Hyphantria cunea eggs and artificial diets were obtained from the Research Institute of Forest Ecology, Environment and Protection, Chinese Academy of Forestry (Beijing, China). Eggs were incubated at 25°C until hatching, and larvae were fed on artificial diets in 250ml transparent plastic bottles, which were maintained at 25±1°C with a 16:8h light:dark photoperiod.

Reverse transcription PCR was initially used to validate the sequences of H. cunea LK and LKR transcripts from the H. cunea genome database. The LK and LKR genes were cloned using the following thermal conditions: 94°C for 3min; followed by 35cycles of 94°C for 30s, 60°C for 30s, and 72°C for 1min; then a final extension at 72°C for 10min. The PCR product was sub-cloned into pMD18-T vector (TaKaRa, Japan) and then verified sequences. The primers used for the PCR cloning of HcLK and Hyphantria cunea leucokinin receptor (HcLKR) are presented in Table 1. The PCR products were directly cloned into the pcDNA-3.1-myc-His vector. The recombinant vectors were verified by sequencing.

The deduced amino acid sequences of LK and LKR orthologs were obtained from GenBank using BLAST searches (blastx and tblastx). Multiple alignment of the amino acid sequences was performed using the ClustalX2 program and BioEdit. A phylogenetic tree was constructed using the neighbor-joining (NJ) method in MEGA 5.0 with 1,000 bootstrap replicates (Tamura et al., 2011). Signal peptides were predicted using Signal P 4.1 Server (Mccarthy et al., 2004), and transmembrane domains were predicted using TMHMM server v2.02 (Sonnhammer et al., 1998). The presence of N-glycosylation sites in predicted protein sequences was assessed using NetNGlyc 1.0¹, and the generation of sequence logos for the C-terminal motifs of LK proteins was created by Weblogo (Crooks et al., 2004).

The human embryonic kidney 293 (HEK293) cell line was cultured in Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum (FBS) and 4mM L-glutamine (Invitrogen) at 37°C in a humidified incubator containing 5% CO₂. HEK293 cells were transfected with LKR cDNA plasmid constructs using Effectene transfection reagent (Qiagen) according to the manufacturer’s instructions. Two days after transfection, stably expressing cells were selected by the addition of 800mg/L G418.

To investigate the interaction between the LKR and LKs in H. cunea, the response of the LKR to chemically synthesized LKs was examined using the Ca²⁺ imaging assay. A fluorescent Ca²⁺-sensitive probe, Fura-4/AM (Beyotime, Shanghai, China), was used to detect the intracellular cytosolic calcium signals according to the manufacturers’ instructions. In brief, HEK293 cells stably expressing LKR were washed twice with phosphate-buffered saline and were suspended at 5×10⁶ cells/ml in Hanks’ balanced salt solution. The cells were then loaded with 2μl Fura-4/AM for 20min and washed twice with HBSS buffered medium. Then, cells were stimulated with 0.1 and 1μM HcLKs (HcLK-1, HcLK-2, and HcLK-3) chemically synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). Each 96-well plate was transferred into a Multi-Mode Microplate Reader (Varioskan Flash Beckman XL-70 F; Thermo Fisher Scientific Inc. Waltham, MA) to monitor the Fluo-4 fluorescence. The excitation wavelength was 485nm, and fluorescence emission was detected at 520nm. Various concentrations of receptor ligands were added when Fluo-4 fluorescence had reached a stable value in each well. The changes in Fluo-4 fluorescence were recorded automatically. Dose–response curves for putative agonists were established in at least three independent experiments.

A 463-bp dsRNA representing the H. cunea LK-encoding gene sequence and a 505-bp dsRNA representing the H. cunea LKR-encoding gene sequence were synthesized using the MEGAscript T7 high-yield transcription kit (Ambion) according to the manufacturers’ protocol. The dsRNA was purified with phenol/chloroform followed by ethanol precipitation. The dsRNA of the enhanced green fluorescent protein gene (pEGFP-N1 plasmid as template, WP_031943942.1, 507-bp dsRNA) was employed as a control. A 2μg/μl dsRNA solution (1μl) was microinjected into the penultimate posterior abdominal section of individual seventh instar H. cunea larvae using an injection needle (MICROLITERTM #65 with 33-gauge needle, Hamilton Co., Reno, NV, United States) under ice anesthesia (Sun et al., 2016). Control H. cunea larvae were microinjected with the EGFP dsRNA. Microinjected H. cunea larvae were allowed to recover for 2h at room temperature and then reared on an artificial diet under a 16:8h light: dark photoperiod at 25±1°C. After 72 and 96h, LK and LKR mRNA levels in the dsRNA-treated seventh instar H. cunea larvae were measured by qRT-PCR technology.

To measure water content, the larvae treated with dsEGFP, dsLK, and dsLKR for 48h were dehydrated at 80°C until a constant weight. Ten H. cunea larvae were weighed before and after dehydration using a Mettler MT5 analytical microbalance (Columbus, OH, United States). Water content was calculated as the difference between the fresh and dry weight. Each replicate contained 10 H. cunea larvae, and the experiment was performed in triplicate.

To study survival under desiccation and starvation, the H. cunea larvae treated with dsRNA were kept in empty vials or vials containing cotton ball with sterile water, respectively. Ten H. cunea larvae were used per replicate, and the experiment was performed in triplicate. The survival was recorded every 24h until all the H. cunea larvae were dead. The vials were placed in an incubator at 25±1°C under normal photoperiod conditions (16:8h light: dark).

On day of the seventh instar stage, H. cunea larvae were microinjected with dsRNA (LK and LKR dsRNA or EGFP dsRNA) and then returned to transparent plastic vials and starved for 24h. After a subsequent 4-day feeding period, the appetite of the larvae was checked by measuring the amount of artificial diet eaten by individual larvae during 24h. The weight of the artificial diet was measured before and after H. cunea larva feeding. Three biological replicates were included for each experiment, and for each biological replicate, 10 H. cunea larvae were kept in transparent plastic vials. The vials were placed in an incubator at 25°C under normal photoperiod conditions (16:8h light: dark).

The RNA was extracted from H. cunea eggs, first to seventh instar larvae, pupae, adults, and tissue samples using the RNeasy Mini Kit (Qiagen, Valencia, CA, United States). The tissues – head, silk glands, midgut, epidermis, testis, ovary, Malpighian tubules, and fat body – were collected from larvae on day 1 of the seventh instar stage. cDNA was synthesized using the total RNA (0.5μg) and the PrimeScript®RT Reagent Kit with gDNA Eraser (Perfect Real Time, TaKaRa, Japan), according to the manufacturer’s protocol. The mRNA levels of LK, LKR and insulin-like peptide (ILP) genes were assessed using RT-qPCR with a SYBR Green kit (Toyobo, Osaka, Japan) and MJ Opticon™² machine (Bio-Rad, Hercules, CA, United States). The reaction mixture (20μl) was composed of SYBR Green Real-time PCR Master Mix (10μl; Toyobo), nuclease-free water (7μl), gene-specific primers (1μl, 0.5μM; Table 1), and cDNA template (2μl; equivalent to 50ng of total RNA). RPL13 and EF-1α were used as internal reference genes (Sun et al., 2019). The conditions for RT-qPCR reactions were as follows: 1cycle at 95°C for 30s, followed by 45cycles at 95°C for 12s, 60°C for 30s, 72°C for 40s, and 82°C for 1s for plate reading. The purity of the amplified products was analyzed by melting curve analysis. qRT-PCR was performed in using independent biological repeats in triplicate to ensure the reproducibility of the results. The expression levels of the clones were calculated using the 2^-ΔΔCt method (Livak and Schmittgen, 2001).

Statistical analysis was performed using SPSS (v17.0, SPSS Inc., Chicago, Illinois). One-way ANOVA was performed using Prism 8.0 (GraphPad Software, La Jolla, CA, United States). Value of p<0.05 was considered to indicate statistical significance for all experiments performed in the present study.

The sequences of LK and LKR genes were identified using transcriptome and genome analysis (Sun et al., 2019; Wu et al., 2019). The LK gene contains a 1,014bp open reading frame (ORF), which encodes a signal peptide (23 residues). The three mature peptide sequences comprise six (YFSPWGamide, HcLK-1), seven (VRFSPWGamide, HcLK-2), and eight (KVKFSAWGamide, HcLK-3) amino acid residues, respectively. The mature peptide cleavage site is a combination of lysine (K) and arginine (R) and has an amidation site “G” (Figure 1A). LK proteins from H. cunea and other insects showed very high sequence similarity (Figure 1B). Phylogenetic analysis revealed that HcLK and LKs from other insect species were clustered in a single group and that HcLK is most closely related to the Danaus plexippus plexippus homologs (Figure 1C).

The full-length HcLKR cDNA consists of 2,186 nucleotides; the predicted ORF encodes 485 amino acids (Figure 2A). The ORF contains an ATG initiation codon, an upstream 608bp 5′ untranslated region (UTR), and a termination (TAA) codon followed by a 120bp 3′ UTR (Figure 2A). The HcLKR protein contains the characteristic seven transmembrane domains (TM, Figure 1B, TMHMM 2.0 server), with a typical signature of rhodopsin-like G protein-coupled receptor (GPCR; Figure 2A). The amino acid residues at positions 49–73, 82–104, 120–142, 163–179, 215–238, 266–291, and 306–331 represented TMI, TMII, TMIII, TMIV, TMV, TMVI, and TMVII, respectively. The predicted three-dimensional model of HcLKR showed a characteristic structure, with seven TM segments with α-helices (TM-I to TM-VII) linked by three intracellular and three extracellular loops, an extracellular amino terminus, and an intracellular carboxyl terminus (Figure 2B). Likewise, the Pfam analysis predicted seven transmembrane passes, and six conserved cysteine residues in the N-terminal extracellular domain. Six potential N-glycosylation sites were predicted for the N-terminal extracellular domain (NetNGlyc 1.0 server). Multiple amino acid sequence alignment between HcLKR and other LKRs showed high overall amino acid homology in the seven transmembrane domains (Figure 2B).

The tissue-specific and developmental mRNA profiles of HcLK and HcLKR in H. cunea were quantified using RT-qPCR (Figure 3). Compared with that at the egg stage, the transcript level of HcLK in the first instar larvae was the highest (1.81-fold that in eggs) and that in the seventh instar larvae was the lowest (0.47-fold that in eggs). The expression level of HcLK in the hindgut was 24.45-fold of that in the head (Figures 3A,B). The expression of HcLK in silk gland, foregut, Malpighian tubules, testis, and ovary was 0.24-fold, 0.74-fold, 0.52-fold, 1.87-fold, and 0.38-fold of that in head tissue, respectively, and did not differ significantly. The HcLKR expression in the first and fifth larval stages was similar but significantly higher than that at other instar stages (p<0.05, Figure 3C). Compared with that in the head, the transcript level of HcLKR in the hindgut was the highest (32.96-fold that in the head and that in the fat body was the lowest 0.0004-fold that in the head). The HcLKR expression in the epidermis, silk gland, foregut, Malpighian tubules, ovary, and testis was 0.0004–1.16-fold that in the head (Figure 3D).

The ORF of the HcLKR was inserted into the expression vector pcDNA3.1-Myc-His to construct a recombinant plasmid for stable expression. The HcLK gene encodes a 338-amino acid polypeptide (Figure 1A), which is a precursor of three LKs – LK-1–3 (Figure 4A). Notably, HEK293 cells expressing HcLKR responded to all HcLKs at a concentration of 1μM. The dose response of LKR to LKs was further investigated (Figure 4B). Of the three tested LKs, LK-2, and LK-3 stimulated LKR at lower concentrations, with EC₅₀ values of 28.0 and 8.44nM, respectively, whereas LK-1 showed a lower activity (EC₅₀ values: 90.44nM).

Considering the induction of HcLK and HcLKR mRNA expression by starvation stress, we investigated whether HcLK and HcLKR gene expression in the systemic silence plays a functional role in organismal stress tolerance employing knockdown of HcLK and HcLKR via dsRNA microinjection. The HcLK and HcLKR knockdown larvae showed ~80% lower LK and LKR mRNA levels than the control dsEGFP larvae after 96h (Figures 5A,B). Next, the survival of H. cunea RNAi larvae was investigated following desiccation and starvation stress.

Under desiccation and starvation stress, HcLK and HcLKR RNAi larvae survived longer than control larvae (Figures 5C,D). To determine whether the difference in survival rates of larvae stems from changes in water content, the water content in H. cunea larvae microinjected with dsEGFP, dsLK, and dsLKR were assayed after 48h of desiccation treatment. As expected, H. cunea larvae with dsLK and dsLKR silencing contained more water than those in control dsEGFP group (Figure 5E).

The expression of ILP genes in H. cunea was altered in dsLK and dsLKR larvae after 48h of starvation. Significant effects on HILP transcription were observed only for HILP2 (except dsLK treatment), HILP3, HILP4, HILP5, HILP6, and HILP8. The transcript levels of HILP2, HILP5, and HILP8 in the dsLKR larvae were significantly higher (1.31–4.42-fold) than those in the dsEGFP group. However, the transcript levels of HILP3, HILP4, and HILP6 in the dsLK larvae were significantly lower (0.33–0.52-fold) than those in the dsEGFP group (Figure 6). Complex results were also observed when LKR and LK were knocked down in H. cunea larvae, the LK signal negatively regulated HILP3, HILP4, and HILP6 expression but positively regulated HILP5 and HILP8 expression and played no significant regulatory role in HILP1 and HILP7 expression (Figure 6).

Our results suggested that LK signaling is associated with starvation stress. Thus, the HcLK and HcLKR knockdown mutants were found to affect food intake over different periods. The food intake of larvae microinjected with dsHcLK and dsHcLKR after starvation for 1day was significantly different from that of larvae microinjected with dsEGFP (Figure 7). During the feeding time tested, the food intake of dsHcLK and dsHcLKR larvae was significantly higher than that of the control dsEGFP larvae. The food intake of dsHcLK and dsHcLKR larvae on the day 1 was 1.61- and 1.62-fold higher than that of the control dsEGFP larvae, respectively (Figure 7). On day 4 of feeding, the food intake of dsHcLK and dsHcLKR larvae was 1.26- and 1.66-fold higher than that of the control dsEGFP larvae, respectively.