Search

The maturation of pseudorabies virus DNA from the replicative concatemeric form to molecules of genome length was examined using nine DNA+ temperature-sensitive mutants of pseudorabies virus, each belonging to a different complementation group. At the nonpermissive temperature, cells infected with each of the mutants synthesized concatemeric DNA. Cleavage of the concatemeric DNA to genome-length viral DNA was defective in all the DNA+ ts mutants tested, indicating that several viral gene products are involved in the DNA maturation process. In none of the ts mutant-infected cells were capsids with electron-dense cores (containing DNA) formed. Empty capsids with electron-translucent cores were, however, formed in cells infected with six of the nine temperature-sensitive mutants; in cells infected with three of the mutants, no capsid assembly occurred. Because these three mutants are deficient both in maturation of DNA and in the assembly of viral capsids, we conclude that maturation of viral DNA is dependent upon the assembly of capsids. In cells infected with two of the mutants (tsN and tsIE13), normal maturation of viral DNA occurred after shiftdown of the cells to the permissive temperature in the presence of cycloheximide, indicating that the temperature-sensitive proteins involved in DNA maturation became functional after shiftdown. Furthermore, because cycloheximide reduces maturation of DNA in wild-type-infected cells but not in cells infected with these two mutants, we conclude that a protein(s) necessary for the maturation of concatemeric DNA, which is present in limiting amounts during the normal course of infection, accumulated in the mutant-infected cells at the nonpermissive temperature. Concomitant with cleavage of concatemeric DNA, full capsids with electron-dense cores appeared after shiftdown of tsN-infected cells to the permissive temperature, indicating that there may be a correlation between maturation of DNA and formation of full capsids. The number of empty and full capsids (containing electron-dense cores) present in tsN-infected cells incubated at the nonpermissive temperature, as well as after shiftdown to the permissive temperature in the presence of cycloheximide, was determined by electron microscopy and by sedimentation analysis in sucrose gradients. After shiftdown to the permissive temperature in the presence of cycloheximide, the number of empty capsids present in tsN-infected cells decreased with a concomitant accumulation of full capsids, indicating that empty capsids are precursors to full capsids.

Intracellular forms of pseudorabies virus parental DNA were examined before and after the onset of viral DNA synthesis. Before initiation of synthesis, parental viral DNA acquires single-stranded ends. Circular and concatemeric molecules are also observed, indicating that the single-stranded ends are complementary. Viral DNA replication is initiated at an internal site within the DNA molecule, giving rise to characteristic replicative loops with single-stranded regions in the trans position. Such replicative loops were seen in unit-size (and smaller than unit-size) linear molecules as well as in circular and concatemeric molecules. These results show that the parental viral DNA molecules that acquire single-stranded ends, and consequently are able to form circles and concatemers, proceed to replicate.

The genome of pseudorabies virus consists of two components, short (S) and long (L). Only the S component is bracketed by inverted repeats, and only the S component inverts itself relative to the L component, giving rise to two isomeric forms of the genome. An attenuated vaccine strain of pseudorabies virus (Norden), however, has a genome which is found in four isomeric forms (B. Lomniczi, M. L. Blankenship, and T. Ben-Porat, J. Virol. 49:970-979, 1984). To determine the basis for the atypical structure of the genome of the Norden strain, we examined more than 40 field isolates of pseudorabies virus; all contained genomes in which the L component was fixed in only one orientation relative to the S component. Several independently generated vaccine strains which have been passaged extensively in chicken embryos and chicken embryo fibroblast (CEF) cell cultures were also analyzed; they possessed an invertible L component. Furthermore, emergence of pseudorabies virus variants with an invertible L component was observed after passage of the virus in CEF, but not in rabbit kidney or pig kidney, cells. The invertibility of the L component was associated consistently with a translocation of sequences from the left end of the genome to a position next to the inverted repeat sequence of the S component. Three observations indicate that genomes with an invertible L component (and the translocation) have a selective growth advantage over standard pseudorabies virus when grown in CEF. The proportion of virions with such genomes does not increase linearly as would be expected if the translocation events occurred repeatedly, most genomes eventually experiencing the translocation. Instead, after a lag, the proportion of such virions in the population increases relatively rapidly. The genome structures that are generated upon independent passage in CEF of each virion population were relatively homogeneous. Some heterogeneity was observed at relatively early stages of the emergence of the genomes carrying the translocation; at later stages, virions with genomes with a specific size translocation predominated in the virus population. Parallel passages in CEF of the same pseudorabies virus strain resulted in the emergence of populations of virions with genomes with different size translocations. However, in each of the passaged populations of virions the majority of virions had genomes with the same size translocation. The most likely interpretation of these results is that virions with genomes carrying the translocations that emerge upon passage of the virus in CEF have a selective advantage when grown in these cells.

As part of a study designed to identify the genes responsible for the virulence of pseudorabies virus, we have mapped the genomes of two independently derived vaccine strains (Bartha and Norden) by restriction enzyme analysis. The structures of these genomes have been compared with that of the genome of a laboratory strain previously mapped, of restriction fragments which had been cloned. The genome of the Bartha strain was found to be very similar to that of other pseudorabies virus strains, except that a deletion of approximately 2.7 X 10(6) daltons was found in the unique short (US) region. This deletion was also observed in the genome of the Norden vaccine strain but was not observed in the genomes of any other pseudorabies virus strains that have been studied (more than 20). The genome of the Norden strain differs from that of other pseudorabies virus strains in several other respects as well. The most important difference is that in contrast to all other pseudorabies virus strains analyzed to date, which contain a type 2 herpesvirus DNA molecule (in which the US region only inverts itself relative to the unique long [UL] region), the genome of the Norden strain is a type 3 molecule in which both the US and the UL regions of the genome invert themselves, giving rise to four isomeric forms of the genome. The ability of the UL region to invert itself is probably related to the fact that a sequence normally present in all other pseudorabies virus strains at the end of the UL only is found also in inverted form at the junction of the UL and the internal inverted repeat in the Norden strain.

During the first round of the replication of pseudorabies virus DNA, replicating DNA is mainly in the form of circles. The main origin of replication is located inthe region of the molecule bearing the inverted repeat. Replication proceeds unidirectionally from the origin.

Adsorption of mutants of pseudorabies virus (PrV) lacking glycoprotein gIII is slower and less efficient than is that of wild-type virus (C. Schreurs, T. C. Mettenleiter, F. Zuckermann, N. Snugg, and T. Ben-Porat, J. Virol. 62:2251-2257, 1988). To ascertain the functions of gIII in the early interactions of PrV with its host cells, we compared the effect on wild-type virus and gIII- mutants of antibodies specific for various PrV proteins. Although adsorption of wild-type virus was inhibited by polyvalent antisera against PrV as well as by sera against gIII and gp50 (but not sera against gII), adsorption of the gIII- mutants was not inhibited by any of these antisera. These results suggest that, in contrast to adsorption of wild-type PrV, the initial interactions of the gIII- mutants with their host cells are not mediated by specific viral proteins. Furthermore, competition experiments showed that wild-type Prv and the gIII- mutants do not compete for attachment to the same cellular components. These findings show that the initial attachment of PrV to its host cells can occur by a least two different modes--one mediated by glycoprotein gIII and the other unspecific. gIII- mutants not only did not adsorb as readily to cells as did wild-type virus but also did not penetrate cells as rapidly as did wild-type virus after having adsorbed. Antibodies against gIII did not inhibit the penetration of adsorbed virus (wild type or gIII-), whereas antibodies against gII and gp50 did. It is unlikely, therefore, that gIII functions directly in virus penetration. Our results support the premises that efficient adsorption of PrV to host cell components is mediated either directly or indirectly by gIII (or a complex of viral proteins for which the presence of gIII is functionally essential) and that this pathway of adsorption promotes the interactions of other viral membrane proteins with the appropriate cellular proteins, leading to the rapid penetration of the virus into the cells. The slower penetration of the gIII- mutants than of wild-type PrV appears to be related to the slower and less efficient alternative mode of adsorption of PrV that occurs in the absence of glycoprotein gIII.

The genome of pseudorabies virus (PrV) consists of two components--a noninvertible long (L) and an invertible short (S) component. The S component is bracketed by inverted repeats. In some variant strains of PrV (which have a selective growth advantage in certain cell lines), a sequence normally present at the left end of the L component has been translocated to the right end of the L component next to the inverted repeat. Consequently, these strains have acquired a genome with an L component that is bracketed by inverted repeats and that also inverts. We have determined the restriction maps and have analyzed the nucleotide sequences of those parts of the genome of three strains with invertible L components that contain the translocated segment of DNA. The results were as follows. The translocated fragments were derived in all cases from the extreme left end of the L component only. The sizes of the translocated fragments were similar, ranging from 1.3 to 1.4 kilobase pairs. The junction between the L and S components in these strains was the same as that in standard viral concatemeric DNA. The translocation of sequences from the left end of the genome next to the inverted repeats was always accompanied by a deletion of sequences from the right end of the L component. The sizes of the deleted fragments varied considerably, ranging from 0.8 to 2.3 kilobase pairs. Sequence homology at the points of recombination, i.e., at the junction between the right end and the left end of the L component, existed sometimes but not always. A model depicting how a virus with a class 2 genome (such as PrV) may acquire a genome with characteristics of a class 3 genome (such as herpes simplex virus) is presented.

Several attenuated strains of pseudorabies virus contain genomes that carry a deletion in their short unique (Us) component. The sizes of the deletions are different in the various attenuated strains; the deletions may include part of one of the inverted repeats as well as part of the Us region of the genome. In most cases, the deletion includes the gene encoding the glycoprotein gI. The attenuated strains with a deletion in their S component have a common history of having been cultivated in chicken embryo fibroblasts (CEF). We show here that passage of wild-type virus in CEF promotes the emergence of populations of virions with a deletion in their S component. The emergence of these mutants is the result of their growth advantage over the wild type and is related to the lack of expression of gI, as shown by the following. (i) The Norden strain (which has a deletion in the Us) was marker rescued to restore an intact Us. The nonrescued Norden strain had a growth advantage over the rescued Norden strain in CEF. (ii) Passage of wild-type (gI+) virus in CEF but not in rabbit kidney or pig kidney cells resulted invariably in the emergence of virions whose genomes had a deletion in the S component. (iii) Passage of a gI- mutant in CEF did not result in the emergence of such virions. The emergence of virions with a deletion in their S component thus appears to be linked to gI expression. We conclude that gI is deleterious to the growth of pseudorabies virus in CEF and that this effect is cell type specific.

The Bartha vaccine strain of pseudorabies virus has a deletion in the short unique (Us) region of its genome which includes the genes that code for glycoproteins gI and gp63 (E. Petrovskis, J. G. Timmins, T. M. Gierman, and L. E. Post, J. Virol. 60:1166-1169, 1986). Restoration of an intact Us to the Bartha strain enhances its ability to be released from infected rabbit kidney cells and increases the size of the plaques formed on these cells (T. Ben-Porat, J. M. DeMarchi, J. Pendrys, R. A. Veach, and A. S. Kaplan, J. Virol. 57:191-196, 1986). To determine which gene function plays a role in virus release from rabbit kidney cells, deletions were introduced into the genomes of both wild-type virus and the "rescued" Bartha strain (Bartha strain to which an intact Us had been restored) that abolish the expression of either the gI gene alone or both gI and gp63 genes. The effect of these deletions on the phenotype of the viruses was studied. Deletion mutants of wild-type virus defective in either gI or gI and gp63 behave like wild-type virus with respect to virus release and plaque size on rabbit kidney cells. Deletion of gI from the rescued Bartha strain, however, strongly affects virus release and causes a decrease in plaque size. We conclude that gI affects virus release but that at least one other viral function also affects this process. This function is defective in the Bartha strain but not in wild-type virus; in its absence gI is essential to efficient release of the virus from rabbit kidney cells.

Two pseudorabies virus vaccine strains (Bartha and Norden) that have a similar deletion in the short unique (Us) region of the genome have been identified previously (B. Lomniczi, M. L. Blankenship, and T. Ben-Porat, J. Virol. 49:970-979, 1984). These strains do not code for the glycoprotein gI, a glycoprotein that has been mapped on the wild type virus genome by T. C. Mettenleiter, N. Lukacs, and H. J. Rziha (J. Virol. 53:52-57, 1985) to the sequences deleted from the vaccine strain. Restoration of these deleted sequences to the Bartha strain genome restores to the virus the ability to specify the gI glycoprotein. The Bartha vaccine strain grows as well as wild-type virus in pig kidney and in rabbit kidney (RK) cells, but is not released efficiently from and forms small plaques in RK cells. The rescued Bartha 43/25a strain (which has an intact Us) is released considerably more efficiently than the Bartha vaccine strain, but less efficiently than wild-type virus from RK cells; it also forms larger plaques on RK cells than does the parental Bartha vaccine strain. The Norden vaccine strain, which has a deletion in the Us, is released better from RK cells than is the Bartha strain, but not as well as is wild-type virus. We conclude that whereas the sequences in the Us that are deleted from the Bartha and Norden strain genomes specify functions that play a role in the release of virions from some cell types, at least one other function (which is defective in the Bartha strain but not in the Norden strain) also affects release of virus from these cells. Since restoration to the Bartha strain of an intact Us restores to the virus both the ability to grow in chicken brains (B. Lomniczi, S. Watanabe, T. Ben-Porat, and A. S. Kaplan, J. Virol. 52:198-205, 1984) and to be released from RK cells, the possibility that the lack of virulence of the Bartha vaccine strain may be related to its limited release from some target cells is discussed.

To ascertain the biological functions of different glycoproteins that are nonessential for pseudorabies virus growth in vitro, we have constructed mutants defective in one (or a combination) of these glycoproteins and have examined various aspects of their role in the infective process. We made the following two observations. (i) Glycoproteins gI and gp63 are noncovalently complexed to each other. They are coprecipitated by antisera against either one of these glycoproteins but do not share antigenic determinants: monoclonal antibodies against gp63 do not immunoprecipitate gI from extracts of gp63- mutant-infected cells, and monoclonal antibodies against gI do not immunoprecipitate gp63 from extracts of gI- mutant-infected cells. (ii) Mutants unable to synthesize either gI or gp63 have some common biological characteristics; they have a growth advantage in primary chicken embryo fibroblasts. Furthermore, we have shown previously that in conjunction with glycoprotein gIII, gI and gp63 are necessary for the expression of virulence (T. C. Mettenleiter, C. Schreurs, F. Zuckermann, T. Ben-Porat, and A. S. Kaplan, J. Virol. 62, 2712-2717, 1988). These results show that the functional entity affecting virus replication in chicken embryo fibroblasts, as well as affecting virulence, is the complex between gI and gp63. The gI-gp63 complex of pseudorabies virus does not appear to have Fc receptor activity as does its homolog, the gI-gE complex of herpes simplex virus.

Deletion mutants of pseudorabies virus unable to express glycoprotein gIII, gI, or gp63 or double and triple mutants defective in these glycoproteins were constructed, and their virulence for day-old chickens inoculated intracerebrally was determined. Mutants of wild-type pseudorabies virus defective in glycoprotein gIII, gI, or gp63 were only slightly less virulent (at most, fivefold) for chickens than was the wild-type virus. However, mutants defective in both gIII and gI or gIII and gp63 were avirulent for chickens, despite their ability to grow in cell culture in vitro to about the same extent as mutants defective in gIII alone (which were virulent). These results show that gIII plays a role in virulence and does so in conjunction with gI or gp63. The effect of gIII on virulence was also shown when the resident gIII gene of variants of the Bartha vaccine strain (which codes for gIIIB) was replaced with a gIII gene derived from a virulent wild-type strain (which codes for gIIIKa); gIIIKa significantly enhanced the virulence of a variant of the Bartha strain to which partial virulence had been previously restored by marker rescue. Our results show that viral functions that play a role in the virulence of the virus (as measured by intracerebral inoculation of chickens) may act synergistically to affect the expression of virulence and that the ability of the virus to grow in cell culture is not necessarily correlated with virulence.

One of the major glycoproteins of pseudorabies virus, gIII, is nonessential for growth in cell culture. Mutants defective in gIII, however, consistently yield lower titers of infectious virus (3- to 20-fold) than does wild-type virus. The interactions of gIII- mutants with their host cells were compared with those of wild-type virus in an attempt to uncover the functions of gIII. We show that gIII plays a major role in the stable adsorption of the virus to its host cell; in the absence of gIII, the rate of adsorption is reduced and adsorption is easily reversed by washing. Thus, adsorption of pseudorabies virus can be said to occur in at least the following two ways: (i) a gIII-mediated rapid adsorption or (ii) a slower and more labile adsorption that is independent of gIII. After virions have been complexed with monoclonal antibodies against gIII (but not some monoclonal antibodies against other glycoproteins), both modes of adsorption were inhibited. Glycoprotein gIII affects virus stability and virus release, as well as adsorption. The effect on virus release is marked when the virus is defective in additional functions. Thus, although we found no obvious difference in the release of virus from gIII- or wild-type virus-infected rabbit kidney cells, release of a gIII-/gI- double mutant from the cells occurred less readily than did release of a gI- mutant. The gIII-/gI- and gIII- mutants, however, adsorbed to cells at a similar rate, indicating that the effects of gIII on adsorption and virus release constitute separate functions. The Bartha vaccine strain of pseudorabies virus has a defective gIII gene and is released poorly from rabbit kidney cells. After the resident Bartha gIII gene was replaced by the gIII gene of wild-type virus, virus release was enhanced considerably. Since inactivation of gIII in wild-type pseudorabies virus did not significantly affect virus release, the Bartha strain must be defective in another function which, in conjunction with gIII, significantly affects virus release. These results indicate again that gIII affects virus release in conjunction with other functions. Also, although the Bartha strain was functionally defective in virus release, it adsorbed to cells as well as wild-type virus did, showing that the effects of gIII on virus adsorption and release constitute separate functions. We conclude that gIII is a multifunctional glycoprotein.

We have shown previously (Lomniczi et al., J. Virol. 52:198-205, 1984) that the Bartha vaccine strain of pseudorabies virus has a deletion in the short unique (Us) region of its genome--a deletion that is related to the absence of virus virulence. This strain is, however, also defective in other genes involved in virulence. We show here that virulence can be restored by marker rescue of the Bartha strain to which an intact Us has been restored (but not to the parental Bartha strain) by sequences derived from approximate map units 0.460 and 0.505 of the wild-type virus genome. No difference in the ability to grow in cell culture was observed between parental Bartha, Bartha 43/25a (Bartha to which an intact Us has been restored), or the doubly rescued Bartha strains. However, only the doubly rescued Bartha strain was virulent for both chickens and pigs and replicated to high titers when inoculated directly into the brains of chickens. The sequences that could restore virulence to the Bartha 43/25a strain encode four genes, all of which are involved in processes leading to the assembly of nucleocapsids. Since these sequences rescue virulence, it appears that a function that plays a role in nucleocapsid assembly is defective in the Bartha strain and that this defect contributes to the lack of virulence of this virus.

Serial passage at high multiplicity of pseudorabies virus generates defective interfering particles (DIPs) whose genomes consist at least in part of reiterations of segments of DNA in which sequences originating from different regions of the genome have become covalently linked (F. J. Rixon and T. Ben-Porat, Virology 97:151-163). To determine whether some cis functions present in these reiterated DNA sequences may be responsible for the amplification of DIP DNA, BamHI restriction fragments of this DNA were cloned. These fragments were analyzed and tested for their ability to promote the amplification of covalently linked pBR325 DNA when cotransfected into cells with helper pseudorabies virus DNA. The cloned DIP BamHI DNA fragments consisted of various combinations of sequences originating from either one or both ends as well as sequences from the middle of the unique long (UL) segment of the genome. Only plasmids with inserts consisting of segments of defective DNA originating from the middle of the UL, as well as from both ends of the genome, were able to replicate and be encapsidated autonomously. This finding indicated that signals present at both ends of the genome may be necessary for efficient cleavage-encapsidation. To confirm this observation, we constructed plasmids in which DNA segments containing an origin of replication and sequences from either one or both ends of the virus genome were linked. These experiments showed that efficient cleavage-encapsidation requires the presence of sequences derived from both ends of the genome. Two origins of replication, one at the end of the UL segment and one in the middle of the UL segment, were also identified.

Serial passage of pseudorabies virus (PrV) at high multiplicity yields defective interfering particles (DIPs), but the sharp cyclical increases and decreases in titer of infectious virus that are observed upon continued passage at high multiplicity of most DIPs of other viruses are not observed with DIPs of PrV (T. Ben-Porat and A. S. Kaplan, Virology 72:471-479). We have studied the dynamics of the interactions of the virions present in a population of DIPs to assess the cis functions for which the genomes of the DIPs are enriched. The defective genomes present in one population of DIPs, [PrV(1)42], replicate preferentially over the nondefective genomes present in that virion population at early stages of infection, indicating that the DIP DNA is enriched for sequences that can serve as origins of replication at early stages of infection. This replicative advantage of the DIP DNA is transient and disappears at later stages of infection. The defective DNA does not appear to be encapsidated preferentially over the nondefective DNA present in this virion population, which might indicate that it is not enriched for cleavage-encapsidation sites. However, the nondefective DNA in the DIP virion population has become modified and has acquired reiterations of sequences originating from the end of the unique long (UL) region of the genome. Furthermore, both the infectious and defective genomes present in the DIP population compete for encapsidation more effectively than do the genomes of standard PrV. These results indicate that the defective genomes in the population of virions studied are enriched not only for an origin of replication but probably also for sequences necessary for efficient cleavage-encapsidation. Furthermore, the nondefective genomes present in this population of DIPs have also been modified and have acquired the ability to compete with the defective genomes for cleavage-encapsidation.

Previously we have reported that among the proteins of purified pseudorabies virions there are four major glycoproteins (T. Ben-Porat and A. S. Kaplan, Virology 41:265-273, 1970). Several minor glycoproteins can also be identified by two-dimensional gel electrophoresis. Removal of the viral envelope with Triton X-100 selectively removes from the virions all of the glycoproteins as well as several non-glycosylated proteins. Sedimentation analysis or chromatography of these proteins reveals that several are complexed with one another, some being covalently linked via disulfide bridges. Analysis of the proteins by immunoprecipitation with monoclonal antibodies reactive with the membrane proteins showed also that three of the four major virus glycoproteins (125K, 74K, and 58K; gIIa, gIIb, and gIIc, respectively) are linked covalently by disulfide bridges. Furthermore, all three share extensive sequence homology as indicated by the identity of their antigenic determinants and by partial peptide mapping; they probably originate from a single protein precursor. The fourth major glycoprotein (98K; gIII) is not complexed to any other protein. Three minor glycoproteins (130K [gI], 98K [gIV], and 62K [gV]), which form a noncovalently linked complex with a 115K nonglycosylated protein, have also been identified. Of the monoclonal antibodies used in this study, only those reactive with the major 98K glycoprotein (gIII) inhibit virus adsorption and neutralize virus infectivity in the absence of complement. However, all react with surface components of the virion, indicating that the proteins with which they react are exposed on the surface of the virions. A nomenclature for the pseudorabies virus glycoproteins is proposed.

Lomniczi et al. (J. Virol. 49:970-979, 1984) have shown previously that two attenuated vaccine strains of pseudorabies virus have a similar deletion in the short unique (US) region of the genome. The region which is deleted normally codes for several translationally competent mRNAs. As expected, these mRNAs are not formed in the cells infected with the vaccine strains. The function specified by these mRNAs is thus not necessary for growth in cell culture. Using intracerebral inoculation of 1-day-old chicks as a test system, we have attempted to determine whether a gene within the region that is missing from the attenuated strains specifies functions that are required for the expression of virulence. An analysis of recombinants between the Bartha vaccine strain and a virulent pseudorabies virus strain (having or lacking a thymidine kinase gene [TK+ or TK-]) revealed the following. None of the recombinant plaque isolates that were either TK- or which had a deletion in the US was virulent. Not all recombinant plaque isolates which were both TK+ and had an intact US were virulent. These results indicate that both thymidine kinase activity and an intact US were necessary but not sufficient for the expression of virulence. Marker rescue experiments involving cotransfection of the Bartha strain DNA and a restriction fragment spanning the region of the genome that was missing from the Bartha strain resulted in the isolation of virions to which an intact US had been restored. These virions were not virulent but had an improved ability to replicate in the brains of chicks compared with that of the parental nonrescued Bartha strain. Our results show that genes in the US region, which are missing from the Bartha strain, were necessary for virulence but that this strain was also defective in other genes required for the expression of virulence. Thus, the virulence of pseudorabies virus, as measured by intracerebral inoculation into chicks, appears to be controlled multigenically.

The rate of pseudorabies virus DNA elongation was measured by three different techniques: density shift experiments, radioautography examined by light microscopy, and radioautography examined by electron microscopy. The rate of the fork movement at 37 degrees C was estimated to be approximately 1 micron/min.