Spontaneous curing of a minute virus of mice carrier state by selection of cells with an intracellular block of viral replication.

We previously described a persistent infection established by the lymphotropic minute virus of mice in mouse L cells at the level of the cell population (D. Ron, P. Tattersall, and J. Tal, J. Virol. 52:63-69, 1984). This carrier state is maintained by a series of consecutive phenotypic changes which take place in both the cells and the virus and is cured spontaneously after 150 to 200 cell generations (D. Ron and J. Tal, J. Virol. 55:424-430, 1985). We show here that the cure was caused by the selection of virus-resistant cells in the culture. The resistance of these survivor cells to virus replication was due to an intracellular block. Infection of a spontaneously cured culture with the fibrotropic parental minute virus of mice resulted in a restrictive infection in which the viral replicative-form DNA was formed and amplified, but the synthesis of single-stranded progeny DNA was markedly reduced. The lymphotropic strain was blocked in these cells at an earlier stage, with little or no amplification of viral replicative-form DNA observed. These data indicate that the replication of minute virus of mice requires host-coded helper functions in at least two stages of its growth cycle.     

Formation of a host range mutant of the lymphotropic strain of minute virus of mice during persistent infection in mouse L cells

Minute virus of mice (i), the lymphotropic strain of minute virus of mice, established a persistent infection in normally restrictive L cells. The carrier state, which lasted 150 days, exhibited three clearly distinguishable stages. During the early stage (days 1 to 10 postinfection), small amounts of virus were formed. A "crisis" then developed that lasted 50 to 60 days and was characterized by massive cell lysis and high titers of virus. This was followed by a 70- to 80-day period in which small but stable quantities of virus were produced. Virus shed by the carrier culture during the latter phase had acquired an altered host range, namely, it had lost its ability to replicate in T-lymphocyte cell lines and had adapted to growth in L cells. Virus isolated at this time from a single plaque in L cells, designated hr301, was shown to possess similar host range properties. No differences, however, could be found between the DNAs of minute virus of mice (i) and of hr301 by restriction enzyme analysis, suggesting that the mutation that affected the viral host range did not involve an extensive region of the viral genome.     

Restricted viral RNA synthesis in establishment of persistent infection in Vero cells with a Sendai virus mutant

It was previously shown that a temperature-sensitive mutant of Sendai virus, ts-23, readily establishes persistent infection in Vero cells at 37 C, a permissive temperature for growth of the mutant. In the present study, it was demonstrated that the virus yield from ts-23-infected Vero cells at 37 C began to decrease 48 to 72 hr postinfection, after an initial phase of high virus production. Before the decrease in virus production, the formation of viral nucleoprotein declined, although synthesis of all species of viral protein continued. It was suggested that the limited formation of viral nucleoprotein and the decrease in virus production were due to the restriction of viral RNA synthesis which began to occur early after infection in ts-23-infected cells at 37 C. The mutant has a temperature-sensitive defect in RNA polymerase activity and the temperature 37 C, used for establishment of persistent infection, would be a semi-permissive temperature for the RNA polymerase activity of the mutant. The ts-23 mutant interfered with the replication of the parental wild virus in Vero cells at 37 C.     

Deletion of the varicella-zoster virus large subunit of ribonucleotide reductase impairs growth of virus in vitro.

Cells infected with varicella-zoster virus (VZV) express a viral ribonucleotide reductase which is distinct from that present in uninfected cells. VZV open reading frames 18 and 19 (ORF18 and ORF19) are homologous to the herpes simplex virus type 1 genes encoding the small and large subunits of ribonucleotide reductase, respectively. We generated recombinant VZV by transfecting cultured cells with four overlapping cosmid DNAs. To construct a virus lacking ribonucleotide reductase, we deleted 97% of VZV ORF19 from one of the cosmids. Transfection of this cosmid with the other parental cosmids yielded a VZV mutant with a 2.3-kbp deletion confirmed by Southern blot analysis. Virus-specific ribonucleotide reductase activity was not detected in cells infected with VZV lacking ORF19. Infection of melanoma cells with ORF19-deleted VZV resulted in plaques smaller than those produced by infection with the parental VZV. The mutant virus also exhibited a growth rate slightly slower than that of the parental virus. Chemical inhibition of the VZV ribonucleotide reductase has been shown to potentiate the anti-VZV activity of acyclovir. Similarly, the concentration of acyclovir required to inhibit plaque formation by 50% was threefold lower for the VZV ribonucleotide reductase deletion mutants than for parental virus. We conclude that the VZV ribonucleotide reductase large subunit is not essential for virus infection in vitro; however, deletion of the gene impairs the growth of VZV in cell culture and renders the virus more susceptible to inhibition by acyclovir.     

Two amino acid substitutions within the capsid are coordinately required for acquisition of fibrotropism by the lymphotropic strain of minute virus of mice.

Nucleotide changes at both codons 317 and 321 in the VP2 capsid gene of the immunosuppressive strain of the murine parvovirus minute virus of mice, MVM(i), are required to create a virus capable of growing in A9 fibroblasts. This double mutant virus, ILB1, has growth characteristics very similar to those of the prototype fibrotropic strain MVM(p) in both single- and multiple-round infections of fibroblasts and is about 100-fold better at infecting fibroblasts than MVM(i). When only one nucleotide position is changed, either in codon 317 (as in ILB2) or in codon 321 (as in ILB3), the resulting viruses are less than twice as efficient as their parent MVM(i) at infecting fibroblasts. In the restrictive infection of A9 cells by the single mutants and MVM(i), gene expression and DNA replication were markedly reduced compared with ILB1 infection of the same cells or compared with infections of permissive hybrid cells by each of the viruses. This suggests that restriction acts predominantly at an early step in the infection. Since the phenotypes of ILB2 and ILB3 are essentially indistinguishable in restrictive infections, it is most likely that the individual loci affect the same step in the viral life cycle. The dramatic increase in fibroblast infectivity shown by ILB1 indicates a synergistic interaction between these two amino acid residues in the same rate-limiting process in fibroblast infection.     

A double deletion prevents replication of the pestivirus bovine viral diarrhea virus in the placenta of pregnant heifers

In contrast to wild type bovine viral diarhea virus (BVDV) specific double deletion mutants are not able to establish persistent infection upon infection of a pregnant heifer. Our data shows that this finding results from a defect in transfer of the virus from the mother animal to the fetus. Pregnant heifers were inoculated with such a double deletion mutant or the parental wild type virus and slaughtered pairwise on days 6, 9, 10 and 13 post infection. Viral RNA was detected via qRT-PCR and RNAscope analyses in maternal tissues for both viruses from day 6 p.i. on. However, the double deletion mutant was not detected in placenta and was only found in samples from animals infected with the wild type virus. Similarly, high levels of wild type viral RNA were present in fetal tissues whereas the genome of the double deletion mutant was not detected supporting the hypothesis of a specific inhibition of mutant virus replication in the placenta. We compared the induction of gene expression upon infection of placenta derived cell lines with wild type and mutant virus via gene array analysis. Genes important for the innate immune response were strongly upregulated by the mutant virus compared to the wild type in caruncle epithelial cells that establish the cell layer on the maternal side at the maternal–fetal interface in the placenta. Also, trophoblasts which can be found on the fetal side of the interface showed significant induction of gene expression upon infection with the mutant virus although with lower complexity. Growth curves recorded in both cell lines revealed a general reduction of virus replication in caruncular epithelial cells compared to the trophoblasts. Compared to the wild type virus this effect was dramtic for the mutant virus that reached only a TCID₅₀ of 1.0 at 72 hours post infection.     

Initiation and maintenance of persistent infection by respiratory syncytial virus.

Propagation of cells infected with temperature-sensitive (ts) mutants of respiratory syncytial (RS) virus at nonpermissive temperature (39 degrees C) resulted in cytolytic, abortive, or persistent infection, depending on the mutant used to initiate infection. Five mutants from complementation group B produced cytolytic or abortive infections, whereas a single mutant (ts1) from group D and a noncomplbmenting mutant produced persistent infections. The persistently infected culture initiated by mutant ts1 (RS ts1/BS-C-1) has been maintained in serial culture for greater than 100 transfers, and infectious-center assays and immunofluorescent staining indicated that all cells harbored the RS virus genome. RS ts1/BS-C-1 cultures were resistant to superinfection by homologous and some heterologous viruses, and interferon-like activity against some heterologous viruses was present in the culture medium. Small amounts (0.002 to 0.2 PFU/cell) of infectious virus were present in the culture fluid, but autointerfering defective particles were not detected. This released virus formed small plaques and produced persistent infection of BS-C-1 cells at 37 degrees C. The RS ts1/BS-C-1 cells contained abundant RS virus antigen internally, but little at the surface, although the cells showed enhanced agglutinability by concanavalin A. Nucleocapsids and the 41,000-molecular-weight nucleoprotein were present in extracts of both nucleated and enucleated cells. No infectious RS virus was obtained by transfection of DNA from RS tsl/BS-C-1 cells to susceptible BS-C-1 or feline embryo cells under conditions allowing efficient transfection of a foamy virus proviral DNA. It was concluded that persistent infection was maintained in part by a non-ts variant of RS virus partially defective in maturation. The karyotype of the RS ts1/BS-C-1 culture differed from that of unifected cells.     

Interaction of minute virus of mice with differentiated cells: strain-dependent target cell specificity is mediated by intracellular factors

The prototype strain of minute virus of mice and the immunosuppressive strain are unable to grow lytically in each other's murine host cell type. To characterize these strain-dependent virus-host cell interactions further, we have compared the early events of both productive and restrictive infections. Each virus binds to specific receptors on the surface of both productive and restrictive cell types. Competition experiments show that both viruses recognize the same receptor on each cell type. Penetration and uncoating are presumed to be similar in both productive and restrictive infections, since incoming viral genomes are converted to parental replicative form DNA independent of the final outcome of the virus-host cell interaction. In contrast to the majority of other systems studied to date, these differences in minute virus of mice target cell specificity are not mediated at the cell surface, but by the interaction of a strain-specific viral determinant with intracellular host factors that are expressed in particular cell types as a function of differentiation. These cellular factors catalyze a step in viral replication which occurs after the initiation of viral DNA synthesis, but before the detectable expression of the viral capsid polypeptide genes.     

Absence or overexpression of the Varicella-Zoster Virus (VZV) ORF29 latency-associated protein impairs late gene expression and reduces VZV latency in a rodent model

Varicella-zoster virus (VZV) ORF29 encodes the viral single-stranded DNA binding protein and is expressed during latency in human ganglia. We constructed an ORF29 deletion mutant virus and showed that the virus could replicate only in cells expressing ORF29. An ORF29-repaired virus, in which ORF29 was driven by a cytomegalovirus promoter, grew to peak titers similar to those seen with the parental virus. The level of ORF29 protein in cells infected with the repaired virus was greater than that seen with parental virus. Infection of cells with either the ORF29 deletion or repaired virus resulted in similar levels of VZV immediate-early proteins but reduced levels of glycoprotein E compared to those observed with parental virus. Cotton rats infected with the ORF29 deletion mutant had a markedly reduced frequency of latent infection in dorsal root ganglia compared with those infected with parental virus (P < 0.00001). In contrast, infection of animals with the ORF29 deletion mutant resulted in a frequency of ganglionic infection at 3 days similar to that seen with the parental virus. Animals infected with the ORF29-repaired virus, which overexpresses ORF29, also had a reduced frequency of latent infection compared with those infected with parental virus (P = 0.0044). These studies indicate that regulation of ORF29 at appropriate levels is critical for VZV latency in a rodent model.     

Immunization with replication-defective mutants of herpes simplex virus type 1: sites of immune intervention in pathogenesis of challenge virus infection.

Replication-defective mutants of herpes simplex virus type 1 (HSV-1) were used as a new means to immunize mice against HSV-1-mediated ocular infection and disease. The effects of the induced immune responses on pathogenesis of acute and latent infection by challenge virus were investigated after corneal inoculation of immunized mice with virulent HSV-1. A single subcutaneous injection of replication-defective mutant virus protected mice against development of encephalitis and keratitis. Replication of the challenge virus at the initial site of infection was lower in mice immunized with attenuated, wild-type parental virus (KOS1.1) or replication-defective mutant virus than in mice immunized with uninfected cell extract or UV-inactivated wild-type virus. Significantly, latent infection in the trigeminal ganglia was reduced in mice given one immunization with replication-defective mutant virus and was completely prevented by two immunizations. Acute replication in the trigeminal ganglia was also prevented in mice immunized twice with wild-type or mutant virus. The level of protection against infection and disease generated by immunization with replication-defective mutant viruses was comparable to that of infectious wild-type virus in all cases. In addition, T-cell proliferative and neutralizing antibody responses following immunization and corneal challenge were of similar strength in mice immunized with replication-defective mutant viruses or with wild-type virus. Thus, protein expression by forms of HSV-1 capable of only partially completing the replication cycle can induce an immune response in mice that efficiently decreases primary replication of virulent challenge virus, interferes with acute and latent infection of the nervous system, and inhibits the development of both keratitis and systemic neurologic disease.     

Neuroblastoma cell-adapted yellow fever 17D virus: characterization of a viral variant associated with persistent infection and decreased virus spread

Serial passage of yellow fever 17D virus (YF5.2iv, derived from an infectious molecular clone) on mouse neuroblastoma (NB41A3) cells established a persistent noncytopathic infection associated with a variant virus. This virus (NB15a) was dramatically reduced in plaque formation and exhibited impaired replication kinetics on all cell lines examined compared to the parental virus. Nucleotide sequence analysis of NB15a revealed a substitution in domain III of the envelope (E) protein at residue 360, where an aspartic acid residue was replaced by glycine. Single mutations were also found within the NS2A and NS3 proteins. Engineering of YF5.2iv virus to contain the E(360) substitution yielded a virus (G360 mutant) whose plaque size and growth efficiency in cell culture resembled those of NB15a. Compared with YF5.2iv, both NB15a and G360 were markedly restricted for spread through Vero cell monolayers and mildly restricted in C6/36 cells. On NB41A3 cells, spread of the viruses was similar, but all three were generally inefficient compared with spread in other cell lines. Compared to YF5.2iv virus, NB15a was uniformly impaired in its ability to penetrate different cell lines, but a difference in cell surface binding was detected only on NB41A3 cells, where NB15a appeared less efficient. Despite its small plaque size, impaired growth, and decreased penetration efficiency, NB15a did not differ from YF5.2iv in mouse neurovirulence testing, based on mortality rates and average survival times after intracerebral inoculation of young adult mice. The data indicate that persistence of yellow fever virus in NB41A3 cells is associated with a mutation in the receptor binding domain of the E protein that impairs the virus entry process in cell culture. However, the phenotypic changes which occur in the virus as a result of the persistent infection in vitro do not correlate with attenuation during pathogenesis in the mouse central nervous system.     

Interaction between parvovirus NS2 protein and nuclear export factor Crm1 is important for viral egress from the nucleus of murine cells

A mutation that disrupts the interaction between the NS2 protein of minute virus of mice and the nuclear export factor Crm1 results in a block to egress of mutant-generated full virions from the nucleus of infected murine cells. These mutants produce wild-type levels of monomer and dimer replicative DNA forms but are impaired in their ability to generate progeny single-stranded DNA in restrictive murine cells in the first round of infection. The NS2-Crm1 interaction mutant can be distinguished phenotypically from an NS2-null mutant and reveals a role for the Crm1-mediated export pathway at a late step in viral infection.     

Demonstration of two distinct cytopathic effects with syncytium formation-defective human immunodeficiency virus type 1 mutants.

The mechanism of human immunodeficiency virus type 1 (HIV-1) cytopathicity is poorly understood and might involve formation of multinucleated giant cells (syncytia), single-cell lysis, or both. In order to determine the contributions of the fusion domain to syncytium formation, single-cell lysis, and viral infectivity and to clarify the molecular details of these events, insertion mutations were made in the portion of env encoding this sequence in the functional HIV-1 proviral clone HXB2. Viruses produced from these mutant clones were found to have a partial (F3) or complete (F6) loss of syncytium-forming ability in acutely infected CEM, Sup T1, and MT4 T-cell lines. During the early stage of acute infection by F6 virus, there was a loss of the syncytial cytopathic effect, which resulted in increased cell viability, and a 1.9- to 2.6-fold increase in virus yield in the cell lines tested. In the late stage of acute infection, the single-cell cytopathic effect of F6 virus was similar to that of the parental HXB2 virus. The F3 and F6 viruses were also found to have a 1.7- to 43-fold reduction in infectivity compared with the HXB2 virus. The mutant F3 and F6 and parental HXB2 envelope proteins were expressed in vaccinia virus, and the mutant envelope proteins were observed to be defective in their ability to form syncytia. BSC-40 cells infected with vaccinia virus recombinants revealed no differences in kinetics of cleavage, cell surface expression, or CD4 binding capacity of the mutant and parental envelope proteins. These results demonstrate that a loss of syncytium formation results in an attenuation of infectivity and a loss of the syncytial cytopathic effect without a loss of single-cell lysis. These mutants may reflect in tissue culture the changes observed in the HIV isolates in vivo during disease progression, which exhibit marked differences in syncytium production.     

Modulation of the type I interferon pathways by culture-adaptive hepatitis C virus core mutants

Hepatitis C virus (HCV) often establishes a persistent infection that leads to chronic liver diseases. The viral core protein modulates various cellular activities involved in this process. We found two mutations, K23E and V31A, in the core gene of the transfected HCV JFH-1 genome, which had been replicated for a prolonged period. The mutant viruses escaped immunochemical detection by a core-specific antibody and demonstrated enhanced RNA replication and protein expression, compared to the parental virus. The mutant core proteins bound less tightly than the parental type core to the DEAD-box RNA helicase DDX3 and attenuated the TBK1-mediated activation of interferon-related promoters. These results suggest a mechanism by which the viruses adapt to attenuate cellular antiviral activity and to establish persistent infection.     

In vivo selection of lymphocyte-tropic and macrophage-tropic variants of lymphocytic choriomeningitis virus during persistent infection.

This study demonstrates cell-specific selection of viral variants during persistent lymphocytic choriomeningitis virus infection in its natural host. We have analyzed viral isolates obtained from CD4+ T cells and macrophages of congenitally infected carrier mice and found that three types of variants are present in individual carrier mice: (i) macrophage-tropic, (ii) lymphotropic, and (iii) amphotropic. The majority of the isolates were amphotropic and exhibited enhanced growth in both lymphocytes and macrophages. However, some of the lymphocyte-derived isolates grew well in lymphocytes but poorly in macrophages, and a macrophage-derived isolate replicated well in macrophages but not in lymphocytes. In striking contrast, the original wild-type (wt) Armstrong strain of lymphocytic choriomeningitis virus that was used to initiate the chronic infection and from which the variants are derived grew poorly in both lymphocytes and macrophages. These three types of variants also differed from the parental virus in their ability to establish a chronic infection in immunocompetent hosts. Adult mice infected with the wt Armstrong strain cleared the infection within 2 weeks, whereas adult mice infected with the variants harbored virus for several months. These results suggest that the ability of the variants to persist in adult mice is due to enhanced replication in macrophages and/or lymphocytes. This conclusion is further strengthened by the finding that the variants and the parental wt virus grew equally well in mouse fibroblasts and that the observed growth differences were specific for cells of the immune system.     

NS2 is required for efficient translation of viral mRNA in minute virus of mice-infected murine cells.

Detailed analysis of five NS2 mutants of the autonomous parvovirus minute virus of mice (MVMp) has revealed the following. At low multiplicities of infection, NS2 mutants killed NB324K cells as well as wild-type (wt) MVM did and grew to high titers, while in contrast they grew poorly and did not readily kill murine A9 cells. Following CaPO4 transfection of murine fibroblasts, NS2 mutant infectious clones generated approximately 10-fold less monomer replicative-form DNA than wt and no detectable progeny single-stranded DNA. On nonmurine semipermissive NB324K cells, however, these mutant plasmid clones generated near wt levels of all replicative DNA forms. After infection of highly synchronized murine fibroblasts by NS2 mutant virus at inputs equivalent to those of the wt, mutant monomer replicative-form DNA was decreased 5- to 10-fold compared with that of the wt, and progeny single-stranded DNA accumulation was decreased to an even greater extent. Both total and cytoplasmic NS2 mutant RNA was decreased, but the amount of total viral mRNA generated, relative to accumulated viral DNA in the same experiments, was similar to that seen in wt infection. The accumulation of virus-generated proteins was also decreased in NS2 mutant infection; however, the magnitude of this decrease, compared with that of wt infections, was significantly greater than the concomitant decrease in mutant-generated levels of accumulated cytoplasmic RNA, and this effect was most dramatic for VP2. There was no such disparity between the relative accumulation of mutant-generated RNA and protein in cells permissive for the growth of these mutants. These results suggest that translation of MVM viral RNA is specifically reduced in NS2 mutant infection of restrictive cells. Because the affected viral proteins are required for the efficient production of viral replicative DNA forms, these results reveal a fundamental, although perhaps not the only, role for NS2 in parvovirus infection.     

Isolation of highly fusogenic variants of simian virus 5 from persistently infected cells that produce and respond to interferon.

A series of experiments were undertaken to examine how interferon and neutralizing antibodies influence the ability of simian virus 5 (SV5) (strain W3) to establish and maintain persistent infections in murine cells. In contrast to the rapid decline in SV5 protein synthesis observed in murine BALB/c fibroblasts (BF cells), which produce and respond to interferon, between 24 and 48 h postinfection there was no inhibition of virus protein synthesis in MSFI- cells, skin fibroblasts derived from alpha/beta-interferon receptor knockout BALB/c mice. Furthermore, the addition of anti-interferon antibodies to the culture medium of infected BF cells significantly reduced the observed decline in virus protein synthesis. Following infection of untreated BF cells, the majority replicated virus but survived the infection and eventually cleared the virus after 8 to 15 days. However, not all the cells were cured, and the cultures became persistently infected. Upon passage of persistently infected cultures, the virus fluxed between active and repressed states as a consequence of interferon production. This resulted in a balance being reached in which only 5 to 20% of the cells were infected at any one time. After 30 passages of the persistently infected cells, highly fusogenic virus variants arose (one of which was isolated and termed W3-f). W3-f remained as sensitive to interferon as the parental W3 isolate but, in the absence of interferon, spread much more rapidly than the parental W3 strain through BF cell monolayers. Sequence analysis revealed no deduced amino acid differences between the F proteins of W3 and W3-f. BF cell cultures persistently infected with W3-f were rapidly cleared of virus by the addition of virus-neutralizing antibodies to the culture medium. In contrast, neutralizing antibodies had little effect on the numbers of cells persistently infected with W3 over several passages. These results suggest that the ability of paramyxoviruses to cause cell-cell fusion may be selected for in vivo as a consequence of their adaptation to the interferon response rather than their need to escape from neutralizing antibodies. The significance of these observations with regard to persistent parainfluenza virus infections in vivo is further discussed.     

Role of the host cell in persistent viral infection: Coevolution of L cells and reovirus during persistent infection

Mutant L cells, designated LR cells, were isolated after “curing” a persistently infected cell line (L/C) with antireovirus serum. The LR cells were shown to be virus-free; no reovirus was detectable by infectious center assays, plaque assays, presence of viral proteins, presence of viral dsRNA and immunofluorescence studies. Persistent infections were readily established in LR cells following infection with either cloned, low passage wild-type reovirus or cloned, low passage reovirus isolated from carrier cultures. Reovirus isolated from carrier cultures, however, grew much better than wild-type reovirus in LR cells and showed complete dominance over wild-type reovirus in coinfection experiments. Infection of LR cells with wild-type reovirus resulted in a low-level persistent infection with inefficient viral replication; these mutant L cells were partially resistant to infection with wild-type reovirus. In contrast, infection of the mutant L cells with virus isolated from the persistently infected cells resulted in a persistent infection accompanied with efficient viral replication. Infection of the original L cells with either wild-type reovirus or reovirus isolated from the persistently infected cells resulted in a lytic infection with no surviving cells. Thus the host cell plays a crucial role in the maintenance of persistent reovirus infection. Our results show that there is a coevolution of both mutant L cells and mutant reovirus during persistent infection.