Study on the effects of polymer modifiers and phloem girdling on cotton in cadmium-contaminated soil in Xinjiang Province, China

The effects of two liquid modifiers (polyacrylate compound modifier and organic polymer compound modifier) and phloem girdling (stem girdling and branch girdling) on cadmium (Cd) content, Cd transport, and photosynthetic parameters of cotton (Xinluzao 60) in Cd-contaminated soil (40 mg kg −1) were studied through barrel experiment. The results showed that the distribution ratios of Cd in stem, leaves, and bolls, leaf net photosynthetic rate (Pn), leaf stomatal conductance (Gs), leaf transpiration rate (Tr), and chlorophyll content were decreased after girdling; and the application of modifiers reduced the Cd content and the Cd transported to the shoot, while alleviating photosynthetic damage caused by girdling. In general, our results indicated that the inhibition of carbohydrate supply caused by girdling reduced the photosynthetic capacity of cotton, while the applications of the two liquid modifiers decrease the influence to cotton photosynthesis. Moreover, Cd and modifiers may be transported to the shoot through both phloem and xylem.

Effects of girdling and modifiers on cadmium transport. The distribution ratio and transport index of Cd in different organs of cotton under different treatments were shown in Table 1. Compared with CK, the distribution ratios of Cd in root and bolls for T1 group decreased by 1.55% and 3.97%, respectively, whereas those in stem and leaves increased by 7.68% (P < 0.05) and 2.45%, respectively; the distribution ratio of Cd in root for T2 group decreased by 8.91%, whereas those in stem, leaves, and bolls increased by 18.95% (P < 0.05),   Table 1. Distribution ratios and transport indexes of Cd in cotton organs. Means in each row or column followed by the same lowercase letters are not significantly different (P < 0.05) by Duncan's multiple range test. Data are means ± SD (n = 3).
14.07% (P < 0.05), and 4.14%, respectively. That is, the application of modifiers reduced the Cd content in root but increased the Cd content in stem, leaves, and bolls. After stem girdling, the distribution ratio of Cd in root for CK-J group increased by 8.43%, whereas those in stem, leaves, and bolls decreased by 31.83% (P < 0.05), 6.85%, and 3.91%, respectively, compared with CK. Compared with CK-J group, the distribution ratios of Cd in root, stem, and bolls for T1-J group increased by 5.57%, 4.18%, and 2.16%, respectively, whereas that in leaves decreased by 20.33% (P < 0.05); the distribution ratios of Cd in root and stem for T2-J group increased by 4.52% and 9.82%, respectively, whereas those in leaves and bolls decreased by 16.93% (P < 0.05) and 2.14%, respectively. After branch girdling, the distribution ratio of Cd in root for CK-G group increased by 11.80% (P < 0.05), whereas those in stem, leaves, and bolls decreased by 28.29% (P < 0.05), 12.89% (P < 0.05), and 15.27% (P < 0.05), respectively, compared with CK. Compared with CK-G group, the distribution ratios of Cd in root, stem, and bolls for T1-G group increased by 3.98%, 1.06%, and 7.11%, respectively, whereas that in leaves decreased by 17.56% (P < 0.05); the distribution ratios of Cd in root, stem, and bolls for T2-G group increased by 2.78%, 14.39% (P < 0.05), and 5.69%, respectively, whereas that in leaves decreased by 17.80% (P < 0.05). As shown in Table 1, compared with CK, the transport indexes of Cd from root to stem for T1 and T2 groups increased by 9.38% and 30.59% (P < 0.05), respectively, whereas those from stem to leaves for T1 and T2 groups decreased by 4.86% and 4.10%, respectively; in addition, the transport indexes of Cd from stem to bolls decreased by 10.82% (P < 0.05) and 12.45% (P < 0.05), respectively. Compared with CK-J group, the transport indexes of Cd from root to stem, from stem to leaves, and from stem to bolls for T1-J group decreased by 1.32%, 23.52% (P < 0.05), and 1.93%, respectively, and those from root to stem and from stem to bolls for T2-J group decreased by 24.36% (P < 0.05) and 10.89% (P < 0.05), respectively. Compared with CK-G group, the transport indexes of Cd from root to stem and from stem to leaves for T1-G group decreased by 2.81% and 18.42% (P < 0.05), respectively, whereas that from stem to bolls increased by 5.98%; the transport index of Cd from root to stem for T2-G group increased by 11.29%, whereas those from stem to leaves and from stem to bolls decreased by 28.14% (P < 0.05) and 7.61%, respectively.

Effect of girdling and modifiers on photosynthesis parameters of cotton leaves.
For the groups without girdling, the application of modifiers increased the net photosynthesis rate (Pn) of cotton leaves (Fig. 3). Compared with CK, the Pn for T1 and T2 groups increased by 14.51% (P < 0.05) and 3.98%, respectively. The stem girdling and branch girdling decreased the leaf Pn. The Pn for CK-J and CK-G groups decreased by 4.30% and 23.02% (P < 0.05), respectively, compared with CK. The application of modifiers alleviated the decreases of Pn caused by girdling. Compared with CK-J group, the Pn for T1-J and T2-J groups increased by 1.19% and 13.39% (P < 0.05), respectively. Compared with CK-G group, the Pn for T1-G and T2-G groups increased by 12.14% (P < 0.05) and 22.34% (P < 0.05), respectively.
Modifiers application and girdling treatments significantly affected the stomatal conductance (Gs), intercellular CO 2 concentration (Ci), and transpiration rate (Tr) of cotton leaves. For the groups without girdling, the application of modifiers increased the Gs and Tr but decreased the Ci (Fig. 3). After stem girdling, the Gs and Tr for CK-J group decreased by 26.86% (P < 0.05) and 20.02% (P < 0.05), respectively, whereas the Ci increased by 3.43% compared with CK; the Gs and Tr for CK-G group decreased by 34.21% (P < 0.05) and 40.96% (P < 0.05), respectively, whereas the Ci increased by 16.73% (P < 0.05) compared with CK. The application of modifiers also significantly changed the Gs, Ci and Tr of cotton leaves. Among them, compared with CK-J group, the Gs for T1-J and T2-J groups increased by 25.03% (P < 0.05) and 42.03% (P < 0.05), respectively, and the Tr for T1-J and T2-J groups increased by 1.53% and 45.44% (P < 0.05), respectively; whereas the Ci for T1-J and T2-J groups decreased by 0.37% and 8.45%, respectively. Compared with CK-G group, the Gs for T1-G and T2-G groups increased by 6.71% and 8.38%, respectively, and the Tr for T1-G and T2-G groups increased by 27.13% (P < 0.05) and 34.90% (P < 0.05), respectively; whereas the Ci for T1-G and T2-G groups decreased by 15.80% (P < 0.05) and 11.20% (P < 0.05), respectively.
Effect of modifiers on microstructure of cotton leaf tissue. The observation of the paraffin sections of leaves in CK, T1, and T2 groups showed that leaves are composed of epidermis, palisade tissue, and sponge www.nature.com/scientificreports www.nature.com/scientificreports/ tissue (Fig. 4). For the control group (CK), leaf tissues were irregularly arranged; some cells contracted, shortened, and even curved, and the gap between cells increased. Compared with T1 and T2 groups, the number of sponge tissue cells for CK was decreased, the arrangement was loose, the intercellular spaces were increased, and the cells were decreased. For T1 and T2 groups, the palisade tissue cells of cotton leaves were long columnar; the cells were arranged neatly and tightly, and there were more chloroplasts compared with CK. The rounded sponge tissue cells were arranged closely.
Redundancy analysis of cotton physiological characteristics. The effects of modifiers on Cd accumulation and leaf photosynthetic parameters for different groups were analyzed by redundancy analysis (RDA). As shown in Fig. 5, PTI root-stem , DR stem , DR leaves , and DR bolls were clearly distributed on the left side of the RDA1 axis. PTI root-stem and DR stem were closely related to the Tr and Gs, whereas DR leaves was closely related to the Pn. STI stem-leaves , STI stem-bolls , and DR root were clearly distributed on the right side of the RDA1 axis. The Ci was closely related to DR root . and PTI root-stem and DR stem were closely related to T1 and T2 groups. After stem girdling, T1-J group was closely related to the Chla/b and Ci, and T2-J group was closely related to the Chla, Chlb, Car, Tr, and Gs. After branch girdling, T2-G group was closely related to the Ci.

Discussion
Most heavy metals such as Pb, Cr, and Cu are absorbed and stored in crop root, while Cd, Ni, and Zn, etc. are easily transported to the upper part of crops [16][17][18] . At present, many plants have been studied in terms of the Cd distribution and accumulation 8,18 , and the ability of xylem to transport Cd into shoots has been considered as a major determinant of Cd accumulation in the shoots of many plants 7 . In addition, the absorption of Cd by peanut and wheat also showed that Cd was mobile in phloem tissue 19 . According to the previous results, we studied the distribution and transport of Cd in cotton organs by girdling, and found that the mobile of Cd in different organs of cotton was obvious. For the control group (CK), the root had the highest Cd accumulation, followed by leaves, stem, and bolls (Fig. 1). From the distribution ratio, we found that stem girdling significantly reduced the accumulation of Cd in stem, and branch girdling not only significantly reduced the accumulation of Cd in stem, but also reduced that in leaves and bolls ( Table 2). It indicated that Cd was of high mobility in both xylem and  www.nature.com/scientificreports www.nature.com/scientificreports/ phloem, and stem may play an important role in transport of Cd for the two pathways. This is same as the research result by Reid et al. 19 : Cd was transported through xylem and phloem.
The accumulation and transport of Cd in crops have been regulated through the application of exogenous materials in previous studies. For example, Huang et al. 20 reported that the application of calcium can reduced the Cd accumulation in soybean and wheat roots. Guiwei et al. 21 found that the application of polyacrylate polymers reduced the bioavailable Cd, which could be employed to enhance productivity of crops in Cd-contaminated soils. In this study, the modifiers used are mainly polymer materials. The carboxyl and amide groups of polyacrylamides can form coordination bonds with metal ions on the surface of soil particles to form complexes 22 . From transport indexes, we found that the polyacrylate compound modifier significantly reduced the transport of Cd from stem to bolls, and the organic polymer compound modifier significantly reduced the transport of Cd from root to stem and from stem to bolls ( Table 2), indicating that polyacrylate compound modifier mainly reduced the mobility of Cd in cotton, and organic polymer compound modifier mainly reduced the absorption of Cd by cotton. Moreover, it was found that the application of polyacrylate compound modifier reduced the transport of Cd from stem to leaves after stem girdling and branch girdling, and the application of organic polymer compound modifier reduced the transport of Cd from stem to leaves after stem girdling and the transport of Cd from stem to bolls after branch girdling ( Table 2), indicating that modifiers could be transported to leaves and bolls after phloem girdling. According to RDA analysis, without girdling, the application of polyacrylate compound modifier could significantly change the distribution and transport of Cd in cotton organs compared with CK, whereas the application of organic polymer compound modifier could significantly change the distribution and transport of Cd in cotton organs compared with CK-J and CK-G groups after girdling.
Many studies have found that phloem girdling reduces the net photosynthetic rate (Pn) of plant leaves 23 . Sala et al. 24 showed that phloem girdling influenced the rate of photosynthetic assimilation, and Poirier-Pocovi et al. 25 showed that phloem girdling reduced the photosynthesis rate of leaves. Similar findings were also shown in this study. After girdling, the Pn, Gs, and Tr all decreased, and only Ci increased, indicating that the decrease of photosynthetic rate was mainly caused by non-stomatal factors 26 . This is consistent with the conclusion of Zhou et al. 27 . Without girdling, the applications of the two modifiers significantly decreased the inhibition from Cd on Pn of cotton leaves. For groups with girdling, only T2-J group significantly decreased the inhibition on Pn, but there was significant difference in the decreases of Pn among the groups treated with modifiers. The results indicated that the modifiers could alleviate the photosynthetic damage caused by girdling. After stem girdling, the Pn for the groups treated with organic polymer compound modifier were higher than that of CK (Fig. 3), which indicated that the application of organic polymer compound modifier after stem girdling had a positive effect on Pn. This might be due to the application of organic polymer compound modifier decreased intercellular CO 2 concentration, but increased stomatal conductance and transpiration rate (Fig. 3).
Consistent with the research results of Tang et al. 28 , the pigment chlorophyll of cotton leaves was significantly decreased by girdling, which helps to determine the compensatory photosynthesis by modifiers (Fig. 2). The decreases of Chla, Chlb, and Car concentrations for the groups treated with branch girdling was greater than those for the groups treated with stem girdling, and the changes of photosynthetic rates also showed the same phenomenon. It indicated that the damage caused by branch girdling was more serious. This may be due to the accumulation of carbohydrates in the upper part above the cut after girdling. Photosynthates could not be transported to the root, thus photosynthesis was inhibited. At this time, the plant might avoid absorbing too much light by reducing the content of photosynthetic pigments. This reaction is the feedback regulation according to the source-sink relation 25 . The application of modifiers improved the photosynthesis of leaves because the chlorophyll content was significantly increased, and the effect of the application of organic polymer compound modifier after stem girdling was more obvious. Moreover, Chla and Chlb play different roles in photosynthesis 29 . The ratios of Chla/b in cotton leaves decreased significantly after girdling, but they were increased after the application of modifiers, indicating that after girdling, the application of modifiers increased the number of molecules involved in photochemical reaction, which led to the improvement of photosynthesis 29 . According to the RDA analysis, the polyacrylate compound modifier has a greater effect on the protection of pigment chlorophyll of cotton without girdling than the organic polymer compound modifier, whereas the organic polymer compound modifier has a greater effect on that of cotton with girdling. In addition, Liu et al. 30 found that Cd distributed in leaf epidermis, spongy tissue, and palisade tissue of crops in Cd-contaminated soils. By observing the microstructure of leaf tissues, we found that Cd did not damage the epidermis cells but the palisade tissue cells in mesophyll and sponge tissue cells, which would affect chloroplasts. Because Cd was absorbed by cotton root and transported from root to shoots, there was no direct contact with epidermis cells. By applying the modifiers, the densities of palisade tissue and sponge tissue were adjusted, which promoted the development of mesophyll cells and chloroplasts and increased the chlorophyll content. Thus, the photosynthesis of cotton leaves was improved.

Conclusions
This study proved that Cd is of high mobility in both xylem and phloem through girdling. The transport pathways of polyacrylate compound modifier and organic polymer compound modifier in cotton were the same as that of Cd. Modifies could still be transported to the shoots after phloem girdling, but the effect of modifiers on alleviating Cd stress was not greater than that without girdling. Moreover, on the one hand, the application of the two modifiers could stabilize Cd in cotton root and reduce the absorption of Cd; on the other hand, it could alleviate the inhibition of photosynthesis caused by phloem girdling.
Test program. Two-year continuous remediation of Cd-contaminated soil using polymer modifiers was conducted in 2017 and 2018. On April 5th, 2017, to maintain the original soil profile, soils were packed into plastic barrels (length × width × height = 30 cm × 30 cm × 80 cm), and then barrels were buried back into the field. Based on previous studies 14 and pre-test results, cadmium chloride (CdCl 2 •2.5H 2 O) solution was added into the barrels and mixed fully with the plough layer. The Cd content in the plough layer reached approximately 40 mg kg −1 after three weeks. On April 26th, urea (345 kg hm −2 ) and compound fertilizer (17-17-17) (795 kg hm −2 ) was applied.
A total of nine groups were set in a randomized block design, and two treatments were employed, including: (1) the application of modifiers (non-modifier, polyacrylate compound modifier, and organic polymer compound modifier (Table 3)); and (2) phloem girdling (non-girdling, stem gridling, and branch girdling) ( Table 2). Each group had three repetitions. On April 30th, two modifiers (8.48 kg hm −2 ) were diluted with water and applied through drip irrigation. Cotton (Xinluzao 60) was sown on May 2th. After emergence, six seedlings were retained in each barrel. The first irrigation was conducted on June 14th. The irrigation cycle was 3 days during the growth period, and the irrigation volume was 4,500 m 3 hm −2 . No fertilizers and modifiers were applied at later stages. Cotton was harvested on September 5, 2017. After that, the stalks were pulled out and the field was plowed.
On April 5th, 2018, the Cd content in plough layer reached 32.44 mg kg −1 in the soil without modifiers, 31.57 mg kg −1 in the soil with polyacrylate compound modifier, and 31.22 mg kg −1 in the soil with organic polymer compound modifier. In 2018, no cadmium was applied into the soils, and the amounts of modifiers applied (8.48 kg hm −2 ) and the other managements were the same as those in 2017.
To evaluate the transport pathway of Cd (phloem or xylem), and analyze the effects of modifiers on Cd content and regulating photosynthesis in cotton, phloem girdling was employed to inhibit the transport of Cd and modifiers through phloem in 2018. Phloem girdling was conducted at the cotton flowering boll-setting stage (the 10th day after manual topping). Girdling sites of stem girdling and branch girdling were located at the main stem of cotton (3-5 cm from the ground) and the top fruit branch, respectively. The width of the cut was 1 cm, and the depth was up to the xylem. Ten days after girdling, functional leaves from each barrel were randomly selected to determine photosynthetic parameters, and three cotton plants were randomly selected from each barrel to determine Cd content.
Experimental determination. Measurement of Cd content. Root, stem, leaves, and cotton bolls of cotton plants were separated, and then they were dried at 105 °C for 0.5 h and dried to constant at 80 °C. Samples of each organ (0.5 g) were acid-digested using sulfuric and nitric acid (1:5, v/v) at 60 °C for 24 h, and treated with HNO 3 / HClO 4 (5/1, v/v). A Hitachi Z2000 graphite atomic absorption spectrophotometer (PinAAcle900T, PerkinElmer, USA) was used for the determination of Cd content 31 .  Table 3. Test design.
= DR(Distribution ratio) Cd uptake in root(stem, leaves)/total Cd uptake of the plant (1) = PTI(Primary transport index) Cd content in stem/Cd content in root (2) = STI(Secondary transport index) Cd content in cotton bolls(leaves)/Cd content in stem (3)