Proc. Natl. Acad. Sci. USA Vol. 73, No. 12, pp. 4324-4328, December 1976 Biochemistry II, apurinic acid endonuclease, and III (chemical carcinogens/DNA repair/ and ) D. M. KIRTIKAR, G. R. CATHCART, AND D. A. GOLDTHWAIT Department of Biochemistry, Case Western Reserve University, Cleveland, Ohio 44106 Communicated by Harland G. Wood, August 23, 1976

ABSTRACT An endonuclease of Escherichia coli active on They found that the two activities copurified and a thermo- a DNA treated with methylmethane sulfonate has been sepa- rated from an endonuclease active on depurinated sites. The sensitive mutant of one activity was thermosensitive for the former is designated here as endonuclease II, while the other activity. Furthermore, in a revertant of one of the mu- latter enzyme is designated as apurinic acid endonuclease. tants, the levels of both activities increased. This evidence Endonuclease II is also active on DNA treated with- methylni- suggested that the two enzyme activities were due to the same trosourea, 7-bromomethyl-12-methylbenz[a anthracene, and protein. Weiss has recently purified a protein of molecular 'y-irradiation. A third fraction which contains activities for both weight 28,000 to homogeneity (13) and showed that it has the depurinated and alkylated sites needs further study. Endonu- clease II, molecular weight 33,000, has been purified 12,500-fold activity expected of exonuclease III and also has an activity on and does not have exonuclease III activity. Apurinic acid en- the heavily alkylated DNA entrapped in polyacrylamide gel. donuclease, molecular weight 31,500, has been purified He has concluded that endonuclease II of E. coil is exonuclease 11,000-fold and does not have exonuclease III activity. Exonu- III. However, we have shown that the three , endo- clease III, molecular weight 26,000, has been purified 2300-fold II, exonuclease III, and apurinic acid endonuclease, and does not have endonucleolytic activity at depurinated re- are separate proteins. Furthermore, we have noted that a pu- duced sites or at alkylated sites in DNA. Therefore, these are rified preparation of exonuclease III can make three separate proteins. Exonuclease III can produce, presum- double-strand ably by its exonucleolytic activity, double-strand breaks in breaks in heavily alkylated DNA under conditions where it is heavily alkylated DNA under conditions where it does not make unable to make single-strand breaks at alkylated or depurinated single-strand endonucleolytic breaks at either depurinated- sites. reduced or alkylated sites. METHODS The first purpose of this paper is to define endonuclease II of Assays. The assays for the endonucleases involved either Escherichia coli as an activity different from the apurinic acid DNA immobilized in acrylamide gel or DNA examined on endonuclease of E. coli. Strauss and Robbins first described an sucrose gradients. [3HjThymidine-labeled T4 DNA was pre- endonucleolytic activity in extracts of Bacillus subtilis that pared as described (2). Preparations of DNA had specific ac- recognized alkylated DNA (1). In this laboratory, an enzyme tivities ranging from approximately 1 to 5 X 103 cpm/nmol. in extracts of E. coli, active on heavily alkylated DNA, was Endonuclease assays, using either the DNA-gel or sucrose partially purified, characterized, and designated endonuclease gradients, and the exonuclease assay are described in the leg- II of E. coli (2, 3). The substrate used for these experiments was ends to Figs. 1 and 2. DNA that was entrapped in a polyacrylamide gel and then al- Enzymes. The E. coli strain used for small-scale enzyme kylated with methylmethane sulfonate [MeSO2OMe (MMS)] purification was AB 1157. This was grown in a fermentor to at an MeSO2OMe-to- ratio of 6000 to 1. A partially mid-logarithmic phase in a modified EM-9 medium (16) sup- purified preparation of endonuclease II was also found to have plemented with L-leucine, L-proline, L-histidine, L-threonine, an endonucleolytic activity on depurinated reduced DNA (4), and L-arginine, each at 20 Asg/ml. For the initial fractionation, and this activity was thought to be due to the same enzyme that see legend of Fig. 1. was active on MeSO20Me-treated DNA. However, Verly et For a large-scale preparation of the apurinic acid endonu- al. (5, 6), using the purification procedure originally described clease and endonuclease II, E. coli JC4583 (endo I-, his-, F-, in this laboratory, obtained an enzyme that was active on gal-, SMs, BI-) was grown at the New England Enzyme Center, depurinated DNA but not on alkylated DNA. Subsequently, and 800 g were used. The purification steps will be described we succeeded in separating the activity on depurinated sites in future publications. Although each purified endonuclease in DNA from the activity on MeSO2OMe-treated DNA (7, 8). preparation, when examined by sodium dodecyl sulfate (Na- The former we designate as the apurinic acid endonuclease of DoSO4)-gel electrophoresis, showed only a single band, it could E. coli, while the latter we designate as endonuclease II of E. not be concluded unequivocally that the preparations were coli. Endonuclease II of E. coli is also active on DNA treated homogeneous because large amounts of enzyme were not with methylnitrosourea, 7-bromomethyl-12-methvlbenz- available to look for small amounts of contaminant protein. [alanthracene, and y-irradiation (7-11). Exonuclease III was purified from 200 g of E. coil (JC 4583) The second purpose of this paper is to demonstrate that en- by a modification of the procedure of Richardson and Kornberg donuclease II, the apurinic acid endonuclease, and exonuclease (15), which included ammonium sulfate fractionation and III are separate proteins. Originally, Yajko and Weiss (12) DEAE-cellulose, phosphocellulose, Sephadex G-100, and hy- demonstrated that a number of E. coli mutants deficient in droxyapatite column chromatography. The enzyme purifica- exonuclease III were also deficient in "endonuclease II" and tion was 2300-fold when the 3'- activity in fraction vice versa. The "endonuclease II" activity was measured with III due to exonuclease III was used for the calculation. The heavily alkylated DNA in acrylamide gel (2), as noted above. preparation was not homogeneous by NaDodSO4 gel electro- phoresis. Dr. C. C. Richardson very kindly provided a sample Abbreviations: MeSO2OMe, methylmethane sulfonate (MMS); Na- of his purified exonuclease III for comparison with this prep- DodSO4, sodium dodecyl sulfate. aration. 4324 Downloaded by guest on September 25, 2021 Biochemistry: Kirtikar et al. Proc. Nati. Acad. Sci. USA 73 (1976) 4325

0 N K b E E 0. 0. U Pt 01

FIG. 1. Chromatographic behavior of apurinic acid endonuclease, endonuclease II, and exonuclease III on DEAE-cellulose. Fraction III (2), a 45-75% ammonium sulfate precipitate (396 mg), was dialyzed against buffer C (0.05 M Tris.HCl at pH 8.0, 0.1 mM dithiothreitol, and 20% glycerol) and then applied to a DE-52 column 2.5 X 48 cm. Elution was with one column volume of buffer C plus 0.03 M NaCl and then as indicated in the figure. Ten-milliliter fractions were collected and 50 AI ofevery fourth fraction was assayed for apurinic acid endonuclease, endonuclease II, and exonuclease III. For the endonuclease assays, labeled DNA was entrapped in a polyacrylamide gel which was forced through a screen to produce gel particles (14). Depurinated reduced DNA gel was prepared by suspending the gel in four volumes of 0.1 M sodium citrate buffer at pH 3.5 plus 0.1 mM EDTA and heating at 450 for 30 min. The gel was then cooled, the pH was adjusted to 6.5 with NaOH, and potassium phosphate buffer at pH 6.5 was added to a final concentration of 0.5 M. Aldehyde groups at depurinated sites were reduced with NaBH4 (4) to prevent spontaneous a-elimination with phosphodiester bond hydrolysis. A final concentration of 0.25 M NaBH4 was attained by three additions at 15-min intervals at room temperature. After an incubation of 60 more min, the DNA gel was washed in 0.05 M Tris-HCl at pH 8.0 and resus- pended in the same buffer. These conditions produce approximately one depurinated site per 1550 . The MeSO2OMe-treated DNA gel was prepared by incubation of the DNA gel for 120 min at room temperature in 0.05 M Tris-HCl at pH 8.0 with MeSO2OMe at a molar ratio of MeSO2OMe to DNA nucleotide of 500 to 1. Gels were washed extensively with 0.05 M Tris.HCl at pH 8.0 and resuspended in the same buffer. The packed DNA gels contained 60-80 nmol of DNA nucleotide per ml of packed volume. Both types ofgels were used on the day of their prep- aration. Incubation mixtures contained 20 nmol of DNA substrate, 0.05 M Tris-HCl at pH 8.0, 10-4 M 8-hydroxyquinoline, 10-4 M dithiothreitol, and 1.5 mg of bovine serum albumin in a volume of 1.5 ml. Incubations were at 37° for 30 min and were stopped with 0.1 ml of 1% NaDodSO4. After dilution with water to 2.0 ml and centrifugation, the radioactivity of a 1.0-ml aliquot was determined in a liquid scintillation counter. One unit represents 1 ,umol of DNA nucleotide released per hr. (0) Depurinated reduced [3H]DNA gel; (0) MeSO2OMe-treated [3H]DNA gel. Exo- nuclease III was assayed by its 3'-phosphatase activity. E. coli [32P]DNA (2 X 104 cpm/nmol) was digested with until 30% was acid soluble; the higher-molecular-weight material remaining after dialysis was used (15). Incubation mixtures contained 50 nmol of DNA, 10-3 M 2-mercaptoethanol, 10-2 M MgCl2, and 0.066 M potassium phosphate at pH 7.0 in a volume of 0.3 ml. After incubation for 30 min at 370, 0.5 mg of calf thymus DNA was added, followed by 0.5 ml of 10% trichloroacetic acid. The radioactivity that did not absorb to Norit charcoal was determined (15). One unit of 3'-phosphatase activity is defined as the amount of enzyme able to release 1 nmol of 32P/30 min at 370 (15). Exonuclease III activity was also measured by determining the 5% trichloroacetic acid-soluble material from the [32P]DNA used for the 3'- phosphatase assay after a 30-min incubation at 37O. (A) 32p; release.

RESULTS treated with MMs, methylnitrosourea, 7-bromomethyl-12- Separation of apurinic acid endonuclease and methylbenz[a]anthracene, or y-irradiation, is eluted with 0.25 endonuclease II M NaCI as peak III (Fig. 1). It is this fraction and not the apurinic acid endonuclease which recognizes the specific The apurinic acid endonuclease can be separated from endo- damage (other than apurinic or apyrimidinic sites) in these nuclease II by DEAE-cellulose chromatography (Fig. 1). When substrates (7, 8). Endonuclease II has been purified in this lab- fraction III, a 45-70% ammonium sulfate precipitate, was ap- oratory over 12,500-fold to a single band on NaDodSO4 gel plied to a DEAE-cellulose column, the apurinic acid endonu- electrophoresis. Its molecular weight by gel filtration is 33,000 clease (peak I) was eluted with 0.1 M NaCI prior to the gradient. and by NaDodSO4 gel is 34,500. This activity was assayed by the gel method using DNA with Peak II (Fig. 1) contains both the apurinic acid endonuclease very few depurinated-reduced sites. This enzyme has been activity and the endonuclease II activity. The molecular weight purified 11,000-fold in this laboratory, to a single band on of the material obtained from the DEAE-cellulose column was NaDodSO4 gel electrophoresis. It has a molecular weight by gel approximately 58,000. The nature of this material is not clear. filtration of 31,500. Verly and Rassart have also purified this A control assay, not shown in Fig. 1, involved native DNA en- enzyme 9450-fold to homogeneity (17) and have found a mo- trapped in the gel and incubated in the presence of 8-hy- lecular weight of 32,000 by gel filtration and 33,000 by Na- droxyquinoline. There was no release of native DNA from the DodSO4 gel electrophoresis. gel in peaks I and III, but there was some activity overlapping Endonuclease II, defined as the enzyme that recognizes DNA peak II with maximum activity in fraction number 145. Thus, Downloaded by guest on September 25, 2021 4326 Biochemistry: Kirtikar et al. Proc. Natl. Acad. Sci. USA 73 (1976)

A B C 14 Q 7 L Apurinic A 20 Endonuclease Endonuclease 11 0.008 unit 0.01 unit I ' 12 3rl 0.04 unit 61_ 0.05 unit Exonuclease III Control 1' -- b 036 unit CY 101 5 by \ Ia ll unit T-0 |0.9 x Control 81 41_ Exonuclease'm * E , 0 M. 10 III,0.9unt Contr\l 0 6 Exonuclease * 31- III, 0.9/1unit ,~~~I 4 Io 2 ! 1@~~~~%00 5 , .. ..A.il 0- 0- 1 -A-A 21 -~~9R \nlz 0A 0.A

- -ir A I a I I A-JL -,'- I IW -- . r a 5 10 15 *20 25 5 10 15 20 25 5 10 15 20 25 30

'- Sedimentation FRACTIONS FIG. 2. Sucrose gradient analysis of exonuclease III action on depurinated-reduced DNA, lightly alkylated DNA, and heavily alkylated DNA. The reaction mixtures contained 0.05 M Tris-HCl at pH 8.0, 10-4 M 2-mercaptoethanol, 10-4 M 8-hydroxyquinoline, plus treated DNA and enzyme units as indicated in a volume of 0.25 ml. Depurinated T4 [3H]DNA was prepared by heating the DNA at pH 3.5 at 370 for 60 min, conditions which produced approximately 1 depurinated site per 1150 nucleotides. Incubations were for 1 hr at 370 and were stopped by the addition of NaDodSO4 and EDTA at final concentrations of 0.05% and 0.02 M, respectively. Centrifugation was through 3.6 ml of 5-20% alkaline or neutral sucrose gradients in an SW 56 Spinco rotor. (A) Alkaline sucrose gradient fractionation of depurinated reduced DNA. The reaction mixture was as described in Methods and contained 10 nmol of depurinated reduced DNA (2.4 X 103 cpm/nmol), plus the indicated units of either exonuclease III or apurinic acid endonuclease purified approximately 2000-fold. Centrifugation was at 35,000 rpm for 3 hr at 200. (B) Alkaline sucrose gradient fractionation of lightly alkylated DNA. T4 DNA was alkylated at a molar ratio of MeSO2OMe to DNA nucleotide of 10:1 as described in Methods. The reaction mixture contained 7 nmol of alkylated DNA (1.3 X 103 cpm/nmol) and the indicated units of en- donuclease II that had been purified 1500-fold. Centrifugation was at 30,000 rpm for 3 hr at 200. (C) Neutral sucrose gradient fractionation of heavily alkylated DNA. The DNA was alkylated at a molar ratio of 6000:1, MeSO2OMe to DNA nucleotide, as described (2). The reaction mixture contained 23.6 nmol of DNA (4.7 X 103 cpm/nmol) and enzyme as indicated. The sedimentation in the neutral gradient was at 28,000 rpm for 3 hr.

this fractionation on DEAE-cellulose is able to separate the this was exonuclease III was as follows: (i) the major purification major apurinic acid endonuclease activity, peak I, from the steps of the published procedure were followed. (ii) The en- major endonuclease II activity, peak III. zyme preparation possessed a 3'-phosphatase activity, as well as the ability to release acid-soluble fragments from labeled Separation of exonuclease III from apurinic acid DNA. (iii) The Pi released by this exonuclease III preparation endonuclease and endonuclease II was comparable to the Pi released from the same substrate by Exonuclease III is an enzyme which can be separated from the . (iv) The enzyme activity was inhibited apurinic acid endonuclease and from endonuclease II. There- 95% by 3 X 10-5 M ZnCl2. (v) Dr. C. C. Richardson very kindly fore, it is not the same enzyme as the apurinic acid endonuclease provided us with a sample of exonuclease III for comparison or endonuclease II (13). On DEAE-cellulose chromatography, with our preparation. The ratio of Pi release to acid-soluble exonuclease III elutes in an area overlapping the apurinic acid nucleotide release was the same for the two preparations. (vi) endonuclease in peak I (Fig. 1). A purified preparation of ex- The chromatographic behavior of both preparations on the onuclease III provided by Dr. Richardson was eluted in the DEAE-cellulose column was similar. (vii) The molecular weight same position. It is apparent that the exonuclease III activity by gel filtration of the exonuclease III purified in this laboratory does not coincide exactly with the apurinic acid endonuclease. was 25,500 to 26,000, while the molecular weight of Dr. Rich- Pi release was also observed in the area of peak II, but does not ardson's preparation was 26,000. Therefore, although the represent either exonuclease III activity or alkaline phosphatase preparation purified in this laboratory was not homogeneous, activity. The nature of this activity is unknown. we can conclude that it is exonuclease III by the above crite- Exonuclease III could be partially separated from the apur- ria. inic acid endonuclease in early stages of purification because Exonuclease III purified in this laboratory had no endonu- of a difference in molecular weights. When a fraction of the cleolytic activity on either apurinic acid sites or alkylated sites apurinic acid endonuclease that had been purified 500-fold was in DNA (Fig. 2A and B). T4 DNA was depurinated and then passed through a Sephadex G-100 column, the two activities reduced with NaBH4 to prevent alkali-catalyzed phosphodiester overlapped, but did not coincide. In this experiment, the mo- bond hydrolysis (4). After incubation with or without exonu- lecular weight of the exonuclease III was 25,500, as opposed to clease III or the apurinic acid endonuclease, samples were ex- that of the apurinic acid endonuclease of 31,500. amined in alkaline sucrose gradients. Fig. 2A shows that exo- Exonuclease III was purified 2300-fold. The evidence that nuclease III was unable to recognize depurinated sites that were Downloaded by guest on September 25, 2021 Biochemistry: Kirtikar et al. Proc. Natl. Acad. Sci. USA 73 (1976) 4327

Table 1. Activities of purified enzymes on various substrates Percent units/unit of enzyme tested MeSO2OMe-treated Depurinated Enzyme Fold purification Units/mg DNA gel reduced DNA gel 32Pi release Endonuclease II 12,500 295 (100) 7.1 0 Apurinic acid endonuclease 11,000 365 12.4 (100) 0 Exonuclease III 2,300 2300 0.046* 0.044* (100) * Determined by the gel assay; no activity when determined by the gradient technique (Fig. 2).

recognized by the apurinic acid endonuclease. Likewise, exo- to be associated with the exonucleolytic activity (13) was ac- nuclease III was unable to recognize sites in the DNA due to tually due to the exonucleolytic action of the purified exonu- alkylation with MeSO2OMe (Fig. 2B). These sites were recog- clease III, which was able to produce double-strand breaks nized by endonuclease II. without making single-strand breaks. The apurinic acid endonuclease purified 11,000-fold has A false assay was also observed in this laboratory with negligible exonuclease III activity, as measured by Pi release depurinated DNA. Exonuclease III, lacking the apurinic acid (Table 1). Thus, the apurinic acid endonuclease activity can be endonuclease activity (Fig. 2A), was chromatographed on Se- shown not to coincide with the exonuclease III activity on phadex G-100 and the fractions were examined both by the Pi DEAE-cellulose and Sephadex G-100 columns, and the purified release assay for exonuclease III and by the gel assay using DNA endonuclease does not have significant levels of exonuclease with a small number of depurinated sites. Both activities par- III. The low activity observed with alkylated DNA (Table 1) alleled each other but the endonucleolytic activity on the gel may be due to depurinated sites in this substrate. Endonuclease was very low compared to the Pi release and very low compared II activity, isolated from the DEAE-cellulose column, did not to the usual release by the apurinic acid endonuclease (see also have any contaminating exonuclease III activity (Fig. 1, peak Table 1). The experiments described in this section indicate that III). Also, in the preparation purified 12,500-fold there was no exonuclease III does not recognize apurinic or alkylated sites significant exonuclease III 3'-phosphatase activity (Table 1). in DNA, but does give false-positive reactions, especially when The low activity observed with depurinated reduced DNA heavily alkylated DNA is used. seems to be an intrinsic property of the enzyme. Exonuclease III, therefore, has no endonucleolytic activity DISCUSSION directed against depurinated or alkylated sites, and as noted The confusion regarding the nomenclature of the phosphodi- above, the purified apurinic acid endonuclease and the endo- of E. coli that recognize depurinated sites and alkyl- nuclease II preparations have no significant exonuclease III ated sites has been extensive. The term endonuclease II was activity. originally used in this laboratory to define an enzyme active on DNA treated with MeSO2OMe (2, 3). We then concluded er- False-positive assays of endonucleolytic activities by exonuclease III roneously that the activity on depurinated DNA was due to the same enzyme (4). However, Verly et al. (5, 6, 17) demonstrated One basis of the claim that exonuclease III and endonuclease that the apurinic acid endonuclease had no activity on alkylated II were the same protein (13) was the use of the DNA-gel assay. DNA. The two activities have now been separated, and each The conclusion was that a homogeneous protein with a mo- has been purified to a single band on NaDodSO4-gel electro- lecular weight of 28,000 had exonuclease, 3'-phosphatase, and phoresis. There is a third fraction, as yet not purified exten- endonuclease activity (13). If DNA is released from the gel by sively, which contains both activities. We propose that the en- exonucleolytic activity, then an erroneous conclusion could be zyme that recognizes depurinated sites be called apurinic acid drawn. To measure the "endonuclease II" activity of the pu- endonuclease and that the enzyme that recognizes alkylated rified protein (13) the method that was originally described in DNA as well as other substrates (8-11) be called endonuclease our laboratory was used in which DNA is entrapped in a poly- II. acrylamide gel and then alkylated heavily with MeSO2OMe Endonuclease II has both a and an N- (2). We show here that exonuclease III can make double-strand glycosidase activity (9, 10). Both of these activities parallel each breaks in heavily alkylated DNA (Fig. 2C) under conditions other during chromatography of fractions that are purified where there is no endonucleolytic activity (Fig. 2A and B). T4 12,500-fold (Kirtikar, unpublished observations). The suggestion DNA was treated at the high MeSO2OMe to nucleotide ratio that these are separate enzymes, an N-glycosidase and a phos- and then used as a substrate. After incubation with the enzyme, phodiesterase, that act sequentially (8) cannot be correct since the DNA was examined in neutral sucrose gradients to look for kinetic experiments with alkylated DNA and with depurinated double-strand breaks (Fig. 2C). It is apparent that double-strand DNA show that the rate of phosphodiester bond breakage with breaks occur with the exonuclease III preparation, which does depurinated DNA is far too slow to account for the rate of not make single-strand breaks at either apurinic (Fig. 2A) or phosphodiester bond breakage with alkylated DNA (Kirtikar, alkylated (Fig. 2B) sites. We suspect that the heavily alkylated unpublished observations). DNA can undergo chemical depurination, and at some of these Mutants, isolated by others, have been shown in this labora- depurinated sites, 3-elimination with phosphodiester bond tory to lack either the apurinic acid endonuclease or endonu- hydrolysis occurs. This creates sites for exonucleolytic action. clease II. The mutant isolated by Yajko and Weiss, BW 2001 If two sites are near but on opposite strands, exonucleolytic (12), lacks the apurinic acid endonuclease in peak I when grown action will result in a double-strand break. Double-strand breaks at 420, but has a normal level of activity for MeSO2OMe-treated are required to release DNA from a polyacrylamide gel (14). DNA in peak III (8). A mutant, AB3027, isolated by Howard- Therefore, we feel that the endonucleolytic activity observed Flanders (19), has no endonuclease II activity (peak III), but has Downloaded by guest on September 25, 2021 4328 Biochemistry: Kirtikar et al. Proc. Natl. Acad. Sci. USA 73 (1976) a normal apurinic acid activity in peak I (8). These observations the hypothesis that these enzymes arise from a larger molecule provide further evidence that these are separate proteins coded by proteolytic cleavage. for by separate genes. Exonuclease III has been claimed to be endonuclease II (as We thank Irene Ukstins for excellent technical assistance and Dr. defined by an activity on heavily alkylated DNA-gel) on the C. C. Richardson for the gift of exonuclease III. This research was basis of genetic evidence (12), and also on the basis of a homo- supported by grants from the National Institutes of Health (CA 11322), geneous protein, purified 1600-fold, which showed both ac- the American Cancer Society, Cuyahoga County Unit, and by a con- tivities (13). We have provided the following evidence to prove tract with ERDA (11-1)2725. D.A.G. is a recipient of National Institutes that exonuclease III is not endonuclease II: (i) the apurinic acid of Health Research Career Award Fellowship (K-6-GM-21444). endonuclease has been purified 1 1,000-fold and has no signif- icant 3'-phosphatase activity characteristic of exonuclease III. 1. Strauss, B. & Robbins, M. (1968) Biochim. Biophys. Acta 161, Its molecular weight is approximately 31,500. (ii) Endonuclease 68-75. 2. Friedberg, E. C. & Goldthwait, D. A. (1969) Proc. Nati. Acad. II has been purified 12,500-fold and also has no significant 3'- Sci. USA 62,934-940. phosphatase activity. Its molecular weight is approximately 3. Friedberg, E. C., Hadi, S. M. & Goldthwait, D. A. (1969) J. Biol. 33,000. (iii) Exonuclease III has been purified 2300-fold and Chem. 244,5879-5889. has no endonucleolytic activity on depurinated or alkylated sites 4. Hadi, S. M. & Goldthwait, D. A. (1971) Biochemistry 10, in DNA. Its molecular weight is approximately 26,000. The 4986-4994. probable explanation for the conclusion that exonuclease III was 5. Verly, W. G. & Paquette, Y. (1972) Can. J. Biochem. 50,217- the same enzyme as endonuclease II (13) lies in the ability of 224. the exonuclease to make double-strand breaks in heavily al- 6. Paquette, Y., Crine, P. & Verly, W. G. (1972) Can. J. Biochem. kylated DNA under conditions where the enzyme is unable to 50, 1199-1209. 7. Kirtikar, D. & Goldthwait, D. A. (1975) Fed. Proc. 34,515. make single-strand breaks at depurinated-reduced or alkylated 8. Kirtikar, D. M., Kuebler, J. P., Dipple, A. & Goldthwait, D. A. sites. It should be emphasized that results such as those shown (1976) in 8th Miami Winter Symposium (Academic Press, New in Fig. 1 were obtained with DNA-gel assays in which the DNA York), in press. was either depurinated sparingly or alkylated lightly and the 9. Kirtikar, D. M. & Goldthwait, D. A. (1974) Proc. Nati. Acad. Sci. gels were used for testing immediately after preparation. Pu- USA 71, 2022-2026. rified exonuclease III has little activity on this depurinated gel 10. Kirtikar, D. M., Dipple, A. & Goldthwait, D. A. (1975) Bio- and no activity on this alkylated gel. chemistry 14, 5548-5553. However, these results do not close the issue. Yajko and Weiss 11. Kirtikar, D. M., Slaughter, J. & Goldthwait, D. A. (1975) Bio- claimed that their thermosensitive mutant, BW 2001, was chemistry 14, 1235-1244. 12. Yajko, D. M. & Weiss, B. (1975) Proc. Natl. Acad. Sci. USA 72, missing both exonuclease III and endonuclease II (12). We have 688-692. examined the chromatographic pattern of this mutant, grown 13. Weiss, B. (1976) J. Biol. Chem. 251, 1896-1901. at elevated temperature, and shown that it is missing both the 14. Melgar, E. & Goldthwait, D. A. (1968) J. Biol. Chem. 243, apurinic acid endonuclease as well as exonuclease III. They also 4401-4408. claimed that AB3027 was missing both exonuclease III and 15. Richardson, C. C. & Kornberg, A. (1964) J. Biol. Chem. 239, endonuclease II. We have observed that the 3'-phosphatase 242-250. activity of exonuclease III as well as the endonuclease II activity 16. Kushner, S. R., Nagaishi, A., Templin, A. & Clark, A. J. (1971) on MeSO2OMe-treated DNA (peak III) are both missing in this Proc. Natl. Acad. Sci. USA 68,824-827. mutant but the apurinic acid endonuclease is normal. Fur- 17. Verly, W. G. & Rassart, E. (1975) J. Biol. Chem. 250, 8214- BW 8219. thermore, a deletion mutant, 9109, is missing all three 18. Lindahl, T. (1976) Nature 259,64-66. enzymes. These data suggest that the genes for endonuclease 19. Ljungquist, S., Lindahl, T. & Howard-Flanders, P. (1976) J. II, apurinic acid endonuclease, and exonuclease III are all Bacteriol. 126, 646-653. clustered in the region at 38.2 min on the revised map (20), as 20. Bachman, B. J., Low, K. B. & Taylor, A. L. (1976) Bacteriol. Rev. determined by Weiss for exonuclease III (21). The data also 40, 116-167. suggest that AB 3027 and BW 2001 are double mutants, but this 21. White, B. J., Hochhauser, S. J., Cintron, N. M. & Weiss, B. (1976) requires further investigation. There is no evidence to support J. Bacteriol., 126, 1082-1088. Downloaded by guest on September 25, 2021