A T4 DNA fragment containing genes for the baseplate central plug: Endonuclease restriction, gene expression and cell wall changes

Summary A fragment of Escherichia coli bacteriophage T4D DNA, containing 6.1 Kbp which included the six genes (genes 25, 26, 51, 27, 28 and 29) coding for the tail baseplate central plug has been partially characterized. This DNA fragment was obtained originally by Wilson et al. (1977) by the action of the restriction enzyme EcoRI on a modified form of T4 DNA and was inserted in the pBR322 plasmid and then incorporated into an E. coli K12 strain called RRI. This plasmid containing the phage DNA fragment has now been reisolated and screened for cleavage sites for various restriction endonucleases. Restriction enzymes Bgl 11 and Xbal each attacked one restriction site and the enzyme Hpa 1 attacked two restriction sites on this fragment. The combined digestion of the hybrid plasmid containing the T4 EcoRI DNA fragment conjugated to the pBR322 plasmid with one of these enzymes plus Bam H1 restriction enzyme resulted in the localization of the restriction site for Bgl 11, Xba 1 and Hpa 1. Escherichia coli strain B cells were transformed with this hybrid plasmid and found to have some unexpected properties. E. coli B cells, which are normally restrictive for T4 amber mutants and for T4 temperature sensitive mutants (at 44°) after transformation, were permissive for 25am, 26am and 26^Ts, 51am, and 51^Ts, 27^Ts, and 28^Ts T4 mutants. Extracts from the transformed E. coli cells were found in complementation experiments to contain the gene 29 product, as well as the gene 26 product, the gene 51 product, and the gene 27 product. The complementation experiments and the permissiveness of the transformed E. coli B cells to the various conditional lethal mutants clearly showed that the six T4 genes were producing all six gene products in these transformed cells. However, these cells were not permissive for T4 amber mutants in genes 27, 28, and 29. The transformed E. coli B cells, as compared to untransformed cells, were found to have altered outer cell walls which made them highly labile to osmotic shock and to an increased rate of killing by wild type T4 and all T4 amber mutants except for T4 am29. The change in cell walls of the transformed cells has been found to be due to the T4 baseplate genes on the hybrid plasmid, since E. coli B transformed by the pBR322 plasmid alone does not show the increase in osmotic sensitivity.     

Various characteristics of the anti-restriction mechanism in bacteriophage T5

Bacteriophage T5 is not confined by the restriction systems of the second type EcoRII and EcoRV. Bacteriophage T5 DNA is not modified by EcoRII and EcoRV methylases in vivo. The sites of recognition for restriction endonuclease EcoRV are mapped at 24.4; 57.6; 68.5; 70.2% of T5 DNA, while the sites at 5.1; 7.6% are recognized by EcoRII, the sites at 5.75; 6.0 and 6.5% are recognized by HpaI in FST. A high activity of restriction endonucleases EcoRI and EcoRV is demonstrated in crude extracts of E. coli B834 (RI) and E. coli B834 (RV) cells infected by bacteriophage T5. The simultaneous infection of E. coli B834 (RI) or E. coli B834 (RV) cells by the amber mutants of bacteriophage T5 and the suppressing phage lambda NM761 does not result in the protection of lambda DNA by the T5 anti-restriction mechanism. The presented data support the hypothesis that the anti-restriction mechanism of bacteriophage T5 is based on prevention of T5 DNA contacts with restriction enzymes by a specific phage protein.     

Mutants in a Nonessential Gene of Bacteriophage T4 Which Are Defective in the Degradation of Escherichia coli Deoxyribonucleic Acid

Wild-type bacteriophage T4 was enriched for mutants which fail to degrade Escherichia coli deoxyribonucleic acid (DNA) by the following method. E. coli B was labeled in DNA at high specific activity with tritiated thymidine ((3)H-dT) and infected at low multiplicity with unmutagenized T4D. At 25 min after infection, the culture was lysed and stored. Wild-type T4 degrades the host DNA and incorporates the (3)H-dT into the DNA of progeny phage; mutants which fail to degrade the host DNA make unlabeled progeny phage. Wild-type progeny are eventually inactivated by tritium decay; mutants survive. Such mutants were found at a frequency of about 1% in the survivors. Eight mutants are in a single complementation group called denA located near gene 63. Four of these mutants which were examined in detail leave the bulk of the host DNA in large fragments. All eight mutants exhibit much less than normal T4 endonuclease II activity. The mutants produce somewhat fewer phage and less DNA than does wild-type T4.     

Effect of chloramphenicol and cyanide on the increase in UV resistance of intracellular bacteriophage T1.

The effects of chloramphenicol and cyanide on the increase in UV resistance of intracellular phage T1 infecting cells of E. coli B or E. coli Bs-1 were investigated. The inhibitiors were added to the cells 3 min prior to infection and to the complexes of phage-bacteria 3.5 and 6.5 min after adsorption of phage by the cells. The data obtained are not in agreement with the suggestion that increase in UV resistance of intracellular phage is mainly due to the accumulation of phage DNA inside the host cells. It is suggested that a very important role in this resistance is played by the interaction of phage DNA with the cell membranes.     

Bacteriophage T4 Inhibits Colicin E2-Induced Degradation of Escherichia coli Deoxyribonucleic Acid I. Protein Synthesis-Dependent Inhibition

The deoxyribonucleic acid (DNA) of Escherichia coli B is converted by colicin E2 to products soluble in cold trichloroacetic acid; we show that this DNA degradation (hereafter termed solubilization) is subject to inhibition by infection with bacteriophage T4. At least two modes of inhibition may be differentiated on the basis of their sensitivity to chloramphenicol. The following observations on the inhibition of E2 by phage T4 in the absence of chloramphenicol are described: (i) Simultaneous addition to E. coli B of E2 and a phage mutated in genes 42, 46, and 47 results in a virtually complete block of the DNA solubilization normally induced by E2; the mutation in gene 42 prevents phage DNA synthesis, and the mutations in genes 46 and 47 block a late stage of phage-induced solubilization of host DNA. (ii) This triple mutant inhibits equally well when added at any time during the E2-induced solubilization. (iii) Simultaneous addition to E. coli B of E2 and a phage mutated only in gene 42 results in extensive DNA solubilization, but the amount of residual acid-insoluble DNA (20 to 25%) is more characteristic of phage infection than of E2 addition (5% or less). (iv) denA mutants of phage T4 are blocked in an early stage (endonuclease II) of degradation of host DNA; when E2 and a phage mutated in both genes 42 and denA are added to E. coli B, extensive solubilization of DNA occurs with a pattern identical to that observed upon simultaneous addition of E2 and the gene 42 mutant. (v) However, delaying E2 addition for 10 min after infection by this double mutant allows the phage to develop considerable inhibition of E2. (vi) Adsorption of E2 to E. coli B is not impaired by infection with phage mutated in genes 42, 46, and 47. In the presence of chloramphenicol, the inhibition of E2 by the triple-mutant (genes 42, 46, and 47) still occurs, but to a lesser extent.     

Specificity and functions of guanine methylase of Shigella sonnei DDVI phage.

DNA methylase methylating adenine with formation of 6-methylaminopurine has been identified in Shigella sonnei 1188 cells which are the natural host of DDVI phage. At the same time, in DNA of DDVI phage replicating both in Sh. sonnei 1188 cells and in Escherichia coli B cells 7-methylguanine was found as the only minor base in amounts of 0.25 and 0.27 mol per 100 mol of nucleotides, respectively. The extract of the infected cells was found to contain both kinds of DNA methylases: virus-specific guanine methylase and cellular adenine methylase. The lack of 6-methylaminopurine in DNA of this phage is explained by reversible inhibition of the cell enzyme in the infected cells. The amount of methyl groups transferred by DDVI-specific methylase on DNA does not depend on the species of the infected cells and is similar in the case of unmodified SD phage DNA and DNA of T2 phage methylated by E. coli B enzyme. Guanine methylase has been shown to be a DDVI-induced modification enzyme and to protect against restriction of B-type. It methylates double-stranded DNAs only and is inhibited by S-adenosylhomocysteine.     

Synthesis of Messenger Ribonucleic Acid After Bacteriophage T1 Infection

Synthesis of host-specific and phage-specific messenger ribonucleic acid (mRNA) was studied in bacteria infected by unmodified (T1 . B) or modified [T1 . B(P1)] bacteriophage T1. In a "standard" infection of Escherichia coli B by T1 . B (no host-controlled modification involved), the rate and amount of T1 mRNA synthesis was intermediate between those values reported for infections by a virulent phage such as T4 or a temperate phage such as lambda. The initial rate of mRNA synthesis was slightly increased after T1 . B(P1) infection of E. coli B in comparison with T1 . B infection of the same host. Little or no phage mRNA synthesis could be detected in T1 . B infection of E. coli B(P1). Phage mRNA synthesis in T1 . B(P1)-infected E. coli B(P1) cells was approximately the same in amount as that seen in T1 . B(P1) infection of E. coli B. Synthesis of host-specific mRNA continued throughout the latent period in all infections studied. However, the enzyme beta-galactosidase could not be induced, except after T1 . B infection of E. coli B(P1). In an attempt to understand the apparent differences in mRNA synthesis after infection of E. coli B by phages T1 . B or T1 . B(P1), the effect of altered T1 deoxyribonucleic acid (DNA) methylation on mRNA synthesis was studied. Methyl-deficient T1 DNA, made in cells infected with ultraviolet-irradiated phage T3, inhibited (14)C-uridine incorporation more strongly than normal T1. One passage of methyl-deficient T1 through E. coli B restored uracil incorporation rates to those seen with ordinary T1. This suggests that methylation of T1 DNA can influence the rate of phage mRNA synthesis. However, attempts to relate the difference in mRNA synthesis seen between T1 . B and T1 . B(P1) in E. coli B to the activity of the P1 modification gene were not conclusive.     

M13-oriC cloning vehicles: use for amplification of high copy lethal (HCL) genetic elements

M13 cloning vehicles have been constructed which contain the Escherichia coli origin for DNA replication (oriC), with and without selectable antibiotic-resistance genes. Since the M13 viral strand origin requires a functional rep gene product, using oriC these vehicles propagate as low-copy-number plasmids in E. coli rep mutants. This property is exploited to amplify cloned "high copy lethal" (HCL) DNA fragments, those containing genetic elements which kill the E. coli host when present at multiple copies in the cell. Following cloning of such fragments in these vehicles and initial selection in E. coli rep cells, the M13-oriC chimeric plasmid DNA is used to transfect appropriate E. coli rep+ cells. The chimeric DNA propagates as M13 viral DNA, yielding double-stranded and single-stranded DNA products and phage particles prior to killing of the host via expression of the HCL element; these events mimic a lytic phage infection. Such amplification will greatly facilitate both DNA "library" constructions (HCL elements are absent a priori from libraries using high-copy-number cloning vehicles) and studies of HCL elements including restriction mapping, DNA sequencing, and physiological studies.     

Some features of the reaction of intracellular bacteriophages T1 to UV irradiation.

The reaction of complexes pf phage T1-cells of E. coli B or E. coli Bs-1 to UV irradiation was investigated. The complexes were irradiated at various stage of infection, and their survival, extent of Hcr and Phr, were evaluated. It was found that the UV resistance of phage DNA in the second half of the latent period fluctuates. Hcr after UV exposure at these stages of infection operates in a small volume. The ability of intracellular phage to photoreactivate when cells of E. coli B were infected is constant after irradiation at many stages of infection, except the early ones. In the complexes of phage T1-bacteria of E. coli Bs-1 this ability declines while infection is promoted. The daughter phage particles released from UV irradiated complexes undergo Phr and Hcr only after irradiation at the late stages of infection. This was not the cases when complexes of phage-bacteria were irradiated during the first half of the latent period. A possible tole of UV-damaged phage DNA in propagation of infection and in maturation of phage particles is discussed.     

Expression of the genome of Mu-like phage D3112 specific for Pseudomonas aeruginosa in Escherichia coli and Pseudomonas putida cells

The behavior of Escherichia coli cells carrying RP4 plasmid which contains the genome of a Mu-like D3112 phage specific for Pseudomonas aeruginosa was studied. Two different types of D3112 genome expression were revealed in E. coli. The first is BP4-dependent expression. In this case, expression of certain D3112 genes designated as "kil" only takes place when RP4 is present. As a result, cell division stops at 30 degrees C and cells form filaments. Cell division is not blocked at 42 degrees C. The second type of D3112 genome expression is RP4-independent. A small number of phage is produced independently of RP4 plasmid but this does not take place at 42 degrees C. No detectable quantity of the functionally active repressor of the phage was determined in E. coli (D3112). It is possible that the only cause for cell stability of E. coli (D3112) or E. coli (RP4::D3112) at 42 degrees C in the absence of the repressor is the fact of an extremely poor expression of D3112. In another heterologous system, P. putida both ways of phage development (lytic and lysogenic) are observed. This special state of D3112 genome in E. coli cells is proposed to be named "conditionally expressible prophage" or, in short, "conex-phage", to distinguish it from a classical lysogenic state when stability is determined by repressor activity. Specific blockade of cell division, due to D3112 expression, was also found in P. putida cells. It is evident that the kil function of D3112 is not specific to recognize the difference between division machinery of bacteria belonging to distinct species or genera. Protein synthesis is needed to stop cell division and during a short time period this process could be reversible. Isolation of E. coli (D3112) which lost RP4 plasmid may be regarded as an evidence for D3112 transposition in E. coli. Some possibilities for using the system to look for E. coli mutants with modified expression of foreign genes are considered.     

Cloning of the recA gene of Bordetella pertussis and characterization of its product.

A recA gene of Bordetella pertussis was identified in a plasmid library by complementation of a recA mutation in E coli and subcloned as a 2.1-kb Sph I DNA fragment. Southern hybridization experiments showed no similarity to the E coli recA gene, but very strong similarity to other Bordetella species. E coli recA mutant cells containing the B pertussis recA gene at high gene dosage were resistant to DNA-damaging agents such as methyl methane sulfonate or 4-nitroquinoline-N-oxide, displayed induction of SOS functions, and were able to promote DNA recombination, but not induction of phage lambda. The latter phenotype distinguishes the B pertussis recA gene product from the corresponding proteins from most other Gram-negative organisms. Amino acid sequence comparisons revealed a high degree of structural conservation between prokaryotic RecA proteins.     

Replication of the mini-R1 plasmid Rsc11 and Rsc11 hybrid plasmids

Summary The DNA polymerase induced by bacteriophage T7 is composed of a phage-specified subunit, the gene 5 protein, and a host-specified subunit, the 12,000 dalton thioredoxin of Escherichia coli. tsnC mutants of E. coli B (Chamberlin, 1974) have no detectable thioredoxin, and thus cannot support the growth of phage T7, although they are killed by phage infection. A mutant of E. coli K12 affecting thioredoxin has been isolated by a modification of the procedure used by Chamberlin (1974) to isolate tsnC mutants of E. coli B. The gene affecting thioredoxin has been designated trxA. This mutant, E. coli JM109, shows the TsnC phenotype in that it is killed by, but cannot support the growth of, bacteriophage T7. T7 DNA replication does not occur in mutantinfected cells. These phenotypic expressions of the tsnC mutation have enabled us to screen recombinants for the trxA allele in HfrxF^- crosses and F-ductants in episome transfer experiments. Extracts of transductants in generalized transduction by P1 phage were screened for their ability to complement partially purified phage T7 gene 5 protein to form T7 DNA polymerase. The trxA gene is located at 84 min on the E. coli linkage map, between uvrE and metE; trxA is 34% co-transducible with metE.     

Stable albicidin resistance in Escherichia coli involves an altered outer-membrane nucleoside uptake system

Albicidin blocked DNA synthesis in intact cells of a PolA- EndA- Escherichia coli strain, and in permeabilized cells supplied with all necessary precursor nucleotides, indicating a direct effect on prokaryote DNA replication. Replication of phages T4 and T7 was also blocked by albicidin in albicidin-sensitive (Albs) but not in albicidin-resistant (Albr) E. coli host-cells. All stable spontaneous Albr mutants of E. coli simultaneously became resistant to phage T6. The locus determining albicidin sensitivity mapped at tsx, the structural gene for an outer-membrane protein used as a receptor by phage T6 and involved in transport through the outer membrane of nucleosides present at submicromolar extracellular concentrations. Albicidin does not closely resemble a nucleoside in structure. However, Albs E. coli strains rapidly accumulated both nucleosides and albicidin from the surrounding medium whereas the Albr mutants were defective in uptake of nucleosides and albicidin at low extracellular concentrations. An insertion mutation blocking Tsx protein production also blocked albicidin uptake and conveyed albicidin resistance. Albicidin supplied at approximately 0.1 microM blocked DNA replication within seconds in intact Albs E. coli cells, but a 100-fold higher albicidin concentration was necessary for a rapid inhibition of DNA replication in permeabilized cells. We conclude that albicidin is effective at very low concentrations against E. coli because it is rapidly concentrated within cells by illicit transport through the tsx-encoded outer-membrane channel normally involved in nucleoside uptake. Albicidin resistance results from loss of the mechanism of albicidin transport through the outer membrane.     

Gene 1.2 protein of bacteriophage T7. Effect on deoxyribonucleotide pools

The gene 1.2 protein of bacteriophage T7, a protein required for phage T7 growth on Escherichia coli optA1 strains, has been purified to apparent homogeneity and shown to restore DNA packaging activity of extracts prepared from E. coli optA1 cells infected with T7 gene 1.2 mutants (Myers, J. A., Beauchamp, B. B., White, J. H., and Richardson, C. C. (1987) J. Biol. Chem. 262, 5280-5287). After infection of E. coli optA1 by T7 gene 1.2 mutant phage, under conditions where phage DNA synthesis is blocked, the intracellular pools of dATP, dTTP, and dCTP increase 10-40-fold, similar to the increase observed in an infection with wild-type T7. However, the pool of dGTP remains unchanged in the mutant-infected cells as opposed to a 200-fold increase in the wild-type phage-infected cells. Uninfected E. coli optA+ strains contain severalfold higher levels of dGTP compared to E. coli optA1 cells. In agreement with this observation, dGTP can fully substitute for purified gene 1.2 protein in restoring DNA packaging activity to extracts prepared from E. coli optA1 cells infected with T7 gene 1.2 mutants. dGMP or polymers containing deoxyguanosine can also restore packaging activity while dGDP is considerably less effective. dATP, dTTP, dCTP, and ribonucleotides have no significant effect. The addition of dGTP or dGMP to packaging extracts restores DNA synthesis. Gene 1.2 protein elevates the level of dGTP in these packaging extracts and restores DNA synthesis, thus suggesting that depletion of a guanine deoxynucleotide pool in E. coli optA1 cells infected with T7 gene 1.2 mutants may account for the observed defects.     

Studies on energy supply for genetic processes. Requirement for membrane potential in Escherichia coli infection by phage T4

In this study the hypothesis considering the requirement for an electrochemical proton gradient in the injection of phage T4 DNA into Escherichia coli cell has been verified experimentally. The phage caused a reversible depolarization of cell membrane, while phage 'ghosts' induced an irreversible depolarization. The phage infection was strictly dependent on E. coli membrane potential value when phage/cell ratio was 5 and higher. When the ratio was close to 1, the decrease in the membrane potential up to -100 mV caused practically no effect on the phage infection. The infection inhibition was observed when the membrane potential was lowered below this 'threshold' value. On the other hand, the decrease in the membrane potential caused no effect on the phage infection under conditions promoting a concomitant increase in the value of the transmembranous pH gradient. The phage DNA transfer through the membrane of ATPase-deficient cells was reversibly inhibited by switching off the respiratory chain - the sole generator of a protonmotive force in these mutant cells. The membrane should be kept in the energized state during the phage DNA entrance into the cell. Adsorption of the phage on E. coli was followed by the reversible release of the respiratory control. Thus the results presented here indicate the requirement of the electrochemical proton gradient across the plasma membrane for injection of phage T4 DNA into E. coli. They support the concept postulating an expenditure of host cell metabolic energy for phage T4 DNA transfer through the membrane.     

Escherichia coli detection by GFP-labeled lysozyme-inactivated T4 bacteriophage

Escherichia coli has been used as an indicator of the fecal contamination of water and food, identifying potential health hazards. In this study, an E. coli-specific bacteriophage, T4, was used to detect E. coli bacteria. The T4 phage small outer capsid (SOC) protein was used to present green fluorescent protein (GFP), an easily detectable marker protein, on the phage capsid. To inactivate phage lytic activity, we used the T4e(-) phage, which does not produce the lysozyme responsible for host cell lysis. Infection of E. coli K12 cells with the GFP-labeled T4e(-) phage (T4e(-)/GFP) enabled the visualization and distinction of E. coli K12 cells from T4 phage-insensitive cells, Pseudomonas aeruginosa. Prolonged incubation of E. coli K12 cells with the T4e(-)/GFP phage did not lead to cell lysis. Propagation of T4e(-)/GFP in host cells increased the intensity of green fluorescence, making the distinction of E. coli cells from other cells simple and effective. This method enables the rapid, conclusive quantitation of E. coli cells within an hour.     

Rescue of abortive T7 gene 2 mutant phage infection by rifampin.

Infection of Escherichia coli with T7 gene 2 mutant phage was abortive; concatemeric phage DNA was synthesized but was not packaged into the phage head, resulting in an accumulation of DNA species shorter in size than the phage genome, concomitant with an accumulation of phage head-related structures. Appearance of concatemeric T7 DNA in gene 2 mutant phage infection during onset of T7 DNA replication indicates that the product of gene 2 was required for proper processing or packaging of concatemer DNA rather than for the synthesis of T7 progeny DNA or concatemer formation. This abortive infection by gene 2 mutant phage could be rescued by rifampin. If rifampin was added at the onset of T7 DNA replication, concatemeric DNA molecules were properly packaged into phage heads, as evidenced by the production of infectious progeny phage. Since the gene 2 product acts as a specific inhibitor of E. coli RNA polymerase by preventing the enzyme from binding T7 DNA, uninhibited E. coli RNA polymerase in gene 2 mutant phage-infected cells interacts with concatemeric T7 DNA and perturbs proper DNA processing unless another inhibitor of the enzyme (rifampin) was added. Therefore, the involvement of gene 2 protein in T7 DNA processing may be due to its single function as the specific inhibitor of the host E. coli RNA polymerase.     

Localization of Parental Deoxyribonucleic Acid from Superinfecting T4 Bacteriophage in Escherichia coli

High-resolution autoradiography has been employed to localize the nonsolubilized but genetically excluded deoxyribonucleic acid (DNA) of T4 bacteriophage superinfecting endonuclease I-deficient Escherichia coli. This DNA was found to be associated with the cell envelope (this term is used here to include all cellular components peripheral to and including the cytoplasmic membrane); in contrast, T4 DNA in primary infected cells, like host DNA in uninfected E. coli, was found to be near the cell center. The envelope-associated DNA from super-infecting phage was not located on the outermost surface of the cell since it was insensitive to deoxyribonuclease added to the medium. These results suggest that DNA from superinfecting T-even phage is trapped within the cell envelope.     

Effect of Hydroxyurea on Replication of Bacteriophage T4 in Escherichia coli

Hydroxyurea inhibited the replication of bacteriophage T4 in Escherichia coli B. The concentration of hydroxyurea required to inhibit net deoxyribonucleic acid (DNA) synthesis 50% was about 50-fold less than that required in uninfected cells. Even in the presence of high hydroxyurea concentrations, phage DNA was readily synthesized from the products of breakdown of the E. coli DNA, and viable phage were made. Deoxyribonucleotide, but not ribonucleotide, synthesis was strongly inhibited in the presence of hydroxyurea. The data indicate that hydroxyurea specifically inhibits de novo DNA synthesis in E. coli infected with bacteriophage T4 by inhibiting the ribonucleoside diphosphate reductase system, but does not affect DNA synthesis at subsequent steps.     

Partial exclusion of bacteriophage T2 by T4: An early function of the second linkage group

Evidence for exclusion as an early function of phage T4 was obtained along the following lines: a) In crosses of T2 and T4am 122, a mutant of an early phage enzyme, T2 is virtually eliminated from the progeny; b) there is no net synthesis of DNA in the mixed complexes of the non-permissive host; furthermore, there is a loss of DNA, not found in the monocomplexes of T4am 122; this suggests a specific breakdown of T2 DNA; c) a cross was made between T4 and a partially non-excludable, otherwise T2-like phage strain that did not exclude standard type T2. The gene responsible for exclusion segregated in an almost normal way and appeared to reside in-betweenh ⁺ andr at the beginning of the map segment of T4 for the early functions.The frequencies with which the genes of T2 were recovered in the progeny of this cross showed polarity, i.e. the frequency of the T2 genes along the segment of the early functions increased gradually from 0.26 for the adsorption properties to 0.44 for the UV sensitivity of T2 in correlation with the position of the genes on the map.There is a stronger effect of exclusion and a more pronounced polarity of the frequencies of recovered genes of T2 whenE. coli CR 63 is used as a host instead ofE. coli B.This work was carried out partially within the frame of the association between Euratom and the University of Leiden embodied in contract nr. 052-64-1 BIAN.