1 Introduction

Olfaction, one of the most important modalities of environmental perception, plays a crucial role in insect foraging, mating, defense, reproduction, and information communication (Carey and Carlson 2011; Mo et al. 2019). Several classes of proteins such as odorant-binding proteins (OBPs), chemosensory proteins (CSPs), odorant-degrading enzymes (ODEs), odorant receptors (ORs), ionotropic receptors (IRs), and sensory neuron membrane proteins (SNMPs) have been implicated in the peripheral olfactory system of insects (Leal 2013). As an essential protein in the normally functioning olfactory system, OBPs initially interact with external lipophilic chemicals in the process of olfactory sensing, transferring them to odorant receptor neurons to activate ORs distributed on the surface of dendritic membranes (Brito et al. 2016). Since the first discovery of OBPs in the antennae of Antheraea polyphemus (Vogt and Riddiford 1981), an increasing number of OBPs have been identified in different taxa, including the Lepidoptera (Yang et al. 2016; Walker et al. 2019), Coleoptera (Dippel et al. 2014; Li et al. 2017), Diptera (He et al. 2016; Zhao et al. 2018), Hymenoptera (Forêt and Maleszka 2006; Liu et al. 2018), Orthoptera (Zhang et al. 2015; Jiang et al. 2018), and Hemiptera (Sun et al. 2017; Qu et al. 2020), through various genome and transcriptome sequencing projects.

Insect OBPs are a class of acidic water-soluble proteins with 130–150 amino acids and usually comprise six α-helical domains folded into a very stable hydrophilic pocket for ligand binding, which is distinct from vertebrate OBPs those belong to the lipid carrier protein family (e.g., retinol binding protein and β-lactoglobulin) characterized by the typical structure of 8 antiparallel β-sheets and 1 short α-helix (Flower et al. 2000). Some vertebrate OBPs, like buffalo OBP, have two terminal alpha helices (Muthukumar et al. 2018). It is widely believed that OBPs essential for a normally functioning olfactory system are specifically expressed in antennae (Li et al. 2016; Iovinella et al. 2018). According to amino acid sequence homology and ligand binding characteristics, insect OBPs can be divided into pheromone binding proteins (PBPs), general odorant binding proteins (GOBPs), and antennal binding proteins (ABPx) (Krieger et al. 1996). PBPs are specifically expressed by non-neuronal trichogen and tormogen cells in trichoid sensilla and may enhance both the specificity and sensitivity of olfactory receptors to insect sex pheromones (Liu et al. 2015; Chang et al. 2015). Studies have shown that inhibiting PBP expression may seriously disrupt the response of males to sex pheromones released by females (Dong et al. 2017). Other OBP subfamilies play a central role in identifying general odors and pheromones (Yin et al. 2013). In recent years, gene knock-out, RNA interference (RNAi), and electrophysiology studies have been used to directly illustrate the essential role of OBPs in the normal operation of the olfactory system (Zhu et al. 2016; Dong et al. 2017; Hu et al. 2019; Wang et al. 2020).

With an increased identification of OBP-family member genes, it has become clear that OBPs are also present in other non-olfactory regions of the body and various functions have been proposed for OBPs (Pelosi et al. 2018); for example, Plus-C AlinOBP14 is highly expressed in the head of adult Adelphocoris lineolatus, where it is proposed to function as an endogenous compound carrier (Sun et al. 2019). Furthermore, a classical OBP over-expressed in the midgut of Rhodnius prolixus is involved in transport of nutrients or other molecules with digestive function (Ribeiro et al. 2014). AlinOBP11 is predominantly expressed in adult mouthparts and display strong binding abilities to host secondary metabolites, thus suggesting a crucial role in feeding behaviors (Sun et al. 2016). The female-biased EoblOBPs expressed in legs are related to host-seeking and oviposition behaviors in Ectropis obliqua (Ma et al. 2016). It has been suggested that OBPs may be associated with distinctively physiological roles, such as a medium for non-volatile compound recognition and transport in gustatory system (Du et al. 2019) or olfactory function where OBPs solubilize, bind, and carry odor compounds through sensillum lymph to ORs (Vogt and Riddiford 1981; Pelosi and Maida 1990). While many studies have examined gustatory behavior in honey bees, little research has been done on OBPs involved in the perception of non-volatile compounds.

Apis cerana cerana Fabricius, with a sensitive olfactory system and strong resistance to environmental stress, is a functionally irreplaceable honey bee species indigenous to China, mainly in mountainous areas (Du et al. 2016). Sevefenteen OBP-encoding genes have been identified from the antennal transcriptome on adult workers of different ages, among which AcerOBP15 expression was significantly upregulated in 25-day-old antennae (Zhao et al. 2016). Current studies on OBPs from A. cerana cerana are focused on the classical OBPs, and there is no report on the study of the expression and function of Minus-C OBPs yet. Here we cloned and analyzed the cDNA sequence of AcerOBP15. The expression profiles in different tissues of different developmental stages were detected using qRT-PCR and western blotting, respectively. The binding affinity of purified recombinant AcerOBP15 to 45 candidate ligands was analyzed using a fluorescence competitive binding assay. Finally, RNAi experiments and electroantennogram (EAG) assays were performed to further investigate the roles of AcerOBP15 in the detection of tested volatiles. These studies provide insights into the dual role of OBPs in chemosensory mechanism of A. cerana cerana.

2 Materials and methods

2.1 Insect rearing and collection

Apis cerana cerana was fed in a laboratory apiary at the College of Animal Science, Shanxi Agricultural University. Capped brood combs close to eclosion from three healthy colonies were maintained in an artificial climate incubator (75–80% relative humidity [RH]; 34 ± 0.4°C). Adult bees were marked with non-toxic paint on the back of the thorax after emergence and returned to their hives until sampling. Workers were collected (on days 1, 5, 10, 15, 20, 25, and 30) and snap-frozen in liquid nitrogen, in pools of n = 100 bees from the same colony per sample. Antennae, head, thorax, abdomen, legs, and wings were dissected on dry ice and ground into powder using a mortar filled with liquid nitrogen and then stored at –80°C.

2.2 RNA isolation and cDNA synthesis

Total RNA was extracted using TRIzol™ reagent (Invitrogen) according to manufacturer’s protocol. First-strand cDNA was synthesized from 1 μg of total RNA using the PrimeScript™ RT Reagent Kit with gDNA Eraser (Perfect Real Time) (TaKaRa).

2.3 Cloning of full-length AcerOBP15 cDNA

Full-length cDNA sequences were amplified using PCR in a reaction volume of 50 μL which comprised 2.5 μL synthesized cDNA of antennae, 2 μL of each primer (10 μΜ), 25 μL of 2×EasyPfu PCR SuperMix (-dye) (TransGen), and 18.5 μL of Nuclease-free water. Primers used for PCR were designed using the Primer 3.0 plus server (http://www.bioinformatics.nl/cgi-bin/primer3plus/primer3plus.cgi) (Table S1). PCR products were analyzed by electrophoresis on 1.5% agarose gel and purified using the E.Z.D.A™ Gel Extraction Kit (Omega) and then subcloned into the pEASY®-Blunt3 cloning vector (TransGen). The adapter-ligated fragments were transformed into Trans1-T1 chemically competent cells. Following blue-white selection, single white colonies were cultured in Luria-Bertani (LB) broth with ampicillin (100 μg/mL) for scale-up propagation. Positive colonies were confirmed using PCR and sequenced at Huada Gene Research Center (Beijing, China).

2.4 Bioinformatic and phylogenetic analyses

Homology analysis was performed using BLASTp (http://blast.ncbi.nlm.nih.gov/Blast.cgi) on NCBI with an e-value cutoff of 0.01. Open reading frame (ORF) was found using ORF Finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html). Conserved domains were located using the NCBI Conserved Domains website (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). N-terminal signal peptides were predicted by SignalP 5.0 Server (http://www.cbs.dtu.dk/services/SignalP/). ProtScale (http://web.expasy.org/cgi-bin/protscale/protscale.pl) was used to perform hydrophobic analysis. The tertiary structure was determined by homology modeling in MOE v2014.0901 (Hu 2016). Amino acid sequences were aligned by ClustalW, and a phylogenetic tree was constructed by MEGA 6.0 using the neighbor-joining method, applying bootstrap values of 1000 replicates (Tamura et al. 2013).

2.5 qRT-PCR analysis

qRT-PCR assay was performed in triplicates using 7500 Real-time PCR instrument (Life Technologies) and SYBR® Premix Ex Taq™ II (Tli RNaseH Plus) (TaKaRa). The procedure included a reaction at 95°C for 30 s, followed by 45 cycles of 95°C for 5 s, and 62°C for 34 s. Relative expression was normalized to that of AcerArp1 (GenBank No: HM640276.1) using the comparative 2-ΔΔCt method. One-way ANOVA and Duncan’s method (SPSS 17.0) were used to determine the significance of differences. Statistical significance of mRNA levels was determined at P < 0.01.

2.6 Purification and antibody preparation of AcerOBP15

The ORF without signal peptide was amplified by PCR with EasyPfu DNA Polymerase. Purified PCR products were subcloned into pET-30a (+) vectors (Novagen) using T4 DNA ligase (Promega) and used to transform BL21(DE3) E. coli competent cells (Transgen). Recombinant colonies were transferred into LB culture medium containing 50 μg/mL kanamycin and cultured with shaking for 2.5 h at 37°C. When OD600 reached 0.6–0.8, the bacterial medium with 0.5 mM IPTG was cultured with shaking overnight at 28°C. After collection and sonication of the bacterial pellet, supernatants and inclusion bodies were separated and detected by SDS-PAGE. Recombinant proteins were identified by LC-MS-MS (Q-TOF) and purified using a Ni-NTA affinity column (BBI). After cleaving His-tag with recombinant bovine enterokinase (rEK) (Sangon), purified proteins were again chromatographed using Ni-NTA affinity columns. Following dialysis and ultrafiltration concentration, the protein was freeze-dried and stored at –80°C. The polyclonal antibody of AcerOBP15 was prepared by immunizing New Zealand rabbits with purified recombinant protein.

2.7 Western blotting analysis

Total proteins were extracted using RIPA lysis buffer containing 1 mM PMSF and protease inhibitor cocktail. Approximately 150 μg of total protein extract was separated on a 12% SDS-PAGE gel and transferred onto a 0.22 μm nitrocellulose filter membrane at 100 V for 1 h. Each membrane was blocked with 5% non-milk dissolved in 1×TBS for 1 h. Next, they were probed with primary antibody (diluted 1: 1000 in TBS with 0.1% Tween-20) at 4°C overnight and then washed four times with TBST and twice with TBS, for 5 min at a time. Membranes were then incubated with secondary antibody IRDye 800CW Goat anti-Rabbit IgG at 1: 20,000 dilution for 1 h, and washed four times in TBST and once in TBS for 5 and 10 min, respectively. Finally, immunoreactive bands were detected using the Odyssey® CLx Imaging System (LI-COR).

2.8 Fluorescence-based competitive binding assay

Binding affinities of the protein with ligands were assessed using the capacity of ligands to displace the fluorescent reporter, i.e., N-phenyl-1-naphthylamine (1-NPN) from the complex (Iovinella et al. 2011). Fluorescence spectra were recorded using a RF-5301PC instrument (Shimadzu) with 1 cm light path quartz and 5 nm slits for both excitation and emission. The excitation wavelength was 337 nm, and emission spectra were recorded between 360 and 550 nm. Firstly, the interaction of AcerOBP15 (1 μM) diluted in 50 mM Tris-HCl (pH 7.4) with 1-NPN was monitored by recording the increase in fluorescence upon titration with aliquots of 1 mM 1-NPN to final concentrations of 1–10 μM. Binding data were collected in three independent measurements. The dissociation constant of 1-NPN to AcerOBP15 (K1-NPN) was estimated from Scatchard plots in GraphPad Prism 6.0. Then, the affinities of 45 ligands were measured in competitive binding assays using 1-NPN as the fluorescent reporter at 1 μM concentration and 1–10 μM final concentrations of each ligand. The binding affinity (Ki) of each ligand for AcerOBP15 was calculated using the equation: Ki=[IC50]/(1+[1-NPN]/K1-NPN), where [IC50] is the ligand concentration at half-maximal fluorescent intensity, [1-NPN] is the concentration of free 1-NPN, and K1−NPN is the dissociation constant of the AcerOBP15/1-NPN complex. All chemicals including 1-NPN and 45 ligand compounds were dissolved in HPLC methanol to obtain a 10 mM as stock solution and stored at –20°C until experiments.

2.9 Synthesis and feeding of dsRNA

Primers for synthesizing double-stranded RNA (dsRNA) were designed using the E-RNAi tool (http://www.dkfz.de/signaling/e-rnai3//) (Table S1). The dsRNA of AcerOBP15 and green fluorescent protein (GFP) were synthesized using the T7 RiboMAX™ Express RNAi System (Promega) and diluted in nuclease-free water to a final concentration of 2.5 μg/μL. Newly emerged worker bees were maintained in an artificial climate incubator (65–75% RH; 28 ± 0.4°C), fed with a 30% honey solution during RNAi experiments, and randomly divided into four groups (n = 30 bees/group). Three biological replicates were performed for each group. Each solution (nuclease-free water, dsAcerOBP15, and dsGFP; 4 μL) was fed to a random individual once on the sixth day. The fourth group was designed as the non-treated group. Next, samples were collected at 24, 48, 72, and 96 h following ingestion, and qRT-PCR was performed to validate the effects of dsAcerOBP15 on AcerOBP15 mRNA expression.

2.10 EAG recording

Electrophysiological responses to odor chemicals in the antenna, before and after RNAi, were detected using EAG at room temperature. The chemicals were dissolved in hexane, which was also used as a blank control, to a final concentration of 500 μg/μL (Guo et al. 2018). Under anesthesia, single antenna was carefully removed from the base of several terminal segments excised at both ends. The prepared antenna was immediately connected to the reference and recording electrodes using Spectra 360 conductive electrode gel (Spectra) and placed 10 mm vertically from an airflow port. A constant flow of active carbon-filtered air (500 mL/min) was delivered by an air stimulus controller (ModelCS-55, Syntech). Prior to odor stimulation, a filter paper strip (1 cm × 5 cm) loaded with 10 μL of a test chemical solution was inserted into a Pasteur pipette. The applied stimulus airflow was 40 mL/min for 0.5 s. Each chemical was tested three times at 30 s intervals on 10 antennae. EAG responses of antennae were analyzed using EAGPro software (Syntech). Significant differences in EAG responses were determined using one-way ANOVA, followed by Duncan’s test, determined at P < 0.05.

3 Results

3.1 Gene cloning and sequence analysis

The amplified fragment cloned from A. cerana cerana was 505 bp in length, which is homologous to AmelOBP15 (GenBank No.: NM_001040208) with 69% amino acid sequence identity. AcerOBP15 ORF consists of 408 nucleotides and encodes a deduced protein of 135 amino acids, containing four conserved cysteines (Figure 1a). The target gene was submitted to GenBank under the accession number KT588076.

Fig. 1
figure 1

Molecular characterization of AcerOBP15. (a) Nucleotide and deduced amino acid sequences of AcerOBP15 cDNA. Non-coding region and ORF are indicated with lowercase and uppercase letters, respectively. Signal peptide is underlined, and conversed cysteine residues are boxed in black. (b) Predicted three-dimensional structure of AcerOBP15. Alpha helices are indicated by α1–α7. Rod-like structure represents the disulfide bond. (c) Ramachandran plot for AcerOBP15. Dark green dots represent the residues in favored regions; yellow dots represent the residues in allowed regions. (d) Amino acid sequence alignments of AcerOBP15 with the known OBPs from other Hymenoptera insects. Conversed cysteine residues were indicated by black triangles. (e) Phylogenetic tree of OBPs reconstructed from Hymenoptera insects based on amino acid sequence alignment. The branch labels are bootstrap values (%), which are based on 1000 replicates. The scale bar is 0.2.

Full size image

3.2 Structural analysis of AcerOBP15

AcerOBP15 has a signal peptide with 16 amino acids at the N-terminus and possesses a PBP_GOBP superfamily domain between residues 30 and 126. According to the algorithm of Kyte and Doolittle (Vogt et al. 1999), AcerOBP15 was predicted to be hydrophilic, containing several obviously hydrophobic regions for binding liposoluble odorant. The crystal structure of AmelOBP14 (PDB ID 3S0G) was selected as the model with 83% coverage and 39% identify to AcerOBP15. The 3D structure of AcerOBP15 possesses seven α-helices (Figure 1b), which were observed at positions Ile3–Thr21 (α1), Gln25–Glu33 (α2), Lys41–Asn53 (α3), Pro65–Met72 (α4), Glu78–Cys88 (α5), Pro96–Ser108 (α6), and Thr114–Leu117 (α7). The core helices of classical OBPs are formed by α1–α6, and α7 is exposed to the surface of the protein at the C-terminus. Two disulfide bridges were observed between Cys17–Cys49 and Cys88–Cys106, connecting α1 to α3 and α5 toα6, respectively. The Ramachandran plot for AcerOBP15 showed that 95.56% of the amino acid residues are located in the favored regions, the remaining 4.44% in allowed regions (Figure 1c), indicating that the predicted structure of AcerOBP15 is stable and reasonable.

3.3 Phylogenetic analysis of AcerOBP15

An alignment of the amino acid sequences of AcerOBP15 and corresponding OBPs from other Hymenoptera species is shown in Figure 1d. AcerOBP15, AmelOBP15, AcerOBP14, AmelOBP14, AcerOBP21, and AdorGOBP56a-like all have four conserved cysteine sites missing the second and fifth cysteine and belong to the Minus-C OBP subfamily (Hekmat-Scafe et al. 2002). BterGOBP56d-like, BimpGOBP56d-like, MmedOBP17, MdemGOBP56d, NvitOBP17, MrotGOBP56a-like, CcosGOBP56h-like, TpreGOBP56a-like, and SinvGOBP69a-like contain a six-cysteine signature and a common pattern, C1-X25-27-C2-X3-C3-X37-39-C4-X7-8-C5-X8-C6, in which X represents any amino acid (Pikielny et al. 1994). The phylogenetic tree showed that all Minus-C OBPs were part of the same group with AcerOBP15 and AmelOBP15 sharing the closest relationship (Figure 1e).

3.4 Expression pattern analysis of AcerOBP15

AcerOBP15 transcript levels in antennae and legs were significantly more abundant than that in head, thorax, abdomen, and wings (P < 0.01), with the lowest expression recorded in the abdomen (Figure 2a). The AcerOBP15 expression pattern in antennae and legs of workers varied with the developmental stage (Figure 2b). The expression profile of AcerOBP15 in antennae was the lowest at 1-day-old, increased gradually at 5- and 10-day-old, and then decreased markedly at 15-day-old. The expression levels in 20-, 25-, and 30-day-old antennae were more stable and significantly higher than other days (P < 0.01). Generally, the expression profile of legs trended upwards as days of age increased, reaching peak expression at 25-day-old. Although protein level of AcerOBP15 in antennae and legs detected by western blotting had a tendency opposite to mRNA level (Figure 2c and 2d), both reached high expression levels in foragers (approximately 20- or 25-day-old).

Fig. 2
figure 2

Expression pattern analysis of AcerOBP15. (a) mRNA expression levels of AcerOBP15 in different tissues of workers at different ages. An: Antennae; H: Head without antennae; T: Thorax; Ab: Abdomen; L: Legs; W: Wings; 1-30 d: age of adult worker bees. Different uppercase letters above bars indicated significant differences in expression levels between different tissues (P < 0.01, Duncan’s multiple range test). (b) Relative expression levels of AcerOBP15 mRNA in the antennae and legs of workers at different ages. Lowercase letters indicated extremely significant differences in the expression levels in the same tissues of different ages (P < 0.01, Duncan’s multiple range test). (c) and (d) Western blotting analysis of AcerOBP15 in antennae and legs at different ages.

Full size image

3.5 Competitive fluorescence binding assay of AcerOBP15

Recombinant AcerOBP15 was expressed in an E. coli expression system and purified using Ni-NTA affinity columns. A single band of approximately 15 kDa was observed on the gel after His-tag removal and then purified again (Figure 3a). AcerOBP15 was titrated with increasing concentrations of 1-NPN as the fluorescence reporter, from which both saturation and linear Scatchard plots were obtained, with a K1-NPN of 3.318 μM (Figure 3b). A total of 45 candidate ligands were selected to measure the binding characteristics of AcerOBP15. The fluorescence competitor assay curve and the values of Ki for each ligand with AcerOBP15 are shown in Figure 3c and Table S2. AcerOBP15 displayed the strongest binding affinity to myrcene, geranyl acetate, and β-citronellol, with Ki values of 6.50 ± 0.07, 7.53 ± 0.33, and 7.54 ± 0.21 μM, respectively. Moreover, alarm pheromones (2-heptanone and isoamyl acetate), Nasonov pheromones (nerol and geraniol), seven plant volatiles (3-carene, eucalyptol, 3,4-dimethylbenzaldehyde, hexyl acetate, 2-phenylethanol, 1-octanol, and ocimene), and two carbohydrates (D-fructose, and D-(+)-trehalose dihydrate) showed relatively stronger binding affinities to AcerOBP15, with a Ki of 10–20 μM. Five volatiles (eugenol, linalool, β-Ionone, methyl salicylate, and myristic acid), four carbohydrates (sucrose, D-(+)-maltose monohydrate, D-(-)-arabinose, and stachyose hydrate), three amino acids (L-Proline, L-Glutamic acid, and L-Lysine), one bitter-tasting substance (Amygdalin), and two salts (KCl and NaCl) had relatively lower affinities (Ki = 20–40 μM) to AcerOBP15. The remaining candidate ligands exhibited very weak or no affinity to AcerOBP15.

Fig. 3
figure 3

Purification and competitive fluorescence binding assay of AcerOBP15. (a) SDS-PAGE analysis of the recombinant AcerOBP15. M: Protein molecular weight marker; 1–3: Induced E. coli AcerOBP15 at 28, 30 and 37°C, respectively. 4: Non-induced E. coli AcerOBP15. 5: Inclusion body; 6: Supernatant; 7: Purified recombinant AcerOBP15; 8 and 9: Purified recombinant AcerOBP15 with His-tag cleaved by rbEK. (b) The binding curve and relative Scatchard plot of 1-NPN and AcerOBP15. (c) Competitive binding curves of AcerOBP15 to 45 ligands. Data are represented as means of three independent experiments.

Full size image

3.6 Effect of dsRNA on AcerOBP15 mRNA expression

RNAi efficiency on AcerOBP15 was estimated according to mRNA expression after dsAcerOBP15 intake. The result of gene silencing is shown in Figure 4a. Ingestion of dsRNA decreased AcerOBP15 expression levels. Generally, mRNA expression levels of AcerOBP15 in the dsAcerOBP15 group at 24, 48, 72, and 96 h were significantly lower than that in the dsGFP, water, and non-treated groups (P < 0.01). Compared with non-treated groups, the gene-silencing efficiencies at the aforementioned four time points were 75.75, 91.83, 94.53, and 86.06%, respectively. Thus, 72 h post-ingestion was chosen for EAG recording.

Fig. 4
figure 4

Relative EAG responses of A. cerana cerana to three volatiles before and after RNAi. Data represents the mean values ± S.E.M of three independent replicates. (a) qRT-PCR analysis of AcerOBP15 transcripts in worker bees after the ingestion of nuclease-free water, dsGFP and dsAcerOBP15 for 24, 48, 72, and 96 h. The expression of AcerOBP15 was normalized using Arp1. Different uppercase letters above bars indicated significant differences in the expression levels under different treatments (P < 0.01). (b), (c), and (d) Electrophysiological responses to 2-heptanone, isoamyl acetate and myrcene after the ingestion of dsRNA for 72 h. Different lowercase letters above bars indicated that there were differences in relative EAG response between dsRNA-treated and control worker bees (P < 0.05).

Full size image

3.7 EAG analysis before and after RNAi

Our previous research identified relatively high EAG values for A. cerana cerana foragers to myrcene, 2-heptanone, and isoamyl acetate, which were significantly higher than those of other compounds (P < 0.05) (Table S3). In conjunction with the results of the competitive binding assays, we selected the above-mentioned three volatiles on which to perform EAG recording before and after RNAi in order to dissect the physiological function of AcerOBP15. There was no significant decrease in response to 2-heptanone and isoamyl acetate in dsAcerOBP15 groups compared with other control groups (P > 0.05) (Figures 4b and 4c). However, EAG response to myrcene was significantly reduced following dsAcerOBP15 ingestion (P < 0.05) (Figure 4d).

4 Discussion

In the present study, we successfully cloned the full-length cDNA of AcerOBP15 encoding a protein that belongs to the Minus-C OBPs subfamily. Two disulfide bridges formed by four conserved cysteine sites play an important role in the formation and stabilization of the protein’s structure. The analysis of amino acid homology and phylogenetic trees showed that AcerOBP15 was closely related to AmelOBP14 (Spinelli et al. 2012) and AcerOBP14 (Du et al. 2016), which had been identified as GOBPs, so we suggested that AcerOBP15 may also be a member of the GOBP family. Temporal-spatial expression patterns indicated that AcerOBP15 was differentially expressed in the antennae, head, thorax, abdomen, legs, and wings of adult worker bees. It is generally believed that OBPs specifically expressed in antennae are related to the recognition of insect pheromones, while OBPs simultaneously expressed in other tissues are related to the recognition of common odor substances (Gu et al. 2010). Hence, it is speculated that AcerOBP15 is a GOBP for sensing and binding general floral substances and a few pheromones, a finding consistent with the results of phylogenetic tree analysis in this study.

Antennae are important olfactory sensory organs, comprising a large number of densely packed olfactory sensilla. Early studies have shown that most insect OBPs have antennae-biased expression, especially PBPs, which are mainly expressed in male antennae (Feng et al. 2017; González-González et al. 2019). Similarly, AcerOBP15 was also expressed at a high level in antennae. As a social insect, workers have to rely on their complex olfactory system to regulate all activities inside and outside the hive, such as tending to larva and the queen, recognizing nestmates, and searching for nectar and pollen plants (Graham 2015). As AcerOBP15 expression levels in the antennae of adults were found to vary with increasing age, this may suggest the regulatory effect of AcerOBP15 on olfactory perception in A. cerana cerana. Besides antennae, high expression of AcerOBP15 was also observed in the legs of worker bees. Insect chemosensillar on legs possess gustatory receptor neurons (GRNs) for the detection of specific non-volatiles after landing on plants (Jeong et al. 2013; de Brito Sanchez et al. 2014). Many studies have demonstrated that OBPs differentially expressed in the legs may be involved in taste recognition, contributing to host-selection and food-foraging in insects (Brito et al. 2016). DmelOBP57d/e were largely expressed in leg taste sensilla and involved in the sensation of hexanoic and octanoic acids, impacting taste perception and host-plant determination in Drosophila sechellia (Matsuo et al. 2007). Immunocytochemical analysis has clearly demonstrated that AlinOBP11 predominately expressed in adult legs is strongly located in the tarsal gustatory sensillum lymph of Sch2 and therefore is a good candidate for non-volatile substance detection in A. lineolatus host plant-seeking behavior (Sun et al. 2017). In situ hybridization and analysis of binding ability indicated that the over-expressed MmedOBP19 in legs was concentrated at the base of the gustatory sensilla on tarsi of the foreleg, and had strong affinities with various non-volatile plant secondary metabolites, thus suggesting a major role in taste perception (Yang et al. 2017). Therefore, we speculated that AcerOBP15 strongly expressed in the legs was a good candidate for non-volatile substance detection, in addition to odorant detection in the olfactory system.

Surprisingly, AcerOBP15 expression in antennae and legs had contrasting trends at the protein and mRNA levels, likely due to many levels of regulation of gene expression in eukaryotes, as well as the potential impacts of post-transcriptional control or translational modifications (James et al. 2016). However, regardless of mRNA or protein level, high expression of AcerOBP15 was consistently recorded in the antennae and legs of foragers (about 20- and 25-day-old). There are plentiful and diverse sources of nectar and pollen in the outside world compared with the enclosed beehive; here it is needed that genes associated with olfaction and gustation are highly expressed to effectively identify and transport chemicals from the external environment. Thus, we speculate AcerOBP15 is probably related with foraging behavior. It is worth noticing that the legs of foragers had the highest expression level at 25- and 30-day-old which were significantly higher than the antennae (P < 0.01). Honey bees have strong learning and memory abilities. They discover the nectar sources from a distance through olfactory and visual systems and form conditioned reflexes on their orientation and distance (Ding 2014). But bees still rely on gustation to estimate the quality of nectar sources at close range, due to nectar concentration variations related to environmental temperature and humidity.

Subsequent competitive fluorescence binding assays provide further insight into the understanding of physiological roles of AcerOBP15. The results clearly indicated that AcerOBP15 could extensively bind a wide range of odor compounds. In this study, 17 flower scents and 11 bee pheromones were selected to assess the binding ability of AcerOBP15. A majority of flower volatiles showed certain binding capabilities to AcerOBP15, and myrcene exhibited the highest binding affinity (Ki = 6.50 ± 0.07 μM), which was similar to a previous finding that SaveOBP9, a classical OBP from Sitobion avenae, had high (Ki < 10 μM) binding ability to myrcene (Ullah et al. 2020). This suggests that AcerOBP15 may be involved in olfactory orientation for discovering nectar and pollen sources, consistent with its high expression in the antennae of foragers. Recent research by Li et al. (2020) found that an additional helix (i.e., the seventh) in Minus-C OBPs with two disulfide bonds could result in greater flexibility and adaptability for this protein to bind to different compound molecules. Similarly, AcerOBP15 belongs to the Minus-C OBP subfamily and has a small α-helix (α7) at the C-terminus. Such structure may add the flexibility and spectra to the binding of AcerOBP15 with small molecules. Nasonov pheromone is produced by Nasonov gland of workers and is used as a long-range attractant for bees to the colony, food sources, or to promote swarm clustering, which plays a guiding role in many behaviors (Trhlin and Rajchard 2011). Studies have also shown that the Nasonov pheromone can maximally maintain the population of a colony and may also lure workers visiting flowers to improve crop pollination efficiency (Gao and Zhao 2014). AcerOBP15 displayed a high binding affinity for the Nasonov pheromone compounds, nerol and geraniol, but not to farnesol. Therefore, we speculate that AcerOBP15 plays a role in colony population maintenance and outdoor orientation to food sources. In contrast to AcerOBP15, AcerOBP11 exhibited a strong binding affinity for farnesol, but not to nerol and geraniol (Song et al. 2018), indicating a functional differentiation among AcerOBPs, suggesting that the synergistic regulation of some physiological behaviors require different AcerOBPs.

Alarm pheromones released by fighting bees (usually foragers) are necessary to promote defensive behavior in honey bees, thus protecting food sources and the brood from invading predators and robbing honey bees from other colonies. There are two specialized kinds of alarm pheromone: isoamyl acetate (found in the dorsal anterior portion of the sting) and 2-heptanone (secreted by the mandibular gland) (Brodmann et al. 2009). In our study, AcerOBP15 strongly bound 2-heptanone and isoamyl acetate with Ki of 10.22 ± 0.18 and 10.97 ± 0.26 μM, respectively, suggesting it may play a role in worker behaviors designed to maintain and defend the colony. Similarly, AcerOBP10 and AcerOBP11 also had strong binding affinities for these two volatiles (Song et al. 2018; Wu et al. 2016), indicating that recognition and transport of alarm pheromone is regulated by several different OBPs in the colony of A. cerana cerana.

The queen mandibular pheromone (QMP) produced by the queen is largely composed of (E)-9-oxodec-2-enoic acid (9-ODA), both enantiomers of (E)-9-hydroxydec-2-enoic acid (+/-9-HDA), methyl-p-hydroxybenzoate (HOB), and 4-hydroxy-3-methyoxyphenylethanol (HVA) (not found in A. cerana), and regulates many aspects of colony organization. 9-ODA is a typical sex attractant for drones and a social regulator for workers, while 9-HAD and HOB may play some role in stimulated drone attraction to virgin queens. HVA is high in laying queens, and strongly tied to laying behavior. Although HVA is a unique QMP component in A. mellifera, behavioral research has identified that it also has a certain attraction to drones of A. cerana. cerana (Yang et al. 2018). AcerASP1 may contribute to the A. mellifera queen recognition by A. cerana workers, due to its affinity to HVA (Weng et al. 2013). AcerOBP11 shows high affinity to HOB, 9-ODA, and HVA, and plays an important role in worker bee perception of QMPs (Song et al. 2018). Yet surprisingly, neither 9-ODA nor HVA show binding affinity to AcerOBP15, implying no effect on the regulation of the reproductive system of workers or mating behaviors between queen and drones. However, whether AcerOBP15 specifically binds to other components of QMP is still unknown. Moreover, as AcerOBP15 had weak or no binding to brood pheromone components, it may have no influence on worker bee behaviors, such as brood attention and cell capping.

In addition, AcerOBP15 had relative binding ability to 12 of the 17 non-volatile compounds derived from nectar and honey, in which D-fructose and D-(+)-trehalose dihydrate showed relatively stronger binding affinities. In Apolygus lucorum, both Plus-C OBP3 and classical OBP35, highly expressed in forelegs of male and female bugs, had strong binding affinity to non-volatile secondary metabolites of host plant (i.e., gossypol), and may be involved in close or contact chemical communication (Li 2020). Hence, coupled with its specific tissue distribution within legs, we suggest AcerOBP15 may also function as a non-volatile compound detector and carrier in the gustatory system, when honey bees land on a suitable food plant, tasting the nectar and pollen prior to foraging. In particular, fructose acts as a source of nutrition in larval development and in caste determination. The content of fructose in nectar is high, and foragers of Apis cerana were found to be more sensitive than A. mellifera (Wang 2013). Thus, A. cerana cerana might employ the legs enriched AcerOBP15 to perceive and discriminate non-volatile substances; however, from the results of the current study, it is difficult to determine where AcerOBP15 is involved in the chemical perception of the legs of honeybee.

EAG, as an effective and reliable detection method, has significant applications in the research of insect olfactory sensitivity. In the present study, the floral compound myrcene elicited the strongest response, in agreement with previous study by Jung et al. (2014). In conjunction with RNAi widely used for gene silencing in gene function investigations, we found that the EAG response elicited by myrcene was significantly reduced (>65%) after the ingestion of dsAcerOBP15 (P < 0.05). Thus, it is possible that AcerOBP15 has an indispensable role in myrcene recognition. However, the EAG response to isoamyl acetate and 2-heptanone in dsAcerOBP15 group showed no significant difference to the control group. This is quite possible since perception of alarm pheromone requires more than two OBPs including AcerOBP10 and AcerOBP11 (Song et al. 2018; Wu et al. 2016).

5 Conclusion

In summary, AcerOBP15, a member of the Minus-C OBP subfamily, is highly abundant in forager antennae and legs of A. cerana cerana. It can bind small hydrophobic compounds including pheromone components, common odor molecules, and non-volatile substances, suggesting dual role in olfactory and gustatory perception when foraging. This study not only prompts the theoretical basis of OBPs-mediated olfactory recognition, but also extends the understanding of different physiological roles of AcerOBPs.