Biological activity of purified bacteriophage T3 prohead and prohead-like structure as precursors for in vitro head assembly

Proheads were purified by differential centrifugation followed by sucrose density gradient centrifugation from Escherichia coli cells infected with a bacteriophage T3 mutant defective in either DNA synthesis (gene 5) or DNA maturation (gene 19). The prohead could be converted to infectious phage very efficiently by incubation in vitro with an extract prepared from cells infected with mutants defective in head genes 8, 9, 10, 13, 14, 15, or 16 (8^? extract and so on) but not with 19^? extract. Under the most favorable condition, 40% of added proheads were converted to infectious phage. The efficiency of the conversion did not depend upon the prohead concentration.Sodium dodecyl sulfate-polyacrylamide gel electrophoresis demonstrated that prohead contained the proteins coded by genes 8, 9, 10, 13, 14, 15, and 16 but did not contain the product of gene 19. Prohead-like structures were isolated from cells infected with double mutants of gene 5 and one of the head protein genes 8, 13, 14, 15, and 16. These structures were not converted to infectious phage when incubated with 10^? extract (lacking the major head protein and had protein compositions differing from biologically active proheads. From these results, we discuss the possible role of head gene products in head assembly of phage T3.     

In vitro assembly of polioviruses. 3. Assembly of 14 S particles into empty capsids by poliovirus-infected HeLa cell membranes

Rough membranes obtained from poliovirus-infected HeLa cells have the capacity to assemble 14 S particles into 73 S empty capsids in vitro. A corresponding fraction from uninfected cells did not possess this activity. When labeled 14 S particles were incubated with smooth membranes obtained from infected cells, a significant amount of radioactivity sedimented as heterogeneous material in the 30–70 S region of the gradient.Sucrose gradient analysis of ¹⁴C-labeled rough and smooth membrane fractions lysed with deoxycholate (DOC) demonstrated the presence of 14 S particles, 73 S empty capsids, and a 110 S structure in the rough membrane fraction. These structures were found only in trace amounts in the smooth membrane fraction. In the absence of DOC treatment, no 14 S particles were found in any of the membrane fractions. The 73 S empty capsids, on the other hand, were detected in the presence or absence of DOC treatment in the rough-membrane fraction. Therefore, it appears that 14 S particles are associated with the rough membranes where they are assembled into 73 S empty capsids and/or complete virions.     

Alterations of the bacteriophage T7 and T3 DNA packaging pathway in Escherichia coli mutant TSN B

Data previously obtained indicate that, during assembly of the related bacteriophages T7 and T3, a DNA-free procapsid (capsid I) is produced and that subsequently capsid I: (1) binds to a longer than mature (concatemeric) DNA and then becomes structurally altered to a particle isolated as a capsid (capsid II) physically resembling the mature bacteriophage capsid more than the procapsid (initiation phase of packaging), (2) draws DNA to its interior (entry phase of packaging), (3) participates in cutting the concatemeric DNA to mature size. It was found that, after infection of Escherichia coli mutant tsnB (selected for a deficiency in plating T7; M. Chamberlin [1974], J. Virol. 14, 509-516), T7 and T3 capsid I is assembled at a rate not significantly different from its rate of assembly in the wild-type host. However, the conversion of capsid I to capsid II was slowed in E. coli tsnB, suggesting that the tsnB mutation interferes with the initiation of DNA packaging. Although some T3 and T7 DNA enters capsids and is cut to mature size in the tsnB mutant, the data further suggest that the entry rate of DNA into capsid II is lower in the tsnB mutant than it is in an unaltered host. T7 capsid II-concatemeric DNA complexes accumulate during infection of the tsnB mutant. These observations suggest that use of the tsnB mutant as a host will simplify studies of bacteriophage T7 and T3 DNA packaging.     

Absence of subviral particles and assembly activity in HeLa cells infected with defective-interfering (DI) particles of poliovirus

HeLa cells infected with defective-interfering (DI) particles of poliovirus were examined for their capacity to synthesize viral proteins, viral-related particles, and the viral factor(s) that promotes the assembly of 14 S particles into empty capsidsin vitro. As reported by C. N. Cole and D. Baltimore (1973,J. Mol. Biol.76, 325–343) cells infected with purified DI particles failed to synthesize the capsid precursor NCVP 1a or any of the capsid polypeptides VP 0, VP 1, VP 2, or VP 3. Consequently, no 14 S particles, empty capsids, or virions were formed at low to moderate multiplicities. Cytoplasmic extracts prepared from cells infected with purified DI particles do not promote the assembly of viral 14 S particles into empty capsids. We conclude that the presence of assembly activity is dependent on the formation of the capsid precursor protein (NCVP-1a) or its cleavage products.     

Intracellular visualization of precursor capsids in phage P22 mutant infected cells

The morphogenesis of the double-stranded DNA Salmonella phage P22 has been studied by electron microscopy of sections of wild type and mutant-infected cells. Previous work had shown that the precursor capsid structure that encapsidates DNA is a complex of two major protein species, the gene 5 coat protein and the gene 8 scaffolding protein. Scaffolding protein exits from the precursor capsid in coupling with DNA encapsidation and then recycles.No organized structures were seen in cells infected with amber mutants of the coat protein gene. Cells infected with an amber mutant of the scaffolding gene contained small numbers of aberrant particles, including empty petit capsids and giant nested or spiral shell structures.In cells infected with mutants blocked in DNA encapsidation (genes 1, 2, and 3) precursor capsids (proheads) accumulate amid the vegetative DNA and not along the membrane, as in phage T4. The proheads appear as double-shell structures about the size of mature phage, with the inner cell diameter about two-thirds the dimension of the outer shell.Mature phage and defective particles containing DNA form paracrystalline arrays within infected cells. Empty capsids, lacking both DNA and the inner shell of proheads, appear within the paracrystalline arrays of filled heads in cells infected with mutants blocked in head completion (genes 10 and 26). These empty capsids are presumably derived from filled but incomplete heads that have lost their DNA intracellularly.Use of temperature-sensitive mutants blocked in the encapsidation steps allowed visualization of the first filled heads upon shift to permissive temperature. These particles tended to appear at the edge of the DNA pool. Partially filled particles with dense central cores often were seen associated with the growing paracrystalline arrays, and they probably represented intermediates in encapsidation.These experiments, in conjunction with others, suggest that the scaffolding protein, which functions in prohead assembly and perhaps in DNA encapsidation, is organized into an inner shell within the precursor capsid.     

Role of gene 8 product in morphogenesis of bacteriophage T3

The product of gene 8 (gp8) of T3 phage is one of the minor head proteins located at the phage head-tail junction. To determine the role of gp8, an amber (8-) and four temperature-sensitive mutants (ts8) were characterized by sedimentation analysis, polyacrylamide gel electrophoresis, and extract complementation. Neither DNA-containing particles nor empty particles were formed in cells infected with 8-. In addition, prohead assembly was greatly reduced. Prohead assembly was also blocked in cells infected with all ts8 mutants at 42 degrees and with some ts8 even at 37 degrees. Proheads containing gpts8 were converted to empty heads when cell lysates were treated with chloroform. The protein compositions of proheads showed that the minor head proteins, gp8, gp15, and gp16, were lost from proheads formed in cells infected with ts8, but these minor proteins were present in proheads formed in cells infected with double mutants of ts8 and 5- or 19-, which are defective in DNA synthesis or DNA maturation, respectively. In vitro complementation experiments suggested that a ts mutation in gene 8 affected not only DNA packaging but also subsequent assembly steps. From these results, it is concluded that gp8 plays multiple roles in T3 phage morphogenesis, including prohead assembly, prohead stabilization, DNA packaging, and subsequent events.     

Electron microscopy of the morphogenesis of Bacillus subtilis bacteriophage SP3.

The capsid of Bacillus subtilis bacteriophage SP3 is assembled via a prohead intermediate which subsequently encapsulates DNA and attaches a tail. The prohead contains a ring-like core structure. The spokes which extend from the core to the inner prohead surface are thought to form a scaffold for the polymerization of the prohead. Ninety percent of the proheads are assembled prior to the onset of DNA encapsulation. The first mature phage particles are observed at 45 min after infection; titres of intracellular phage demonstrate their infectivity. The core is visible in phage ghosts.     

Cell-free assembly of a polyoma-like particle from empty capside and DNA.

A polyoma-like particle (PLP) is formed when polyoma DNA and purified empty capsids are incubated in a cell-free system. The DNA of this new particle is protected against the action of pancreatic DNase. The density of the purified PLP in CsCI is 1.32 g/cm³, which is intermediate between that of polyoma virions (1.34 g/cm³) and empty capsids (1.29 g/cm³). Purified PLP sediments at 190 S in sucrose and is stable in solutions of high ionic strength. When the DNA is extracted from PLP by the use of detergent and phenol, it is found to be doublestranded with a molecular weight of approximately 1.1 × 10⁶. The particles are stable in CsCl at 4° for at least 5 months. Electron micrographs indicate that highly purified PLPs stained with 2% PTA have the same appearance as polyoma capsids. Neither aggregates nor complexes bound by loose ionic bonds appear reasonable to explain these results. The evidence indicates that the DNA of this new polyoma-like particle, made under cell-free conditions, is protected by the capsid.     

Complexes between bacteriophage T7 capsids and T7 DNA

Heat inactivation of bacteriophage T7 (sedimentation coefficient 453 S) yields a nucleoprotein structure sedimenting at 40 S (in vitro 40 S structure). Lysates of T7-infected Escherichia coli, made with the nonionic detergent Brij 58, also contain a 40 S nucleoprotein structure (in vivo 40 S structure). Both structures consist of a phage capsid bound to a mature length of T7 DNA which is outside the phage head. The two differ in that the capsid of the in vitro 40 S structure was located at a random distance from the nearest end of the DNA while the capsid from the in vivo structure was always located less than 7% of the mature DNA length from the nearest end. The nearest end for the in vivo structure seems to be the genetic right end of the DNA. Kinetic labeling experiments suggest that the in vivo 40 S structure is either an intermediate in DNA packaging or a breakdown product from such an intermediate.The fast side of the in vivo 40 S peak contained structures in which T7 capsids were attached to DNA which was longer than mature T7 DNA. The largest DNA detected was close to twice the length of mature T7 DNA.     

Alterations in virus protein synthesis and capsid production in infection with DI particles of herpesvirus

High multiplicity, undiluted passage of equine herpesvirus type 1 (EHV-1) in L-M cells resulted in the rapid production of virus particles whose genome was genetically less complex, contained more reiterated DNA sequences and exhibited a greater buoyant density (rho = 1.724 g/ml) than the DNA (rho = 1.716 g/ml) of standard virus. These data and the finding that these particles inhibited the replication of standard virus in interference assays confirmed that these were defective interfering (DI) particles (Henry et al. 1979). Additional evidence for this has been obtained from the pattern of cyclic fluctuation in infectious virus titre through 17 serial passages as well as from the pronounced variation in the particle to plaque ratio for each passage. Total particle production was markedly reduced in cells infected with virus preparations containing DI particles and quantification of major cell-associated EHV-1 capsid species by electron microscopy and analysis in Renografin density gradients indicated that this reduction occurred at the level of capsid assembly. Although total capsid production was reduced in cells infected with DI particle preparations, the synthesis of I (immature) capsids increased relative to that of L (empty) capsids and these alterations in the assembly of capsid species could be related to changes in the synthesis of capsid proteins. In cells infected with EHV-1 preparations rich in DI particles, the synthesis of major capsid protein 150000 was greatly reduced, whereas core protein 46000, a major component of I capsids, was overproduced as compared to standard virus infection. Capsids produced in cells infected with virus preparations rich in DI particles were identical in polypeptide composition to those made in standard virus infection.     

Bacteriophage T4 head morphogenesis. V. The role of DNA synthesis in maturation of an intermediate in head assembly

On the basis of metabolic-inhibitor studies it was found that maturation of T4 head intermediates, which had accumulated in tsC9 (gene 49) mutant infected cells, was apparently dependent on continued DNA synthesis. This requirement for DNA synthesis in head maturation was confirmed by using a gene 43 mutant (tsP36) blocked in the enzyme, T4 DNA polymerase. In temperature shift-up experiments with tsP36-infected cells it was found as previously reported (Luftig and Ganz, 1972) that 300s particles accumulated when enzyme synthesis was halted. These latter structures appear identical to the gene 49-defective intermediates since, when isolated, they both: (i) had an average sedimentation coefficient of 310 ± 20s and a nucleoprotein density of 1.34 g/cc; (ii) had the major capsid polypeptide in a cleaved state; (iii) were quantitatively rescued (to a >50% level in 40 min) after the mutant-blocked function was recovered, and (iv) showed a reduced ability to convert to phage when an inhibitor of DNA synthesis such as fluorodeoxyuridine (FUdR) was added at the time the function was rescued.The maturation of rapidly labeled 300s wild-type (T4D⁺) head intermediates was also found to be sensitive to inhibitors of DNA synthesis. Upon isolation these heads had the same properties as the above gene 49 and gene 43 mutant-defective intermediates. This strongly suggests that DNA synthesis is a normal requirement for maturation of a T4 capsid intermediate, whose properties are defined above.     

Characterization of Aleutian mink disease virus.

Aleutian mink disease virus (ADV), propagated in normal mink kidney cell cultures, was resolved by centrifugation on CsCl density gradients into two components of densities 1.334 and 1.295 gm/cm³ which were, respectively, about 85 and 15% of the total virus yield. Electron microscopic examination revealed that the major component contained predominantly intact virions of 25 nm in diameter which resembled virus particles with icosahedral structure while the minor component contained mainly empty capsids. Both the empty capsids and intact virus particles absorbed to immunoabsorbents prepared with IgG from chronically infected mink. The buoyant density, size, and structure of purified ADV are in common with viruses belonging to the picornavirus group.     

Studies on the in vitro formation of infectious DNA-proteins aggregates from SV40 components.

Complete SV40 virions, disrupted in vitro at pH 10.6, were reassociated by HCl neutralization into a heterogeneous group of DNA-protein aggregates. These contained lower, varying ratios of DNA:protein than control virions. After “reassociation” a portion of the DNA was sensitive to DNase treatment. This treatment led to a loss of the infectivity which had been found in the reassociated aggregates. Nucleic acid is not essential for physical reassociation since aggregates reconstituted from disrupted complete virions had varying amounts of DNA and further aggregates were readily assembled from disrupted empty shells. Disruption by the relatively mild conditions used indicates that the capsid is held in its conformation in part by noncovalent linkage. The rapid reassociation as well as the absence of low molecular weight protein after disruption suggests that there is not a total breakdown by the disruption procedure used. The nucleic acid is probably bound to the capsomeres by noncovalent linkages since disruption does lead to loss of a large amount of the DNA. The rapid in vitro reconstitution of disrupted empty particles to protein aggregates similar to native empty shells strongly suggests that orientation of capsomeres into a tertiary conformation is fairly specific when unencumbered by the presence of DNA. Successful attempts were made to synthesize infectious aggregates in vitro using disrupted [¹⁴C]amino acid-labeled empty shells and [³H]thymidine-labeled SV40 nucleoprotein complex obtained from infected cells 25 hr postinfection.     

Display of a PorA peptide from Neisseria meningitidis on the bacteriophage T4 capsid surface.

The exterior of bacteriophage T4 capsid is coated with two outer capsid proteins, Hoc (highly antigenic outer capsid protein; molecular mass, 40 kDa) and Soc (small outer capsid protein; molecular mass, 9 kDa), at symmetrical positions on the icosahedron (160 copies of Hoc and 960 copies of Soc per capsid particle). Both these proteins are nonessential for phage infectivity and viability and assemble onto the capsid surface after completion of capsid assembly. We developed a phage display system which allowed in-frame fusions of foreign DNA at a unique cloning site in the 5' end of hoc or soc. A DNA fragment corresponding to the 36-amino-acid PorA peptide from Neisseria meningitidis was cloned into the display vectors to generate fusions at the N terminus of Hoc or Soc. The PorA-Hoc and PorA-Soc fusion proteins retained the ability to bind to the capsid surface, and the bound peptide was displayed in an accessible form as shown by its reactivity with specific monoclonal antibodies in an enzyme-linked immunosorbent assay. By employing T4 genetic strategies, we show that more than one subtype-specific PorA peptide can be displayed on the capsid surface and that the peptide can also be displayed on a DNA-free empty capsid. Both the PorA-Hoc and PorA-Soc recombinant phages are highly immunogenic in mice and elicit strong antipeptide antibody titers even with a weak adjuvant such as Alhydrogel or no adjuvant at all. The data suggest that the phage T4 hoc-soc system is an attractive system for display of peptides on an icosahedral capsid surface and may emerge as a powerful system for construction of the next generation multicomponent vaccines.     

Functions of two new genes in Salmonella phage P22 assembly.

The capsid of bacteriophage P22 is composed of at least eight protein species, six of which had been identified as the products of known P22 genes. The two smallest proteins, pX of 18,000 daltons, and pa of 15,000 daltons, were not the products of any of the known P22 genes. We have isolated and characterized phage carrying amber mutations in the two genes which code for pX and pα (designated gene 7 and gene 4, respectively). The products of these genes are essential for the formation of viable phage. The gene 7 protein is not needed for particle assembly, but rather for particle infectivity. Restrictive cells infected with a mutant in gene 7 accumulate structures with the sedimentation properties and electron microscopic appearance of wild-type phage. Although the defective 7^? particles adsorb to cells, they are unable to recombine with a defective prophage inside the cell or express phage-specific protein synthesis. The 7^? phenotype is similar to the phenotypes of amber mutants in genes 16 and 20 [Botstein et al., J. Mol. Biol.80, 669–695 (1973)]. Genes 7, 16, and 20 map contiguously. The gene 4 product seems to be necessary for the stabilization of newly DNA-filled phage heads. Restrictive cells infected with a mutant of gene 4 accumulate structures with the sedimentation properties and electron microscopic appearance of the “empty heads” that occur in smaller relative amounts in wild-type lysates. These 4^? empty heads are derived from unstably packaged heads inside the cell. The product of gene 4 is not required for cutting phage DNA into mature length pieces. The 4^? phenotype is similar to the 10^? and 26^? phenotypes [Lenk et al., Virology68, 182–199 (1975)]. Genes, 4, 10, and 26 map contiguously. All the proteins found in P22 phage and phage precursor particles have been matched with their genes.     

Characterization of morphogenetic intermediates and progeny of normal and alkylated bacteriophage T7

Analysis of thin sections of Escherichia coli B cells infected by normal (nonalkylated) or alkylated bacteriophage T7 showed that alkylation altered phage morphogenesis. To understand these morphogenetic alterations, we have isolated phage-related particles from infected-cell lysates by differential and sucrose gradient centrifugation. Cells infected by normal and by alkylated phage produced mature phage particles, empty heads, and proheads; however, production of proheads and mature phage particles was less in the case of alkylated phage. These lysates also contained sedimentable material which migrated more slowly than empty heads on sucrose gradients. In the case of alkylated phage, this peak contained radioactive material in amounts nearly equal to that in either proheads or empty heads; for normal phage, this peak represented a smaller fraction of the total radioactivity. Examination of the gradient fractions by electron microscopy revealed appreciable quantities of phage tails and tail-related particles. The same gradient fractions contained phage tail proteins: gene products (gps) 11, 12, and 17 as well as smaller amounts of gp 8, the head-tail connector. In addition, these fractions contained two other proteins which we believe to be of bacterial origin. These proteins may be related to tail formation or function as part of the phage receptor. On the basis of our data, we propose an alternative morphogenetic pathway for T7 tail formation, a pathway which would involve formation of a complex of tail proteins prior to association with the phage head.     

Structural proteins of bacteriophage T5

Analysis by SDS polyacrylamide gel electrophoresis has revealed that bacteriophage T5 contains at least 15 different polypeptides, five of which are associated with the head. The head has one major protein with a molecular weight of about 32,000 daltons, representing about 65% of the total protein. The major protein of the tail has a molecular weight of about 51,000 daltons and represents approximately 17% of the total protein of the phage. No detectable differences were noted between the structural proteins of T5⁺ and its heat-stable mutant, T5st. Purified capsids or “ghosts,” prepared by treatment with lithium chloride, lacked at least two minor head proteins. However, these proteins were still found in ghosts prepared by freeze-thawing. Amber mutants defective in structural genes were isolated and partially characterized. Some of these mutants produced functional heads or tails under restrictive conditions. T5stamN5 induced the production of empty defective heads, which were normal in size, but lacked a protein, which normally accounts for about 6% of the total protein of the phage.     

Endonuclease of T4 ghosts.

An endonuclease was found to associate with purified T4 particles. Its activity was low with intact phages but became distinct when phage particles were disrupted by osmotic shock, and more distinct when they were disassembled by guanidine. The enzyme cleaved all DNA species tested so far, including glucosylated and nonglucosylated T4 DNA, T3 DNA and fd DNA. The cleavage was more efficient with doublestranded DNA than with single-stranded DNA. Defective capsids of mutants of genes 4, 10, 14, 16, and 49 had enzyme activity which was comparable to capsids of the wild type, suggesting that the enzyme is located on capsids. The nucleolytic activity was barely detectable, if at all, with four out of five mutants of gene 17.     

DNA arrested mutants of gene 59 of bacteriophage T4. II. Replicative intermediates

The abnormal DNA replication of bacteriophage T4 containing amber nonsense mutants in gene 59 (amC5, amHL628) has been studied. Zonal centrifugal analysis of the mutant intracellular DNA molecules isolated from a nonpermissive host, Escherichia coli B, reveals the following: (1) Parental mutant DNA as well as nascent mutant DNA can be recovered by sucrose gradient centrifugation as a fast sedimenting fraction (>1500 S) which is presumably a cell-membrane DNA replication complex. (2) At 2–3 min after initiation of viral DNA synthesis, a gradual accumulation of mutant DNA with a sedimentation coefficient of 63 S is observed. The molecular weight of this DNA is close to, or a little less than, that of a mature viral DNA, suggesting that less than one round of replication occurs before the complete arrest of DNA synthesis. (3) In contrast to what is found in wild-type infected cells, little or no long single-stranded DNA (concatemer) appears in mutant infected cells. Concatenated molecules of DNA, which have been proposed as intermediates in viral replication, are normally found in wild-type infected cells 7 min after infection. (4) Unlike the case of cells infected with mutants of gene 46 and 47, addition of chloramphenicol prevents premature release of the mutant DNA of gene 59 from the fast-sedimenting fraction (>1500 S).