Molecular characterization and transcriptional modulation of stress-responsive genes under heavy metal stress in freshwater ciliate, Euplotes aediculatus

Heavy metal pollutants in the environment are increasing exponentially due to various anthropogenic factors including mining, industrial and agricultural wastes. Living organisms exposed to heavy metals above a certain threshold level induces deleterious effects in these organisms. To live in such severe environments, microbes have developed a range of tolerance mechanisms which include upregulation of stress-responsive genes and/or antioxidant enzymes to detoxify the metal stress. Single cell eukaryotic microorganisms, i.e., ciliates, are highly sensitive to environmental pollutants mainly due to the absence of cell wall, which make them suitable candidates for conducting ecotoxicological studies. Therefore, the present investigation describes the effects of heavy metals (cadmium and copper) on freshwater ciliate, Euplotes aediculatus. The activities of antioxidant enzymes, i.e., catalase and glutathione peroxidase in E. aediculatus were determined under heavy metal exposure. Besides, the expression of stress-responsive genes, namely, heat-shock protein 70 (hsp70) and catalase (cat), has also been determined in this freshwater ciliate species under metal stress. The present study showed that the enzyme activity and the expression of these genes increased with an increase in the heavy metal concentration and with the duration of metal exposure. Also, these stress-responsive genes were sequenced and characterized to comprehend their role in cell rescue.


Introduction
Heavy metal contamination is increasing exponentially and rapidly in the environment especially in aquatic ecosystems due to their high rate of solubility (Gheorghe et al. 2017, Hameed et al. 2020. Anthropogenic forces, including mining and industrial activities act as major source of heavy metal pollution (Gutiérrez et al. 2008, Jin et al. 2018. Heavy metals are highly toxic as they induce initiation of reactive oxygen species (ROS). This ultimately affects genomic material by inducing DNA or protein damage and lipid peroxidation, gradually leading to cell death (Ali et al. 2019, Gutiérrez et al. 2008, Leonard et al. 2004, Valko et al. 2005. Microorganisms are known to express a range of tolerance mechanisms to combat metal stress to survive such unfavorable conditions (Gutiérrez et al. 2015, Igiri et al. 2018, Somasundaram et al. 2018. Over the past few decades, the interest towards heavy metal interaction with microorganisms has been increasing , Arora et al. 1999, Jin et al. 2018, Kim et al. 2011, Makhija et al. 2015, Somasundaram et al. 2019. The predominant molecular defence mechanisms displayed by microorganisms include the activation of antioxidant enzymes and bioaccumulation of heavy metals by metal-binding proteins such as metallothionein , Emamverdian et al. 2015, Ghori et al. 2019, Somasundaram et al. 2019. For the present study, cadmium (Cd) and copper (Cu) were selected for determining the heavy metal toxicity on freshwater ciliate collected from Sanjay Lake, Delhi, India. Cd (non-essential heavy metal) and Cu (essential heavy metal) were preferred for this study since these two heavy metals have been previously reported to be highly toxic to the ciliate species , Madoni 2010, Pudpong and Chantangsi 2015. Also, these heavy metals are reported to be present in higher concentrations in the environment which exceeds the permissible limit set by Bureau of Indian Standards (BIS) (Bhardwaj et al. 2017) as shown in Table 1. The permissible limit given by BIS for Cu and Cd are 0.05 mg/L and 0.003 mg/L, respectively (Bhardwaj et al. 2017) whereas in water samples the concentration of Cu and Cd varied from 0.07 to 5.02 mg/L and 0.016 to 0.300 mg/L, respectively. The sources of Cd in the environment (especially aquatic ecosystem) are Ni-Cd batteries, plastic coloration, and various electronic products (Idrees et al. 2018). Whereas the sources of Cu are electrical industry, chemical weathering, steel production, copper mining/smelting, and agriculture wastes (Hussain et al. 2017).
The concentration of heavy metals (Cd and Cu) in freshwater bodies of Delhi, such as River Yamuna, increased exponentially from 2014 to 2021 (Table 1). Also, the concentrations of Cd and Cu reported from the other freshwater bodies of the Delhi, i.e., Sanjay Lake and Okhla Bird Sanctuary were observed to exceed the permissible limit given by BIS (Table 1). To control this prevailing condition, it is essential to evaluate heavy metal toxicity. For early detection of environmental toxicity, living organisms are nowadays used as environmental biomarkers (Lionetto et al. 2019). It has been reported that since 1950s, ciliated protists are being used in various ecotoxicological studies (Grebecki andKuznicki 1956, Vilas-Boas et al. 2020). Ciliates are present in diverse habitats such as aquatic and terrestrial ecosystems (Abraham et al. 2019a). They play a crucial role in shaping the microbial diversity in aquatic ecosystems (Abraham et al. 2019a). Since ciliates do not have cell walls, they show accurate and constant sensitivity to environmental pollutants. Also, they exhibit a short life cycle and can be easily cultured and maintained under laboratory conditions (Somasundaram et al. 2018). Hence, the diversity of ciliates and their abundance are investigated in several studies to study the environmental changes (Abraham et al. 2019b, Maurya et al. 2020, Xu et al. 2014) and in assessment of heavy metal pollution , Gutiérrez et al. 2003, Somasundaram et al. 2018, Vilas-Boas et al. 2020, Yeomans et al. 1997.
In this study, the freshwater spirotrich ciliate, Euplotes aediculatus (Abraham et al. 2021) which was isolated from Sanjay Lake (Delhi), has been selected as model system to investigate the molecular responses exhibited by the ciliate under heavy metal (Cd and Cu) stress. Euplotes aediculatus was selected for this study since this ciliate species is observed in almost all freshwater bodies of Delhi throughout the year. Also, the genus Euplotes has been previously reported to be more tolerant to environmental stress (Chen et al. 2019, Rehman et al. 2006, 2008. Previously, it has been reported that the property of bioaccumulation, and involvement of antioxidant enzymes act as major cell defense mechanisms in different species of genus Euplotes, i.e., E. mutabilis (Rehman et al. 2006(Rehman et al. , 2008(Rehman et al. , 2009) and E. crassus (Kim et al. 2011, Kim et al. 2017, Mori et al. 2003. Toteja et al. (2017) reported that the superoxide dismutase activity was enhanced in E. aediculatus in the presence of heavy metal stress. The morphology of E. aediculatus has been reported in Abraham et al. (2021) where the body size of this species was observed to be107-119 × 72-82 µm (in vivo), and size after protargol impregnation was 81-107 × 52-74 µm. Its body is broadly obovate, rectangular in shape, and dorsoventrally flattened. Around 69% of the body length was covered by the adoral zone which contained 42-46 membranelles. It has six frontal, three ventral, five transverse, two left marginal, two caudal cirri, and eight dorsolateral kineties. The cytoplasm is colourless containing many refractive granules. It has double eurystomus type of silverline system. It has one large 3-shaped macronucleus and a micronucleus distinctly separated from the macronucleus (Abraham et al. 2021). Thus, in the present study, the activity of catalase and glutathione peroxidase, and expression of heat-shock protein 70 (hsp70) and catalase (cat), were investigated in E. aediculatus after exposing the cells to various concentrations of Cd and Cu. Besides, for the first time hsp70 and cat genes were characterized with respect to their structure and functions in the Indian population of E. aediculatus and were compared with the other reported and closely related ciliate species.

Water sample collection
Freshwater samples for the present study were taken from Sanjay Lake (28°36′51.12″N, 77°18′14.04″E), Delhi, India. While collecting the samples, pH and temperature of the freshwater samples were observed to be 7.2 and 23°C, respectively. This artificial lake was constructed by the Delhi Development Authority (DDA). The surface area of the lake was around 0.17 km 2 with a depth of 1-2.5 m.

In vitro culturing and identification of ciliates
The collected freshwater samples were passed through a mesh with the pore size of around <200 µm and the concentrate was transferred to large troughs in the laboratory. Mixed planktonic cultures were grown at room temperature by adding freshly boiled cabbage pieces which promoted the growth of bacteria, and that served as the food organisms for the ciliates. These samples were periodically examined for 5-10 days to obtain the maximum growth of ciliates. Clonal cultures of ciliates were finally raised by isolating single cells by micropipettes. These cultures were then raised in Pringsheim's medium (Chapman-Andresen 1958) and the temperature was maintained at 22-23°C (Abraham et al. 2021).
Cliates were identified after observing under a stereoscopic microscope, phase-contrast microscope, and by molecular tools (Abraham et al. 2019a). The ciliary positioning and structures were observed by protargol and silverline staining methods (Abraham et al. 2019a(Abraham et al. , 2021. The nuclear morphology was determined by Feulgen staining (Chieco andDerenzini 1999, Feulgen andRossenbeck 1924). For molecular analysis, 18S rRNA gene was sequenced and submitted to the GenBank database with the accession number KX867114 (Abraham et al. 2021). The details of primers used and PCR protocol followed to amplify 18S rRNA gene are available in Abraham et al. (2021).

Evaluation of heavy metal toxicity
Toxicity assays were carried out for Cd and Cu to obtain the range of tolerance limits (i.e., 0% to 100% survivability) in E. aediculatus. The stock solutions (1000 mg/L) of cadmium chloride (CdCl 2 ), and copper sulfate (CuSO 4 ), were prepared in Pringsheim's medium to determine their lethal concentrations. LC 30 , LC 50 , and LC 70 doses of Cd and Cu were subsequently determined in E. aediculatus as given in Abraham et al. (2017). The cells were accordingly exposed to varying concentrations (control, LC 30 , LC 50 , and LC 70 ) of heavy metals (Cd and Cu) and maintained at 23°C for 24 h to study the activity of antioxidant enzymes. For determining the gene expressions, the cells were maintained for 24 h and 48 h after exposing them to the LC 50 doses of heavy metals (Cd and Cu). All the experiments were conducted with respective controls and done in triplicates.

Enzyme assays
The activities of antioxidant enzymes, i.e., catalase (CAT) and glutathione peroxidase (GPx) were investigated in the Indian population of E. aediculatus after heavy metal exposure. The CAT activity was assayed according to the method given by Luck (1963) and GPx activity by Fortress kit. The control (untreated) and heavy metal treated cells (100 cells/ml) were pelleted separately. The pelleted cells were lysed in pre-cooled mortar and pestle. Lysates were transferred to pre-cooled eppendorfs and were coldcentrifuged (4°C) at 10,000 rpm for 20 min. The supernatants containing the enzyme extracts, were collected to determine the enzyme activities . All the experiments were conducted in triplicates.

CAT enzyme assays
To determine the enzyme activity, around 100 µl of enzyme extract (collected supernatant) was added to 1 ml of 10 mM H 2 O 2 . This was measured at 240 nm in a UV-VIS spectrophotometer (Genesys) and the time required for the decrease in absorbance was recorded at an interval of every 30 s for 2 min. The activity of this enzyme was thus calculated in the International unit where one unit implies the amount of CAT enzyme required to break down one µmoles of H 2 O 2 per minute.
Catalase activity was calculated by: ΔA= min Â volume of hydrogen peroxide Molar coefficient of hydrogen peroxide Â volume of enzyme extract GPx enzyme assays 50 µl of enzyme extract is mixed with the reaction reagents given in Fortress kit and absorbance was taken at 340 nm using UV-Vis spectrophotometer by taking readings at an interval of every 30 s for 2 min. GPx converts reduced glutathione (GSH) to oxidized (GSSG) where cumene hydroperoxide (provided by the kit) acts as substrate. The oxidized glutathione is converted back to reduced glutathione by glutathione reductase and NADPH cofactor. Thus, in this reaction, NADPH gets converted to NADP + where this decrease in NADPH absorbance is measured at 340 nm using Genesys UV-Vis spectrophotometer.
The concentration of GPx was calculated by: where, 8412 ¼ Total volume Â 1000 Sample volume Â millimolar extinction coefficient Â light path Isolation of total RNA and cDNA preparation Qiagen RNeasy Mini Kit (QIAGEN, India) was used to isolate total RNA from the control and heavy metals exposed live cells of E. aediculatus (100 cells/ml). This was further analyzed by running the isolated RNA on 1.2% formaldehyde agarose (FA) gel. With the help of Qiagen Reverse Transcriptase kit (QIAGEN, India), cDNA was prepared from the isolated RNA where 7-9 μl of RNA constituting to 350-500 ng of RNA, was used.

Quantitative real-time PCR (qRT-PCR)
Primers with amplicon size ranging between 100-200 bp were designed for qRT-PCR using Primer3 online software and custom synthesized by M/s Biolinkk Pvt. Ltd. The amplicon size of the primers designed for 18S rRNA gene (taken as an endogenous control for the reaction) from E. aediculatus (GenBank accession number: KX867114), was 103 bp (forward: 5′-TGT CAG AGG TGA AAT TCT CG-3′, reverse: 5′-GTC TTT GAT CCC CTA ACT TTC-3′). Similarly, 177 bp amplicon size of hsp70 gene from Sanjay Lake (SL) population of E. aediculatus (forward: 5′-GCT GGA GTC ATT GCA GGA TT-3′, reverse: 5′-CTG CAG TTG CCT TAA CTT CG-3′), and 121 bp amplicon size of cat gene from SL population of E. aediculatus (forward: 5′-TAA CCA GGG AGC TTG GGA CT-3′, reverse: 5′-GTG GGA TGA ATA TCC GTT C-3′) were designed using Primer3 online software. Quantitative real-time PCR was performed to investigate the expression patterns of hsp70 and cat genes in control and heavy-metal (Cd and Cu) treated cells. For each PCR reaction, 10 μM of primer set and 1 μl of cDNA were used. The reaction conditions set for this qRT-PCR were as follows: 95°C/ 3 min; 40 cycles of 95°C/30 s, 60°C/30 s with a 0.5°C increase for every 5 s. This PCR reaction was performed using SYBR Green dye (Applied Biosystems, life technologies, Invitrogen), and on the Applied Biosystems ViiA6 system. All the experiments were performed in triplicates. The obtained data was in threshold cycle (CT) values and the fold change in the relative gene expression between control and treated cells was calculated according to the 2 −ΔΔCT method (Livak and Schmittgen 2001). The change of threshold cycle number (ΔCT) was determined by calculating the difference between the CT values of the target genes and the reference genes for each sample. Similarly, the ΔΔCT value was calculated by determining the difference between the ΔCT values of treated cells with controls. The specificity of each primer was tested by quantitative PCR which was determined by melting curve analysis. For all the genes including endogenous control gene (18S rRNA gene, cat, hsp70) in control and for 24 h and 48 h heavy metal exposure standard curves were obtained using tenfold cDNA dilutions and by determining their CT values. All the experiments were done in triplicates. To check the efficiency of the primers, the standard line parameters, i.e., the slope of the trendline, coefficient of determination (R 2 ), and amplification efficiency (E) [where E = 10 (−1/slope) ] (Rasmussen 2001) of each gene were determined. The slope, R 2 , and E ranged from −3 to −3.75, 0.81 to 1, and 1.86 to 2.19, respectively. The PCR efficiency percent [(E-1)x100] ranged from 109% to 119% for 18S rRNA gene, 95% to 119% for cat gene, and 86% to 104% for hsp70 gene (Fig. 1).

Statistical analysis
Results of enzyme assays were analyzed statistically using One Way ANOVA applying Post-hoc Tukey's test with the help of IBM SPSS 22.0 statistics software and by Student's t-test (parametric tests) for which XLSTAT software was used. Results of qPCRs were analyzed statistically where all the results were considered to be significant with P-value showing < 0.05 and values of fold change were mentioned as mean ± standard deviation (SD).

DNA isolation, amplification, and sequencing
Qiagen DNeasy blood & tissue kit (QIAGEN, India) was used to isolate total genomic DNA from 50 cells/ml culture of E. aediculatus. Primers for isolating stressresponsive genes were designed by Primer3 online software and synthesized by M/s BioLinkk Pvt. Ltd. For hsp70 gene, the primers were selected from Euplotes aediculatus cytosolic hsp70 gene (GenBank accession number AF031354) and for cat gene, the primers were designed from Tetmemena sp. SeJ-2015 contig 00995 (GenBank accession number LASU02000995). For hsp70 gene isolation from E. aediculatus, the primers used were: forward primer 5′-TCA TGT GTC GGA GTA TGG GTT A-3′ and reverse primer 5′-TGA TGA GTT GTT GGA CTT TTG G-3′. Additionally, two internal primers: one forward 5′-ATG CAG TCG TCA CAG TTC CA-3′ and one reverse 5′-TTC TTC AAA TTT GCC CCT TG-3′ were used. For cat gene isolation, custom synthesized forward primer 5′-GAG TTG TTC ACG CCA AGG G-3′ and revere primer 5′-GCA CCA AGT CTG TGT CTA TGG GT-3′ were used. The PCR conditions used for this reaction were as follows: 1st cycle including 95°C for 5 min, 50°C (for cytosolic hsp70)/ 56°C (for cat) for 1 min, 72°C for 1 min followed by 30 cycles showing denaturation at 95°C for 45 s, annealing at 50°C/56°C for 45 s and extension at 72°C for 45 s. And the last cycle with 95°C for 45 s, 50°C/56°C for 45 s and 72°C for 10 min. The final PCR products of cytosolic hsp70 and cat genes were run on 0.8% agarose gel and these genes were subsequently eluted using Qiagen QIAquick gel extraction kit (QIAGEN, India). Finally, these purified products of cytosolic hsp70 and cat genes were sequenced by Sangers' di-deoxy method using Sequence Scanner Software 1.0 of Applied Biosystems, Inc. (AB1).

Sequence analysis and characterization of stressresponsive genes
The obtained nucleotide sequences were subjected to Basic Local Alignment Search Tool (BLAST) homology searches from the NCBI database (http://www.ncbi.nlm.nih.gov/bla st). Besides, the amino acid sequences were retrieved from ExPASy translate tool (http://web.expasy.org/translate/). In addition, conserved domains were identified using NCBI (National Center for Biotechnology Information) conserved domain search (CD-search; http://www.ncbi.nlm.nih.gov/ Structure/cdd). I-TASSER online server (Yang and Zhang 2015) and Swiss-Model (http://swiss model.expasy.org/inter active) were used to predict the tertiary structures of HSP70 and CAT protein. The stereochemical quality of these predicted protein structures was determined from Ramachandran plot using PROCHECK (Laskowski et al. 2001) and Swiss-PDB viewer (Guex et al. 2009). The conserved domains in HSP70 and CAT protein sequences of E. aediculatus were compared with the other reported ciliates by aligning the multiple protein sequences using BioEdit 7.2.1 sequence alignment editor software (Hall 1999).

In silico molecular docking studies of HSP70 and CAT proteins
The predicted structures of the target proteins involved in cell defense mechanisms were used for molecular docking. Their respective substrates were retrieved from metal PDB (http:// metalweb.cerm.unifi.it/) and/or PDB (Protein data bank) (https://www.rcsb.org/) which acted as ligands. The PDB files of the target and its respective ligand were read as inputs for AutoDock4.2 software (http://autodock.scripps.edu/dow nloads/autodock-registration/autodock-4-2-download-page/) to carry out the docking simulation. Kollman united atom charges, and polar hydrogens were added to the target protein for docking simulation. AutoDock requires pre-calculated grid maps, one for each atom type, present in the ligand being docked as it stores the potential energy arising from the interaction with the target. This grid must surround the active site of the target protein. For the best conformers, the Lamarckian Genetic Algorithm (LGA) was selected. These were then further analyzed using Cygwin command (http://www.cygwin. com/install.html) to obtain around 10 conformations of the target-ligand interaction files. These conformations were finally visualized in Discovery Studio Visualizer (http://accelrys.com/products/discovery-studio/ visualization-download.php) and the best conformation of target-ligand file was selected from the 10 conformations with respect to their inhibition constant and binding energy and saved as the final image file (Rizvi et al. 2013). With these online bioinformatic tools, a comparative molecular docking was conducted between E. aediculatus and other closely related species of the genus Euplotes to determine the difference in their active sites and functioning of the proteins.

In vitro culturing of Euplotes aediculatus
For the present study, Euplotes aediculatus (Fig. 2) was identified and cultured at the laboratory conditions. The detailed morphology and culture conditions of this ciliate species are available in Abraham et al. (2021).

Heavy metal toxicity
In Euplotes aediculatus, Cu was observed to be more toxic as compared to Cd . The LC 30 , LC 50 , and LC 70 values for Cd were 1 mg/L, 2 mg/L, and 3 mg/L, respectively. For Cu, the LC 30 , LC 50 , and LC 70 values were 0.1 mg/L, 0.2 mg/L, and 0.4 mg/L, respectively ).

CAT and GPx activity
The CAT and GPx activities were determined in E. aediculatus after exposing the cells to varying heavy metal concentrations, i.e., 0 mg/L (control), 1 mg/L (LC 30 ), 2 mg/L (LC 50 ) and 3 mg/L (LC 70 ) of Cd and 0 mg/L (control), 0.1 mg/L (LC 30 ), 0.2 mg/L (LC 50 ), and 0.4 mg/L (LC 70 ) of Cu. The CAT activity increased in E. aediculatus with an increase in concentration of heavy metals (Cd and Cu) but at higher doses, the activity of CAT enzyme decreased moderately, yet significantly higher than control (Fig. 3a). The concentration of CAT in control, LC 30 , LC 50 , and LC 70 doses of Cd were 1.53 U/ml, 3.90 U/ml, 2.22 U/ml, and 2.14 U/ml, respectively. The concentration of CAT in control, LC 30 , LC 50 , and LC 70 doses of Cu were 3.06 U/ml, 3.82 U/ml, 5.05 U/ml, and 4.05 U/ml, respectively. Similarly, the GPx activity was dose-dependent where the enzyme activity increased proportionally with heavy metal concentration (Fig. 3b). The concentration of GPx in control, LC 30 , LC 50 , and LC 70 doses of Cd were 33.65 U/l, 36.65 U/l, 46.27 U/l, and 54.68 U/l, respectively. The concentration of GPx in control, LC 30 , LC 50 , and LC 70 doses of Cu were 22.43 U/l, 24.11 U/l, 35.61 U/l, and 40.94 U/l, respectively.

Transcriptional regulation of stress-responsive genes under metal stress
When the cells were exposed to heavy metals, the expressions of the stress-responsive genes (cytosolic hsp70 and cat) in E. aediculatus increased significantly as compared to the control/ unexposed cells.

Cytosolic hsp70 gene
When the cells were exposed to LC 50 dose of Cd, the expression of the cytosolic hsp70 gene increased by 3.25 fold after 24 h exposure and by 2.13 fold after 48 h exposure. Similarly, when the cells were exposed to LC 50 dose of Cu, the expression of the cytosolic hsp70 gene increased by 1.85 fold and 2.48 fold after 24 h and 48 h of metal exposure, respectively (Fig. 4a).

Cat gene
When the cells were exposed to LC 50 dose of Cd, the expression of the cat gene increased by 1.75 fold after 24 h exposure and by 15.24 fold after 48 h exposure. Similarly, when the cells were exposed to LC 50 dose of Cu, the expression of the cat gene increased by 4.86 fold and 2.89 fold after 24 h and 48 h of metal exposure, respectively (Fig. 4b).

Cytosolic hsp70 gene
The length of the partially sequenced macronuclear cytosolic hsp70 gene of E. aediculatus was 871 bp and encoded a putative polypeptide of 290 aa (Fig. 5a) (Fig. 5b, c). Amino acid composition indicated that alanine content was predominantly present constituting 10% of the total protein and methionine and histidine were found to be relatively low (1%). Ramachandran plot for this predicted protein model exhibited 94.7% in the most favorable region which indicates that the model has a good stereochemical quality.  to SSU rRNA (18S rRNA) gene which was used as a reference housekeeping gene. Data represent mean ± SD of three replicates of exposed cell. Asterisks on the bar show the significance level [*P < 0.05; **P = 0.01; ***P < 0.01] The HSP70 protein sequence of E. aediculatus was aligned with other reported HSP70 sequences of genus Euplotes and with Tetmemena sp., Sterkiella nova, Stylonychia lemnae belonging to the class Spirotrichea, and with Paramecium and Tetrahymena thermophila belonging to the class Oligohymenophorea (Fig. 6). It was found that the cytosolic signature sequence VFDA was conserved in all the species of the genus Euplotes. In addition, the universally conserved sequence of ATP binding domain of HSP70 protein, i.e., IFDLGGGTFDVSLLT, was identical in all the ciliate species except for Tetrahymena thermophila where Val was observed instead of Ile (Fig. 6).
Also, the predicted structure of HSP70 in E. aediculatus was compared with the above-mentioned ciliate species. The structure was observed to be similar in all the species with the involvement of Arg, Glu, and Lys residue near the ATP binding site (Fig. 7).

Catalase gene (cat)
The length of the partially sequenced cat gene of E. aediculatus was 858 bp long coding for 286 aa. This sequenced gene was submitted to the GenBank database (accession number MN044623). BLAST result of nucleotide sequences showed 81.64% similarity with E. vannus (accession number JN601111). BLAST result of the predicted protein sequence of cat gene showed 81.47% similarity with E. vannus (accession number AEZ02310). The putative molecular weight and pI of the predicted CAT protein were 32.89 kDa and 6.54, respectively. The amino acid residue contained high content of aspartate and glycine and low content of cysteine. The amino acids that are involved in binding with the heme group have been highlighted in Fig. 8a. The predicted homotetrameric structure of CAT protein contained heme group at the catalytic site of each monomer as shown in Fig. 8b-e. Ramachandran plot for this predicted protein model showed 96.04% of residues in the most favorable region indicating that the model has a good stereochemical quality.
The CAT protein sequence of E. aediculatus was compared with the reported CAT protein sequence of E. vannus, Tetmemena sp., and Tetrahymena thermophila. The distal heme-binding ligand i.e., VVHAKGAG, is highly conserved in all the ciliates whereas amino acid variation was observed in the proximal heme-binding ligand, i.e., RPF/RLF/RSF where the sequence contained Arg, Pro, and Phe in E. aediculatus. In Tetmemena sp., Ser was observed instead of Pro residue. Whereas, Leu was observed in place of Pro in E. vannus and Tetrahymena thermophila (Fig. 9) which resembled that of human catalase (Mashhadi et al. 2016).
In addition, their predicted protein structure were compared where involvement of His, Asn, and Tyr residues were observed at the catalytic site in all the other ciliate species but in E. aediculatus though the involvement of distal His residue was observed, the presence of Lys could be observed instead of Asn at the catalytic site (Fig. 10).

In vitro ciliate culturing and heavy metal toxicity
In the present study, Euplotes aediculatus (spirotrich ciliate) isolated from freshwater source, i.e., from Sanjay Lake, has been selected to study its molecular response to essential (Cu) and non-essential (Cd) heavy metals. The concentration of Cd and Cu in freshwater bodies of Delhi has been reported to be 0.016-0.047 mg/L and 0.085-5.02 mg/L, respectively, where Sanjay Lake contained 0.017 mg/L of Cd and 0.085 mg/L of Cu (Table 1). This exceeded the permissible limits given by BIS for Cu and Cd which are 0.05 mg/L, and 0.003 mg/L, respectively (Bhardwaj et al. 2017). In the present investigation, E. aediculatus was exposed to 1-4 mg/L of Cd and 0.1-0.3 mg/L of Cu to study the enzyme activity and transcriptional modulation of stressresponsive genes. Cu is found to be more toxic to E. aediculatus as compared to Cd similar to other reported studies , Gutiérrez et al. 2008, Kim et al. 2011, Madoni and Romeo 2006. In the environmental freshwater bodies, although the concentrations of Cu is reported to be high (more than Cd concentration) as given in Table 1, ciliates such as E. aediculatus could have survived such harsh condition by forming cysts or with the help of other adaptations and survival strategies which cannot be compared to the favourable and optimal laboratory conditions. Hence, even the low concentrations of Cu in laboratory conditions is detected to be toxic to the cells. Additionally, Cd has been reported to be sequestered by ciliates by metal-binding proteins such as metallothionein whereas Cu is observed to affect the electron transport chain directly, thereby acting as a more toxic heavy metal as compared to Cd . Therefore, the range of LCs values obtained in the present study for Cd was observed to be higher than the reported environmental concentration of Cd (Table 1). Whereas the range of LCs values was low in Cu, and found to be even lower than the metal concentration reported in the various environmental freshwater bodies in India (Table 1).

Enzyme assays
The activity of CAT enzyme was observed to be dosedependent in E. aediculatus though, at higher doses (LC 70 ) of Cd and Cu, there was a slight drop in the enzyme activity. A similar kind of result has been reported in Tetmemena sp. (freshwater ciliate species) where the activity of CAT increased with an increase in the concentration of heavy metal but decrease in the enzyme activity has been reported at higher concentration (Somasundaram et al. 2019). In Euplotes vannus, the activity of CAT enzyme increased when treated with the chemical, nitrofurazone (Hong et al. 2015). But significant decrease in the enzyme activity was observed at a higher chemical concentration and with increase in duration of exposure to nitrofurazone (Hong et al. 2015). Simiarly, CAT activity increased in Paramecium sp. after Cd exposure (Benlaifa et al. 2016). Also, in green microalgae such as Scenedesmus sp. and Chlorella pyrenoidosa, increase in CAT activity has been reported under heavy metal (chromium, copper, lead, and zinc) stress (Ajayan and Selvaraju 2012). Similarly, under heavy metal (Cd and Cu) stress, an increase in CAT activity was observed in brown mussels (Perna perna) (Boudjema et al. 2014) and in mangrove plant seedlings (Kandelia candel) where in the presence of Cd, the activity of CAT enzyme increased with heavy metal concentration but started decreasing at a higher metal concentration (Zhang et al. 2007).
In this study, the activity of GPx was observed to increase significantly in E. aediculatus with increase in heavy metal concentration. Similar studies have been reported earlier where the activity of this enzyme increased significantly under heavy metal stress in plant, (Salvinia auriculata), in freshwater gammarid (Gammarus pulex), and in freshwater snail (Lymnaea natalensis) (Mnkandla et al. 2019, Vestena et al. 2011). But at a higher concentrations of Cd and Cu, GPx activity was also affected but not as frequently as CAT enzyme (Somasundaram et al. 2019). GPx is most abundantly present in the cytoplasm of living organisms and is relatively less prone to the inhibitory effect of oxidative stress as compared to other antioxidant enzymes (Zitka et al. 2012, Zoidis et al. 2018). Since GPx belongs to the selenoprotein family, it has selenium (Se) as cofactor, and this cofactor increases the stability of the enzyme and helps to fight against oxidative stress effectively (Ferro et al. 2020, Zoidis et al. 2018). Se has an important role in fighting against the oxidative damage induced by heavy metals (Malik et al. 2012). Therefore, GPx appears to be a promising antioxidant enzyme in ROS detoxification. Earlier studies have reported that heavy metal stress increases the enzyme activity in the living organisms immediately after exposure (Bhaduri and Fulekar 2012, Somasundaram et al. 2019. Since heavy metals are main sources for generating reactive oxygen species (ROS), the activity of superoxide dismutase (SOD) increases to convert ROS to hydrogen peroxide ). This increases the activity of catalase to reduce the production of H 2 O 2 (Somasundaram et al. 2019) . But at higher concentrations of heavy metals or prolonged exposure to heavy metals, the enzyme (SOD and CAT) activity was affected and started decreasing (Somasundaram et al. 2019. At high concentrations of heavy metals, especially redox inactive metals such as Cd, the activity of CAT enzyme is affected (Boudjema et al. 2014). Cd, at higher concentration, is known to induce toxic effect by binding to the sulfhydryl (-SH) group of heme (porphyrin ring) present at the active site of CAT enzyme, thereby lowering enzyme activity (Boudjema et al. 2014, Radhakrishnan 2008, Vestena et al. 2011. However, GPx and GR are known to show decreased or low activity at the beginning of heavy metal exposure but gradually their activities are reported to increase with increase in metal concentration (Bhaduri andFulekar 2012, Gomes-Junior et al. 2006). Heavy metals especially redox inactive metals such as Cd, decrease the concentration of GSH in the living organism for the synthesis of phytochelatin enzyme which acts as a metal chelating enzyme (Bhaduri andFulekar 2012, Gomes-Junior et al. 2006). At a higher concentration of heavy metals, where CAT activity starts to decrease, GSH concentration has been reported to increase gradually followed by GPx and GR activity to detoxify endogenous ROS such as hydrogen peroxide (Fang et al. 2019).
Comparison of antioxidant enzyme activity in E. aediculatus (present study) with Tetmemena sp. (Somasundaram et al. 2019) The CAT and GPx activities were observed to increase with an increase in the concentration of metals in both E. aediculatus (present study) and in Tetmemena sp. (Somasundaram et al. 2019). When further analyzed, it was observed that E. aediculatus showed high GPx activity in comparison to Tetmemena sp. whereas CAT activity was relatively less in the presence of Cd and Cu. The CAT activity observed in Tetmemena sp. ranged from 2.52-4.36 U/ml (under Cd stress) and 2.83-7.11 U/ml (under Cu stress) with respect to the increasing concentration of heavy metals [control to LC 70 ] whereas in E. aediculatus, the activity ranged from 1.53-2.14 U/ml (under Cd stress) and 3.06-4.05 U/ml (under Cu stress). However, GPx activity in Tetmemena sp. was observed to range from 21. respectively,whereas,.94 U/l under Cd and Cu stress, respectively. It has been reported that CAT enzyme usually involved in simplifying only H 2 O 2 (hydrogen peroxide) to water and oxygen whereas GPx enzymes are involved in the breakdown of both hydrogen peroxide and lipid peroxides (Góth et al. 2004, Ighodaro andAkinloye 2018). Also, GPx enzyme is present in cytosol and mitochondria (highly prone to oxidative stress) showing that these enzymes have a prime role in protecting the cells from various oxidative stresses (Gill andTuteja 2010, Ighodaro andAkinloye 2018). Since E. aediculatus exhibits high GPx activity, this species may show better metal tolerance as compared to Tetmemena sp.

Comparison of enzyme biomarkers in ciliate species
The activity of superoxide dismutase (SOD) has been reported to increase in freshwater ciliates, namely Tetmemena sp., Euplotes sp., and Notohymena sp. after heavy metal exposure . At Cu concentration ranging from 0.1 to 0.25 mg/L, the activity of SOD in Tetmemena sp. increased from 0.46 to 0.91 U/ml. At Cu concentration ranging from 0.1 to 0.30 mg/L, the SOD activity of Euplotes sp. increased from 0.42 to 3.34 U/ml. In Notohymena sp., the activity of SOD increased from 0.42 to 1.87 U/ml under Cu concentration ranging from 0.5 to 1.5 mg/L. Similarly, under Cd concentration ranging from 1 to 4 mg/L, an increase in SOD activity was observed in both Tetmemena sp. and Euplotes sp. which ranged from 0.29 to 2.34 U/ml. Whereas in Notohymena sp., the SOD enzyme activity increased from 0.28 to 1.15 U/ml at Cd concentration ranging from 2 to 7 mg/L . This shows that the sensitivity of enzyme biomarkers varies from species to species. The sensitivity of SOD enzyme was high in both Tetmemena sp. and Euplotes sp. as compared to Notohymena sp. as their enzyme activity was highly expressed at even lower concentrations of heavy metals (Cd and Cu). This was similar to the present study where increase in CAT and GPx activities was observed at lower concentrations of heavy metals, i.e., 0.1 mg/L of Cu and 1 mg/L of Cd in E. aediculatus.

Transcriptional regulation of stress-responsive genes under heavy metal stress
The expression of hsp70 gene was observed to increase in E. aediculatus with increase in duration of metal exposure. In Cd treated cells, there was a significant increase in the expression of cytosolic hsp70 gene after 24 h (LC 50 dosage) but after 48 h of metal exposure, a slight drop in the gene expression level was observed, yet significantly higher than control. Whereas, the expression of stress-responsive genes in Cu-treated cells increased significantly with an increase in the duration of metal exposure, i.e., after 24 h and 48 h of exposure (at LC 50 dosage). Similar to the present study, the expression of hsp70 gene was highest in Tambaqui fish (Colossoma macropomum) after 3 h of Cu treatment whereas, under Cd exposure, the hsp70 gene expression was upregulated after 1 h but downregulated after 3 h (Casanova et al. 2013).
E. aediculatus, when exposed to LC 50 doses of heavy metals (Cd and Cu), the expression of cat gene increased significantly after 24 h, but started decreasing after 48 h of metal exposure. Similar studies have been reported earlier where the expression of cat gene increased under heavy metal stress in various organisms, but at higher metal concentrations, a decrease in the gene expression has been reported (Aydin et al. 2016, Azpilicueta et al. 2008, Radhakrishnan 2008, Roh et al. 2006. Heavy metals, at higher concentration, cause a sudden increase in intracellular ROS levels which escape the scavenging activities of antioxidant enzymes and can significantly damage cell structure (Huang et al. 2019). These oxidative radicals, especially H 2 O 2 and . OH radical, can react with all biological molecules and induce DNA singlestrand breakage, thus affecting the transcription of stressresponsive genes at higher concentrations of metals (Hiramoto et al. 1996, Huang et al. 2019).
Comparison of transcriptional modulation of stressresponsive genes of E. aediculatus (present study) and Tetmemena sp. (Somasundaram et al. 2019) In both the spirotrich ciliates (i.e., E. aediculatus and Tetmemena sp.), the expression of hsp70 gene increased within 24 h of exposure of Cd but decreased after 48 h of metal exposure. Whereas under Cu stress, the expression of hsp70 gene increased with an increase in the duration of metal exposure, i.e., the expression was observed to be high at 48 h of exposure of Cu. In Tetmemena sp., the transcriptional expression increased up to 46 fold after 24 h of Cd exposure (LC 50 dose) whereas it decreased to nine-fold after 48 h Cd exposure (LC 50 dose) (Somasundaram et al. 2019). However, in the case of Cu, the transcriptional expression of cytosolic hsp70 in Tetmemena sp. was observed to increase with increase in metal duration, i.e., around threefold increase after 24 h and 29 fold increase after 48 h of Cu exposure (Somasundaram et al. 2019). Similarly, in E. aediculatus (present study), the expression of cytosolic hsp70 gene increased by 3.25 fold and by 2.13 fold after 24 h and 48 h exposure to LC 50 dose of Cd, respectively. Whereas, there was 1.85 fold increase and 2.48 fold increase after 24 h and 48 h exposure to LC 50 dose of Cu, respectively. This fairly indicates that the rate of protein degradation induced by Cd could be much faster than Cu since Cd is a non-essential heavy metal (Somasundaram et al. 2019). Also, Cd is highly capable of inhibiting many important functional metalloproteins by binding and replacing essential ions acting as cofactors resulting in protein damage (Tamás et al. 2014) and thereby expressing hsp70 gene at a faster rate as compared to Cu.

Molecular characterization of stress-responsive genes
The alignment of HSP70 protein sequences of the different species of Euplotes and other ciliates belonging to class Spirotrichea and Oligohymenophorea (Fig. 6) showed that the highly conserved sequence of ATP binding domain (ABD), i.e., IFDLGGGTFDVSLLT except for Tetrahymena thermophila belonging to class Oligohymenophorea where Val was observed in place of Ile, and cytosolic signature domain, i.e., VFDA, are identical in all the species. The main difference was observed in the conserved sequence of NBD (N-binding domain) which is involved in dimerization, i.e., IADAAYNQVARN. In E. aediculatus (present study), Ala (A) residue present in the beginning of the sequence, is different from the protein sequences of other species of genus Euplotes which have Gly (G) residue. Also, Tyr (Y) in the conserved sequence was replaced with basic amino acids, i.e., Asn (N) in E. aediculatus (accession no. AAC33419), E. eurystomus (accession number AAA99875), and by Lys (K) in other ciliate species.
The ATP binding domain of HSP70 protein was predicted in different ciliate species. It showed the presence of Arg, Lys, and Glu residues bound to the ATP molecule in all the ciliate species irrespective of different classes (Fig.  7). This observation could be well supported with previously reported data where Arg and Glu have been observed to help in the proper binding of ATP at the ATP binding site in HSP70 (Brehmer et al. 2001, Mayer andGierasch 2019). Lysine at the catalytic site interacts with the phosphate group of ATP and helps in proper ATP hydrolysis (Brehmer et al. 2001, Mayer andGierasch 2019). Lys and Glu residues are known to form salt bridge across the nucleotide-binding (ATP binding) cleft in both prokaryotes and eukaryotes (Brehmer et al. 2001, Mayer andGierasch 2019). Arg present in ABD stabilizes the interaction of ABD with substrate-binding domain (SBD) thereby helping in the proper binding of substrate and enhancing HSP70 protein activity (Vogel et al. 2006). Since the ATP binding domain of HSP70 protein is highly conserved, the predicted ATP binding site of HSP70 was similar in all the species of genus Euplotes.
The predicted protein sequence of CAT of E. aediculatus was compared with E. vannus, Tetmemena sp., and Tetrahymena thermophila. CAT protein has Tyr (Y) residue at the C-terminal domain that acts as proximal heme ligand and His (H) and Asn (N) residues at the N-terminal domain (NTD) that act as catalytically active distal residues (Mashhadi et al. 2016, Zámocký andKoller 1999). But in the present study, Lys (K) was observed instead of Asn (N) residue in E. aediculatus (Fig. 9).
In the present study, the predicted active sites of CAT protein were compared between E. aediculatus and E. vannus. In both the species, histidine residue was observed at the active site. Histidine generally binds to the porphyrin ring of heme at the catalytic site which is further stabilized by a cross-link with tyrosine residue (Mashhadi et al. 2016, Zámocký andKoller 1999). In addition to histidine at the active site, presence of Asn was observed in E. vannus whereas, Lys was observed in E. aediculatus (Fig. 10). It has been reported that Asn creates more polarity at the active site and hence enhances the enzyme function (Zámocký and Koller 1999). Due to this feature, the activities of CAT enzyme observed in E. aediculatus under Cd and Cu stress could have been relatively low as compared to Tetmemena sp. reported in Somasundaram et al., 2019.

Conclusion
In the present investigation, for the first time, the activity of catalase and glutathione peroxidase, and expression of hsp70 and catalase gene were studied after heavy metal (Cd and Cu) exposure in Euplotes aediculatus (SL population). The enzyme activity and the gene expression were observed to increase with an increase in concentration of heavy metals depicting that these genes may be used to evaluate the heavy metal toxicity. Besides, hsp70 and cat (stressresponsive genes) were characterized in E. aediculatus and compared with the other reported ciliate species. It was observed that, despite the similarity in their tertiary protein structures, amino acid variations could be observed in both the genes in E. aediculatus (SL population) when compared with other ciliates. In hsp70 gene, amino acid variations were noticed at the N-binding domain and the nuclear localization sequence of the ATP binding domain. In CAT protein, variation in the amino acid residue was observed in E. aediculatus where the presence of Lys was noticed instead of Asn at the catalytic site. Also, the enzyme activity and transcriptional modulation of stress-responsive genes were compared with the previously reported spirotrich ciliate, i.e., Tetmemena sp. and was found that though the transcriptional modulation was similar in both the species, the difference in the activity of CAT and GPx have been noticed where the activity of GPx was relatively high and CAT activity was observed to be relatively low when compared with Tetmemena sp. which could be due to the predicted structural changes in the CAT protein. Whereas, GPx activity was relatively high in E. aediculatus. Thus, in this investigation, the detailed study on enzyme activity, transcriptional expression of stress-responsive genes after heavy metal exposure, and the molecular characterization of these genes in E. aediculatus help in better understanding of the stress-regulation mechanism in this freshwater ciliate species that can be used to assess heavy metal toxicity.

Data availability
The nucleotide sequences obtained in this study have been deposited to GenBank, NCBI.