The SP100 component of ND10 enhances accumulation of PML and suppresses replication and the assembly of HSV replication compartments

Generation and Properties of SP100^−/−CL5-3 and PML/SP100^−/−CL18-3 Derived from HEp-2 Cells and the ID2 PML^−/− Cell Line.

SP100 locating at human chromosome 2q37 encodes at least four isoforms through mRNA splicing. These isoforms share their N-terminal 476 amino acids, encoded by the first 15 exons (GenBank: NG_029842.1). SP100^−/− and PML/SP100^−/− clones were derived from wild-type (WT) HEp-2 cells or PML^−/− cell clone 1D2 (31) by transfection of a clustered regularly interspaced short palindromic repeats (CRISPR)/cas9 cassette targeting exon 3, 4, or 5 of human SP100. Experimental procedures are described in Materials and Methods. Single-cell–derived clones SP100^−/−CL5-3 and PML/SP100^−/−CL18-3 (simplified as SP100^−/− and PML/SP100^−/− hereinafter) were selected.

The knockout of SP100 from the two cell lines was validated in two sets of experiments. Because SP100 is an IFN-stimulated gene, all cell lines were exposed to IFN-β at 1,000 U/mL for 24 h to elevate SP100 expression. Cell lysates from HEp-2, SP100^−/−, and PML/SP100^−/− cells were solubilized, subjected to electrophoresis in denaturing gels, transferred to a nitrocellulose sheet, and reacted with rabbit polyclonal antibody anti-SP100 and antibody recognizing β-actin (Fig. 1A). Although a nonspecific band running at 100 kDa was present in all three samples, an SP100 protein band at 80 kDa was detected only in WT HEp-2 cells.

In the second set of experiments, cells were fixed and reacted with Texas Red-labeled polyclonal rabbit anti-SP100 antibodies (Fig. 1B). Uniform nuclear punctate structures (arrow in Fig. 1B, a) representing ND10 bodies were noted in HEp-2 cells, whereas PML^−/− cells contained an array of structures of different sizes. The arrow in Fig. 1B, c points to dense bodies larger than the ND10 bodies in HEp-2 cells. Nuclear structures reacting with anti-SP100 antibody were not detected in the SP100^−/− (Fig. 1B, b) or PML/SP100^−/− (Fig. 1B, d) cell line.

PML Accumulation Is Reduced in SP100^−/− Cells.

The results of two series of experiments suggest that the accumulation of PML is reduced in SP100^−/− cells, but that PML is induced and accumulates to high levels on exposure to IFN in both HEp-2 and SP100^−/− cells (39).

In the first series of experiments, replicate cultures of HEp-2 or SP100^−/− cells were mock-treated or exposed to IFN-β (1,000 U/mL). After 24 h, the cells were harvested, processed as described above, and reacted with anti-PML antibody. The results, shown in Fig. 2B, suggest that the accumulation of PML was reduced by as much as fourfold in SP100^−/− cells. Pretreatment of cells with IFN-β stimulated the accumulation of PML. Our previous study showed that IFN enhances the accumulation of SP100 mRNA in the absence of PML. The results presented here suggest that the synthesis of PML is stimulated by IFN in the absence of SP100, and thus both PML and SP100 are induced by IFN independently of each other, although the basal level of PML is affected by SP100.

In the second series of experiments, we examined the distribution of PML in intranuclear bodies. Fig. 2A presents images of representative cells from cultures fixed and reacted with anti-PML antibody. The striking feature of these images is the reduced amount of PML and number of nuclear bodies containing PML (compare the cells in a and b). Cultures treated with IFN-β (1,000 U/mL for 24 h) exhibited a large increase in dense structures of varying size (cells c and d).

Replication of WT Virus in SP100^−/− and SP100/PML^−/− Cells.

In this series of experiments, replicate cultures of HEp-2, PML^−/−, SP100^−/−, and PML/SP100^−/− were exposed to either 0.01 or 5 PFU of HSV-1(F) per cell. The yields were titered on Vero cells. The results, shown in Fig. 3, can be summarized as follows. The virus harvested from cultures exposed to 5 PFU/cell were within a twofold range and were not significantly different (Fig. 3A). The yields from cultures exposed to 0.01 PFU/cell varied depending on cell genotype (Fig. 3B). As would be expected based on previously reported results, the yields from PML^−/− cells were significantly lower than those obtained from HEp-2 cells [P = 0.017, comparison of yields obtained at 48 h postinfection (hpi)]. In contrast, the yields obtained from infected SP100^−/− cells were twofold to ninefold higher than those obtained from infected HEp-2 cells in repeat experiments (P = 0.011 for the representative experiment shown in Fig. 3B). The positive and negative effects of depletion of PML and SP100 appear to balance out, given that the yields from infected PML/SP100^−/− cells did not differ significantly from those obtained from HEp-2 cells.

These results suggest that in cells exposed to low ratios of virus per cell, a likely reflection of natural infections, the key constituents of ND10 bodies have contrasting roles. Thus, PML appears to be beneficial, whereas SP100 appears to have a negative effect on virus replication.

At a Low Ratio of Virus per Cell, ∆ICP0 Virus Replicates to Higher Titers in SP100^−/− Cells Than in Parental HEp-2 Cells.

ICP0 is a major multifunctional protein created immediately after infection. Among its major functions are the recruitment of CLOCK histone deacetylase to the viral transcriptome (40), the dissociation of the CoREST/REST/LSD1 repressor complex from its cognate sites on viral DNA (41, 42) and the degradation of PML and SP100 (11, 15, 43). ∆ICP0 mutants replicate in U2OS cells but poorly in numerous cell lines, including human (e.g., HEp-2) and African green monkey kidney (Vero) cells (44–46). The question posed here is whether the replication of ∆ICP0 virus is affected by PML or SP100. In this series of experiments, cultures of HEp-2, PML^−/−, SP100^−/−, and PML/SP100^−/− cells were exposed to 0.01 PFU of HSV-1(F) and ∆ICP0 mutant R7910. The cells were harvested at 48 hpi and titered on both U2OS and Vero cells.

As illustrated in Fig. 4, human and primate cell lines differ with respect to their ability to support HSV replication and, by extension, the formation of plaques. Thus, the titers of the viruses grown in the HEp-2 cells and in mutant cell lines varied depending on the cell lines in which the titrations were done. The titers of WT [HSV-1(F)] virus were higher in Vero cells than in the U2OS cell line by as much as 30- to 100-fold. In contrast, the yields of ΔICP0 mutant were 10–100 to higher in U2OS cells than in Vero cells. These results, which were reproduced in several experiments, suggest that the U2OS cells are more permissive to ∆ICP0 virus and at the same time restrict the replication of HSV-1(F).

The key result is the similar patterns of replication of HSV-1(F) in the four cell lines observed in Vero cells and U2OS cells. Specifically, the highest yields were obtained in SP100^−/− cells, and the lowest yields were obtained in PML^−/− cells. Although the yields of ΔICP0 measured in Vero cells were orders of magnitude lower than those of HSV-1(F), the basic patterns were similar. Again, the highest yields were obtained in SP100/^−/− cells. In U2OS cells, the yields were highest in SP100^−/− cells and nearly equal in HEp-2 and PML^−/− cells.

PML and SP100 Do Not Have Significant Roles in the Transcription of Viral mRNAs.

A central and as-yet unresolved question is whether PML or SP100 plays a role in the early transcription of viral genes after infection (14, 42, 47–51). Both PML and SP100 are degraded within the first 6 hpi, and thus the major effect of PML or SP100 on the transcription of viral DNA would occur during that period (11). The experiments designed to answer this question are based on the observation that the first new progeny viruses become apparent within the first 6 hpi. Implicit in this observation is that measurements done during the first 6 h after exposure of the virus to the cell would detect only transcripts accumulating during the first round of replication and, moreover, would capture the accumulation of transcripts made both early and late in the replicative cycle.

In this series of experiments, cultures of HEp-2, PML^−/−, SP100^−/−, and PML/SP100^−/− cells were exposed to 0.1 PFU of HSV-1(F) per cell. At 1, 3, and 5 hpi, the cells were harvested, and the mRNAs representative of α, (ICP27), β (TK), and ϒ (VP16) genes were assayed as described in Materials and Methods. The results, shown in Fig. 5, did not show any significant differences in the accumulation of various classes of mRNAs during the early stages of infection. These results indicate that the differences in virus replication in the cell lines studied in this report are not due to variation in the rates of accumulation of viral mRNAs.

Formation of Viral Replication Compartments Is Hampered in the Absence of PML and Facilitated in the Absence of SP100.

Viral DNA synthesis and some early stages of virion assembly occur in nuclear structures and, at least in the initial stages of formation, at ND10 bodies (9, 52, 53). In this series of experiments, we used immunostaining with an antibody to ICP8 to visualize the presence of replication compartments. ICP8 is viral protein that binds DNA and is essential for viral DNA synthesis (54–57).

To facilitate analyses of infected cells, we classified the appearance of ICP8 in infected cells into four categories, as shown in Fig. 6A. These images, collected at 6 hpi, were classified into uninfected or very early stage of infection (cell 1) or diffuse distribution of ICP8 from very faint to dense (cells 2 and 3) or forming very dense aggregates (cells 4 and 5). The aggregates seen in cells 4 and 5 were classified as replication compartments (34, 52, 58, 59).

We next quantified the frequency of occurrence of replication compartments in the various cell lines. In these experiments, cultures of HEp-2, PML^−/−, SP100^−/−, and PML/SP100^−/− cells were exposed to 20 PFU of HSV-1(F) per cell. At 6 hpi, the cells were fixed and reacted with antibodies against ICP0 (red) and ICP8 (green). At least 20 different fields comprising on average 200 positive cells for ICP8 in each of the four cell lines were photographed and classified as described above. Error bars were calculated based on data obtained in two independent experiments in which the cells were counted and classified three times. As shown in Fig. 6C, infected cells exhibiting replication compartments were readily detected in all four cell lines. The number of replication compartments shown in this figure is not representative of the cultures.

To combine three independent calculations from two sets of independently conducted experiments, the exact percentage of replication compartment-positive cells out of the total number of ICP8-positive cells of each cell line was normalized to that of HEp-2 cells (set as 1), combined, and plotted as shown in Fig. 6B. The data show that ablation of PML hampered the efficient formation of replication compartments (P = 0.001), whereas knockout of SP100 facilitated this process (P = 0.027). The difference between SP100^−/− and SP100/PML^−/− did not reach statistical significance (P = 0.065).

One hypothesis that could account for the results is that ICP8 accumulates at different rates in the four cell lines. To test this hypothesis, we exposed cultures of HEp-2, PML^−/−, SP100^−/−, and PML/ SP100^−/− cells to 5 PFU of HSV-1(F) per cell. The cells were harvested at 3, 5, and 8 hpi, processed as described above, and immunoblotted with antibody to ICP8 and β-actin. The results (Fig. 6D) indicate no correlation between the accumulation of ICP8 and the assembly of replication compartments in these cells.

The Degradation of ICP0 Early in Infection Is Unrelated to Its Role in the Degradation of SP100 or PML.

ICP0 functions as a ubiquitin ligase that in conjunction with ubiquitin-conjugating enzyme UbcH5a degrades PML and SP100 (11, 15, 43). A curious property of ICP0 is that the protein made early in infection coincident with the time course of degradation of SP100 and PML has a relatively short half-life, whereas that made at later times after infection is stable (60). The objective of the experiments reported herein was to test the hypothesis that the decay of ICP0 early in infection is related to its activity as an ubiquitin ligase in the degradation of PML or SP100. The design of these studies is illustrated schematically in Fig. 7A. In brief, replicate cultures of the four cell lines were exposed to 10 PFU of HSV-1(F) per cell for 2 h. At 3 hpi, the cells were rinsed and overlaid with medium containing cycloheximide (100 μg/mL). The cells were harvested at 3, 5, or 7 hpi, solubilized and processed as described above, and then immunoblotted with antibody against ICP0 and β-actin. Fig. 7B shows the accumulation of ICP0 in the four cell lines. Fig. 7C shows the amounts of ICP0 normalized with respect to loading control (β-actin) and the amount of ICP0 at 3 hpi (set as 100%). The results show that the rates of decay of ICP0 vary but that ultimately, by 7 hpi, the residual amounts of ICP0 are similar. The results do not support the hypothesis that the degradation of ICP0 is causally related to the degradation of PML, SP100, or both.

Viral Replication Is Suppressed in SP100^−/− Cells Pretreated with IFNβ.

Previous studies have shown that PML is a potent effector of IFN in that it is induced by IFN and that the suppression of viral replication in PML^−/− cells is significantly reduced compared with that observed in parental PML^+/+ cells (5, 31, 32, 61). Because SP100 is independently induced by IFN, the question arose concerning its role as an effector of IFN. In these experiments, replicate cultures of HEp-2, PML^−/−, SP100^−/−, and PML/SP100^−/− cells were incubated for 24 h in medium only or medium containing IFN-β (1,000 U/mL), then rinsed and exposed to 0.1 PFU of HSV-1(F). Fig. 8 shows the virus yields from cells harvested at 24 hpi. As expected, the reduction in virus yields was significantly lower in IFN-β–pretreated PML^−/− cells compared with parental HEp-2 cells. The reduction in virus yields from IFN-pretreated SP100^−/− cells was not significantly different from that seen in parent HEp-2 cells. These results suggest a different role of SP100 as an IFN effector compared with PML.

Cells and Viruses.

The sources and maintenance of HEp-2, Vero, and U20S cells have been reported elsewhere (31). PML^−/− cells, generated from HEp-2 cells using CRISPR/cas9 (31), were maintained in 5% FBS DMEM (Gibco) plus 2 μg/mL puromycin. HSV-1(F) is the prototype strain used in our laboratory, and ∆ICP0 mutant (R7910) was generated by deletion of both copies of ICP0 from HSV-1(F) (45). HSV-1(F) was amplified in HEp-2 cells, and ∆ICP0 mutant was propagated in U20S cells. For determination of virus yields, all infected cells were collected, resuspended in 2% milk, briefly sonicated to release all viral particles, and then titrated on Vero or U2OS monolayer cell cultures.

RNA Isolation and Real-Time PCR.

RNA extraction and assays were performed according to the manufacturer’s protocol (Thermo Fisher Scientific). mRNA levels of HSV-1(F) viral genes ICP27, VP16, and TK were measured using gene-specific primers as described previously (31). All targets were normalized to host internal RNA 18s (Ambion).

Generation of the SP100^−/− Cell Line Using CRISPR/cas9.

Three sets of CRISPR/cas9 systems were designed to target the N-terminal exons (exons 3, 4, and 5) of the human SP100 gene: gRNA 5′-TTGTGATGAGATCACGATCA-3′ and gRNA 5′-GCAGCCTGTCATCTACACCC-3′ on exon 3, gRNA 5′-TTGTACACCACTCTCTGTAC-3′ and gRNA 5′-GTACAGAGAGTGGTGTACAA-3′ on exon 4, and gRNA 5′-CTCCAACTAAGTCTTGAACA-3′ and gRNA 5′-TACCTTGTTCAAGACTTAGT-3′ on exon 5. Generation of SP100^−/− single-cell–derived lines were generated as described previously (31). In brief, parental HEp-2 or PML^−/− cells were cotransfected with one set of gRNAs (two gRNAs) and their donor vectors containing a selection cassette expressing GFP-blasticidine flanked by sequences homologous to the target site of the gRNAs. Surviving cells under blasticidine selection (2–10 μg/mL) were serially diluted to form single-cell–derived colonies. The colonies were further screened for expression of SP100 by immunoblotting and immunofluorescence staining.

Antibodies and Drugs.

The antibodies used in this study included monoclonal anti-PML (Santa Cruz Biotechnology; sc-966), rabbit polyclonal anti-PML (Abcam; 179466), rabbit polyclonal anti-SP100 (H-60) (Santa Cruz Biotechnology; sc-25568), rabbit polyclonal anti-SP100 (Abcam; ab4315), mouse monoclonal anti-Daxx (Abcam; ab9091), mouse monoclonal anti–β-actin (Thermo Fisher Scientific; MA5-15739), rabbit polyclonal anti-ICP0 (laboratory stock), and mouse monoclonal anti-ICP8, -US11, and -ICP4 (laboratory stock). The drugs used in this study included blasticidine (Sigma-Aldrich; 15205-25MG), puromycin (Sigma-Aldrich; p8833), and cycloheximide (Sigma-Aldrich; C7698-5G).

Immunoblot Analyses.

At the indicated times, the cells were lysed, heat-denatured, electrophoretically separated in denaturing gels, transferred to PVDF membranes, and probed with the appropriate primary antibody as described previously (31). The membranes were then incubated with alkaline phosphatase-conjugated goat anti-mouse or goat anti-rabbit secondary antibody (Sigma-Aldrich) and visualized with BCIP/NBT Western Blotting Detection Reagent (GE Healthcare); to detect PML protein, the membranes were incubated with horseradish peroxidase-conjugated goat anti-mouse (Sigma-Aldrich) and developed using ECL western blotting detection reagent (GE Healthcare).

ImageJ Quantification and Statistics.

ImageJ software was used to quantify band intensity. To quantify the relative PML protein level in HEp-2 and SP100^−/− cells shown in Fig. 2B, the intensity of the major band of PML (marked by a rectangle) was normalized to the loading control (β-actin), and the relative protein amount of PML in SP100^−/− cells was normalized to that in HEp-2 (set as 100) under each condition (untreated and treated with IFN-β). For measuring ICP0 protein stability in early infection, the ICP0 protein band intensity in Fig. 7B was quantified, normalized to β-actin, and normalized to the relative ICP0 level at 3 h postinfection (hpi) (set as 100%).

Statistics.

Error bars represent SE/SD of the mean. Two-tailed P values were calculated using a standard t test.

Immunofluorescence Staining and Confocal Microscopy.

In brief, cells in four-well slides were fixed and permealized, and then blocked and incubated with primary antibodies at 4 °C overnight. The next day, they were rinsed three times with blocking solution and then subjected to reaction with FITC-conjugated goat anti-mouse (Invitrogen) or/and Texas Red-conjugated goat anti-rabbit (Sigma-Aldrich) secondary antibodies. Images were obtained with a Leica SP5 II STED-CW super-resolution laser scanning confocal microscope.

Assessment of Replication Compartment Formation.

HEp-2, PML^−/−, SP100^−/−, and PML/SP100^−/− cells were seeded onto microslides at 24 h before infection, then exposed to 20 PFU of HSV-1(F) per cell. At 6 hpi, the cells were fixed and reacted with anti-ICP8 and anti-ICP0 antibodies. On average, 20 different fields were photographed at random at either 40× or 60× magnification for each of the four cell lines. Approximately 200 ICP8-positive cells were counted for each cell line, and three independent countings were conducted for two independent experiments. The percentage of replication compartments forming cells out of the total number of ICP8-positive cells was calculated according to the classification scheme shown in Fig. 6A and normalized to that of HEp-2 cells. Error bars represent SE/SD of the mean.