DNA degradation in minicells of Escherichia coli K-12

Summary The properties of minicell producing mutants of Escherichia coli deficient in gentic recombination were examined. Experiments were designed to test recombinant formation in conjugal crosses, survival following UV-irradiation in cells, and the state of DNA metabolism in minicells. The REC^- phenotypes are unaffected by min ⁺/^- genotypes in whole cells. In contrast to minicells produced by rec ⁺ parental cells, minicells from a recB21 strain have limited capacity to degrade linear, Hfr transferred DNA. The lack of a functional recA gene product, presumably involved in inhibiting the recBC nuclease action(s), permits unrestricted Hfr DNA breakdown in minicells produced by a recA1 strain. This results in an increase in TCA soluble products and in the formation of small DNA molecules that sediment near the top of an alkaline sucrose gradient. Unlike the linear DNA, circular duplex DNA from plasmids R64-11 or dv, segregated into the minicells, is resistant to breakdown. By using in vitro criteria, and [³²P]-labelled linear DNA from bacteriophage T₇ for substrate, we found that the ATP-dependent exonuclease of the recBC complex (exo V) is present in rec ⁺ and recA^- minicells, and is lacking in the recB21 mutant. In fact, the absence of a functional exo V in recBC^- minicells results in isolation of larger than average Hfr DNA from minicells. We suggest that recombination (REC) enzymes segregate into the polar minicells at the time of minicell biogenesis. This system should be useful for studies on DNA metabolism and functions of the recBC and recA gene products.Paper 1 in series, see Khachatourians et al., 1974.     

Mechanism of conjugation and recombination in bacteria. XVIII. Polarity of donor DNA strands transferred to the recipient as determined by DNA-RNA hybridization.

Polarity of donor DNA strand transferred into recipient during conjugation in Escherichia coli K-12 was determined by DNA-3H-RNA hybridization. Lambda prophage was used as a marker. The defective lysogen Hfr H (lambdat11) as a donor and thermosensitive F- CR34 dnaB strain as recipient were used. Two sets of hybridization experiments, with 1-strand specific lambda mRNA and lambda mRNA specific for both phage strands but with large excess of r-strand specific mRNA, were carried out. Strand 1 of lambda DNA was detected preferentially in recipient cells mated at restrictive temperature, when Hfr transferred its genophore in the order gal-lambda-bio. Thus the genophore is transferred with 5'OH at its origin.     

Gaps in DNA synthesized by ultraviolet light-irradiated WI38 human cells.

DNA replication in ultraviolet-irradiated human cells was examined by treatment of the extracted DNA with a single-strand specific endonuclease from Neurospora crassa. WI38 cells were uniformly labeled with 32Pi for two generations before irradiation and then labeled with [3H]thymidine after irradiation. The isolated DNA was sedimented in neutral sucrose gradients after incubation with the endonuclease. The endonuclease treatment had no effect on the sedimentation profiles of either [32P]DNA or [3H]DNA from unirradiated control cultures. The endonuclease treatment also did not significantly alter the profile of [32P]DNA from irradiated cultures but did introduce breaks in the 3H pulse-labeled DNA synthesized after irradiation. These results indicate that DNA synthesis after ultraviolet irradiation proceeds in such fashion that gaps are formed along the newly made strand, leaving regions of single strandness in template DNA. As replication proceeds these gaps disappear and 2 h after irradiation (100-250 ergs/mm2) they are barely detectable by the endonuclease assay.     

Molecular Studies on Entry Exclusion in Escherichia coli Minicells

Minicells produced by abnormal cell division in a strain of Escherichia coli (K-12) have been employed here to investigate the phenomenon of "entry exclusion." When purified minicells from strains containing F' or R factors, or both, are mated with radioactive thymidine-labeled Hfr or R(+) donors, the recipient minicells can be conveniently separated from normal-sized donors following mating, and the products of conjugation can be analyzed in the absence of donors and of further growth of the recipients. Transmissible plasmids or episomes are transferred less efficiently to purified minicells derived from strains carrying similar or related elements than to strains without them. Measurement of deoxyribonucleic acid (DNA) degradation and determination of weight-average molecular weights following transfer indicate that degradation of transferred DNA or transfer of smaller pieces cannot account for the comparative reduction in transfer to entry-excluding recipients. Therefore, we conclude that entry exclusion operates to prevent the physical entry of DNA into recipients expressing the exclusion phenotype. The R-produced repressor (product of the drd(+) gene), which represses fertility (i.e., ability to act as donor), reduces exclusion mediated by R or F factor, or both, in matings between strains carrying homologous elements. Furthermore, the data suggest that the presence of the F pilus or F-like R pilus on recipient cells ensures maximum expression of the exclusion phenotype but is not essential for its expression. In contrast to previous suggestions, we found no evidence for a reduction of entry exclusion attributable to the DNA temperature-sensitive chromosomal mutation dnaB(TS).     

Deoxyribonucleic Acid Transferred from Ultraviolet-Irradiated Excision-Defective Hfr Cells of Escherichia coli K-12

Deoxyribonucleic acid (DNA) transfer from (3)H-thymine-labeled Hfr cells has been measured by determining the amount of radioactivity remaining after selective lysis of the donor cells in the mating mixture. DNA transfer was less effectively reduced by ultraviolet irradiation of excision-defective Hfr cells than was the yield of recombinants. The buoyant density of DNA transferred from unirradiated and irradiated Hfr cells was equivalent to that of double-stranded DNA. Mating-dependent DNA synthesis in the recipient has been measured by mating Hfr cells deficient in thymidine kinase with irradiated thymine-requiring F(-) cells in the presence of (3)H-thymine. The extent of such DNA synthesis approximated the amount of DNA transferred from unirradiated donors. Neither DNA transfer nor mating-dependent DNA synthesis could be reliably measured when both parents were irradiated. It is proposed that transferred Hfr DNA is replicated in the recipient and that this replication still occurs when the Hfr DNA contains dimers.     

Minicell production and bacteriophage superinducibility of thymidine-requiring strains of Haemophilus influenzae.

Aminopterin- or trimethoprin-resistant thymidine-requiring strains of Haemophilus influenzae produce minicells, and the ratio of minicells to cells increases during the stationary phase of growth. Strain LB11, isolated after mutagenesis of a thymidine-requiring strain (Rd thd), produces more minicells than the parent strain. The mutations involved in high frequency minicell production have been transferred into the wild type (strain Rd) by transformation. The thymidine requirement in the resulting strain, MCl, is essential for minicell production, since spontaneous revertants of MCl to prototrophy do not produce minicells. The ratio of minicells to cells was increased more than 10(3)-fold by differential centrifugation. The minicells contain little or no deoxyribonucleic acid (DNA). Phage HPlcl apparently cannot attach to minicells. Competent cells of LB11 and its thymidine-requiring parent strain produce defective phage as a result of exposure to transforming DNA, whereas only LB11 produces many defective phage in response to the competence regime alone. Competent HP1c1 and S2 lysogens of MC1 and Rd thd are also superinducible by transforming DNA, but competent LB11 lysogens produced about the same amount of HP1c1 or S2 phage with or without exposure to transforming DNA possibly because of competition between the induced defective phage and Hp1c1 or S2 phage.     

DNA synthesis in toluene-treated bacteriophage-infected minicells or Bacillus subtilis.

Bateriophage (phi29, SPP1, or SPO1)-infected, toluene-treated minicells of Bacillus subtilis are capable of limited amounts of non-replicative DNA synthesis as measured by incorporation of [3H]dTTP into a trichloroacetic acid-precipitable form. The [3H]dTTP is covalently incorporated into small DNA fragments which result from the degradation of a small percentage of the infecting phage genomes (molecular weights in the range of 2 . 10(5)). Short exposure of the DNA molecules containing the incorporated [3H]dTMP to Escherichia coli exonuclease III results in over 90% of the E13H]dTMP being converted to a trichloroacetic acid-soluble form. The synthesis is totally dependent on host-cell enzymes and is not inhibited by the addition of chloramphenicol, rifampicin, nalidixic acid and mitomycin C and only slightly (approx. 20%) inhibited by the addition of 6-(p-hydroxyphenylazo)-uracil.     

Minicells of Bacillus subtilis

After nitrosoguanidine (N-methyl-N'-nitro-N-nitrosoguanidine) mutagenesis, two Bacillus subtilis mutants (div IV-A1 and div IV-B1) were isolated that are defective in the location of division site along cell length. Both mutations were transferred into strain CU403 by transformation, and their properties were studied in the CU403 genetic background. Location of divisions in close proximity to cell pole regions in both mutants results in minicell production. Purified minicells contain a ratio of ribonucleic acid to protein comparable to that found in the parent cells. Autoradiographs of (3)H-thymine incorporation into deoxyribonucleic acid (DNA), thymine-2-(14)C incorporation into DNA, electron micrographs, and chemical analyses for DNA all fail to demonstrate DNA in the minicells. Minicells produced by both mutants are highly motile, an indication of functional energy metabolism. Electron micrographs reveal that minicells are produced by a structurally normal division mechanism and that minicells contain a normal cell surface. The div IV-A1 mutation has been mapped by PBS1 transduction linked to ura. The div IV-B1 mutation is closely linked to pheA by both PBS1 transduction and by co-transformation.     

Photoreactivation, Excision, and Strand-Rejoining Repair in R Factor-Containing Minicells of Escherchia coli K-12

Chromosomeless “minicells” are formed by misplaced cell fissions near the polar extremities of an Escherichia coli K-12 mutant strain. Resistance (R)-factor deoxyribonucleic acid (DNA) can be introduced into minicells by segregation from an R⁺ (R64-11) derivative of the original mutant. We have assessed the ability of R⁺ minicells to correct defects produced in their plasmid DNA by ultraviolet (UV) and gamma radiations. Minicells harboring plasmid DNA, in comparison with their repair-proficient minicell-producing parents, possess (i) an equal competence to rejoin single-strand breaks induced in DNA by gamma rays, (ii) a reduced capacity for the photoenzymatic repair of UV-induced pyrimidine dimers, and (iii) a total inability to excise dimers, apparently owing to a deficiency in UV-specific endonuclease activity responsible for mediating the initial incision step in excision repair. Assuming that the DNA repair properties of R⁺ minicells reflect the concentration of repair enzymes located in the plasmid-containing polar caps of entire cells, these findings suggest that: (i) the enzymes responsible for rejoining single-strand breaks are distributed throughout the cell; (ii) photoreactivating enzyme molecules tend to be concentrated near bacterial DNA and to a lesser extent near plasmid DNA; and (iii) UV-specific endonuclease molecules are primarily confined to the central region of the E. coli cell and, thus, seldom segregate with R-factor DNA into minicells.     

Repair of daughter strand gaps in nascent DNA from mouse epidermal cells treated with dihydrodiol epoxide derivatives of benzo[a]pyrene

Alkaline sucrose gradient analysis of [methyl-3H]thymidine-pulse-labeled DNA was used to study the effect of (+/-)-7 beta,8 alpha-dihydroxy-9 alpha,10 alpha-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (benzo[a]pyrene-diol epoxide I), a potent mutagen and carcinogen, and (+/-)-7 beta,8 alpha-dihydroxy-9 beta,10 beta-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (benzo[a]pyrene-diol epoxide II), a weaker mutagen and carcinogen, on the size of newly synthesized DNA in primary cultures of mouse epidermal cells. Both isomers caused a dose-dependent decrease in the size of newly synthesized DNA and in the rate of [methyl-3H]thymidine incorporation into DNA. When the pulse time was increased in the treated cells so that the amount of [methyl-3H]thymidine incorporation was equal to the control, newly synthesized DNA from exposed cells was still considerably smaller than DNA from control cells. The low molecular weight of the nascent DNA from treated cells was consistent with, but not indicative of, the presence of gaps in the nascent DNA from the treated cells. Evidence of gapped DNA synthesis was obtained by treatment of extracted DNA with a single-strand specific endonuclease from Neurospora crassa. The endonuclease treatment did not significantly alter the profile of [methyl-3H]thymidine prelabeled DNA from benzo[a]pyrene-diol epoxide-treated cultures but did introduce double-stand breaks in pulse-labeled DNA from treated cultures. The numbers of [14C]benzo[a]pyrene-diol epoxide I or [3H]benzo[a]pyrenediol epoxide II-DNA-bound adducts and daughter strand gaps were compared at several dose levels. Treatment with either isomer yielded one gap in the nascent DNA/DNA-bound adduct. Pulse-chase experiments showed that gaps in the nascent DNA were closed with time.     

Manganese Stimulates the Incorporation of [3H]Inositol into a Pool of Phosphatidylinositol in Brain That Is Not Coupled to Agonist-Induced Hydrolysis

Mn2+ greatly increases the incorporation of myo-[3H]inositol into phosphatidylinositol (PI) of brain and other tissues by stimulating the activity of a PI-myo-inositol exchange enzyme. This study examined the ability of norepinephrine (NE) and carbachol to stimulate the hydrolysis of [3H]PI formed in the absence and presence of Mn2+-stimulated [3H]inositol exchange. Rat cerebral cortical slices were incubated with myo-[3H]inositol for 60 min in an N-2-hydroxyethyl piperazine-N'-2-ethanesulfonic acid (HEPES) buffer without or with MnCl2 (1 mM). The tissue was washed and further incubated with unlabeled myo-inositol and LiCl (10 mM). Prelabeled slices were then incubated with NE (0.1 mM) or carbachol (1 mM) to induce agonist-stimulated [3H]PI hydrolysis. Mn2+ treatment resulted in eight- and sixfold increases in control levels of [3H]PI and [3H]inositol monophosphate [( 3H]IP), respectively. Both NE and carbachol stimulated [3H]IP formation in tissue prelabeled without or with manganese. However, the degree of stimulation (percentage of control values) was greatly attenuated in the presence of Mn2+. In the absence of Mn2+ treatment, NE decreased [3H]PI radioactivity in the tissue to 80% of control values. However, NE did not decrease [3H]PI radioactivity in the Mn2+-treated tissue. These data demonstrate that Mn2+ stimulates incorporation of myo-[3H]inositol into a pool of PI in brain that has a rapid turnover but is not coupled to agonist-induced hydrolysis.     

Properties of receptors for the Ca2+-channel blocker verapamil in transverse-tubule membranes of skeletal muscle. Stereospecificity, effect of Ca2+ and other inorganic cations, evidence for two categories of sites and effect of nucleoside triphosphates

The verapamil receptor associated with the voltage-dependent calcium channel of rabbit skeletal muscle transverse tubule membranes has the following properties. (i) This receptor is stereospecific and discriminates between the different stereoisomers of verapamil, gallopamil and diltiazem. (ii) Inorganic divalent cations inhibit the binding of [3H]verapamil to its receptor in an apparently non-competitive fashion. The rank order of potency is: Ca2+ = Mn2+ greater than Mg2+ greater than Sr2+ greater than Ba2+ much greater than Co2+ much greater than Ni2+. Ca2+ and Mn2+ have inhibition constants of 0.3 mM. Binding of [3H]verapamil is also sensitive to monovalent cations such as Cs+, K+, Li+ and Na+. The most active of these cations (Cs+ and K+) have inhibition constants in the range of 30 mM. (iii) Binding of [3H]verapamil is pH-dependent and reveals the presence on the verapamil receptor of an essential ionizable group with a pKa of 6.5. (iv) A low-affinity binding site for verapamil and for some other Ca2+ channel blockers is detected by studies of dissociation kinetics of the [3H]verapamil receptor in the presence of high concentrations of verapamil, gallopamil, bepridil and diltiazem. (v) GTP and nucleoside analogs change the properties of [3H]verapamil binding to verapamil binding sites. High-affinity binding sites seem to be transferred into low-affinity sites. Dissociation constants obtained from inhibition studies of [3H]verapamil binding are in the range of 0.1-0.3 mM for GTP, ATP and Gpp(NH)p.     

UV-induced pyrimidine hydrates in DNA are repaired by bacterial and mammalian DNA glycosylase activities

Escherichia coli endonuclease III and mammalian repair enzymes cleave UV-irradiated DNA at AP sites formed by the removal of cytosine photoproducts by the DNA glycosylase activity of these enzymes. Poly(dG-[3H]dC) was UV irradiated and incubated with purified endonuclease III. 3H-Containing material was released in a fashion consistent with Michaelis-Menten kinetics. This 3H material was determined to be cytosine by chromatography in two independent systems and microderivatization. 3H-Containing material was not released from nonirradiated copolymer. When poly(dA-[3H]dU) was UV irradiated, endonuclease III released 3H-containing material that coeluted with uracil hydrate (6-hydroxy-5,6-dihydrouracil). Similar results are obtained by using extracts of HeLa cells. There results indicate that the modified cytosine residue recognized by endonuclease III and the mammalian enzyme is cytosine hydrate (6-hydroxy-5,6-dihydrocytosine). Once released from DNA through DNA-glycosylase action, the compound eliminates water, reverting to cytosine. This is consistent with the known instability of cytosine hydrate. The repairability of cytosine hydrate in DNA suggests that it is stable in DNA and potentially genotoxic.     

Measurement of interstrand cross-link frequency and distance between interruptions in DNA exposed to 4,5',8-trimethylpsoralen and near-ultraviolet light

Bifunctional psoralens react photochemically with DNA to form single-strand adducts and interstrand, chemical cross-links. Cross-link formation is first order with [P], the concentration of added psoralen, when [P] much less than Kd, the psoralen-DNA dissociation constant. DNA molecules containing interstrand cross-links are reversibly bihelical and so are readily detected. It was not heretofore possible to determine cross-link frequency in polydisperse DNA from the mass F of DNA spared cross-linkage. We have derived a statistical relation to calculate cross-link frequency at fixed light exposure and variable [P]. We show here that S, the initial slope of the curve described by -ln F as a function of [P], is proportional to Mw, the weight-average molecular weight of nick-free DNA. The cross-link frequency at any [P] can be determined from k, a constant measured for DNA of known Mw at low cross-linkage. This relation is valid for DNA of any molecular weight distribution. In experiments with uniform length DNA, -ln F (cross-link frequency) increased in simple proportion to [P]. Intact and restriction endonuclease HindIII digested phage lambda DNA molecules have discrete lengths. S for each was proportional to Mw of the twin helix even though the molecular weight distribution of the restriction fragments was skewed. S was proportional to Mw and to the median molecular weight of sheared cellular DNA over a wide range. Also, we found that 1/S was linear with exposure of cellular DNA to gamma radiation. S can therefore be used to calculate L, the average distance between interruptions in the double helix.     

Incorporation of 3H-labeled nucleosides and 3H-labeled deoxynucleosides into detergent soluble DNA

Detergent soluble DNA from splenocytes of immunologically stimulated mice has been shown to incorporate [3H]dThd more rapidly than detergent insoluble DNA. In this report we compare the incorporation of other 3H-labeled nucleosides and 3H-labeled deoxynucleosides and the distribution of 3H in the different size classes of detergent soluble DNA. The order of incorporation into DS DNA is [3H]dThd greater than [3H]dCyd greater than [3H]Ado greater than [3H]dGuo approximately equal to [3H]Cyd greater than [3H]dAdo greater than [3H]Guo. We also show that the previously reported slight enrichment in Gua + Cyt content is not due to preferential incorporation of dGuo or of dCyd into any one size class.     

The interaction of ( 3 H)ethyl carbamate with nucleic acids of regenerating mouse liver

The binding of [³H]urethane to liver DNA and RNA has been examined in partially hepatectomised and intact male Crackenbush mice. A single dose of [³H]urethane (50 μCi) was given to non-hepatectomised mice (group A) and to 3 groups of partially hepatectomised mice at 18 (B), 28 (C) and 38 (D) hours postoperatively, respectively. The binding was examined over the subsequent 16 h. The maximum levels of binding to DNA declined in the order, group A > B > C > D, although the binding to DNA persisted longest in group B. The binding to RNA was greater in groups B, C and D than in group A. Neither the restoration of liver mass nor an alteration in the metabolism of urethane appeared to account for the different levels of binding. In normal and partially hepatectomised mice a single dose of urethane (20 mg) was followed by an inhibition of mitosis and of the incorporation of [³H]thymidine into liver DNA and of [³H]uridine into liver RNA.     

Changes in deoxyribonucleic acid polymerase activities in synthesis of deoxyribonucleic acid during sporulation of Bacillus subtilis.

The deoxyribonucleic acid (DNA) polymerase activities in Bacillus subtilis strains Marburg 168 (thy-trp2) and D22, a DNA polymerase I-deficient mutant, were measured at various stages of sporulation. The DNA polymerase I activity, which had decreased after the exponential growth, began to increase at the early stage of sporulation, reached a maximum and then again decreased. The activity of neither DNA polymerase II nor III was observed to change so drastically as that of DNA polymerase I during sporulation. The incorporation of [3H]deoxythymidine 5'-triphosphate ([3H]dTTP) into Brij 58-treated permeable cells increased during sporulation. The stimulation of [3H]dTTP incorporation into the cells by irradiation with ultraviolet light was also observed to coincide with DNA polymerase I activity. In strain D22 the activities of DNA polymerase II and III were almost constant with time. Neither change of [3H]dTTP incorporation into Brij 58-treated cells nor stimulation of incorporation by irradiation with ultraviolet light was observed.     

Titration of integrated simian virus 40 DNA sequences, using highly radioactive, single-stranded DNA probes.

Nick-translated simian virus 40 (SV40) [32P]DNA fragments (greater than 2 X 10(8) cpm/micrograms) were resolved into early- and late-strand nucleic acid sequences by hybridization with asymmetric SV40 complementary RNA. Both single-stranded DNA fractions contained less than 0.5% self-complementary sequences; both included [32P]-DNA sequences that derived from all regions of the SV40 genome. In contrast to asymmetric SV40 complementary RNA, both single-stranded [32P]DNAs annealed to viral [3H]DNA at a rate characteristic of SV40 DNA reassociation. Kinetics of reassociation between the single-stranded [32P]DNAs indicated that the two fractions contain greater than 90% of the total nucleotide sequences comprising the SV40 genome. These preparations were used as hybridization probes to detect small amounts of viral DNA integrated into the chromosomes of Chinese hamster cells transformed by SV40. Under the conditions used for hybridization titrations in solution (i.e., 10- to 50-fold excess of radioactive probe), as little as 1 pg of integrated SV40 DNA sequence was assayed quantitatively. Among the transformed cells analyzed, three clones contained approximately one viral genome equivalent of SV40 DNA per diploid cell DNA complement; three other clones contained between 1.2 and 1.6 viral genome equivalents of SV40 DNA; and one clone contained somewhat more than two viral genome equivalents of SV40 DNA. Preliminary restriction endonuclease maps of the integrated SV40 DNAs indicated that four clones contained viral DNA sequences located at a single, clone-specific chromosomal site. In three clones, the SV40 DNA sequences were located at two distinct chromosomal sites.     

Escherichia coli endonuclease III is not an endonuclease but a beta-elimination catalyst.

The oligonucleotide [5'-32P]pdT8d(-)dTn, containing an apurinic/apyrimidinic (AP) site [d(-)], yields three radioactive products when incubated at alkaline pH: two of them, forming a doublet approximately at the level of pdT8dA when analysed by polyacrylamide-gel electrophoresis, are the result of the beta-elimination reaction, whereas the third is pdT8p resulting from beta delta-elimination. The incubation of [5'-32P]pdT8d(-)dTn, hybridized with poly(dA), with E. coli endonuclease III yields two radioactive products which have the same electrophoretic behaviour as the doublet obtained by alkaline beta-elimination. The oligonucleotide pdT8d(-) is degraded by the 3'-5' exonuclease activity of T4 DNA polymerase as well as pdT8dA, showing that a base-free deoxyribose at the 3' end is not an obstacle for this activity. The radioactive products from [5'-32P]pdT8d(-)dTn cleaved by alkaline beta-elimination or by E. coli endonuclease III are not degraded by the 3'-5' exonuclease activity of T4 DNA polymerase. When DNA containing AP sites labelled with 32P 5' to the base-free deoxyribose labelled with 3H in the 1' and 2' positions is degraded by E. coli endonuclease VI (exonuclease III) and snake venom phosphodiesterase, the two radionuclides are found exclusively in deoxyribose 5-phosphate and the 3H/32P ratio in this sugar phosphate is the same as in the substrate DNA. When DNA containing these doubly-labelled AP sites is degraded by alkaline treatment or with Lys-Trp-Lys, followed by E. coli endonuclease VI (exonuclease III), some 3H is found in a volatile compound (probably 3H2O) whereas the 3H/32P ratio is decreased in the resulting sugar phosphate which has a chromatographic behaviour different from that of deoxyribose 5-phosphate. Treatment of the DNA containing doubly-labelled AP sites with E. coli endonuclease III, then with E. coli endonuclease VI (exonuclease III), also results in the loss of 3H and the formation of a sugar phosphate with a lower 3H/32P ratio that behaves chromatographically as the beta-elimination product digested with E. coli endonuclease VI (exonuclease III). From these data, we conclude that E. coli endonuclease III cleaves the phosphodiester bond 3' to the AP site, but that the cleavage is not a hydrolysis leaving a base-free deoxyribose at the 3' end as it has been so far assumed. The cleavage might be the result of a beta-elimination analogous to the one produced by an alkaline pH or Lys-Trp-Lys. Thus it would seem that E. coli 'endonuclease III' is, after all, not an endonuclease.