GUO Wen-fang, Kevin Yueju Wang, WANG Nan, LI Jun, LI Gang-qiang, LIU De-hu

1 Biotechnology Research Institute, Chinese Academy of Agricultural Sciences, Beijing 100081, P.R.China

2 Department of Natural Sciences, Northeastern State University, Broken Arrow, Oklahoma 74014, USA

Abstract Cotton plants are recalcitrant with regards to transformation and induced regeneration. In the present study,5-enolpyruvylshikimate-3-phosphate (EPSPS), a glyphosate resistant gene from the bacterium Agrobacterium sp. strain CP4, was introduced into an elite Bt transgenic cotton cultivar with a modified technique involving in planta Agrobacteriummediated transformation of shoot apex. Primary transformants were initially screened using a 0.26% glyphosate spray and subsequently by PCR analysis. Five out of 4 000 transformants from T1 seeds were obtained resulting in an in planta transformation rate of 0.125%. Four homozygous lines were produced by continuous self-fertilization and both PCR-based selection and glyphosate resistance. Transgene insertion was analyzed by Southern blot analysis. Gene transcription and protein expression levels in the transgenic cotton lines were further investigated by RT-PCR, Western blot, and ELISA methods. Transgenic T3 plants were resistant to as much as 0.4% of glyphosate treatments in field trials. Our results indicate that the cotton shoot apex transformation technique which is both tissue-culture and genotype-independent would enable the exploitation of transgene technology in different cotton cultivars. Since this method does not require sterile conditions, the use of specialized growth media or the application of plant hormones, it can be conducted under the greenhouse condition.

Keywords: shoot apex transformation, CP4-EPSPS, glyphosate, in planta, transgenic cotton

1. Introduction

Cotton genus (Gossypium L.) consists of 45 diploid(2n=2x=26) and five allotetraploid (2n=4x=52) species. It belongs to the tribe Gossypieae, in the family Malvaxceae(Van Deynze et al. 2009). Cotton is the world’s mostimportant textile crop to provide fiber for apparel, linen,paper, adhesive tape, and other products. Cottonseed is also one of the major oil source for consumable goods and food industry. The by-products after oil crushing is mainly used as livestock food or fertilizers (He et al. 2015). Insect pests and weed competition, however, badly affect cotton yield. Genetic modification using plant transformation technologies have significantly improved the efficiency of cotton production (Witjaksono et al. 2014). For example,genetically modified Bt cotton has been adopted globally to control the larvae of cotton bollworm, beetles, moths,and butterflies (Wu et al. 2008). Transgenic herbicidetolerant cotton has been rapidly adopted to control weed competition since it was first commercialized in 1997 (Latif et al. 2015). Genetically-modified lines have significantly reduce pesticide application, herb use, and labor costs, thus increasing profits of cotton farmers (Li et al. 2014).

Agrobacterium-mediated transformation has been the main approach used to produce transgenic cotton. This method involves an elaborate tissue culture process, and thus suffers from embryogenesis-related problems; such as chimeric variation (Keshamma et al. 2008), poor quality of somatic embryos (Wilkins et al. 2004), cytogenetic changes and phenotypic abnormalities (Stelly et al. 1989).Furthermore, most cotton cultivars are also recalcitrant,and as a result, are difficult to regenerate in tissue culture.Therefore, the approach has been limited to the Coker or Coker-related varieties. Generally, elite transgenic cultivars are produced from lengthy and labor-intensive backcrossing of transgenic Coker with other varieties (Keshamma et al.2008). Biolistic transformation has also been used to generate transgenic cotton lines (Rech et al. 2008). In mostcases, only the epidermal cell layer receives the insertion of the foreign gene. Additionally, biolistic transformation often causes multiple complex insertions resulting in instability or silencing of the desired transgene (Harwood 2012).Biolistic transformation also requires an arduous process of tissue culture and regeneration. Biolistic equipment and consumables are expensive and therefore limit its application in routine plant transformation.

Tissue culture-independent in planta cotton transformation methods have been developed to eliminate the need for regeneration backcrossing. A pollen tube pathway-mediated method for cotton transformation has been investigated(Wang et al. 2013). In this method, exogenous DNA reaches the embryo sac via a pollen tube pathway and the egg cell or zygote may then subsequently take up the applied DNA. Although various lines of transgenic cotton plants have been produced by this method (Huang et al.1999; Song et al. 2007; Liu et al. 2013), it requires highly trained technicians and approximately 3 mon from seed germination to the ability to conduct the transformation. In addition to the aforementioned complications, the timing of when the transformation is conducted is also very critical.The insertion of foreign DNA along the pollen tube pathway has to be accomplished during a narrow window of time because fertilization occurs after 24 h of pollination. This method still remains controversial and unproven.

An additional method of transforming cotton utilizing shoot apex in planta was developed by Gould and Magallanes(1998). The method is very convenient since it utilizes newly germinated seeds as the source of transformation material. Since the apical meristem is growing very rapidly,it is undergoing high rates of mitosis. This eliminates the need for callus or somatic embryo cultures and avoids the potential problems of somatic mutations and somaclonal variation that can occur in long-term tissue culture. Some transgenic cotton lines have been produced using the shoot apex transformation method (Zapata et al. 1999; Bazargani et al. 2011; Ma et al. 2013). The protocol requires the use of various specialized media for co-cultivation, selection, and rooting. However, sterile conditions are required during the apex transformation process until rooted plants are ready for transfer to soil conditions. In the present study, we use a modified in planta shoot apex transformation method. The herbicide resistance gene (CP4-EPSPS) was successfully introduced into the elite transgenic Bt cotton variety Yinkang P53.Glyphosate spray was used to identify transgenic seedlings,i.e., those surviving to the treatment. Importantly, this technique did not require sterile conditions, specialized plant media or hormones during the entire process.

2. Materials and methods

2.1. Plant materials and chemicals

Seeds of the transgenic Bt cotton variety, Yinkang P53,possessing a kanamycin-resistance gene, were purchased from Silver Land Biotech Co. (Beijing, China). Escherichia coli Trans10 competent cells and DNA gel purification and PCR kits were purchased from TransGen Biotech (Beijing,China). DNA primers synthesis and gene fragment sequencing were provided by Sangon Biotech Co., Ltd. (Shanghai,China). The LA PCR and RT-PCR kits were acquired from TaKaRa Bio (Otsu, Shiga, Japan). All other chemicals and reagents, unless otherwise stated, were purchased from BioDee Biotech Co., Ltd. (Beijing, China). Agrobacterium tumefaciens strain LBA4404, an intermediate cloning vector pTΩ4A, the plant expression vector pBI121, and the plasmid pEASY-CP4 (containing the CP4-EPSPS gene fragment)were maintained in our laboratory at –80°C ultra-low freezer.

2.2. Plant expression vector construction

The plasmid, pEASY-CP4 (TransGen Biotech, Beijing,China), was used as a PCR template to amplify the CP4-EPSPS gene fragment utilizing a pair of primers CP-F1 and CP-R1 (Sangon Biotech Co., Ltd.). All of primers were shown in Appendix A. Underlined bases indicate KpnI and XhoI sites. A Kozak sequence (AACA) was added in front of the start codon (ATG). The resulting PCR product(1.4 kb) was cloned into the TA cloning vector, pEASY-T1,and sequenced by Sangon Biotech Co., Ltd. (Shanghai,China). The confirmed gene fragment was isolated using the restriction enzymes, KpnI and XhoI, and inserted into the corresponding sites in the pTΩ4A vector to yield the intermediate vector, pTΩ4A-CP4. The Ω-CP4-EPSPS gene fragment was digested from pTΩ4A-CP4 using HindIII/EcoRI and was subsequently ligated into the pBI121 vector to obtain the final binary vector, pBI-35SCP4 (Fig. 1). The final expression vector was confirmed by restriction enzyme digestion analysis and introduced into Agrobacterium tumefactions LBA4404 using the tri-parental mating method.A single colony selected from an LB plate supplemented with kanamycin (50 mg L–1) was inoculated in 5 mL of liquid LB medium (supplemented with 50 mg L–1kanamycin) and cultured at 280 r min–1for 2 to 3 days at 28°C until the optical density (OD) 660 value reached 0.9–1.0.

2.3. In planta Agrobacterium-mediated transformation of shoot apex

The in planta shoot apex method described by Gould and Magallanes (1998) was used for cotton transformation with slight modification. Cotton seeds were germinated and grown in a greenhouse until two cotyledons were fully expanded(Fig. 2-A). Cotyledons were used for the transformation.Two-true leaves were used for the test of resistance to glyphosate. One of the two cotyledons of each germinated seed of elite Bt cotton (Yinkang P53) was removed, resulting in exposure of the shoot apical meristem (in bud) (Fig. 2-B).The shoot apex embedded between the remaining cotyledon and stem was pricked and wounded with a sterile needle to facilitate the infiltration of Agrobacterium. A sterile cotton ball was dipped in a previously prepared suspension of Agrobacterium and applied to the wounded shoot apex,thus transferring the Agrobacterium cells to the wounded shoot apex. Seedlings were maintained in the dark for 3 days and then transferred to the light for around 10 days.A lethal concentration (0.26%, Appendix B) of glyphosate was sprayed on the fully expanded leaves of treated plants(Fig. 2-C). Treated seedlings that survived the application of the initial glyphosate treatment received additional glyphosate sprays every 3 wk for a total of three applications.Glyphosate-resistant (GR) plants were transferred to larger pots to produce T1transformant seeds (Fig. 2-D).

2.4. Selection of homozygous transgenic plants

Fifty T1seeds from independent primary transformed lines were germinated and seedlings obtained planted in a greenhouse. These plants were sprayed with 0.26% of glyphosate every 3 wk for a total period of 9 wk. Genomic DNA was extracted from GR plants using the CTAB method(Paterson et al. 1993) and analyzed by PCR amplification with a pair of primers CP4-F2 and CP4-R2. The PCR condition was pre-denaturation at 96°C for 4 min, followed by 35 cycles of 95°C for 30 s, 56°C for 1 min, 72°C for 45 s, and final extension at 72°C for 5 min. PCR-positive seedlings were self-pollinated for two consecutive generations and each generation was treated with glyphosate and further analyzed by PCR. Homozygous lines were screened with glyphosate. Transgene insertions were further confirmed by DNA Southern blot. A 1.5-kt PCR probe was synthesized with DIG Necleic Acid Kit (Sigma-Aldrich, St. Louis, MO,USA). A total of 30 µg of total genomic DNA was digested with 20 U of the unique restriction enzymes on T-DNA,EcoRI and HindIII, respectively. DNA fragment separation,membrane transfer, and hybridization and detection process were performed following the kit’s instructions.

Fig. 1 The plant expression vector, pBI-35SCP4. nos P, diagrammatic representation of the plant binary vector constructed for use in Agrobacterium-mediated transformation. nos P and nos T, nopaline synthase promoter and terminator, respectively; NPTII (Kanr),neomycin phosphotransferase (npt)-kanamycin resistancegene; CaMV 35S-pro, Cauliflower mosaic virus (CaMV) 35S promoter;CP4-EPSPS (1.4 kb), glyphosate resistant gene (EPSPS) from the bacterium Agrobacterium sp. strain CP4. RB, the right border of T-DNA; LB, the left border of T-DNA.

Fig. 2 In planta Agrobacterium-mediated transformation of shoot apex in cotton. A, seedlings with fully expanded cotyledons prior to transformation. B, removal of one cotyledon from each plant and placement of a sterile cotton ball that had been dipped in a transformation medium containing Agrobacterium tumefaciens LBA4404 on the wounded shoot apex of each plant. C, transformants selected using a foliar application of glyphosate (0.26%). The plant (1, positive transgenic cotton) survived treated by 0.26%glyphosate, and other plants were negative transgenic cotton. D, glyphosate-resistant cotton plants.

2.5. RT-PCR, Western blot, and ELISA analyses

For RT-PCR analysis, total plant RNA was extracted using a RNA PCR Kit (AMV) v. 3.0 (TaKaRa Bio, Japan) according to instructions provided by the manufacturer. Subsequent cDNA amplification was performed using CP4-F2 and CP4-R2 primer pair following the manufacturer’s instructions.The ubiquitin7 gene (GhUBQ7), used as an internal control, was amplified with the primer pair GhUBQ7-F and GhUBQ7-R. The sizes of the RT-PCR products were confirmed by gel electrophoresis. Immunoblot (Western blot)and ELISA analyses were performed to confirm the presence of the EPSPS recombinant protein. A total of 0.2 g of young leaf tissue was thoroughly homogenized using a Mini Bead Beater (Cole-Parmer, IL, USA) in a 1.5-mL Eppendorf tube containing glass mill beads and 200 µL of deionized water. Homogenates were centrifuged at 12 000 r min–1for 5 min. A total of 10 µL of supernatant previously prepared was subjected to 12% SDS-PAGE analysis. Western blot analysis was performed according to the manufacturer’s protocol (Thermo Fisher Scientifc., MA, USA). Anti-mouse CP4-EPSPS antibody (Kedawe Co., Beijing, China) was used as the primary antibody. For the ELISA analysis,100 µL of supernatant was transferred to an ELISA plate(Agdia Co., IL, USA) that was designed to determine the presence or absence of CP4-EPSPS using an absorbance reading at 660 nm (OD660). The detailed steps of the ELISA reaction were instructed by the manufacturer.

2.6. Analysis of the resistance of transgenic cotton to glyphosate

In field, five concentrations (0.25 to 0.50%) of glyphosate were used for testing the resistance of transgenic plants.Seedlings with two true leaves were sprayed with glyphosate.There were 250 plants in total and 50 plants were tested by each concentrations of glyphosate. And 2 wk after testing with different concentrations of glyphosate, we madestatistics to dead plants. After 1 mon, we observed visually injury levels of plants.

3. Results

3.1. Plant expression vector and in planta shoot apex transformation

CP4-EPSPS driven by constitutive 35S promoter was successfully introduced to a plant binary vector (Fig. 1).The expression vector was transformed into A. tumefaciens LBA4404 using a tri-parental mating method. A. tumefaciensstrain LBA4404, harboring the plant binary vector pBI-35SCP4, was used to transform cotton (var. Yinkang P53)employing a modified in planta shoot apex transformation protocol (Fig. 2). When two cotyledons were fully expanded(Fig. 2-A), one of the cotyledons was removed at the base of the stem. The stem apical meristem, which was now readily accessible, was wounded and then covered with a cotton ball that had been soaked in a suspension of Agrobacterium containing the plant binary vector (Fig. 2-B). Approximately 4 000 shoot apexes were treated in this manner.

The lethal concentration of glyphosate was determined prior to the transformation studies (Appendix B). In the current study, glyphosate (0.26%) was used to screen putative transformants and their progenies. A total of 10 d after the transformation attempt, glyphosate (0.26%)was sprayed on the plants to identify transformants.Approximately 95% of the seedlings were killed by the firstround of the herbicide treatment (Fig. 2-C). The surviving plants were sprayed again with glyphosate at 20 d after the first glyphosate treatment. A total of three applications of glyphosate were applied at 20-day intervals to identify putative transformants. At the end of the screening, five glyphosate-resistant transformants were obtained and allowed to set seeds (Fig. 2-D).

3.2. PCR and Southern blot analysis of transgenic T3 seedlings

Each independent transformant produced between 130 to 200 T1seeds. Fifty seeds from each primary transformant line were subsequently planted. PCR analysis indicated that 57 to 65% of the T1plants produced the expected 1.5 kb fragment matching the size of the CP4-EPSPS gene fragment (Appendix C). These data indicated that all five of the surviving plants had been transformed with the foreign CP4-EPSPS gene. The in planta shoot apex transformation rate was determined to be 0.125%. Twenty PCR-positive T1progenies from each primary transformant were self-fertilized in successive generations to produce T3seeds that were homozygous for the CP4-EPSPS gene. PCR analysis of 20 individuals from each T3transgenic line indicated that all plants contained the foreign gene fragment (Appendix D).Two independent homozygous T3lines were confirmed single insertion after Southern blot analysis (Fig. 3). A field test of T3plants indicated that the plants were resistant to 0.26%of glyphosate when they had formed their first true leaves.In contrast, wild type, non-transformed plants became chlorotic and the shoot apical meristem was seriously injured. Glyphosate treatment (0.26%) caused the leaves to wilt and defoliate in the wild-type plants, resulting in whole plant death (Fig. 4).

3.3. RT-PCR, Western blot, and ELlSA analysis of transgenic T3 seedlings

RNA expression of CP4-EPSPS in transgenic lines was confirmed by RT-PCR analysis. While expression of the native gene GhUBQ7 was detected in both WT and transgenic lines, CP4-EPSPS RNA expression was not detected in WT plants (Fig. 5-A). Immunoblotting (Western blot) using the Anti-Mouse CP4-EPSPS antibody revealed～50 kDa bands in transgenic plants but not in wild-type plants. Both RT-PCR and Western blot results revealed varying levels of CP4-EPSPS gene and protein expression in the transgenic lines but not in wild-type plants (Fig. 5-B).CP4-EPSPS protein levels in T3homozygous lines were evaluated by ELISA (Fig. 6). Based on the absorbance values of positive and negative control, OD660values higher than 0.3 were interpreted as CP4-EPSPS positive plants,whereas OD660values less than 0.1 were considered as nontransgenic plants. OD660values in four of the homozygous transgenic lines were higher than 0.3. The ELISA results demonstrated that CP4-EPSPS protein levels were indicative of glyphosate resistance.

3.4. Analysis of resistance to glyphosate

Fig. 3 Southern blot analysis of transgenic T3 seedlings. CK+,positive control (kit); lane 1, pBI-35SCP4/EcoRI (14.3 kb); lane 2, WT/EcoRI; lane 3, WT/HindIII; lanes 4 and 5, two independent T3 homozygous transgenic cotton lines cut with EcoRI; lanes 6 and 7, two independent T3 homozygous transgenic cotton lines cut with EcoRI. EcoRI and HindIII are two single unique external sites of cp4-epsps.

Fig. 4 Glyphosate-resistance assay. A, foliar application of glyphosate (0.26%) was administered to both wild type (WT) and transgenic (T3 homozygous) cotton plants.

Fig. 5 RT-PCR (A) and Western blot (B) analysis of transgenic T3 seedlings. A, lanes 1–4, independent T3 transgenic cotton lines;lane 5, wild-type (WT) cotton; M, DNA molecular weight DNA markers. B, lane 1, purified CP4-EPSPS protein (positive control);lane 2, WT cotton; lanes 3–5, independent T3 transgenic cotton lines.

Glyphosate has detrimental effects on plant growth even it does not completely kill the plant (Fig. 7). It inhibits the aromatic amino acid biosynthesis pathway which leads to a reduction in protein synthesis and photosynthesis(Steinrucken and Amrhein 1980). Glyphosate can also form stable complexes with other metals, resulting in micronutrient deficiencies (Bott et al. 2008). We respectively made statistics about cotton plants after 2 wk and 1 mon(Table 1). When treated with 0.25 and 0.30% glyphosate(Fig. 7-A and B, respectively), plants did not show any obvious symptoms of injury relative to the untreated control plants. Application of 0.35 and 0.40% glyphosate (Fig. 7-C and D, respectively), however, significantly affected plant growth. Multiple branches emerged, resulting in dense bush.These higher concentrations of glyphosate caused leaves to yellow, reduced growth, induced the formation of lesions on the leaves, and resulted in wilted plants at the 7th day after they had been sprayed with glyphosate. These higher concentrations (0.35 and 0.40%), however, did not kill the transgenic plant lines. The plants were capable of recovering from the glyphosate sprays within 20 d after treatment butstill exhibited some alterations in growth. Transgenic lines exposed to 0.35 or 0.40% exhibited more branching. Higher concentrations of glyphosate (≥0.5%) caused leaf necrosis and ultimately had a lethal effect (Fig. 7-E).

Fig. 6 ELISA analysis of transgenic T3 seedlings. 1–4, randomly selected independent T3 homozygous lines; CK–, wild-type plant; CK+, positive control (kit). Absorbance values above 0.3 were considered to be evidence of CP4-EPSPS protein. The values indicate the mean±standard error of three replicates(n=3).

Fig. 7 Analysis of glyphosate resistance. Five concentrations (A, 0.25%; B, 0.30%; C, 0.35%; D, 0.40%; E, 0.50%) of glyphosate were used to evaluate glyphosate resistance in T3 transgenic lines. F, untreated (1, normal) and treated (2, abnormal) plants.

4. Discussion

EPSPS (5-enolpyruvyl-shikimate-3-phosphate synthase, EC 2.5.1.19) is an enzyme that catalyzes the reaction between phosphoenolpyruvate (PEP) and 3-phosphoshikimate to synthesize 5-enolpyruvylshikimate-3-phosphate (EPSP)and phosphate. The resulting product is involved in the synthesis of products utilized in the aromatic amino acid biosynthesis pathway. Therefore, EPSPS exists widely in microorganisms and higher plants. Glyphosate, a potent inhibitor of EPSPS (Steinrucken and Amrhein 1980), is one of the most widely applied herbicides in the world. In recent years, many EPSPS genes have been cloned from plants and microorganisms (Priestman et al. 2005; Kahrizi et al. 2007; Taravat et al. 2014). Constitutive expression of EPSPS in several transgenic crops has been used to convey resistance to glyphosate (Agapito-Tenfen et al. 2014;Wang et al. 2014; Liu et al. 2015). In the present study, the glyphosate-resistance gene, CP4-EPSPS, was introduced into the commercially available Bt cotton variety, YinkangP53 and glyphosate were used to identify putative transgenic seedlings. Independent transgenic lines exhibited tolerance to glyphosate and homozygous plants were obtained by several generations of self-fertilization.

Table 1 The statistics of resistance under different concentrations of glyphosate in field

The PCR results indicated that although the initial transformants were chimeric (Appendix C), the transgene was successfully transmitted to successive generations in a non-chimeric manner (Appendix D). RT-PCR results indicated that CP4-EPSPS gene expression, and Western blot and ELISA analysis confirmed that varying levels of CP4-EPSPS protein expression in the transgenic lines.The lethal concentration (0.26%) of glyphosate was used to screen putative transformants. Under normal field use,high concentrations of glyphosate are not needed and a concentration of 0.15% glyphosate, applied as a foliar spray,is used to kill weeds. In the present study, transgenic cotton plants tolerated up to 0.40% glyphosate. Therefore, the recommended dose of glyphosate (0.15%) for weed control would not result in any obvious damage to the transgenic cotton plants that were produced in this study. This study finds that high concentrations of glyphosate have an enormous effect on the normal growth and development of transgenic cotton, which makes concentration of glyphosate controlled within a certain range. The tolerance of transgenic cotton to glyphosate is not high, and the use of glyphosate will be greatly limited. Therefore we suggest that the new herbicide-resistant genes should be isolated and cloned to improve resistance of crop to herbicide.

Although the method employed in the current study was easy and simple to use, the transformation efficiency was very low (0.125%). Higher rates of transformation may be possible to achieve using a more virulent strain of A. tumefaciens, such as EHA105 or AGL1, which are strains that have shown higher transformation frequencies than other stains (Chetty et al. 2013). Acetosyringone is a wound response molecule that plays a critical role in promoting Agrobacterium infection and thus increases the efficiency of transformation of planet explants by A. tumefaciens(Baker et al. 2005). Therefore, future studies will explore adding different concentrations of acetosyringone to the transformation medium. Since a vacuum is capable of generating a negative pressure that can assist the Agrobacterium to more readily penetrate into the wounded plant tissue and cells, this method will also be explored in future modifications of this technique (Arun et al. 2015).We expect to increase the transformation efficiency of the in planta plant shoot apex transformation technique to 5% or greater by using an optimal combination of the aforementioned modifications.

5. Conclusion

The present study describes a simplified genotypeindependent method of stable genetically transformed cotton varieties that stand glyphosate-based control of weeds in the fields. In fact, CP4-EPSPS gene was successfully introduced into a cotton variety using an in planta wounded shoot apex by Agrobacterium sp. The insertion of EPSPS gene in the modified cotton was confirmed with PCR analyses and confirmed by the resistance of transgenic T3plants to glyphosate sprays. The use of this genetic transformation technique, along with modifications that may further increase transformation efficiency, provides new prospects for the transformation of recalcitrant cotton varieties for improved weed control in the field.

Acknowledgements

This work was supported by the National Biotechnology Development Plan, China (2016ZX08005-004).

Appendicesassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm