Many candidate genes have been discovered and tried to develop high-yield crop thanks to rapid cloning technics including NGS. More specific traits such as timing of flowering or senescence is often mentioned as an important factor in qualitative and quantitative seed development, since the timing of plant senescence is related to the retention and repositioning of nutrients. Plants also undergo aging processes with survival strategies for seasonal or unexpected changes in the external environment. The senescence process progresses according to the change of expression of several genes. The expression of photosynthesis and the basal metabolism-related genes is decreased, and the expression of cell death, stress-responsive genes, and hydrolase-related genes is increased (Hopkins et al. 2007; Lim et al. 2007). Regulation of the plant senescence can be a decisive contribution to the increase of crop productivity. It has been reported that suppression of leaf senescence in tobacco has increased productivity by up to 50% (Gan and Amasino 1995). In soybean crop, the productivity was increased by more than 30% due to senescence delay (Guiamét et al. 1990). However, many applications of this aging regulatory gene provided a problem with crop harvest timing. Therefore, it is necessary to develop a high-yield crop similar to that of a wild-type harvesting season (Kim et al. 2017a).

In plants, AT-hook proteins have been shown to have multiple roles in growth and developmental processes including leaf longevity, flowering transition, seed formation, and extended post-harvest storage life. These proteins are known to co-regulate transcription of genes through the modification of chromatin architecture. Recently, many genes related to the regulation of chromatin architecture have been discovered to induce specific phenotypic traits, such as plant organ size and seed yield increase (Lim et al. 2006; Street et al. 2008; Xiao et al. 2009; Kim et al. 2017a). The technology of the modulation of chromatin architecture using the gene encoding AT-hook protein provided the stability of chromatin during the aging process of the plants, thus showing the phenotypic characteristics of increased seed yield or delayed senescence in plants. These phenotypic features were also related to the expression level of the applied gene. The high level of gene expression showed strong phenotype of senescence delay. When the expression of gene was maintained at low level, it had a characteristic phenotype for increased biomass and seed yield. In particular, the general phenotypic characteristics of transgenic plants are similar to those of wild type.

Soybean [Glycine max (L.) Merr] is a major economic crop with abundant vegetable oil and protein, and the consumption is continuously increasing. Soybean transgenic plants with various useful genes have been developed using cotyledonary-node (CN) method via Agrobacterium-mediated transformation (Hinchee et al. 1988). Recently, the production of stable transformants using half-seed explants has been improved (Paz et al. 2006). This improved method does not need to germinate soybean seeds, so it is less time-consuming and can be handled simply. In addition, the use of additional thiol compounds, such as L-cysteine, sodium thiosulfate, and dithiothreitol (DTT), and the application of sonication and vacuum treatment provided a positive improvement in transformation efficiency. (Olhoft et al. 2003; Dan 2008; Verma et al. 2014; Kim et al. 2012, 2013, 2016, 2017b, and 2018).

In this study, we generated soybean transgenic plants overexpressing chromatin architecture-controlling ATPG8 and ATPG10 genes by Agrobacterium-mediated transformation and investigated yield components.

Vector construction and Agrobacterium preparation

ATPG8 (AT-hook protein of Genomine 8) and ATPG10 (AT-hook protein of Genomine 10) genes were provided by Dr. D.H. Lee (Genomine Inc., Korea) and were amplified by PCR using ATPG8 primers (forward 5′-ATGGATGAGGTATCTCGTTCTCAT-3′ and reverse 5′-AATCTTTCTGCCAGCAACGCAAGG-3′) and ATPG10 primers (forward 5′-ATGAAAGGTGAATACAGAGAGCAA-3′ and reverse 5′-TTAGTATGGCGGTGGAGCTCTGGC-3′), respectively. The desired destination vector, pB2GW7.0 (VIB-Ghent University, Ghent, Belgium) was used for ATPG8 gene, and vector pCSEN (provided by Dr. D.H. Lee) was used for ATPG10 gene for the vector construction. The plasmids, pB2GW7.0-ATPG8 and pCSEN-ATPG10 (Fig. 1A) were constructed and transformed into Agrobacterium tumefaciens strain EHA105 (Karimi et al. 2002), following the protocol described by Kim et al. (2012, 2013, 2016, 2017b, and 2018).

Soybean transformation

Mature Korean soybean seeds (Glycine max L. cv. Kwangankong) were used for soybean transformation by following the protocol described by Kim et al. (2012, 2013, 2016, 2017b, and 2018). Herbicide paint assay was used in two trifoliate leaves of T₀ plants to identify putative transformants. The upper part of the leaf surface was painted using a brush with the mixture of 100 mg/L PPT and Tween 20. The response to the herbicide was screened at 3–5 days after PPT application. Plants with herbicide-resistance were grown in greenhouse until maturity seeds were harvested (Fig. 1B).

Genomic DNA analysis of transgenic plants

Total genomic DNA was extracted from non-transgenic (NT) and transgenic plants using the cetyltrimethylammonium bromide. The polymerase chain reaction (PCR) was conducted to confirm the introduced genes using KOD FX (TOYOBO, Japan) according to the manufacturer’s instructions with a thermal cycler (Takara, Japan). To identify gene insertion in transgenic plants, primer sets were designed as shown in Table 1.

RNA analysis of transgenic plants

Total RNAs were isolated from NT and transgenic plants using Plant RNA Purification Reagent (Invitrogen, USA), and reverse transcriptase PCR (RT-PCR) was also performed using the RT-PCR Premix Kit (Genetbio, Korea) according to the manufacturer’s instructions. The primer sets used in the RT-PCR are shown in Table 1. The constitutively expressed TUB (primer sets, 5′-TGAGCAGTTCACGGCCATGCT-3′/5′-TCATCCTCGGCAGTGGCATCCT-3′) was used as an internal control to normalize the amount of RNA in soybean leaves.

Investigation of agronomic traits in GMO field

NT and transgenic plants including T₃ pB2GW7.0-ATPG8 and T₁ pCSEN-ATPG10 were planted in a seedling tray, and these seedlings were than transplanted in GMO field (Gunwi, Gyeongsangbuk-do) after the screening of herbicide painting assay. To evaluate yield components, agronomic traits including plant height, the number of branches per plant and the number of nodes per plant were investigated. The number of pods per plant and the total seed weight were also determined to investigate the relative yield of transgenic plants. Statistical analysis was performed using the Excel t-test program to identify significant differences.

Production of soybean transgenic plants via Agrobacterium-mediated transformation

To produce soybean transgenic plants expressing ATPG8 and ATPG10 genes, pB2GW7.0-ATPG8 and pCSEN-ATPG10 plasmids (Fig. 1A) were used for soybean transformation with half-seed explants of the Korean soybean cultivar Kwangankong (Fig. 1B). The transformation procedure followed the modified protocol described by Kim et al. (2012, 2013, 2016, 2017b, and 2018). After 3–5 days of observing the PPT treatment, a total of 3 pB2GW7.0-ATPG8 and 13 pCSEN-ATPG10 transgenic soybean plants were produced from each batch of experiment where 400 explants were transformed for each gene. T₁ seeds were harvested from all herbicide resistant plants in greenhouse. To verify the transgene integration in soybean transformants, genomic DNAs were isolated from T₀ pB2GW7.0-ATPG8 and T₀ pCSEN-ATPG10 transgenic plants. Only healthy and well grown transgenic lines from pB2GW7.0-ATPG8 (lines #1 and #2) and pCSEN-ATPG10 (lines #8 and #9) were analyzed by PCR for the introduced genes ATPG8 (798 bp) and ATPG10 (831 bp), respectively. In addition, Bar gene primer was also used to amplify the DNA fragment of 548 bp in size (Fig. 2). To confirm the expression of transgenes, RNAs were extracted from transgenic plants, and reverse transcriptase-PCR (RT-PCR) was conducted using ATPG8, ATPG10, and Bar primer sequences (Fig. 3). The transformed genes expressed in all transgenic lines as expected, while no expression was observed in NT plants.

Investigation of yield components in transgenic soybean

To examine agronomic characteristics of ATPG8 (T₃) and ATPG10 (T₁) transgenic plants, plant height, the number of branches per plant, nodes per plant, pods per plant, and total seed weight of plants were investigated, compared with those of NT plants in GMO field (Fig. 4). Transformation of ATPG8 gene into soybean carried earlier than ATPG10 gene, so generation gap exists between two genes. The plant height of ATPG8 transgenic lines (#1 and #2) and ATPG10 transgenic lines (#8 and #9) was higher than NT plants with statistical significance (P < 0.01). Their number of pods was also higher than NT plants by showing approximately 60% and 59% increase in ATPG8 line #1 and #2, and 6% and 31% increase in ATPG10 line #8 and #9, respectively (P < 0.05 in ATPG10 line #9, and P < 0.01 in ATPG8 line #1 and #2). In total seed weight, line #1 and #2 of ATPG8 transgenic plants showed about 47% and 74% greater than NT plants with statistical significance (P < 0.05 in line #1 and P < 0.01 in line #2). ATPG10 transgenic lines also showed yield increase when compared with NT plants, indicating that total seed weight of line #8 and #9 showed about 41% and 48% increase, respectively (P < 0.01). The result of yield components shown above is well coincident with typical characteristics for high yield soybean. In comparison of total seed weight, ATPG8 transgenic lines showed relatively higher than ATPG10 transgenic lines. The result came from different level of gene expression (Fig. 3). It was frequently observed that strong expression of AT-hook binding protein gene causes many troubles such as growth retardation, stay-green phenotype, no bolting, or yield decrease (Kim et al. 2017a). Direct comparison between two different genes driven by different promoters was not possible in this study. But selection of high yield lines turned out to be possessed such expression level.

An efficient and stable soybean transformation method is important for understanding the function of the transgene and for developing new soybean varieties by molecular breeding (Verma et al. 2014; Li et al. 2017). In this study, soybean transformation has been developed based on the half-seed (Paz et al. 2006) and cotyledonary-node (Hinchee et al. 1988) method. Instead of germinating soybean seeds, an alternative cotyledonary explant derived from mature seed following an overnight imbibition was used to improve soybean transformation (Paz et al. 2006; Kim et al. 2012, 2013, 2016, 2017b, and 2018). This half-seed method does not require complicated techniques to produce transgenic plants. There are several factors that affect the efficiency during the transformation process. An important factor to introduce foreign genes by Agrobacterium is tissue culture response of individual soybean cultivar. Multiple shoot formation is crucial for the production of soybean transformants after initial transfection. Multiple shoot formation during the transformation and selection process has a great impact on shoot elongation, a very important process during soybean transformation.

In this study, careful consideration was taken to obtain a certain level of efficiency. However, we observed substantial difference in two separate experiments with two genes. Especially, transformation with ATPG8 only obtained less than 1% of transformation efficiency. As a result, only 3 independent transgenic lines were regenerated from 400 explants. Similar results were often observed from various genes transformed in our lab (data not shown). We believed that each gene transformed has its own effect on tissue culture process through their unique physiological function in plant cell. Some genes may have positive effect on plant regeneration and other may not. Another reason to show different efficiency in transformation may come from different promoters. We did not directly compare two different promoters, constitutively expressed 35S promoter and inducible pSEN promoter. However, relatively weak pSEN promoter has better effect on tissue culture process, considering the fact that strong senescence driven by 35S promoter must lead to problematic growth of explants. To clearly understand and evaluate the effect of different promoters more results should be collected from field test.

Prevention of enzymatic browning and cell death in wounded area is also significant in Agrobacterium infection procedure at early stage of experiment. Addition of thiol compounds including L-cysteine, sodium thiosulfate, and dithiothreitol (DTT) in co-cultivation medium improved T-DNA delivery by inhibiting the activity of tissue browning of transformed cells (Olhoft et al. 2001 and 2003). By combining these factors with our modified transformation protocol, we improved the production of transgenic soybean plants with agronomically important genes. But this modified method also needs precaution to harvest satisfied result in soybean transformation. We hired all three additional treatments during Agrobacterium infection procedure. The modification is likely to cause high copy insertion of gene of interest into soybean. It is not likely recommended that strong expression of AT-hook protein genes in plant cell. Senescence has dual features; relatively low level of senescence will delay aging and help extending photosynthesis and cellular activity or higher level of senescence will give plant abnormal greening only.

Plant biotechnology is emerging as a new strategy to develop superior varieties of improved traits after discovering useful genes from various plants and introducing the genes into economically important crops. In recent years, functional genomics has been used to discover genes, and the introduction of these useful genes into economic crops will enable the creation of high value-added products. As trials, we repeatedly exploited the potential of AT-hook binding protein gene in practical use (Kim et al. 2017a). This study confirmed the high possibility of AT-hook binding protein gene. However, delicate manipulation of gene expression is needed because of complexity of senescence. More sophisticate vector design or precise promoter should be developed to use these genes.

This work was supported by the Next-Generation BioGreen 21 Program, Rural Development Administration (PJ01366501 granted to Y.S. Chung), and Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2017R1A6A3A11028883 granted to H.J. Kim) in the Republic of Korea.