Improvement of Soybean Agrobacterium-Mediated Transformation Efficiency by Adding Glutamine and Asparagine into the Culture Media

As a genetically modified crop, transgenic soybean occupies the largest global scale with its food, nutritional, industrial, and pharmaceutical uses.Efficient transformation is a key factor for the improvement of genetically modified soybean. At present, the Agrobacterium-mediated method is primarily used for soybean transformation, but the efficiency of this method is still relatively low (below 5%) compared with rice (above 90%). In this study, we examined the influence of l-glutamine and/or l-asparagine on Agrobacterium-mediated transformation in soybean and explored the probable role in the process of Agrobacterium-mediated transformation. The results showed that when the amino acids l-glutamine and l-asparagine were added separately or together to the culture medium, the shoot induction frequency, elongation rate, and transformation frequency were improved. The combined effects of l-glutamine and l-asparagine were better than those of l-glutamine and l-asparagine alone. The 50 mg/L l-glutamine and 50 mg/L l-asparagine together can enhance the transformation frequency of soybean by attenuating the expression level of GmPRs (GmPR1, GmPR4, GmPR5, and GmPR10) and suppression of the plant defense response. The transgene was successfully transmitted to the T1 generation. This study will be useful in genetic engineering of soybean.


Introduction
Soybean (Glycine max (L.) Merrill) is one of the world's most important oil and protein crops, serving as a food source for humans and for animal forage, and it has industrial uses [1]. Transgenic soybean is a genetically modified (GM) crop that occupies the largest global scale with its food, nutritional, industrial, and pharmaceutical uses. From 1996 to 2015, approximately 1030 million hectares of GM soybean were planted and more than 50 billion US dollars were generated as income to farmers. In 2015, approximately 92.1 million hectares of GM soybean were planted around the world in 11 countries, which accounted for 51% of all GM crop hectares and 83% of the soybean grown all over the world [2].
The frequency was enhanced by 10-15% in media supplemented with L-asparagine or L-glutamine. The shoot induction frequency was above 90% when media was supplemented with both L-asparagine and L-glutamine. Compared with no L-asparagine or L-glutamine was added to the media, the frequency was enhanced remarkably when media was supplemented with both L-asparagine and L-glutamine (Table 1). Three independent experiments were performed for each treatment. No. of explants: the total number of explants for the three experiments; No. of explants with shoots: the total number of explants produced shoots for the three experiments. The explants with shoots were defined as the shoots with green leaves after 21 days transferred to selection media. The shoot induction frequency is expressed as the mean ± standard deviation. The different small letters represent significant differences between treatments by ANOVA (p < 0.05). (-) represents no L-amino acid added into the media. (+) represents L-amino acid added into the media.
All evaluations in transformed soybean plants were conducted using the expression cassette which encodes for a GUS reporter gene and Bar resistance gene for selection. The reporter transgene expression (GUS) of multiple shoots was tested and classified according to staining intensity as measured by ELISA. According to the GUS activity, the GUS staining was classified into four classes: negative (−) (Figure 1a-d), the GUS activity was below 0.45 U/g; low (+) (Figure 1e-h), the GUS activity was around 0.45 to 0.6 U/g; medium (++) (Figure 1i-l), the GUS activity was around 0.6 to 0.7 U/g; and high (+++) (Figure 1m-p), the GUS activity was around 0.7 to 0.8 U/g. For each treatment, 30 shoots were evaluated for GUS staining as classified in Figure 1 and GUS activity as measured by ELISA.
As shown in Table 2, in the presence of L-aspargine the percentage of GUS activity was 66.7%, while in the presence of L-glutamine activity rose to 70%, in comparison to shoots in the absence of L-amino acids tested. In the presence of L-asparagine and L-glutamine overall levels of GUS-positive shoots was 70%. It suggested that L-asparagine and L-glutamine could increase the frequency of highly-expressing GUS shoots.  For each treatment, 30 shoots was for GUS staining. −: negative, GUS activity was below 0.45 U/g; +: low GUS staining, GUS activity was around 0.45 to 0.6 U/g; ++: medium GUS staining, GUS activity was around 0.6 to 0.7 U/g; +++: high GUS staining, GUS activity was around 0.7 to 0.8 U/g. 0.1 g tissues were used to measure GUS activity by ELISA Kit. The GUS activity (U/g) is determined by comparing the Optical Density of the samples to the standard curve generated from standard with kit. Percentage % = No. of shoots in each class/total number × 100%. a Total GUS-positive shoots means the shoots with GUS staining, including +, ++ and +++. (-) represents no L-amino acid added into the media. (+) represents L-amino acid added into the media.
2.2. L-Asparagine and L-Glutamine Improved the Elongation Efficiency and Prolonged Transfer for Shoot Elongation (SE) After selecting for shoot induction, the explants produced multiple herbicide-resistant buds. The multiple buds without cotyledons were transferred to the shoot elongation medium (SEM) and subcultured for 3-5 times. Therefore, we recorded the number of elongated plants cultured every time (Table 3). When the media with no L-asparagine or L-glutamine added, 28.4% of the elongated shoots were produced in the first time (SE1); 37.8% and 33.8% of the elongated shoots were produced in the second and third times (SE2 and SE3), respectively. There were no elongated shoots in the fourth and fifth times (SE4 and SE5). When the media was supplemented with L-asparagine, 34.6% of the elongated shoots were produced in SE1; 43.8% and 19.1% of the elongated shoots were produced in SE2 and SE3, respectively. Only 2.5% of the elongated shoots were produced in SE4. When the media was supplemented with L-glutamine, 13.8%, 37.1%, 34.5%, 11.2%, and 3.4% of the elongated shoots were produced in SE1-SE5, respectively. When both L-asparagine and L-glutamine were added to the media, 13.3%, 57.7%, 19.5%, and 9.5% of the elongated shoots were produced in SE1-SE4, respectively. At this stage, after the length of the shoots had elongated to more than 4 cm, the shoots were cut from the multiple buds and then transferred for rooting. The more elongated shoots available, the more transgenic plants can be cultured. Table 3 shows the total number of elongated plants and the elongation rate. When the media with no L-asparagine or L-glutamine added, the elongation rate was approximately 20.2%. When the media was supplemented with L-asparagine or L-glutamine, the elongation rate was approximately 39.8% and 26.6%, respectively. When both L-asparagine and L-glutamine were added to the media, it was evidently enhanced and reached to 41.6%. It suggested that when supplemented with either L-asparagine, L-glutamine or in combination, the elongation rates were significantly higher than in the absence of the L-amino acids.

L-Asparagine and L-Glutamine Improved the Transformation Frequency
We cut leaves from the elongated shoots to detect the positive shoots, and verification was performed using GUS staining, PAT strips and leaf painting ( Figure 2). Strips were used to determine the presence of PAT protein of transgenic plants, and leaf painting was used to check the resistance of bar gene. When the media with no L-asparagine or L-glutamine added, 74 elongated shoots were detected, and among them, 12 shoots tested positive by PAT strips. The positive rate of the elongated shoots was approximately 16.2%. The positive shoots were 14.8% and 36.0% in media supplemented with L-asparagine or L-glutamine, respectively. 46 of 180 elongated shoots were tested positive, and the positive rate of the elongated shoots was approximately 30.2% in when media was supplemented with both L-asparagine and L-glutamine ( Table 4). The leaf painting and GUS staining were performed in order to corroborate results between the two methods and there was no difference between methods, so only PAT results are presented in Table 4.   From the detection of the elongated shoots, we calculated the transformation frequency. As shown in Table 4, the transformation frequency was approximately 3.5% with no L-asparagine or L-glutamine was added to the media. The transformation frequency values were approximately 5.9% and 8.8% in media supplemented with L-asparagine or L-glutamine, respectively. The transformation frequency reached 11.1% in when media was supplemented with both L-asparagine and L-glutamine. The results indicated that L-asparagine and L-glutamine could remarkably enhanced transformation frequency.

The Transformation Frequency in Four Soybean Cultivars Utilizing the New Conditions
According to the results of the different treatments, the transformation frequency was the highest when media was supplemented with both L-asparagine and L-glutamine for cultivar Jack. Four soybean cultivars Jack, Williams 82 (standard cultivar), Zigongdongdou (early-maturity cultivar) and Heihe 27 (late-maturity cultivar) were infected with Agrobacterium and cultured on the improved medium supplemented with both L-asparagine and L-glutamine. The media with no L-asparagine or L-glutamine was the control. Shoots were regenerated with 6 mg/L glufosinate for all four cultivars. Under the new media supplemented with both L-asparagine and L-glutamine, the shoot induction frequency in Jack reached 92.3%, and in Williams 82 and Heihe 27, it was above 85%. The shoot induction frequency was approximately 62.7% in Zigongdongdou. Compared with the media with no L-asparagine or L-glutamine, the shoot induction frequency was enhanced in all cultivars tested. In addition, the transformation frequency was also improved. The transformation frequency in Jack was the highest. In Williams 82 and Heihe27, the transformation frequency was above 8%. It was lower in Zigongdongdou ( Table 5). The results demonstrated that the improved conditions can be applied to more soybean cultivars. Table 5. Shoot induction frequency and transformation frequency in four soybean cultivars.  Three independent experiments were performed for each genotype. The shoot induction frequency and transformation frequency were expressed as the mean ± standard deviation. The different small letters represent significant differences between treatments by ANOVA (p < 0.05). (-) represents no L-amino acid added into the media. (+) represents L-amino acid added into the media.

Analysis of T1 Plants
Five independent T0 transformants of each treatment were randomly selected, and the T1 plants from the five independent T0 transformants were examined for the presence of the bar gene by PCR. Some PCR detection results were showed in Figure 3. The plants were positive by presence of the amplified fragment of 427 bp, and the negative plants were not amplified in any fragment ( Figure 3). The T1 plants were also examined the segregation ratios by Chi-square analysis ( Table 6). The results indicated that only three lines (line 1, line 7, line 17) showed a 3:1 ratio (bar + :bar -), and other lines were not fit to this ratio. We further confirmed the copy number of T1 transgenic plants (line 1-1, line 7-5, line 17-2, line 17-9) using Southern blot. The results showed the T1 plants in these three T0 lines had only one copy of the bar gene ( Figure 4).    Next, the seeds collected from all the other T0 transformants were planted and tested. If only one of the T1 plants was detected positive which was confirmed both by PCR and strip, we considered the T0 transformant was inherited to T1 generation. We determined that 66.7% (8 out of 12) in treatment with no L-asparagine or L-glutamine, 68.2% (15 out of 22) in treatment with only L-asparagine, 71.9% (23 out of 32) in treatment with only L-glutamine, and 73.9% (34 out of 46) in treatment with both L-asparagine and L-glutamine bar positive T0 transformants had inherited in T1 plants, respectively ( Table 7). Some of the T0 transformants in different treatments did not produce positive plants in T1 plants. Transgene positive T0 soybean plants did not necessarily pass the transgene to the next generation. There was no obvious difference in inheritance frequency between treatments. This indicated that L-asparagine and L-glutamine could not improve chimera.  46 34 73.9 ± 5.5 a Three independent experiments were performed for each genotype. The inheritance frequency was expressed as the mean ± standard deviation. The different small letters represent significant differences between treatments by ANOVA (p < 0.05). (-) represents no L-amino acid added into the media. (+) represents L-amino acid added into the media.

L-Asparagine and L-Glutamine Reduce Plant Defense Response by Regulating Pathogen-Related Genes
Some soybean PRs genes were reported that they were induced by wounding and wall glucan elicitor treatment which is a pathogen-derived general defense elicitor [38,39]. To evaluate whether the plant defense response was altered with different treatments, the expression of soybean PRs, GmPR1, GmPR4, GmPR5 and GmPR10 were analyzed during co-cultivation, recovery and selection culture.
The expression of these GmPRs in response to the treatments during different culture stage was different. The results showed that the expression of GmPR1 was weak with different treatments during co-cultivation and recovery culture, and strongly induced during selection culture in all treatments. The level of GmPR1 in treatment with no L-asparagine or L-glutamine was higher than that in other treatments with L-asparagine or/and L-glutamine. The expression level of GmPR4 was induced strongly in recovery and selection culture. The level of GmPR4 in treatment with no L-asparagine or L-glutamine was highest, and in treatment with both L-asparagine and L-glutamine it was lowest. The expression of GmPR5 has no obvious difference among treatments in co-cultivation culture, and it was obviously different between the treatment with no L-asparagine or L-glutamine and other three treatments in recovery and selection culture. The expression level of GmPR10 was the highest in treatment with no L-asparagine or L-glutamine in co-cultivation and selection culture. There was no obvious difference among treatments in recovery culture ( Figure 5). The remarkable difference in expression of these GmPRs in various treatments suggested that L-asparagine and L-glutamine could play a role in mitigating plant defense responses by attenuating the expression level of GmPRs. The expression level of GmPR1, GmPR4, GmPR5, and GmPR10 were analyzed during co-cultivation, recovery and selection culture.GmActin was used as an internal reference. The different small letters represent significant differences between treatments by ANOVA (p < 0.05).

Discussion
Agrobacterium-mediated soybean transformations have been previously reported, but the transformation frequency has remained low. The effects of the great complexity of soybean genotypes and different explants selections on soybean transformation have been studied by different laboratories [24][25][26][27]31,32]; however, none of the reported protocols appear to be sufficient.
L-Asparagine and L-glutamine are the two major nitrogen transport compounds in most plants [40]. Under normal growth conditions, the C/N ratio in the plant is probably the most important determining factor for whether asparagine or glutamine is selected as the major nitrogen. A high C/N ratio can lead to the production of glutamine for nitrogen transport, and a low C/N ratio lead to the production of asparagine for nitrogen transport [41,42]. In addition to providing carbon and nitrogen, glutamine is also involved in other cellular processes, including antioxidative stress, mTOR signalling, autophagy, and plant defence reactions [34]. The Agrobacterium-mediated transformation efficiency in perennial ryegrass (Lolium perenne L.) was highly improved by myo-inositol removal, cold shock pretreatment and L-glutamine supplementation, which enhanced Agrobacterium binding to the cell surface and decreased H 2 O 2 production [34]. As the pathogen competes for host nitrogen reserves during infection, the depletion of L-glutamine in infected tissues could activate plant defence responses. In addition, a direct application of glutamine on leaves can strongly inhibit H 2 O 2 production and could suppress cell death during pathogen infection [43].
In some labs, L-asparagine and L-glutamine have been added into the culture medium of at certain stages, like in shoot elongation medium or rooting medium [24,44]. However, in our study, we added the L-asparagine and/or L-glutamine from co-culture to rooting, the results showed that L-asparagine or L-glutamine added alone to culture media could enhance the regeneration rate, but better results were seen with the combination of L-asparagine and L-glutamine (Table 1). When media was supplemented with both L-asparagine and L-glutamine, the shoot induction efficiency was enhanced to 90%. The elongation rate was also increased when the medium had both L-asparagine and L-glutamine; the elongation rate was above 40%. For soybean, not all multiple shoots could elongate, only the elongated shoots can further grow into transformant plant. A high elongation rate is the basis for obtaining a high transformation frequency.
For the typical Jack cultivar, utilizing L-asparagine and L-glutamine in the culture medium increased the soybean transformation efficiency to approximately 11%, an increase from approximately 3.5-5.5% when no asparagine or glutamine was added to the media (Tables 4 and 5). In addition, this increase in overall transformation efficiency was applicable to other soybean cultivars, with the exception of Zigongdongdou (Table 5) which for reasons unknown is relatively recalcitrant to Agrobacterium transformation. Based on this improved transformation, the soybean genome editing (CRISPR/Cas9) was successfully achieved in our lab [45]. Although the improved transformation efficiency was still not high, we can see that it has the potential to enhance the transformation efficiency by improving the positive rate of elongated shoots. At present, through this transformation process, high shoot induction and elongation rates can be achieved, and elongated shoots can be produced, although many were not positive. Further optimization of glufosinate selection may be the key factor to improving the soybean transformation efficiency.
The increased transformation efficiencies resulted in an overall larger number of T0 and T1 plants under new conditions tested. The number of inheritance transformants was 8 for no supplement versus 34 for L-asaparagine and glutamine (Table 7). A larger number of genetically stable transformants can be screened by a piece of leaf for the presence of PAT (strip tested) and confirmed through either PCR or activity of the transgene (GUS) during early seedling development. Perhaps even growing T0 seeds on selection media could help screen transformants. So the T1 plants with the desired outcome can be quickly detected from the larger number of T1 plants.
Plant defense responses can lead to cell death at the sites of Agrobacterium infection and lower transformation frequencies. Our results showed that the expression of GmPRs in various treatments was remarkable difference during co-cultivation, recovery and selection culture. The expression level of GmPRs among treatments was relative low during co-cultivation culture. The expression level of GmPR4 and GmPR5 was induced during recovery culture, but it was higher in treatment with no L-asparagine or L-glutamine than in other treatments. During selection culture, the expression level of GmPR1, GmPR4, GmPR5 and GmPR10 was higher in treatment with no L-asparagine or L-glutamine than in other treatments. These results suggested that L-asparagine and L-glutamine could play a role in mitigating plant defense responses by attenuating the expression level of GmPRs. The transformation frequency was enhanced in treatments with L-asparagine or/and L-glutamine. And this enhancement in transformation frequency was likely attributable to attenuation in the expression of GmPRs and suppression of plant defense response by L-asparagine and L-glutamine.

Plant Materials
The soybean cultivars (Jack, Zigongdongdou, Williams 82 and Heihe 27) were utilized for Agrobacterium-mediated transformation in the present study, and Jack was used for the culture medium experiments. Healthy seeds were surface-sterilized by exposure to chlorine gas for 16 h as described by Di [46]. Sterilized seeds were placed in germination culture medium (GCM) containing 3.

Agrobacterium Strain and Vector
A. tumefaciens strain EHA101 containing the binary vector PTF102 was used in the experiments. The vector carried T-DNA containing the β-glucuronidase gene (uidA) driven by the 35S CaMV promoter and the bar gene as an herbicide resistance marker [44,47].

Infection and Co-Cultivation
Explants were prepared from one-day-old seedlings following the method described by Paz [24]. A longitudinal cut along the hilum was made to separate the cotyledons, and the seed coat was removed. The embryonic axis found at the junctions of the hypocotyls and the cotyledon was excised to obtain the half-seed explants. The explant cuttings were immersed in Agrobacterium for 2 h at 50 rpm. After inoculation, each of the 9 cotyledons were placed in solid co-culture medium (CCM) containing1/2 Murashige&Skoog Basal Salt Mixture, 3.9 g/L MES, 30 g/L sugar, 1mL/L Gamborgs Vitamin Solution, 150 mg/L DTT, 40 mg/L AS, 2 mg/L zeatinand 7 g/L agar, pH 5.4, with a piece of Whatman filter paper and then incubated at 22 • C in the dark for 5 days (Figure 6e,f).

Shoot Elongation and Rooting
After selection culture, the cotyledons and brown leaves were cut from the explants, and the remaining tissues were transferred to shoot elongation medium (SEM) containing 4.0 g/L Murashige&Skoog Basal Salt Mixture, 0.6 g/L MES, 30 g/L sucrose, 1 mL/L Gamborgs Vitamin Solution, 150 mg/L cefotaxime, 450 mg/L timentin, 0.1 mg/L 3-Indoleacetic acid(IAA), 0.5 mg/L Gibberellic acid(GA), 1 mg/L zeatin, 7 g/L agar, and 6 mg/L glufosinate, pH 5.6, and incubated at 28 • C (Figure 6k). The culture medium was changed every two weeks. Simultaneous with changing the SEM, the elongated shoots (5-8 cm) were cut from the base of the buds, and the stems were dipped in 1 mg/L Indole-3-Butytric acid (IBA) for 1 min, placed ina rooting culture medium (RCM) containing 1 /2 Murashige&Skoog Basal Salt Mixture, 0.6 g/L MES, 20 g/L sucrose, 1 mL/L Gamborgs Vitamin Solution, and 7 g/L agar,3 mg/L glufosinate, pH 5.7, and incubated at 28 • C for 7 days (Figure 6l). After root production, the plants were transferred topots and grown in the greenhouse.

Adding L-Asparagine and L-Glutamine
To determine the effects of L-asparagine and L-glutamine on Agrobacterium-mediated transformation efficiency, different experimental treatments were prepared. Starting with the co-cultivation medium, L-asparagine and/or L-glutamine was added to five different medium types (CCM, SIM0, SIM6, SEM, RCM): (1) with no L-asparagine or L-glutamine; (2) with 100 mg/L L-asparagine; (3) with 100 mg/L L-glutamine; (4) with 50 mg/L L-asparagine and 50 mg/L L-glutamine.

Detection of the Shoot Induction Rate and Elongation Rate
During the selection stage, the shoot induction rate was calculated as follows: shoot induction rate = explants with shoots/total explants for infection × 100%. During the elongation stage, the elongation rate was calculated as follows: elongation rate = total elongated shoots/total explants for infection × 100%.

Detection of Positive Plants and Transformation Frequency
The transgenic plants were verified by GUS staining, leaf painting, and strip analysis. The leaf was incubated in GUS staining buffer (50 mM sodium phosphate, pH 7.0, 0.5 mM potassium ferrocyanide, 0.5 mM potassium ferricyanide, 0.5 mg/mL 5-bromo-4-chloro-3-indolyl-β-D-glucuronide (X-Gluc), 0.1% Triton X-100 and 20% methanol) for 24 h at 37 • C and then washed in 95% ethanol before being photographed [48]. Next, 160mg/L of glufosinate was brushed on the leaf for a leaf painting test, and Liberty Link strips were used to detect PAT proteins in the transgenic plants. The plants used to calculate transformation frequency were all positive as determined by these three detection tests. The transformation frequency was calculated as follows: transformation frequency = the number of positive plants/the total explants for infection × 100%.

Detection of GUS Activity
0.1 g tissue samples were homogenized with 1 mL Phosphate buffer saline (PBS) buffer (pH 7.4) at 8000 rpm and 4 • C for 30 min, and the 50 µL supernatant was used for the reaction. The GUS activity was measured using an enzyme-linked immunosorbent assay (ELISA) kit (SU-B91122) according to the kit instructions. The GUS ELISA Kit includes a set of calibration standards. The calibration standards are assayed at the same time as the samples and produce a standard curve of Optical Density versus GUS activity. The GUS activity (U/g) is determined by comparing the Optical Density of the samples to the standard curve generated from standard with kit.

Plant Defence Analysis by qRT-PCR
The expression of GmPRs was used for analysis of plant defense by qRT-PCR. The 5 days co-cultivated explants, 7 days recovery explants and 21 days selected explants were harvested for RNA extraction. Total RNA was isolated from frozen tissue using TransZolUp Plus RNA Kit (TransGen Biotech, Beijing, China). For reverse transcription, 1 µg of total RNA was used to synthesize single-stranded cDNA using TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech). The GmActin gene was used as internal control. The GmPR1, GmPR4, GmPR5 and GmPR10 were used for analysis of plant defense. The primers used for GmActin amplification were 5 -GAGCTATGAATTGCCTGATGG-3 (forward) and 5 -CGTTTCATGAATTCCAGTAGC-3 (reverse). The primers used for GmPR1 amplification were 5 -ACACAGGTCGTTTGGGCTAA-3 (forward) and 5 -CAACAAAGTTGCCAGGGGGA-3 (reverse). The primers used for GmPR4 amplification were 5 -TGAACGCCGTGAGTGCTTAT-3 (forward) and 5 -CTGTATTCGTCACCCGGAGG-3 (reverse). The primers used for GmPR5 amplification were 5 -GTGCTTGGCGTTGAATACGG-3 (forward) and 5 -TGTGGGACACGCATTCTTGA-3 (reverse). The primers used for GmPR10 amplification were 5 -ACAACGTCATCCCAAAGGCT-3 (forward) and 5 -TCCATTCATTAACATCAGCCTCA-3 (reverse). qRT-PCR reactions were performed according to the three-step method using SYBR Green I dye and the ABI7500 instrument for qRT-PCR. For qRT-PCR, a total volume of 20 µL was used that contained 10 µL of SYBR Premix Ex Taq (2×), 0.4 µL of dye, 0.4 µL of 10 µM of the upstream or downstream primers, 2 µL of cDNA template, and 6.8 µL of ddH 2 O. qRT-PCR amplification using the standard three-step procedure for denaturation was performed as follows: 95 • C for 30 s, followed by 40 cycles of 95 • C for 5 s, and 60 • C for 30 s. The relative expression level was calculated using the comparative 2 −∆∆Ct method.

Molecular Analysis
PCR, strip and southern blot were used to analysis T1 generations of transgenic lines. Total genomic DNA was extracted from the leaves of the T1 transgenic plants using CTAB method. The bar gene detection in transgenic plants was demonstrated by PCR amplification of a 427 bp fragment using the primer pair 5 -GCACCATCGTCAACCACTACATC-3 and 5 -CAGAAACCCACGTCATGCCAGTT-3 . For southern blot analysis, the 10 µg genomic DNA was digested with HindIII, the restriction products were separated on 0.8% agarose gel and transferred onto a nylon Hybond-N + membrane (Roche, Mannheim, Germany) with a membrane transfer instrument (Model 785, Bio-Rad, Hercules, CA, USA). The bar (427 bp) PCR products were labelled with Digoxigenin and used as probes to hybridize with the digested DNA on the membrane. The hybridization and detection steps were performed according to the instructions for the DIG High Prime DNA Labeling and Detection Starter KitII (Roche, Mannheim, Germany).

Statistical Analysis
For all experiments, each treatment contained three replicates. The regeneration frequency, elongation rate and transformation frequency were expressed as the mean ± standard deviation, and the data shown represent the mean of three independent experiments. The data were analysed by ANOVA. The different small letters represent significant differences at p<0.05.

Conclusions
Adding 50 mg/L L-asparagine and 50 mg/L L-glutamine in culture medium enhanced shoot induction frequency and elongation rate in soybean transformation and finally improved the transformation frequency. The transformation frequency of Jack increased from a baseline of approximately 3.5-5.5% to 11% in the presence of both L-aparagine and L-glutamine. In addition, the improved conditions can be applied to more soybean cultivars. L-asparagine and L-glutamine could play a role in mitigating plant defense responses by attenuating the expression level of GmPRs. The enhancement in transformation frequency was likely attributable to attenuation in the expression of GmPRs and suppression of plant defense response. The transgene can be successfully transmitted to the T1 generation. This study may be useful in genetic engineering of soybean.