The role of bacteriophage T7 exonuclease (gene 6) in genetic recombination and production of concatemers

Genetic analysis reported here shows that bacteriophage T7 exonuclease (gene 6) is necessary for intragenic and intergenic recombination in several areas of the T7 genetic map. This supports our previous conclusion (Lee & Miller, 1974) that the enzyme is necessary for T7 molecular recombination.Results of sucrose gradient analysis show that DNA concatemers are formed when both the T7 exonuclease (gene 6) and the T7 endonuclease (gene 3) are absent. Further results show that concatemers cannot be maintained in the absence of the exonuclease unless the endonuclease is also eliminated. Therefore, concatemers are formed by a process other than normal phage recombination. Selective defects in the recombination system do interfere with the stability of concatemers, however.     

Molecular Recombination in T4 Bacteriophage Deoxyribonucleic Acid I. Tertiary Structure of Early Replicative and Recombining Deoxyribonucleic Acid

A replicative hybrid resulting from the infection of heavy (substituted with 5-bromodeoxyuridine) bacteria with light (not substituted with 5-bromodeoxyuridine) radioactive bacteriophage was isolated from a CsCl density gradient. Sedimentation studies indicate that 60% of the deoxyribonucleic acid (DNA) behaves as if it were in units more than four times as large as an intact reference molecule. Under the electron microscope, hybrid molecules appeared tangled, showed puffs and loops, occupied a small area, and often had a total length twice that of mature phage. This indicates that sucrose gradient sedimentation is not applicable as a method for estimating the relative molecular size of replicative forms of DNA. After denaturation, the separated strands of hybrid were of the same size as those of reference DNA. CsCl density gradient analysis revealed no terminal covalent addition of new material to the old parental strand. The possibility of a continuous growth of the DNA molecule, either on a single-stranded level or as a double helical structure, is disproved. When chloramphenicol (CM) was added at critical times after infection, DNA synthesis continued at a constant rate. The parental label soon assumed and retained a hybrid density, despite concomitant synthesis of DNA, throughout the rest of the period of incubation in CM. The hybrid moiety, however, actively participated in replication and exchanged its partner strand for a new one; this was demonstrated by changing the density label during incubation in CM. A new enzyme synthesized shortly after infection introduced single-stranded "nicks" into the parental DNA. Since nicking can be inhibited by chloramphenicol, the responsible enzyme is not of host origin. The time of the appearance of this enzyme coincided with the onset of molecular recombination. Another enzyme, which mediates the repair of the continuity of the polynucleotide chain after recombination, appeared after recombination. If selectively inhibited by chloramphenicol, recombinant molecules remained unrepaired, and, upon denaturation, the parental fragment was liberated in pure form.     

RNA-linked nascent DNA pieces in T7 phage-infected Escherichia coli cells. I. Role of gene 6 exonuclease in removal of the linked RNA.

Summary The presence of RNA-linked nascent DNA pieces in T7 phage-infectedEscherichia coli cells has been shown by the selective degradation of the 5-hydroxyl-terminated nascent DNA, produced by alkali or RNase treatment, with spleen exonuclease. At 43°C, the proportion of RNA-linked DNA pieces in nascent short DNA is 50 to 60% in T7ts136 (ts mutant of gene 6) phage-infectedE. coli, whereas that in T7 wild-type phage-infected cells is less than 6%. Joining of the nascent pieces is greatly retarded in T7ts136-infectedE. coli temperature sensitivepolA mutants at 43° C. These results suggest that gene 6 exonuclease plays a role in removal of the linked RNA during the discontinuous replication of T7 DNA.     

Multiple and Specific Initiation of T4 DNA Replication

Partially replicated T4 DNA molecules (PRM) whose parental or progeny DNA was labeled with bromodeoxyuridine BUdR was analyzed by gradual shearing followed by CsCl banding of the sheared product. Analysis of PRM containing 18-mum replicated DNA showed that each replicated region was 3- to 6-mum long, indicating three to 6 replicative sites per molecule. Analysis of PRM containing 9-mum replicated DNA similarly indicated two to three replicated regions per molecule. DNA from the replicated regions of PRM containing 10-mum replicated DNA ("donor") was hybridized to DNA from mature phage ("recipient"), and the resulting hybrid was subjected to digestion with exonuclease I. The extent of protection of the recipient and more efficient self-annealing of progeny fragments from PRM indicated that the replicated regions represented 8 to 10 nonrandom locations of the genome. Possible significance of multiple sites for initiation of DNA replication is discussed.     

Purification of bacteriophage T7 DNA-membrane complex and its application to the in vitro recombination reaction

In order to construct an in vitro recombination system of T7 DNA, the reaction products of which resemble those in vivo in structure, T7 DNA-membrane complex which is free from concomitant DNase activity was purified from T7 phage-infected cells. T7-infected cells were lysed with T4 lysozyme/Brij58, and T7 DNA-membrane complex was purified through three successive density gradient centrifugations. The properties of the complex on exposure to defined nucleases and observation of the complex by electron microscopy revealed that in T7 DNA-membrane complex, both ends of a linear T7 DNA are bound with membrane components. A mixture of 32P-labeled T7 DNA-membrane complex and BU-labeled T7 DNA-membrane complex was incubated with T7 exonuclease and T7 DNA-binding protein, and the reaction products with intermediate density were purified. Most of the products were found to have structures similar to that of the recombination intermediate found in T7-infected cells upon electron microscopic examination.     

Association of Replicative T4 Deoxyribonucleic Acid and Bacterial Membranes

Experiments utilizing CsCl density gradient analysis and radioactive labels specific for bacteriophage T4 deoxyribonucleic acid (DNA) and membranes have shown that replicative T4 DNA is associated with host membranes. The association is inhibited by chloramphenicol and takes place just prior to semi-conservative replication of the phage DNA.     

Role of gene 6 exonuclease in the replication and packaging of bacteriophage T7 DNA

When bacteriophage T7 gene 6 exonuclease is genetically removed from T7-infected cells, degradation of intracellular T7 DNA is observed. By use of rate zonal centrifugation, followed by either pulsed-field agarose gel electrophoresis or restriction endonuclease analysis, in the present study, the following observations were made. (1) Most degradation of intracellular DNA requires the presence of T7 gene 3 endonuclease and is independent of DNA packaging; rapidly sedimenting, branched DNA accumulates when both the gene 3 and gene 6 products are absent. (2) A comparatively small amount of degradation requires packaging and occurs at both the joint between genomes in a concatemer and near the left end of intracellular DNA; DNA packaging is only partially blocked and end-to-end joining of genomes is not blocked in the absence of gene 6 exonuclease. (3) Fragments produced in the absence of gene 6 exonuclease are linear and do not further degrade; precursors of the fragments are non-linear. (4) Some, but not most, of the cleavages that produce these fragments occur selectively near two known origins of DNA replication. On the basis of these observations, the conclusion is drawn that most degradation that occurs in the absence of T7 gene 6 exonuclease is caused by cleavage at branches. The following hypothesis is presented: most, possibly all, of the extra branching induced by removal of gene 6 exonuclease is caused by strand displacement DNA synthesis at the site of RNA primers of DNA synthesis; the RNA primers, produced by multiple initiations of DNA replication, are removed by the RNase H activity of gene 6 exonuclease during a wild-type T7 infection. Observation of joining of genomes in the absence of gene 6 exonuclease and additional observations indicate that single-stranded terminal repeats required for concatamerization are produced by DNA replication. The observed selective shortening of the left end indicates that gene 6 exonuclease is required for formation of most, possibly all, mature left ends.     

T7 gene 6 exonuclease has an RNase H activity.

T7 gene 6 exonuclease has been shown to have an RNase H activity as well as a double-strand specific DNase activity by the following experiments: The RNase H activity coelutes with the DNase activity from DEAE-cellulose, phosphocellulose, hydroxyapatite, and Sephadex G-200 columns. Gene 6 exonuclease specified by a T7 strain with a temperature sensitive mutation in gene 6 has an extremely heat-labile RNase H activity as well as a heat-labile DNase activity. T7 gene 6 exonuclease degrades the RNA region of a poly(A) . poly(dT) hybrid polymer exonucleolytically from the 5' terminus, releasing a ribonucleoside 5'-monophosphate product. When the RNA strand of a 0X174 RNA . DNA hybrid molecule synthesized with E. coli RNA polymerase is degraded, a ribonucleoside triphosphate is produced from the 5'-triphosphate terminus. Participation of T7 gene 6 exonuclease in the removal of primer RNA in discontinuous replication of T7 DNA is discussed.     

T7-induced DNA polymerase. Characterization of associated exonuclease activities and resolution into biologically active subunits.

Bacteriophage T7-induced DNA polymerase has been isolated by a procedure suitable for large scale use and which yields near homogeneous enzyme. In addition to previously described DNA polymerase activity and 3' to 5' exonucleolytic activity on single stranded DNA (Grippo, P., and Richardson, C. C. (1971) J. Biol. Chem. 246, 6867-6873), the enzyme also possesses a highly active exonuclease which hydrolyzes duplex substrates with 3' to 5' directionality. The native polymerase has been dissociated using 6 M guanidine HCl and resolved into biologically active subunits: T7 gene 5 protein and Escherichia coli thioredoxin. The phage-specified subunit obtained by this procedure is deficient in DNA polymerase and double strand exonuclease activities, with deficiencies in these activities being apparent at the level of a single turnover. However, it possesses near normal levels of a single strand hydrolytic activity which is identical to that associated with the native polymerase with respect to substrate specificity and suppression of hydrolysis by low levels of deoxyribonucleoside 5'-triphosphates. Thioredoxin forms a molecular complex with the T7 gene 5 protein, and addition of the host protein restores restores DNA polymerase and double strand exonuclease activities to near normal levels.     

In vitro packaging of bacteriophate T7 DNA synthesized in vitro.

An in vitro DNA packaging system was used to encapsulate T7 DNA that had been synthesized by extracts prepared from gently lysed Escherchia coli infected with bacteriophage T7 carrying amber mutations in gene 3 or in both genes 3 and 6. Isopycnic centrifugation of density-labeled wild-type DNA was employed in an effort to separate product from template; suppressor-free indicator bacteria were used to eliminate contributions from endogenous DNA or contaminating phage. Additional controls indicated that fragmented DNA is packaged in vitro only with very low efficiency and that the frequency of recombination during packaging is too low to affect interpretation of these experiments. T7 DNA replicated by extracts prepared using T7 mutants deficient in both genes 3 and 6 could be packaged in vitro with an efficiency comparable to that found when highly purified virion T7 DNA was used. When T7 deficient in the gene 3 endonuclease but with normal levels of the gene 6 exonuclease was used, fast-sedimentingconcatemer-like DNA structures were formed during in vitro DNA synthesis. Electron microscopy revealed many branched and highly complex DNA structures formed during this reaction. This concatemer-like DNA was encapsulated in vitro with an efficiency significantly greater than that found for DNA the length of a single T7 genome.     

T4 DNA polymerase (gene 43) is required in vivo for repair of gaps in recombinants.

Experiments with a mutant of T4, tsL97, temperature sensitive for gene 43, showed that T4 DNA polymerase was necessary in vivo to repair gaps in recombinant molecules. CsCl density gradient experiments showed that molecular recombinants were not repaired when the T4tsL97-infected cells were shifted to 42 C after replication and recombination had taken place. Repair was almost complete when the same procedure was followed with the wild-type T4, or when the T4tsL97-infected cells were incubated at the permissive temperature, 36 C. Long-single-strand production was also affected similarly by the T4tsL97 mutation. All the results were consistent with the theory that gaps exist in many recombinant molecules at the recombinant joint, that T4 DNA polymerase is the enzyme that repairs these gaps in vivo, and that covalent repair of the recombinants leads to extensive long-single-strand production.     

Parent-to-Progeny Transfer and Recombination of T4rII Bacteriophage

Transfer of parental, light (not substituted with 5-bromodeoxyuridine) (32)P-deoxyribonucleic acid (DNA) from rII(-) mutants of T4 bacteriophage to heavy (5-bromodeoxyuridine-substituted) progeny in Escherichia coli B was less homogeneous than in wild phages. The net transfer was 5 to 20% of the value for wild T4 phage, and the parental contribution per progeny DNA molecule amounted to 7 to 100% of the genome. Three classes could be distinguished, based on the density distribution of parental label in CsCl analysis of the progeny phages. "Far recombined" phages contain parental material only in semiconservatively replicated subunits covalently attached to progeny DNA, amounting to 5 to 10% parental contribution per genome. "Intermediate recombinants" contain, aside from conventional recombinant DNA, parental DNA banding at the original, light density. This DNA may be unattached to heavy progeny DNA or attached by weak bonds which are very sensitive to shearing during the extraction procedure. The parental contribution is 10 to 50% per progeny DNA molecule in this class. "Conservative" phages band close to the parental, light density in CsCl; their DNA is purely light. When the parental phage is labeled with both (3)H-leucine (capsid) and (32)P (DNA), the specific activity of (3)H/(32)P in the "conservative progeny" is 10 to 40% of that in the parental, showing that at least some of the (32)P in this area belongs to phages with parental DNA as the sole DNA component inside an unlabeled capsid, i.e., parental DNA which has been injected into the host and matured in a new capsid without replication or recombination. This phenomenon occurs to about the same extent in both single and multiple infection.     

Genetic recombination of bacteriophage T7 in vivo studied by use of a simple physical assay.

A new physical method was developed to assay genetic recombination of phage T7 in vivo. The assay utilized T7 mutants that carry unique restriction sites and was based on the detection of a new restriction fragment generated by recombination. Using this assay, we reexamined the genetic requirements for recombination of T7 DNA. Our results were in total agreement with previous findings in that recombination required the products of genes 3 (endonuclease), 4 (primase), 5 (DNA polymerase), and 6 (exonuclease). Recombination was found to be independent of DNA ligase and DNA packaging and maturation functions.     

On the Direction of Reading of Bacteriophage T4 Gene 43 (Deoxyribonucleic Acid Polymerase)

Amber (am) mutants of the two closely linked sites, B22 and C125, in bacteriophage T4 gene 43 [deoxyribonucleic acid (DNA) polymerase] synthesize in the nonpermissive (su(-)) Escherichia coli host gene 43 products which are devoid of DNA polymerase activity, but which retain a 3'-exonuclease activity. Diethylaminoethyl-cellulose chromatographic analysis of DNA polymerase and deoxyribonuclease activities from extracts of su(-) cells infected with single- and double-am mutants of T4 gene 43 showed that the exonuclease activity which is observed with amB22 is not seen with double mutants carrying, in addition to amB22, am mutations which map to the clockwise side of the B22 site on the circular genetic map of T4. Similarly, am mutations which map to the clockwise side of the C125 site abolish the exonuclease activity which is observed with an am mutant (amE4335) of this site. It was concluded that in these double mutants termination signals to the clockwise side of amB22 and amE4335 are encountered before the amB22 and amE4335 signals during translation of the messenger ribonucleic acid from T4 gene 43. Thus, it seems that the T4 DNA polymerase is synthesized in vivo in a direction which corresponds to a counterclockwise reading of gene 43.     

A T7 amber mutant defective in DNA-Binding protein

Summary A T7 amber mutant, UP-2, in the gene for T7 DNA-binding protein was isolated from mutants that could not grow on sup ⁺ ssb-1 bacteria but could grow on glnU ssb-1 and sup ⁺ ssb ⁺bacteria. The mutant phage synthesized a smaller amber polypeptide (28,000 daltons) than T7 wild-type DNA-dinding protein (32,000 daltons). DNA synthesis of the UP-2 mutant in sup ⁺ ssb-1 cells was severely inhibited and the first round of replication was found to be repressed. The abilities for genetic recombination and DNA repair were also low even in permissive hosts compared with those of wild-type phage. Moreover, recombination intermediate T7 DNA molecules were not formed in UP-2 infected nonpermissive cells. The gene that codes for DNA-binding protein is referred to as gene 2.5 since the mutation was mapped between gene 2 and gene 3.     

Visualization of repair of double-strand breaks in the bacteriophage T7 genome without normal DNA replication

An in vitro system based on extracts of Escherichia coli infected with bacteriophage T7 is able to repair double-strand breaks in a T7 genome with efficiencies of 20% or more. To achieve this high repair efficiency it is necessary that the reaction mixtures contain molecules of donor DNA that bracket the double-strand break. Gaps as long as 1,600 nucleotides are repaired almost as efficiently as simple double-strand breaks. DNA synthesis was measured while repair was taking place. It was found that the amount of DNA synthesis associated with repair of a double-strand break was below the level of detection possible with this system. Furthermore, repair efficiencies were the same with or without normal levels of T7 DNA polymerase. However, the repair required the 5'-->3' exonuclease encoded by T7 gene 6. The high efficiency of DNA repair allowed visualization of the repaired product after in vitro repair, thereby assuring that the repair took place in vitro rather than during an in vivo growth step after packaging.     

Replicative Intermediates of Bacteriophage T7 Deoxyribonucleic Acid

After infection with bacteriophage T7, parental and newly synthesized deoxyribonucleic acid (DNA) exhibit an extremely fast sedimentation rate in neutral sucrose gradients. This fast-sedimenting component (intermediate I) has a sedimentation constant of about 1,500S and contains T7 DNA as determined by DNA-DNA hybridization experiments. Pulse-chase experiments indicate that the fast-sedimenting material is metabolically active and serves as a precursor to the formation of T7 DNA. Intermediate I contains about 2.5 to 7% of the total ³H-labeled protein formed between 3 and 9.5 min after T7 infection. Treatment of intermediate I with Pronase results in the release of the DNA from the complex. At early times after infection, a second intermediate (intermediate II) can be detected which contains both parental and newly synthesized DNA sedimenting slower than intermediate I but 2 to 3 times as fast as mature T7 DNA. Intermediates I and II containing parental DNA are formed after infection of the nonpermissive host with an amber mutant in gene 1, a gene whose expression is necessary for the synthesis of most T7 proteins. The two intermediates are also observed when infection with T7 wild type is carried out in the presence of chloramphenicol.     

The essential role of bacteriophage T7 endonuclease (gene 3) in molecular recombination

Density transfer and shearing experiments show that the bacteriophage T7 endonuclease (gene 3) is necessary for the dispersion of parental DNA in the newly replicated DNA. These experiments on parental to progeny recombination support previous genetic data (Powling & Knippers, 1974; Kerr & Sadowski, 1975) that the gene 3 protein is essential for T7 recombination. Concatemers containing the newly replicated DNA have been sheared to the size of mature phage DNA and also to quarter molecules. In the presence of gene 3 protein, parental DNA and newly replicated DNA are interspersed. In the absence of gene 3 protein, the parental strand of each sheared DNA molecule is usually found intact.     

Specific stimulation of the T7 gene 6 exonuclease by the phage T7 coded deoxyribonucleic acid binding protein

Bacteriophage T7 codes for a single-stranded DNA binding protein. This protein is the product of gene 2.5 and has been found previously to stimulate specifically the activity of the phage-coded DNA polymerase. We report here that the T7 DNA binding protein also stimulates the activity of the phage-coded exonuclease. The gene 6 exonuclease is a double-stranded DNA specific 5'-exonuclease that has been implicated in destruction of bacterial DNA, removal of RNA primers during DNA replication, genetic recombination, and DNA maturation. The enzyme is markedly inhibited by physiological concentrations of NaCl. This inhibition, which is due to a marked reduction in the Vmax of the enzyme, can be largely overcome by the phage-coded DNA binding protein. This stimulation is specific since the Escherichia coli DNA binding protein is without effect. The stimulation by the binding protein is apparently not due to its coating of the 3' single-stranded tails generated during the digestion. Kinetic studies show that the stimulation is due to a combined effect on both the Km and Vmax of the exonuclease. These studies are consistent with a loose binding of the binding protein to either the DNA or the exonuclease.