Methanosarcina acetivorans C2A Topoisomerase IIIα, an Archaeal Enzyme with Promiscuity in Divalent Cation Dependence

Topoisomerases play a fundamental role in genome stability, DNA replication and repair. As a result, topoisomerases have served as therapeutic targets of interest in Eukarya and Bacteria, two of the three domains of life. Since members of Archaea, the third domain of life, have not been implicated in any diseased state to-date, there is a paucity of data on archaeal topoisomerases. Here we report Methanosarcina acetivorans TopoIIIα (MacTopoIIIα) as the first biochemically characterized mesophilic archaeal topoisomerase. Maximal activity for MacTopoIIIα was elicited at 30–35°C and 100 mM NaCl. As little as 10 fmol of the enzyme initiated DNA relaxation, and NaCl concentrations above 250 mM inhibited this activity. The present study also provides the first evidence that a type IA Topoisomerase has activity in the presence of all divalent cations tested (Mg2+, Ca2+, Sr2+, Ba2+, Mn2+, Fe2+, Co2+, Ni2+, Cu2+, Zn2+ and Cd2+). Activity profiles were, however, specific to each metal. Known type I (ssDNA and camptothecin) and type II (etoposide, novobiocin and nalidixic acid) inhibitors with different mechanisms of action were used to demonstrate that MacTopoIIIα is a type IA topoisomerase. Alignment of MacTopoIIIα with characterized topoisomerases identified Y317 as the putative catalytic residue, and a Y317F mutation ablated DNA relaxation activity, demonstrating that Y317 is essential for catalysis. As the role of Domain V (C-terminal domain) is unclear, MacTopoIIIα was aligned with the canonical E. coli TopoI 67 kDa fragment in order to construct an N-terminal (1–586) and a C-terminal (587–752) fragment for analysis. Activity could neither be elicited from the fragments individually nor reconstituted from a mixture of the fragments, suggesting that native folding is impaired when the two fragments are expressed separately. Evidence that each of the split domains plays a role in Zn2+ binding of the enzyme is also provided.


Introduction
Topoisomerases manage DNA supercoiling and are classified according to the mechanism of action employed [1][2][3][4]. In general, topoisomerases form a phosphotyrosine intermediate while cleaving one strand (type I or odd-numbered) or two strands (type II or even-numbered) of DNA. Before the initial transesterification is reversed and a ligated DNA backbone regenerated, type I and II enzymes change the DNA linking number (Lk) by 1 and 2, respectively. Type I topoisomerases are ascribed further distinction based on the polarity of the covalent intermediate formed. Type IA (like Type II) forms a 59-phosphotyrosine adduct while type IB forms 39-phosphotyrosine covalent adduct. Thus, the structure and mechanism of each class are distinct.
The characterization of the first topoisomerase E. coli TopoI (EcoTopoI), originally titled v protein, was an earnest exploration into a field that has grown to be of burgeoning biological importance [5]. Since then, topoisomerases have been demon-strated to be vital components of the cellular machinery in a wide array of processes that are essential for life, such as DNA replication, DNA repair and chromosome segregation [6][7][8]. While these molecular machines have been identified in every genome sequenced to date, irrespective of domain, eukaryotic and bacterial topoisomerases have received greater scrutiny given their role in diseased states such as Cancer and infections [9,10].
Interestingly, studies from bacteria and both lower and higher eukaryotes have demonstrated that ablation of TopoIII (a subfamily of type IA) activity in vivo without some compensatory mechanism leads to genomic instability and/or abnormal growth [11][12][13][14][15][16]. In spite of this direct evidence that TopoIII is vital for normal development, the majority of archaeal studies have focused on the non-essential Reverse Gyrase (a subfamily of type IA) given its unique phylogenetic distribution in thermophiles and hyperthermophiles [17][18][19][20]. Furthermore, the only data published on archaeal TopoIII are from hyperthermophilic organisms of the archaeal subdomain Crenarchaeota [21][22][23].
The archaeal subdomain Euryarchaeota contains a large group of economically important organisms, including both the mesophilic and hyperthermophilic methane-producing organisms. The methane-producing genus Methanosarcina is known to harness all known substrates (including acetate, H 2 and CO 2 , formate, methylamine and methanol) to produce methane, a greenhouse gas [24]. Recent reports also demonstrate that methanogens may be important in human health, especially from the nutritional standpoint [25,26]. With the largest sequenced archaeal genome, Methanosarcina acetivorans C2A provides a fertile ground for studies on archaeal DNA metabolism. This organism has four putative topoisomerases ( [24]. Based on sequence homology, MacTopoIIIa is classified as a type IA topoisomerase III protein. In the present report, we characterize MacTopoIIIa, as the first mesophilic archaeal TopoIII studied. The attributes shared with orthologs from the other domains and those potentially specific to Archaea are discussed.

Expression and purification of MacTopoIIIa
In order to characterize MacTopoIIIa, the encoding gene was amplified, cloned, expressed, and purified from E. coli cells as a fusion protein with an N-terminal hexa-Histidine (6-His) tag. Purification of the recombinant protein was achieved through affinity chromatography, ion-exchange chromatography and sizeexclusion chromatography. Based on the polypeptide sequence, the recombinant protein has an estimated molecular mass of ,86 kDa. As expected, the highly purified MacTopoIIIa migrated between the 66 and 116 kDa molecular mass markers via a 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) according to the method of Laemmli (Fig. 1A, lane 2) [27].

Catalytic properties of MacTopoIIIa
The putative topoisomerase was tested for the ability to relax negatively supercoiled pUC18 (300 ng). As little as 10 fmol of enzyme initiated relaxation upon addition to the reaction mixture ( Fig. 2A). Whereas 200 fmol of enzyme showed modest activity, 5 pmol of enzyme elicited maximal activity under standard assay conditions. Analysis of aliquots of samples collected from the last step (gel filtration column) demonstrated similar activities in all fractions.
Maximal relaxation activity was observed around 30uC, and at temperatures of 50uC and above, no relaxation was observed (Fig. 2B). M. acetivorans C2A is an anaerobic mesophile that grows optimally at 35-40uC. Therefore, the optimal temperature for activity of MacTopoIIIa falls within the expected range of activity [28]. Both Na+ and Mg2+ are required for growth of this archaeon, and the optimal concentration of NaCl required was 200 mM. In the present study, at NaCl concentrations of 400 mM or higher, the activity of MacTopoIIIa was almost not observable (Fig. 2C). A similar observation was made for the effect of KCl concentration on MacTopoIIIa activity. Initiation of activity by the enzyme was observable as early as 15 seconds after addition of MacTopoIIIa to the reaction mixture, and maximal activity was observed within 5 min (Fig. 2D).

The effects of different divalent cations on MacTopoIIIa activity
Topoisomerase activity was assessed as a function of divalent cation concentration in a semi-logarithmic range of concentrations from 50 mM to 10 mM (Fig. 3). We observed that each divalent cation elicited a unique activity profile. As in previous studies on topoisomerases, Mg 2+ was able to elicit maximal activity at concentrations greater than or equal to 1 mM. Ca 2+ was the only other cation that maintained a profile comparable to Mg 2+ , with concentrations greater than 1 mM showing near maximal relaxation. In fact, relaxation activity of MacTopoIIIa at concentrations above 1 mM appeared to be higher for Ca 2+ compared with Mg 2+ . Ba 2+ and Sr 2+ profiles were similar, but were not able to relax the supercoiled substrate as effectively as Mg 2+ and Ca 2+ . The ability of these Group II alkaline metals to elicit DNA relaxation activity at low cation concentration was, therefore, as follows: Mg 2+ . Ca 2+ .. Ba 2+ . Sr 2+ .
In the presence of Zn2+, activity was observed in a narrow window of 100 mM. Zn2+ concentrations of 50 mM or less and 500 mM or more elicited no activity even when the reaction was performed in a 1:1 enzyme:DNA molar ratio (data not shown). To a lesser extent, Cd2+ was also able to elicit activity within a similar narrow range. Relaxation activity was observed at 0.1 mM and 0.5 mM concentrations of Co2+, and at a concentration of 1 mM or higher, either no activity was detectable or divalent cationcatalyzed DNA degradation was observed. Fe2+ required a concentration of 1 mM or higher for activity. However, DNA degradation was also significant at concentrations above 1 mM. These results are consistent with metal catalyzed DNA degradation seen with certain transition metals that are capable of promoting the Haber-Weiss reaction [29,30]. Relaxation activity in the presence of Mn2+ or Ni2+ or Cu2+ was also elicited under a narrow range (100 mM -1 mM).

Classification of MacTopoIIIa
To determine the class of MacTopoIIIa in the topoisomerase family, its response to known topoisomerase inhibitors, with different mechanisms of action, was investigated. Table 1 shows a summary of the inhibitors tested. Ethidium bromide (EtBr) and M13 ssDNA inhibited activity. The type I prokaryotic inhibitor spermidine and the type I eukaryotic inhibitor camptothecin were unable to inhibit the activity of MacTopoIIIa. Furthermore, the type II inhibitors (etoposide, nalidixic acid and novobiocin) also had no effect on the relaxation activity of MacTopoIIIa. Relaxation activity was, however, inhibited by high concentrations of KCl and NaCl (.400 mM). Based also on preliminary analytical gel filtration experiments, MacTopoIIIa was estimated to exist as a monomer in solution, which is consistent with the oligomerization state of type 1A topoisomerases [3].

Catalytic Tyrosine of MacTopoIIIa is Y317
A sequence alignment using Sulfolobus solfataricus topoisomerase III (SsoTopoIII) and the entire archaeal TopoIII subfamily II (of which MacTopoIIIa is a member) was created to provide a hint at the potential catalytic tyrosine (Fig. 1C), because SsoTopoIII is the only archaeal TopoIII with a characterized active site tyrosine [9,22]. The alignment suggested Tyr317 as the potential active site residue of interest. Site-directed mutagenesis was used to create a Y317F mutant. Size exclusion chromatography of MacTopoIIIa Y317F yielded similar elution volume to that of the wild-type protein, suggesting that the mutant also exists as a monomer in solution (data not shown). The purified wild-type and mutant proteins (Fig. 1A, lanes 2 and 3) were subjected to circular dichroism scan to determine whether the mutation grossly impacted the structure of MacTopoIIIa Y317F (Fig. 1C). No gross differences between the structures could be discerned, suggesting that the secondary structures of both enzymes are comparable, outside of the point mutation. By subjecting the  . The GenBank accession numbers are in brackets. The conserved and similar amino acids are shaded black and gray, respectively. The putative catalytic residue is shaded red. (D) Circular Dichroism (CD) spectra of purified MacTopoIIIa wild-type and the Y317F mutant. Triplicate data sets were collected from samples at a concentration of 0.5 mg/ml in a buffer containing 50 mM Tris-HCl pH 7.5, 200 mM NaCl, 2 mM DTT. All data sets were normalized against baseline readings from buffer containing no protein. doi:10.1371/journal.pone.0026903.g001 purified mutant protein to the relaxation reaction, it was revealed that MacTopoIIIa Y317F was unable to relax the negatively supercoiled substrate under the same conditions where the wildtype protein demonstrated activity (Fig. 4A). The products seen in Lane 5 are likely to be due to iron catalyzed DNA degradation via the Haber-Weiss reaction (refer to Fig. 3, last lane under Fe2+).

Truncation mutants
Two truncations of MacTopoIIIa were constructed based on sequence alignment with the 67 kDa fragment of EcoTopoI [31]. Based on this sequence alignment, the MacTopoIIIa sequence was spliced at a position between two small aliphatic residues (A586 and I587) where no secondary structure was apparent based on secondary structure prediction (PredictProtein: http://www. predictprotein.org/). We constructed two fragments, N1-586 and C587-752, containing the N-terminal 586 amino acid residues and the C-terminal 166 amino acid residues, respectively. Both truncations were cloned, overexpressed and highly purified. Neither the N1-586 nor the C587-752 fragment was able to relax the same substrate as the wild-type under standard assay conditions ( Fig. 4B and 4C). To determine whether relaxation activity could be reconstituted by mixing the two fragments, they were incubated briefly prior to addition to the reaction mixture at varying ratios (50:1, 5:1 and 1:1). However, no relaxation activity was observed (Fig. 4C). To determine whether higher concentrations of the two fragments will elicit topoisomerase activity, we examined concentrations up to five times the maximum concentration used for the wild-type protein; however, no

Zinc Content Measurements
In an effort to assess the Zn 2+ binding ability of this enzyme, samples of wild-type and the two truncation mutants were subjected to ICP-MS analysis. The wild-type enzyme was found to contain 1.2960.01 mol of Zn 2+ . The MacTopoIIIa N1-586 truncation was found to contain 0.6260.02 mol, and the C587-752 truncation mutant was found to contain 0.4960.01 mol of Zn 2+ (Fig. 5).

MacTopoIIIa is promiscuous
A fundamental property of type IA topoisomerases is the absolute requirement of Mg 2+ for the large conformational change required during DNA relaxation, but other divalent cations can substitute [5,32]. Recent publications have proposed a two metal mechanism for type IA and type II topoisomerases wherein one may bind acidic residues within the TOPRIM motif causing the necessary large conformational changes while the other may play a transient role in stabilizing the transition state in preparation for  attack on the scissile phosphate [33,34]. Fluorescence and ICP-MS measurements on E. coli TopoI (EcoTopoI) reveal that one Mg 2+ plays a role influencing the overall structure of the enzyme, while a second Mg 2+ is required to achieve full catalytic activity [35]. Mutation of three acidic residues of EcoTopoI within the TOPRIM domain only partially inhibited activity [36]. But even when crystals were soaked in MgCl 2 , no Mg 2+ was detected in the crystal structures of EcoTopoIII whether the wild-type was covalently linked or whether the wild-type or mutant were primed for attack at a scissile bond of ssDNA in a manner akin to S. cerevisiae TopoII (SceTopoII) [37,38]. The data in total suggests that two transient divalent cations are required for maximal activity with one divalent cation binding in order to prime the enzymatic structure for large conformational changes while the second cation primes the phosphate linkage of interest for attack before immediate removal from the active site.
And while the role of Mg 2+ is usually well examined, it is commonplace for other divalent cations to receive less scrutiny ( Table 2). As one example, EcoTopoI DNA relaxation activity was investigated in the presence of 0-10 mM Mg 2+ while Ca 2+ , Co 2+ , Mn 2+ and Zn 2+ were investigated at 2 mM only [39]. In another study, while the activity of human topoisomerase IIIa (hTopoIIIa) was again tested in the presence of a range of MgCl 2 concentrations, hTopoIIIa was deemed inactive in the presence of other divalent cations based on one concentration (5 mM) [40]. We demonstrate for the first time that MacTopoIIIa is promiscuous with regard to divalent cation (Ba 2+ , Ca 2+ , Cd 2+ , Co 2+ , Cu 2+ , Fe 2+ , Mg 2+ , Mn 2+ , Ni 2+ , Zn 2+ ) and that each elicits a unique activity profile. Unlike results with other topoisomerases, we observe DNA relaxation activity in the presence of Zn 2+ albeit under a narrow range [22,32]. These results are not plasmidspecific as confirmed in the presence of 300 ng of either pUC18 or pGEM-T (data not shown). While further studies are needed to determine if all divalent cations may substitute in the mechanism for all topoisomerases, our data clearly demonstrates that there is a need to assess bivalent metal ion profiles and not simply examine activity at a singular concentration. As a result, it is possible that different divalent cations not previously reported to elicit activity for any given topoisomerase indeed do elicit activity, although it may be under a narrow range of concentrations. It also remains to be seen whether these divalent cation profiles can act as a tool to distinguish topoisomerases or whether this activity is specific to Archaea or to this particular archaeon in the present study.

MacTopoIIIa is a type IA topoisomerase
Nalidixic acid, etoposide and novobiocin are specific inhibitors of type II topoisomerases with different mechanisms of action [10]. Camptothecin is a known eukaryotic type I inhibitor that stabilizes the binary protein-DNA complex intermediate thereby impeding DNA relaxation [41]. None of these inhibitors were able to inhibit the DNA relaxation activity of MacTopoIIIa. Ethidium bromide induces positive supercoiling [42]. Type IA topoisomerases have a preference for ssDNA and preferentially relax negatively supercoiled DNA. Ethidium bromide and M13 ssDNA were each able to inhibit activity. Thus, the sequence homology in combination with the results from the inhibitor studies clearly demonstrates that MacTopoIIIa is a type IA topoisomerase.  Table 2. Summary of divalent cation studies.

Enzyme
Mg Spermidine had no effect on MacTopoIIIa at concentrations up to 100 mM. Spermidine was initially posited as a prokaryotic polyamine topoisomerase inhibitor given its inhibitory effect on both EcoTopoI and EcoTopoIII at physiological concentrations [43,44]. However, the data actually suggests that it is a case by case scenario within the domain Bacteria and the only archaeal TopoIII tested to-date, SsoTopoIII, demonstrated no susceptibility [22,[45][46][47][48][49]. The data reveal that there is no definite trend for polyamine susceptibility amongst bacterial type IA topoisomerases and no demonstrated effect to-date in the domain Archaea.

Catalytic tyrosine is located at Y317
We assayed the Y317F mutant of MacTopoIIIa for activity under conditions that were demonstrated to elicit relaxation activity for wild-type. After analyzing the CD spectra of MacTopoIIIa Y317F to ensure that there were no gross structural differences compared to the wild-type, we demonstrated that MacTopoIIIa Y317F was inactive under conditions where wildtype had activity. This result confirmed our prediction that Tyr317 is indeed essential in the mechanism of DNA relaxation by MacTopoIIIa.

N-terminal Zn 2+ -binding activity
An open question about TopoI and TopoIII is the role that the C-terminal (Domain V) plays in the context of Zn 2+ -binding. While lacking the C-terminal (Domain V), the EcoTopoI fragment (Domains I-IV) retained the ability to cleave oligonucleotides, but was not able to relax negatively supercoiled DNA [31]. With three Zn 2+ -binding sites in Domain V (the domain absent in the original crystal structure), methyl methanethiosulfonate (MMTS)-treated EcoTopoI abolished Zn 2+ -binding and rendered the enzyme incapable of catalyzing DNA relaxation [32]. Meanwhile, a double mutant (Cys559A/Cys561A) or truncation that abolished Zn 2+binding in Thermotoga maritima TopoI (TmTopoI) only minimally affected DNA relaxation activity [4]. No TopoI crystal structures have been identified with Zn 2+ positioned even when it was detected in the mother crystal liquor [50]. The story is further confounded by the fact that there are several bacterial type IA topoisomerases that contain no homologous Zn 2+ -binding motif or Zn 2+ -binding ability [37,48,51]. A reasonable conclusion from the current data available is that type IA topoisomerase Zn 2+ -binding motifs, when present in a given bacterial organism, play a structural role that may or may not be essential for DNA relaxation activity.
Given that MacTopoIIIa contains a C-terminal Cys-X 2 -Cys-X 17 -Cys-X 5 -Cys motif and that there have not been any studies to date examining the role of Zn 2+ -binding in archaeal TopoIII, we set out to determine whether MacTopoIIIa contains Zn 2+ in order to delineate any potential differences with the well-characterized bacterial topoisomerases. The results demonstrated that MacTo-poIIIa binds to approximately one mole of Zn 2+ per mole of protein. We then constructed, expressed and purified N-terminal and C-terminal fragments to assess the Zn 2+ -binding ability of each segment. To our surprise, both the C-terminal fragment and the N-terminal fragment demonstrated the ability to chelate approximately a half mole of Zn 2+ . To our knowledge, this is the first report of an N-terminal fragment of a TopoIII or TopoI having this ability to coordinate Zn 2+ . It is likely that residues in the two fragments coordinate Zn 2+ in the wild-type. Mutational studies are currently underway to investigate the Zn 2+ -binding role that any of the eight His residues in the N-terminal and eleven Cys/His residues in the C-terminal may play.
In this report, MacTopoIIIa, a mesophilic archaeal TopoIII, was biochemically characterized in the context of well-established properties of eukaryotic and bacterial type IA topoisomerases. We demonstrate that MacTopoIIIa is a monomer in solution, binds Zn 2+ and exhibits susceptibility to inhibitors in a manner similar to other type IA topoisomerases. MacTopoIIIa is a distributive topoisomerase that has a high affinity for ssDNA and is inhibited by high salt concentrations. Properties that may be unique to this enzyme are the ability of all divalent cations tested to elicit DNA relaxation activity and the ability of this enzyme to bind Zn 2+ via the N-and C-terminal. Further studies are needed in the domain Archaea and beyond to determine if these attributes are domain or organism specific.

Materials and Methods
Cloning and expression of M. acetivorans C2A TopoIIIa (MacTopoIIIa) gene The gene for the putative topoisomerase, MacTopoIIIa (NP_617416), was amplified from the M. acetivorans C2A genomic DNA using PCR primers containing restriction sites for NdeI and XhoI in the forward (59-CATATGCACCTTATCGTAACG-GAAAAAAATATA-39) and reverse (59-CTCGAGCTATAGG TCCTCAACTATGCCGCCGTTACA-39) primers, respectively. The size of the PCR product was verified through agarose gel electrophoresis and cloned into pGEM-T, a TA-cloning vector (Qiagen). The recombinant plasmid was purified using a Qiagen gel extraction kit, and after confirming the integrity of the sequence of the DNA insert, the MacTopoIIIa gene was transferred into a modified pET28a expression vector. The pET28a vector has its kanamycin resistance gene replaced with an ampicillin resistance gene [52]. Therefore, the ligated products were transformed into E. coli JM109 cells and transformants were selected on lysogeny broth (LB) plates supplemented with ampicillin at a final concentration of 100 mg/mL. The plates were incubated overnight at 37uC, and a single colony was picked and cultured in 10 mL LB containing the same antibiotic and grown at 37uC for 8 hours. The plasmid was extracted from the E. coli culture and examined for the presence of the MacTopoIIIa gene through DNA sequencing. The recombinant plasmid was named pET28a/MacTopoIIIa. All DNA sequencing in the present report were carried out at the W. M. Keck Center for Comparative and Functional Genomics (University of Illinois at Urbana-Champaign).
The recombinant plasmid pET28/MacTopoIIIa was used in transforming E. coli BL21 codonplus RIPL cells (Stratagene) and plated on LB plates supplemented with ampicillin (100 mg/mL) and chloramphenicol (50 mg/mL). A transformant was cultured at 37uC in 10 ml LB medium supplemented with both antibiotics at the same concentrations stated above until the optical density (O.D.) at 600 nm reached 0.3. The expression of the MacTo-poIIIa gene was then induced by adding isopropyl b-Dthiogalactopyranoside (IPTG) at a final concentration of 1 mM. The temperature was decrease to 16uC and cell culturing was continued for 12-16 hours. The recombinant E. coli cells were then collected through centrifugation at 6500 rpm for 10 minutes.

Site-directed mutagenesis
Site-directed mutagenesis was carried out using the Quikchange site-directed mutagenesis kit (Stratagene) according to the instructions of the manufacturer. The PCR primer (59-AACGGG-TATATATCTTTTCCCAGGACCGACAAT-39) was utilized to convert MacTopoIIIa Tyr317 into Phe317. Mutagenized plasmid was transformed into JM109 cells after digestion of the parental plasmid with DpnI. The E. coli transformants were selected on LB agar plates supplemented with ampicillin [100 mg/mL] overnight at 37uC. Plasmids were extracted from individual colonies after growth in liquid LB cultures supplemented with ampicillin [100 mg/mL]. Plasmids were sequenced as described above, and the desired plasmid containing the Y317F mutation was selected for gene expression as described above for the wild-type gene.

Truncation mutagenesis
A truncated gene containing the DNA sequence for the Nterminal 586 amino acids (N1-586) was generated using an iProof TM High-Fidelity PCR kit with a forward primer (59-CATATGCACCTTATCGTAACGGAAAAAAATATA-39) and reverse primer (59-CTCGAGCTATGCCTGCAGGGACTC-TATGATTTTGTC-39) according to the manufacturer's instructions. A truncated gene containing the DNA sequence for the Cterminal 166 amino acids (C587-752) was generated in the same manner with the forward primer (59-CATATGGGTCTCAGG-GAAGACAAGATCATAGGCAAC-39) and the reverse primer (59-CTCGAGCTATAGGTCCTCAACTATGCCGCCGTTACA-39). The size of the PCR product was verified through agarose gel electrophoresis and cloned into the pGEM-T vector (Qiagen). Plasmids were transformed into JM109 cells after ligation. The E. coli transformants were selected on LB agar plates supplemented with ampicillin [100 mg/mL] overnight at 37uC. Plasmids were extracted from individual colonies after growth in liquid cultures supplemented with ampicillin [100 mg/mL]. Plasmids were sequenced as described above, and the correct inserts of N1-586 and C587-752 were transferred to the modified pET28a for expression as described above.

Purification of MacTopoIIIa and mutant proteins
For MacTopoIIIa wild-type protein and the Y317F mutant, the harvested recombinant E. coli cells were resuspended in Buffer A (50 mM Sodium phosphate, pH 7.0, 300 mM NaCl), and the cell contents were released by a French pressure cell (American Instruments Co). The cell debris was removed through centrifugation at 9500 rpm for 15 min at 4uC. The supernatant was subsequently filtered using a 0.22 mm filter and applied to a His-Trap TM HP column (GE Healthcare) pre-equilibrated with 90% buffer A and 10% buffer B (50 mM Sodium phosphate, pH 7.0, 300 mM NaCl, 500 mM imidazole), and eluted using a step-wise gradient of Buffer B.
Fractions containing the protein of interest were pooled and dialyzed against buffer A (50 mM Sodium phosphate, pH 7.0, 300 mM NaCl) overnight at 4uC. Samples were applied to a HiTrap TM SP column (GE Healthcare) pre-equilibrated with buffer A, and to elute the bound protein, a linear gradient using 100% buffer C (50 mM Sodium phosphate, pH 7.0, 1 M NaCl) was utilized. Concentrated fractions of each protein were directly applied to a size exclusion column (Superdex HR 200 HR 10/30 column) pre-equilibrated with buffer GF (50 mM sodium phosphate, 150 mM sodium chloride, pH 7.0), and the chromatography was developed with the same buffer. Highly purified proteins were dialyzed against a storage buffer (50 mM Tris-HCl pH 7.5, 200 mM NaCl, 2 mM DTT and 50% Glycerol) and stored at 280uC until used.
In the case of MacTopoIIIa C587-752 truncated mutant, the harvested recombinant E. coli cells were re-suspended in Buffer A, and the cell contents were released by a French pressure cell (American Instruments Co). The cell debris was removed through centrifugation at 9500 rpm for 15 min at 4uC. The supernatant was subsequently filtered using a 0.22 mm filter and applied to a His-Trap TM HP column (GE Healthcare) pre-equilibrated with 70% buffer A and 30% buffer B (50 mM Sodium phosphate, pH 7.0, 300 mM NaCl, 500 mM imidazole), and eluted using a linear gradient of buffer B.
The protein fractions from the HisTrap TM HP column were pooled and dialyzed against buffer A (50 mM Sodium phosphate, pH 7.0, 300 mM NaCl) overnight at 4uC. Samples were applied to a HiTrap TM SP column (GE Healthcare) pre-equilibrated with buffer A and eluted with a linear gradient of 100% buffer C. Highly purified fractions were dialyzed against the storage buffer and stored at 280uC until used.
In the case of MacTopoIIIa N1-586 truncated mutant, the harvested recombinant E. coli cells were re-suspended in Buffer A, and the cell contents were released by a French pressure cell (American Instruments Co). The cell debris was removed through centrifugation at 9500 rpm for 15 min at 4uC. The supernatant was subsequently filtered using a 0.22 mm filter and applied to a His-Trap TM HP column (GE Healthcare) pre-equilibrated with 75% buffer A and 25% buffer B (50 mM Sodium phosphate, pH 7.0, 300 mM NaCl, 500 mM imidazole), and eluted using a linear gradient of buffer B.
The fractions of the MacTopoIIIa N1-586 truncated mutant from the His-Trap column were pooled and dialyzed against buffer A overnight at 4uC. Samples were then applied to a HiTrap TM heparin HP column (GE Healthcare) pre-equilibrated with 90% buffer A and 10% buffer C. The MacTopoIIIa N1-586 truncated mutant was eluted with a step gradient using 75% buffer A and 25% buffer C. Highly purified fractions were dialyzed against the storage buffer and stored at 280uC until used. Aliquots of eluted fractions from all chromatographies were examined through SDS-PAGE.
Preparation of DNA substrates for topoisomerase assays E. coli JM109 cells were heat shocked to uptake the pUC18 plasmid, and transformants were cultured in ampicillin supplemented (100 mg/mL) LB medium to amplify the plasmid. The pUC18 plasmid was purified from cell pellets with a commercial kit (Qiagen), and electrophoresed on a 1.5% agarose gel containing 1X TBE (89 mM Tris, 89 mM Boric acid, 2 mM EDTA). Subsequently, the gel was stained with ethidium bromide and imaged with a UV illuminator. The plasmid DNA migrating the fastest was excised and purified utilizing a gel extraction kit (Qiagen) according to the manufacturer's specifications. The purified negatively supercoiled DNA was verified using a 2-Dimensional gel with EcoRI linearized pUC18 and ethidium bromide treated pUC18 as standards. The M13 ssDNA was from a commercial source (New England Biolabs).

DNA relaxation assay
Unless otherwise stated, the method is the same as previously described [22]. Briefly, reaction volumes contained 300 ng negatively supercoiled pUC18 DNA in the presence of 5 pmol of MacTopoIIIa wild-type (or its mutants) in 50 mM Tris-HCl, pH 8.8, 1 mM DTT, 0.1 mM EDTA, 90 mM NaCl, 30 mg of BSA/ml, and 12% (vol/vol) ethylene glycol with the indicated divalent cation. After 30 min incubation at 37uC (unless indicated otherwise), reactions were terminated by adding 4 mL of DNA loading dye (25% Glycerol, 0.2% Bromophenol Blue, 50 mM EDTA) and the products were resolved with 1.5% agarose gel in 0.5X TPE (44.6 mM Tris, 0.13% Phosphoric acid, 1 mM EDTA) buffer. DNA bands were visualized through ethidium bromide staining and imaging with a UV illuminator.
The levels of negatively scDNA that remained or were relaxed (partially and completely) were determined by densitometry using AlphaEaseFC TM software. For each divalent ion, the values for a given lane were normalized to that obtained without any divalent cation present (Fig 3, Lane 1). Activity for each ion was assessed according to the maximal percentage of substrate that was relaxed (partially and completely) relative to the percentage of negatively scDNA when no ion was present. High activity (Y) was utilized to denote greater than or equal to 95% activity while medium activity (M) denotes activity falling below 95%.

Determination of zinc content by ICP-MS
The zinc content in purified wild-type MacTopoIIIa and variants was determined at the University of Illinois Microanalysis Laboratory using the SCIEX ELAN DRCe ICP-MS (PerkinElmer Life Sciences). Briefly, two aliquots of purified protein, 1 mL each, at a concentration of approximately 0.5 mg/mL, were digested in nitric acid. The resulting solutions were diluted to 25 mL and analyzed. Two isotopes of zinc, 64 Zn and 66 Zn, were analyzed with gallium as the internal standard.

Estimation of subunit organization by gel filtration
Since MacTopoIIIa was expressed with an N-terminal 6-His tag, the protein eluted from the HiTrap TM SP column was incubated with the protease thrombin (1 unit/mg/mL) during dialysis in buffer GF (50 mM sodium phosphate, 150 mM sodium chloride, pH 7.0) in order to remove the tag. The dialyzate was loaded onto a Superdex 200 HR 10/300 gel filtration column preequilibrated with buffer GF. The chromatography was developed with the same buffer at a flow rate of 0.5 ml/min and fractions of 0.5 ml were collected and analyzed by SDS-PAGE. To generate a standard curve, the column was calibrated by analyzing a set of gel filtration standards (thyroglobulin, 669 kDa; bovine c-globulin, 158 kDa; chicken ovalbumin, 44 kDa; equine myoglobin, 17 kDa; and vitamin B12, 1.35 kDa) under the same conditions as the MacTopoIIIa.

Materials
Methanosarcina acetivorans C2A genomic DNA was used for amplification of the topoisomerase. E. coli JM109 competent cells were used as host cells for amplification of the plasmids pGEM-T and pUC18. The negatively supercoiled pGEM-T and pUC18 DNA were isolated with the Qiagen Mini Plasmid kit according to the specification of the manufacturer and used in the topoisomerase assay as substrate DNA. Reagents for SDS-PAGE, 1 protein molecular weight standards, 1 DNA molecular mass standards, etoposide, camptothecin, EDTA, nalidixic acid, novobiocin, spermidine, ethidium bromide and DTT were purchased from Sigma. HiTrap Q HP, HiTrap S HP, and Superdex 200 HR 10/ 30 columns were from GE Healthcare. BSA, single-stranded M13 ssDNA was from New England BioLabs. Gel Filtration standards were obtained from Bio-Rad.