Structure and Protein-ligand Binding Studies of Pheromone Binding Protein-like Sol g 2.1 Protein from Solenopsis geminata Fire Ant Venom

doi:10.21203/rs.3.rs-3319477/v1

Download PDF

Research Article

Structure and Protein-ligand Binding Studies of Pheromone Binding Protein-like Sol g 2.1 Protein from Solenopsis geminata Fire Ant Venom

https://doi.org/10.21203/rs.3.rs-3319477/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Sol g 2 is the major protein in Solenopsis geminata fire ant venom. It shares the highest sequence identity with Sol i 2 (S. invicta), and shares high structural homology with LmaPBP (pheromone binding protein (PBP) from the cockroach, Leucophaea maderae). We examined the specific Sol g 2 protein ligands from the fire ant venom. The results revealed that the protein naturally formed complexes with hydrocarbons, including decane undecane, dodecane, and tridecane, in aqueous venom solutions. We found that decane has a higher affinity binding with the recombinant Sol g 2.1 protein (rSol g 2.1) than other specific ligands. Surprisingly, the mixture of the alkanes showed higher binding affinity with rSol g 2.1 protein than any single one, a positive blend effect. The ligands were examined further by molecular docking simulations, which showed allosteric binding sites in the Sol g 2.1 protein model. We also performed a trail-following bioassay and found that rSol g 2.1 and the mixture of hydrocarbons elicited S. geminata worker ants to follow the trails for a longer time and distance than only a mixture of hydrocarbons, which suggests that Sol g 2.1 protein may delay the evaporation of the hydrocarbons. Interestingly, the fully reconstituted venom, which contained piperidine alkaloids and trail pheromones, has the highest attraction to the ants. Therefore, the mixture of hydrocarbons and the piperidines have a synergistic effect on the trail-following of the ants when both were added to the protein.

Solenopsis geminata

Venom protein Sol g 2.1

Pheromone-binding proteins (PBPs)

Trail pheromones

The tropical fire ant (Solenopsis geminata) is one of the ubiquitous ant species in Thailand. They are known as an aggressive ant species, due to their behavior and venomous painful sting. The venom is produced in the venom glands, stored in the poison sac, and then secreted through a sting at the tip of the abdomen. The venomous secretion comprises 90–95% basic piperidine alkaloids and four major allergen proteins (Shi et al. 2015). The piperidine alkaloids are mainly 2-methyl-6-alkyl piperidines with different lengths of alkyl or alkenyl chains (Chen et al. 2019). The alkaloids are used in defense against possible predators. Furthermore, venom alkaloids help these ants control and avoid competition for their hosts (Tschinkel 2006). Derivatives of solenopsins which are the fire ant potent piperidine alkaloids have been found to have antibacterial, antifungal, insecticidal, and antiangiogenics activities (Hoffman 1995; Touchard et al. 2016; Blum 1992; Carvalho et al. 2019). As mentioned, the S. geminata venom also contains four major proteins, including Sol g 1, 2, 3, and 4, which are responsible for allergenic activity (dos Santos Pinto et al. 2012). Importantly, Sol g 2 is one of the major protein components in the venom (Sukprasert et al. 2012).

Sol g 2.1 protein (GenBank: UYX46120.1) shares 83.04% sequence identity with Sol i 2 (S. invicta); both consist of five α-helices and three intramolecular disulfide bridges, forming a hydrophobic cavity. Interestingly, Sol i 2 has a high binding affinity with hydrophobic molecules such as (E)-β-farnesene aphid alarm pheromone, plant volatiles, analogs of ant trail pheromones like decane and undecane, and short fatty acids (Borer et al. 2012). Moreover, the three-dimensional structure of the Sol i 2 is highly similar to odorant-binding proteins (OBPs), which are located in the olfactory organs of various insect species (Fan et al. 2011). OBPs are small soluble proteins consisting of 130–150 amino acids, and they are found in the sensillum lymph of insects (Leal 2013; Vieira and Rozas 2011). These proteins are mainly involved in the peripheral olfactory system, in the sensillum lymph fluid, by acting as a mediator between odorants and their membrane receptors (Honson et al. 2005; Terrado et al. 2021). The OBP family also includes proteins that specifically bind semiochemicals, e.g., pheromone binding proteins (PBPs), which transport and scavenge pheromones, to activate and protect pheromone receptors (PRs). PBPs are acidic proteins that contain 120–150 amino acids and six conserved cysteine residues. Interestingly, Sol g 2.1 and LmaPBP (PBP from the cockroach, Leucophaea maderae) share high structural homology and an inner hydrophobic cavity (Borer et al. 2012). Thus, Sol g 2.1 protein may be involved in binding and transporting hydrophobic molecules like ant pheromones or straight-chain alkyl substituents of piperidine alkaloids, to solubilize them in aqueous environments (Fan et al. 2011; Vieira and Rozas 2011; Borer et al. 2012; Das et al. 2018). Nevertheless, specific Sol g 2.1 protein ligands in S. geminata crude venom have not been reported.

In this research, we isolated endogenous ligands of Sol g 2.1 protein in the fire ant crude venom by gel filtration and then investigated the binding of these endogenous compounds to Sol g 2 and their function further. Using rSol g 2.1, we studied its binding of endogenous ligands from venom using a competitive binding assay. In addition, we predicted the structure of the protein complexed with various ligands by molecular docking simulations. The function of Sol g 2.1 protein and its endogenous ligands was examined by a trail-following bioassay.

Harvesting and Extraction of S. geminata Crude Venom. To obtain S. geminata venom for the studies, adult workers were harvested from Khon Kaen City in Thailand. After stimulation, ants secreted their venom through the stinger. The venom was collected drop-by-drop using a capillary tube, and collected venom was then dissolved in PBS buffer pH 7.4 (1.8 mM KH₂PO₄, 2.7 mM KCl, 137 mM NaCl, and 10 mM Na₂HPO₄). Next, 25 µL of the crude venom was extracted in 50 µL of 1% ethyl acetate in distilled hexane. The mixture was then shaken until completely separated into two layers. An organic phase was transferred into a new test tube with Na₂SO₄ for drying. The extraction process was repeated again. Next, the extracted solution was analyzed by gas chromatography coupled with mass spectrometry detection (GC/MS) on a Clarus 690-GC interfaced with a Clarus SQ8T-MS (Perkin Elmer, MA, USA). The GC was equipped with a 30 m fused silica SBP-5 column (0.25 mm i. d., 25 µm film thickness, Supelco, Bellefonte, PA, USA), He was used as a carrier gas (1.0 mL/min). The GC was programmed as follows: 80°C for 5 min at the initial temperature, 10°C/min to 250°C (hold for 15 min). The split/splitless injector was set to 250°C and operated in splitless mode. The ionization was accomplished using electron impact (EI), and the MS Scan was collected between m/z = 50–450.

Binding Assay. Binding of endogenous ligands to Sol g 2.1 in the crude venom was investigated by size exclusion chromatography (P-2 gel, BioRad, molecular size limit 2 kDa) (Plettner et al. 2000). This approach is used for separating large molecules (protein-ligands complexes) from smaller molecules (free ligands), which were retained on the column. In this experiment, 50 mg of gel beads were added into a small pipette tip (200 µL capacity) fitted with a cotton plug. PBS buffer (50 µL) was then added to swell the gel. We applied 50 µL of the S. geminata crude venom onto the column, and the elution fraction from a column bed was then collected. After the column was washed with 50 µL PBS buffer, the flow-through solution was pooled into the existing filtrate. The combined flow-through solution was carried out for protein identification and analysis of endogenous ligands and Sol g 2 protein on GC/MS and SDS-PAGE, MALDI-TOF MS, LC-MS/MS, respectively. Specifically, the first, 25 µL of the flow-through was extracted in 1% ethyl acetate in distilled hexane. After that, the extract was analyzed on GC/MS as described above.

To identify the Sol g 2 protein, the flow-through solution was analyzed on one-dimensional SDS-PAGE. The molecular mass of proteins was then determined using the matrix-assisted laser desorption/ionization coupled to time-of-flight mass spectrometry (MALDI-TOF MS) (Bruker Launches autoflex (TM) speed MALDI-TOF(/TOF)). The autoflex TOF/TOF (Bruker Daltonics flexAnalysis) was used to examine the mass spectra. The linear operation mode was used to determine the acquisition settings, and the positive polarity and total 3000 spectra were summed. Afterward, the expected band of Sol g 2 protein on the SDS-PAGE was cut, extracted, and proteolyzed by trypsin, followed by liquid chromatography and tandem mass spectrometry, by LC-MS/MS (Sukprasert et al. 2012). Peptides were separated on a nano-liquid chromatography system (EASY-nLC II, Bruker). The sample was loaded onto an EASY-Colum (10 cm, i.d., 75 µm, 3 µm, C18-A2, Thermo Scientific), using 0.1% formic acid in water and acetonitrile for mobile phase A and B, respectively. The LC was coupled to a nano-spray ESI-Ion trap MS (Sciex tripletof® 6600+) and a Time of Flight (TOF) analyzer, for the MS/MS scan mode. The PEAKS DB Protein Identification LC-MS/MS Software was used to identify peptides from the peaks. As a precursor mass search type, fragment mass tolerance is 0.1 Da and monoisotopic (Astiti et al., 2021).

Expression of Sol g 2.1 Protein in E. coli. The Sol g 2.1 coding sequence (GenBank: UYX46120.1) in the pProEx-HTB expression vector, which is composed of His₆-tagged protein at the C-terminus was expressed in E. coli BL21 (DE3) pLysS competent cells (Promega, Malaysia) as described previously (Srisong et al. 2018). A single colony was inoculated in Luria-Bertani (LB) medium containing 50 µg/mL Ampicillin at 37°C, overnight. The cell culture was induced with isopropyl β-D-1-thiogalactopyranoside (IPTG) after the OD₆₀₀ reached 0.5. After harvesting the cells by centrifugation (10,000 x g for 30 minutes at 4°C), the cell pellets were extracted in a lysis buffer (80 mM Tris, 200 mM NaCl, 1 mM EDTA, and 4% glycerol, pH 7.2). The protein accumulated largely in insoluble inclusion bodies, which were isolated by centrifugation at 10,000 x g for 30 min at 4 ºC. The inclusion bodies were then refolded using 8 N guanidinium HCl, according to Terrado et al (2020). After denaturation and renaturation, the soluble protein solution was purified using a nickel affinity column (His-Bind resin, Novagen), using 20 mM Tris-HCl pH 7.4, 500 mM NaCl, and 20 mM imidazole as a binding buffer. His₆-tagged proteins were eluted by increased concentrations of imidazole (50, 100, 250, and 500 mM in 20 mM Tris-HCl pH 7.4, 500 mM NaCl), following desalting. For delipidation, the purified protein was incubated with methyl-functionalized methacrylate HIC resin (hydrophobic interaction chromatography, Bio-Rad, Hercules, CA) in 50 mM Tris-HCl pH 7.4 at 4°C for three days on a rotary mixer (Srisong et al. 2018; Terrado et al. 2020).

Fluorescence Competitive Binding Assay. The fluorescence binding assay of rSol g 2.1 protein was conducted using the fluorescent probe N-phenyl-1-naphtylamine (NPN). To investigate affinity binding, a stock of 1 mM NPN in methanol was titrated into 2 µM rSol g 2.1 in 50 mM Tris-HCl (pH 7.4) to a final concentration range of 0–12 µM by using methanol as a negative control. The fluorescence intensity was measured on a PTI QuantaMaster fluorometer (Horiba Ltd., Kyoto, Japan) with 337 nm excitation wavelength and emission scanned ranging from 300–500 nm, at room temperature, in triplicate. To obtain the equilibrium dissociation (K_d) value, the fluorescence intensity at maximal emission wavelength 400 nm was plotted for each NPN concentration. Data were fitted to a specific allosteric binding model using GraphPad Prism 9 (GraphPad Software, San Diego, California). Y = B_max*X^h/(K_d^h + X^h) was used as an equation in the fitting model, where B_max is maximal specific binding in the same unit as Y, X is NPN concentration (µM), and Y is fluorescence intensity counts/second. Moreover, h is the Hill slope which is 1.0 for cases with one site. If there is more than one binding site per protein and there is cooperativity, then h > 1.0 and the graph takes on a sigmoidal appearance.

Binding of the endogenous ligands, including decane, undecane dodecane, and tridecane to rSol g 2.1 protein was measured by a competitive binding assay. In this procedure, 1 mL of rSol g 2.1 protein (2 µM) in 50 mM Tris-HCl (pH 7.4) containing NPN (4 µM) was titrated with each ligand (1 mM in methanol stock) to final concentrations of 0, 0.125, 0.375, 0.625, 0.875, 1.125, 2, 3, and 4 µM. Additionally, the same concentrations of the blended ligands, which consisted of decane (45%), undecane (36%), dodecane (15%), and tridecane (4%) were used to investigate the blend effect of Sol g 2.1 protein. The resulting isotherms were fitted (GraphPad Prism 9) to both a one-site model and an allosteric model. Since the latter gave a better fit (R² = 0.98). Where K_d is the equilibrium dissociation constant (µM) of modulator binding. The ternary complex constant is alpha; when alpha = 1.0, the modulator has no effect on binding, and when alpha is less than 1.0, the modulator decreases ligand binding.

In silico Studies, Homology Modeling & Molecular Docking of Sol g 2.1 Protein and Ligands. The homology models of Sol g 2.1 protein, which are based on Sol i 2 crystallized (S. invicta, PDB ID: 2ygu.1.A, 2.60 Å resolution) and LmaPBP (PDB ID: 1org.1.A, 1.7 Å resolution) templates were generated by using the SWISS-MODEL program (https://swissmodel.expasy.org/). The sequence identity of proteins was aligned on Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). The docking of Sol g 2.1 and the endogenous ligands was simulated using MOE version 2019 (Molecular Operating Environment). MOE protonate 3D, the three-dimensional structural (3D) model of Sol g 2.1 protein was protonated by ionization state and adding hydrogen atoms to the structure, as described previously (Terrado et al. 2020). Afterward, in the energy minimization step, the protein was energy minimized with the rigid water molecule constraints in the Amber 10 force field. Next, the ligand site finding was based on the Alpha Shapes center approach using the MOE Site Finder (Edelsbrunner and Mucke 1994). Each ligand was placed at the top of positive PLB ranks via dummy atoms. Each site was placed at the Triangle Matcher method, and the complex was scored at 30 poses of the London dG score tool. The induced Fit energy minimization model was refined at 5 poses by GBVI/WSA force field in which the ligand's free energy binding was computed in the S score (Labute 2009). The structure with the lowest S score is the best pose of the binding affinity model (Attique et al. 2019). The top rank of MOE docking S scores with the lowest RMSD (root-mean-square deviation of atomic positions) value of the Sol g 2.1 protein and various ligands were chosen (triplicates, average ± SEM (standard error of the mean)).

Trail-following Bioassay. The S. geminata colony was collected from Mueang Khon Kaen District, Khon Kaen, Thailand. The ants were acclimatized by being placed in a plastic cage box at room temperature in an open-air environment. They also were fed with 20% w/v of sugar in water and frozen crickets (Chalissery et al. 2019). Venom was collected drop-by-drop from 60 individual ants (each ant had an average of 20 drops) and was then dissolved in 60 µL PBS buffer as described previously (Sukprasert et al. 2012). After harvesting, the protein concentration of crude venom was measured by Bradford’s method. In this procedure, a positive control group was prepared from 1 µL of the crude venom in 25 µL PBS buffer (treatment C; 1 ant equivalent (AE)). Next, 40 µL of the crude venom stock was aliquotted and extracted in 80 µL of 1% ethyl acetate in hexane as described above. The upper phase was taken out into a new vial, which was called piperidine alkaloids and organic compounds. Afterward, 2 µL of the extracted solution was dissolved in 25 µL hexane (treatment P; 1 AE). Moreover, a reconstituted venom was constructed from 2 µL of the piperidine alkaloids extracted in 25 µL rSol g 2.1 protein in PBS to 1 ng/µL final concentration (treatment P + S; 1 AE). A mixture of medium-chain hydrocarbons, including decane (45%), undecane (36%), dodecane (15%), and tridecane (4%) was used as representative of the endogenous trail pheromone in S. geminata venom. Each compound was dissolved in hexane to 1 ng/µL final concentration (treatment M; 1 AE). In addition, the mixture of hydrocarbons was aliquotted into rSol g 2.1 protein in PBS buffer to 1 ng/µL final concentration, which gave the artificially reconstituted trail pheromone with the protein (treatment M + S). A fully reconstituted S. geminata venom, there was composed of 2 µL piperidine alkaloids extracted in rSol g 2.1 protein in PBS and the mixture of hydrocarbons, which were 1 ng/µL in a final concentration for each compound (treatment P + M + S; 1 AE). There were three negative control groups, including hexane, PBS buffer, and cleaned 1 ng/µL rSol g 2.1 protein in PBS buffer (S).

In the trail-following bioassay, each test stimulus was administered as 1 µL per arc (or 26 drops for the full circle) by using a micro syringe along the perimeter of a circular Whatman filter paper (90 mm in diameter, Sigma-Aldrich) (David Morgan 2009). The circular filter paper was marked with a circle 1 cm from the edge, and the circle was divided into 26 arcs. To begin the bioassay, the treated paper was placed in the center of an acrylic arena (30 cm × 17 cm × 10 cm), and a Falcon tube containing a single worker ant was placed 2.5 cm from the edge of the paper. Each ant was given 5 minutes in the arena to settle down before filming. The ant's movement was tracked for 10 minutes (N = 10) (Chalissery et al. 2019; Billen 1987; Campos Rda et al. 2016).

Statistical Analysis. GraphPad Prism 9 (GraphPad Software, San Diego, California) was used to analyze all data. R studio (version 2022.07.1) was used to visually generalize the ant-following distance response to all treatments for the bioassay analysis. ANOVA was used to examine the variance and significance, with Tukey's honest significant difference (HSD) test, p < 0.01.

Piperidine Alkaloid Profiles from S. geminata Venom Extraction. The venom contained approximately 2 µg/µL of total protein concentration. After extraction, the organic solution was then analyzed on GC/MS. The results revealed that there were 11 peaks detected (Fig. 1A). Peaks 1 to 4 were hydrocarbons, including decane, undecane, dodecane, and tridecane, which were identified using straight-chain hydrocarbon standards by comparing retention times and mass spectra. Piperidine alkaloids were detected at peaks 5–11, Peaks 5 and 6 had mass spectra corresponding to cis and trans-C9 2-methyl-6-n-nonylpiperidines, respectively. Peaks 7 and 9 had base and molecular mass ions at m/z 98 and 252 [M+], respectively, corresponding to cis- and trans-2-methyl-6-n-undecylpiperidines, also known as solenopsin A and isosolenopsin A, both prominent compounds in fire ant venom (90% and 9.2%, respectively, Table 1). Peak 8 had ions m/z 96 and 111, consistent with 1,6-didehydro-2-methyl-6-undecylpiperidine. The mass spectrum also showed base peak ion at 98 m/z and mass at 252 m/z (Chen et al. 2012; Hussein et al. 2016). Moreover, the mass spectra of peaks 10 and 11 showed a base peak at ion 98 m/z and mass at 281 m/z. After comparing with previous reports we found that these peaks could correspond to cis and trans-C13 (2-methyl-6-n-tridecylpiperidines), respectively (Table 1) (Chen et al. 2010).

Binding Assay. After complexes of proteins and ligands and any free ligands in the crude venom were separated using gel filtration column, the flow-through solution was then extracted and analyzed on GC/MS. The chromatogram showed that there were 4 peaks, including 1ʹ, 2ʹ, 3ʹ, and 4ʹ detected (Fig. 1B). These compounds were decane, undecane, dodecane, and tridecane with peak area ratios as 45%, 36%, 15%, and 4%, respectively. To identify protein content in the flow-through solution, MALDI-TOF MS was performed on this procedure. We found that the molecular weights of the proteins forming the fire ant crude venom were major at 13,274.48 followed by 14,112.86, 24,054.88, and 26,721.56 Da parallel of Sol g 2, Sol g 4, Sol g 3, and the fragments of Sol g 1, respectively. The results showed that Sol g 2 is a major protein in S. geminata venom (Fig. 2). Furthermore, the through-flow solution was separated using SDS-PAGE. From the results, we found that at approximately 15 kDa (band C), 37 kDa (band A), and 26 kDa (band B), which corresponded to Sol g 2, Sol g 1, and Sol g 3, respectively (Fig. S4) (Hoffman et al. 1988; Sukprasert et al. 2012). The expected band of Sol g 2 protein (band C) was found to be identical to the venom protein Sol g II (Accession AAY32926.1), which is an allergen protein in S. geminata venom (Table 2).

Table 2. Identification of the Sol g 2 protein contained in the flow-through solution after separating S. geminata crude venom by gel filtration column.

Band

Matched protein

Accession

-10lgP^a

Average mass^b

Peptide sequence

Coverage (%)^c

species

Venom protein Sol g II

AAY32926.1

129.29

15,370

KDIAECARTLPK

CENQPDDPLAR

RGVFDDPAPAAIKKK

S. geminata

^a -10lgP value was determined after LC-MS/MS analysis by PEAKS DB Software

^b An average mass of protein after the LC-MS/MS analysis

^c Percent coverage of amino acid sequences

Fluorescence Binding Assay. Fluorescent emission spectra revealed a maximum emission peak at 337 nm for all conditions (Terrado et al. 2020; Srisong et al. 2018). However, when the cleaned recombinant Sol g 2.1 was combined with different doses of NPN, a significant emission peak at 400 nm was seen. The fluorescence spectra at maximum signal intensities at 400 nm were obtained from titration of various concentrations of NPN ranging from 0 to 12 µM. As increased NPN concentration, the isotherm reached saturation, and the data were then fitted to a specific binding with Hill slope model. The K_d and h slope of rSol g 2.1 protein and NPN were 1.90 ± 0.08 µM and 1.64 ± 0.12, respectively (Fig. 3A). In our finding, affinity value of Sol g 2.1 protein with NPN is within the range seen with other insect OBPs (Campanacci et al. 2001). The reduction in fluorescence intensity at 400 nm was evaluated to assess the binding affinities of Sol g 2.1 protein with the competitive ligands. The results of decane, undecane, dodecane, and tridecane as the NPN displacing ligands were shown as percentages of NPN fluorescence reduction (Fig. 3B). The K_d values of decane, undecane, dodecane, and tridecane of rSol g 2.1 protein binding were 0.32, 0.33, 0.39, and 0.38 µM, respectively (Fig. 3C). According to the findings, decane had the highest affinity for interacting with the rSol g 2.1 protein, followed by undecane, dodecane, and tridecane. This is consistent with the gel filtering results, which showed that the hydrocarbons eluted were 45% decane, 36% undecane, 15% dodecane, and 4% tridecane. Interestingly, the K_d value of the mixture of hydrocarbons binding to rSol g 2.1 protein was reduced to 0.24 µM. These findings imply that the protein has a stronger affinity for the combination than the individual ligands, indicating a positive blend effect. This result is relative to the equilibrium constant fitting with the Hill slope, which has an h value higher than 1.0 meaning that there is more than one binding site with positive cooperativity between the protein and ligands (El-Sharif et al. 2017).

sample of Sol g 2.1 protein. The points represent the average fluorescence intensity at the maximal emission wavelength (400 nm) ± SEM, triplicates. The curve was fitted using nonlinear least squares fitting model for a single binding. (B): Competitive binding curves of selected ligands. (C): Columns showed K_d values of competitor ligands with Sol g 2.1 protein.

Sol g 2.1 Protein Homology Modeling and Molecular Docking. To predict the binding sites of the endogenous ligands (decane, undecane, dodecane, and tridecane) in Sol g 2.1 protein, we used molecular docking (Fig. 4). At the internal binding site 1 (PLB = 2.38), all alkanes ligands were surrounded by mostly non-polar amino acids, including Trp36, Met40, Val61, Ile65, Ile79, Ile104, Val109, and Val110 of Sol g 2.1. For the longer hydrocarbon chains, there were more non-polar amino acid residues that were in contact with these ligands. Val45 interacted with undecane, dodecane, and tridecane as well as Ile66, contacted with decane and tridecane. Moreover, Leu105 also surrounded the dodecane ligand. However, Tyr46, Asn58, Cys62, Cys75, Thr101, and Thr113 amino acid residues of Sol g 2.1 protein pocket also interacted with all ligands (Fig. S5). The average S scores of decane, undecane, dodecane, and tridecane binding to Sol g 2.1 were − 7.58 ± 0.02, -7.51 ± 0.01, -7.96 ± 0.02, and − 8.30 ± 0.03, respectively (triplicates, mean ± SEM). At the external binding site 2 (PLB = 0.61), decane and undecane were in contact with mostly polar amino acids, including His37, Tyr46, Asp47, Ans93, and Arg94 as well as a non-polar amino acid residue which were Ala41 and Pro49 (undecane). This binding site was located around α1-α2 and α4 regions. Next, dodecane and tridecane interacted with both polar and non-polar amino acids which were lined between the loop among α2-α4 regions. Tyr46, Asp47, Asn48, Thr87, Asn93, Arg94, and Lys96 were polar residues at this binding site. There also were some non-polar residues consisting of Pro49, Ile54, Ala97, and Ile100 (dodecane). At this binding site on the Sol g 2.1 protein model, the average S scores of decane, undecane, dodecane, and tridecane ligands with the protein complexes were − 4.31 ± 0.01, -4.30 ± 0.01, -5.66 ± 0.01, and − 5.72 ± 0.01, respectively (triplicates, mean ± SEM). At binding site 3 (PLB = 0.03), all ligands were in contact with polar amino acids that included Arg85, Glu86, Asn99, Gly102, Cys103, and Arg106, additionally Arg89 for undecane and tridecane. There also was Val33 for non-polar residue interaction of all ligands. Moreover, Ile98 and Leu105 bound with decane and two ligands, including dodecane and tridecane, respectively. This binding site was located near the C-terminus and helices α3-α4. The S scores of all ligands were − 4.68 ± 0.02, -4.41 ± 0.01, -5.33 ± 0.01, and − 5.12 ± 0.01, respectively (Fig. 4E). Prediction of Sol g 2.1 and ligand complexes, we found that the lowest S score of the protein and each ligand was at the internal binding site (site 1), suggestion that these ligands are more stabilized in the inner hydrophobic pocket of the protein by hydrophobic interaction than at other sites.

Trail-Following Bioassay. The average distances (cm ± SEM) of the trail-following by the ant workers to M, P, C, M + S, P + S, P + M + S were 60.0 ± 16.9, 221.1 ± 51.5, 197.5 ± 28.8, 205.6 ± 66.4, 191.1 ± 39.4, and 303.4 ± 99.6 respectively. The fully reconstituted venom (P + M + S) is the most attractive for the worker S. geminata ants following only piperidine alkaloids (P), the reconstituted treatment (M + S), crude venom (C), and piperidine in rSol g 2.1 solution (P + S). Moreover, the ants followed the full reconstituted treatment for longer distances than the negative control groups, which did not elicit any response from the ants (Fig. 6).

The crude venom from the fire ant S. geminata consists of various components such as piperidine alkaloids, pheromones, fatty acids, small hydrophobic compounds, and proteins, among them the Sol g 2 protein which is the major allergen protein in Solenopsis spp. venom. This protein has an inner hydrophobic pocket, which can bind with hydrophobic compounds. Moreover, there are many previous studies reporting that the three-dimensional structure and physiochemical properties of Solenopsis venom allergen 2 proteins like Sol i 2 and Sol g 2.1 are similar to PBPs (Das et al. 2018; Borer et al. 2012; Sukprasert et al. 2012). The protein may act as a pheromone transporter protein from the site of pheromone biosynthesis to the sting apparatus and beyond, after sting (Borer et al. 2012; Das et al. 2018). However, there is no report about specific endogenous ligands of this protein in S. geminata venom. Here is the first report to investigate the specific binding activity and ligands of the protein.

The alkaloid peaks are highly dominating in the hexane extracts of fire ants. Because the chemical structures and GC profiles of piperidine and piperideine alkaloids in fire ant venom are well-defined, the chemical identities of major peaks can be determined by comparing peak characteristics with previously published profiles of alkaloids of the two parental species (Brand et al. 1972). From piperidine alkaloid profiles, we found that solenopsin A and isosolenopsin A were the major components in the venom from S. geminata, consistent with results from previous studies (Chen et al. 2010; Chen et al. 2012; Hussein et al. 2016). The binding assays showed that decane was the major endogenous ligand followed by undecane, dodecane, and tridecane, respectively, which was an unexpected finding. We hypothesized that these medium-chain hydrocarbons could be components of the known trail pheromone, which consists of piperidine alkaloids (Igwe and Offiong 2015; Choi and Vander Meer 2012). Surprisingly, the piperidine alkaloids did not bind to this protein. In a previous report, the structural model of Sol i 2 showed that the C-terminal tail of the protein prevents access to the inner cavity compartments, resulting in large molecules like Solenopsin A not being able to access the protein interior.

In in vitro binding assays with NPN, we found that the ligand with highest affinity binding with rSol g 2.1 was decane, followed by undecane, dodecane, and tridecane. Interestingly, after applying the mixture of ligands, we found that the equilibrium dissociation constant, K_d of the mixture to Sol g 2.1 protein was prominently decreased when compared with individual ligands alone. This is due to a positive blend effect of the protein, wherein the mixture of ligands binds more strongly than the individual ligands. This finding is related to the equilibrium constant fitting with the Hill slope, which has an h value greater than 1.0, indicating that the protein and ligands have positive cooperativity (El-Sharif et al. 2017).

Molecular docking of the endogenous alkanes to one internal and two external binding sites of Sol g 2.1 showed that the longer hydrocarbon chains, including dodecane and tridecane had the highest affinity binding with the protein at both the internal and external binding sites. Nevertheless, from the competitive binding assay, decane had the strongest binding affinity with rSol g 2.1 protein followed by undecane, dodecane, and tridecane. Even though, there were no significantly different in K_d values of all ligands (Fig. 3C). This is because the shorter hydrocarbon chains may easily move into the inner hydrophobic cavity of the protein, which may be blocked by the C-terminal region of the protein, preventing access to longer (Park et al. 2015). Related to the peak area ratio after approaching the binding assay, the result revealed that decane had the highest amount followed by undecane, dodecane, and tridecane, respectively, consistent with affinities seen in vitro. Interestingly, the computational model revealed that there are three binding sites on the Sol g 2.1. All of the ligands were most stabilized in the hydrophobic inner pocket of the protein by hydrophobic interactions. Therefore, the possible way to explain the positive blend effect of Sol g 2.1 with the mixture of endogenous hydrocarbons is that there is positive allostery between the external binding sites and the internal one. In the mixture, dodecane and tridecane may strongly bind at the external binding sites, whereas decane and undecane bind at the internal site (Honson et al. 2003; Plettner et al. 2000). Sol g 2.1 may bind to other cryptic ligands not detected here, because it has some polar residues, including Ser and Tyr (Fig. S5), that could interact with various functional groups. E.g., the residues equivalent to Ser58 and Ser46 of Sol i 2, Ser52 and Thr57 of LUSH (odorant-binding protein in Drosophila melanogaster), are in contact with the polar part of vaccenyl acetate via hydrogen-bond donation from the amino acid OH to the pheromone (Laughlin et al., 2008; Borer et al. 2012).

Normally, the fire ant venom comprises piperidine alkaloids and pheromones like trail pheromones which are insoluble in water. These compounds are the most abundant components in the venom (≥ 90%). Although some of the alkaloids bind to Sol g 2 proteins, which act as hydrophobic moiety protection, others possibly can not bind (Borer et al. 2012; Das et al. 2018). Surprisingly, we found that the piperidine alkaloids we detected in the complete venom extract did not bind to Sol g 2.1 strongly enough to emerge bound to the protein from a gel filtration column. We tested if the complete venom extract functions as a trail marking pheromone and found that it does. Interestingly, the fully reconstituted venom (P + M + S), which was calculated between the protein and ligands assuming a 1:1 ratio, had a higher ant-following response than crude venom. We believe that the mixture of hydrocarbon might bind with the protein leading to a reduced evaporation rate of the hydrocarbon. Trail following behavior has been found to be elicited by the alkaloids from crude venom (Vander Meer et al. 1981; Tillman et al. 1999), consistent with our results with the piperidine extract which resulted in trail following responses. Importantly and unexpectedly, the reconstituted group of the mixture of hydrocarbons and Sol g 2.1 protein can also elicit the trail-following response. The composition between Sol g 2.1 and the mixture can attract the ants for following the trails for a longer time and distance than the mixture in hexane only. Therefore, the Sol g 2.1 protein may act as a sticker, delaying the evaporation of the hydrocarbons.

The dominating piperidine alkaloids in S. geminata crude venom were solenopsin A and isosolenopsin A, respectively. The endogenous ligands of Sol g 2 protein were decane, undecane, dodecane, and tridecane corresponded to the percent of peak area ratio, respectively. The key potential chemicals in the venom, piperidine alkaloids, did not bind to this protein. This is because the C-terminal tail of Sol g 2 may inhibit entrance to the inner cavity compartments or obstruct large molecules like Solenopsin A accessing the protein inner cavity. In this state, Sol g 2 protein may protect the hydrophobic moiety of hydrophobic compounds and plunge them into the hydrophobic pocket of the protein. In vitro binding fluorescent assay, the results revealed that decane had the highest affinity for rSol g 2.1, followed by undecane, dodecane, and tridecane. According to molecular docking, the longer hydrocarbon chains, including dodecane and tridecane, have the highest affinity interaction with the protein at both internal and exterior binding sites. Larger ligands, on the other hand, may actually be blocked or difficult to move into the internal binding site, resulting in lower affinity binding compared to shorter ligands. It is supported by the binding assay, which has decane as a major compound. The trail-following behavior to completely reconstituted venom is greater than that of crude venom. The combination of hydrocarbons and Sol g 2.1 protein can also induce trail-following behaviors for extended periods of time and over greater distances.

Acknowledgments The instruments in this research were provided by the Faculty of Pharmaceuticals Science and Research instrument center Khon Kaen University, Khon Kaen, Thailand. The authors also thank the Department of Chemistry, Faculty of Science, Simon Fraser University, Burnaby, British Columbia, Canada for the facilities supporting.

Funding This research was supported by “The Royal Golden Jubilee Ph.D. Program, Thailand (PHD0076/2559)” and “National Research Council of Thailand (NRCT)”. The research also was partially funded by “The Fundamental Fund of Khon Kaen University (KKU)” with funding support from the National Science, Research and Innovation Fund (NSRF), Thailand. Funding in Canada (to E. Plettner) came from the Natural Sciences and Engineering Council of Canada, discovery grant.

Conflicts of Interest We confidently declare that there is no conflict of interest in this research.

Author Contributions All authors contribute conceptualization and design all experiments in this manuscript. S. Nonkhwao and E. Plettner performed data analysis as well as statistical analysis. Software for molecular docking was supported by E. Plettner. All experiments were investigated and validated by S. Nonkhwao. Original draft has been written and reviewed by S. Nonkhwao. This manuscript was revised and corrected by E. Plettner and S. Daduang. This study was visualized and administrated by S. Daduang and E. Plettner. S. Daduang is responsible for funding grants. All authors have read and agreed to the published version of the manuscript.

Astiti MA, Jittmittraphap A, Leaungwutiwong P, Chutiwitoonchai N, Pripdeevech P, Mahidol C, Ruchirawat S, Kittakoop P (2021) LC-QTOF-MS/MS Based Molecular Networking Approach for the Isolation of α-Glucosidase Inhibitors and Virucidal Agents from Coccinia grandis (L.) Voigt. Foods 10(12):3041. https://doi.org/10.3390/foods10123041
Attique SA, Hassan M, Usman M, Atif RM, Mahboob S, Al-Ghanim KA, Bilal M, Nawaz MZ (2019) A Molecular Docking Approach to Evaluate the Pharmacological Properties of Natural and Synthetic Treatment Candidates for Use against Hypertension. Int J Environ Res Public Health 16(6):923. https://doi: 10.3390/ijerph16060923
Billen JPJ (1987) New structural aspects of the Dufour’s and venom glands in social insects. Naturwissenschaften 74:340-341. https://doi.org/10.1007/BF00367931
Blum MS (1992) Ant venoms: chemical and pharmacological properties. J Toxicol Toxin Rev
11:115-164. https://doi.org/10.3109/15569549209033107
Borer AS, Wassmann P, Schmidt M, Hoffman DR, Zhou JJ, Wright C, Schirmer T, Markovic-Housley Z (2012) Crystal structure of Sol i 2: A major allergen from fire ant venom. J Mol Biol 415:635-648. https://doi:10.1016/j.jmb.2011.10.009
Brand JM, Blum MS, Fales HM, MacConnell JG (1972) Fire ant venoms: comparative analyses of alkaloidal components. Toxicon 10(3):259-71. https://doi:10.1016/0041-0101(72)90011-6
Campanacci V, Krieger J, Bette S, Sturgis JN, Lartigue A, Cambillau C, Breer H, Tegoni M (2001) Revisiting the specificity of Mamestra brassicae and Antheraea polyphemus pheromone- binding proteins with a fluorescence binding assay. J Biol Chem 276(23):20078-84. https://doi: 10.1074/jbc.M100713200
Campos Rda S, Mendonça Ade L, Cabral CR Jr, Vaníčková L, Do Nascimento RR (2016) Chemical and behavioural studies of the trail-following pheromone in the leaf-cutting ant Atta opaciceps, Borgmeier (Hymenoptera: Formicidae). J Insect Physiol 86:25-31. https://doi: 10.1016/j.jinsphys.2015.12.008
Carvalho DB, Fox EGP, Santos DGD, Sousa JS, Freire DMG, Nogueira FCS, Domont GB, Castilho LVA, Machado EA (2019) Fire Ant Venom Alkaloids Inhibit Biofilm Formation. Toxins (Basel) 11(7):420. https://doi: 10.3390/toxins11070420
Chalissery JM, Renyard A, Gries R, Hoefele D, Alamsetti SK, Gries G (2019) Ants Sense, and Follow, Trail Pheromones of Ant Community Members. Insects 10(11):383. https://doi: 10.3390/insects10110383
Chen J, Zhao Y, Li X, Zhao J (2019) Pyridine Alkaloids in the Venom of Imported Fire Ants. J Agric Food Chem 67:11388-11395. https://doi:10.1021/acs.jafc.9b03631
Chen L, Hu QB, Fadamiro HY. (2010) Reduction of venom alkaloids in Solenopsis richteri×Solenopsis invicta hybrid: an attempt to identify new alkaloidal components. J Agric Food Chem 58(22):11534-42. https://doi: 10.1021/jf103402f
Chen L, Lu YY, Hu QB, Fadamiro HY (2012) Similarity in venom alkaloid chemistry of alate queens of imported fire ants: implication for hybridization between Solenopsis richteri and S. invicta in the Southern United States. Chem Biodivers 9(4):702-13. https://doi: 10.1002/cbdv.201100109
Choi MY, Vander Meer RK (2012) Ant trail pheromone biosynthesis is triggered by a neuropeptide hormone. PLoS One 7(11):e50400. https://doi: 10.1371/journal.pone.0050400
Das T, Alabi I, Colley M, Yan F, Griffith W, Bach S, Weintraub ST, Renthal R (2018) Major venom proteins of the fire ant Solenopsis invicta: insights into possible pheromone-binding function from mass spectrometric analysis. Insect Mol Biol 27:505-511. https:// doi:10.1111/imb.12388
David Morgan E (2009) Trail pheromones of ants. Physiol Entomol 34(1):1-17. https://doi: 10.1111/j.1365-3032.2008.00658.x
dos Santos Pinto JR, Fox EG, Saidemberg DM, Santos LD, da Silva Menegasso AR, Costa-Manso E, Machado EA, Bueno OC, Palma MS (2012) Proteomic view of the venom from the fire ant Solenopsis invicta Buren. J Proteome Res 11(9):4643-53. https://doi: 10.1021/pr300451g
Edelsbrunner H, Mucke EP (1994) Three-Dimensional Alpha Shapes. ACM Transactions on Graphics 13(1):43-72. https://doi.org/10.1145/174462.156635
El-Sharif HF, Stevenson D, Reddy SM (2017) MIP-based protein profiling: A method for interspecies discrimination. Sens Actuators B: Chem 241:33-39. https://doi.org/10.1016/j.snb.2016.10.050
Fan J, Francis F, Liu Y, Chen JL, Cheng DF (2011) An overview of odorant-binding protein functions in insect peripheral olfactory reception. Genet Mol Res 10(4):3056-3069. http://dx.doi.org/10.4238/2011.December.8.2
Hoffman DR (1995) Fire ant venom allergy. Allergy 50:535-544. https://doi.org/10.1111/j.1398- 9995.1995.tb01196.x
Hoffman DR, Dove DE, Moffitt JE, Stafford CT (1988) Allergens in Hymenoptera venom. XXI. Cross-reactivity and multiple reactivity between fire ant venom and bee and wasp venoms. J Allergy Clin Immunol 82:828-34. https://doi.org/10.1016/0091-6749(88)90085-1
Honson NS, Gong Y, Plettner E (2005) Structure and Function of Insect Odorant and Pheromone- Binding Proteins (OBPs and PBPs) and Chemosensory-Specific Proteins (CSPs). Recent Adv Phytochem 39:227-268. https://doi.org/10.1016/S0079-9920(05)80010-3
Honson NS, Johnson MA, Oliver JE, Prestwich GD, Plettner E (2003) Structure-activity studies with pheromone-binding proteins of the gypsy moth, Lymantria dispar. Chem Senses 28(6):479- 489. https://doi.org/10.1093/chemse/28.6.479
Hussein H, Hameed I, Ubaid J (2016) Analysis of the Secondary Metabolite Products of Ammi majus and Evaluation Anti-Insect Activity. Int J Pharmacogn Phytochem 8(8):1403-1411.
Igwe OU, Offiong SP (2015) Chemistry of Semiochemicals Used as Trail Pheromones in Tropical Fire Ant (Solenopsis geminata). Int j biol chem Sci 7:35-40.
Labute P (2009) Protonate3D: assignment of ionization states and hydrogen coordinates to macromolecular structures. Proteins 75(1):187-205. https://doi.org/10.1002/prot.22234
Laughlin JD, Ha TS, Jones DNM, Smith DP (2008) Activation of pheromone-sensitive neurons is mediated by conformational activation of pheromone-binding protein. Cell 133(7):1255-1265. https://doi.org/10.1016/j.cell.2008.04.046
Leal WS (2013) Odorant reception in insects: roles of receptors, binding proteins, and degrading enzymes. Annu Rev Entomol 58:373-391. https://doi.org/10.1146/annurev-ento-120811- 153635
Park J, Pham HV, Mogensen K, Solling TI, Bennetzen MV, Houk KN (2015) Hydrocarbon binding by proteins: structures of protein binding sites for ≥C10 linear alkanes or long-chain alkyl and alkenyl groups. J Org Chem 80(2):997-1005. https://doi.org/10.1021/jo502488e
Plettner E, Lazar J, Prestwich EG, Prestwich GD (2000) Discrimination of pheromone enantiomers by two pheromone binding proteins from the gypsy moth Lymantria dispar. Biochemistry 39(30):8953-62. https://doi.org/10.1021/bi000461x
Shi Q, Hu L, Wang W, Vander Meer RK, Porter SD, Chen L (2015) Workers and alate queens of Solenopsis geminata share qualitatively similar but quantitatively different venom alkaloid chemistry. Front Ecol Evol 76. https://doi.org/10.3389/fevo.2015.0007
Srisong H, Sukprasert S, Klaynongsruang S, Daduang J, Daduang S (2018) Identification, expression and characterization of the recombinant Sol g 4.1 protein from the venom of the tropical fire ant Solenopsis geminata. J Venom Anim Toxins Incl Trop Dis 24:23. https://doi.org/10.1186/s40409-018-0159-6
Sukprasert S, Uawonggul N, Jamjanya T, Thammasirirak S, Daduang J, Daduang S (2012) Characterization of the allergen Sol gem 2 from the fire ant venom, Solenopsis geminata. J Venom Anim Toxins Incl Trop Dis 18:325-334. https://doi.org/10.1590/S1678- 91992012000300010
Terrado M, Okon M, McIntosh LP, Plettner E (2020) Ligand-and pH-Induced structural transition of Gypsy Moth Lymantria dispar pheromone-binding protein 1 (LdisPBP1). Biochemistry 59(37):3411-3426. https://doi.org/10.1021/acs.biochem.0c00592
Terrado M, Pinnelli GR, Plettner E (2021) Mechanism and dynamics of insect odorant-binding proteins. In: Blomquist GJ and Vogt RG (ed) Chapter 18 Insect Pheromone Biochemistry and Molecular Biology, 2nd edn. Academic Press, pp 591-618
Tillman JA, Seybold SJ, Jurenka RA, Blomquist GJ (1999) Insect pheromones-an overview of biosynthesis and endocrine regulation. Insect Biochem Mol Biol 29(6):481-514. https://doi.org/ 10.1016/s0965-1748(99)00016-8
Touchard A, Aili SR, Fox EGP, Orivel J, Nicholson GM, Dejean A (2016) The Biochemical Toxin Arsenal from Ant Venoms. Toxins 8. https://doi.org/10.3390/toxins8010030
Tschinkel WR (2006) The Fire Ants; Harvard University Press: Cambridge, MA, USA
Vander Meer RK, Williams FD, Lofgren CS (1981) Hydrocarbon components of the trail pheromone of the red imported fire ant, Solenopsis Invicta. Tetrahedron Lett 22(18):1651-1654. https://doi.org/10.1016/S0040-4039(01)90401-0
Vieira FG, Rozas J (2011) Comparative genomics of the odorant-binding and chemosensory protein gene families across the arthropoda: origin and evolutionary history of the chemosensory system. Genome Biol Evol 3:476-490. https://doi.org/10.1093/gbe/evr033
Yu Y, Plettner E (2013) Enantiomer and conformer recognition of (+) and (-)-disparlure and their analogs by the pheromone binding proteins of the gypsy moth, Lymantria dispar. Bioorg Med Chem 21(7):1811-22. https://doi.org/10.1016/j.bmc.2013.01.043

Table 1 is available in the Supplementary Files section.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Structure and Protein-ligand Binding Studies of Pheromone Binding Protein-like Sol g 2.1 Protein from Solenopsis geminata Fire Ant Venom

Status:

Version 1

Abstract

Figures

INTRODUCTION

METHODS AND MATERIALS

RESULTS

DISCUSSION

CONCLUSION

Declarations

References

Table

Additional Declarations

Supplementary Files

Status:

Version 1