|Home | About | Journals | Submit | Contact Us | Français|
ICP0, a promiscuous transactivator that enhances the expression of genes introduced by infection or transfection, functions in both nucleus and cytoplasm. The nuclear functions include degradation and dispersal of ND10 bodies and suppression of silencing of viral DNA. Subsequently, ICP0 shifts to the cytoplasm. Transfection of DNA prior to infection has no effect on the localization of ICP0 in cells that are efficient expressers of transgenes (e.g., Vero and HEK293) but results in delayed cytoplasmic localization of ICP0 in cells (e.g., HEp-2 and HEL) that are poor transgene expressers. Here, we examined by real-time PCR (qPCR) the accumulation of a transgene and of viral gI mRNAs in Vero or HEp-2 cells that were transfected and then infected with wild-type or ΔICP0 mutant viruses. The accumulation of transgene mRNA was unaffected by a ΔICP0 mutant, gradually increased in HEp-2 cells, but increased and then decreased in Vero cells infected with wild-type virus. In both cell lines, accumulation of gI mRNA increased with time and was less affected by the transfected DNA in Vero cells than in HEp-2 cells. The relative kinetics of mRNA accumulation reflected continued synthesis and degradation of the transgene and gI mRNAs. We conclude that the role of ICP0 is to render the DNA templates introduced by transfection or infection accessible by transcriptional factors, that the two cell lines differ with respect to the transcription-ready status of entering foreign DNA in the nucleus, and that ICP0 is not per se the recruiter of transcriptional factors to the accessible DNA templates.
Numerous studies carried out shortly after the discovery of infected cell protein 0 (ICP0) of herpes simplex virus type 1 (HSV-1) described the function of the protein as a promiscuous transactivator of genes introduced by transfection or infection (1, 4, 5, 9, 10, 12, 17, 24, 26, 27, 29). The enhancement of gene expression by ICP0 was particularly puzzling, since it does not bind DNA and is not known to recruit transcriptional factors to specific promoters (11). This article is a reappraisal of its function as a transactivator. We propose that its function is that of a DNA template remodeler, that the extent to which it functions in that capacity is cell type dependent, and that in an absolute sense it is not an obligate component of the transcriptional apparatus used by the virus to transcribe its genes. The experiments we report are based briefly on the following studies.
Early in infection, ICP0 is located in the nucleus. At later times, ICP0 is in the cytoplasm. During its nuclear sojourn, ICP0 performs several functions. Studies of 3 of these functions are particularly illuminating. The first involves the degradation of several components of the ND10 nuclear bodies by the ubiquitin ligase function of ICP0 in conjunction with the UbcH5a ubiquitin-conjugating enzyme, followed by the dispersal of ND10 bodies (3, 13, 16). The second function of ICP0, executed in tandem with the first, is the suppression of silencing of viral DNA (6, 14, 15). Thus, on infection, DNA entering the nucleus localizes at ND10 structures and recruits host proteins, including the repressor complex. ICP0 displaces histone deacetylase 1 (HDAC1) or HDAC2 from the HDAC1-2-CoREST-REST-LSD1 repressor complex (14, 15). Lastly, ICP0 binds and recruits cyclin D3 to ND10 structures that ultimately evolve into replication compartments (20, 21). The substantive finding that led to these studies is that failure in the execution of any one of these functions leads to the retention of ICP0 in the nucleus (14, 15, 20, 21).
To examine more thoroughly the relationship between the suppression of silencing of viral DNA and the retention of ICP0 in the nucleus, we transfected a variety of DNAs into cells prior to infection with wild-type virus (19). We noted that the outcome was dependent on the cell type and the quantity of transfected DNA, but not on the type of DNA (19). Thus, in cells that fail to efficiently express transfected DNA (e.g., HEp-2 and human embryonic lung [HEL] cells), ICP0 was retained in proportion to the amount, but not the type, of transfected DNA. In contrast, ICP0 was not retained in the nuclei of cells (e.g., Vero, rabbit skin, or HEK-293 cells) that efficiently express transfected DNA (19). In simplistic terms, one hypothesis that could explain these results is that ICP0 has to work harder to render the DNA available for transcription in cells in which it is retained in the nucleus than in cells in which it is not retained. A more explicit statement of the hypothesis is that transfected DNA or viral DNA introduced by infection requires less modification and is more readily available for immediate transcription in cells in which ICP0 is not retained but must be extensively “processed” by ICP0 in cells in which ICP0 is retained in the nucleus following infection/transfection.
To test this hypothesis, we transfected Vero and HEp-2 cells with a plasmid encoding ampicillin and then infected them with either wild-type or ΔICP0 mutant viruses. The transgene mRNA and the mRNA encoding the gI protein of HSV-1 were then measured by real-time PCR (qPCR). The expectation was that comparisons of infected cells to transfected/infected HEp-2 cells would show that the transgene competes and interferes with the expression of gI. We report three fundamental conclusions. First, the findings that the accumulation of the transgene mRNA is unaffected by ΔICP0, slowly upregulated by wild-type virus in HEp-2 cells, and rapidly upregulated and then downregulated in Vero cells suggest that in Vero cells, DNAs entering the cells are in a nearly expression-ready state whereas in HEp-2 cells they must be extensively remodeled with the participation of ICP0. Second, the observation that the accumulation of viral mRNAs is upregulated in both cell lines but that the starting values in ΔICP0-infected cells are vastly lower suggests that the function of ICP0 is to provide suitable DNA substrates to the transcriptional machinery but that ICP0 does not per se recruit transcriptional factors to the DNA. Lastly, we show that the levels of transgene and viral mRNAs detected even at late times after infection reflect both synthesis and degradation of the mRNAs.
The sources, properties, and propagation of HEp-2 and Vero cells have been described elsewhere (18-20). HSV-1(F), a limited-passage isolate, is the prototype strain used in this laboratory (7). The construction and the phenotypic properties of R7901, lacking both copies of ICP0, have been described elsewhere (21).
For the expression of ampicillin (Amp) in mammalian cells, the BspHI open reading frame (ORF) of Amp from pcDNA 3.1 Zeo(+) was inserted into the SpeI-XhoI sites of pRB5162, generating the pRBEgr-1/Amp plasmid, where the ampicillin ORF is driven from the Egr-1 promoter and flanked from the bidirectional UL21/UL22 poly(A) (21).
Cells grown on 25-cm2 flasks were transfected when 60 to 70% confluent with 1.5 μg DNA/flask in mixtures with 4 μl Lipofectamine and 6 μl Plus per flask, as specified by the supplier (Invitrogen). For the immunofluorescence studies, cells grown in four-well slides (Erie Scientific) were transfected when 60 to 70% confluent with 250 ng per well of pRBEgr-1/Amp DNA in mixtures of 1 μl of Lipofectamine and 1.5 μl of Plus reagents per well. At 3 h after transfection, the cells were rinsed extensively with culture medium and further incubated for 18 h. In all experiments, the cells were exposed to 10 PFU of either HSV-1(F) or ΔICP0 per cell in mixture 199 (Sigma) supplemented with 10% fetal bovine serum at 18 h after mock transfection or transfection with DNA. For the mRNA decay experiments, actinomycin D (Sigma) was added to the cells 2.5 h after infection at a final concentration of 10 μg/ml, and the cells were harvested 0.5 h, 1 h, 2.5 h, or 5 h after the addition of the drug. The quantification of all mRNAs was done as detailed below.
Total RNA was extracted with the aid of TRIzol reagent (Invitrogen) according to the manufacturer's instructions. DNase treatment (Promega), phenol-chloroform extraction (Ambion), and ethanol precipitation (Fisher Scientific) were carried out to remove possible DNA contamination. First-strand cDNA synthesis using 2 μg of total RNA and oligo(dT) was done with the SuperScriptIII First-Strand Synthesis System for RT-PCR (Invitrogen), according to the suppliers’ instructions. Equal volumes of the cDNA synthesis mixtures were used for quantification of transgene (ampicillin) transcripts or the viral-gene (gI) transcripts with the SYBR GreenER qPCR SuperMix Universal (Invitrogen), according the manufacturer's instructions. Samples without RT-PCR were tested as controls. The primers for ampicillin were forward, 5′ GATACGGGAGGGCTTACCAT 3′, and reverse, 5′ GATAACACTGCGGCCAACTT 3′; for gI, they were forward, 5′ CCCACGGTCAGTCTGGTATC 3′, and reverse, 5′ TTTGTGTCCCATGGGGTAGT 3′. The transgene and viral transcripts were normalized to the 18S rRNA levels. The 18S rRNA primers (universal primers from Ambion) were modified according to the supplier's instructions to be suitable as internal controls for mRNA species at any abundance. The fold change of transgene or viral-gene transcripts was expressed relative to their value at the earliest time point, as detailed in the text and the figure legends. The assays were performed on an ABI 7300 system or a StepOnePlus system (Applied Biosystems) and analyzed with software provided by the supplier.
The procedures for immunofluorescence studies have been described elsewhere (18-20). Briefly, the cells, seeded in four-well slides (Erie Scientific), were fixed in 4% paraformaldehyde at the times indicated in Results; permeabilized; blocked with phosphate-buffered saline (PBS)-TBH solution consisting of 0.1% Triton X-100 in PBS, 10% human serum, and 1% bovine serum albumin (BSA); and reacted with the ICP0 exon 2 rabbit polyclonal antibody diluted 1:2,000 in PBS-TBH (21). The slides were then rinsed several times with PBS-TBH and reacted with Alexa-Fluor 594-conjugated goat anti-rabbit immunoglobulin diluted 1:1,000 in PBS-TBH. After several rinses, first with PBS-TBH and then with PBS, the samples were mounted and examined with a Zeiss confocal microscope equipped with software provided by Zeiss.
The objective of these studies was to compare the accumulation of transgene and viral gI mRNAs in cells transfected or infected with those accumulating in transfected/infected cells. We tested HEp-2 cells in which ICP0 is retained in the nuclei of transfected/infected cells and Vero cells in which ICP0 is not retained (19). The experiments described here were designed to verify the localization of ICP0 during the time intervals after transfection/infection examined in these studies.
In infected cells, the duration of the sojourn of ICP0 in the nucleus is cell type dependent. In this study, we examined the accumulation of transcripts at 1.5, 3, 6, and 12 h after infection. Earlier studies have shown that at 3 h after infection, ICP0 was in the nuclei of all cells tested to date (22). For the purpose of the studies described in this report, we examined the distribution of ICP0 at 6 and 12 h after infection (Fig. (Fig.1).1). At each time point, cells grown in 4-well slides and either infected or transfected and infected, as described in Materials and Methods, were tabulated according to the location of ICP0. The results, shown in Fig. Fig.1,1, were as follows. In both infected Vero and HEp-2 cells, ICP0 was partially in the cytoplasm at 6 h and in the cytoplasm of nearly all cells at 12 h. In transfected/infected Vero cells, the number of cells containing ICP0 in the cytoplasm closely matched the number of cells that were only infected. In contrast, in transfected/infected HEp-2 cells, ICP0 was retained in the nucleus in more than 95% of the cells at both time points.
The objective of these experiments was to examine the accumulation of transgene mRNA in cells that were only transfected with those cells that were both transfected and infected. We report the results of 3 experiments. In one, the cells were harvested at 1.5, 3, 6, and 12 h after infection. In the other two, the cells were harvested at 3, 6, and 12 h after infection. The mRNAs were then extracted and measured for ampicillin and gI mRNAs by qPCR. The results were normalized with respect to the 18S rRNA and then to the mRNA levels of cells that were only transfected and were harvested at 1.5 h (experiment 1) or at 3 h after mock infection (experiments 2 and 3). The values assigned to mRNAs to which all other mRNAs were normalized (equal to 1) are identified by arrows in the figures. Although the assays of transgene mRNAs were carried out in tandem with assays of viral gI mRNAs, it is convenient to consider them separately. Figure Figure22 shows the results of the experiments in each of the two cell lines. Briefly, they were as follows.
In the absence of infection, the amounts of transgene mRNAs in HEp-2 cells fluctuated approximately 3-fold over the course of the experiment (Fig. (Fig.2B).2B). The levels of transgene mRNA increased with time after infection with wild-type virus. The highest levels were observed at 12 h after infection.
In the absence of infection, the levels of ampicillin mRNAs in Vero cells (Fig. (Fig.2C)2C) fluctuated approximately 2- to 3-fold over the course of the experiment. In all experiments, the levels of transgene mRNAs in cells infected with wild-type virus were 8- to 12-fold higher at 3 h than in cells that had only been transfected. In all experiments, the mRNAs attained their highest levels 3 h after infection and tended to decrease at later time points.
In HEp-2 cells at 3 and 6 h after infection with ΔICP0 mutant virus (Fig. (Fig.2B),2B), the levels of transgene mRNAs were similar or slightly higher than those of cells that had only been transfected. At 12 h, the levels of transgene mRNA decreased relative to those of tranfected/infected cells. In Vero cells transfected and then infected with the ΔICP0 mutant virus, there was an increase in the transgene mRNA levels. The discernible trend was toward slight increases at late times after infection.
The results led us to conclude that in Vero cells the highest levels of transactivation of the ampicillin gene occurred at 3 h after transfection/infection or before and that past 3 h, the levels of the transgene mRNAs were declining concordant with the disappearance of ICP0 from the nucleus. In contrast, the levels of mRNAs continued to increase, and the highest levels of transgene mRNAs in HEp-2 cells were not reached until 12 h after infection with wild-type virus, again concordant with the maintenance of ICP0 in the nucleus.
In Vero cells, there was a nearly 100-fold increase in the amounts of gI mRNA between 3 and 6 h after infection with wild-type virus in two of the three experiments. This increase was matched in cells that were transfected and then infected (Fig. 3A and C). In the 3rd experiment, the increase in the accumulated mRNA was only 10-fold, but the starting amount at 3 h in this experiment was much higher than in experiments 1 and 2. Again, we noted no differences in the accumulation of gI mRNAs in infected versus transfected/infected cells.
In cells infected with ΔICP0 mutant virus (Fig. 3B and D) carried out in the same experiments shown in Fig. 3A and C, the levels of gI mRNA were approximately 100-fold lower than those detected in cells infected with wild-type virus. The amounts increased 20-fold at 6 h and more than 100-fold at 12 h after infection. Here again, the differences between the amounts of mRNAs recovered from infected cells and those recovered from transfected/infected cells were 2-fold or less.
The results of the experiments discussed above indicate that the DNA transfected into Vero cells did not significantly interfere with the accumulation of gI mRNA.
An entirely different picture emerged from analysis of the accumulation of gI mRNAs in infected or transfected/infected HEp2 cells (Fig. (Fig.4).4). Thus, in the 3 experiments reported here, the amounts of gI mRNAs accumulating in transfected/infected cells at 1, 5, and 3 h after infection were 5- to 20-fold lower than in corresponding cells that were only infected. At later time points, the amounts of gI mRNA were nearly 10-fold lower in experiment 1 (Fig. (Fig.4A)4A) but reached levels nearly comparable to those of infected cells in experiments 2 and 3 (Fig. 4C and E).
The differences between the levels of gI mRNAs in cells infected with ΔICP0 virus and wild-type virus were more striking in HEp-2 cells than in Vero cells. Thus, in ΔICP0 mutant-virus-infected or -transfected/infected HEp-2 cells, there was 100-fold less gI mRNA than in corresponding cells infected or transfected/infected with wild-type virus (compare Fig. 4A and B, and C and D). The amounts of gI mRNAs in ΔICP0 mutant-virus-infected or -transfected/infected cells increased with time, but even at 12 h after infection, they did not reach the levels of gI mRNA at 3 h after infection with wild-type virus.
The key conclusions to be drawn from these experiments are 2-fold. Foremost, in HEp-2 cells, the transfected DNA interferes with the accumulation of gI mRNA at early times after infection. The interference decreases with time (e.g., 6 and 12 h after infection).
The second observation is that the accumulation of gI mRNA encoded by the ΔICP0 mutant virus is more significantly affected in HEp-2 cells than in Vero cells. In each instance, the mRNAs continued to accumulate at rates that paralleled the rates of increase in the accumulation of gI mRNAs in cells infected with wild-type virus.
The impetus for this series of experiments was based on the observation that in Vero cells at early times after transfection/infection, the amounts of accumulated transgene mRNA declined with time, whereas that of gI mRNA increased. The amounts of mRNA at any given time reflect the outcome of two contrasting phenomena: continued synthesis of mRNA on one hand versus decay of the mRNA due to degradation. In the case of infected cells, a powerful endoribonuclease encoded by the UL41 gene of HSV significantly alters the landscape of accumulating mRNAs after infection. The question we posed is, to what extent does the steady-state decrease in accumulation of transgene mRNA reflect decreased synthesis as opposed to degradation of the accumulating mRNAs in the environment of the infected cell? The experiments we report were done as follows. As illustrated briefly in Fig. Fig.5A,5A, Vero cells were transfected at −18 h with respect to the time of mock infection or infection with HSV-1(F). At 2.5 h after infection, duplicate cultures were mock treated or exposed to actinomycin D. The cultures were harvested 30 min, 1 h, 2.5 h, and 5 h after mock treatment or exposure to actinomycin D. The mRNAs were extracted and analyzed by qPCR. The amounts of mRNA were normalized with respect to the 18S rRNA of each sample and then to the mRNA levels recovered at 0.5 h after addition of actinomycin D. The results for transgene mRNA and gI mRNAs are shown in Fig. 5B and C, respectively. The salient features of the results can be summarized as follows.
In uninfected cells, the transgene mRNA levels accumulating in the presence or absence of actinomycin D were relatively similar, indicating that the mRNAs were relatively stable and, as shown in Fig. Fig.5,5, at low, relatively steady levels.
In infected cells treated with actinomycin D (Fig. (Fig.5,5, T+I+ACT), there was a precipitous decline in the accumulation of transgene mRNA. This precipitous decline in transgene mRNA was not observed in mock-treated cells infected with wild-type virus (T+I). The hypothesis that best fits the results is that the levels of mRNA reflect both synthesis and decay, and therefore, that the synthesis of mRNA continues even while the accumulation of mRNA exhibits a steady decline.
As expected from the results shown in Fig. Fig.3,3, the levels of HSV gI mRNA increased at rates similar to those of both infected (I) and transfected/infected (T+I) cells. After exposure to actinomycin D in this experiment, the gI mRNA decreased at a rate slightly lower than that of transgene mRNA. In two other experiments (not shown), the rate of decrease of transgene mRNA was significantly higher than that of gI mRNA.
An intriguing characteristic of ICP0 is that it performs its functions sequentially in several compartments of the cell. At early stages, it is in the nucleus, starting first at ND10 bodies and later filling the nucleus. At later stages, it is located in the cytoplasm (22). As noted in the introduction, the duration of its stay in the nucleus is to some extent cell type dependent and ceases only after at least 3 of the investigated nuclear functions are executed. These function are degradation of key components and dispersal of ND10 bodies, suppression of silencing of viral DNA, and recruitment of cyclin D3 to the emerging replication compartments at ND10 structures (3, 6, 13-15, 20). The focus of this report is on suppression of silencing of DNA introduced into cells by transfection and or infection. Specifically, earlier reports have shown that transfected DNA and HSV DNA introduced into cells by infection accumulate at ND10 bodies (2, 19, 28). This laboratory reported that a function of ICP0 is to bind CoREST and dislodge HDAC1 or -2 from the CoREST-REST-LSD1 repressor complex (14, 15). The link between the displacement of HDACs from the repressor complex and suppression of silencing of viral DNA emerged from studies showing that a dominant-negative CoREST that bound REST but not HDAC1 rescued ΔICP0 mutant virus in Vero cells and, to a lesser extent, in HEp-2 and other cell lines (14). Since transfected DNA also accumulates at ND10 bodies, the question arose as to whether ICP0 also suppresses the silencing of genes introduced into cells by transfection.
To answer this question, we transfected cells with DNA and then superinfected the cells with wild-type virus. The expectation was that if ICP0 suppresses silencing of cellular and viral DNAs, its burden may increase, and therefore, the time required to execute its function would be extended in proportion to the amount of transfected DNA. In the published studies, we found that in some cell lines, transfected DNA blocked the export of ICP0 from the nucleus in a DNA dose- but not DNA type-dependent manner (19). Since the DNAs tested included promoter-driven open reading frames, promotorless sequences, and G+C copolymers, it could be concluded that ICP0 did not discriminate between bona fide open reading frames and noninformational DNA (19). The cells in which the export of ICP0 was affected were cell lines that are poor expressers of transfected DNA (19). A second set of cells characterized by efficient expression of transgenes did not delay the export of ICP0 from the nuclei of cells transfected with any of the DNAs tested (19). There are two possible explanations for this observation. Thus, in these cells, ICP0 discriminates between viral and nonviral DNA sequences. An alternative consistent with efficient expression of transgenes is that the silencing mechanism employed by these cells is qualitatively or quantitatively different and requires less effort for suppression of silencing by ICP0.
The studies described in this report had three objectives. The first was to verify that the transgene contained in transfected DNA was activated both in a cell line in which transfected DNA retains ICP0 in the nucleus and in a cell line in which there is no retention of ICP0 in the nucleus. The second objective was to determine whether the transgene interferes with the expression of viral DNA. Lastly, the question arose as to whether ΔICP0 mutant virus induces the expression of the transgene or whether, in turn, it is affected by the transfected DNA. It is convenient to discuss the results and their significance in the context of three hypotheses.
The first hypothesis concerns the role of ICP0 in enabling viral-gene expression. For example, the transition from α to β and γ gene expression in cells infected at low ratios of virus per cell requires both suppression of silencing and recruitment of transcriptional factors to appropriate promoters (4, 14, 27). The key finding reported here is that in ΔICP0 mutant-virus-infected cells there was an increase in the accumulation of viral mRNA and, to a much lesser extent, of transgene mRNA. With respect to the gI mRNA, the starting level was orders of magnitude lower than in wild-type-virus-infected cells but the rate of increase was similar to that observed in wild-type-virus-infected cells. These findings support the hypothesis that in the process of suppression of silencing and activation of transcription, ICP0 performs only one of the two functions. Thus, in the absence of ICP0, fewer templates are in a “transcription-ready” state and other viral genes (e.g., ICP4) perform the actual recruitment of transcription factors to promoters.
The second hypothesis concerns the suppression of the DNA-silencing function by ICP0. Briefly stated, (i) suppression takes place on DNA in proximity to ND10 bodies, (ii) it does not discriminate between viral and cellular DNAs, (iii) cells differ qualitatively or quantitatively with respect to the silencing machinery that is deployed to safeguard the cell from foreign DNA, and (iv) the duration of remodeling hinges on the quality or quantity of silencing machinery and the quantity of DNA that needs to be remodeled. Thus, in Vero cells, the accumulation of gI mRNA is already on the upswing 3 h after infection. By 6 h after transfection/infection, a large fraction of cells exhibit cytoplasmic ICP0 and the accumulation of transgene mRNA is on the decline. In contrast, in HEp-2 cells, the accumulation of transgene mRNA is on the upswing and does not plateau. Furthermore, at least at early times, there is evidence of competition between transgene DNA and viral DNA in that the former interferes with the accumulation of gI mRNA. Additional arguments supporting differences in the quality or quantity of the silencing machinery in the two cell lines include (i) retention of ICP0 in the nuclei of transfected HEp-2 cells but not in transfected Vero cells and (ii) the observation that dominant-negative CoREST increased the yield of ΔICP0 mutant virus 100-fold to near wild-type virus levels in Vero cells but only 10-fold in HEp-2 cells (14, 19).
The third hypothesis evolved from the consequences of the fundamental conclusion that the suppression of silencing is not template specific. In the absence of constraints, a lack of template specificity would result in activation of the transcription of myriad cellular genes. The hypothesis we pose is that at least two distinct mechanisms reduce the potential impact of suppression of silencing of cellular genes. The first is spatial, inasmuch as the cellular genes most likely enabled to be expressed by HSV are those in the immediate vicinity of or abutting ND10 bodies. Even so, the numbers are likely to be high and dependent on cell stress response, as noted from the observation that even low levels of HDAC inhibitors block the translocation of ICP0 from the nucleus (19). A second, nonexclusive mechanism is the fact that the degradation of mRNAs in infected cells is not sequence independent, but rather, the degradation of cellular mRNAs proceeds at a rate distinctly different from that of viral mRNAs. Thus, selectivity of degradation of cellular mRNAs has already been reported (8). In the one experiment discussed in this report, the transgene mRNA in infected cells was slightly less stable than a viral mRNA made at the same time. Even though the differences were higher in other experiments, a more thorough reexamination of the relative rates of degradation of viral and various classes of cellular mRNAs is in order. The combination of localized suppression of silencing and an enhanced rate of degradation of cellular mRNAs would more readily account for the ability of HSV to suppress host protein synthesis and host responses to infection.
Lastly, a strategic feature of the evolution of HSV is that multiple and sometimes totally diverse gene functions target the same facet of the innate immune defenses. In fact, the importance of the particular defense mechanism can be deduced from the number of viral functions directed against it. The primary function of ICP0 in blocking silencing appears to be in the dislocation of HDACs from the repressor complex, in tandem with the degradation of PML and dispersal of ND10 bodies. HSV also posttranslationally modifies HDAC1 and -2 by the US3 kinase and CoREST through the actions of both UL13 and US3 kinases (23, 25). From a virocentric point of view, suppression of silencing is an important step in evading host defenses, and undoubtedly, additional viral functions bearing on this target will eventually emerge.
These studies were aided by a grant from the National Cancer Institute (R37 CA78766).
Published ahead of print on 17 February 2010.