1 INTRODUCTION

Lemna minor L. (duckweed) is a free-floating macrophyte species extensively distributed from freshwater to brackish estuaries. It can grow well in eutrophic and polluted waters rich in heavy metals and organic compounds, and tolerates stresses caused by pollutants. This species has been used as a potential scavenger in the reclamation of waste water contaminated by heavy metals, herbicides pesticides and other pollutants (Arthur et al., 2005). It is a convenient experimental model system for investigating defensive strategies against a variety of toxic substances, as well as biotic and abiotic stresses, including heavy metals, salt stress, chilling, UV radiation and pathogen attack (Oron, 1994; Yilmaz, 2007) and for monitoring toxic pollutants in water areas by assessing traditional biomarkers, such as malondialdehyde (MDA), superoxide dismutase (SOD), and catalase (CAT) (Gasdaska et al., 2003; Cox et al., 2006). Cadmium, as a non-essential element, is a highly toxic and widespread heavy metal pollutant due to its high solubility, absorption and accumulation in many organisms, including plants (Buchet et al., 1990; Romero-Puertas et al., 2004; Lux et al., 2010; Gill et al., 2011). Recently, the level of heavy metal pollutants in soil, aquatic and wetland ecosystems has been gaining great attention globally (Rai et al., 2008). A range of studies related to cadmium pollution have been carried out in a variety of plant species.

The accumulation of cadmium in plant tissues can interfere with biochemical and physiological processes and induce a serious of secondary oxidative stresses, including membrane lipid peroxidation, protein oxidation, enzyme inhibition and DNA and RNA damage, and can eventually result in cell damage and cell death (Gratão et al., 2005; Foyer and Noctor, 2005; López et al., 2006). In response to heavy metal stresses and the secondary oxidative stresses, plants have developed effective antioxidant defense systems comprised of both enzymatic and non-enzymatic components. The primary enzymatic antioxidant scavengers of reactive oxygen species (ROS) include ascorbate peroxidase (APX), SOD, glutathione S-transferase (GST), CAT and peroxidase (POD). Ascorbic acid (AA) or glutathione (GSH) are the major non-enzymatic components in the response to the secondary oxidative stresses (Gratão et al., 2005). High levels of enzymatic and non-enzymatic antioxidants in plant tissues usually reflect high stress tolerance and resistant capacities to oxidative damage (Shalata and Tal, 1998; Bor et al., 2003). A consistent and close correlation between biotic/abiotic stresses and the enzymatic/non-enzymatic responses to oxidative stresses has been shown in many plant, microalgae and animal species, such as Mesembryanthemum crystallinum, Arabidopsis thaliana, Helianthus annuus, Chlorella vulgaris, Mollusc species and L . minor (Prasad et al., 2001; Broetto et al., 2002; Scragg and Bonnett, 2002; Di Baccio et al., 2004). Therefore, both enzymatic and non-enzymatic components are widely used as biomarkers in assessing the toxicity of environmental pollutants, such as heavy metals and herbicides, and in indicating and monitoring environment pollution (Sharma and Dietz, 2009).

In L . minor, the responses to xenobiotics and the bioaccumulation of heavy metals in aquatic and wetland ecosystems have been studied using a number of approaches that directly measured both enzymatic and non-enzymatic components as biomarkers. These works showed that several stresses can induce changes, in varying degrees, in enzymatic antioxidant activity levels (Cummins et al., 1999; Hatton et al., 1999; Tombuloglu et al., 2012).

A native-PAGE analysis of the multiple isoenzyme activities of SOD, CAT and APX showed specific isozyme patterns in response to different stresses (Tkalec et al., 2008; Han, 2013), suggesting that while the response of specific isoenzyme is closely related to a specific stress, these responses cannot be revealed through ordinary enzymatic assays. Therefore, these enzymatic and non-enzymatic biomarkers have the obvious disadvantages of low specificity and sometimes low sensitivity in indicating a specific stress or pollutant (Nakamori et al., 2010). GSTs are abundant and diverse dimeric enzymes that exist extensively in bacteria, fungi, animals and plant species. The GST proteins share as little as 10% amino acid identity and their roles, especially in endogenous plant metabolism, are unclear. Genetic and genomics approaches showed that there are 25 or more GST-coding genes in plants. Recently, the cloning and functional analysis of different GST gene family members in a number of plants, animals, microalgae and micro-organisms allowed us to gain a better functional understanding of these gene family members and the encoded enzymes (Lei et al., 2009). Some GST members had detoxifying enzyme activities and played important roles in the detoxification of xenobiotics and oxidative stress resistance in cells (Hayes and Mclellan, 1999). They were able to catalyze the nucleophilic attachment of GSH to molecules that present an electrophilic carbon, nitrogen or sulfur atoms, to reduce lipid hydroperoxides to their corresponding alcohols, to reduce free hydrogen peroxide to water (Bhabak and Mugesh, 2010) and/or conjugate GSH to the toxic reactive compounds, 4-hydroxynonenal and cholesterol-oxide, which are generated during membrane oxidation (Edwards et al., 2000). In plants, GSTs (Phi, Tau and Theta) are able to act as GSH peroxidases (GPXs) and are important in oxidativestress tolerance (Roxas et al., 1997; Cummins et al., 1999). The activities of these GSTs can be induced by a variety of intracellular xenobiotics, including pesticides, polychlorinated biphenyls, metals and polycyclic aromatic hydrocarbons (Edwards et al., 2000). Similar to many other enzymatic antioxidants, GST activity is also widely used as an indicator when monitoring xenobiotics and stress responses in various organisms.

At the molecular level, the responses of plants to biotic and abiotic stresses are closely associated with the expression levels of specific gene(s) triggered by specific environmental factors. This gene expression is generally considered to be a useful biomarker of biotic and abiotic stresses due to its high sensitivity and specificity to particular stress exposures (Dix et al., 2006; Ankley et al., 2007; van Straalen and Roelofs, 2008; Nakamori et al., 2010). The investigation into the molecular mechanisms of plants responding to cadmium stress allows us to gain a better understanding of its mode of action and to facilitate the exploitation of biomarkers to monitor cadmium pollution with a high degree of sensitivity and specificity. It may also assist in controlling, particularly bio-controlling, pollution.

In this work, a putative GST (phi) gene induced by cadmium was cloned from L . minor . Its expression profiles and enzymatic activities under cadmium stress were studied.

2 MATERIAL AND METHOD 2.1 Culture and preparation of plant materials

twice with 0.01 mol/L NaOCl for 20 s to prevent algal and aphid growth, and cultured axenically in 3-L plastic containers of Datko medium with sparging (Datko et al., 1980) at 25±1℃ under cool white fluorescent light (160 μmol/(m²∙s) photo-synthetically active radiation) with a 18 h/6 h (light/dark) cycle. The growth medium was replaced weekly. For each treatment, ~3 g (~100 colonies) of L . minor plants were exposed to 75 μmol/L (~15 mg/L) CdCl ₂ for 0 h (ck), 3 h, 6 h, 12 h, 24 h, 36 h and 48 h. After treatment, the plant samples were washed thoroughly with distilled water, and dried with absorbent paper. Samples of 250 mg each were briefly frozen in liquid nitrogen and stored at -80℃ until use. For each treatment, two samples were analyzed separately as biological replicates, each with three technical replicates.

2.2 Cloning of the GST gene in L . minor 2.2.1 Total RNA extraction and cDNA synthesis

Total RNA was extracted using Trizol (TianGen, Beijing, China) following the manufacturer’s instructions. The integrity of RNA was evaluated through electrophoresis. The 260/280 absorbance ratio was measured using Nanovue plus for RNA quantity and quality estimations. The cDNA was synthesized using PrimeScript l ^st strand cDNA Synthesis Kit (TaKaRa, Dalian, China) according to manufacturer’s instructions.

2.2.2 LmGST (phi) gene cloning

A fragment of the LmGST (phi) gene containing the initiation codon was first amplified using degenerate primers. Then, the 3′ end of the LmGST (phi) gene was cloned using 3′ RACE through three round of PCR amplifications.

To clone the 5′ end of the LmGST (phi) gene, the degenerate primers Lm GST -F1-degenerate and LmGST-R1-degenerate (Supplementary materials 1, Table 1) were designed based on the conserved sequence regions of GST (phi) in other plant species that were published in the NCBI EST databases. Primers were synthesized by BGI Tech Co. Ltd. (Beijing, China). All of the primers used for PCR amplification were dissolved in double-distilled water to a final concentration of 10 μmol. PCR products were ligated into the pMDl8-T vector (TaKaRa, Dalian, China), and the positive recombinant, LmGST - pMD18-T, was selected for sequencing.

To clone the full length LmGST (phi), the 3′ RACE forward primer (Supplementary materials 1, Table 1) was designed based on the sequence information of the newly cloned 5′-end fragment of LmGST (phi). A BD SMART RACE cDNA amplification kit (TaKaRa Biomedical Technology (Beijing) Co., Ltd.) was used following the manufacturer’s instructions. For the 1 ^st round amplification, cDNA obtained through reverse transcription was used as the template. LmGST -3′ RACE-F110 and Oligo(dT) 25-AP were used as upstream and downstream primers, respectively. KOD Taq polymerase (Baori Biotechnology Co. Ltd.) was used with the PCR conditions: 94℃ for 5 min; 30 cycles of 94℃ for 45s, 55℃ for 45s and 68℃ for 90 s, followed by a final extension at 68℃ for 10 min.

For the 2 nd round of amplification, the LmGST -3′ RACE-F230 and AP primers were used, with the annealing temperature (Ta) raised to 56℃. The 1 ^st round PCR products were used as templates. The 3 rd round PCR reaction was performed at Ta 56℃, using the LmGST -3′ RACE-F513 and AP primers with the 2 nd round PCR products as templates.

After three rounds of PCR amplifications, the amplified products of LmGST (phi) 3′-end fragments were ligated into the vector pMDl8-T, and the positive recombinant of 3′-end- LmGST -pMDl8-T was selected for sequencing.

2.3 Quantitative analysis of GST (phi) gene expression

The LmGST -rt-F227 and LmGST -rt-R507 primers were designed based on the newly isolated LmGST (phi) sequence (Supplementary materials Table 1) and were used to reveal the expression profiles of LmGST (phi) with a Rotor - Gene Q instrument (Qiagen).

2.4 Enzyme extractions

Plant tissue (1 g) from control and treated L . minor were homogenized in a mortar with 5 mL of 100 mmol/L potassium phosphate buffer (pH 7.5) containing 1 mmol/L of EDTA in the presence of polyvinyl polypyrrolidone. Phenylmethyl sulphonyl fluoride was added to ensure enzyme stability. The homogenate was centrifuged at 15 000× g at 4℃ for 30 min (Hou et al., 2007). The supernatant was used to measure the activities of SOD, APX, POD and CAT. All steps in the preparation of the enzyme extracts were carried out at 0-4℃.

2.5 Activity assay of antioxidant enzymes

The enzymatic activities of SOD, APX, CAT and POD were measured spectrophotometrically. The activity of SOD (EC 1 . 15 . 1 . 11) was assayed by measuring the inhibition of the photochemical reduction of nitroblue trazolium according to Beauchamp and Fridovich (1971). The activity of APX (EC 1 . 11 . 1 . 11) was measured according to Nakano and Asada (1981), by measuring the decreased absorbance of ascorbate at 290 nm. The activity of CAT (EC 1 . 11 . 1 . 6) was determined by measuring spectrophotometrically the consumption of H ₂ O ₂ at 240 nm according to the method of Beers (Beers and Sizer, 1952). The activity of POD (EC 1 . 11 . 1 . 7) was measured as an increased absorbance at 470 nm, which occurs when guaiacol polymerizes to tetraguaiacol according to Putter (1974).

2.6 Determination of GST (EC 2 . 5 . 1 . 18)

The GST activity was defined as the amount in micromoles of GS-DNB conjugated /min under the conditions of the assay according to the method of Habig et al. (1974).

2.7 GSH assay

The GSH content was estimated by measuring the absorbance at 412 nm according to the method of Anderson (1985).

2.8 Assay of lipid peroxidation by MDA

MDA, a by-product of lipid peroxidation that reacts with thiobarbituric acid (Ortega-Villasante et al., 2005), was measured according to the method of Razinger et al. (2008).

2.9 Total soluble protein estimation

The total soluble protein content was measured according to Bradford (1976), using a commercial Bradford reagent (Sigma), and bovine serum albumin (Sigma) was used as the standard. Supernatants and dye were pipetted into spectrophotometer cuvettes (P General T6; 200-1 000 nm) and absorbances were measured at 595 nm.

2.10 Sequence analysis

The protein sequences of GST were aligned using ClustalX (1.83) software (Thompson et al., 1994) with the default settings. An un-rooted neighborjoining (NJ) tree was constructed using Mega version 4.1 with 1 000 bootstrap replicates based on the deduced amino acid sequence of Lm GST and other GST proteins.

2.11 Statistical analysis

The statistical analysis was carried out by a oneway analysis of variance (ANOVA) using SPSS version 13.0 (SPSS, Cary, NC, USA), and the individual comparisons were obtained using Duncan’s Multiple Range Test (DMRT). Significance values (^* P<0.5, ^** P<0.01) were determined using an ANOVA with Dunnett’s comparison (Duncan et al., 1957).

3 RESULT AND DISCUSSION 3.1 Effect of cadmium toxicity on lipid peroxidation activity

MDA is produced by the ROS decomposition of polyunsaturated fatty acids (PUFA) in cell membranes. The MDA content is an oxidative stress level indicator. In this experiment, the MDA content showed a significant linear increase with treatment time when exposed to 75 μmol cadmium (Fig. 1). This indicates that L . minor suffered severe oxidative stress over an extended time when exposed to 75 μmol cadmium, which is in accordance with previous reports (Pryor and Stanley, 1975; Moore and Roberts, 1998; Prasad et al., 2001).

3.2 Effect of cadmium toxicity on CAT, SOD, POD and APX activities

Heavy metal toxicity is usually the result of ROS generation, which causes constant enzymatic and non-enzymatic antioxidant responses. Cadmium stress also enhanced CAT, SOD, POD and APX activities in L . minor during the early stage of 75 μmol cadmium exposure in the present study. However, the prolonged cadmium treatment caused a reduction of CAT, SOD, POD and APX enzymatic activities (Figs.2-5), possibly by inactivating the existing enzymes, reducing enzyme synthesis or accelerating the enzymes’ degradation. The acceleration of enzyme degradation was supported indirectly by the total soluble protein assay shown in Fig. 6, which was in agreement with previous reports in wheat, Folsomia candida, sorghum and Lemna trisulca (Zhang and Kirkham, 1994; Prasad et al., 2001; Domínguez et al., 2010; Nakamori et al., 2010).

3.3 Effect of cadmium toxicity on the GSH content

GSH is the major intracellular low-molecular-weight thiol compound. It is an important nonenzymatic antioxidant in maintaining the cellular thiol-disulfide balance and in enhancing cellular defense against oxidative stress and the metabolism of xenobiotics (Jakoby and Ziegler, 1990). As an electron donor for sulfydryl- and thiol-containing enzymes, such as GSH reductase, GPX, GST and glutaredoxin, reduced GSH levels play a critical role in protecting the GSH-dependent enzymes from oxidization and in maintaining their enzyme activity. Therefore, the GSH level is closely associated with the potential GSH-dependent peroxidase activity and the organism’s oxidative stress tolerance induced by various abiotic and biotic stresses (Edwards et al., 2000). In the present study, the content of reduced GSH increased during the early stage of 75 μmol cadmium exposure (up to 24 h), and decreased in the later periods (Fig. 7), which is in accordance with GSH-dependent enzyme activities, such as POD and APX.

Our results showed that the activity of many tested antioxidants, such as APX, CAT, GST, SOD, POD and GSH, in L . minor increased in the early period of cadmium exposure and decreased with prolonged cadmium stress. In contrast, MDA contents substantially increased in response to cadmium expose. Similar to enzymatic and non-enzymatic antioxidant activities, the GST gene’s expression level also increased under cadmium exposure. These results indicate that the 75 μL cadmium exposure caused oxidative stress and damages (with an increased MDA content), and that antioxidants are important and effective defense mechanisms in response to cadmium induced oxidative stress and oxidative damages, which is in accordance with most previous reports (Cummins et al., 1999; Hatton et al., 1999; Dixon et al., 2001, 2010; Gong et al., 2005; Gratão et al., 2005; Huseyin et al., 2012).

3.4 Sequence analysis of the LmGST gene

A putative GST gene fragment was obtained from L . minor through RACE PCR. The Sanger sequencingconfirmed fragment was 866 bp in length with a full open reading frame of 645 nucleotides and a 3′UTR region of 198 bp. The deduced amino acid sequence of 212 amino acids contained a GST N-terminal domain and a GST C-terminal alpha-helical domain, which are shared by the GST protein family. A BLASTp result using the deduced GST protein as the query sequence revealed an 88% identity to the GST(phi) protein in Cynodon dactylon and a 62% identity to the GST(phi) protein in Alopecurus myosuroides .

A multiple sequence alignment analysis (Fig. 8) showed that the deduced GST protein sequence shared a high similarity with the GST(phi) protein sequences to the 11 selected plant species. The neighbor-joining phylogenetic tree (Fig. 9) showed that the deduced protein sequence in L . minor was closely related to the GST(phi) proteins in the 11 selected plant species since they formed a single cluster, while the 6 GST(theta) proteins clustered in a separate group.

The above results suggest that the cloned sequence is a putative GST gene, subgroup phi, in L . minor, designated as LmGST (phi). The nucleotide and deduced amino sequences of LmGST (phi) are shown in Supplementary materials 6 and 7. The sequence data reported here have been deposited in the GenBank database, accession number KM247621.

3.5 The effect of cadmium toxicity on GST activity and LmGST gene expression

Our experimental results showed that GST activity in L . minor is influenced and induced by cadmium exposure, which showed a similar pattern to those of the other tested enzymatic antioxidants, such as SOD and CAT. This indicates that GST is involved in the cadmium stress response (Fig. 10).

GSTs are abundant proteins presented in all eukaryotic species, including bacteria, fungi and plants. They collectively account for 1% of the total soluble protein in maize leaves (Marrs et al., 1995). These dimeric enzymes are associated with the diversity of physiological and metabolic processes, and are encoded by a multigene family, GST . Different GST forms represent different substrate-specific isomers. Little is known about the cellular locations and distinct functions of each GST in responses to specific stress signals (David et al., 1999). Research at the gene level may provide more information. AmGSTF1, an herbicide-induced GST member in blackgrass (Alopecurus myosuroides), is a highly active GPX that appears to contribute to herbicide resistance by preventing the accumulation of cytotoxic hydroperoxides (Cummins et al., 1999). Roxas et al. (1997, 2000) reported that the over-expression of t-GST (M107) enhanced the activities of GST and GPX in tobacco. Milligan’s works (Milligan et al., 2001) showed that ZmGSTIV in maize enhanced the herbicide tolerance of wheat. Similar results have been reported in various animal, algae, microorganism and plant species.

In this work, the quantitative expression results of the GST (phi) gene in L . minor in response to cadmium stress (Fig. 11) showed that LmGST (phi) was strongly induced by cadmium stress. Compared with the untreated L . minor, the expression level increased substantially and reached its peak shortly after cadmium exposure. Additionally, shortly after cadmium exposure, 6 h, the increase of LmGST (phi) expression level was much more remarkable than the change in GST enzymatic activity. This suggests that the LmGST (phi) gene expression is highly sensitive to cadmium exposure. However, its specificity and sensitivity to other pollutants were not tested in the current work and remain to be studied further.

4 CONCLUSION

In this work, we cloned the GST (phi) gene in L . minor and investigated its characteristics, expression and enzymatic activity under high cadmium concentrations in comparison with other scavengers of ROS, including APX, CAT, SOD and POD.

When exposed to cadmium stress, a number of enzymatic antioxidants in duckweeds, including SOD, CAT, APX and GST, are involved in the enhanced resistance to heavy metal stresses and the secondary oxidative stress. The activities of these enzymatic antioxidants increased substantially during the initial phase of cadmium treatment, while a prolonged cadmium treatment reduced the enzymatic activities of CAT, SOD, POD and APX.

Collectively, the results described in this work indicate that the activities of enzymatic and nonenzymatic components, and the expression of the LmGST (phi) gene, are closely associated with cadmium stress and the resulting secondary oxidative stress, and could be used as biomarkers for assessing cadmium pollution more sensitively and rapidly than the tested enzymatic antioxidants. Understanding the molecular mechanisms of plants responding to various stresses/pollutants at the gene level could be helpful for controlling and monitoring heavy metal pollution more efficiently, and for facilitating the exploitation of biomarkers with high sensitivities and specificities to specific pollutants.

5 ACKNOWLEDGEMENT

We thank Professor SONG Jiancheng for his contribution in editing the manuscript.