Volume 3 no.11 November 1976
Nucleic Acids Research
A restriction endonuclease from Staphylococcus aureus
J. S. Sussenbach, C. H. Monfoort, R. Schiphof and E. E. Stobberingh*
Laboratory for Physiological Chemistry, Vondellaan 24 a, Utrecht, The Netherlands
Received 15th September 1976 ABSTRACT A specific endonuclease, Sau 3AI, has been partially purified from Staphylococcus aureus strain 3A by DEAE-cellulose chromatography. The enzyme cleaves adenovirus type 5 DNA many times, SV40 DNA eight times but does not cleave double-stranded 4X174 DNA. It recognizes the sequence 3 and cleaves as indicated by the arrows. Evidence is present' ed that this enzyme plays a role in the biological restriction-modification system of Staphylococcus aureus strain 3A.
-C-T-cA-G
INTRODUCTION Recently, a number of endonucleases has been isolated which recognize specific nucleotide sequences in double-stranded DNA and cleave the DNA at these sites (1). These so-called r'estriction endonucleases facilitate the specific fragmentation of double-stranded DNA and are very useful for DNA sequence analysis and the unraveling of *the genetic organization of viral chromosomes. Although their name suggests that these enzymes play a role in restriction-modification it is noteworthy that for most specific endonucleases the physiological function is still unknown and that only for a few enzymes the actual involvement in biological restriction-modification has been demonstrated (2, 3, 4). Staphylococcus aureus is a Gram-positive bacterium, which possesses a biological restriction-modification system as shown by the analysis of the propagation of different types of bacteriophages in a number of strains of this bacterium (5; Stobberingh and Winkler, in press). The presence of a specific restriction endonuclease, however, has not yet been demonstrated. This communication describes the partial purification and characterization of a specific endonuclease from Staphylococcus aureus strain 3A, which is
involved in restriction-modification.
G Information Retrieval Limited 1 Falconberg Court London Wl V 5FG England
3193
Nucleic Acids Research MATERIALS AND METHODS Materials
Lysostaphin (Schwarz-Mann, New York) was dissolved in 0.05 M Tris-HCl, 0.145 M NaCl, pH 7.4 as a stock solution of 200 U/ml and stored at -200 C. Strains of bacteria and bacteriophages Staphylococcus aureus strains 3A (NCTC 8319) and 3AR1 (a restrictiondeficient mutant) were examined for the presence of specific endonucleases. Partial characterization of the enzymes was performed with DNA's isolated from Staphylococcal phages 6 (NCTC 8403) and 3A (NCTC 8408), adenovirus type 5, SV40 and 4X174.
DNA
prearations
DNA from adenovirus type 5 was isolated as described previously (6). Double-stranded fX174 DNA and SV40 DNA were generous gifts from Dr.P.D. Baas (Utrecht) and Dr. J. ter Schegget (Amsterdam), respectively. Staphylococcal phage DNA was isolated from purified virions. Phage 3A was propagated in propagating strain (PS) 3A or PS 6 and phage 6 in PS 6 as described by Blair and Williams (7). Phage suspensions were centrifuged at 2000 x g for 30 min to remove cellular debris and subsequently sedimented for 2 h at 90,000 x g. The pellet was resuspended in phosphate buffered saline (PBS) plus 0.02% EDTA, pH 7.6, then pronase (0.1 ml; 10 mg/ml) was added to 5 ml of the concentrated phage suspension and the mixture was incubated for I h at 370 C. Subsequently 0.8 ml 2% sodium dodecylsulphate solution was added and the mixture kept for 5 min at room temperature. Finally, the DNA was purified by extraction with phenol saturated with PBS and remaining phenol was removed by dialysis against 20 mM Tris-HCl, 0.5 mM EDTA, pH 7.5.
Preparation
lYs2staphin-treated cells The method employed was a modification of a procedure described by Klesius and Schuhardt (8). Nutrient Broth (Difco; 3600 ml) was inoculated of
with 400 ml of an overnight culture of Staphylococcus aureus. After incubation for 4 h at 370 C, the cells were spun down and washed three times with 0.05 M Tris-HC1, 0.015 M trisodiumcitrate, pH 7.4 and resuspended in 4 ml of the same solution. Lysostaphin was added to a final concentration of 5 U/ml. After incubation for 15 min at 370 C on a rotating table the cells were resuspended in 0.01 M Tris-HCl, 0.01 M a-mercaptoethanol, pH 7.4.
3194
Nucleic Acids Research Endonuclease assay and gel electroEhoresis Digestions were made in 15 mM MgC12, 6 mM Tris-HCl, pH 7.5, 6 mM 0-mercaptoethanol, 60 mM NaCl at 300 C. After incubation sodium acetate was added to 0.1 M and the DNA fragments were deproteinized by extraction with chloroform-isoamylalcohol (24:1). Then DNA was precipitated by addition of 2.7 volumes of ethanol and centrifugation at -5° C in a Spinco SW65 rotor at 35,000 rpm for 30 min. The DNA pellet was dissolved in a small volume of 20 mM Tris-HCl pH 7.5, 1 mM EDTA and subjected to electrophoresis in 1.4% agarose gels (Sussenbach and Kuijk, submitted).
Seg2uence analysis
of the 5'-termini of Sau 3AI fra&ments 3A_ Sau were produced by digestion of 7.5 AdS DNA with fragments _g Sau 3AI. After deproteinization by extraction with chloroform-isoamylalcohol (24:1) and ethanol precipitation of the DNA, the fragmented DNA was brought in 100 pl 40 mM Tris-HCl pH 8.5, 10 mM MgC12 and 10 mM dithiothreitol. Then 2 pg of bacterial alkaline phosphatase (Worthington, Freehold, N.J.) was added and the DNA was incubated for I h at 370 C. Subsequently the fragmented DNA was deproteinized and precipitated as described above and again taken up in 100 pl of 40 mM Tris-HCl pH 8.5, 10 mM MgC12 and 10 mM dithiothreitol. Then 10 pl (10 U) T4 polynucleotide kinase (Biolabs, Beverly, Ma) and 5 pl y-32P-ATP (5 pCi; 16 C/mmole) (The Radiochemical Centre, Amersham, England) was added and the mixture incubacted for 1 h at 370 C (9). DNA was then precipitated by addition of 2 ml of 10% trichloroacetic acid, 0.01 M PPi at 0° C and was centrifuged for 25 min at 30,000 rpm in a Spinco SW65 rotor. The supernatant was discarded and the precipitate washed with trichloroacetic acid as described above. This procedure was repeated five times. Finally, trichloroacetic acid was removed by three washings with 90% ethanol. The pellet was dissolved in 50 pl H20. For the analysis of the 5'-terminal nucleotide 0.15 pg 32P-phosphorylated DNA was digested with 10 pg pancreatic DNase (Worthington, Freehold, N.J.) in 10 mM Tris-HCl, 10 mM MgC12, pH 7.5 for 2 h at 370 C. Subsequently pH was raised to 9.0 and 2 pg of snake venom phosphodiesterase (Boehringer, Mannheim) was added and incubation was continued for another 2 h at 370 C. Under these conditions DNA is completely digested to 5'-mononucleotides. Separation of the mononucleotides was performed by paper chromatography on Whatman no. I paper (10). The chromatogram was developed using saturated ammonium sulphate-1.0 M sodium acetate-isopropanol (80:18:2) as solvent. Detection of the mononucleotides was facilitated by cochromatography of
3195
Nucleic Acids Research unlabeled mononucleotides. The four nucleotide spots were visualized by UV-light, cut from the chromatogram and counted in a liquid scintillation counter. For the analysis of the recognition site 32P-phosphorylated DNA was partially digested with pancreatic DNase. The products were fractionated according to a standard two-dimensional fractionation procedure developed by Brownlee and Sanger (11) in a modification according to Volckaert, Min Jou and Fiers (12). In the 1st dimension fractionation was performed by electrophoresis at pH 3.5 on cellulose acetate and in the 2nd dimension by homochromatography on PEI-cellulose using an RNA homomix. The RNA homomix was prepared by hydrolysis of a 3% RNA (yeast RNA, sodium salt, BDH, Poole, U.K.) solution at pH 12.8 for 30 min at room temperature.
RESULTS
Isolation of an endonuclease Staphylococcus aureus 3A was grown as described in Materials and Methods. About 3 grams of packed cells were treated with lysostaphin and after centrifugation resuspended in 8 ml 0.01 M Tris-HCl, 0.01 M 8-mercaptoethanol, pH 7.4 (see Materials and Methods). It appeared that the length of the lysostaphin treatment is rather critical for the optimal detection of endonucleases. Extensive treatment of the cells with lysostaphin causes lysis of the cells and very low levels of endonuclease. After the short lysostaphin treatment the cells were disrupted by sonication (6 x 1 min with a Branson Sonifier) at 00 C and centrifuged for I1 h at 40,000 rpm in a Spinco SW41 rotor at 40 C. The supernatant was collected and streptomycin sulphate was added for the precipitation of nucleic acids (1.8 ml of a 10% streptomycin sulphate solution in 0.01 M Tris-HCl, 0.01 M 8-mercaptoethanol, pH 7.4). The precipitate was removed by centrifugation at 35,000 rpm for 30 min in a Spinco SW41 rotor at 40 C and the supernatant was dialyzed against 0.01 M Tris-HCl, 0.01 M a-mercaptoethanol, pH 7.4. During the dialysis a precipitate arose which was removed by lowspeed centrifugation. The crude enzyme preparation (9 ml) was brought on a DEAE-cellulose column (30 ml) which was equilibrated with 0.01 M KPO4 buffer, 0.01 M a-mercaptoethanol, 0.0001 M EDTA, pH 7.4. Elution was performed with a KC1 gradient (200 ml; 0.0-0.6 M KC1). Fractions of 5 ml were collected and tested for the presence of endonuclease as described in Materials and Methods.
3196
Nucleic Acids Research
Figure I DEAE-cellulose chromatography of Sau 3AI. Extract of Staphylococcus aureus strain 3A were prepared as described in Materials and Methods and applied to a 30 ml DEAE-cellulose column and eluted with 200 ml 0-0.6 M KC1 gradient. Fractions of 5 ml were collected and tested alternately for endonucleolytic activity. The gradient starts as indicated by the arrow. Elution is from left to right.
Endonucleolytic activity was eluted between 0.20 and 0.33 M KC1 (Fig. 1). Concentration of the enzyme containing fractions was performed by dialysis against 50% glycerol in 0.01 M KP04 buffer, pH 7.4, 0.01 M S-mercaptoethanol, 0.0001 M EDTA. The enzyme preparations were stored at -200 C and no detectable loss in endonucleolytic activity was detected after storage for several months. The yield of purified enzyme from I gram of packed cells was equal to about 1000 units (1 unit is the amount of enzyme required for complete digestion of I ig of adenovirus type 5 DNA in 2 h at 300 C).
Characterization of Sau 3AI The observation that extracts of Staphylococcus aureus strain 3A possess endonucleolytic activity led us to a further characterization of this specific endonuclease which was called Sau 3AI according to the proposed nomenclature for this category of enzymes (13). Sau 3AI was characterized by its action on adenovirus type 5 DNA, SV40 DNA and double-stranded *X174 DNA. The results are shown in Fig. 2. It
3197
Nucleic Acids Research 1
2
3
4
Figure 2 Gel electrophoresis of DNA's of different sources cleaved by Sau 3AI. DNA's were cleaved and subjected to gel electrophoresis in 1.4% agarose gels as indicated in Materials and Methods. Lane I shows digested SV40 DNA and lane 2 digested adenovirus type 5 DNA. Lane 3 and 4 display digested and undigested fXI74 RF DNA,respectively. The faster moving band represents component I and the slower moving band component II. appears that adenovirus type 5 DNA is cleaved at least 35 times, while double-stranded 4X174 DNA is not cleaved by this enzyme. In the latter case there is a conversion of the closed circular form (component I) to the open circular form (component II) detectable but this is probably due to an
aspecific contamination in the Sau 3AI preparation. SV40 DNA is cleaved into eight fragments of 27.2, 25.2, 17.9, 10.4, 6.7, 6.4, 4.2 and 2.0% of genome size, respectively. To investigate whether this enzyme was actually involved in biological restriction-modification, the action of Sau 3AI on phage 3A DNA propagated in PS 3A, on phage 6 DNA propagated in PS 6 and phage 3A DNA propagated in PS 6 was determined. Furthermore, the action of crude extracts of a restriction-deficient mutant 3ARl was studied for the presence of endonucleolytic activity (Fig. 3). Sau 3AI cleaves phage 6 DNA many times as well as phage 3A DNA propagated in PS 6, but does not cleave phage 3A DNA propagated in 3198
Nucleic Acids Research 1
2
3
4
Figure 3 Digestion of different staphylococcal phage DNA's with enzyme from Staphylococcus aureus strains 3A and 3AR . Electrophoresis was in 1.4% agarose as described in Materials and Methods. Lane 1: phage 6 DNA x extract 3AR-; lane 2: phage 6 DNA x Sau 3AI; lane 3: phage 3A DNA x Sau 3AI; lane 4: DNA from phage 3A propagated in PS 6 x Sau 3AI. PS 3A. Further, the crude extract of the restriction-deficient mutant 3ARdid not contain Sau 3AI activity as indicated by the fact that phage 6 DNA was not cleaved. These results indicate that Sau 3AI is required for res-
triction in PS 3A and that PS 3A and PS 6 contain modification enzymes with different specificities. Sau 3A1 cleaves DNA's with the PS 6 modification but is not able to cleave DNA's with the PS 3A modification. Characterization of the recognaition site The recognition site of Sau 3A1 was characterized by determination of the 5'-nucleotide sequence of the DNA fragments produced with this enzyme. For this purpose adenovirus type 5 DNA cleaved with Sau 3AI was first digested with bacterial alkaline phosphatase to remove 5'-phosphate groups, and subsequently labeled with 32P-phosphate using T4 polynucleotide kinase and y-32P-ATP leading to introduction of a terminal 5'-32P-phosphate group at the ends of the fragment strands (see Materials and Methods). Complete digestion of this 32P-labeled DNA to 5'-mononucleotides with pancreatic DNase and snake venom phosphodiesterase followed by paper chromatography of
3199
Nucleic Acids Research
pG
Figure 4 Two dimensional fractionation of oligonucleotides obtained after partial pancreatic DNase digestion of adenovirus type 5 DNA fragments produced by cleavage with Sau 3MI. The 5'-terminal nucleotides of the fragment strands were phosphorylated with 32P-phosphate using T4 polynucleotide kinase. In the first dimension (I fractionation is achieved by electrophoresis at pH 3.5 on cellulose acetate and in the second dimension (II) by homochromatography on PEI-cellulose. The nucleotide s.equence was derived from the known relative mobilities (14).
the nucleotides showed that 97% of the radioactivity was present in pG and 3% in pA. For a further analysis of the 5'-terminal nucleotide sequences of the DNA fragment 5'-terminal 32p-labeled DNA was partially digested with
pancreatic DNase. The digestion products were separated two-dimensionally by electrophoresis on cellulose acetate (1st dimension) followed by homochromatography on PET-cellulose (2nd dimension). The autoradiogram displayed unique mono-, di-, tri- and tetranucleotide spots and multiple penta- and hexanucleotide spots (Fig. 4). Analysis showed that the fragment strands terminated specifically with the sequence 5' G-A-T-C-N-N-. DISCUSSION
Staphylococcus
aureus strain 3A possesses a
biological restrictionmodification system as shown by phage typing experiments (5; Stobberingh and Winkler, in press). Extracts of Staphylococcus aureus 3A appear to conrtain a specific endonucleolytic activity which can be partially purified by 3200
Nucleic Acids Research DEAE-cellulose chromatography. This enzyme, Sau 3AI, is absent in extracts of a restriction-deficient mutant, indicating that it is involved in biological restriction-modification. This notion is supported by the observation that Sau 3AI does not cleave DNA from phage 3A, which has been propagated in PS 3A, while passage of phage 3A in PS 6 makes phage 3A DNA accesible for Sau 3AI cleavage. These observationsjustify the conclusion that Sau 3AI is a restriction endonuclease which plays a role in the restriction-modification system of this bacterium. Partial purification of Sau 3AI can easily be achieved by DEAE-cellulose chromatography while even crude extracts prepared as described in Results are almost free of aspecific nucleases. The enzyme is very stable at 40 C in crude extracts and at -20° C in 50% glycerol. Analysis of the 5'-termini of Sau 3AI fragments revealed the sequence 5' G-A-T-C-N-N-. This suggests that the enzyme recognizes the sequences 5'
-C-T-CA- 3,
and cleaves
as
indicated
by
the
arrows.
An identical recog-
nition site and cleavage has also been found for an endonuclease from Moraxeila bovis (Mbo I) (R. Gelinas, G.A. Weiss, R.J. Roberts, A. Morrison and K. Murray, in preparation). Comparison of DNA fragments of adenovirus type 5 and SV40 obtained by cleavage with Sau 3AI and Mbo I, respectively confirmed that these enzymes recognize the same nucleotide sequence. A specific restriction endonuclease from Diplococcus pneumoniae (Dpn II) also recognizes the above sequence (S. Lacks, personal communication).
ACKNOWLEDGEMENTS The authors thank J.M. Vereijken, A.D.M. van Mansfeld, P.H. Steenbergh and M.G. Kuijk for assistance and Profs. K.C. Winkler and H.S. Jansz for interesting discussions and critical reading of the manuscript. This investigation was supported in part by the Netherlands Foundation for Chemical Research with financial aid from the Netherlands Organization for
the Advancement of Pure Research.
*Laboratory of Microbiology, State University of Utrecht, Utrecht, The Netherlahds REFERENCES
1. Nathans, D. and Smith, H.O. (1975). Ann. Rev. of Biochemistry 44, 273293 2. Smith, H.O. and Wilcox, K.W. (1970). J. Mol. Biol. 51, 379-391 3. Yoshimori, R. (1971). A genetic and biochemical analysis of the restric-
3201
Nucleic Acids Research
4. 5. 6. 7.
8. 9. 10. 11.
12. 13. 14.
tion and modification of DNA by resistance transfer factors, PhD thesis Univ. Calif. San Francisco Bron, S., Murray, K. and Trautner, T.A. (1975). Molec. gen. genet. 143, 13-23 Stobberingh, E.E. and Winkler, K.C. (1975). Antonie van Leeuwenhoek 41, 212-213 Sussenbach, J.S. (1971). Virology 46, 969-972 Blair, J.E. and Williams, R.E.O. (1961). Bull. World Health Organization 24, 771 Klesius, P.H. and Schuhardt, V.T. (1968). J. Bacteriol. 95, 739-743 Richardson, C.C. (1971) in Procedures in Nucleic Acid Research Vol 2, p. 815-828, Harper and Row, New York Markham, R. and Smith, J.D. (1952). Biochem. J. 52, 552-565 Brownlee, G.G. and Sanger, F. (1969). Eur. J. Biochem. 11, 395 Volckaert, G., Min Jou, W. and Fiers, W. (1976). Anal. Biochem. 72, 433-445 Smith, H.O. and Nathans, D. (1973). J. Mol. Biol. 81, 419-423 Galibert, F., Ziff, E. and Sedat, J. (1974). J. Mol. Biol. 87, 377-407
3202
Nucleic Acids Research
A restriction endonuclease from Staphylococcus aureus
J. S. Sussenbach, C. H. Monfoort, R. Schiphof and E. E. Stobberingh*
Laboratory for Physiological Chemistry, Vondellaan 24 a, Utrecht, The Netherlands
Received 15th September 1976 ABSTRACT A specific endonuclease, Sau 3AI, has been partially purified from Staphylococcus aureus strain 3A by DEAE-cellulose chromatography. The enzyme cleaves adenovirus type 5 DNA many times, SV40 DNA eight times but does not cleave double-stranded 4X174 DNA. It recognizes the sequence 3 and cleaves as indicated by the arrows. Evidence is present' ed that this enzyme plays a role in the biological restriction-modification system of Staphylococcus aureus strain 3A.
-C-T-cA-G
INTRODUCTION Recently, a number of endonucleases has been isolated which recognize specific nucleotide sequences in double-stranded DNA and cleave the DNA at these sites (1). These so-called r'estriction endonucleases facilitate the specific fragmentation of double-stranded DNA and are very useful for DNA sequence analysis and the unraveling of *the genetic organization of viral chromosomes. Although their name suggests that these enzymes play a role in restriction-modification it is noteworthy that for most specific endonucleases the physiological function is still unknown and that only for a few enzymes the actual involvement in biological restriction-modification has been demonstrated (2, 3, 4). Staphylococcus aureus is a Gram-positive bacterium, which possesses a biological restriction-modification system as shown by the analysis of the propagation of different types of bacteriophages in a number of strains of this bacterium (5; Stobberingh and Winkler, in press). The presence of a specific restriction endonuclease, however, has not yet been demonstrated. This communication describes the partial purification and characterization of a specific endonuclease from Staphylococcus aureus strain 3A, which is
involved in restriction-modification.
G Information Retrieval Limited 1 Falconberg Court London Wl V 5FG England
3193
Nucleic Acids Research MATERIALS AND METHODS Materials
Lysostaphin (Schwarz-Mann, New York) was dissolved in 0.05 M Tris-HCl, 0.145 M NaCl, pH 7.4 as a stock solution of 200 U/ml and stored at -200 C. Strains of bacteria and bacteriophages Staphylococcus aureus strains 3A (NCTC 8319) and 3AR1 (a restrictiondeficient mutant) were examined for the presence of specific endonucleases. Partial characterization of the enzymes was performed with DNA's isolated from Staphylococcal phages 6 (NCTC 8403) and 3A (NCTC 8408), adenovirus type 5, SV40 and 4X174.
DNA
prearations
DNA from adenovirus type 5 was isolated as described previously (6). Double-stranded fX174 DNA and SV40 DNA were generous gifts from Dr.P.D. Baas (Utrecht) and Dr. J. ter Schegget (Amsterdam), respectively. Staphylococcal phage DNA was isolated from purified virions. Phage 3A was propagated in propagating strain (PS) 3A or PS 6 and phage 6 in PS 6 as described by Blair and Williams (7). Phage suspensions were centrifuged at 2000 x g for 30 min to remove cellular debris and subsequently sedimented for 2 h at 90,000 x g. The pellet was resuspended in phosphate buffered saline (PBS) plus 0.02% EDTA, pH 7.6, then pronase (0.1 ml; 10 mg/ml) was added to 5 ml of the concentrated phage suspension and the mixture was incubated for I h at 370 C. Subsequently 0.8 ml 2% sodium dodecylsulphate solution was added and the mixture kept for 5 min at room temperature. Finally, the DNA was purified by extraction with phenol saturated with PBS and remaining phenol was removed by dialysis against 20 mM Tris-HCl, 0.5 mM EDTA, pH 7.5.
Preparation
lYs2staphin-treated cells The method employed was a modification of a procedure described by Klesius and Schuhardt (8). Nutrient Broth (Difco; 3600 ml) was inoculated of
with 400 ml of an overnight culture of Staphylococcus aureus. After incubation for 4 h at 370 C, the cells were spun down and washed three times with 0.05 M Tris-HC1, 0.015 M trisodiumcitrate, pH 7.4 and resuspended in 4 ml of the same solution. Lysostaphin was added to a final concentration of 5 U/ml. After incubation for 15 min at 370 C on a rotating table the cells were resuspended in 0.01 M Tris-HCl, 0.01 M a-mercaptoethanol, pH 7.4.
3194
Nucleic Acids Research Endonuclease assay and gel electroEhoresis Digestions were made in 15 mM MgC12, 6 mM Tris-HCl, pH 7.5, 6 mM 0-mercaptoethanol, 60 mM NaCl at 300 C. After incubation sodium acetate was added to 0.1 M and the DNA fragments were deproteinized by extraction with chloroform-isoamylalcohol (24:1). Then DNA was precipitated by addition of 2.7 volumes of ethanol and centrifugation at -5° C in a Spinco SW65 rotor at 35,000 rpm for 30 min. The DNA pellet was dissolved in a small volume of 20 mM Tris-HCl pH 7.5, 1 mM EDTA and subjected to electrophoresis in 1.4% agarose gels (Sussenbach and Kuijk, submitted).
Seg2uence analysis
of the 5'-termini of Sau 3AI fra&ments 3A_ Sau were produced by digestion of 7.5 AdS DNA with fragments _g Sau 3AI. After deproteinization by extraction with chloroform-isoamylalcohol (24:1) and ethanol precipitation of the DNA, the fragmented DNA was brought in 100 pl 40 mM Tris-HCl pH 8.5, 10 mM MgC12 and 10 mM dithiothreitol. Then 2 pg of bacterial alkaline phosphatase (Worthington, Freehold, N.J.) was added and the DNA was incubated for I h at 370 C. Subsequently the fragmented DNA was deproteinized and precipitated as described above and again taken up in 100 pl of 40 mM Tris-HCl pH 8.5, 10 mM MgC12 and 10 mM dithiothreitol. Then 10 pl (10 U) T4 polynucleotide kinase (Biolabs, Beverly, Ma) and 5 pl y-32P-ATP (5 pCi; 16 C/mmole) (The Radiochemical Centre, Amersham, England) was added and the mixture incubacted for 1 h at 370 C (9). DNA was then precipitated by addition of 2 ml of 10% trichloroacetic acid, 0.01 M PPi at 0° C and was centrifuged for 25 min at 30,000 rpm in a Spinco SW65 rotor. The supernatant was discarded and the precipitate washed with trichloroacetic acid as described above. This procedure was repeated five times. Finally, trichloroacetic acid was removed by three washings with 90% ethanol. The pellet was dissolved in 50 pl H20. For the analysis of the 5'-terminal nucleotide 0.15 pg 32P-phosphorylated DNA was digested with 10 pg pancreatic DNase (Worthington, Freehold, N.J.) in 10 mM Tris-HCl, 10 mM MgC12, pH 7.5 for 2 h at 370 C. Subsequently pH was raised to 9.0 and 2 pg of snake venom phosphodiesterase (Boehringer, Mannheim) was added and incubation was continued for another 2 h at 370 C. Under these conditions DNA is completely digested to 5'-mononucleotides. Separation of the mononucleotides was performed by paper chromatography on Whatman no. I paper (10). The chromatogram was developed using saturated ammonium sulphate-1.0 M sodium acetate-isopropanol (80:18:2) as solvent. Detection of the mononucleotides was facilitated by cochromatography of
3195
Nucleic Acids Research unlabeled mononucleotides. The four nucleotide spots were visualized by UV-light, cut from the chromatogram and counted in a liquid scintillation counter. For the analysis of the recognition site 32P-phosphorylated DNA was partially digested with pancreatic DNase. The products were fractionated according to a standard two-dimensional fractionation procedure developed by Brownlee and Sanger (11) in a modification according to Volckaert, Min Jou and Fiers (12). In the 1st dimension fractionation was performed by electrophoresis at pH 3.5 on cellulose acetate and in the 2nd dimension by homochromatography on PEI-cellulose using an RNA homomix. The RNA homomix was prepared by hydrolysis of a 3% RNA (yeast RNA, sodium salt, BDH, Poole, U.K.) solution at pH 12.8 for 30 min at room temperature.
RESULTS
Isolation of an endonuclease Staphylococcus aureus 3A was grown as described in Materials and Methods. About 3 grams of packed cells were treated with lysostaphin and after centrifugation resuspended in 8 ml 0.01 M Tris-HCl, 0.01 M 8-mercaptoethanol, pH 7.4 (see Materials and Methods). It appeared that the length of the lysostaphin treatment is rather critical for the optimal detection of endonucleases. Extensive treatment of the cells with lysostaphin causes lysis of the cells and very low levels of endonuclease. After the short lysostaphin treatment the cells were disrupted by sonication (6 x 1 min with a Branson Sonifier) at 00 C and centrifuged for I1 h at 40,000 rpm in a Spinco SW41 rotor at 40 C. The supernatant was collected and streptomycin sulphate was added for the precipitation of nucleic acids (1.8 ml of a 10% streptomycin sulphate solution in 0.01 M Tris-HCl, 0.01 M 8-mercaptoethanol, pH 7.4). The precipitate was removed by centrifugation at 35,000 rpm for 30 min in a Spinco SW41 rotor at 40 C and the supernatant was dialyzed against 0.01 M Tris-HCl, 0.01 M a-mercaptoethanol, pH 7.4. During the dialysis a precipitate arose which was removed by lowspeed centrifugation. The crude enzyme preparation (9 ml) was brought on a DEAE-cellulose column (30 ml) which was equilibrated with 0.01 M KPO4 buffer, 0.01 M a-mercaptoethanol, 0.0001 M EDTA, pH 7.4. Elution was performed with a KC1 gradient (200 ml; 0.0-0.6 M KC1). Fractions of 5 ml were collected and tested for the presence of endonuclease as described in Materials and Methods.
3196
Nucleic Acids Research
Figure I DEAE-cellulose chromatography of Sau 3AI. Extract of Staphylococcus aureus strain 3A were prepared as described in Materials and Methods and applied to a 30 ml DEAE-cellulose column and eluted with 200 ml 0-0.6 M KC1 gradient. Fractions of 5 ml were collected and tested alternately for endonucleolytic activity. The gradient starts as indicated by the arrow. Elution is from left to right.
Endonucleolytic activity was eluted between 0.20 and 0.33 M KC1 (Fig. 1). Concentration of the enzyme containing fractions was performed by dialysis against 50% glycerol in 0.01 M KP04 buffer, pH 7.4, 0.01 M S-mercaptoethanol, 0.0001 M EDTA. The enzyme preparations were stored at -200 C and no detectable loss in endonucleolytic activity was detected after storage for several months. The yield of purified enzyme from I gram of packed cells was equal to about 1000 units (1 unit is the amount of enzyme required for complete digestion of I ig of adenovirus type 5 DNA in 2 h at 300 C).
Characterization of Sau 3AI The observation that extracts of Staphylococcus aureus strain 3A possess endonucleolytic activity led us to a further characterization of this specific endonuclease which was called Sau 3AI according to the proposed nomenclature for this category of enzymes (13). Sau 3AI was characterized by its action on adenovirus type 5 DNA, SV40 DNA and double-stranded *X174 DNA. The results are shown in Fig. 2. It
3197
Nucleic Acids Research 1
2
3
4
Figure 2 Gel electrophoresis of DNA's of different sources cleaved by Sau 3AI. DNA's were cleaved and subjected to gel electrophoresis in 1.4% agarose gels as indicated in Materials and Methods. Lane I shows digested SV40 DNA and lane 2 digested adenovirus type 5 DNA. Lane 3 and 4 display digested and undigested fXI74 RF DNA,respectively. The faster moving band represents component I and the slower moving band component II. appears that adenovirus type 5 DNA is cleaved at least 35 times, while double-stranded 4X174 DNA is not cleaved by this enzyme. In the latter case there is a conversion of the closed circular form (component I) to the open circular form (component II) detectable but this is probably due to an
aspecific contamination in the Sau 3AI preparation. SV40 DNA is cleaved into eight fragments of 27.2, 25.2, 17.9, 10.4, 6.7, 6.4, 4.2 and 2.0% of genome size, respectively. To investigate whether this enzyme was actually involved in biological restriction-modification, the action of Sau 3AI on phage 3A DNA propagated in PS 3A, on phage 6 DNA propagated in PS 6 and phage 3A DNA propagated in PS 6 was determined. Furthermore, the action of crude extracts of a restriction-deficient mutant 3ARl was studied for the presence of endonucleolytic activity (Fig. 3). Sau 3AI cleaves phage 6 DNA many times as well as phage 3A DNA propagated in PS 6, but does not cleave phage 3A DNA propagated in 3198
Nucleic Acids Research 1
2
3
4
Figure 3 Digestion of different staphylococcal phage DNA's with enzyme from Staphylococcus aureus strains 3A and 3AR . Electrophoresis was in 1.4% agarose as described in Materials and Methods. Lane 1: phage 6 DNA x extract 3AR-; lane 2: phage 6 DNA x Sau 3AI; lane 3: phage 3A DNA x Sau 3AI; lane 4: DNA from phage 3A propagated in PS 6 x Sau 3AI. PS 3A. Further, the crude extract of the restriction-deficient mutant 3ARdid not contain Sau 3AI activity as indicated by the fact that phage 6 DNA was not cleaved. These results indicate that Sau 3AI is required for res-
triction in PS 3A and that PS 3A and PS 6 contain modification enzymes with different specificities. Sau 3A1 cleaves DNA's with the PS 6 modification but is not able to cleave DNA's with the PS 3A modification. Characterization of the recognaition site The recognition site of Sau 3A1 was characterized by determination of the 5'-nucleotide sequence of the DNA fragments produced with this enzyme. For this purpose adenovirus type 5 DNA cleaved with Sau 3AI was first digested with bacterial alkaline phosphatase to remove 5'-phosphate groups, and subsequently labeled with 32P-phosphate using T4 polynucleotide kinase and y-32P-ATP leading to introduction of a terminal 5'-32P-phosphate group at the ends of the fragment strands (see Materials and Methods). Complete digestion of this 32P-labeled DNA to 5'-mononucleotides with pancreatic DNase and snake venom phosphodiesterase followed by paper chromatography of
3199
Nucleic Acids Research
pG
Figure 4 Two dimensional fractionation of oligonucleotides obtained after partial pancreatic DNase digestion of adenovirus type 5 DNA fragments produced by cleavage with Sau 3MI. The 5'-terminal nucleotides of the fragment strands were phosphorylated with 32P-phosphate using T4 polynucleotide kinase. In the first dimension (I fractionation is achieved by electrophoresis at pH 3.5 on cellulose acetate and in the second dimension (II) by homochromatography on PEI-cellulose. The nucleotide s.equence was derived from the known relative mobilities (14).
the nucleotides showed that 97% of the radioactivity was present in pG and 3% in pA. For a further analysis of the 5'-terminal nucleotide sequences of the DNA fragment 5'-terminal 32p-labeled DNA was partially digested with
pancreatic DNase. The digestion products were separated two-dimensionally by electrophoresis on cellulose acetate (1st dimension) followed by homochromatography on PET-cellulose (2nd dimension). The autoradiogram displayed unique mono-, di-, tri- and tetranucleotide spots and multiple penta- and hexanucleotide spots (Fig. 4). Analysis showed that the fragment strands terminated specifically with the sequence 5' G-A-T-C-N-N-. DISCUSSION
Staphylococcus
aureus strain 3A possesses a
biological restrictionmodification system as shown by phage typing experiments (5; Stobberingh and Winkler, in press). Extracts of Staphylococcus aureus 3A appear to conrtain a specific endonucleolytic activity which can be partially purified by 3200
Nucleic Acids Research DEAE-cellulose chromatography. This enzyme, Sau 3AI, is absent in extracts of a restriction-deficient mutant, indicating that it is involved in biological restriction-modification. This notion is supported by the observation that Sau 3AI does not cleave DNA from phage 3A, which has been propagated in PS 3A, while passage of phage 3A in PS 6 makes phage 3A DNA accesible for Sau 3AI cleavage. These observationsjustify the conclusion that Sau 3AI is a restriction endonuclease which plays a role in the restriction-modification system of this bacterium. Partial purification of Sau 3AI can easily be achieved by DEAE-cellulose chromatography while even crude extracts prepared as described in Results are almost free of aspecific nucleases. The enzyme is very stable at 40 C in crude extracts and at -20° C in 50% glycerol. Analysis of the 5'-termini of Sau 3AI fragments revealed the sequence 5' G-A-T-C-N-N-. This suggests that the enzyme recognizes the sequences 5'
-C-T-CA- 3,
and cleaves
as
indicated
by
the
arrows.
An identical recog-
nition site and cleavage has also been found for an endonuclease from Moraxeila bovis (Mbo I) (R. Gelinas, G.A. Weiss, R.J. Roberts, A. Morrison and K. Murray, in preparation). Comparison of DNA fragments of adenovirus type 5 and SV40 obtained by cleavage with Sau 3AI and Mbo I, respectively confirmed that these enzymes recognize the same nucleotide sequence. A specific restriction endonuclease from Diplococcus pneumoniae (Dpn II) also recognizes the above sequence (S. Lacks, personal communication).
ACKNOWLEDGEMENTS The authors thank J.M. Vereijken, A.D.M. van Mansfeld, P.H. Steenbergh and M.G. Kuijk for assistance and Profs. K.C. Winkler and H.S. Jansz for interesting discussions and critical reading of the manuscript. This investigation was supported in part by the Netherlands Foundation for Chemical Research with financial aid from the Netherlands Organization for
the Advancement of Pure Research.
*Laboratory of Microbiology, State University of Utrecht, Utrecht, The Netherlahds REFERENCES
1. Nathans, D. and Smith, H.O. (1975). Ann. Rev. of Biochemistry 44, 273293 2. Smith, H.O. and Wilcox, K.W. (1970). J. Mol. Biol. 51, 379-391 3. Yoshimori, R. (1971). A genetic and biochemical analysis of the restric-
3201
Nucleic Acids Research
4. 5. 6. 7.
8. 9. 10. 11.
12. 13. 14.
tion and modification of DNA by resistance transfer factors, PhD thesis Univ. Calif. San Francisco Bron, S., Murray, K. and Trautner, T.A. (1975). Molec. gen. genet. 143, 13-23 Stobberingh, E.E. and Winkler, K.C. (1975). Antonie van Leeuwenhoek 41, 212-213 Sussenbach, J.S. (1971). Virology 46, 969-972 Blair, J.E. and Williams, R.E.O. (1961). Bull. World Health Organization 24, 771 Klesius, P.H. and Schuhardt, V.T. (1968). J. Bacteriol. 95, 739-743 Richardson, C.C. (1971) in Procedures in Nucleic Acid Research Vol 2, p. 815-828, Harper and Row, New York Markham, R. and Smith, J.D. (1952). Biochem. J. 52, 552-565 Brownlee, G.G. and Sanger, F. (1969). Eur. J. Biochem. 11, 395 Volckaert, G., Min Jou, W. and Fiers, W. (1976). Anal. Biochem. 72, 433-445 Smith, H.O. and Nathans, D. (1973). J. Mol. Biol. 81, 419-423 Galibert, F., Ziff, E. and Sedat, J. (1974). J. Mol. Biol. 87, 377-407
3202
