ecori endonuclease

THE JOURNAL OF B~OLOCICAL CHEMISTRY Vol. 251, No. 19, Issue of October 10, pp. 5866-5874, 1976

Printed in U.S.A.

EcoRI Endonuclease

PHYSICAL AND CATALYTIC PROPERTIES OF THE HOMOGENEOUS ENZYME*

(Received for publication, March 2, 1976)

PAUL MODRICH+ AND DONNA ZABEL

From the Department of Chemistry, University of California, Berkeley, California 94720

A procedure for large scale isolation of Escherichia coli RI endonuclease in high yield has been developed. The purified enzyme is homogeneous as judged by polyacrylamide gel electrophoresis and analytical sedimentation. The denatured and reduced form of the enzyme has a molecular weight of 28,500 A 500. In solution the enzyme exists as a mixture of dimers and tetramers of molecular weights 57,000 and 114,000, respectively. We estimate the dissociation constant for tetramer to dimer transition to be less than or approximately equal to 1 x lo-’ M.

Steady state kinetic analysis of the endonuclease with ColEl DNA as substrate showed that the enzyme obeys Michaelis-Menten kinetics. At 37” the turnover number is four double strand scissions per min, and the K, for ColEl molecules is 8 x 10m9 M. At 0” the major product of endonuclease action contains only one single strand break in the RI site, and such molecules can dissociate from the enzyme. In contrast, at 30” or 37’, two single strand breaks are introduced into the RI sequence prior to dissociation of the enzyme. A transient enzyme-bound intermediate containing only one break in the RI site was observed in studies of a single turnover at 30”. Kinetic analysis of this reaction indicates that the first break is introduced into the RI site with a first order rate constant of at least 40 min-‘, while the second cleavage occurs with a rate constant of 14 min-‘. Since the turnover number of the enzyme at 30” is only 0.72 min-‘, these results indicate that the rate-limiting step is release of endonuclease from its DNA product.

Escherichia coli RI (EcoRI) DNA restriction and modification enzymes are responsible for the host specificity of E. coli strains harboring the fi+ drug resistance transfer factor RTFI (l).’ The enzymes recognize unique sites on DNA, with the minimal recognition site being a hexanucleotide sequence characterized by 2-fold symmetry (Fig. 1) (3). Restriction results from introduction of two staggered single strand breaks within the recognition sequence to generate 3’-hydroxyl and 5’-phosphoryl termini (3, 4). Modification is a consequence of methylation of adenine residues adjacent to the axis of symmetry, rendering the site resistant to cleavage by the endonuclease (5).

The catalytic requirements of the EcoRI enzymes are relatively simple. In vitro restriction requires only unmodified DNA and a divalent cation, while unmodified DNA and S-adenosyl-L-methionine are sufficient for in vitro modifica-

* This investigation was supported by United States Public Health Service Grant GM-21610 and a Biomedical Support Grant from the University of California.

f To whom reprint requests should be sent. Present address, Department of Biochemistry, Duke University Medical Center, Dur- ham, N. C., 27710.

‘In accordance with the nomenclature suggested by Smith and Nathans (2), EcoRI designates the DNA restriction and modification system specified by the extrachromosomal element in Escherichia coli strains harboring the drug resistance transfer factor RTFI.

tion (6). Thus, the EcoRI enzymes are among the simplest sequence-specific DNA enzymes known and are well suited to a study of DNA sequence recognition by proteins. Furthermore, since the endonuclease and methylase are separable in oitro (6)

1 5’ ~GpApipTpTpC

CPTPTP$PAPGP 5’ t

FIG. 1. Minimal recognition sequence of EcoRI restriction and modification enzymes. Arrows indicate sites of endonucleolytic cleavage (3, 4). Methylation occurs on adenine residues indicated by asterisks (5).

and since genetic studies indicate that the two activities do not share a common subunit (l), this system may offer a unique opportunity to compare two proteins which interact with the same DNA sequence.

EcoRI endonuclease has been purified extensively by Greene et al. (6). and the final preparation exhibited a single band on polyacrylamide gels in the presence of sodium dodecyl sulfate; however, it was not demonstrated that this species possessed endonuclease activity. Furthermore, little information is avail- able on the physical properties of the enzyme or on its mechanism of action. Since this information is essential to understanding the interaction of the enzyme with its DNA

5866

by guest on January 31, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Physical and Catalytic Properties of EcoRI Endonuclease 5867

substrate, we have undertaken a study of the endonuclease with emphasis on these points.

Inasmuch as previous procedures for isolation of the endonuclease have been characterized by low yields and have required addition of nonionic detergent to prevent precipitation of the enzyme (6), we have developed a new procedure that does not require detergent and which can be used for large scale isolation of the enzyme in high yield. We also describe oligomeric states of the enzyme in solution and studies on its catalytic mechanism.

EXPERIMENTAL PROCEDURE

Materials

Bacterial Strains-Escherichia coli RY13 gal end (RI) was provided by Dr. H. Boyer (University of California, San Francisco), E. cob JC411 thy (ColEl) (7) was obtained from Dr. D. Helinski (University of California, San Diego).

Enzymes and Proteins-EcoRI methylase was purified according to the method of Rubin and Modrich.‘The enzyme was at least 97% pure as judged by polyacrylamide gel electrophoresis and was free of methylase activity on bacteriophage T7 and PM2 DNAs which do not contain RI sites (4, 6). E. coli DNA ligase and DNA polymerase I were isolated as described (8, 9). Polynucleotide kinase (Fraction VI (10)) was the generous gift of Dr. I. R. Lehman (Stanford University). E. coli recBC DNase (11, 12) was purified through the phosphocellulose step of Eichler and Lehman.3 Ovalbumin, DNase I, lysozyme, and catalase were from Worthington. Crystalline bovine serum albumin (A grade) was purchased from Calbiochem, and bovine hemoglobin and yeast alcohol dehydrogenase from Sigma.

DNA-Salmon sperm DNA (A grade) was obtained from Calbio- them. Covalently closed circular bacteriophage PM2 DNA was isolated from purified virions (13) by phenol extraction and purified by CsCl-ethidium bromide centrifugation (14). Bacteriophage T7 DNA was isolated according to the method of Hinkle and Chamberlin (15).

‘H-labeled (‘7 to 10 cpm/pmol) or 3ZP-labeled (6 cpm/pmol) covalently closed circular colicin El (ColEl) DNA was isolated from JC411 (ColEl) after chloramphenicol amplification of the plasmid (7). For preljaration of unlabeled or ‘H-labeled DNA, cells were grown at 37” in M9 medium (16) containing 1% glucose, 0.5% casamino acids, 4 pglml of thymine, and 10 fig/ml of thiamin. For 3ZP-labeling, growth was in Medium II (17) containing 1% glucose, 0.25% vitamin-free casamino acids, 10 @g/ml of thymine, and 10 rg/ml of thiamin. When the A,,, of the culture reached 1.0 to 1.2, chloramphenicol (Calbiochem) was added to 150 rg/ml, followed 90 min later by the addition of [3H]thymidine (2 rCi/ml, SchwarziMann, 11 Ci/mmol) or ‘T-labeled inorganic phosphate (20 rCi/ml, New England Nuclear, carrier free). After incubation at 37” for an additional 16 h, the culture was chilled to O”, and cells collected by centrifugation. ColEl DNA was purified by minor modifications of published procedures. Chromosomal DNA present in lysozyme-EDTA lysates was removed by NaCl-sodium dodecyl sulfate precipitation (18). After treatment of the supernatant fraction with autodigested pronase (0.14 mg/ml) for 60 min at 37”, the plasmid DNA WSS isolated by two cycles of CsCl-ethidium bromide equilibrium density centrifugation (14). DNA isolated in this manner was at least 95% closed circular and was free of contaminating RNA.

ColEl DNA methylated at the RI site was prepared in a reaction (1.2 ml) containing 0.1 M Tris/HCl (pH 8.0), 0.01 M EDTA (pH 8.0), 1 mM dithiothreitol, 0.56 mM ColEl DNA, 2 pM S-adenosyl-L-[methyl- ‘Hlmethionine (11.9 Ci/mmol, New England Nuclear), and 3.6 pg of EcoRI methylase. Incubation was at 37”. and additional enzyme was added at 15-min intervals. As judged by incorporation of ‘H into DNA, reaction was complete in 30 min. After 60 min the reaction was terminated by extracting twice with redistilled phenol equilibrated with 0.1 M Tris/HCl (pH 7.6), 2 mM EDTA. The DNA was dialyzed exhaustively against 0.02 M Tris/HCl (pH 7.6), 0.05 M NaCl, 1 mM EDTA. The final preparation contained 1.54 methyl groups per mol of ColEl DNA, assuming the concentration of the methyl donor cited by the manufacturer to be accurate.

ColEl DNA containing a single strand break at the RI site was prepared in reactions containing 0.02 M potassium phosphate (pH 7.4),

*R. Rubin and P. Modrich, manuscript in preparation. 3D. Eichler and I. R. Lehman, personal communication.

0.05 M NaCl, 1 rnM EDTA, 5 mM MgCl,, 0.06 mM ColEl DNA, and EcoRI endonuclease at 0.75 to 6.7 units/ml. At the higher endonuclease concentrations the yield of circles with a single strand break was higher, but the fraction of linear molecules in the product was also increased. After 15 min at O”, the reaction was terminated by addition of EDTA to 0.02 M, extracted with phenol, and dialyzed as described above.

All reagents and glassware used in preparation of DNA were sterile, and dialysis tubing was boiled twice for 20 min in sterile H,O prior to use with DNA solutions.

Other Materials-Phosphocellulose (Whatman Pll) was precycled and equilibrated as described previously (8). Sephadex G-200 was purchased from Pharmacia. DNA-cellulose (8 mg of heat-denatured salmon sperm DNA/g of cellulose) was prepared by the ultraviolet irradiation procedure of Litman (19). Hydroxylapatite (Bio-Gel HTP), acrylamide, and agarose were from Bio-Rad. “Ultrapure” ammonium sulfate was obtained from Schwarz/Mann. Streptomycin sulfate was the generous gift of Merck and Co. Where indicated spectral grade glycerol (Eastman Spectra) was employed.

Methods

Growth of Cells-E. coli RY13 was grown in 200.liter cultures of L-broth (per liter: 10 g of Bacto-tryptone, 5 g of yeast extract, 10 g of NaCl, 5 g of glucose, 5 mM potassium phosphate (pH 7.0)) in a New Brunswick fermentor with maximum aeration. Culture pH was main- tained at 7.0 by addition of NaOH. When the culture entered early stationary phase as judged by no further increase in cell mass in 20 min (A,,, = 15), the cells were chilled to 5” and harvested with a refrigerated Sharples centrifuge. Cell paste was frozen at -20” and showed no loss of EcoRI endonuclease activity over a period of 4 months.

Gel Electrophoresis-Protein electrophoresis in the presence of 0.1% sodium dodecyl sulfate was performed according to the method of Weber and Osborn (20). The Tris system of Jovin et al. (21) was employed for electrophoretic analysis under native conditions on gels of 7.5% polyacrylamide. Gels were stained with Coomassie brilliant blue and protein quantitated by scanning at 550 nm.

Electrophoresis of DNA on 1% agarose slab gels (0.3 cm thick) was performed in an apparatus similar to that described by Studier (22) using the Tris/borate system of Peacock and Dingman (23). After electrophoresis for 2 h at 12.7 V/cm, DNA bands were visualized by staining for 1 h in 1 rg/ml of ethidium bromide. DNA content of bands was determined by liquid scintillation counting if [‘T]DNA was employed, or by scanning photographic negatives with a Joyce-Loebl microdensitometer according to the method of Depew and Wang (24). Although tedious, the latter method was highly sensitive, capable of detecting 1 ng per band.

EcoRI Endow&ease Reactions-The standard assay for the endonuclease measures conversion of closed circular ColEl DNA, which contains a single EcoRI site (25), to the linear form which is sensitive to degradation by recBC DNase (11, 12). Reactions (0.20 ml) contained 0.1 M Tris/HCl (pH 7.6), 0.05 M NaCl, 5 mM MgCl,, 0.2 mM EDTA, 0.2 mM dithiothreitol, 0.1 mM ATP, 0.02 mM ColEl [SH]DNA, 6.8 units of recBC DNase, and EcoRI endonuclease. After incubation at 37” for 10 min, the reaction was terminated by addition of 0.05 ml of 3 mM salmon sperm DNA and 0.05 ml of 30% trichloroacetic acid. After 10 min at O”, the precipitate was removed by centrifugation at 23,000 x g for 10 min, and acid-soluble nucleotide determined in Bray’s scintillation fluid (26). Reaction was linear with time, and control experiments showed recBC DNase to be present in excess. One unit of endonuclease activity converts 1 pmol of ColEl molecules (12.8 nmol of nucleotide) to a form which is sensitive to recBC DNase in 1 min under assay conditions. ’ The assay was linear up to 0.01 unit, and the enzyme was diluted as necessary into 0.02 M potassium phosphate (pH 7.4), 0.2 M NaCl, 0.5 mM dithiothreitol, 0.2 mM EDTA, 0.2 mg/ml of bovine serum albumin, and 10% (w/v) glycerol.

Unless indicated otherwise, endonuclease activity with T7, PM2, or methylated ColEl DNA as substrate was determined at a DNA concentration of 0.06 mM in 0.1 M Tris/HCl (pH 7.6), 0.05 M NaCl, 5 mM MgCl,, 100 kg/ml of bovine serum albumin at 37”. Reaction

‘As described under “Results,” this assay measures double strand scission at the RI site with 1 unit corresponding to hydrolysis of 2 pmol of phosphodiester bonds/min. As defined by Greene et al. (6), 1 unit of endonuclease cleaves 1 pmol of phosphodiester bonds/min. Hence, the unit employed here corresponds to 2 units of Greene et al. (6).


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

5868 Physical and Catalytic Properties of EcoRI Endonuclease

products were separated and quantitated by agarose gel electrophoresis as described above.

The optimal buffer for EcoRI endonuclease at temperatures of 25-37” is 0.1 M Tris/HCl (pH = 7.6 at 23”), 0.05 M NaCl, 5 mM MgCl, (6). However, this buffer is not suitable at O-10” due to premature termination of reaction. This effect, which may reflect the elevated pH of Tris/HCl buffers at low temperature, can be overcome by substitu- tion of 0.02 M potassium phosphate (pH 7.4 at 23”) for the Tris/HCl component of the above buffer. In experiments on the mechanism of double strand cleavage, which include low temperature incubations, the phosphate-buffered reaction was employed exclusively. Although the activity of the endonuclease in phosphate was only 40% of that in Tris/HCl at 37”, reaction products were identical.

Amino Acid Analysis-Amino acid analyses were performed on a Beckman model 121 analyzer. We are grateful to Dr. George Stark (Stanford University) for use of his instrument. Samples of the endonuclease were hydrolyzed in 6 N HCl containing a crystal of phenol at 110’ in uac~o for 24, 48, and 72 h. Values for serine and threonine were obtained by extrapolating to zero hydrolysis time, while those cited for valine and isoleucine are averages of 4% and 72-h determina- tions. Half-cystine was determined as cysteic acid after performic acid oxidation (27). Tryptophan was determined spectrophotometrically under alkaline conditions (28).

Sedimentation Analysis-Prior to sedimentation, EcoRI endonuclease was dialyzed against two changes of 2000 volumes (8 to 12 h per change) of 0.02 M potassium phosphate (pH 7.4), 0.25 M KCl, 0.2 mM EDTA, 0.1 mM dithiothreitol.

Analytical centrifugation was performed in a Beckman model E equipped with photoelectric scanner. A wavelength of 278 nM was employed in all experiments. Boundary sedimentation was performed at an initial protein concentration of 0.25 mg/ml at 43,630 rpm at 20.0”. The sedimentation coefficient was corrected to standard conditions as described (29). A protein partial specific volume of 0.736 at 25” was calculated from amino acid composition (30, 31) and for use in calculations was corrected for an assumed temperature dependence of 5 x lo-’ ml g-’ deg-’ (32). Conventional sedimentation equilibrium centrifugation was performed according to the method of Chervenka (29) at initial enzyme concentrations of 0.29 to 0.54 mg/ml. Duplicate scans were used in all equilibrium calculations.

Velocity sedimentation in sucrose density gradients was performed according to the method of Martin and Ames (33). Linear gradients of 10 to 30% sucrose contained 0.02 M potassium phosphate (pH 7.4), 0.1 mM dithiothreitol, 0.25 M KCl. Centrifugation was at 39,000 rpm at 4’ for 24 h in an SW 50.1 rotor in a Beckman L5-50 ultracentrifuge. Sedimentation markers were present in all gradients and included catalase (11.3 S), yeast alcohol dehydrogenase (7.4 S), and bovine hemoglobin dimer (2.8 S).

Other Methods-Cary 14 and Beckman model 25 spectrophotome- ters were employed for optical measurements. Protein spectra at neutral pH were determined at 5” in 0.02 M potassium phosphate (pH 7.4), 0.25 M NaCl, 0.1 mM EDTA. Protein content of partially purified fractions was determined according to the method of Lowry et al. (34), while the concentration of homogeneous EcoRI endonuclease was determined by amino acid analysis. Unless specified otherwise, DNA concentrations are expressed as equivalents of nucleotide. Buffer pH was determined at room temperature at a concentration of 0.05 M. Radioactivity was determined by liquid scintillation counting in a Beckman LS230 spectrometer.

RESULTS

Purification of EcoRI Endonuclease

A summary of a purification from 3.2 kg of Escherichia coli RY13 is presented in Table I. All steps were performed at O-4”, and centrifugation was at 12,000 x g for 20 to 30 min in Sorvall GSA or GS3 rotors.

Preparation of Extract and Streptomycin Fractionation

After thawing overnight at 4”, 3.2 kg of E. coli RY13 cell paste were suspended in 9.6 liters of 0.02 M potassium phosphate (pH 7.0), 7 mM 2-mercaptoethanol, 1 mM EDTA. Cells were disrupted by two passages through a Manton- Gaulin homogenizer at 9000 pounds. The homogenizer effluent

TABLE I

Purification of EcoRI endonuclease from 3.2 kg of Escherichia coli RY13

Fraction Total protein

Specific activity Recovery

I

II

III IV V

Streptomycin supernatant

Ammonium sulfate

Phosphocellulose Hydroxylapatite DNA cellulose

mg un1tslmg I. protein

160,000 19 (100)

32,000 51 55

640 2030 43 37 20,000 25 27.2” 22,400 21

a Protein determined by amino acid analysis.

had a temperature of 12-15”, and was quickly cooled to 4” in an ice-salt bath. The extract was clarified by centrifugation and adjusted to an A,,, of 215 with the above buffer. The extract (16.1 liters) was treated with 3.22 liters of freshly prepared 25% streptomycin sulfate. After stirring for 30 min, the precipitate was removed by centrifugation. The supernatant (Fraction I) had a ratio of A,,, to A,,, of 0.81.

Ammonium Sulfate Fractionation-Solid (NH,),SO, (6.75 kg) was added to Fraction I (17.3 liters) over a period of 20 min. The suspension was stirred for an additional 30 min, and placed at 0” overnight. The precipitate was collected by centrifugation, and then extracted successively with solutions of 45,40,35,30, and 25% saturation in (NH,) $0, (prepared by dissolving 277, 242, 208, 176, and 144 g of (NH,),SO,, respectively, per liter of 0.02 M potassium phosphate (pH 7.4), 5 mM

2-mercaptoethanol, 1 mM EDTA). Back-extraction was performed by suspending the precipitate in 5.2 liters of the appropriate (NH,),SO, solution, stirring for 20 to 30 min, followed by removal of insoluble material by centrifugation. EcoRI endonuclease was recovered in 25, 30, and 35% (NH,),SO, solutions.5 These fractions were pooled (15.7 liters) and concentrated by precipitation with (NH,),SO, (3.16 kg). The precipitate was suspended in 1300 ml of 0.02 M potassium phosphate (pH 7.4), 1 mM EDTA, 5 mM 2-mercaptoethanol. It was dialyzed against two 40.liter portions (2.5 h per change) of 0.02 M potassium phosphate (pH 7.4), 5 mM 2-mercaptoethanol, 10% (w/v) glycerol (Buffer A) containing 0.2 M KC1 and 0.5 mM EDTA (Fraction II).

Phosphocellulose Chromatography-Fraction II (1400 ml) was diluted to 5.6 liters with Buffer A containing 0.067 M KCl, 0.5 mM EDTA, and immediately applied at 600 ml/h to a phosphocellulose column (34 cm x 30 cm’) equilibrated with Buffer A containing 0.1 M KCl, 0.5 mM EDTA. The column was washed with 3 liters of Buffer A containing 0.2 M KCl, 0.5 mM

EDTA, and then eluted with a 28liter linear gradient of KC1 (0.2 to 0.75 M) in 0.02 M potassium phosphate (pH 7.4), 5 mM 2-mercaptoethanol, 0.1 mM EDTA, 10% glycerol. Fractions containing EcoRI endonuclease activity, which eluted about 0.6 M KCl, were pooled (Fraction III).

Hydroxylapatite Chromatography-Fraction III (8.0 liters) was applied at 150 ml/h to a column of hydroxylapatite (23.5 cm x 13.6 cm*) equilibrated with Buffer A containing 0.2 M

KCl. After washing with 280 ml of 0.15 M potassium phosphate (pH 7.4), 5 mM 2-mercaptoethanol, 10% glycerol (spectral

‘This step resolves EcoRI endonuclease from EcoRI methylase which is recovered primarily in 40% and 45% (NH,)*SO, solutions,


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


grade glycerol was employed in this and subsequent steps), the column was eluted at 120 ml/h with a 2.8-liter linear gradient

of potassium phosphate (pH 7.4, 0.15 to 0.80 M) containing 5 mM P-mercaptoethanol, and 10% glycerol. Active fractions, which eluted about 0.45 M phosphate, were pooled, and EDTA added to 1 mM (Fraction IV).

DNA-cellulose Chromatography-Fraction IV (265 ml) was

dialyzed against two 4-liter portions (2 h/change) of Buffer A containing 0.2 M NaCl, 5 mM 2.mercaptoethanol. Dialyzed enzyme was diluted with 0.33 volume of Buffer A and immediately applied at 65 ml/h to a column of denatured DNA-cellulose (10 cm x 3 cm*) equilibrated with Buffer A containing 1 mM EDTA and 0.1 M NaCl. After washing with 40 ml of the equilibrating buffer, the column was step eluted with 60-ml portions of Buffer A, 1 mM EDTA containing 0.25, 0.50, and 0.75 M NaCl. The endonuclease, which was recovered in 14 ml of the 0.50 M NaCl step, was made 50% (v/v) in glycerol

(Fraction V). Fraction V lost no detectable activity ( < 10%) over a period

of 7 months when stored at -20’. EcoRI endonuclease purified by this procedure was free of double strand endonuclease activity on T7 DNA which contains no RI sites (<2 pmol of double strand scissions/min/mg). The enzyme was also free of detectable DNA methylase activity on ColEl DNA ( <1 pmol/min/mg). However, Fraction V did contain low levels of endonuclease activity capable of introducing single strand breaks into ColEl DNA methylated at the RI site (50 pmol/ min/mg) or into PM2 DNA which lacks an RI site (400 pmol/min/mg). It is not yet clear whether this activity is due to a contaminant or to activity of EcoRI endonuclease at second- ary sites in these DNA molecules.

In addition, the enzyme was free of significant exonuclease activity as judged by two criteria. Unit length linears of ColEl DNA generated by cleavage with a 2-fold molar excess (as dimer) of Fraction V contained dAMP almost exclusively

(90%) as the 5’.terminal nucleotide (3), and could be extensively (85%‘) converted to covalently closed circles and linear oligomers (4) upon treatment with DNA ligase (not shown). Fraction V was used in all studies described below.

Physical and Chemical Properties of EcoRI Endonuclease

Electrophoretic Analysis-Fraction V was subjected to electrophoretic analysis under native and denaturing conditions. Only a single protein zone was detected on 5 or 7.5% polyacrylamide gels containing sodium dodecyl sulfate (Fig. 2), with a mobility of 0.88 * 0.01 or 0.71 i 0.01, respectively. Comparison with proteins of known molecular weight (20) indicates an apparent molecular weight for the denatured and reduced protein of 28,500 * 500 which is similar to the value of 30,000 reported by Greene et al. (6).

Electrophoresis under native conditions showed that this protein species does in fact possess EcoRI endonuclease activity (Fig. 3). Although Fraction V exhibited a single major protein zone on native gels, a significant fraction of the applied protein was consistently found between the top of the gel and the major band. A similar distribution was observed in the case

of catalytic activity recovered from unstained gels. Although endonuclease activity was clearly associated with the major

protein zone, significant activity was also found trailing to the top of the gel. As shown below, EcoRI endonuclease undergoes a concentration-dependent dissociation reaction, and we at- tribute the anomalous electrophoretic behavior under native conditions to this phenomenon. Based on these electrophoretic

Dye

1

FIG. 2. Sodium dodecyl sulfate electrophoresis of EcoRI endonuclease. Upper panel, 10 pg of Fraction V was subjected to electrophoresis on a gel of 7.5% polyacrylamide in the presence of sodium dodecyl sulfate. The gel was stained and scanned at 550 nM as described under “Methods.” Lower panel, the molecular weight of denatured and reduced EcoRI endonuclease was determined by comparison with protein standards of known molecular weight. Mobilities (relative to bromphenol blue) of protein standards are averages of three deter- minations, while those indicated for EcoRI endonuclease (closed circles) are averages of three (5% gels) or five (7.5% gels) measurements. Mobilities of the endonoclease determined in the presence of protein standards were identical with those observed in the absence of marker proteins. Pollis Escherichia coli DNA polymerase I and BSA is bovine serum albumin.

results, we have concluded that Fraction V is at least 97% pure. Additional evidence for homogeneity is presented below.

Amino Acid Analysis and Ultraviolet Spectra-The amino

acid composition of homogeneous EcoRI endonuclease is shown in Table II, and except for a slight deficiency of half-cystine, it is not unusual. From these data we have calculated a partial specific volume U of 0.736 for the protein at 25” (30, 31). The purified endonuclease exhibited typical protein ultraviolet spectra under native or alkaline conditions (not shown). The native spectrum was characterized by an absorption maximum at 278 nm, and using protein concentrations determined by amino acid analysis, we have determined the extinction coefficient (E:‘:,,, ) of the native protein to be 8.30 at this wavelength. The ratio of absorbance at 280 nm to that at 260 nm for the native protein was 1.86 to 1.90, ruling out significant contamination of Fraction V by nucleic acid.

Analytical Sedimentation of EcoRI Endonuclease-The native molecular weight of EcoRI endonuclease was determined by sedimentation equilibrium centrifugation at initial protein concentrations of 0.29 to 0.54 mg/ml. Plots of logarithm

of protein concentration versus r* were linear (Fig. 4) at several rotor speeds indicating physical homogeneity. A weight-aver-


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Physical and Catalytic Properties of EcoRI Endonuclease

FIG. 3. Correspondence of EcoRI endonuclease activity with major protein species. Samples of Fraction V (9.8 rg) were run on parallel gels under native conditions as described under “Methods.” One gel was stained and protein quantitated by scanning at 550 nm. The second gel was cut into 2-mm slices. After elution by overnight incubation at 0” in 0.2 ml of endonuclease dilution buffer, activity was determined as described under “Methods.” Recovery of activity was 25%.

TABLE II

Amino acid composition of EcoRI endonuclease

Amino acid Moles per A4, = 28,500

Alanine 15.8 Arginine 12.2 Aspartic 36.0 Half-cystine 1.50 Glutamic 25.3 Glycine 20.2 Histidine 4.90 Isoleucine 19.1 Leucine 24.8 Lysine 19.6 Methionine 4.50 Phenylalanine 9.93 Proline 5.84 Serine 19.5 Threonine 9.40 Tryptophan 1.86 Tyrosine 8.21 Valine 16.6

age molecular weight of 110,000 + 4,000 was calculated for the native enzyme from the slopes of such plots (Table III).

The sZo,W of the endonuclease determined by analytical sedimentation at an initial protein concentration of 0.25 mg/ml was 5.35 S. A single boundary was observed during sedimentation velocity analysis (not shown) providing additional evidence for purity of Fraction V. Therefore, under conditions of analytical sedimentation, the stable form of EcoRI endonuclease is a 5.4 S tetramer composed of four 28,500 M, subunits.

Concentration-dependent Dissociation of EcoRI Endonu- clease-Since analytical sedimentation was performed at endonuclease concentrations on the order of lo5 times greater than those used in catalytic assays, we tested the possibility that a different oligomeric state of the enzyme might predomi- nate under more dilute conditions. In these experiments, the enzyme was analyzed by band sedimentation in sucrose density gradients over a broad range of concentration (Fig. 5). Two distinct forms of the enzyme were observed. The more rapidly sedimenting species had an sZ,,,, of 5.3 S, a value identical with that observed in the analytical ultracentrifuge,

I I I I I I I I I I I

46.0 46.5 47.0 47.5 40.0 40.5 49.0 49.5 50.0 r*

FIG. 4. Sedimentation equilibrium analysis of EcoRI endonuclease. Sedimentation was performed as described under “Methods” at 8240 rpm at 10.3’ for 43 h.

TABLE III

Sedimentation equilibrium centrifugation of EcoRI endonuclease

Centrifugation was performed as described under “Methods” for 24 h (Experiment 1) or 43 h (Experiments 2 to 4). Molecular weights were calculated from plots of In c versus rZ (29, 35), which in all cases were linear throughout the cells.

Experiment Speed Temperature M

rpm 9211 8250 8250 8240

14.1” 114,000 10.4 110,000 10.4 111,000 10.3 104,000

6

FIG. 5. Concentration dependence of sedimentation coefficient of EcoRI endonuclease. Samples of EcoRI endonuclease at various dilutions along with sedimentation markers were subjected to velocity centrifugation on sucrose density gradients as described under “Methods.” Recovery of endonuclease activity from gradients ranged from 65 to 95% except in the case of the three most dilute samples where irecovery was only 5%. The concentration of EcoRI endonuclease shown on the abscissa is the weight-average concentration present in the endonuclease peak at the time of collection, and was calculated from endonuclease activity assuming a specific activity of 22,000 units/mg. In gradients represented by open circles, 3.8 S and 5.3 S forms of the endonuclease were both present. In such gradients the minor component was present as a partially resolved shoulder of the major peak; however, for clarity only the sedimentation coefficient of the predominant species is indicated.

and predominated at enzyme concentrations above 25 pg/ml (concentrations cited are weight-average concentrations in the endonuclease peak at termination of centrifugation). In contrast, at enzyme concentrations of 1 rig/ml to 10 pg/ml the endonuclease sedimented more slowly, with an average sedimentation coefficient of 3.8 S.

If spherical, the 3.8 S form would correspond to a protein of molecular weight of 42,000 (35), clearly too large for a monomer. However, if we make the reasonable assumption that frictional coefficients of 5.4 S and 3.8 S forms of the enzyme are


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


similar, then with a sedimentation coefficient of 5.4 S for the tetramer, the approximate relation (33)

M, S, 3J2 -= -

0 M, &

indicates a molecular weight for the 3.8 S species of 67,000. Since this value is reasonably close to 57,000, we have concluded that the 3.8 S form is a dimer composed of two A4, = 28,500 subunits.

A striking feature of the sucrose gradient experiments is that at intermediate protein concentrations, 3.8 S and 5.4 S forms of the enzyme were both present in individual gradients (Fig. 5, open circles), suggesting that equilibration of the two forms is slow relative to separation by sedimentation. Slow interconver- sion of the two forms in these gradients would result if dimer association were slow relative to transport, or if dimer association and tetramer dissociation were both slow reactions (35). At present, however, we cannot distinguish between these possibilities. Although sucrose gradient centrifugation suggests that the 3.8 S form predominates under dilute conditions, the apparent nonequilibrium nature of these experiments pre- cludes calculation of the equilibrium constant relating the two forms of the enzyme.

Thus, EcoRI endonuclease can exist in at least two forms in solution: a M, = 114,000 5.4 S tetramer or a M, = 57,000 3.8 S dimer. Furthermore, the results of sedimentation equilibrium centrifugation (Fig. 4 and Table III) place an upper limit on the equilibrium constant for tetramer dissociation (&). At the endonuclease concentrations employed in these experiments, a K, larger than 1 x lo-’ M would have resulted in a weight-average molecular weight significantly less than the value of 110,000 + 4,000 obtained, and in addition would have imparted detectable curvature to plots of In c uersus r2.

For example, the lowest enzyme concentration achieved at the meniscus during sedimentation equilibrium centrifugation was 56 fig/ml at a rotor speed of 9211 rpm (Table III). In this experiment, the concentration at the center of the fluid column was 170 @g/ml while at the cell bottom it was 490 fig/ml. Assuming molecular weights of 114,000 for the tetramer and 57,000 for the dimer, a K, of 1 x lo-’ M predicts that at these concentrations the endonuclease would be 20, 12, and 7% dimer by weight, respectively. These values correspond to weight-average molecular weights of 102,000 at the meniscus, 107,000 at the center, and 110,000 at the cell bottom. The corresponding molecular weights for a K,, of 2 x lo-’ M would be 98,000, 104,000, and 108,000; while for a constant of 4 x lo-’ M they would be 93,000, 101,000, and 106,000. Although molecular weights calculated using a Kd of 1 x 10m7 M may be within experimental error of the measured value (110,000 f 4,000), those calculated using a Kd of 4 x lo-’ M are clearly too low. Furthermore, the degree of dependence of molecular weight on concentration associated with a Kd of 4 x lo-’ M

would have imparted detectable curvature to the plot of In c uersus rz. This was not observed. Consequently, the equilibrium constant for the tetramer to dimer transition is less than or approximately equal to 1 x lo-’ M under conditions of sedimentation.

Catalytic Properties of EcoRI Endonuclease

Steady State Kinetic Analysis-The restriction enzyme responsible for E. coli B host specificity does not turn over in the endonuclease reaction in vitro (36). In contrast, published

specific activities for highly purified EcoRI endonuclease (6) suggest that this type II enzyme does function catalytically. To determine the extent of turnover in vitro and to explore the kinetic properties of the enzyme, we have examined the steady state kinetics of EcoRI endonuclease using covalently closed ColEl DNA circles (one EcoRI recognition site) as substrate. As shown in Fig. 6, the endonuclease obeys Michaelis-Menten kinetics with a K, for DNA recognition sites of 8 nM. The enzyme turns over in uitro, with a turnover number, calculated on the basis of the dimer, of 3.8 double strand scissions per min at 37’.

The assay employed for kinetic analysis measures conversion of ColEl DNA circles to a form sensitive to digestion by recBC DNase. Since this nuclease cannot initiate attack at single strand scissions (12), the assay only scores double strand cleavage to yield linear molecules. However, as shown below, there is no significant accumulation of substrate molecules containing a single strand break at 37’ under standard steady state reaction conditions. Hence, the recBC assay scores all endonucleolytic events under these conditions.

Mechanism of Double Strand Cleauage-Using ColEl DNA as substrate, we have examined reactions performed at limiting endonuclease concentrations for the presence of a DNA intermediate containing only one single strand break in the RI site. At temperatures of 15” or lower, DNA circles containing at least one single strand scission did accumulate, and at 0” were the major reaction product (Fig. 7). Since these breaks could be quantitatively (95%) closed by E. coli DNA ligase (not shown), single strand scission produced 3’.hydroxyl and 5’-phosphoryl termini (37). Furthermore, there was no detectable endonuclease activity on PM2 DNA circles (no RI site) or on ColEl DNA circles with a methylated RI site at 0” under conditions where extensive cleavage of ColEl DNA occurred (Table IV). Since DNA sites which are substrates for EcoRI endonuclease are also substrates for EcoRI methylase (5), we infer that single strand breaks introduced at low temperature are located within the RI recognition sequence, with one break being introduced per RI site.6 In addition, since molecules containing a single strand break accumulated at 0” to concentrations exceeding that of the endonuclease (as dimer) present, it is clear that they can be released from the enzyme under these conditions.

In contrast to results obtained at low temperature, there was no significant accumulation of a DNA intermediate containing one single strand interruption under steady state conditions at temperatures of 22-37”. A typical result obtained at 30” is shown in Fig. 7. Therefore, the two endonucleolytic events required for double strand cleavage are tightly coupled at higher temperatures.

Two distinct mechanisms can account for this coupling. Subsequent to introduction of one single strand break into the RI site, the DNA intermediate may be released from the

‘We have attempted to determine whether. the single strand break introduced into ColEl DNA at low temperature is in the RI site by the more direct method of labeling the 5’-phosphate at the break with “P using polynucleotide kinase after phosphatase treatment (2). I f the break is at the RI site, then this phosphate residue should be present at the 5’ terminus of the unit length linear duplex generated by subsequent cleavage with the endonuclease. Unfortunately, our at- tempts to label this phosphate have not been successful. This was not a consequence of inadequate phosphatase treatment since phosphatase- treated molecules could no longer be closed by DNA ligase, indicating successful removal of the 5’-phosphate at the single strand break. It is possible that this break is of the class resistant to polynucleotide kinase observed by Weiss et al. (38).


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


l/[ColEl] hM-‘1

FIG. 6. Steady state kinetic analysis of EcoRI endonuclease reaction. Assays were performed by a two-step modification of the procedure described under “Methods.” Reaction mixtures (0.2 ml) contained 0.1 M Tris/HCI (pH 7.6), 0.05 M NaCI, 100 &ml of bovine serum albumin, 5 mM MgCl*, 1.15 ng of EcoRI endonuclease, and indicated concentrations of ColEl [3H]DNA. After 10 min at 37”, the reaction was terminated by adding 5 ~1 of 0.3 M EDTA and heating at 6.5” for 3 min. Samples (0.02 ml) were then treated with an excess of recBC DNase in reactions (0.2 ml final volume) containing the follow- ing additional components: 0.1 M Tris/HCl (pH 7.6), 6 mM MgC12, 100 pM ATP, 0.25 mM dithiothreitol, and 13.5 units of recBC DNase. After 20 min at 37”, acid-soluble nucleotide was determined (“Methods”). Concentration of DNA is expressed in terms of molecules.

enzyme to be preferentially attacked by other enzyme molecules in solution. Alternatively, both breaks may be introduced by a single enzyme molecule without release of substrate from the enzyme surface. To distinguish between these possibilities, we have compared the rate of digestion of ColEl DNA containing a single strand break in the RI site with that of covalently closed molecules present in the same reaction (Table V). At 30”, covalently closed circles were cleaved at a rate 1.3 times that of circles containing a break in the RI site. Since the concentration of closed circles in this experiment was 7-fold lower than that used in the experiment shown in Fig. 7, the mechanism involving preferential attack of a DNA intermediate free in solution can clearly be ruled out. Thus, at 30”

EcoRI endonuclease cleaves both strands at the recognition site prior to dissociation from DNA.

Although DNA containing a single break in the RI site does not accumulate free in solution at 30”, it can be detected as a transient intermediate during a single turnover of the enzyme at this temperature. In these experiments ColEl DNA circles were incubated with a Y-fold molar excess of endonuclease (as dimer) in the presence of EDTA, conditions under which the enzyme binds to its substrate, but is unable to initiate endonucleolytic attack.’ Cleavage was then initiated by addition of magnesium ion. As shown in Fig. 8, 98% of the closed circles were cleaved within 5 s after addition of MgCl,, although 49% of the molecules had suffered only a single strand break. Circles containing a single strand interruption then decayed rapidly to yield the linear form. Since there was no detectable cleavage of methylated ColEl DNA under identical

‘P. Modrich and D. Zabel, unpublished experiments

16 32 40 64 00

60 l Form I

60

k

40

Form III

20 Form I[

0 0 2 4 6 6 IO

Time (min)

FIG. 7. Single strand scission by EcoRI endonuclease at low temperature. Reaction mixtures (0.15 ml) contained 0.02 M potassium phosphate (pH 7.4), 0.05 M NaCl, 1 mM EDTA, 5 mM M&l,, 0.060 mM ColEl [3T]DNA, and 0.5 unit/ml of EcoRI endonuclease (0.084 mol of endonuclease as dimer/mol of DNA circles). Incubation was at 0” or 30” as noted. Samples (0.02 ml) were removed at indicated times and added to 0.02 ml of cold 0.02 M EDTA (pH S.O), 20%) sucrose, 0.025% bromphenol blue to terminate the reaction. DNA species were separated by agarose gel electrophoresis and “T quantitated as described under “Methods.” Form I, covalently closed duplex DNA circles; Form II, circles containing at least one single strand break; Form III, unit length linear DNA. Data are not corrected for 5% Form II present in the DNA preparation used as substrate.

conditions during the time period shown, cleavage of ColEl DNA presumably occurred at the RI site.

If the sodium dodecyl sulfate-EDTA procedure for terminat- ing the reaction was fast relative to catalytic steps, single turnover experiments provide kinetic parameters characteris- tic of enzyme-bound intermediates in the endonuclease reaction. Two control experiments (not shown) indicate that the procedure was fast and effective. When the sodium dodecyl sulfate-EDTA stop solution was added simultaneously with a magnesium ion to endonuclease.DNA complexes, there was no detectable ( <4R) endonucleolytic cleavage of the DNA circles, even after a 30-min incubation at 30”. More significantly, when magnesium ion was added simultaneously with only the detergent component of the stop solution, again there was no detectable single or double strand cleavage of substrate. The latter result clearly shows that sodium dodecyl sulfate inacti-

vates DNA-bound endonuclease much more rapidly than the enzyme can introduce the first single strand break. It therefore seems likely that the stop solution was effective for rapid termination of the reaction and consequently, that the data of

Fig. 8 provide a valid picture of a single turnover of enzyme. Since introduction of the first single strand break by DNA

bound enzyme was 98% complete within 5 s after initiation of the reaction, the half-time for this cleavage must be 1 s or less, corresponding to a first order rate constant of at least 40 min’. Introduction of the second single strand break occurred more slowly. As expected for an enzyme-bound intermediate, DNA circles containing one single strand break decayed with first order kinetics with a half-life of 3 s, corresponding to a first


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


TABLE IV I I I ,

Single strand scission occurs at RI site

Reactions were performed at a DNA concentration of 0.06 mM at 0” for 80 min. Endonuclease at 1 unit/ml corresponds to 0.17 mol of enzyme (as dimer)/mol of ColEl molecules. DNA species were separated by agarose gel electrophoresis as described under “Methods,” except in the case of PM2 DNA where electrophoresis was performed at 2.7 V/cm for 12 h. DNA was quantitated by the microdensitometer procedure (“Methods”). Other details were as described in the legend to Fig. 7. Control experiments in which methylated and unmethylated ColEl DNA were present in the same reaction showed that the inability of the endonuclease to attack methylated DNA was not due to the presence of an inhibitor of the enzyme in such preparations. Although not shown, the endonuclease was similarly inactive on PM2 or methylated ColEl DNA at 30”.

DNA

ColEl ColEl PM2 PM2 EcoRI methylated

ColEl EcoRI methylated

ColEl

EcoRI Form Form Form endonuclease I II III

unit/ml % total

0 95 5 <l 1 14 66 20 0 97 3 <l 1 97 3 <l 0 93 7 <l

1 93 7 <l

Time (set)

TABLE V

Relative rate of digestion of CoJEJ DNA and CoJEJ DNA containing single strand break in RI site

Reactions (0.20 ml) contained 0.02 M potassium phosphate (pH 7.41, 0.05 M NaCl, 1 mM EDTA, 5 mM MgC12, 100 pglml of bovine serum albumin, 9.2 pM covalently closed ColEl DNA (Form I), 13.4 KM ColEl DNA containing a single strand break at the RI site (Form II), 5.3 PM linear ColEl DNA, and 0.2 unit/ml (Experiment 1) or 1.4 unit/ml (Experiment 2) EcoRI endonuclease. Incubation was at 30”. Samples (0.02 ml) were removed and added to 0.02 ml of 0.04 M EDTA (pH 8), 2% sodium dodecyl sulfate, 20% sucrose, 0.025% bromphenol blue. DNA species were separated by agarose gel electrophoresis and DNA quantitated by the microdensitometer procedure (“Methods”). Rates shown were determined from the linear portion of the time course.

FIG. 8. Enzyme-bound intermediate containing a single strand break. The reaction mixture (0.50 ml) contained 0.02 M potassium phosphate (pH 7.4), 0.05 M NaCl, 1 mM EDTA, 0.060 mM ColEl DNA, and 2.45 rg/ml of EcoRI endonuclease. After incubation at 30” for 3 min, a zero time sample was removed and then reaction initiated by addition of 0.01 ml of 0.25 M MgCl,. To ensure rapid mixing, this addition was made while vigorously mixing the reaction on a Vortex Mixer. Using two automatic pipettes (Gilson P200), 0.05ml samples were removed about 2.5 s prior to the indicated times. At the indicated times they were added to 0.05 ml of 0.04 M EDTA (pH 81, 2% sodium dodecyl sulfate, 20% sucrose, 0.025% bromphenol blue while mixing as above. We estimate sampling times to be accurate to + 1 s. Samples were immediately subjected to electrophoresis on agarose gels (“Methods”). DNA was quantitated by the microdensitometer procedure. The zero time sample contained 96% closed circles and 3.9% circles containing at least one single strand break (Form II). These Form II molecules were present in DNA not previously exposed to EcoRI endonuclease, and we assume their single strand breaks to be located at sites other than the RI site. Except in the case of the 5-s point where 2% closed circles remained, the only detectable DNA species present after the reaction was initiated were Form II circles and linear molecules. Although not shown, a parallel control reaction was performed with methylated ColEl DNA as substrate. There was no detectable cleavage of this DNA during the 20-s period shown; however, by 60 s, 5% of the molecules had suffered at least one single strand break.

Experiment

1

2

DNA

Form I Form II

Form I Form II

Rate

pmollmin

0.033 0.026

0.40 0.31

require the addition of detergent to prevent precipitation of the

enzyme (6), and hence yields a preparation suitable for physical analysis.”

order rate constant of 14 min’. At 30” in the phosphate buffer

in which these experiments were performed, the turnover number of the endonuclease was determined to be 0.72 double strand scissions per min (not shown). If the majority of enzyme molecules in our preparation are active, then this turnover number accurately reflects the catalytic constant of the enzyme. If this is assumed to be the case, then our results demonstrate that the two endonucleolytic events are fast relative to turnover, thus indicating that neither cleavage can be rate-limiting in catalysis.

Under denaturing conditions, the endonuclease behaves as a single species of molecular weight 28,500; a result in agreement with that of Greene et al. (6). While this suggests that the enzyme is composed of a single polypeptide species, the presence of two distinct polypeptide chains of similar molecular weight cannot be ruled out. The enzyme is present in the cell in relatively large quantities. Based on a subunit molecular weight of 28,500 and a specific activity of 22,000 units/mg, we estimate that there are 2,000 monomer equivalents per cell.

The identification of endonuclease dimers and tetramers in solution raises questions concerning the biological activities of these two forms of the enzyme. Although both forms can give rise to endonuclease activity (Fig. 5), it is not clear whether both are, in fact, catalytically active. The upper limit of 1 x

lo-’ M that we can place on the Kd for tetramer dissociation is not sufficiently low to permit identification of the ther-

DISCUSSION

The purification procedure for EcoRI endonuclease described here is simple, reproducible, and leads to isolation of

homogeneous enzyme in good yield. The method does not

‘Although we have not observed precipitation of the endonuclease, we did find that 25 to 40% of the protein was consistently lost when Fraction V was dialyzed against glycerol-free buffers in preparation for sedimentation analysis. However, the protein remaining had a specific activity identical with that of Fraction V, was stable at 0” for at least 1 week, and as evidenced by results of sedimentation analysis, showed no tendency to precipitate.


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/


modynamically stable form of the enzyme at the low concentrations employed for catalytic assay. The results of sucrose gradient centrifugation of the enzyme (Fig. 5) suggest that the dimer predominates under dilute conditions; however, since the two forms of the endonuclease apparently were not in equilibrium in these experiments, a conclusion on this point is not possible at present. It is interesting to note, however, that if the limit we place on Kd is valid in uiuo, then the tetramer would constitute a large fraction of the endonuclease within the cell. Assuming the volume of the E. coli cell to be 2 x 10-“cm3 (39, 40), 2,000 monomer equivalents per cell correspond to an intracellular endonuclease concentration of 50 pg/ml. At this concentration, a Kd of 1 x lo-’ M predicts 80% tetramer by weight.

While the endonuclease clearly turns over in uitro, the turnover number is surprisingly low. This finding does not reflect inactivation of enzyme during purification since recovery of activity at each step was 55% or better, with significant activity being discarded in choosing the most active fractions. A more reasonable interpretation is that the low turnover number reflects an inherently low catalytic constant of the enzyme itself. Additional support for this conclusion has been provided by Greene et al. (41) who recently demonstrated that the enzyme turns over on an SV40 DNA substrate. They determined the turnover number at 15” to be 1.5 double strand events per min with a K, for substrate of 3 x 1O-8 M. Although strict comparison of these kinetic parameters with those reported here for the ColEl substrate is not possible due to differences in reaction temperature, the turnover numbers obtained with the two substrates appear to be in reasonable agreement.

Our results show that the endonuclease reaction at 30” proceeds by a two-step mechanism via an enzyme-bound intermediate containing a single phosphodiester bond interruption in the RI site. Although introduction of the second break into the recognition sequence is slow relative to the first, neither endonucleolytic event is slow enough to be rate-limiting. Therefore, the rate-determining step must occur after introduction of the second break within the RI site. Thus, at 30” the slow step in the reaction can be viewed as release of enzyme from its DNA product.

In contrast to the reaction path at 30”, EcoRI endonuclease dissociates from the DNA intermediate in reactions performed at 0”. This implies that the second endonucleolytic cleavage within the RI site is slow relative to enzyme dissociation at low temperature. While the reason for this dependence of reaction mechanism on temperature is not yet clear, the low temperature effect does extend the utility of the enzyme as a reagent by providing a simple method for introduction of site-specific single strand scissions into DNA.

REFERENCES 1. Yoshimori, R. N. (1971) Ph.D thesis, University of California, San

Francisco, Dissertation Abstracts Order No. 73.3662 2. Smith, H. O., and Nathans, D. (1973) J. Mol. Viol. 81, 419-423 3. Hedgpeth, J., Goodman, H. M., and Boyer, H. W. (1972) Proc.

Natl. Acad. Sci. U. S. A. 69, 3448-3452

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14. 15.

16.

17.

18.

19. 20. 21.

22. 23.

24.

25.

26. 27. 28.

29.

30.

31.

32. 33.

34.

35.

36. 37. 38.

39.

40.

41.

Mertz, J. E., and Davis, R. W. (1972) Proc. Natl. Acad. Sci. U. S. A. 69, 3370-3374

Dugaiczyk, A., Hedgpeth, J., Boyer, H. W., and Goodman, H. W. (1974) Biochemistry 13. 503-512

Greene, P. J., Betlach, M. C., Goodman, H. M., and Boyer, H. W. (1974) Methods Mol. Biol. 7, 87-111

Clewell, D. B., and Helinski, D. R. (1972) J. Bacterial. 110, 1135-1146

Modrich, P., Anraku, Y., and Lehman, I. R. (1973) J. Biol. Chem. 248, 7495-7501

Jovin, T. M., Englund, P. T., and Bertsch, L. L. (1969) J. Biol. Chem. 244, 2996-3008

Richardson, C. C. (1965) Proc. Natl. Acad. Sci. U. S. A. 54, 158-165

Wright, M., Buttin, G., and Hurwitz, J. (1971) J. Biol. Chem. 246, 6543-6555

Karu, A. E., MacKay, V., Goldmark, P. J., and Linn, S. (1973) J. Biol. Chem. 248, 4874-4884

Hinnen, R., Schafer, R., and Franklin, R. M. (1974) Eur. J. Biochem. 50, l-14

Wang, J. C. (1971) in l’roc. Nucleic Acid Res. 2, 407-416 Hinkle, D. C., and Chamberlin, M. J. (1972) J. Mol. Biol. 70,

157-185 Benzer, S., and Champe, S. (1961) Proc. Natl. Acad. Sci. U. S. A.

47, 1025-1038 Hanawalt, P. C., and Cooper, P. K. (1971) Methods Enzymol. 21,

221-230 Guerry, P., LeBlanc, D. J., and Falkow, S. (1973) J. Bacterial. 116,

1064-1066 Litman, R. M. (1968) J. Biol. Chem. 243, 6222-6233 Weber, K., and Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412 Jovin, T., Chrambach, A., and Naughton, M. A. (1964) Anal.

Biochem. 9, 351-369 Studier, F. W. (1973) J. Mol. Biol. 79, 237-248 Peacock, A. C., and Dingman, C. W. (1968) Biochemistry 7,

668-674 Depew, R. E., and Wang, J. C. (1975) Proc. Natl. Acad. Sci. U. S.

A. 72, 4275-4279 Lovett, M. A., Guiney, D. G., and Helinski, D. R. (1974) Proc.

Natl. Acad. Sci. U. S. A. 71, 3854-3857 Bray, G. A. (1960) Anal. Biochem. 1, 279-285 Moore, S. (1963) J. Biol. Chem. 238, 235-237 Beaven, G. H., and Holliday, E. R. (1952) Adu. Protein Chem. 7,

319-386 Chervenka, C. H. (1969) A Manual of Methods for the Analytical

Ultracentrifuge, Spinco Division, Beckman Instruments, Palo Alto, Cal.

Cohen, E. J., and Edsall, J. T. (1943) Proteins, Amino Acids and Peptides, Reinhold, New York

McMeekin, T. L., Groves, M. L., and Hipp, N. J. (1949) J. Am. Chem. Sot. 71, 3298

Bull, H. B., and Brese, K. (1968) J. Phys. Chem. 72, 1817 Martin, R. G., and Ames, B. N. (1961) J. Biol. Chem. 236,

1372-1379 . Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

(1951) J. Biol. Chem. 193, 265-275 Schachman, H. K. (1959) Ultrac&rifugation in Biochemistry, Aca-

demic Press, New York Eskin, B., and Linn, S. (1972) J. Biol. Chem. 247, 6183-6191 Lehman, I. R. (1974) Science 186, 790-797 Weiss, B., Live, T. R., and Richardson, C. C. (1968) J. Biol. Chem.

243, 4530-4542 Lundquist, R., and Olivera, B. M. (1971) J. Biol. Chem. 246,

1107-1116 Watson, J. D. (1965) Molecular Biology of The Gene, W. A.

Benjamin, Inc., New York Greene, P. J., Poonian, M. S., Nussbaum, A. L., Tobias, L.,

Garfin, D. E., Boyer, H. W., and Goodman, H. M. (1975) J. Mol. Biol. 99, 237-261


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

P Modrich and D ZabelEcoRI endonuclease. Physical and catalytic properties of the homogenous enzyme.

1976, 251:5866-5874.J. Biol. Chem.

http://www.jbc.org/content/251/19/5866Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/251/19/5866.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/content/251/19/5866

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;251/19/5866&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/251/19/5866

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=251/19/5866&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/251/19/5866

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/251/19/5866.full.html#ref-list-1

http://www.jbc.org/

ecori endonuclease

Documents