A T4-phage deoxycytidylate deaminase mutant that no longer requires deoxycytidine 5'-triphosphate for activation.

A deoxycytidylate (dCMP) deaminase encoded in T4-bacteriophage DNA that is induced on phage infection of Escherichia coli was shown earlier (Maley, G. F., Duceman, B. W., Wang, A. M., Martinez, J. M., and Maley, F. (1990) J. Biol. Chem. 265, 47-51) to be similar in size, properties, and amino acid composition to the T2-phage-induced deaminase. Neither enzyme is active in the absence of dCTP or its natural activator, 5-hydroxymethyl-dCTP. However, on changing the arginine (Arg) at residue 115 of the T4-deaminase to either a glutamate (R115E) or a glutamine (R115Q), the resulting mutant enzymes were active in the absence of dCTP, with each mutant possessing a turnover number or k(cat) that is about 15% that of the wild-type deaminase. When compared on the basis of specific activity, however, the mutants are about 40-50% of the wild-type (WT)-enzyme's specific activity. Molecular weight analysis on the wild-type and mutant deaminases using HPLC size exclusion chromatography revealed that the wild-type deaminase was basically a hexamer, particularly in the presence of dCTP, regardless of the extent of dilution. Under similar conditions, R115E remained a dimer, whereas R115Q and F112A varied from hexamers to dimers particularly at concentrations normally present in the assay solution. Activity measurements appear to support the conclusion that the hexameric form of the enzyme is activated by dCTP, while the dimer is not. Another feature emphasizing the difference between the WT and mutant deaminases was observed on their denaturation-renaturation in EDTA, which revealed the mutants to be restored to 50% of their original activities with the WT deaminase only marginally restored.

type of interaction between dCTP and dTTP has been described by us for the eukaryote dCMP deaminases, but in these cases the enzyme is active in the absence of dCTP, but allosterically regulated by the ratio of dCTP to dTTP (4,5). The cooperative nature of this ligand-protein interaction suggested that the phage and eukaryotic deaminases are oligomeric proteins, a hallmark of allosteric proteins (6), and each was found subsequently to be composed of six identical M r 20,000 subunits.
We will demonstrate in the present study that on mutating Arg 115 of T4-phage dCMP deaminase to a Glu, the resulting enzyme is highly active in the absence of dCTP-Mg 2ϩ and contrary to the WT 2 -deaminase exists as a dimer. A comparison of the protein structure and enzyme activity of R115E with other mutants (R115Q, F112A) and the WT deaminase will also be presented.

EXPERIMENTAL PROCEDURES
Chemicals and Reagents-High purity dCMP, dCTP, and dTTP were purchased from Sigma. Other chemicals and reagents used for bacterial cell growth, buffers, enzyme purification and assay, and polyacrylamide electrophoresis were of reagent grade and purchased from the following commercial sources: Sigma, Fisher, Mallinckrodt Baker, Inc., (Paris, KY), or Life Technologies, Inc. Restriction enzymes were purchased from New England Biolabs (Beverly, MA). DE52 and cellulose phosphate were obtained from Schleicher and Schuell, Inc.
Mutagenesis of T4-dCMP Deaminase-T4-dCMP deaminase mutants were generated using the QuikChange Site-Directed Mutagenesis kit from Stratagene (La Jolla, CA) following the manufacturer's instructions. The plasmid, pET3c-CD5 (7), containing the entire 581-base pair coding region of the wild-type deaminase enzyme was the DNA template used in the QuikChange polymerase chain reaction-based mutagenesis. The codon CGA encoding arginine 115 of the deaminase was changed to GAG (glutamic acid), or CAG (glutamine), to generate the enzyme mutants R115E, or R115Q, respectively. DNA sequence analysis verified that the changed codon sequence at amino acid position 115 in the mutant genes was the only change made in the entire coding region. Wild-type or mutated pET3c-CD5 plasmids were used to transform E. coli BL21(DE3)/pLysS cells (Novagen, Madison WI).
Enzyme Induction and Purification-Wild-type and mutant deaminases were overexpressed from a pET3c-CD5 transformed BL21(DE3)/ pLysS host cells as described earlier (7), with the following modifications; bacterial cultures (5 ml) were grown overnight at 37°C in tryptone-phosphate medium containing 50 g/ml ampicillin or carbenicillin, and 25 g/ml chloramphenicol. Stationary phase cultures were added to Fernbach flasks containing 500 ml of the above medium and were shaken (250 rpm) at 30°C to an A 600 of 0.6. After adding isopropropyl-1-thio-␤-D-galactopyranoside to a final concentration of 0.4 mM the flasks were shaken for another 4 -6 h before harvesting the cells by centrifugation. Cell pellets were resuspended in buffer, recentrifuged, and stored at Ϫ70°C. All purification steps, modified from an earlier procedure (8), were carried out at 4°C unless stated otherwise.
The thawed cell pellet derived from 0.5 liter of induced cells was resuspended in 24 ml of ice-cold TME (10 mM Tris-HCl, pH 8.0, 1.0 mM MgCl 2 , 0.1 mM EDTA) containing 20 mM 2-mercaptoethanol, 1.0 g/ml leupeptin, and 0.1 mM phenylmethanesulfonyl fluoride. Resuspended cells were disrupted using a model VC600 Vibra Cell sonicator (Sonics * This work was supported in part by National Institutes of Health/ NCI Grant CA44355 (to F. M.) and National Science Foundation Grant MCB-9724071 (to G. F. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ To whom correspondence should be addressed: Wadsworth Center, New York State Department of Health, Empire State Plaza, Box 509, Albany, New York 12201-0509. Tel.: 518-474-4184; Fax: 518-473-2900; E-mail: maley@wadsworth.org. 1 5-Hydroxymethyl-dCTP is the most probable activator of the T4phage deaminase, since it is incorporated into the phage DNA in place of dCTP. The latter is broken down to dCMP by a phage-induced enzyme and subsequently hydroxymethylated by another phage-induced enzyme. Because of its commercial availability dCTP is used in place of 5-hydroxymethyl-dCTP. For a more complete discussion of the physiology of T-even phage pyrimidine-nucleotide metabolism, see Ref.

5.
& Materials, Inc., Danbury, CT) equipped with a 12 mm diameter probe (duty cycle set at 80% with an output control setting of 9). Cells were sonicated five times, 30 s per sonication, in an alcohol/ice bath. The disrupted cell mixture was centrifuged for 30 min at 34,500 ϫ g and the pellet discarded. The supernatant fraction contained about 2500 R115E units of deaminase activity with a specific activity of 10 units/mg of protein. Ammonium sulfate was added to the supernatant fraction to 35% saturation, and the mixture was stirred for 20 -30 min, after which the turbid solution was centrifuged for 20 min at 34,500 ϫ g and the resulting pellet discarded. Ammonium sulfate was then added to the supernatant fraction to 80% saturation and the mixture stirred for 30 min, at which time the precipitated protein was collected by centrifugation for 30 min at 34,500 ϫ g. The protein pellet was dissolved in 10 -15 ml of Buffer A (10 mM Tris-HCl, pH 8.5, 10% (w/v) ethylene glycol, 20 mM 2-mercaptoethanol, 0.2 mM MgCl 2 , 0.1 mM EDTA) and dialyzed overnight against two 1.0-liter changes of Buffer A. The dialysate was sonicated three times at 20 s each in an ice bath using a 3-mm diameter probe (50% duty cycle; with an output control setting of 5) to reduce its viscosity and then loaded onto a 2.5 ϫ 14-cm DE52 anion exchange column equilibrated with Buffer A. The DE52 column was washed with 3 column volumes of Buffer A (or until the OD 280 was at background levels), then with 3 column volumes of Buffer A ϩ 0.1 M NaCl. Most of the R115E and R115Q deaminase activity eluted in the 0.1 M NaCl wash, while the wild-type deaminase eluted with 0.2 M NaCl. Fractions containing the deaminase peak were pooled, concentrated in an Amicon (Amicon Division, W. R. Grace & Co., Beverly, MA) ultrafiltration device (10 kDa cut-off filter), and dialyzed over night against two 1.0-liter changes of Buffer B (10 mM potassium phosphate, pH 7.1, 10% ethylene glycol, 0.2 mM MgCl 2 , 0.1 mM EDTA, 2 mM DTT 2 ). Although most of the wild-type and R115Q enzymes usually precipitated from solution during the course of the dialysis, R115E remained in solution. The precipitates were centrifuged for 30 min at 34,500 ϫ g, and the pellets were dissolved in 0.15-0.25 M potassium phosphate, pH 7.1, 10% ethylene glycol, 2 mM DTT, and then diluted in 10% ethylene glycol, 2 mM DTT until the phosphate concentration was about 20 mM. The dissolved protein was applied to a 1.5 ϫ 12-cm cellulose phosphate cation exchange column equilibrated with Buffer B, which was then washed with 3-5 column volumes of Buffer B, followed by 4 -5 column volumes of Buffer B ϩ 50 mM phosphate. Most of the R115E and R115Q deaminase activities eluted in the 50 mM phosphate wash, in contrast to the wild-type deaminase, which required 0.2 M phosphate for elution. About 1500 units of R115E were recovered, usually with a specific activity of 90 -100 units/mg of protein. The enzyme eluates were concentrated in an Amicon as above and when stored at 0 -4°C for extended periods 2-mercaptoethanol was added to 0.1 M. For longer periods of storage the enzyme was precipitated by addition of solid ammonium sulfate to 80% saturation and the centrifuged protein was frozen at Ϫ40°C. Enzyme purity at this stage was usually Ͼ95% as determined by Coomassie blue staining of sodium dodecyl sulfate-polyacrylamide gels following electrophoresis of the purified protein samples.
Enzyme Activity and Protein Assays-Measurement of enzyme activity and the quantitation of protein in purified deaminase preparations were performed at 30°C as described earlier (9). The assay solution in which deaminase activity was measured contained 10 mM Tris-HCl, pH 8.0, 0.5 mM dCMP, 0.5 mM MgCl 2 , 5 mM 2-mercaptoethanol, with or without 20 M dCTP, depending on the enzyme to be assayed. One unit of activity refers to the deamination of 1 mol of dCMP/min under the conditions of the assay. Specific activity refers to units of activity/mg of protein.
Enzyme Kinetic Analysis-dCMP deaminase activity was measured at 5, 10, 50, 100, 250, and 500 M dCMP in the presence and absence of dCTP. When MgCl 2 was omitted from the assay solution, wild-type enzyme activity was negligible at 0.5 mM dCMP. Substrate K m and V max values were calculated for the WT dCMP deaminase and the R115Q, and R115E mutants by nonlinear fitting of the kinetic data (catalytic velocities and substrate concentrations) to the Michaelis-Menten equation using the Enzfitter program of Robin Leatherbarrow (version 1.05). Due to the relatively high turnover of the WT deaminase (ϳ2.5 ϫ 10 4 mol of dCMP deaminated min Ϫ1 mol Ϫ1 hexamer), enzyme samples were routinely diluted to 15-35 g/ml protein with a solution of 0.2 M potassium phosphate, pH 7.5, 10% ethylene glycol, 0.1 M 2-mercaptoethanol, and 20 M dCTP, from which aliquots were removed for the measurement of enzyme activity.

Mutation of Arg 115 Results in Loss of Activation by dCTP-
The effect of replacing Arg 115 in WT T4-dCMP deaminase with either a glutamine (R115Q) or a glutamate (R115E) residue was similar in that neither enzyme, unlike the WT deaminase, required dCTP and Mg 2ϩ for activity to be observed. Omission of dCTP yielded only marginal activity for the WT enzyme as is evident in the first column in Fig. 1. When Mg 2ϩ was omitted too, the enzyme was essentially inactive. In enzyme assays where dCTP was absent from the assay solution, activity was observed immediately upon the addition of R115E or R115Q, as shown earlier for F112A (10), whereas wild-type enzyme activity could not be detected until dCTP was added to the reaction cuvette (data not shown). In the case of R115Q, it should be noted that while this mutant was about as active as R115E in the absence of dCTP (Table I), its activity was increased by 30 -40% in the presence of 20 M dCTP (Fig. 1). When 0.1 mM dTTP was included in the WT deaminase assay the enzyme was inhibited by 20%, with no effect on the mutants. However, in the presence of 0.3 mM dTTP the WT deaminase was inhibited by about 90%, while the mutants were inhibited by about 30% (Fig. 1). Thus, although some inhibition of the mutants can be effected by dTTP, they appear to be much less sensitive to inhibition by dTTP than the WT deaminase, consistent with their lesser sensitivity to dCTP.
Kinetic Characterization of the Deaminase Mutants-To determine whether the kinetic properties of the mutants differed, V max and K m values of each of the enzymes was evaluated in the presence and absence of dCTP using a nonlinear fitting of the kinetic data to the Michaelis-Menten equation. Table I reveals that the maximal activity (V max ) of R115Q and R115E was about 45-55% that of the WT deaminase when measured in the absence of dCTP. At the concentration of dCTP normally employed for the WT deaminase assay (20 M) the activity of R115E was perhaps slightly inhibited, while that of R115Q was increased by 30 -40% (data not shown). A similar type of comparative analysis of enzyme activity with the WT deaminase in the presence and absence of dCTP could not be conducted due to the fact this enzyme is basically inactive in the absence of dCTP. The higher V max (about 2-fold) of the WT enzyme relative to R115Q and R115E could be due in part to its 4 -5-fold lower K m relative to that of the mutants. When the mutants and WT deaminase are compared on the basis of k cat , the differences become even more exaggerated (Table I), since M r values become a factor in this parameter, and as indicated below the WT enzyme is most likely a hexamer, while the mutants appear to be in their dimeric state during the course of the assay. The differences are even more exaggerated when their specificity constants (k cat /K m ) are compared (22 s Ϫ1 M Ϫ1 for the WT deaminase versus 0.60 for the mutants).
Molecular Weight Analysis of T4-phage Deaminase and Mutants-Earlier studies with the chick embryo (4) and T2-phage deaminases (11) indicated that under conditions of dilution or inhibition by dTTP the enzyme was reduced in size. With the WT and mutant deaminases available to us now, and the improvement in M r determinations via size exclusion HPLC, a more careful evaluation of the apparent M r values of these proteins was undertaken using the protocol described under "Experimental Procedures." A plot of the elution times of the WT T4-deaminase, R115E, R115Q, and F112A, versus log M r using the standards indicated under "Experimental Procedures" yielded the following respective M r values, 127,000, 44,300, 124,500, and 111,900 (data not shown). It should be indicated that 20 M dCTP was included in the elution buffer. Considering that the estimated M r of a subunit of WT T4phage-dCMP deaminase is 21,200 (12), R115E would be a dimer, WT T4-dCMP deaminase and R115Q would be hexamers, and F112A would most probably be a hexamer.
However, while the elution times of WT deaminase and R115E did not vary with protein concentration over a 400-fold dilution (4 to 0.001 mg/ml) in the presence or absence of dCTP (Table II), that of R115Q was affected dramatically, indicating that it had undergone a marked change in its oligomeric structure on dilution. This effect is clearly seen in Table II and Fig.  2d, where R115Q appears to be mostly a hexamer, but also contains a component that elutes as a dimer. When the protein was diluted from 4 to 0.5 mg/ml, the dimer increased somewhat (Fig. 2e), but on further dilution to about 0.04 mg/ml the major peak by far was the dimer (Fig. 2f). Complete transformation to the dimer was observed to occur in the presence or absence of dCTP at 1 g/ml (Table II). Under similar conditions, WT T4-dCMP deaminase retained its hexameric structure (Fig. 2,  a-c), and R115E remained as a dimer (Fig. 2, g-i). In the case of F112A (Table II), this protein's elution time suggested it to behave anomalously in that its M r was about 113,000 at 4 mg/ml, which decreased to about 63,000 at 0.04 mg/ml whether in the presence of dCTP or not (data not shown). The broadness of the elution profile in the latter case suggests the presence of multiple species, but on further dilution to 1 g/ml, however, F112A clearly eluted as a dimer (Table II). It should be emphasized that at the concentrations employed in the assay for each protein, the only one that retained its hexameric structure, regardless of the presence or absence of dCTP or extent of dilution, was the WT deaminase. R115Q appeared to retain much of its hexameric structure in the presence of dCTP until diluted to 1 g/ml, where like F112A it eluted as a dimer. However, in the absence of dCTP, R115Q migrated with an M r of 60,000 at the higher levels of protein, but on further dilution (1 g/ml) it migrated as a dimer (Table II). Since the protein concentrations given in Table II are those prior to injection onto the column, it should be noted that even further dilution occurs during the process of elution.
It is of interest to note that R115Q, like the WT deaminase, was markedly activated by dCTP when its hexameric form was assayed following isolation from the sizing column. The extent of activation was at least five times the activity of the enzyme in the absence of dCTP. By contrast the column-isolated dimer, like R115E, showed little if any activation by dCTP. A similar result was obtained with the M r 60,000 form of R115Q that was isolated in the absence of dCTP. However, as indicated in Fig.  1, R115Q when assayed directly showed only a 30 -40% activation by dCTP, which suggests that when the two forms of R115Q are not separated prior to assaying, the dimer predominates.
Enzyme Denaturation-Renaturation and Zinc Chelation Experiments-Previous work (9) in our laboratory demonstrated that when T4-phage-dCMP deaminase was denatured in the presence of 6 M guanidine HCl, almost complete restoration of activity could be obtained on removal of the denaturant by dilution or dialysis. Similar studies were conducted with the b This value refers to mol of dCMP deaminated per min/mg of protein at 30°C, which can also be defined as specific activity.
c WT deaminase showed no activity in the absence of dCTP ϩ Mg 2ϩ , but in the presence of Mg 2ϩ alone a small amount of activity was noted.
d Activity was determined in the absence of dCTP. R115E and R115Q mutants and comparable with the earlier WT-deaminase denaturation-renaturation results, their respective activities were restored with time (Fig. 3A). It is of interest to note that the WT-deaminase, when refolded, was even more active than prior to the denaturation-renaturation process, which indicates that the renatured enzyme's structure is more supportive of enzyme activity than that before denaturation. However, when the denaturation of WT-enzyme was conducted in the presence of EDTA to remove its two resident Zn 2ϩ atoms/subunit (9), little activity could be restored on denaturant removal (Fig. 3B), even when Zn 2ϩ was added to the renaturation solution. By contrast, when similar studies were conducted with R115E and R115Q, 54 and 60% of their respective original activities were detected 40 min after renaturation (Fig. 3B).

DISCUSSION
Since our initial observation that dCMP deaminase, at least in eukaryotes, is an allosteric enzyme activated by dCTP and inhibited by dTTP (13), a major objective has been to determine how these effectors manage to regulate this enzyme. The T4-(and also T2-and T6-) phage dCMP deaminase may not be considered a classical example of an allosteric enzyme (6), since it does not show kinetic cooperativity, or activity for that matter, in the presence of substrate alone. However, in the presence of dCTP and Mg 2ϩ the enzyme assumes an active conformation and demonstrates a heterotrophic interaction between dCTP and dTTP that is sigmoidal in nature (8). Thus, the phage deaminase could be considered a unique example of an allosteric enzyme, one that is related to its eukaryote counterpart, but differing somewhat in the way its six subunits interact in the presence of ligands, relative to the manner in which the six subunits of the chick embryo (4) or human deaminase interact (14). The photofixation of dTTP to the T2-deaminase (15), a process inhibited by dCTP, suggests that the regulators probably bind to the same site or to sites that are in proximity to one another. Subsequent studies revealed that dTTP photo-fixes to Phe 112 in the T4-phage enzyme and that mutating this residue to an Ala results not only in the loss of photofixation, but in a partial loss in the enzyme's dCTP requirement for activity (10). In contrast to the WT-deaminase, F112A is active in the absence of dCTP, but it possesses only about 10% as One-hundred g of each protein was applied to the column in each case except for the 0.035 mg/ml samples, where a total of 7 g of protein was applied. much activity. The F112A activity is increased greatly by the presence of 100 -300 M dCTP, reflecting this mutants greatly reduced binding of dCTP (10). The fact that dCTP, in addition to dTTP binding, is reduced considerably suggests a common binding site for these nucleotides. McIntosh and Haynes (16) also implicated this site in the binding of dTTP to several proteins, including dCMP deaminase and ribonucleotide reductase. Thus, the conservation of an Ala-Ala-Arg site in these regulated proteins suggested that the positively charged guanidinium group of Arg might be involved in the binding of dCTP and dTTP through an electrostatic interaction with their negatively charged phosphate groups. In support of this thesis it is seen in this paper that on replacing Arg 115 of the T4-phage dCMP deaminase with a Glu, a mutant is obtained that no longer depends on dCTP for activity and that it is much less sensitive to inhibition by dTTP. In this instance as much as 40% of the activity of the R115E mutant was retained (based on specific activity) relative to the WT-deaminase, with little or no response on addition of dCTP to the assay (Fig. 1). The lack of responsiveness of R115E to dCTP contrasts with R115Q, which although as active as R115E, can still be activated by 30 to 40% by the presence of 20 M dCTP (Fig. 1).
It is perhaps deceptive to compare the activities of the various deaminases on the basis of specific activities alone because of the difference in their molecular weights. Thus, when size is factored in as a result of employing k cat values, that of the WT T4-deaminase is 467 s Ϫ1 , while in the case of R115E a value of 69 s Ϫ1 is obtained. This 7-fold difference contrasts with their specific activities where only a 2-fold difference (Table I) is encountered. In the case of R115E it is clear that this protein is a dimer, but for R115Q its oligomeric state is less certain as it can be a dimer or hexamer or something in between (Fig. 2, Table II). As a dimer the k cat of R115Q is 86 s Ϫ1 (Table I), but as a hexamer its k cat would be 255 s Ϫ1 . It is therefore important to know what the structural status of a protein is in calculating k cat . At the protein concentration of 0.05-1.0 g employed in the assay of each of the mutant enzymes, all would appear to be dimers (Table II) whether dCTP is present or not. It is not entirely clear that the size of the holoenzyme is affected by the charge of the amino acid residue at 115 but is suggested by the fact that R115Q has less of a tendency to dissociate at protein concentrations comparable with R115E (Fig. 3, Table II), and the presence of dCTP appears to stabilize the hexameric state of this protein. However, when diluted sufficiently R115Q will dissociate to a dimer whether dCTP is present or not (Table II). By contrast the WT-deaminase retains its hexameric state at the highest levels of dilution (1 g/ml), particularly in the presence of dCTP.
There appears to be a complex relationship between Phe 112 and Arg 115 in their interaction with dCTP and dTTP, as evidenced by the fact that the mutation of either amino acid affects the binding of these nucleotides. As shown earlier (10) conversion of Phe 112 to an Ala greatly reduces the binding of dCTP and dTTP and impairs the photofixation of the latter. The importance of Phe in this process is emphasized by the fact the human dCMP deaminase, which contains a Thr-Ala-Ala-Arg sequence (17) in a sequence comparable with that of the phage deaminase, does not photofix dTTP (10). Replacement of Phe 112 in the phage enzyme with a Thr, as occurs in the case of the human deaminase, might explain why the human enzyme (14) is not regulated as tightly by dCTP and dTTP. Both residues would thus appear to be important in nucleotide binding, Phe possibly through van der Waals interactions with the pyrimidine ring and Arg via electrostatic interaction with the phosphate of the nucleotides. Whether this proposal is realistic or not might be established through x-ray crystallographic analysis, if reasonable crystals of the WT-deaminase or R115E can be obtained. Meanwhile it would be of interest to determine whether converting the Thr in the human deaminase to a Phe enables this enzyme to now photofix dTTP, or on converting the Arg corresponding to Arg 115 to a Glu promotes dimer formation in the human enzyme, as it does with phage deaminase.
Another difference between the human and phage deaminases is in the number of Zn 2ϩ atoms associated with each subunit. While it was anticipated that Zn 2ϩ would be associated with the active sites of the human and phage deaminases, based on their similarity to other aminohydrolases such as adenosine (18) and cytidine deaminase (19), the presence of a second Zn 2ϩ in each subunit of the phage deaminase was unexpected (9). As indicated previously this Zn 2ϩ may be involved in promoting protein-protein interactions between T4-phage dCMP deaminase and T4-phage thymidylate synthase (20). The role of this second Zn 2ϩ site, as well as its location, is still to be clarified and may provide an explanation for why R115E and R115Q retain at least 50% of their activity on denaturation-renaturation in the presence of EDTA, while the WTdeaminase is at least five times more sensitive to this treatment (Fig. 3B). It is possible that the zinc ions at the active and secondary sites of the WT-deaminase are more exposed to EDTA during denaturation than R115E and R115Q, although preliminary data obtained from sizing analysis has shown that the hexameric state of the WT-deaminase is not restored following denaturation-renaturation in the presence of EDTA, while it is in the absence of EDTA.