African green monkey kidney Vero cells require de novo protein synthesis for efficient herpes simplex virus 1-dependent apoptosis

During HSV-1 infection, IE gene expression triggers apoptosis, but subsequent synthesis of infected cell proteins blocks apoptotic death from ensuing. This "HSV-1-dependent" apoptosis was identified in HEp-2/HeLa cells infected with wild-type HSV-1 in the presence of an inhibitor of protein synthesis or a virus lacking ICP27 {HSV-1(vBSDelta27)}. Unlike HEp-2/HeLa cells, vBSDelta27-infected Vero cells fail to exhibit dramatic apoptotic morphologies at times prior to 24 hpi. Here, we examined the basis of these different apoptotic responses to HSV-1. We found that infected Vero cells take substantially longer than HEp-2/HeLa cells to display membrane blebbing, chromatin condensation, DNA laddering, and PARP cleavage. Vero, but not HEp-2/HeLa, cells required de novo protein synthesis to exhibit efficient HSV-1-dependent apoptosis, which included changes in mitochondrial membrane potential, and these factors were produced prior to 3 hpi. Vero cells infected with recombinant viruses devoid of the ICP27 and ICP4 proteins alone or both the ICP27 and ICP22 proteins were apoptotic. These results indicate a requirement for cellular or other viral protein synthesis in Vero cells and provide insight into cell type differences in HSV-1-dependent apoptosis.     

Some characteristics of an early protein (ICP 22) synthesized in cells infected with herpes simplex virus

In Vero cells incubated at 40 degrees C or treated with azetidine at 37 degrees C, synthesis of a polypeptide ('C') of apparent mol. wt. 66000 was stimulated. It was not phosphorylated and was found in the cytoplasmic fraction of cell lysates. In cells infected with herpes simplex virus type 1 (HSV-1) in the presence of azetidine, synthesis of cellular proteins, including polypeptide C, was suppressed and infected cell polypeptides ICP 4, 0, 22 and 27 (apparent mol. wt. 170000, 120000, 75000 and 60000, respectively) were made. All were phosphorylated and accumulated in the nucleus. Messenger RNA for the same four polypeptides was made in cells infected in the presence of cycloheximide. Thus, ICP 22 is distinct from cellular polypeptide C and is probably a virus-specific alpha polypeptide, although it differs from alpha ICP 4, 0 and 27 in that its rate of synthesis does not decline rapidly when later polypeptides are produced. It is modified after synthesis in at least two steps, the second of which may require a later virus-specific polypeptide. In cells infected with HSV-2 the synthesis of a polypeptide analogous to ICP 22 could not be detected.     

Identification of two nuclear import signals in the alpha-gene product ICP22 of herpes simplex virus 1

The herpes simplex virus 1 (HSV-1) infected cell protein 22 (ICP22) is a multifunctional viral regulator that localizes in the nucleus of infected cells. ICP22 is required for optimal virus replication in certain cell types and is subject to extensive posttranslational modification. To map the signals in ICP22 which mediate its efficient nuclear localization, we investigated the nuclear import of fusion proteins comprising various fragments of ICP22 fused to green fluorescent protein (GFP) or beta-galactosidase (beta-Gal). These data demonstrated that ICP22 contains two independent regions with nuclear localization signal (NLS) activity. NLS1 maps to ICP22 amino acid position 16-31 and closely resembles the classical bipartite NLS of the type originally identified in nucleoplasmin. In contrast, NLS2 maps to ICP22 amino acid position 118-131 and contains multiple critical basic residues. Furthermore, fusion of both NLSs to chimeric glutathione-S-transferase (GST)-GFP protein and subsequent cytoplasmic microinjection of the respective transport substrates allowed us to monitor nuclear import in real-time. These data demonstrated that both ICP22-derived NLSs mediated efficient nuclear import with identical kinetics, resulting in complete nuclear accumulation of the chimeric transport cargoes at approximately 30 min postinjection. Finally, our data provide new insights into the domain structure of the multifunctional alpha-gene product ICP22 of HSV-1.     

Herpes simplex virus 2 modulates apoptosis and stimulates NF-kappaB nuclear translocation during infection in human epithelial HEp-2 cells

Virus-mediated apoptosis is well documented in various systems, including herpes simplex virus 1 (HSV-1). HSV-2 is closely related to HSV-1 but its apoptotic potential during infection has not been extensively scrutinized. We report that (i) HEp-2 cells infected with HSV-2(G) triggered apoptosis, assessed by apoptotic cellular morphologies, oligosomal DNA laddering, chromatin condensation, and death factor processing when a translational inhibitor (CHX) was added at 3 hpi. Thus, HSV-2 induced apoptosis but was unable to prevent the process from killing cells. (ii) Results from a time course of CHX addition experiment indicated that infected cell protein produced between 3 and 5 hpi, termed the apoptosis prevention window, are required for blocking virus-induced apoptosis. This corresponds to the same prevention time frame as reported for HSV-1. (iii) Importantly, CHX addition prior to 3 hpi led to less apoptosis than that at 3 hpi. This suggests that proteins produced immediately upon infection are needed for efficient apoptosis induction by HSV-2. This finding is different from that observed previously with HSV-1. (iv) Infected cell factors produced during the HSV-2(G) prevention window inhibited apoptosis induced by external TNFalpha plus cycloheximide treatment. (v) NF-kappaB translocated to nuclei and its presence in nuclei correlated with apoptosis prevention during HSV-2(G) infection. (vi) Finally, clinical HSV-2 isolates induced and prevented apoptosis in HEp-2 cells in a manner similar to that of laboratory strains. Thus, while laboratory and clinical HSV-2 strains are capable of modulating apoptosis in human HEp-2 cells, the mechanism of HSV-2 induction of apoptosis differs from that of HSV-1.     

Infected cell protein No. 22 is subject to proteolytic cleavage by caspases activated by a mutant that induces apoptosis

Earlier reports have shown that the d120 mutant of herpes simplex virus 1 lacking both copies of the gene encoding the infected cells protein No. 4 (ICP4) induces apoptosis in a variety of cell lines. The programmed cell death induced by this mutant is blocked by overexpression of Bcl-2 or by transduction of infected cells with the gene encoding the viral U(S)3 protein kinase. HEp-2 cells infected with the d120 mutant express predominantly alpha proteins. Studies on these proteins revealed the accumulation of a M(r) 37,500 protein that reacted with antibody directed against the carboxyl-terminal domain of ICP22. We report that the M(r) 37,500 protein is a product of the proteolytic cleavage of ICP22 by a caspase activated by the d120 mutant. Thus the accumulation of the M(r) 37,500 protein was blocked in cells transduced with the U(S)3 protein kinase, in cells overexpressing Bcl-2, or in infected cells treated with the general caspase inhibitor zVAD-fmk. Exposure of ICP22 made in wild-type virus-infected cells to caspase 3 yielded two polypeptides, of which one could not be differentiated from the M(r) 37,500 protein with respect to electrophoretic mobility. We conclude that the cellular apoptotic response targets at least one viral protein for destruction.     

Differences in the morphology of herpes simplex virus infected cells

The two types of herpes simplex virus (HSV-1, HSV-2) induced significantly different alterations in the morphology and permeability of infected cells. HEp-2 cells infected with HSV-1 (strain THEA) were characterized by the formation of polynuclear syncytia. In contrast, after infection with HSV-2 (strain D316, DD), the cells were rounded up. The HSV-1 strains KOS and LS5039 and the HSV-2 strain 196 induced both types of cytopathic effect. As shown by comparative scanning and transmission electron microscopy newly synthesized virus particles of the various strains of HSV-1 were generally found to be restricted to smooth areas of the cell surface. In these areas the number of microvilli was reduced in comparison to uninfected cells. However, the progeny viruses of the strains of HSV-2 were mainly connected with protrusions of the cell membrane (microvilli and filopodia). The morphological changes in cells infected with either type of HSV were associated with different functional alterations of the cell membrane. The membranes of HEp-w cells became more stable after infection with HSV-1. This is characterized by a reduced permeability for 51Cr as well as by a decreased sensitivity to the detergent Triton-X-100. HSV-2 induced opposite effects on the stability of the membrane in infected cells. In contrast to these findings with HEp-2 cells, opposite results were obtained with primary chick embryo fibroblasts: Infection with HSV-1 rendered the cell membrane more permeable for 51Cr and a reduction of the 51Cr-release was achieved by infection with HSV-2. The results show that HSV-cell interactions depend on the type of the virus as well as on the type of the infected cell.     

Induction of apoptosis in HEp-2 cells by infection with herpes simplex virus type 2

Summary.  Although herpes simplex virus type 1 (HSV-1) does not induce apoptosis in infected HEp-2 cells, herpes simplex virus type 2 (HSV-2) did induce apoptosis in a small but significant fraction of the same cells. Apoptosis was not observed in Vero or HeLa cells infected with HSV-2. In addition, HSV-2 infection in the presence of cycloheximide induced extensive apoptosis of HEp-2 or HeLa cells. Received June 9, 1998 Accepted July 15, 1998     

Regulation of herpesvirus macromolecular synthesis. V. Properties of alpha polypeptides made in HSV-1 and HSV-2 infected cells.

Previous reports from this laboratory have defined as α those viral polypeptides whose production in infected cells does not require prior protein synthesis. Two subsequent groups, β and γ, depend on prior a and β polypeptide synthesis, respectively. Comparison of the synthesis and properties of a polypeptides specified by herpes simplex viruses (HSV) 1 and 2 showed the following: (i) The three earliest virus-specific infected cell polypeptides (ICP) made in HSV-2 infected cells migrated slightly more slowly in polyacrylamide gels than the HSV-1 α polypeptides, ICP 4.0 and 27. (ii) Cells treated with canavanine from the time of infection with HSV-2 produced all α, a subset of β, and a small amount of one γ polypeptide; the synthesis of these polypeptides continued for many hours, at rates related to the multiplicity of infection. The transition from a to this subset of β polypeptide synthesis did not appear to be affected by the arginine analogue, whereas transition to the other sets of β and to γ polypeptide synthesis was blocked. A similar discrimination between different subsets of β proteins was seen in treated HSV-1 infected cells. (iii) All a and a number of β polypeptides observed in this study underwent translocation into the nucleus, posttranslational modification resulting in a reduced electrophoretic mobility, and phosphorylation. For example, the modification of HSV-1 and HSV-2 ICP 4 was in at least two steps from the translational product ICP 4a, labeled during a pulse, to slower-migrating forms ICP 4b and 4c. All three forms were phosphorylated but only 4b and 4c were found in the nucleus. In untreated infected cells, ICP 4 ultimately accumulated in form ICP 4c. ICP 4a made in the presence of canavanine was not processed efficiently into ICP 4c. In another instance, the polypeptide made in the presence of canavanine were not translocated into nuclei.     

A conserved carboxy-terminal domain in the major tegument structural protein VP22 facilitates virion packaging of a chimeric protein during productive herpes simplex virus 1 infection

Recombinant virus HSV-1(RF177) was previously generated to examine tegument protein VP22 function by inserting the GFP gene into the gene encoding VP22. During a detailed analysis of this virus, we discovered that RF177 produces a novel fusion protein between the last 15 amino acids of VP22 and GFP, termed GCT-VP22. Thus, the VP22 carboxy-terminal specific antibody 22-3 and two anti-GFP antibodies reacted with an approximately 28 kDa protein from RF177-infected Vero cells. GCT-VP22 was detected at 1 and 3 hpi. Examination of purified virions indicated that GCT-VP22 was incorporated into RF177 virus particles. These observations imply that at least a portion of the information required for virion targeting is located in this domain of VP22. Indirect immunofluorescence analyses showed that GCT-VP22 also localized to areas of marginalized chromatin during RF177 infection. These results indicate that the last fifteen amino acids of VP22 participate in virion targeting during HSV-1 infection.     

In vitro characterization of a herpes simplex virus type 1 ICP22 deletion mutant

We report the construction of a deletion mutant (del22Z) that is unable to synthesize any detectable messenger RNA or protein products from the herpes simplex virus type 1 (HSV-1) immediate early ICP22 gene upon infection. The del22Z deletion mutant lacks all but 18 nucleotides of the ICP22 coding sequence and carries the bacterial lacZ gene at the site of the deletion. No other known open reading frames or flanking sequences were disrupted. Del22Z was able to infect Vero cells productively but was severely restricted in human and rodent cells that were permissive for the parental HSV-1(F). The yield of del22Z was not enhanced significantly, either by increasing the multiplicity of infection or by increasing the duration of the infection. There was a prolonged expression of some early gene products and a delayed appearance of some late gene products in both permissive and restrictive cells. This phenotype of cell-line restricted growth and alteration of the normal gene expression cascade maps specifically to the ICP22 coding region.     

The immediate-early protein, ICP0, is essential for the resistance of herpes simplex virus to interferon-alpha/beta

Herpes simplex virus type 1 (HSV-1) is resistant to the antiviral effects of interferon (IFN)-alpha, -beta, or -gamma. The fact that ICP0(-) mutants replicate like wild-type virus in IFN-alpha/beta receptor knockout mice (Leib et al., 1999, J. Exp. Med. 189, 663) suggested that ICP0 may serve a direct role in the resistance of HSV-1 to IFN. To test this hypothesis, the effects of IFN-alpha, -beta, and -gamma were compared against wild-type HSV-1 and an ICP0(-) mutant virus, 7134. In Vero cells, 7134 was more sensitive to inhibition by low doses of type I IFN (-alpha/beta) or type II IFN (-gamma) than vesicular stomatitis virus, a well-studied IFN-sensitive virus. At a concentration of 100 U/ml, IFN-alpha, -beta, or -gamma reduced the efficiency of 7134 plaque formation by 120-, 560-, and 45-fold, respectively. In contrast, none of the IFNs reduced wild-type HSV-1 plaque formation by more than 3-fold. Even when Vero cells were infected with 10 pfu per cell, IFN-alpha and -beta inhibited 7134 replication by over 100-fold, but inhibition by IFN-gamma decreased to less than 10-fold. While IFN-beta efficiently inhibited 7134 replication in primary mouse kidney and SK-N-SH cells, IFN-gamma did not inhibit 7134 to a comparable extent in these cells. ICP0 provided in trans from an adenovirus vector allowed 7134 to replicate efficiently in Vero cells in the presence of IFN-alpha, -beta, or -gamma. While IFN-beta or -gamma efficiently repressed the ICP0 promoter-lacZ reporter gene in 7134 (i.e., approximately 60-fold reduction in beta-galactosidase activity), ICP0 provided in trans almost completely reversed IFN-mediated repression of the lacZ gene in 7134. The results suggest that the rate of ICP0 expression in infected cells in vivo may be critical in determining whether host IFNs repress the HSV-1 genome. This concept is discussed in light of its potential relevance to the establishment of latent HSV-1 infections.     

Conversion of trichosanthin-induced CD95 (Fas) type I into type II apoptotic signaling during Herpes simplex virus infection

Trichosanthin (TCS) is a type I ribosome-inactivating protein with wide spectrum of pharmacological activities. It inhibits human immunodeficiency virus type 1 (HIV-1) and Herpes simplex virus type 1 (HSV-1) replication but the mechanism is not clear. From a previous study, TCS was found to be more cytotoxic to HIV-1 infected cells than uninfected cells. Similar finding was confirmed with HSV-1 in the present study. TCS induced cell death in HEp-2 cells and the EC(50) was 24.64μg/mL. When the same experiment was performed in HSV-1 infected HEp-2 cells, the EC(50) decreased to 3.01μg/mL. TCS appeared to cause more death and apoptosis in viral infected cells. This study explored plausible mechanism with respect to the apoptosis signal pathways. In uninfected cells, TCS induced CD95 (Fas)-mediated and caspase-8-dependent type I apoptosis. When cells were infected with HSV-1, apoptosis induced by TCS clearly switched to a more potent type II pathway. This involved mitochondrial depolarization and caspase-9 activation. The major evidences arose from studying the individual signals of the two apoptosis pathways in infected and uninfected cells. In addition, over expression of Bcl-2, which mainly affected the type II pathway reduced TCS induced apoptosis mostly in infected cells. This further demonstrated that the type II pathway was operating in infected cells. The reason for the switching is not entirely clear but it is well known that viral infection affects signal pathways especially those related to apoptosis. In conclusion, TCS selectively induces more apoptosis in HSV-1 infected cells than uninfected cells. The consequence of infection switches the TCS-induced apoptosis pathway from a CD95 (Fas) dependent type I to a more potent type II pathway mediated by mitochondrial depolarization and caspase-9 activation.     

Phosphorylation of a ribosomal protein and of virus-specific proteins in cells infected with herpes simplex virus.

In cells infected with herpes simplex virus a protein associated with the small subunit of ribosomes became phosphorylated. It was not detectably labelled with 14C-amino acids added after infection and is therefore probably a cellular protein. The phosphorylated ribosomal proteins from HSV-1- and HSV-2-infected cells were indistinguishable electrophoretically and had an apparent mol. wt. of about 48 000. Phosphorylation of the 48K protein was detected 2 to 3 h after infection and reached a maximum rate at 4 to 5 h. It was prevented by adding cycloheximide at 2 h, or actinomycin at 1.5 h p.i., or azetidine at the beginning of infection. The phosphorylation did not occur on reversal of a cycloheximide block in the presence of actinomycin, confirming that it is not caused by a virus alpha-polypeptide. Virus that had been irradiated with u.v. light, although still able to suppress synthesis of cellular protein and DNA, did not induce phosphorylation of the 48K ribosomal protein. Therefore the phosphorylation is not responsible for the suppression of host synthesis. The alpha polypeptides ICP 4, 0, 22 and 27 are also phosphorylated but, in contrast to that of the ribosomal protein, their phosphorylation does not depend on the synthesis of beta and gamma polypeptides. It is probably mediated by a host enzyme.     

ICP34.5-dependent and -independent activities of salubrinal in herpes simplex virus-1 infected cells

The small molecule salubrinal has antiviral activity against herpes simplex virus-1 (HSV-1) and inhibits dephosphorylation of eIF2α mediated by the HSV-1 protein ICP34.5. We investigated whether salubrinal's activities in infected cells depend on ICP34.5. An ICP34.5 deletion mutant was as sensitive as wild type HSV-1 to salubrinal inhibition of plaque formation in Vero cells. However, salubrinal induced formation of syncytia in infected Vero cells, which was enhanced by ICP34.5 mutations. Expression of HSV-1 US11 with immediate early kinetics, which is known to suppress the effects of ICP34.5 mutations, resulted in slight resistance to salubrinal in murine embryonic fibroblasts, and substantial resistance in those cells when ICP34.5 was additionally mutated. ICP34.5 mutations, but not immediate early expression of US11, prevented salubrinal's ability to increase phosphorylation of eIF2α during HSV-1 infection of Vero cells. Taken together, our data indicate that salubrinal has both ICP34.5-dependent and -independent activities in HSV-1 infected cells.     

Antiserum to the recombinant truncated VP22 protein of herpes simplex virus type 1 that also recognizes full-length VP22

The herpes simplex virus type 1 (HSV-1) tegument protein VP22 encoded by the UL49 gene is essential for HSV-1 infection. However, its precise functions in the virus life cycle are unknown. A relatively important tool for disclosing these functions is an antiserum specifically detecting VP22 in the infected cell. To this end, a recombinant truncated VP22 protein consisting of C-terminal 45 aa fused to EYFP (enhanced yellow fluorescent protein) and His-tag was expressed in Escherichia coli, purified by the Ni2+-NTA affinity chromatography, and used for the preparation of antiserum in rabbits. Western blot and immunofluorescence assay showed that this antiserum specifically detected purified truncated VP22 as well as full-length VP22 in the HSV-1 infected cells. These results indicate that the prepared antiserum could serve as a valuable tool for further studies of VP22 functions.     

Dependence of herpes simplex virus type 1-induced cell fusion on cell type

Syncytial mutants of herpes simplex virus type 1 (HSV-1), such as syn20, cause extensive fusion of human embryonic lung (HEL) cells but only a small amount of fusion of human epidermoid carcinoma No. 2 (HEp-2) cells. In order to determine the cellular basis of this difference in fusion, sparse cultures of syn20-infected HEL or HEp-2 cells, previously labeled with [³H]thymidine, were surrounded with uninfected, unlabeled HEL or HEp-2 cells. The fusion of radioactive with nonradioactive cells was determined at different times after infection using radioautography. syn20-infected HEL cells fused extensively with surrounding uninfected HEL or HEp-2 cells, while syn20-infected HEp-2 cells fused poorly with surrounding uninfected HEL or cells. Therefore, the major difference in the fusion capacity of HEL and HEp-2 cells was not due to a difference in cell-surface receptors for a fusion factor in the two cell types. The process of infection of HEp-2 cells did not cause the plasma membranes of the cells to become refractory to fusion, because syn20-infected HEL cells fused equally well with either uninfected or infected HEp-2 cells. The capacity for a mutant virus to express the syncytial phenotype in mixed infection with a wild-type virus is also dependent on cell type. In a mixed infection with equal numbers of MP and its nonsyncytial parent, mP, extensive fusion was observed for infected HEL cells and significantly less fusion was observed for infected African green monkey kidney (CV-1), baby hamster kidney (BHK-21), and HEp-2 cells.     

Differences in the morphology of herpes simplex virus infected cells: I. Comparative scanning and transmission electron microscopic studies on HSV-1 infected HEp-2 and chick embryo fibroblast cells.

Infection with herpes simplex virus type 1 (HSV-1) induces different morphological changes in different cell lines. This is demonstrated by comparative scanning (SEM and transmission (TEM) electron microscopic investigations of cell cultures prepared under identical conditions. SEM of HSV-1 infected HEp-2 cells reveals a slightly altered cell surface: only the number of the microvilli is reduced. Large amounts of released virions are detectable adhering to the outer plasma membrane. Ultra-thin sections show typical virus maturation steps in the nuclei (formation of nucleocapsids and virus budding from the inner lamella of the nuclear membrane) and in the cytoplasm (egress of enveloped nucleocapsids through membranous structures). HSV-infected primary chick embryo fibroblast (CEF) cells are characterized by crumpled and rough surfaces without virus particles adhering to the membrane. Ultra-thin sections exhibit atypical virus maturation with many unenveloped nucleocapsids within the cytoplasm. The distribution of HSV-induced antigen(s) on the surface of the infected cells is identical in the two cell systems as determined by the peroxidase labelling technique. The c.p.e. (as seen by phase contrast light microscopy) is similar in both HEp-2 and CEF cells: both fusion and rounding up is induced in the infected cells.     

Genomic characterisation of Wongabel virus reveals novel genes within the Rhabdoviridae

ICP0 is a multi-functional herpes simplex virus type 1 (HSV-1) immediate-early (IE) gene product that contributes to efficient virus growth and reactivation from latency. Here we show that HSV-1-induced cell-cycle arrest at the G2/M border requires ICP0 and Chk2 kinase and that ICP0 expression by transfection or infection induces ATM-dependent phosphorylation of Chk2 and Cdc25C. Infection of cells with a replication-defective mutant virus deleted for all the regulatory IE genes except ICP0 (TOZ22R) induced G2/M arrest whereas a mutant virus deleted in addition for ICP0 (QOZ22R) failed to do so. Chk2-deficient cells and cells expressing a kinase-deficient Chk2 did not undergo cell-cycle arrest in response to TOZ22R infection. Chk2 deficiency diminished the growth of wild-type HSV-1, but not the growth of an ICP0-deleted recombinant virus. Together, these results are consistent with the interpretation that ICP0 activates a DNA damage response pathway to arrest cells in G2/M phase and promote virus growth.     

Purification of herpes simplex virus-specified major DNA-binding proteins and detection of antibodies to them in human and rabbit sera

We purified herpes simplex virus type 1 and 2-specified nonstructural DNA-binding proteins (ICP8 and ICSP 11/12, respectively) from infected Vero cells and applied them in ELISA test to analyse human and rabbit sera. Only a weak immune response to the ICP8 and ICSP 11/12 could be demonstrated following experimental HSV-1 or HSV-2 infections of rabbits. Low levels of ICSP8 antibodies were found also in acute HSV infections of man.