A putative merR family transcription factor Slr0701 regulates mercury inducible expression of MerA in the cyanobacterium Synechocystis sp. PCC6803

Abstract In cyanobacteria, genes conferring mercury resistance are not organized as mer‐operon, unlike in other bacterial phyla. Synechocystis contains only a putative MerR regulator, Slr0701, and a mercury reductase, MerA, located aside from each other in the genome. The slr0701‐mutant showed reduction in photosynthetic activity and reduced tolerance to mercury compared to the wild‐type. The incubation of wild‐type cells with HgCl2 resulted in the upregulation of slr0701 and slr1849 genes whereas mercury‐induced expression was not observed in the slr0701‐mutant. Slr0701 binds to a conserved cis‐regulatory element located in the upstream of slr1849 and slr0701 ORFs. The same element was also identified in the upstream of other cyanobacterial homologs. Slr0701 binds to cis‐regulatory element with faster association and slower dissociation rates in the presence of HgCl2. Although these genes were constitutively expressed, the addition of HgCl2 enhanced their promoter activity suggesting that mercury‐bound Slr0701 triggers induced expression of these genes. The enhanced promoter activity could be attributed to the observed secondary structural changes in Slr0701 in the presence of HgCl2. For the first time, we demonstrated the mechanism of merA regulation in a cyanobacterium, Synechocystis. Although merA and merR genes are distantly located on the cyanobacterial genome and distinct from other bacterial mer‐operons, the transcriptional regulatory mechanism is conserved.

readily accumulated by higher plants and causes severe damage to the cellular metabolism and physiology of the plants (Cargnelutti et al., 2006). Mercury hinders the photosynthetic electron transport and is exceedingly harmful as a result of the capacity of Hg 2+ to bind with proteins (Rai, Agrawal, & Agrawal, 2016). Once, in the cell, Hg 2+ forms covalent bonds with cysteine residues of proteins, it drains antioxidants. These mercuric ions in the aquatic environment persist for very long period in the sediments (Randall & Chattopadhyay, 2013). However, the microorganisms evolved with a molecular mechanism to resist the inorganic and organic forms of mercurial compounds involving mer-operon.
Mercury-tolerant bacterial strains reported till to date contain mer-operon in their genomes, which code for proteins involved in mercury detoxification. The main open reading frames, merR, merA, and merP, code for a transcriptional regulator, mercuric reductase, and a periplasmic mercury binding protein, respectively. In some bacterial species, additional genes such as merB code for organomercurial lyase, merD an additional transcriptional regulator, merE and merF, auxiliary transporters are present. The enzymatic reduction of the mercuric ions to elemental mercury, catalyzed by mercury reductase, is the main detoxification mechanism in bacteria (Barkay, Miller, & Summers, 2003). This key detoxification protein reduces the toxic Hg (II) to Hg (0). As a result of its low dissolvability in water and moderately high-vapour weight, the elemental mercury is discharged from the cell (Barkay et al., 2003;Silver, 1996). The biochemical basis of protection from inorganic-mercury compounds, for example, HgCl 2 seems to be similar in few bacterial species. Studies have shown that a wide range of microorganisms such as Shigella flexneri, Pseudomonas aeruginosa, Serratia marcescens, Xanthomonas sp., and Staphylococcus aureus possess mer-operon. However, when mer genes were compared, the number and their operonic organization differ among various bacterial genera (Barkay et al., 2003;Hobman, Wilson, & Brown, 2000).
Cyanobacteria, which occur in almost all environmental niches contain genes that code for a putative MerR-like transcriptional regulator and MerA, mercury reductase. In Synechocystis sp. PCC6803 (hereafter referred as Synechocystis), two putative MerR family transcription factors are predicted. One of them, Slr0794 has been reported to be involved in the regulation of genes related to Ni 2+ toxicity (García-Domínguez, Lopez-Maury, Florencio, & Reyes, 2000).
Several MerR transcription factors, though they are called as mercury-resistant regulators and classified under MerR family, are also known to be involved in resistance to toxic metals other than mercury. For example, ZntR (Brocklehurst et al., 1999) in Escherichia coli and PbrR in Ralstonia metallidurans CH34 (Harley & Reynolds, 1987) are MerR family members involved zinc and lead resistance, respectively. In Synechocystis, Slr1849, codes for MerA and biochemically it was demonstrated to be a mercury reductase (Marteyn et al., 2013).
However, the mer-operon organization and regulation in cyanobacteria are poorly understood. Being photosynthetic bacteria, progenitors of higher plant chloroplasts and occurring in almost all habitats on the earth, it is important to unravel the regulatory mechanism of mercury tolerance in cyanobacteria. In this study, we show that the mer genes of cyanobacteria are phylogenetically distinct from other bacterial phyla. We also demonstrate a mercury-induced regulation of merA gene by a putative MerR family protein in Synechocystis.

| Bacterial strains and culture conditions
Synechocystis, a glucose-tolerant strain that was initially acquired from Dr. J.G.K. Williams (Dupont de Nemours, Wilmington, DE, U.S.A.), served as the wild-type. Wild-type cells were cultured photoautotrophically at 34°C in BG-11 added with 20 mM HEPES/ NaOH, pH 7.5 under consistent light at 70 μE m −2 s −1 of photons as presented earlier (Krishna et al., 2013). The slr0701-mutant cells in which the slr0701 ORF was inactivated by inserting a spectinomycin (sp r ) cassette was also cultured similarly as described above with an exception that the BG-11 medium contained spectinomycin at 25 μg/ml in pre-culture. The growth of the cells was monitored by estimating the absorbance at 730 nm in a spectrophotometer (NanoDrop™, 2000C, Thermo Fisher Scientific).

| Identification and phylogenetic analysis of putative mercury responsive genes
Protein sequences involved in mercury resistance from Bacillus megaterium and E. coli were used as queries for searching homologs in Synechocystis genome, which is publicly available at "cyanobase" http://genome.microbedb.jp/cyanobase/ (Kaneko et al., 1996). Slr0701 and Slr1849 protein sequences from Synechocystis were used to search for homologues from other cyanobacteria. The homolog protein sequences were obtained from the KEGG and NCBI databases (http://www.genome.jp/dbget-bin/ and http://www.ncbi. nlm.nih.gov/gquery/) for building phylogenetic tree. Phylogenetic connections were deduced by phylogeny examination using http:// www.phylogeny.fr/advanced.cgi (Dereeper et al., 2008).

| Generation of slr0701-mutant and slr0701 + complement strains
We have generated a slr0701-mutant of Synechocystis by inserting a Ω-spectinomycin resistant (sp r ) cassette within the slr0701 ORF. A DNA fragment containing the slr0701 ORF with 602 bp upstream and 325 bp downstream flanking regions were amplified by PCR using sequence-specific primers: slr0701-FP (5′-CAC CCT GGT TTG ATC AAT ACT CC-3′) and slr0701-RP (5′-CGA TCG CCC ATC TGT GTT GAA G-3′). The PCR amplified fragment (1340bp) was ligated to a linear T-vector (InsTAclone™ PCR Cloning Kit, #K1214). The resultant plasmid pTslr0701 was used to inactivate the slr0701 gene by performing restriction digestion at HpaI site. The sp r gene cassette was PCR amplified with specific primers Sp-F (5′ AAACTTTTTAAATCCTTAATTATTTGCCCACTAAAC 3′), Sp-R (5′ ATCAAAGTT TAA AACTCC CCC AGG GTC TTA GTT C 3′) using ΔcrhR genomic DNA in which sp r cassette was previously used to inactivate crhR gene (Prakash et al., 2010). The DraI site in the Sp r specific primers was underlined. The DraI-digested Sp r cassette was cloned at the HpaI-digested pTslr0701 DNA construct by blunt end ligation. The final plasmid, pTslr0701::sp r in which the slr0701 ORF was disrupted by Sp r cassette was used to transform wild-type Synechocystis cells. The site of insertion of the Sp r cassette was confirmed by sequencing the pTslr0701::sp r DNA construct using the slr0701-F primer. The segregation analysis was performed by PCR amplification to check the extent of replacement of wild-type copies of slr0701 with the pslr0701::sp r in slr0701-mutant strain was verified by PCR amplification using genomic DNA as a template.
The slr0701 + complement strain was generated to check whether the phenotype change observed in slr0701 mutant was due to inactivation of slr0701 gene. The PCR-amplified slr0701 ORF with flanking regions as mentioned above was used for making complement. The resulting 1,340 bp DNA fragment was cloned by blunt end ligation at SmaI site located within the kanamycin (kan r ) cassette into a cyanobacterial replicative vector pVZ321vector (Zinchenko, Piven, Melnik, & Shestakov, 1999). The recombinant vector, pVZ-slr0701, was introduced into slr0701-mutant by triparental mating as described in Zinchenko et al., 1999. We transferred the recombinant replicative pvz-slr0701 into the recipient slr0701-mutant strain by conjugal mating with donor and helper E. coli strains. E. coli-DH5 α carrying pvz-slr0701 was served as a donor strain. The recipient, donor, and helper strains were mixed in a ratio of 10:1:1, on a membrane filter (Millipore, Catalog No: GSWP 04700), and incubated overnight on BG-11 solid agar supplemented with 5% LB under dim light for conjugal transfer. The colonies thus developed were selected with antibiotic selection pressure of 35 µg/ml of chloramphenicol (Cm r ) on BG-11 agar medium in addition to 25 µg/ml spectinomycin. The strain thus generated was named slr0701 + .

| Growth and viability of Synechocystis strains under HgCl 2
The phenotypic characterization of wild-type and the slr0701 mutant strains was performed in the presence of different concentration of iron, copper, zinc, cobalt, and mercury. About 50 ml of the wild-type, slr0701-mutant and slr0701 + strains were grown till mid-log phase, that is, OD 730nm equals to 0.6 and were collected by centrifugation at 3,500 rpm for 5 min using swing-bucket rotor (Eppendorf, R5804).
The cell pellets were washed twice with liquid BG-11 medium and resuspended in a small volume of fresh BG-11 medium to adjust the final density of the resuspended cells equivalent to 10 at OD 730nm .
The cell suspensions were serially diluted five times, taking each time 500 µl of resuspended cells into 500 µl BG-11 medium (in 1:1 ratio).
Finally, 50 µl of diluted cell suspensions was spotted onto solid BG-11 agar plate containing different concentrations of trace elements as mentioned above. Similarly, 50 µl of diluted cell suspensions was spotted onto solid BG-11 agar plate containing different concentrations of HgCl 2 (0-500 nM). To check the effect of mercury on their growth and viability, growth was monitored for 5 days from the day of spotting on the BG-11 agar plates and photographed on 5th day.
PSII activity was measured in presence of 1.0 mM p-benzoquinone (PBQ) as described earlier (Sireesha et al., 2012). Photosynthetic oxygen evolution was recorded at 1,000 μE m −2 s −1 light. Three independent cultures of wild-type and slr0701-mutant cells were estimated with or without HgCl 2 . Cultures were grown till 0.6 OD at 730 nm and measured the oxygen evolution and the data was used as a reference.
Then 500 nM of HgCl 2 was added to a final concentration of the cultures and measured oxygen evolution at 12 hr and 24 hr.

| Measurements of chlorophyll a fluorescence
We recorded chlorophyll a fluorescence of Synechocystis cell suspensions with a continuous excitation PEA fluorometer (PEA, Hansatech, King's Lynn, Norfolk, UK). The PEA fluorometer provides continuous excitation at 650 nm (3,000 μE m −2 s −1 ; Δλ = 22 nm). It perceives fluorescence at wavelengths above 700 nm (50% transmission at 720 nm) and records it continuously from 10 μs to 300 μs. The fluorescence curves were recorded using wild-type and slr0701-mutant cells as described (Sireesha et al., 2012). Cultures were grown till OD at 730 nm reached to 0.6, and then measured the chlorophyll a fluorescence. These data were used as a reference. Then 500 nM of HgCl 2 was added to a final concentration of the cultures and measured fluorescence at 12 hr and 24 hr.

| Transcript analysis of mercury responsive genes
Wild-type and slr0701-mutant cells were cultivated at 70 μE m −2 s −1 of light. 50 ml of the cells were killed immediately by the addition of 50 ml equal volume of cold 5% w/v, phenol in ethanol, and then total RNA was isolated as described earlier (Srikumar et al., 2017).
The RNA was treated with DNase I (Cat. No. 89836, Thermo Fischer Scientific) to remove the DNA contaminants. One-microgram of RNA from total RNA was converted to cDNA using Takara Kit (Cat. no. 6110A). RNA was isolated from wild-type and slr0701-mutant cells before and after treatment with HgCl 2 . This RNA was utilized for cDNA preparation with the Prime Script™ first strand cDNA Synthesis Kit (Cat. no. 6110A). qRT-PCR was done using the SYBR ® Premix Ex Taq™ II (TliRNase H Plus; feline. no. RR820A). Every reaction was completed in a 25 μl volume containing 12.5 μl of Power SYBR Green Master Mix, 0.2 μM of specific primers. (Table 1) and 5 μl of diluted cDNA was kept in triplicates for running qRT-PCR (Mx3005P, Agilent Technologies). The instrument was conditioned for 95°C for 10 min, and then 40 cycles of 30 s at 95°C, 60 s at 60°C, and 60 s at 72°C. For every reaction, the melting curves were investigated and the PCR product was analyzed on an agarose gel with the end goal to affirm the specificity of the RT-PCR.
Expression levels were normalized using the gap1 gene as an internal reference.

| Overexpression and purification of Slr0701
PCR amplification of slr0701 ORF was performed, with the specific primers slr0701-ExF (5′-gtacGC TAG CGT GAG CAT TAT GTT AAC CGT CAG C-3′) and slr0701-ExR (5′-gcatGAA TTC CTA AGT CAA CTG CTC ATT TAA CAA AC-3′). The NheI and EcoRI restriction enzyme sites are underlined. The PCR product was purified and then cloned into pET-28a (+) at the NheI and EcoRI sites to produce pET-slr0701 by site-specific restriction cloning. The C-terminally 6x-His-labeled At each round of buffer replacement, urea and imidazole concentrations were gradually decreased to attain the final buffer concentration of 50 mM Tris and 300 mM NaCl.

| Western blotting analysis
Western blotting experiment was performed as presented earlier with some modifications (Prakash et al., 2010). Polyclonal antibodies induced in rabbit against 6x-His-Slr0701 protein was utilized as primary antibody, and an HRP-linked antibody induced in goat against rabbit IgG was utilized as the secondary antibody. Wild-type cells were grown with and without 500 nM HgCl 2, and the cells were harvested for protein extraction at different time intervals. Protein samples were separated using SDS-PAGE and were blotted onto PVDF membrane (Cat. No. IPVH00010, Immobilon-P; Merck Millipore) with a semi-dry transfer apparatus (TE77-PWR semi-dry transfer unit, GE Healthcare). The levels of Slr0701 were detected immunologically with the ECL Plus, immunoblotting system (Cat. No. 1705060-61, Bio-Rad). We used anti-rabbit secondary antibodies conjugated to horseradish peroxidase (1:5,000) for detection. Blot was scanned using Bio-Rad chemidoc (XRS + ) and analyzed with Image Lab software (Bio-Rad).

| Prediction of cis-regulatory elements using MEME suite
Intergenic DNA regions of cyanobacterial orthologs obtained from blastP were taken and submitted to MEME (Motif Extraction by Multiple Expectation Maximizations) version 4.3.0. MEME was run using the default parameters to find maximum of three motifs per sequence with the distribution of zero or one occurrence per sequence (Bailey & Elkan, 1994).

| Gel retardation assays
The slr0701 and slr1849 upstream DNA regions were PCR ampli-

| Biomolecular interaction analysis by SPR
All the experiments in surface plasmon resonance (SPR) analysis were performed at 25°C using Biacore T200 instrument (GE Healthcare Life Sciences). To study the interactions under real-time conditions between 6x-His-Slr0701 protein and P slr0701 , a DNA fragment

| Analysis of P slr0701 and P slr1849 promoter activity
To monitor the activity of promoters, we fused LuxAB, reporter gene coding for luciferase to the promoter of slr0701 and slr1849. The (50%, v/v), the final concentration in the suspension was 1 mM). We used the LuxAB plasmid without P slr0701 and P slr1849 in the wild-type and slr070-mutant as a control to evaluate the metal-induced regulation of slr0701 and slr1849 promoters.

| CD spectroscopy
The upstream of slr0701 was amplified by PCR using specific primers

| Diversity in the genetic organization of mer genes in cyanobacteria
To identify the mercury responsive genes in Synechocystis, sequences of well-characterized mer proteins from certain gram-positive and gram-negative bacteria were used as queries and blastP search was performed against Synechocystis genome database as described in the experimental methods. The best hits of blastP search using wellcharacterized MerR and MerA were a putative merR transcriptional regulator and a mercury reductase encoded by ORF numbers slr0701 and slr1849, respectively. The evolutionary relationship of Slr0701 and Slr1849 with their respective orthologs is shown in Appendix information Figure A1. The phylogenetic trees demonstrate their presence across different bacterial and cyanobacterial genera. The homologs for Slr0701 and Slr1849 proteins were identified mainly in freshwater cyanobacterial species and some other non-photosynthetic bacteria isolated from soil as well. Though several of these cyanobacterial species were originally isolated from freshwater ponds, they were observed to exist in soil crusts also (Tirkey & Adhikary, 2005). It was also observed that both, MerR and MerA homologs from cyanobacterial species formed a separate clade ( Figure A1).
Homologs for Slr0701 and Slr1849 proteins were also present in Geobacter pickeringii, a metal-reducing isolate from sedimentary kaolin deposits (Shelobolina et al., 2007), and Acidihalobacter prosperus, isolated from geothermally heated sea sediments (Khaleque, 2017 In bacteria, proteins involved in resistance to mercury are encoded by mercury responsive transcription factor; merR, an inorganic-mercury reductase; merA, alternative transcription factor; merD, periplasmic transporters; merP, merT and merC, and organomercurial lyase; merB. These genes are organized as mer operon in most of the bacterial species reported to date (Barkay et al., 2003).
Comparison of the organization of mer genes from a diverse range of bacterial species has revealed considerable similarity in their genetic organization ( Figure A2). Almost all these operons contain a regulatory gene, merR. In gram-positive bacteria, the mer genes including merR form a single operon. Downstream to the regulator, merR other mer structural genes merA, merD, merP, and merT are located in gram-positive bacteria. In some gram-positive bacteria, an additional ORF, merC, is also part of the operon. In some gram-negative bacteria, merR is divergently located to other mer structural genes which form an operon without merR. In certain bacterial species, such as B. megaterium and Pseudomonas sp. K62, an organo-mercurial lyase, merB was also reported as a part of the mer-operon (Kiyono, Omura, Inuzuka, Fujimori, & Pan-Hou, 1997;Wang et al., 1989). In some bacterial genomes, more than one copy of the structural gene for merA, merB and merR were reported (Khaleque, 2017;Kiyono et al., 1997;Wanget al., 1989). In Synechocystis, only merA and merR genes exist and were distantly located in the genome and all other mer genes, such as merC, merD, merP, merT, merG, merE, and merB were not detected. It is important to note that not only in Synechocystis, but also in all other cyanobacterial species reported to date, merR and merA are distantly located in their genomes unlike in other bacterial species ( Figure A2). However, the mechanism of gene regulation in response to mercury metal toxicity has not been reported in any cyanobacteria. Since homologs of the Slr0701 and Slr1849 appear to be well conserved among different aquatic cyanobacteria and relatively more similar to the metal-reducing bacterial species, it appears likely that the Slr0701 might be a regulatory gene in metal stress response.

| Complete targeted inactivation of a gene, slr0701 coding for putative MerR
To elucidate the regulatory role of Slr0701, the ORF was inactivated as shown in the schematic representation ( Figure 1a). The extent of replacement of wild-type copies of the slr0701 with that of disrupted copies of slr0701::sp r was confirmed by comparing the sizes of the amplified PCR products. When the genomic DNA from wild-type was used as a template for PCR amplification with specific primers (slr0701-F and slr0701-R), a PCR product of 1,340 bp representing the slr0701 ORF was amplified including the upstream and downstream regions. In contrast, when the genomic DNA isolated from slr0701-mutant were used as a template with the same set of primers, a 3,423 bp PCR product was amplified corresponding to the wild-type fragment of 1,340 bp including the inserted Ω-spectinomycin gene (sp r ) cassette of 2083 bp (Figure 1b).

| slr0701-mutant is more sensitive to HgCl 2 treatment compared to wild-type
Since, slr0701 codes for a putative MerR type transcriptional regulator, we investigated the effect of trace metals on the viability of slr0701-mutant cells in comparison with wild-type. Initially, to find the involvement of Slr0701 in metal resistance, we tested the growth of wild-type and slr0701-mutant strain in which slr0701 gene was inactivated on a solid BG-11 media containing various metals as described in experimental procedures. Both wild-type and slr0701mutant cells revealed similar profiles of growth at different metal ions tested except the growth on the BG-11 solid plate containing HgCl 2 (data not shown). In order to confirm that the slow growth phenotype seen in slr0701-mutant was due to HgCl 2 , we compared the growth profiles of wild-type, slr0701-mutant and slr0701 + -complemented cells in the absence and presence of HgCl 2 (Figure 2). All the three strains exhibited similar growth profiles when serially diluted cultures were spotted on a BG-11 agar plate without HgCl 2 .
However, with increasing concentrations of HgCl 2 in the BG-11 solid agar plate, slr0701 mutant exhibited a slow growth phenotype compared to wild-type. At 500 nM of mercuric chloride, almost no growth of the mutant cells was observed in the spot area, at which 5-times diluted cells were deposited (Figure 2). The complementation of slr0701-mutant cells with a functional Slr0701 expressed from pVZ321-slr0701 restored growth, and appeared similar to that of wild-type growth in the presence of mercuric chloride. We also tested the effect of HgCl 2 using the liquid cultures. We observed a significant difference in the growth profiles of wild-type and slr0701mutant cultures, when incubated with 500 nM HgCl 2 . However, at 750 nM concentration both the strains exhibited similar growth phenotypes ( Figure A3). This indicated that mercury-induced expression of slr0701 gene seems to be crucial for Synechocystis cells to tolerate inorganic mercury. The functional complementation of slr0701-mutant cells by the slr0701 gene explains that the slow growth phenotype observed in the presence of mercury was due to the inactivation of slr0701.

| Effect of HgCl 2 on photosynthetic activity due to mutation in slr0701
We measured the PSII activity in the wild-type and slr0701-mutant cells to analyze the extent of damage caused by mercury on photosynthetic performance due to inactivation of slr0701 gene. We used both oxygraph as well as relative Chl a fluorescence kinetics to analyze PSII activity. At optimal growth conditions, no significant difference in the PSII activity was observed (from water to PBQ) between wild-type and slr0701-mutant cells (Figure 3a). The PSII activity was  (Lu & Vonshak, 2002). Such inhibition of electron flow, that is, reduced PSII activity due to incubation of cells with HgCl 2 was much higher in the slr0701-mutant than wildtype cells. Data from both oxygraph and Chl a fluorescence kinetics are consistent and a rapid decline in PSII activity under HgCl 2 stress in slr0701-mutant clearly indicates severe damage in the photosynthetic machinery as compared to wild-type cells.

| Inactivation of slr0701 changes the mercuryinduced expression mer genes
The slr0701

| Slr0701 binds to upstream of slr0701 and slr1849
As the expression of distantly located slr0701 and slr1849 genes were affected by the slr0701 inactivation, a common cis-regulatory binding element for Slr0701 protein is expected in the upstream of these genes. Using MEME motif discovery tool, we identified a common conserved cis-regulatory element not only in the upstream of slr1849 and slr0701 genes, but also in the upstream of their homologs from other cyanobacterial species (Figure 6a).
F I G U R E 3 Changes in the photosystem II activity in Synechocystis wild-type and slr0701 mutant cells during incubation with HgCl 2 . Synechocystis wild-type and slr0701 mutant cells were grown to 0.6 at OD 730nm and then added 500 nM HgCl 2 . Collected cells before addition and 12 and 24 hr after addition of HgCl 2 for measuring PSII activity. with Slr0701 protein was studied using EMSA. Purified 6x-His-Slr0701 protein retarded the electrophoretic mobility of the DNA fragments having P slr0701 and P slr1849 cis-acting elements, and intensities of retarded DNA bands were increased with the amount of protein used for incubation (Figure 6b,c). Importantly, in the presence of 50 µM of HgCl 2 , the intensity of retarded DNA-protein complex was more compared to the intensity of the same in the absence of HgCl 2 . This result indicated that Hg 2+ enhances the binding of Slr0701 to its target DNA binding site. In order to confirm the specificity of binding to slr1849 and slr0701, we used an unrelated DNA upstream (sll1920) in the presence and in the absence of HgCl 2 ( Figure A4). The 6x-His-Slr0701 did not bind to the unrelated-DNA fragment confirming its specificity to the upstream of slr1849 and slr0701.

| Binding kinetics of Slr0701 in the upstream of slr0701
We further analyzed the interaction between the upstream of slr0701 and Slr0701 protein using SPR, as there was a significant increase in Slr0701 binding to P slr0701 in the presence of HgCl 2 . We performed a binding analysis with varying concentrations of Slr0701 protein against the immobilized P slr0701 DNA (Figure 7a). When the titration was repeated in the presence of 62 µM mercuric chloride, there was an increase in the response (RU). It is clear that association rate (k a ) of the Slr0701 with immobilized DNA fragment was faster and the dissociation rate (k d ) was observed to be slower in the presence of HgCl 2 (Figure 7b) . The equilibrium dissociation constant (K D ) evaluated from the kinetic traces showed that Slr0701 has nearly 1,000 times greater affinity to the promoter in the presence of HgCl 2 (Figure 7c). This complements with the EMSA results where mercury enhanced the binding between the Slr0701 protein and upstream of slr0701 (Figure 6b).

| HgCl 2 induces transcription of mercury responsive genes
We monitored the activities of P slr0701 and P slr1849 promoters in the wild-type and slr0701-mutant cells in the presence and the absence of HgCl 2, in order to examine the role of HgCl 2 and Slr0701 in transcriptional activation of merA. The P slr0701 and P slr1849 promoters were independently fused to a luxAB reporter gene and introduced into both Synechocystis wild-type and slr0701-mutant and slr1849 including their cyanobacterial homologs using MEME suite (http://meme.sdsc.edu/meme/cgi-bin/meme.cgi). Conserved inverted repeat region is highlighted in the alignment. Consensus representation of the inverted repeat is shown as a Logo below the alignment. Binding of Slr0701 protein to the upstream DNA fragment of slr0701 (b) and slr1849 (c) ORFs. A gel mobility shift assay was performed with 0.5 and 1 µg of 6x-His-Slr0701 and upstream DNA fragments covering the cis-acting element were used in binding reaction. A 231 bp DNA upstream of slr0701 starting from −1 to −231 bp with respect to the translation start site was used in (b). A 230 bp DNA upstream of slr1849 starting from −1 to −230 bp with respect to the translation start site was used in (c) cells. These modified strains were used for monitoring activities of the promoters. A weak luminescence signal was detected in the wild-type cells for P slr0701 and also for P slr1849 promoters even in the absence of HgCl 2 indicated their constitutive expression ( Figure 8a). This result is consistent with previous reports where merR promoter found to be a weak promoter (Lund & Brown, 1989).
The P slr0701 -luxAB in the wild-type and in slr0701-mutant showed 4.3 ± 2.4 and 1.3 ± 0.1 relative luminescence units (RLU) prior to the treatment with HgCl 2. The P slr1849 -luxAB showed 4 ± 1.3 and 0.6 ± 0.3 RLUs prior to HgCl 2 treatment indicating that the basal activity of P slr0701 and P slr1849 promoters even in the absence of mercury. In the wild-type cells, after the addition of HgCl 2 there was an increase in luminescence signal suggesting that HgCl 2 enhances the expression of mer genes in Synechocystis (Figure 8a).
Upon addition of HgCl 2 to wild-type cells, within 30 min the relative luminescence units were increased by four-fold (22.2 ± 2.0) and three-fold (10.3 ± 1.3) for P slr0701 and P slr1849 , respectively indicating that both the promoters could get activated immediately in the presence of HgCl 2 (Figure 8a). However, P slr1849 activity was gradually increased as compared to the P slr0701 , which is consistent with the qRT-PCR results (see Figure 4a).
In contrast, the slr0701-mutant cells harbouring plasmid DNA constructs having either P slr0701 or P slr1849 promoter fused to luxAB genes did not show any luminescence signal either in the absence or in the presence of HgCl 2 suggests that Slr0701 protein is necessary for mercury-induced activation of its own gene expression as well as slr1849 transcription (Figure 8b).

| Circular dichroism analysis of structural changes in Slr0701
The structural changes in Slr0701 upon interaction with HgCl 2 were analyzed using CD. In the far-UV region (200-250 nm), a gradual decrease in negative ellipticity was observed, when the protein Slr0701 was titrated with increasing concentration of HgCl 2 . This suggests the relaxation in the secondary structure of the protein upon binding to HgCl 2 (Figure 9a) . The ellipticity changes followed at 222 nm upon addition of HgCl 2 is presented in Figure 9b. The initial changes saturating at 100 µM of HgCl 2 might be attributed to the binding of HgCl 2 whereas further sharp decline in the negative ellipticity could arise from denaturation of the protein.
In addition, the tertiary structural change in Slr0701 protein was analyzed in the near-UV region (250-300 nm) upon binding with P slr0701 with and without HgCl 2 . As compared to DNA-Protein alone, there was a significant increase in ellipticity in the presence of HgCl 2 suggesting stronger binding of the protein with DNA in the presence of HgCl 2 (Figure 9c). (c) Near UV-CD spectra of Slr0701, P slr0701 , P slr0701 +Slr0701, and P slr0701 +Slr0701+HgCl 2 F I G U R E 1 0 Schematic representation of mercury-induced merA gene regulatory mechanism by a putative transcriptional regulator, Slr0701 in Synechocystis sp. PCC6803. (a) Slr0701 and Slr1849 genes are located far apart from each other in Synechocystis genome. Slr0701 is constitutively expressed and regulates its own gene expression. Slr0701 is transcribed from a weak promoter. (b) When cyanobacterial cells experience inorganic Hg 2+ , Slr0701-Hg 2+ complex binds to cis-regulatory element located upstreams of its own ORF as well as slr1849 with greater affinity and leads to an induced expression. Thus Hg 2+ bound Slr0701 activates the mercury-dependent expression of Slr1849. Slr1849 being an inorganic-mercury reductase converts Hg 2+ to Hg 0 and the volatile mercury is sent out of the cell us to elucidate the regulatory mechanism of mercury detoxification in Synechocystis.
We inactivated the slr0701 that code for a putative MerR transcriptional regulator (Figure 1). Mercury is known to damage photosynthesis in cyanobacteria and higher plants (Bernier, Popovic, & Carpentier, 1993;Tangahu et al., 2011). We analyzed the extent of damage caused by HgCl 2 on photosystem II activity. The slr0701-mutant strain showed a severe reduction in photosynthetic activity and became relatively more sensitive to HgCl 2 than wild-type cells emphasizing its role in Hg 2+ detoxification ( Figures   2 and 3). Rapid and gradual upregulation of slr0701 and slr1849 mRNA levels were observed in the presence of HgCl 2 (Figure 4).
This induced gene expression was observed to be due to the binding of Slr0701 protein to the cis-regulatory element in the upstream of its own ORF and slr1849 followed by transcriptional activation (Figures 5, 6 and 10). Moreover, the cis-regulatory element is found to be well conserved among most cyanobacterial mer homologs. In the presence of HgCl 2 , Slr0701 protein showed faster association and slower dissociation constants to the target DNA binding element (Figures 7 and 10). The slr0701 and slr1849 are constitutively expressed. Upon addition of HgCl 2 , mercurybound Slr0701 enhances the transcription of these genes ( Figures   8 and 10). In addition HgCl 2 induced conformational changes in the protein and enhanced affinity to DNA leads to upregulation of gene expression (Figures 9 and 10). In the slr0701-mutant, the mercury-induced expression of mercury reductase, merA was not observed due to inactivation of slr0701 gene. Hence, the slr0701-mutant became more sensitive to Hg 2+ than wild-type.
Cyanobacteria contain only a well-conserved transcriptional regulator, MerR and a mercury reductase, MerA. They are found to be distantly located in the genome which is a unique feature when compared to other bacterial phyla. However, MerR could still regulate the mercury-dependent expression of MerA. The mechanism of gene regulation involving MerR is well-conserved among all bacterial phyla.

ACK N OWLED G EM ENTS
We gratefully acknowledge the Department of Science and

CO N FLI C T O F I NTE R E S T S
None declared.

AUTH O R S CO NTR I B UTI O N
D.K.S performed majority of the experiments and wrote the paper.
L.B generated the mutant strain. K.T.N and D.K.S performed CD spectral experiments together. N.P.P analyzed the CD and SPR results and edited the paper. J.S.S.P conceived the idea for the project and wrote the paper with D.K.S.

E TH I C S S TATEM ENT
None required.

DATA ACCE SS I B I LIT Y
The data that support the findings of this study are available from the corresponding author upon reasonable request.

R E FE R E N C E S APPENDIX
F I G U R E A 1 Phylogenetic relationship of Slr0701 and Slr1849 with their closely related orthologs. The complete amino acids sequences of Slr0701 and Slr1849 and their orthologs were taken, phylogenetic relationship was deduced. All investigated cyanobacterial and bacterial strains are presented in the figure. The well-studied Bacillus megaterium, MerR and MerA with five other proteins from different organisms more closely related to these Mer proteins were also included. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (500 replicates) is shown next to the branches. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances are given in the number of amino acid substitutions per site as indicated by the scale bar representing 0.7 for slr0701 (a) and 0.3 for slr1849 (b) substitutions per amino acid F I G U R E A 2 Comparison of mer-operon organization in different bacterial and cyanobacterial species. merR ( ), Transcprional regulator; merG ( ), merT( ), merP( ), merC( ), merE( ), periplasmic proteins; merA( ), mercury reductase; merB ( ), organo-mercurial lyase and merD ( ) alternative transcriptional regulator. The reading frames are shown as arrows to indicate ORFs. The mer genes shown below the bold dash line belong to cyanobacterial species. ( ) indicates that the genes are located distantly from each other in the genome F I G U R E A 3 Effect of HgCl 2 on growth of Synechocystis wildtype and slr0701-mutant strains. For monitoring the growth in the presence of HgCl 2, cultures were initially allowed to grow till 16 hr as described in methods, and then HgCl 2 was added to a final concentration of 250 nm, 500 nm and 750 nm to the respective culture tubes. Optical density at 730 nm was measured at regular intervals. Filled symbols represent growth of wild-type Synechocystis cultures and open symbols represent slr0701-mutant cultures. The mean of three experimental values is presented with standard error F I G U R E A 4 Binding of Slr0701 protein to the upstream DNA fragment of sll1920 ORF. A gel mobility shift assay was performed with 0.5 and 1 µg of 6x-His-Slr0701 protein and 80 nmoles of upstream DNA fragment of sll1920 in a binding reaction. A 201 bp DNA upstream of sll1920 starting from −1 to −201 bp with respect to the translation start site was used in the binding reaction