Activity and Transcriptional Responses of Hepatopancreatic Biotransformation and Antioxidant Enzymes in the Oriental River Prawn Macrobrachium nipponense Exposed to Microcystin-LR

Microcystins (MCs) are a major group of cyanotoxins with side effects in many organisms; thus, compounds in this group are recognized as potent stressors and health hazards in aquatic ecosystems. In order to assess the toxicity of MCs and detoxification mechanism of freshwater shrimp Macrobrachium nipponense, the full-length cDNAs of the glutathione S-transferase (gst) and catalase (cat) genes were isolated from the hepatopancreas. The transcription level and activity changes in the biotransformation enzyme (glutathione S-transferase (GST)) and antioxidant enzymes (superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GPx)) in the hepatopancreas of M. nipponense exposed to MC-LR (0.2, 1, 5, and 25 μg/L) for 12, 24, 72 and 96 h were analyzed. The results showed that the isolated full-length cDNAs of cat and gst genes from M. nipponense displayed a high similarity to other crustaceans, and their mRNAs were mainly expressed in the hepatopancreas. MC-LR caused significant increase of GST activity following 48–96 h (p < 0.05) and an increase in SOD activity especially in 24- and 48-h exposures. CAT activity was activated when exposed to MC-LR in 12-, 24- and 48-h exposures and then it was inhibited at 96-h exposure. There was no significant effect on GPx activity after the 12- and 24-h exposures, whereas it was significantly stimulated after the 72- and 96-h exposures (p < 0.05). The transcription was altered similarly to enzyme activity, but the transcriptional response was generally more immediate and had greater amplitude than enzymatic response, particularly for GST. All of the results suggested that MC-LR can induce antioxidative modulation variations in M. nipponense hepatopancreas in order to eliminate oxidative damage.

12-, 24-, 72-and 96-h exposure to assess the toxicity of MC-LR and the aquatic invertebrate species' defense mechanism.

Animals and Chemicals
One-year-old adult M. nipponense (weight, 4.4 ± 0.2 g) were bought from Huzhou fish farm (Zhejiang, China) and raised in 50 L glass aquaria with aerated water, a controlled temperature of 25 ± 2 °C, and the photoperiod was a 14 h:10 h light/dark cycle. Males were chosen to minimize possible interfering effects of sex. MC-LR (purity ≥ 98%) was purchased from Alexis Biochemicals (Lauseanne, Switzerland).

MC-LR Exposure
M. nipponense were exposed to MC-LR or a solvent control (0.05% methanol, v/v) in a 20 L glass tank for 12, 24, 48, 72 and 96 h (15 individuals per tank in triplicate and approximately 3.3 g shrimp/L). The MC-LR doses were 0.2, 1, 5, and 25 μg/L, according to environmental content of MCs. Half of the exposure solution was changed every day in order to guarantee the concentration of MC-LR and water quality. They were fed with commercial pellet food once a day. Water quality was regularly monitored and no differences were recorded between tanks during the entire experimental period. No crayfish died during the exposure period. M. nipponense in each group were immediately sacrificed following exposure, and tissues were dissected, frozen in liquid nitrogen, and maintained at −80 °C until use.

Sampling, RNA Isolation, and Reverse Transcription (RT)
M. nipponense hepatopancreas, muscle, stomach, intestine, and gill tissues were assayed for cat and gst mRNAs. Tissue samples for cDNA cloning and quantitative real-time polymerase chain reaction (qRT-PCR) analysis were homogenized in TRIZOL reagent (Invitrogen, Carlsbad, CA, USA), total RNA was extracted as described by Wang et al. [21], and treated with RNase-free DNase I (Fermentas, Burlington, ON, Canada) to remove genomic DNA contamination. Total RNA concentrations were calculated at an absorbance of 260 nm. Total RNA quality was verified on 1% agarose gels by visual inspection of the 18 S and 28 S ribosomal RNA bands and by the A260 nm/A280 nm ratio (range 1.90-2.05) measured with a Nanodrop spectrophotometer (Thermo Electron Corp., Waltham, MA, USA). The cDNAs were synthesized from 5 μg total RNA with M-MLV reverse transcriptase (Invitrogen) and the oligo (dT)18 primer in a 20 μL final volume.

Cloning of Full-Length Cat and Gst cDNAs from M. nipponense
Partial cDNA fragments for the two target genes in M. nipponense were amplified using primers designed from the conserved regions of their counterparts in other crustaceans (Table S1). PCR was performed with 2 μL of the cDNA template synthesized by RT-PCR in 20 μL containing 5 pmol of each specific primer and 1 unit of Taq DNA polymerase (TaKaRa Bio, Shiga, Japan). The purified PCR products of these cDNA fragments were inserted into the pMD18-T plasmid using a TA cloning kit (TaKaRa Bio) following the manufacturer's instructions. Three confirmed recombinant plasmid clones for each gene were sequenced separately by Genscript Corp. (Nanjing, China), using the ABI 3730 automated DNA sequencer (BigDye Terminator Chemistry) and the BigDye Terminator Cycle Sequencing Kit (PE Biosystems, Foster City, CA, USA). The nucleotide sequences were determined for both sense and antisense strands of the plasmid inserts. The 3′ and 5′ ends of both cDNAs were obtained with several pairs of gene-specific primers designed for the target genes to overlap the sense or antisense regions of their amplified partial fragments sequenced above. The 5′ and 3′ rapid amplification of cDNA ends (RACE) procedures were performed with the RACE cDNA Amplification Kit (Invitrogen), using total RNA and following the manufacturer's instructions, respectively. The nested 5′-and 3′-RACE PCR products of the expected size were processed and sequenced as described above. The cat and gst full-length cDNAs were assembled by aligning the partial cDNA fragments and the 5′-and 3′-RACE fragments using SeqMan in Lasergene software (DNASTAR, Madison, WI, USA).

Protein Alignment and Phylogenetic Analyses
The putative amino acid sequences of the M. nipponense cat and gst genes were aligned with their counterparts from other species using the Megalign program in Lasergene software. We aligned diverse invertebrate cat and gst genes at the amino acid level using the ClustalX (1.83) sequence alignment program to establish phylogenetic trees for the two M. nipponense genes [22]. The neighbor-joining algorithm method [23] in the Mega 4.0 program was used to construct the phylogenetic trees [24]. Bootstrap analyses were conducted using 1000 replicates.

qRT-PCR
The qRT-PCR analysis was performed on a CFX96 real-time PCR detection system (Bio-Rad, Hercules, CA, USA) thermocycler with the SYBR Premix ExTaq II kit (TaKaRa Bio). The qRT-PCR reactions were carried out in a final volume of 25 μL using 1 × SYBR Premix Ex Taq, 0.4 μM of each primer, and 2.5 μL RT reaction solution. The reactions were denatured at 95 °C for 30 s, followed by 40 cycles of denaturation at 95 °C for 5 s, and annealing at 60 °C for 30 s. To assess the specificity of each amplicon, a melting curve analysis was performed at the end of each PCR thermal profile. CFX Manager software (Bio-Rad) was used to analyze SYBR Green I density and to determine the threshold cycle (Ct) value. All samples were run in triplicate. qRT-PCR efficiency (E) was calculated from the given slopes using CFX Manager software and a 10-fold diluted cDNA sample series with five dilution points measured in triplicate. E was determined using the equation: E = 10 [25].

Tissue Distribution of the Cat and Gst Genes
qRT-PCR was conducted with male M. nipponense hepatopancreas, intestine, gill, stomach, and muscle tissues to analyze tissue distribution of the cat and gst transcripts, with the β-actin gene as the endogenous control [26]. The qRT-PCR primers are listed in Table S1. The primers were tested during normal PCR amplification, and the PCR products were visualized on a 1.5% agarose gel before qRT-PCR to verify primer specificity. The relative changes in mRNA levels of the two genes in different tissues were calculated using the 2 −ΔΔCt method [27]. The analysis was performed on tissues from nine males. All data are expressed as mean ± standard deviation (SD).

Enzyme Activity Assays Following MC-LR Exposure
Nine hepatopancreatic samples from each group were defrosted twice and homogenized on ice with 10 volumes of cold buffer (250 mM sucrose, 5 mM Tris-HCL, and 0.1 mM edetic acid-2Na; pH 7.5). The homogenate was centrifuged at 13,000 g and 4 °C for 10 min to obtain the supernatant for the assays. Protein concentration was determined using a commercial protein assay kit (Nanjing Jiancheng Bioengineering Institute). GST, SOD, CAT, and GPx activities were measured using commercial kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). One unit (U) of GST, CAT, and GPx activity was defined as the amount of enzyme consuming 1 μmol of substrate or generating 1 μmol of product/min/mg soluble protein (U/mg prot). One U of SOD activity was defined as the amount of enzyme required to inhibit the oxidation reaction by 50% and was also expressed as U/mg prot.

mRNA Expression of the Enzymes Following MC-LR Exposure
The gene expression patterns of hepatopancreatic gst, Cu/Zn-sod, cat, and gpx were detected by qRT-PCR in M. nipponense after exposure to 0.2-25 μg/L MC-LR for 96 h. Each transcript was analyzed in 9 individuals. The β-actin gene was used as an endogenous control. Relative mRNA levels were calculated using the 2 −ΔΔCt method and the formula F = 2 −ΔΔCt , ΔΔCt = (Ct, target gene − Ct, β-actin) MC-LR − (Ct, target gene − Ct, β-actin) control [27]. Data are expressed as mean ± SD. Statistical differences were tested with analysis of variance and the least significant difference test.

Cloning and Molecular Characterization of the Cat and Gst Gene cDNAs
The full-length cDNA sequences of cat (GenBank accession No. KC485002) and gst (GenBank accession No. KC485003) genes were determined by the RT-PCR and RACE methods. The cat cDNA was 1699 bp and contained a 1554 bp open reading frame (ORF) encoding a 518 amino acid polypeptide with a theoretical pI of 6.66 and a calculated molecular mass of 58.57 kDa and 135 bp 5′-untranslated region (UTR) ( Figure S1). Meanwhile, the gst cDNA was 1122 bp and contained a 651 bp ORF encoding a 217 amino acid polypeptide with a theoretical pI of 5.35 and a calculated molecular weight of 25.45 kDa, 173 bp 5′-UTR and 298 bp 3′-UTRs with a polyadenylation signal (AATAAA) ( Figure S2).

Homology and Phylogenetic Analyses of the Putative Cat and Gst Amino Acid Sequences
The putative CAT of M. nipponense contains NADPH binding site, proximal heme-ligand signature, and proximal active site signature ( Figure S3). The amino acid sequence identities of M. nipponense CAT with its counterparts from other crustaceans were more than 80%. The M. nipponense GST protein sequence showed 53.3%-72.1% similarity with the GST mu class from different organisms and had several conserved GSH binding sites and substrate binding sites ( Figure S4). Two phylogenetic trees were constructed from the amino acid sequence alignments for CAT and GST in crustaceans, fish, and mammals using the neighbor-joining method to better understand the positions of M. nipponense CAT and GST in the evolutionary history of the respective proteins. M. nipponense CAT was more similar to its counterpart in Macrobrachium rosenbergii (Figure 1). GST was more similar to its counterpart in Lepeophtheirus salmonis than those of other crustaceans and fish ( Figure 2).

Tissue Distributions for Cat and Gst mRNAs
The cat and gst tissue distributions in M. nipponense are shown in Figure 3. Both cat and gst mRNAs were expressed predominantly in the hepatopancreas. Meanwhile, cat and gst mRNAs were expressed at low level in other tissues with decreasing order of stomach, muscle, intestines and gill. The cat and gst mRNA levels in hepatopancreas were 80.1-fold and 16.8-fold higher than in intestine, respectively.

Effect of MC-LR on Biotransformation Enzyme and Antioxidant Enzymes
MC-LR had no effect on GST activity after the 12-and 24-h exposures ( Figure 4A), but GST activity increased significantly following MC-LR exposure at four concentrations for 48-96 h, except MC-LR exposure at 1 μg/L for 48 h and 0.2 μg/L for 96 h ( Figure 4A).
It was interesting that none of the MC-LR concentrations had a significant effect on GPx activity in the 12-and 24-h exposures ( Figure 4D). MC-LR at 0.2 and 5 μg/L significantly enhanced GPx activity after the 48-h exposure, whereas 1 and 25 μg/L MC-LR did not change GPx activity. In the 72-and 96-h exposures, MC-LR at 0.2-25 μg/L all significantly stimulated GPx activity (p < 0.05, Figure 4D). Values are expressed as mean ± standard deviation. Different lowercase letters indicate significant differences (p < 0.05), and different capital letters indicate highly significant differences (p < 0.01).

Expression Profiles of M. nipponense Gst, Cu/Zn-sod, Cat, and Gpx in the Hepatopancreas After MC-LR Exposure
The changes in gst, Cu/Zn-sod, cat and gpx mRNA expression are shown in Figure 5  The values were calibrated with the endogenous control β-actin. Each transcript abundance value is expressed as mean ± standard deviation (n = 9 for each value). Different lowercase letters indicate significant differences (p < 0.05), and different capital letters indicate highly significant differences (p < 0.01).

Discussion
In this study, CAT and GST cDNA sequences cloned from M. nipponense were presented, and sequence alignment analysis demonstrated this CAT shared high identity (>80%) with catalase from other crustaceans. The predicted size of the crayfish CAT polypeptide was 58.57 kDa, very close to swimming crab Portunus trituberculatus (58.57 kDa) [28], Chinese shrimp Fenneropenaeus chinensis (58.8 kDa) [29] and white shrimp Litopenaeus vannamei (57.6 kDa) [30]. The M. nipponense catalase proximal heme-ligand signature sequence (RLFSYNDTH) and the proximal active site signature (FDRERIPERVVHAKGAGA) were highly conserved with other species [31,32], which suggested that this CAT should have the same function with other species due to the conserved sequences and motif. Two ligand binding sites were present in GST. The G-sites were highly specific for glutathione binding sites and contained 11 highly conserved amino acids: Tyr 7 , Arg 13 , Trp 38 , Lys 46 , Gln 53 , Leu 54 Pro 55 , Gln 66 , Ser 67 , Glu 98 and Asp 99 (alignment positions in GST of M. nipponense 7, 11, 46, 50, 59, 60, 61, 72, 73, 101 and 102, respectively) [33,34]. The second ligand binding site, H-sites, interacting with electrophilic xenobiotic substrates, contained eight conserved residues: Tyr 7 , Phe 8 , Val 10 , Arg 13 , Val 104 , Tyr 108 , Asn 204 and Gly 205 [35]. Most of these characteristic amino acids were found in GST of M. nipponense, with the exception of Asn 53 , Asn 99 in the G-sites and Try 8 , Ile 10 in the H-sites, which were respectively substituted for Gln 53 , Asp 99 and Phe 8 , Val 10 . The predominant cat and gst expression in the hepatopancreas agreed with previous studies on white shrimp, Litopenaeus vannamei [36] and tilapias [37]. Therefore, the hepatopancreas was selected to detect cat and gst gene expression following exposure to MC-LR as the main antioxidative tissue of M. nipponense.
The rate or amounts of reactive oxygen species (ROS) generation can be induced by the presence of a wide range of natural and anthropogenic xenobiotics or toxins [38]. Much evidence suggested that MCs-induced oxidative damage could be a result of ROS generation [39,40]. The SOD-CAT system provides the first defense against oxygen toxicity. SOD catalyzes the dismutation of the superoxide radical to molecular oxygen and hydrogen peroxide [41]. Increased SOD activity is part of the adaptive mechanism to oxidative stress [42]. In the present study, the SOD activity was all stimulated by MC-LR at 0.2-25 μg/L in the early stage of exposure (12-72 h), and with an exposure duration from 12-48 h, the increase of SOD activity was shown to be time-dependent at almost four concentrations. The results indicated that the antioxidant defense system of M. nipponense was rapidly activated after being exposed to MC-LR and had high efficiency to restore oxidative balance. In 96-h exposure, MC-LR tended to inhibit SOD activity, especially at 25 μg/L, which was in line with the results in a previous study on freshwater mussel Dreissena polymorpha [43]. It is suggested that the exposure duration or concentration for MC-LR is not sufficient to provoke an increase in activity of the species. Pigeolet et al. [44] demonstrated that SOD was itself susceptible to oxidation and was suppressed by peroxides and oxygen derived free radicals. Therefore, in the present study, the high level of ROS at the end of exposure (96-h) might be detrimental to SOD activity. CAT is a key enzyme to remove the H2O2 produced by SOD catalyzing dismutation of highly reactive superoxide anions [45]. In the present study, the CAT activity was increased by MC-LR in the early stage of exposure (12-48 h), which indicated an activation of antioxidant defense system and non-lagged direct or indirect ROS production upon MC-LR exposure. However, at the end of exposure (96-h), the CAT activity had decreased with significant suppression at 1 and 25 μg/L. In estuarine crab Chasmagnathus granulates, seven-day MC exposure also caused a decrease of CAT activity [46]. Previous study suggested that peroxides and oxygen derived free radicals inhibited CAT activity [44,47]. So, we speculate that the CAT might be suppressed by extra ROS at the end of exposure.
GSTs are a group of multifunctional dimeric proteins involved in cellular detoxification reaction, catalyzing the enzymatic conjugation of GSH to toxic compounds, including MCs [48]. The reaction will neutralize the electrophilic sites of MCs, increasing their water solubility and favoring their excretion [49]. In mussel Dreissena polymorpha, crustacean Daphnia magna and zebrafish Danio rerio, MC-LR glutathione conjugates were formed enzymatically via soluble GST [50], which suggested that GSTs played a key role in detoxification of MC-LR. GSTs contain two forms: the soluble form (sGST) and the membrane form (mGST). The sGST is found in the cell cytosol and mGST is bound to membrane of microsome, peroxisome, mitochondria and endoplasmic reticulum. In the present study, the sGST and mGST were not separated, so the GST activity signifies a combination of sGST and mGST activities. In the freshwater shrimp Palaemonetes argentines exposed to MCs, activity of sGST and mGST had a similar regulatory trend at any exposure time point [51]. So, we speculate that modulation of sGST in hepatopancreas of M. nipponense upon MC-LR may have a similar trend to the regulation of total GST activity in the present study. In the present study, mu-GST activity was significantly stimulated by MC-LR at 0.2-25 μg/L from 48-96 h, which was consistent with the results of studies on other aquatic organisms such as crab and fish [52,53].
The function of GPx is to reduce free hydrogen peroxide to water and to reduce lipid hydroperoxides to their corresponding alcohols. GPx activity in the M. nipponense hepatopancreas was not significantly affected during the early stage of MC-LR exposure (12-24 h), while it was significantly stimulated by MC-LR even at 0.2 μg/L in the later stage of exposure (72-96 h). It seems that the enzymatic activity stimulation for GPx and CAT in response to MC-LR complements each other in the early and later stage of exposure. It suggested that the activation of GPx by MC-LR exposure might lag behind CAT.
Although increasing numbers of researches have focused on the oxidative stress responses arising from MCs, most of them were limited and mainly examined the changes in antioxidative enzymatic activity. Recently, differential expression of the genes encoding these antioxidant proteins has been used to detect biological toxicity [54,55]. In the present study, the transcriptional changes for gst, Cu/Zn-sod, cat, and gpx genes generally followed a similar trend to the changes in the respective enzyme activities in M. nipponense exposed to MC-LR. gst, Cu/Zn-sod, cat, and gpx transcription generally changed with higher amplitude compared with the activities of their respective enzymes during MC-LR exposure, suggesting that changes in transcription of antioxidant enzymes could serve as sensitive molecular biomarkers for exposure to MC-LR in M. nipponense, particularly for gst genes. In several exposure scenarios, the enzyme transcriptional changes were inconsistent with the changes in enzyme activities, such as CAT after the 12-and 24-h MC-LR exposures at 25 μg/L and GST and GPx after the 12-and 24-h MC-LR exposures at 5 and 25 μg/L. This discrepancy suggested that the mRNA level of an enzyme signified a snapshot of its activity at any given time point, whereas enzyme activity may be regulated post-transcriptionally [56]. A study on common carp, Cyprinus carpio, exposed to MC showed that a time window probably exists between transcription and translation, as enzyme activity and mRNA expression were detected simultaneously [57].

Conclusions
In the present study, the full lengths of gst and cat cDNAs were firstly isolated in M. nipponense, and their putative amino acid sequences displayed high similarity to their respective counterparts of crustacean species. The mRNAs of gst and cat were mainly expressed in the hepatopancreas. The activity and transcriptional level variations of biotransformation enzyme (GST) and antioxidant enzymes (SOD, CAT, and GPx) indicated that MC-LR induced ROS production and crayfish made corresponding reaction to restore oxidative balance. The transcriptional response was generally more immediate and with greater amplitude than the enzyme activity responses, particularly for GST, which suggested that transcription of these genes could be a sensitive molecular biomarker for acute MC-LR exposure.

Acknowledgments
The work was support by the Science Development Foundation of Zhejiang Province, China (No. 2012C12907-2).

Author Contributions
Zaizhao Wang and Zhiming Gu conceived and designed the study. Julin Yuan and Xueqin Wang performed the experiments. Julin Yuan wrote the paper. Zaizhao Wang and Yingying Zhang reviewed and edited the manuscript. All authors read and approved the manuscript.

Conflicts of Interest
The authors declare no conflict of interest.