Establishment of a Rapid and Stable Transgenic System by Agrobacterium-Mediated Transformation of the Germination Seeds in Diploid Strawberry.

Although much in the identication of gene using traditional, stable and transient genetic transformation systems in the limitation is, more or present. Thus, development of a and system Here, using diploid Hawaii-4 (Fragaria vesca) seeds and a reporter gene of CHLH (the H subunit of magnesium chelatase magnesium chelatase) , we rst develop a new, rapid, ecient, and stable transgenic system by the Agrobacterium-mediated seed transformation to silence the reporter gene, obtaining a transformation frequency with 10 % through a series of optimization conditions, including full imbibition and initial germination, shaking infection for 24 h, dark cultivation on MS medium for 3 d at 24 ℃ , light culture on MS-Tim medium for 1 week at 24 ℃ , and vector construction carried uorescence label. Taken together, radicle-emergence germination seeds, appropriate Agrobacterium concentration and infection time are critical for successful transformation, obtaining transgenic kanamycin-resistant seedlings within 1 month and T2 generation transgenic plants within 4 months.


Background
Strawberry (Fragaria), as a distinct member of Rosaceae, is not only an important fruit crop worldwidely, but is also regarded as a model for studying non-climacteric fruit ripening [1]. In the past years, a large of genes related to strawberry fruit ripening have been identi ed via traditional stable and transient transgenic systems in strawberry; however, more or less, the limitation is existed [2][3][4]. Thus, development of a rapid, e cient, and stable transformation system is required for strawberry research and breeding.
It is known that strawberry leaf tissue has been extensively studied to develop e cient transformation and regeneration methods through somatic tissue or callus formation [5][6][7][8][9][10][11][12][13][14]. However, the regeneration of transgenic strawberry plants not only remains di cult, but also takes a long time from transformation to identi cation [3,15,16]. Such as, it takes at least 15 months from transformation to gaining transgenic fruits for gene function analysis in ripening [17,18]. In addition, although the transient transformation technology in strawberry fruits provides a rapid tool to study gene function in ripening, some limits are existent in fruit transient transgenic system [2,4].
In addition to the widely-used leaf-disk transformation [14,19,20], other tissues are also used for gene transformation21-30. The mature and immature embryo-generated calli are used to develop transgenic plants in rice, maize, wheat and Barley [21][22][23]27]. Notably, a non-tissue culture-based method, oral dip transformation, is widely and e ciently used in Arabidopsis as a model system [24][25][26]. Notably, the Agrobacterium-mediated transformation of germinating seeds (AMTGS) of Arabidopsis, another nontissue culture approach is also early reported [31]. In recent years, the AMTGS of kenaf (Hibiscus cannabinus) seeds have been developed within 3 months with 6% transformation e ciency by the e cient, fast, reliable stable, non-tissue culture transgenic method [32]. Therefore, to rapidly identify gene function in strawberry, an AMTGS of diploid strawberry 'Hawaii-4' (Fragaria vesca) is to be established.
It is known that selection of a good reporter gene contributes to early identi cation of transgenic lines. In the past years, the H subunit of magnesium chelatase magnesium chelatase (CHLH), an enzyme involved in chlorophyll biosynthesis, is chosen as a good reporter gene, in that its silencing phenotype was clearly observed through chimeric phenotypes with yellow/white leaves [33][34][35][36]. In the present study, on the basis of CHLH function in chlorophyll biosynthesis, an Agrobacterium-mediated transformation of germinating seeds in diploid strawberry was rst established.

Results
Determination of seed imbibition extent and optimum kanamycin screening concentration To establish rapid and stable transgenic system in strawberry, the Agrobacterium-mediated transformation of germinating seeds in diploid strawberry was rst studied based on the previous reports [31,32]. We rst nd that there was no signi cant difference in the germination rate of strawberry seeds under different light culture conditions, including light culture, dark culture, and light culture after dark culture; the three treatments had a few, half, and 90% of seed germination respectively with 4-d, 5-d and 6-d imbibition (Fig. 1a).
To con rm optimal kanamycin concentration for screening, the sterilized strawberry seeds were cultured in MS selection medium with kanamycin concentration at 0, 25, 50, 75, 100 mg/L for 15 days. The results showed that when the concentration of kanamycin was 0 mg/L, 95% of the seeds germinated, the plants grew normally and the leaf color was dark green. Compared with no selection pressure, when the concentration of kanamycin increased to 25 mg/L, 91.7% of the seeds germinated, except for the growth of the root system is slightly slow, the germinated seedlings continued to show as the same as the wildtype growth; when the concentration of kanamycin was 50 mg/L, 41.7% of the seeds germinated, but the rst leaf of the germinated seedlings could not fully unfold; when the selection pressure was increased to 75 mg/L, only 20% of the seeds could germinate and the color of the two cotyledons of the seedlings were yellow and white; when the selection pressure increased to 100 mg/L, 5% of the plants germinated and died of browning (Fig. 1b, c). Taken together, 75 mg/L of kanamycin in MS-selection medium was used as an antibiotic selection.

Optimization of infection parameters
After 3-d culture in the selection medium, the GFP could be observed in the successfully-transgenic plants ( Fig. 2a, b). Similarly, the unsuccessfully-infected seedlings had no GUS expression (Fig. 2c, e), while the successfully-infected seedlings showed blue in the roots and true leaves with GUS expression (Fig. 2d, f).
Base on the observation of both GFP and GUS expression in transgenic seedlings, a 1.5-OD600 value of the Agrobacterium infection solution was optimum infection concentration (Fig. 2g). The seeds with just sprouted the radicles had highest infection e ciency (Fig. 2h) during 24-h infection (Fig. 2i). Taken together, the optimum parameters of Agrobacterium-mediated transformation of germinating seed in diploid strawberry are as follows: 75 mg/L of kanamycin in MS-selection medium, the seeds with just sprouted the radicles, 24-h infection in 1.5-OD600 value of the Agrobacterium infection solution, and then it was dark cultured at 24 °C for 3 d.
Establishment of a rapid and stable transgenic system by the Agrobacterium-mediated transformation of germinating seeds by a reporter gene of FvCHLH in strawberry The CHLH encoding the H subunit of magnesium chelatase magnesium chelatase involved in chlorophyll biosynthesis is chosen as a reporter gene to verify the effectiveness of the seed transformation system owing to its silencing phenotypes with yellow/white leaves easy to be observed [33][34][35][36].
Based on the optimal transformation conditions obtained above, we have transformed the germination seeds with just sprouted the radicles of diploid strawberry 'Hawaii-4' through the GV3101 strain carrying the pK7GWIWG2 (II) RR -FvCHLH vector. After 3-day infection under dark culture, the seeds were transferred to MS-selection medium. After 3 days, the seeds were screened by DsRed uorescence. After 2 weeks, the uorescent seedlings were transferred to quartz sand and watered using nutrient solution containing 75% kanamycin. The whole processes for transformation of germination strawberry seeds were shown in Fig. 3. We found that after observation with uorescence microscope, the uorescence of DsRed + plants showed an enhanced trend. The uorescence of 7-day-old plants ( Fig. 4c, d) cultured on MS-selection medium was signi cantly higher than that of 3-day-old plants ( Fig. 4a, b). DsRed uorescence was also expressed in the roots (Fig. 4e, f) and stolons ( Fig. 4i, j) of the plants after three months, but not in the wild-type plants of the roots (Fig. 4g, h) and stolons (Fig. 4k, l).
One month after transplanting, in comparison to the control, the FvCHLH-RNAi plants showed yellow spots (Fig. 5a, phenotype-1), or half green and half white in a leaf (Fig. 5a, phenotype-2). The PCR was performed to detect both kanamycin and DsRed gene fragments in the T-DNA inserted, transgenic plants. The expected 400-bp Kana and 964-bp DsRed products were detected in ve FvCHLH-RNAi plants but not in the wild-type plants (Fig. 5b, c). Compared with the wild-type plants, the expression of FvCHLH gene in the FvCHLH-RNAi plants was signi cantly down regulated (Fig. 5d). In the process of transformation, phenotype-1 plants accounted for 10% of the total infected plants, and phenotype-2 plants accounted for about 20% of the total infected plants, it can be concluded that the conversion e ciency is over 10% (Fig. 5e). These results showed that through Agrobacterium-mediated transformation of germinating seed, the pK7GWIWG2 (II) RR-FvCHLH vector was successfully transformed and inhibited FvCHLH expression, and the FvCHLH-RNAi transgenic plants have chlorosis phenotypes.
T2 seedlings of FvCHLH-RNAi plants retains kanamycinresistance T1 seeds of FvCHLH-RNAi transgenic plants were spread on the quartz sand soaked in the nutrient solution containing 75 mg/L kanamycin. In the presence of kanamycin, the seeds of sensitive to kanamycin cannot germinate or the cotyledons turn yellow 10 days after germination, and the true leaves could not develop. T2-resistant seedlings could germinate on quartz sand with kanamycin and continued to grow normally (Fig. 6a). The Kanamycin resistant bands and DsRed bands were detected in T2 generations (Fig. 6b). These results demonstrate that the transgenic plants can obtain by seed transformation and the relative genetic characteristics may be transferred to T1 progeny. A rapid and stable transgenic system by the Agrobacterium-mediated transformation of germinating seed is successfully established.

Discussion
Strawberry is a model plant for both Rosaceae and non-climacteric fruit ripening. Although much progress has been made underlying the molecular mechanism of strawberry fruit ripening using transiently-genetic fruit transformation and stably-genetic transformation systems, in which some limits are existed [1-14, 16, 37]. Notably, it is previously reported that the Agrobacterium-mediated seed transformation may facilitate the entry of Agrobacterium tumefaciens into the seeds to integrating a target gene into the embryo cells [31][32], similar to the pollen-tube mediated genetic transformation method widely used in Arabidopsis thaliana [24,30]. Reference of the seed transformation method [31][32], we rst establish the Agrobacterium-mediated transformation of the germination seeds in diploid strawberry by a series of optimization conditions. First, some factors key to transformation e ciency were con rmed, including seed imbibition time, suspension Agrobacterium concentrations, Agrobacterium infection time, kanamycin concentration in selection of transgenic plants. (1) Full imbibition is requisite for seed transformation. It is early reported that in Arabidopsis seed transformation, 12-h imbibition has the highest transformation e ciency, less than 9 h showing no transformants [31]. In strawberry, we demonstrate that seed imbibition to reach both testa rupture and, radicle emerging is vital for transformation e ciency (Fig. 2). In addition, Agrobacterium concentrations and infection time are also critical to transformation e ciency [38]. We also con rm that 24-h infection time has maximum transformation e ciency, in consistent with the previous reports in seed transformation [31,32]. Generally, in the conventional leaf disk method, Agrobacterium with concentration at 0.3-0.5 of OD600 has high vigor and is used as an optimal concentration for infection [39], in that a higher Agrobacterium concentration causes harm to explant differentiation including browning even death, and a lower concentration is not enough Agrobacterium for T-DNA integration of explants with lower transformation e ciency. In the present study, we nd that the germination seeds were more tolerant to Agrobacterium tumefaciens, and OD600 of 1.5 is an optimal concentration for highest transformation e ciency (Fig. 2). Finally, the appropriate concentration of kanamycin is also pivotal for selection of transgenic positive plants. In Arabidopsis seed transformation, T1 and T2 generation positive seedlings could be selected on medium containing 100 mg/L kanamycin [31]. In Kenaf seed transformation, F0 and F1 plants were successfully obtained on medium containing 50 mg/L kanamycin [32]. In the present study, we nd that strawberry seedlings were selected on medium containing 75 mg/L kanamycin, in which positive transgenic plants could developed true leaves, whereas not in the control (Fig. 2). Taken together, we con rm high e ciency transformation parameters in the Agrobacterium-mediated transformation of the germination seeds of diploid strawberry, including radicle initially-emerging seeds, 1.5-OD 600 Agrobacterium, 24-h infection, and 75 mg/L kanamycin in transgenic plant selection.
In fact, how to early select positive transgenic seedlings is more important in genetic transformation. It is previously demonstrated that the reporter gene is a good strategy for early transformation selection, such as neomycin phosphotransferase gene (NPT ), green uorescent protein (GFP), red uorescent protein (DsRed), and β-glucuronidase (GUS) genes [31,40]. In the present study, apart from the GFP and GUS genes (Fig. 2), we also used the reporter gene encoding magnesium chelatase H subunit (CHLH) functioned in chlorophyll biosynthesis, in that when its expression is downregulated, resulting in yellow or white leaf phenotype and thus using as a good report gene [33][34][35][36]. We construct the recombination pK7GWIWG2 (II) RR-FvCHLH vector with two reporter genes, DsRed and CHLH, which used as early selection. On the basis of DsRed uorescence, it is easy to pick out early potential transgenic seedlings, which is further con rmed to be positive transgenic plants by kanamycin selection (Figs. 3 and 4). The rst selection by DsRed uorescence may bypass a large quantity of seeds used for kanamycin selection, not only saving more time and labor, but also and making it easier to obtain transgenic plants. One month after the transformation, through the reporter gene CHLH, we may observe the leaves of transgenic plants with a loss-of-green phenotype (Fig. 5). In all, through the two reporter genes, the Agrobacteriummediated transformation of strawberry seeds was successfully established.
To further con rm the Agrobacterium-mediated transformation of the germination seeds, we also carried out kanamycin resistance and PCR analysis in T2 seedlings (Fig. 6): (a) when strawberry seedlings were selected on medium containing 75 mg/L kanamycin, the true leaves of kanamycin-resistant T2 seedlings can develop normally, whereas the growth of wild-type seedlings and kanamycin-sensitive T2 seedlings was partially inhibited and true leaves failed to develop; (b) through PCR test, the integration of DsRed gene and Kanamycin resistance gene into the genome of kanamycin-resistant T2 seedlings have been con rmed, whereas in wild-type seedlings and kanamycin-sensitive T2 seedlings, corresponding gene integration has not been detected. By analysis of the transgenic characteristics of transformed progeny (T1 and T2 generation), we provide strong evidence to con rm that the Agrobacterium-mediated transformation of strawberry seeds was successfully established.
In summary, we have rst established a fast and e cient protocol for Agrobacterium-mediated transformation of strawberry seeds: (1) seed imbibition time (just to the point of radicle emergence) is a key step; (2) Agrobacterium concentration and the infection time are also critical for successful transformation; (3) the selection of kanamycin-resistant seedling takes 1 month, T2 generation transgenic plants are obtained within 4 months; (4) the transformation e ciency is 10%. Thus, this method can greatly shorten the experimental cycle and simplify the operation processes. Give that strawberry is a model plant for both Rosaceae and non-climacteric fruit ripening, the Agrobacteriummediated transformation of strawberry seeds is to be widely used and will facilitate fruit development research and breeding.

Conclusions
On the basis of the CHLH (the H subunit of magnesium chelatase magnesium chelatase) as a reporter gene and the germinating seeds with radicle appearance, the Agrobacterium-mediated transformation of germinating seeds (AMTGS) in diploid strawberry (Fragaria vesca) has rst been established, especially nding that appropriate Agrobacterium concentration and infection time are critical for the transformation. The successful transgenic kanamycin-resistant seedlings (T1) are gained within 1 month and T2 transgenic plants within 4 months. In all, this new method is to provide a powerful tool for rapid identi cation of gene function in strawberry.

Plant materials and seeds germination
The diploid strawberry 'Hawaii-4' (Fragaria vesca) plants were grown in green house at 25 ± 1 ℃. The seeds were taken from ripe fruits and residual receptacle was removed, then the seeds were put at room temperature to dry for 1-2 d. The dried seeds are put into 2-mL tube and stored at 4 ℃ for use. Seeds were sterilized by 1% sodium hypochlorite for 8 min and washed with sterile water three times. The sterilized seeds were transferred onto MS-germination medium Vector construction and Agrobacterium tumefaciens culture Two Agrobacterium strains were used to optimize transformation parameters, namely GV3101 strain carrying pCAMBIA 1300 vector with a Green uorescent protein (GFP) reporter gene and carrying pBI121 vector with a β-glucuronidase (GUS) reporter gene. The RNAi recombinant plasmid was constructed using the Gateway system and target genes were silenced in strawberry fruit by intron splicing RNA (ihpRNA). The 465-bp FvCHLH were ampli ed and cloned into the pDONRTM 221 vector (Invitrogen, Beijing China) then into the pK7GWIWG2 (II) RR vector using Gateway LR Clonase II enzyme (Invitrogen 11791-020, USA) combination technology. The primers for constructing FvCHLH-RNAi were: forward, 5'-GGGGACAAGTTTGTACAAAAAAGCAGGCTTCGCTGCCAACAATCCATCT-3' and reverse, 5'-GGGGACCACTTTGTACAAGAAAGCTGGGTCTCTTCCAACTGTCTCAGCCA-3'. The pK7GWIWG2 (II) RR vector contains the red uorescence protein (DsRed) reporter gene and kanamycin resistance gene.

Determination of appropriate Kanamycin concentration for screening
The sterilized seeds were cultured on the MS-selection medium containing different concentrations of kanamycin with 0, 25, 50, 75 and 100 mg/L in order to determine most suitable antibiotic screening concentration. Twenty seeds were placed in each petri dish and the experiment were done with three repetitions.

Optimization of parameters related to infection e ciency
The pre-imbibed seeds were placed in Agrobacterium infection solution and co-cultured at 120 rpm, 28 ℃ for 24 h, then the seeds were placed in MS-germination medium and wrapped in tin foil at 25 ℃ for 3 d. Subsequently, the seeds were further placed in MS-selection medium [4.43 g/L MS, 2% sucrose, 75 mg / L kanamycin, 200 mg /L Timentin (Phyto Technology Laboratories T869), 0.8% agar, pH 5.8]. After 3 days, GFP/DsRed uorescent tag and GUS reporter gene were used for early screening and statistics. After 2 weeks, the seedlings with uorescence were transplanted into quartz sand and covered with cling lm for three days to prevent water loss, and watered with the kanamycin-added nutrient solution to allow the seedlings to grow under appropriate selection pressure at condition with a greenhouse at 24 ℃, light/dark 16/8 h.
To determine factors related to infection e ciency, we performed a series of experiments, including OD values (optical density at 600 nm, OD600 = 0.5, 1, 1.5, and 2), infection time (12 h, 24 h, and 36 h), and seed germination status (1 represents the seed that has not yet exposed the radicle; 2 represents the seed that has just sprouted the radicle; 3 represents that the seed has two cotyledons,). The optimal infection parameters were determined by investigation of GFP using stereo uorescence microscope (ZEISS-Axiocam 506 color) and GUS Histochemical Stain Kit (Solarbio, G3060). Infection e ciency was calculated by the proportion of GFP uorescent protein expression plants to the total infected plants or Gus expression plants to the total infected plants. During the infection process, 20 seeds were placed in each petri dish, and three replicates were performed.

Identi cation of transgenic plants by PCR
In order to con rm transformation of T-DNA in diploid strawberry genome, leaves of one-month-old transgenic strawberry plants with red light were collected and genomic DNA was extracted by Plant genomic DNA Extraction Kit (Aidlab Biotech, DN15, China). The leaves of wild type plants were used as control. The fragments of Kanamycin resistance genes and DsRed genes were ampli ed by PCR. Primers for PCR identi cation were: Kan forward, 5'-cgatagaaggcgat gcgctg-3' and reverse, 5'cgcttgatccggctacctg-3'; Red forward, 5'-cgcccttggtcaccttcagcttcac-3' and reverse, 5'caatgcagtgggacccacggttc-3'.

Real-time PCR
Total RNA from leaves of transgenic and wild-type plants was extracted using an OMEGA RNA extraction kit (OMEGA biotek, USA) according to manufacturer's protocols. To generate rst-strand cDNA, 400 ng of total RNA was reverse transcribed using the Trans kit (Transgen, China) according to manufacturer's protocols. The rst-strand cDNA was used as a template for PCR ampli cation for real-time PCR on a Light Cycler 96 Real Time PCR System (Roche Diagnostics GmbH, Mannheim City, Germany) using TransStart Top Green qPCR SuperMix (Transgen, China). The reactions of 10 µL contained 5 µL qPCR SuperMix, 0.25 µL forward speci c primer (10 µM; Sangon, China), 0.25 µL reverse speci c primer (10 µM; Sangon, China), 2 µL cDNA template and 2.5 µL ddH2O. Actin was used as a reference gene. Relative gene expression was analyzed by LightCycler® 96 SW 1.1. The primers used for real-time PCR were: forward, 5'-TGGGTCCCCTGATAAC-3'and reverse, 5'-CCAAATCCCACTGTCC-3'.
Kanamycin resistance test of T1 seeds Twenty seeds were obtained from a single inbred T1 generation, sprinkled on quartz sand, and irrigated with kanamycin (75 mg/L). After 10 days of culture, the sensitivity of the plants to kanamycin resistance was observed, and DNA molecules were analyzed for the resistant seedlings.  has not yet exposed the radicle; 2 represents the seed that has just sprouted the radicle; 3 represents that the seed has two cotyledons). i Effect of infection time on infection e ciency. Each treatment was treated with 20 seeds. The error bars represent the standard error (n=3). Different letters indicate statistically signi cant differences at P < 0.05 as determined by Duncan's test. Bars = 0.2 cm.