Skip to main content
Science Advances logoLink to Science Advances
. 2026 Feb 13;12(7):eaea4413. doi: 10.1126/sciadv.aea4413

Opposing insulin signals underlie the same developmental switch across hemipteran insects

Jin-Li Zhang 1,2,†, Yong-Kang Liu 1,†, Yi Wan 1,†, Hui-Jie Wu 1,†, Ni-Tong Xu 1,†, Zhuo-Qi Liu 1, Heng-Guang Huang 1, Jia-Peng Yang 1, Xin-Yang Liu 1, Xu-Mei Luo 1, Yi-Song Li 3, Qi-Sheng Song 4, Frederik H Nijhout 5, Ehab Abouheif 6,*, Hai-Jun Xu 1,*
PMCID: PMC12904164  PMID: 41686885

Abstract

Across metazoans, the insulin/insulin-like signaling (IIS) pathway is assumed to promote organ growth. We found that genes within the IIS pathway evolved opposite roles in organ growth across wing-dimorphic insects, playing opposite roles in regulating the developmental switch between short and long wings. In the firebug, the InR-Chico-PI3K-PDK1-Tsc1/2-TORC1 cascade redirects the switch from short to long wings by inactivating IIS, whereas in the planthopper, the InR-Chico-PI3K-Akt-FoxO cascade redirects this same switch by activating IIS. Activation or inactivation of IIS results in low ecdysteroid titers in both species by regulating different genes within the ecdysteroid biosynthesis pathway, which in turn induces long wing development. Our findings reveal a negative relationship between IIS and organ growth and a mechanism for how a hormone pathway can translate opposing IIS activities into the same developmental switch across species.


The ecdysteroid pathway translates opposing insulin signaling activities into the same developmental switch across species.

INTRODUCTION

Signal transduction pathways, which transmit signals from a receptor within the membrane of the responding cell to the nucleus to regulate gene expression, underlie the development and evolution of adaptive phenotypes (1–3). Two general patterns in the evolution of signaling pathways have emerged: First, a few signal transduction pathways—such as the Wnt, Janus kinase–signal transducers and activators of transcription, and the receptor tyrosine kinase pathway—have been repeatedly co-opted to generate the remarkable diversity of cell types and phenotypes across animals (4); and second, these signaling pathways have evolved the ability to produce similar developmental outcomes and switches despite showing a remarkable degree of rewiring across species (4–8). While the mechanisms underlying the first pattern have received much attention (9–11), the mechanisms through which an evolutionarily flexible signaling pathway can produce a similar developmental outcome or switch remains poorly understood.

Here, we address this challenge by investigating the evolution and development of the insulin/insulin-like signaling (IIS) pathway, an evolutionarily conserved receptor tyrosine kinase pathway across insect species. This pathway mainly involves the homolog of insulin receptor (InR), mammalian InR substrates 1 to 4 (Chico), phosphatidylinositol 3-kinase (PI3K), phosphoinositide-dependent kinase (PDK1), protein kinase B (Akt), phosphatase and tensin homolog (Pten), and the transcription factor FoxO (12). The IIS pathway is assumed to positively modulate growth and development in response to nutrients across metazoans (13, 14). For example, in Drosophila, the activation of IIS via disruption of genes acting negatively within IIS (e.g., Pten) results in flies with larger body and organ size (15), whereas inactivation of IIS via disruption of genes acting positively (e.g., InR, Chico, and Akt) results in smaller body and organ size (16–18). Because of its role in the regulation of growth and size of the body and appendages, IIS has been co-opted to play a widespread role in regulating insect polyphenisms, in which alternative phenotypes develop from a single genome in response to environmental cues. Examples include promoting the growth of mandibles in the soldiers of termites and growth of horns in males of rhinocerous beetles (19–21). Here, we address the question of how an evolutionarily flexible signaling pathway can produce similar alternative phenotypes (developmental switch) across species by using, as a model, the role of the IIS pathway in regulating wing polyphenism across hemipteran insects.

Wing polyphenism within the insect order Hemiptera produces either long-winged or short-winged/wingless morphs during development in response to environmental cues (22, 23). The same developmental switch between long and short wing morphs evolved several times independently within the Hemiptera, such as in the European firebug Pyrrhocoris apterus (order Heteroptera) and the brown planthopper Nilaparvata lugens (order Auchenorrhyncha). In both species, nymphs bear bipotential wing buds that can either develop into short-winged or long-winged adults (Fig. 1). However, wing buds in these species are externally indistinguishable during this nymph stage and become visible only once the adults emerge (Fig. 1). The repeated evolution of this trait within the Hemiptera may be related to their colonization of a diversity of ecological habitats (fig. S1). Previous studies suggest differences in the expression of the IIS pathway between short and long wing morphs across hemipteran species. For example, unlike the two InR paralogs (NlInR1 and NlInR2) described in the planthopper (24), the firebug P. apterus has three InR paralogs (PaInR1a, PaInR1b, and PaInR2); two InR paralogs (PaInR1a and PaInR2) switches the developmental trajectory from long to short wings, while the third (PaInR1b) has only a neutral effect (25). InR homologs with neutral or marginal effects on wing polyphenism have also been found in the soapberry bug Jadera haematoloma and in the water strider Gerris buenoi (26, 27). Another concrete example is that disruption of FoxO leads to the switch from short to long wings in the planthopper (24, 28) but long to short wings in the soapberry bug (26). How the highly conserved IIS pathway has evolved yet, at the same time, produces the same developmental switch between wing morphs across hemipteran insects remains unknown.

Fig. 1. Development of alternative wing morphs in the firebug P. apterus and planthopper N. lugens.

Fig. 1.

Both P. apterus (A) and N. lugens (B) have five nymphal stages, and its wing buds grow gradually along with the increasing stage of the nymph, but the short- and long-winged morphs are externally indistinguishable until the adults emerge. According to environmental cues, bipotential nymphs can alternatively develop into short-winged or long-winged adults. The short-winged morph has underdeveloped forewings and rudimentary hindwings, whereas the long-winged morph has fully developed forewings and hindwings. Arrowhead, hindwing; arrow, forewing.

To address this question, we conducted a comparative functional analysis of IIS components between the firebug and the planthopper. We first applied a multimethod approach, including CRISPR-Cas9–mediated gene editing, RNA interference (RNAi), and pharmacological treatments to systematically characterize the role of key IIS genes in wing-morph determination in both species. Further, we used comparative transcriptomics to identify conserved downstream regulatory modules that might be shared to mediate the divergence in the IIS pathway between these two insects. Last, through liquid chromatography tandem mass spectrometry (LC-MS/MS) and metabolic rescue experiments, we sought to uncover the potential mechanisms that integrate species-specific IIS inputs to ultimately produce a common wing-polyphenic switch.

RESULTS

Inactivation of IIS promotes growth of long wings in the firebug

RNAi-based knockdown of PaInR1a or PaInR2 in the firebug results in long-winged morphs, whereas knockdown of PaInR1b has no effect (25). Although this shows that PaInR1a and PaInR2 play a role in the developmental wing morph switch, the mechanism through which these paralogs to do so remains unclear. There are three possibilities: first, PaInR1a and PaInR2 function as negative regulators of the IIS pathway thereby leading to short wings by inhibiting long wing development. Second, because InR is generally assumed to be a positive regulator of the IIS pathway in metazoans, knockdown of one InR might enhance the activity of the second, which in turn stimulates wing growth. Lastly, PaInR1a and PaInR2 act as positive regulators of the IIS pathway, but the IIS pathway may negatively regulate wing growth. However, this last possibility would contrast sharply with the current paradigm in which IIS signaling promotes organ growth across metazoans.

To distinguish between these possibilities, we assembled a chromosome-level genome of the firebug P. apterus (GenBank accession number: JBBVMJ000000000), which allowed us to create frameshifts of the coding region for each of the three InR paralogs by using CRISPR-Cas9 system and thus obtained PaInR1aIn13, PaInR1bDe5, and PaInR2De4 loss-of-function homozygous mutants (fig. S2 and Supplementary Text). Compared to their wild-type counterparts, which are greater than 99% short-winged morph, PaInR1aIn13 and PaInR2De4 gave rise to 100% long-winged morphs with normally patterned wings (Fig. 2, A and B). PaInR2 regulates long wing development mainly through extending the duration of wing cell proliferation (fig. S3 and Supplementary Text). In contrast, greater than 99% of PaInR1bDe5 mutants developed into short-winged morphs (Fig. 2C), similar to their wild-type counterparts. Notably, double-gene dysfunction experiments revealed that PaInR1a and PaInR2 act independently to regulate the wing-morph switch from short to long, whereas PaInR1b has a neutral effect (fig. S4 and Supplementary Text). Furthermore, we found that PaInR1aIn13and PaInR2De4 nymphs but not PaInR1bDe5 nymphs had longer developmental time to adulthood compared to their wild-type counterparts, indicating that the long wing development might be correlated with extended duration of the nymphal stage (fig. S5). All three PaInR mutants displayed body (fig. S6) and compound eyes (fig. S7) of comparable size to their wild-type counterparts, suggesting that wing growth is particularly sensitive to the expression of PaInR1a and PaInR2.

Fig. 2. Inactivation of IIS redirects the developmental trajectory from short to long wings in the firebug.

Fig. 2.

(A) Wing-morph of PaInR1a mutants (PaInR1aIn13). (B) The wing-morph of PaInR2 mutants (PaInR2De4). PaInR1aIn13 and PaInR2De4 mutants are 100% long-winged (LW) morphs, in contrast of wild-type counterparts that are greater than 99% short-winged (SW) morphs. (C) Wing-morph of PaInR1b mutants (PaInR1bDe5). Similar to wild-type counterparts, greater than 99% of PaInR1bDe5 nymphs develop into short-winged morphs. (D) Wing morphs of short wing–destined the wild-type firebugs (WtSW) after dsRNA treatments. PaChicoRNAi, PaPi3kRNAi, and PaPdk1RNAi, but not PaAktRNAi and PaFoxORNAi, redirects the developmental trajectory from short-to-long wings. ***P < 0.001 (Pearson’s χ2 test: χ2 = 39.955 and df = 1 for PaChicoRNAi, χ2 = 36.146 and df = 1 for PaPi3kRNAi, and χ2 = 19.888 and df = 1 for PaPdk1RNAi). (E) Short-winged morph of GfpRNAi-treated firebugs. (F) IMW morph of PaChicoRNAi-treated WtSW firebugs. (G) IMW morph of PaPi3kRNAi-treated WtSW firebugs. (H) The IMW morph of PaPdk1RNAi-treated WtSW firebugs. (I) Wing morph (short-winged) of PaAktRNAi-treated firebugs. (J) Wing morph (short-winged) of PaFoxORNAi-treated firebugs. (K) PaPtenRNAi but not PaAktRNAi and PaFoxORNAi redirects the developmental trajectory from long-to-short wings in the context of PaInR2De4. ***P < 0.001 (Pearson’s χ2 test: χ2 = 64 and df = 1); ns, not significant. (L) LW PaInR2De4 adults after Gfp knockdown (PaInR2De4::GfpRNAi) and the SW PaInR2De4 pharate adults after PaPten knockdown (PaInR2De4::PaPtenRNAi). (M) SW wild-type pharate adults after PaPten knockdown (WtSW::PaPtenRNAi). Arrows, forewings. Arrowheads, hindwings; IMW, intermediate size wings; WtSW, the short wing–destined wild type.

In the planthopper, knockdown of Chico, Pi3k, Pdk1, and Akt (positive regulators of IIS) switches long to short wings, whereas knockdown of FoxO (a negative transcription factor downstream of IIS) has the opposite phenotype (24, 28–30). To understand how signaling is transmitted by PaInR1a and PaInR2 to regulate wing morphs in the firebug, we knocked down the components acting positively in the IIS pathway using RNAi (data S1). Relative to 100% short-winged morphs of GfpRNAi controls, 85% of PaChicoRNAi, 88% of PaPi3kRNAi, and 54% of PaPdk1RNAi developed into adults with long wings or intermediate-size wings (Fig. 2, D to H). The intermediate size–winged morphs are defined by the range of individuals with hindwings, forewings, or both wings that are greater in size than those of short-winged morphs but smaller than those of long-winged morphs (fig. S3B). These observations closely resemble the characteristic phenotype of PaInR1aIn13 and PaInR2De4 mutants. However, PaAktRNAi or PaFoxORNAi developed into short wings similar to the GfpRNAi controls (Fig. 2, D, I, and J). To rule out that the Ras/mitogen-activated protein kinase signaling, another branch of the IIS pathway, plays a role in regulating wing polyphenism in the firebug, we performed RNAi of Raf and Erk, two main genes within this pathway. We found that knockdown of these genes had no effect on the wing morph developmental switch (fig. S8). Therefore, our findings show that PaInR1a and PaInR2 transmit signals via Chico, PI3K, and PDK1, but not Akt and FoxO, to switch the developmental trajectory from short to long wings. To determine whether the Akt-FoxO axis mediates the IIS activity in regulating the development of additional tissues in the firebug, we measured the phosphorylation level of Akt (P-Akt), an indicator of the IIS activity, in the thorax of PaPi3kRNAi nymphs. We found that inactivation of IIS via PaPi3kRNAi significantly decreases P-Akt (fig. S9), suggesting that the Akt-FoxO axis is indeed involved in tissue development in the fire bug. Together, our results support the third possibility that the IIS pathway negatively regulates wing growth in the firebug, which is contrast to the planthopper and Drosophila, where IIS signaling promotes organ growth.

Hyperactivation of IIS inhibits long wing growth in the firebug

Given that inactivation of the IIS pathway via knockdown of positive regulators (e.g., PaPi3k) switches the developmental trajectory from short to long wings in the firebug (Fig. 2D), we predicted that hyperactivation of this pathway via knockdown of negative regulators would produce the opposite phenotype, leading to short-winged morphs. To test this prediction, we knocked down a negative regulator of IIS PaPten, which is an antagonist of PaPI3K located downstream of PaInR2 in the IIS pathway. Because the PaInR2De4 mutants leads to 100% long- winged morphs (Fig. 2B), knocking down PaPten in the context of PaInR2De4 mutants (PaInR2De4::PaPtenRNAi) should redirect the developmental trajectory from long to short wings. We found that PaInR2De4::PaPtenRNAi all nymphs developed into pharate adults (those that have failed to fully close) with short wings as compared to 100% long-winged morphs of Gfp controls (PaInR2De4::GfpRNAi) (Fig. 2, K and L). To rule out the possibility that the resulting pharate phenotype is due to the PaInR2De4 mutant background, we knocked down PaPten in a wild-type background and found that they also developed into pharate adults (Fig. 2M). In line with the observed short-wing phenotype, PaPten knockdown in PaInR2De4 mutants resulted in an earlier arrest of wing cell proliferation compared to the Gfp-RNAi control (fig. S10), further supporting the role of IIS activation in promoting this phenotype. These findings indicate that PaInR2 is likely mediated by the Chico-PI3K-PDK1 signaling cascade to redirect the developmental trajectory from long to short wings. Together, our results show that the IIS pathway bidirectionally regulates wing-morph switches from short to long and from long to short wings and, therefore, is both necessary and sufficient in regulating wing polyphenism in the firebug.

Last, because Akt and FoxO are not required for the wing-morph switch from short-to-long wings (Fig. 2D), we also asked whether the Akt-FoxO axis is involved in the wing-morph switch from long to short wings. We therefore conducted PaAktRNAi and PaFoxORNAi in the context of the PaInR2De4 mutant background. We found that all PaInR2De4 nymphs developed into long-winged morphs regardless of PaAkt and PaFoxO RNAi (Fig. 2K). These findings confirm that, in the firebug, the inactivation of the IIS pathway promotes long wing growth primarily through the Chico-PI3K-PDK1 cascade rather than the Akt-FoxO axis.

TORC1 mediates the IIS pathway to inhibit long wing growth in the firebug

In model systems such as Drosophila and the mouse, the target of rapamycin complex 1 (TORC1) acts as a positive regulator of cell and organ growth. The TORC1 is a composed of TOR in association with the proteins Raptor and LST8, and it mediates the Chico/PI3K/PDK1/Akt signaling cascade (31). In the planthopper, the inactivation of TORC1 has no effect on the wing-morph developmental switch (24). We therefore investigated, in the firebug, whether the TOR pathway is involved in the wing-morph switch by inactivating TOR activity in two ways: First, we fed wild-type nymphs with rapamycin, a pharmological agent that inhibits the TORC1 activity, and second, PaTor (PaTorRNAi) and its binding partners PaRaptor (PaRaptorRNAi) in the wild-type nymphs were knocked down. We found that at 10 μM concentration of rapamycin, 92% of nymphs developed into long-winged and intermediate size–winged adults compared to 100% short-winged morphs of its H2O control (Fig. 3, A and B). Most PaTorRNAi nymphs failed to develop into adults due to high mortality, and only one of the 25 that survived developed into the long-winged morph compared to its GfpRNAi control (Fig. 3C). Because it remains unclear whether the PaTorRNAi result is due to high mortality or whether there is no effect on the wing morph switch, we performed RNAi of PaRaptor (PaRaptorRNAi), a component of TORC1. We found that, similar to rapamycin treatment, 57% of the PaRaptorRNAi nymphs developed into long-winged and intermediate size–winged adults compared with 100% short-winged morphs of the GfpRNAi control (Fig. 3, C and D). Our results show that, in the firebug, the TOR pathway inhibits wing growth, leading to short-winged morphs.

Fig. 3. The TOR pathway inhibits long wing growth.

Fig. 3.

(A) Wing morphs after rapamycin treatment. **P < 0.01 and ***P < 0.001 (Pearson’s χ2 test: χ2 = 41.59 and df = 1 for 10 μM and χ2 = 7.438 and df = 1 for 25 μM). (B) Phenotype of long-winged and intermediate size–winged firebugs after feeding on rapamycin. (C) Wing morphs of short wing–destined P. apterus (WtSW) after treatment with PaRaptorRNAi and PaTorRNAi. ***P < 0.001 (Pearson’s χ2 test: χ2 = 38.272 and df = 1). (D) Phenotype of PaRaptorRNAi. Both long-winged and intermediate size wings are shown. The intermediate size hindwings in (B) and (D) are indicated by arrowheads. (E) PaTsc1RNAi or PaTsc2RNAi in the context of PaInR2De4 leads to short-winged morphs. PaInR2De4 nymphs develop into long-winged morphs after GfpRNAi. ***P < 0.001 (Pearson’s χ2 test: χ2 = 93 and df = 1 for PaTsc1RNAi and χ2 = 88 and df = 1 for PaTsc2RNAi). (F) Phenotype (short-winged morphs) of PaInR2De4 after PaTsc1RNAi or PaTsc2RNAi.

Since inactivation of TORC1 redirects the developmental trajectory from short to long wings in the firebug, we predicted that hyperactivation of TOR pathway should have the opposite phenotype, redirecting the developmental trajectory from long to short wings. In mammals and Drosophila, Tsc1 and Tsc2 repress TORC1 activity and themselves are repressed by the IIS pathway through PI3K/Akt (32, 33). In the planthopper, the knockdown of Tsc1 or Tsc2 does not affect wing morph switches (fig. S11), consistent with the observation from knockdown of other components (Tor and Raptor) within the TOR pathway (24). In the firebug, because PaInR2De4 mutants lead to 100% long-winged morphs through the Chico-PI3K-PDK1 signaling cascade (Fig. 2D), we asked whether knockdown of PaTsc1 or PaTsc2 in the context of PaInR2De4 mutants (PaInR2De4::PaTsc1RNAi) and (PaInR2De4::PaTsc2RNAi) could redirect the developmental trajectory from long to short wings. We found that all PaInR2De4::PaTsc1RNAi and PaInR2De4::PaTsc2RNAi nymphs developed into short-winged adults (Fig. 3, E and F) compared with GfpRNAi controls (PaInR2De4::GfpRNAi), which were long-winged morphs. Altogether, our results indicate that TOR pathway signaling transmits the PaInR2-Chico-PI3K-PDK1 signaling to regulate alternative wing morphs in the firebug. Furthermore, because Akt is not involved in the wing morph switch in the firebug, we infer that another, yet unknown, regulatory protein may functionally substitute for Akt to repress Tsc1 and Tsc2 activity through mediating the PaInR2-Chico-PI3K-PDK1 signaling activity.

IIS inactivation in the firebug or IIS activation in the planthopper results in low ecdysteroid titers

Thus far, together with our previous studies (24), our findings show that the relationship between the wing morph switch and IIS activity underlying the evolution of wing polyphenism in the firebug and planthoppers is dramatically different. First, the firebug implements IIS inactivation to trigger long wing development, whereas the planthopper employs IIS activation to achieve the same effect. Second, the firebug uses the Chico-PI3K-PDK1-TORC1 cascade to transmit the IIS signal, whereas the planthopper uses the Chico-PI3K-PDK1-Akt-FoxO cascade. One possibility for how firebugs and planthoppers achieve a similar wing-morph switch despite different inputs from the IIS pathway is the integration of these differences by a common downstream modulator.

The firebug PaInR2De4 mutants result in 100% long-winged morphs via inactivation of IIS. In the planthopper, because the noncanonical insulin receptor NlInR2 is a negative regulator of NlInR1, NlInR2E4 mutants result in 100% long-winged morphs via activation of IIS (24, 29). To identify a potential common downstream modulator of IIS for both species, we conducted comparative transcriptomic analyses of PaInR2De4 and NlInR2E4 mutants versus wild type to identify commonly expressed genes involved in long wing development (Fig. 4A, data S2 and S3, and Supplementary Text). We used the fifth-instar stage for our analysis because this stage is the critical period for switching the developmental trajectory of wing morphs in both firebugs (fig. S3A) and planthoppers (30). Our analysis identified “insect hormone biosynthesis” as one of the three shared Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways between the two mutants, which is composed of genes within the ecdysteroid biosynthesis pathway (Fig. 4A and fig. S12). Upon further analysis, we found that different genes within the ecdysteroid biosynthesis pathway are differentially expressed between the mutant and wild type in the firebug and planthopper. In the firebug, Torso, Phantom, and Disembodied are down-regulated (Fig. 4B), whereas in the planthopper, Spook, Neverland, and Cyp314A1 genes are down-regulated (Fig. 4C). To confirm the RNA sequencing (RNA-seq) data, we quantified the expression of the six genes by quantitative real-time polymerase chain reaction (qRT-PCR). Consistent with the RNA-seq data, PaInR2De4 mutants significantly down-regulated the expression of Torso, Phantom, and Disembodied compared wild-type firebugs (fig. S13A). Similar to the firebug, NlInR2E4 mutants significantly down-regulated the expression of Spook, Neverland, and Cyp314A1 compared to wild-type planthoppers (fig. S13B). Together, our results raise the possibility that the ecdysteroid biosynthesis pathway is a common downstream modulator of IIS in both species, which integrates opposing IIS activities likely through different ecdysteroid biosynthesis genes.

Fig. 4. Both activation (in N. lugens) or inactivation (in P. apterus) of the IIS pathway compromises ecdysteroid biosynthesis.

Fig. 4.

(A) Identification of ecdysteroids as a target of IIS by comparative transcriptomics. Fifth-instar WtSW (short-winged–destined) and PaInR2De4 (long-winged–destined) P. apterus nymphs, and fifth-instar WtSW (short-winged–destined) and NlInR2E4 (long-winged–destined) N. lugens nymphs were collected for RNA sequencing (RNA-seq). Three KEGG pathways are commonly regulated in the context of inactivation of IIS in the firebug (via PaInR2De4) and activation of IIS in the planthopper (via NlInR2E4). The ecdysteroid biosynthesis pathway is outlined on the right. (B) Heatmap of gene regulation by PaInR2De4. Three genes key to ecdysteroid biosynthesis are down-regulated in the context of IIS inactivation (via PaInR2De4). (C) Heatmap of gene regulation by NlInR2E4. Three genes key to ecdysteroid biosynthesis are down-regulated in the context of IIS activation (via NlInR2E4). (D) PaInR2De4 firebugs decrease titers of the ecdysteroid makisterone A. Fifth-instar nymphs at 1, 3, 5, 7, and 9 days (d) after eclosion were collected for makisterone A determination. The duration of fifth-instar PaInR2De4 and WtSW nymphs is indicated at the bottom. (E) NlInR2E4 planthoppers decrease titers of the ecdysteroid 20E. Fifth-instar nymphs at 2, 24, 48, and 72 hours (h) after eclosion were collected for 20E determination. The duration of fifth-instar PaInR2De4 and WtSW nymphs is indicated at the bottom. Bars in (D) and (E) represent means ± SEM derived from three independent biological replicates. *P < 0.05, **P < 0.01, and ***P < 0.001 (two-tailed unpaired t test).

To test this possibility, we used LC-MS/MS to determine the titers of the major ecdysteroid molting hormones in firebugs (makisterone A) and planthoppers [20-hydroxyecdysone (20E)] in PaInR2De4 and NlInR2E4 mutants (34, 35). In firebug wild-type controls, makisterone A was present at the basal level for up to 3 days after fifth-instar eclosion, where the peak levels were reached around 5 days after eclosion (Fig. 4D). In contrast, PaInR2De4 mutants showed a significantly lower makisterone A titer than the wild type at 5 days after eclosion, where the timing of the pulse for makisterone A production is delayed to 7 days after eclosion (Fig. 4D). Similar to the firebug PaInR2De4 mutant, NlInR2E4 planthopper mutants had a significantly lower 20E titer than wild-type planthoppers around 48 and 72 hours after fifth-instar eclosion (Fig. 4E). Our results indicate that the ecdysteroid biosynthesis pathway may integrate opposing IIS activities through decreased titers of ecdysteroid molting hormones regardless of whether IIS is inactivated in the firebug or activated in the planthopper.

Exogenous administration of ecdysteroids nullify the IIS effect in wing-morph switch

If low ecdysteroid levels mediate opposing IIS activities in firebugs and planthoppers to develop into long wings, then the exogenous administration of ecdysteroids before the timing of ecdysteroid secretion during the fifth-instar should redirect the development trajectory in both species from long to short wings. We found that injection of makisterone A into PaInR2De4 firebug nymphs at 72 to 84 hours after eclosion significantly redirected their development trajectory from long to short wings as compared to dimethyl sulfoxide (DMSO) controls, which developed into long-winged morphs (Fig. 5, A to C). A similar phenotype was observed in the planthopper after injection with 20E, in which NlInR2E4 nymphs developed into short-winged and intermediate size–winged adults (Fig. 5, D to F). Together, our findings show that ecdysteroid hormones mediate the opposing activities of IIS to regulate the same developmental switch between wing morphs in the firebug and the planthopper.

Fig. 5. Exogenous administration of ecdysteroids nullifies the IIS effect in wing-morph switch.

Fig. 5.

(A) Wing morphs of PaInR2De4 after makisterone A microinjection. **P < 0.01 and ***P < 0.001 (Pearson’s χ2 test: χ2 = 7.501 and df = 1 for 0.1 μg/μl and χ2 = 63.188 and df = 1 for 0.5 μg/μl). (B) Phenotypes of PaInR2De4 microinjected with DMSO (long-winged morphs). (C) Phenotypes of PaInR2De4 microinjected with makisterone A (short-winged morphs). (D) Wing morphs of NlInR2E4 after 20E microinjection. ***P < 0.001 (Pearson’s χ2 test: χ2 = 40.675 and df = 1 for 0.2 μg/μl, χ2 = 32.125 and df = 1 for 1 μg/μl, and χ2 = 32.178 and df = 1 for 2.5 μg/μl). (E) Phenotypes of NlInR2E4 microinjected with DMSO (long-winged morphs). (F) Phenotypes of NlInR2E4 microinjected with 20E (short-winged and intermediate size–winged morphs). Forewings and hindwings are indicated by arrows and arrowheads, respectively.

DISCUSSION

Our study reveals a mechanism for how a rewired signaling pathway with opposing activities can produce a similar developmental outcome across species. The core of this mechanism lies in the ability of a hormone pathway, the ecdysteroid biosynthesis and signaling pathway, to integrate opposite inputs of the IIS pathway in two hemipteran species (the planthopper N. lugens and the firebug P. apterus) to produce the same developmental switch. Our results provide evidence for the following model (Fig. 6). First, two alternative cascades within the IIS pathway in these species lead to the same developmental wing morph switch. In the firebug, the InR-Chico-PI3K-PDK1-Tsc1/2-TORC1 cascade determines the switch from short to long wings by inactivating IIS, whereas in the planthopper, the InR-Chico-PI3K-Akt-FoxO cascade determines this switch by activating IIS. Second, our results suggest that the ecdysteroid biosynthesis pathway integrates the activation or inactivation of IIS across these two species by targeting different genes within this pathway. In the firebug, IIS inactivation results in low ecdysteroid titers by down-regulating three genes (Torso, Phantom, and Disembodied) required for ecdysteroid biosynthesis. By contrast, in the planthopper, IIS activation results in low ecdysteroid titers by down-regulating another three different genes (Neverland, Spook, and Cyp314A1), which are also required for ecdysteroid biosynthesis. Third, in both species, low ecdysteroid titers subsequently direct bipotential nymphs toward a long wing developmental trajectory. Conversely, high ecdysteroid titers inhibit long wing growth, leading to short-winged morphs.

Fig. 6. Proposed model for the underlying mechanisms of wing polyphenism in hemipteran insects.

Fig. 6.

The IIS pathway acts as both a positive and a negative regulator to control ecdysteroid titers in P. apterus and N. lugens via the InRs-Chico-PDK1-Tsc1/2-TORC1 and InRs-Chico-PDK1-Akt-FoxO cascades, respectively. In P. apterus, inactivation of the IIS pathway decreases ecdysteroid (makisterone A) biosynthesis via targeting Torso, Phantom, and Disembodied genes. In N. lugens, activation of the IIS pathway decreases ecdysteroid (20E) biosynthesis via targeting Neverland, Spook, and Cyp314A1 genes. Low ecdysteroid titers may guide the bipotential wing buds to develop into long wings. In contrast, high levels of ecdysteroid titers may guide wing buds to develop into short wings. Down and up arrows in colors indicate inactivation and activation, respectively.

There are several unexpected findings inherent in this model. First, we were surprised to find that the IIS pathway acts as both a positive or negative regulator of organ growth to produce long wings. The current paradigm across metazoans assumes that activation of the IIS pathway positively regulates body and organ growth. For example, in Drosophila, the activation of IIS results in flies with larger body and organ size (15), whereas inactivation results in flies with smaller body and organ size (16–18). Similar to Drosophila, the inactivation of the IIS in zebrafish, mice, and humans also leads to organisms with smaller body and organ size (36–38). The planthopper fits within the current paradigm in that activation of IIS stimulates wing growth and, consequently, directs bipotential nymphs toward long wing development (24). However, we found that inactivation of IIS also promotes organ growth (i.e., long wing development) in the firebug, which is inconsistent with the current paradigm. This antagonistic relationship between IIS and wing growth is mediated by the ecdysteroid biosynthesis and signaling pathway. Thus, the IIS pathway likely indirectly regulates wing growth in the firebug. The scenario differs from the finding in Drosophila, in which tissue-specific manipulation of the IIS pathway directly affects tissue growth. These findings indicate that the relationship between IIS regulation, ecdysteriod hormones, and organ growth is more evolutionarily flexible than previously thought.

Our second unexpected finding is that in the planthopper activation of IIS via the InR-Chico-PI3K-Akt-FoxO cascade results in low ecdysteroid levels. In Drosophila, by contrast, the activation of IIS in the prothoracic gland results in increased levels of ecdysteroids and inactivation of IIS results in decreased levels of ecdysteroids (39). This positive regulation of ecdysteroid biosynthesis by IIS pathway is similar to that found in the firebug but is opposite to that found in the planthopper. We show that the ecdysteroid biosynthesis integrates these opposite IIS effect to achieve the same developmental wing morph switch by regulating different genes within the ecdysteroid biosynthesis pathway between species. Previous studies have shown that in Drosophila, the activation of the IIS/TOR pathways in the prothoracic gland increased the transcriptional level of Phantom and Disembodied, two genes required for ecdysteroid biosynthesis, whereas the activation of negative regulators within the IIS/TOR pathway (such as FoxO, Pten, and Tsc1/2) delays the induction of Phantom and Disembodied (39–42). In the firebug, same as Drosophila, both Phantom and Disembodied are also down-regulated after inactivation of IIS via CRISPR mutation, suggesting that the role of these genes in responding to IIS is evolutionarily conserved. In contrast, a different set of genes of the planthopper are down-regulated (Neverland, Spook, and Cyp314A1) after activation of IIS via CRISPR mutation. Therefore, a second set of targets evolved within the ecdysteroid biosynthesis pathway of the planthopper to integrate IIS inputs, resulting in a similar wing morph developmental switch mediated by low levels of ecdysteroids. Future studies should focus on how genes within the IIS pathway evolved to regulate different genes within the ecdysteroid biosynthesis pathway.

Our results are consistent with the predictions of the “layer cake model” sensu Wagner (7), where signaling pathways and their inputs are predicted to be more evolutionarily flexible or variable than the highly conserved core gene regulatory networks (GRNs) underlying the phenotype. Here, we expand this layer cake model by adding an additional layer that we call the “integrator layer” (Fig. 6). In this case, the integrator layer is the ecdysteroid biosynthesis pathway that transforms opposing IIS inputs (via the targeting different genes in different species) into the same output: low ecdysteriod levels. This promotes wing growth through activation of the wing GRN, resulting in a similar developmental wing morph switch in both species. A key implication for the existence of an integrator layer is the dissociability of these different layers from one another. We revealed one such dissociation in the firebug—once inactivation of the IIS pathway is integrated by the ecdysteroid biosynthesis pathway resulting in low ecdysteroid levels, the IIS pathway is no longer needed (i.e., dissociated) from wing growth, thereby allowing more freedom for the IIS pathway to evolve. This dissociation of layers in this model may on the one hand ensure the stable formation of similar developmental outcomes or switches through the integrator and core GRN layers and, on the other hand, allow a rapid evolution of the transient signaling pathway inputs at the top to respond to and accommodate heterogeneous environments. Wing polyphenism in hemipteran bugs evolved independently in very different ecological environments. For example, the firebug evolved wing polyphenism in trees and bushes, whereas the planthopper evolved it in a gramineous plant environment (fig. S1). Evolutionary flexibility in the IIS pathway may therefore facilitate the processing of different resources and different selective pressures in these different environments, which are subsequently integrated through the ecdysteroid biosynthesis and signaling pathway to produce similar alternative wing morphs.

Our results suggest that the IIS pathway is evolutionary flexible and can be integrated with the ecdysteroid biosynthesis and signaling pathway in the control of other insect polyphenisms in addition to wing polyphenism. For example, in the rhinoceros beetle Trypoxylus dichotomus, InR (a positive regulator within the IIS pathway) is involved in regulating horn size (19), whereas in the dung beetle Onthophagus taurus and Digitonthophagus gazelle, InR has no effect on horn size. Instead, FoxO (a negative regulator within the IIS pathway) is involved in regulating horn size in these dung beetles (43, 44). This indicates that the IIS pathway in beetles has been rewired. Therefore, future studies are needed to test whether the ecdysteroid biosynthesis mediates these differences in IIS activity to regulate horn plasticity across beetles. In addition, most insects rely on dietary sterols as precursors for ecdysteroid biosynthesis. Since insulin signaling is commonly recognized as a nutrient-sensing pathway, it is plausible that insects adjust ecdysteroid production in response to nutritional availability and internal metabolic status, thereby influencing phenotypic plasticity. Furthermore, in insects, the ecdysteroid biosynthesis is potentially regulated by other signaling networks in addition to the IIS pathway, including the PTTH/Torso signaling pathways, transforming growth factor β, dopamine signaling, octopamine signaling, hedgehog signaling, autocrine factors, neuropeptides, and sesquiterpenoid juvenile hormones (45, 46). For example, in the dung beetle O. taurus, the hedgehog signaling pathway was implicated in regulating horn size (47). This raises the possibility that other pathways may be evolutionarily flexible and the ecdysteroid biosynthesis pathway may integrate these other transient signaling pathways across different insect polyphenisms.

More generally, the evolutionary flexibility of signaling pathways extends far beyond the world of insect polyphenisms, and is an emerging pattern across organisms. Notable examples include: the development of the vulva (the egg-laying and copulatory organ) in nematodes (48, 49), dorsal-ventral axis establishment in insects (50), and sex determination in invertebrates and vertebrates (5, 7, 8, 51–53). In the case of nematode worms within the genus Caenorhabditis, the vulva is formed from vulval precursor cells that are regulated by the EGF/Ras and Delta/Notch signaling transduction pathways. Across four different species within Caenorhabditis, components within these signaling pathways have been rewired, yet produce a similar developmental outcome—the vulva (48, 49). Another example is the case of dorsal-ventral (D-V) axis establishment in insects, in which D-V axis formation is regulated by the Toll and BMP signal transduction pathways (50). Although a similar D-V axis is stably produced across insects, the components within these two signal transduction pathways have been altered and rewired. Finally, although sex determination similarly produces male and female sexes across vertebrate species, the signaling genes within the sex determination pathway show notable degree of rewiring across species (51, 53). In each of these cases, however, the mechanisms that mediate this flexibility in a signaling pathway to produce a similar developmental outcome or switch across species remains poorly understood. We predict that our finding of an integrator layer will be a general feature of biological systems that allow for this evolutionary flexibility, whether evolving through divergence, as in the case nematode vulva or sex determination pathways, or independent evolution as is this case of wing polyphenism in hemipterans. Although we identified this integrator layer as a hormone signaling pathway, it remains unclear whether all integrators are hormone signaling pathways or whether integrators can be found in other forms of genetic regulation.

MATERIALS AND METHODS

Insects

The colony of the firebug P. apterus population, which are greater than 99% short-winged morph, was established from colonies collected during 2020 in Urumqi, Xinjiang Uygur autonomous region, China. The P. apterus population was reared at 28° ± 0.5°C under a photoperiod of 16:8 hours (Light:Dark) at a relative humidity of 50 ± 5% on red clover seeds (Trifolium pratense L.) and moistened cotton wools.

The colony of the planthopper N. lugens population, which is greater than 90% short-winged morph, was originally collected from a rice field in Hangzhou, China, in 2008, and used for genome assembly as described in our previous report (54). The long wing–destined NlInR2E4 mutants were constructed and maintained in our laboratory as described previously (29). The N. lugens population was reared at 26° ± 0.5°C under a photoperiod of 16:8 hours (Light:Dark) at a relative humidity of 50 ± 5% on rice seedlings (rice variety: Xiushui 11).

Phylogenetic analysis and divergence time estimation

To infer the phylogenetic relationships of the wing dimorphic insects in hemipteran, protein sequences of the planthoppers (N. lugens, Sogatella furcifera, and Laodelphax striatellus), the water strider G. buenoi, the firebug P. apterus, the pea aphid Acyrthosiphon pisum, and the fruit fly Drosophila melanogaster were downloaded from InsectBase 2.0 (http://v2.insect-genome.com/Genome). The RNA-seq data of the soapberry bug J. haematoloma were downloaded from the National Center for Biotechnology Information (NCBI) and assembled by Trinity (v2.8.5) (55), and the redundancy was removed by the cd-hit (v4.8.1) (56). The amino acid sequence of each protein was deduced using TransDecoder (v5.5.0; https://github.com/TransDecoder/TransDecoder). To remove redundancy caused by alternative splicing, only gene models encoding the longest protein sequence were retained for each gene locus. In total, 169,575 orthologs were identified from all eight species using OrthoFinder v2.5.5 with the default parameters (57). A single-copy orthologous group containing 607 genes was selected and aligned by using MAFFT v7.520 with the default parameters (58). After filtering with trimAl v1.4.1 (59), the maximum likelihood phylogenetic tree was constructed using IQ-TREE v2.2.3 with 1000 replicates for ultrafast bootstrap analysis (60). Divergence time estimation was performed using the TimeTree 5 (61) and mcmctree program in PAML v4.10.7 (62), and the divergence times were inferred and calibrated using the TimeTree database (www.timetree.org/). Last, the phylogenetic tree was visualized and rendered using FigTree v1.4.2.

sgRNAs, CRISPR-Cas9, and embryo injection

The gene structures of PaInR1a, PaInR1b, and PaInR2 were predicted by searching their cDNA sequences against the genome of P. apterus using the Splign algorithm (www.ncbi.nlm.nih.gov/sutils/splign/splign.cgi). The single guide RNA (sgRNA) sequences were designed using CHOPCHOP (http://chopchop.cbu.uib.no/) and prepared as described previously (63) with the corresponding primers (data S4). sgRNAs were in vitro transcribed with a T7 high-yield transcription kit (Vazyme, TR101-01) according to the manufacturer’s instructions.

A mixture of Cas9 proteins (~500 ng/μl) and sgRNAs (~300 ng/μ;) was microinjected into the posterior of pre-blastoderm P. apterus embryos with a FemtoJet microinjection system (Eppendorf, Hamburg, Germany) as described previously (63, 64). After microinjection, embryos were transferred to petri dishes covered with damp filter papers and subsequently placed in a chamber at 28° ± 0.5°C for P. apterus with 50 ± 5% relative humidity until hatching.

Genotyping and establishment of homozygous mutant lines

To determine the genotypes of mutants, antennas from P. apterus were homogenized in a 1.5-ml sterile Eppendorf tube containing 100 μl of DNA extraction buffer [10 mM tris-HCl (pH 8.2), 1 mM EDTA, 25 mM NaCl, and proteinase K (20 μg/ml)] (65). The lysate was then incubated at 37°C for 30 min, followed by 2 min at 95°C to inactivate proteinase K. The precipitate was removed by centrifuging at 12,000g for 2 min at 4°C, and the supernatant was retained for PCR amplification with the corresponding primers (data S4). The PCR products were used directly for Sanger sequencing.

To obtain homozygous mutants, newly enclosed G0 adults were mated with wild-type adults to produce G1 offspring. G1 individuals with the same allelic mutation were sibling-crossed to obtain G2 offspring, followed by self-crossing G2 individuals to generate a homozygous mutant.

Determination of nymphal duration

To measure the duration of nymph development, newly hatched first-instar PaInR1aIn13, PaInR1bDe5, PaInR2De4, and wild-type nymphs were collected and placed in plastic boxes containing red clover seeds. The durations of PaInR1aIn13 (n = 16), PaInR1bDe5 (n = 24), PaInR2De4 (n = 21), and wild-type (n = 17) nymphs were monitored every 12 hours.

Preparation of dsRNA and RNAi

The templates for double-stranded DNA (dsRNA) synthesis were amplified by PCR using primers with the T7 RNA polymerase promoter at both ends (data S4) and confirmed by Sanger sequencing. dsRNA was synthesized and microinjected as described previously (24). Briefly, dsRNAs were synthesized using a T7 high-yield RNA transcription kit (Vazyme, TR101-01) according to the manufacturer’s instructions. The quality and size of dsRNAs were verified by in 1% agarose gel electrophoresis.

Microinjection was performed using a FemtoJet microinjection system (Eppendorf) equipped with a borosilicate glass capillary needle. Nymphs were anesthetized with CO2 for 10 to 15 s. Each fourth- or fifth-instar P. apterus nymph (data S1) was microinjected with about 1 or 2 μg of dsRNAs. After microinjection, all insects were maintained in transparent polycarbonate boxes and provided with red clover seeds. Two days after injection, insects (n = 3 for each of the three replicates) were collected to examine the efficiency of RNAi by qRT-PCR. All emerged adults were collected for calculating wing-morphs ratios.

qRT-PCR assay

Total RNA was isolated from samples using RNAiso Plus (Takara, 9109) according to the manufacturer’s protocol. First-strand cDNA was synthesized from total RNA (600 ng) using HiScript QRT Super Mix (Vazyme, R123-01). The synthesized cDNAs were fivefold diluted and used as templates for qRT-PCR with primers specific to each gene (data S4). qRT-PCR was conducted using a CFX96 real-time PCR detection system (Bio-Rad, Hercules, CA, USA) under the following conditions: denaturation for 3 min at 95°C, followed by 40 cycles at 95°C for 10 s, and then 60°C for 30 s. The ribosomal protein 48 (rp49) and ribosomal protein S11 (rps11) genes were used as the internal reference genes for P. apterus and N. lugens, respectively. The 2−∆∆Ct method (Ct represents the cycle threshold) was used to determine relative expression levels.

Determination of critical period for the SW-LW transition in P. apterus

Fifth-instar wild-type nymphs were collected at 24-hour interval after ecdysis and subjected to dsRNA microinjection. Each individual was microinjected with ~2 μg of dsRNAs targeting Gfp or PaInR2. The dsRNA-treated nymphs were allowed to emerge into adults, and the numbers of P. apterus adults with distinct wing morphs were counted.

Determination of wing cell proliferation

Wing cells were stained with 5-ethynyl-2′-deoxyuridine (EdU) as described previously (29). Briefly, fourth-instar wild-type P. apterus nymphs were microinjected with dsRNAs targeting PaInR2 and Gfp to construct long wing–destined and short wing–destined morphs, respectively. At 72 to 96 hours after eclosion and 96 to 120 hours after eclosion, each individual fifth-instar nymph was microinjected with ~1 μl of EdU (0.5 mM). After 24 hours, the forewings were dissected and fixed in formalin–acetic acid–alcohol solution (5 ml of 37% formaldehyde, 5 ml of acetic acid, 81 ml of ethanol, and 9 ml of H2O) before staining with Click-iT EdU assay (Invitrogen, C10337) according to the manufacturer’s instructions. To determine whether PaPten knockdown affects wing cell proliferation in PaInR2De4 mutants, fifth-instar PaInR2De4 at 0 to 24 hours after eclosion were collected for microinjection with Gfp-RNAi or Pten-RNAi. Four days postmicroinjection, fifth-instar nymphs were subjected to EdU microinjection. Following a 24-hour incorporation period, wing buds were dissected and process for EdU staining. Fluorescent images were acquired using a Zeiss LSM 810 confocal microscope (Carl Zeiss, Oberkochen, Germany).

Western blotting

Fifth-instar wild-type P. apterus nymphs within 24 hours after ecdysis were microinjected with dsRNAs targeting PaPten, PaPi3k_21b, or Gfp. After 72 hours, thoraxes were dissected and homogenized in lysis buffer supplemented with phosphatase inhibitor cocktail (Sigma-Aldrich, P0044). In parallel, thorax tissues (n = 20) dissected from fifth-instar wild-type nymphs (72 to 96 hours after eclosion) were treated with 50 μM PI(3)K inhibitor LY294002 (Sigma-Aldrich, CL9908) for 30 min and subsequently homogenized in lysis buffer supplemented with phosphatase inhibitor cocktail. Equal amounts of protein (100 μg) were loaded per lane on SDS–polyacrylamide gel electrophoresis gels. Western blotting was performed using primary antibodies against phospho-Akt [Cell Signaling Technology (CST), 9271] and Akt (CST, 9272). A β-actin monoclonal antibody (Huabio, EM21002) was used as a loading control. Band intensity was quantified using Image Lab 6.0 (Bio-Rad). Tukey’s multiple comparisons test was performed using GraphPad Prism 8.0.1.

Determination of wing cell size

Fifth-instar nymphs of long wing–destined PaInR2De4 and short wing–destined wild-type nymphs were collected at 48 hours after eclosion, and the forewings (n = 50) were dissected in cold phosphate-buffered saline (PBS) solution. To disassociate cells, the forewings were transferred to 1 × trypsin-EDTA solution (Sigma-Aldrich, T3924) and kept at 37°C for 10 min with pipette mixing every 2 min. The digested cells were washed twice with ice-cold SF9 medium and fixed with 4% paraformaldehyde at room temperature (RT) for 15 min. Following fixation, the cells were stained with Hoechest 33342 (Yeasen, 40731ES10) and tetramethyl rhodamine isothiocyanate–labeled phalloidin (Yeasen, 40734ES75) in PBS for 5 min at RT. Fluorescence images were acquired using a Zeiss LSM 810 confocal microscope (Carl Zeiss). The diameters of forewing cells from PaInR2De4 and wild-type were measured using ImageJ (v. 1.47). Plotting and statistical analyses were performed with GraphPad Prism (v8.0.1).

Rapamycin feeding assay

Rapamycin (MedChemExpress, HY-10219) was first dissolved in DMSO before adding sterile water to obtain concentrations of 5, 10, and 25 μM. Cotton wool was dipped in the rapamycin solution and then provided as the water source for fourth-instar P. apterus nymphs. Cotton wool rinsed with water only was used as the control. All emerged adults were collected for calculating wing-morph ratios.

RNA-seq of PaInR2De4 and NlInR24E4 mutants

To screen for potential wing-morph regulator in P. apterus and N. lugens, thoraxes and abdomens with guts removed were dissected from fifth-instar PaInR2De4 and wild-type P. apterus nymphs at 48 to 72 hours after eclosion, and from fifth-instar NlInR2E4 and wild-type N. lugens nymphs at 36 to 48 hours after eclosion. Samples were homogenized, and the total RNA was extracted using RNAiso Plus (Takara, 9109) according to the manufacturer’s protocol. The quality and integrity of RNA were assessed using an RNA Nano 6000 assay kit for the Agilent Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA, USA).

For RNA-seq analysis, 1 μg of RNA per sample was used to construct a sequencing library using a VAHTS universal V6 RNA-seq library prep kit (Vazyme, NR604-01) for Illumina following the manufacturer’s recommendations. The quality of the constructed library was assessed by using the Agilent Bioanalyzer 2100 system. Subsequently, the cDNA libraries were sequenced with the Illumina NovaSeq 6000 platform from Annoroad Gene Technology Co., Ltd. (Beijing, China), and 150–base pair paired-end reads were generated.

Read mapping and DEG analysis

After Illumina sequencing, clean reads were obtained from raw reads by trimming adaptors, poly-N, and low-quality reads. Subsequently, the clean reads were aligned to the P. apterus and N. lugens (66) reference genomes using Hisat2 (v2.1.0) (67), and transcript abundances were quantified with StringTie (v1.3.5) (68).

The expression level of each transcript was quantified as the Fragments Per Kilobase of transcript per million mapped reads (FPKM). DESeq2 package (69) was used for differentially expressed gene (DEG) analysis, and gene expression changes were considered to be significant when the adjusted P value < 0.05 and |log2 fold change| > 1. KEGG pathway enrichment analysis was performed using the online OmicShare tool (www.omicshare.com/tools/home/report/goenrich.html). For heatmap, the expression level of each gene was shown after z-score normalization.

Determination of 20E and makisterone A titers by LC-MS/MS

Ecdysteroids were isolated from N. lugens and P. apterus as described in previous studies with minor modifications (70, 71). Briefly, fifth-instar NlInR2E4 and wild-type N. lugens nymphs were collected at 2, 24, 48, and 72 hours after eclosion. Thirty-five individuals were pooled for each of the three replicates. Samples were homogenized in 200 μl of sonic buffer (50 mM tris, 150 mM NaCl, and 2 mM EGTA, pH = 7.5), followed by sonication for 10 min in ice water. Debris was removed by centrifugation at 1000g for 5 min at 4°C, and the supernatant was mixed with two times the volume of 1-butanol. The solution was mixed by shaking violently for 5 min at RT. After centrifugation at 1000g for 10 min, the upper phase of the solution was retained for 20E quantification. To efficiently isolate 20E, the lower phase of the solution was mixed with 1-butanol again, and the isolation process was repeated. The solvent was evaporated in a centrifugal evaporator (Eppendorf), and the residue was redissolved in 100 μl of 50% methanol. After centrifugation at 10,000g for 10 min at 4°C, the supernatant was transferred to a glass vial for LC-MS/MS analysis.

To isolate makisterone A from P. apterus, fifth-instar PaInR2De4 and wild-type P. apterus nymphs were collected at designed times (1, 3, 5, 7, and 9 days after eclosion). Three nymphs were pooled for each of the three replicates. Samples were homogenized and sonicated in 400 μl of sonic buffer for 20 min. Subsequently, the solution was mixed with times the volume of 1-butanol and shaken violently for 5 min before sonicating for 20 min. After centrifugation at 1000g for 10 min, the supernatant was retained and the debris was mixed with 1-butanol again before repeating the isolation process. The solvent obtained was blow dried with nitrogen gas before dissolution in 500 μl of 70% methanol. After centrifugation at 16,000g for 10 min at 4°C, the supernatant was transferred to a glass vial for LC-MS/MS analysis.

The separation of 20E or makisterone A in aliquots was conducted using an Acquity Premier Xevo TQ Absolute system (Waters, MA, USA) with an Atlantis Premier BEH C18 chromatographic column (1.7 μm, 2.1 mm by 100 mm). The mobile phase comprised ultrapure water (0.1% formic acid) in phase A and acetonitrile in the phase B. The elution gradient was 2 to 95% B for 14 min. The flow rate was set at 0.3 ml/min. Standard 20E and makisterone A were purchased from MedChemExpress (HY-N6979) and Macklin (M912023), respectively. The concentrations of the components corresponding to 20E and makisterone A were calculated on the basis of the peak areas obtained by LC-MS/MS analysis.

Microinjection with makisterone A and 20E

To microinject makisterone A into PaInR2De4 P. apterus mutants, fifth-instar nymphs at 72 to 84 hours after eclosion were microinjected with ~0.3 μl of makisterone A (Yuanye Biotech, B74700) solution (0.1 and 0.5 μg/μl) three times at 24-hour intervals. In parallel, nymphs microinjected with 0.2% DMSO solution served as a control.

To treat NlInR2E4 N. lugens mutants with 20E, fifth-instar nymphs at 0 to 24 hours after eclosion were microinjected with ~0.1 μl of 20E solution (0.2, 1, and 2.5 μg/μl), followed by another two times of microinjection at 12-hour intervals. In parallel, nymphs microinjected with 0.4% DMSO solution were used as the control. Wing morphs were counted after adult emergence.

Image acquisition and statistical analyses

Images of insects were acquired using a DVM6A digital microscope (Leica Microsystems, Germany) with LAS X software. Differences in wing ratios between two strains were statistically compared using Pearson’s χ2 test. Differences in qRT-PCR data, ecdysteroid titers, developmental durations, and cell sizes were analyzed using two-tailed Student’s t tests. Differences in comparisons were considered significant at *P < 0.05, **P < 0.01, or ***P < 0.001.

Acknowledgments

We thank X. Wu and M. Fang from the Zhejiang University for the technical assistance in the LC-MS/MS analysis and C. Zhang from the Ningbo University for the valuable suggestions.

Funding:

This work was funded by the National Key Research and Development Program of China 2023YFF1000500 (to H.-J.X.) and 2021YFD1401100 (to H.-J.X.) and the National Natural Science Foundation of China 32272519 (to H.-J.X. and J.-L.Z.) and 32472541 (to H.-J.X.).

Author contributions:

Conceptualization: Q.-S.S., E.A., and H.-J.X. Methodology: J.-L.Z., Y.-S.L., Q.-S.S., and H.-J.X. Investigation: J.-L.Z., Y.-K.L., Y.W., H.-J.W., N.-T.X., Z.-Q.L., H.-G.H., J.-P.Y., X.-Y.L., X.-M.L., and Y.-S.L. Visualization: J.-L.Z. and H.-J.X. Resources: H.-J.X. Validation: J.-L.Z. and H.-J.X. Data curation: J.-L.Z. Formal analysis: J.-L.Z. Supervision: H.-J.X. Project administration: H.-J.X. Writing—original draft: E.A. and H.-J.X. Writing—review and editing: Q.-S.S., F.H.N., E.A., and H.-J.X.

Competing interests:

The authors declare that they have no competing interests.

Data and materials availability:

All data and code needed to evaluate and reproduce the results in the paper are present in the paper and/or the Supplementary Materials. NCBI accession numbers have been acquired for sequences generated in this study and are listed as follows: KX087103.1 (PaInR1a; www.ncbi.nlm.nih.gov/nuccore/KX087103.1), KX087104.1 (PaInR1b; www.ncbi.nlm.nih.gov/nuccore/KX087104.1), KX087105.1 (PaInR2; www.ncbi.nlm.nih.gov/nuccore/KX087105.1), PP813829 (PaChico; www.ncbi.nlm.nih.gov/nuccore/PP813829), PP813830 (the subunit p60 of PaPi3k; www.ncbi.nlm.nih.gov/nuccore/PP813830), PP813831 (PaPten; www.ncbi.nlm.nih.gov/nuccore/PP813831), PP813841 (PaPdk1; www.ncbi.nlm.nih.gov/nuccore/PP813841), PP813832 (PaAkt; www.ncbi.nlm.nih.gov/nuccore/PP813832), PP813833 (PaFoxO; www.ncbi.nlm.nih.gov/nuccore/PP813833), PP813834 (PaTsc1; www.ncbi.nlm.nih.gov/nuccore/PP813834), PP813835 (PaTsc2; www.ncbi.nlm.nih.gov/nuccore/PP813835); PP813836 (PaTor; www.ncbi.nlm.nih.gov/nuccore/PP813836), PP813837 (PaRaptor; www.ncbi.nlm.nih.gov/nuccore/PP813837), PP813838 (PaRaf; www.ncbi.nlm.nih.gov/nuccore/PP813838), PP813839 (PaErk; www.ncbi.nlm.nih.gov/nuccore/PP813839), PRJNA1107695 (The RNAseq data of P. apterus and N. lugens; www.ncbi.nlm.nih.gov/bioproject/PRJNA1107695), PRJNA1091548 (the raw sequencing genomic data of P. apterus; www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA1091548), and JBBVMJ000000000 (the genome assembly data of P. apterus; www.ncbi.nlm.nih.gov/nuccore/JBBVMJ000000000.1/). This study did not generate any unique code or materials.

Supplementary Materials

The PDF file includes:

Supplementary Text

Figs. S1 to S13

Legends for data S1 to S4

sciadv.aea4413_sm.pdf (2.3MB, pdf)

Other Supplementary Material for this manuscript includes the following:

Data S1 to S4

REFERENCES

  • 1.Karin M., Smeal T., Control of transcription factors by signal transduction pathways: The beginning of the end. Trends Biochem. Sci. 17, 418–422 (1992). [DOI] [PubMed] [Google Scholar]
  • 2.Blüthgen N., Legewie S., Robustness of signal transduction pathways. Cell. Mol. Life Sci. 70, 2259–2269 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.M. Barresi, S. Gilbert, Developmental Biology (Oxford Univ. Press, 2023). [Google Scholar]
  • 4.Pires-daSilva A., Sommer R. J., The evolution of signalling pathways in animal development. Nat. Rev. Genet. 4, 39–49 (2003). [DOI] [PubMed] [Google Scholar]
  • 5.Wagner G. P., The developmental genetics of homology. Nat. Rev. Genet. 8, 473–479 (2007). [DOI] [PubMed] [Google Scholar]
  • 6.Halfon M. S., Perspectives on gene regulatory network evolution. Trends Genet. 33, 436–447 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.G. P. Wagner, “Devo-Evo of cell types,” in Evolution Developmental Biology: A Reference Guide, L. Nuño de la Rosa, G. B. Müller, Eds. (Springer Nature, 2021), pp. 511–528. [Google Scholar]
  • 8.McColgan Á., DiFrisco J., Understanding developmental system drift. Development 151, dev203054 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Ma S., Meng Z., Chen R., Guan K. L., The Hippo pathway: Biology and pathophysiology. Annu. Rev. Biochem. 88, 577–604 (2019). [DOI] [PubMed] [Google Scholar]
  • 10.Philips R. L., Wang Y., Cheon H., Kanno Y., Gadina M., Sartorelli V., Horvath C. M., Darnell J. E. Jr., Stark G. R., O’Shea J. J., The JAK-STAT pathway at 30: Much learned, much more to do. Cell 185, 3857–3876 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Rim E. Y., Clevers H., Nusse R., The Wnt pathway: From Signaling mechanisms to synthetic modulators. Annu. Rev. Biochem. 91, 571–598 (2022). [DOI] [PubMed] [Google Scholar]
  • 12.Taniguchi C. M., Emanuelli B., Kahn C. R., Critical nodes in signalling pathways: Insights into insulin action. Nat. Rev. Mol. Cell Biol. 7, 85–96 (2006). [DOI] [PubMed] [Google Scholar]
  • 13.Edgar B. A., How flies get their size: Genetics meets physiology. Nat. Rev. Genet. 7, 907–916 (2006). [DOI] [PubMed] [Google Scholar]
  • 14.Hietakangas V., Cohen S. M., Regulation of tissue growth through nutrient sensing. Annu. Rev. Genet. 43, 389–410 (2009). [DOI] [PubMed] [Google Scholar]
  • 15.Gao X., Neufeld T. P., Pan D., Drosophila PTEN regulates cell growth and proliferation through PI3K-dependent and -independent pathways. Dev. Biol. 221, 404–418 (2000). [DOI] [PubMed] [Google Scholar]
  • 16.Böhni R., Riesgo-Escovar J., Oldham S., Brogiolo W., Stocker H., Andruss B. F., Beckingham K., Hafen E., Autonomous control of cell and organ size by CHICO, a Drosophila homolog of vertebrate IRS1-4. Cell 97, 865–875 (1999). [DOI] [PubMed] [Google Scholar]
  • 17.Verdu J., Buratovich M. A., Wilder E. L., Birnbaum M. J., Cell-autonomous regulation of cell and organ growth in Drosophila by Akt/PKB. Nat. Cell Biol. 1, 500–506 (1999). [DOI] [PubMed] [Google Scholar]
  • 18.Brogiolo W., Stocker H., Ikeya T., Rintelen F., Fernandez R., Hafen E., An evolutionarily conserved function of the Drosophila insulin receptor and insulin-like peptides in growth control. Curr. Biol. 11, 213–221 (2001). [DOI] [PubMed] [Google Scholar]
  • 19.Emlen D. J., Warren I. A., Johns A., Dworkin I., Lavine L. C., A mechanism of extreme growth and reliable signaling in sexually selected ornaments and weapons. Science 337, 860–864 (2012). [DOI] [PubMed] [Google Scholar]
  • 20.Corona M., Libbrecht R., Wheeler D. E., Molecular mechanisms of phenotypic plasticity in social insects. Curr. Opin. Insect Sci. 13, 55–60 (2016). [DOI] [PubMed] [Google Scholar]
  • 21.Nijhout H. F., McKenna K. Z., The distinct roles of insulin signalling in polyphenic development. Curr. Opin. Insect Sci. 25, 58–64 (2018). [DOI] [PubMed] [Google Scholar]
  • 22.Nijhout H. F., Development and evolution of adaptive polyphenisms. Evol. Dev. 5, 9–18 (2003). [DOI] [PubMed] [Google Scholar]
  • 23.Zhang C. X., Brisson J. A., Xu H. J., Molecular mechanisms of wing polymorphism in insects. Annu. Rev. Entomol. 64, 297–314 (2019). [DOI] [PubMed] [Google Scholar]
  • 24.Xu H. J., Xue J., Lu B., Zhang X. C., Zhuo J. C., He S. F., Ma X. F., Jiang Y. Q., Fan H. W., Xu J. Y., Ye Y. X., Pan P. L., Li Q., Bao Y. Y., Nijhout H. F., Zhang C. X., Two insulin receptors determine alternative wing morphs in planthoppers. Nature 519, 464–467 (2015). [DOI] [PubMed] [Google Scholar]
  • 25.Smýkal V., Pivarči M., Provazník J., Bazalová O., Jedlička P., Lukšan O., Horák A., Vaněčková H., Beneš V., Fiala I., Hanus R., Doležel D., Complex evolution of insect insulin receptors and homologous decoy receptors, and functional significance of their multiplicity. Mol. Biol. Evol. 37, 1775–1789 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Fawcett M. M., Parks M. C., Tibbetts A. E., Swart J. S., Richards E. M., Vanegas J. C., Cenzer M., Crowley L., Simmons W. R., Hou W. S., Angelini D. R., Manipulation of insulin signalling phenocopies evolution of a host-associated polyphenism. Nat. Commun. 9, 1699 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Gudmunds E., Narayanan S., Lachivier E., Duchemin M., Khila A., Husby A., Photoperiod controls wing polyphenism in a water strider independently of insulin receptor signalling. Proc. Biol. Sci. 289, 20212764 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Zhang J.-L., Liu K.-L., Cai X.-Y., Liu X.-Y., Xu H.-J., FoxO is required for optimal fitness of the migratory brown planthopper, Nilaparvata lugens (Hemiptera: Delphacidae). Insect Sci. 30, 1352–1362 (2022). [DOI] [PubMed] [Google Scholar]
  • 29.Xue W.-H., Xu N., Chen S.-J., Liu X.-Y., Zhang J.-L., Xu H.-J., Neofunctionalization of a second insulin receptor gene in the wing-dimorphic planthopper, Nilaparvata lugens. PLOS Genet. 17, e1009653 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Zhang J.-L., Fu S.-J., Chen S.-J., Chen H.-H., Liu Y.-L., Liu X.-Y., Xu H.-J., Vestigial mediates the effect of insulin signalling pathway on wing-morph switching in planthoppers. PLOS Genet. 17, e1009312 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Martin D. E., Hall M. N., The expanding TOR signalling network. Curr. Opin. Cell Biol. 17, 158–166 (2005). [DOI] [PubMed] [Google Scholar]
  • 32.Gao X., Zhang Y., Arrazola P., Hino O., Kobayashi T., Yeung R. S., Ru B., Pan D., Tsc tumour suppressor proteins antagonize amino-acid-TOR signalling. Nat. Cell Biol. 4, 699–704 (2002). [DOI] [PubMed] [Google Scholar]
  • 33.Inoki K., Li Y., Zhu T., Wu J., Guan K. L., TSC2 is phosphorylated and inhibited by Akt and suppresses mTOR signalling. Nat. Cell Biol. 4, 648–657 (2002). [DOI] [PubMed] [Google Scholar]
  • 34.Kaplanis J. N., Dutky S. R., Robbins W. E., Thompson M. J., Lindquist E. L., Horn D. H., Galbraith M. N., Makisterone A: A 28-carbon hexahydroxy molting hormone from the embryo of the milkweed bug. Science 190, 681–682 (1975). [DOI] [PubMed] [Google Scholar]
  • 35.Kelly T. J., Aldrich J. R., Woods C. W., Borkovec A. B., Makisterone A: Its distribution and physiological role as the molting hormone of true bugs. Experientia 40, 996–997 (1984). [Google Scholar]
  • 36.Louvi A., Accili D., Efstratiadis A., Growth-promoting interaction of IGF-II with the insulin receptor during mouse embryonic development. Dev. Biol. 189, 33–48 (1997). [DOI] [PubMed] [Google Scholar]
  • 37.Abuzzahab M. J., Schneider A., Goddard A., Grigorescu F., Lautier C., Keller E., Kiess W., Klammt J., Kratzsch J., Osgood D., Pfäffle R., Raile K., Seidel B., Smith R. J., Chernausek S. D., IGF-I receptor mutations resulting in intrauterine and postnatal growth retardation. N. Engl. J. Med. 349, 2211–2222 (2003). [DOI] [PubMed] [Google Scholar]
  • 38.Schlueter P. J., Peng G., Westerfield M., Duan C., Insulin-like growth factor signalling regulates zebrafish embryonic growth and development by promoting cell survival and cell cycle progression. Cell Death Differ. 14, 1095–1105 (2007). [DOI] [PubMed] [Google Scholar]
  • 39.Colombani J., Bianchini L., Layalle S., Pondeville E., Dauphin-Villemant C., Antoniewski C., Carré C., Noselli S., Léopold P., Antagonistic actions of ecdysone and insulins determine final size in Drosophila. Science 310, 667–670 (2005). [DOI] [PubMed] [Google Scholar]
  • 40.Layalle S., Arquier N., Leopold P., The TOR pathway couples nutrition and developmental timing in Drosophila. Dev. Cell 15, 568–577 (2008). [DOI] [PubMed] [Google Scholar]
  • 41.Mirth C., Truman J. W., Riddiford L. M., The role of the prothoracic gland in determining critical weight to metamorphosis in Drosophila melanogaster. Curr. Biol. 15, 1796–1807 (2005). [DOI] [PubMed] [Google Scholar]
  • 42.Koyama T., Rodrigues M. A., Athanasiadis A., Shingleton A. W., Mirth C. K., Nutritional control of body size through FoxO-Ultraspiracle mediated ecdysone biosynthesis. eLife 3, e03091 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Casasa S., Moczek A. P., Insulin signalling’s role in mediating tissue-specific nutritional plasticity and robustness in the horn-polyphenic beetle Onthophagus taurus. Proc. Biol. Sci. 285, 20181631 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Rohner P. T., Casasa S., Moczek A. P., Assessing the evolutionary lability of insulin signalling in the regulation of nutritional plasticity across traits and species of horned dung beetles. J. Evol. Biol. 36, 1641–1648 (2023). [DOI] [PubMed] [Google Scholar]
  • 45.Yamanaka N., Rewitz K. F., O’Connor M. B., Ecdysone control of developmental transitions: Lessons from Drosophila research. Annu. Rev. Entomol. 58, 497–516 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Pan X., Connacher R. P., O’Connor M. B., Control of the insect metamorphic transition by ecdysteroid production and secretion. Curr. Opin. Insect Sci. 43, 11–20 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Kijimoto T., Moczek A. P., Hedgehog signalling enables nutrition-responsive inhibition of an alternative morph in a polyphenic beetle. Proc. Natl. Acad. Sci. U.S.A. 113, 5982–5987 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Félix M. A., Cryptic quantitative evolution of the vulva intercellular signaling network in Caenorhabditis. Curr. Biol. 17, 103–114 (2007). [DOI] [PubMed] [Google Scholar]
  • 49.Hoyos E., Kim K., Milloz J., Barkoulas M., Pénigault J. B., Munro E., Félix M. A., Quantitative variation in autocrine signalling and pathway crosstalk in the Caenorhabditis vulval network. Curr. Biol. 21, 527–538 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Roth S., Neofunctionalization of Toll signalling in insects: From immunity to dorsoventral patterning. Annu. Rev. Cell Dev. Biol. 39, 1–22 (2023). [DOI] [PubMed] [Google Scholar]
  • 51.Graham P., Penn J. K., Schedl P., Masters change, slaves remain. Bioessays 25, 1–4 (2003). [DOI] [PubMed] [Google Scholar]
  • 52.Bachtrog D., Mank J. E., Peichel C. L., Kirkpatrick M., Otto S. P., Ashman T. L., Hahn M. W., Kitano J., Mayrose I., Ming R., Perrin N., Ross L., Valenzuela N., Vamosi J. C., Tree of sex consortium, sex determination: Why so many ways of doing it? PLOS Biol. 12, e1001899 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Bertho S., Herpin A., Schartl M., Guiguen Y., Lessons from an unusual vertebrate sex-determining gene. Philos. Trans. R. Soc. Lond. B Biol. Sci. 376, 20200092 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Xue J., Zhou X., Zhang C. X., Yu L. L., Fan H. W., Wang Z., Xu H. J., Xi Y., Zhu Z. R., Zhou W. W., Pan P. L., Li B. L., Colbourne J. K., Noda H., Suetsugu Y., Kobayashi T., Zheng Y., Liu S., Zhang R., Liu Y., Luo Y. D., Fang D. M., Chen Y., Zhan D. L., Lv X. D., Cai Y., Wang Z. B., Huang H. J., Cheng R. L., Zhang X. C., Lou Y. H., Yu B., Zhuo J. C., Ye Y. X., Zhang W. Q., Shen Z. C., Yang H. M., Wang J., Wang J., Bao Y. Y., Cheng J. A., Genomes of the rice pest brown planthopper and its endosymbionts reveal complex complementary contributions for host adaptation. Genome Biol. 15, 521 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Grabherr M. G., Haas B. J., Yassour M., Levin J. Z., Thompson D. A., Amit I., Adiconis X., Fan L., Raychowdhury R., Zeng Q., Chen Z., Mauceli E., Hacohen N., Gnirke A., Rhind N., di Palma F., Birren B. W., Nusbaum C., Lindblad-Toh K., Friedman N., Regev A., Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat. Biotechnol. 29, 644–652 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Fu L., Niu B., Zhu Z., Wu S., Li W., CD-HIT: Accelerated for clustering the next-generation sequencing data. Bioinformatics 28, 3150–3152 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Emms D. M., Kelly S., OrthoFinder: Solving fundamental biases in whole genome comparisons dramatically improves orthogroup inference accuracy. Genome Biol. 16, 157 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Katoh K., Standley D. M., MAFFT multiple sequence alignment software version 7: Improvements in performance and usability. Mol. Biol. Evol. 30, 772–780 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Capella-Gutiérrez S., Silla-Martínez J. M., Gabaldón T., trimAl: A tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics 25, 1972–1973 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Minh B. Q., Schmidt H. A., Chernomor O., Schrempf D., Woodhams M. D., von Haeseler A., Lanfear R., IQ-TREE 2: New models and efficient methods for phylogenetic inference in the genomic era. Mol. Biol. Evol. 37, 1530–1534 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Kumar S., Suleski M., Craig J. M., Kasprowicz A. E., Sanderford M., Li M., Stecher G., Hedges S. B., TimeTree 5: An expanded resource for species divergence times. Mol. Biol. Evol. 39, msac174 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Yang Z., PAML 4: Phylogenetic analysis by maximum likelihood. Mol. Biol. Evol. 24, 1586–1591 (2007). [DOI] [PubMed] [Google Scholar]
  • 63.Xue W. H., Xu N., Yuan X. B., Chen H. H., Zhang J. L., Fu S. J., Zhang C. X., Xu H. J., CRISPR/Cas9-mediated knockout of two eye pigmentation genes in the brown planthopper, Nilaparvata lugens (Hemiptera: Delphacidae). Insect Biochem. Mol. Biol. 93, 19–26 (2018). [DOI] [PubMed] [Google Scholar]
  • 64.Kotwica-Rolinska J., Chodakova L., Chvalova D., Kristofova L., Fenclova I., Provaznik J., Bertolutti M., Wu B. C., Dolezel D., CRISPR/Cas9 genome editing introduction and optimization in the non-model insect Pyrrhocoris apterus. Front. Physiol. 10, 891 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Reding K., Pick L., High-efficiency CRISPR/Cas9 mutagenesis of the white gene in the milkweed bug Oncopeltus fasciatus. Genetics 215, 1027–1037 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Ye Y. X., Zhang H. H., Li D. T., Zhuo J. C., Shen Y., Hu Q. L., Zhang C. X., Chromosome-level assembly of the brown planthopper genome with a characterized Y chromosome. Mol. Ecol. Resour. 21, 1287–1298 (2021). [DOI] [PubMed] [Google Scholar]
  • 67.Kim D., Paggi J. M., Park C., Bennett C., Salzberg S. L., Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat. Biotechnol. 37, 907–915 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Pertea M., Pertea G. M., Antonescu C. M., Chang T. C., Mendell J. T., Salzberg S. L., StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat. Biotechnol. 33, 290–295 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Love M. I., Huber W., Anders S., Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Nakaoka T., Iga M., Yamada T., Koujima I., Takeshima M., Zhou X., Suzuki Y., Ogihara M. H., Kataoka H., Deep sequencing of the prothoracic gland transcriptome reveals new players in insect ecdysteroidogenesis. PLOS ONE 12, e0172951 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Zhou X., Ye Y. Z., Ogihara M. H., Takeshima M., Fujinaga D., Liu C. W., Zhu Z., Kataoka H., Bao Y. Y., Functional analysis of ecdysteroid biosynthetic enzymes of the rice planthopper, Nilaparvata lugens. Insect Biochem. Mol. Biol. 123, 103428 (2020). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Text

Figs. S1 to S13

Legends for data S1 to S4

sciadv.aea4413_sm.pdf (2.3MB, pdf)

Data S1 to S4

Data Availability Statement

All data and code needed to evaluate and reproduce the results in the paper are present in the paper and/or the Supplementary Materials. NCBI accession numbers have been acquired for sequences generated in this study and are listed as follows: KX087103.1 (PaInR1a; www.ncbi.nlm.nih.gov/nuccore/KX087103.1), KX087104.1 (PaInR1b; www.ncbi.nlm.nih.gov/nuccore/KX087104.1), KX087105.1 (PaInR2; www.ncbi.nlm.nih.gov/nuccore/KX087105.1), PP813829 (PaChico; www.ncbi.nlm.nih.gov/nuccore/PP813829), PP813830 (the subunit p60 of PaPi3k; www.ncbi.nlm.nih.gov/nuccore/PP813830), PP813831 (PaPten; www.ncbi.nlm.nih.gov/nuccore/PP813831), PP813841 (PaPdk1; www.ncbi.nlm.nih.gov/nuccore/PP813841), PP813832 (PaAkt; www.ncbi.nlm.nih.gov/nuccore/PP813832), PP813833 (PaFoxO; www.ncbi.nlm.nih.gov/nuccore/PP813833), PP813834 (PaTsc1; www.ncbi.nlm.nih.gov/nuccore/PP813834), PP813835 (PaTsc2; www.ncbi.nlm.nih.gov/nuccore/PP813835); PP813836 (PaTor; www.ncbi.nlm.nih.gov/nuccore/PP813836), PP813837 (PaRaptor; www.ncbi.nlm.nih.gov/nuccore/PP813837), PP813838 (PaRaf; www.ncbi.nlm.nih.gov/nuccore/PP813838), PP813839 (PaErk; www.ncbi.nlm.nih.gov/nuccore/PP813839), PRJNA1107695 (The RNAseq data of P. apterus and N. lugens; www.ncbi.nlm.nih.gov/bioproject/PRJNA1107695), PRJNA1091548 (the raw sequencing genomic data of P. apterus; www.ncbi.nlm.nih.gov/bioproject/?term=PRJNA1091548), and JBBVMJ000000000 (the genome assembly data of P. apterus; www.ncbi.nlm.nih.gov/nuccore/JBBVMJ000000000.1/). This study did not generate any unique code or materials.


Articles from Science Advances are provided here courtesy of American Association for the Advancement of Science

RESOURCES