Crop Science Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 8 September 2006
Published in Crop Sci 46:2022-2032 (2006)
© 2006 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hossain, M. A.
Right arrow Articles by Sen, S. K.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Hossain, M. A.
Right arrow Articles by Sen, S. K.
Agricola
Right arrow Articles by Hossain, M. A.
Right arrow Articles by Sen, S. K.
Related Collections
Right arrow Gene Expression
Right arrow Cell Biology & Molecular Genetics
Right arrow Insect Resistance

GENOMICS, MOLECULAR GENETICS & BIOTECHNOLOGY

Transgenic Expression of Onion Leaf Lectin Gene in Indian Mustard Offers Protection against Aphid Colonization

Munshi Azad Hossain, Mrinal K. Maiti, Asitava Basu, Supriya Sen, Arnab K. Ghosh and Soumitra K. Sen*

IIT-BREF Biotek, Indian Institute of Technology, Kharagpur 721302, India

* Corresponding author (sk_sen55{at}yahoo.co.in)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Genes of three naturally occurring monocot mannose-binding lectins from snowdrop (Galanthus nivalis L. agglutinin, GNA), garlic (Allium sativum L. leaf agglutinin, ASAL), and onion (Allium cepa L. agglutinin, ACA) and a recombinant fusion lectin between ASAL and ACA genes were expressed in a bacterial system. The pure and active form of the recombinant lectin peptides were utilized for estimation of their sensitivity potential against feeding nymphs of mustard aphid [Lipaphis erysimi (Kaltenbach)], a major sap-sucking insect pest of Indian mustard [Brassica juncea (L.) Czern.], an oilseed crop. The artificial diet bioassay revealed that ACA and the fusion lectin contained higher toxicity potential than GNA and ASAL. Ectopic expression of these lectins in mustard plants confirmed their protective capacity on the development of the population of aphids on transgenic plants. Based on the strong possibilities that lectins originating from diverse sources would have differential insecticidal potential against different insects, deployment of the appropriate lectin gene figures as crucial in the transgenic approach to protect crop plants against sap-sucking insect pests.

Abbreviations: ACA, Allium cepa L. agglutinin • ASAL, Allium sativum L. leaf agglutinin • GNA, Galanthus nivalis L. agglutinin • MBP, maltose-binding protein • ORF, open reading frame


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
WE ARE BEGINNING to realize that some plant lectins are capable of offering a defense against sap-sucking insects either by adversely affecting the survivability of the insects, or by affecting the fecundity of the insects and/or the ability to transmit pathogens. Since none of the toxins of Bacillus thuringiensis have been found to be toxic to sap-sucking insects, much research has centered on finding any effective insecticidal agent for the purpose. Monocot mannose-binding lectins have earned special significance due to their antiviral (Balzarini et al., 1992), antifungal (Wang et al., 2001), and anti-insect (Gatehouse et al., 1995; Peumans and Van Damme, 1995) properties. When such lectins are expressed transgenically in plant systems, the plants have been shown to develop the capacity to resist the attack of the sap-sucking insect pests (Hilder et al., 1995; Gatehouse et al., 1996; Rao et al., 1998; Stoger et al., 1999; Foissac et al., 2000; Tang et al., 2001; Sun and Tang, 2002; Nagadhara et al., 2004; Dutta et al., 2005a, 2005b). This is considered significant because no other insecticidal agents for transgenic use against sap-sucking insects have proven to be adequately effective. Thus, the current understanding has been that transgenic plants with lectins offering even partial resistance against any sap-sucking insect would still find acceptance in agriculture (Ferry et al., 2004) and could be a useful tool in integrated pest management strategy (Banerjee et al., 2004).

The yield loss of mustard crop caused by aphid (Lipaphis erysimi), a sap-sucking insect, is very high in India (Kumar, 1999; Patel et al., 2004). The nymphs and the adult insects are both responsible for the damage. The insects suck the plant sap during flowering and fruition of the plant, preventing the flow of nutrients from reaching the flowering shoot and thereby affecting the crop yield. Additionally, many of these sap-sucking insects act as vectors for pathogens. Mustard aphids are known to act as vectors for disease-causing luteoviruses. Plant-to-plant transmission of these luteoviruses is in fact completely dependent on such vectors. Thus, an efficient crop protective strategy would be the one that can stop such a destructive cycle. The predatory behavior of the aphids toward the mustard crop has a close relationship with diurnal temperature and the attainment of the flowering stage of the crop. The mustard aphids predominantly breed through parthenogenesis. A female aphid can, within 7 to 10 d, give rise to more than 100 nymphs in the same breeding cycle. As a consequence, the population of the pests increases within a very short time to cover the entire flowering shoot, and unless there is sustained application of chemical pesticides several times in a cropping season, no crop yield of any extent materializes. Although the application of the pesticide provides temporary relief, such liberal dispersion of the chemicals is costly for the farmers and environmentally hazardous. Transgenic attempt to address this problem has been encouraging recently. It has been shown that when the gene for monocot mannose-binding ASAL is expressed, mustard can partially withstand aphid attack (Dutta et al., 2005a). Other plant lectins such as the wheat germ lectin (WGA) and Oryzacystatin gene (OC-1) expressed in mustard and rapeseed respectively can also provide some level of protection against aphid damage (Kanrar et al., 2002; Rahbe et al., 2003). However, many of the lectins are not safe for biosafety reasons and the scope of any role of protease inhibitors in homopteran insect gut physiology remains unclear. For effectiveness and biosafety reasons, the monocot mannose-binding lectins appear as prospective candidate genes in respect of their suitability for the purpose. The snowdrop bulb lectin (GNA) has been proven to be nontoxic to higher animals (Pusztai et al., 1990) and naturally occurring food lectins of garlic (ASAL) and onion (ACA) possessing biopesticidal property (Bandyopadhyay et al., 2001) are expected also to be safe for mammals.

Keeping the above in view, the present study was undertaken to find out the most effective monocot mannose-binding lectins amongst GNA, ASAL, and ACA for transgenic use against the aphids in mustard crop. From past experience with toxin genes of Bacillus thuringiensis, we have learned that the success of a transgenic approach for insect control is heavily dependent on the deployment of the most appropriate toxin gene as determined by the high entomocidal specific activity potential of the gene product against the target insect. In that respect, the study revealed that the onion leaf lectin ACA contains the highest toxic potential against mustard aphids (Hossain, 2005). So, the onion ACA gene and a fusion gene between garlic ASAL and onion ACA lectin were transferred into mustard plants to test their capacity to protect plants against aphid attack. It was documented that the transgenic mustard plant line so developed was significantly resistant to aphids, in greenhouse trials. Thus, the study for the first time demonstrated that selective deployment of a strongly insecticidal lectin gene may form an important component of plant biotechnological strategy to control yield loss of crop due to sap-sucking insect pests. The results of the present study are expected to enrich further the body of literature on plant lectins and their potential role for development of insect-resistant transgenic crop plants.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Source of Materials
Plant Material
Leaf tissues of onion cultivar Muthi and garlic cultivar Desi were utilized as sources for isolation of lectin genes. For plant transformation studies, seeds of an elite mustard cultivar, PCR 7, made available by Dr. P.R. Kumar (Oilseed Rape & Mustard Directorate, Indian Council of Agricultural Research, India) were utilized for the present study.

Lectin Genes
Cellular RNA was isolated from fresh leaf tissues of field-grown onion and garlic using the hot phenol extraction method (Verwoerd et al., 1989). The isolated RNA was then treated with RNase-free DNase I (Boehringer Mannheim, Mannheim, Germany). The first strand cDNA was synthesized using oligo-dT primer and reverse transcriptase (Superscript-II RT, GIBCO-BRL) following manufacturer's instructions. The cDNA-encoding mature part of the garlic and onion leaf lectin peptide was PCR amplified using forward primer 5'-AGCGGATCCATGAGGAACCTACTGACGAAC-3' and the reverse primer 5'-GGCCTGCAGTCATCTTCTGTAGGTACCAGT-3' for garlic; and for onion forward primer 5'-GGCGGATCCATGAGAAACGTATTGGTGAA-3' and the reverse primer 5'-GGCCTGCAGTCATTTCCTGTACGTACCAGTAGACCA-3'. The primers were designed based on already-known gene sequences (GenBank accession no. U58947 for ASAL and accession no. L12171 for ACA). The design of the primers ensured that the coding sequence of both the genes corresponding to the mature peptide start with an ATG (methionine) codon. The underlined sequences of the primers indicate the position of the incorporated BamHI and PstI sites, in 5' and 3' primers, respectively. The gene encoding mature snowdrop bulb lectin, GNA was chemically synthesized in a 392 DNA/RNA synthesizer (Applied Biosystems, Foster City, CA) based on LECGNA2 sequence (GenBank accession no. A18023). This was further reconstructed by recombinant PCR using two sets of long overlapping primers and one set of 5' and 3' end primers with incorporated BamHI and PstI sites, respectively. The oligonucleotides for the two end primers (restriction site underlined) were: 5'-ATCGCGGATCCATGGACAATATTTT GTACTCC-3' as forward and 5'-CGAAAACTGCAGTTATCCGGTGTGAGTTCCAGTAG-3' as reverse primers. The forward primer contained the introduced ATG as the initiation codon. Two rounds of PCR were performed to get the final product. In the first round, two separate PCRs were performed using Deep Vent DNA polymerase (New England Biolab, Beverly, MA) taking each set of long primers with the following thermal profile: initial denaturation at 94°C for 4 min, followed by 30 cycles of 94°C for 30 s, 55°C for 40 s, and 75°C for 1 min, and a final extension at 75°C for 7 min. The second round PCR was done with 5' and 3' end primers using equimolar mixture of two first round PCR products as template, following similar thermal profile as in case of first round PCR. After digestion of the final PCR products (RT-PCR of garlic and onion and recombinant PCR for snowdrop) with BamHI and PstI, the DNA fragments were cloned in pUC18 vector using E. coli DH10B (GIBCO BRL) as the host. Furthermore, construction of a fusion lectin gene of garlic (ASAL) and onion (ACA) was accomplished by taking advantage of the existing unique KpnI site near the 3' end of the garlic ASAL gene. For this, the ACA gene was PCR amplified with the help of two primers; the forward primer 5'-GGGGTACCTACAGAGAAGAAGAACCGTATGGTGACAA-3' contained the overlapping sequence of near-3' end (bold) of ASAL gene as well as the 5' region (italic) of ACA gene, and the reverse primer is ACA gene specific 3'–end primer 5'-GGCCTGCAGTCATTTCCTG TACGTAC CAGTAGACCA-3'. The PCR amplified fragment was digested with KpnI and PstI (underlined sequences in the primers); and ligated with BamHI–KpnI fragment of ASAL gene to clone in pUC18 at BamHI–PstI sites. Sequencing of the randomly selected clones confirmed the desired open reading frame (ORF) starting with methionine codon of ASAL gene and ending with stop codon of ACA gene. The fusion gene thus developed was designated as ASAL::ACA.

Generation of Recombinant Lectins in Bacterial Expression System
The genes ACA, ASAL, GNA, and ASAL::ACA were expressed individually under Ptac promoter based vector, pMAL-c2E (New England Biolab) as maltose-binding protein–lectin (MBP-lectin) fusion protein in Escherichia coli, strain TB1. All four lectins were expressed in soluble form (>90%) at 28°C with 0.1 mM IPTG induction, though a small amount of protein still remained as insoluble aggregates. It is known that the molecular mass of the MBP–ß-galactosidase {alpha} fragment fusion product without the foreign gene insert is 50.8 kDa, whereas the molecular mass of the MBP partner of the recombinant fusion protein is only 42.5 kDa when the vector contains an insert (Riggs, 1997). Thus, the molecular mass of the MBP-lectin fusion proteins viz., MBP-ACA, MBP-ASAL, and MBP-GNA, were about 55 kDa. However, the molecular mass of MBP-ASAL::ACA was 67.5 kDa. Sufficient amount of pure recombinant proteins could be recovered through MBP-amylose affinity chromatography. Affinity-purified MBP-lectin fusion proteins were dialyzed against 20 mM Tris pH 8.0, 50 mM NaCl, 2 mM CaCl2. Thereafter, the recombinant proteins were digested with enterokinase (New England Biolab) at 23°C for different enzyme–protein (w/w) combinations. The enterokinase cleaved the four recombinant lectins at the enterokinase digestion site of the fusion protein. It was found that ~95% of the total protein was digested at 23°C for 30 h and the enterokinase needed was in the range of ~0.05% of the total protein for different lectins. Digestion was stopped using a protease inhibitor cocktail (Roche Applied Science, Mannheim, Germany) and each set was extensively dialyzed against 20 mM PBS pH 7.4.

Hemagglutination Assay
Heparin-treated rabbit erythrocytes were washed extensively with PBS and finally resuspended in the required volume of PBS to generate 1% (v/v) erythrocyte suspension. Agglutination assays were performed in microtiter plates in a total volume of 150 µL containing 50 µL of erythrocyte suspension per well with addition of serially diluted measured amount of protein samples. The agglutination reaction was monitored visually after 1-h incubation at room temperature.

Laboratory Culture of Mustard Aphids
First instar nymphs of mustard aphids were collected in Petri plates with the help of a soft brush from the field-grown plants. The rearing in the laboratory was performed on standard artificial diet (Dadd and Mittler, 1966) supplemented with green tissue extracts of young shoots of mustard plant mixed in 1:3 (v/v) ratio with artificial diet dissolved in 7% sucrose solution and membrane sterilized. The rearing condition of the insects comprised of a layer of sterile absorbent cotton placed in the bottom half of a Petri plate (35-mm diameter), covered by a fine greenish yellow linen tightly fixed at the top. A fixed volume of the liquid diet was mixed with different concentration of bacterially expressed recombinant lectins (lyophilized) and then poured drop by drop onto the top of the linen. The cotton layer beneath the linen was soaked in the liquid artificial diet. The linen surface was covered with a monolayer of a stretched (four times its original length) parafilm membrane. The insect could access the diet by using its proboscis to pierce the membrane. For the insect-feeding test, 16 nymphs were placed on the top of the parafilm layer of each of the Petri plates. The rearing of the insects was performed in a growth chamber at 22°C, 75% RH, and under fluorescent light (140 µE m–2 s–1) at a photoperiod of 16 h. Each test was conducted three times, with three replicates. Every third day a fresh set-up of the Petri plates containing the feed was provided, replacing the old one. The nymphs could develop normally into complete adults with no apparent developmental constraints.

Agrobacterium-Mediated Transformation and Generation of Transgenic Mustard Plant Lines
For plant transformation with lectin genes (ACA and ASAL::ACA), molecular constructs were developed by fusing a 2x35S CaMV promoter element (476 bp) at the 5' end of the coding sequence of the lectin gene and the nopaline synthase (nos) terminator sequences (270 bp) at the 3' end (Fig. 1 ). The DNA fragment containing the chimeric gene was inserted at the multiple cloning site of the Agrobacterium-mediated plant transformation vector pCAMBIA 1300 and finally introduced into Agrobacterium tumefaciens strain LBA4404 Vir GN54D (Van der Fits et al., 2000). Mature seeds of mustard, PCR 7 were surface sterilized, germinated in dark on MS (Murashige and Skoog, 1962) basic medium supplemented with 3% sucrose (pH 5.8). Apical meristem from 3- to 4-d-old seedlings were preconditioned for 72 h on MS medium supplemented with 3% sucrose, 1 mg L–1 BAP and 1 mg L–1 NAA at 25°C under 16 h light/8 h dark cycle. The preconditioned explants were then incubated in Agrobacterium suspension for 10 to 15 min, removed and then laid out on sterile blotting paper and incubated in MS medium supplemented with 3% sucrose, 1 mg L–1 BAP, 1 mg L–1 NAA at 25 ± 1°C for 72 h. The explants were transferred for growth on the regeneration medium containing antibiotics: cefotaxime 250 µg mL–1 and augmentin 60 µg mL–1 for 15 d and 25 µg mL–1 hygromycin for selection. Three to four rounds of subculture were performed in the same medium at 28 ± 2°C with 16 h light/8 h dark photoperiod for 15 d in each case. Emerging green shoots were selectively transferred to fresh basic MS medium with 3% sucrose, 1 mg L–1 BAP, 1 mg L–1 NAA. Each regenerated plant lines was numbered and then multiplied in several copies in culture. A part of in vitro grown plants were transferred to glasshouse contained environment and grown to maturity.


Figure 1
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 1. Schematic representation of the T-DNA region of (A) pCAMBIA-ACA which carried the Allium cepa agglutinin (ACA) gene; and (B) pCAMBIA-ASAL::ACA which carried the fusion lectin (ASAL::ACA) gene; 2x35S, doubly enhanced CaMV35S promoter; hypt II, hygromycin resistance gene; Nos, nopaline synthase transcriptional terminator; RB, right border; LB, left border. ASAL, Allium sativum leaf agglutinin.

 
Molecular Analysis of Plant Transformants
For identification of the transgenic plant lines, genomic DNA was extracted following CTAB method (Doyle and Doyle, 1990) from leaves of the putative transgenic plants. PCR analysis of the genomic DNA was performed using ACA and fusion lectin gene ASAL::ACA specific primers. The genomic DNA from untransformed mustard plant and the plasmid DNA of the binary vector (pCAMBIA-ACA/pCAMBIA-ASAL::ACA) were used as negative and positive controls, respectively. Initial screening was conducted by PCR-based test for the presence of the lectin gene.

Western analysis was carried on the proteins, from the homogenized leaf tissue of presumptive transgenic lines (PCR-based selection) extracted in buffer (20 mM Tris, 100 mM NaCl, 0.02% Tween 20, 10 mM PMSF). The supernatant was collected after centrifugation of the extract at 12.8 x 103 g for 10 min. The total protein content in the extract was estimated (Bradford, 1976). About 100 µg of total protein from each plant sample was separated in 15% SDS- PAGE and then transferred to PVDF membrane (Boehringer Mannheim). Membranes were probed with rabbit anti-ACA antibody at 1:5 x 103 dilution. The ASAL::ACA and ACA blots were processed and developed by the ECL method (Amersham Biosciences) according to the vendor's instruction. Quantitative estimation of the expressed protein on the Western blot was performed with the help of a Kodak Digital Science ID Image Analysis Software (9/97–1b5430103 win version 2.0.1; New York).

Southern blot analyses were conducted of selected transformed plant lines that expressed the transferred lectin gene at detectable levels. About 10 µg of genomic DNA was digested with BamHI and fractionated on 0.8% agarose gels and blotted on Hybond-N+ membrane (Amersham Biosciences). Hybridization in solution (10% dextran sulfate, 50% formamide) was performed for 18 to 20 h using [{alpha}-32P]dCTP-labeled 342-bp coding region of ACA gene and 681 bp of ASAL::ACA fusion gene for ACA and ASAL::ACA blot, respectively. After hybridization, the blots were washed and autorads were developed on Kodak AR film after exposure at –70°C.

Test for Survivability and Fecundity of Aphids Feeding on Transgenic Plants
Micropropagated copies of transgenic T0 lines of ACA#2, 8, 16; ASAL::ACA#11, 4; and untransformed PCR7 mustard plants were raised for experimental use. Young plants (8-cm height) in small pots (10 by 8 cm) of sterilized sand, soil, and vermiculite in 1/2 MS solution were enclosed within a vertically placed transparent plastic tunnel, and the top of the tunnel was covered with a muslin cloth. Each plant was inoculated with 20 live first instars nymphs with the help of a brush. A set of 12 potted plants inoculated with insects were placed in a plastic tray and reared in a growth chamber at 24°C, 75% RH, and a 16 h light/8 h dark photoperiodic regime. Observations on survivability of the insects were made after 8 d, and the extent of development of the insect population was estimated after 14 d. The test was conducted three times with two replicates in each case.

Test for Build Up of Population Flux Density when Aphids Colonized a Transgenic Plant Line
In this test mustard aphids were made to infest a T0 ACA#8 plant line and an untransformed plant. Each of the plant lines was multiplied in four copies maintained in separate pots. The individual potted plants were raised in nylon mesh cages in the glass house. When the plants initiated flowering, one flowering twig per plant was enclosed within a porous (60-µm pore diameter) nylon bag (35 cm long by 10 cm broad), and 30 live adult aphids were introduced within the nylon bag with the help of a suction gadget. The nylon bag was sealed at the base with the stalk so that the entrapped aphids could not move beyond the plant part they were exposed to. A wooden stick was additionally fixed to provide mechanical support to the flowering shoot with the nylon bag. Observations were made on the 10th, 20th, 30th, and 40th day of the experiment. The experiment was performed only once in a cropping season with one replicate.

Statistical Analysis
Insect Mortality
Considering the natural death of the insect, the percentage of insect mortality was determined using Abbott's formula

Formula 1[1]
(where X = percentage of survivability in the control line where no toxin is present and Y = percentage of survivability in a treatment line). Percentage of mortality values was further converted to probability unit (probit). A linear regression could be obtained by plotting probit values vs. log10 of doses of toxin; finally LC50 (median lethal concentration) values could be derived at probit value of 5 for each treatment line (Finney, 1971). Furthermore, since the insect species used for bioassay were initially collected from the field, it is likely that the insect populations represent for a heterogeneous group, so their tolerance limit would be different. Keeping this in mind, the confidence limit (upper and lower) of LC50 values were also determined at 95% probability. The true value of LC50 was asserted to lie between them.

Estimation of Standard Error
Standard error of the mean was estimated to measure the precision of the estimate of the mean. This was obtained by taking the standard deviation of the sample and dividing it by the square root of the number of observations.


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Characteristics of the Lectin Genes
Nucleotide sequence analysis of ASAL and ACA genes revealed that they both contained a 333-bp ORF, whereas the chemically synthesized GNA gene contained a 330-bp ORF, including the introduced initiation codon. There were six nucleotide changes in the ACA gene sequence of the onion cultivar from the earlier reported nucleotide sequence of ACA gene (Van Damme et al., 1993). These changes predicted differences in two amino acids, V18E and I24Y. In the case of ASAL, one nucleotide change was observed when compared to the published ASAL gene sequence (Smeets et al., 1997) resulting in amino acid difference at position 85 from N to S. All changes were located outside the conserved amino acid sequence for the mannose-binding domain (QXDXNXXXY). However, in case of ACA the changes were quite close (eighth and fifth amino acid upstream) to the first mannose-binding domain (QDDCNLVLY). The DNA sequences corresponding to ACA and ASAL have been deposited into GenBank as accession no. AY376826 and AY376827. The deduced amino acid sequence of the chemically synthesized GNA gene confirmed the presence of three mannose-binding domain per lectin peptide as in the case of onion and garlic leaf lectins. The mature ACA, ASAL, and GNA lectin peptides contained 111, 111, and 110 amino acid residues respectively, including the introduced starting methionine (Fig. 2 ). Each of them encoded a protein having molecular mass of about 12.5 kDa, similar to the mass of the native protein. Comparison of the amino acid sequences of ACA and ASAL lectins (Fig. 2) revealed that the extent of similarity was 86.3%. However, the two Alliaceae leaf lectins possessed only 49% of similarities with the GNA of Amaryllidaceae family. The fusion lectin ASAL::ACA encoded a polypeptide of 221 amino acid residues with six predicted mannose-binding domains. The estimated molecular mass of the fusion protein was about 25 kDa.


Figure 2
View larger version (53K):
[in this window]
[in a new window]
  		Fig. 2. Amino acid sequences corresponding to mature peptides of Galanthus nivalis agglutinin (GNA), Allium sativum leaf agglutinin (ASAL), and Allium cepa agglutinin (ACA). The sequences were aligned using Biowire Jellyfish Version 3.0 www.biowire.com). Underlined amino acids indicate the mannose-binding domain (QXDXNXXXY), and the shaded regions show identical amino acids among the three sequences.

 
Recombinant Lectins are Biologically Active
Agglutination assays were performed with the same molar ratio of enterokinase-digested (MBP+lectin) and undigested (MBP–lectin) fusion recombinant proteins. The digested protein showed enhanced agglutination activity compared to the uncut counterpart. The four different recombinant lectins possessed varying agglutination activities with rabbit erythrocytes and required different concentration levels of mannose (145 mM for ACA; 135 mM for ASAL::ACA; 185 mM for ASAL; and 175 mM for GNA) to inhibit agglutination. Inhibition of agglutination assay with mannose proved the presence of the active mannose-binding groove in the recombinant lectin. The test proved that the recombinant lectins were biologically active.

Insect-Feeding Tests for Toxicity of the Lectins
The four recombinant lectins, GNA, ASAL, ACA, and ASAL::ACA were mixed individually with the artificial diet for insect-feeding tests. Insect mortality was monitored at 48-h intervals and varying concentration levels of each lectin were mixed with the artificial diet. They were found to be variably toxic to the feeding nymphs. No toxicity was observed when the diet was not supplemented with any of the lectins. It was found that the survivability decreased linearly with the levels of dosage (Fig. 3A ). The LC50 mean values as determined were 20.5 µg for ACA, 33.7 µg for ASAL, 33.4 µg for GNA, and 25.5 µg for ASAL::ACA per milliliter of the artificial diet. Among the four lectins, ACA was found to be the most toxic. The ASAL::ACA was the secondmost toxic to the insect. The effectiveness of ACA in causing mortality to insects was found to be highest in comparison to other lectins (Fig. 3B). When 20 µg of each kind of lectin was mixed with the artificial diet and mortality of the insects was monitored every day for 5 d, it was evident (Fig. 3B) that highest mortality could be observed in case of ACA followed by ASAL::ACA.


Figure 3
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 3. Insect bioassay test for sensitivity of four lectins when present with the artificial diet. (A) Different concentration levels of lectin (10–50 µg mL–1) were offered to 16 insects and surivivability was monitored after 2 d. (B) Specific activity of different lectins causing mortality to the insects monitored every day for 5 d, when 20 µg mL–1 of different lectins was mixed with artificial diet.

 
Generation of Transgenic Plants with ACA and ASAL::ACA Lectin Genes
Several lines of transgenic plants showing resistance to hygromycin were generated. After initial PCR selection, 22 putative transgenic plants with ACA gene were grown to maturity. In the similar manner, 12 putative transformant lines with ASAL::ACA gene were identified. Thereafter, Western blot analysis of all the presumptive transformants was performed to identify the most expressive transgenic lines. Some of the primary transgenic lines showed expression of the transferred gene to be at detectable levels. Presence of a polypeptide of 12.5 kDa was detected in those lines that corresponded to the ACA protein (Fig. 4A ). It revealed that the T0 plants ACA#2 (0.39% of total protein) and ACA#8 (0.61% of total protein) showed fairly high levels of expression of the transferred ACA lectin gene in the respective transgenic leaf tissue. However, a transformant, ACA#16, did not show any detectable trace of expression (Fig. 4A) of the transferred gene product. Likewise, Western blot analysis of a transgenic line (ASAL::ACA #11) with fusion lectin gene documented detectable level of expression (0.46% of total protein) and in transgenic line ASAL::ACA#4 the expression (Fig. 4B) of the transferred gene was quite low (0.04% of total protein). The expressed polypeptide ASAL::ACA showed higher molecular mass (~25 kDa) than the corresponding mass of the ACA lectin peptide used as the positive control (Fig. 4B). No signal was detected in the untransformed control plant.


Figure 4
View larger version (52K):
[in this window]
[in a new window]
  		Fig. 4. Western blot analysis of total protein from transformed B. juncea plants using Allium cepa agglutinin (ACA) specific antibody. (A) Expression of ACA gene product amongst transgenic lines (T0). Lane 1, untransformed plant (–ve); lane 2, ACA peptide as positive control (+ve); lanes 3–9, transgenic plant lines ACA#2, 16, 12, 8, 14, 18, and 4, respectively. (B) Expression of fusion lectin amongst transgenic lines (T0) ASAL::ACA. Lanes 1–7, transformed plants with ASAL::ACA#2, 4, 5, 8, 9, 7, 11, respectively; lane 8, untransformed plant as negative control (–ve); lane 9, ACA peptide as positive control (+ve). ASAL, Allium sativum leaf agglutinin.

 
Southern blot analysis of the expressive transgenic plant lines was conducted. For this, 10 µg of genomic DNA of each experimental plant line was digested with BamHI restriction endonuclease. It is evident from Fig. 1 that a single BamHI site is present in the T-DNA of the transformation vector. When the plant genomic DNA is cleaved by BamHI restriction endonuclease, different primary transgenic plant lines would generate DNA fragments of varying sizes containing the transferred lectin gene. This is because each event of random integration of T-DNA in the plant genome would be in a unique site and the BamHI site beyond the left border of T-DNA would be different in each case. After random integration of T-DNA expression cassette in plant genome after BamHI cleavage, hybridized DNA fragments would normally be larger than 2.8 kb in the case of ACA and 3.1 kb in case of ASAL::ACA as that would be the minimum size of the T-DNA present in the plant genomic DNA after cleaving with BamHI. The autoradiograms showed (Fig. 5A and 5B) a single copy of integration events in most cases, while there were some cases showing more than one site of integration. No such signal could be detected in the case of the untransformed plants. Thus, the results of the Southern blot analysis provided evidence for integration of the transferred full-length ACA and ASAL::ACA gene in the plant nuclear genome. The transgenic lines ACA#2 and ACA#8 and ASAL::ACA#11 contained a single copy of the transferred gene in each case.


Figure 5
View larger version (47K):
[in this window]
[in a new window]
  		Fig. 5. Southern blot analysis of some of the primary transgenic plant lines. (A) Transgenic plant lines with Allium cepa agglutinin (ACA) gene. Lane 1, untransformed plant as negative control (–ve); lanes 2–8, transgenic lines ACA#7, 6, 8, 16, 2, 10, 12, respectively; lane 9, plasmid DNA (pCAMBIA-ACA) as positive control (+ve). (B) Transgenic plant lines with ASAL::ACA fused lectin gene; lane 1, untransformed plant as negative control (–ve); lanes 2–9 transgenic lines ASAL::ACA# 2, 4, 5, 7, 6, 8, 9, 11, respectively; and lane 10, plasmid DNA (pCAMBIA- ASAL::ACA) as positive control (+ve). ASAL, Allium sativum leaf agglutinin.

 
Effect on Survivability and Fecundity of Aphids when Feeding on Transgenic Plants
The experimental treatment lines consisted of one untransformed plant line PCR 7 as the negative control line; two transformed plant lines, ACA#16 and ASAL::ACA#4, that expressed the transferred lectin gene either at nondetectable level as in case of ACA#16 or at low level as in case of ASAL::ACA#4; and three transgenic plant lines (ACA#2, ACA#8 and ASAL::ACA#11) that expressed the transferred lectin gene fairly well. It is evident from Fig. 6A that the insects feeding on different transgenic plant lines showed varying levels of mortality. Relatively higher mortality levels were observed in the case of the expressive transgenic plant lines viz., ACA#2 and #8; and ASAL::ACA#11 when compared to the untransformed plant line as well as nonexpressive transgenic plant lines. The majority of the dead insects were adults indicating that the nymphs could by and large metamorphose to adulthood. After 14 d, it was expected that the nymphs would develop into adults and the adults would give rise to a cycle of fresh generation of nymphs. It was, however, observed that the increase in the insect population was adversely affected (Fig. 6B) by insects' feeding on plants that expressed the transferred lectin gene. The relatively lower number of insects present (Fig. 6B) in the case of transgenic plant lines ACA#2, ACA#8, and ASAL::ACA#11 indicated that the presence of the lectin peptide was responsible for preventing the increase in the population of the insects. The reduction in the nymph production rate of the aphids feeding on transgenic plants containing the lectin peptide, implied that the latter does adversely affect the fecundity of the insect.


Figure 6
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 6. Adverse effects on the survivability and fecundity of insects were observed when the insects colonized on transgenic plants containing the Allium cepa agglutinin (ACA) peptide. First instar nymphs of aphids were placed on young transgenic and control (untransformed) plants. (A) After 8 d, the survivability of the insects was monitored. (B) After 14 d, the total number of insects (adults and nymphs) was estimated to judge the fecundity of the insects as the result of feeding on transgenic plants expressing the lectin peptide.

 
Population Flux Density with Time when Aphids Colonize Transgenic Plants Expressing Lectin
The result of aphids harboring for 40 d on a flowering shoot of transgenic ACA#8 plant line made for an interesting observation. It was observed that the shoot bearing the developing fruit in the case of the untransformed control plants was covered entirely with young and adult aphids after 40 d (Fig. 7B ). In contrast, the ACA #8 plant showed the presence of only a few insects (Fig. 7A). The magnitude of infestation of the insects on the flowering shoot and the fruit of the control plant was conspicuously high when compared to the presence of the few insects on the transgenic plant. Within the first 10 d, the aphids could increase in number. In the case of transgenic plants the increase has been slower than the nontransgenic plants (Fig. 8 ). However, it became gradually evident that the rate of increase of the insect population in the case of transgenic plant was much slower than in the untransformed control plant (Fig. 8). After 20 d, the insect population in the case of the untransformed plant was again far more in number when compared to the transgenic plant. The same trend continued in case of the transgenic plant where the insect population had ceased to increase (Fig. 8).


Figure 7
View larger version (106K):
[in this window]
[in a new window]
  		Fig. 7. The figure demonstrates accumulation of aphid population after 40 d on a transgenic plant ACA #8 expressing Allium cepa agglutinin (ACA) lectin peptide versus an untransformed plant. On the 40th day, the flowering twig shows (A) only a few aphids are present in the transgenic ACA #8; (B) the twig is fully infested with aphids on the untransformed control plant parts.

 

Figure 8
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 8. The normal flow of increase of population of aphids with time was severely affected when observed after 40 d on the transgenic plant line expressing ACA lectin peptide, when compared to the increase of aphid population after 40 d on the untransformed plant. The figure documents that the number of aphids colonizing on the transgenic plant, ACA# 8 had shown an initial increase until the d 20, but ceased to increase further. On the contrary, the number of aphids infesting the untransformed plant has shown rapid increase with passage of time. The ({diamond} line indicates insects on untransformed PCR7 plant and the {triangleup} line indicates insects on transgenic ACA#8 plant.

 
Test for Inheritance of the Transferred Lectin Gene in T1 Generation
T1 seeds of primary transgenic T0 ACA#2, #8 and ASAL::ACA#11 lines were obtained after self fertilization of each of these lines. Randomly chosen 50 T1 seeds of each line were germinated in basic MS medium in presence of hygromycin, along with seeds of untransformed control lines. Although some of the T1 seeds in presence of hygromycin (30 mg L–1) had ceased to grow after germination and turned brown, the majority of the T1 seeds continued to grow in presence of hygromycin. Seeds of untransformed plants showed full sensitivity to hygromycin. Simultaneously, in another experimental set up, 50 T1 seeds of each transgenic line were germinated without hygromycin. Presence of the transferred ACA or ASAL::ACA gene in their genomic DNA was monitored in these seedlings through PCR analysis. Both hygromycin resistance and presence of ACA gene traits were observed to be present three times more often than plants showing hygromycin sensitivity and absence of the ACA gene amongst T1 progenies. The inheritance of the two traits was thus found to follow a Mendelian pattern of inheritance ratio of 3:1. The two characters co-inherited in the T1 progenies.


   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Enrichment of the armory of insecticidal agents with entomocidal mannose-binding lectins originating from diverse biological sources are expected to contain differential potency against diverse sap-sucking insects. Evidence for different lectins showing variation in their potency against a single insect system was not available. The present study has fulfilled that requirement. It was observed that ACA is more potent than ASAL and GNA against mustard aphid. An analogy of this situation can be found in case of varying insecticidal potency of {delta}-endotoxins of Bacillus thuringiensis against lepidopteran insects. The different classes of CryIA toxin peptides show a wide range of variation of toxicity against lepidopteran insects in spite of the presence of considerable degree of structural similarity amongst the toxin molecules. A specific receptor molecule in the gut of the mustard aphid has been observed to bind with the lectin molecule to initiate the insecticidal effect in the insect system (Banerjee et al., 2004), a pathway similar to the mechanism of action of Bacillus thuringiensis toxin against lepidopteran insects for toxicity. Existence of a positive relationship between binding of lectins to glycoprotein receptors in the gut of the insect and entomocidal activity has already been proposed in the past (Harper et al., 1995; Bandyopadhyay et al., 2001). Since plant–aphid–virus association maintains high specificity, any lectin capable to interrupt this association essentially also has to be specific. Thus, entomocidal potency of a lectin is suspected to depend on the interactive capacity of the lectin molecule being able to disrupt this association. In support of this prediction it is known that the insecticidal property of a lectin varies with different insect systems and that contrasting responses of aphids to various lectins do exist (Gatehouse et al., 1995; Rahbe et al., 1995; Carlini and Grossi-de-Sa, 2002). Furthermore, GNA has been found to cause mortality to rice brown plant hopper [Nilaparvata lugens (Stäl)] and green leaf hoppers [Nephotettix virescens (Distant)] (Rao et al., 1998; Foissac et al., 2000; Sun and Tang, 2002), whereas it only reduces the fecundity of peach potato aphids [Myzus persicae, (Sulzer)] (Down et al., 1996; Gatehouse et al., 1996). Thus, identification of the most potent lectin to be used in controlling damage caused by the target sap-sucking insects is an essential component of any meaningful transgenic approach to control crop loss due to insect colonization. Moreover, a lectin with high specific activity will require low transgenic expression level, which technically should be achievable. Thus, it is the first report to establish that the monocot mannose-binding lectins, known to be safe for human consumption, differ in their entomocidal potency in respect to a sensitive insect. Thus, this phenomenon offers interesting perspectives to improve the resistance or quality of transgenic crop plants.

The specific activity of ACA as an insecticidal agent against mustard aphids on an artificial diet was found to be higher than ASAL and GNA. The low dose requirement of the lectins for the feeding insects to be sensitive in the present study might have been accentuated by the purity of the lectin used. Also, sensitivity of the first instar nymphs to the toxin than adults might have contributed additionally toward suitability of the test system. Setting up of an artificial diet bioassay system in the laboratory for sap-sucking insects has always been a tricky proposition. The test, however, demonstrated that all the four recombinant lectins were differentially toxic. The results of the test were reproducible and found to be sensitive enough to reveal that the homopteran insects were indeed more sensitive to mannose-binding plant lectins than shown in earlier results (Powell et al., 1993, 1998; Bandyopadhyay et al., 2001).

The only form of the lectin gene (ACA) expressed in leaves of onion and one gene amongst a few isoforms present in the leaves of garlic was isolated with the application of RT-PCR technique. These genes were bacterially expressed for utilization of the pure form of the gene product. Longstaff et al. (1998) in the past had expressed the GNA gene in E. coli and obtained 50% of biologically active GNA from inclusion bodies after purification through affinity chromatography and refolding of the denatured protein. A different bacterial expression protocol was successfully adopted in the present study. It was possible to generate over 90% of the total expressed protein in soluble form (Kapust and Waugh, 1999). The recovered four recombinant MBP-fusion proteins were thereafter cleaved with enterokinase before their use. That the biological activity of the enterokinase-digested recombinant lectins could be retained was confirmed by their ability to agglutinate rabbit erythrocytes. The sensitivity test of insects to the lectin has been performed for the first time in the present study with purified recombinant toxin peptide free form any contamination from other isoforms of the peptide. The number of active sites per monomer amongst monocot mannose-binding lectins varies from zero to three. A good correlation does exist between the number of active binding sites and the biological activity (Barre et al., 1996). A functional lectin in case of ASAL is a dimer, whereas in the case of ACA it is a tetramer. Hemagluination test revealed that the fusion between ASAL and ACA lectin contained higher hemaglutination activity than any of the single lectins. Although the precise mode of action of entomotoxic lectin is not yet fully understood, binding of lectins to glycoprotein receptors in the gut of target insects seems to be associated with its activity (Harper et al., 1995; Bandyopadhyay et al., 2001; Banerjee et al., 2004). Our studies with hybrid lectin, a fusion of garlic and onion lectin, indicated that there is no direct correlation between the carbohydrate (mannose) binding capacity and efficiency of insect (sap-sucking) toxicity of the lectin. The binding of the carbohydrate moiety to a lectin is mediated by a complex network of hydrogen bonds and by hydrophobic interactions between the side chains of the amino acid residues, comprised of the binding site and the sugar residues (Weis and Drickaner, 1996).

High levels of ectopic expression of ACA and the fusion lectin gene could be achieved in some of the transgenic plant lines. The expression could be achieved under the control of a modified 2x35SCaMV constitutive promoter. In the past, a number of constitutive promoters have been used (Rao et al., 1998; Kanrar et al., 2002; Dutta et al., 2005a). In certain cases, phloem specific promoters have also been used. However, no apparent advantage could be gained through the use of the phloem specific promoter (Dutta et al., 2005a). It was observed in the present study that the ectopically expressed lectins could negatively influence development of population of aphids on the transgenic plants. It became thus evident that the onion leaf lectin is indeed a potent biopesticidal agent for developing aphid-resistant transgenic mustard plant. The lectin, on ingestion by the insects, caused mortality and impairment of fecundity of the surviving insects. We further speculated that the continuous feeding on the same plant expressing the lectin peptide might have been instrumental in arresting the insect population (Fig. 8). Nevertheless, the survival of some aphids, albeit few in number, could not be explained. We believe that the experiment mimicked fairly closely what would have happened in a real-life situation. This would imply that the plant expressed ACA lectin could prevent increase of the aphid population in a significant way. The mechanism that affects the fecundity of the insects appears to be different than the mechanism that results in mortality, a process mediated through binding of lectin with the receptor molecule in the insect gut (Banerjee et al., 2004). The effect on fecundity seemed to be positively correlated with the influence of the lectin to the physiology of the sensitive insects. Localization of GNA throughout the hemolymph of the rice brown plant hopper (Powell et al., 1998) and accumulation of ASAL in the ovarioles of peach potato aphid have been observed (Dutta et al., 2005b). The effect of lectin in causing mortality and/or lowering fecundity of insect are both desirable as each of these effects will lower the stress imposed on the host plant through predation. Thus, any lectin that can contribute most in offering relief to the plants from the stress, should be preferred.

The present finding represents for the first report on ACA as insecticidal agent different from ASAL and GNA, in terms of transgenic use in offering protection of a plant against sap-sucking insects. The earlier held reservation about the prospect of lectins as insecticidal agent in transgenic use (Estruch et al., 1997) seems to be not tenable any more. The protective capacity contained in GNA against a number of sap-sucking insects, ASAL against mustard and peach potato aphids, and now ACA against the mustard aphid have been experimentally established. The evidence provided in the present study for ACA as an anti-insect agent adds to the repertoire of plant lectins suitable for biotechnological use. It is expected that the transgenic plants with these lectins will be a cost-effective option to defend our crop plants against sap-sucking insects to a greater degree than any of the insecticidal agents until now.


   ACKNOWLEDGMENTS
 
This work was partly supported by grants received by the laboratory from the Council of Scientific and Industrial Research, New Delhi, and the Department of Biotechnology, Government of India. Special thanks to Dr. S. Das, Bose Institute, Calcutta for making some of their research results available before publication. Skilled technical supports received from Mr. Bijon Ghosh in insect bioassay; Ms. Dolly Jana for transformation experiments; and Mr. Meghnath Prasad for secretarial help are acknowledged gratefully.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Munshi Azad Hossain and Mrinal K. Maiti contributed equally to this work.

Received for publication March 18, 2006.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hossain, M. A.
Right arrow Articles by Sen, S. K.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Hossain, M. A.
Right arrow Articles by Sen, S. K.
Agricola
Right arrow Articles by Hossain, M. A.