/ . Biochem. 82, 1567-1573 (1977)

Action of T4 Endonuclease V on DNA Containing Various Photoproducts1 Fumiko MAKINO,* Hiroshi TANOOKA,* and Mutsuo SEKIGUCHI** •Radiobiology Division, National Cancer Center Research Institute, Chuo-ku, Tokyo 104, and "Department of Biology, Faculty of Science, Kyushu University, Fukuoka 812 Received for publication, June 15, 1977

The action of T4 endonuclease V on DNA containing various photoproducts was investigated. (1) The enzyme introduced strand breaks in DNA from ultraviolet-irradiated vegetative cells of Bacillus subtilis but not in DNA from irradiated spores of the same organism. DNA irradiated with long wavelength (360 nm peak) ultraviolet light in the presence of 4,5',8trimethylpsoralen was not attacked by the enzyme. These results indicate that 5-thyminyl5,6-dihydrothymine (spore photoproduct) and psoralen-mediated cross-links in DNA are not recognized by T4 endonuclease V. (2) DNA of phage PBS1, containing uracil in place of thymine, and DNA of phage SPO1, containing hydroxymethyluracil in place of thymine, were fragmented by the enzyme when the DNA's had been irradiated with ultraviolet light. T4 endonuclease V seems to act on DNA with pyrimidine dimers whether the dimers contain thymine residues or not.

Irradiation of native DNA with UV results in/ the formation of dimers between adjacent pyrimidine residues (1, 2). T4 endonuclease V recognizes such dimers in DNA and introduces a singlestranded break on the 5'-side of the dimer (3, 4). Genetic studies with UV-sensitive mutants of T4 showed that the enzyme is responsible for the first incision step in excision repair (5, 6). Furthermore, the use of the enzyme has provided a fairly sensitive and specific tool for measuring the number of pyrimidine dimers in the DNA of 1

This work was supported by Grants-in-aid from the Ministry of Education, Science and Culture and the Ministry of Health and Welfare, Japan. Abbreviation: TDHT, 5-thyminyl-5,6-dihydrothymine. Vol. 82, No. 6, 1977

1567

irradiated organisms (7). Although cyclobutane-type pyrimidine dimers are the most representative lesions induced by UV, other types of photoproducts are also produced in DNA. 5-Thyminyl-5,6-dihydrothymine (TDHT) is formed after UV irradiation of spore or dried DNA (8, 9). The photosensitized reaction of DNA with psoralen yields adducts of thymine, which have some structural resemblance to cyclobutane-type dimers (10-12). To understand the repair mechanism and also for the practical use of the enzyme in estimating the number of dimers in DNA, it is important to know whether T4 endonuclease V acts on photoproducts other than cyclobutane-type dimers. The present studies were undertaken to elucidate this.

F. MAKINO, H. TANOOKA, and M. SEKIGUCHI

1568

MATERIALS AND METHODS Bacteria—A thymine-requiring mutant of Bacillus subtHis, 168T (13), was used to prepare DNA from vegetative cells. A strain defective in both excision repair and spore repair UVSSP-42-1 (hcrlt~ ssp~) (14) was used to extract DNA from spores. A UV-sensitive strain UVS-1 (hcr^) (14) was used as a host for phages. Preparation of DNA from Vegetative Cells, Spores, and Phages—Vegetative cells of B. subtilis were grown to the middle of their log growth phase in glucose-minimal medium supplemented with 10 fig of thymine, 100 fig of required amino acids, and 1 fid of methyl-'H-thymidine (5 Ci/mmol; Radiochemical Centre, Amersham) per ml. Spores were produced by incubation for 38-40 h with shaking in Schaeffer liquid sporulation medium (15) containing 0.5 ftCi of *H-thymidine per ml. Spores were treated with lysozyme and sodium dodecyl sulfate to remove vegetative cells (16), and were further purified by centrifugation in a 30-60-70% step gradient of Urografin (Schering AG) for 45 min at 15,000 rpm in a swingingbucket rotor (Hitachi, RPS 40T-2). The spores were recovered from the boundary between the 30 and 60% layers. This procedure eliminates lysozyme-resistant spore-like bodies in which pyrimidine dimers can be formed (17). The purified spores were washed several times with 0.15 M NaCl by centrifugation. DNA was isolated from vegetative cells according to Marmur (18) and from spores according to Sakakibara et al. (19) up to the deproteinization step. DNA solutions were dialyzed overnight in the cold against 0.15 M NaCl-0.015 M sodium citrate to remove sodium dodecyl sulfate. J H-Deoxyuridine-labeled phage PBS1 was prepared as described elsewhere (20). Phages were purified by equilibrium centrifugation in CsQ and DNA was isolated by the phenol method according to Yamagishi (21). SPO1 DNA was prepared from SPO1 phage which had been labeled with 6-3H-deoxyuridine and purified by equilibrium centrifugation in CsQ. Irradiation with UV and Near-UV—Five ml of vegetative cells (2xlO 7 cells/ml) in phosphatebuffered saline (pH 7.2) were irradiated in a Petri dish (9 cm in diameter) with UV from an 18.5 W

germicidal lamp (peak wavelength, 253.7 nm, Matsushita Electric Co., Osaka) at a distance of 20 cm. UV flux was estimated to be 4.85 J/m'/s from the UV survival curve for UVSSP-42-1 spores (22). Spores ( 8 x l 0 7 p e r ml, suspended in distilled water) and phages (about 1010 p.f.u./ml, suspended in Spizizen glucose-minimal medium) were similarly irradiated. To produce DNA crosslinks, *H-thymidine-labeled DNA of B. subtilis (23 figjm\) in a 0 . 1 5 M NaCI-0.015M sodium citrate solution was irradiated with two 20 W black light tubes (peak wavelength: 360 nm, Matsushita Electric Co., Osaka) at a distance of 10 cm for 30 min in the presence of 2.5 or 0.1 fig of 4,5',8-trimethyl-psoralen (prepared by Shiseido Laboratory and supplied by Dr. M.S. Sasaki of Tokyo Medical and Dental University) per ml. Enzyme Preparation and Assay—T4 endonuclease V was prepared from T4-infected Escherichia coli 1100 according to the procedure described by Yasuda and Sekiguchi (23). The enzyme was purified by phase partition in dextran500/polyethylene glycol-6000 and column chromatography on CM-Sephadex C-25 (Fraction HI). Fraction III was purified about 400-fold over the crude extract. The enzyme was further purified by two successive chromatographies on hydroxylapatite (Fraction V) and then by chromatographies on UV-irradiated (Fraction VI) and non-irradiated DNA-cellulose columns (Fraction VII). Fraction VII was purified approximately 3500-fold over the crude extract. One unit of the enzyme is defined as the amount catalyzing the conversion of 1 nmol of " P from irradiated DNA into phosphomonoesterase-sensitive forms in 60 min under standard assay conditions (23). The hydrolysis of UV DNA's was examined under the following conditions. The reaction mixture (0.25 ml) contained 50 mM Tris-HCl, 10 mM EDTA, 2-3 fig of 3H-labeled DNA, and 20 fi\ of enzyme solution (5.3 units/ml) at pH 7.5. The reaction was carried out for 30 min at 37°C and then terminated by immersing the tubes in a water bath at 60°C for 5 min. Two-tenths ml of the mixture was placed on 0.2 ml of lysing solution (0.25 N NaOH) layered on top of 4.6 ml of a linear gradient of 5 to 20% sucrose (w/v, pH 12). The sucrose gradient solution was centrifuged in a Hitachi RPS 40T-2 swinging-bucket rotor for 2.5 h at 40,000 rpm at 20°C in most experiments.

J. Biochem.

ACTION OF T4 ENDONUCLEASE V ON DNA PHOTOPRODUCTS After centrifugation, the samples were collected, washed, and counted as described previously (24). T4 and X DNA's were used as markers for molecular weight. Photoproduct Analysis—Photoproduct analysis of DNA from spores, vegetative cells, and phages was carried out by paper chromatography as described by Munakata and Rupert (21). The solvent was butanol-acetic acid-water (80:12: 30) (25). RESULTS TDHT-Containing DNA—On irradiation of B. subtilis spores with 291 J/m 1 of UV, 8.5% of radioactive thymine residues of the spore DNA was converted to TDHT; no cyclobutane-type dimers were detected on paper chromatography (Fig. 1). When such DNA was incubated with T4 endonuclease V, no fragmentation of the DNA took place (Fig. 2a). Under the conditions used, DNA prepared from UV-irradiated vegetative cells of B. subtilis, which contained cyclobutane-type dimers, was fragmented by the enzyme. To eliminate the possibility that the spore DNA preparation contains an inhibitor of the

enzyme action, the DNA from irradiated spores was re-irradiated in vitro and then treated with T4 endonuclease V. As shown in Fig. 2b, this DNA was fragmented extensively. Thus, T4 endonuclease V, which makes cuts near pyrimidine dimers, is unable to induce strand breaks in TDHTcontaining DNA. DNA with Psoralen-Induced Cross-Links— DNA was irradiated with long wavelength UV in the presence of 2.5 /Jg/ml of 4,5', 8-trimethylpsoralen. The psoralen-treated DNA sedimented more rapidiy in alkaline sucrose gradients than did normal DNA, indicating the formation of interstrand cross-Iinkes. Treatment of such DNA with T4 endonuclease V did not result in a decrease in the sedimentation rate of the DNA (Fig. 3a). Since a large number of cross-links might hinder the detection of strand breaks, less cross-linked DNA was prepared by illuminating sonicated DNA in the presence of 0.1 /ig/ml of psoralen and subjected to the enzyme action. In this case, too, no fragmentation of DNA was observed (Fig. 3b). Degradation of DNA by the combined action of T4 endonuclease V and exonucleases was ex-

Fig. 1. UV-induced photoproducts in DNA of B. subtilis spores (UVSSP-42-1). DNA extracted from cells or spores after UV irradiation (291 J/m*) was hydrolyzed in trifluoroacetic acid for 90 min at 175°C. Samples were subjected to paper chromatography in butanol-acetic acid-water (80 : 12 : 30). A , DNA from non-irradiated spores; • , DNA from irradiated spores. Vol. 82, No. 6, 1977

1569

1570

F. MAKINO, H. TANOOKA, and M. SEKIGUCH1

20

FRACTION NO. Fig. 2. Sedimentation patterns of DNA from UVirradiated B. subiilis spores with or without T4 endonuclease V treatment. Two to three ^g of ["HJthymidine-labeled DNA from UV-irradiated (291 J/m1) spores was incubated for 30min at 37°C with 0.1 unit of T4 endonuclease V (Fraction HI) in 0.05 M Tris-HCI (pH 7.5>-10 mix EDTA (total volume, 0.25 ml). The reaction was terminated by immersing samples in a water bath at 60°C for 5 min. Two-tenths ml of the reaction mixture was placed on 0.2 ml of 0.2 N NaOH layered on top of a 5-20% alkaline sucrose gradient (pH12.0) and centrifuged for 2.5 h at 40,000 rpm at 20°C. (a) Spores were irradiated (291 J/m1) and then DNA was extracted, (b) DNA was irradiated (291 J/m') after the extraction from spores which had been irradiated (291 J/m*). • , Without enzyme; • , with enzyme. amined to confirm the above results. DNA treated with psoralen or irradiated with UV was incubated with an extract of T4 Kj-infected cells in the presence or absence of T4 endonuclease V. The T4^-infected cell extract contains 5'-+3' exonucleases but not endonuclease V (5). As shown in Fig. 4, there was no or only slight T4 endonuclease V-dependent degradation of psoralentreated D N A whereas similar treatment of UVirradiated DNA resulted in extensive degradation. From these results, it is evident that T4 endonuclease V does not act on psoralen-mediated D N A cross-links.

DNA of UV-irradiated PBS1 andSPOl Phages —DNA of PBS1 and SPO1 contains uracil and 5-

FRACTION NO. Fig. 3. Sedimentation patterns of psoralen-mediated cross-linked DNA with or without T4 endonuclease V treatment, (a) fHl-labeled B. subiilis DNA (23 /