Single-stranded DNA ligation and XLF-stimulated incompatible DNA end ligation by the XRCC4-DNA ligase IV complex: influence of terminal DNA sequence

The double-strand DNA break repair pathway, non-homologous DNA end joining (NHEJ), is distinctive for the flexibility of its nuclease, polymerase and ligase activities. Here we find that the joining of ends by XRCC4-ligase IV is markedly influenced by the terminal sequence, and a steric hindrance model can account for this. XLF (Cernunnos) stimulates the joining of both incompatible DNA ends and compatible DNA ends at physiologic concentrations of Mg²⁺, but only of incompatible DNA ends at higher concentrations of Mg²⁺, suggesting charge neutralization between the two DNA ends within the ligase complex. XRCC4-DNA ligase IV has the distinctive ability to ligate poly-dT single-stranded DNA and long dT overhangs in a Ku- and XLF-independent manner, but not other homopolymeric DNA. The dT preference of the ligase is interesting given the sequence bias of the NHEJ polymerase. These distinctive properties of the XRCC4-DNA ligase IV complex explain important aspects of its in vivo roles.     

The topography of lambda DNA: polyriboguanylic acid binding sites and base composition

The positions of the ten poly(rG) binding sites on the two strands of λ DNA have been determined with the aid of the families of overlapping DNA fragments described in the preceding paper and the DNA of deletion mutants. Nine sites are on the r-strand ^‡: two are located in the right half, 0·84 ± 0·03 and 0·91 ± 0·02 molecular length unit from the left end; and seven are evenly distributed within 0·4 unit from the left end. The tenth site is on the l-strand, 0·60 ± 0·03 unit from the left end. Another site was mapped on the r-strand of the bio segment of Escherichia coli DNA found in λ bio variants; the l-strand of this segment contains no sites.     

Kinetic control of Mg2+-dependent melting of duplex DNA ends by Escherichia coli RecBC

Escherichia coli RecBCD is a highly processive DNA helicase involved in double-strand break repair and recombination that possesses two helicase/translocase subunits with opposite translocation directionality (RecB (3′ to 5′) and RecD (5′ to 3′)). RecBCD has been shown to melt out ∼ 5-6 bp upon binding to a blunt-ended duplex DNA in a Mg²⁺-dependent, but ATP-independent reaction. Here, we examine the binding of E. coli RecBC helicase (minus RecD), also a processive helicase, to duplex DNA ends in the presence and in the absence of Mg²⁺ in order to determine if RecBC can also melt a duplex DNA end in the absence of ATP. Equilibrium binding of RecBC to DNA substrates with ends possessing pre-formed 3′ and/or 5′ single-stranded (ss)-(dT)_n flanking regions (tails) (n ranging from zero to 20 nt) was examined by competition with a fluorescently labeled reference DNA and by isothermal titration calorimetry. The presence of Mg²⁺ enhances the affinity of RecBC for DNA ends possessing 3′ or 5′-(dT)_n ssDNA tails with n < 6 nt, with the relative enhancement decreasing as n increases from zero to six nt. No effect of Mg²⁺ was observed for either the binding constant or the enthalpy of binding (ΔH_obs) for RecBC binding to DNA with ssDNA tail lengths, n ≥ 6 nucleotides. Upon RecBC binding to a blunt duplex DNA end in the presence of Mg²⁺, at least 4 bp at the duplex end become accessible to KMnO₄ attack, consistent with melting of the duplex end. Since Mg²⁺ has no effect on the affinity or binding enthalpy of RecBC for a DNA end that is fully pre-melted, this suggests that the role of Mg²⁺ is to overcome a kinetic barrier to melting of the DNA by RecBC and presumably also by RecBCD. These data also provide an accurate estimate (ΔH_obs = 8 ± 1 kcal/mol) for the average enthalpy change associated with the melting of a DNA base-pair by RecBC.     

Catalysis of dehydrogenation of 4-trans-(N,N-dimethylamino)cinnamaldehyde by aldehyde dehydrogenase

4-trans-(N,N-dimethylamino)cinnamaldehyde (DACA) is a chromophoric and fluorogenic substrate of aldehyde dehydrogenase. Fluorescence of DACA is enhanced by binding to aldehyde dehydrogenase in the absence of catalysis both in the presence and absence of the coenzyme analogue 5′AMP. DACA binds to aldehyde dehydrogenase with a dissociation constant of 1–3 μM and stoichiometry of 2 mol mol⁻¹ enzyme. Incorporation of DACA during catalysis was also investigated and found to be 2 mol DACA mol⁻¹ enzyme. Effect of pH on the stoichiometry of DACA incorporation during catalysis has shown that DACA incorporation remained constant at 2 mol DACA mol⁻¹ enzyme, despite a 74-fold velocity enhancement between pH 5.0 and 9.0. Increase of pH increased decomposition of enzyme–acyl intermediate without affecting the rate-limiting step of the reaction. At pH 7.0 the pH stimulated velocity enhancement was 10-fold over that at pH 5.0; further velocity enhancement (11.5-fold that of pH 7.0) was achieved by 150 μM Mg²⁺ ions. The velocity at pH 7.0 with Mg²⁺ exceeded that of pH 9.0, and that at maximal pH stimulation at pH 9.5. It was observed that level of intermediate decreased to about 1 mol mol⁻¹ enzyme, indicating that Mg²⁺ ions increased the rate of decomposition of the enzyme–acyl intermediate and shifted the rate-limiting step of the reaction to another step in the reaction sequence.     

Variables affecting the selectivity and efficiency of retention of DNA fragments by E. coli RNA polymerase in the nitrocellulose-filter-binding assay

In this paper we characterize the effect of varying the solution conditions and filter-binding protocols on the extent and selectivity of DNA retention on nitrocellulose filters by DNA-binding proteins. These effects are illustrated by the binding interaction of Escherichia coli RNA polymerase with λ and T7 phage DNA restriction fragments. We present procedures which will help enhance the selective retention of some DNA restriction fragments over others. These include increasing the pH and salt concentration, decreasing the enzyme-to-DNA ratio, and including an appropriate washing step. Selective binding is not dependent on the presence of Mg²⁺. Although we only show data for RNA polymerase-DNA interactions, many of the principles discussed are likely to find practical applications in studying selective DNA-protein binding in general.     

In vitro methylation of bacteriophage lambda DNA by wild type (dam+) and mutant (damh) forms of the phage T2 DNA adenine methylase.

The wild-type (dam⁺) and mutant (dam^h) forms of the bacteriophage T2 DNA adenine methylase have been partially purified; these enzymes methylate the sequence, 5/t' … G-A-Py … 3′ (Hattman et al., 1978a). However, in vitro methylation studies using phage λ DNA revealed the following: (1) T2 dam⁺ and dam^h enzymes differ in their ability to methylate λ DNA; under identical reaction conditions the T2 dam^h enzyme methylated λ DNA to a higher level than did the dam⁺ enzyme. However, the respective methylation sites are equally distributed on the l and r strands. (2) Methylation with T2 dam^h, but not T2 dam⁺ protected λ against P1 restriction. This was demonstrated by transfection of Escherichia coli (P1) spheroplasts and by cleavage with R·EcoP1. (3) T2 dam⁺ and dam^h were similarly capable of methylating G-A-T-C sequences on λ DNA; e.g. λ·dam₃ DNA (contains no N⁶-methyladonine) methylated with either enzyme was made resistant to cleavage by R·DpnII. In contrast, only the T2 dam^h modified DNA was resistant to further methylation by M·EcoP1 (which methylates the sequence 5′ … A-G-A-C-Py … 3′; Hattman et al., 1978b). (4) λ·dam₃ DNA was partially methylated to the same level with T2 dam⁺ or T2 dam^h; the two enzymes produced different patterns of G-A-C versus G-A-T methylation. We propose that the T2 dam⁺ enzyme methylates G-A-C sequences less efficiently than the T2 dam^h methylase; this property does not entirely account for the large difference in methylation levels produced by the two enzymes.     

Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I

Circular (e.g. simian virus 40) and linear (e.g. λ phage) DNAs have been labeled to high specific radioactivities (>10⁸ cts/min per μg) in vitro using deoxynucleoside [α-³²P]triphosphates (100 to 250 Ci/mmol) as substrates and the nick translation activity of Escherichia coli DNA polymerase I. The reaction product yields single-stranded fragments about 400 nucleotides long following denaturation. Because restriction fragments derived from different regions of the nick-translated DNA have nearly the same specific radioactivity (cts/min per 10[su3] bases), we infer that nicks are introduced, and nick translation is initiated, with equal probability within all internal regions of the DNA. Such labeled DNAs (and restriction endonuclease fragments derived from them) are useful probes for detecting rare homologous sequences by in situ hybridization and reassociation kinetic analysis.     

Mechanism of DNA denaturation in the presence of manganese ions

The unwinding of DNA strands in the presence of small concentrations of Mn²⁺ ions (2 × 10^?4?4 × 10^?4M) has been studied. The process of unwinding is nonequilibrium; the DNA strands are gradually unwound at a constant temperature corresponding to the beginning of the melting curve. There is no true renaturation in the partially melted DNA. It is shown in the paper that these effects are due to the aggregation of the unwound DNA regions. The Mn²⁺ ions are responsible for the binding of the unwound strands. The aggregation precludes renaturation, shifts the equilibrium towards the melted state, and causes slow unwinding at a constant temperature. The binding of denaturated regions seems to occur through the guanines.     

Processing of bacteriophage lambda DNA during its assembly into heads

DNA purified from bacteriophage λ added to a cell-free extract derived from induced λ lysogens can be packaged into infectious phage particles (Kaiser & Masuda, 1973). In this paper the structure of the DNA which is the substrate for in vitro packaging and head assembly is described. The active precursor is a multichromosomal polymer that contains covalently closed cohesive end sites. Neither circular or linear DNA monomers nor polymers with unsealed cohesive ends are packaged efficiently into heads. The unit length monomer is packaged when it is either contained in the interior of a polymer (both of its ends are in cos sites) or when it has a free left end and a cos site on its right. The monomer unit with a free right end is not a substrate for packaging.A procedure is given for the purification of λ DNA fragments that contain either the left or the right cohesive end. The fragments are produced by digesting λ DNA with the site-specific Escherichia coli R₁ endonuclease; the left and right ends are separated by sedimentation through a sucrose gradient. These fragments are used to construct small polymers that have a unit length λ monomer with (1) a free left end and a closed right end, (2) a free right end and a closed left end, or (3) both ends closed in cos sites.     

Biochemical evidence for Ku-independent backup pathways of NHEJ

Cells of higher eukaryotes process within minutes double strand breaks (DSBs) in their genome using a non-homologous end joining (NHEJ) apparatus that engages DNA-PKcs, Ku, DNA ligase IV, XRCC4 and other as of yet unidentified factors. Although chemical inhibition, or mutation, in any of these factors delays processing, cells ultimately remove the majority of DNA DSBs using an alternative pathway operating with an order of magnitude slower kinetics. This alternative pathway is active in mutants deficient in genes of the RAD52 epistasis group and frequently joins incorrect ends. We proposed, therefore, that it reflects an alternative form of NHEJ that operates as a backup (B-NHEJ) to the DNA-PK-dependent (D-NHEJ) pathway, rather than homology directed repair of DSBs. The present study investigates the role of Ku in the coordination of these pathways using as a model end joining of restriction endonuclease linearized plasmid DNA in whole cell extracts. Efficient, error-free, end joining observed in such in vitro reactions is strongly inhibited by anti-Ku antibodies. The inhibition requires DNA-PKcs, despite the fact that Ku efficiently binds DNA ends in the presence of antibodies, or in the absence of DNA-PKcs. Strong inhibition of DNA end joining is also mediated by wortmannin, an inhibitor of DNA-PKcs, in the presence but not in the absence of Ku, and this inhibition can be rescued by pre-incubating the reaction with double stranded oligonucleotides. The results are compatible with a role of Ku in directing end joining to a DNA-PK dependent pathway, mediated by efficient end binding and productive interactions with DNA-PKcs. On the other hand, efficient end joining is observed in extracts of cells lacking DNA-PKcs, as well as in Ku-depleted extracts in line with the operation of alternative pathways. Extracts depleted of Ku and DNA-PKcs rejoin blunt ends, as well as homologous ends with 3′ or 5′ protruding single strands with similar efficiency, but addition of Ku suppresses joining of blunt ends and homologous ends with 3′ overhangs. We propose that the affinity of Ku for DNA ends, particularly when cooperating with DNA-PKcs, suppresses B-NHEJ by quickly and efficiently binding DNA ends and directing them to D-NHEJ for rapid joining. A chromatin-based model of DNA DSB rejoining accommodating biochemical and genetic results is presented and deviations between in vitro and in vivo results discussed.     

Differences between Ca2+ and Mg2+ in DNA binding and release by the SfiI restriction endonuclease: implications for DNA looping

Many enzymes acting on DNA require Mg²⁺ ions not only for catalysis but also to bind DNA. Binding studies often employ Ca²⁺ as a substitute for Mg²⁺, to promote DNA binding whilst disallowing catalysis. The SfiI endonuclease requires divalent metal ions to bind DNA but, in contrast to many systems where Ca²⁺ mimics Mg²⁺, Ca²⁺ causes SfiI to bind DNA almost irreversibly. Equilibrium binding by wild-type SfiI cannot be conducted with Mg²⁺ present as the DNA is cleaved so, to study the effect of Mg²⁺ on DNA binding, two catalytically-inactive mutants were constructed. The mutants bound DNA in the presence of either Ca²⁺ or Mg²⁺ but, unlike wild-type SfiI with Ca²⁺, the binding was reversible. With both mutants, dissociation was slow with Ca²⁺ but was in one case much faster with Mg²⁺. Hence, Ca²⁺ can affect DNA binding differently from Mg²⁺. Moreover, SfiI is an archetypal system for DNA looping; on DNA with two recognition sites, it binds to both sites and loops out the intervening DNA. While the dynamics of looping cannot be measured with wild-type SfiI and Ca²⁺, it becomes accessible with the mutant and Mg²⁺.     

The 3'-terminal nucleotide sequences of lambda DNA

The base sequences of the 3′-termini of coliphage λ DNA have been analyzed by a new technique. Escherichia coli DNA polymerase I was used to add a single radioactive nucleotide to the 3′-OH terminus of one of the DNA strands. The DNA was then digested with pancreatic DNase I, and the resulting oligonucleotides were separated by two dimensional ionophoresis. Terminal oligonucleotides were identified by the presence of the radioactive label, and the base sequence of the labelled terminus was deduced from the base compositions of the terminal di-, tri-, tetra-, etc., oligonucleotides. It is found that the left 3′-terminus of λ DNA ends with the sequence d(pCpGpCpG) and the right 3′-terminus ends with the sequence d(pCpG).     

Tests of spool models for DNA packaging in phage lambda

Experiments are reported which bear on two spool models proposed for packaging the DNA of phage lambda. Both spool models fill an assumed spherical cavity with DNA wrapped in cylindrical or quasi-cylindrical layers composed of adjacent circular turns. In the curved-spool model, a single continuous segment of DNA, about 20% of the DNA length and probably located near the left end of the DNA, is in contact with the coat protein of the phage capsid. In the straight spool model, there are several DNA segments in contact with the capsid; they are concentrated in one half (probably the left half) of lambda DNA. We have identified the loci on the DNA which are in contact with the capsid by chemical crosslinking, induced by ultraviolet-irradiation of phage containing 5-bromodeoxyuridine in place of thymine.In an electron microscope experiment, phage are first lysed with EDTA, and then spread in a cytochrome c film by the formamide method. The disrupted capsid, which has the appearance of a phage ghost, serves as a marker showing where the DNA is crosslinked to the coat. The left end of the DNA is not distinguished from the right end, and so the map of DNA-capsid contacts is folded over on itself. Contacts are found nearly randomly over the entire map.In a second experiment, DNA from lysed, crosslinked phage is cut either with EcoRI or HindIII restriction endonucleases and the cut restriction fragments are labeled at their ends with ³²P. Density centrifugation in a CsCl gradient separates free DNA from restriction fragments crosslinked to protein. After digestion with proteinase k, the DNA fragments previously crosslinked to protein are identified by size after agarose gel electrophoresis. DNA fragments from all parts of the genome are found.These two experiments show that, if the DNA of each phage is packaged identically, then the curved-spool model is ruled out and the straight spool model is unlikely. Alternatively, the manner of packaging the DNA may vary from one phage to the next. These results agree with other recent experiments on λ DNA packaging by Hall & Schellman (1982a,b), and by Haas et al. (1982).A different experiment is also reported. The psoralen derivative aminomethyltrioxalen (AMT) is allowed to intercalate into λ phage and then the DNA strands are crosslinked by ultraviolet-irradiation after the rapid phase of AMT intercalation is complete. The DNA is subsequently denatured by glyoxal modification and spread for electron microscopy in a cytochrome c film by the formamide method. Sites of AMT crosslinking appear duplex; uncrosslinked regions appear as single-stranded loops. AMT is found to intercalate throughout the λ DNA. Patterns of reacted sites appear different from one DNA molecule to the next, and no consistent pattern can be found. More extensive intercalation occurs with the deletion mutant λb221 than with phage of wild-type DNA length, and free DNA shows much more reaction than the DNA inside either phage type. In order for intercalation to occur, the DNA helix must unwind and become further extended. This experiment shows that regions throughout the entire DNA molecule can unwind and be extended by intercalation, which is not confined to a single DNA segment or to segments in one half of the DNA molecule, as would be expected for the two spool models if only the DNA in contact with the capsid were accessible to the dye.     

Initiation of recA+-dependent recombination in Escherichia coli (lambda). I. Undamaged covalent circular lambda DNA molecules in uvrA+ recA+ lysogenic host cells are cut following superinfection with psoralen-damaged lambda phages.

When λ bacteriophages were treated with a photosensitizing agent, psoralen or khellin, and 360 nm light, monoadducts and interstrand crosslinks were produced in the phage DNA. The DNA from the treated phages was injected normally into Escherichia coli uvrA^? (λ) cells and it was converted to the covalent circular form in yields similar to those obtained in experiments with undamaged λ phages. In excision-proficient host cells, however, there was a dose-dependent reduction in the yield of rapidly sedimenting molecules, and a corresponding increase in slow sedimenting material, the extent of this conversion corresponding to about one cut per two crosslinks. Presumably, the damaged λ DNA molecules were cut by the uvrA endonuclease of the host cell, but were not restored to the original covalent circular form.The presence of psoralen damage in λ phage DNA greatly increased the frequency of genetic exchanges in λ phage-prophage crosses in homoimmune lysogens (Lin et al., 1977). As genetic recombination is thought to depend on cutting and joining in DNA molecules, experiments were performed to test whether psoralen-damaged λ DNA would cause other λ DNA in the same cell to be cut. E. coli (λ) host cells were infected with ³²P-labeled λ phages and incubated to permit the labeled DNA to form covalent circles. When these host cells were superinfected with untreated λ phages, there was no effect upon the circular DNA. When superinfected with λ phages that had been treated with psoralen and light, however, many of the covalent circular molecules were cut. The cutting of undamaged molecules in response to the damaged DNA was referred to as “cutting in trans”. It required the uvrA⁺ and recA⁺ host gene functions, but neither recB⁺ nor any phage gene functions. It occurred normally in non-lysogenic hosts treated with chloramphenicol before infection. Cutting in trans may be one of the steps in recA-controlled recombination between psoralen crosslinked phage λ DNA and its homologs.     

Packaging of coliphage lambda DNA. I. The role of the cohesive end site and the gene A protein

The cohesive ends of the DNA of bacteriophage λ particles are normally formed by the action of a nuclease on the cohesive end sites (cos) of concatemeric λ DNA (reviewed by Hohn et al., 1977). The nuclease also cuts the cos site of an integrated prophage, and DNA located to the right is preferentially packaged into phage particles. This process occurs with approximately the same efficiency and rate in a single lysogen as in a tandem polylysogen. Thus, the rate of cos cutting does not increase when the number of cos sites per molecule increases, an hypothesis that has been proposed to explain why cohesive ends are not formed in circular monomers of λ DNA. We propose instead that the interaction of Ter with cos is influenced by the configuration of the DNA outside of cos during packaging, and that this configuration is different for circular monomers than for other forms of λ DNA. A model that gives rise to such a difference is described.We also found that missense mutations in the λ A gene changed the efficiency of packaging of phage relative to host DNA. This was not the case for missense mutations in several phage genes required for capsid formation. Thus, the product of gene A plays a role in determining packaging specificity, as expected if it is or is part of the nuclease that cuts λ DNA at cos.     

Role of metal ions in catalysis by HIV integrase analyzed using a quantitative PCR disintegration assay

Paired metal ions have been proposed to be central to the catalytic mechanisms of RNase H nucleases, bacterial transposases, Holliday junction resolvases, retroviral integrases and many other enzymes. Here we present a sensitive assay for DNA transesterification in which catalysis by human immunodeficiency virus-type 1 (HIV-1) integrase (IN) connects two DNA strands (disintegration reaction), allowing detection using quantitative PCR (qPCR). We present evidence suggesting that the three acidic residues of the IN active site function through metal binding using metal rescue. In this method, the catalytic acidic residues were each substituted with cysteines. Mn²⁺ binds tightly to the sulfur atoms of the cysteine residues, but Mg²⁺ does not. We found that Mn²⁺, but not Mg²⁺, could rescue catalysis of each cysteine-substituted enzyme, providing evidence for functionally important metal binding by all three residues. We also used the PCR-boosted assay to show that HIV-1 IN could carry out transesterification reactions involving DNA 5′ hydroxyl groups as well as 3′ hydroxyls as nucleophiles. Lastly, we show that Mn²⁺ by itself (i.e. without enzyme) can catalyze formation of a low level of PCR-amplifiable product under extreme conditions, allowing us to estimate the rate enhancement due to the IN-protein scaffold as at least 60 million-fold.     

Involement of supertwisted DNA in integrative recombination of bacteriophage lambda.

Negatively supertwisted closed circular DNA is the primary substrate for integrative recombination of phage λ DNA in vitro. Closed circular λ DNA without supertwists must be converted to the supertwisted form by the action of Escherichia coli DNA gyrase before efficient recombination can occur. When negatively supertwisted substrate is provided, E. coli DNA gyrase and its cofactors are dispensable components of recombination reaction mixtures. In the absence of DNA gyrase activity, circular DNA considerably less negatively twisted than naturally occurring supercoils is an effective substrate, but positively supertwisted DNA appears to be an ineffective substrate.The predominant products of integrative recombination in vitro are covalently closed circles. The closure of the recombined sites appears to occur without appreciable DNA synthesis and without the action of E. coli DNA ligase. No detectable difference can be observed between the degree of supertwisting of product DNA and that of unrecombined DNA. These facts suggest that the resealing of broken DNA strands is an integral part of the recombination reaction mechanism and is closely coupled with the breakage and realignment steps of recombination.     

A sequence specific endonuclease from Micrococcus radiodurans

A new sequence specific endonuclease, MraI has been purified from Micrococcus radiodurans. This enzyme cleaves bacteriophage λ DNA at three sites, adenovirus type 2 DNA at more than 12 sites and has a unique site on ΦX174 DNA. It has no sites on SV40, PM2 and pBR322 DNA. The three sites on phage λ DNA are different from those cleaved by SmaI, XmaI and XorII. The sites of cleavage are located at 0.424, 0.447 and 0.834 fractional lengths on the physical map of λ DNA. MraI is shown to be an isoschizomer of SacII and SstII recognizing the palindromic nucleotide sequence ′5-CCGC↓GG-3′. The enzyme shows an absolute requirement of Mg²⁺, but is active in the absence of added 2-mercaptoethanol. The enzyme shows activity at a broad range of temperature and pH with an optimum at 45°C and pH 7.0. MraI represents the first restriction enzyme from a bacterium whose DNA lacks modified methylated bases.     

Mechanism of degradation of duplex DNA by the DNase induced by herpes simplex virus

Reaction intermediates formed during the degradation of linear PM2, T5, and λ DNA by herpes simplex virus (HSV) DNase have been examined by agarose gel electrophoresis. Digestion of T5 DNA by HSV type 2 (HSV-2) DNase in the presence of Mn²⁺ (endonuclease only) gave rise to 6 major and 12 minor fragments. Some of the fragments produced correspond to those observed after cleavage of T5 DNA by the single-strand-specific S1 nuclease, indicating that the HSV DNase rapidly cleaves opposite a nick or gap in a duplex DNA molecule. In contrast, HSV DNase did not produce distinct fragments upon digestion of linear PM2 or λ DNA, which do not contain nicks. In the presence of Mg²⁺, when both endonuclease and exonuclease activities of the HSV DNase occur, most of the same distinct fragments from digestion of T5 DNA were observed. However, these fragments were then further degraded preferentially from the ends, presumably by the action of the exonuclease activity. Unit-length λ DNA, EcoRI restriction fragments of λ DNA, and linear PM2 DNA were also degraded from the ends by HSV DNase in the same manner. Previous studies have suggested that the HSV exonuclease degrades in the 3′ → 5′ direction. If this is correct, and since only 5′-monophosphate nucleosides are produced, then HSV DNase should “activate” DNA for DNA polymerase. However, unlike pancreatic DNase I, neither HSV-1 nor HSV-2 DNase, in the presence of Mg²⁺ or Mn²⁺, activated calf thymus DNA for HSV DNA polymerase. This suggests that HSV DNase degrades both strands of a linear double-stranded DNA molecule from the same end at about the same rate. That is, HSV DNase is apparently capable of degrading DNA strands in the 3′ → 5′ direction as well as in the 5′ → 3′ direction, yielding progressively smaller double-stranded molecules with flush ends. Except with minor differences, HSV-1 and HSV-2 DNases act in a similar manner.