was discovered in Drosophila melanogaster as a gene with structural similarity to the segmentation gene Krüppel, with which it shares the zinc-finger motifs and amino acid spacers connecting them. The first evidence that connected Kr-h1 and JH was also obtained in D. melanogaster. In this fly, the adult abdominal epidermis derives from larval histoblasts, which start proliferating after puparium formation. The experiments of Ashburner in 1970 showed that administration of JH prior to the prepupal stage prevents the normal differentiation of the abdominal epidermis, and the bristles that should be formed in the adult are shorter or lacking. In 2008, the experiments of Minakuchi and coworkers indicated that Kr-h1 expressed ectopically in the abdominal epidermis during metamorphosis of D. melanogaster also resulted in missing or short bristles, thereby suggesting that Kr-h1 mediates the antimetamorphic action of JH. New experiments of Minakuchi and coworkers carried out in the beetle Tribolium castaneum in 2009 showed that RNAi depletion of Kr-h1 in young larvae caused a precocious larval–pupal transformation, providing clear evidence that Kr-h1 represses metamorphosis and works downstream from Met in the JH signaling pathway. The antimetamorphic action of Kr-h1 was generalized to hemimetabolans in two parallel papers published in 2011 and conducted, respectively, in the cockroach Blattella germanica by Lozano and Belles, and the bugs Pyrrhocoris apterus and Rhodnius prolixus by Konopová and coworkers. In these two studies, RNAi experiments showed that Kr-h1 depletion in nymphs in the penultimate or antepenultimate nymphal stage triggers precocious metamorphosis [1] (Fig. 1A).

E93

E93 is an early gene in the ecdysone signaling cascade that is specifically expressed in late prepupae of D. melanogaster. As shown by the groups of Thummel and Baehrecke in the 1990 decade, the gene encodes for a protein with RHF domains significantly similar to pipsqueak motifs, which was found to be a key player in the degeneration process of the salivary glands during D. melanogaster metamorphosis. However, the action of E93 in metamorphosis is not restricted to the regulation of degeneration processes, given that it also plays morphogenetic roles. In 2012 Mou and coworkers observed that E93 is widely expressed in adult cells of the pupa of D. melanogaster, where it is required for patterning processes. Studying the induction of the Distal-less (Dll) gene within bract cells of the pupal leg using epidermal growth factor (EGF) receptor signaling, Mou and coworkers found that E93 causes Dll to become responsive to EGF receptor signaling, thus indicating that E93 is both necessary and sufficient for determining this switch. These results suggested that E93 controls the responsiveness of many other target genes and that it is generally required for patterning during metamorphosis. Subsequent RNAi experiments reported by Ureña and coworkers in 2014 showed that E93-depleted D. melanogaster larvae are able to pupate but die at the end of the pupal stage. In T. castaneum, E93 depletion by RNAi prevented the pupal–adult transition, resulting in the formation of a supernumerary second pupa. Similar results were obtained in the cockroach B. germanica, where E93 depletion in nymphs prevented the nymphal–adult transition, giving rise to repeated supernumerary nymphal instars (Fig. 1B). The same year, Belles and Santos showed that the expression of E93 in juvenile nymphs of B. germanica is inhibited by the transcription factor Kr-h1, thus uncovering the essential mechanism by which JH represses metamorphosis, which was named MEKRE93 pathway [3].

The MEKRE93 pathway in hemimetabolan species

The observation that Kr-h1 represses E93 expression led to propose the MEKRE93 pathway as the essential axis regulating insect metamorphosis. Accordingly, in nymph–nymph transitions, JH acts through its receptor Met-Taiman to induce the expression of Kr-h1, while Kr-h1 represses the expression of E93. In contrast, the decline of JH production in the final juvenile stage interrupts Kr-h1 expression, E93 becomes de-repressed, thus triggering adult morphogenesis (Fig. 2A) (Belles and Santos, 2014). RNAi experiments in B. germanica by the same authors also revealed that E93 depletion increases Kr-h1 expression, thus indicating that Kr-h1 and E93 are reciprocally repressed. The inhibitory action of Kr-h1 upon E93 expression was corroborated two years later in the holometabolan T. castaneum by Ureña and coworkers, which extended the MEKRE93 pathway framework to holometabolan metamorphosis.

The MEKRE93 pathway in holometabolan species

The main difference between the hemimetabolan and holometabolan metamorphoses is the regulation and function of the Broad complex (BR-C) zinc-finger transcription factors. In hemimetabolan species, BR-C is mainly involved in promoting the growth of wing primordia. For example, this was shown in B. germanica by Huang and coworkers in 2013, who additionally reported that BR-C expression is induced by JH and Kr-h1 during juvenile stages. One year later, Ureña and coworkers found that BR-C is repressed by E93 in the metamorphic transition. Furthermore, RNAi studies made in 2019 by the group of Mito and Noji in the cricket Gryllus bimaculatus, a hemimetabolan species, confirmed the mentioned interactions, and additionally discovered that BR-C and Kr-h1 are reciprocally activated. In sharp contrast, and as shown mainly by the group of Riddiford in the decade of 1990, BR-C triggers the formation of the pupal stage in holometabolan species, where JH inhibits the expression of BR-C during larval stages and stimulates BR-C expression after pupal commitment (Fig. 2B). In 2019, Chafino and co-workers showed that E93 is involved in triggering the pupal stage, as it promotes BR-C expression in T. castaneum. The whole data indicates that the MEKRE93 pathway is conserved in the holometabolan species, which added the E93/BR-C interaction loop to the ancestral (hemimetabolan) pathway during the evolutionary transition from hemimetaboly to holometaboly (Fig. 2C).