Improving crops genome through genetic engineering of the key metabolic pathways

Soybean loss due to pests and pathogens is a serious problem worldwide. Soybean producers have few options to manage diseases caused by general pathogens where major genes for full resistance have not been discovered. The innate defense of soybean plants could be enhanced by improving content and composition of lignin by genetic engineering of the phenylpropanoid pathway. We used a novel technique of germ-line genetic transformation of soybean plants via natural pollen tubes as vectors. This technique uses Agrobacterium tumefaciens to mediate transfer of genes of interest to the zygote to introduce the key lignification genes (PtMYB4, PAL5, F5H, CAD1) into soybean genome. We observed 5.6% average transformation efficiency in the first generation of transgenic plants and in the second generation the presence of the transgene constructs was confirmed in more than 50% (for CsVMV/PtMYB4sens, 35SVTM/PAL5, C4H/F5H, CsVMV/CAD1 constructs) transgenic soybean lines. We confirmed the expression of the introduced genes at transcriptional level using RT-PCR and Northern blot. Functional analysis using lignin content determination and the activity of PAL5 and CAD1 enzymes demonstrated that the transgenes perform their function in planta. The proposed technique is effective and inexpensive and can be used to create novel stress and disease resistant soybean genotypes.


Introduction
The world-wide soybean (Glycine max, L.) production has increased several fold starting from 1960s' and currently soybean is one of the largest source of vegetable oil and animal feed (Pagano & Miransari, 2016). The demand for soybean will increase in future due to the rising population and the increased demand in both animal feed and products for human consumption. Growing of soybean meets a number of challenges such as biotic stress (micropathogens, pests and weeds). It is estimated that only in the US soybean diseases decrease the yield by up to 11% (Hartman et al., 2016).
Lignin is a phenolic compound providing strength to the secondary cell wall in vascular plants, that is assigned to a broad range of physiological processes participating in plant growth, providing the rigidity to the cell walls, the natural mechanical barrier and defense against pathogen penetration. It acts as a barrier against pathogen penetration as well as a toxins produced by the pathogens, promoting plant immunity Acta Scientiarum. Biological Sciences, v. 42, e52272, 2020

Plant material
Three cultivars of USA breeding: Jack and Jack x 4 (resistant to biotic stress), and Spencer (sensitive to micropathogens infection), 7 varieties of Kazakhstan breeding (Kazakhstan, Eureka, Vita, Tazhan, Zara, Perizat, Dannaya) were used for introduction of lignification genes. Ten biological replicates of each homozygous transgene and wild-type were grown simultaneously in a random block design, in the same environment in greenhouse. Plants were grown under long-day conditions (16 hours light, 28°C, and 65% humidity) to allow the development of health soybean plants.

Molecular vectors
Germ-line transformation was performed using A. tumefaciens, strain EHA 105 transformed with following vectors (Table 1). Transformation in A. tumefaciens or E. coli (for glycerol plasmid stocks and plasmid propagation) was performed using electroporation.

Preparation of Agrobacterium tumefaciens suspension for flower stigma pipetting
Single colony of transformed A. tumefaciens was transferred from Petri dishes to 3 mL of LB medium with 100 mg L -1 kanamycin (Sigma Aldrich), and incubated for 1 -1.5 days at 28°C on a shaker at 150 rpm. Cells were harvested by centrifugation at 4500 g for 10 minutes and resuspended in LB medium without antibiotics. Pipetting of flower stigma was performed with the bacterial suspension 10 10 cells mL -1 (equal OD=1 when measured at ƛ=600 nm using Beckman Coulter DU 800 spectrophotometer (USA)). Surfactant Pluronic F68 (0.001%) (Thermo Fisher) was added to bacterial suspension to activate vir-zone of T-plasmid and enhance transformation efficiency.

Agrobacterium tumefaciens suspension pipetting in soybean flower stigma
Soybean flower petals were separated and flower was cut in the carpel remove 1/3 of stigma with small scissors. About 2 µL of A. tumefaciens suspension transformed with a vector containing a gene of interest was placed in the center of the open flower to the stigma by a microsyringe or a micropippete (Figure 1a). After 30 minutes flowers were inspected and if the suspension was not enough the second pipetting was performed. Treated flowers were marked by color threads correspondent to the genes of interest (Figure 1b). Pipetting time was optimized to introduce transformed A. tumefaciens at 3-4 stages of soybean flower development, when the anthers were fully formed and pollen tubes are growing into mature, but not yet dividing zygote (Figure 1c). To synchronize pipetting time all soybean flowers opened the day before were removed at 8:00. It was determined that the stigma allocation of substances that stimulate the germination of pollen starts at 9:00, at around 9:15 pollen is attached to the stigma and the pollen germination begins and continues till 9:30 (Figure 1d-f). Therefore, the best time for pipetting in a greenhouse conditions (at 28°C) was 9:00 am -12:00 pm. In the field the time range can be 11:00 -13:00 the same day, when the flowers have opened and the resulting pollen tubes are able to pass the desired genetic material (at the time of the pollen activity maximum). The timing of pipetting in the field may be dependent on temperature and solar radiation.

Molecular analysis
Plasmid DNA was extracted from an overnight culture by using QIAprep Spin Mini prep Kit QIAGEN (Qiagen, Valencia, CA, USA). Genomic DNA was extracted from fresh mature but young transgenic soybean leaves using the method (Weigel & Glazebrook, 2009) and was used in PCR. Following primers were used (Table 2).  Transgene presence in plant leaves tissues was confirmed by the PCR reaction using Taq DNA polymerase (New England Biolabs, Ipswich, MA, USA) with denaturation at 94 °C (1 min.), annealing at 48°C or 54°C (40 s) for PtMYB4., PAL5, F5H, CAD1 constructs and extension at 72°C (90 s), for 35 cycles.
Multiplex PCR was performed similarly, with the exception of the temperature of the annealing stage (60°C). RNeasy Plant Mini Kit (Qiagen, Valencia, CA, USA) was used for extraction of total RNA from examples. DNA was digested with DNase I, using the TURBO DNA-free kit (Ambion/Life technologies, Carlsbad, CA, USA). cDNA was synthesized using oligo (dT) primer and reverse transcriptase enzyme, Promega GoScript kit, (Madison, WI, USA) for positive control or excluding reverse transcriptase enzyme as a negative control for genomic DNA contamination. To amplify PtMYB4, PAL5, F5H and CAD1 transcripts from cDNA same primers as for genomic DNA detection were used.

Southern and northern blot
Total genomic DNA of transgenic plants was digested with BamHI endonuclease (Fermentas, ThermoFisher, USA). Fragment size was 1,5 kB for PAL5 transformants and 1,2 kb for CAD1 transformants. DNA was precipitated in isopropanol and sodium acetate solution, precipitate was washed with ethanol and dissolved in water. After DNA electrophoresis DNA was transferred to a nylon membrane (Amersham, UK) using denaturation alkaline buffer (1 N NaOH + 0.4 M NaOH + 0.6 M NaCl). Membrane was dried for 30 min., DNA was fixed in UV-light (265 nm) for 5 minutes. Membrane was used for subsequent hybridization with P 32 -labeled probes prepared from PAL5 and GUS PCR products (as described in molecular analysis section) using NEBlot® Kit (New England Biolabs, Ipswich, MA, USA). After a prehybridization at 50°C hybridization was performed for 16h at 50°C in hybridization oven (Amersham, UK). Membrane was washed with 2 times in 1X SSC and 0.1% SDS at 50°С for 15 minutes and 3 times in 0.2X SSC and 0.1% SDS at 58°C. The intensity of radioactive label was measured using radioactivity counter. Exposure of the membrane with the X-ray film (Hyperfilm TMMP, Amersham, UK) was carried out with an amplifying screen. Same membrane was hybridized with the gus-probe.
For Northern blot total RNA was extracted from the same plants using guanidinium thiocyanate method. 5 ug RNA was subjected to agarose gel electrophoresis and was transferred to a nylon membrane (Amersham, UK). Membrane was prepared similarly to the membrane used for Southern blotting. Membrane was used for subsequent hybridization with P 32 -labeled probes. Probes were prepared from PCR products (as described in the molecular analysis section) using NEBlot® Kit (New England Biolabs, Ipswich, MA, USA). Membrane was washed with 0.2X SSC and 1% SDS at 58°C for 1 hour.

Lignin analyses
Lignin content and composition was performed as described in (Van Acker et al., 2013). The lignin composition was investigated with thioacidolysis (Robinson & Mansfield, 2009). The monomers involved in β-O-4-ether bonds, released upon thioacidolysis, were detected with gas chromatography (GC) as their trimethylsilyl (TMS) ether derivatives on a Hewlett-Packard HP 6890 Series GC system (Agilent, Santa Clara, CA, USA) coupled with a HP-5973 mass-selective detector. The GC conditions were as described in the protocol. The quantitative evaluation was carried out based on the specific prominent ions for each compound. Response factors for H (hydroxyphenyl), G (guaiacyl), and S (syringyl) units were taken from. The sum of H, G, and S is a good estimate of the total thioacidolysis yield and, thus, the condensation degree of the lignin polymer.

Enzyme assays
PAL activity was determined by the conversion of L-phenylalanine to trans-cinnamic acid essentually as described previously (Edwards & Kessman, 1992) with a few modifications: 300 mg of fresh leaves were ground in liquid nitrogen and extracted with 1200 µL of the extraction buffer, containing 0.05 M TrisHCl (pH-8.8), 0.5% ascorbate, 10% glycerol and 10 mM 13-mercaptoethanol. 100 µL of crude enzyme extract was combined with 400 µL of 0.05 M Tris buffer (pH 8.8), containing 0.2 mM phenylalanine as substrate. The reaction mixture was incubated for 30, 60, 120 and 180 min. at 37°C. The reaction was stopped by adding of 100 µL 0.5M HCl. Cinnamic acid was extracted with 1 mL of toluene and the OD was measured at 290 nm, with toluene used as a blank. The calibration curve was obtained using trans-cinnamic acid from Sigma. Total protein was measured using Bio-Rad Protein Assay (Bradford method).
For CAD1 enzyme activity determination plant tissue (300 mg) was ground in liquid nitrogen and crude extract was prepared using the extraction buffer containing Tris HCL 0.05M, pH 8.8, 10% glycerol and 10 mM β-mercaptoethanol. The CAD assay was performed according to (Mansell, Gross, Stöckigt, Franke, & Zenk, 1974) with a few modifications. The enzyme activity was determined by the conversion of alcohol to aldehyde (reverse reaction). The formation of hydroxycinnamaldehyde at 37°C was monitored at 400 nm using the following molar extinction coefficients: 1.78 x 10 4 M -1 cm -1 for coniferyl aldehyde and 14.7 x 10 3 M -1 cm -1 for sinapyl aldehyde (Hawkins & Boudet, 1994) . The reaction mixture contained in a total volume of 500 µL : 25 mM Tris-HCl, pH 8.8, 200 µl NADP (2mM), 200 µL (2mM) coniferyl alcohol/sinapyl alcohol and 100 µL crude extract. The same mixture, except 100 µL of TrisHCl buffer instead of the crude extract, was used as a blank. Protein concentrations were determined by the Bradford dye-binding assay.

Genetic transformation of soybean plants by Agrobacterium tumefaciens pipetting for introduction of genes of lignification
We performed A. tumefaciens-mediated transformation of 1070 flowers of 3 USA and 7 established and zoned in Kazakhstan soybean varieties. Overall we obtained 833 pods (78% efficiency) containing 2000 putative transgenic soybean seeds of zero generation (T 0 ) , expressing the genes of interest (PtMYB4, PAL5, F5H, and CAD1), respectively. Seeds forming efficiency on average was 187%. We did not observe any significant differences by size, shape or mass of pods and seeds between T 0 and control plants (data is not shown). The average number of seeds in a pod at T 0 was 2.5 seeds. T 0 seeds were planted to obtain plants of the first generation (T 1 ) for the further molecular analysis.

Molecular analysis of T 1 transgenic soybean plants
We analyzed the presence of the transgenes in the DNA and the presence of the transgene transcripts in T 1 soybean plants with PCR using genomic DNA and RT-PCR respectively. As a positive control for PCR of transgenic soybean plants, we used vector plasmid DNA. As a negative control for PCR, genomic DNA of untransformed wild parental plants was used.
The results of the standard PCR analysis of PtMYB4 gene insertion into soybean genome of T 1 plants shown in Figure 2a  analyzed putative transgenic soybean seeds we generated 112 T 1 transgenic plants with average efficiency of transformation 5,6 %. Importantly we did not observe any significant difference between the varieties in term of transformation efficiency, which proves that the developed method of germ-line transformation is not genome-specific. Data for other lignification genes is not shown.
Seeds of T 1 transgenic soybean plants were planted to obtain of second-generation plants (T 2 ) for confirmation of stable transformation by molecular and biochemical analyses.

PCR molecular detection of T 2 transgenic soybean plants
About 1000 T 2 transgenic soybean plants, obtained from confirmed by PCR T 1 transgenic plants , were selected randomly and screened by PCR and multiplex PCR (Figure 3). As shown in Figure 3 a-c, a large number of T 2 transgenic soybean lines demonstrates the presence of the gene of interest (PtMYB4 and PAL5) (Figure 3 a, b), and marker gene nptII (Figure 3c). PCR of soybean plants of second generation T 2 have confirmed stable insertion of valuable PtMYB4 and PAL5 genes and marker gene nptII (Figure 3 c) into soybean genome.
As shown in Figure 3d, all genetic elements of CsVMV/PtMYB4sens. construct: the valuable gene, reporter LicB -(lichenase), marker nptII, T-nos terminator confirmed to be inserted into soybean genome in second generation (T 2 ) of transgenic plants. Multiplex PCR shows the main genetic elements included in gene construction through alignment at annealing temperature 48°C for all genetic elements (valuable gene, reporter, marker, t-nos terminator) of the CsVMV/PtMYB4sens, 35SVTM/PAL5, C4H/F5H, CsVMV/CAD1 constructs. Multiplex PCR demonstrated the integration of the whole transgenic construction into genome of transgenic plant. This confirms the stable insertion of whole genetic construction containing the genes of interest into genome of soybean transgenic lines. Overall, we confirmed the stable integration of all elements of genetic constructions of CsVMV/PtMYB4sens., 35S/PAL5, С4H/F5H, Cs/CAD1 genes into soybean genome with the efficiency more than 50%, in second generation T 2 of transgenic soybean lines.  fig.) and marker gene nptII (bottom part of fig.). b: samples 12, 11a, 13, 7, 11, 4, 7, 1 ,9, 10 -

Southern and northern blot analyses of genomic DNA and total RNA
The same putative transgenic plant examples were subjected to the Southern and Northern blot analysis (Figure 4). Total genomic DNA digested with BamHI. Hybridization with the PAL-probe confirmed the presence of the transgenes in all tested lines except PBI 121 PAL5 #1 and #3, which is in agreement with the PCR data (Figure 4a). Further we hybridized the same membrane with the gus-probe allowed seeing the pattern of gene insertion and estimating the transgene copy number (Figure 4b). All tested lines have multiple copies of the transgenes. The fact that bands are much wider in PBI 101 PAL5 line suggests the presence of a tandem repeats. There were no bands in negative control lines.
The same transgenic lines were used for total genomic RNA isolation and a Northern blot analysis. Northern blot with total genomic RNA of PBI 101 PAL5 and PBI 121 PAL5 lines was performed with untreated samples and samples treated with water (Figure 4 c). Hybridization with the pal probe revealed the presence of the genomic PAL in all samples, including control and the presence of transgenic band in PBI 121 PAL5 line #4 (Figure 4 c). Hybridization of the same membrane with the gus probe resulted in a band at the same place, suggesting the presence of bisictronic mRNA due the expression of a transgene at the transcriptional level (Figure 4 d). This band appears much weaker in untreated samples due to the significantly lower amount of RNA. Hybridization with the nptII probe revealed the presence of the nptII band in all samples, except the control as expected (Figure 4 e). Thus, the soybean transgenic lines are capable of the expression of introduced genes at the transcriptional level.

Lignin content and composition
Further we analyzed the content of lignin in the cell wall in CAD1 and PAL5 transformants. Lignin content in control lines was 23.17% of the dry weight, 28.8% in Cs/CAD1 transformants and 29.99% in 35S/PAL5 transgenic lines. This confirms that lignin content is increased in the presence of the target lignification genes expression.
Furthermore, we analyzed lignin composition in these transgenic lines. We observed an overall increase of lignin monomer content in 35S/PAL5 transgenic lines compared to the control lines. In the Cs/CAD1 transformants we observed an increase of S-monomers content compared to the control lines ( Figure 5 a, b).
Additionally, we analyzed the activity of PAL and CAD enzymes expressed in the transgene plants to convert L-phenylalanine to trans-cinnamic acid and the conversion of coniferyl and synapil alcohols to the corresponding aldehydes, respectively. We observed a significant increase of PAL5 activity in PBI 101 PAL5 line #11 (Figure 5 c) as well as an increase of CAD1 activity in PBI 101 CAD1 line #5 (Figure 5 d). Correspondent with the transgene expression this proves the functional activity of the inserted genes of interest. All tested lines have multiple copies of the transgenes. gene. Error bars represent (n=8) (Weissgerber, Milic, Winham, Garovic, 2015)

Discussion
We developed a novel germ-line technique for soybean genetic transformation using natural pollen tubes to introduce the key genes of lignification into mature but not yet divided zygote. This technique was used to perform a genetic transformation of 1070 flowers of 3 USA and 7 established and zoned in Kazakhstan soybean varieties. We obtained 833 pods (78% efficiency) with 2000 putative transgenic soybean seeds of zero generation (T 0 ). Seeds forming efficiency on average was 187%. We confirmed the expression of the key lignification genes as well as the expression of all elements in the constructs (CsVMV/PtMYB4sens., 35S/PAL5, С4H/F5H, Cs/CAD1) using PCR, RT-PCR, multiplex PCR in all 10 soybean varieties with the average efficiency of 5.6%. In the second generation of transgenic plants T 2 we observed a stable integration of all elements of genetic constructions of these genes into soybean genome with the efficiency more than 50% . All tested lines of PAL5 transformants have multiple copies of the transgenes as verified by Southern Blot and confirmed expression of transformed genes into soybean genome. Northern Blot analysis demonstrated expression of introduced genes (PAL, gus, nptII) at the transcriptional level. We observed the increase of lignin content from 23.17% of the dry weight in control to 28.8% in Cs/CAD1 transformants and 29.99% in PAL5 transformants. Additionally, PAL5 transformants demonstrated elevated levels of all lignin monomers (H-, G-, and S), and observed an increase of S-monomers in CAD1-transgenic lines compared to the control.
So, we have developed a novel soybean germ-line genetic transformation technique using natural pollen tubes as vectors for the Agrobacterium-mediated introduction of target transgenes into soybean genome. This method is based on previously described method of pollen-tube pathway transformation which is successfully used for crops like cotton or rice (Ali et al., 2015), however seem to be not efficient and irreducible in soybean (Shou et al., 2002). The major difference of the used method is using A. tumefaciens to introduce the vectors containing the genes of interest into the newly fertilized but not yet dividing zygote cell via pollen tubes.
The method has a high efficiency of T 1 plants generation containing the genes of interest (5.6% on average, compared to 2.8% dependent on genotype in transformation in cotyledonary nodes (Jia et al., 2015), or 3.8 (1.4-8.7%) (Paz et al., 2006). We confirmed the presence of the transgenes using standard molecular biology techniques such as PCR, RT-PCR and Southern and Northern blotting. One of the important findings is that the transformation efficiency seems to be independent on the genotype which is a big advantage of the method. However, further investigations on the efficiency of A. tumefaciens-mediated transduction using natural pollen tubes should be performed, including the selection of the most transductable genotypes and more rigorous analysis of the selected plants. We found several important advantages of the suggested method. Firstly, this method is similar with natural pollen hybridization and transgenic plants are generated within a short period of time (90-120 days). No tissue culture and plant regeneration steps are required which makes this technique much less expensive and simpler compared to the conventional approaches. The plants derived in this study are able to produce normal health seeds, comparable with control plants. We observed that the transgenic plants are healthy and are very similar to the plants derived using conventional breeding. Also, due to the transformation of the germ-line these plants are less likely chimeric therefore the genes of interest are transmitted to the next generation.
Importantly, as the method uses only single and simple transformation step it can be performed in greenhouses or in the field. This can greatly reduce the soybean transformation expenses on tissue culture and recovering steps thus the overall cost of transgenic plant production.
The described method was used to produce soybean plants with altered lignin concentrations by introduction of the genes related to the lignin synthesis. We created a number of transgenic cell lines stably expressing PtMYB4, PAL5, F5H, and CAD1. Importantly, we observed the an increased amount of lignin in the plant lines harboring PAL5 and CAD1 genes which are required for proper lignin formation. This demonstrates that the used technique is capable of producing transgenic lines with the desired phenotype. The next step is to determine if the changes in lignin content and lignin composition will provide the resistance to various pathogenes. Earlier it was shown that soybean plants with low stem lignin concentrations are more resistant to sclerotinia stem rot -one of the most common soybean fungal diseases (Peltier, Hatfield, & Grau, 2009). Therefore more studies should be performed to characterized the resistance profile of the derived transgenic plant lines. Other important line of the further studies is to identify the transgenic constructs that will increase soybean resistance and will not compromise plant qualities such as yield and nutritional value.

Conclusion
In summary, we developed a novel simple and effective germ-line transformation method for soybean genetic transformation and were able to introduce genes related to lignification which may be important for soybean biotic stress and disease resistance improvement. This simple and effective genetic transformation technique is expected to be beneficial for the Kazakhstan and US soybean growers and also world-wild and will help to generate great variation in phenotype, increase of biodiversity and establishment of different valuable soybean genotypes with enhanced stress tolerance and productivity for breeders.