Cell Suspension Culture-Mediated Incorporation of the Rice Bel Gene into Transgenic Cotton

Cotton plants engineered for resistance to the herbicides, glyphosate or glufosinate have made a considerable impact on the production of the crop worldwide. In this work, embryogenic cell cultures derived from Gossypium hirsutum L. cv Coker 312 hypocotyl callus were transformed via Agrobacterium tumefaciens with the rice cytochrome P450 gene, CYP81A6 (bel). In rice, bel has been shown to confer resistance to both bentazon and sulfanylurea herbicides. Transformed cells were selected on a liquid medium supplemented alternately or simultaneously with kanamycin (50mg/L) and bentazon (4.2 µmol). A total of 17 transgenic cotton lines were recovered, based on the initial resistance to bentazon and on PCR detection of the bel transgene. Bel integration into the cotton genome was confirmed by Southern blot and expression of the transgene was verified by RT-PCR. In greenhouse and experimental plot trials, herbicide (bentazon) tolerance of up to 1250mg/L was demonstrated in the transgenic plants. Transgenic lines with a single copy of the bel gene showed normal Mendelian inheritance of the characteristic. Importantly, resistance to bentazon was shown to be stably incorporated in the T1, T2 and T3 generations of self-fertilised descendents and in plants outcrossed to another upland cotton cultivar. Engineering resistance to bentazon in cotton through the heterologous expression of bel opens the possibility of incorporating this trait into elite cultivars, a strategy that would give growers a more flexible alternative to weed management in cotton crops.


Introduction
Cotton (Gossypium hirsutum L.) has long been the world's most important source of natural textile fiber. Although the cotton plant is susceptible to a wide range of environmental stressors, over 180 million people worldwide depend on it for their livelihood. Many advances have been made through traditional selection and breeding programs. Biotechnologies have now been introduced into these programs. Cotton was one of the first cultivated crops to include transgene-mediated herbicide tolerance (Ht). Weed control based on the application of herbicides to eliminate susceptible weeds within a herbicide-resistant transgenic crop has been widely adopted by growers [1,2,3]. Herbicide-resistant transgenic crops have now been generated in a number of species for non-selective herbicides such as glyphosate, bromoxynil, triazolopyrimidine sulfonanilides, imidazolinones and 2,4-dichlorophenoxyacetic acid (2,4-D) [1,3]. A flexible approach is vital for any weed management program and, amongst other benefits, the development of new herbicide-tolerant crops would minimize the potential for the evolution of tolerance to the herbicides currently used with Ht transgenic lines. In this paper, we describe the novel integration into cotton of a rice gene, CYP81A6 (bel), which has been shown to confer resistance to the herbicides, bentazon and sulfonylurea.
Bentazon, known commercially as Basagran, is a benzothiadiazinone contact herbicide that controls broadleaf weeds and sedges by disrupting photosystem II electron transfer [4,5,6,7]. Resistance to bentazon in non-susceptible species is primarily based on metabolic detoxification, mediated by oxidative reactions that are catalyzed by cytochrome P450 mono-oxygenases [8,9,10]. Sulfonylurea herbicides act by inhibiting the essential enzyme, acetolactate synthase. The herbicide resistant, CYP81A6, bel gene was originally cloned from two rice bentazon lethal mutants, bel a and bel b , that were more susceptible to the herbicide than wildtype plants. The bentazon-and sulfonylurea-lethal phenotypes of the mutants were the result of defects in the ability of the plants to metabolize these xenobiotics [11].
Traditional breeding programs for commercial cotton fiber production are based on the upland cotton species, Gossypium hirsutum L. These breeding programs have led to a steady improvement in agronomic traits. However, a paucity of useful economic characteristics in the breeding stock and the availability of potentially useful genes in other organisms have led to the incorporation of genetic transformation technologies into cotton breeding programs [24,25,26]. Although several genetically modified cotton lines have been widely adopted, difficulties with the transformation and regeneration of cotton remains an impediment to its widespread adoption. A number of improvements to the transformation procedure have been published that increase transformation efficiency and reduce somaclonal variation. A reliable and effective step-by-step protocol that incorporates numerous improvements and modifications was published by Wilkins et al. (2004) [25]. We have refined the protocol for cotton transformation further to improve its efficiency and reducing the time required to develop new transgenic lines. Through the use of our protocol, we have developed transgenic cotton lines heterologously expressing the rice bel gene, that confers resistance to the herbicide, bentazon. This provides the potential for further flexibility for weed management in cotton crops.

Plant Materials
G. hirsutum L. Coker 312 seeds were de-coated and sterilized by dipping in 70% (v/v) ethanol prior to a 10 min exposure to 0.1% (w/v) HgCl 2 . They were subsequently rinsed in sterile distilled water and germinated on 1/2 MS (Murashige & Skoog medium) [27] medium with 10% (w/v) glucose and 0.25% (w/v) Phytagel (Sigma, USA) at 28uC in the dark for 3 d and transferred to the culture room (28uC, 14-h photoperiod, irradiance of 135 mmol m -2 s -1 provided by cool-white fluorescent lamps) for 4 days.

Bacterial Strains and Plasmids
The Agrobacterium tumefaciens-disarmed helper strain LBA4404 [29], harbouring the plasmid pC450-2 that carries p35SCYP81A6 and the nptII gene as the selectable marker and was used for transformation and regeneration experiments. CYP81A6 was cloned from rice by Pan et al (2006) [11] (Figure 1) and had been shown to confer resistance to the bentazon and sulfonylurea herbicides. The binary vector was transferred to Agrobacterium by the heat-shock method [30]. The transformed Agrobacterium was grown on LB medium plates containing rifampicin (25 mg/l) and kanamycin (50 mg/l) (Sigma, USA) (LB medium [31]: Bacto Tryptone 10 g/L, Bacto Yeast Extract 5 g/L and NaCl 10 g/L). Five single colonies were inoculated individually in 20 ml LB medium containing rifampicin (25 mg/l) and kanamycin (50 mg/ l) in flasks and these were grown for 36 h at 28uC with 200 rpm shaking. Another 10 ml of MGL medium (Tryptone 5 g/L, NaCl 5 g/L, MgSO 4 .7H 2 O 0.1 g/L, KH 2 PO 4 0.25 g/L, Gly 1.0 g/L, pH 5.8) was added to the Agrobacterium suspension for Agrobacterium activation on a shaker at 200 rpm to A6 00 1.0-1.5 for 2-3 h at 28uC in a 100 ml flask. Just prior to co-cultivation, the Agrobacterium were diluted by fresh MGL medium to OD 600 to 0.8-1.0. Additional acetosyringone (AS) (pH 5.8) was added to the culture at a final concentration of 100 mM.

Co-cultivation
One, 15, 30, 45 and 60 day-old suspension cultures were prepared for infection by Agrobacterium. Friable embryogenic calluses was cultured in liquid MS medium for one day was also prepared for infection and used as a control. 3-4 day old postsubculture embryogenic cell suspension cultures growing on liquid MSB medium with IBA (0.5 mg/l) were used for transformations. The process of Agrobacterium-mediated transformation of suspension cultures is shown in Figure 2. Somatic cell suspension cultures were collected by sterile pipette and sieved through a 100 mm filter to remove large cell clumps and then desiccated for 15-20 min. The suspension cultures were incubated with Agrobacterium culture LBA4404 (OD 600 0.8-1.0) for 30 min. The suspension cultures were then transferred onto sterile filter paper and placed on the solid co-culture medium (CM medium, MSB medium, 0.25% (w/ v) phytagel, 3% (w/v) Glucose) for 36 to 48 hours in the dark at 21uC. 1 ml of 50 mg/L As was added to the sterile filter paper. Three flasks of suspension cultures with different culture times (1, 15, 30, 45 and 60 day, respectively) were co-cultivated with Agrobacterium with OD 600 value of 0.8-1.0. The co-cultivation with cultures of different ages was repeated three times and the number of bentazon-resistant colonies per flask was determined.
After co-cultivation, the cells along with the filters were transferred to sterile water supplemented with cefotaxime (200 mg/l) (Sigma, USA) and washed 2-3 times. The cells were then cultured in liquid MSB medium supplemented with cefotaxime (500 mg/l) for 3-4 days. The liquid medium was then removed and the same liquid medium supplemented with kanamycin (50 mg/l) and bentazon (3 mM) (Sigma, USA) was applied. The cultures were then incubated at 28uC at 110 rpm on a rotary shaker. The liquid medium was changed every 7 days for 4 times to eliminate the false positive cells and proliferate the resistant cells, when no more Agrobacterium contamination appeared, the cefotaxime was eliminated from the medium. The suspension cells were then transferred to solid MSB medium supplemented with kanamycin (50 mg/l) and bentazon (4.2 mM) for proliferation.  Mature embryos (cotyledon embryo and torpedo embryo) were transferred onto the filter paper on the DM medium for germination for 30 days. After root and shoot elongation, the germinated embryos were transferred to solid medium (MSB, 3% (w/v) Glucose, 0.25% (w/v) Phytagel, IBA 0.5 mg/L, KT 0.1 mg/L) to conversion into plantlets with normal roots. Then the plantlets were planted into pots containing a 1:1:1 mixture of vermiculite, top soil, and peat, and covered to increase humidity. Plantlets were gradually exposed to ambient humidity over a two week period and then placed in the greenhouse. Regenerated plants with minimal root systems were grafted onto wild-type Coker 312 rootstock material.

Screening and Analysis of Transgenic Plants
PCR and southern blot analysis. Cotton genomic DNA was extracted from young leaves using methods described by Paterson et al. (1993) [32]. Polymerase chain reaction (PCR) and Southern blotting were used to confirm the presence of the bel gene and the copy number in transgenic plants. PCR analysis for detection of the nptII gene was performed using the primers 59-TCG GCT ATG ACT GGG CAC AAC AGA-39 (forward) and 59-AAG AAG GCG ATA GAA GGC GAT GCG-39 (reverse) that amplify a section of the nptII gene coding region. PCR analysis for detection of the bel gene was performed with the primers 59-GAA GTT CAT GCC GGA GAG-39 (forward) and 59-ATT GCG GGA CTC TAA TCA TA-39 (reverse). PCR was performed in 25 ml reaction mixtures consisting of 10 6 reaction buffers, 50 ng DNA templates, 15 mM MgCl 2 , 10 mM dNTPs, 50 ng of each primer and 1 unit Taq DNA polymerase in an ABI thermal cycler (Veriti 9902, USA) using the following conditions: initial denaturation at 94uC for 2 min followed by 35 cycles of denaturing at 94uC for 1 min, annealing at 55uC for nptII and   annealing at 55uC for bel gene for 1 min, extension at 72uC for 3 min and final extension at 72uC for 5 min. PCR products were analyzed by gel electrophoresis on 1% (w/v) agarose gels. Southern blot analysis was carried out according to Sambrook et al. (1989) [33] using the DIG DNA Labeling and Detection Kit (Roche, Switzerland) according to the manufacturer's recommendations. Plasmid DNA was used as a positive control and nontransgenic Coker 312 was used as a negative control. Approximately 10 mg of total genomic DNA was digested with EcoR I at 37uC overnight in a total volume of 100 ml and separated by electrophoresis on 0.8% (w/v) agarose gel at 30 V for 8 h. DNAs were blotted onto charged nylon membranes (Hybond N + , Bedford, MA01730) and hybridized following manufacturer's instructions of Roche kit. PCR-amplified bel fragments were used as a probe with boitin-labelling with the DIG Detection Kit (DIG DNA Labeling and Detection Kit, Roche, Switzerland).
In 2007, T1 transgenic cotton plant families derived from selffertilized T0 transgenic plants were grown in a greenhouse at the Huajiachi Campus of Zhejiang University in Hangzhou, China. Bentazon (1250 mg/L) was sprayed during the seedling stage, to test for herbicide resistance of the transgenic plants. Strongly herbicide resistant lines with normal gross morphological phenotypes were self-fertilised. The copy number of bel transgenes inserted was determined in the T1, herbicide tolerant lines using Southern blot analysis. Mature seeds were harvested from individual high-Bentazon-resistant plants with a single-copy insertion. Homozygous transgenic cotton lines were selected by molecular and genetic analysis of T2 lines in 2008 to get T3 seeds. T2 homozygous transgenic cotton plants were crossed with an elite cultivar for hybrid selection and testing with Bentazon sprays in the experimental plot.
RT-PCR analysis. In order to test for the expression level of the bel transgene, total RNA was isolated from young leaves of untransformed control and transformed single copy T2 and T3 plants from using the RNeasy Plant Mini Kit (Qiagen, Germany) following manufacturer's instructions. cDNA obtained from this RNA (Eppendorf, Germany) was then used as template for PCR amplification of bel transcripts using the primer pair: 59 AGAAGAAGAGCATGATCGCC 39 (forward), 59 TGGTTCAGCAGTAGCGACAT 39 (reverse). PCR conditions were: 94uC for 5 min for initial melting followed by 35 cycles of amplification with each cycle consisting of the following steps: 94uC for 1 min, 56uC for 1 min and 72uC for 1 min 15 s with a final extension at 72uC for 10 min. Primers specific to the actin gene were used as internal control: 59 CCGATGCCTTGATGAAGATT 39 (forward), 59 GCAGTCTCCAGTTCCTGCTC 39 (reverse). PCR products were analyzed by gel electrophoresis on 1% (w/v) agarose gels.

Generation of Herbicide Resistant Transgenic Cotton Plants
Although glyphosate, glufonidase and a number of other herbicides have previously been used to effectively control dicot and monocot weeds in cotton crops, it is desirable to develop additional transgenic cotton varieties resistant to other herbicides, as no one herbicide is sufficient for an effective weed control program. In addition, the availability of transgenic lines with tolerance or resistance genes to different herbicides can reduce the danger of developing acquired herbicide resistance in weed species. We have used an improved transformation system to develop transgenic cotton lines resistant to the herbicide bentazon through the incorporation of the rice Bel gene into the cotton genome. Bel is a cytochrome P450 gene, originally identified from work conducted on bentazon and sulfonylurea-lethal rice mutants [11].
Using our protocol, embryogenic callus and embryos were formed from the cell suspension cultures after 4 weeks. After cocultivation with Agrobacterium, transformed cells were selected and putative positive colonies were transferred onto solid medium (Figure 3 A) containing 50 mg/L kanamycin and 4.2 mM bentazon. Many of the cell masses proliferating on the kanamycin and bentazon selection media browned and died. However, in many instances, yellow living calluses formed from the browning cell mass. These were subcultured onto fresh selection media (Figure 3 B, C) and 5 to 6 weeks after the co-cultivation with Agrobacterium, multiple distinct living colonies were obtained. A total of 504 transformed colonies were obtained from four of the suspension culture treatments, with frequencies of 3.7% in the 1day suspension cultures, 10.3% in the 15-day cultures, 10.6% in the 30-day cultures, 18.6% in the 45-day n cultures, and 18.1% in the 60-day cultures, indicating a significant improvement in transformation efficiency between 30 and 45 day cultures (Table 1). After 4 weeks of cell proliferation and differentiation on the selective media, many somatic embryos appeared on the surface of the resistant calluses (Figure 3 C, D, E). Mature embryos were transferred to fresh solid media for conversion into plantlets via In order to verify that the colonies were comprised of transgenic cells, PCR testing was undertaken for the presence of the the nptII and bel gene. Material from 17 of 29 lines tested was confirmed as containing the npt II and bel genes incorporated in the cotton genome (Figure 4 A, B). The 17 positive lines were subsequently carried forward through the cell proliferation, somatic embryogenesis and plant regeneration processes. Twenty-six primary (T0) transgenic cotton plantlets were obtained from the 17 lines and re-confirmed by PCR analysis as containing the bel and nptII transgenes. The regenerated, transgenic cotton plants were grafted as scions onto Coker 312 rootstocks. After the graft union had formed, the grafted plants were sprayed with the herbicide (Bentazon 500 mg/ L) on the leaves of the scion and the cotyledons of the rootstock. The leaves of transgenic scion remained green, whereas the rootstock cotyledons wilted, indicating that the rootstocks retained sensitivity to the herbicide, while the transgenic scion maintained its acquired resistance (Figure 5 A). Grafted transgenic plants (T0) were transferred to pots in the greenhouse for further growth and self-fertilization (Figure 5 B).
The grafted T0 transgenic lines were allowed to self-fertilize and the resulting T1 plants were tested by spraying with bentazon (1250 mg/L). Surviving plants were classed as putatively bentazon-resistant. These were screened by PCR for the presence of the nptII and bel genes. The PCR screen confirmed that all of the tested plants were positive for the transgenes. Finally, the T1 plants were examined by Southern blot for determination of transgene copy number. The Southern blots indicated that 7 lines out of the 17 tested had one copy of the T-DNA inserted into the cotton genome. Other lines were shown to have two or three copies, and four of the lines showed no signals on the Southern blot (Figure 4 C). The 7 lines with a single copy of the bel gene were further selffertilized to obtain T2 seeds. Resistant seedlings with one copy of the bel gene appeared at a ratio of approximately 3:1 (Figure 5 C). Similarly, the T2 plants were self-fertilized to obtain T3 seeds. Testing of T2 and T3 plants showed that the bentazon resistance was carried through to these subsequent generations. RT-PCR analysis was also conducted on T2 and T3 plants from the same lines. RT-PCR results showed that all lines had high levels of expression of the bel gene (Figure 4 D). Importantly, hybrids between the T2 transgenic cotton lines and an elite varieties retained resistance to bentazon. Seedlings of the T2 and T3 generations of transgenics and wild-type Coker 312 were sprayed with Bentazon (1250 mg/L) for two times 3 days later at the two true leaf stage of development in the field. All of the Coker 312 plants subsequently wilted and died. The T2 and T3 seedlings, however, continued to grow normally ( Figure 5 D). Similarly, the F1 hybrid seedlings between the transgenic and the elite lines retained their resistance in the face of the same herbicide treatment used for the T2 and T3 transgenic cotton lines.
In this study, we further refined the procedure for Agrobacteriummediated cotton transformation, using suspension cell cultures as explants to produce transgenic cotton lines with resistance to the herbicide, bentazon. Overall, the process of transformation, from suspension cultures to transgenic cotton plants took approximately 3-5 months. Previously published cotton transformation protocols take 6-10 months for cotton in order to obtain transgenic plants [25,34]. Wilkins et al. [25] published a detailed protocol using hypocotyls as explant material for co-cultivation with Agrobacterium, the protocol requires approximately 8-10 months for the production of transgenic plantlets [25]. Our improvements to the published protocols therefore make the process less time consuming. One of the difficulties has been that many of the previously published protocols have proven difficult to replicate in other laboratories. In our protocol, cell suspension cultures are used as explants material for transformation and for the selection of transgenic clones because it has the advantages of: improving the survival of transgenic colonies; decreasing the percentage of false positive resistant calluses; and reducing the occurrence of deleterious somaclonal variations that are common in material derived from solid selection media. Certainly, our method is efficient in terms of the number of transgenic lines produced. It is known that in cell suspension cultures, many of the single cells and small cell masses become growth stage synchronous. It is possible that the success of our method is at least partially due to a uniformity of cell physiology because of this synchrony [35,36,37]. We believe that widespread adoption of our cotton transformation protocol has the potential to significantly accelerate the generation of transgenic lines with other agronomical useful traits.
We also expect that the bentazon resistance incorporated in the lines developed in this study will be incorporated into elite line breeding programs and will eventually give growers a useful additional tool in their arsenal.