In vitro regeneration and Agrobacterium-mediated genetic transformation of Dragon’s Head plant (Lallemantia iberica)

Dragon’s head plant (Lallemantia iberica), is a flowering species belongs to the mint family (Lamiaceae). The species contains valuable essential oils, mucilage and oil which are used in pharmaceutical and food industries. Tissue culture is a feasible strategy to attain large‐scale production of plantlets with a huge potential to produce plants with superior quality. The objective of this study was to develop a simple and efficient method for regeneration and transformation of L. iberica. To reach this goal, the regeneration ability of various explants including leaf, cotyledonary node, hypocotyl and cotyledon segments was investigated in MS medium supplemented with diverse concentrations of NAA (Naphthalene acetic acid) and BAP (6-Benzyl Amino Purine). According to the results, cotyledonary nodes showed the best regeneration response. The maximum rate of regeneration (and number of induced shoots was achieved in 1 mg l−1 BAP in combination with 0.05 mg l−1 NAA from the cotyledonary nodes. Additionally, through the optimized regeneration technique Agrobacterium-mediated transformation of L. iberica was successfully accomplished. Gene transfer was assessed on leaf samples from regenerated plantlets under a fluorescent microscope to detect the GFP signals. Moreover, transgene integration and its expression were confirmed by PCR and RT-PCR analysis, respectively. The establishment of these efficient regeneration and genetic transformation methods paved the way for further application such as plant improvement, functional analysis and gene editing.

www.nature.com/scientificreports/ enhanced adaptability, disease and stress resistance based on plant biotechnological tools such as transformation is mostly depended on tissue culture technique [10][11][12] . Genetic transformation enables the introduction of foreign genes into L. iberica or efficient modification of target genes to create new cultivar with desirable traits such as higher valuable secondary metabolites or omega 3 fatty acid and alpha-linolenic acid. Moreover, in vitro regeneration and genetic transformation are important tools to enhance the production of secondary metabolites and engineer the valuable oil content in plants such as L. iberica. Actually Agrobacterium-mediated transformation has been frequently used to improve many plant species for several valuable traits 13 . Due to significant advantages of transformation using Agrobacterium, such as integration of transgene in the genome with low copy number, transgene stability in the host genome and efficient transformation, this is a preferred method comparing with other transformation systems 14,15 .
To the best of our knowledge, no well-organized method for transformation of L. iberica medicinal plant via Agrobacterium tumefaciens is available and only very limited studies have been performed on its regeneration 6,16 and so far,. Therefore, for the first time an appropriate technique for in vitro regeneration and gene delivery using Agrobacterium are presented. This allows genetic improvements to be put into practice for Lallemantia iberica. In summary, the major factors which affect the regeneration and transformation such as, type of explants, concentrations or combinations of PGRs, acetosyringone, Agrobacterium inoculation duration, have been successfully optimized and a viable regeneration and transformation protocol well established.

Material and methods
In vitro seed germination. Dragon's Head plant seeds were procured from Pakanbazr an Iranian private joint stock company, providing deferent kinds of seeds to the customers (www. pakan bazr. com/ en/ lalle mantiaiberi ca-seed). Using these seeds for research are not restricted according the role of University of Kurdistan, Sanandaj, Iran. This study meets national and international guidelines for research. The seeds were surface-sterilized by immersion in 70% ethanol for 30s and immediately washing three times with sterilized distilled water, followed by soaking in 5% sodium hypochlorite for 15 min and rinsing three times with sterilized distilled water. Next, the seeds were transferred into half strength Murashige and Skoog (MS) basal medium 17 and incubated for germination under a photoperiod of 8/16 h dark/light using a 2400 lx light regime at 25 ± 1 °C.
In vitro plant regeneration. 21-day old seedlings grown in vitro, were used as source of explants for regeneration potential of L. iberica plants. The cotyledonary node, cotyledon, hypocotyl and leaf pieces were prepared in small segments about 0.5-1 cm 2 . The explants were separately cultured on MS medium containing four combinations of 6-Benzylaminopurine (BAP; 0, 0.5, 1 and 2 mg l −1 ) and 1-naphthaleneacetic acid (NAA; 0, 0.05, 0.1 and 0.2 mg l −1 ). Afterwards, all cultures were incubated under a photoperiod of 8/16 h dark/light using a 2400 lx light regime at 25 ± 1 °C. For all plant growth regulator (PGR) combinations, 10 explants were incubated in each glass vessels. The explants were sub-cultured every two weeks and the number of regenerated shoots were finally recorded.

Root induction.
The well regenerated shoots were cultured on MS supplemented with 4 concentrations of NAA (0, 0.1, 0.5 and 1 mg/l) and incubated in culture room at 25 ± 1 °C. To evaluate the optimal concentrations of NAA for root induction all samples were daily monitored and data of rooting were collected.
Acclimatization of plantlets. The completely rooted plantlets were washed to remove the remains of MS medium, then transferred to small plastic pots containing 1:1 ratio of coco peat and perlite mix and totally covered with a transparent plastic shield. To adapt them to environmental condition, 2-3 small holes in the plastic wrap were daily created. Meanwhile, the plants were regularly irrigated with diluted liquid MS basal medium. After 3-4 weeks, the shield was entirely removed. Finally, the plantlets were transferred into the larger pots filled with 1: 1: 1 ratio of farm soil, compost and sand for more growth and evaluation.
Agrobacterium strain and binary vector. Agrobacterium tumefaciens strain C58C1Rif R (pMP90) containing pXK2FS7 binary vector has been utilized for transformation. It was kindly provided by Department of Plant Systems Biology (VIB, Ghent, Belgium). The genetic map of pXK2FS7 plasmid is shown in Fig. 1. It carries two reporter genes green fluorescent protein (GFP) and β-glucuronidase (GUS) that both are controlled under the CaMV 35S promoter 18 . To prepare the suspension culture of the Agrobacterium, a single colony obtained from the streaked LB agar petri plate incubated at 28 °C, was first inoculated in liquid LB selective medium containing 50 mg l −1 rifampicin, 50 mg l −1 streptomycin and 25 mg l −1 gentamicin. Next the culture was incubated at 28 °C on a shaker with 200 rpm to achieve desired OD600 = 0.5-0.8. Later on, the cells were collected by centrifugation at 5000 rpm for 10 min. The pellet was finally re-suspended in appropriate volume of liquid MS medium (pH 7.8) for further use.
Agrobacterium-mediated transformation of explants. The cotyledonary node pieces were submerged in A. tumefaciens suspension (OD 600 = 0.6) for inoculation. The inoculated explants blotted dry on sterile Whatman paper under sterile conditions, then immediately transferred to co-cultivation medium (MS containing no PGRs and different concentrations of acetosyringone). The cultures were incubated in a dark chamber (25 ± 1 °C) for 3 days. Subsequently, the infected explants were washed three times with sterile distilled water, following with 50 ml of cefotaxime solution 500 mg l −1 for 20 min to remove any trace of bacteria. The explants were gently blotted dry on sterile Whatman paper and directly cultured on regeneration medium (MS containing 1 mg l −1 BAP and 0.05 mg l −1 NAA) having kanamycin at concentration of 60 mg l −1 . After regeneration and Effect of acetosyringone on transformation. To study the effect of acetosyringone on transformation frequency, in a separate experiment, the suspension of Agrobacterium cells was supplemented with different concentration (0, 100, 200 and 300 µM) of acetosyringone for 15 min. Following the transformation of explants, transformation efficiency was calculated.
Green fluorescent protein visualization. Green fluorescent protein visualisation was inspected in regenerated putative transgenic lines. Leaf samples were prepared from putative transgenic and non-transgenic plants. The leaf explants were exposed to blue ultraviolet light at excitation bandpass of 400 nm with a fluorescence Olympus BX51 microscope (Olympus, Japan) equipped with narrowband filter of 460-490 nm. Images were captured using a digital camera (Olympus, Japan) attached to a computer. Assessment of transgene expression. Total RNA was extracted from leaves of positive PCR transgenic and control plants (non-transgenic) using Mazzara and James method 20 . To avoid PCR amplification from genomic DNA, the RNA was treated with DNase I (DNAbiotech). First-strand cDNA was synthesized from total RNA using 2X HyperScript RT premix (GeneAll, South Korea) according to the manufacturer instructions. The cDNA was then used for PCR amplification of a 300 bp of GFP using specific primers (F: 5'TCG TGA CCA CCC TGA CCT AC3' , R: 5' ACC TTG ATG CCG TTC TTC TGC3'). PCR products were finally visualised on 1.2% (w/v) agarose gel.

In vitro verification of stable transformation. To evaluate transient and stable transformation, in vitro
regeneration of T 0 generation was performed. The stem's nodes of confirmed transgenic plants were isolated, and cultured on the MS medium containing PGRs and 60 mg l −1 kanamycin. Following regeneration, a number of regenerated shoots were assessed using observation of GFP or molecular assessment (PCR and RT-PCR).

Data analysis.
Treatments are assigned to a completely randomized design (CRD) with three replicates of ten explants. The data were analysed with SPSS 15.0 software package. Comparison of means was performed using Duncan's multiple range test (p ≤ 0.05).

Results
In vitro plant regeneration. In regeneration experiment, cotyledonary node, cotyledon, hypocotyl and leaf explants were examined on 16 different PGR combinations (Table 1). Cotyledonary nodes were able to directly regenerate shoots in all PGR combinations (Fig. 2a) except control (0 concentration of PGR) or media without BAP. But, no shoots or calli were obtained on cotyledon and leaf explants in the presence of the 16 combinations, nonetheless hypocotyls produced a few shoots only on the medium supplemented with 2 mg l −1 BAP + 0.1 mg l −1 NAA (Fig. 2, b and c). Shoot regeneration frequency (Table 1) and the number of induced shoots per explant (Table 2) in cotyledonary nodes were significantly varied in different combinations of NAA/ BAP. The highest regeneration rate (100%) and number of induced shoots (23 ± 3.60) were observed in combination of 1 mg l −1 BAP + 0.05 mg l −1 NAA. However, other combinations resulted 100% regeneration occurrence but the number of induced shoots was significantly lower than of 1 mg l −1 BAP + 0.05 mg l −1 NAA combination ( Table 2). The lowest regeneration rate (59.66) was obtained from medium containing 2 mg l −1 BAP plus 0.05 mg l −1 NAA and the lowest number of induced shoots (3.33 ± 1.15%) was resulted in the medium containing 2 mg l −1 BAP (Table 1 and 2). Overall, different responses were observed in different explant types and bet-  www.nature.com/scientificreports/ ter regeneration performance for cotyledonary node was obtained using 0.5 or 1 mg l −1 BAP in combination with some concentrations of NAA (Tables 1 and 2). Shoot regeneration from cotyledonary node and hypocotyl explants are presented in Fig. 2.

Root induction.
Roots were initiated after nearby two weeks on in vitro-regenerated shoots cultured on MS medium with different concentrations of NAA (Fig. 2d). Maximum (23 ± 2.00) and minimum (8.33 ± 0.57) number of roots were induced in the medium supplemented with 0.1 mg l −1 NAA and 0 mg l −1 NAA, respectively. Furthermore, by increasing the concentration of NAA over 0.1 mg l −1 , number of induced roots was reduced (p ≤ 0.05) (Fig. 3).
Acclimatization of plantlets. The in vitro-regenerated shoots with well-developed root system were maintained in the acclimatization conditions for 4 weeks in the small plastic pots (Fig. 4a,b,c) contain coco peat and perlite mix (1:1) and presented the best results with 100% survival rates. The plantlets were next transferred to the greenhouse into the larger pots filled with 1: 1: 1 ratio of farm soil, compost and sand (Fig. 4d), compost and sand and they grew and developed normally and morphologically were identical to the mother plant. Overall, the regenerated plants were morphologically normal, phenotypically stable, fertile, reproductive and able to set viable seeds (Fig. 4e).
Kanamycin tolerance assay. In the control (0 concentration of kanamycin), all explants exhibited normal growth and were healthy and green. On medium containing 30 mg l −1 kanamycin, some explants stopped growing (Fig. 5a) and their tissues turned brown after 30 days but most of them survived (66.66 ± 0.57%). In all other applied levels of kanamycin (60, 90 and 120 mg l −1 , the explants were in poor condition and after two weeks they showed gradual whitening and finally stopped growing (Fig. 5b). Therefore, the concentration of 60 mg l −1 was designated as the effective level of kanamycin for selection of putative transgenic plants.
Plant transformation. Using optimized regeneration system, explant transformation was performed and finally several shoots were successfully induced from cotyledonary nodes in regeneration medium (1 mg l −1 BAP + 0.05 mg l −1 NAA) containing 60 mg l −1 kanamycin (Fig. 6a,b). Four weeks after culture, shoots were entirely developed (Fig. 6c). The induced shoots were gently transferred into rooting medium (including 0.1 mg l −1 NAA) (Fig. 6d,e,f). Finally, the rooted plants were acclimatized and hardened before moving them to the greenhouse for further analysis (Fig. 6g). There were no obvious phenotype differences between normal wild type and transgenic plants. They were fertile, reproductive and able to set viable seeds (Fig. 6h).

Effect of infection time on transformation efficiency.
Analysis of variance showed that the effect of exposure time (treatment) of explants to Agrobacterium was significant. It means that the different infection time     (Fig. 7a). By increasing of the infection period in Agrobacterium suspension, the transformation efficiency rates were considerably reduced (Fig. 7a).
Effect of acetosyringone on transformation efficiency. Maximum percentage of transformation (6.66 ± 0.33%) was achieved when acetosyringone was applied at 100 µM, however, presence or absence of acetosyringone did not significantly affect the transformation efficiency (p ≤ 0.05; Fig. 7b). Higher concentrations of acetosyringone (200 and 300 µM) caused overgrowth of agrobacteria and browning of explants, ultimately leading to death of some explants.

Evaluation of transgenesis by GFP visualization.
To confirm the transgene expression in regenerated plants after transformation, leaf samples of putative transgenic and non-transgenic plants (control) were investigated under fluorescence Olympus BX51 microscope (Olympus, Japan). All the collected sample leaves from   (Fig. 8a), while no signal of green fluorescence was observed in the non-transgenic leaf samples and they appeared red (Fig. 8b).

Confirmation of putative transgenic lines by PCR.
PCR reactions were carried out using primers designed specifically for the DNA region of interest. Agarose gel profile representing the PCR products showed exact amplicons of 310 bp for GFP (Fig. 9a) and 710 bp for GFP-GUS (Fig. 9b)

Analysis of transgene expression in transgenic plants. Transcript transgene expression was studied
by RT-PCR. First, cDNA was synthesized from RNA samples of transgenic and non-transgenic plants which previously confirmed by PCR. RT-PCR was performed using specific primers to detect GFP transcript from transgenic plants. The RT-PCR product after agarose gel electrophoresis revealed an expected amplified fragment of 300 bp (Fig. 9c). While in non-transgenic plants no band was detected. This result confirmed successful integration and transcription of the genes on T-DNA in selected transgenic plants.   www.nature.com/scientificreports/ Verification of stable transformation. The stem's nodes of confirmed T 0 transgenic plants were subcultured onto the medium containing 60 mg l −1 kanamycin. Following the regeneration, a number of shoots were obtained. Leaf samples of the regenerated plants were evaluated under a fluorescence microscope. The GFP signals were effectively observed. In addition, the transgenic status of regenerated shoots was confirmed by PCR and RT-PCR and the targeted transgene fragments were detected. Additionally, PCR (Fig. 9d) was carried to detect a 711 bp GFP-GUS fragment on the genomic DNA of a few T 1 plants (progeny plants) derived from primary transgenic plants (T 0 ). T 1 transgenic progeny plants did not show any considerable difference of morphological traits over the wild type plants (Fig. 6i).

Discussion
Plant tissue culture method has been used as an important approach for secondary metabolites production, micropropagation and it can assist breeding for the genetic modification and improvement of medicinal plants. Therefore, establishment of an efficient tissue culture system with several applications could be a prerequisite for L. iberica. In the present study, for the first time, we achieved in vitro regeneration and A. tumefaciens-mediated transformation of L. iberica. Regeneration experiment revealed that the cotyledonary nodes exhibited the highest regenerative capacity. Other explants except hypocotyl failed to develop any callus or shoot and turned brown at last. These results support the importance of evaluating different explant types when developing a plant regeneration protocol. Nodal sections normally exhibit a high regeneration response. The cotyledonary node usually has many meristematic cells; therefore, this enables cotyledonary node to produce directly a large number of shoots. Several studies have demonstrated that cotyledonary node was the most appropriate explant for regeneration [21][22][23] and it performed best for in vitro regeneration or Agrobacterium-mediated transformation in several plant species such as Lupinus albus 24 , Passiflora caerulea L. 25 , Cicer arietinum L. 26 , Lallemantia iberica 6 , Glycine max 27 and Stevia rebaudiana Bertoni 28 .
The BAP in combination with NAA has been usually used for the shoot regeneration in various plant species [29][30][31] . BAP appears to be one of the most efficient cytokinins for shoot generation from nodal segments and the success of shoot induction using BAP is very high 32,33 . In L. iberica the BAP/NAA combination, at certain concentration, was successfully induced in vitro shoots from cotyledonary nodes. The optimal concentration of BAP/NAA combination for shoot regeneration was 1 mg l −1 BAP plus 0.05 mg l −1 NAA which led to induction of the highest number of shoots. In regeneration experiment of L. iberica the number of shoots were reduced by increasing the concentration of BAP to 2 mg -l , such results have also been reported in grain legumes 34,35 . Although the higher concentration of cytokinin can induce the signal of shoot formation, it may also inhibit the normal growth of shoots and can form vitrified tissues because of their toxicity as this has been reported previously in other plant species [36][37][38][39] . We found different performance of various explants and different PGR combinations in L. iberica regeneration. Previous studies in several plant species have also discovered that many factors such as genotype, the nature of explant, culture media type, combination and concentration of plant growth regulators can influence shoot regeneration 10,40,41 .
In root induction study, cluster roots were produced at all concentrations of NAA, but the maximum number of roots were induced in the medium supplemented with 0.1 mg l −1 NAA. On this NAA concentration vigorous roots developed in normal shape with branched and whitish colour. Furthermore, by increasing NAA concentration, the number of roots were significantly decreased. These results are consistent with study of Hsieh and colleagues 42 which reported that the lowest concentration of NAA is appropriate to induce in vitro roots in Arachis hypogaea. Additionally, in Siratia grosvenorii the maximum roots observed in medium containing 0.1 mg l −1 NAA 43 . Indirect plant regeneration from callus is normally laborious and rely on different factors and additionally it might lead to somaclonal variation. Several researchers have declared that establishment of a tissue culture method for propagation and transformation from callus may result in genetic abnormalities [44][45][46][47] . Therefore, availability of a direct regeneration procedure is valuable for micropropagation and genetic transformation.
In Agrobacterium-mediated transformation experiment, two factors affecting gene transfer such as infection time and acetosyringone concentration were studied. The results showed that transformation efficiency of L. iberica was substantially affected by infection time, nonetheless the different concentrations of acetosyringone had no significant effect on transformation efficiency, although lower concentration of acetosyringone (100 µM ) induced higher transformation efficiency (Fig. 7b). Phenolic compounds such as acetosyringone and hydroxyacetosyringone, which attracts the agrobacteria to the wounded explants, also could induce the virulence genes, thus facilitating the transfer of the T-DNA region. Moreover, the transformation frequency might be negatively affected with the higher concentration of acetosyringone but the data was not statistically significant. Transformation efficiency was decreased at higher concentrations in other plant species such as Anthurium andraeanum 48 and Vanda Kasem's Delight 49 . Moreover, by increasing the time of explant inoculation, the number of obtained transgenic plants were intensely decreased, and no transgenic was obtained in the time of 30 min. Such effects of suitable time of inoculation on transformation efficiency was also reported in Ocimum gratissimum 39 .
Effect of kanamycin at four concentrations (0, 30, 60, 90 and 120 mg l −1 ) on regeneration revealed that the concentration of 60 mg/l −1 resulted in a 66.66% shoot induction frequency. Regeneration on this kanamycin concentration was effective and gene transformation was finally confirmed by PCR and RT-PCR. Similar to our result, in Salvia miltiorrhiza, kanamycin at concentration of 30 and 60 mg l −1 was used to screen the transformant plantlets and the highest regeneration efficiency was obtained in the medium containing 60 mg l −150 .
In addition to nptII selectable marker, which is frequently used to select transformed plants, pXK2FS7 vector contains the GFP reporter gene. One advantage of GFP is that its expression analysis does not involve a destructive assay and visualization of GFP expression is performed with fluorescent microscopy allowing tracking progress throughout different stages 51 . In this study the putative transformant plants were initially screened using GFP signals. The use of green fluorescent protein (GFP) marker gene can reduce the time required for selection  52,53 . Fluorescent marker genes allow visual detection and monitoring of transgene expression in transformed tissues and manual selection of transgenic plants even without using antibiotic or herbicide selection markers 54,55 . Detection of GFP signals under fluorescence microscope provided successfully the initial screening of transgenic plants that were finally confirmed by PCR and RT-PCR.

Conclusion
In this study, we established a preliminary in vitro procedure for direct regeneration of L. iberica using cotyledonary node and also a transformation system using A. tumefaciens. To our knowledge, this is the first report of Agrobacterium-mediated transformation of L. iberica which could provide feasible practical basis for genetic transformation and breeding of this valuable medicinal plant and it is expected to be of commercial value either by improvement of L. iberica seed oil or secondary metabolites. Also, genetic transformation provides a powerful tool to study of gene function and gene editing that will contribute to trait improvement, functional genomics and metabolite engineering.