Fresh tubers of Colocasia esculenta and nymphs of Aphis gossypii, Dysdercus cingulatus & Aphis craccivora were collected from institutional farm as per requirement.

Lectin was isolated from tuber tissue extracts of Colocasia esculenta by affinity chromatography using affinity matrix [α-D-Mannose, immobilized on cross linked 4% beaded agarose, Sigma-Aldrich], pre-equilibriated with 20 mM Tris-HCl (pH 7.2) following the protocol of Van Damme et al. [13] with some modification as described by Roy et al. [11]. The matrix bound lectin was desorbed with 20 mM unbuffered 1,3-diaminopropane [DAP] and lyophilised.

The lyophilised semi-purified sample from affinity purification was dialyzed against phosphate-buffered saline [PBS], pH 7.2 and loaded onto a DEAE-Sephadex ion exchange column equilibrated with PBS, pH 7.2. A gradient of 0 - 500 mM NaCl solution was passed through and the eluted fractions were collected, pooled & lyophilised. Bradford assay was carried out to quantify the purified protein [14].

Lectin preparations after chromatographic separation were resolved in 15% SDS-PAGE according to Laemmli [15]. Western blot analysis was carried out by using an antiCEA polyclonal primary antibody developed in our laboratory following the protocol of Harlow et al. [16] and anti-rabbit IgG-horse radish peroxidase conjugates as the secondary antibody obtained from Sigma-Aldrich, USA.

Colocasia lectin was allowed to agglutinate rabbit erythrocytes as described by Mondal et al. [9]. Rabbit erythrocytes were collected, washed extensively in a 0.9% saline solution, and finally made up to 20% (v/v) with 0.9% saline. Twenty microliters of erythrocyte suspension was dispensed in each well of microtiter plate. Eighty microliters of serially diluted lectin solutions (starting from 4 to 1 μg per well) were added to each well. Control well was supplemented with same amount of 0.9% saline water. Agglutination reaction was monitored visually, after incubation for 1 h at 37˚C.

Bioassay has been set up on an artificial diet formulated with some modifications [6] from the original description by Dadd and Mittler [17]. Second instar nymphs of Aphis gossypii [Cowpea aphid], Dysdercus cingulatus [Red cotton bug] and Aphis craccivora [Cowpea aphid] were incubated for each set in liquid diet of 200 μl supplemented with different doses of CEA (0, 5, 10, 15, 20, 25 μg/ml each). The experimental set up was designed according to Mondal et al. [9]. Entire experiment was done in triplicate. Data on survival of the insects were collected at every 24 h. The LC₅₀ value of each toxin corresponding to each insect had been determined by statistical probit analysis [18]. Statistical significance level used was 0.05 [α ≤ 0.05]. The statistical evaluation was done using the χ² method as specified by the Probit analysis software [11,18].

CEA protein band form SDS PAGE gel was excised and was reduced with TCEP [tris(2-carboxyethyl)phosphine], alkylated by using iodoacetamide and finally digested with trypsin. Generated peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) on a Orbitrap Velos. MASCOT searching UNIPROT_ 2011_08 was used for protein identification with a minimum of 5 peptide match and over 95% probability cut off. The results were visualized with scaffold.

For getting better sequence coverage the identified proteins were selected and searched again with an error tolerant search allowing for amino acid mutations and modifications.

Genomic DNA was isolated from young tubers of C. esculenta using Qiagen DNeasy Plant Mini kit (GmbH, Hilden, Germany) following manufacturer’s instructions.

RNA was isolated from young tubers of C. esculenta following the protocol described by Kumar et al. [19]. Diethylpyrocarbonate (DEPC) treated water was used to prepare all the chemicals and reagents. Two grams of fresh tuber tissue of C. esculenta was ground to a fine powder in liquid nitrogen using a mortar pestle and transferred to a Falcon tube. RNA was extracted by the addition of 1.0 ml 5 M NaCl, 0.5 ml 10% (w/v) SDS, 1.65 ml 1.95% (w/v) Na₂SO₃, 1.75 ml borate-Tris buffer (0.2 M, pH 8.0 containing 10 mM EDTA) and 0.1 ml β-mercaptoethanol in a sequential manner. Following vortexing the extract was incubated at 65˚C for 5 min. Then the extract was centrifuged (1800 g, 5 min, room temperature), equal volume of Tris-saturated phenol (pH 7.9) was mixed with the supernatant and centrifuged further for phase separation. The upper phase was extracted with equal volume of chloroform-isoamyl alcohol (24:1) and centrifuged again for phase separation. To 500 µl of the upper phase 450 µl of isopropyl alcohol was added in a microfuge tube and incubated at 4˚C for 1 h. RNA was pelleted at 20,000 g for 15 min at 4˚C. The RNA pellet was washed with 70% ethanol; residual ethanol was evaporated in laminar air flow and the pellet was solubilised in DEPC treated water.

cDNA was prepared from the total RNA isolated from C. esculenta tubers using SuperScript^TM III Reverse Transcriptase (Invitrogen, CA, USA) following manufacturer’s instructions.

Genome walking was performed using BD Genome Walker^TM Universal Kit (BD Biosciences Clontech, California, USA) following manufacturer’s instruction. Primers were designed from the corresponding peptide data obtained from LC-MS/MS analysis from the protein which showed best coverage (Table 1; primers PF and PR were designed from the peptide information generated from the LC-MS/MS analysis of the protein, primers

Table 1. Sequence of primers used.

5’ GSP1 and 2 and 3’ GSP1 and 2 were used for 5’ and 3’ genome walk respectively, CDSF and CDSR were used to finally clone the complete coding sequence). PCR product after purification from agarose gel using Qiaquick Gel Extraction kit (Qiagen, Hilden, Germany) was cloned into pGEMT Easy Vector (Promega, Wisconsin, USA) and nucleotide sequencing was performed using Applied Biosystems 3130 xl Genetic Analyzer at the sequencing facility of Bose Institute, Kolkata, India.

Colocasia tuber lectin purified through affinity chromatography and subsequent ion-exchange chromatography was subjected to 15% SDS-PAGE. Two bands of ~25 kDa and ~12.5 kda resolved after affinity purification (Figure 1: Panel 1, lane 1) and a single band of purified CEA (~12.5 kDa) is visible in ion-exchange chromatography purified fraction (Figure 1: Panel 1, lanes 2 & 3). Western blot analysis authenticated the purified lectin fraction by showing its cross-reactivity to anti-CEA antibody (Figure 1: Panel 2, lane 4).

The agglutination assay was performed with different concentrations of purified CEA (4 µg to 1 µg per well) on rabbit erythrocytes that demonstrated agglutination activity at all the given concentrations after the incubation time of 1 h confirming the functional activity of the purified protein (Figure 2).

From insect bioassay experiments, it was evident that the effect of CEA was more with increasing dose over a period of 72 hours [Figures 3(a)-(c)].

The LC₅₀ values of CEA calculated by Probit analysis against Aphis gossypii, Dysdercus cingulatus and Aphis craccivora were 9.98 ± 0.239 μg∙ml^–1, 16.95 ± 0.279 μg∙ml^–1 and 15.21 ± 0.274 μg∙ml^–1, respectively (Table 2).

LC-MS/MS analysis of the lectin under study shows best 3 protein hits (Figure 4). Best coverage (81%) was observed with Colocasia esculenta lectin (A5HMM7_ COLES, GenBank Accession: EF541132.1) followed by 73% coverage in case of (B5LYJ9_9ARAE, GenBank Accession: EU924066.1, [20]) which is a mannose-binding lectin from Remusatia vivipara. The third best hit with 50% coverage was found with Q39487_COLES (GenBank Accession: D16173.1 [21]) corresponding to a partial cds of a 12 kD storage protein from tubers of C. esculenta.

Complete coding sequence was derived from the corresponding peptide mass fingerprinting data of purified CEA. Subsequently the full length CEA gene was cloned following genome walking method and the 795 bp sequence was submitted to GenBank under accession No. JX435122 (Figure 5).

Comparison of the sequence information obtained from the genomic DNA and cDNA showed that they are identical and the genomic sequence is devoid of any intron (data not shown).

The alignment of 5’ upstream region with tar1 gene from C. esculenta gave a possible assumption of the putative cis-acting elements like TATA box, major and minor transcription start sites reported in case of tar1 gene by Bezerra et al. [22] (Figure 6).

From the 3’ downstream sequence putative polyadenylation motifs [23] were also found (data not shown) similar to those of tar1 gene from C. esculenta [22].

CEA vs. Aphis gossypii

(a)

CEA vs. Dysdercus cingulatus

(b)

CEA vs. Aphis craccivora