The PA1b (Pea Albumin 1, subunit b) peptide is an entomotoxin extract from Legume seeds with lethal activity on several insect pests, such as mosquitoes, some aphids and cereal weevils. This 37 amino-acid cysteine-rich peptide has been, until now, obtained by biochemical purification or chemical synthesis. In this paper, we present our results for the transient production of the peptide in Nicotiana benthamiana by agro-infiltration, with a yield of about 35 µg/g of fresh leaves and maximum production 8 days after infiltration. PA1b is part of the PA1 gene which, after post-translational modifications, encodes two peptides (PA1b and PA1a). We show that transforming tobacco with the PA1b cDNA alone does not result in production of the toxin and, in fact, the entire cDNA is necessary, raising the question of the role of PA1a. We constructed a PA1-cassette, allowing for the quick “cut/paste” of different PA1b mutants within a conserved PA1 cDNA. This cassette enabled us to produce the six isoforms of PA1b which exist in pea seeds. Biological tests revealed that all the isoforms display similar activity, with the exception of one which is inactive. The lack of activity in this isoform led us to conclude that the amphiphilic nature of the peptide is necessary for activity. The possible applications of this expression system for other cysteine-rich biomolecules are discussed.
Citation: Eyraud V, Karaki L, Rahioui I, Sivignon C, Da Silva P, Rahbé Y, et al. (2013) Expression and Biological Activity of the Cystine Knot Bioinsecticide PA1b (Pea Albumin 1 Subunit b). PLoS ONE 8(12): e81619. https://doi.org/10.1371/journal.pone.0081619
Editor: Vladimir N. Uversky, University of South Florida College of Medicine, United States of America
Received: September 11, 2013; Accepted: October 23, 2013; Published: December 11, 2013
Copyright: © 2013 Eyraud et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Funding was provided by Institut National de le Recherche Agronomique (INRA, France) and Institut National des Sciences Appliquées (INSA de Lyon, France). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Chemical pesticides are increasingly used around the world but they are also increasingly stigmatized because of their persistence and their toxicity to non-target organisms. Crop protection against two very important pests, namely cereal weevils and aphids, is currently carried out almost exclusively by chemical treatments. A few alternative methods do exist to combat these insects, but they are either much less effective or prohibitively expensive, compared with chemical control.
Finding new plant-derived molecules, which have less impact on the environment, is a major challenge in order to develop biopesticides for sustainable and healthy agriculture. The PA1b (Pea Albumin 1, subunit b) peptide is an entomotoxin extract from pea seeds and, more generally, from Legume seeds . PA1b is one of the few orally active peptide toxins currently known. It displays outstanding insecticidal activity against certain insects, such as grain weevils (genus Sitophilus) and the mosquitoes Culex pipiens and Aedes aegyptii (for review see .
PA1b consists of 37 amino acids, including six cysteines forming three intramolecular disulfide bridges (Figure 1B). The structure of the pea albumin 1 gene (PA1 gene) is depicted in Figure 1A. The PA1 gene is transcribed as a single mRNA encoding the 13-kDa preproprotein PA1. After cleavage of its signal peptide, the proprotein precursor is endoproteolytically cleaved to yield two peptides which represent the mature forms of PA1a (6 kDa, 53 aa) and PA1b (3.8 kDa, 37 aa) .
A. The PA1 gene is transcribed as a single mRNA encoding the preproprotein PA1 (molecular weight 13.9 kDa; 130 amino-acids). After cleavage of its signal sequence (SS), the proprotein precursor (11.2 kDa, 104 aa) is endoproteolytically cleaved to yield two peptides which, after removal of two propeptides (PP), represent the mature forms of PA1a (6 kDa, 53 aa) and PA1b (3.8 kDa, 37 aa) . B. PA1b primary structure. The cysteine residue pairing is shown below .
The three-dimensional structure of the peptide reveals that it belongs to the cystine knot inhibitor (ICK or knottin) family . The ICK family contains numerous peptides and, as well as their related cyclized homologues (named cyclotides), they exhibit numerous biological activities presenting, potentially, a structural core for the design of new drugs or pesticides , , . Using chemical synthesis and in vitro folding of PA1b , a collection of alanine mutants has been constructed. The three-dimensional structure of PA1b is particularly tolerant of any modifications. Receptor binding and insecticidal activities are both dependent on common well-defined clusters of residues located on one single face of the toxin, with Phe10, Ile23, and Leu27 as the key residues of the binding interaction corresponding to a hydrophobic area for bioactivity. An additional amino-acid, Arg21, represents a second strong area of interaction with the receptor .
The insecticidal activity of PA1b has been patented . Indeed, PA1b has many attributes for use on an industrial scale: it is extracted from a commonly grown plant which is regularly consumed by humans and other mammals, it is suitable for use in transgenic plants, and it can withstand many stages of extraction and purification without losing its activity. Nevertheless, the current major disadvantage of PA1b is that it is only active at relatively high doses (for example, the dose used for routine laboratory tests on weevils is 400 µg/g of food). Hence, an important objective is to optimize the PA1b sequence in order to increase its activity towards insects. This research is based on the results of a structure/function study  and also on the natural diversity of A1b, found in a large number of Legumes . The cDNA sequence of A1b isoforms from Fabaces, obtained from a gene or an EST bank, or by cloning in other Legumes, needs to be toxicologically tested on weevils in order to correlate the sequences with the activity.
To test these isoforms, it is necessary to produce them within a heterologous system. PA1b production in bacteria (E. coli) or yeast (Pichia pastoris) has never been successful. To date, only production by chemical synthesis and in vitro folding has enabled us to synthesize a biologically active PA1b. However, this system is not suitable for the expression of a large number of isoforms. Thus, we chose to investigate the expression of the toxin in a plant system, the tobacco plant Nicotiana benthamiana, with the aim of synthesizing and testing the activity of the six PA1b isoforms produced naturally by the pea, which cannot be purified by biochemical methods. Until now, the individual activities of each isoform have not been investigated.
Expression of the Peptide(s) PA1b (and PA1a) in N. benthamiana
The PA1 cDNA from pea, coding for the signal peptide and the PA1 preproprotein, was inserted into the plant expression vector pMDC32. A. tumefasciens containing pMDC32 and A. tumefasciens containing the p19 plasmid (a suppressor of silencing) were co-infiltrated into N. benthamania leaves, and the inoculated leaves were harvested five days later. The HPLC chromatogram, presented in Figure 2, shows that the peptide PA1b was found in the pMDC32-(PA1cDNA) inoculated leaves but was not detectable in the control leaves inoculated with only the p19 plasmid. Using the chromatogram, we measured 21.4 µg of PA1b per g of fresh leaves. The mass spectrometry analysis corroborated the HPLC analysis: the only detectable peptide was PA1b in the pMDC32-(PA1cDNA) inoculated leaves extract, and the results confirmed that three disulfide bridges exist in the molecule. Nevertheless, three forms of the same peptide exist: the peptide with an additional Val38 (m/z = 3855.0), the pea native peptide (m/z = 3857.2) and, predominantly, the peptide lacking Gly37 (m/z = 3700.3). In addition, all forms have the oxidized Met12 peptide (+16 in mass). The PA1a peptide was not detectable on the HPLC chromatogram and the mass spectrometry analysis of the harvested leaves, at the 10–12 min elution time, displayed no m/z within the range of the PA1a mass.
The figure shows the HPLC chromatograms of extracts from tobacco leaves infiltrated either with the p19 plasmid only (control, blue curve) or with p19 and the pMDC32 plasmid carrying the entire PA1 cDNA (red curve). The figure shows the HPLC chromatograms of extracts from tobacco leaves infiltrated either with the p19 plasmid only (control, blue curve) or with p19 and the pMDC32 plasmid carrying the entire PA1 cDNA (red curve). 5 days after infiltration, the leaves were harvested and, after partial purification, the mixture was injected into a C18 HPLC column and the elution was monitored at 210 nm. The green line represents the acetonitrile gradient.
To ensure that the expressed PA1b was in a biologically active form, we next performed toxicity assays on weevils (S. oryzae) and on Sf9 insect cells. The TL50 of the tobacco-expressed PA1b was similar to that of the same amount of pure PA1b peptide from pea seeds (7.8+/−1.13 and 7.23+/−1.1 days, respectively) on the WAA42 sensitive weevil strain. There was no mortality on the PA1b resistant strain ISOR3, confirming that the toxicity on the WAA42 strain was PA1b-like. Similarly, the LD50 values on Sf9 cells were 84+/−21 and 68+/−15 nM for the tobacco-expressed PA1b and for the PA1b from pea seeds, respectively.
Kinetics of Expression of the Peptide PA1b in N. benthamiana
A kinetic analysis, to investigate the expression of PA1b in N. benthamiana, was carried out. The leaves were inoculated on day 0 and were harvested, over a range of time points, from day 1 to day 20 after inoculation. The PA1b and PA1a peptide were analyzed and quantified by HPLC. The results confirmed that the PA1a peptide was never detectable, and neither was the PA1b peptide in the control (p19) experiment. As regards the assay experiment (Figure 3), the PA1b peptide was visible two days after inoculation (2.5+/−0.2 µg/g of fresh leaves) and it increased to reach a maximum level 8 days after inoculation (38.7+/−4.8 µg/g of fresh leaves). Subsequently, the amount of PA1b found in the tobacco leaves decreased rapidly.
The data are the mean of three independent experiments. For each experiment, a set of 120 leaves were inoculated with the pMDC32 plasmid containing PA1 cDNA. After extraction of the peptide from harvested leaves (5 leaves per day), the toxin was quantified by HPLC.
Mass spectrometry analysis of the PA1b produced in tobacco showed that, whilst 3 to 6 days after infiltration all three forms of the PA1b were present, 8 days after inoculation only one form was detectable (the peptide without Gly37 and with oxidized Met12, m/z = 3701.1).
In the next experiment, we harvested all the parts of the tobacco plant (i.e. transformed leaves, non-transformed leaves, stems and roots) on days 5, 8 and 12 after inoculation. Following extraction, the PA1b peptide was detected only in the transformed leaves and not in the other organs, using both HPLC and mass spectrometry.
Role of the Different Parts of PA1 cDNA on the Expression of PA1b
The relative importance of each part of the PA1 cDNA was analyzed next. Different constructions of PA1 cDNA were inserted into the pMDC32 plasmid (see Figure 4). Each modified plasmid was then introduced into A. tumefasciens, and the tobacco leaves were infiltrated. The results, presented in Figure 4, show that the amount of PA1b peptide extracted from tobacco leaves 8 days after inoculation was at a maximum level and was similar for the entire PA1 cDNA and for the PA1 cDNA lacking the terminal propeptide (37.4 and 36.6 µg PA1b/g of fresh leaves, respectively). In contrast, the constructions lacking the PA1a part of the gene showed a dramatic decrease in PA1b production (2.4 µg PA1b/g of fresh leaves). Finally, when only the PA1b cDNA was present, the peptide was undetectable by HPLC or by mass spectrometry.
The first (control) construction was the entire PA1 cDNA. The other constructions were the cDNA coding PA1b alone; the signal sequence (SS) – PA1b; the SS – PA1b – propeptide(PP); and the SS – PA1b – PP – PA1a cDNAs. The PA1 cDNA constructions were inserted in the pMDC32 plasmid under the control of the double CaMV35S promoter, and Agrobacterium were transformed with the different plasmids. After agro-infiltration of tobacco leaves, the PA1b expression, following HPLC purification, is indicated.
Construction of a PA1 Gene Cassette to Rapidly Express various PA1b Isoforms
Following our results, which indicated that the PA1a sequence is necessary for the production of the PA1b peptide, we next constructed a PA1 cDNA allowing for the quick and easy replacement of the PA1b sequence by a modified one. First, two restriction sites, not present in the pMDC32 plasmid, were introduced by directed mutagenesis at the borders of the PA1b sequence (SnaBI and Acc651). This modified PA1 sequence was expressed in tobacco: PA1b was expressed as for the native gene as regards the quantity (39.8 µg PA1b/g of fresh leaves) and the mass (m/z = 3700.9). The peptide was purified by HPLC and biological tests on weevils were then performed. The results showed that the toxin expressed from the PA1-cassette displayed a LD50 (8.41+/−1.42 days) similar to that of native PA1b on the WAA42 strain, and no mortality was observed on the ISOR3 resistant strain.
Expression of Pea Peptides using the PA1-cassette
We next used the cassette system for expressing the peptides in tobacco plants and for biological testing of the others pea PA1b isoforms. The results, presented in Table 1, show that all the isoforms (numbered from 1 to 6) can be successfully produced in tobacco, with yields ranging from 17.61 to 53.33 µg/g of fresh leaves, except for isoform 3 (8.7 µg/g of fresh leaves). The mass spectrometry results showed that, in all the isoforms, the terminal Gly37 is lacking and the Met12 is oxidized (with the exception of isoform 2, where both forms were produced (numbered 2.1 and 2.2), and isoform 3, which was lacking the Met). Regarding biological activity, all isoforms have similar CL50 values on Sf9 insect cells (ranging from 53 to 99 nM) and similar TL50 values on the sensitive weevil strain WAA42 (ranging from 8.9+/−0.67 to 9.5+/−0.7 days), except for isoform 4 (269+/−50 nM on Sf9 and 12.9+/−1.1 days on weevils) and isoform 3 which was completely inactive on the Sf9 cells and on weevils. None of the isoforms had an effect on the resistant weevil strain ISOR3, confirming that this is a PA1b type of activity.
We next constructed a modified cDNA which encodes a peptide with two mutations: a Gly instead of the initial Ala1 and an Asn instead of the terminal Gly37. These modifications, in theory, produce a peptide which could be cyclized in the tobacco plants . After purification, by HPLC, in the PA1b retention time range, the mass spectrometry results showed that only the linear peptide exists (m/z = 3670.2 and m/z = 3686.3 for the form with oxidized Met12). Based on measurement of the mass, this peptide is lacking the Asn37 terminal.
Molecular Modeling of the Modified PA1b
The lipophilic potentials at the Connolly surfaces of isoforms 1 (control) and 3 were calculated. These molecular properties can be derived from the 3-D structures and are related to the functions of the proteins. The distribution of the hydrophobic potentials, calculated with MOLCAD at the surface of PA1b isoform 1, is characteristic of an amphipathic structure (Figure 5A). There is clearly a hydrophobic face formed by the residues of the hydrophobic loop (Val25, Leu27, Val28, and Ile29) together with residue Phe10, which is opposite. The hydrophilic face, situated at the other pole of the molecule, arises from the following polar residues: Ser2, Asn4, Thr17, Ser18, Asn34, and Ser36. When the hydrophobic potentials are calculated, using the same hydrophobic scale, the surface of PA1b isoform 3 (Figure 5C) appears to be mainly hydrophobic, with a hydrophobic face replacing the hydrophilic face of PA1b due to the Pro18, Ile19, and Tyr36 mutated residues.
Lipophilic potentials, calculated with the MOLCAD option of SYBYL, at the Connolly surfaces of (A) PA1b isoform 1 and (C) isoform 3, and (B) ribbon representation of the PA1b isoform 1 NMR structure (the disulfide bridges are shown in green). Figures A and C are the same orientation as figure B, using a common hydrophobic scale. Hydrophobic and hydrophilic areas are displayed in brown and blue, respectively. Green surfaces represent an intermediate hydrophobicity.
PA1b is an entomotoxin with considerable potential for applications on an industrial scale, the main advantage being that it is derived from an edible plant and, consequently, it can be used in organic farming. However, the main concern is the relatively high doses required for its activity. To overcome this problem, we have developed a program with the aim of determining the optimum PA1b sequence with maximum biological activity. This program is based on established data regarding the structure/function of the peptide  and on the existence of numerous isoforms in Legumes , . Various peptide sequences are currently produced by chemical synthesis with in vitro folding . However, this method is both expensive and long, with a variable yield depending on the isoforms. In this study, we succeeded, for the first time, in producing PA1b within a heterologous system, and we used this system to synthesize and test all the PA1b isoforms present in pea seeds.
In the extracts of pea seeds, the PA1b isoforms are always found in a unique form consisting of 37 amino acids (as indicated by mass spectrometry analysis). In tobacco plants, we extracted three forms of the peptide made up of 36, 37 or 38 amino acids. The cleavage of the proprotein seems to be different in tobacco, occurring after Val38 instead of after Gly37, as in the pea, indicated by the presence of the 38-mer peptide. After cleavage, the two terminal amino-acids are probably cleaved by a carboxypeptidase (not present in pea seeds), as indicated by a decrease in the 38-mer form which had been totally replaced, by day 8 after infiltration, by the 36-mer peptide. Similarly, replacing the Gly37 by an Asn results in a 36-mer, the Asn being cleaved in the same way as the Gly. Oxidation of Met12 is often found in PA1b, regarding the biological activity, we found that the lack of Gly37 and the oxidation of Met12 did not alter the toxicity of the peptide.
The kinetics of PA1b production in tobacco shows that production is at a maximum level on day 8, and it decreases rapidly thereafter. The optimum production time for other proteins is often earlier (days 4 to 6 after infiltration) , . It has been demonstrated that the 3-D structure of PA1b (the cystine knot fold) confers a very high stability, especially regarding peptidase activities , . This stability probably explains the increase in the production of PA1b right up to day 8. As PA1b has not been found in other plant organs (stem, root or leaves) on day 5, 8 or 13 after infiltration, we can hypothesize that PA1b is either ultimately degraded in the tobacco leaves by (induced?) proteases, or inactivated by coupling to another molecule.
Although, in theory, the two peptides PA1b and PA1a are produced in equimolar quantities, PA1a was never detected. No HPLC peaks were seen with a retention time corresponding to that of PA1a purified from pea seeds and even mass spectrometry, performed on the HPLC samples at the PA1a retention time, failed to detect any PA1a peptides. PA1b is a very stable peptide but this is not the case for PA1a, which has been shown to be less stable than PA1b in pea seeds . Moreover, the 3D-structure of PA1a is much less organized than that of PA1b (L. Jouvensal and P. Da Silva, personal communication). It is thought that, in tobacco plants, PA1a has a very short half-life and is, therefore, not detectable.
We observed, in this study, that the entire cDNA is necessary for the expression of the PA1b peptide and that the presence of the PA1a sequence is also essential. These results raised the question of the role of PA1a, possibly at the level of RNA and involving a more stable PA1 RNA rather than just the PA1b RNA alone. An alternative hypothesis of interest is that the PA1a peptide could act as a chaperone for the correct folding of PA1b. This activity could occur at the level of the proprotein (PA1a participating in the correct folding of the whole protein) or after the post-translational cleavage (PA1a perhaps assists in the folding of the PA1b peptide (and that of the PA1a itself?)). The importance of PA1a is supported by the recent discovery of a cyclotide (a cyclized cystine knot peptide called Cter M), in Clitoria ternatea, which is embedded in the PA1 gene, i.e. the cyclotide sequence replaces the PA1b sequence in the PA1 gene. Remarkably, this Cter M cyclotide displayed insecticidal activity against the cotton budworm Helicoverpa armigera .
The main objective of our study was to establish a quick and reliable expression system to test the many isoforms of the peptide for their toxicity, and to elucidate the effect of each isoform found in pea seeds. Nevertheless, although cloning and sequencing the PA1b part of the cDNA on various Legumes is relatively simple, cloning the entire PA1 cDNA is often more complicated . To simplify the expression of the PA1b isoforms, we constructed a PA1-cassette, with specific restriction sites bordering the PA1b part of the cDNA in order to easily change the PA1b-cDNA sequence whilst conserving the essential propeptides and PA1a sequences. The results demonstrate that this cassette allows for the expression of significant quantities of a biologically active toxin.
It has long been established that the mixture of isoforms in pea seeds is toxic for insects , , but the role of each individual isoform was not known due to the difficulty of separating these isoforms with HPLC or any other technique. Hence, our initial aim, using the Nicotiana system, was to produce all the pea seed isoforms in order to test them individually. The results show that all the isoforms have similar levels of activity, except for isoform 3 which was found to be inactive. This isoform is the one that differs most and three mutations (E11D, M12I and V25A) were previously found to have no, or only a slight, effect on activity . Using molecular modeling, we found that three other mutations (S18P, A19L and S36Y) lead to the loss of the amphiphilic character of PA1b by changing the hydrophilic face to a hydrophobic one. This result completes the previous structure/function work . The biological activity of PA1b depends on the presence of a hydrophilic area in the peptide (this work), and on the presence of the Arg21 residue and the hydrophobic nature of the PA1b loop opposite the hydrophilic loop . Moreover, the replacement of Leu28 by a Phe, in isoform 2, confirms that the overall hydrophobicity of this loop is more important than the nature of the amino acid side chain.
The cyclization of a peptide only occurs rarely, but this leads to a very stable structure  and we suggest that a cyclized PA1b may show stronger activity. Moreover, cyclotides have been proven to be useful scaffolds for stabilizing bioactive peptides . The cyclization of PA1b could help to identify new insecticides. It has been demonstrated that peptide cyclization depends on an enzyme, asparaginyl endopeptidase, which is present in tobacco . In addition, this cyclization depends on the peptide sequence, and the presence of a C terminal Asn and an initial Gly were found to be very important , . For this reason, the A1G and G37N mutations were inserted in the PA1b sequence, but this did not result in the production of cyclic PA1b. The linear peptide produced by the tobacco plant lacked the terminal, and essential, Asn. It is difficult to establish whether cyclization does not occur because of the proteolysis of Asn or because PA1b cannot be cyclized in tobacco. Studies on the cyclization of peptides have been conducted on naturally cyclized peptides, the cyclotides , . PA1b may be more difficult to cyclize because it is longer (37 vs. 29–30 amino acids) and its primary sequence is very different from those of cyclotides, although it has the same cystine knot fold.
However, this work enabled us to easily express a sulfur-rich plant peptide in tobacco, which was previously shown to be not feasible in E. coli or in P. pastoris. Also, the entire PA1 gene is shown to be necessary for effective expression and we constructed a PA1-cassette for the quick and reliable expression of a large number of PA1b peptides. Subsequently, in our study, this cassette system expression allowed us to determine the real effect of each pea isoform, providing important information about their structure/function. Such an expression system will help in determining the best sequence for producing the most efficient toxin.
PA1b belongs to the ICK family, which contains numerous peptides with various biological activities . This cystine knot core is increasingly considered to be a valuable structural platform for creating compounds with numerous biological targets, particularly in the field of therapeutics and of agriculture, and the PA1b-cassette expression could be useful in other research involving ICK peptides. Until now, ICK peptides have been produced either by a heterologous system or by chemical synthesis. The success of production in E. coli depends greatly on the peptide and, for chemical synthesis, the limiting step is often the folding of the cystine knot and its three disulfide bridges . It is possible that our expression system could provide a means of producing cystine knot peptides which are otherwise difficult to produce, in E. coli or by chemical synthesis, as well as a quick and valuable tool for any structure/function studies.
Materials and Methods
N. benthamiana Culture
Nicotiana plants were grown in a greenhouse with a 16 h-day photoperiod, at a controlled temperature of 24°C (day) or 21°C (night). 8 to 10 week-old plants were chosen for Agrobacterium infiltration.
PA1b Extraction, Identification and Quantification
Tobacco leaves were harvested and immediately ground, in a mortar, in liquid nitrogen. The resulting powder was incubated in 10 volumes of 60% ethanol for 2 h, at 4°C. Next, the mixture was centrifuged at 10 000 g for 10 min and the ethanol was evaporated under vacuum. The remaining solution was dried by lyophilisation and the powder was stored at −20°C until required.
The powder was re-suspended in 60% EtOH (10 mg.mL−1) and injected onto a reverse-phase HPLC column (Jupiter C18 5 µm, 300 A, 250×4.6 mm, Phenomenex), in an Agilent 1200 apparatus with elution, as described in . The peptides were harvested according to their retention times (typically between 18 and 20 min for PA1b and between 10 and 12 min for PA1a) and the samples were then lyophilised. PA1b was quantified, using HPLC, by comparison with known quantities of pure PA1b from pea seeds.
Peptides were identified by mass spectrometry. The samples were analyzed on a MALDI-TOF MS (Autoflex, Bruker Daltonics), operated in reflectron positive mode.
Rice weevils (Sitophilus oryzae, Coleoptera) were reared on wheat seeds at 27.5°C and 70% RH. Tests and survival analysis were performed on adults feeding on food pellets (composed of wheat flour and water) incorporating PA1b (extracted from pea seeds for control tests or produced in tobacco) at 400 µg.g−1, as described in detail in . LT50 values were calculated using the SIMFIT software (http://www.simfit.man.ac.uk). The PA1b resistant strain, ISOR3, was reared in an identical way except that the weevils were fed on pea seeds rather than on wheat seeds.
Spodoptera frugiperda Sf9 cells were grown and used for viability assays, according to .
The mixture for PCR (20 µl) consists of 2 µL of reaction buffer 10X (670 mM Tris-HCl, pH8.8, 160 mM (NH4)2SO4, 0.1% Tween-20) (Ozyme), 1 µl of 50 mM MgCl2, 1 µL of 10 mM dNTP, 1 µL of 1 µM forward and reverse primers, 2 µL of DNA template and 0.25 µL of 5 U.µl−1 high fidelity Taq-polymerase (Platinium Taq DNA polymerase High fidelity, Invitrogen), and PCR assays were performed using a BIO-Rad C1000 thermal cycler.
The PCR conditions were 5 min at 95°C, then 30 cycles of [95°C, 30 sec/50°C, 1 min/72°C, 1 min] and, finally, 72°C for 10 min.
For directed mutagenesis, the PCR conditions were 95°C for 5 min, then 28 cycles of [95°C, 30 sec/50°C, 1 min/72°C, 1 min] for the first and second PCR whereas, for the third PCR, we used 95°C for 5 min, then 30 cycles of [95°C, 30 sec/50°C, 1 min/72°C, 1 min] and 72°C, for 10 min.
Construction of the PA1 and PA1 Truncated Variants Expression Plasmid
The sequences of all the primers used in this paper are given in the Table S1 and were synthesized by Eurogentec (France). The cDNA used for PCR corresponds to the pea gene with accession number AJ574795.
The entire PA1 cDNA codes for a peptide with a signal sequence (SS), the PA1b peptide, an initial propeptide (PP), the PA1a peptide, and a second propeptide (PP).
The entire PA1 cDNA (SS-PA1b-PP-PA1a-PP) was amplified, by PCR, with two specific primers (Sac1PA1F and BamH1PA1R) to incorporate the enzyme restriction sites Sac1 and BamH1. In the same way, the SS-PA1b-PP-PA1a, the SS-PA1b-PP, the SS-PA1b, and the PA1b (with inclusion of a start codon) cDNAs were amplified with, respectively, the specific primers Sac1PA1F and BamH1PA1R, BamH1PA1R and Sac1PA1-P3F; BamH1PA1R and Sac1PA1b+P2F; BamH1PA1R and Sac1PA1bF; and BamH1PA1-P1 and Sac1PA1bF. Next, they were sub-cloned into the pCR2.1 vector (TOPO TA Cloning, Invitrogen). With enzyme BamH1/Sac1 restriction cutting, PA1 cDNA and the truncated variants were inserted into the pMDC32 vector, provided by http://www.arabidopsis.org/abrc/catalog/vector_1.html , between the cauliflower mosaic virus (CaMV) double 35S constitutive promoter and the nos transcriptional termination region.
For the pea gene (with the intron), the primers used for PCR amplification were the same as for the entire PA1 cDNA and genomic DNA from P. sativum was used as a template.
Transformation of N. benthamiana
Agrobacterium C58pMP90 cells, carrying the pMDC32 plasmid with PA1 cDNA or PA1 truncated variants, and Agrobacterium carrying the p19 plasmid  were grown in Luria Broth (LB) media, containing 50 µg.ml−1 kanamycin and 50 µg.ml−1 rifampicin, at 28°C. When the A600 reached 0.6 the cells were pelleted (3500 g for 10 min, room temperature) and then twice re-suspended in an equal volume of MES buffer (50 mM MES pH 5.6, 2 mM NaH2PO4, 100 µM acetosyringone, 0.5% glucose). Each suspension of Agrobacterium, carrying pMDC32 with a different PA1 construct, was incubated with the suspension of Agrobacterium carrying p19 (50/50), at room temperature for 1–2 h, and then infiltrated into N. benthamiana leaves, following the methods of , .
Construction of the PA1-cassette
Two new enzyme restriction sites (SnaBI and Acc651) were inserted into the pMDC32– PA1 vector, framing the PA1b sequence in codons 23 and 69, respectively. Mutagenesis was carried out, by PCR, using oligonucleotides containing the required mutations. To generate the necessary site-directed mutagenesis constructs, a three-round asymmetric PCR strategy was used for each enzyme restriction site, as previously described . The first two reactions were carried out using the pMDC32-PA1 vector as the DNA template, and for the SnaBI restriction site mutation, respectively, primers BamH1PA1R and SnaBIPA1R, and Sac1PA1 F and SnaBIPA1F were used. The two resulting DNA products were purified and used as templates for the third PCR, with primers Sac1PA1F and BamH1PA1R. This new PA1-SnaBI was inserted into the pCR2.1 plasmid and used as a template for the insertion of the Acc651 enzyme restriction site. Following the same pattern, the Acc651 mutation was inserted, using a three-step PCR, with the following primer couples: PA1Acc651F and PA1BamH1R; Sac1PA1F and PA1Acc651R; PA1BamH1R and Sac1PA1F. This new sequence was then inserted, by enzymatic cutting (BamH1, Sac1), into the pMDC32 vector.
Insertion of PA1b Pea Isoforms into a PA1-cassette
The mutated sequences of PA1b were obtained from a pMDC32 PA1-cassette by PCR, using specific long primers as described in S1. The newly created sequences were cloned in pCR2.1 and were inserted into the pMDC32-PA1 cassette by SnaBI and Acc651 enzymatic cutting. N. benthamiana leaves were transformed and toxicity tests were carried out as explained above.
DNA Quantification and Sequencing
DNA was quantified using a NanoDrop ND-1000 spectrophotometer and DNA sequences were performed by BioFidal (Vaux-en-Velin, France).
Sequence Alignments and Comparative Modeling
Sequence alignments were performed using CLUSTAL OMEGA . Comparative modeling was performed using the ORCHESTRAR homology modeling program in the SYBYL-X 2.0 software package (TRIPOS Inc., St Louis, MO). The PA1b NMR structure (PDB code: 1P8b) was used as a template to build the 3D structure model of the PA1b isoforms. Lipophilic potentials were calculated and represented using the MOLCAD option of the SYBYL-X 2.0 software package.
We are grateful to Dr. Lucas SIGNOR (CEA, Grenoble, France) for the MALDI-TOF/MS analysis. We thank Dr. Valerie James for the English corrections. We thank Prof. David Baulcombe (University of Cambridge) and the SL/The Gatsby Charitable Foundation for providing the p19 plasmid.
Conceived and designed the experiments: VE PDS YR CR FG. Performed the experiments: VE LK IR CS PDS FG. Analyzed the data: VE PDS CR FG. Contributed reagents/materials/analysis tools: VE IR CS CR FG. Wrote the paper: VE YR PDS CR FG.
- 1. Delobel B, Grenier AM, Gueguen J, Ferrasson E, Mbaiguinam M (1998) Utilisation d’un polypeptide dérivé d’une albumine PA1b de légumineuse comme insecticide. French Patent 98 05877.
- 2. Gressent F, Da Silva P, Eyraud V, Karaki L, Royer C (2011) Pea Albumin 1 subunit b (PA1b), a promising bioinsecticide of plant origin. Toxins (Basel) 3: 1502–1517.
- 3. Higgins TJ, Chandler PM, Randall PJ, Spencer D, Beach LR, et al. (1986) Gene structure, protein structure, and regulation of the synthesis of a sulfur-rich protein in pea seeds. J Biol Chem 261: 11124–11130.
- 4. Jouvensal L, Quillien L, Ferrasson E, Rahbe Y, Gueguen J, et al. (2003) PA1b, an insecticidal protein extracted from pea seeds (Pisum sativum): 1H-2-D NMR study and molecular modeling. Biochemistry 42: 11915–11923.
- 5. Craik DJ (2012) Host-defense activities of cyclotides. Toxins (Basel) 4: 139–156.
- 6. Craik DJ, Swedberg JE, Mylne JS, Cemazar M (2012) Cyclotides as a basis for drug design. Expert Opin Drug Discov 7: 179–194.
- 7. Kolmar H (2009) Biological diversity and therapeutic potential of natural and engineered cystine knot miniproteins. Curr Opin Pharmacol 9: 608–614.
- 8. Da Silva P, Strzepa A, Jouvensal L, Rahioui I, Gressent F, et al. (2009) A folded and functional synthetic PA1b: an interlocked entomotoxic miniprotein. Biopolymers 92: 436–444.
- 9. Da Silva P, Rahioui I, Laugier C, Jouvensal L, Meudal H, et al. (2010) Molecular requirements for the insecticidal activity of the plant peptide pea albumin 1 subunit b (PA1b). J Biol Chem 285: 32689–32694.
- 10. Louis S, Delobel B, Gressent F, Duport G, Diol O, et al. (2007) Broad screening of the legume family for variability in seed insecticidal activities and for the occurrence of the A1b-like knottin peptide entomotoxins. Phytochemistry 68: 521–535.
- 11. Gillon AD, Saska I, Jennings CV, Guarino RF, Craik DJ, et al. (2008) Biosynthesis of circular proteins in plants. Plant J 53: 505–515.
- 12. Louis S, Delobel B, Gressent F, Rahioui I, Quillien L, et al. (2004) Molecular and biological screening for insect-toxic seed albumins from four legume species. Plant Science 167: 705–714.
- 13. Chen Q, He J, Phoolcharoen W, Mason HS (2011) Geminiviral vectors based on bean yellow dwarf virus for production of vaccine antigens and monoclonal antibodies in plants. Hum Vaccin 7: 331–338.
- 14. Wydro M, Kozubek E, Lehmann P (2006) Optimization of transient Agrobacterium-mediated gene expression system in leaves of Nicotiana benthamiana. Acta Biochim Pol 53: 289–298.
- 15. Poth AG, Colgrave ML, Lyons RE, Daly NL, Craik DJ (2011) Discovery of an unusual biosynthetic origin for circular proteins in legumes. Proc Natl Acad Sci U S A 108: 10127–10132.
- 16. Gressent F, Duport G, Rahioui I, Pauchet Y, Bolland P, et al.. (2007) Biological activity and binding site characteristics of the PA1b entomotoxin on insects from different orders. Journal of Insect Science 7: -.
- 17. Conlan BF, Gillon AD, Craik DJ, Anderson MA (2010) Circular proteins and mechanisms of cyclization. Biopolymers 94: 573–583.
- 18. Eliasen R, Daly NL, Wulff BS, Andresen TL, Conde-Frieboes KW, et al. (2012) Design, synthesis, structural and functional characterization of novel melanocortin agonists based on the cyclotide kalata B1. J Biol Chem 287: 40493–40501.
- 19. Saska I, Gillon AD, Hatsugai N, Dietzgen RG, Hara-Nishimura I, et al. (2007) An asparaginyl endopeptidase mediates in vivo protein backbone cyclization. The journal of biological chemstry 282: 29721–29728.
- 20. Conlan BF, Colgrave ML, Gillon AD, Guarino R, Craik DJ, et al. (2012) Insights into processing and cyclization events associated with biosynthesis of the cyclic Peptide kalata B1. J Biol Chem 287: 28037–28046.
- 21. Reinwarth M, Glotzbach B, Tomaszowski M, Fabritz S, Avrutina O, et al. (2013) Oxidative folding of peptides with cystine-knot architectures: kinetic studies and optimization of folding conditions. Chembiochem 14: 137–146.
- 22. Gressent F, Rahioui I, Rahbe Y (2003) Characterization of a high-affinity binding site for the pea albumin 1b entomotoxin in the weevil Sitophilus. Eur J Biochem 270: 2429–2435.
- 23. Da Silva P, Eyraud V, Carre-Pierrat M, Sivignon C, Rahioui I, et al. (2012) High toxicity and specificity of the saponin 3-GlcA-28-AraRhaxyl-medicagenate, from Medicago truncatula seeds, for Sitophilus oryzae. BMC Chem Biol 12: 3.
- 24. Curtis MD, Grossniklaus U (2003) A gateway cloning vector set for high-throughput functional analysis of genes in planta. Plant Physiol 133: 462–469.
- 25. Voinnet O, Rivas S, Mestre P, Baulcombe D (2003) An enhanced transient expression system in plants based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus. Plant J 33: 949–956.
- 26. Batoko H, Zheng HQ, Hawes C, Moore I (2000) A rab1 GTPase is required for transport between the endoplasmic reticulum and golgi apparatus and for normal golgi movement in plants. Plant Cell 12: 2201–2218.
- 27. Sparkes IA, Runions J, Kearns A, Hawes C (2006) Rapid, transient expression of fluorescent fusion proteins in tobacco plants and generation of stably transformed plants. Nat Protoc 1: 2019–2025.
- 28. Perrin S, Gilliland G (1990) Site-specific mutagenesis using asymmetric polymerase chain reaction and a single mutant primer. Nucleic Acids Res 18: 7433–7438.
- 29. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, et al. (2011) Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 7: 539.