ABSTRACT

Agrobacterium-mediated transformation through floral dip and rapid selection process after transgenic event had become a preference as it will overcome the difficulties faced in tissue culturing procedures and lengthy time for screening transformed progenies. Therefore, in this study, three constructs, p5b5 (14,289 bp), p5d9 (15,330 bp) and p5f7 (15,380 bp) in pDRB6b vector which has hygromycin as a selectable marker gene were introduced individually into Agrobacterium tumefaciens strain (AGL1). The cell suspension was applied to Amaranthus inflorescence by drop-by-drop technique and was left to produce seeds (T₁). The T₁ seeds were germinated and grown to produce seedlings under non-sterile condition. Hygromycin selection on seedling cotyledon leaves results in identification of 12 putative transformants, three from p5b5, four from p5d9 and five from p5f7. All positive putative transformants that were selected at the first stage through hygromycin spraying showed positive result in leaf disk hygromycin assay and in a construct specific polymerase chain reaction-based assay. A ~750 bp amplified hygromycin gene was further verified through sequencing. Our results suggest that Amaranthus inflorescences were able to be transformed and the transformed progenies could be verified through a combination of simple and rapid methods.

1. INTRODUCTION

Amaranthus is one of the most important leafy staple crops which can be found throughout the tropics and in many warm temperate regions [1]. Studies have been conducted to develop technologies that aim in exploiting Amaranthus due to its high nutritional value [2]. Technological innovation in plant biotechnology is an important catalyst in any crop enrichment [3]. Stable Agrobacterium-mediated genetic transformation offers advantages in transferring one or few copies of DNA fragments carrying the genes of interest. It can be carried out on the whole plant by using either tissue infiltration [4] or floral dip [5]. This method is applied in preference to tissue culture-based techniques as it directly produces transformed seed circumventing and avoiding the lengthy tissue culture period and somaclonal variation [6].

The success in the transformation of Amaranthus floral would widen the possibilities for the exploitation of the available reported method on many other alternative crops. Agrobacterium-mediated floral dip transformation has been developed and successfully applied in the popular model plant such as Arabidopsis thaliana [7] and Medicago truncatula [8]. Since both Amaranthus and Arabidopsis are dicot plants with bisexual flowers, it is understandable that they share a common ancestor and similar anatomy and therefore amenable to the same transformation technique.

In addition, successful production of transgenic plants is also heavily dependent on a suitable selection technique to screen for transformants. The effectiveness and ease of use influence the choice of suitable selectable marker gene. A simple leaf spraying method developed for selection using Basta [9] had enabled large scale screening of plantlets, and similarly in the present study, hygromycin was sprayed on the seedlings during the screening process. The homogentisate phytyltransferase gene (HPT) was introduced for future manipulation of vitamin E biosynthesis in successfully transformed Amaranthus.

2. MATERIALS AND METHODS

2.1. Vectors

The pDRB6b vector digested with HindIII/KpnI served as the backbone vector for p5b5, p5f7 and p5d9 recombinant vectors. The recombinant vector was constructed in between RB2 and LB of pDRB6b. p5b5 contained Ubi1P driving HPT; p5f7 contained Ubi1P, Ubi1-intron and HPT while p5d9 contained LHCB, Ubi1-intron and HPT. The NosT site was located at the 3` end of the HPT gene in all the three constructs. All the recombinant vectors contained hygromycin gene driven by CaMV35S promoter located in between RB1 and RB2 (Figure 1). The constructs were transformed separately into AGL1 strain using freeze-thaw method [10].

2.2. Floral Transformation in Amaranthus

Amaranthus seeds (Leckat Corporation Sdn Bhd, Kepong, Malaysia) were sown in a pot containing soil, peatgro (Peatgro™, Batu Caves, Selangor, Malaysia), and sand in a ratio of 1:2:1 in the Transgenic Glasshouse, UPM under natural daylight. The plants were allowed to grow into maturity until the length of the inflorescences was in the range of between 4 and 6 cm.

A. tumefaciens strain (AGL1) carrying the recombinant vectors from the glycerol stocks were streaked onto LB agar containing 25 mg×L⁻¹ rifampicin and 50 mg×L⁻¹ spectinomycin and was incubated at 28˚C for two to three days in the dark. The bacterial suspension was prepared according to Lee and Yang [11] except Silwet L-77 was replaced with 0.01% Tween 20.

Amaranthus inflorescences were inoculated according to Martinez-Trujillo et al. [12]. The plants were injected in a shaded area away from direct sunlight and were allowed to remain at similar condition overnight. The treated plants were placed back to the green house and allowed to grow normally [13]. Watering of the plants was continued until the siliques started to lose their green colour and became yellow. Seeds (T₁) were harvested and stored according to Curtis [8] and Bent [14]. Seeds were sorted to identify those that are black in colour to determine percentage of seed productivity (w/w) prior to expression analysis.

2.3. Expression Studies and Transgene Analysis Using the hph Selectable Marker Genes

Seeds from floral dipped plants which were left for maturation until 50% of the plant became yellow were sown into soil and the origin where the seeds were collected was noted. Twenty days after sowing, the young seedlings were sprayed with 150 mg×L⁻¹ hygromycin containing 0.01% Tween 20 daily for five days. Young seedlings that showed no necrosis on the leaves after the 5^th day were considered to contain a functional hph gene. Hygromycin resistant plants were transferred to new pots and allowed to grow bigger. To further confirm the transgenicity of the hph resistant plants, leaf pieces of one month old plants were excised and analysed by hph histochemical assay [9]. Leaf disk assay in hygromycin solution to detect the expression of the hph gene was conducted to further distinguish the transformed plants [15] and untransformed control plants.

Fresh plant leaf materials were obtained by using FLOTING PUNCH (adoro^TMA). The leaf sample was crushed with 100 μL pipette tip by pressing against the microcentrifuge wall until there were no large pieces of tissues left. Five hundred microlitre of dilution buffer (0.5 N NaOH) [16] was added to every disk according to 3(π/j²); j² = d = 4.5 mm (π = 22/7; j² = d = diameter). The supernatant was centrifuged using Eppendorf Centrifuge 5810R and 0.5 μL of fresh plant tissue solution was used for preparation of a 20 μL Phire^® Plant Direct Polymerase Chain Reaction (PCR) (Finnzymes, Finland). The PCR reaction consist of 7.6 μL of distilled water, 10 μL 2× Phire^® Plant PCR buffer, 1 μL of each 0.5 μΜ primer HyG (F:5`TCCGGATGCCTCCGCTCGAA3`) and primer HyG (R:5`ATGCAGCTCTCGGAGGGCGA3`) and 0.4 μL Phire Hot Start II DNA polymerase (F-130X). The cycling reaction was as follows: Initial denaturation at 98˚C for 5 min, second denaturation at 98˚C for 45 sec, annealing at 72˚C for 45 sec, extension at 72˚C for 40 sec and final extension was at 72˚C for 1 min, the cycle was hold at 4˚C. The PCR product was analysed on 0.8% (w/w) agarose gel and was sequenced. The sequencing

result was analysed using nucleotide blast NCBI.

2.4. Statistical Analysis

Bacterial growth, transformation event and percentage of seed productivity value were expressed as mean values accompanied by standard error (SE) from triplicate reading performed repeatedly on the same subjects.

3. RESULTS

Based on the condition mentioned Table 1 shows that AGL1 transformed with p5b5, p5d9 and p5f7 reached the target growth (OD₆₀₀) between 0.7 and 1.4. The results in Table 2 show that all the inoculated plant with the transformed AGL1 containing p5b5, p5d9 and p5f7 produced more than 95% of seed productivity. This data was better than the value obtained for the non-transformed plant.

The seedlings of non-transformed plants began to show yellow spots (necrosis effect) on the third day after being sprayed, and the effects worsened on the fifth day. While, the transformed plants (T₁) remained green and healthy (Figure 2). A total of 12 progenies were identified as hygromycin resistant (hph^R). Further tests on the T₁ progeny seedlings using leaf disk assay and PCR analysis also showed positive results. In the leaf disk assay, all the 12 putative transformants remained green for five days in hygromycin solution. Figure 3 showed the size (~750 bp) of amplified transgene hygromycin fragments were detected in all the 12 transformant plants (T₁). No transgene was detected in the non-transformed plants by PCR.

The identity of the hygromycin gene was confirmed by sequencing analysis which showed more than 95% of maximum identity with pTEXL-Hyg (Accession No: JN596098.1), pTcR-GA-Hyg (Accession No: JN596081.1) and pTcR-HG-Hyg (Accession No: JN596072.1). The transformation efficiency of approximately 1% to 2% was determined for p5b5, p5d9 and p5f7, respectively (Table 2).

4. DISCUSSION

The rapid ongoing process in engineering metabolites has increased the demands to utilise reported protocols/technology in in-vivo plant transformation to plants that are taxonomically related as their anatomical characteristics are similar [17]. A limited number of bisexual plants such as Arabidopsis thaliana [18], Medicago truncatula [8], Thellungiella halophila [19], and Chenopodium rubrum [20] have been successfully transformed through floral dip transformation.

Amaranthus is a genus in the family of Amaranthaceae. The flowers of most species in the Amaranthaceae are bisexual and it is similar to Arabidopsis, where they have both male and female reproductive organs within the same flower [21]. Therefore, a successful attempt was made to utilise standard floral dip procedure which was reported for Arabidopsis on Amaranthus as both plants fall into the same group (eudicots).

In order to reduce the amount of AGL1 solution needed, infiltration solution containing transformed AGL1 was pipetted rather than dipping the whole plant [22]. This was preferred as flowering Amaranthus reaching up to 90 cm tall are difficult to be dipped. The developing female reproductive organ, the ovary, is the place where female genotype cell lineages formed [23]. The stage of open florets during anthesis was selected to inoculate the transformed AGL1. This will subsequently produce transformed (T₁) seeds.

The AGL1 inoculums were prepared in infiltration medium containing Tween 20 rather than Silwet L-77. Most of the reports so far used Silwet L-77 as a unique surfactant for transformation procedures [24,25]. However, this can cause some restrictions for researchers in small laboratories with limited source of reagents and budget. In addition, the transformation efficiencies de

Table 1. Growth of AGL1 carrying the recombinant plasmid in LB broth after 48 hours at 28˚C with aeration at 180 rpm.

Values are the mean of three replicates followed by standard error (SE).

Table 2. Percentage of seed productivity after floral dip transformation compared to non-transformed plant and transformation efficiencies by AGL1 carrying p5b5, p5d9 and p5f7.

Seed productivity represents as mean replicates ± standard error (SE).

tected in this study were similar to using Silwet L-77 [18].

Construct specific PCR is being used widely for the detection of specific DNA sequences that is not found in plants [26]. According to Wang et al. [16] younger tissues will gave better amplification. Amplification of a specific sequence such as hygromycin (sequence that is not found naturally in the plants) was carried out through construct specific amplification, thus will provide reliable information on a transgenic event [27]. In this study, the first stage of selection contains two steps 1) hygromycin spraying and 2) leaf disk assay. This was highly sensitive and accurate to be used in the selection of transformants as all putative hygromycin positive transformants that were selected through the first stage showed positive result in the construct specific PCR amplification.

Transformation frequency that was determined in this study was similar to the one reported by Clough and Bent [18] for A. thaliana which was in the range of 0.5% to 3%. This confirmed the hypothesis of the study that the floral dip transformation method can be applied to any bisexual plants albeit some suitable modifications. In addition, the transformation rate achieved was comparable to that reported by other researchers [23,25]. With this in mind, available technology for transformation can be applied in Amaranthus. All transformed Amaranthus have to be analysed and expression levels of vitamin E have to be ascertained in order to produce nutritional modified crop.

5. ACKNOWLEDGEMENTS

The authors would like to thank Dr. N. M. Upadhyaya from CSIRO Plant Industry for providing a series of pDRB vectors and A. tumefaciens (AGL1). The senior author would like to express appreciation to Universiti Malaya for providing financial assistance in the form of Skim Latihan Akademik IPTA scholarship.

REFERENCES

NOTES

^*Corresponding author.