The Role of the Methyltransferase Domain of Bifunctional Restriction Enzyme RM.BpuSI in Cleavage Activity

Restriction enzyme (REase) RM.BpuSI can be described as a Type IIS/C/G REase for its cleavage site outside of the recognition sequence (Type IIS), bifunctional polypeptide possessing both methyltransferase (MTase) and endonuclease activities (Type IIC) and endonuclease activity stimulated by S-adenosyl-L-methionine (SAM) (Type IIG). The stimulatory effect of SAM on cleavage activity presents a major paradox: a co-factor of the MTase activity that renders the substrate unsusceptible to cleavage enhances the cleavage activity. Here we show that the RM.BpuSI MTase activity modifies both cleavage substrate and product only when they are unmethylated. The MTase activity is, however, much lower than that of M1.BpuSI and is thought not to be the major MTase for host DNA protection. SAM and sinefungin (SIN) increase the Vmax of the RM.BpuSI cleavage activity with a proportional change in Km, suggesting the presence of an energetically more favorable pathway is taken. We further showed that RM.BpuSI undergoes substantial conformational changes in the presence of Ca2+, SIN, cleavage substrate and/or product. Distinct conformers are inferred as the pre-cleavage/cleavage state (in the presence of Ca2+, substrate or both) and MTase state (in the presence of SIN and substrate, SIN and product or product alone). Interestingly, RM.BpuSI adopts a unique conformation when only SIN is present. This SIN-bound state is inferred as a branch point for cleavage and MTase activity and an intermediate to an energetically favorable pathway for cleavage, probably through increasing the binding affinity of the substrate to the enzyme under cleavage conditions. Mutation of a SAM-binding residue resulted in altered conformational changes in the presence of substrate or Ca2+ and eliminated cleavage activity. The present study underscores the role of the MTase domain as facilitator of efficient cleavage activity for RM.BpuSI.


Introduction
Restriction endonucleases (REases) have been central to cloning and mapping of DNA sequences [1]. In nature, REases are normally accompanied by methyltransferases (MTases) that recognize the same target sequences and modify specific adenines or cytosines by methylation so as to protect the host DNA from cleavage by the cognate REase. Restrictionmodification (R-M) systems can be classified into four main types. Type I R-M systems are large protein complexes consisting of three unique subunits each contributing to a functional property of the enzymes: an M subunit (HsdM) that contributes to the methylation activity, an S subunit (HsdS) that contributes to the sequence specificity and an R subunit (HsdR) that contributes to the DNA cleavage activity. In methylation mode, Type I enzymes function as trimers in the form of M 2 /S 1 whereas in cleavage mode they function as pentamers in the form of R 2 /M 2 /S 1 . Type II R-M systems normally contain separate MTases and REases which cleave at defined sites [2]. Type III R-M systems consist of Res and Mod subunits that associate to form a complex in order for DNA cleavage to occur, although the Mod subunits modify target bases on their own [3][4][5]. Type IV R-M systems encode proteins that recognize and cleave methylated cytosinecontaining DNA [6,7] Research on Type II R-M systems has led to designations of subtypes that are descriptive of the properties of the REases. Type IIP REases, such as BamHI (G↓GATCC), recognize palindromic sequences and create symmetric double strand breaks, and are the most commonly used REases for molecular cloning. Type IIM REases recognize and cleave methylated, hydroxymethylated or glucosyl hydroxymethylated DNA [8][9][10]. Type IIS REases, such as BsaI (GGTCTC 1/5), cleave outside of the recognition site and are used in unconventional cloning methodologies such as Golden Gate Assembly [11]. This binary classification does not always capture all of the properties of the REases. For example, RM.BpuSI recognizes GGGAC and cleaves 10 and 14 nt downstream on the top strand and bottom strand, respectively (GGGAC 10/14). It contains both MTase and REase activity in one polypeptide and its cleavage activity can be enhanced by S-adenosyl-methionine (SAM). It can therefore be described as Type IIS/C/G for its cleavage site outside of the recognition sequence (Type IIS), bifunctional restriction/modification polypeptide (Type IIC), cleavage activity affected by Sadenosyl-methionine (SAM) (Type IIG).
Several Type IIS/C/G REases enzymes have been reported, namely, GsuI [12,13], Eco57I ( [12,13]), BseMII [14], Tth111II [15], MmeI [16,17] TaqII [18], TspDTI, TthHB27I, TsoI and TspGWI [15]. The stimulatory effect of SAM on cleavage activity has been well documented and presents a major paradox: a cofactor of the MTase activity that can modify the substrate such that it is no longer susceptible for cleavage enhances the cleavage activity. To date, the biochemical basis of the stimulatory effect remains largely unknown. The cloning of the BpuSI R-M system and the crystal structure of RM.BpuSI, the first of a Type IIS/C/G REase has recently been reported [19]. The BpuSI R-M system consists of three enzymes: two MTases and a dual function RM.BpuSI. The RM.BpuSI structure shows distinct functional domains arranged in the order of REase domain, a helix-rich connector domain, the MTase domain and target recognition domain (TRD) from the N-to C-terminus. The structure adopts an "idle" conformation where the REase active site is buried at the interface with the MTase domain and the TRD is expected to rotate relative to the MTase domain to assume the closed conformation adopted by M.TaqI as observed in co-crystal structures with substrate DNA [20]. This suggests that large movements of the REase domain and the TRD are involved for RM.BpuSI to carry out its catalytic activities. Here we report the biochemical properties of the RM.BpuSI MTase activity and provide evidence to suggest that SAM enhances cleavage activity by facilitating domain movements.

Methyltransferase activity assays
To determine the target base specificity of M1.BpuSI, M1.BpuSI was produced by using PURExpress in vitro transcription/translation system (NEB) in a 25 μl reaction according to manufacturer's instructions. 0.6 pmol of a 2 kb fragment derived from pSYX20 (NEB) containing an HgaI site overlapping with a BpuSI site (GGGACGC; HgaI site being GACGC) was incubated with 2 μl of the PURExpress product in 20 μl reactions containing 20 mM Tris HCl, pH 8.0, 50 mM NaCl and 160 μM SAM at 37°C for 2 h. 2 μl of 10x NEBuffer 1 and 2 U of HgaI (NEB) were added to the reaction and incubation at 37°C was continued for 1 h. The cleavage reactions were then analyzed by electrophoresis through a 0.8% agarose gel. HgaI cleavage activity is blocked by methylation at the N6 position of the lone adenine of GACGC (Stickel, S.K. & Roberts, R.J., REBASE). The absence of HgaI cleavage on the BpuSI/HgaI overlapping site is indicative of methylation activity of N6A methylation activity on GGGAC by M1.BpuSI.
For endpoint MTase activity assays, 1 pmol of annealed oligonucleotide duplexes was modified by 8 pmol of purified M1.BpuSI or RM.BpuSI in 100 μl reactions containing 20 mM Tris HCl, pH 8.0, 100 mM NaCl, 0.5 μM of 3 H-SAM (Perkin-Elmer) and 5 mM EDTA or MgCl 2 at 37°C for 1 h in triplicate. The reactions were stopped by immersing in a dry ice-ethanol bath and stored at -20°C until analyzed. It has been determined separately that the amount of modified sub01 DNA plateaued after 30 minutes of incubation with molar excess of RM.BpuSI (data not shown). Ten microliters of the stopped reactions were spotted onto Whatman DE81 discs (2.4 cm; GE Healthcare) in triplicate. The discs were air-dried and washed three times by 0.2 M ammonium bicarbonate, water and then ethanol. After air-drying, radioactivity remained on the discs was detected by scintillation counting.

Endonuclease activity assays
RM.BpuSI endonuclease activity was routinely assayed by incubating 5 μl of 2-fold serially diluted enzyme in 50 μl reactions containing 1x NEBuffer 4, 1 μg λ DNA, with or without 160 μM SAM at 37°C for 1 h. The reactions were stopped by adding 10 μl of 6x DNA loading dye (NEB) and analyzed by electrophoresis through 1% agarose in TBE.
Forty-nine bp long oligonucleotide duplexes containing a unique BpuSI site (sub01) were synthesized by IDT (Coralville, Iowa) with or without the indicated N6m A on the top strand or m5 C on the bottom strand. The oligonucleotide duplexes were annealed and incubated with 0.2 μM of RM.BpuSI in 20 mM Tris HCl, pH 8.0, 100 mM NaCl, 5 mM MgCl 2 , at 37°C for 1 h. The reactions were stopped by adding a 6x gel loading dye (NEB) and analyzed by electrophoresis using 10% polyacrylamide gels (Invitrogen) followed by SYBR Gold staining (Invitrogen). The stained gels were scanned and analyzed using Typhoon 9400 and the accompanying ImageQuant TL software (GE Healthcare).
Steady state kinetics experiments were carried out by incubating 200 nM of purified RM.BpuSI protein with 1.25 to 25 μM of sub01with a 5' FAM label [21] on the top strand in the presence or absence of 160 μM of SAM or sinefungin (SIN) in 100 μl reactions containing 20 mM Tris HCl, pH 8.0, 100 mM NaCl, 5 mM MgCl 2 at 37°C for 1 h. Five microliter samples were removed at designated time points and analyzed on a 10% TBE polyacrylamide gel. The gels were analyzed using Typhoon 9400 and ImageQuant TL. Protein concentration was determined by measuring the OD at 280 nm and using an extinction coefficient of 100000 M -1 cm -1 and a molecular weight of 99.5 kDa.

Spectrophotometric analysis of RM.BpuSI-ligand interactions
8-anilino-1-naphthalene sulfonic acid (ANS; Sigma-Aldrich) interacts with proteins with dissociation constant in the submillimolar range [22]. It interacts with amino acids through electrostatic and hydrophobic interactions [22] and has been used as a probe to detect conformational changes of RM.MmeI upon interactions with SAM [23]. Purified RM.BpuSI WT or mutant N406A (2.5 μM) which had undergone the SAMremoval treatment was incubated with 400 μM ANS in a buffer containing 20 mM Tris HCl, pH 8.0, 100 mM NaCl at 25°C for 5 min before ligands were added. Ligands include SAM (160 μM), SIN (160 μM), CaCl 2 (2 mM), sub01 (80 μM) and pdt01 (80 μM). Fluorescence spectra were collected from 450 to 600 nm with excitation at 360 nm. The binding reactions were assembled with utmost attention to variability and each data point represents the average of 6 scans. The change in fluorescence ΔF was calculated as ΔF = F enz -F no_enz . For the calculation of equilibrium dissociation constant, the ligands were titrated into 100 μl of 2.5 μM of RM.BpuSI in the same buffer. After subtracting from the zero ligand value, the ΔF values at 490 nM were plotted against ligand concentrations. The data points were fitted to a hyperbolic function: ΔF = (ΔF max where ΔF max is the maximum change in fluorescence, [L] is ligand concentration and K d is the dissociation constant.

Target bases of M1.BpuSI and M2.BpuSI
RM.BpuSI is a fusion of MTase and REase [19]. It recognizes a 5-bp sequence GGGAC and cleaves 10 nt and 14 nt downstream at the top strand and the bottoms strand, respectively (GGGAC 10/14). Curiously, the BpuSI R-M system contains two more MTases: M1.BpuSI, M2.BpuSI [19] The target base of M2.BpuSI was determined to be the first cytosine of the bottom strand (GTCCC) through bisulfite sequencing of a BpuSI site on a plasmid carrying the M2.BpuSI ORF under the control of a T7 promoter (data not shown). These results were consistent with the target base specificity of the two MTases of the homologous BspLU11III system [24,25].
To verify the functional role of the modifications, synthetic oligonucleotide duplexes containing the M1-and/or M2modifications ( Figure 1A) were subjected to RM.BpuSI cleavage. Figure 1B shows that the M2-and M1/M2modification completely blocked RM.BpuSI cleavage in vitro. A very small amount of cleavage product can be observed wit the M1-modified sub01.
Although M2.BpuSI modification alone blocks RM.BpuSI cleavage in vitro ( Figure 1B), omission of M1.BpuSI expression in E. coli lead to unstable RM.BpuSI clones (data not shown), indicating that both M1 and M2.BpuSI are required for the establishment of a stable RM.BpuSI system in vivo. It is in contrast to RM.MmeI, another well-studied Type IIS/C/G REase and its homologous REases which only require modification on one DNA strand for host protection and do not have a separate companion MTase [17].

Both cleavage substrate and cleaved product are substrates for RM.BpuSI MTase activity
To understand the role of the MTase domain in RM.BpuSI, we first investigated the substrate preference of the M1.BpuSI and RM.BpuSI MTase. Because RM.BpuSI cleaves downstream of the recognition sequence and does not destroy the recognition site, methylation is possible before and after RM.BpuSI cleavage. In vitro endpoint MTase activity assays were performed using oligonucleotide duplexes as cleavage substrate and cleavage product with or without the M1 or M2 modification ( Figure 1A). Excess enzyme was used over substrate so that substrate binding and catalysis are considered simultaneously, assuming the rate of product release is the same for all substrates. The reactions were carried out in the presence of 5 mM EDTA to inhibit the RM.BpuSI endonuclease activity. Endpoint experiments showed that both RM.BpuSI and M1.BpuSI methylated both the unmodified cleavage substrate and the unmodified cleavage product (Figure 2A). However, the MTase activity of RM.BpuSI is 4-6 fold lower than that of M1.BpuSI. While M1.BpuSI modified M2-modified substrate, RM.BpuSI did not modify this substrate to any observable level. Moreover, M1.BpuSI modified the cleavage substrate and cleavage product to a similar level. RM.BpuSI appeared to modify the cleavage product to a slightly higher level than the cleavage substrate, but kinetic experiments performed in the presence of excess enzyme over substrate did not show significant difference of specific MTase activity between the two substrates (data not shown). Compared to M1.BpuSI, the relatively low MTase activity towards unmodified substrates and the lack of activity toward M2-modified substrates suggests that the RM.BpuSI MTase activity is not the major MTase for the protection of the host BpuSI sites from cleavage in vivo.
Next we asked if RM.BpuSI performs both cleavage and modification reactions if both SAM and Mg 2+ are present. Our results showed that in the presence of MgCl 2 , RM.BpuSI modified the DNA to a similar extent as in the presence of EDTA (Figure 2A). Analysis of the MTase assay reactions by 10% polyacrylamide gel electrophoresis showed that sub01 was cleaved into the expected products when 5 mM of MgCl 2 was present ( Figure 2B). In the presence of SAM and Mg 2+ , it is possible that both cleavage and methylation reactions take place in parallel. Our conformational studies showed that in the presence of substrate, SIN and Ca 2+ , RM.BpuSI adopts a cleavage but not MTase conformation (see below), suggesting the enzyme exhibits MTase activity in the cleavage conformation.

SAM and SIN enhance the cleavage reaction by facilitating conformational changes
Many Type IIG REases exhibit higher cleavage activity in the presence of cofactor SAM [15,26] but not much is known about how it is achieved. We tried to answer this question by investigating how SAM affects the REase activity of RM.BpuSI. Cleavage reactions were carried out by incubating FAMlabeled sub01 with 200 nM purified RM.BpuSI that had undergone the SAM-removal treatment under steady state conditions (see Material and Methods). A plot of the rate of product formation against substrate concentration and the derived kinetic parameters are shown in Figure 3. In the presence of SAM, the V max of the RM.BpuSI REase reaction increased by ~2 fold with the K m value increased by the same magnitude compared to the kinetics in the absence of SAM (Figure 3). The enhancement of V max is independent of the MTase activity because SIN, a non-permissive analog of SAM, caused a similar increase in the V max and K m values. The concerted increase in V max and K m suggests that the enzyme carries out the cleavage reaction through a more favorable energy landscape when SAM or SIN is present.
To investigate the role of SAM-binding to the cleavage activity of RM.BpuSI, the SAM-binding residue N406 within the NPPY motif was mutated to alanine. Spectroscopic analysis of protein conformation using ANS as a probe showed that mutant N406A adopted a conformation highly similar to that of WT RM.BpuSI ( Figure 4A). Titration of SAM or SIN showed that the WT RM.BpuSI binds to SAM and SIN at a K d of 76.4±18.2 and 71.1±14.2 μM, respectively ( Figure 4B). Mutant N406A did not bind to SAM or SIN specifically (data not shown) and was inactive in MTase as expected (Figure 2A). To our surprise, this SAM-binding deficient mutant is also inactive in cleavage ( Figure 4C). Two possibilities can reconcile the presence of cleavage activity after SAM-removal treatment of WT RM.BpuSI observed in Figure 4C and elsewhere (data not shown): 1. Our SAM-removal treatment is not complete. The presence of cleavage activity is due to the presence of a small amount of SAM-bound RM.BpuSI molecules; 2. The mutant N406A adopts conformations that are not competent of cleavage upon binding to divalent metal ion and/or substrate.

RM.BpuSI undergoes conformational changes in the presence of activity-related ligands
We further investigated the conformational changes of WT RM.BpuSI and mutant N406A by measuring the fluorescence spectrum of bound ANS in the presence of ligands. SIN and Ca 2+ are used instead of SAM and Mg 2+ , respectively, because they are non-permissive co-factors for MTase and cleavage activity, respectively, and therefore allow us to measure the equilibrium state of the enzyme-ligand complexes. Figure 5A shows the fluorescence spectra of RM.BpuSI alone or in the presence of sub01, CaCl 2 or both. The presence of either of the ligands resulted in substantial reduction of maximum emission intensity, suggesting that RM.BpuSI undergoes significant conformational changes upon binding to Ca 2+ or sub01. The fluorescence spectra in the presence of sub01 and CaCl 2 are highly similar in shape and emission maximum, suggesting precleavage states of slightly different conformation. In the presence of both sub01 and Ca 2+ , the fluorescence spectrum undergoes a small red shift, suggesting another conformational change, possibly to a cleavage conformation. The difference, however, is rather small and we therefore interpret the results as an indication of a pre-cleavage/cleavage conformation distinct from the apo enzyme conformation.
We next examined the effect of SIN in the conformation of RM.BpuSI in the presence of sub01. Figure 5B shows that in the presence of SIN, RM.BpuSI adopts a conformation drastically different from the sub01 or Ca 2+ -bound or apo conformation. This unique conformation is inferred as an important state to achieve the cleavage state (see below). The RM.BpuSI-sub01-SIN spectrum exhibits a higher fluorescence intensity than the RM.BpuSI-sub01. Further experiments using pdt01 as ligand showed that the RM.BpuSI-pdt01-SIN and RM.BpuSI-sub01-SIN spectra are indistinguishable ( Figure  5C), suggesting that RM.BpuSI adopts the same conformation when sub01 or pdt01 was bound in the presence of SIN. Our biochemical experiments showed that both sub01 and pdt01 are substrate for the RM.BpuSI MTase activity in the presence of SAM and EDTA (absence of divalent metal ion). This sub01- We further examined the effect of SIN on the RM.BpuSI-sub01-Ca 2+ cleavage complex. Figure 5D shows that the RM.BpuSI-sub01-Ca 2+ -SIN spectrum is highly similar to that without SIN, especially in the lower emission wavelengths. This indicates that SIN does not induce significant conformation changes in the RM.BpuSI-Ca 2+ -sub01 cleavage complex. Furthermore, the RM.BpuSI+pdt01+Ca 2+ +SIN spectrum is significantly different from any of the spectra, suggesting that the enzyme undergoes more conformational changes after the cleavage reaction and remains bound to the product ( Figure  5D), suggesting that further conformational changes occur after the cleavage chemistry, and that the enzyme remains bound to the product.
To find out if SAM-bound RM.BpuSI-substrate-divalent metal ion complex prefers cleavage or MTase activity, we compared the fluorescence spectra of RM.BpuSI in the presence of SIN, Ca 2+ and sub01 to those represent the MTase or cleavage conformations. Figure 5E shows that in the presence of sub01, SIN and Ca 2+ , the fluorescence spectrum of the bound ANS resembles the RM.BpuSI+sub01+Ca 2+ spectrum (the cleavage conformation) than to the RM.BpuSI+sub01+SIN spectrum (the MTase conformation). This suggests that the majority of the enzyme is engaged in the cleavage mode. The fact that RM.BpuSI exhibits similar level of MTase activity in the presence or absence of Mg 2+ (Figure 2A) suggests that  RM.BpuSI can carry out MTase activity in both MTase and cleavage conformation.
As a control, the RM.BpuSI cleavage activity was assayed in the presence or absence of 0.4 mM ANS. We found that the presence of 0.4 mM ANS does not affect the RM.BpuSI cleavage activity under standard conditions on λ DNA (data not shown).

SIN increases the binding affinity of substrate to RM.BpuSI under cleavage conditions
To further investigate the role of SIN in the cleavage activity, equilibrium dissociation constant of sub01 to WT RM.BpuSI under cleavage conditions was determined in the presence or absence of SIN. As shown in Figure 5F, the K d of sub01 binding to Ca 2+ -bound RM.BpuSI is 0.17 and 1.6 μM in the presence or absence of SIN, respectively. The presence of SIN therefore increases the binding affinity of the substrate to RM.BpuSI under cleavage conditions.

SAM-binding deficient mutant N406A adopts cleavageincompatible conformations in the presence of metal ions or substrate
To determine if the absence of cleavage activity of mutant N406A is due to alternative conformational changes upon substrate binding, fluorescence spectra of N406A were obtained in the presence and absence of sub01 or CaCl 2 . Figure 6 shows that mutant N406A adopts a very different conformation than the WT when sub01 ( Figure 6A) or Ca 2+ ( Figure 6B) is present. Similarly, when both Ca 2+ and sub01 are present, mutant N406A adopts yet another different conformation ( Figure 6C), supporting the notion that mutant N406A adopts conformations that are not compatible to the cleavage activity upon substrate or Ca 2+ binding.

Discussion
SAM has been shown to affect the cleavage activity of Type I and Type III REases [27][28][29]. It induces conformational changes required for cleavage activity in EcoKI, a prototypical Type I REases [5,27], and acts as a positive allosteric effector for the cleavage activity of EcoP1I, a prototypical Type III REase [29]. Although SAM does not normally affect monofunctional Type II REases, cleavage activity of Type IIG REases is affected by SAM [15,26]. The mechanism through which SAM influences Type IIG REases cleavage activity, however, remains unclear.
In the crystal structure, RM.BpuSI adopts an "idle" configuration: the cleavage active site is buried at the interface between the REase domain and the MTase domain. The TRD is in an "open" conformation with a S-adenosyl-homocysteine, the byproduct of MTase reaction, bound to the MTase active site [19]. Molecular modeling predicted that a backbone rotation at a hinge point located at or near T580 would bring the RM.BpuSI TRD into the same position of the M.TaqI TRD bound to the substrate DNA (PDB 2IBS). A further rotation at or near A170 and G171 would position the REase active site to cleave the substrate DNA at the cleavage site [19].
In this report, we showed that under steady state conditions, SAM and SIN increased the V max of the RM.BpuSI cleavage activity with a proportional change in the K m value. Without the knowledge of the oligomerization state during cleavage, we cannot calculate the k cat or the k cat /K m values from the steady state kinetics experiments. Nonetheless, if we assume that SAM and SIN do not alter the enzyme's oligomerization state during cleavage, the V max value and the V max /K m ratio can been seen as a surrogate of the k cat and the k cat /K m value (specificity constant), respectively. In that case the V max /K m value of RM.BpuSI cleavage activity in the absence or the presence of SAM or SIN would be 0.00017, 0.00016 and 0.00016 s -1 , respectively. SAM and SIN therefore do not alter the V max /K m , and hence the k cat /K m value. Therefore, SAM and SIN increase the rate of cleavage through a higher turnover rate and an energetically more favorable pathway. A higher turnover rate is also evident by comparing the equilibrium dissociation constant K d and the K m values from the steady state kinetics experiments. In the presence of SIN, the K d value decreases from 1.6 μM to 0.17 μM; K m increases from 0.82 μM to 1.55 μM. In contrast to the K d value that represents the mass partitioning of substrate-bound enzyme versus apo enzyme under equilibrium conditions, K m is a kinetic representation of the enzyme turnover rate divided by the effective rate of substrate binding (K m = k cat /(k cat /K m )). Whereas a decrease in K d value indicates that SIN increases the binding affinity of the substrate DNA to the enzyme, an increase of K m at the same time is conceivable when the turnover rate increases at a larger extent than the increase in the rate of substrate binding.
Using bound ANS as a probe to protein conformation, we identified distinct functional conformations adopted by RM.BpuSI in the presence of relevant ligands: the precleavage/cleavage state in the presence of substrate or Ca 2+ , substrate DNA or both; the MTase state in the presence of SIN and substrate, SIN and cleavage product or cleavage product alone. Interestingly, RM.BpuSI adopts a unique conformation when SIN is bound. This unique conformation is inferred as an intermediate of an energetically favorable pathway responsible for the increase in V max in cleavage. Our results of increased enzyme turnover rate and binding affinity of sub01 to RM.BpuSI in the presence of SIN in the presence of Ca 2+ ( Figure 5F) support this hypothesis.
Finally, we created and purified a SAM-binding deficient mutant N406A to investigate if the absence of SAM-binding would obviate the increase the V max for RM.BpuSI cleavage activity. To our surprise, this mutant is inactive in cleavage. After verifying that this mutant adopts the same conformation as the WT protein in the absence of ligands, we went on to show that this mutant adopts conformations different from the cleavage conformations of the WT when Ca 2+ or sub01 is present ( Figure 6). The results with this SAM-binding deficient mutant show that the SAM-binding domain of RM.BpuSI does not only increase the V max of cleavage activity by providing a energetically favorable pathway but also making sure the correct conformational changes take place. It has been reported that SAM induces conformational changes in RM.MmeI and enhances the binding of the enzyme to target DNA [23]. The role of the MTase domain could therefore be a common regulatory mechanism for the cleavage activity of Type IIC/G REases.