Common and distinct roles of juvenile hormone signaling genes in metamorphosis of holometabolous and hemimetabolous insects

Abstract

Insect larvae metamorphose to winged and reproductive adults either directly (hemimetaboly) or through an intermediary pupal stage (holometaboly). In either case juvenile hormone (JH) prevents metamorphosis until a larva has attained an appropriate phase of development. In holometabolous insects, JH acts through its putative receptor Methoprene-tolerant (Met) to regulate Krüppel-homolog 1 (Kr-h1) and Broad-Complex (BR-C) genes. While Met and Kr-h1 prevent precocious metamorphosis in pre-final larval instars, BR-C specifies the pupal stage. How JH signaling operates in hemimetabolous insects is poorly understood. Here, we compare the function of Met, Kr-h1 and BR-C genes in the two types of insects. Using systemic RNAi in the hemimetabolous true bug, Pyrrhocoris apterus, we show that Met conveys the JH signal to prevent premature metamorphosis by maintaining high expression of Kr-h1. Knockdown of either Met or Kr-h1 (but not of BR-C) in penultimate-instar Pyrrhocoris larvae causes precocious development of adult color pattern, wings and genitalia. A natural fall of Kr-h1 expression in the last larval instar normally permits adult development, and treatment with an exogenous JH mimic methoprene at this time requires both Met and Kr-h1 to block the adult program and induce an extra larval instar. Met and Kr-h1 therefore serve as JH-dependent repressors of deleterious precocious metamorphic changes in both hemimetabolous and holometabolous juveniles, whereas BR-C has been recruited for a new role in specifying the holometabolous pupa. These results show that despite considerable evolutionary distance, insects with diverse developmental strategies employ a common-core JH signaling pathway to commit to adult morphogenesis.

Figure 1. Developmental expression of Pyrrhocoris JH signaling genes.

A) Semi-quantitative RT-PCR on total RNA from embryos (26 temperature cycles) or larvae (28 cycles) of indicated stages. d3, d5, mid- and late-L4 instar larvae, respectively. B) Relative mRNA levels on successive days of larval development were assessed with qRT-PCR and normalized to rp49 mRNA. Dashed lines mark ecdyses to L4, L5 and adult stages. Values are mean ± SD from three measurements on RNA isolated from individual animals.

Figure 2. Loss of Met or Kr-h1 causes precocious metamorphosis in Pyrrhocoris larvae.

A) Larvae newly ecdysed to the L4 instar were injected with control (egfp) or Met dsRNA. B) Efficacy of Met mRNA depletion and its effect on expression of Kr-h1 and BR-C expression were determined by qRT-PCR three days after injection of Met dsRNA (gray columns) relative to egfp dsRNA controls (open columns) arbitrarily set to 100%. Values are mean ± SD from n = 5 animals. C) Met, Kr-h1 and BR-C RNAi phenotypes after ecdysis to the L5 stage as compared to control L5 larvae and adults (left two columns). Animals in the top row are to the same scale; the middle and bottom rows show details of wings and of the ventral abdomen, respectively. Precocious adult attributes upon Met and Kr-h1 RNAi include color-patterned articulated wings (arrows) separated from the scutellum (s), extended notum (open arrowheads) with two posterior black spots, external male genitals (solid arrowheads), and dark abdominal cuticle (asterisks). Compared to control L5, BR-C(RNAi) larvae display retarded wing growth but no precocious adult development. For quantitative data see Table 1. (Scale bars: A and C, top row, 3 mm; C, middle row, 1 mm; C, bottom row, 2 mm).

Figure 3. Precocious development of adult features after Kr-h1 RNAi in the blood sucking bug, Rhodnius prolixus.

When subjected to Kr-h1 dsRNA injection as L3 larvae, the animals displayed abnormal growth and venation of wing lobes after molting to the L4 stage (right) as compared to control L4 larvae. We also noticed premature development of external genitals in Kr-h1(RNAi) animals (arrowhead).

Figure 4. Met and Kr-h1 mediate the anti-metamorphic effect of exogenous JH mimic.

A–D) Animals received either control (egfp), Met or Kr-h1 dsRNA as late-L4 larvae, followed by mock (acetone) or JH mimic (methoprene) treatment early at the L5 instar. Bottom row shows ventral view of the abdomens. Acetone-treated larvae produced normal adults (A). Methoprene induced a supernumerary larval instar (L6) whose wing pads remained black and attached to the tergites, while the abdominal cuticle tanned only partly (B). Knockdown of Met (C) or Kr-h1 (D) prior to methoprene treatment restored normal adult development. E) Ectopic Met-dependent induction of Kr-h1 and BR-C by methoprene at the L5 instar. Relative mRNA levels of Kr-h1 and BR-C in the abdominal epidermis of animals injected with control (egfp) or Met dsRNA and subjected to hormonal treatment as described above were assessed on day 4 of the L5 instar. Values are mean ± SD from n = 5 animals. Note that the much higher Kr-h1 induction (left) is on the logarithmic scale.

Figure 5. Regulation of hemimetaboly and holometaboly.

Pyrrhocoris (left) and Tribolium (right) cartoons signify the main innovations – postponement of wing development and the resting pupal stage in holometabolans. The absence of JH-dependent Kr-h1 expression in pupae and final instar hemimetabolous larvae (gray shaded areas) is prerequisite to adult development in both types of metamorphosis, supporting the view that these final juvenile stages of both insects types may be homologous , . The orange shaded area marks a period of low Kr-h1 activity in the absence of JH, which is necessary to permit partial metamorphosis during the pupal molt, specified by the newly acquired function of BR-C in holometabolans. Gene expression profiles for Pyrrhocoris and Tribolium are from Figure 1B and from , respectively. JH and ecdysteroid titers are from Blattella germanica (left) and Manduca sexta (right).