Identity of Genes Coding for Soluble and Structural Dihydrofolate Reductases in Bacteriophage T4

By recombination between bacteriophage T4 wh2, a dihydrofolate reductaseless mutant, and T6, I have prepared T4 wh(T6), a T4 strain which codes for the T6-specific soluble dihydrofolate reductase. This strain has the heat sensitivity of T6, not T4, which provides direct evidence that the wh gene codes for both the soluble dihydrofolate reductase and the structural dihydrofolate reductase which is a constituent of T-even phage tail plates.     

Bacteriophage Tail Components: I. Pteroyl Polyglutamates in T-Even Bacteriophages

A pteroylpolyglutamate has been found to be a constituent of all Escherichia coli T-even bacteriophages and has been characterized with regard to its oxidation state, molecular weight, origin, and location on the phage particle. The phage compound has been shown to be a dihydropteroyl penta- or hexaglutamate on the basis of its chemical and physical properties. Analyses of extracts of uninfected and T2L-infected E. coli have indicated that the phage dihydropteroyl polyglutamate was present only in infected cells. Its synthesis was sensitive to the addition of chloramphenicol before infection, and the compound appeared to be specifically induced by phage infection. Analyses of isolated phage ghosts and tail substructures have shown that each phage particle contains between two and six phage-specific pteroyl derivatives and that the juncture of the phage tail plate with the tail tube is the most likely site of binding of the phage-induced pteroyl compound.     

Bacteriophage T4 baseplate components. II. Binding and location of bacteriophage-induced dihydrofolate reductase.

The location of T4D phage-induced dihydrofolate reductase (dfr) has been determined in intact and incomplete phage particles. It has been found that phage mutants inducing a temperature-sensitive dfr (dfrts) procude heat-labile phage particles. The structural dfr produced by these ts mutants was shown to assume different configurations depending on the temperature at which the phage is assembled. Morphogenesis of incomplete phage particles lacking the gene 11 protein on their baseplates was found to be inhibited by reagents binding to dfr, such as antibodies to dfr. Further, cofactor molecules for dfr, such as reduced nicotinamide adenine dinucleotide phosphate and reduced nicotinamide adenine dinucleotide, also inhibited the step in morphogenesis involving the addition of gene 11 product. On the other hand, inhibitors of dfr, such as adenosine dephosphoribose, stimulated the addition of the gene 11 protein. It has been concluded that the phage-induced dfr is a baseplate component which is partially covered by the gene 11 protein. The properties of phage particles produced after infection of the nonpermissive host with the one known T4D mutant containing a nonsense mutation in its dfr gene suggested that these progeny particles contained a partial polypeptide, which was large enough to serve as a structural element.     

Bacteriophage T4 baseplate components. III. Location and properties of the bacteriophage structural thymidylate synthetase.

Two T4D thymidylate synthetase (td) temperature-sensitive mutants have been isolated and characterized. Both mutants produce heat-labile phage particles. This observation supports the view that this viral-induced protein is a phage structural component. Further, antiserum to td has been shown to block a specific step in tail plate morphogenesis. The results indicated that the td protein is largely covered by the T4D tail plate gene 11 protein. Since the phageinduced dihydrofolate reductase (dfr) also is partially covered by the gene 11 protein, it appears that td was adjacent to the tail plate dfr. This location has been confirmed by constructing a T4D mutant which is dfrtstdts and showing that these two tail plate constituents interact and give altered physical properties to the phage particles produced. A structural relationship for the tail plate folate, dfr, and td has been reported.     

Bacteriophage Tail Components: III. Use of Synthetic Pteroyl Hexaglutamate for T4D Tail Plate Assembly

The assembly of T4D tail plates occurring during in vitro complementation reactions was found to be stimulated by pteroyl hexaglutamate. Neither the pteroyl pentaglutamate nor the pteroyl heptaglutamate substituted for the hexaglutamate. A small stimulation of the rate and amount of T4D tail plate assembly was observed in untreated extracts. A greater stimulation occurred when activated charcoal-treated bacterial extracts were used. Charcoal treatment inhibited complementation only when no preformed tail plates were present in the extracts, and the inhibition was reversed by the addition of 9 x 10(-6)m chemically synthesized pteroyl hexaglutamate. The stimulation is apparently due to a requirement for the pteroyl hexaglutamate for tail plate assembly.     

Inactivation of T4D Bacteriophage by Antiserum Against Bacteriophage Dihydrofolate Reductase

Antiserum was prepared against highly purified T4D bacteriophage-induced dihydrofolate reductase (DFR). This serum not only inactivated the enzyme but also inactivated all strains of T4D examined. T6 was inactivated to a lesser extent, and T2L, T2H, and T5 were unaffected by the antiserum. The phage-killing power of the serum could be blocked by prior incubation with partially purified T4D dfr obtained from host cells unable to make phage structural proteins. These observations confirm earlier results that the phage dfr is a structural component of the phage particle, and they offer new evidence on the manner in which this enzyme in incorporated into the tail structure.     

Chemically Induced Cofactor Requirement for Bacteriophage T4D

The treatment of bacteriophage T4D with 2-hydroxy-5-nitrobenzyl bromide, a specific reagent for alkylating the indole ring of tryptophan residues, converts these particles from a cofactor-independent form to a cofactor-sensitive form. These treated T4D particles phenotypically resemble T4B particles in certain respects. Their ability to form plaques on minimal medium plates is increased by the addition of l-tryptophan and is inhibited by the addition of indole. In liquid medium, their rate of adsorption is dependent on the presence of the cofactor l-tryptophan. l-Tryptophan-requiring phage have been produced by in vitro assembly of treated tail-fiberless particles of a T4D amber mutant plus untreated tail fiber preparation. When treated tail fibers were used with untreated tail-fiberless particles, the newly assembled particles did not require cofactor. A model of the tail structure of all the T-even bacteriophages is presented which postulates that the active configuration of the tail fibers requires that there be either (i) an endogenous tryptophan residue of the phage particle itself or (ii) an exogenously added l-tryptophan molecule complexed with a specific tryptophan receptor site, most likely on the phage base plate.     

Bacteriophage T4-coded dihydrofolate reductase: synthesis, turnover, and location of the virion protein.

Dihydrofolate reductase plays a dual role in bacteriophage T4, first, as an enzyme of thymidylate metabolism, and second, as a protein component of the tail baseplate. Antibody to the purified enzyme has been used to study its synthesis and intracellular turnover. The antibody specifically precipitates one protein from T4D-infected cell extracts. This has been identified as dihydrofolate reductase, although the polypeptide molecular weight (22,000) is lower than that earlier determined for this enzyme. The protein comigrates on gels with pY, a genetically undefined protein component of the baseplate. However, it is not pY, for pY is synthesized late in infection, whereas virtually no dihydrofolate reductase synthesis occurs later than 10 min after infection at 37 degrees C. Dihydrofolate reductase, once formed, is neither degraded nor converted to proteins of higher or lower molecular weight. Thus, it is probably incorporated into virions at the same molecular weight as that of the soluble enzyme. 125I-radiolabeled antibody binds to the wedge substructure of the baseplate, and this binding is blocked by preincubation with purified T4 dihydrofolate reductase. Thus, the enzyme protein seems to be a component of the wedge.     

Function of T4D Structural Dihydrofolate Reductase in Bacteriophage Infection

Various properties of the bacteriophage structural dihydrofolate reductase (DFR) have been examined to determine its function during phage infection. It has been found that a binding site for reduced nicotinamide adenine dinucleotide phosphate (NADPH), most likely on the DFR present in the phage tail plate, is required for phage viability. Attachment of adenosine diphosphoribose, an analogue of NADPH, to this site prevents phage adsorption and injection. This adenosine diphosphoribose inhibition can be competitively reversed by the addition of NADPH or oxidized nicotinamide adenine dinucleotide phosphate. It is suggested that, during phage infection, the host bacterial cell might leak compounds functionally similar to the pyridine nucleotides. These compounds have been shown to nonenzymatically change the conformation of the phage tail plate DFR which is apparently necessary for successful injection.     

Bacteriophage T4 baseplate components. I. Binding and location of the folic acid.

Two different proteins with high affinities for the pteridine ring of folic acid have been used to determine the location of this portion of the folate molecule in the tail plate of T4D and other T-even bacteriophage particles. The two proteins used were (i) antibody specific for folic acid and (ii) the folate-binding protein from bovine milk. Both proteins were examined for their effect on various intact and incomplete phage particles. Intact T2H was weakly inactivated by the antiserum but not by the milk protein. No other intact T-even phage, including T4D, was affected by these two proteins. When incomplete T4D particles were exposed in an in vitro morphogenesis system, it was found that neither of the two proteins affected either the addition of the long tail fibers to fiberless particles or the addition of tail cores to tail plates. On the other hand, these two proteins specifically blocked the addition of T4D gene 11 product to the bottom of T4D baseplates. After the addition of the gene 11 protein, these two reagents did not inhibit the further addition of the gene 12 protein to the baseplate. It can be concluded that the phage folic acid is a tightly bound baseplate constituent and that the pteridine portion of the folic acid is largely covered by the gene 11 protein.     

Morphogenesis of Bacteriophage φ80: Identification of the Cistron 13 Product

An in vitro complementation reaction leading to the assembly of bacteriophage phi80 tails from component proteins is described. Tail assembly occurs when a lysate of any mutant in cistron 13 is mixed with a second lysate of a mutant in any of the other cistrons involved in tail formation. Lysates of mutants that are blocked in tail formation contain phage heads that can unite with free tails to form infective particles. The rate of the complementation reaction shows little dependence upon temperature, suggesting that the assembly depends largely upon the kinetic encounter of the interacting components. The tail component missing in cistron 13 mutant lysates was purified approximately 55-fold and shown to be, at least in part, a protein having a molecular weight of approximately 22,000. This protein was also released from highly purified infective phi80 particles after osmotic shock followed by heattreatment, suggesting that it most probably is an integral structural protein of the phage tail. Lysates of mutants of bacteriophage lambda that are defective in tail formation were shown to contain a tail component identical with or similar to the phi80 cistron 13 product.     

Analysis of T4 bacteriophage deletion mutants that lack td and frd genes.

The roles of bacteriophage T4-encoded thymidylate synthase and dihydrofolate reductase as virion structural components have been further investigated. Two mutants, del(63-32)7 and del(63-32)9, bearing deletions in the gene 63 to 32 region of the T4 genome, were characterized by Southern blotting analysis, as well as by enzyme and immunological assays. Our results have confirmed the original report of Homyk and Weil (Virology 61:505-523, 1974) that del7 and del9 each carries a deletion of about 4.0 kilobases, which totally eliminates the frd gene, encoding dihydrofolate reductase, and the td gene, encoding thymidylate synthase. With the well-characterized deletion mutants, along with newly prepared antisera against T4-encoded thymidylate synthase and dihydrofolate reductase, we have reevaluated the experimental results supporting the idea that T4-induced dihydrofolate reductase and thymidylate synthase are essential T4 baseplate components and antigenic determinants of phage particles. These deletion mutant phages are not targets for neutralization by antisera against either dihydrofolate reductase or thymidylate synthase purified from cloned genes. Furthermore, these newly prepared antisera also cannot neutralize the infectivity of T4D. Those results suggest that the phage-neutralizing components in the old antisera used in the earlier studies were not antibodies against either dihydrofolate reductase or thymidylate synthase but were antibodies against minor components of the purified enzyme preparations. Study of the biological properties of the deletion mutants indicates that T4-induced thymidylate synthase and dihydrofolate reductase play significant roles in growth of the phage beyond their known roles in nucleotide biosynthesis, even though they are apparently not essential for phage viability. The deletion mutants should be useful in defining these roles.     

Critical Arginine Residue for Maintaining the Bacteriophage Tail Structure

The addition of 0.2 m l-arginine to various T-even bacteriophage preparations inactivated the virus preparations irreversibly. The virus particles were even more sensitive to added d-arginine and l-homoarginine than to l-arginine but were unaffected by arginine analogues with either an altered carboxyl group or guanidyl group. Treatment of phage T2H with 2,3-butanedione, a reagent which specifically reacts with the guanidyl portion of arginine residues, resulted in the apparent in-activation of most of the virus particles. However, after incubation of the treated particles at pH 7.5 at 37 C for 1 hr in the absence of butanedione, the original virus titer almost completely returned. The reactivation was completely inhibited by the presence of 0.2 m d-arginine. It appeared that the virus protein coat was sufficiently plastic so that the initial conformational change resulting from the alteration of an arginine residue (to possibly an ornithine residue) was at least partially reversible and that the virus tail proteins then refolded to produce a stable and active virus particle. These reactivated virus particles were not sensitive to inactivation by d-arginine but could now be rapidly inactivated by l-ornithine. Virus particles inactivated by arginine have altered tail structures. They have contracted tail sheaths still attached to tail plates and still contain tail cores. These properties of virus particles indicate that there is a free carboxyl group and a guanidyl group spatially equivalent to an arginine residue on one component of the virus tail which bind reversibly by means of polar linkages to another tail component. These bonds maintain the integrity of the virus tail. Added arginine appears to compete with this endogenous viral arginine for the binding sites and then to favor an irreversible conformational change.     

Characterization of New Regulatory Mutants of Bacteriophage T4

Plating techniques which eliminate T4 plaque formation on Escherichia coli by folate analogue inhibition of dihydrofolate (FH(2)) reductase (EC 1.5.1.3) allowed the isolation of folate analogue-resistant (far) mutants of T4. One class of far mutants overproduces the phage-induced FH(2) reductase. Deoxycytidylate deaminase (EC 3.5.4.12), thymidine kinase (EC 2.7.1.21), and deoxycytidine triphosphatase (EC 3.6.1.12) are also overproduced by 20 min after infection at 37 C. The overproduction of FH(2) reductase by these far mutants is not affected by the absence of DNA synthesis. Other types of mutations that affect the synthesis of early enzymes cause overproduction in the absence of DNA synthesis of some of the above enzymes but not of FH(2) reductase. Therefore, overproducing far mutants apparently have mutations in previously undescribed genes controlling the expression of the T4 genome. Three of four mutants under study map near gene 56, and one maps near gene 52. All of these mutants show delays in DNA synthesis, phage production, and lysis and appear to show decreased levels of RNA synthesis based on the cumulative incorporation of uridine.     

Bacteriophage T4 Virion Baseplate Thymidylate Synthetase and Dihydrofolate Reductase

Additional evidence is presented that both the phage T4D-induced thymidylate synthetase (gp td) and the T4D-induced dihydrofolate reductase (gp frd) are baseplate structural components. With regard to phage td it has been found that: (i) low levels of thymidylate synthetase activity were present in highly purified preparations of T4D ghost particles produced after infection with td⁺, whereas particles produced after infection with td⁻ had no measurable enzymatic activity; (ii) a mutation of the T4D td gene from td^ts to td⁺ simultaneously produced a heat-stable thymidylate synthetase enzyme and heat-stable phage particles (it should be noted that the phage baseplate structure determines heat lability); (iii) a recombinant of two T4D mutants constructed containing both td^ts and frd^ts genes produced particles whose physical properties indicate that these two molecules physically interact in the baseplate. With regard to phage frd it has been found that two spontaneous revertants each of two different T4D frd^ts mutants to frd⁺ not only produced altered dihydrofolate reductases but also formed phage particles with heat sensitivities different from their parents. Properties of T4D particles produced after infection with parental T4D mutants presumed to have a deletion of the td gene and/or the frd gene indicate that these particles still retain some characteristics associated with the presence of both the td and the frd molecules. Furthermore, the particles produced by the deletion mutants have been found to be physically different from the parent particles.     

Relationship of Bacillus subtilis DNA polymerase III to bacteriophage PBS2-induced DNA polymerase and to the replication of uracil-containing DNA.

In vivo studies of PBS2 phage replication in a temperature-sensitive Bacillus subtilis DNA polymerase III (Pol III) mutant and a temperature-resistant revertant of this mutant have suggested the possible involvement of Pol III in PBS2 DNA synthesis. Previous results with 6-(p-hydroxyphenylazo)-uracil (HPUra), a specific inhibitor of Pol III and DNA replication in uninfected cells, suggest that Pol III is not involved in phage DNA replication, due to its resistance to this drug. Experiments were designed to examine possible explanations for this apparent contradiction. First, assays of the host Pol III and the phage-induced DNA polymerase activities in extracts indicated that a labile Pol III did not result in a labile phage-induced enzyme, suggesting that this new polymerase is not a modified HPUra-resistant form of Pol III. Indeed the purified phage-induced enzyme was resistant to the active, reduced form of HPUra under all assay conditions tested. Since in vitro Pol III was capable of replicating the uracil-containing DNA found in this phage, the sensitivity of the purified enzyme to reduced HPUra was examined using phage DNA as template-primer and dUTP as substrate; these new substrates did not affect the sensitivity of the host enzyme to the drug.     

Bacteriophage-coded thymidylate synthetase. Evidence that the T4 enzyme is a capsid protein

It has previously been shown that T4 bacteriophage-coded dihydrofolate reductase is a capsid protein, specifically an element of the tail plate. This paper presents evidence that thymidylate synthetase is also a structural protein. Antiserum prepared against purified T4 thymidylate synthetase neutralizes T4 infectivity. Evidence is presented that structural thymidylate synthetase is the target of the antiphage component of the serum.The td gene in T4 codes for thymidylate synthetase. We have crossed the td gene from phage T6 into T4 and eliminated other T6 genetic material from the hybrid phage by extensive backcrossing. The hybrid phage, T4td^T6, is inactivated at 60 °C significantly more rapidly than the parent phage, T4D. Thus, the td gene is a determinant of a physical property of the virion, providing direct confirmation that thymidylate synthetase is a capsid protein. At present the role of the virion-bound enzyme is unknown.     

T6r+-induced Proteins and Nucleic Acids in Escherichia coli Infected in the Presence of Streptomycin

Streptomycin does not strongly inhibit T-even phage multiplication in the streptomycin-susceptible polyauxotroph, Escherichia coli strain T(-)H(-)U(-). The relatively slight inhibition, observed earlier, on production of late proteins has now been studied further. The phage-induced ribonucleic acid, synthesized in T6 phage infection in the presence of streptomycin, has been characterized by its base composition, size distribution, and behavior in hybridization tests. Comparison of these properties to those of control samples, taken during either early or late periods of infection, have not shown any significant differences. Phage-induced proteins, synthesized at different times during infection, were studied by disc-gel electrophoresis. Staining and autoradiography of the patterns of pulse-labeled proteins, formed in the absence and presence of the antibiotic showed only slight quantitative changes in the appearance of early proteins. More marked quantitative effects were detected later in infection. Nevertheless, changes in the mobilities of the different proteins were not observed in the streptomycin-treated cultures at any time after infection, suggesting the absence of gross misreading sufficiently great to alter the distinctive electrophoretic patterns of the extracts. Cells infected and incubated in the presence of the antibiotic were found to contain intact virus particles, as shown by electron microscopy. Such infected cells contained extensive deoxyribonucleic acid pools and did not develop the rounded nucleoids with enclosed dense bodies characteristic of the lethal action of the antibiotic. On the other hand, infected bacteria previously exposed to lethal concentrations of streptomycin were unable to synthesize the early enzymes, deoxycytidylate (dCMP) hydroxymethylase and dihydrofolate reductase, or to make phage deoxyribonucleic acid and phage. Such previously killed cells contained the rounded and clotted nucleoids and were unable to unravel this pathological structure after phage infection.     

Identification of bacteriophage T4D gene 29 product, a baseplate hub component, as a folylpolyglutamate synthetase

An assay for folylpolyglutamate synthetase activity in extracts of uninfected and bacteriophage T4D-infected Escherichia coli B has been developed. T4D infection induced the formation of a new synthetase raising the total synthetase activity three-fold. Extracts obtained after infection with T4 gene 51, 27 or 28 amber mutants showed increased synthetase activities while extracts obtained from cells infected with a T4D gene 29 amber mutant did not show any increase in synthetase activity. The phage-induced synthetase was found to copurify with the gene 29 product and a 100-fold purified synthetase of molecular size of 74,000 daltons has been obtained. The purified synthetase has a folate substrate specificity different from the host synthetase since it added glutamate residues to dihydrofolate as well as to the usual tetrahydrofolate substrate.     

The role of thioredoxin in filamentous phage assembly. Construction, isolation, and characterization of mutant thioredoxins

Filamentous phage assembly in vivo shows an absolute requirement for thioredoxin and a partial requirement for thioredoxin reductase. Mutants in which one or both of the active site cysteine residues of thioredoxin were changed to alanine or serine were constructed and shown to support filamentous phage assembly. Some of the mutants were almost as effective as wild-type thioredoxin, while others supported phage assembly only when high levels of the mutant protein were present in the infected cell. The mutant proteins were all inactive in an assay which couples oxidation of NADPH to reduction of 5,5'-dithiobis-2-nitrobenzoic acid) via thioredoxin reductase and thioredoxin. These active site mutants make phage assembly completely independent of thioredoxin reductase, which suggests that the phage needs, and the active site mutants provide, the proteins in the reduced conformation. Other mutants were isolated on the basis of their failure to support filamentous phage growth. These specified mutant thioredoxin proteins with varying levels of redox activity in vivo and in vitro. The locations of these mutations suggest that the surface of thioredoxin thought to interact with thioredoxin reductase also interacts with the filamentous phage assembly machinery. An in vivo assay for thioredoxin redox function, based on the ability of cells to utilize methionine sulfoxide, was developed. Met- cells containing mutant thioredoxins that are inactive in vitro do not form colonies on plates containing methionine sulfoxide as the sole methionine source.