NAA at a high concentration promotes efficient plant regeneration via direct somatic embryogenesis and SE-mediated transformation system in Ranunculus sceleratus

The novel methods for efficient plant regeneration via direct somatic embryogenesis (SE) and SE-mediated transformation system under high concentration of NAA in Ranunculus sceleratus were established. On MS media containing a high concentration of NAA (10.0 mg/L) in the dark, all inoculated explants (root, stem and leaf) formed somatic embryos at high frequencies, respectively, 66.03, 126.47 and 213.63 embryoids per explant, and 100% of the embryoids developed into plantlets on 1/2 MS rooting media. Morphological and histological analyses revealed that SE in R. sceleratus followed a classical pattern. All inoculated explants can be used as receptors for genetic transformation in R. sceleratus, through direct SE-mediated method after Agrobacterium infection. RcLEC1-B, as a marker gene, changed the number and morphology of flower organs and the development of cuticle in R. sceleratus, which indicated that the efficient transgenic system of R. sceleratus was established. To our knowledge, this is the first observation that both direct SE and transgenic transformation system, via induction of a single plant growth regulator, have been successfully constructed in R. sceleratus.

There are a variety of medicinal plants and ornamental flowers in Ranunculaceae family, especially the diploid plants, such as Aquilegia coerulea and Nigella damascene, which are ideal recipient system for the molecular mechanism research of flower development. Ranunculus sceleratus is a high rich source of valuable secondary metabolites, including alkaloids, tannins, flavonoids, isoscopoletin and protocatechuyl aldehyde 1 . It has been used as a traditional medicine in China for a long history 2 . R. sceleratus was always employed to prevent the replications of HBV (Hepatitis B virus) and HSV-1 (Herpes simplex virus type-1) 1 , and in the treatments of jaundice, rheumatic pains, asthma and urinary incontinence 3 . R. sceleratus is very effective as an insecticide against Drosophila melanogaster and Tribolium castaneum 4 . In addition, R. sceleratus has a high capacity for sewage disposal, it can absorb large amounts of nitrogen and phosphorus, accumulate and monitor the variation of heavy metals, such as Cu, Pb, Fe and Zn 2,5 , etc. However, until now, it has no a universal and stable genetic transformation method used in Ranunculaceae family, which greatly limits the research on the basic research and application research of Ranunculaceae plants.
In the process of plant somatic embryogenesis (SE), the induced somatic cells promote proliferation and dedifferentiation under an appropriate stimulus, for instance, plant growth regulators (PGRs), stress factors, spontaneous factors, etc., then initiate embryo development through redifferentiation 6,7 . Somatic embryogenesis structures can be used as recipient explant for genetic transformation systems 8 , which can generate transgenic plants with potential as industrial sources of useful compounds and thus act as plant bioreactors 9 ; for example the hairy Induction of SE. Leaf discs about 1-1.5 cm 2 in area, and root and stem segments about 1-2 cm long (without axillary buds on stem segments) were excised from R. sceleratus plantlets for use as explants for the induction of somatic embryogenesis. The explants were put on MS media with 30 g/L sucrose and 3.6 g/L gellan gum (pH 5.8) supplemented with 1-naphthaleneacetic acid (NAA) at concentrations of 1.0, 2.5, 5.0, 10.0 and 20.0 mg/L. The inoculated explants were cultivated at 25 ± 1 °C in the dark or under low light (60 µmol·m −2 s −1 ) conditions to induce SE. The process of SE development was recorded using a digital camera (EOS 600D, Canon Inc., Japan)  and a stereomicroscope (SMZ800, Nikon Corporation, Japan). To evaluate the frequency of SE from explants, thirty replicates of each of 10 explants were used.
Histological analyses of SE. Frozen sections of somatic embryos at different developmental stages were prepared for microscopy according to a previously published method 13 , and imaged using an optical microscope (BX 41, Olympus Corporation, Japan).
Plantlet formation from somatic embryos. Without no change in the media, the PGR supplement and the dark conditions, somatic embryos spontaneously developed into plantlets. When plantlets had reached a length of 1-2 cm they were separated and transferred to rooting medium (1/2 MS salt + 1/2 B5 vitamins + 10 g/L sucrose + 8.0 g/L agar, pH 5.8) to induce root formation. To evaluate the frequency of regeneration of plantlets from somatic embryos, thirty replicates of each of 10 somatic embryos were set.

Isolation and overexpression vector construction of RcLEC1-B.
The RNA isolation, DNase treatment, reverse transcription and isolation of RcLEC1-B (LEAFY COTYLEDON1-B) were made according to the published method 14 . The nucleotide sequence was submitted to GenBank, was deposited in GenBank under accession number KM115581.1. The RcLEC1-B ORF was cloned into the KpnI and BamHI sites of pCAMBIA2300 to produce the overexpression vector pCAMBIA2300-RcLEC1-B. This vector was introduced into Agrobacterium tumefaciens (GV3101) by means of freeze-thawing 14 . In addition, the binary expression vector pBI121 carrying GUS reporter gene was used to analysis of transgenic efficiency via GUS histochemical assay.
Plant transformation via direct SE. The root, stem and leaf explants were immersed in Agrobacterium suspension (1/2 MS salt + 1/2 B5 vitamins + 50 g/L sucrose, pH 5.8) and uniformly shocked in the dark for 8-10 min. The explants were dried with sterile tissue paper and transferred to co-culture medium (MS www.nature.com/scientificreports www.nature.com/scientificreports/ salt + B5 vitamins + 40 mg/L acetosyringone (AS) + 30 g/L sucrose + 8.0 g/L agar, pH 5.8) and kept in the dark for 72 h. Then the explants were placed on the selective culture medium, with composition MS salt + B5 vitamins + 10 mg/L NAA + 100 mg/L Kan + 500 mg/L Carb + 30 g/L sucrose + 3.6 g/L gellan gum (pH 5.8). Formed embryoids on explants were isolated and transferred to rooting medium.
Statistical analysis. Using SPSS 16.0 software, analyses of variance (ANOVA) with 99% and 95% confidence intervals were applied to the digital data.

Results
Induction of SE from root, stem and leaf explants on media with NAA in the dark. To establish the SE induction system, NAA at concentrations of 1.0, 2.5, 5.0, 10.0 and 20.0 mg/L was tested in order to optimize PGR conditions, and different types of explants from root, stem and leaf material were investigated. The results showed that with all the investigated supplementary concentrations of NAA in the media, SE was successfully induced from all of the explant types tested, though at different induction frequencies. However, no SE induction was observed from root, stem or leaf explants on media without NAA (Table 1). Of the concentrations tested, 1.0 mg/L NAA resulted in the smallest average numbers of induced somatic embryos (9.33 from leaf, 6.57 from stem and 3.90 from root) and only a small amount of rhizoids 16 per explant (Table 1). However, a concentration of www.nature.com/scientificreports www.nature.com/scientificreports/ 2.5 mg/L NAA also resulted in a relatively small amount of SE and produced a few frog egg-like bodies (FELBs) 17 per explant (Table 1). A concentration of 10.0 mg/L NAA resulted in the highest average numbers of somatic embryos (66.03 from leaf, 126.47 from stem and 213.63 from root) induced per explant (Table 1). These results showed that for R. sceleratus leaf explants are the best type to use (Fig. 1), and that 10.0 mg/L NAA is the optimal PGR concentration. We also tested the effect of light; this experiment showed that under light conditions no SE was induced from any of the types of explant tested on media containing any of the above-mentioned concentrations of NAA, suggesting that incubation under dark conditions is necessary for SE induction in R. sceleratus. We found that the embryoids induced exhibited the five classic developmental stages in terms of morphological structure: multicellular pro-embryo, globular embryo, heart-shaped embryo, torpedo-shaped embryo, and cotyledon embryo (Fig. 2). However, there were many kinds of cotyledon embryos with different numbers of cotyledons, with two, three, four and multiple cotyledons all being observed in cotyledon embryos (Fig. 1A,B,A1,B1).

Plantlet formation from somatic embryos in vivo and GUS histochemical assay. Plantlets could
be induced in vivo from R. sceleratus somatic embryos. Somatic embryos produced in vivo on the induction medium was able to spontaneously develop into plantlets; each somatic embryo structure generally developed only one plantlet (Fig. 3B,D,E). The formed somatic embryos were stained in blue after GUS-staining (Fig. 4C1).

Overexpression of RcLEC1-B results in dwarfing and abnormal flower organs. Overexpression of
RcLEC1-B in R. sceleratus resulted in slightly dwarfed plants with folded leaves (Fig. 4F,F1,F2  www.nature.com/scientificreports www.nature.com/scientificreports/ (Fig. 5B,B2-B6,B9,B10). In the statistics of the efficiency of genetic transformation, we found that the highest transformation efficiency was approximately 70%, was obtained in leaf explant; that of the stem explant was 40%, and the minimum transformation efficiency was only 20% in root explant.

Discussion
Somatic embryogenesis is model system for reveal the developmental mechanism of plant zygotic embryo and a valuable tool for plant regeneration 6,18 . The results of previous studies showed that various treatments could be successfully applied to the induction of SE in higher plants; these conditions include PGRs, such as 2,4-D 17,19 , NAA 16 , TDZ 16,19 , and IAA 20 ; desiccation 20 ; osmotic stress 21 ; polyamines 22 , sugars [23][24][25][26] , salts 27 , and metal ions 28 . Always, the PGRs with suitable concentration, variety and combination are very important to SE and plant propagation 29 . In our study, it showed that the induced effect of SE under high concentration NAA (10.0 mg/L) is optimal, however, it needs to be strictly dark. The induction of SE in R. sceleratus under light was failure, which was consistent with the results of FELB in Solanum nigrum 17 . On the contrary, the induction of PLB structure in Rosa canina need involve high light condition 19 , and in Trichosanthes kirilowii the second stage of induction of rhizoid tubers (RTB), after rhizoid induction, also required light 16 . Optimization of light conditions is therefore important for the induction of different somatic embryo structures in different plant species. In this study, we found that the optimum concentration of NAA on induction of SE in R. sceleratus was 10.0 mg/L, which was much higher than that (1.0 mg/L) on induction of RTB in T. kirilowii 16 . The same PGR with different concentrations always can successfully induce different SE structures, which indicates that the concentration of PGR plays a key role in plant SE activation, however, the specific mechanism is still unclear. In this study, all types of explants have the potential to successfully induce SE, nevertheless, the induction results indicated that the leaf was the best explant with a higher frequency on SE induction compared with stem and root explants.
We conclude that the SE induction of R. sceleratus in this report, possess some traits based on the following reasons. In summary: (1) To our knowledge, this is first report of a method for the direct induction of SE in R. sceleratus; (2) There are five classic stages in the development of somatic embryo structures: pro-embryo, heart-shaped embryo, globular embryo, torpedo-shaped embryo and cotyledon embryo (Fig. 2). (3) We observed many kinds of cotyledon embryos with cotyledon numbers ranging from two to four, and also multiple cotyledons on polymeric cotyledon embryos (Figs. 1 and 3); (4) The results showed that SE often form individually on the surface of explants; (5) Darkness is necessary for induction; (6) The regeneration system via SE has a high regeneration efficiency (66.03-213.63 plantlets per explant), and it is therefore reasonable to predict that R. sceleratus transformation based on this regeneration system might also be high-efficiency. Further study is needed to determine whether the induction pathway developed here for R. sceleratus can be established in other species of Ranunculaceae.
The transcription factor LEC1-B 14 , one of the CCAAT box-binding factors (CBFs) known as the HAP3 (HEME-ACTIVATED PROTEIN3) group, was isolated in protocorm-like body (PLB) 19 , a special SE structure of Rosa canina. It was proved in our previous studies that RcLEC1-B-OE significantly changed the number and morphology of floral organs, formed the transition state structures and regulated the development of cuticle in