An insulin-like peptide mediates trehalose storage in the western tarnished plant bug, Lygus hesperus

Abstract

Insulin signaling controls many physiological processes in insects, and it has a demonstrated role in cellular uptake of circulating sugars. Although model insects have pioneered much of our understanding of insulin signaling, high throughput genetic sequencing has enabled opportunities for the physiological study of less explored insect species. Lygus hesperus, the western tarnished plant bug, is a significant agricultural pest of numerous crops and recent efforts have focused on molecular approaches for identifying new pest management strategies. In this work, three insulin-like peptides (LhILP1, LhILP2, and LhILP3) were characterized from the transcriptome of L. hesperus. LhILP1 and LhILP2 structurally resemble classic insulin-like peptides while LhILP3 resembles arthropod Insulin-like Growth Factors (aIGFs). All three LhILPs were primarily expressed in the head, and were observed throughout development. We examined their function by observing clearance rates of injected trehalose from the hemolymph in adult L. hesperus males following RNAi knockdown of the respective LhILPs. Untreated males cleared the trehalose within 6 h and showed a corresponding increase in glycogen content. While knockdown of LhILP1 did not impact clearance rate, knockdown of LhILP2 prevented clearance of circulating trehalose and glycogen accumulation. Knockdown of LhILP3 also prevented trehalose clearance, but this appears to be influenced by co-knockdown of LhILP2, rather than a direct effect. Head ligation stopped the release of head-produced LhILPs, preventing the clearance of injected trehalose, a condition that could be rescued with co-injection of heterologous insulin. Stage-specific expression in LhILP RNAi insects suggest that LhILPs have multiple additional roles besides hemolymph carbohydrate homeostasis in L. hesperus that remain to be explored.

1. Introduction

Insulin-like peptides (ILPs) are involved in multiple functions in invertebrates, including metabolism, reproduction, growth, immunity, behavior, diapause, and lifespan (Okamoto and Yamanaka, 2015; Biglou et al., 2021; Chowanski et al., 2021). Insulin signaling has been widely demonstrated as a signal for the cellular uptake of circulating hemolymph sugars, which in insects is principally trehalose (Wyatt and Kalf, 1957; Matsuda et al., 2015; Leyria et al., 2021a,b). ILP signaling is particularly well studied and understood in Drosophila melanogaster (Grönke et al., 2010; Biglou et al., 2021), yet ILPs in other insects also have a demonstrated role in regulating circulating sugars (Wu and Brown, 2006; Brown et al., 2008; Okamoto and Yamanaka, 2015; Xue et al., 2020). Typically, these absorbed sugars are converted to intracellular glycogen for ready use or into lipids for longer-term storage (Brown et al., 2008; Pooraiiouby et al., 2018). Comparatively fewer studies have investigated the role of ILP signaling in circulating sugars in Heteroptera, despite this suborder’s impact on human health (Leyria et al., 2021a) and agricultural production (Xue et al., 2022). In Rhodnius prolixus, knockdown of an ILP (Rhopr-ILP1) resulted in elevated hemolymph carbohydrates and lipids, particularly after a blood meal (Defferrari et al., 2016a) while knockdown of an insulin-like growth factor (Rhopr-IGF) reduced circulating carbohydrates and lipids (Defferrari et al., 2016b). In addition, increased expression of ILP, IGF, and the insulin receptor IR1 occurs as starvation in R. prolixus progresses, possibly to create a sensitized state that can rapidly respond to release of ILPs after a blood meal (Leyria et al., 2021b). Incubation of Rhopr-ILP1 or porcine insulin with fat body or ovaries (and injection, for porcine insulin) increased expression of trehalose-specific facilitated transporter (TRET) in these tissues, suggesting that this may be part of the mechanism by which insulin signaling regulates circulating trehalose in R. prolixus (Leyria et al., 2021a).

ILP precursors are synthesized as a single peptide consisting of a secretory signal peptide, a B-chain, and an A-chain interconnected by a C-peptide (Krieger et al., 2004; Okamoto and Yamanaka 2015). Six consistently spaced cysteine residues form three disulfide bonds, joining the A- and B-chains. Further post-translational modifications occur by prohormone convertase cutting of conserved dibasic cleavage sites flanking the A- and B-chains (Brogiolo et al., 2001; Krieger et al., 2004; Okamoto and Yamanaka, 2015; Biglou et al., 2021). The two-chain structure, consistent spacing of cysteine residues, and occurrence of dibasic cleavage sites delineating the A- and B- chains are characteristic features of ILPs. Many insect ILPs are produced in neurosecretory brain cells (Broughton and Partridge, 2009), but for some species they can be expressed in other tissues, such as the fat body (Badisco et al., 2008; Okamoto et al., 2009; Mizoguchi and Okamoto, 2013).

The western tarnished plant bug, Lygus hesperus Knight, is a polyphagous pest insect distributed across western North America (Scott 1977; Wheeler 2001; Antwi and Rondon, 2018). It feeds on numerous field, fruit, and vegetable crops, causing significant annual losses (Hagler et al., 2016; Joseph and Bolda, 2016; Zalom et al., 2018). Due to changes in agricultural practices over the past three decades, namely, a decrease in use of broad-spectrum chemical insecticides and a switch to transgenic and classically selected plant varieties targeted at non-Lygus pests, Lygus spp. have become significant pests and present one of the more considerable challenges for insect pest management to date (Lu et al., 2010). Lastly, insecticide resistance to multiple chemistries in Lygus lineolaris (Snodgrass, 1996; Snodgrass et al., 2009; Parys et al., 2018; Du et al., 2024a,b) suggests L. hesperus could develop similar resistance as well, which jeopardizes current Lygus species control strategies that are dependent on broad-spectrum insecticides (Parys et al., 2017). Therefore, new insecticides with novel modes of action and more specificity to Lygus spp. are needed to provide management of these insects. Neuropeptides and other signaling molecules have been suggested as new mode of action targets for pesticide development (Audsley and Down 2015; Nuss et al., 2015; Conley et al., 2015).

Transcriptomes of L. hesperus suggest that at least three putative ILPs may be expressed (Hull et al., 2021). In this study, the expression of the three ILPs (termed LhILPs) was characterized throughout development and in various tissues of the adult stage. Also, the knockdown of individual LhILPs was examined for its impact on the ability of L. hesperus to clear circulating trehalose from the hemolymph. This work demonstrates the hallmark functions of insulin signaling in L. hesperus and provides a foundation to explore other functions associated with LhILPs in L. hesperus.

2. Methods

2.1. Insect rearing

A colony of L. hesperus derived from the USDA-ARS, US Arid Land Agricultural Research Center (Maricopa, AZ, USA) was used for this study and was maintained under laboratory conditions at 28 ± 2°C, 60 ± 5 % relative humidity, with a 16:8h light:dark photoperiod. Insects were provisioned with water, green beans (Phaseolus vulgaris L.), and artificial food diet (Portilla et al., 2011) packaged with Parafilm M (Pechiney Plastic Packaging, Chicago, IL), all of which were replenished every 2–3 days. Starting at the end of the 5th instar, daily checks were made for molting, and newly eclosed adults were transferred to known age cohort cages. Successive generations were maintained by creating mixed-sex cages. Females use the green beans as their primary oviposition site so the beans could be separately saved and maintained under standard rearing conditions to allow nymphs to hatch.

2.2. Identification of L. hesperus ILPs

A transcriptome of L. hesperus (Tassone et al., 2016) was searched by BLASTp and tBLASTn (https://blast.ncbi.nlm.nih.gov/Blast.cgi), using known ILP sequences from other insects (Xu et al., 2015; Veenstra, 2020; Hull et al., 2021; Huygens et al., 2022). All hits were manually checked for homology to the targeted query. Oligonucleotide primers were designed to amplify the targeted sequence based on the nucleotide sequences of the transcripts of the L. hesperus ILP genes using Primer3 (Koressaar and Remm, 2007; Untergasser et al., 2012; Koressaar et al., 2018) (Table S1). Total RNA was isolated from a pool of 5 male and 5 female, 5–7 day old adult L. hesperus homogenized by pestle grinding in 250 μl Trizol^™ reagent (Invitrogen, Waltham, MA, USA) in a 1.7 ml tube, and then using a Direct-zol RNA miniprep kit (Zymo, Irvine, CA, USA) following the manufacturer’s instructions. RNA quantity and quality was determined using a spectrophotometer (Thermo-Scientific, Waltham, MA, USA). To remove any residual genomic DNA, total RNA was treated with DNase I [Invitrogen (Ambion), Waltham, MA, USA] according to manufacturer instructions and was then heat inactivated at 85 °C for 5 min.

A total of 1 μg of DNA-free RNA was used as a template for cDNA synthesis using the iScript cDNA synthesis kit (Biorad, Hercules, CA, USA) following the manufacturer’s instructions. The cDNA was subsequently diluted 1:5 in deionized water. The gene-specific paired primers were used with polymerase chain reaction (PCR) amplification using Apex 2.0x Taq Master Mix (Apex BioResearch, Houston, TX, USA) with 1 μl of 1:5 diluted cDNA with the following thermocycler program: 40 cycles at 94 °C for 30 s; 64 °C for 30 s; 72 °C for 1 min. This included an initial 5 min denaturing step at 94 °C, and a final 10 min, 72 °C extension step. These reactions were separated by gel electrophoresis. Bands of the predicted size were extracted and sent for Sanger sequencing (Genewiz, South Plainfield, NJ).

2.3. Quantitative PCR of L. hesperus ILPs: Tissue and developmental profile

Quantitative reverse transcription PCR (qRT-PCR) was performed to determine the relative expression level of ILPs in the head, gut tract, and reproductive tissues (gonads) of 7 day old male and female adult L. hesperus. Tissues were dissected in phosphate-buffered saline (PBS), then placed in 250 μl Trizol. The remaining thorax and abdomen portions of the carcass (containing cuticle and fat body) were also collected separately in 250 μl Trizol. All dissected tissues were stored at −80 °C until total RNA extraction and cDNA preparation, as described above. Each qRT-PCR reaction was performed on a CFX Touch Real-Time PCR Detection system (BioRad, Hercules, CA, USA), using 1 μl 1:5 diluted cDNA, 5 μl SYBR green, 1 μl (10 μM) forward and reverse gene-specific primers (Table S1), and 3 μl H₂O (final volume of 10 μl). The constitutively expressed L. hesperus 60S ribosomal protein L32 (Rpl32, Accession: GDHC01014441) was used as a housekeeping gene for normalization. All reactions were performed with a two-step program, with an initial denaturation for 5 min at 95 °C, followed by 40 cycles of 10 s at 95 °C, 15 s at 66 °C, and 15 s at 72 °C. A melt curve was analyzed at 70–95 °C. Relative expression was calculated using the 2^−ΔΔCt method (Livak and Schmitthen, 2001), normalizing to gonad tissue. All qRT-PCR analyses were performed on three biological cohorts, each run with three technical replicates. Technical replicates were averaged before analysis of biological cohorts.

To examine ILP expression across L. hesperus development, RNA was extracted from eggs, 1st through 5th instar nymphs, and day 0 and 7 adults. Male and female samples were collected separately for 5th instars and adults. All samples were collected individually into 250 μl TRIzol and kept at −80 °C before total RNA extraction. Total RNA extraction, cDNA synthesis, qRT-PCR reaction conditions, and analysis were as described above, normalizing expression to 5th instar males.

2.4. Trehalose and glycogen assays

To test whether L. hesperus ILPs signal for circulating sugar uptake from the hemolymph, an assay was used to determine whether injected trehalose was converted into glycogen in vivo in L. hesperus males. Adult males, collected 3–6 days post-eclosion, were used for these experiments due to their consistent weight, and content of glycogen and trehalose. Females, in contrast, were much more variable. Male L. hesperus were injected into the side of the thorax with 150 μg trehalose, dissolved in 500 nl of insect saline (Wen et al., 2010). Injections were made with a Nano-injector III (Drummond Scientific Company, Broomall Pa, USA) fitted with pulled-glass capillary needles. Samples were collected and stored at −80 °C as follows: L. hesperus males were either collected without injection (“pre-injection”) or were injected and collected immediately (“0 h”) or after a recovery period of 6 h (“6h post-injection”). All insects were individually weighed before the microseparation procedure to calculate percentage trehalose or glycogen by body weight. All trehalose uptake analyses were performed on three biological cohorts, each run with three technical replicates.

To quantify trehalose and glycogen, a microseparation of nutrients through a sodium sulfate titration process was performed, followed by a modified anthrone assay, using a procedure previously developed for mosquitoes (Van Handel, 1965; Telang and Wells, 2004; Pooraiiouby et al., 2018). After microseparation, samples were pelleted, dried, then resuspended in 200 μl of 25 % EtOH. Trehalose samples were additionally treated with 25 μl of 1 M HCl and heated at 90 °C for 7 min, followed by addition of 75 μl, 1 M NaOH and heating for 7 min at 90 °C. Stocks of 1 mg/ml glycogen or trehalose (Sigma-Aldrich, St. Louis, MO, USA) in 25 % EtOH were used to set up a standard line calibration series. Samples and standards were placed in glass tubes, and 1 ml of anthrone reagent was added to each. Samples and standards were heated at 90 °C for 20 min in darkness, cooled for 20 min at 4 °C, and optical densities were read on a spectrophotometer (CLARIOstar, BMG Labtech, Cary NC, USA) at 625 nm against a reagent blank. A regression curve was generated (Microsoft Excel) using readings from standards to calculate concentrations from samples.

2.5. LhILP RNAi knockdown

To determine which, if any, LhILPs were responsible for signaling the uptake of circulating sugars, individual dsRNAs were synthesized for LhILP1, LhILP2, and LhILP3. Briefly, distinct, ~500 bp regions of DNA encoding each LhILP were amplified with PCR using a pooled sample of L. hesperus cDNA and specific primers (Table S1) containing a T7 promoter sequence with the following two-step program: an initial 5 min at 98 °C, followed by five cycles of 10 s at 98 °C, 30 s at 60 °C, 30 s at 72 °C, and 35 cycles of 10 s at 98 °C, 30 s at 64 °C, 30 s at 72 °C. Using these templates, dsRNAs were prepared and purified using the MEGAscript T7 synthesis kit [Invitrogen (Ambion), Waltham, MA, USA] according to manufacturer instructions. In addition, dsRNA was synthesized for Venus, a yellow fluorescent protein mutant of green fluorescent protein initially derived from the jellyfish Aequorea victoria (Nagai et al., 2002), which served as an injection control. Pelleted dsRNAs were resuspended in insect saline to a concentration of 2 μg/μl and were aliquotted and stored at −80 °C until use. To eliminate batch-specific effects, dsRNA was re-synthesized multiple times and used on at least three different biological cohorts of L. hesperus.

To confirm knockdown efficiency, recently (0–3 day) emerged adult male L. hesperus were injected with 1 μg (0.5 μl) of individual LhILP or Venus dsRNA. These were collected 3 days after treatment, along with saline-only injected males. Whole-body RNA was extracted, cDNA was synthesized, and qRT-PCR was performed as described above, using RNAi-specific primers to avoid overlapping with dsRNA regions (Table S1) (Brent and Hull, 2019).

To determine the impact of LhILP knockdown on circulating trehalose, dsRNA-treated males were were injected with 150 μg of trehalose three days after RNAi. These were allowed to recover for 6 h, then stored at −80 °C. RNAi and control insects that were not trehalose injected or were injected with trehalose and harvested immediately (0 h) were also collected. These samples were processed through the microseparation procedure and anthrone assays for trehalose and glycogen, as described above.

Statistical analysis using the Tukey-Kramer posthoc test was performed to compare the average means of the experimental and control groups and determine statistical significance (Microsoft Excel). A p-value < 0.05 was considered statistically significant.

2.6. Insulin injection assays

To determine the activity of an ILP on trehalose clearance in L. hesperus in the absence of potentially incomplete or co-knockdown induced by RNAi, a subset of adult males were neck-ligated by securing a human hair tightly around the neck (Zhang et al., 2017), then injected with trehalose. Additional males were ligated then injected with trehalose in insect saline containing 100 mM human insulin (hIns, Sigma-Aldrich). Samples were collected 6 h after injection and processed through the microseparation and anthrone assays, as described above. Experimental and control groups were compared using the Tukey-Kramer post hoc test as noted above.

3. Results

3.1. Confirmation and characteristics of L. hesperus ILPs

Three ILPs were previously annotated in a peptidome screen of L. hesperus (Hull et al., 2021). The predicted open reading frames of these sequences were confirmed in the current study with RT-PCR and Sanger sequencing (Genbank Accession # PV763575 (LhILP1), # PV763576 (LhILP2), # PV763577 (LhILP3)). All three ILPs have a predicted N-terminal signal peptide (identified using SignalP (https://www.cbs.dtu.dk/services/SignalP/) (Petersen et al., 2011)), indicating extracellular secretion (Sharma et al., 2019a). They also contain consistently spaced, conserved cysteines predicted to form disulfide bridges, characteristic of ILPs (Figs. 1, 2A) (Brogiolo et al., 2001; Riehle et al., 2006; Okamoto et al., 2009; Veenstra, 2020). In addition, putative B-chains, C-peptides, and A-chains were identified and were flanked by dibasic cleavage sites that are predicted to be cleaved by prohormone convertases. For the C-peptides of LhILP1 and LhILP2, a putative internal dibasic cleavage site was detected. LhILP1 contained a “RYCGRDI” sequence in the B-chain, resembling a relaxin-like binding motif (RXXXRXXI/V) (Xue et al., 2020). Neither LhILP1 nor LhILP2 had an extended C-terminus beyond the predicted A chain. In contrast, LhILP3 had characteristics similar to vertebrate insulin-like growth factor (IGF), noted in other arthropods (Okamoto et al., 2009; Sharma et al., 2019b), and these peptides have been called arthropod insulin-like growth factors (aIGFs) (Veenstra, 2020). Namely, the predicted C-peptide is truncated (only 9 amino acids) compared to other ILPs (LhILP1 and LhILP2 have C-peptides with 31, and 23 amino acids, respectively). There is an extended C-terminal sequence (162 amino acids) beyond the predicted A-chain, also a characteristic of aIGFs (Fig. 2C) (Riehle et al., 2006; Okamoto et al., 2009; Veenstra, 2020). As many as 15 putative dibasic cleavage sites occur along the entire LhILP3 sequence, two of which may permit cleavage of the C-peptide by prohormone convertases. The remaining sites occur in the extended C-terminal sequence. No dibasic cleavage site follows the predicted signal peptide, unlike LhILP1 and LhILP2.

Fig. 1.

Amino acid translations and structural predictions of three L. hesperus ILPs confirmed by Sanger sequencing. Cysteine residues are indicated in bold text. Putative dibasic cleavage sites are in bold red text; B-chain: Blue highlight; A-chain: Yellow highlight; Signal peptide underlined. Putative C-peptides are between the B and A chains.

Fig. 2.

Comparison of LhILP amino acid sequences with selected hemipterans and D. melanogaster ILPs (DILPs). A) Amino acid sequence alignment of B- and A-chains of LhILPs (truncated around core Cys residues for the sake of space) with other invertebrates (particularly heteropterans). Colors indicate amino acids with similar properties: green = hydrophobic; dark blue = aromatic; pink = small; dark red = basic; yellow = cysteine; tan = hydroxyl; gray = tryptophan; light blue = proline; black = acidic; orange = amide. Predicted disulfide bridges are connected by black bars. These alignments were performed in Clustal Omega (Madeira et al., 2024) and colored manually. B) Alignment of the heteropteran aIGF C-terminal extension. Ascension numbers of sequences used for this figure can be found in Table S2.

3.2. Expression profiles

As determined by qRT-PCR, all three LhILPs were expressed predominantly in the heads of reproductively mature adult males and females (Fig. 3A, C, E). However, most tissues displayed traces of ILP transcripts. The expression of LhILPs throughout development was generally stable (Fig. 3B, D, F). Expression of LhILP1 was highest in eggs (Fig. 3B). No significant differences in LhILP2 expression were detected (Fig. 3D). LhILP3 expression was highest in 5th instar females (Fig. 3F).

Fig. 3.

qRT-PCR quantification of LhILP expression in body tissues for adult female (black bars) and male (white bars) and through development for A-B) LhILP1, C-D) LhILP2, and E-F) LhILP3 (average ± SE). Expression is scaled to gonads for each gene and sex for tissue expression, and to 5th instar males for the developmental profile. Abbreviations: Head: HD; digestive tract: DT; gonads: GD; thoracic carcass: TX; abdominal carcass: AB. For developmental expression of LhILPs 1–3, eggs through 4th instars were pooled regardless of sex. Sex-specific expression was tracked for 5th instar (5thM, 5thF), newly emerged adults (D0M, D0F), and reproductively mature adults (D7M, D7F). Different letters indicate significant differences (α = 0.05). N/S = not significant.

3.3. Influence of LhILPs on hemolymph sugar homeostasis

Activating a signaling cascade that results in the cellular uptake of circulating sugars from the hemolymph is one of the most basic hallmarks of insulin signaling. To determine whether LhILPs have this function, RNAi was used to knock down each individual LhILP and the ability of L. hesperus to clear injected trehalose from the hemolymph was assayed. The knockdown efficiency of LhILP1, LhILP2, and LhILP3 dsRNAs were 93 ± 5 %, 79 ± 2 %, and 70 ± 12 %, respectively (Table 1) compared to Venus dsRNA controls, as determined by qRT-PCR. In addition, knockdown efficiency was assayed on non-target LhILPs to assess whether individual LhILP knockdown influenced expression of other LhILPs. Expression of non-target LhILPs dropped in all dsRNA treatments, with the most robust responses for dsLhILP1, knocking down LhILP3 by 72 ± 15 % and dsLhILP3 knocking down LhILP2 by 62 ± 10 %.

Table 1.

Average LhILP RNAi knockdown efficiency (±SE) and off-target LhILP knockdown compared to Venus controls as determined by qRT-PCR.

	LhILP1 % knockdown	LhILP2 % knockdown	LhILP3 % knockdown
LhILP1 RNAi	93 ± 5 %	17 ± 30 %	72 ± 15 %
LhILP2 RNAi	58 ± 12 %	79 ± 2 %	37 ± 43 %
LhILP3 RNAi	50 ± 18 %	62 ± 10 %	70 ± 12 %

To examine impacts on circulating sugar uptake from the hemolymph, trehalose was injected into dsLhILP and dsVenus control L. hesperus, and soluble sugars and glycogen were assayed 6 h post-injection. In all insects, trehalose levels were approximately 0.5 % of body weight before trehalose injection (Fig. 4A). Trehalose-injected dsVenus insects that were immediately harvested (0 h) had significantly higher trehalose levels (~2.3 % of body weight). Trehalose levels dropped in dsVenus insects that were allowed to recover for 6 h (slightly, but significantly, above pre-injection levels). Insects with RNAi knockdown of LhILP1 displayed a similar trehalose phenotype to the dsVenus insects. In contrast, RNAi knockdown of either LhILP2 or LhILP3 resulted in significantly higher trehalose levels, even after 6 h of recovery.

Fig. 4.

Macronutrient levels of trehalose-injected male L. hesperus, 3 days post-RNAi knockdown with LhILP dsRNAs. Whole body content of A) trehalose and B) glycogen as a percentage of wet body weight (±SE). A set of RNAi knockdown insects was collected prior to trehalose injection (Pre-injection), immediately after injection for a Venus RNAi control (Venus 0 h), and 6 h post-trehalose injection. Different letters indicate significant differences (α = 0.05).

Glycogen levels were significantly higher in 6 h dsVenus samples compared to non-trehalose injected and 0 h dsVenus controls (Fig. 4B), suggesting cellular uptake of trehalose and conversion to glycogen. However, levels of glycogen at 6 h post-trehalose injection in dsLhILP2 and dsLhILP3 knockdowns were not significantly higher than non-trehalose injected dsLhILP and dsVenus controls. In contrast, dsLhILP1 knockdown insects at 6 h post-injection had significantly higher glycogen levels than all other samples, even though circulating trehalose levels were not significantly different from dsVenus controls at 6 h (Fig. 4B).

3.4. Insulin and ILP injection assays

Since RNAi introduced confounding LhILP co-knockdown effects, a heterologous insulin, hIns, was used to examine whether an insulin-like molecule could induce trehalose clearance. Because all identified LhILPs were predominantly expressed in the head, neck ligations were performed on L. hesperus to prevent release and circulation of head-produced LhILPs prior to trehalose injection into the thorax. Ligated insects did not convert injected trehalose to glycogen (Fig. 5). However, co-injection of human insulin with trehalose into ligated insects rescued the effects of ligation, resulting in trehalose clearance and increased glycogen after 6 h. In fact, the insulin-injected insects stored significantly more glycogen than intact insects (Fig. 5B).

Fig. 5.

Macronutrient levels of trehalose-injected 3–6 day old male L. hesperus, immediately after injection (0 h), 6 h after injection, or non-injected (colony baseline). Whole body content of A) trehalose and B) glycogen as a percentage of wet body weight (average ± SE).

4. Discussion

Insulin signaling has many conserved roles in eukaryotes and is well known as a trigger for cellular uptake of circulating sugars (Masumura et al., 2000; Wu and Brown, 2006; Mizoguchi and Okamoto, 2013; Chowanski et al., 2021). However, work in Hemiptera regarding insulin signaling is scant in comparison to other insects. In the current work, three putative ILPs in L. hesperus were characterized that display structural similarities to ILPs in other insects (Fig. 1). RNAi knockdown of individual ILPs demonstrated a role for LhILP2 as a signal for the uptake of circulating hemolymph sugars, similar to ILPs in other organisms (Defferrari et al., 2016a; Kawabe et al., 2019; Xue et al., 2020; Thorens, 2024).

In the current work, the primary sequences of L. hesperus ILPs were compared to other insects. Alignments with ILPs and aIGFs from other insects showed conserved sequence motifs in the B and A chains (Fig. 2A), and also the C-terminal extension observed in LhILP3 and aIGFs (Fig. 2B). Despite amino acid similarity being notoriously variable in ILPs between insects of different groups (Okamoto and Yamanaka, 2015; Sharma et al., 2019a,b; Veenstra, 2020), the ILPs identified from L. hesperus can putatively be classified as structurally related to ILPs (LhILP1, LhILP2) or aIGFs (LhILP3) based on their length and spacing of dibasic cleavage sites (Brogiolo et al., 2001; Riehle et al., 2006; Okamoto et al., 2009; Slaidina et al., 2009; Veenstra, 2020). Outside of a slight resemblance to a relaxin-like binding motif in LhILP1 (RXXXRXXI/V, as reported in Xue et al., 2020), none of the identified LhILPs resembled insect relaxins, which is consistent with the reported absence in Heteroptera of relaxins and their predicted LGR4 receptor (Veenstra, 2020).

The LhILP transcripts were predominantly head-expressed (Fig. 3A, C, E). This is consistent with many other insects in which ILPs are mainly localized to the medial neurosecretory cells of the brain (Marquez et al., 2011; Defferrari et al., 2016a). In other insects, ILP and aIGF transcripts are highly expressed in the fat body (Okamoto et al., 2009; Veenstra et al., 2021; Biglou et al., 2021), but none of the LhILPs in the current study, including the aIGF-like LhILP3, were abundantly expressed in the L. hesperus abdominal body wall that contained fat body. However, high aIGF expression in the head, rather than in the fat body, has been noted in other insects, such as Locusta migratoria (Veenstra et al., 2021). In R. prolixus, IGFs are expressed in both the fat body and central nervous system (Defferrari et al., 2016b). Despite being localized principally to the head in L. hesperus, ILP expression did vary significantly over the course of development, suggesting changing functional roles. LhILP1, for instance, is more highly expressed in egg and early growth stages, possibly suggesting a developmental or growth factor role (Fig. 3B). However, more precise developmental sampling (e.g., early versus late instar development), or a response to specific conditions (e.g., starvation) might be required to pinpoint more obvious links to LhILP functions.

Several ILPs have been identified in different insect species that signal for the uptake of circulating sugars from the hemolymph (Wu and Brown, 2006; Brown et al., 2008; Okamoto and Yamanaka, 2015; Xue et al., 2020; Defferrari et al., 2016a). In the current study, RNAi knockdown of LhILP2 and LhILP3 influenced clearance of injected trehalose (Fig. 4) and an accumulation of glycogen. However, an examination of co-knockdown (Table 1) revealed that trehalose clearance was only disrupted when a high level of LhILP2 transcripts were knocked down (>60 %). In contrast, when LhILP3 transcripts were knocked down as a byproduct of LhILP1 knockdown, trehalose clearance was not impacted. This suggests that LhILP3 is not directly regulating trehalose uptake from the hemolymph. The reasons for co-knockdown are unclear, but are unlikely to be due to overlap in dsRNA sequence. Percent nucleotide sequence identity between LhILPs is low, ~44–50 %. Alignment is also fragmentary, with the longest contiguous stretch of shared sequence at 9 bp between LhILP1 and LhILP2. In the future, siRNA might be attempted for more precise knockdown of individual LhILPs, but other co-regulatory factors may be involved. Co-knockdown of ILPs was noted in other insects, such as Nilaparvata lugens, another hemipteran, although the mechanism driving co-knockdown was unclear (Xue et al., 2020). Both co-knockdown and compensatory ILP expression has been reported in D. melanogaster, when individual ILPs are knocked down or knocked out (Broughton et al., 2008; Grönke et al., 2010). DILP3 expression, for instance, is likely influenced by insulin signaling itself through the downstream signaling component FOXO, a conclusion supported by multiple putative FOXO binding sites on the DILP3 promoter (Broughton et al., 2008). Currently, there is no publicly available L. hesperus genome to examine LhILP promoter regions for FOXO or other potential regulatory element binding sites, but co-regulation of LhILPs in L. hesperus and their impact on physiological functions warrants further investigation.

To examine insulin signaling in the absence of confounding RNAi co-knockdown, LhILP release from the brain into the body cavity hemolymph was prevented with neck-ligations, providing an in vivo assay system that could be injected with trehalose and heterologous hIns. Ligated L. hesperus injected with trehalose were unable to clear the sugar from the hemolymph after 6 h, but ligated insects co-injected with hIns were able to do so (Fig. 5). Taken together with the RNAi experiments, this suggests that an insulin-like head factor is largely responsible for regulating hemolymph trehalose levels. However, exactly which LhILP mediates these levels remains to be resolved. Often, commercially available vertebrate insulin is used in insect studies as a proxy for endogenous insulin, albeit at high concentrations to achieve biological activity (Riehle and Brown, 2002, 2003; Brown et al., 2008; Gu et al., 2009; Defferrari et al., 2018; Wang et al., 2025). There are very few examples of successful synthesis of insect ILPs, mainly due to the challenge and costs of arranging the precise, correct disulfide bonds (Maruyama et al., 1992; Brown et al., 2008; Sajid et al., 2010; Nuss and Brown, 2018). It remains to be demonstrated whether LhILPs, if they can be efficiently synthesized, can also directly stimulate trehalose clearance, but this is an avenue for future work.

5. Conclusion

This study presents the first exploration of ILP functions in L. hesperus. Three ILPs were characterized that display structural similarities to ILPs or aIGFs in other insects. Our results suggest that at least one of these LhILPs plays a role in regulating and storing circulating sugars, as with ILPs in other organisms (Defferrari et al., 2016a; Chowanski et al., 2021; Kawabe et al., 2019; Thorens, 2024). However, RNAi knockdown of individual ILPs suggests that co-regulation of ILP expression may occur. As with other insects, ILPs likely have multiple additional roles in L. hesperus that remain to be explored. Further, the dynamics of yet to be identified cognate receptors for LhILPs are an avenue for future investigation. Although not explored in the current study, several studies have advanced the potential of mimics of peptide hormones, neuropeptides, or other signaling molecules as insecticides (Audsley and Down, 2015; Nuss et al., 2015; Conley et al., 2015) and may be an avenue for further investigation for management of L. hesperus.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.jinsphys.2025.104885.

Data availability

Data will be made available on request.