Different Growth and Physiological Responses to Cadmium of the Three Miscanthus Species

Miscanthus has been proposed as a promising crop for phytoremediation due to its high biomass yield and remarkable adaptability to different environments. However, little is known about the resistance of Miscanthus spp. to cadmium (Cd). To determine any differences in resistance of Miscanthus to Cd, we examined plant growth, net photosynthetic rate (Pn), activities of anti-oxidant and C4 photosynthetic enzymes, concentrations of Cd in leaves and roots, and observed the chloroplast structure in three Miscanthus species treated with 0, 10, 50, 100 or 200 μM Cd in solutions. Miscanthus sinensis showed more sensitivity to Cd, including sharp decreases in growth, Pn, PEPC activity and damage to chloroplast structure, and the highest H2O2 and Cd concentrations in leaves and roots after Cd treatments. Miscanthus sacchariflorus showed higher resistance to Cd and better growth, had the highest Pn and phosphoenolpyruvate carboxylase (PEPC) activities and integrative chloroplast structure and the lowest hydrogen peroxide (H2O2) and leaf and root Cd concentrations. The results could play an important role in understanding the mechanisms of Cd tolerance in plants and in application of phytoremediation.


Introduction
Soil cadmium (Cd) pollution has posed a serious threat to our soil quality and food security as well as to human health. The sources of Cd contamination is not only introduced through geogenic processes but also derive from anthropogenic activities, such as the by-product of smelting, mining and refining of metal works [1], industrial waste from electroplating, manufacturing of plastics and paint pigments processes [2] and agriculture pollutions including impurities of fertilizers and irrigation with wastewater [3,4]. The toxicant of Cd is higher than that of organic toxic compounds due to its greater mobility and harder degraded and thus resulting in difficult to remove from the environment [5].
Cd is not a necessary element for plant growth and excess Cd has a series of harmful effects. Cd is known to inhibit plant growth, disorder nutrient uptake, affect chloroplast ultrastructure, 0.5; ZnSO 4 Á7H 2 O, 1.0; MnCl 2 , 1.25; H 3 BO 3 , 7.5; (NH 4 ) 6 Mo 7 O 24 , 0.25 and NaFeEDTA 50. To ensure proper growth, the solutions were aerated and renewed weekly. Following 32 days of hydroponic growth, seedlings were subjected to aerated nutrient solution including 0, 10, 50, 100 or 200 μM CdCl 2 . Each treatment was replicated six times and each replicate included eight seedlings. The solutions were renewed every week. The entire experiment was conducted in an environmentally controlled growth room with a 14 h/26°C day (white fluorescent light intensity of 1200 μmol photons m −2 s −1 ) and 10 h/22°C night regime with relative humidity kept at 65%.

Growth analysis and Cd contents measurement
Growth such as plant height, root length, aerial part and hypogeal-part dry weight were measured after 16 d of treatment. Plant height and the length of the below ground part (root length) were measured on a centimeter scale. Dry weight was determined after drying the samples in an oven at 80°C to a constant weight. The root:shoot ratio was computed as the hypogeal part divided by the aerial part on a dry weight basis.
For Cd content measurement, roots and shoots were separately harvested, and the roots were washed with deionized water for three times. Then shoots and roots were dried at 80°C for 72 h, weighed, ground to fine powder and 0.2 g of each was digested with nitric acid/H 2 O 2 (30:1, v/v) and total Cd content was measured by inductively coupled plasma atomic emission spectrometer (ICP-AES; Fisons ARL Accuris, Ecublens, Switzerland).

Determination of photosynthetic and chlorophyll fluorescence parameters
Photosynthetic parameters of leaves were measured with a Li-6400 portable photosynthesis system (LI-COR, Lincoln, NE, USA). These parameters consisted of net photosynthetic rate (Pn), stomata conductance (g s ) and intercellular CO 2 concentration (Ci). The data were recorded at 16 d of treatment using the most fully expanded youngest leaves. The light intensities were maintained at 2000 μmol m -2 s -1 , and the temperature and external CO 2 concentration were maintained at 30°C and 400 μmol L -1 , respectively. Five representative plants of each treatment were selected randomly at each measured time-point. For light response curves measurements, a series of light intensities were set as 2500, 2000, 1500, 1200, 800, 600, 400, 300, 200, 100, 50, 30, 10, 0 μmol m -2 s -1 PPFD at an ambient CO 2 concentration 400 μmol mol -1 with the LI-COR CO 2 mixer. Minimum time and maximum time were respectively set to 1min and 2 min for each given PPFD. Before the measurement, each leaf was adapted at a PPFD of 2500 μmol m -2 s -1 for about 5 min until the stability state of Pn. According to the modified rectangular hyperbola model [22], light compensation point (LCP), the maximum photosynthetic rate (Pn MAX ), apparent quantum yield (AQY) and dark respiratory rate (DR) were calculated as: P(I) = αÁIÁ(1-βÁI)/(1+γÁI)-R d . Where P(I) is Pn, I is light intensity, R d is dark respiratory rate, and α, β and γ are coefficients which are independent of I. Once Pn was obtained, the leaf tissue was freeze-clamped quickly at liquid N 2 temperature and stored at -80°C for chlorophyll, malondialdehyde (MDA), hydrogen peroxide (H 2 O 2 ) contents and enzyme activity analysis.
The chlorophyll fluorescence parameters were measured with an chlorophyll fluorescence imaging system (CF imager, Technologica Ltd., Colchester, UK) according to the method of Liu et al. [23] with minor modification. The first fully grown leaves of Miscanthus seedlings treated with different concentrations of Cd were dark-adapted for 20 min with leaf clips, then the leaves were cut off and arranged neatly underneath the fluorometer for recording the minimum fluorescence (F 0 ) and maximum fluorescence (Fm) parameters and getting the false-color images of maximal photochemical efficiency (Fv/Fm) images. The Fv/Fm was calculated as (Fm-F 0 )/Fm. Then leaves were light-adapted for approximately 15 min prior to measurement of the effective PSII quantum yield [Y(II)] which was calculated as Y(II) = (Fm'-F)/Fm', where Fm' and F were fluorescence at maximum fluorescence and steady-state photosynthesis in the light, respectively.

Determination of photosynthetic pigment contents
Photosynthetic pigments were extracted by soaking 0.1 g of frozen leaf tissues in 80% (v/v) acetone in darkness at room temperature for 45 h. Chlorophyll and carotenoid contents in supernatants were determined with a spectrophotometer (UV-2550, Shimadzu, Kyoto, Japan) at 665, 649 and 470 nm, and calculated using the method of Lichtenthaler and Wellburn [24].

Determination of C 4 photosynthetic enzyme activities
The phosphoenolpyruvate carboxylase (PEPC), NADP-malate enzyme (NADP-ME) and NADP-malate dehydrogenase (NADP-MDH) activities of leaves were determined using a commercial chemical assay kit (Jiangsu Keming Biotechnology Institute, Suzhou, China). For the measurement of PEPC and NAD-MDH activities, about 0.1 g of frozen leaf tissues were homogenized in 1 ml buffer I [0.4 M Tris-HCl buffer (pH 8.0), 15 mM EDTA, 10 mM DTT, 5 mM MgCl 2 and 2% (w/v) polyethylene pyrrole (PVP)], which is contained in the commercial chemical assay kit, at 4°C with an ice-chilled pestle and mortar, centrifuged at 10,000 rpm at 4°C for 10 min and the supernatant was used for the enzymes activity analysis according to the manufacturer's instructions. For the measurement of NADP-ME activity, about 0.1 g of frozen leaf tissues were extracted using 1 ml buffer I [0.1 mM KH 2 PO 4 /KOH buffer (pH 7.5), 10 mM DTT, 5 mM MgCl 2 and 2% (w/v) polyethylene pyrrole (PVP)] according to the above-mentioned method, then analyzed according to the manufacturer's instructions.

Chloroplast ultrastructure
The chloroplast ultrastructure of bundle sheath cells were observed according to Shao et al. [25]. After 16 days treatment, the fully expanded youngest leaves were immediately fixed in 2.5% (v/v) glutaraldehyde (0.1 mol L -1 phosphate buffer, pH 7.2) for 24 h. Then the samples were post-fixed for 30 min in 1% (v/v) osmium acid, dehydrated in a graded ethanol series (30%-100%, v/v), embedded in Spurr resin and ultrathin-sectioned for transmission electron microscopy (H7650, Hitachi, Tokyo, Japan).

Determination of MDA, H 2 O 2 contents and anti-oxidant enzymes activities
For the determination of SOD, CAT and POD activities, about 0.5 g of frozen leaf tissues were ground at 4°C in a mortar with 5 ml of 50 mM phosphate buffer solution (pH 7.8) containing 1% PVP. The homogenate was centrifuged at 10,000 rpm at 4°C for 30 min. Supernatants were collected for measuring enzyme activities according to Hong et al. [26]. The MDA, H 2 O 2 contents, GR and APX activities, were determined using a commercial chemical assay kit (Jiangsu Keming Biotechnology Institute, Suzhou, China). For the measurement of MDA content and GR and APX activities, about 0.1 g of frozen leaf tissues were homogenized in 1 ml buffer I [50 mM phosphate buffer (pH 7.8), containing 0.1 mM EDTA, 0.5% (w/v) Triton-100 and 2% PVP], which is contained in the commercial chemical assay kit, at 4°C with mortar and pestles, centrifuged at 10,000 rpm at 4°C for 10 min and the supernatant was used for content or enzyme ability analysis according to the manufacturer's instructions. For the measurement of H 2 O 2 contents, the extraction buffer I was replaced by 1 ml acetone according to the abovementioned method, then analyzed according to the manufacturer's instructions.

Statistical analysis
Statistical analysis was carried out by one-way or two-way analysis of variance using SPSS (SPSS Inc., USA, version 13.0) and OriginPro (OriginLab Corp., USA, v8.0724). Differences between treatments were evaluated at P < 0.05.

Plant growth and Cd accumulation
The sensitivities of Miscanthus spp. to Cd varies, and roots were more sensitive than shoots (Fig 1 and Table 1). The growth parameters such as root length, dry weight of the hypogeal part and the entire plant dry weight were significantly inhibited (p<0.05) by all Cd concentrations and plant height was significantly decreased (p<0.05) by ! 50 μM Cd concentrations in Miscanthus sinensis (Table 1). For M. floridulus there were no significant differences (p<0.05) in root length between 10 μM Cd treatment and control (0 μM Cd), and no significant differences (p<0.05) in plant height, hypogeal part and entire plant between 10 and 50 μM Cd; however, there was a significant difference (p<0.05) in dry weight of aerial parts between all Cd treatments. For M. sacchariflorus, however, in comparison with control 10 μM Cd treatment slightly promoted plant growth according to all growth indexes (Table 1) and 50 μM Cd treatment significantly increased (p<0.05) in dry weight of the hypogeal part and the entire plant, and in root:shoot ratio. Therefore, M. sacchariflorus was more resistant to Cd than the other Miscanthus spp.
The Cd contents of roots and leaves of Miscanthus spp. were extremely different, although they significantly increased (p<0.05) with increasing Cd concentration (Fig 2A and 2B). In roots, M. sinensis exhibited the highest Cd concentration, followed by M. floridulus and then M. sacchariflorus (Fig 2A).     Fig 3D) and AQY dramatically decreased (p<0.05) in all species with increased Cd concentrations (Fig 3E), but the degree of increase in LCP and decrease in AQY was least in M. sacchariflorus (Fig 3D and 3E). Moreover, DR gradually increased and reached a maximum for 50-100 μM Cd treatments in all Miscanthus spp. (Fig 3F).

Photosynthetic pigment contents
Chlorophyll and carotenoid contents were significantly decreased (p<0.05) by increased Cd concentrations for all Miscanthus spp., and the reductions were always lower in M. sacchariflorus than for the other species (Fig 4).  (Fig 4A and 4B).

C4 photosynthetic enzymes activities and Chloroplast structure
PEPC activity differed markedly between the Miscanthus spp., although it significantly decreased (p<0.05) under Cd stress. The highest activity was in M. sacchariflorus, whereas the lowest in M. sinensis (Fig 5A). The inhibitory effect of Cd on PEPC activity was more evident for M. sinensis and M. floridulus than for M. sacchariflorus. The PEPC activities were decreased   Fig 5A). NADP-ME activity also decreased significantly (p<0.05) for all Miscanthus spp. under Cd stress with slightly higher activity for M. floridulus than the other species for all Cd concentrations (Fig 5B). The greatest reduction in NADP-ME activity was for 100 and 200 μM Cd treatments in M. sacchariflorus, with inhibition ratios of 68.1% and 81.6%, respectively ( Fig 5B).  Fig 5C). The structural changes in chloroplasts markedly differed between Miscanthus spp. under Cd stress (Fig 6 and S2 Fig). With zero Cd treatment, chloroplasts in all species showed welldeveloped structures with normal granal and stromal thylakoids and some small osmiophilic globules (Fig 6A-6C). Treatment ! 10 μM Cd dramatically increased production of starch grains and enlarged osmiophilic globules in M. sinensis (Fig 6D, 6G and 6J); 100 μM Cd caused accumulation of small starch grains and enlargement of osmiophilic globules in M. floridulus (Fig 6K and 6N); but in M. sacchariflorus, only 200 μM Cd resulted in accumulation of small starch grains and enlargement osmiophilic globules (Fig 6L and 6O). The chloroplast envelope  became indistinct in M. sinensis treated with ! 50 μM Cd (Fig 6G, 6J and 6M) and in M. floridulus treated with ! 100 μM Cd. Higher concentrations of exogenous Cd caused the granal and stromal lamellae of chloroplasts to condense and a loss of connection between both lamellae in M sinensis (Fig 6G, 6J and 6M) and M. floridulus (Fig 6N). The chloroplast structure in M. sacchariflorus did not change significantly for all Cd concentrations.

Activities of anti-oxidant enzymes
All anti-oxidant enzymes including SOD, CAT, POD, APX and GR had markedly increased activity in leaves of Miscanthus spp. treated with Cd (Fig 8A-8E (Fig 8D). GR activity was lower in M. sacchariflorus than in M. sinensis and M. floridulus with increasing Cd concentrations, except for 200 μM Cd treatment where GR activity increased by 2.4, 1.2 and 1.4 times, respectively, compared with their controls (Fig 8E).

Different response of growth and Pn in Miscanthus spp. to exogenous Cd concentrations
Cd is a trace pollutant that is toxic to plants, animals and humans. In the present study, all Cd levels negatively influenced plant growth of M. sinensis and M. floridulus, causing significant reductions (p<0.05) in plant growth and dry biomass, while, there was a slight increase in growth of M. sacchariflorus at < 50 μM Cd and lower degree of reduction compared to M. sinensis and M. floridulus at 200 μM Cd, suggesting that M. sacchariflorus had greater Cd stress tolerance than M. sinensis and M. floridulus. It has also been reported that plant genotypes differ in their tolerance to Cd toxicity [6]. At high Cd concentrations, the leaves of M. sinensis and M. floridulus became yellow and roots became soft and brown, while the leaves remained green and the roots white for M. sacchariflorus, even at 200 μM Cd (Fig 1), thus further confirming greater Cd tolerance of M. sacchariflorus among the three species. Arduini et al. [15] find that, even for long-term, low Cd (0.5 mg L -1 ) application stimulates Miscanthus growth. Gill et al. [27] reported that at 25 mg kg -1 soil Cd, co-ordination of S and N metabolism can still complement to the antioxidant mechanism to protect the growth and photosynthesis of Lepidium sativum plants. However, high Cd doses (50-100 μM) cause growth inhibition and even plant death owing to inhibiting photosynthesis, respiration, water and nutrient uptake [28,29].
The inhibitory effect of Cd on Pn, g s and chlorophyll content was more evident in M. sinensis and M. floridulus than M. sacchariflorus. The growth inhibition may be a consequence of Cd interference with the main metabolic processes such as photosynthesis and translocation of photosynthetic products and nutrient elements [30]. In the present study, the decrease in whole plant dry weight was in accordance with the decrease of Pn (Fig 3A and Table 1), suggesting that Pn played an important role in biomass accumulation during Cd stress. The Cdinduced reduction in Pn and AQY (Fig 3A and 3E) could be partially due to the decrease in g s and chlorophyll content of the Miscanthus species (Figs 3B and 4A), as reported for maize (Zea mays L.) [31,32] and sugarcane (Saccharum officinarum L.) [33]. The mechanism of photosynthetic response involves both stomatal and non-stomatal effects under environmental stress in C 4 crops [34][35][36]. The results showed that the decrease in Pn was accompanied by increasing Ci concentration in M. sinensis and M. floridulus, suggesting that the factor limiting photosynthesis was mainly non-stomatal under Cd stress [34][35][36][37]. However, such changes were absent in M. sacchariflorus, implying different mechanisms for Pn depression due to Cd in different Miscanthus spp.
Difference in Cd accumulation and transfer is relative to resistance among Miscanthus spp.
Gill et al. [6] reported that the uptake and transport of Cd differed with plant species and genotypes. Cd accumulation in leaves directly leads to damage to the photosynthetic apparatus and decreases in Pn [28,29]. The different concentrations of Cd in roots and leaves of Miscanthus spp., even for the same concentration of exogenous Cd treatment (Fig 2A and 2B) reflects the difference in absorption by roots and transport from roots to shoot, and explains the difference in resistance of Miscanthus spp. to Cd. The highest Cd concentration in roots (Fig 2A) and medium Cd concentration in leaves (Fig 2B) indicated restricted transport and more absorption for M. sinensis. The highest Cd concentrationin leaves ( Fig 2B) and medium Cd concentration in roots (Fig 2A) suggested stronger transport and absorption for M. floridulus. The lowest Cd concentrations both in leaves and in roots confirmed much less absorption of exogenous Cd for M. sacchariflorus (Fig 2) and this low absorption is not only a characteristic but could be the main cause of the higher resistance of M. sacchariflorus to Cd.
The decrease in Pn was due to lower activities of C 4 photosynthetic enzymes and damage to chloroplast structure Exogenous Cd treatment resulted in depression of Pn and AQY of all species, and the depression was much greater in M. sinensis and M. floridulus (Fig 3A and 3E). To determine the reason for this depression in Miscanthus spp. under different Cd concentrations, we determined the activities of key enzymes of the C 4 photosynthetic pathway-PEPC, NADP-ME and NADP-MDH-that participate in the process of concentrating CO 2 in C 4 photosynthesis [38]. We found significant decreases (p<0.05) in PEPC, NADP-ME and NADP-MDH activity in all Miscanthus spp. exposed to Cd stress. However, PEPC activity was much higher in M. sacchariflorus than in the other two species for all Cd concentrations (Fig 5A). This was consistent with the highest Pn and higher Cd tolerance in M. sacchariflorus. The Pn of Miscanthus spp. were closely related to PEPC content rather than ribulose 1,5 bisphosphate carboxylase/oxygenase (Rubisco) under higher nitrogen content [36]. Moreover, in maize leaves it was found to be inactivated by Cd [39]. The NADP-ME is a key enzyme in the NADP-ME subtype of C 4 plants and helps enrich the CO 2 for Rubisco, thus lowering photorespiration and improving photosynthetic efficiency [40]. NADP-MDH is particularly abundant in C 4 plants, where it functions photosynthetically in the NADP-dependent reduction of oxaloacetate to malate [41]. It was also reported that higher MDH activity and malate accumulation in companion with higher Pn were found in the drought-resistant Sorghum bicolor genotype compared with a sensitive genotype [42]. In the present study, the higher activity of NADP-MDH in M. sacchariflorus under Cd stress favored conversion of oxaloacetate to malate (Fig 5C) which is then transported into adjacent bundle sheath cells to enhance Calvin cycle in bundle sheath cells [43]. The increase in malate synthesis can cause a significant increase in root malate exudation, thus improving toxic metal resistance in C 3 plants [44,45], but it is still unclear in C4 plant whether or not the malate, except for C 3 CO 2 fixation in leaf, can be transported from leaf to root. If a part of malate resulted from the higher activity of NADP-MDH in M. sacchariflorus under Cd stress can be transported out of leaf and reaches to root, it is possible to confer high Cd tolerance in this plants.
Chloroplast ultrastructure could provide important information concerning the biochemical properties of the thylakoids, which suffer the greatest changes during adverse environmental conditions, such as salt [46], drought [47] or heavy metal [48] stresses. The decrease of Pn is related to changes in the membrane structure of chloroplasts [49] and degradation of chloroplasts [48]. The production of starch grains, enlargement of osmiophilic globules and loss of the chloroplast envelope showed large differences among the Miscanthus species under various Cd concentrations (Fig 6D, 6G and 6J-6O). Cd caused the grana and stroma lamellae of chloroplasts to condense and the loss of connection between both lamellae in M. sinensis (Fig 6G,  6J and 6M) and M. floridulus (Fig 6N) but did not induce significant change in M. sacchariflorus. These results not only indicated the different resistance of Miscanthus spp. to Cd, but also confirmed that the different decreases in Pn, photosynthetic pigment, the maximal photo- Stronger anti-oxidant system may alleviate the damage to photosynthetic apparatus Cd exposure initially results in severe oxidative stress, which in turn caused lipid peroxidation and H 2 O 2 accumulation [9]. MDA is a product of lipid peroxidation and is considered an indicator of oxidative damage [50]. The present study showed that MDA accumulation increased most in M. sinensis under all Cd stress (Fig 7A), indicating that Cd induced stronger peroxidation and caused more serious damage to the cell membrane in M. sinensis. A certain amount of H 2 O 2 accumulation during Cd stress may act as an oxidative agent and a local or systemic signal that activates various anti-oxidant enzymes, but over-accumulation of H 2 O 2 induces peroxidative reactions that damage plant cells [51]. is restricted by the ascorbate-glutathione (ASH-GSH) cycle, where APX uses ASH as a hydrogen donor and GR catalyzes the NADPH-dependent reduction of oxidized glutathione (GSSG) to reduced GSH [52]. The activities of these enzymes were increased by Cd stress in wheat and tobacco [53,54]. In the present study, M. sinensis showed a greater increase in SOD activity (Fig 8A), but lesser increase in POD and APX activities (Fig 8C and 8D), resulting in greater H 2 O 2 accumulation ( Fig 7B) and the most serious damage to chloroplasts (Fig 6D, 6G, 6J and 6M), compared with the other Miscanthus spp. In M. sacchariflorus, for all concentrations of Cd treatment, the SOD, POD and APX activities were higher (Fig 8A, 8C and 8D) and the H 2 O 2 accumulations much lower than that in the other Miscanthus spp. (Fig 7B). These results indicated that M. sacchariflorus possessed a better anti-oxidative system, which could scavenge ROS and maintain integrity of the chloroplast (Fig 6F, 6I, 6L and 6O). The results support the view that genotypic difference in the anti-oxidative system could partially account for the genotypic difference in Cd accumulation, tolerance and an increase in tolerance to Cd stress is positively correlated with anti-oxidant capacity [55].

Conclusions
The present study revealed the effects of Cd on plant growth, photosynthesis characteristics, chloroplast ultrastructure, Cd-uptake and translocation and physiological responses of three Miscanthus species. The results showed that the effects of different Cd concentrations on growth and Pn in Miscanthus spp. differed. The inhibitory effect of Cd on growth characteristics was more evident for M. sinensis whereas, least for M. sacchariflorus. The resistance of M. sacchariflorus to Cd was mainly due to a lower Cd absorption and translocation, thus keeping more effective activities of C 4 photosynthetic enzymes and better chloroplast structure. Furthermore, hyperactivity of anti-oxidant enzymes also played an important role in protecting M. sacchariflorus from Cd toxicity.