Agrobacterium-mediated Transformation of Cotton (Gossypium hirsutum) Using a Synthetic cry1Ab Gene for Enhanced Resistance Against Heliothis armigera

INTRODUCTION

Cotton (Gossypium hirsutum L.) is the world’s most important source of natural fiber with approximate annual plantation of 35 million ha worldwide (Wilkins et al., 2000). In recent years, genetic engineering of plants has facilitated the production of agronomically desirable crops that exhibit increased resistance to pests, herbicides, pathogens, and environmental stresses as well as enhanced quality and quantity. Cotton has attracted much interest in the field of gene transfer with the aim of introducing agronomically interesting new traits (Wilkins et al., 2000; John and Keller, 1996). The level of losses due to insect damage is such that cotton protection represents approximately 24% of the global insecticide market (Chen et al., 2002). Cotton Bollworm (Heliothis armigera) is one of the most important insect pests limiting crop yield under certain environmental conditions that include the cotton growing areas of Iran.
 Cotton breeding for insect resistance has been limited by a lack of sufficient genetic variation in the existing germplasms. Therefore, efforts to improve productivity by genetic engineering of cotton plants are critical in enhancement of resistance to insect pests. The advent of genetic engineering approaches to insect resistance in crop plants now raises the possibility of achieving high levels of resistance to bollworm in cotton. Lepidopteran insects are sensitive to many of the Cry1 and Cry2 proteins produced by different Bacillus thuringiensis strains (Hofte and Whiteley, 1989). Cry1Ab is among the proteins known to be toxic when incorporated into artificial diets supporting bollworm growth in culture. These proteins are characterized by their safety with respect to non-target insects, vertebrates, and biodegradability (Kuiper et al., 2001).
 In the year 2007, from 114.3 million ha of genetically engineered crops, 23.0 million hectares were devoted to Bacillus thuringiensis (Bt) crops and 14.1 million hectares to crops containing both insecticide (Bt) and herbicide tolerance traits. Biotech cotton occupies 15.1 million hectares equivalent to one third of the global area of cotton cultivation and 13.2% of the global biotech crop area (James, 2007). Several laboratories have been able to transfer synthetic cry1Ab and cry1Ac genes into plants and develop transgenic crop with enhanced resistance to insects. These include broccoli (Xiang et al., 2000), rice (Ye et al., 2000; Ghareyazi et al., 1997), corn (Koziel et al., 1993), potato (Davidson et al., 2002), soybean (Walker, 2002) tobacco (Shi et al., 1999) and cotton (Wu et al., 2006; Guo et al., 2003; Chen et al., 2002; Perlak et al., 1990). Due to worldwide cultivation of Bt transgenic cotton plants, there has been a considerable decrease in the use of chemical insecticides, and as a result, reductions in environmental pollution and operator exposure to toxins have been achieved (Bennett et al., 2004). Bt cotton cultivation has increased dramatically in the recent years. For example in India, an area of approximately 3.8 million ha has been devoted to Bt cotton (James, 2007).  As a result of Bt cotton cultivation in India, there has been a 78.8% increase in the due to yield and a 14.7% reduction in pesticide cost as compared to non-Bt cotton (APCoAB, 2006).
 Among the advantages of this approach are: (1) the availability of several diverse mechanisms of resistance including crystal (Cry) proteins from the soil bacterium B. thuringiensis (Gould, 1996) and proteinase inhibitors from plants such as cowpea and soybean (Zhao et al., 1998; Roush, 1994); (2) the ability to introduce one or several such genes directly into popular cultivars without the disadvantages associated with sexual hybridization; and (3) the availability of three different methods of cotton transformation,  i.e. biolistic transformation (Rajasekaran et al., 2000), Agrobacterium-mediated transformation (Tohidfar et al., 2005; Zapata et al., 1998) and a combination of both methods (MaJeed et al., 2000).
 Cotton is a major cash crop in the North east of Iran. Cotton bollworm is an important pest insect and is considered as a major pest in the cotton growing areas of the Golestan province. As Iranian cotton varieties have low regeneration potential and the Coker variety responds very well to regeneration and gene transfer, most of the desirable genes are initially introduced into latter and then back crossed into other varieties. The objective of this study was to transform cotton var. Coker with a synthetic cry1Ab gene using the Agrobacterium system for enhancing resistance to Lepidopteran insects.

MATERIALS AND METHODS

Plasmid construction: A 1.8 kb cry1Ab gene from pGEM-4z (Kindly provided by Professor I. Altosaar, University of Ottawa, Canada) was subcloned into the BamHI site of pBI121 (Clontech, USA) yielding pBI121-Cry1Ab. Gene orientation was verified by double restriction digestion (EcoRI and Hind III). The expression of the cry1Ab gene was under the control of the constitutive CaMV 35S promoter (Fig. 1). The plasmid was transfered into competent cells of Agrobacterium tumefaciens (strain LBA 4404) by the freeze-thaw method (An, 1987).

Plant material and transformation procedure: Seeds of Gossypium hirsutum var. Coker were provided by the Cotton Research Institute of Iran, Gorgan, Golestan province. Cotton hypocotyl sections (0.5 cm2) obtained from 7 to 10 days old sterile seedlings were dipped into Agrobacterium cell suspensions grown to the late log phase (OD600 nm = 0.6-0.8). Explants were gently shaken in the bacterial suspension for 5 seconds to ensure contact of all hypocotyls edges with bacterial cells. The hypocotyl pieces were blot dried and placed on a Whatman No. 1 filter paper and were subsequently transferred to  MS medium (Murashige and Skoog, 1962) co-cultivation medium followed by incubation for 2 days at 26 ±2ºC in the dark (Tohidfar et al., 2005). After co-culture, hypocotyl pieces were transferred to callus induction medium (MS1) composed of MS salts supplemented with vitamin B5 and containing 0.75 g/l of  MgCl2, 30 g/l of glucose, 0.46 µM kinetin, 0.45 mM (2,4-Dichlorophenoxyacetic), pH 5.9, solidified with 0.2% (w/v) phytagel (Sigma,USA). Fifty mg/l of kanamycin and 200 mg/l of cefotaxime were also included for selection. Plates were incubated at 28ºC with a 16- h photoperiod (90 µmol/m2/s). After 3-4 weeks, calli were excised from original explants and  transferred to fresh MS medium containing kanamycin. After another 2-3 weeks, calli were placed onto embryo maturation medium (MS2), composed of MS salts containing 0.75 g/l of MgCl2, 1.9 g/l of KNO3, 0.25% (w/v), 30 g/l of glucose, pH 5.8 and supplemented with vitamin B5 and 25 g/l of  kanamycin for selection. Mature embryos were picked up and transferred to embryo germination medium (MS3), composed of MS salts supplemented with vitamins B5, 0.45 mM zeatin, 30 g/l of sucrose and 25 mg/l of kanamycin (Zhang et al., 2001; Kumar and Pental, 1998). Germinated somatic embryos were placed in 500 ml jars containing MS4 medium composed of MS salts supplemented with 100 mg/l of myo-inositol, 0.5 mg/l of  thiamin-HCl, 0.5 mg/l of nicotinic acid, 0.5 mg/l of pyridoxine HCl, 3% (w/v) Sucrose and 0.15% (w/v) Gelrite (Sigma, USA), pH 5.7. Rooting medium did not contain kanamycin in order to allow formation of roots on the plantlets (Zapata et al., 1998). Individual rooted plantlets were transferred to pots containing a (1:1 v/v) mixture of sterile soil and sand in plastic cups covered with polyethylene bags.
 Transgenic cotton Lines were successfully self-pollinated, leading to the production of T1 and T2 seeds. Since the aim of this investigation was to produce homozygous pure lines, at least in the T2 and later generations, analysis of T1 lines was not performed but the T1 and T2 plants were self-crossed and subjected to molecular analysis.

Polymerase chain reaction (PCR): PCR was carried out using specific primer pairs to amplify the nptll, and cry1Ab transgenes from transgenic cotton plants. The sequences of the primer pairs used in this assay were as follows: nptII 1: 5´-GAA CAA GAT GGA TTG CAC GC-3´, nptII 2: 5´-GAA GAA CTC GTC AAG AAC GC-3´, and cry1Ab F: 5´-AGG AAG TTC ATT CAT TTG CAG-3´cry1Ab R: 5´-TAA CTT CGG CAG GCA CAA AC-3´.
 Genomic DNA was extracted and purified from immature leaves based on the protocol of Li et al. (2001). PCR was performed in a total reaction mixture volume of 25 ml consisting of 1X reaction buffer, 15 ng of DNA template, 1.5  mM MgCl2, 1 mM of each of the dNTPs, 60 ng of each primer and one unit of Taq DNA polymerase (Cinagen Co., Iran).
 PCR was carried out in a thermal cycler using the following conditions: initial denaturation at 94ºC for 4 min followed by 35 cycles of denaturation at 94ºC for 1 min, annealing at 55ºC (for the nptll) and 60ºC (for the cry1Ab gene) for 1 min, extension at 72ºC for 3 min, followed by a final extension at 72ºC for 5 min. Amplified DNA fragments were electorphoresed on 1.0% (w/v) agarose gel and visualized by ethidium bromide staining under ultraviolet (UV) light.

Southern blot hybridization: Fifty μg of DNA extracted from young leaves was completely digested with BamHI and EcoRI at 37ºC, overnight. Digested DNA fragments were separated on 0.8% (w/v) agarose gels at 30 V for 8 h. DNA was transferred onto a nylon membrane (Hybond N+, Amersham, UK) by capillary blotting. Coding sequence of the cry1Ab gene (BamHI fragment ~2.0 kb) was labeled with the DIG DNA labeling kit (Roche) and used as a probe. Detection was carried out using the DIG Detection Kit and according to the manufacturer’s instructions (Roche, Germany).

Western blot analysis: Segments of leaves of transgenic and control plants were ground to a fine powder with a mortar and pestle in liquid nitrogen. Soluble proteins were extracted with 1 ml of extraction buffer [40% (w/v), SDS, 5% (v/v) 2-mercaptoethanol, 20% (v/v) glycerol, 68 mM Tris- HCl (pH 6.8)]. Ten μg of protein from each sample was fractionated by 13% (w/v) SDS-polyacrylamide gel electrophoresis, as described by Laemmli (1970). Western immunoblot analysis of the cry1Ab gene was performed as described by Ghareyazie et al. (1997). After transferring the proteins onto a Hybond-C membrane (Amersham, England) by a semi-dry trans-electroblotter (Sigma Co., UK), the membrane was probed with the anti-Cry1Ab anti-serum (1:2000 v/v) (a gift from Prof. Altossar, University of Ottawa, KIN 6N5 Canada). The goat-anti rabbit IgG alkaline phosphatase conjugate (1:2000 v/v) (Gibco, USA) was used as a secondary antibody. Total soluble protein was measured using the Bradford method (1976).

Insect Bioassays: Adult cotton bollworms (Heliothis armigera) were collected from the cotton fields of Gorgan, Golestan province. The moths were caged on cotton plants. Egg masses were collected from the plants and held in plastic cups. One day prior to the expected egg hatch, the egg masses were placed in vials with artificial insect diets to provide food for the early hatching larvae. Neonate larvae were collected from the vials and used in the bioassay, in the entomology testing chamber. The temperature in this chamber was maintained at 25-28ºC and 85% humidity.
 Prior to the blooming stage, fully expanded young leaves on the top of the transgenic cotton plants were used for the cut leaf assay. This experiment was performed in 20 cm Petri dishes in three replications. The leaves were infested with five larvae and the plates were sealed with Parafilm and were kept in 25-27ºC in entomology testing chamber. The numbers of live and dead insects were recorded 3 and 7 days after infestation, and the weight of live insects were measured on the 7th day. The mortality and degree of the leaf damage was measured according to a previously described method (Gallie et al., 1988).
      Mortality% = Number of dead larva/total Number of larva placed in Petridish × 100. Degree of leaf damage: little or no damage; 1: mildly damaged (10-20%); 2: moderately damaged (20-40%); 3: severely damaged (40-70%); 4: completely destroyed (70-100%).
 Transgenic cotton lines were successfully self-pollinated and T1 and T2 seeds were produced as a result. Twenty T1 seeds of each PCR- positive T0 plant showing a higher resistance against insect pests with mortality higher than 80%, were collected for inheritance analysis and selection of the homozygous positive plants.

RESULTS

Transformation and selection: Kanamycin- resistant microcalli (0.5 mm in diameter) emerged at the wounded sites of hypocotyl segments co-cultivated with Agrobacterium 3 weeks after incubation on MS medium (Fig.  2A), while no callus was obtained from the untransformed explants. The transformed calli were placed and maintained on the embryogenic selection medium. The calli produced light yellow-colored granular embryogenic structure, which eventually developed into somatic embryos (Fig. 2B). Germinated embryos were transferred to Jars on MS4 medium.
 The plantlets were regenerated after two to three months (Fig. 2C). Twelve putative independent transgenic plants with well developed leaves and root systems were transferred to soil and left to flower and set seeds under green house conditions (Fig. 2D).

Integration of the cryIA(b) gene: Twelve randomly selected T0 plants exhibiting some degree of kanamycin resistance were analyzed by PCR. Ten of these plants yielded a single DNA fragment of 785 bp for the nptII gene (Fig. 3). All 10 transformants were also positive for the cry1Ab gene, as indicated by the amplification of the predicted 1.8 kb internal fragment of the cry1Ab gene. Untransformed cotton plants were negative for both nptII and cry1Ab genes (Fig. 4a). Transgenic cotton lines were successfully self-pollinated, thus setting seeds. Similar PCR products obtained from the T0 plants were also amplified in their T1 progenies (Fig. 4b).
 Figure 5 shows southern blot analysis of T0 and T1 transgenic lines (31, 57 and 61). Analysis of DNA blotting established that plants 31, 57 and 61 were independent transformants. The 600 bp PCR product of the coding sequence of the cry1Ab gene was used as probe. Hybridization to undigested DNA was occurring exclusively at high molecular weights, indicating the integration of the gene into the cotton genome. When the DNA was digested with BamHI, most of the hybridization was to an expected fragment of 2.0 kb that included the whole coding sequence of cry1Ab gene and the nos terminator. This result shows that at least one intact copy of the integrated cry1Ab gene is present (Fig. 5a). When EcoRI digested DNA was hybridized to the probe, only one band with different sizes (~4.5 kb for line 31, larger than 12.0 kb for line 57 and ~3 kb for line 61) was detected in each of the different transgenic lines examined. Since there is only one EcoRI site in the T-DNA, this result indicates the presence of only one single copy of the transgene in each of the three independent transgenic events (Figs. 5b and 5c). No hybridization signal could be detected for the DNA extracted from the untransformed plants.

Expression of the cryIA (b) gene: For technical reasons immunoblot analysis was not carried out for the T0 generation. Immunoblot analysis of the T2 transgenic plants (lines 61 and 57) and the untransformed control showed high levels of Cry1Ab protein production in both plants 61 and 57 (Fig. 6). The size of the Cry1Ab protein accumulating in leaves of T2 transgenic plants (lines 61 and 57) was 67 kDa. Untransformed control did not show any positive signal for Cry1Ab protein.

Bioassay of T0 plants with cotton bollworm: Bioassay was carried out on transgenic cotton plants by using larvae of the cotton bollworm. After being infested for 7 days, lines 61 and 57 proved to be the plants on which all recovered larvae were dead. The dead larvae possessed  black head capsule and pronotum, characteristic of the first instar. The larvae recovered from untransformed control plants were alive and relatively larger. They had developed into the second instar and caused considerable damage.

Bioassay of T2 plants with cotton bollworm: Transgenic cotton lines were successfully self- pollinated, thus setting seeds. The seeds were indistinguishable in appearance from non-transgenic seeds.
 Table 1 shows the development (mortality and degrees of damage) of larvae fed on the transgenic lines. The plants were infested with first instar cotton bollworm larvae and dissected 7 days after infestation. All larvae recovered from the Cry1Ab-positive T2 plant (plant 61) were dead and had not progressed beyond the first instar (Fig. 7). Progeny analysis by PCR showed that the T2 line 61 was homozygous for the cry1Ab gene and the T1 lines 31 and 57 were hemizygous for the cry1Ab gene (data not shown). By contrast, all larvae recovered from the control plants were alive and had progressed to the second or third instar (Fig. 8). The majority of larvae on the Cry1Ab positive plants was not recovered and is presumed to have died and decomposed or to have dispersed from the plants. In contrast, all recovered larvae fed on non-transgenic control plants were alive.
 As judged by the damage rate and mortality of the recovered larvae we conclude that line 61 was a transgenic line which could be further studied for commercial applications since they were significantly resistant against the studied insect pest.

DISCUSSION

Transgenic cotton plants using the transformation system as described previously (Zhang et al., 2001; Kumar and Pental, 1998; Zapata et al., 1998) were readily obtained in this investiagtion. Also, in this study, it was observed that the presence of kanamycin was disadvantageous to rooting, so the use of antibiotics was abolished during rooting and plant regeneration. Rapid callus initiation (on MS medium containing 0.46 mM kinetin and 0.4 mM 2,4-Dichlorophenoxyacetic) was critical for the recovery of a high number of transformed microcalli at the periphery of the inoculated tissues. This could be due to a better recovery of transformed cells on this medium and high competency of dividing cells for transformation with A. tumefaciens, or both (Firoozabady and Galbraith, 1984 and 1983).
 In this research, by using the hypocotyl as an explant for inoculation (Firoozabady et al., 1987), only a few untransformed calli proliferated rapidly. These tissues were highly chimeric due to lack of complete contact between explants and kanamycin containing medium. Excision of calli from the explants was essential for promoting growth of calli and avoiding Agrobacterium contamination. Small sized calli had a lower rate of survival. This situation is believed to be similar to the requirement for a critical minimum cell density, as reported for the growth of cells or protoplasts (Shneyour et al., 1984). Glucose was used as a sole carbon source for callus induction and embryo maturation since sucrose encouraged production of phenolics by cotton explants. Using rooting powder along with liquid MS medium was found to be more effective for root induction. Of the 12 putative the T0 plants infected by Agrobacterium at different periods and analyzed by PCR, 10 were found to be positive, indicating that the rate of escape occurred during the selection procedure of our experiment.
 Thirty transgenic T0 plants were produced from 50 different inoculations (data not shown). Of these, two lines were analyzed in detail and shown that they were independent events and were significantly more resistant against cotton bollworm. The transgenic lines 61 and 57 were normal in appearance and fertile, but a few of the other independent events were male sterile (data not shown).
      Majority of lines that gave positive results for PCR analysis were further confirmed by southern hybridization (data not shown). In most of the analyzed plants, only one copy of the transgene was integrated into the cotton genome. The different sizes of hybridization signals also indicated that they resulted from the stable T-DNA integration into the cotton genome and not from endophytic Agrobacterium contamination. Lines 61 and 57 had integrated one copy of the cry1Ab gene into their genome. However, presence of the expected hybridization signals in the majority of the transformed plants showed that the probed gene and coding sequence of cry1Ab remained intact when integrated into the cotton genome.
 Lines 61 and 57 were also the plants that produced sufficient Cry1Ab protein to be detected by immunoblots. Immunoreaction of the protein blot with the Cry1Ab antiserum produced a specific band with a molecular weight of 67 KDa.
      Feeding assays with T0 plants from two independent transformants confirmed that the Cry1Ab protein produced in the transgenic Line 61 was highly toxic to cotton bollworm. The larvae began to die 1 or 2 days after feeding on the transgenic leaf tissues. A mortality of 100% was reached 4-5 days after infestation. The Cry1Ab-positive T1 plants showed enhanced resistance to cotton bollworm. Differences in larval mortality and development between the Cry1Ab-positive and control plants were generally apparent 3-4 days after infestation.
     Taken together, these observations suggest that the toxin levels in cotton line 61 (T2 generation) are sufficient to confer a high degree of cotton bollworm resistance. Transgenic plants are currently being grown in the greenhouse and will be crossed with the Iranian cotton cultivars for the purpose of introgression of the cry1Ab gene.
 
Acknowledgments

We wish to thank Prof. lllimar Altossar, for providing us the anti-Cry1Ab anti-serum. The pGEM-4z plasmid (containing the cry1Ab gene) was kindly provided by the Department of Biochemistry, University of Ottawa, KIN 6N5 Canada, for the purpose of sole scientific research.