1 Selenium Supply Alters Subcellular Distribution and Chemical Forms of Cadmium and Expression of Transporter Genes Involved in Cadmium Uptake and Translocation in Winter Wheat (Triticum Aestivum)

Cadmium (Cd) accumulation in crops will affect the yield and quality of crops, and also harm human health. The application of selenium (Se) can reduce the absorption and transport of Cd in winter wheat. The result showed that increasing Se supply signicantly decreased Cd concentration and accumulation in shoots and roots of winter wheat, and the root to shoot translocation of Cd. The Se supply increased the root length, surface area and root volume, but decreased the root average diameter. Increasing Se supply signicantly decreased Cd concentration in cell wall, soluble fraction and cell organelle in roots and shoots. An increase of Se supply inhibited Cd distribution in the organelle of shoot and root, but enhanced Cd distribution in the soluble fraction of shoot and the cell wall of root. The Se supply also decreased the proportion of active Cd (ethanol-extractable (FE) Cd and deionized water-extractable (FW) Cd) in roots. In addition, the expression of TaNramp5-a, TaNramp5-b, TaHMA3-a, TaHMA3-b and TaHMA2 were signicantly increased with the increase of Cd concentration in roots, and the expression of TaNramp5-a, TaNramp5-b and TaHMA2 in roots were down-regulated by increasing Se supply, regardless of Se supply or Cd stress, respectively. The expression of TaHMA3-b in root was signicantly down-regulated by Se 10 treatment at both Cd 5 and Cd 25 but up-regulated by Se 5 treatment at Cd 25 . The expression of TaNramp5-a, TaNramp5-b, TaHMA3-a, TaHMA3-b and TaHMA2 in shoot were down-regulated by increasing Se supply at Cd 5 , and Se 5 treatment up-regulated the expression of those genes in shoot at Cd 25 .


Background
Cadmium (Cd) is one of the most dangerous heavy metals due to its detrimental effect on agricultural soil and potential harm to human health [1]. It is generally believed that plants are the major source of Cd uptake by human beings. Thus, Cd can harm human health through the enrichment effect of food chain.
Wheat is not only one of the principal food in the north of China but also the most important grain crop in the world [2]. Cd-polluted wheat accumulation in humans may cause many diseases, such as anemia, osteoporosis, kidney damage and hypertension [3]. Therefore, it has become an urgent public health problem to reduce the accumulation of Cd in wheat and maintain food safety [4].
Although Cd has no essential biological function in plants, the accumulation of Cd in plants will produce obvious toxic effects, including destroying chlorophyll, inhibiting photosynthesis and crop growth and development, reducing yield and quality [5]. The intracellular and extracellular mechanisms for detoxi cation in plants have gradually developed in the process of adapting to heavy metal stress.
Binding in the cell wall and the transfer to vacuoles may be associated with metal tolerance [6]. The toxicity and migration ability of heavy metals are closely related to their chemical forms. This suggests that Cd chemical forms could affect the movement of Cd in plants and be one of the major mechanisms of heavy metal detoxi cation [7]. The total amount of Cd entering plants is determined by the absorption capacity of Cd in root. Cd in soil is absorbed by plant root and transported to other parts through transporters for some essential elements, such as manganese (Mn), zinc (Zn) and iron (Fe) [8]. At least seven families of transporters participate in the Cd transport in plants, including natural resistanceassociated macrophage proteins (NRAMP), heavy metal ATPases (HMA), ATP-binding cassette transporters(ABC), Zrt/Irt-like proteins (ZIP), H + /cation exchanger (CAX), LCT transporter and cation e ux family (CE) [9]. Se is an essential trace element for humans, animals and plants [10]. The Se can promote the growth and development of plants by improving antioxidant function and regulating photosynthesis.
In addition, Se plays a vital role in plant resistance to adversity stress and the alleviation of the toxicity of heavy metals [11]. The Se is also a bene cial element for human, which can maintain human health by improving immunity, resisting aging and reducing cancer risk [12]. In recent years, many research results showed that Se and Cd in plants are antagonistic. Sun et al. [13] found that Se could reduce Cd concentration in maize and promote maize growth under Cd stress. Wan et al. [14] also reported that the translocation of Cd from root to shoot reduced effectively with the increase of Se supply in rice seedlings.
In addition, Ahmad et al. [15] found that Se reduces Cd toxicity by regulating antioxidative system in Brassica juncea. Shanker et al. [16] revealed that Se and Cd can be combined to form a complex, thus reducing the toxicity of Cd. These studies suggest that applying Se fertilizer is an effective measure to reduce Cd accumulation in plants.
The aims of the present study was to: i) re-examine the effects of different Se supply rates on Cd uptake and translocation; ii) investigate the subcellular distribution and chemical forms of Cd in response to different Se supply rates; and iii) investigate the expression of Cd transporter genes regulated by different Se supply rates, under two levels of Cd stress using a hydroponic trial. Our results will contribute to a better understanding of the mechanism of Se inhibiting Cd uptake and translocation in winter wheat.

Plant material and experimental designs
Winter wheat (Triticum aestivum cv Zhengmai379, obtained from Henan Agricultural High Tech Group Co., Ltd.) seeds were sterilized for 15 min with 10% NaClO, rinsed by deionized water, and then cultured at 25 ℃ for 5 days. Then, 20 seedlings of the same size were transferred to the plastic pot containing 4L nutrient solution. The composition of nutrient solution was: 6.0 mM KNO 3  CdCl 2 at two levels: 5 and 25 μM and Se was added as Na 2 SeO 3 at three levels: 0, 5, and 10 μM after seedlings transferred for one week. Six treatments were included: Cd 5 Se 0 , Cd 5 Se 5 , Cd 5 Se 10 , Cd 25 Se 0 , Cd 25 Se 5 and Cd 25 Se 10 . Each treatment was replicated three times. The quarter and half strength nutrient solutions were provided at the rst and second weeks, respectively, followed by full-strength nutrient solutions. The greenhouse conditions were as follows: relative humidity 70%, 14h light / 10h dark at 25 / 18 °C, light intensity of 400 μmol m −2 s −1 .
The seedlings were harvested after 21 days, and the shoot and root were separated. Ten seedlings were dried in an electric oven at 60°C to analyze the Cd concentration in plant issues. The others were frozen in liquid nitrogen immediately, and then stored at -80℃for further subcellular fractions, chemical forms and gene expression analysis.

Determination of Cd concentration
The Cd concentration in plant issues were determined by the method of Liu et al. [17]. Dry samples were powdered and digested in the mixture of HNO 3 :HClO 4 (4:1, v/v). The Cd concentration in solution were determined using the ame atomic absorption spectrophotometer (ZEEnit 700, Analytik Jena AG, Germany).

Determination of root morphology
After 14 days of seedling growth, one seedling from each pot was taken for the root morphological analysis. The root length, root surface area, root volume, and average root diameter of wheat were measured by using the root imaging analysis software WinRHI-ZO Version 2009 PRO (Regent Instruments, Quebec City, Canada) .

Determination of Subcellular fractions
According to Zhao et al. [18], frozen samples were homogenized in pre-cold extraction buffer containing 50 mM Tris-HCl (pH 7.5), 1.0 mM dithiothreitol (C 4 H 10 O 2 S 2 ) and 250 mM sucrose at the ratio of 1:20 (w/v). The mixtures were centrifuged at 924×g for 15min and the cell wall fraction was obtained in the residue. After centrifuging the supernatant at 20,000×g for 45 min, the supernatant solution and precipitate were called soluble fraction and cell organelle fraction, respectively. All steps were carried out at 4 ℃. The mixture of HNO 3 :HClO 4 (4:1, v/v) was used for wet digestion of different fractions and Cd concentration in digestion solution was determined using the ame atomic absorption spectrophotometer (ZEEnit 700, Analytik Jena AG, Germany).
About 0.5 g of frozen samples were added to the extraction solution at the ratio of 1:10 (w/v), and then it was shaken at 25℃ for 22 h, and centrifuged at 5000×g for 10 min. The precipitate was re-suspended with the same extractive solution twice, shaken at 25 ℃for 2 h , and then centrifuged at 5000×g for 10 min. Pooled the supernatant after three centrifugations, and evaporated to 1-2 mL on an electric plate. Each form of Cd was digested with HNO 3 :HClO 4 (4:1, v/v) and Cd concentration were analysed by ame atomic absorption spectrophotometer (ZEEnit 700, Analytik Jena AG, Germany).
Expression of TaNramp5-a, TaNramp5-b, TaHMA3-a ,TaHMA3-b , TaHMA2 Total RNA was extracted from seedling shoot and root, and then used for rst-strand cDNA synthesis using a PrimeScript™ RT reagent Kit (TakaRa) in accordance with the manufacturer's protocol. The expression was determined with TB green premix Ex Taq™ II (TakaRa). Relative gene expression was calculated by the 2 −△△Ct method. The primers for TaHMA2 was obtained from Tan et al. [20], and the primers for TaNramp5-a, TaNramp5-b, TaHMA3-a and TaHMA3-b were designed by GenScript Real-time PCR (TaqMan) Primer Design Online (https://www.genscript.com.cn/ based on the mRNA sequences obtained from the Ensembl database (http://plants.ensembl.org/). Primers for the genes of interest and reference genes are detailed in Table S1.

Statistics analysis
The main effects and interactions of Cd and Se were statistically examined by two-way ANOVA using SPSS 7.05 software (Chicago, USA). Tukey-test was used for multiple comparisons at a 5% signi cance level (P<0.05).

Cd concentrations, accumulation and migration rate
The Cd and Se treatments had signi cant effects on Cd concentration and accumulation in shoot and root as well as Cd migration rate from root to shoot (P < 0.01; Table S2); Their interaction had signi cant effects on the Cd concentration (P < 0.01; Table S2) and migration rate from root to shoot (P < 0.05; Table  S2).
The Cd concentration in root was higher than that in shoot ( Fig. 1A and B). At Se 0 treatment, Cd concentration in shoot and root was signi cantly increased by increasing Cd stress level; at Se 5 and Se 10 treatments, Cd concentration in root was also signi cantly increased by increasing Cd stress level. Compared with Se 0 , Se 5 and Se 10 signi cantly decreased the shoot Cd concentration at each Cd stress level, with the range of decreased degree from 27.6% to 67.7% (Fig. 1A). Similarly, Se 5 and Se 10 signi cantly decreased the root Cd concentration at each Cd stress level, with the range of decreased degree from 18.6% to 53.6%, except for the no obvious effect of Se 5 on the root Cd concentration at Cd 5 ( Fig. 1B).
The Cd accumulation in root was also higher than that in shoot ( Fig. 1C and D). At Se 0 and Se 5 treatments, with the increase of Cd stress level, Cd accumulation in shoot was signi cantly decreased.
Compared with Se 0 , Se 5 and Se 10 signi cantly decreased the shoot Cd accumulation at each Cd stress level, with the range of decreased degree from 33.3-71.6% (Fig. 1C). The Se 10 signi cantly decreased the root Cd accumulation at each Cd stress level by 46.9% and 61.5% (Fig. 1D).
Compared with Cd 5 treatment, Cd migration rate from root to shoot was signi cantly decreased by Cd 25 treatment at Se 5 ( Fig. 2). At each Cd stress level, Se 5 and Se 10 signi cantly reduced the Cd migration rate from root to shoot, with the range of decreased degree from 18.8% to 30.3%.

Root morphology
The Cd, Se treatments and their interaction had signi cant effects on root length, root total surface area and root volume (P < 0.01; Table S3). The Se treatments had signi cant effects on the average root diameter (P < 0.01; Table S3).
The root length, root volume and root surface area were reduced signi cantly with increasing Cd stress (Fig. 3A, C and D). At Cd 5 , Se 5 and Se 10 signi cantly increased the root length, surface area and root volume, but decreased the average root diameter in winter wheat, with the range of decreased or increased degree from 12.3% to 89.2%. At Cd 25 , Se 5 and Se 10 signi cantly reduced the average root diameter by 11.0% and 19.3%, respectively; but Se 5 and Se 10 signi cantly increased the root volume by 57.2% and 46.9%, respectively. Cd subcellular fraction and distribution The Cd, Se treatments and their interaction had signi cant effects on subcellular fractions of Cd in tissues of wheat seedlings (P < 0.01; Table S4).
The Cd concentration in each fraction of shoot and root was signi cantly increased by increasing Cd stress level, except for the Cd concentration in cell wall of shoot at Se 10 , that in soluble fraction and cell organelle of shoot at Se 5 and Se 10 , and that in cell organelle of root at Se 10 (Table 1)  Values are means of three independent replicates (± sd). For each trait, means followed by different letters are signi cantly different from each other according to two-way ANOVA followed by Turkey multiple comparison (P < 0.05).
In both of shoot and root, the proportion of Cd in soluble fraction was higher than that in cell organelle or cell wall (Fig. 4). Cd proportion in cell organelle of shoot at Se 0 and Se 10 , and that in cell wall of shoot at  Values are means of three independent replicates (± sd). For each trait, means followed by different letters are signi cantly different from each other according to two-way ANOVA followed by Turkey multiple comparison (P < 0.05).
The Cd proportion in each chemical forms of shoot and root was signi cantly increased by increasing Cd stress level, except for FNaCl-Cd of shoot and root, FW-Cd of root, FHAC-Cd of root at Se 10 ,FE-Cd of shoot

Se inhibits Cd absorption via altering root morphology in winter wheat
In our study, Se supply decreased Cd concentration and accumulation in both shoot and root (Fig. 1), indicating Se supply could inhibit Cd absorption in winter wheat. Huang et al. [21] found that Se application reduced Cd concentration in brown rice via a pot experiment, and Lin et al. [22] reported that Se decreasing the toxicity and accumulation of Cd in rice was related to the reduced Cd uptake. Plants absorb nutrients mainly through the root system [23]. Many studies showed that Cd stress could lead to the short root length, thick root diameter and reduced lateral root [24]. Our results observed that Se alleviated the toxic effect of Cd on the root growth of winter wheat, especially at low Cd stress, showing the increased root length, root surface area, root volume and the decreased root diameter by Se supply at Se 10 and FW-Cd of shoot at Se 5 ( Fig. 5). In root, Se 5 and Se 10 increased the proportion of FE-Cd, FNaCl-Cd and FHAC-Cd, with the range of increased degree from 9.38% to 135%; but Se 5 and Expression of TaNramp5-a, TaNramp5-b, TaHMA3-a, TaHMA3-b and TaHMA2 The Cd, Se treatments and their interaction had signi cant effects on the transcript levels of TaNramp5-a, TaNramp5-b, TaHMA3-a, TaHMA3-b, TaHMA2 in shoot and root (P < 0.05 or P < 0.01; Table S6). The transcript levels of TaNramp5-a, TaNramp5-b, TaHMA3-a, TaHMA3-b and TaHMA2 in root were higher than those in shoot, except for the transcript level of TaHMA2 in the treatments of Cd 25 Se 5 and Cd 25 Se 10 (Fig. 6). In root, the transcript levels of TaNramp5-a, TaNramp5-b, TaHMA3-a and TaHMA3-b were signi cantly increased with the increase of Cd stress level; increasing Cd stress signi cantly increased the transcript level of TaHMA2 at Se 0 but decreased that at Se 5 and Se 10 ( Fig. 6A, C, E, G and I).
In shoot, increasing Cd stress signi cantly decreased the transcript levels of TaNramp5-b and TaHMA2 at Se 0 , but increased the ve genes transcript levels at Se 5 as well as the transcript levels of TaHMA3 level in root and TaNramp5-a transcript level in shoot, but increased transcript levels of TaHMA3-a and TaHMA2 in shoot. (Fig. 3). However, Ding et al. [25] found that the addition of 0.8 mg L − 1 Se to the treatments containing 4 mg L − 1 Cd increased the root length, surface area, volume, and average diameter of rice. The root morphology have a great in uence on the absorption of minerals [5]. And the ne root are the most active part of the root system for mineral absorption [26]. Nazar et al. [27] also noted that plant nutrients such as iron (Fe), manganese (Mn) and zinc (Zn) and Cd compete for the same transporters. Therefore, the inhibited Cd uptake by Se application in this experiment may be related to the decreased root diameter and the increased mineral nutrient uptake by root.
Se inhibits Cd transport via altering the distribution of Cd in subcellular fraction, chemical forms in tissues of winter wheat Our study suggested Se 5 and Se 10 signi cantly decreased Cd migration rate from root to shoot, and Cd concentration in cell wall, soluble fraction and cell organelle of shoot and root (Table 1 and Fig. 2). The decreased Cd concentration in subcellular fraction was due to the decreased Cd concentration in winter wheat by Se supply. It was also suggested that most of Cd accumulated in the soluble fraction, followed by that in the cell wall (Table 1 and Fig. 4). Our results are consistent with the results of Li et al. [28], who found that the majority of Cd was compartmentalized in the soluble fraction (53-75%) and bound to the cell wall (19-42%) in Agrocybe aegerita. Cd in soluble fraction and cell wall is easily chelated and xed by organic substances, so it is di cult to transfer to other parts [29]. Li et al. [30] found that Cd in the soluble fraction of wheat root tended to combine with heat-stable protein (HSP), thus reducing the mobility and toxicity of Cd. In addition, vacuole (involved in the soluble fraction) is considered to accumulate the greatest amount of Cd and is the place where waste and by-products are accumulated [31]. Heavy metals can be separated in vacuoles through bounding with various proteins, organic acids and organic bases [32]. In our study, Se supply enhanced Cd accumulation in soluble fraction of shoot (Fig. 4A), indicating that Se supply could inhibit Cd migration to other organs thus to alleviate the Cd toxicity. Cell wall fraction can bind Cd ions reduce the transport to other parts, which is the rst barrier to protect the protoplast from Cd toxicity [33]. Cd proportions in cell wall of root were increased by Se 5 and Se 10 , respectively (Fig. 4B), suggesting Se supply enhanced Cd accumulation in root thus to inhibit Cd transport form root to shoot.
Different chemical forms of Cd have distinct migration capacity. For example, Compared with undissolved Cd phosphate (FHAC-Cd) and Cd oxalate (FHCl-Cd), inorganic and organic water-soluble Cd (FE-Cd and FW-Cd, respectively) have higher migration ability and greater harm to plant cells [7]. Some studies showed that FNaCl-Cd played an important role in the alleviation of Cd toxicity [18,34]. In our study, Cd was mainly integrated with pectates and protein (FNaCl-Cd) in shoot and existed in the form of FW-Cd and FNaCl-Cd in root (Table 2 and Fig. 5). It indicates that Cd easily migrate from root to shoot in the water-soluble form but the toxicity of Cd also can be alleviated via converting Cd into undissolved pectate and protein-bound form. Qiu et al. [35] found that the majority of Cd in both the root and shoot of cabbage was in the extraction of 1 M NaCl. Some speci c polar compounds contain hydroxyl or carboxyl which can combine with Cd to form a non-toxic complex [18]. Se supply signi cantly decreased the total proportion of active Cd (FE-Cd and FW-Cd) but increased the proportion of FNaCl-Cd and FHAC-Cd in root, suggesting that Se supply reduced the mobility of Cd from root to shoot via promoting the transformation of Cd from active form to inactive form in root. The total proportion of active Cd (FE-Cd and FW-Cd) in shoot was decreased by high Se (Se 10 ) supply at Cd 25 , suggesting that high level of Se supply could inhibit the mobility of Cd in shoot at high Cd stress level. Down-regulation of Cd transporter genes might be responsible for Se-decreased Cd accumulation in winter wheat It is widely believed that Cd enters plant root mainly through the Mn channel protein Nramp5 [36].
Nramp5 is a member of the Nramp family, located on the plasma membrane of plant roots [36]. In our study, the expression of TaNramp5-a and TaNramp5-b was found in both root and shoot, and that were signi cantly increased with the increase of Cd concentration (Fig. 6A, B, C and D), suggesting that Nramp5 might be involved in the absorption and transport of Cd in wheat plants. It is in agreement with the results of Ma et al. [37] showing that the expression of OsNramp5 was signi cantly increased with increasing Cd concentration. Tang et al. [38] and Sasaki et al. [36] observed that knockout of OsNramp5 can signi cantly reduce the Cd concentration in root and shoot of rice. In our study, Se supply signi cantly decreased the expression of TaNramp5-a and TaNramp5-b in shoot as Cd stress was low ( Fig. 6B and D), indicating that Se supply might inhibit the remobilization of Cd in shoot. In addition, Se supply signi cantly decreased the expression of TaNramp5-a and TaNramp5-b in root ( Fig. 6A and C), indicating that the down-regulation of TaNramp5-a and TaNramp5-b by Se supply might be helpful to decrease Cd uptake in wheat. Cui et al. [39] also found that Se pretreatment decreased the expression of OsNramp5 thus to inhibit Cd uptake.
Heavy metal ATPases (HMAs) is responsible for the transmembrane transport of cations and play an important role in Cd transport. HMA3 (heavy metal ATPase3), is located on the vacuole membrane in the root. And it is involved in the sequestration of Cd into the vacuoles of root cells, thereby decreasing the transport of Cd to the shoot and reducing the toxicity of Cd [40]. Sasaki et al. [41] reported that overexpression of OsHMA3 leaded to decreased root-to-shoot translocation of Cd. In our study, the expression of TaHMA3-a and TaHMA3-b was found in both root and shoot, and that were signi cantly increased with the increase of Cd concentration (Fig. 6E, F, G and H), suggesting that HMA3 might be responsible for the transport of Cd in wheat plants. Se supply down-regulated the expression of HMA3 in shoot at Cd 5 but up-regulated that at Cd 25 ( Fig. 6F and H ), also indicating that Se supply could inhibit the remobilization of Cd in shoot by enhancing the sequestration of Cd into the vacuoles as Cd stress was high. Cui et al. [39] also showed that Se pretreatment activated the expression of OsHMA3 thus to enhance the transport of Cd into vacuoles.
HMA2 (heavy metal ATPase2), which is homologous with HMA3 and belongs to the heavy metal ATPase family. HMA2 plays a role in the loading of Cd and Zn into xylem and get involved in the root-to-shoot translocation of Cd and Zn [20]. Our results showed that the expression of TaHMA2 was found in both root and shoot, and that were signi cantly increased with the increase of Cd concentration ( Fig. 6I and J).
It suggested that HMA2 might be involved in the transport of Cd in wheat plants. This is consistent with the results of Tan et al. [20] showing that the overexpression of HMA2 in wheat and rice increased the root-shoot translocation of Zn/Cd. A recent report showed that La decreased Cd accumulation in wheat, which may be related to the TaHMA2 down-regulation [42]. In our study, Se supply signi cantly decreased the expression of TaHMA2 in root, indicating that the down-regulated TaHMA2 by Se supply might contribute to the inhibited Cd root-to-shoot translocation and nal decreased Cd accumulation in shoot of winter wheat. The expression of TaHMA2 in shoot was signi cantly increased by Se supply at Cd 25 (Fig.   J), suggesting that Se supply might promote the remobilization of Cd in shoot by up-regulating the expression of TaHMA2 as Cd stress level was high.

Conclusions
Our results showed that TaNramp5, TaHMA3 and TaHMA2 might be responsible for the uptake and transport of Cd in wheat plants. Se supply could inhibit Cd absorption and root-to-shoot transport in winter wheat. Our results suggested that Se supply inhibit Cd absorption via reducing the root diameter

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.  Cd concentration and accumulation in shoot (A and C, respectively) and root (B and D, respectively) of winter wheat (Triticum aestivum cv Zhengmai379) seedlings.

Figure 2
Cd migration rate from the root to shoot of winter wheat seedlings (Triticum aestivum cv Zhengmai379) seedlings.

Figure 4
Proportions of Cd in subcellular fractions of winter wheat seedlings (Triticum aestivum cv Zhengmai379) seedlings.

Figure 5
Proportions of Cd in chemical forms of winter wheat seedlings (Triticum aestivum cv Zhengmai379) seedlings.