Eur. J. Biochem. 206,833-839 (1992)
0FEBS 1992
Purification to homogeneity and characterization of a redoxyendonuclease from calf thymus Ikramul HUQ, Bjerrn-Ivar HAUKANES and Dag E. HELLAND Center for Biotechnology, University of Bergen, Norway (Received May 30/November 13, 1991) - EJB 91 0701
A redoxyendonuclease from calf thymus was purified to apparent homogeneity. The redoxyendonuclease recognized and induced cleavage of DNA damaged by ultraviolet light. The enzyme preparation produced a single band of a relative molecular mass of approximately 34 kDa upon SDS/ PAGE. The apurinic/apyrimidinic endonuclease and the DNA glycosylase activities remained associated in the apparently homogeneous preparation of the enzyme. The redoxyendonuclease activity displayed a broad pH optimum between pH 5.0-8.5 and exhibited no requirement for divalent cations. By application of FPLC columns Mono-S, Mono-Q and Mono-P, the isoelectric point (PI) of the enzyme was found to be approximately 8.0. Using the DNA sequencing procedure of Maxam and Gilbert [Maxam, A. M. & Gilbert, W. (1980) Methods Enzymol. 65, 499-5601 the purified enzyme was found to incise ultraviolet-light-irradiated DNA at pyrimidine sites as observed previously with a more crude form of the enzyme. While the most frequently cleavaged sites for the crude preparation were at cytosine residues, the apparently homogeneous enzyme preparation frequently induced cleavage sites at both cytosine and guanine residues. Predominant incision induced by the apparently homogeneous preparation was observed at guanine residues when a particular DNA sequence was used as substrate. Furthermore, the 16 N-terminal amino acid residues of the purified enzyme were identified. The sequence did not show any significant similarity to other known proteins.
Living organisms have evolved repair systems to cope with the large variety of different lesions which can be induced in DNA and thereby maintain the genetic integrity of the cell. In mammalian and other eukaryotic cells, several endonuclease activities have been identified which recognize and cleave DNA altered by oxidizing agents, ionizing radiations, or high doses of ultraviolet light [l - 71. In fact, similar enzyme activities have been identified in all systems studied so far (reviewed by Wallace [S]}. The name redoxyendonuclease has been suggested for this type of enzymes, since they act on DNA modified by oxidation or reduction [2]. The redoxyendonucleases act as both DNA glycosylases and apurinic/apyrimidinic endonucleases, thus mediating the first two steps in the base-excision repair pathway [9]. An endonuclease from calf thymus was found to possess such a repair function and was capable of acting on both supercoiled and relaxed DNA [lo, 111. Studies with a heterogeneous preparation of the calf thymus redoxyendonuclease revealed that the enzyme had a somewhat similar substrate specificity as the Escherichia coli endonuclease 111 [7, 11- 131. Both enzymes Correspondence to B. I. Haukanes, Laboratory of Biotechnology, University of Bergen, P.O. Box 3152 Arstad, N-5029 Bergen, Norway Ahhreviutions. Ade, adenine; Cyt, cytosine; Gua, guanine; Thy, thymine, (pX174 RFI DNA, the replicative form I DNA which is circular, double-stranded and covalently closed. Enzymes. Redoxyendonuclease (EC 4.2.99. -); DNA glycosylase (EC 3.2.2. -); apurinic/apyrimidinic endonuclease (EC 3.1.25.2); endonuclease I11 (E. coli) (EC 3.1.25.1); calf intestinal phosphatase (EC 3.1.3.1); T4-polynucleotide kinase (EC 2.7.1.78).
were found to incise DNA at sites of damaged pyrimidines following exposure to high doses of ultraviolet light, osmium tetroxide and other oxidizing agents. However, the E. coli endonuclease I11 preparation used [2] appeared to induce cleavage at some Gua residues in addition to the pyrimidine residues [2]. Furthermore, the E. coli endonuclease 111 produced a 3' deoxyribose residue and a 5' phosphoryl releasing only the damaged base. The calf thymus redoxyendonuclease activity, however, produced both 3' and 5' phosphoryl groups, thus releasing both the damaged base residue and the deoxyribose residue [Ill. In the present report we describe the purification of the calf thymus redoxyendonuclease to apparent homogeneity using different chromatographic and electrophoretic techniques, and identification of the N-terminal amino acid residues of the purified enzyme. Furthermore, we have studied the incision pattern using end-labeled DNA fragments of defined sequences as substrate and analyzed the cleavage products on DNA sequencing gels. This procedure made it possible to detect and analyse each single cleavage site created by the enzyme. We have also studied the ultraviolet light dose response with respect to incision pattern and frequency of incision induced by the enzyme.
MATERIALS AND METHODS Enzymes and chemicals DEAE-cellulose (DE 32 microgranular) and P11 phosphocellulose were from Whatman. Sephadex G-75 superfine,
834 Mono-S, Mono-Q and Mono-P were obtained from Pharmacia A/S. Nitrocellulose membrane filters were purchased from Gelman (Metricell GN-6). [MethyL3H]thymidine (48 Ci/ mmol), T,-polynucleotide kinase and [Y-~'P]ATP were obtained from Amersham. SaA and PvulI restriction endonucleases were from New England Biolabs. Calf intestinal phosphatase was purchased from Boehringer-Mannheim.
3 - 5 niin before application on the polyacrylamide gel. The resolving gel contained 10- 20% acrylamide and 2.5% bisacrylamide, and the stacking gel contained 3.5% acrylamide with the same concentration of crosslinker. The gel was run at 200 V. The preparative SDS/PAGE was run at the same conditions as described above, except that the gel was not fixed and stained after electrophoresis or before electroblotting.
Preparation of nuclear and chromatin extracts
Definition of enzyme activity
Calf thymus from young calves were collected and frozen at - 80 "C. After thawing, blood vessels, fats and connective tissue were removed. Washed tissue was cut into small pieces, suspended in homogenization buffer containing 20 mM Tris/ HCl, pH 8.0, 5 mM MgC12, 10 mM KCl, 0.5 mM phenylmethylsulphonylfluoride, 10 mM 2-mercaptoethanol and 0.1 YONonidet P-40, and homogenized in a Potter-Elvehjem homogenizer with a teflon piston. The NaCl concentration was adjusted to 0.7 M to dissociate chromatin-bound enzyme. The suspension was then centrifuged at 8000 rpm for 30 min at 0-2°C using a Sorvall GSA rotor. The supernatant was dialyzed against a buffer containing 10 mM Tris/HCl, pH 7.5 - 8.0,0.25 M NaCI, 1 mM EDTA, 5% glycerol and 10 mM 2-mercaptoethanol.
1 I J calf thymus redoxyendonuclease activity is defined as the amount of enzyme required to produce one nick/DNA molecule with 0.1 pg 3H-labeled (pX174 DNA using the standard assay described. The average number of breaks introduced, DNA molecule was estimated by assuming a Poisson distribution [6].
Preparation of (pX174RFI ['HIDNA The 3H-labelled qX174 RFI DNA was prepared principally according to Nes [6] and Helland et al. [14].
Reparation of defined-sequence DNA substrates Plasmid pUC18 was isolated and purified according to previously established methods [17]. The plasmid was digested with SaA restriction endonuclease prior to phosphatase treatment and labelled with T,-polynucleotide kinase and \?32P]ATP.A second digest with PvuII restriction endonuclease resulted in two labeled fragments of 109 and 21 1 bp in length; both 32P-labeledin one strand only. Both of the fragments were employed in the DNA sequencing analysis. DNA sequencing reactions and gel electrophoresis
The DNA was ultraviolet-light irradiated (254 nm) with a dose of 10000 J/m2 using a G8 TS germicidal lamp as described by Helland et al. [14]. The DNA substrates were subjected to enzyme treatment immediately after irradiation.
The Gua + Ade and Cyt + Thy base-specific DNA sequencing reactions were run alongside each set of samples analysid on the sequencing gels. Reprecipitated DNA samples from enzyme digestion and chemical treatment, were loaded onto denaturing 20% polyacrylamide-urea gels and subjected to electrophoresis and autoradiography [ 181.
Calf thymus redoxyendonucleasestrand-scission assay
N-tenninal amino acid sequence analysis
The standard assay mixture consisted of 10 mM Tris/HCl, pH 8.0, 10mM 2-mercaptoethanol, 1 mM EDTA, 40mM NaCl, 0.05-0.1 pg substrate DNA. 1 U calf thymus redoxyendonuclease was added and incubated at 37 "C for 30 min. The amount of nicked DNA was determined by the nitrocellulose filter method [6].
Protein determination
Samples for N-terminal analysis were obtained by preparative gel electrophoresis (SDS/PAGE) of fraction V, followed by electroblotting [191 on a polyvenylidenedifluoride membrane (Durapore, Millipore). The proteins on the polyvenylidenedifluoride membrane were visualized by staining with 0.1% Coomassie brilliant blue. The 34-kDa band was excised and kept in 50% ethyl alcohol. Approximately 75 pmol protein was blotted and about 4.6% could be recovered for N-terminal amino acid sequence analysis. The analysis was performed using the Applied Biosystem model 477A ProteinIPeptide Sequencer. The polyvenylidenedifluoride membrane stripe was cut into small fragments and placed on the glass-fibre filter in the reaction chamber. Duplicate analysis were run as otherwise described [20].
The protein content was estimated using the Biorad Protein assay h t with y-globulin as standard.
RESULTS
Ultraviolet-light damage of DNA substrate
Preparation of apurinic DNA Apurinic sites were introduced into qX174 RFI [3H]DNA by treatment with 0.1 M NaOAc, pH 5.1 and incubated for 30 min at 70°C [15].
Sodium-dodecylsulphate polyacrylamide gel electrophoresis (SDS/PAGE) Proteins were separated according to the method of Laemmli [16]. A sample of 20 p1 was mixed with 10 p1 sample buffer containing 2% SDS, 5% glycerol, 16% 2-mercaptoethanol in Tris/HCl, pH 6.7 and 0.002% bromophenol blue as front marker. The sample mixture was heated at 100"C for
Purification Zon-exchange chromatography
Fraction I, the dialyzed material from the calf thymus extract, was applied on a 9.6 x 2.7 cm DEAE cellulose column equilibrated with a buffer containing 10 mM Tris/HCl, pH 7.5 - 8.0,0.25 M NaCl, 1 mM EDTA, 5% glycerol and 10 mM 2-mercaptoethanol. The same buffer was used to elute the
active-enzyme-containing material, almost all of which came in the flow through. Most of the nucleic acids and some of the proteins present in fraction I were retained on the DEAE column. The redoxyendonuclease-activity-containing fractions were pooled and dialyzed against elution buffer containing 20 mM Tris/HCl, pH 8.0 and 50 mM NaCl (fraction 11). Fraction I1 was subsequently loaded onto a phosphocellulose column (8 x 6 cm). The column was equilibrated with the same buffer as for dialysis of fraction 11. The enzyme activity was eluted with a gradient of 50 mM - 1 mM NaCl in elution buffer with a total volume of 650 ml. Fractions of 12 ml were collected and assayed for the enzyme activity. Roughly two main peaks of the activity were found (Fig. 1A), eluted at approximately 400 and 500rnM NaCl. The latter peak contained 12 fractions which were pooled and dialysed against the storage buffer (20 mM Tris/HCl, pH 8.5, 50% glycerol, 50 mM NaCl, 1 mM EDTA, 10 mM 2-mercaptoethanol) and the final volume was 55 ml (fraction 111).
i2ou
z s
6
1000 -
3
A
x
'Oo0
f
800 2
800
9 B
600
3
600
d
400
I
0FEBS 1992
Purification to homogeneity and characterization of a redoxyendonuclease from calf thymus Ikramul HUQ, Bjerrn-Ivar HAUKANES and Dag E. HELLAND Center for Biotechnology, University of Bergen, Norway (Received May 30/November 13, 1991) - EJB 91 0701
A redoxyendonuclease from calf thymus was purified to apparent homogeneity. The redoxyendonuclease recognized and induced cleavage of DNA damaged by ultraviolet light. The enzyme preparation produced a single band of a relative molecular mass of approximately 34 kDa upon SDS/ PAGE. The apurinic/apyrimidinic endonuclease and the DNA glycosylase activities remained associated in the apparently homogeneous preparation of the enzyme. The redoxyendonuclease activity displayed a broad pH optimum between pH 5.0-8.5 and exhibited no requirement for divalent cations. By application of FPLC columns Mono-S, Mono-Q and Mono-P, the isoelectric point (PI) of the enzyme was found to be approximately 8.0. Using the DNA sequencing procedure of Maxam and Gilbert [Maxam, A. M. & Gilbert, W. (1980) Methods Enzymol. 65, 499-5601 the purified enzyme was found to incise ultraviolet-light-irradiated DNA at pyrimidine sites as observed previously with a more crude form of the enzyme. While the most frequently cleavaged sites for the crude preparation were at cytosine residues, the apparently homogeneous enzyme preparation frequently induced cleavage sites at both cytosine and guanine residues. Predominant incision induced by the apparently homogeneous preparation was observed at guanine residues when a particular DNA sequence was used as substrate. Furthermore, the 16 N-terminal amino acid residues of the purified enzyme were identified. The sequence did not show any significant similarity to other known proteins.
Living organisms have evolved repair systems to cope with the large variety of different lesions which can be induced in DNA and thereby maintain the genetic integrity of the cell. In mammalian and other eukaryotic cells, several endonuclease activities have been identified which recognize and cleave DNA altered by oxidizing agents, ionizing radiations, or high doses of ultraviolet light [l - 71. In fact, similar enzyme activities have been identified in all systems studied so far (reviewed by Wallace [S]}. The name redoxyendonuclease has been suggested for this type of enzymes, since they act on DNA modified by oxidation or reduction [2]. The redoxyendonucleases act as both DNA glycosylases and apurinic/apyrimidinic endonucleases, thus mediating the first two steps in the base-excision repair pathway [9]. An endonuclease from calf thymus was found to possess such a repair function and was capable of acting on both supercoiled and relaxed DNA [lo, 111. Studies with a heterogeneous preparation of the calf thymus redoxyendonuclease revealed that the enzyme had a somewhat similar substrate specificity as the Escherichia coli endonuclease 111 [7, 11- 131. Both enzymes Correspondence to B. I. Haukanes, Laboratory of Biotechnology, University of Bergen, P.O. Box 3152 Arstad, N-5029 Bergen, Norway Ahhreviutions. Ade, adenine; Cyt, cytosine; Gua, guanine; Thy, thymine, (pX174 RFI DNA, the replicative form I DNA which is circular, double-stranded and covalently closed. Enzymes. Redoxyendonuclease (EC 4.2.99. -); DNA glycosylase (EC 3.2.2. -); apurinic/apyrimidinic endonuclease (EC 3.1.25.2); endonuclease I11 (E. coli) (EC 3.1.25.1); calf intestinal phosphatase (EC 3.1.3.1); T4-polynucleotide kinase (EC 2.7.1.78).
were found to incise DNA at sites of damaged pyrimidines following exposure to high doses of ultraviolet light, osmium tetroxide and other oxidizing agents. However, the E. coli endonuclease I11 preparation used [2] appeared to induce cleavage at some Gua residues in addition to the pyrimidine residues [2]. Furthermore, the E. coli endonuclease 111 produced a 3' deoxyribose residue and a 5' phosphoryl releasing only the damaged base. The calf thymus redoxyendonuclease activity, however, produced both 3' and 5' phosphoryl groups, thus releasing both the damaged base residue and the deoxyribose residue [Ill. In the present report we describe the purification of the calf thymus redoxyendonuclease to apparent homogeneity using different chromatographic and electrophoretic techniques, and identification of the N-terminal amino acid residues of the purified enzyme. Furthermore, we have studied the incision pattern using end-labeled DNA fragments of defined sequences as substrate and analyzed the cleavage products on DNA sequencing gels. This procedure made it possible to detect and analyse each single cleavage site created by the enzyme. We have also studied the ultraviolet light dose response with respect to incision pattern and frequency of incision induced by the enzyme.
MATERIALS AND METHODS Enzymes and chemicals DEAE-cellulose (DE 32 microgranular) and P11 phosphocellulose were from Whatman. Sephadex G-75 superfine,
834 Mono-S, Mono-Q and Mono-P were obtained from Pharmacia A/S. Nitrocellulose membrane filters were purchased from Gelman (Metricell GN-6). [MethyL3H]thymidine (48 Ci/ mmol), T,-polynucleotide kinase and [Y-~'P]ATP were obtained from Amersham. SaA and PvulI restriction endonucleases were from New England Biolabs. Calf intestinal phosphatase was purchased from Boehringer-Mannheim.
3 - 5 niin before application on the polyacrylamide gel. The resolving gel contained 10- 20% acrylamide and 2.5% bisacrylamide, and the stacking gel contained 3.5% acrylamide with the same concentration of crosslinker. The gel was run at 200 V. The preparative SDS/PAGE was run at the same conditions as described above, except that the gel was not fixed and stained after electrophoresis or before electroblotting.
Preparation of nuclear and chromatin extracts
Definition of enzyme activity
Calf thymus from young calves were collected and frozen at - 80 "C. After thawing, blood vessels, fats and connective tissue were removed. Washed tissue was cut into small pieces, suspended in homogenization buffer containing 20 mM Tris/ HCl, pH 8.0, 5 mM MgC12, 10 mM KCl, 0.5 mM phenylmethylsulphonylfluoride, 10 mM 2-mercaptoethanol and 0.1 YONonidet P-40, and homogenized in a Potter-Elvehjem homogenizer with a teflon piston. The NaCl concentration was adjusted to 0.7 M to dissociate chromatin-bound enzyme. The suspension was then centrifuged at 8000 rpm for 30 min at 0-2°C using a Sorvall GSA rotor. The supernatant was dialyzed against a buffer containing 10 mM Tris/HCl, pH 7.5 - 8.0,0.25 M NaCI, 1 mM EDTA, 5% glycerol and 10 mM 2-mercaptoethanol.
1 I J calf thymus redoxyendonuclease activity is defined as the amount of enzyme required to produce one nick/DNA molecule with 0.1 pg 3H-labeled (pX174 DNA using the standard assay described. The average number of breaks introduced, DNA molecule was estimated by assuming a Poisson distribution [6].
Preparation of (pX174RFI ['HIDNA The 3H-labelled qX174 RFI DNA was prepared principally according to Nes [6] and Helland et al. [14].
Reparation of defined-sequence DNA substrates Plasmid pUC18 was isolated and purified according to previously established methods [17]. The plasmid was digested with SaA restriction endonuclease prior to phosphatase treatment and labelled with T,-polynucleotide kinase and \?32P]ATP.A second digest with PvuII restriction endonuclease resulted in two labeled fragments of 109 and 21 1 bp in length; both 32P-labeledin one strand only. Both of the fragments were employed in the DNA sequencing analysis. DNA sequencing reactions and gel electrophoresis
The DNA was ultraviolet-light irradiated (254 nm) with a dose of 10000 J/m2 using a G8 TS germicidal lamp as described by Helland et al. [14]. The DNA substrates were subjected to enzyme treatment immediately after irradiation.
The Gua + Ade and Cyt + Thy base-specific DNA sequencing reactions were run alongside each set of samples analysid on the sequencing gels. Reprecipitated DNA samples from enzyme digestion and chemical treatment, were loaded onto denaturing 20% polyacrylamide-urea gels and subjected to electrophoresis and autoradiography [ 181.
Calf thymus redoxyendonucleasestrand-scission assay
N-tenninal amino acid sequence analysis
The standard assay mixture consisted of 10 mM Tris/HCl, pH 8.0, 10mM 2-mercaptoethanol, 1 mM EDTA, 40mM NaCl, 0.05-0.1 pg substrate DNA. 1 U calf thymus redoxyendonuclease was added and incubated at 37 "C for 30 min. The amount of nicked DNA was determined by the nitrocellulose filter method [6].
Protein determination
Samples for N-terminal analysis were obtained by preparative gel electrophoresis (SDS/PAGE) of fraction V, followed by electroblotting [191 on a polyvenylidenedifluoride membrane (Durapore, Millipore). The proteins on the polyvenylidenedifluoride membrane were visualized by staining with 0.1% Coomassie brilliant blue. The 34-kDa band was excised and kept in 50% ethyl alcohol. Approximately 75 pmol protein was blotted and about 4.6% could be recovered for N-terminal amino acid sequence analysis. The analysis was performed using the Applied Biosystem model 477A ProteinIPeptide Sequencer. The polyvenylidenedifluoride membrane stripe was cut into small fragments and placed on the glass-fibre filter in the reaction chamber. Duplicate analysis were run as otherwise described [20].
The protein content was estimated using the Biorad Protein assay h t with y-globulin as standard.
RESULTS
Ultraviolet-light damage of DNA substrate
Preparation of apurinic DNA Apurinic sites were introduced into qX174 RFI [3H]DNA by treatment with 0.1 M NaOAc, pH 5.1 and incubated for 30 min at 70°C [15].
Sodium-dodecylsulphate polyacrylamide gel electrophoresis (SDS/PAGE) Proteins were separated according to the method of Laemmli [16]. A sample of 20 p1 was mixed with 10 p1 sample buffer containing 2% SDS, 5% glycerol, 16% 2-mercaptoethanol in Tris/HCl, pH 6.7 and 0.002% bromophenol blue as front marker. The sample mixture was heated at 100"C for
Purification Zon-exchange chromatography
Fraction I, the dialyzed material from the calf thymus extract, was applied on a 9.6 x 2.7 cm DEAE cellulose column equilibrated with a buffer containing 10 mM Tris/HCl, pH 7.5 - 8.0,0.25 M NaCl, 1 mM EDTA, 5% glycerol and 10 mM 2-mercaptoethanol. The same buffer was used to elute the
active-enzyme-containing material, almost all of which came in the flow through. Most of the nucleic acids and some of the proteins present in fraction I were retained on the DEAE column. The redoxyendonuclease-activity-containing fractions were pooled and dialyzed against elution buffer containing 20 mM Tris/HCl, pH 8.0 and 50 mM NaCl (fraction 11). Fraction I1 was subsequently loaded onto a phosphocellulose column (8 x 6 cm). The column was equilibrated with the same buffer as for dialysis of fraction 11. The enzyme activity was eluted with a gradient of 50 mM - 1 mM NaCl in elution buffer with a total volume of 650 ml. Fractions of 12 ml were collected and assayed for the enzyme activity. Roughly two main peaks of the activity were found (Fig. 1A), eluted at approximately 400 and 500rnM NaCl. The latter peak contained 12 fractions which were pooled and dialysed against the storage buffer (20 mM Tris/HCl, pH 8.5, 50% glycerol, 50 mM NaCl, 1 mM EDTA, 10 mM 2-mercaptoethanol) and the final volume was 55 ml (fraction 111).
i2ou
z s
6
1000 -
3
A
x
'Oo0
f
800 2
800
9 B
600
3
600
d
400
I
