Characterization of the pyruvate kinase gene family in soybean and identification of a putative salt responsive gene GmPK21

Background As a key regulatory enzyme in the glycolysis pathway, pyruvate kinase (PK) plays crucial roles in multiple physiological processes during plant growth and is also involved in the abiotic stress response. However, little information is known about PKs in soybean. Results In this study, we identified 27 PK family genes against the genome of soybean cultivar Zhonghuang13. They were classified into 2 subfamilies including PKc and PKp. 22 segmental duplicated gene pairs and 1 tandem duplicated gene pair were identified and all of them experienced a strong purifying selective pressure during evolution. Furthermore, the abiotic stresses (especially salt stress) and hormone responsive cis-elements were present in the promoters of GmPK genes, suggesting their potential roles in abiotic stress tolerance. By performing the qRT-PCR, 6 GmPK genes that continuously respond to both NaCl and ABA were identified. Subsequently, GmPK21, which represented the most significant change under NaCl treatment was chosen for further study. Its encoded protein GmPK21 was localized in the cytoplasm and plasma membrane. The transgenic Arabidopsis overexpressing GmPK21 exhibited weakened salinity tolerance. Conclusions This study provides genomic information of soybean PK genes and a molecular basis for mining salt tolerance function of PKs in the future. Supplementary Information The online version contains supplementary material available at 10.1186/s12864-023-09929-7.


Background
As an important metabolic pathway, glycolysis widely exists in higher plants and animals.Pyruvate Kinase (PK) is a key regulatory enzyme in the glycolysis pathway.It catalyzes the last step of glycolysis by transferring the high-energy phosphate group of phosphoenolpyruvates to adenosine diphosphate (ADP), and then produces ATP and pyruvate [1].Pyruvate then acts as an important substance in organisms to participate in multiple metabolic reactions [2].
In plants, there are two isoenzymes of PKs in cytoplasm and plastid respectively, namely PKc and PKp [3].They are different in physics, dynamics, immunology and expression regulation [3][4][5][6][7].The PKs acting as PKc or PKp were found to be involved in multiple development and growth processes.It was reported that several PK genes in Arabidopsis were involved in oil and fatty acid biosynthesis in seeds [8,9].In tobacco, absence of leaf PKc resulted in a striking decrease in root biomass and root: shoot ratios [10].Similarly, GhPK6 in cotton played an important role in regulating cotton fiber elongation [11].Compared with other plants, PK genes have been studied more systematically in rice.For instance, downregulation of OsPK1 causes dwarfism and panicle enclosure in rice [12]; OsPK5 is involved in rice glycolytic metabolism and GA/ABA balance for improving seed germination [13].Furthermore, several PK genes were reported to be involved in the grain filling of rice [14][15][16].
With the development of transcriptomics and proteomics technology, the expression of PK genes or the accumulation of PK protein in plants suffering from abiotic stress were found to be affected.AhABI4s negatively regulate salt-stress response in peanut.The transcriptomics and quantitative proteomics analyses showed that PKs and other genes/proteins were affected by silencing of AhABI4s in peanut leaf and root after NaCl treatment.Furthermore, AhABI4s were able to bind to the promoters of pyruvate kinase (PK) coding genes in vitro [17].In soybean, pyruvate kinase was increased under flooding, but gradually decreased during post-flooding recovery period at protein abundance, mRNA, and enzyme activity levels [18].In pepper, the expression of CaPKc1 gene was not only increased when the plants were infected by Tobacco mosaic virus (TMV), but also triggered by NaCl and wounding [19].In Brassica napus, PKs showed an increased abundance under freezing stress after cold acclimation [20].These reports suggest that PKs might be involved in the abiotic stress responses.However, the study of the PKs function in the abiotic stress tolerance is very limited.The only report is about a rice PK gene OsPK5.It could interact with OsSAP6, a member of stress-associated protein (OsSAP) gene family which could positively regulates saline-alkaline tolerance.Furthermore, overexpression of OsPK5 in rice enhanced tolerance to soda saline-alkaline stress at seedling stage [21].
Soybean (Glycine max [L.] Merr.) is the leading oilseed crop produced and consumed in the world [22].However, the production of soybean is severely threatened by the abiotic stresses, such as high salt [23].Although the PK gene family has been identified and analyzed in Arabidopsis and rice [24,25], little information is known about PK proteins in soybean.In this study, we performed genome-wide characterization of the soybean PK family genes against a reference soybean genome of a soybean cultivar, Zhonghuang13 [26] and analyzed their expression patterns under NaCl and ABA treatment.Furthermore, GmPK21, which was represented the most significant change under NaCl treatment was chosen for further study.The gene function analysis of GmPK21 showed that it had a negative regulatory effect on salt stress tolerance in Arabidopsis.This research lays a foundation for further function investigations of PK genes in soybean.

Identification and phylogenetic analysis of PK genes in soybean
We identified 27 soybean PK genes in this study (Table 1; Table S1).Those 27 soybean PK genes were named from GmPK1 to GmPK27.The predicted protein products of these 27 GmPK genes varied from 184 (GmPK3) to 582 (GmPK16 and GmPK24) amino acids.The protein MW (Molecular Weight) of these 27 GmPK proteins ranged from 20.06 kDa (GmPK3) to 63.68 kDa (GmPK20).Their predicted PI (isoelectric points) varied from 4.52 (GmPK3) to 8.01 (GmPK7).The protein sequences of 27 GmPK genes were shown in Table S1.
We then built a phylogenetic tree by analyzing the amino acid sequences with the PK proteins from soybean, Arabidopsis and rice.Based on the phylogenetic analysis and previous research [24,25], all of the PK proteins were divided into 2 subfamilies, PKc and PKp.The PKc subfamily was further divided into PKc-1 and PKc-2 clade.Similarly, the PKp subfamily was divided into PKp-α and PKp-β clade.As shown in Fig. 1, 11 GmPKs belonged to the PKp subfamily.Among them, 7 belonged to PKp-β clade and the other 4 belonged to PKp-α clade.The PKc subfamily contained 16 PKs.Of them, 7 were PKc-1 genes, and the other 9 were PKc-2 genes (Fig. 1).
Tandem and segmental duplications play important roles in the expansion of plant gene families [27].To reveal the expansion mechanism of the GmPK gene family, the gene duplication events of the soybean PK genes were investigated.We found that only 1 pair of soybean PK genes was tandem repeat.22 pairs of soybean PK genes were found to be segmental duplications (Fig. 3; Table S2).The nonsynonymous (Ka) and synonymous (Ks) substitution rates between these duplicated gene pairs were also calculated.The Ka/Ks of all the 23 duplicated gene pairs was found to be less than 1, suggesting that the soybean PK gene family might have experienced a strong purifying selective pressure during evolution (Table S2).
To detect the synteny of GmPK genes, a collinearity analysis between soybean and the other 2 plant species, including 1 monocot (rice) and 1 dicot (Arabidopsis) was performed.A total of 15 and 4 soybean PK genes showed a syntenic relationship with those in Arabidopsis and rice, respectively (Table S3).This finding suggests that soybean PK genes display a higher evolution divergence with monocotyledonous plants.Notably, we found 2 GmPK genes (GmPK19 and GmPK22) were collinear with the PK genes from both of the Arabidopsis and rice (Table S3), suggesting that they might play important roles during the evolution of GmPK genes.

Gene structure and conserved protein motif analysis of GmPK genes
By performing the multiple sequence alignment, we found that all of the 27 GmPKs shared the typical PK domain (Figure S1).Then the MEME program was used to detect conserved motifs in the GmPK gene family.10 distinct motifs (named motif 1-10) were identified (Fig. 4b, Figure S2).Among them, Motif 1, 2, 3, 6, 7 and 10 belonged to the PK domain, motif 4 (or 5), 7 and 8 belonged to PK_C domain.In general, proteins from the same subfamily or clade were characterized by a similar motif type and distribution (Fig. 4a and b).Almost all of the PKc (including PKc-1 and PKc-2 clade) subfamily members (except GmPK2, GmPK3, GmPK7 and GmPK8) contained motif 3, 7, 6, 10, 2, 1, 9, 4 and 8.All the members in PKp-α clade contained the motif 3, 7, 6, 10, 2, 1 and 4. We also noticed that the motif 8 and 9 can only be detected in PKc subfamily, and motif 5 was only observed in the PKp-β clade.
In order to examine the gene structures, the exonintron structures of the 27 GmPK genes were analyzed.As shown in Fig. 4c, the number of introns varied greatly between the clades of the same subfamily.The introns number of PKc-2 clade ranged from 8 to15.However, the introns number of PKc-1 clade members ranged from 1 to 4. Almost all of the GmPK genes in PKp-β clade contained 11 introns (except GmPK10 and GmPK17), while the GmPK genes in PKp-α clade only contained 5 or 6 introns.

Expression profiling of PK genes in different tissues of soybean
In order to investigate the transcript abundance of PK genes in different tissues, expression of the 27 GmPK genes covering 7 tissues of soybean was performed by qRT-PCR.Based on their expression pattern, the GmPK genes could be generally classified into 4 groups (Fig. 5).Genes in group 1 were barely detected in more than 5 soybean tissues.The genes in group 2 could be detected in more than 5 tissues, but the expression level was relative lower.There was a trend that all of those 3 genes in group 3 were highly expressed in all of the 7 tissues, and the 10 genes in group 4 were expressed with moderate level in almost all of the tissues (Fig. 5).Furthermore, several GmPK genes were found to display higher expression level in particular tissues.For example, GmPK22, GmPK6, GmPK26 and GmPK14 exhibited higher expression levels in roots than other tissues.The expression of GmPK24, GmPK1 and GmPK12 were relatively higher in cotyledon than in other tissues (Fig. 5).

Cis-acting regulatory element analysis
To further understand the transcriptional regulation mechanisms of GmPK genes, we characterized the cisacting regulatory elements within a 1500 bp upstream region from the transcription start site.As shown in Fig. 6 and supplementary Table S4, cis-elements related to abiotic stresses were identified, including the anaerobic responsive element (ARE), salt-regulated elements (SRE, GT1GMSCAM4), drought responsive elements (MYB binding site involved in drought-induction (MBS)) and low temperature responsive element (LTR).Several other regulatory elements associated with plant hormones were also detected, such as abscisic acid responsive element (ABRE), Auxin response element (AuxRR-core), gibberellin-responsive element (GARE-motif ), methyl jasomonate (MeJA)-responsive element (CGTCA motifs), salicylic acid responsive element (TCA-element) and ethylene-responsive element (ERE).Of these abiotic stresses and hormone-responsive elements, the salt-regulated element (SRE, GT1GMSCAM4) was present in 25 out of the 27 promoters, being the most prevalent cis-element (Fig. 6b, Table S4).These results might imply that the GmPK genes could be regulated by hormones and abiotic stresses, especially the salt stress.Furthermore, the promoters of the GmPK genes contained predicted binding sites for several transcription factors, including MYB (23 out of 27 promoters), MYC (24 out of 27 promoters) and WRKY (11 out of 27 promoters) (Fig. 6; Table S4).

Expression profiles of GmPK genes in response to salt stress and ABA treatment
In order to identify the salt-responsive GmPK genes in soybean, we performed qRT-PCR of all the 27 GmPK genes when the roots of soybean suffered from NaCl treatment at seedling stage.As shown in Fig. 7, the expression of 24 GmPK genes were found to be significantly changed in at least 1 time point.Among them, 16 genes were up-regulated by NaCl treatment, while 7 genes were down-regulated.1 gene (GmPK1) was down-regulated at 3 h, but was up-regulated at 6 and 12 h.Notably, 15 GmPK genes were differently expressed at all of the 3 time-points (Fig. 7).
We next examined the expression of these 15 GmPK genes which respond to NaCl at all of the 3 time point under 100 µM ABA treatment.The result displayed that the expressions of 6 GmPK genes (GmPK4, GmPK11, GmPK14, GmPK17, GmPK21 and GmPK26) which were continuously regulated by salt were also significantly changed at all of the 3 time-points under ABA treatment (Fig. 8).Among them, the expression of GmPK21 was represented the most significant change under NaCl Fig. 2 The chromosomal locations of PK genes in soybean.Green bars represent the 20 soybean chromosomes.The chromosome numbers are presented at the tops of the green bars.Scale bar on the left represents the lengths of each chromosome (Mb).Chr: Chromosome treatment (Fig. 7), being the best candidate for further study.

Expression of GmPK21 and subcellular localization of its protein GmPK21
In order to clarify the details of the expression of GmPK21, the tissue expression pattern of GmPK21 was analyzed by histochemical GUS staining of transgenic Arabidopsis in which the GUS gene was driven by the promoter of GmPK21.As shown in Fig. 9 a-c, the GUS signal could be detected in leaves, roots, shoot apical, calyx and silique, which is consistent with the expression pattern of GmPK21 in different tissues of soybean measured by qRT-PCR (Fig. 5).Moreover, GmPK21 was found to be highly expressed in vascular of the tested tissues.We then performed the GUS staining of the root when the plants were treated by 100 mM NaCl.The result showed that the GUS signal was significant stronger than control Fig. 3 Circle plot of soybean chromosomes and the GmPK genes displayed as segmental duplicated gene pairs.The GmPK gene pairs are indicated by the red lines.Scale bar in marked on the chromosome indicates the chromosome lengths (Mb).Chr: Chromosome (Fig. 9 d-e), which is consist with the fact that GmPK21 could be upregulated by NaCl treatment (Fig. 7).
In order to explore the subcellular localization of GmPK21, the GmPK21-GFP fusion protein was transformed into Arabidopsis protoplast (Fig. 9f ) and tobacco protoplast (Figure S3).The result showed that the green fluorescent signal of GmPK21 was detected in the cytoplasm and plasma membrane (Fig. 9f, Figure S3).

Ectopic expression of GmPK21 in Arabidopsis reduced the plant salt tolerance
To investigate the role of GmPK21 in salt tolerance, GmPK21 was overexpressed in Arabidopsis.3 transgenic lines displayed higher PK enzyme activity compared to WT were further used to perform NaCl treatment assay (Fig. 10a).The seedlings of the transgenic line 4 and 8 displayed similar growth status with wild type (WT), and Fig. 6 Cis-acting regulatory element analysis a Diagram of cis-acting regulatory elements in promoters of GmPK genes; b The distribution of the cis-elements related to abiotic stresses, plant hormones and transcription factors binding site.ARE: anaerobic responsive element; GT1GMSCAM4: salt-regulated element; MBS: drought responsive element; LTR: low temperature responsive element; ABRE: abscisic acid responsive element; AuxRR-core: Auxin response element; GARE-motif: gibberellin-responsive element; CGTCA-motif: methyl jasomonate (MeJA)-responsive element; TCA-element: salicylic acid responsive element; ERE: ethylene responsive element.W-box: WRKY binding site the L10 showed slightly longer primary roots compared with WT on 1/2 MS medium (Fig. 10 b-c).However, when the 3-day-old seedlings were transfer to the 1/2 MS medium containing 200 mM NaCl, the growth status of all the 3 transgenic lines was weaker than WT, and the primary roots were shorter than WT (Fig. 10 b-c).
We further verified the salt tolerance of transgenic Arabidopsis in soil.The 21-day-old seedlings were subject to 200 mM NaCl for 19 days.The growth status and the fresh weights of the up-ground part of the seedlings were similar between the transgenic Arabidopsis and WT under normal condition.However, after 19 days NaCl treatment, the seedlings of transgenic lines grew significantly slower and weaker than WT, and the fresh weight of the aboveground parts of transgenic seedlings were much lower than that of WT (Fig. 10 d-e).The accumulations of proline could assist plants to resist the environmental stresses [28], and the MDA content is an important indicator of the degree of membrane lipid peroxidation and plasma membrane damage in plant cells.Therefore, we measured the Proline (Pro) and malondialdehyde (MDA) contents of the transgenic and WT seedlings.The Pro and MDA contents of all the seedlings didn't show significant difference between transgenic Arabidopsis and WT under normal condition.However, 19 days after NaCl treatment, the accumulation of Pro and MDA in transgenic plants and WT was significantly promoted compared to that under normal condition.Moreover, the Pro content in transgenic Arabidopsis were significantly lower than that in WT seedlings, while the MDA content in transgenic Arabidopsis were significantly higher than that in WT (Fig. 10 f-g).

Discussion
Pyruvate Kinase is a key regulatory enzyme in the glycolysis pathway.However, little information is known about PK genes in soybean.In this study, we identified 27 PK genes which were distributed in 13 out of the 20 chromosomes of soybean.The number of soybean PK genes is more than that in Arabidopsis (14), rice (10), potato (11) [29] and cotton (33) [11].By examining the genome size and the density (number/Mb) of PK genes of the species above, we found that the density of PK genes in the soybean genome (0.03) is less than that in Arabidopsis (0.10) and potato (0.11), but similar to that in rice (0.03) and cotton (0.02).The differences in PK gene number and density between species might due to gene duplication events or the genome size [27,30].The soybean PK family genes were classified into 2 distinctive subfamilies PKc and PKp and each of them was further divided into two subclades.The result is consistent with the classification reported in Arabidopsis and rice [24,25].
Gene duplication is one of the major evolutionary mechanisms for gene expansion [27,31].In general, Fig. 7 The expression of all the 27 GmPK genes under 200 mM NaCl treatment The error bar represents the mean ± SD of three biological replicates.Student's t-test was used to examine the statistical significance (** p < 0.01, *p < 0.05) gene families expand mainly by tandem and segmental duplications [27].In our study, 22 pairs segmental duplicated gene (involved in 20 genes) representing 74.07% of the total 27 GmPK genes were identified.Therefore, the segmental duplication is the main mechanism of GmPK genes duplication.The soybean genome has undergone two rounds of whole genome duplication events, including the Legume WGD at around 59 million years ago (Ks < 0.3) and the Glycine WGD at around 13 million years ago (0.3 < Ks < 1.5) [32].In the present study, half of the duplication events occurred during the Glycine genus WGD (whole-genome duplication) event, 8 gene pairs were separated during the legume WGD event.The duplication has contributed to adaptive evolution in plants [33].Duplication gene pairs will experience different selection process, including subfunctionalization through purifying selective pressure (Ka/Ks < 1) [34] and neofunctionalization through positive selective pressure (Ka/Ks > 1) [35].In soybean, most of the duplicated genes are subfunctionalized, only a small proportion of the duplicated genes have been neofunctionalized or nonfunctionalized [36].In this study, all of the Ka/Ks ratios in different PK gene pairs were less than 1, suggesting that all of those duplicated PK genes were experienced a strong purifying selective pressure and subfunctionalized during evolution.
GmPK proteins from the same subfamily displayed similar motif type and distribution (Fig. 4).That is, the GmPKs of PKc-1 and PKc-2 had similar motif, and the GmPKs from PKp-α and PKp-β also had similar motif.However, the gene structures between each clade were different.The introns of PKc-2 genes are much more than those of PKc-1 genes, and the introns of PKp-β genes are much more than those of PKp-α genes (Fig. 4).This finding is consistent with that of the PKs in rice [25].The results might indicate that the PK genes have changed the numbers and the length of introns to adapt to the environment during their evolutionary process but retained their conserved domains, which enabled the genes to perform their functions stably [25].The genes expression patterns in different tissues provide clues to mining the potential function of these genes.In this study, a number of GmPK genes were constitutively expressed in most of the tested tissues, suggesting that they might  and seed set [12].In the present study, 2 PKc-2 subclade member GmPK4 and GmPK12 were expressed in all of these tested tissues with a remarkable transcription level.This finding suggests that GmPK4 and GmPK12 might also be involved in the regulation of soybean morphological development.In Arabidopsis, AtPKp1 (At3g22960) was reported to be involved in seed oil accumulation, embryo development and seed storage compounds mobilization upon germination [8,9,37].Here, a soybean PKp gene GmPK13 which is homologous to AtPKp1 exhibited a higher expression in pod, suggesting its potential roles in the seed development.
Gene promoters are DNA sequences located upstream of gene coding regions and contain multiple cis-acting elements, which are specific binding sites for proteins involved in the initiation and regulation of transcription [38].The promoters of GmPK were densed in transcription factors biding sites, including MYB, MYC and WRKY, indicating that these transcription factors may be relevant to GmPK expression.It was reported that the cis-acting element W-box (TTG ACC ) was present on the promoter of PK gene DkPK1 in persimmon.By using the yeast one-hybrid method, DkWRKY3 and DkWRKY15 were found to interact with the promoter of DkPK1 [39].In this study, 11 GmPK genes possess the W-box (Fig. 6; Table S4) in their promoters.These GmPK genes probably could be upregulated by WRKY genes in soybean.Further study could use the yeast one-hybrid system to verify this speculation.
92.6% of the GmPK gene promoters contained the saltregulated element (SRE, GT1GMSCAM4) (Fig. 6).Therefore, we speculated that GmPK genes might play roles on the soybean salt tolerance.However, studies on the involvement of soybean PK genes in salt stress response have rarely been reported.Based on the qRT-PCR, 16 GmPK genes were found to be up-regulated by NaCl (Fig. 7).Previously, Huang [40] screened a population of soybean accessions using GWAS analysis and identified QTL associated with salt tolerance.Within these QTL regions, 10 candidate genes including the PK gene GmPK6 have been identified.In this study, the expression of GmPK6 was induced at 6 and 12 h after NaCl treatment, indicating that GmPK6 might be a candidate gene responsible for the salt tolerance.Furthermore, we identified 6 GmPK genes were continuously regulated by salt and ABA.It is noteworthy that 4 of them belonged to the PKc subfamily, suggesting that PKc subfamily might play important roles in the salt tolerance regulation.GmPK21, which showed the most sensitive to NaCl was selected to further analysis by ectopic expressed in Arabidopsis.The result showed that the GmPK21-overexpression plants were more sensitive to the salt stress, suggesting that GmPK21 might negatively regulate salt tolerance.
Notably, GmPK21 is homologous to the rice PK gene OsPK5 (LOC_Os04g58110) (Fig. 1).Zhu et al. reported that OsSAP6, a saline-alkaline tolerance regulator in rice could interacts with OsPK5.Furthermore, overexpression of OsPK5 in rice improved soda saline-alkaline tolerance [21].In our study, GmPK21 is also involved in the salt stress tolerance regulation although it performed opposite function to its homologs in rice.This initial research on the function of GmPK21 should be followed by further work that focuses on the soybean transformation, and then thoroughly explore its function on salt tolerance.

Conclusions
This study provided a comprehensive characterization of the soybean PK family genes, highlighting their structures, expression patterns and the potential function in salt tolerance.The results indicated that GmPK genes could respond to the NaCl treatment.GmPK21, which represented the most significant change under NaCl treatment was found to negatively regulate the salinity tolerance.This research could have significant implications for the development of genetic improvement strategies to increase soybean tolerance to salt, thus contributing to food security and agricultural sustainability.

Genome-wide identification of PK genes in soybean
All of the genome sequences data of soybean cultivar Zhonghuang13 were downloaded from the Genome Warehouse (GWH) database in the BIG Data Center under Accession Number GWHAAEV00000000.1 [26].The Hidden Markow Model (HMM) of PK and PK_C domain (PF00224 and PF02887) was retrieved from the Pfam database (http:// pfam.xfam.org) [41], and then was used to search against the Zhonghuang13 genome.All of the putative proteins were extracted and confirmed by the Pfam and NCBI-CDD database (https:// www.ncbi.nlm.nih.gov/ cdd).The information of soybean PK proteins including molecular weights and isoelectric points were calculated using the online program ExPASy (https:// www.expasy.org) [42].

Chromosomal location and gene duplication analysis of soybean PK genes
The physical positions of soybean PK genes were extracted from the annotation file downloaded from GWH database.Then, the chromosomal location map was drawn using the MapChart (v.2.32) software [44].Potential gene duplications were determined by two major criteria: length of aligned sequence covers ≥ 75% of longer gene and similarity of aligned regions is ≥ 75%.Ka and Ks values were calculated using KaKs Calculator (http:// code.google.com/p/ kaks-calcu lator/ wiki/ KaKs_ Calcu lator) [45].MCSscanX was used to detect the synteny of PK genes between soybean and other plants [46].

Gene structure and conserved motifs analysis
The gene structures of GmPK genes were retrieved from the GFF3 annotation file downloaded from GWH database and then visualized by the Gene Structure Display Server (http:// gsds.cbi.pku.edu.cn).To identify the conserved motifs of GmPK proteins, the online program MEME (http:// meme-suite.org/ tools/ meme) was used.

Expression analysis of soybean PK genes in different tissues
Soybean cultivar Zhonghuang13 was grown in a controlled culture room under short day condition (12 h light/12 h dark).The root, stem, leaf, shoot apical, cotyledon was sampled at 12 days after emergence (DAE), and the flower and pod were sampled at 34 DAE and 50 DAE, respectively.Each sample contained 3 whole roots from 3 independent plants.Three biological replicates were performed.The total RNA of these samples was extracted and reverse-transcribed for subsequent quantitative realtime PCR (qRT-PCR) analysis.Data were presented as means ± SD, and Student's t-tests were used by SPSS statistics 19 to assess the significance of differences.

Cis-element analysis of GmPK genes
For the cis-element analysis, the 1500 bp upstream sequences of soybean PK family genes were obtained from the GWH database.The PLANTCARE database (http:// bioin forma tics.psb.ugent.be/ webto ols/ plant care/ html/) was used to predict the cis-elements of the GmPK gene promoters.

Identification of soybean PK genes in response to the NaCl and ABA treatment
The seedlings of soybean cultivar Zhonghuang13 were grown in a controlled culture room at 25 °C with the photoperiod of 12 h light/12 h dark.For the NaCl and ABA treatment, 14-day-old seedlings were gently removed from the soil and cultivated in Hoagland liquid medium for 2 days.After that, the seedlings were transferred into new Hoagland liquid medium containing 200 mM NaCl or 100 μM ABA.The roots from control and treated seedlings were sampled at 0 h, 3 h, 6 h and 12 h after treatment and then immediately immersed in liquid nitrogen and stored at -80 °C for RNA extraction and qRT-PCR.
Each sample contained 3 whole roots from 3 independent plants.Three biological replicates were performed.Data were presented as means ± SD, and Student's t-tests were used by SPSS statistics 19 to assess the significance of differences.

qRT-PCR analysis
The total RNA of the tested samples was extracted using Trizol reagent.qRT-PCR with 3 technical replicates for each of the triplicate biological samples was conducted to measure the expression levels of the GmPK genes.qRT-PCR was performed by using Roche 480 Light Cycler (Roche, Mannheim, Germany) and the Takara SYBR Premix Extaq (Takara, Japan).Gene expression levels were then calculated according to the method (2 −ΔΔCT ) described by Livak and Schmittgen [47].For the expression pattern of those 27 GmPK genes in different tissues and under NaCl treatment, GmActin (Glyma.18G290800)was used as the reference gene.For the expression level of GmPK21 in the transgenic Arabidopsis, AtActin2 (AT3G18780) was used as the reference gene.The details of the primers were shown in Table S5.

Histochemical GUS assays
The promoter of GmPK21 was amplified by genomic PCR (The primers GmPK21pro-F and GmPK21pro-R were shown in Table S5) and then was inserted into the linearized DX2181 vector (digested by HindIII and BamHI) using the EasyGeno Assembly Cloning kit (Tiangen, VI201), generating the GmPK21pro-GUS fusion plasmid.The plasmid was then introduced into the Agrobacterium tumefaciens strain GV3101 and then transferred to Arabidopsis ecotype Columbia by using the floral dip method [48].To detect GUS activity in different tissues of transgenic Arabidopsis, the seedlings, flowers and siliques were sampled and incubated in the X-Gluc solution for 16 h at 37 °C [49].To detect the GUS activity after the NaCl treatment, the Arabidopsis seedlings were submerged in the 1/2 MS medium containing 100 mM NaCl.The roots of the Arabidopsis treated by NaCl were sampled after 6 h treatment, and the roots of the Arabidopsis which submerged in the 1/2 MS medium without NaCl were used as control.Then all of the samples were incubated in the X-Gluc solution for 16 h at 37 °C.In order to remove the chlorophyll, the samples were transferred into the ethanol (70% v/v) and then observed under stereomicroscope (Leica S9i, Germany).The samples were collected from 3 different transgenic Arabidopsis lines.

Subcellular localization of GmPK21
The CDS without the termination codon of GmPK21 was amplified by PCR using the primer GmPK21-F1 and GmPK21-R1 (Table S5) and then was introduced into the plasmid linearized pAN580 vector (digested by XbaI and BamHI) using the EasyGeno Assembly Cloning kit (Tiangen, VI201), generating the GmPK21-GFP fusion plasmid.The empty vector pAN580 was used as the control.The GmPK21-GFP and empty vector pAN580 were transformed into the Arabidopsis or tobacco protoplasts [50] respectively.The GFP signal was then analyzed using the LSM710 confocal microscope (Zeiss, Oberkochen, Germany).

Overexpression of GmPK21 in Arabidopsis
The full length of the GmPK21 CDS was amplified by PCR using the primer GmPK21-F2 and GmPK21-R2 (Table S5) and then was introduced into the linearized plasmid pC3300s (digested by SacI and BamHI) using the EasyGeno Assembly Cloning kit (Tiangen, VI201), generating the pC3300s-35S::GmPK21.The construct was then transformed into Arabidopsis (Col-0) using Agrobacterium GV3101.The seeds of transformants were selected by 1/2 MS medium containing glufosinate solution.The RT-PCR and qRT-PCR was then performed to confirm the presence of the GmPK gene in the glufosinate resistant lines (The RT-PCR primers RT-GmPK21-F/ RT-GmPK21-R and qRT-PCR primers qGmPK21-F (Ara)/qGmPK21-R(Ara) were shown in Table S5).The PK activity of 20-day-old transgenic Arabidopsis and WT determined by using the Pyruvate Kinase (PK) Activity Assay Kit (BC0540, Solarbio life sciences).3 biological replicates were performed and 10 seedlings (aboveground parts) were pooled per biological replicate.Data were presented as means ± SD, and Student's t-tests were used by SPSS statistics 19 to assess the significance of differences.

Salt stress treatment of the GmPK21 transgenic Arabidopsis
Three lines of the T3 transgenic Arabidopsis plants were used for NaCl treatment assay.The Arabidopsis were germinated in the 1/2 MS medium.The 3-day-old seedlings were then transferred to the new 1/2 MS medium containing 200 mM NaCl.The seedlings in the 1/2 MS without NaCl were used as control.6 days later, the status of the plants was observed and the length of the primary roots were measured.At least 20 seedlings' roots were measured for each line.Data were presented as means ± SD, and Student's t-tests were used by SPSS statistics 19 to assess the significance of differences.
Meanwhile, 3-day-old seedlings were transferred to the soil, and grown in an incubator (23℃) under short day condition (8 h light/16 h dark).21 days later, seedlings were treated with 200 mM NaCl for 19 days.The status of the plants was observed and the fresh weight of aboveground parts of the seedlings were measured.
At least 20 seedlings' fresh weight were measured for each line.Data were presented as means ± SD, and Student's t-tests were used by SPSS statistics 19 to assess the significance of differences.Then the aboveground parts of the Arabidopsis seedlings were used to measure the proline (Pro) and malondialdehyde (MDA) content.The content of Pro and MDA was determined by Proline (Pro) Content Assay Kit (Solarbio, BC0290) and Malondialdehyde (MDA) Content Assay Kit (Solarbio, BC0025) cording to the kit instructions.3 biological replicates were performed and 10 seedlings were pooled per biological replicate.
• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year

•
At BMC, research is always in progress.

Learn more biomedcentral.com/submissions
Ready to submit your research Ready to submit your research ?Choose BMC and benefit from: ? Choose BMC and benefit from:

Fig. 1
Fig.1Phylogenetic tree based on protein sequences of PK genes in soybean, Arabidopsis and rice (At, Arabidopsis thaliana; Os, Oryza sativa)

Fig. 4 Fig. 5
Fig. 4 Conserved motifs and gene structures of GmPKs.a Phylogenetic tree based on protein sequences of PK genes in soybean.b Conserved motifs of GmPKs.The details of the motif were shown in Figure S2.c The gene structures of GmPK genes.CDS, introns, and untranslated regions (UTRs) are indicated by yellow boxes, black lines, and green boxes, respectively

Fig. 8 Fig. 9
Fig.8 The expression of the 15 NaCl responsive GmPK genes under 100 µM ABA treatment.The error bar represents the mean ± SD of three biological replicates.Student's t-test was used to examine the statistical significance (** p < 0.01, *p < 0.05)

10 (
See legend on previous page.)

Table 1
List of pyruvate kinase family genes from