Synergistic Action of D-Glucose and Acetosyringone on Agrobacterium Strains for Efficient Dunaliella Transformation

An effective transformation protocol for Dunaliella, a β-carotene producer, was developed using the synergistic mechanism of D-glucose and Acetosyringone on three different Agrobacterium strains (EHA105, GV3101 and LBA4404). In the present study, we investigated the pre-induction of Agrobacterium strains harboring pMDC45 binary vector in TAP media at varying concentrations of D-glucose (5 mM, 10 mM, and 15mM) and 100 μM of Acetosyringone for co-cultivation. Induction of Agrobacterium strains with 10 mM D-glucose and 100 μM Acetosyringone showed higher rates of efficiency compared to other treatments. The presence of GFP and HPT transgenes as a measure of transformation efficiency from the transgenic lines were determined using fluorescent microscopy, PCR, and southern blot analyzes. Highest transformation rate was obtained with the Agrobacterium strain LBA4404 (181 ± 3.78 cfu per 106 cells) followed by GV3101 (128 ± 5.29 cfu per 106 cells) and EHA105 (61 ± 5.03 cfu per 106 cells). However, the Agrobacterium strain GV3101 exhibited more efficient single copy transgene (HPT) transfer into the genome of D. salina than LBA4404. Therefore, future studies dealing with genetic modifications in D. salina can utilize GV3101 as an optimal Agrobacterium strain for gene transfer.


Introduction
Dunaliella salina is a well-established natural producer of β-carotene in which synthesis of the pigment is enhanced under extreme environmental conditions, such as illumination intensity, temperature fluctuations and nutrition (depletion or limitation). Although research has yielded several expression systems for exploitation of the algae for its products, hitherto available systems possess several disadvantages in the production of recombinant products, nutraceutical by-products, biodiesel or other value added products. Regardless of the existence of several techniques for gene transformation in D. salina, such as electroporation [1], glass bead [2], biolistic gun [3], silicon carbide whiskers [4], protoplast transformation [5], microinjection [6] and PEG-mediated transformation [7], Agrobacterium-mediated transformation remains one of the most promising method for transient expression and stable integration of foreign gene into the algal host. Extensive work has been carried out in the development of efficient Agrobacteriummediated transformation protocols for microalgae with various types of inducers [8,9]. Unfortunately, the efficiency of the transformation is found to be low due to two potential reasons: poor induction of Vir genes for T-DNA transfer and species specificity of the Agrobacterium strains. Inducers are compounds which play a crucial role in the activation of the Vir gene which prompts genome-integration resulting in an overall increase in the rate of transformation efficiency [10]. Few investigations have reported Agrobacterium-mediated transformation in microalgae in the presence or absence of phenolic compounds acting as inducers [11,12] and some studies have also investigated pre-induction of Agrobacterium with Acetosyringone (AS) and plant-derived compounds for efficient transformation in microalgae under low pH [9,13,14,8].
Agrobacterium detects and responds to plant-derived sugars through a distinct signaling pathway involving VirA and a chromosomally encoded periplasmic protein (ChvE). ChvE mediates a sugar-induced increase in Virulence (Vir) gene expression through activities of the VirA/VirG two-component regulatory system [15][16][17]. This forms a part of an operon encoding an ABC-type transport system which is transcriptionally induced by sugar molecules [18]. ChvE mediates Agrobacterium chemotaxis in response to aldose monosaccharides such as galactose, glucose, arabinose, fucose, xylose, and other sugar acids which interact with VirA [19]. Expression of ChvE is regulated by the transcriptional regulator glucose/galactose-binding protein regulator (GbpR) in the presence of sugars [20]. Monosaccharide molecules involved in the activation of the Vir genes have resulted in the induction of ChvE to a maximum of eightfold during their absence; GbpR represses its expression [17]. From the understanding of these reports, we developed an efficient transformation protocol for Dunaliella by pre-induction of Agrobacterium strains (EHA105, GV3101and LBA4404) with D-glucose in the presence of AS.

Algal strain and culture conditions
Algal strain D. salina V-101 was acquired from the Centre for Advance Science Botany, Madras University, Chennai. Axenic cultures were maintained in De Walne's medium [21]. Initially, Algal cells were grown in TAP medium containing 2.0 M concentration NaCl. Later, Dunaliella culture was transferred every second day into fresh TAP medium, with gradual reduction of 0.1 M from preceding concentration of NaCl every time until cells were acclimated to 0.15 M NaCl. Then, Dunaliella cells were maintained in both liquid and solid TAP medium containing 0.15 M for the remaining experimental studies. Dunaliella cells were incubated at 24 ± 1°C in a growth chamber with an illumination of 22 μmol m −2 s −1 under a 16:8-h photoperiod.

Agrobacterium strains and vector transformation
Three strains of A. tumefaciens (EHA105, GV3101 and LBA4404) were obtained from the Indian Council of Agricultural Research (ICAR), Delhi and transformed using the binary vector pMDC45 (The Arabidopsis Information Resources). The characteristics of these Agrobacterium strains are explained in Supporting information, S1 Table. The vector pMDC45 harbors GFP as a reporter gene and, kanamycin and hygromycin resistant genes as selective markers which are driven by the 2xCaMV 35S promoter (Fig 1A). Selected single colonies were analyzed by colony PCR using a GFP-specific primer (S2 Table). The conditions for PCR were as follows: initial denaturation at 95°C for 5 min, 35 cycles of 94°C for 30 s, 60°C for 30 s and 72°C for 1 min, a final extension at 72°C for 10 min and hold at 4°C. PCR-positive colonies were inoculated into 5 ml of YEP medium containing 50 mg/L kanamycin and 25 mg/L rifampicin and incubated at 28°C for 48 h with shaking at 200 rpm. Agrobacterium strains were maintained as a glycerol stock in -80°C for further use.

Experimental study
Co-cultivation of D. salina and pre-induced Agrobacterium strains were divided into six different treatments: Group I, designated as control (without AS and D-glucose); Group II, Agrobacterium-induced with 100 μM AS alone; Group III, Agrobacterium-induced with 10 mM Dglucose alone; Group IV, V and VI, Agrobacterium-induced with 100 μM AS with D-glucose at a concentration of 5 mM, 10 mM and 15 mM respectively.

Induction of Agrobacterium strains
Single colonies of the three Agrobacterium strains were inoculated in LB media supplemented with 50 mg/L kanamycin and 25 mg/L rifampicin and incubated overnight at 28°C (OD 600 = 0.8 to 1). Following this, the Agrobacterium cultures were centrifuged at 4000 rpm for 5 min and the pellets obtained were re-suspended in induction TAP medium containing only 100 μM AS (Group II), only 10mM D-glucose (Group III) and 100 μM AS along with different concentration of D-glucose (5, 10 and 15 mM) for Vir gene induction at pH-5.2 (Groups IV, V and VI) [9]. The cultures were incubated at 25°C for 24 h and used for co-cultivation.

Co-cultivation of D. salina and Agrobacterium strains
Co-cultivation of Agrobacterium with D. salina and antibiotic sensitivity test for D. salina with different antibiotics were carried out as per the protocol [11]. Twenty milliliters of log phase cultures of D. salina (0.8 to 1.0 OD at 520 nm) were spread onto TAP medium and incubated under 16 h light irradiance for a week until algal lawn formations were observed. Two hundred microliters of pre-induced Agrobacterium strains harboring pMDC45 in TAP medium was plated over the lawn of D. salina cells and incubated under same conditions for 48 h. Following co-cultivation, cells were scraped and re-suspended in liquid TAP medium after which they were centrifuged at 3500 rpm for 2 min. Cells were then washed thrice with TAP medium containing 500 mg/L cefotaxime with 3 mg/L hygromycin and further centrifuged at 4000 rpm for 5 min. The cell pellets obtained were suspended in TAP liquid medium and spread onto a solid TAP medium containing 3 mg/L of hygromycin. After 4-7 weeks, individual colonies were picked and spotted onto a separate plate containing 3mg/L of the selective antibiotic, hygromycin.

Expression of GFP reporter gene in transgenic lines and wild-type
Selected transgenic lines and wild-type cells were treated with methanol:tetrahydrofuran (1:1) for 20 min to partially remove the chlorophyll content [11] and cells were washed thrice with distilled water. The cells were then examined under a Weswox optic LED fluorescent microscope FM-3000 at an excitation B filter wavelength ranges between 460 and 490 nm.

Analysis of putative transgene in Dunaliella cells by PCR
Ten milliliters of selected transgenic lines and wild-type cells (0.8 to 1.0 OD at 520 nm) were harvested by centrifugation at 3500 rpm and genomic DNA was isolated using the CTAB (Cetyltrimethylammonium bromide) method [22]. Algal cells were dissolved in 500 μl of extraction buffer containing 1 M Tris-Cl (pH 8.0), 0.5 M EDTA (pH 8.0), 5 M NaCl, 2% CTAB and 0.2% β-mercaptoethanol and incubated at 60°C for 20 min. After incubation, samples were centrifuged at 10,000 rpm at 4°C. Then, DNA was extracted from the pellet using ice-cold isopropanol and dissolved in TE buffer containing 10 mM Tris-Cl (pH 7.5), 1 mM EDTA. PCR was carried out using GFP and HPT specific primers to detect the presence of specific genes in transgenic lines and wild-type cells (S2 Table). PCR conditions were as follows: initial denaturation at 95°C for 5 min, 35 cycles of 94°C for 30 s, 60°C for 30 s and 72°C for 1 min, a final extension at 72°C for 10 min and termination at 4°C. The PCR products were evaluated bygel electrophoresis on a 1% agarose gel.

Southern analysis of transgenic lines and wild-type
Southern blot analysis was carried out to further confirm the copy number of gene integrated to the genome in transgenic lines using DIG-DNA Labeling and Detection Kit (Roche, USA) and by following the manufacturer's instructions. Genomic DNA was isolated from transformed and wild-type cells by CTAB method [22]. Samples containing 5 μg of genomic DNA and control plasmid pMDC45 were digested with KpnI. The 600 bp long fragment of HPT was used to prepare a DIG-labelled probe.

Identification of bacterial contamination in the transformed cells
Identification of Agrobacterium contamination in the genomic DNA of the transformed cells was analyzed using PCR for kanamycin resistant gene, (S2 Table) which lies outside of right and left border of pMDC45. PCR conditions included initial denaturation at 95°C for 5 min, 35 cycles of 94°C for 30 s, 60°C for 30 s and 72°C for 1 min, a final extension at 72°C for 10 min and hold at 4°C. pMDC45 binary vector was used as a positive control for the study.

Long-term stability analysis of transgenic microalgae
To study the stability of transgenes, transgenic Dunaliella cells were sequentially subcultured for more than six months in TAP medium in the absence of hygromycin. To select stable transgenes, the transgenic algal cells were exposed to different concentration of hygromycin. The survival rate was evaluated by calculating the optical density of hygromycin tolerant Dunaliella cells. To confirm the transgene, DNA was isolated and analyzed for the presence of hygromycin resistant gene in the genome of algal transformants by PCR using HPT gene-specific primers (S2 Table).

Statistical analysis
Three replicates per group were used in the experimental study and the values have been expressed in Mean ± SD. The analysis of variance (Two-way ANOVA) between Agrobacterium strains and each group (p<0.05) was conducted using GraphPad Prism (Version 5.0).

Results
Binary vector pMDC45 containing HPT resistant marker and GFP as reporter gene driven by CaMV35S promoter was used in this study (Fig 1). The binary vector was successfully transformed into Agrobacterium strains EHA105 (Lane 2 and 3), GV3101 (Lane 4 and 5) and LBA4404 (Lane 6 and 7) using the freeze-thaw method. Selected transformed cells were screened for the presence of pMDC45 in Agrobacterium strains using colony PCR, with GFPspecific primer (Fig 1B).

Effect of hygromycin and cefotaxime on the growth of Dunaliella salina
The growth of the algae in TAP medium containing 0.1 M NaCl with different concentrations of hygromycin is represented in the Fig 2A and 2B and the result indicates a complete growth inhibition at 3 mg/L hygromycin (Table 1). This concentration of hygromycin was used for the selection of transgenic lines, and it was observed that cefotaxime did not affect the growth of algae even at a concentration as high as 2000 mg/L in the TAP medium. Thus, 500 mg/L of cefotaxime is enough to eliminate the Agrobacterium strains from the transgenic lines (S3 Table).
Co-cultivation of D. salina and pre-induced Agrobacterium strains  combined with 100 μM AS (S1 Fig). From the three strains, LBA4404 presented a significantly higher number of transformants (181 ± 3.78 cfu per 106 cells) followed by GV3101 and EHA105 (EHA105>GV3101>LBA4404). The inducer AS and glucose alone do not cause any difference in transformation efficiency compared to control (without AS) during co-cultivation (Fig 3).

Expression of GFP reporter gene in Dunaliella cells
Transformants and wild-type strains were visualized under a fluorescent microscope for the detection of GFP reporter gene expression through Agrobacterium strains EHA105, GV3101, and LBA4404. All transformants emitted a distinctive greenish fluorescence in the cells which is characteristic of GFP. The greenish fluorescence was completely absent in wild-type cells which were treated similarly. However, autofluorescence (red color) was observed in wild-type, which may have been due to a partial evacuation of chlorophyll (Fig 4).

Analysis of putative transgene in the genome of Dunaliella cells by PCR and Southern blot
PCR amplification of transgene fragments of GFP (493 bp) and HPT (744 bp) using gene specific primers indicated the presence of the genes in the genomic DNA of randomly selected transformants (T1-T18) (Fig 5A) whereas no amplification was observed with DNA from  wild-type cells (W). Plasmid pMDC45 (P) containing both GFP and HPT fragments was used as a positive control. Thus, the result indicates the presence of transgene GFP ( Fig 5B) and HTP (Fig 5C) within the genome of transformed Dunaliella cells (Fig 5). The presence of the transgene in the genome of transformants was clearly indicated by PCR and the copy number was obtained using the Southern blot analysis.    Single KpnI site, which is at 2767 bp of T-DNA region and an expected a release of, is more than 5.1 kb in size as per the sequence of pMDC45 that contains the probe region of HPT. In our results, we observed two copies of the fragments in the transgenic lines mediated by LBA4404 and a single copy of the fragment in transgenic lines mediated by GV3101 and EHA105. No fragment was observed in the lane containing the DNA of the wild type cells digested with the same enzyme. The integration of the transgene copy number (HPT) in the transformed algal genome mediated by the three different Agrobacterium strains is given in Table 2.

Detection of Agrobacterium contamination in the transgenic lines
All the transformants used for molecular characterization had been additionally assessed for the presence of bacterial contamination. Preliminary screening of the transformants involved the production of sub-cultures in enriched medium (LB) to check for the possibility of bacterial contamination as bacteria tends to grow at a faster rate than the algal cells [23]. Additional screening was carried out by performing PCR using primer specific for kanamycin which lies external to T-DNA. No amplification was noticed in the genome of the transformants (T1-T18). This data indicates the absence of Agrobacterium strain within the transformants (Fig 7).

Long-term stability of DNA integration and gene expression in transgenic Dunaliella
Transgenic Dunaliella cells (0.8 to 1.0 OD at 520 nm) which were able to retain their phenotype and grow well when exposed up to 8 mg/L of hygromycin were also able to tolerate when hygromycin concentration was increased up to 16 mg/L, after a six months of sequential subculture in TAP medium (S4 Table). S2 Fig also shows the presence of HPT transgene in the genome of hygromycin resistant microalgae. Thus, the results indicated that stable integration and expression of heterologous transgene mediated by Agrobacterium strains without the addition of antibiotics. Table 2. Integration of number of transgene (HPT) in the genome of transgenic lines mediated by each Agrobacterium strains as analyzed by Southern blot hybridization. Transgenic lines with two copies  Transgenic lines with single copy  No result  Total transgenic lines   EHA105  0  15  3  18   GV3101  0  13  3  16   LBA4404  16  2  1  19 doi:10.1371/journal.pone.0158322.t002

Discussion
This study describes the procedure for the development of an efficient Dunaliella transformation system using pre-induced Agrobacterium strain with D-glucose and AS. Induction of virulence in Agrobacterium has been well-known and demonstrated with various low molecular weight compounds such as phenolic compounds (AS) and monosaccharides (glucose) [24]. He and coworkers [17] provided a direct evidence that ChvE specifically binds to the Vir geneinducing sugar D-glucose with high affinity in the presence of AS. The ChvE protein is homologous to several periplasmic sugar-binding proteins of E.coli. A further mutation in ChvE associated with the latter two phenotypes lies in two overlapping solvent-exposed site adjacent to the sugar-binding cleft. This may bring about an antagonistic impact on ChvEs interaction with its distinct protein. Currently available data indicate that under acidic conditions, plantderived compounds are able to trigger Vir genes which are involved in the transfer of T-DNA [25][26][27][28]. In Chlamydomonas, Agrobacterium-mediated transformation was achieved with the help of 1 mM glycine betaine with AS at a pH of 5.2 [9]. In our study, we also used different concentrations of D-glucose (5 mM, 10 mM, and 15 mM) with AS to trigger the Vir genes of Agrobacterium strains for infection in the Dunaliella cells.
The growth of D. salina was completely suppressed at a hygromycin concentration of 3 mg/ L of in TAP medium. Thus, the minimal inhibitory concentration of hygromycin for D. salina was much lower as compared to those reported in previous studies for D. bardawil (100 mg/L) [11], Porphyra yezoensis [29], Parachlorella kessleri [12] and Schizochytrium sp. [30]. This may be because of the facilitation of the growth of D. salina in a TAP medium with low concentration of salt (0.1M). It has been stated that a higher convergence of salt promotes the utilization of higher amount of hygromycin for the complete termination of D. salina cell development because of the inclusion of H + -ATPase [31]. However, the antibiotic sensitivity of many algal species has not been studied in depth and requires further investigation to pinpoint the exact mechanisms. 500 mg/L of cefotaxime was found to be inhibitory for Agrobacterium; however, this concentration did not affect the D. salina cells and thus was used for the elimination of the bacteria after co-cultivation. Co-cultivation studies revealed that the combined use of D-glucose and AS stimulate the virulence capacity of Agrobacterium strains and thus result in a significant increase in the transformation frequency of D. salina mediated by LBA4404, EHA105 and GV310 at the concentration of 10 mM D-glucose in the presence of 100 μM AS. But, Dglucose and AS alone do not cause any significant increase in the transformation rate. Several studies have reported that microalgae are able to secrete several small molecules, including phenolics into the culture medium during their growth [32][33][34]. It is likely that some of these molecules are capable of inducing the Agrobacterium Vir genes, explaining the transformation as well as Vir gene induction. Thus, the use of D-glucose with AS for pre-induction of Agrobacterium increases the transformation frequency up to 4 times that of Agrobacterium-mediated transformation in Dunaliella sp with or without AS (42 ± 3 per 10 6 cells) [11]. Despite extensive studies on Agrobacterium-mediated transformation of microalgae, the data inferred on transformation frequencies between investigations vary significantly [30,[35][36][37]. In our study, induction of the Vir genes in Agrobacterium strains by D-glucose appears to work synergistically with induction by AS.
The gene coding for the green fluorescent protein (GFP) from the Aequorea victoria has been utilized as a marker for gene expression [38] and in vivo protein localization in microalgae [39][40][41]. In this study, the expression levels of GFP in the cells also varied with the Agrobacterium strains. LBA4404 mediated Dunaliella transformants indicated a higher level of expression compared to GV3101 and EHA105 mediated transformants. Expression levels of the reporter gene may have varied due to random insertion of one or more transgenes into the genome of transgenic algae, [2]. Preliminary study revealed the presence of transgenes (HPT and GFP) in the genome of the transgenic lines. Transgene copy number may influence the levels of expression of integrated genes, gene silencing, and stability. Gene copy number is also an important parameter for plant biotechnology, since transgenic plants with single gene insertions are preferentially approved by regulatory agencies [42]. Thus, it results in a different number of transgenes in the HPT resistant transformants and this number depends on Agrobacterium infection on algae. Most of the transformants mediated by LBA4404 showed two copies of the transgene and those mediated by EHA105 and GV3101 showed a single copy respectively. Identical results from the prior studies reveal that single or multiple copies of transgenes can be seen from transformants mediated by EHA105, EHA101 and LBA4404 respectively in microalgae [11,30,43]. Fang and his coworkers studied the expression of sedoheptulose-1,7-bisphosphatase from C. reinhardtii in D. bardawil mediated by GV3101 but unfortunately, did not evaluate the transformation frequency of GV3101 [44]. In other algal species, LBA4404 mediated transformants contained a single-copy of T-DNA in the genome [9,35]. Differences in copy number obtained with either assay may have resulted from partial restriction digestion of genomic DNA, insertion of tandem repeats, re-arrangements or truncations of the inserted T-DNA [45][46][47][48]. A recent study indicated that the highest transformation rate was observed in GV3101 followed by EHA105, AGL1, and MP90 but the mortality rate was quite lower with the strain GV3101. Even EHA105 was more efficient in the transfer of single-copy transgene than GV3101 in the genome of tomato [42]. The outcomes presented here demonstrate that super virulent A. tumefaciens GV3101 strain seemed to have an ideal combination of transformation efficiency and capacity to create a single copy of transgene in Dunaliella sp. Thus, the study uncovers that Agrobacterium strains differ not only in their transformation proficiency and transgene copy number, but these values may vary from species to species. To our knowledge, this is the first report on pre-induction of Agrobacterium strains with D-glucose as an inducer in the presence of AS in D. salina.   Table. Growth and tolerance of wild and transgenic Dunaliella cells in the TAP medium containing different concentration of hygromycin. Hygromycin resistant phenotypes are able to grow when exposed upto 8 mg/L and also tolerant upto 16 mg/L of selective antibiotic containing medium. (DOCX)