Xiao Anwen, Chen Danting, Li Wai Chin, Ye Zhihong

Research Paper

Root Morphology and Anatomy Affect Cadmium Translocation and Accumulation in Rice

Xiao Anwen1, 2, Chen Danting1, Li Wai Chin2, Ye Zhihong1

(School of Life Sciences, Sun Yat-sen University, Guangzhou 510006, China; Department of Science and Environmental Studies, The Education University of Hong Kong, 999077 Hong Kong, China)

Paddy fields contaminated with cadmium (Cd) present decreased grain yield and produce Cd-contaminated grains. Screening for low-Cd-accumulating cultivars is a useful method to reduce the amount of Cd in the grains. The present study aimed to examine the roles of the root morphology and anatomy in Cd translocation and accumulation in rice plants. Twenty-two rice cultivars were used in the first experiment, after which two cultivars [Zixiangnuo (ZXN) and Jinyou T36 (JYT36)] were selected and used in subsequent experiments under hydroponic conditions. The results showed that there were significant differences in Cd concentrations in the shoots (ranging from 4 to 100 mg/kg) and the Cd translocation rates (shoot/root) (from 7% to 102%) among the 22 cultivars, and the shoot Cd concentration was significantly correlated with the Cd translocation rate of the 22 cultivars under 0.1 mg/LCd treatment.Compared with cultivar ZXN, JYT36 had greater root Cd uptake and accumulation but lower shoot Cd accumulation and Cd translocation rate.The number of root tips per surface area of cultivar ZXN was greater than that of JYT36, while the average root diameter was lower than that of JYT36. Compared with ZXN, JYT36 had stronger apoplastic barriers, and the Casparian bands and suberin lamellae in the root endodermis and exodermis were closer to the root apex in both the control and Cd treatments, especially for suberin lamellae in the root exodermis with Cd treatments, with a difference of 25 mm. The results also showed that, compared with ZXN, JYT36 had greater percentages of Cd bound in cell walls and intracellular Cd but lower Cd concentrations in the apoplastic fluid under the Cd treatment. The results suggested that Cd translocation, rather than root Cd uptake, is a key process that determines Cd accumulation in the rice shoots. The root morphological and anatomical characteristics evidently affect Cd accumulation in the shoots by inhibiting Cd translocation, especially via the apoplastic pathway. It was possible to pre-screen low-Cd-accumulating rice cultivars on the basis of their rootmorphology, anatomical characteristics and Cd translocation rate at the seedling stage.

apoplastic pathway; Cd stress; Cd translocation; Cd accumulation; rice; root morphology and anatomy

Owing to mining and other human activities, large areas of paddy fields in Asia have been contaminated with cadmium (Cd) (Williams et al, 2009; Wang et al, 2018). Compared with other food crop and vegetable species, rice can accumulate Cd more effectively from contaminated soils (Grant et al, 2008; Shi et al, 2019). Paddy fields contaminated with Cd produce rice grains containing Cd, which is a major contributor to the intake of Cd by residents (Zhuang et al, 2014). As a nonessential element, Cd is toxic and can result in human diseases, such as cancer and itai-itai disease (Nordberg et al, 2002). To avoid damage from Cd, it is necessary and critical to reduce Cd accumulation in the rice grains. Screening for rice cultivars that accumulate low amounts of Cd in the grains has been considered a useful method to reduce the health risk of rice consumption (Chi et al, 2018; Chiao et al, 2019, 2020).

In rice plants, the major transport processes that influence Cd accumulation in the grains include root uptake and transfer of Cd from the roots to the shoots by xylem loading, redistribution by the nodes of Cd through intervascular transfer, and leaf blade remobilization of Cd to the grains by the phloem stream (Uraguchi and Fujiwara, 2013). Of these three processes, the first process is considered the key process for determining grain Cd accumulation (Liu et al, 2007; Rodda et al, 2011; Uraguchi and Fujiwara, 2012, 2013).

The differences in Cd uptake and translocation among rice cultivars point to an interesting phenomenon.Huang et al (2017) reported that some rice cultivars exhibit greater Cd uptake by their roots but lower Cd accumulation in their aboveground parts.These differences may be due to different abilities to translocate Cd from the roots to the shoots among the cultivars (Liu et al, 2003, 2007).Cd translocation from the roots to the shoots occurs mainly through two pathways: the cell-to-cell pathway (including the symplastic and transmembrane pathways) and the apoplastic pathway (Kreszies et al, 2018). Research has focused on the cell-to-cell pathway in rice (Uraguchi and Fujiwara, 2012), but few studies have focused on the apoplastic pathway. Plants can regulate the apoplastic pathway via the formation of apoplastic barriers, which decreases and hinders the transport of water and ions by apoplastic pathway (Ma and Peterson, 2003; Tao et al, 2017). Because of the negative effect on ion transport, the formation of apoplastic barriers probably affects ion accumulation in the aboveground parts of plants (Lux et al, 2011). Compared with those of maize roots grown in hydroponics, the apoplastic barriers of maize roots grown in aeroponics developed much stronger, resulting in greater Cd accumulation in maize grown in hydroponics (Redjala et al, 2011). Similarly, inand, the clones/ecotypes have more mature apoplastic barriers and accumulate less Cd in the aboveground parts of the plants (Lux et al, 2004; Tao et al, 2017). In rice plants, apoplastic barriers are also considered to have an inhibitory effect on Cd translocation (Huang et al, 2019; Qi et al, 2020). In addition to the root anatomy, the root morphology may also influence Cd translocation. For example, maize cultivars with a greater root average diameter more effectively inhibit Cd translocation (Maksimovi? et al, 2007). Rice plants with fewer tips per surface area may have lower Cd translocation (Huang et al, 2019).However, currently, whether variation in the root anatomy and morphology of rice cultivars leads to differences in Cd translocation and accumulation in aboveground parts is unclear.

The present study aimed to investigate the relationship between Cd translocation and Cd accumulation in the shoots of rice plants and the roles of the root morphology and anatomy in rice Cd translocation and accumulation. The hypotheses in the present study were as follows: 1) variation in the root morphology and anatomy resulted in differences in Cd translocation from the roots to the shoots and Cd accumulation in shoots among rice cultivars, and 2) lower Cd- accumulating rice cultivars tended to have stronger root apoplastic barriers to reduce Cd translocation from the roots to the shoots compared with greater Cd-accumulating cultivars. To test these hypotheses, hydroponic experiments were set up to 1) study Cd accumulation, translocation and their relationship in 22 rice cultivars in experiment 1, after which two cultivars with obvious differences in Cd accumulation in the shoots and translocation ability from the roots to the shoots were selected, and 2) investigate the changes in root morphology and anatomy in the two selected cultivars in different Cd treatments and analyze their roles in rice Cd translocation and accumulation. When rice plants are grown in low-Cd-contaminated soil, hydroponic screening is recognized as a more efficient method for pre-screening cultivars (Rout and Das, 2002). Moreover, in previous research, the modelling parameters from hydroponic experimentswere correlated with the Cd concentrations in brown rice grown in the field, indicating that the results of hydroponic experiments have certain guiding significance for the field (Chiao et al, 2019). Therefore, for both hydroponics and fields, this study would provide an improved understanding of the roles of the root morphology and anatomy in Cd translocation and accumulation in rice plants.

RESULTS

Accumulation and translation rates of Cd in 22 rice cultivars

After 21 d of treatment with 0.1 mol/L Cd, the plant growths of eight rice cultivars were inhibited with their tolerant indexes ranged from 90% to 100% (Fig. S1), and the growing parameters of 22 rice cultivars were shown in Table S1.Cd tended to accumulate in the roots rather than in the shoots (Fig. 1). The ranges of Cd concentration in the shoots of the 22 rice cultivars tested were 4 to 100 mg/kg. The concentration of Cd in the shoots of rice cultivar Zixiangnuo (ZXN) was significantly greater than those of the other cultivars, and the Cd concentration of Jinyou T36 (JYT36) was one of the lowest recorded (Fig. 1-A).

The Cd concentrations in the roots ranged from 75 to 142 mg/kg. Interestingly, for cultivar ZXN, the Cd concentration was relatively high in the shoots, but was relatively low in the roots, while in cultivar JYT36, the situation was opposite (Fig. 1-A and -B).For cultivars Liangyoupeijiu, JYT36, Nanjin 44, Nuoyou 1, Suyunuo, Tianyou 2168, Xinnuo 1 and Xiangwannuo 1, the Cd translocation rates (shoot/root) were lower than 7%, but that of cultivar ZXN was 102%, which was significantly greater than those of the other cultivars (< 0.05) (Fig. 1-C).The Cd concentrations in the shoots of the 22 cultivars were significantly and positively correlated with their Cd translocation rates (shoot/root) (Fig. 1-D). However, the Cd concentrations in the shoots of the same cultivars were not correlated with the Cd concentrations in the roots (Fig. S2).

According to the Cd concentration in rice plant tissues and their translocation ability, two cultivars (JYT36, with higher Cd concentration in the roots, and lower Cd concentration in the shoots and Cd translocation rate, and ZXN, with the opposite properties) were selected for further research.

Hydroponic experiment using two cultivars

Cd tolerance and accumulation in rice

The growth of the two rice cultivars was inhibited under Cd stress, with the longest root length, shoot height and biomass decreasing with increasing Cd concentrations. At 2.5 mg/L Cd concentration, the growth of the two rice cultivars was inhibited significantly (< 0.05) compared with the control (Table S2). The concentrations of Cd in the roots of JYT36 were greater than those of ZXN in both the 0.5 and 2.5 mg/L Cd treatments. The Cd concentration in the shoots of ZXN was significantly greater than that of JYT36 in the 0.5 mg/L Cd treatment (< 0.05) (Fig. S3).

Concentration-dependent and time-dependent Cd uptake kinetics

The concentration-dependent uptake kinetics experiment assessed both low- and high-affinity uptake systems. The Cd uptake ability of cultivar JYT36 was significantly greater than that of ZXN in both the low- and high-affinity uptake systems (Fig. 2-A and -B). When the Cd concentration was greater than 2500 nmol/L, the uptake curves of the two cultivars tended to be horizontal, which indicated that Cd absorption tended to be saturated (Fig. 2-A). Compared with cultivar ZXN, the uptake curve of JYT36 tended to be a linear correlation (Fig. 2-B). In the time-dependent uptake kinetics experiment, the Cd uptake ability of cultivar JYT36 was also significantly greater than that of ZXN (Fig. 2-C). Within 0?10 min, the Cd uptake level increased rapidly with increasing treatment time. After 30 min, the uptake curves of the two cultivars tended to be horizontal. Themaxandmof JYT36 were 134.3% and 78.2% greater than those of ZXN, respectively (Table S3), which together indicated that cultivar JYT36 had a greater uptake ability than ZXN did, which was consistent with the results presented in Fig. 2.

Fig. 1. Cd concentrations in shoots (A) and roots (B), Cd translocation rate (shoot/root) (C), and correlation between Cd concentration in shoots and Cd translocation rates (shoot/root) (D) of 22 rice cultivars with 0.1 mg/L Cd treatment for 3 weeks.

LYPJ, Liangyoupeijiu; JYT36, Jinyou T36; SYN, Suyunuo; XN1H, Xinnuo 1; TY2168, Tianyou 2168; NY1H, Nuoyou 1; QXY200, Qianxiangyou 2000; TY122, Tianyou 122; NJ44, Nanjing 44; TY998, Tianyou 998; XWN1H, Xiangwannuo 1; WFY128, Wufengyou 128; TXZ, Texianzhan;ZY808, Zhongyou 808; SY402, Shanyou402; PZ163, Peiyou 163; TY196, Tianyou 196; HY665, Huayou 665; ZD097, Zhongdao 097; SHN, Suihongnuo; ST1H, Shengtai 1; ZXN, Zixiangnuo.

Data are Mean ± SE (= 4). Different lowercase letters indicate significant differences among the cultivars at< 0.05.

Fig. 2. Concentration-dependent and time-dependent kinetics for Cd uptake by rice roots of cultivars Zixiangnuo (ZXN) and Jinyou T36 (JYT36).

A, High-affinity uptake kinetics of Cd in the two cultivars.

B, Low-affinity uptake kinetics of Cd in the two cultivars.

C, Time-dependent kinetics of Cd in the two cultivars.

The 10 d old intact seedlings were treated for 30 min in A and B, and treated with 180 nmol/L Cd in C. Data are Mean ± SE (= 3). Data with different lowercase letters mean significant differences between the treatments at< 0.001 (A) and< 0.05 (B).

Characteristics of root morphology and anatomical structure under Cd stress

Root morphology

The number of root tips per cm2(surface area) of the two cultivars significantly and rapidly decreased with increasing Cd concentrations (Table 1). In the control and 0.5 mg/L Cd treatments, the numbers of root tips per surface area of cultivar ZXN were 82.4% and 130.0% greater than those of JYT36, respectively.The average root diameter of ZXN increased with increasing Cd concentrations and that of JYT36 significantly increased in the 0.5 mg/L Cd treatment compared with the control.The average root diameter of JYT36 was 43.0% and 28.0% greater than that of ZXN in the control and 0.5 mg/L Cd treatments, respectively (Table 1).

Root anatomy

Apoplastic barriers contain the Casparian bands and suberin lamellae. The present results showed that the Casparian bands and suberin lamellae of the endodermis developed earlier than those of the exodermis (Fig. 3). Moreover, the Casparian bands developed before or at the same time as the suberin lamellae (Fig. 3). The sites of the Casparian bands and suberin lamellae in the endodermis and exodermis of the two cultivars were closer to the root apex in the Cd treatment compared with the control (Figs. 3, 4 and 5).

Table 1. Parameters of root morphology in rice with treatments of 0.5 and 2.5 mg/L Cd for 4 weeks.

ZXN, Zixiangnuo; JYT36, Jinyou T36.

Data are Mean ± SE (= 8). Data with different lowercase letters mean significant differences between the treatments in the same cultivar. **,< 0.01 and ***,< 0.001 mean significant differences between the cultivars in the same treatment.

Whether under Cd stress or not, the development of the apoplastic barriers was different between the two cultivars (Figs. 3, 4 and 5). Compared with cultivar ZXN, the development of Casparian bands and suberin lamellae in the endodermis and exodermis of JYT36 occurred earlier in both the control and 2.5 mg/L Cd treatments (Fig. 3). In cultivar JYT36, the zones of endodermal Casparian band development were between 1 and 5 mm from the root apex, while in cultivar ZXN, the zones were further away from the apex, ranging between 10 and 30 mm (Fig. 3). Similarly, the deposition of the endodermal suberin lamellae in the roots of JYT36 occurred 5 mm from the root apex in the control and 1 mm from the root apex in the 2.5 mg/L Cd treatment, while that of ZXN occurred at 40 mm in the control and at 10 mm in the 2.5 mg/L Cd treatment (Fig. 3). In the control, Casparian bands and suberin lamellae in the exodermis of JYT36 occurred at 30 and 40 mm, which were 20 and 10 mm closer to the root apex than those of ZXN, respectively. When treated with 2.5 mg/L Cd, the exodermal Casparian bands and suberin lamellae of JYT36 matured at 5 mm, which were 15 and 25 mm closer than those of ZXN, respectively.

Fig. 3. Schematic representation of endodermal and exodermal apoplastic barriers of two rice cultivars in control (CK) and Cd (2.5 mg/L) treatments.

ZXN, Zixiangnuo; JYT36, Jinyou T36. Casparian bands and suberin lamellae in endodermis are represented by red and green lines, respectively. Casparian bands and suberin lamellae in exodermis are represented by brown and blue lines, respectively. The dotted lines indicate the early and immature deposition of the barriers and solid lines represent the Casparian bands or suberin lamellae which have developed maturely in the zone.

To display the differences in the root anatomy structure between the two cultivars in the different Cd treatments, pictures of the section at 20 mm from the root apex were selected. In the control, the endodermal Casparian band of cultivar ZXN did not appear, but cultivar JYT36 developed an endodermal Casparian band and suberin lamellae (Figs. 4-A, 4-B, 5-A and 5-B). Moreover, the two cultivars had no exodermal Casparian bands or suberin lamellae in the control (Figs. 4-E, 4-F, 5-E and 5-F).

Cd subcellular distribution

The results from the Cd subcellular distribution experiment showed that Cd in the roots was transformed into bound Cd (Cd bound in the cell wall and membrane), intracellular Cd and Cd in the apoplastic fluid.Cd in the roots was distributed mainly in the cell and was bound to the cell wall, while the amount of Cd in the apoplastic fluid was lower (Fig. 6-A).Compared with cultivar ZXN, JYT36 had more bound and intracellular Cd in the roots in the Cd treatment. With Cd treatment, the bound Cd in JYT36 was 17%?21% greater than that in ZXN. In the 0.5 mg/L Cd treatment, the intracellular Cd in the roots of JYT36 was 1.55 times greater than that of ZXN (Fig. 6-A), while the Cd concentration in the apoplastic fluid in JYT36 roots was significantly lower than that in ZXN roots (Fig. 6-B).

Fig. 4. Comparison of development of Casparian bands at 20 mm distance from rice root tips with treatments of control and 2.5 mg/L Cd for 4 weeks.

A to D are endodermis sections, and E to H are exodermis sections; A, C, E and G are cultivar Zixiangnuo, and B, D, F and H are cultivar Jinyou T36; A, B, E and F are in the control treatment, and C, D, G and H are in 2.5 mg/L Cd treatment. Arrow heads refer to the Casparian bands. Scale bars are 20 μm in A, B, E, F and G, and 50 μm in C, D and H.

Fig. 5. Comparison of development of suberin lamellae at 20 mm distance from rice root tips with treatments of control and 2.5 mg/L Cd for 4 weeks.

A to D are endodermis sections, and E to H are exodermis sections; A, C, E and G are cultivar Zixiangnuo, and B, D, F and H are cultivar Jinyou T36; A, B, E and F are in control treatment, and C, D, G and H are in 2.5 mg/L Cd treatment. Scale bars are 20 μm in A, B, E and F, and 50 μm in C, D, G and H.

DISCUSSION

Cd translocation determined Cd accumulation in shoots

The accumulation of Cd in the shoots is affected by root Cd uptake and Cd translocation from the roots to the shoots (Uraguchi and Fujiwara, 2013). In the present study, the Cd concentrations in the shoots of the 22 cultivars were not correlated with the Cd concentrations in the roots (Fig. S2), which indicated that Cd uptake by rice roots was not a determining factor in the Cd concentration in the shoots. In addition, in both the concentration-dependent and time-dependent uptake kinetics experiments, the Cd uptake ability of cultivar JYT36 was significantly greater than that of ZXN (< 0.05), and themaxof JYT36 was 2.34 times over that of ZXN (Fig. 2 and Table S3), which was consistent with the fact that the Cd concentration in the roots of JYT36 was significantly higher than that of ZXN (Fig. 1-B). However, the Cd concentration in the aboveground parts of cultivar ZXN was higher than that of JYT36 (Figs. 1-A and S3), which indicated that the Cd uptake ability by the roots was not directly responsible for the differing Cd accumulations in the aboveground parts in the two rice cultivars. Uraguchi et al (2009) also reported that the Cd uptake ability of rice cultivar Habataki (high Cd concentration in the grains) is lower than that of cultivar Sasanishiki (low Cd in the grains), which also indicated that Cd uptake by the roots may not be as important to Cd accumulation in the aboveground rice parts (e.g., the grains). In summary, the root Cd uptake kinetics was not directly responsible for the differing Cd accumulation in the aboveground parts in the rice cultivars studied.

Fig. 6. Distribution of Cd in rice roots with treatments of 0.5 and 2.5 mg/L Cd for 4 weeks.

A, Cd bound in cell wall and intracellular Cd.

B, Cd concentration in the apoplastic fluid.

ZXN, Zixiangnuo; JYT36, Jinyou T36. Data are Mean ± SE (= 3). * and ** mean significant differences between the cultivars in the same treatment at the 0.05 and 0.01 levels, respectively.

In this study, the positive correlation between the Cd translocation rate (shoot/root) and Cd concentration in the shootssuggested that Cd translocation from the roots to the shoots, rather than root Cd uptake, plays a key role in Cd accumulation in the aboveground parts of the rice plants (Fig. 1-D), which was consistent with the results of Uraguchi et al (2009). Therefore, Cd translocation from the roots to the shoots can explain the difference in Cd accumulation in the aboveground parts of the rice plants in the two cultivars, while the Cd translocation rate (shoot/root) of cultivar ZXN was significantly greater than that of JYT36 (< 0.05) (Figs. 1-A, 1-C and S3). The present results showed that Cd translocation from the roots to the shoots, rather than root Cd uptake ability, is a key step in determining Cd accumulation in the above- ground parts of rice plants.

Strong apoplastic barriers, a large average root diameter and a small number of root tips resulted in low Cd translocation

Cadmium translocation from the roots to the shoots is influenced largely by the channel between the root epidermis and xylem, and the formation of apoplastic barriers can limit Cd translocation to the xylem (Lux et al, 2011; Huang et al, 2019). Compared with ZXN, the formation sites of the Casparian bands and suberin lamellae in the endodermis and exodermis of JYT36 occurred closer to the root apex, and the development of the apoplastic barriers occurred earlier (Fig. 3). At the same site in the roots, whether in the control or Cd treatment, the Casparian bands and suberin lamellae in the endodermis and exodermis of JYT36 were more mature than those of ZXN, which indicated that, compared with ZXN, JYT36 had stronger apoplastic barriers (Figs. 4 and 5).In the root endodermis and exodermis, root apoplastic barriers developed from the cell wall owing to the deposition of hydrophobic biomacromolecule polymer lignins and suberins (Schreiber et al, 1999; Man et al, 2018), indicating that more lignin and suberin depositions usually built a stronger apoplastic barrier. On account of the hydrophobic properties, the strong apoplastic barriers are physical barriers, which make the cell wall less penetrable and protect against Cd radial transport to the xylem in the stele. This would slow apoplastic fluid and decrease the Cd concentration in apoplastic fluid, resulting inthe Cd transported by the apoplastic pathwaybeing intercepted and absorbed by the cell near the barrier (Redjala et al, 2011; Yamaguchi et al, 2011), indicating that the intracellular Cd in the roots may increase. These previous results were consistent with the present results: compared with ZXN, JYT36, which had stronger apoplastic barriers, had higher concentrations of intracellular Cd in the roots and obviously lower Cd concentrations in the apoplastic fluid (< 0.05) (Fig. 6-B). Moreover, lignin in root apoplastic barriers contains many functional groups, which can effectively integrate Cd into the cell wall (Parrotta et al, 2015; Loix et al, 2017). The present results showed that roots with stronger apoplastic barriers (cultivar JYT36) can effectively combine and retain more Cd in the form of bound Cd (Fig. 6-A). These two processes greatly decreased Cd mobility and retained more Cd in the roots of JYT36 compared with ZXN (Fig. S3).These results suggested that rice roots with stronger apoplastic barriers have a greater ability to limit Cd translocation, which was consistent with the results of other plant species (e.g., maize) reported by Redjala et al (2011). The inhibitory effect of apoplastic barriers on Cd was realized mainly through affecting Cd translocation via the apoplastic pathway.

The average root diameter reflects the distance of water and Cd radial flow from the epidermis to the xylem directly, and an increase in the root diameterenhances the resistance to Cd translocation (Maksimovi? et al, 2007). In the present study, the average root diameter of cultivar JYT36 (low Cd-translocation ability) was significantly larger than that of ZXN (high Cd-translocation ability), which reflected that JYT36 better inhibited Cd translocation (Table 1). The number of root tips is an important index to measure root ion uptake because the apical zone of the root is the region where cation absorption is the most active (Boominathan and Doran, 2003). Generally, only in the root apex and the region where lateral roots emerge can cations enter the xylem through the apoplast pathway, which can translocate Cd cations well (White, 2001; Moore et al, 2002). The number of root tips reflects the effective area of Cd absorption directly and influences the Cd load in the root xylem. The number of root tips per cm2of surface area of cultivar JYT36 was significantly lower than that of ZXN, which reflected that Cd mobility in the JYT36 roots was less than that in ZXN, allowing for the retention of more Cd (Table 1).

Compared with ZXN, the Cd uptake ability of JYT36 was greater, resulting in greater Cd accumulation in the roots. However, JYT36 had more mature apoplastic barriers in the roots, a larger average root diameter and a smaller number of root tips per surface area in both the control and Cd treatments to inhibit Cd translocation (especially by the apoplastic pathway), resulting in more Cd retained in the rice roots in the form of bound and intracellular Cd,alow Cd translocation rate (shoot/root)and low Cd accumulation in the aboveground parts. It was possible to pre-screen low- Cd-accumulating rice cultivars according to their rootmorphology, anatomical characteristics and Cd translocation rate (shoot/root) at the seedling stage.

To further test the conclusions of this study, it would be meaningful to further study whether the effect of the root morphology and anatomy on Cd translocation is identical in other rice cultivars from differentsubspecies (includingand),and field experiments are needed. Moreover, searching for root morphology- and anatomy-determining genes and their related underlying mechanisms could contribute to the screening of low-Cd-accumulating cultivars.

METHODS

Experiment 1. Cd accumulation in 22 rice cultivars

Twenty-two rice cultivars with different Cd accumulation levels in their grains were chosen according to our previous study (Chi et al, 2018), and their information is summarized in Table S4. The seeds were surface-sterilized with 30% H2O2solution for 15 min and washed with deionized water, and then germinated in sterilized petri dishes containing moistened filter paper in a controlled chamber (28 oC and 70% relative humidity) (Wu et al, 2011). When radicles appeared, the seeds were transferred to moist acid-washed quartz sand, after which they were allowed to grow unabated for 2 weeks (Armstrong and Armstrong, 2005).

Uniform rice seedlings were transplanted into plastic vessels. To simulate paddy soil conditions, the rice seedlings were incubated with deoxygenated 25%-strength Hoagland’s nutrient solution comprising 0.1% agar (Hoagland and Arnon, 1938; Kotula et al, 2009). Half of rice seedlings were incubated with 0.1 mg/L CdCl2(sample for Cd accumulation, translocation and Cd tolerance index determination), and half were incubated without Cd (control, sample for Cd tolerance index determination) after a 7-day pre-culture period. The nutrient solutions were adjusted to pH 5.5 and renewed every 4 d during the growth period. All vessels were arranged randomly in a controlled greenhouse with natural light supplemented by sodium light, a day/night temperature of 28 oC/22 oC, a day/night photoperiod of 14 h/10 h and a relative humidity of 70%. Four replicates were prepared for each treatment. After 21 d, the plants were carefully washed with deionized water and separated into shoots and roots.

Rice growing parameters and Cd tolerance index

The longest root length, plant height, and shoot and root fresh weights of the rice plants were determined. The tolerance index was used to measure the Cd tolerance of different rice cultivars (Wilkins, 1987) and was calculated according to the following equation: Cd tolerance index (%) = The longest root length (Cd treatment) / The longest root length (control) × 100%.

Cd accumulation and translocation

Rice shoot and root samples were oven-dried at 60 oC and then milled into powder for Cd analyses. The powder of the rice sample was microwave-digested in 5 mL concentrated HNO3(16 mol/L), and the digests were filtered and diluted with 25 mL deionized water. The total concentrations of Cd in the digests of the rice samples were analyzed by an atomic absorption spectroscopy (AAS) (Hitachi, Japan).The Cd translocation rate (shoot/root) was used to measure the Cd translocation abilities of different rice cultivars, which were determined according to the following equation: Cd translocation rate (shoot/root) = (Cd concentration in the shoots / Cd concentration in the roots) × 100%.

Experiment 2. Root properties of two selected cultivars

On the basis of the results of Experiment 1, two rice cultivars (ZXN and JYT36) with obvious differences in their Cd- translocation ability were selected forfurther study. Cultivar ZXN had high Cd accumulation in the aboveground parts, a high Cd translocation rate (shoot/root) and low Cd accumulation in the roots, while cultivar JYT36 had a low Cd translocation rate (shoot/root) and Cd accumulation in the aboveground parts and high Cd accumulation in the roots. The seedling preparation was similar to Experiment 1. The seedlings were divided into four parts in hydroponics for the following tests.

1) Uniform rice seedlings of the two selected cultivars were incubated in deoxygenated 1/4-strength Hoagland’s nutrient solution comprising 0.1% agar (Hoagland and Arnon, 1938; Kotula et al, 2009). Every 4 d, the nutrient solution was renewed. After 4 weeks, uniform rice seedlings were rinsed with deionized water and then used to determine the root uptake kinetics.

2) Uniform rice seedlings were incubated in deoxygenated 1/4-strength Hoagland’s nutrient solution (Hoagland and Arnon, 1938). Cadmium (CdCl2) was applied to the nutrient solution at 0 (control), 0.5 and 2.5 mg/L. The nutrient solution was renewed every 4 d. After 28 d, the rice roots were washed with deionized water and harvested for root morphology determination.

3) To analyze theroot anatomy, uniform seedlings of the two cultivars were incubated with deoxygenated 1/4-strength Hoagland’s nutrient solution (Hoagland and Arnon, 1938) comprising 0 (control) and 2.5 mg/L Cd for 4 weeks, and then the roots were washed with deionized water and harvested.

4) To determine the Cd accumulation in rice tissues and Cd subcellular distribution in the roots, uniform seedlings of the two cultivars were incubated with deoxygenated 1/4-strength Hoagland’s nutrient solution (Hoagland and Arnon, 1938) comprising 0.5 and 2.5 mg/L Cd for 4 weeks. The samples were divided into shoots and roots and washed with deionized water.

Concentration-dependent and time-dependent uptake kinetics

The Cd uptake kinetics in the roots was assessed according to the modified method described by Uraguchi et al (2009). The roots were excised and incubated with test solution that contained Cd. All the test solution contained 5.0 mmol/L 2-(-morpholino)-ethanesulphonic acid (MES) and 0.5 mmol/L Ca(NO3)2, with the pH adjusted to 5.6 using KOH. The concentration-dependent uptake kinetics experiment assessed low-affinity uptake systems (0, 89, 178, 445 and 890 nmol/L Cd) and high-affinity uptake systems (0, 132, 660, 2 640 and 6 600 nmol/L Cd), in which the treatment time was 30 min. In the time-dependent uptake kinetics experiment, the Cd concentration in the test solution was 180 nmol/L, and the treatment times were 5, 10, 30, 60, 120 and 180 min. The test solution was renewed every 30 min. After incubation, the rice roots were rinsed for 15 min with ice-cold phosphate solution that consisted of 1 mmol/L K2HPO4, 5 mmol/L MES and 0.5 mmol/L Ca(NO3)2to remove absorbed Cd from the root-free space. Finally, the roots were washed with deionized water and oven-dried at 60 oC, after which the Cd concentrations in the roots were analyzed by AAS. Four rice seedlings composed one replicate, and three replicates were prepared for each treatment.

Root morphology

Uniform rice seedlings from the control, 0.5 and 2.5 mg/L Cd treatments were selected. Tomeasure the root morphology, the fresh roots were washed, scanned and analyzed by a WinRHIZO root scanner (Regent, STD4800, Canada) (Redjala et al, 2011). The number of root tips per cm2of surface area and the average diameter were determined.Eight replicates were prepared for each treatment.

Root anatomy (apoplastic barriers)

Healthy and uniform adventitious roots (with similar length of approximately 10 cm) of the two cultivars from the control and 2.5 mg/L Cd treatments were selected. To analyze the development of apoplastic barriers (Casparian bands and suberin lamellae), the roots were embedded in tissue-freezing medium (Sakura Finetek, USA) and frozen. Afterwards, cross-sections of each root were taken by a freezing microtome (Thermo Microm HM560, Germany) at the following distances from the root tip: 1, 5, 10, 20, 30, 40, 50 and 60 mm. The root sections were attached to glass slides treated with polylysine. To detect the development of Casparian bands, sections were stained for 1 h with 0.1% berberine hemisulphate and then for 30 min with 0.5% aniline blue (Lux et al, 2005; Kotula et al, 2009). For suberin lamellae, sections were stained with 0.1% Sudan red 7B for 5 h (Lux et al, 2005). After being stained, the sections were viewed and documented with a fluorescence microscope (Zeiss Imager, Z1, Germany) and a digital camera (Nikon, D1, Japan). Five replicates were prepared for each treatment.

Cd accumulation in rice plant tissues and Cd subcellular distribution in rice roots

Healthy and uniform rice seedlings from the 0.5 and 2.5 mg/L Cd treatments were selected. The Cd concentrations in the shoot and root tissues were determined by AAS. The Cd concentrations in the apoplastic fluid, symplast and cell wall were determined by the method described by Ye et al (2012). Fresh rice roots were washed with deionized water and then soaked in ice-cold 5 mmol/L Na2-EDTA solution for 10 min to remove Cd adsorbed onto root surfaces. After being washed with deionized water again, the roots were cut transversely into 1 cm segments and gently rocked for 1 h in 25 mL deionized water at room temperature. The total Cd accumulation in the filtrate was the Cd in the apoplastic fluid. After collecting the apoplastic fluids, the sample was milled to powder in liquid nitrogen and then homogenized with 10 mL ice-cold extraction buffer [50 mmol/L HEPES (C8H18N2O4S), 1.0 mmol/L DTT (C4H10O2S2), 500 mmol/L sucrose and 5.0 mmol/L ascorbic acid, and adjusted to pH 7.5 with NaOH]. The homogenate was centrifuged at 4 000 ×at 4 oC for 15 min. After centrifugation, the supernatant was designated as the symplast fraction, and the pellet was designated as the cell wall fraction. The cell wall fractions were then dried at 70 oC and digested by concentrated HNO3. The Cd concentrations in the apoplastic fluid, symplast (intracellular Cd) and cell wall (bound Cd) were analyzed by AAS.

Quality control and statistical analyses

Blanks and plant standard material (GBW-07603, China Standard Materials Research Center, Beijing, China) were used for quality control. Recoveries from the reference materials ranged from 84.1% to 92.3% for Cd analysis. Origin 8.0 and Excel 2007 softwares were used to create the artwork. The data were analyzed using the SPSS 19.0 statistical package, and the data are presented as the Mean ± SE. The treatment means were compared via the Duncan multiple comparison test.

ACKNOWLEDGEMENTS

This study was supported by the National Key Research and Development Program of China (Grant No. 2018YFD0800700), the National Natural Science Foundation of China (Grant No. 31670409), the General Research Fund Proposal of Hong Kong, China (Grant No. RG21/2020-2021R), and the Faculty of Liberal Arts and Social Sciences of the Education University of Hong Kong, China (Grant No. 04548 (IRS-10)).

SUPPLEMENTAL DATA

The following materials are available in the online version of this article at http://www.sciencedirect.com/journal/rice-science; http://www.ricescience.org.

Fig. S1. Cd tolerance index of 22 rice cultivars grown in solution with treatment of 0.1 mg/L Cd for 3 weeks.

Fig. S2. Correlation between Cd concentration in shoots and roots of 22 rice cultivars with treatment of 0.1 mg/L Cd for 3 weeks.

Fig. S3. Cd concentration in roots and shoots of two rice cultivars with 0.5 and 2.5 mg/L Cd treatments for 4 weeks.

Table S1. Growing parameters of 22 rice cultivars grown in solution with treatments of 0 (control) and 0.1 mg/L Cd for 3 weeks.

Table S2. Effects of Cd stress on growth of two rice cultivars with treatments of 0.5 and 2.5 mg/L Cd for 4 weeks.

Table S3. Kinetic parameters for Cd influx into rice roots of cultivars ZXN and JYT36.

Table S4. Rice cultivars used.

Armstrong J, Armstrong W. 2005. Rice: Sulfide-induced barriers to root radial oxygen loss, Fe2+and water uptake, and lateral root emergence., 96(4): 625?638.

Boominathan R, Doran P M. 2003. Cadmium tolerance and antioxidative defenses in hairy roots of the cadmium hyperaccumulator,., 83(2): 158?167.

Chi Y, Li F, Tam N F Y, Liu C, Ouyang Y, Qi X, Li W C, Ye Z H. 2018. Variations in grain cadmium and arsenic concentrations and screening for stable low-accumulating rice cultivars from multi-environment trials., 643(1): 1314?1324.

Chiao W T, Syu C H, Chen B C, Juang K W. 2019. Cadmium in rice grains from a field trial in relation to model parameters of Cd-toxicity and -absorption in rice seedlings., 169: 837?847.

Chiao W T, Chen B C, Syu C H, Juang K W. 2020. Aspects of cultivar variation in physiological traits related to Cd distribution in rice plants with a short-term stress., 61: 27.

Grant C A, Clarke J M, Duguid S, Chaney R L. 2008. Selection and breeding of plant cultivars to minimize cadmium accumulation., 390: 301?310.

Hoagland D R, Arnon D I. 1938. The water culture method for growing plants without soil., 347: 1?39.

Huang G, Ding C, Guo F, Li X G, Zhou Z G, Zhang T L, Wang X X. 2017. The role of node restriction on cadmium accumulation in the brown rice of 12 Chinese rice (L.) cultivars., 65(47): 10157?10164.

Huang L, Li W C, Tam N F, Ye Z H. 2019. Effects of root morphology and anatomy on cadmium uptake and translocation in rice (L.)., 75: 296?306.

Kotula L, Ranathunge K, Schreiber L, Steudle E. 2009. Functional and chemical comparison of apoplastic barriers to radial oxygen loss in roots of rice (L.) grown in aerated or deoxygenated solution., 60(7): 2155?2167.

Kreszies T, Schreiber L, Ranathunge K. 2018. Suberized transport barriers in, barley and rice roots: From the model plant to crop species., 227: 75?83.

Liu J G, Liang J S, Li K Q, Zhang Z J, Yu B Y, Lu X L, Yang J C, Zhu Q S. 2003. Correlations between cadmium and mineral nutrients in absorption and accumulation in various genotypes of rice under cadmium stress., 52(9): 1467?1473.

Liu J G, Qian M, Cai G L, Yang J C, Zhu Q S. 2007. Uptake and translocation of Cd in different rice cultivars and the relation with Cd accumulation in rice grain., 143(1/2): 443?447.

Loix C, Huybrechts M, Vangronsveld J, Gielen M, Keunen E, Cuypers A. 2017. Reciprocal interactions between cadmium- induced cell wall responses and oxidative stress in plants., 8: 1867.

Lux A, Sottniková A, Opatrná J, Greger M. 2004. Differences in structure of adventitious roots inclones with contrasting characteristics of cadmium accumulation and sensitivity.,120(4): 537?545.

Lux A, Morita S, Abe J, Ito K. 2005. An improved method for clearing and staining free-hand sections and whole-mount samples., 96(6): 989–996.

Lux A, Martinka M, Vaculik M, White P J. 2011. Root responses to cadmium in the rhizosphere: A review., 62(1): 21?37.

Ma F, Peterson C A. 2003. Current insights into the development, structure, and chemistry of the endodermis and exodermis of roots., 81: 405?421.

Maksimovi? I, Kastori R, Krsti? L, Lukovi? J. 2007. Steady presence of cadmium and nickel affects root anatomy, accumulation and distribution of essential ions in maize seedlings., 51(3): 589–592.

Man Y, Zhao Y Y, Ye R, Lin J X, Jing Y P. 2018.cytological and chemical analysis of Casparian strips using stimulated Raman scattering microscopy., 220: 136?144.

Moore C A, Bowen H C, Scrase-Field S, Knight M R, White P J. 2002. The deposition of suberin lamellae determines the magnitude of cytosolic Ca2+elevations in root endodermal cells subjected to cooling.,30(4): 457–466.

Nordberg G, Jin T Y, Bernard A, Fierens S, Buchet J P, Ye T T, Kong Q H, Wang H F. 2002. Low bone density and renal dysfunction following environmental cadmium exposure in China., 31(6): 478?481.

Parrotta L, Guerriero G, Sergeant K, Cai G, Hausman J F. 2015. Target or barrier? The cell wall of early- and later-diverging plants vs cadmium toxicity: Differences in the response mechanisms., 6: 133.

Qi X L, Tam N F, Li W C, Ye Z H. 2020. The role of root apoplastic barriers in cadmium translocation and accumulation in cultivars of rice (L.) with different Cd- accumulating characteristics., 264: 114736.

Redjala T, Zelko I, Sterckeman T, Legué V, Lux A. 2011. Relationship between root structure and root cadmium uptake in maize., 71(2): 241?248.

Rodda M S, Li G, Reid R J. 2011. The timing of grain Cd accumulation in rice plants: The relative importance of remobilisation within the plant and root Cd uptake post- flowering., 347: 105–114.

Rout G R, Das P. 2002. Rapid hydroponic screening for molybdenum tolerance in rice through morphological and biochemical analysis., 48: 505?512.

Schreiber L, Hartmann K, Skrabs M, Zeier J. 1999. Apoplastic barriers in roots: Chemical composition of endodermal and hypodermal cell walls., 50: 1267–1280.

Shi Y J, Xu Y F, Ni Z Y, Wang J W, Li D, Zhang M K. 2019. Difference of Cd accumulation in main crops in Hangzhou and its influencing factors., 60(7): 1230?1233. (in Chinese with English abstract)

Tao Q, Jupa R, Luo, J P, Lux A, Kovac J, Wen Y, Zhou Y M, Jan J, Liang Y C, Li T Q. 2017. The apoplasmic pathway via the root apex and lateral roots contributes to Cd hyperaccumulation in the hyperaccumulator., 68(3): 739?751.

Uraguchi S, Mori S, Kuramata M, Kawasaki A, Arao T, Ishikawa S. 2009. Root-to-shoot Cd translocation via the xylem is the major process determining shoot and grain cadmium accumulation in rice., 60(9): 2677?2688.

Uraguchi S, Fujiwara T. 2012. Cadmium transport and tolerance in rice: Perspectives for reducing grain cadmium accumulation., 5(1): 5.

Uraguchi S, Fujiwara T. 2013. Rice breaks ground for cadmium- free cereals., 16(3): 328–334.

Wang F J, Zhang Y T, Guo Q X, Tan H F, Han J H, Lin H R, Wei H W, Xu G W, Zhu C. 2018. Effects of exogenous 5-aminolevulinic acid and 24-epibrassinolide on Cd accumulation in rice from Cd-contaminated soil., 25(6): 320?329.

White P J. 2001. The pathways of calcium movement to the xylem., 52: 891–899.

Wilkins D A. 1987. The measurement of tolerance to edaphic factors by means of root growth., 80: 623–633.

Williams P N, Lei M, Sun G, Huang Q, Lu Y, Deacon C, Meharg A A, Zhu Y G. 2009. Occurrence and partitioning of cadmium, arsenic and lead in mine impacted paddy rice: Hunan, China.,43(3): 637?642.

Wu C, Ye Z H, Shu W S, Zhu Y G, Wong, M H. 2011. Arsenic accumulation and speciation in rice are affected by root aeration and variation of genotypes.,62(8): 2889–2898.

Yamaguchi N, Mori S, Baba K, Kaburagi-Yada S, Arao T, Kitajima N, Hokura A, Terada Y. 2011. Cadmium distribution in the root tissues of solanaceous plants with contrasting root-to-shoot Cd translocation efficiencies., 71(2): 198?206.

Ye J, Yan C L, Liu J C, Lu H L, Liu T, Song Z F. 2012. Effects of silicon on the distribution of cadmium compartmentation in root tips of(S., L.) Yong., 162: 369–373.

Zhuang P, Lu H, Li Z, Zou B, McBride M B. 2014. Multiple exposure and effects assessment of heavy metals in the population near mining area in South China., 9(4): e94484.

Ye Zhihong (lssyzhh@mail.sysu.edu.cn); Li Wai Chin (waichin@eduhk.hk)

14 November2020;

1 March2021

Copyright ? 2021, China National Rice Research Institute. Hosting by Elsevier B V

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Peer review under responsibility of China National Rice Research Institute

http://dx.doi.org/10.1016/j.rsci.2021.03.003

(Managing Editor: Wu Yawen)