Effects of amendments on heavy metal immobilization and uptake by Rhizoma chuanxiong on copper and cadmium contaminated soil

An improved method was applied for remediating cadmium and copper co-contaminated soil and reducing the metal concentration in Rhizoma chuanxiong. Pot experiments were conducted with six amendments (composed with bentonite, phosphate, humic acid, biochar, sepiolite powder, etc.). The results showed that soil pH, biological activities (soil enzymatic activities and microbial counts) and R. chuanxiong biomass were greatly improved with the addition of amendments in all treatments, especially in T3 and T6. Also, amendments effectively decreased the concentration of malondialdehyde and H2O2 in R. chuanxiong. In the T3 treatment, the bio-available Cd and Cu in soil were significantly decreased by 0.53 and 0.41 mg kg−1, respectively. Meanwhile, the amendment in T3 reduced Cd and Cu accumulation in R. chuanxiong about 45.83 and 39.37%, respectively, compared to T0. Moreover, the Fourier transform infrared spectroscopy spectra showed the surface functional groups of every amendment. To conclude, this study offers an effective and environmental method to reduce metal accumulation in R. chuanxiong on heavy metal co-contaminated soil.


Introduction
With the development of industrialization, urbanization and construction, heavy metals have generated several serious environmental problems. Heavy metals such as aluminium (Al), copper (Cu), lead (Pb), cadmium (Cd) and chromium (Cr) are easy to transfer into soil and water, which will pose extreme & 2018 The Authors. Published by the Royal Society under the terms of the Creative (table 2). Two plants of R. chuanxiong (bought from Sichuan Academy of Agricultural Sciences) were planted in each pot, then cultivated from August to November with an average temperature of 258C indoors.
Approximately three months later, R. chuanxiong were harvested successively and washed with deionized water, then wet weights were recorded. Rhizoma chuanxiong were dried at 608C for 3 days in an oven.

pH of soils
One gram of air-dried soil was weighed in a 5 ml breaker, and added to 2.5 ml of deionized water without CO 2 [24]. The solution was stood for 30 min and measured by pH meter (SevenCompact-s210).

Soil biological activities
Soil samples were collected from each pot after R. chuanxiong was harvested to estimate soil enzyme activities and bacteria counts. The measurement of dehydrogenase was through a spectrophotometer as described by Zhou et al. [25] with minor modifications by prolonging the reaction time to 48 h. Activities of fluorescein diacetate (FDA) hydrolysis, urease, acid phosphatase and invertase were assayed by the methods of Adam & Duncan [26], Yan [27], Alef & Nannipieri [28] and Gu [29], respectively. Dehydrogenase activity was determined spectrophotometrically at 492 nm and expressed as microgram triphenylformazan (TPF) per soil per hour. FDA activity was determined spectrophotometrically at 490 nm and presented as the content of fluorecien in dry soil (mg g 21 ). The activity of urease was spectrophotometrically determined at 578 nm and presented as mg NH 4 þ g 21 soil. The activity of invertase was determined spectrophotometrically at 508 nm and presented as milligram glucose per gram soil per 24 h. Bacterial counts in soil were counted on LB agar medium through the spread-plate method as described by Liu et al. [30].

Heavy metals analysis
Heavy metals in R. chuanxiong and soil samples was detected by atomic absorption spectroscopy (VARIAN, SpecterAA-220Fs) as described by Wu et al. [31]. Dried powdered and sieved samples of 0.2 g soil and plant were digested by HCLO 4 (5 : 4 : The soil samples were collected to measure metal bioavailability by a toxicity characteristic leaching procedure (TCLP) and 0.01 M CaCl 2 extraction [32,33]. Two grams of air-dried sieved (2 mm) soil were added to 40 ml of 0.11 mol l 21 HAc at pH 2.88 + 0.05 and shaken for 18 h. Then the solution was filtered with a 0.45 mm filter membrane for Cd and Cu analysis. At room temperature (about 258C), 2.5 g of airdried soil samples were incubated for 3 h in a shaker at 250 rpm with 12.5 ml of 0.1 mol l 21 CaCl 2 . Then the solution was centrifuged at 3000 rpm for 10 min and filtered with a 0.45 mm filter membrane. All the extraction was determined by flame atomic absorption spectrometry.

Malondialdehyde determination
According to the method of Liu et al. [30] with some modifications, R. chuanxiong tissues (1 g) were ground with 2 ml of 10% trichloroacetic acid (TCA) and quartz sand, then homogenized with 8 ml of 10% TCA. The absorbance was determined at 450, 532 and 600 nm. MDA was quantified with equation (

Fourier transform infrared spectra amendment assay
The Fourier transform infrared spectroscopy (FTIR) spectra of amendments prepared as KBr discs were recorded in a Perkin-Elmer Spectrum 100 Model Infrared Spectrophotometer to examine functional groups of amendments [34]. FTIR spectra were recorded in the range of 400-4000 cm 21 at a resolution of 2 cm 21 .

Data analysis
The mean and standard deviation of the three replicates were calculated in this study. Statistical significance was performed through using one-way ANOVA in SPSS 17.0, and the mean values were compared using the least significant difference calculated at a significant level of p , 0.05. All figures were performed by using ORIGIN 8.5 software. Also, SPSS version 21.0 for Win was used to reflect the multivariate linear relationship among contaminated soil physicochemical characteristic, microbial abundance and metal fractions.

Soil pH change
The pH values of soil with different treatments at the end of the experiment are shown in figure 1. The soil pH increased compared to the initial pH ( pH ¼ 7.93), except for T3 ( pH ¼ 7.77) and T6 (pH ¼ 7.87), but there were no significant differences ( p . 0.05) among T0, T3 and T6. T2 presented the highest increase in soil pH. Bentonite which was the same component in all amendments had a positive effect on soil pH [35], therefore causing the slight increase in soil pH in this experiment.

Analysis of metal availability
TCLP is always used for heavy metal leached fraction concept [33] and CaCl 2 extraction can remove metals by ion exchange with Ca 2þ and/or complexation with the chloride species [32]. In this experiment, bio-available metals were extracted by TCLP and CaCl 2 extraction methods. It could be observed that the available metals (Cu, Cd) decreased compared with T0, especially for Cd extraction by CaCl 2 (figure 2 as exceptional data. In the T3 treatment, the immobilization effect on Cu was also excellent with the decrease rate of 78.67% in CaCl 2 -extracted Cu. The order of CaCl 2 extractable Cu from soils was T6 , T3 , T1 , T5 , T4 , T2, while the order of TCLP extracting Cu was T6 , T3 , T5 , T1 , T2 , T4. Above all, with the least content of available heavy metals, T6 showed the best immobilized effect on heavy metals. Many studies have documented a negative correlation between soil pH and heavy metal availability [36]. The increase in pH in this experiment can benefit precipitation and sorption of heavy metals. Besides, some compositions of amendments, such as bentonite [37], phosphate [38], had strong adsorption capacity towards heavy metals, consequently reducing the toxicity of heavy metals.  T0 T1 T2  T4 T5 T6treatments   T3  T0 T1 T2  T4 T5

Microbial counts and respiration intensity of soil
The microbial counts including bacteria and fungi counts and soil respiration are shown in figure 3a. It could be observed that bacterial counts had a slight increase in all treatments compared to T0 (log6.12 CFU g 21 soil). Bacterial counts in T3 (log6.89 CFU g 21 soil) increased most, followed by T4 (log 6.72 CFU g 21 soil) compared to T0. Also, the highest and lowest fungi number were presented in T3 (log 5.50 CFU g 21 soil) and T0 (log5.11 CFU g 21 soil), respectively. On the whole, the number of microbes increased after the immobilization process, indicating that the addition of amendments contributed to soil micro-environment improvement.
The respiration intensity of soil is shown in figure 3b. It was clear that the respiration intensity in T6 (0.45 CO 2 mg g 21 dried soil 24 h 21 ) was around three times that of T0 (0.14 CO 2 mg g 21 dried soil 24 h 21 ); meanwhile, the respiration intensity of T3 (0.28 CO 2 mg g 21 dried soil 24 h 21 ) and T4 (0.22 CO 2 mg g 21 dried soil 24 h 21 ) were both almost two times more than in T0. Besides, the respiration intensity in other treatments also increased more or less.
The available heavy metals decreased, which reduced the environmental stress to microbes. Thus, the microbial counts and respiration intensity increased. Soil bacterial and fungal population were significantly and positively correlated with soil organic matter [39]. Amendments in this experiment provided amount of organic matter that could be used by microbes, which might be a reason for the microbial counts increase. For example, biological matrix (mainly straw powder), one of the same components of amendments in T3 and T4, could motivate soil microbial activity owing to the richness of N, P, K and trace elements [40]. Besides, other components also have a positive effect on microbial activity. For example, biochar in T1, T2, T5 and T6 could improve soil aeration through enhancing oxygen (O 2 ) diffusion, thereby stimulating aerobic activity in soil [41].

Soil enzymatic activities
The activity of dehydrogenase in soil is shown in figure 4a. Dehydrogenase activity showed different degrees of increase in all treatments compared to T0 (262.46 mg TPF g 21 ). It indicated that a light fluctuation happened in T2 (326.69 mg TPF g 21 ) and T4 (278.65 mg TPF g 21 ). Also, T3 and T6 presented the largest activities of dehydrogenase, about 472.86 and 461.90 mg TPF g 21 , respectively.
For urease activity, there was no significant difference ( p , 0.05) among all treatments with comparison to T0. But higher activity presented in T3, T4 and T6 (all above 300 mg NH 4 -N g 21 ) compared to T2 ( just about 190 mg NH 4 -N g 21 ) (figure 4a).
Invertase activity is by far less sensitive to heavy metals than dehydrogenase and urease [42]. Yet, it was observed that a significant change arose in this experiment (figure 4b). The highest activity of invertase presented in T3 and T6, in specifically, about 108 and 112 mg glucose g 21 , respectively, both of which were four times than that of T0 (27 mg glucose g 21 ). Compared to T0, invertase activity increased to approximately 50 glucose g 21 in T1, and nearly 70 mg glucose g 21 in T4. The activities of invertase in T2 and T5 increased slightly, compared to the control. The activity of FDA presented increased trends with the addition of amendments compared to T0 (figure 4b). More concretely, higher FDA activity presented in T3, T4 and T6 treatments, which were around two times greater than the activity in T0. Besides, the FDA activity showed a slight increase in T1, T2 and T5 in comparison to the control.
The increase in soil enzymatic activities cooperated with the increase in soil microbes [39,43]. With the reduction in heavy metal toxicity, microbial activity increased as a result of enzymatic activity increasing. Besides, the addition of amendments improved soil physicochemical properties, such as high cationic exchange capacity of bentonite [44], and phosphate increasing extractable soil inorganic P [45]. Above all, it was obvious that amendments in T3, T4 and T6 showed a more positive effect on enzymatic activity.

Biomass of Rhizoma chuanxiong
To evaluate the effect of amendments on the growth of R. chuanxiong in metal-contaminated soil, biomass of R. chuanxiong was recorded (figure 5). Compared to T0, the growth of R. chuanxiong was promoted in every treatment. Moreover, the biomass of R. chuanxiong in T3 was five times that of T0, and the growth in T1 was the least among treatments. These results indicated amendments played a positive role in the growth of R. chuanxiong in polluted soil.
The amendment provided nutrition for plants and adjusted soil pH, which could be propitious to R. chuanxiong growth. Meanwhile, plant biomass increase, in turn, can dilute the account of heavy  T6  T5  T4  T3  T2  T1  T0  T6  T5  T4  T3  treatments   T2  T1  T0  T6  T5  T4  T3  T2  T1  T0   metals with a 'dilution effect' [46]. The results in this experiment indicated that application of amendments could promote R. chuanxiong growth.

Heavy metal concentrations in Rhizoma chuanxiong
The concentrations of Cd and Cu in R. chuanxiong are presented in figure 6. On the whole, metals declined in R. chuanxiong with the addition of amendments compared to T0. Especially in the T3 treatment, the contents of Cd and Cu showed the remarkable declining rates by 45.83% and 39.37%, respectively. It was interesting that metal uptake by plants had a different tendency as enzymatic activity and microbial biomass. The difference between T3 and T6 on metal availability in soil and metal concentration in plants might indicate that the amendment in T3 had a more positive influence on reduction in metal bioavailability in soil and uptake by plants. Besides, although the amendment of T6 could immobilize metals, it had little effect on preventing plants uptake in this study. The difference of metal availability in soil and plant between T3 and T6 could be explained as follows. It could be observed that the bentonite in T3 and sepiolite powder in T6 had the same proportion of amendments, about 40%. The sepiolite powder had greater specific surface area and greater ion exchange capacity, while montmorillonite as the main component of the bentonite had low crystallinity values [47], which made the sepiolite powder have a stronger adsorption capacity to heavy metals. Thus, the available heavy metals in T6 were less than those in T3. However, the sepiolite powder was rich in Mg 2þ and Ca 2þ [48] whose concentrations in soil had a positive relation to the accumulation of heavy metal in plants. Therefore, the concentrations of Cd and Cu in R. chuanxiong in T3 were lower than those in T6. The phenomena about the reduction in heavy metal concentrations in R. chuanxiong might be explained as follows: (i) the reduction in available heavy metals made it difficult for metals to enter the plant; (ii) amendments used in this experiment were full of organic matter which played an important role in retaining soil Cd and in decreasing its availability to plants through cation exchange capacity (CEC) [49]. As Marchand et al. researched [50], soil organic matter is pivotal for controlling metal (e.g. Fe, Cu and Pb) bioavailability from strong organometallic complexes. For example, in the research of Sun et al. [51], bentonite can reduce metal exchangeable fraction, in alkaline soils, resulting the decrease in metal bioavailability; (iii) some research indicated that phosphate could reduce metal dissolution and transport from contaminated soil by increasing the efficiency of metal-phosphate mineral formation, then precipitating [52]; and (iv) the better effect on decreasing metal availability in 0  T5  T4  T3 treatments Cd concentration (mg kg -1 ) Cu concentration (mg kg -1   )   T2  T1  T0  T6  T5  T4  T3  T2  T1  T0  T6   1

Antioxidant stress of Rhizoma chuanxiong
H 2 O 2 is one of the ROS molecules which is formed upon incomplete reduction in oxygen [53]. The overproduction of H 2 O 2 may lead to cell death and oxidative damage [54]. MDA is produced from cell membrane peroxidation [55]. The metals can increase the concentration of H 2 O 2 and MDA in plants [56]. Thus, the production of H 2 O 2 and MDA can indicate metal stress on plants [57]. Environmental stress, such as salinity, temperature extremes and high metal contamination, could decrease the activities of superoxide dismutase (SOD) and increase the activities of catalase (CAT) and peroxidase (POD) [58 -60]

The Fourier transform infrared spectra amendments
The results of FTIR are shown in figure 8. The absorption band that appeared at 3620 cm 21 might detect the inner hydroxyl groups [61], and it was not observed in T6. Besides, the absorption of 2856 cm 21 are ascribed to stretching vibration of methylene (2CH 2 ) groups in antisymmetric and asymmetric [62], which are mainly obvious in T1, T3 and T6. The sorption bands at 1499 and 1465 cm 21 expressed the C-H bend from CH 2 and C-O bend from carboxylate ions, and the P ¼ O from phosphate, C-O-P and P-O-P were absorbed at 1241, 1197 and 1148 cm 21 [63][64][65]. The bands below 500 cm 21 represent the C a ¼ C a torsion, C-OH 3 torsion of the methoxy group and ring torsion of phenyl [66].  T2  T1  T0  T6  T5  T4  T3  T2  T1  T0  T6   1

Conclusion
In the present study, the addition of amendments reduced the bioavailability of heavy metals and concentration in R. chuanxiong. Also, the amendments slightly improved soil pH, and soil biological activities (enzymatic activities, microbial counts). In all kinds of treatments, T3 and T6 showed a better effect on reducing metal availability in soils, which indicated that the recipes of amendment in T3 and T6 are more suitable to immobilizing Cd and Cu, compared to other treatments; whereas, amendment of T1 and T3 appeared to work better to reduce metal concentration in R. chuanxiong. In summary, results from this experiment declared that the amendment in T3 was the best treatment to reduce soil metal toxicity and concentration in R. chuanxiong. This experiment provided an effective and environmentally friendly pathway to remediate heavy metal-contaminated soil in situ.
Data accessibility. This article does not contain any additional data.