Introduction

Traditional method of Agrobacterium mediated genetic transformation involves rigorous tissue culture and plant-regeneration. Tissue culture is skill intensive and influenced by a large number of uncharacterized physiological factors. Therefore, reproducibility of transformation protocols, being confounded with regeneration efficiency, remained an enigma for most of the crop species. As a result, plant transformation experiments require repeated optimization of the published protocols. Development of floral dip method in Arabidopsis by Clough and Bent was a revolutionary breakthrough in plant transformation (Clough and Bent 1998). In this method floral buds are dipped in Agrobacterium suspension for transfer of T-DNA in the reproductive tissues. This technique has been proven to be the simplest and most successful in Arabidopsis giving rise to transformants at a frequency of ~ 0.04–14% (Davis et al. 2009; Das and Joshi 2011). Primarily due to ease of the method and relatively high efficiency of transformation, numerous attempts have been made either for further improvement in efficiency (Das and Joshi 2011; Chung et al. 2000) or for extrapolating this method in other crop plants. This list of crop plants included several Brassicaceae members such as, B. napus, B. carinata and Camelina sativa (Verma et al. 2008; Li et al. 2010; Liu et al. 2012), as well as a few other crops like flax (Bastaki and Cullis 2014), maize (Mu et al. 2012), Raphanus sativus (Curtis and Nam 2001), Oryza sativa (Rod-in et al. 2014), Setaria viridis (Martins et al. 2015), and wheat (Zale et al. 2009). Though proof of concept has been reported, low frequency of transformation and lack of reproducibility in most of these crop species restricted application of floral dip method beyond Arabidopsis.

Rapeseed-mustard is a globally important group of oilseed crops. Canada, Australia and European countries account for more than 50% of its consumption. In India, among the Brassica spp. B. juncea is predominantly cultivated as oilseed crop. Many of the breeding objectives in B. juncea remained unfulfilled because they require large throughput genetic transformation in this crop. In most of the research projects and publications involving transgenic B. juncea only limited number of transgenic lines have been developed, which reflects difficulty in transformation. A few attempts through floral dip method also led to low transformation efficiency mostly ranging from 2 to 3% (Verma et al. 2008; Li et al. 2010).

In the present study, we have developed a high efficiency in planta transformation method in Indian mustard (B. juncea). For further improving the efficiency certain key parameters were optimized in a systematic way. Combining all the optimized parameters we have demonstrated substantially high transformation efficiency ranging from 10 to 30% that remains unmatched to any of the existing reports. The described method thus potentially overcomes the bottleneck of low transformation efficiency in B. juncea. It will be immensely beneficial to functional genomics and transgenic related projects which have recently regained priority after the reference genome sequence of this crop has been available.

Materials and methods

Plant material and growth conditions

Seeds of Brassica juncea cultivar Varuna and one of its indigenous germplasm lines Bio-YSR (INGR No. 04099) were sown in 8-inch pots filled with fine soil and manure, during the Brassica growing season in transgenic net house facility at Indian Agricultural Research Institute campus, New Delhi. After 15 days, the density of the plants was reduced to 3–4 plants in each pot. The plants were grown to inflorescence stage in 90–120 days. The plants were regularly irrigated and monitored for their proper growth.

Agrobacterium strain and plasmid construct

Agrobacterium tumefaciens strain GV3101 was transformed with binary vector pCAMBIA1305.1 by electroporation. The pCAMBIA1305.1contains GUS as the reporter gene under the control of a CaMV35S promoter and hptII as plant selectable marker conferring resistance to hygromycin. The transformed Agrobacterium cells were maintained on yeast extract mannitol (YEM) plates supplemented with 50 mg/l rifampicin, 50 mg/l gentamycin and 50 mg/l kanamycin.

Physiological and physical parameters

Two physiological parameters viz. germplasm and age of the plants before treatment and one physical parameter viz. presence or absence of light immediately following the application of Agrobacterium were optimized for their likely influence on the efficiency of the method. Plants of 90 and 120 days after sowing were used in the treatments.

Plant transformation

A single bacterial colony of the Agrobacterium strain harboring pCAMBIA1305.1 was inoculated in 10 ml of LB media supplemented with 50 mg/l rifampicin, 50 mg/l gentamycin and 50 mg/l kanamycin and grown to a cell density of OD600 = 2.0 at 28 °C. Aliquot of 2 ml of this near saturated culture was inoculated again in 200 ml of fresh LB broth and grown till the cell density of OD600 = 1.0 was obtained. The cells were harvested by centrifugation at 6000 rpm for 5 min at 22 °C and resuspended in 200 ml of transformation medium (1/2 MS with 5% sucrose and 0.05% Silwet L-77). The plants used for transformation were selected based on synchronous stage of the inflorescences bearing maximum number of flower buds. The plants were shifted to a growth chamber prior set at 24 ± 1 °C with 16/8 h day-night cycle and stabilized for 24 h. Approximately, 20–25 inflorescences from 5 to 7 plants were tied together, and the open flowers were excised. The unopened buds were opened by loosening the outer sepals with the help of a fine forceps so as to increase the accessibility of anthers and stigma to any external spray. Immediately after the mechanical maneuvering, the bunches of the inflorescences were tied together and sprayed with the Agrobacterium suspension inside the plant growth chamber. Following the spray, the inflorescences were wrapped with saran wrap for preventing rapid desiccation. After 24 h, the saran wraps were removed, the inflorescences were rinsed with sterile distilled water and the plants were shifted back to the transgenic net house. The same plants could be treated again for next set of buds after 7 d. The treated inflorescences were covered with paper bags till the siliques are formed and matured. The seeds were harvested from the mature pods and stored at room temperature for 2 months to break the dormancy before they were sown.

Screening and selection of primary transformants

The harvested seeds were surface sterilized in 0.1% HgCl2 and 0.1% SDS for 8–10 min followed by repeated washing with sterile double distilled water for 4–5 times. The treated seeds were sown in ½ MS medium containing 15 mg/l hygromycin. After the preliminary screening based on complete or partial decoloration of the cotyledonary leaves the seedlings were transplanted in pots for further growth in net house. The plantlets were allowed to grow to 4-leaf stage for molecular analysis.

PCR based confirmation of transformants

PCR based detection of the transgene was carried out at 4–6 leaf stage of the plants. The leaves from the individual plants were collected and genomic DNA was isolated by CTAB method (Murray and Thompson 1980). The quantity and quality of the genomic DNA was checked by agarose gel electrophoresis and spectrometric estimation in NanoDrop (Thermo Fisher Scientific, USA). The primer sequences used in PCR are given in Table S1. A 20 μl of PCR cocktail contained 1 μl of genomic DNA (100 ng), 2.5 μM of each gene-specific primers, 10 μl of 2X PCR master mix (TAKARA, Bio Inc. Japan) and 7 μl of nuclease-free water for making up the volume. The PCR cycling was carried out by initial denaturation of 95 °C for 2 min followed by 30 repeated cycles at 95 °C for 1 min, 55 °C for 30 s and 72 °C for 45 s and ended by a final extension step of 10 min at 72 °C. Amplification was assessed on 1% agarose gel containing 0.5 μg/ml ethidium bromide visualized under an UV trans-illuminator.

RNA isolation and cDNA synthesis

Total RNA was isolated from leaf samples by using RNAiso Plus reagent (Takara Bio Inc.Tokyo, Japan) according to manufacturer’s protocol. The isolated RNA samples were treated with TURBO DNA free kit (Ambion Incl., UK) for removing any traces of DNA contaminants. The quality and quantity of RNA was determined by NanoDrop ND-1000 spectrophotometer (Nanodrop technologies, USA). First strand cDNA synthesis was carried out from 5 μg of total RNA by using Superscript III First-Strand cDNA Synthesis Kit (Invitrogen, USA) in a 20 μl reaction volume as per the manufacturer’s instructions.

qRT-PCR based analysis

For qRT-PCR, each cDNA sample was diluted 10 times with nuclease free water. All the qRT-PCR reactions were performed using SYBR green detection chemistry, in a StepOne plus real time PCR machine (Applied Biosystems, USA). A reaction cocktail of 20 μl containing 2 μl diluted cDNA, 10 μl 2X SYBR Premix ExTaq II (Takara Bio Inc, Japan), 0.4 μl of ROX reference dye and 0.4 μl each of the forward and reverse primers was set up. PCR cycling was carried out at an initial denaturation at 95 °C for 30 s, followed by 40 repeated cycles each consisting of 95 °C for 10 s, 60 °C for 30 s and 72 °C for 30 s. For normalization in qPCR, actin was used as the reference gene (Table S1).

GUS analyses of the transformed seeds

GUS assay of the transgenic seeds were carried out as described by Jefferson et al. (1987). The seeds were fixed in a fixative consisting of 4% paraformaldehyde in 0.1 M phosphate buffer, followed by incubation with 1 mM of X-Gluc substrate solution (1 mg 5-bromo-4-chloro-3-indolyl β-d-glucuronide (X-Gluc) in 0.1 ml methanol, 1 ml 2× buffer, 20 μl 0.1 M potassium ferrocyanide, 20 μl 0.1 M potassium ferricyanide, 10 μl 10% (w/v) solution of Triton X-100, 0.85 ml water) and incubated at 37 °C for overnight.

Statistical analyses

For each experiment mean value was derived from two independent set of experiment each with 3 technical replicates. In each technical replicate seeds from 3 to 4 treated plants were pooled. Significant difference between means was analysed by Student’s t test. Different letters on bars were used for indicating significant difference in mean values at P < 0.05.

Results and discussion

Preparation of the binary construct and Agrobacterium strain

The binary vector pCAMBIA1305.1 was chosen as the transformation vector as it contains a GUS reporter gene under the CaMV35S promoter. In this vector, the coding sequence of GUS possesses an intron within it. Presence of intron eliminates any possibility of GUS expression in Agrobacterium host. The binary construct pCAMBIA1305.1 was mobilized into the Agrobacterium strain GV3101 and the transformed colonies were screened by PCR amplification of the GUS gene as well as restriction digestion of the isolated plasmid (data not shown). The purity of Agrobacterium strain harboring the gene construct was important for transformation efficiency. Therefore, it was maintained on YEM plates continuously in presence of the three selective antibiotics through a cycle of streaking and picking up single isolated colony at every 2–3 weeks interval.

Floral spray method of transformation in B. juncea

Since flower buds were the desired tissues for Agrobacterium treatment in this method, 3–4 months old plants with ample buds on the inflorescences were chosen (Fig. 1a, b). Before spraying the Agrobacterium suspension, tightly closed flower buds were opened by loosening the tightly wrapped sepals using forceps (Fig. 1c, d). Such mechanical intervention was favorable for the transformation frequency primarily because of two reasons: for increasing accessibility of the reproductive tissues by Agrobacterium suspension, and secondly, subtle mechanical injury to the floral tissues presumably helped in Agrobacterium infection process. The processed floral buds were sprayed with Agrobacterium suspension contained in a 500 ml volume plastic bottle that can generate aerosol in the form of a fine mist. To avoid any unwanted spillage of the Agrobacterium suspension during the spray, the plants were transferred to a growth chamber before treatment with Agrobacterium and transferred back to the net house only after completion of the treatment. Following the spray, the inflorescences were wrapped with saran wraps for preventing fast desiccation (Fig. 1e). Alternatively, for co-cultivation under dark condition, black polythene sheet was wrapped over the saran wraps covering the treated inflorescences of the plants (Fig. 1f). After 24 h, the saran wraps were removed and the treated inflorescences were rinsed with sterile water for removing excess bacterial growth, if any, and the surfactant. The treated inflorescences were covered with butter paper bags eliminating any subsequent cross-pollination of the flowers. Upon maturity of the pods, the seeds from the covered inflorescences were harvested. The steps of the transformation method through floral spray has been depicted in the form of a flowchart in Fig. 2.

Fig. 1
figure 1

Steps in floral spray based Agrobacterium treatment in B. juncea. a Mustard plants at flowering stage; b floral buds used in the treatment; c opening of floral buds by forceps; d opened floral buds ready for spray; eAgrobacterium treated inflorescence covered with saran wrap; fAgrobacterium treated inflorescence covered with dark polythene over saran wrap

Full size image
Fig. 2
figure 2

Schematic workflow of the steps. See “Material and methods” section for details

Full size image

Screening of T0 seeds on selective media

The harvested seeds were dried and stored for ~ 2 months at room temperature to break the seed dormancy. The T0 seeds were screened initially by germinating them in selective media followed by PCR analysis of the seedlings for identifying the transgenic plants. The seeds were surface sterilized and placed on ½ strength MS media supplemented with 15 mg/l hygromycin. The putative transformants, named as T1 transformants, were identified based on the green color of the cotyledons in contrast to cotyledons turned pale yellow in case of non-transformants within 2 weeks of growing on the selective media (Fig. 3). However, the screening based on cotyledon color was not very definitive because of lack of uniform and distinct qualitative difference in colour and therefore, further PCR based confirmation was warranted. Nevertheless, this preliminary screening was helpful in narrowing down to the putative transformants most of which, albeit not all, turned out to be PCR positive later.

Fig. 3
figure 3

Screening of T1 seedlings on selective growth media. a Seedlings of untransformed wild type seeds with cotyledonary leaves turning pale yellow within 2 weeks on selective media; b segregation of seedlings from the treated plants on selective media. Seedlings with fully green cotyledonary leaves turned out to be mostly positive in PCR analysis

Full size image

PCR based analysis of the transgenics

The putative T1 transgenics identified based on their normal growth and cotyledon color on the selective media were hardened, transplanted into pots and grown further. Presence of the transgene in the transformants was detected by PCR using a set of GUS-specific primers (Fig. 4a). The ~ 400 bp GUS specific amplification was further validated by sequencing. The specificity of the amplicon also validated applicability of the gene specific primers in screening the positive plants. The transformation efficiency was calculated based on the number of PCR confirmed T1 transformants obtained out of 100 T0 seeds collected from the mother plant and subjected to screening by germinating on MS based selective media. We did not prefer using hptII specific primers for screening because of our earlier experience of occasional non-specific amplification by the hptII specific primers in B. juncea.

Fig. 4
figure 4

Molecular analyses of B. juncea transgenic lines. a PCR based screening of transgenics with GUS-specific primers. N = Negative control without template, W = Untransformed wild type plants, P = pCAMBIA1305.1 as positive control, lanes 1–11 = independent transgenic lines; b RT-qPCR analysis of GUS expression in independent transgenic lines; c GUS assay of seeds harvested from T2 plants of transgenic B. juncea

Full size image

qRT-PCR based transcript analysis and gus assay

The transgenic plants confirmed by PCR were further evaluated at the transcript level by qRT-PCR. The quality of the isolated RNA samples analysed by nanodrop was found to be A260/280 ~ 2.0 and were further used for cDNA synthesis. The analyses of the qRT-PCR data revealed the presence of the GUS transcripts and its relative level, which varied among the independent transgenics (Fig. 4b). The varied level of transcripts observed among the individual lines was due to the position effects as in Agrobacterium mediated transformation the site of transgene integration is random. Analysis of qRT-PCR based transgene expression also assist in event selection in case of agronomically important traits. GUS expression in the segregating seeds harvested from the PCR positive plants were also analysed. In GUS assay, the transformed seeds appeared blue in color, whereas the non-transformants appeared colorless (Fig. 4c).

Optimization of parameters for increasing the transformation efficiency

Effect of genotype

We anticipated likely influence of few crucial parameters on the transformation efficiency in this method. Therefore, the effects of genotypes, co-cultivation in light vs dark, and age of the plants during the treatment were studied for their varied levels (Fig. 5). Transformation efficiency was found to vary significantly between and among the treatments. Among the two genotypes, Varuna and Bio-YSR, the transformation efficiency was significantly higher in Varuna as compared to Bio-YSR irrespective of variable ages of the plants and cocultivation conditions studied (Fig. 5). In B. juncea, response to Agrobacterium mediated transformation is known to significantly depend on genotypes (Verma et al. 2008; Yadav et al. 1991; Pental et al. 1993; Ono et al. 1994; Phogat et al. 2000; Rani et al. 2013). It is largely because of genotype dependence of regeneration efficiency for any hormonal regime used in tissue culture. This might also be attributed to the difference in production of the inducer molecules that vary in their cellular concentration in host and ability to induce vir genes (Karami 2008). In floral dip transformation method performed in cultivars of B. napus and B. carinata, the transformation efficiency in B. napus (1.86%) was higher compared to B. carinata (1.49%) (Verma et al. 2008). In another attempt of floral dip transformation in B. napus, though the transformation was successful the efficiency was not reported (Li et al. 2010). The observed transformation efficiency in the present study was significantly higher compared to the similar attempts not only within the Brassica spp. but also among the several other crops (Table 1). We believe that such large increase in the transformation efficiency could be due to mechanical maneuvering of a forceps for opening the floral buds prior to application of the Agrobacterium suspension. Conventionally, such mechanical maneuvering has not been done in any of the similar studies reported in B. napus and B. carinata. The process in addition to increasing the accessibility of the reproductive tissues to Agrobacterium was also expected to cause mild injury to the reproductive tissues. Both these phenomena were important in enhancing the transformation efficiency.

Fig. 5
figure 5

Transformation efficiencies in optimizing two variables viz. co-cultivation in light or dark and age of the plant before treatment in two genotypes. Values (2n = 6) represent mean ± SD from two independent experiments each with three technical replicates. Different lower-case letters on bars indicate statistically significant difference in mean values at P < 0.05

Full size image
Table 1 Transformation efficiency of floral dip transformation method in different plant species
Full size table

Effect of light during co-cultivation

The transformation efficiency increased by more than two folds when the cocultivation for 24–36 h, immediately after the application of Agrobacterium suspension, was carried out under dark compared to in light irrespective of the genotypes. For instance, in Bio-YSR cocultivation in dark increased the transformation frequency to 22% from 10% obtained in case of cocultivation in light (Fig. 5). Dark condition is known to trigger induction of vir-gene in Ti-plasmid of Agrobacterium and favors the transformation process. Cocultivation under dark condition is a routine practice in Agrobacterium mediated transformation. It has been reported that incubation under darkness improves the morphogenic capacity of the explants (Compton 1999), in addition to preserving endogenous light sensitive hormones (Compton 1999; Padua et al. 1998), and preventing accumulation of phenolic compounds (Arezki et al. 2001). As a result, dark condition during the pretreatment improved the frequency of shoot regeneration from cotyledonary nodal explants in watermelon (Compton 1999). In Agrobacterium mediated transformation of Typha latifolia dark cultured calli showed higher transient activity compared to the light cultured ones (Nandakumar et al. 2004). In contrary, incubation at dark inhibited T-DNA transfer in intact tobacco seedlings sprayed with Agrobacterium and kept in dark (Escudero and Hohn 1997). Light is associated with a number of physiological factors, such as plant hormone levels, cell proliferation and cell cycle stage etc. and therefore likely to influence plant regeneration process significantly (Villemont et al. 1997; Zambre et al. 2003).

Effect of plant age

In B. juncea, the plants initiate to flower after 30–78 days of vegetative growth though the cultivars may differ depending on the prevailing temperature, photoperiod, moisture level, nutrient status of the soil and sowing period (Rabbani et al. 1997; Devi and Sharma 2017). Two variable ages of the plants viz. 90 days and 120 days after sowing and bearing the primary as well as secondary branches of inflorescences were used for the treatment. Irrespective of the two different genotypes used in the study, the transformation efficiency was higher when the plants of 120 days were used for the treatment instead of 90 days old plants (Fig. 5). The physiological conditions depend on age of the plant. The plants of different ages have different levels of endogenous hormones which decides responses of the plant to any biotic and abiotic agent. The age and type of the explants are critical for tissue culture mediated regeneration and in turn genetic transformability (Bhuiyan et al. 2011). Even the chemical cues attracting the Agrobacterium cells are known to vary with the growth stages and meristematic nature of the plant cells (Gelvin 2003; Subramoni et al. 2014). Therefore, it was likely and empirically proven in our study that, the later time-points within the flowering stage would be better for achieving higher transformation efficiency through this method. However, beyond 120 days after sowing the density of the buds in the inflorescences were diminished.

Conclusions

Being a self-pollinated crop B. juncea has lost genetic variability for most of the agronomically important traits within the primary gene pool. The narrow genetic base necessitates the use of genetic transformation in crop improvement programme for many of the traits. Though it is considered as an amenable species for Agrobacterium mediated transformation, a large throughput method of transformation is not available. Moreover, cumbersome tissue culture dependent methods require skilful hands which are scarce and even staggered in research institutes. Here, we have described a streamlined in planta method of large throughput genetic transformation of B. juncea. The transformation frequency obtained in this method after integrating the optimized level of a few pivotal parameters was much higher compared to all the protocols available so far in this crop. Because of the ease of this method it will be immensely beneficial for functional genomics and transgenic development programme in Brassica juncea particularly in a time when its genome sequence has recently been decoded.