The polish wheat (Triticum polonicum L.) TpSnRK2.10 and TpSnRK2.11 meditate the accumulation and the distribution of cd and Fe in transgenic Arabidopsis plants

Background The SnRK2s (Plant specific protein kinase) are involved in various biological processes, such as plant defense and environmental challenges. In Arabidopsis, AtSnRK2s regulate the expression of some metal transporters. For example, AtSnRK2.4 plays a role in the regulation of Arabidopsis tolerance to Cd; AtSnRK2.2 and AtSnRK2.3 are involved in Cd uptake and translocation. However, the functions of their homologs, TpSnRK2.10 and TpSnRK2.11 from dwarf Polish wheat are unknown. Results TpSnRK2.11 encodes a cytoplasmic protein. TpSnRK2.10 and TpSnRK2.11 have different expression patterns at different growth stages. Expression of TpSnRK2.10 increased yeast’s sensitivity to Cd; conversely, expression of TpSnRK2.11 enhanced yeast’s tolerance to Cd. Overexpression of TpSnRK2.10 or TpSnRK2.11 did not affect Cd sensitivity in Arabidopsis, but significantly increased Cd accumulation in roots and shoots, and Cd translocation from roots to shoots. While, Fe accumulation was significantly increased in roots but decreased in shoots by overexpression of TpSnRK2.10; opposite results were observed in TpSnRK2.11-overexpressing lines. Subcellular distribution analysis found that overexpression of TpSnRK2.10 and TpSnRK2.11 increased Cd concentration in cell wall and organelle fractions of roots and shoots; meanwhile, they also differentially influenced Fe distribution. Conclusions These results indicated that TpSnRK2.10 and TpSnRK2.11 are involved in the uptakes and the translocations of Cd and Fe, possibly by regulating the expression of AtNRAMP1 and AtHMA4, and other genes involved in the synthesis of phytochelatins or hemicellolosic polysaccharides. Electronic supplementary material The online version of this article (10.1186/s12864-019-5589-1) contains supplementary material, which is available to authorized users.


Background
More arable land soils have been contaminated by cadmium (Cd) through various industrial processes and/or agricultural practices [1]. As one of the toxic heavy metals, Cd is readily taken up by plants, and then entries into food chain, which potentially affects human health [2]. In plants, Cd is absorbed via basal roots, loaded in the xylem, then transported to shoots [3,4], which negatively affects mineral nutrition and homeostasis in tissues and root growth and development [2,[5][6][7]. Characterization of numerous Cd transporters responsible for Cd accumulation and transport found that they are regulated by intricate stress signal transduction pathways [2,5]. Thus, our knowledge about Cd accumulation and transport remains limited; the investigation is still needed to understand the molecular mechanisms.
The Cd signal transduction pathway is mediated by plant protein kinases that are major signal transduction elements [8,9]. For example, Cd stress induces the activities of mitogen-activated protein kinases and sucrose nonfermenting-1 (SNF1)-related protein kinase 2 (SnRK2) in plants [6,8,9]. As plant specific protein kinase, ten SnRK2s are individually identified from Arabidopsis. Molecular analysis of them indicated that they are involved in various biological processes, such as plant defense to biotic and/or abiotic challenges [6,[10][11][12]. AtSnRK2.2 and AtSnRK2.3 regulate abscisic acid (ABA) response element (ABRE)-driven gene expression through the phosphorylation of ABRE binding factors (ABFs) [10]. Knockout of AtSnRK2.2 and AtSnRK2.3 enhances the expression of IRON-REGULATED TRANS-PORTER1 (IRT1) under 10 μM Cd stress [12]. IRT1 is a key transporter responsible for Cd uptake [3,13]. The knockout of IRT1 reduced Cd concentration in roots and increases in shoots [12], which is not consistent with its enhanced expression. Thus, knockout of AtSnRK2.2 and AtSnRK2.3 may regulate other metal transporters responsible for Cd uptake and translocation, such as AtNRAMPs whose promoters have an ABF or ABFs in Arabidopsis [14,15]. In contrast, although knockout of AtSnRK2.4 did not affect the expression of IRT1 and the concentration of Cd under 20 μM Cd stress, it enhanced Cd tolerance and reduced phytochelatins concentration [6]. Phytochelatins chelate Cd to form phytochelatin-Cd complexes, which are transported into vacuoles to enhance Cd tolerance [16]. Meanwhile, AtSnRK2.6 probably participates in multiple signaling pathways. Inactivation of AtSnRK2.6 impairs stomatal close to enhance transpiration [17]. Overexpression of AtSnRK2.6 increases the content of fructose and glucose [18]. Hemicellolosic polysaccharides of cell wall, consisted of fructose, glucose and other sugars, are major Cd binding sites, and positively correlate with the acquire capacity of Cd [19]. These results indicate that SnRK2s could be involved in Cd uptake and translocation; meanwhile, their mechanisms and functions may be different.
Several SnRK2s from rice and wheat, including OsSAPK8, OsSAPK9, OsSAPK10 and TaPKABA1, can also phosphorylate ABFs [20,21]. Thus, they also potentially regulate the expression of genes with an ABRE or ABREs in their promoters. In wheat, 10 SnRK2s (TaSnRK2.1-TaSnRK2.10) are isolated and grouped into three subclasses. Subclass III TaSnRK2s are regulated by ABA stress, suggesting that these genes are involved in ABA regulated stress responses; however, subclass I and II TaSnRK2s are regulated by PEG, NaCl and cold stress, but not by ABA stress, suggesting that these genes responded to various abiotic stressors in an ABA-independent manner [22]. In tetraploid wheat, eight SnRK2s (TpSnRK2.1, 2.2, 2.3, 2.5, 2.7, 2.8, 2.10 and 2.11) are cloned from Polish wheat (Triticum polonicum L.); except of TpSnRK2.7, all of the genes are differentially regulated by ABA, NaCl, PEG and cold stress [23,24]. However, whether these genes involved in Cd uptake and translocation are still unknown. Since TpSnRK2.10 is homologous with AtSnRK2.2 and AtSnRK2.3, and TpSnRK2.11 is homologous with AtSnRK2.4, we hypothesized that they are involved in Cd uptake, translocation and tolerance. To test these hypotheses, we investigated their biological functions by analyzing expression pattern, subcellular localization in Arabidopsis leaf protoplast, Cd tolerance in yeast, and the concentration and subcellular distribution of Cd, Fe and Zn in Arabidopsis overexpressing lines.

Materials
Dwarf Polish wheat (Triticum polonicum L., 2n = 4X = 28, AABB, DPW) used in the present study was collected from Tulufan of Xinjiang province, China. Previous study indicated that the seedling of DPW exhibits high tolerance to Cd [25]. Thus, it is a desirable material for analyzing the mechanism of high tolerance to Cd. The seed of Arabidopsis thaliana (wild type, WT) was provided by Professor Yan Huang (College of Life Science, Sichuan Agricultural University).
Expression analysis of TpSnRK2.10 and TpSnRK2.11 For expression pattern in DPW grown in normal growth season, tissues were collected from three growth stages with three biological replicates, including jointing stage (root, basal stem, leaf sheath, new leaf and young leaf ), booting stage (root, basal stem, second stem, first node, second node, third node, lower leaf, flag leaf sheath, flag leaf blade and peduncle), and grain filling stages (root, stem, first node, leaf sheath, flag leaf blade, lemma and grain). All tissues were snap frozen in liquid nitrogen and stored at − 80°C for RNA extraction.
For response to Cd stress and Fe-deficient in DPW seedlings, sterilized DPW seeds were germinated in dark at room temperature for four days. Uniform seedlings were cultured in Hoagland's nutrient solution in a growth chamber for two weeks. All seedlings were treated with two stresses including 80 μM CdCl 2 (Cd), and Fe-free Hoagland's nutrient solution (Fe-deficient). Seedlings grown in Normal Hoagland's nutrient solution were used as control (CK). After a week of treatments, roots were collected with three biological replicates and snap frozen in liquid nitrogen for RNA extraction.
Total RNA of each sample was isolated using the Total RNA Kit II (Omega, USA) according to the user manual. RNA concentration was measured by NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies). The first-strand cDNA was synthesized from 2 μg total RNA using the M-MLV First Strand cDNA Synthesis Kit (Omega, USA).

Subcellular localization of TpSnRK2.11 in Arabidopsis protoplasts
Subcellular localization of TpSnRK2.11 was analyzed as described by Wang et al. [4] and Peng et al. [27]. Briefly, the open reading frame of TpSnRK2.11 was sub-cloned into the KpnI and XbaI sites of the Arabidopsis protoplast expression vector HBT95-GFP using TpSnRK2.11 primers (F:CGGGGTACCATGGACAAGT ACGA GGAGG, and R: GCTCTAGATTAGATGTGCAACGCG CTCA). Arabidopsis leaf mesophyll protoplast was prepared and transformed according to the method of Yoo et al. [28]. Signal of GFP was detected by a confocal laser scanning microscope (Olympus, Tokyo, Japan).

Cd sensitive analysis in yeast
Cd sensitivity in yeast was tested as described by Peng et al. [27]. Briefly, the open reading frames of TpSnRK2.10 and TpSnRK2.11 were individually sub-cloned into the KpnI and XbaI sites of yeast expression vector pYES2 using TpSnRK2.11 primers and TpSnRK2.10 primers (F: CGGGGTACCATGGACCGGGCGGCGC, and R: TGCT C TAGATCACATAGCATACAC). The S. C. EasyComp Transformation kit (Invitrogen, USA) was used to transform plasmid into yeast strains including wild type (BY4743) and Cd sensitive strain (ydr135c) according to its user manual.
Functional analysis of TpSnRK2.10 and TpSnRK2.11 in overexpressing Arabidopsis lines Biological functions of TpSnRK2.10 and TpSnRK2.11 were analyzed in Arabidopsis overexpressing lines as described by Peng et al. [27]. Briefly, the open reading frames of TpSnRK2.10 and TpSnRK2.11 were sub-cloned into the KpnI and XbaI sites of the expression vector pCAMBIA1305.1 driven by the 35S promoter. TpSnRK2.10 and TpSnRK2.11 were individually transformed into WT Arabidopsis thaliana plants using floral infiltration via Agrobacterium-mediated transformation [29] (two time-independent transformations). Homozygous overexpressing lines were selected using hygromycin selection and PCR with gene-specific primers [23]. The relative expression of TpSnRK2.10 and TpSnRK2.11 in their individually transgenic lines was analyzed as described as the section of "Expression analysis of TpSnRK2.10 and TpSnRK2.11". Meanwhile, the relative expression of AtIRT1, AtNRAMP1 and AtHMA4 was also investigated in WT, vector line, TpSnRK2.10-and TpSnRK2.11-expressing lines using AtActin2 as reference gene. The qPCR primers of AtIRT1, AtNRAMP1, AtHMA4 and AtActin2 were used as described by Ihnatowicz et al. [30], Boonyaves et al. [31] and Chen et al. [32], respectively.
For investigation of Cd sensitive in TpSnRK2.10-and TpSnRK1.11-expressing lines, sterilized seeds of WT, an empty vector line and independent homozygous lines (each line contained three plant lines selected from an independent transformation; totally six plant lines) were germinated on 1/2 MS solid plates containing CdCl 2 (0 μM and 40 μM) in a light incubator with 120 μE m − 2 s − 1 illumination intensity, 16/8 h light/dark period, 22°C temperature and 50% humidity. On the tenth day, root length of each line was measured.
To investigate whether TpSnRK2.10 and /or TpSnRK2.11 influenced metal (Cd, Fe and Zn) concentration in TpSnRK2.10-and TpSnRK1.11-overexpressing lines, uniform seedlings of WT, an empty vector line and independent homozygous lines were cultured in soil. After three weeks, soil was added once with CdCl 2 (0 mg/kg and 40 mg/kg) dissolved in water. After four weeks, dry weight of each plant was measured. Part of roots and shoots was dried at 80°C for two days to measure metal concentration as described by Cheng et al. [7]. Remaining roots and shoots were used to measure the subcellular distribution of Cd and Fe as described by Cheng et al. [7].

Data analysis
All data was statistically analyzed using Tukey's test at Previously phylogenetic analysis revealed that TpSnRK2.10 grouped with AtSnRK2.2 and AtSnRK2.3, which belong to the subgroup 3 kinases that are strongly activated by ABA; TpSnRK2.11 grouped with AtSnRK2.4, which belongs to the subgroup 1 kinases that are not activated by ABA.
Expression patterns of TpSnRK2.10 and TpSnRK2.11 Expression patterns of TpSnRK2.10 and TpSnRK2.11 were investigated in different tissues at jointing, booting and grain-filling stages of wheat grown in a field. At jointing stage, expression of TpSnRK2.10 was the highest in new leaves and young leaves, followed by leaf sheathes, and the lowest in basal stems (Fig. 2a). At booting stage, TpSnRK2.10 was expressed the highest in lower leaves, flag leaf sheathes and flag leaf blades, and the lowest in basal stems and second stems (Fig. 2a). At grain-filling stage, the highest expression of TpSnRK2.10 was observed in grains, and the lowest was found in roots and first nodes (Fig. 2a).
The expression pattern of TpSnRK2.11 was different from that of TpSnRK2.10 (Fig. 2b). At jointing stage, the highest expression of TpSnRK2.11 was found in leaf sheathes, followed by in basal stems and new leaves, and the lowest was in roots and young leaves (Fig. 2b). At booting stage, TpSnRK2.11 was mainly expressed in flag leaf sheathes, flag leaf blades and third nodes, then in lower leaves, roots and first nodes, and the lowest in basal stems and peduncels (Fig. 2b). At grain-filling stage, expression of TpSnRK2.11 was the highest in basal stems and first nodes, and the lowest in roots and grains (Fig. 2b).
Meanwhile, the responses to Cd-supply and Fe-deficient were investigated (Fig. 2c). The expressions of TpSnRK2.10 and TpSnRK2.11 were dramatically down-regulated by Cd stress when compared with their individually CK (Fig. 2c). The expression of TpSnRK2.10 was not regulated by Fe-deficient; while the expression of TpSnRK2.11 was down-regulated by Fe-deficient (Fig. 2c). These results indicate that TpSnRK2.10 and TpSnRK2.11 are response to Cd and Fe.

Subcellular localization of TpSnRK2.11
TpSnRK2.10 and TpSnRK2.11 were predicated to localize to the cytoplasm and the nucleus. To confirm the predicated subcellular localizations, a HBT95-TpSnRK2.11-GFP fusion vector was transiently transformed into Arabidopsis leaf protoplasts. Our previous study indicated that green fluorescence of the empty vector HBT95-GFP is located to the cytoplasm, the nucleus and the plasma membrane [4]. In this study, the green fluorescence of HBT95-TpSnRK2.11-GFP in Arabidopsis leaf protoplasts was detected in the cytoplasm, but was not in the nucleus (Fig. 3). Thus, TpSnRK2.11 encodes a cytoplasmic protein. Unfortunately, we failed to detect the subcellular localization of TpSnRK2.10.

Cd sensitivity in yeast
To investigate whether TpSnRK2.10 and TpSnRK2.11 alter Cd sensitivity in yeast, we expressed TpSnRK2.10, TpSnRK2.11 or pYES2 in BY4743 and ydr135c. In the presence of 2% glucose that represses gene expression with the GAL1 promoter in pYES2 vector [33,34], Cd stress significantly inhibited the growth of ydr135c when compared with BY4743 (Fig. 4a); meanwhile, similar growths were detected among ydr135c individually transformed with pYES2, TpSnRK2.10 and TpSnRK2.11 (Fig. 4a). In the presence of 2% galactose that induces gene expression with the GAL1 promoter in pYES2 vector, expression of TpSnRK2.11 dramatically increased the growth of ydr135c when compared with ydr135c transformed with pYES2 under both of 80 μM and 100 μM CdCl 2 stresses (Fig. 4b); while expression of  TpSnRK2.10 did not change the growth of ydr135c when compared with ydr135c transformed with pYES2 under CdCl 2 stresses (Fig. 4b).
To confirm the results produced from plate testing, the growth curves of transformed yeast grown in liquid medium with 80 μM CdCl 2 were investigated ( Fig. 4c  and d). Compared with ydr135c transformed with pYES2, expression of TpSnRK2.11 significantly increased the OD 600 values of ydr135c starting at 16 h (Fig. 4c); while, expression of TpSnRK2.10 dramatically reduced the OD 600 values of ydr135c staring at 32 h (Fig. 4d). These results indicated that expression of TpSnRK2.11 enhances Cd tolerance in yeast; while, expression of TpSnRK2.10 increases Cd sensitivity in yeast.
Functional overexpression of TpSnRK2.10 and TpSnRK2.11 in Arabidopsis Since expression of TpSnRK2.10 and TpSnRK2.11 altered Cd sensitivity in yeast, it is interesting to investigate whether they would play roles in metal transport in plant. Thus, we individually overexpressed TpSnRK2.10 and TpSnRK2.11 in Arabidopsis. Two independent homozygous of TpSnRK2.10-or TpSnRK2.11-overexpressing lines were developed (Additional file 1: Figure S1). When grown on 1/ 2 MS medium with 40 μM CdCl 2 , overexpression of TpSnRK2.10 or TpSnRK2.11 did not change root length compared with WT ( Fig. 5a and b); Cd stress significantly inhibited root length when compared with CK (Fig. 5b). When grown in soil with 40 mg/kg CdCl 2 , overexpression of TpSnRK2.10 or TpSnRK2.11 also did not change the dried weight of roots and shoots compared with CK ( Fig.  5c and d). These results indicated that overexpression of TpSnRK2.10 or TpSnRK2.11 did not influence Cd sensitivity in plant.
Meanwhile, we analyzed Cd, Fe and Zn concentrations in roots and shoots grown in soil. Under 40 mg/kg CdCl 2 stress, overexpression of TpSnRK2.10 or TpSnRK2.11 significantly increased Cd concentration in roots and shoots when compared with WT and vector line ( Fig. 6a and b). Under none metal stress, overexpression of TpSnRK2.10 significantly enhanced Fe concentration in roots (Fig. 6c), but reduced in shoots when  (Fig. 6d); in contrast, overexpression of TpSnRK2.11 significantly reduced Fe concentration in roots (Fig. 6c), but enhanced in shoots when compared with WT and vector line (Fig. 6d). However, overexpression of TpSnRK2.10 or TpSnRK2.11 did not change Zn concentration in roots and shoots (Fig. 6e  and f). Compared with WT, overexpression of TpSnRK2.10 or TpSnRK2.11 significantly enhanced Cd translocation factor (TF, the shoot-to-root Cd concentration ratio) (Fig. 7a); expression of TpSnRK2.11 enhanced Fe TF, while overexpression of TpSnRK2.10 reduced that (Fig. 7b). These results indicated that TpSnRK2.10 and TpSnRK2.11 mediate Cd and Fe uptake and translocation.
To understand the physiological mechanisms that TpSnRK2.10 and TpSnRK2.11 participate in Cd and Fe uptake and translocation, we analyzed Cd and Fe subcellular distribution in overexpressing lines, WT and vector line. Compared with WT and vector line, overexpression of TpSnRK2.10 and TpSnRK2.11 dramatically increased Cd concentration of cell wall fraction and organelle fraction in roots and shoots ( Fig. 8a and b). Overexpression of TpSnRK2.10 increased Fe concentration of cell wall fraction and organelle fraction in roots (Fig. 8c), but decreased that of cell wall fraction in shoots when compared with WT and vector line (Fig. 8d). However, overexpression of TpSnRK2.11 reduced Fe ; e-f Zn concentration in roots (e) and shoots (f). Asterisk indicated significant difference when compared with WT at P < 0.05; values were means ± standard deviation (three biological replicates) concentration in soluble fraction in roots (Fig. 8c), and cell wall fraction in shoots (Fig. 8d). These results implied that TpSnRK2.10 and TpSnRK2.11 probably influence Cd and Fe uptake and translocation via changing Cd and Fe subcellular distribution in roots and shoots.

Discussion
In Arabidopsis, Cd can activate AtSnRK2.4 and AtSnRK2.10; meanwhile, knockout of AtSnRK2.4 enhances Cd tolerance [6]. Overexpression of wheat genes TaSnRK2.3, TaSnRK2.4, and TaSnRK2.7 or TaSnRK2.8 in Arabidopsis noticeably enhances drought, salinity and cold tolerance [18,[35][36][37]. Meanwhile, SnRK2s from tetraploid wheat are differently regulated by several abiotic stresses including drought, salt, ABA and cold [23,24]. However, the roles of these SnRK2s from wheat in heavy metal responses and transport are still elusive. In this study, expression of TpSnRK2.11 restored the growth of Cd sensitive strain; while expression of TpSnRK2.10 increased the growth inhibition of Cd. The results suggested that TpSnRK2.10 and TpSnRK2.11 play different roles in Cd tolerance in yeast. Overexpression of TpSnRK2.10 or TpSnRK2.11, however, did not change Cd tolerance in Arabidopsis, which was concluded from similar root growth and length of seedling under 40 μM CdCl 2 , and root and shoot dried weight of maturation under 40 mg/kg CdCl 2 . The discrepant results might result from the different genome buffering between yeast and Arabidopsis. Meanwhile, these results implied that the functions of TpSnRK2.10 and TpSnRK2.11 differ from that of AtSnRK2.4, which mediates Cd tolerance [6]. These differences might result from: (1) the low identity (68.6%) between TpSnRK2.11 and AtSnRK2.4, which includes different residues in several crucial motifs (ATP binding site, activation loop, and serine/threonine protein kinases activity site); (2) the different subcellular localizations that TpSnRK2.11 was localized to the cytoplasm, but AtSnRK2.4 was localized to the cytoplasm and nucleus [6]; and (3) the different responses to Cd, such as the expression of TpSnRK2.11 was down-regulated by Cd, and the expression of AtSnRK2.4 was up-regulated by Cd [6].
Many studies indicate that SnRK2 mediates the regulation of sucrose metabolism, phytochelatins synthesis, and sulfur metabolism [6,11,38], which play crucial roles in heavy metal binding, transport and tolerance [16,19,39]. Meanwhile, SnRK2s also regulate the expression of some genes that encode metal transporters and metal chelation, such as IRT and plant metallothionein [6,12,40]. Thus, SnRK2s would be involved in heavy metal uptake, translocation and tolerance. In Arabidopsis, knockout of AtSnRK2.2 and AtSnRK2.3 reduces Cd concentration in roots but increases that in shoots [12]; however, knockout of AtSnRK2.4 did not affect Cd concentration, although it enhanced Cd tolerance [6]. In this study, overexpression of TpSnRK2.10 or TpSnRK2.11 significantly increased Cd concentration in roots and shoots, and enhanced Cd translocation. These results indicated that TpSnRK2.10 and TpSnRK2.11 are Fig. 9 Relative expression of AtIRT1, AtNRAMP1 and AtHMA4 in transgenic lines. a Relative expression of AtIRT1; b relative expression of AtNRAMP1; c relative expression of AtHMA4. Samples were collected from shoot of Arabidopsis without metal stress. Asterisk indicated significant different when compared with WT at P < 0.05; value was mean ± standard deviation (three biological replicates) involved in Cd uptake and translocation in overexpressing Arabidopsis, which are different from the functions of AtSnRK2.2 and AtSnRK2.3 [12], and AtSnRK2.4 [6]. In Arabidopsis, AtIRT1 and AtNRAMP1 are major transporters for Cd uptake [13,41]; their expression could be potentially mediated by SnRK2s through phosphorylation of ABFs [12]. In this study, overexpression of TpSnRK2.10 or TpSnRK2.11 in Arabidopsis enhanced the expression of AtNRAMP1 to increase Cd uptake. Meanwhile, overexpression of TpSnRK2.10 or TpSnRK2.11 in Arabidopsis also enhanced the expression of AtHMA4 that is responsible for Cd translocation in Arabidopsis [42], and enhanced Cd translocation from root-to-shoot.
Except transportation of Cd, AtNRAMP1 also transport Fe for its uptake [41]. Thus, overexpression of TpSnRK2.10 or TpSnRK2.11 in Arabidopsis enhanced Fe uptake and translocation. However, overexpression of TpSnRK2.10 enhanced Fe concentration in roots, but reduced that in shoots, which implied that root retained more Fe to reduce Fe translocation. Conversely, overexpression of TpSnRK2.11 reduced Fe concentration in roots, but increased that in shoots, which indicated that more Fe was transported into shoots. Thus, overexpression of TpSnRK2.10 and TpSnRK2.11 would differentially regulate other metal transporters that specifically transport Fe to mediate Fe translocation. For example, OsYSL15 only transports Fe for Fe uptake and translocation in rice [43]; HvYS1 specifically transports Fe for Fe uptake in barely [44]; and AtNRAMP3 pumps Fe from the vacuoles to the cytosol [45]. The different influences in Fe uptake and/or translocation implied that TpSnRK2.10 and TpSnRK2.11 have different biological functions, such as their different expression patterns.
Knockout of AtSnRK2.4 reduced the synthesis of PCs [6], and overexpression of AtSnRK2.6 increased the content of hemicellolosic polysaccharides of cell wall [11]. In plant, PCs chelate metals and then sequestrated into the vacuoles [16]; hemicellolosic polysaccharides chelate metals and binds in cell wall [19]. They affect metal binding and mobilization. In this study, overexpression of TpSnRK2.10 and TpSnRK2.11 might enhance the content of PCs and hemicellolosic polysaccharides. Thus, higher Cd concentration in cell wall fraction of roots and shoots was observed in overexpressing lines. The results implied that overexpression of TpSnRK2.10 and TpSnRK2.11 enhanced the require capacity of Cd in roots and shoots, resulting in the increased Cd concentration did not influence the growth. Meanwhile, increased Fe concentrations in cell wall and organelle fractions of roots were observed in TpSnRK2.10-overexpressing lines. Conversely, reduced Fe concentrations in soluble fraction of roots and cell wall fraction of shoots were detected in TpSnRK2.11-overexpressing lines. The changes of Fe concentration in roots and shoots were consistent with Fe translocation from roots to shoots. Thus, these results confirmed that TpSnRK2.10 and TpSnRK2.11 are involved in Cd and Fe distribution.

Conclusion
In summary, our results indicate that TpSnRK2.10 and TpSnRK2.11 are involved in Cd and Fe accumulation and distribution in overexpressing lines, possibly by regulating AtNRAMP1 and AtHMA4 who have ABF or ABFs element in their promoters, and/or regulating the synthesis of phytochelatins and hemicellolosic polysaccharides. Since overexpression of TpSnRK2.10 or TpSnRK2.11 enhances Cd uptake and translocation from roots to shoots, knockout or knockdown of these two genes could help reduce Cd entrance into plant; or overexpression of these two genes could be used to phytoextract Cd from polluted soils.

Additional file
Additional file 1: Figure S1.