Deletion of pUL16 or pUL21 results in capsid accumulation at the cytoplasmic face of the nuclear envelope late in infection

Vero cells were infected with HSV-2 strains containing mCherry (mCh) fused to the capsid protein VP26, stained for the nuclear lamina protein, lamin A/C, and DNA (Hoechst 33342), and examined by confocal microscopy at 18 hours post infection (hpi). Unlike what was observed in cells infected with the WT mCh-VP26 strain, cells infected with mCh-VP26 viruses deleted for pUL16 (Δ16), or pUL21 (Δ21), displayed an aberrant accumulation of mCh-VP26 fluorescence at the nuclear periphery (Fig 1A, arrowheads). Measurement of fluorescence intensities across the nuclear envelope demonstrated that, as expected, the peak DNA fluorescence intensity (Hoechst 33342) was inside the peak of nuclear lamina (lamin A/C) fluorescence (Fig 1B, 1C and 1D). Interestingly, the peak mCh-VP26 fluorescence intensity in Δ16 and Δ21 infected cells localized to the outside of the nuclear lamina suggesting that, unlike WT capsids (Fig 1B), Δ16 and Δ21 capsids were accumulating on the cytoplasmic face of the nuclear envelope (Fig 1C and 1D).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. The localization of HSV-2 186 mCh-VP26 WT, Δ16, and Δ21 capsids late in infection.

Vero cells were infected with HSV-2 186 mCh-VP26 WT, Δ16, or Δ21 virus at an MOI of 0.1 and were fixed at 18 hpi. After fixation, cells were permeabilized with TX-100 and stained for lamin A/C. A) Representative confocal images of HSV-2 infected cells. Merge shows the overlay of Hoechst 33342 (DNA) (blue), lamin A/C (green), and HSV-2 capsids (red). The scale bar indicates 5μm. Arrowheads indicate sites that displayed an aberrant accumulation of mCh-VP26 fluorescence at the nuclear periphery. The white lines indicate the location of fluorescence intensity measurements shown in B, C and D. B, C and D) Representative fluorescence intensity measurements for Hoechst, lamin A/C, and mCh-VP26 capsids in a WT (B) Δ16 (C) and Δ21 (D) infected cell starting within the nucleus (distance 0) and extending into the cytoplasm (n = 10 individual cells examined per condition). Peak intensities are indicated by the vertical black lines.

https://doi.org/10.1371/journal.ppat.1011832.g001

To verify that these observations were not due to the mCh-VP26 fusion altering capsid properties, we monitored capsid localization in cells infected with viruses expressing unfused VP26 using antisera against the HSV major capsid protein, VP5. Similar to what was seen using mCh-VP26 capsids, confocal microscopy of cells infected with HSV-2 Δ16, and Δ21 strains showed capsid accumulations at the nuclear periphery whereas cells infected with the HSV-2 WT strain did not (Fig 2A).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. The localization of HSV-2 and HSV-1 capsids in the absence of pUL16 and/or pUL21.

Vero cells were infected with either HSV-2 186 or HSV-1 KOS WT, Δ16, Δ21, or Δ21/Δ16 at an MOI of 1 and fixed at 18 hpi. After fixation, samples were permeabilized with TX-100, or saponin, and stained for the major capsid protein VP5. Merge in confocal images shows the overlay of Hoechst 33342 (blue) and HSV-2, or HSV-1, capsids (red). A and B) Representative confocal images of TX-100 permeabilized HSV-2 and HSV-1 infected cells, respectively, stained for VP5. Scale bars indicate 5μm. Arrowheads indicate nuclei with capsids at the nuclear envelope. C) Representative confocal images of saponin permeabilized HSV-1 infected cells stained for VP5. The scale bar in the lower magnification images is 20μm. The scale bar for zoom panels is 5μm. Arrowheads indicate nuclei with capsids at the nuclear envelope. Insets show the absence (WT panel) or presence (Δ16, Δ21, or Δ21/Δ16 mutant panels) of VP5 capsid fluorescence at the nuclear envelope. D) The quantification of the percentage of HSV-1 infected cells with capsids accumulated at the nuclear envelope. Three biological replicates with n = 198–258 infected cells examined per biological replicate. A one-way ANOVA with Tukey’s HSD Test for multiple comparisons was performed between all viruses. **** represents p ≤ 0.0001.

https://doi.org/10.1371/journal.ppat.1011832.g002

We next asked if the accumulation of Δ16 and Δ21 capsids at the nuclear periphery was restricted to HSV-2, or if it would also be applicable to HSV-1 mutants. Vero cells were infected with HSV-1 WT, Δ16, Δ21, or a pUL21/pUL16 double mutant (Δ21/Δ16), and capsids were visualized by staining for the HSV major capsid protein, VP5. Similar to what was observed with HSV-2 mutant strains, HSV-1 Δ16, Δ21, as well as Δ21/Δ16 infected cells, accumulated capsids at the nuclear periphery (Fig 2B).

To verify that capsid accumulation was occurring at the cytoplasmic face of the nuclear envelope rather than at its nucleoplasmic face, we utilized saponin to permeabilize infected cells prior to staining the cells with VP5 antisera. Saponin permeabilizes the plasma membrane but does not permeabilize the nuclear envelope [59], enabling VP5 antisera to bind cytoplasmic capsids, while nuclear capsids are not detected because they are inaccessible to the antisera. We frequently observed cytoplasmic capsids surrounding the nuclei of HSV-1 Δ16, Δ21, and Δ21/Δ16 infected cells whereas WT infected cells rarely displayed this capsid localization pattern (Fig 2C). Importantly, despite the abundance of capsids in the nuclei of virally infected cells (e.g., Fig 1A), no VP5 signal was evident within the nuclei of cells permeabilized with saponin, confirming that the nuclear envelope had not been breached in these cells. The percentage of cells with an abundance of capsids at the cytoplasmic face of the nuclear envelope was quantified. HSV-1 Δ16, Δ21, and Δ21/Δ16 infections had significantly more cells with capsids at the nuclear envelope in comparison to WT infected cells (Fig 2D). Interestingly, Δ21/Δ16 infected cells had significantly more cells with capsid accumulation at the nuclear envelope in comparison to Δ16 or Δ21 infected cells suggesting that the deletion of both pUL16 and pUL21 has an additive effect on capsid accumulation at the nuclear envelope.

Capsids accumulating at the cytoplasmic face of the nuclear envelope colocalize with NPCs

Considering that cytoplasmic Δ16 and Δ21 capsids accumulate at the periphery of infected nuclei, we asked if these capsids colocalize with NPCs. To examine this, we infected Vero, life-extended human foreskin fibroblasts (T12), HeLa, and HaCaT cells with HSV-2 mCh-VP26 viruses, fixed cells at 18 hpi and stained for NPCs. In all cell types, HSV-2 mCh-VP26 Δ16 and Δ21 capsids, but not WT capsids, colocalized with NPCs (Figs 3A and S1). The colocalization of Δ16 and Δ21 capsids with NPCs was also assessed by measuring the fluorescence intensity of the capsids and NPCs across the nuclear envelope of infected Vero cells (Fig 3B and 3C). In this analysis, the peak of capsid fluorescence intensity and NPC fluorescence intensity coincided raising the possibility that capsids were docked at NPCs. As a complementary approach, HSV-1 WT, Δ16, Δ21 and Δ21/Δ16 infected Vero cells were analyzed by transmission electron microscopy (TEM). HSV-1 Δ16, Δ21, and Δ21/Δ16 capsids were commonly seen docked at NPCs at 18 hpi by TEM, but rarely in WT infected cells (Fig 3D and 3E). Additionally, capsids docked at NPCs late in infection uniformly lacked genomes suggesting that these capsids have ejected their genomes back into the infected nuclei. These findings indicate that pUL16 and pUL21 are important for preventing capsids from docking to NPCs at late stages of infection, irrespective of HSV species or cell type.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. The colocalization of HSV-2 capsids with NPCs at late times in infection.

Vero cells were infected with HSV-2 186 mCh-VP26 WT, Δ16, or Δ21 virus at an MOI of 0.1 and fixed at 18 hpi. After fixation, cells were permeabilized with TX-100 and stained for NPCs. A) Representative confocal images of HSV-2 infected Vero cells. Merge shows the overlay of Hoechst 33342 (DNA) (blue), NPCs (green), and HSV-2 capsids (red). The scale bar indicates 5μm. Arrowheads indicate cells with mCh-VP26 capsid fluorescence colocalized with NPC fluorescence. The white lines indicate the location of fluorescence intensity measurements shown in B and C. B and C) Representative fluorescence intensity measurements for Hoechst, NPCs, and HSV-2 capsids in a Δ16 and Δ21 infected Vero cell, respectively, starting within the nucleus (distance 0) and extending into the cytoplasm (n = 10 individual cells examined per condition). Peak intensities are indicated by the vertical black lines. D) TEM micrographs of capsids in HSV-1 infected cells. White arrowheads indicate capsids docked at NPCs and black arrowheads point to the nucleoplasmic face of NPCs in the zoomed inset panels. E) Quantification of the number of A-capsids docked at NPCs per nuclear section. n = 10–26 nuclear sections examined per condition. A one-way ANOVA with Tukey’s HSD Test for multiple comparisons was performed between all viruses. * and **** represents p ≤ 0.05 and p ≤ 0.0001, respectively.

https://doi.org/10.1371/journal.ppat.1011832.g003

Superinfection exclusion is impaired at the level of viral genome delivery in Δ16 and Δ21 infected cells

Similar to the findings described in Fig 3D, an analysis of the HSV-1 temperature-sensitive mutant, 50B, by Roizman and colleagues revealed the accumulation of empty capsids at the cytoplasmic face of NPCs late in infection [60]. These authors suggested that the 50B strain bore mutations in viral glycoprotein genes that were responsible for conferring resistance of infected cells to superinfection by progeny virions; a process called superinfection exclusion (SIE). Thus, the origin of the empty capsids accumulating at NPCs was suggested to be from re-infection of cells by progeny virions, leading to the delivery of capsids to NPCs and ejection of their genomes into the infected cell nucleoplasm.

To assess whether SIE was altered in cells infected with pUL16 and pUL21 mutant strains, we measured the entry of a second, superinfecting virus and the delivery of capsids to the nuclear envelope. Vero cells were mock infected or infected with HSV-2 186 WT, Δ16, or Δ21 viruses. At 6 hpi, cells were superinfected with WT HSV-2 186 mCh-VP26 virus. One hour after superinfection, cells were stained with fluorescent phalloidin to visualize actin at the cell periphery and the entry of mCh-VP26 capsids into cells was quantified by confocal microscopy (Fig 4A). Importantly, when HSV-2 186 mCh-VP26 virus was absorbed to cells on ice, immediately fixed, and stained with phalloidin, the majority of mCh-VP26 capsids were observed at the cell surface (S2A and S2B Fig). HSV-2 infected cells had significantly fewer cytoplasmic superinfecting mCh-VP26 capsids compared to the mock infected cells (Fig 4B). These findings suggest that HSV-2 infected cells display some degree of SIE at the point of viral entry; however, there was no significant difference in the numbers of mCh-VP26 capsids in the cytoplasm of Δ16 and Δ21 infected cells compared to WT infected cells. This suggests that Δ16 and Δ21 infected cells inhibit entry of a superinfecting virus comparably to WT infected cells. Additionally, the localization of superinfecting capsids to the nuclear periphery was similar in all conditions tested, suggesting that the transport of capsids from the cell periphery towards the nucleus was unaffected (Fig 4C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Examination of SIE in HSV-2 WT, Δ16 and Δ21 infected cells.

A) Representative z-projections of cells that were initially mock infected or infected with HSV-2 186 WT, Δ16 or Δ21 virus at an MOI of 1 and at 6 hpi superinfected with HSV-2 186 mCh-VP26 at an MOI of 3 for one hour. z-projections show the overlay of Hoechst 33342 (DNA) (blue), phalloidin staining of actin filaments (green) and HSV-2 186 mCh-VP26 capsids (red). The scale bar is 10μm. B) Quantification of cytoplasmic mCh-VP26 capsids within mock, WT, Δ16 and Δ21 infected cells. n = 10 z-projections of infected cells per condition. C) Quantification of the percentage of mCh-VP26 capsids within the nuclear periphery of mock, WT, Δ16 and Δ21 infected cells. n = 10 z-projections of infected cells per condition. Capsids at the nuclear periphery were defined as those abutting the Hoechst 33342 signal. D) Representative confocal images of HSV-2 186 WT, Δ16 and Δ21 infected cells infected with HSV-2 186 WT EdC labelled virus at 6 hpi and fixed at 3h after WT EdC infection. Fluorescent puncta from WT EdC ejected genomes are seen in the nuclei of WT, Δ16 and Δ21 infected cells as indicated by the arrowheads in the figure. The scale bar indicates 20μm. E) Quantification of nuclear WT EdC labelled genomes within WT, Δ16 and Δ21 infected cells. n = 177–470 infected cells examined for fluorescent EdC puncta per biological replicate. One-way ANOVAs with Tukey’s HSD Test for multiple comparisons were performed between all viruses. **** represents p ≤ 0.0001.

https://doi.org/10.1371/journal.ppat.1011832.g004

Next, we measured the efficiency with which superinfecting capsids could dock to NPCs and eject their genomes into the nucleoplasm of infected cells. To do this, Vero cells were first infected with HSV-2 186 WT, Δ16, or Δ21 strains and at 6 hpi were superinfected with HSV-2 186 WT virus that had the nucleoside analog, 5-ethynyl-2’-deoxycytidine (EdC), incorporated into its genome. Three hours after superinfection, click-chemistry was performed to label the EdC with AlexaFluor-488 to enable the visualization of viral genomes that had been delivered to the nucleoplasm by superinfecting virions. Interestingly, Δ16 and Δ21 infected cells had significantly more nuclear genomes from the superinfecting virions compared to cells that had been initially infected with the WT strain (Fig 4D and 4E). Importantly, when HSV-2 186 WT EdC virus was absorbed to cells at 4°C, immediately fixed, and click-chemistry was performed, no nuclear EdC-labelled viral genomes were observed (S2C Fig). Consistent with the results shown above, these data suggest that the expression of pUL16 and pUL21 during infection prevents genome ejection into infected nuclei from superinfecting nucleocapsids.

Analysis of pUL16 and pUL21 prevention of capsid recruitment to NPCs

We propose three mechanisms that could explain how pUL16 and pUL21 prevent capsid docking to NPCs. First, pUL16 and/or pUL21 bound to cytoplasmic capsids may mask capsid proteins responsible for NPC docking. Second, pUL16 and/or pUL21 may influence capsid interactions with microtubules and/or microtubule motor proteins that enable their efficient transport away from the nuclear periphery towards sites of secondary envelopment. Finally, pUL16 and/or pUL21 may bind to, or alter, NPCs to prevent capsid docking.

During HSV infection, pUL16 and pUL21 are not expressed until between 4 and 6 hpi when nuclear egress and capsid transport to sites of secondary envelopment are also occurring [61]. Since incoming capsids dock at NPCs before new pUL16 and pUL21 are synthesized, we determined if expression of pUL16 and pUL21 prior to infection blocked incoming capsids from being recruited to the nuclear envelope. HeLa and PK15 cells were transfected with HSV-2 pUL16 or pUL21 expression plasmids, alone or in combination. At 24 hours post transfection, the recruitment of incoming HSV-2 or pseudorabies virus (PRV) capsids to the nuclear periphery was monitored by measuring the distance between mCh-VP26 capsids and the perimeter of the Hoechst 33342 signal (Fig 5A–5C). In these experiments, co-transfection of an EGFP expression plasmid enabled the identification of transfected cells. Interestingly, expression of pUL16 or pUL21 alone did not significantly affect the average distance of incoming HSV-2 or PRV capsids from the nuclear periphery in comparison to control cells that had been transfected with an EGFP expression plasmid alone (Fig 5B and 5C). Only the co-expression of pUL16 and pUL21 significantly increased the average distance of HSV-2 and PRV capsids from the nuclear periphery raising the possibility that a pUL16/pUL21 complex may be required to prevent capsid recruitment.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. The effect of prior ectopic HSV-2 pUL16 and pUL21 expression on HSV-2 and PRV capsid recruitment to nuclei and subsequent virus gene expression.

A) Representative images of HeLa cells transfected with EGFP (control) or HSV-2 pUL16 and pUL21 expression plasmids 24 hours prior to infection with HSV-2 186 mCh-VP26. Cells were imaged at 2 hpi. Hoechst 33342 (DNA) (blue), HSV-2 186 mCh-VP26 capsids (red). Scale bar is 10μm. B) Average distance of HSV-2 capsids from the nuclear periphery in HeLa cells that had been transfected with EGFP alone (control) or in combination with HSV-2 pUL16 and/or pUL21 expression plasmids and infected for 2h with HSV-2 186 mCh-VP26. n = 43–55 capsids were examined in each transfection condition. A one-way ANOVA with Tukey’s HSD Test for multiple comparisons was performed between all transfection conditions. **** represents p ≤ 0.0001. C) Average distance of PRV capsids from the nuclear periphery in PK15 cells that had been transfected with EGFP alone (control) or in combination with HSV-2 pUL16 and/or pUL21 expression plasmids and infected for 2h with PRV765. A one-way ANOVA with Tukey’s HSD Test for multiple comparisons was performed between all transfection conditions. **** represents p ≤ 0.0001. n = 46–60 capsids were examined in each transfection condition. D) Confocal images showing ICP27 expression in cells transfected with EGFP alone (control) or in combination with HSV-2 pUL16 and pUL21 expression plasmids and infected for 2h with HSV-2 186 mCh-VP26. Arrows indicate cells expressing ICP27. Arrowheads indicate transfected and infected cells lacking ICP27 expression (mCh signal). The inset image shows mCh-VP26 capsids inside a cell, that does not express ICP27 and has been co-transfected with pUL16 and pUL21 expression plasmids. The scale bar is 20μm. E and F) Quantification of the percentage of transfected cells positive for HSV-2 ICP27 (E) or PRV Us3 expression (F) at 2 hpi. n = 40–110 transfected cells examined per biological replicate. Cells positive for ICP27 and Us3 expression were scored by visualization directly under the microscope. A one-way ANOVA with Tukey’s HSD Test for multiple comparisons was performed between all transfection conditions. *, **, and **** represent p ≤ 0.05, p ≤ 0.01, and p ≤ 0.0001, respectively.

https://doi.org/10.1371/journal.ppat.1011832.g005

As a complementary approach, viral protein expression was examined to determine if incoming capsids were successfully delivering their genomes into nuclei when pUL16 and/or pUL21 were expressed prior to infection. Only the co-expression of pUL16 and pUL21 significantly decreased the number of cells producing the HSV-2 immediate early protein ICP27 in comparison to the EGFP control (Fig 5D and 5E). Importantly, mCh-VP26 capsids were visible in the cytoplasm of co-transfected cells lacking ICP27 expression indicating that virus had entered these cells (Fig 5D, inset). Ectopic expression of pUL16 alone and in combination with pUL21 significantly decreased the number of cells expressing the PRV Us3 protein in comparison to the EGFP control (Fig 5F). These data suggest that co-expression of both pUL16 and pUL21 prior to infection effectively prevents capsids from docking at NPCs and the subsequent delivery of viral genomes to the nucleus. Because co-expression of HSV-2 pUL16 and pUL21 also disrupted the recruitment of the distantly related PRV capsids to the nuclear periphery, the data suggest that ectopic expression of pUL16 and pUL21 interfere with nuclear envelope factors that influence capsid recruitment rather than blocking NPC-interaction surfaces on capsids.

To explore this idea further, we investigated if ectopically expressed pUL16-EGFP and pUL21-EGFP could be recruited to incoming capsids. HeLa cells were transfected with combinations of plasmids encoding pUL16, pUL21 and their EGFP fusions. At 24 hours post transfection, cells were infected with HSV-2 WT mCh-VP26, and the EGFP signal associated with mCh-VP26 capsids at 2 hpi was evaluated by confocal microscopy (Fig 6A and 6B). No significant differences in capsid-associated EGFP signal were observed in cells transfected with pUL16-EGFP, pUL21-EGFP or EGFP expression plasmids (Fig 6B). Additionally, we noted that when pUL16-EGFP was expressed alone, it had a pancellular distribution; however, when co-expressed with pUL21, a fraction of pUL16-EGFP also localized to the nuclear envelope (Fig 6A, inset). This finding further supports the idea that pUL16 and pUL21 function together at the nuclear envelope to prevent capsid recruitment.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. The colocalization of pUL16-EGFP and pUL21-EGFP with HSV-2 mCh-VP26 capsids.

A) HeLa cells were transfected with EGFP, pUL16-EGFP, pUL21-EGFP, pUL16-EGFP and pUL21, or pUL21-EGFP and pUL16 expression plasmids. At 24 hours post transfection, transfected cells were infected on ice with HSV-2 186 mCh-VP26 at a MOI of 3 then shifted to 37°C and fixed at 2 hpi. Scale bar indicates 5μm. Arrowheads indicate cells with pUL16-EGFP/pUL21-EGFP nuclear rim staining. B) Colocalization between EGFP signal and mCh-VP26 capsid signals was compared to the average EGFP background signal taken from two points 0.5μm from the analyzed capsid. n = 40 capsids per transfection condition were examined. A one-way ANOVA with Tukey’s HSD Test for multiple comparisons was performed between all transfection conditions. No significant differences were seen between any of the transfection conditions when comparing the EGFP signal associated with mCh-VP26 capsids to the EGFP control. C and D) HeLa cells were transfected with EGFP, pUL16-EGFP or pUL21-EGFP expression plasmids. At 24 hours post transfection, cells were infected with HSV-2 186 mCh-VP26 WT, Δ16 or Δ21 virus at a MOI of 3 and fixed at 18 hpi. Fixed cells were permeabilized and stained for NPCs. Δ16 and Δ21 capsids colocalized with NPCs only in EGFP transfected cells (grey arrowheads). White arrowheads and arrows depict EGFP signal colocalized with mCh-VP26 capsid signal in the nucleus and cytoplasm, respectively. Scale bars indicate 5μm.

https://doi.org/10.1371/journal.ppat.1011832.g006

To confirm that the pUL16 and pUL21 EGFP fusion proteins had the capacity to interact with capsids, we showed that ectopically expressed pUL16-EGFP and pUL21-EGFP colocalized with mCh-VP26 capsids at late times post infection (Fig 6C and 6D). In agreement with previous reports, pUL21 was associated with capsids in both the nucleus and cytoplasm, whereas pUL16 was associated with cytoplasmic capsids but not nuclear capsids [58, 62–64] (Fig 6C and 6D). Consistent with experiments shown above, Δ16 and Δ21 capsids colocalized with NPCs only in EGFP transfected cells (Fig 6C and 6D, grey arrowheads). Collectively, the data shown in Fig 6 suggest that pUL16 and pUL21 are not loaded onto incoming capsids when ectopically expressed and it is, therefore, unlikely that the binding of pUL16 and pUL21 to capsids prevents their recruitment to the nuclear envelope.

After nuclear egress, cytoplasmic capsids are transported away from the nuclear periphery to membranes that serve as sites of secondary envelopment. HSV-1 and PRV pUL21 interact with components of the microtubule-transport machinery raising the possibility that pUL21, or a pUL16/pUL21 complex, may enhance nascent capsid transport [58, 65]. Thus, we hypothesized that in the absence of pUL16 and pUL21, nascent cytoplasmic capsids might linger at the nuclear periphery, providing a greater opportunity for these capsids to dock at NPCs. If this were the case, we predicted that disruption of microtubule-based transport with nocodazole might enable nascent WT capsids to dock at NPCs.

Nocodazole reversibly interferes with microtubule polymerization [66]. Treating cells with this agent disrupted the microtubule network (Fig 7A) and prevented the delivery of incoming HSV capsids to NPCs and subsequent viral gene expression (Fig 7B). Cells were infected with WT, Δ21, or Δ16 HSV-2 strains bearing mCh-VP26 capsids and at 6 or 12 hpi, treated with nocodazole or vehicle (DMSO) until 18 hpi. Cells were stained for NPCs and colocalization between mCh-VP26 capsids and NPCs assessed by confocal microscopy (Fig 7C and 7D). Whereas mCh-VP26 capsids colocalized with NPCs in Δ16 and Δ21 infected cells, WT infected cells showed limited colocalization of capsids with NPCs either in the presence or absence of nocodazole suggesting that microtubule de-polymerization does not enhance nascent cytoplasmic capsid recruitment to NPCs (Fig 7C–7E). Importantly, the efficient colocalization of Δ21 and Δ16 mCh-VP26 capsids with NPCs in the presence of nocodazole suggests that the origin of these capsids was the infected cell nucleus rather than from superinfecting virions that would neither be efficiently released from cells, nor be able to reach the nuclear envelope in the absence of a microtubule network.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 7. The colocalization of HSV capsids with NPCs in cells treated with nocodazole.

A) Representative confocal images of mock infected cells treated with DMSO or 10μM nocodazole for 6 hours. Merge shows the overlay of Hoechst 33342 (DNA) (blue) and α-tubulin (red). The scale bar indicates 10μm. B) Representative confocal images of cells treated with DMSO or 10μM nocodazole for 30 mins prior to infection on ice for 1 hour at an MOI of 3. Samples were shifted to 37°C and fixed at 3 hpi. Merge shows the overlay of Hoechst 33342 (DNA) (blue), ICP27 (green) and mCh-VP26 capsids (red). Arrowheads indicate cells with nuclear ICP27 expression. As seen in the nocodazole-treated cells, expression of ICP27 is absent as capsid transport to the nucleus is prevented. The scale bar indicates 10μm. C and D) Representative confocal images of HSV-2 infected cells treated with DMSO or nocodazole at 6 hpi or 12 hpi, respectively and fixed at 18 hpi. Merge shows the overlay of NPCs (green) and HSV-2 mCh-VP26 capsids (red). Scale bars indicate 5μm. Arrowheads indicate cells with mCh-VP26 capsid signal colocalized with NPC signal. E) Quantification of nuclei with capsids colocalizing with NPCs in HSV-2 infected cells treated with DMSO or nocodazole at 12 hpi and fixed at 18 hpi. Three biological replicates with n = 100–140 infected cells examined per biological replicate. A two-way ANOVA with Tukey’s HSD Test for multiple comparisons was performed between all viruses and treatment conditions. *, **, *** and **** represent p ≤ 0.05, p ≤ 0.01, p ≤ 0.001 and p ≤ 0.0001, respectively.

https://doi.org/10.1371/journal.ppat.1011832.g007

The data so far suggest that the expression of pUL16 and pUL21 prior to infection prevents incoming capsids from docking to NPCs (Fig 5), and that when pUL16 and pUL21 are co-expressed, they localize to the nuclear envelope (Fig 6A). Additionally, previous work from our own and other laboratories demonstrated that a population of pUL21 exhibits a nuclear rim like localization in HSV-1 and HSV-2 infected cells [37, 43]. Considering this, we wanted to explore if pUL16 and pUL21 localize to the cytoplasmic face of nuclei, placing them in an appropriate location to interfere with nucleocapsid/NPC interactions. Vero cells were infected with an HSV-2 186 recombinant virus expressing mCh fused to the carboxy terminus of pUL21 (pUL21-mCh) for 18 h. Stimulated emission depletion (STED) microscopy was utilized to examine the localization of pUL21-mCh with respect to lamin A/C and RanBP2 to further understand pUL21 localization within the nuclear envelope. Lamin A/C is a component of the nuclear lamina and found adjacent to the inner nuclear membrane. RanBP2 forms the NPC filaments found on the cytoplasmic face of NPCs and is important for HSV nucleocapsid docking to NPCs [17, 20]. Cross sections through the centre of an infected cell nucleus showed pUL21-mCh in the same plane as lamin A/C and RanBP2; suggesting that pUL21-mCh localizes to both the nuclear and cytoplasmic faces of the nuclear envelope (Fig 8A and 8B, cross section). When the top of the infected cell nucleus was examined to visualize pUL21-mCh in respect to RanBP2, pUL21-mCh was colocalized (arrowheads) and adjacent to RanBP2, however, the majority of pUL21-mCh was seen in a lattice-like distribution on the nuclear envelope (Fig 8B, top). Interestingly, the spaces between pUL21-mCh signals were 120nm to 180nm in size, consistent with the size of NPCs (~120nm in diameter) [26] (Fig 8B, 10X zoom). As a complementary approach, we investigated the localization of pUL21-mCh in relation to other NPC components by staining infected cells with MAb414 that reacts with the FG-repeats found in several nucleoporins [67]. Similar to the findings in RanBP2 stained cells, pUL21-mCh localized in a lattice-like distribution in the nuclear envelope and was localized adjacent to, and with (arrowheads), NPC components in both cross section and at the nuclear surface (Fig 8C). These data suggest that, during infection, pUL21 is localized to the nuclear envelope and is in proximity to NPC components.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 8. pUL21 localization during infection.

Cells were imaged by confocal microscopy and stimulated emission depletion (STED) microscopy. A) Cross section of an HSV-2 186 pUL21-mCh infected nuclei stained with antisera against mCh and lamin A/C is shown. B and C) Cross section and top views of HSV-2 186 pUL21-mCh infected nuclei stained with antisera against mCh and RanBP2 (B) or MAb414 (C) that reacts with multiple NPC components. The 2.5X and 10X zoomed regions are indicated in the merge panels. Scale bars in the lower magnification images are 5μm. Scale bars in the 2.5X and 10X zoom images are 1μm. Arrowheads indicate pUL21-mCh signal colocalized with RanBP2 or NPC signal. Measurements in 10X zoom panels indicate distances between pUL21-mCh signals.

https://doi.org/10.1371/journal.ppat.1011832.g008

Since a fraction of pUL16 localized to the nuclear envelope by confocal microscopy, when pUL21 was co-expressed (Fig 6A), we investigated the localization of pUL16 within the nuclear envelope at higher resolution by STED microscopy. Because our pUL16 antisera does not work in fluorescence microscopy applications, we examined pUL16-mCh localization when co-expressed with pUL21 in transfected cells. Vero cells were transfected with pUL16-mCh and pUL21, or with pUL21-mCh and pUL16 expression plasmids. At 24 hours post transfection, samples were fixed and stained for mCh and NPCs (MAb414). Similar to pUL21 localization in infected cells (Fig 8), pUL16-mCh localized in the same plane as NPCs and in a lattice-like distribution in the nuclear envelope (Fig 9A). Additionally, pUL16-mCh was seen adjacent to and colocalized (arrowheads) with NPC components (Fig 9A). pUL16-mCh localization in respect to lamin A/C was also examined (S3 Fig). A fraction of pUL16 localized in the same plane as lamin A/C, suggesting that pUL16 also localizes to the nucleoplasmic face of the nuclear envelope (S3 Fig). Importantly, when pUL16-mCh was expressed in cells in the absence of pUL21, pUL16 did not localize to the nuclear envelope and had a pancellular distribution (S3 Fig). In cells co-expressing pUL21-mCh and pUL16, pUL21 was seen adjacent to and colocalized with NPC components in a manner indistinguishable from what was observed in infected cells (Figs 9B and 8C). Collectively, these data suggest that both pUL16 and pUL21 are in proximity to NPCs and could interfere with NPC components that mediate capsid docking. A model describing how pUL16 and pUL21 might prevent nascent cytoplasmic nucleocapsids from engaging NPCs is shown in Fig 10.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 9. pUL16 and pUL21 localization in transfected cells.

Cells were imaged by confocal microscopy and stimulated emission depletion (STED) microscopy. Cross section and top views of nuclei co-transfected with pUL16-mCh and pUL21 expression plasmids (A) or pUL21-mCh and pUL16 expression plasmids (B). Cells were stained with antisera against mCh and MAb414 that reacts with multiple NPC components. The 2.5X and 10X zoomed regions are indicated in the merge panels. Scale bars in the lower magnification images are 5μm. Scale bars in the 2.5X and 10X zoom images are 1μm. Arrowheads indicate pUL16-mCh (A) and pUL21-mCh (B) signal colocalized with the NPC signal. Measurements in 10X zoom panels indicate distances between pUL16-mCh/pUL21-mCh signals.

https://doi.org/10.1371/journal.ppat.1011832.g009

Download:

• PNG
  larger image
• TIFF
  original image

Fig 10. Model describing the role of pUL16 and pUL21 in preventing nascent nucleocapsid recruitment to NPCs.

A) To initiate infection, the virion envelope fuses with a cellular membrane, such as the plasma membrane (PM), and the nucleocapsid and virion-associated tegument are released into the cytoplasm. Most of the virion-associated tegument dissociates from the nucleocapsid and the capsid is transported along microtubules towards the microtubule organizing center (MTOC) using the minus-end directed motor protein dynein. After reaching the MTOC, nucleocapsid transport to the nucleus is facilitated by a plus-end directed kinesin motor protein family member. Once at the nuclear periphery, nucleocapsids dock at NPCs, triggering the delivery of the viral genome into the nucleoplasm. B) During virion assembly, nucleocapsids transit from the nucleoplasm to the cytoplasm by acquiring a primary envelope at the inner nuclear membrane (INM) followed by fusion of this primary envelope with the outer nuclear membrane (ONM). Once in the cytoplasm, nucleocapsids are transported away from the nuclear periphery towards the MTOC using dynein. Nucleocapsid associated pUL21, acquired during capsid maturation in the nucleoplasm, may aid the recruitment of dynein to the nucleocapsid. pUL16 and pUL21 form a lattice-like array on the cytoplasmic face of the ONM where these proteins are adjacent to, and co-localized with, NPC components. We hypothesize that the localization of pUL16 and pUL21 to the cytoplasmic face of the ONM prevents nascent cytoplasmic nucleocapsids from docking to NPCs and ejecting their genomes into the nucleoplasm. C) In cells infected with mutant viruses that do not express pUL16 and/or pUL21 (the double-mutant scenario is shown), the cytoplasmic face of the ONM and NPC are not modified through association with pUL16 and pUL21 and attachment of nascent cytoplasmic nucleocapsids to NPCs is not prevented, nor is the delivery of their genomes to the infected cell nucleus. Created with BioRender.com.

https://doi.org/10.1371/journal.ppat.1011832.g010

Cells and viruses

African green monkey kidney cells (Vero), porcine kidney cells (PK15), HeLa cells, life-extended human foreskin fibroblasts (T12) human keratinocytes (HaCaT) and HaCaT cells stably expressing pUL21 or pUL16 (HaCaT21 and HaCaT16, respectively) [38, 44] and life-extended primary human fibroblasts (T12), a gift from Dr. W. A. Bresnahan (University of Minnesota) [73], were maintained in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS) in a 5% CO₂ environment. HSV-1 KOS mutants deficient in pUL21 (Δ21), pUL16 (Δ16), or pUL21 and pUL16 (Δ21/Δ16) have been described previously [38, 42, 64]. HSV-2 186 mutants deficient in pUL21 (Δ21), pUL16 (Δ16) have been described previously [34, 43]. Recombinant HSV-2 strain 186 containing mCherry (mCh) fused to the minor capsid protein VP26 and the corresponding Δ21 and Δ16 mutants have been described previously [34, 43]. Recombinant PRV vaccine strain Bartha containing monomeric RFP (mRFP) fused to VP26 (PRV765), a gift from Dr. L. W. Enquist (Princeton University), has been previously described [74] and was propagated in PK15 cells. All pUL21 deficient and pUL16 deficient viruses were propagated in HaCaT21 and HaCaT16 cells, respectively. The HSV-1 KOS Δ21/Δ16 virus was propagated on a 1:1 mixture of HaCaT21 and HaCaT16 cell monolayers as previously described [64]. 5-ethynyl-2’-deoxycytidine (EdC) was incorporated into the genome of HSV-2 186 WT virus as previously described [64] to produce HSV-2 186 WT EdC virus stocks. Times post infection, reported as hours post infection (hpi), refers to the time elapsed following a one-hour inoculation period.

HSV-2 strain 186 carrying pUL21 with mCh fused to its carboxy terminus was constructed by two-step Red-mediated mutagenesis [75], using bacterial artificial chromosome (BAC) clone pYEbac373 [43] in Escherichia coli GS1783 [75]. Forward primer 5’-GCTTACCGTTTGCCTGGCTCGCGCCCAGCACGGCCAGTCTGTGGTGAGCAAGGGCGAG-3’ and reverse primer 5’- TGGGTTAGAAAACGACTGCACTTTATTGGGATATCTCACTTGTACAGCTCGTC-3’ were used to amplify a PCR product from pEP-mCh-in, a gift from Dr. G.A. Smith (Northwestern University). Restriction fragment length polymorphism analysis following digestion with EcoRI was used to confirm the integrity of BAC DNA. Additionally, PCR was used to amplify products spanning UL21-mCh and these products were sequenced to confirm that mCh was in-frame with UL21 and to ensure that no unintended mutations were introduced. Virus was reconstituted from BAC DNA as described previously [43].

Plasmids

The following plasmids were used in this study and previously described: enhanced green fluorescent protein (EGFP) pEGFP-N1 (Clontech Laboratories, Mountain View, CA), pCI-neo HSV-2 UL21 (UL21) [43], pCI-neo HSV-2 UL16 (UL16) [34], EGFP fused to the C-terminus of HSV-2 UL21 (UL21-GFP) [43]. To construct a plasmid encoding UL21-mCherry, the EGFP portion of UL21-EGFP was excised by digestion with BamHI and BsrGI and replaced with a BamHI/BsrGI fragment encompassing mCherry excised from pJR70 [76]. To construct a plasmid encoding UL16-EGFP, PCR utilizing forward primer 5’-AGTTCGAATTCTTATGGCACAGCGGGCACTCTGGCGTCC-3’ and reverse primer 5’-GATCGTCGACGGTTTGTAATCGGACGATGAGGCTCTGG-3’ were used to amplify the UL16 gene using purified HSV-2 strain HG52 DNA as a template. The product was digested with EcoRI and SalI and ligated into similarly digested pEGFP-N1. To construct a plasmid encoding UL16-mCherry, the UL16 portion of UL16-EGFP was excised by digestion with HindIII and BamHI and ligated to similarly-digested pJR70.

Immunological reagents and chemicals

Mouse monoclonal antibody against lamin A/C (EMD, Millipore Temecula, CA) was used at a dilution of 1:300, mouse monoclonal antibody against nuclear pore complex proteins (MAb414) (Abcam, Cambridge, MA) was used at a dilution of 1:500, mouse monoclonal antibody against HSV ICP5 (Virusys, Taneytown, MD) was used at a dilution of 1:500, mouse monoclonal antibody against HSV ICP27 (Virusys, Taneytown, MD) was used at a dilution of 1:500, mouse monoclonal antibody against PRV Us3 [77], a gift from Dr. L. W. Enquist (Princeton University) was used at a dilution of 1:500, mouse monoclonal antibody against α-tubulin (GT114) (Invitrogen/Thermo Fisher Scientific, Ottawa, ON) was used at a dilution of 1:500, mouse monoclonal antibody against RanBP2 (D-4) (sc-74518) (Santa Cruz Biotechnology, Dallas, TX) was used at a dilution of 1:50 and rabbit polyclonal antibody against mCherry (Rockland Immunochemicals, Burlington, ON) was used at a dilution of 1:200 for immunofluorescence microscopy. Alexa Fluor 568-conjugated donkey anti-mouse, Alexa Fluor 488-conjugated donkey anti-mouse, Alexa Fluor 647-conjugated donkey anti-mouse were used at a dilution of 1:1000 and Alexa Fluor 488-conjugated donkey anti-rabbit (Thermo Fisher Scientific, Ottawa, ON) was used at a dilution of 1:500, Alexa Fluor 532-conjugated goat anti-mouse (Thermo Fisher Scientific, Ottawa, ON) was used at a dilution of 1:400 and Alexa Fluor 488-conjugated phalloidin (Invitrogen, Ottawa, ON) was used at 165nM for immunofluorescence microscopy. Nocodazole (10mM in DMSO) was used to interfere with cellular microtubule polymerization at a final concentration of 10μM and cells were treated for 6 or 12 hours. AlexaFlour 488 picolyl azide and 5-ethynyl-2’-deoxycytidine (EdC) (Click Chemistry Tools, Scottsdale, AZ) were used for click-chemistry according to the manufacturer’s instructions.

Transmission electron microscopy

Vero cells were seeded onto 100mm dishes one day prior to infection and were infected with HSV-1 KOS WT, Δ21, Δ16 or Δ21/Δ16 virus at a multiplicity of infection (MOI) of 3. At 18 hpi infected cells were processed for transmission electron microscopy as previously described [64]. All micrographs used for data quantification have been uploaded to Dryad and are accessible here [78].

Immunofluorescence microscopy

Cells that were mock infected or infected with HSV-2 or HSV-1 viruses were fixed at the indicated times post infection by rinsing the cells three times with PBS, followed by fixation in 4% paraformaldehyde in PBS for 15 minutes at room temperature (RT). Cells were then washed three times with PBS/1% BSA and stored in PBS/1% BSA at 4°C until staining. Samples were washed three times in PBS and permeabilized by adding 0.5% Triton X-100 (TX-100) in PBS for 15 minutes at RT. Samples prepared for quantification of nuclei with cytoplasmic capsids surrounding the outer nuclear membrane were permeabilized by adding 0.01% saponin in PBS for 10 minutes at RT. Samples permeabilized with saponin had 0.01% saponin in all subsequent washes and antibody dilutants. After permeabilization, samples were washed three times with PBS/1% BSA and primary antiserum diluted in the appropriate volume of PBS/1% BSA was applied to samples for one hour at RT. A blocking step was implemented prior to the addition of rabbit primary antibodies using human serum blocking buffer (5% human serum, 0.5% Tween 20 in PBS) for 1 hour at RT. After primary antibody incubation, samples were washed three times with PBS/1% BSA and Alexa Fluor conjugated secondary antibody diluted in the appropriate volume of PBS/1% BSA was applied to samples for 30 minutes at RT. Samples were washed three times with PBS/1% BSA and then incubated with 0.5 μg/mL Hoechst 33342 (Sigma, St. Louis, MO) in PBS for seven minutes at RT. Samples were then washed three times with PBS/1% BSA and stored at 4°C in PBS/1% BSA until imaging. Samples were imaged through a 60X (1.42 NA) oil immersion objective using an Olympus FV1000 confocal laser scanning microscope and FV10 ASW 4.01 software. Average signal intensity was measured using FV10 ASW 4.01 software.

Stimulated emission depletion (STED) microscopy

Vero cells were infected with HSV-2 186 UL21-mCh at an MOI of 0.1. At 18 hpi, cells were rinsed three times with PBS followed by fixation in 4% paraformaldehyde in PBS for 15 minutes at RT. Cells were washed three times in PBS and permeabilized in 0.5% Triton X-100 in PBS for 15 minutes at RT. Cells were rinsed three times with PBS/1% BSA and stained for mCh, NPCs (MAb414), or RanBP2, as outlined above. Because mCh was unable to be used STED microscopy, an anti-mCherry antibody was used and detected using an Alexa Fluor 488-conjugated secondary antibody. Confocal and STED microscopy were performed on an SP8 tandem scanning white light laser confocal platform (Leica Microsystems, Richmond Hill, ON) equipped with STED lasers at 592nm and 660nm. A 100X STED objective, NA 1.4, on a DMI 6000 inverted microscope platform with a motorized x, y and z-glavo stage were employed. We used the 590 and 660 STED depletion lasers for Alexa Fluor 488 and Alexa Fluor 532 respectively. All images were deconvolved using Huygens deconvolution software (Hilversum, Netherlands).

Ectopic expression of pUL16 and pUL21

HeLa or PK15 cells were seeded onto 60mm dishes one day prior to transfection. These cells were then transfected with i) EGFP; ii) EGFP and UL16; iii) EGFP and UL21 or iv) EGFP, UL16 and UL21 expression plasmids using X-treme GENE HP DNA transfection reagent (Roche, Laval, QC) following the manufacturer’s instructions. At 6 hours post transfection, samples were incubated in 500μL of complete medium containing 1μL benzonase (250U/μL) for 30min at 37°C. Cells were then harvested by trypsinization and seeded onto glass coverslips. At 24 hours post transfection, cells were infected on ice for one hour with HSV-2 186 WT mCh-VP26 or PRV765 at an MOI of 3. After one hour, complete medium was added and the cells shifted to 37°C. At 2 hpi, cells were washed three times in PBS and fixed in 4% paraformaldehyde in PBS for 15 minutes at RT. Samples were then permeabilized with TX-100 and stained with ICP27/Us3 antisera, as outlined in the immunofluorescence microscopy section. The average capsid distance from the nuclear periphery was determined using FV10 ASW 4.01 software. 43–60 capsids per condition were examined.

Evaluation of pUL21 and pUL16 loading onto incoming capsids

HeLa cells were transfected with EGFP, UL16-EGFP, UL21-EGFP, UL16-EGFP and UL21 or UL21-EGFP and UL16 expression plasmids and prepared for infection as outlined above. At 24 hours post transfection, samples were infected on ice for one hour with HSV-2 186 WT mCh-VP26 at an MOI of 3, shifted to 37°C following inoculation and fixed at 2 hpi or 18 hpi. Additionally, samples were also infected on ice for one hour with HSV-2 186 Δ16 mCh-VP26 or HSV-2 186 Δ21 mCh-VP26 at an MOI of 3, shifted to 37°C following inoculation and fixed at 18 hpi. Cells were permeabilized and stained for NPCs, as described above. EGFP signal associated with capsids at 2 hpi, as well as the average background EGFP signal (EGFP signal from two points 0.5 μm from the capsid) was measured using FV10 ASW 4.01 software.

SIE analysis

Vero cells were infected with HSV-2 186 WT, Δ21 or Δ16 virus at an MOI of 1. To examine SIE of superinfecting virus entry and capsid transport to infected nuclei, infected samples were placed on ice for 30 minutes at 6 hpi. At 6.5 hpi, infected samples were superinfected on ice with HSV-2 186 WT mCh-VP26 at an MOI of 3 for 1 hour then shifted to 37°C. At 1 h after the superinfection (8.5 hpi after the initial infection), cells were fixed and stained with Alexa Fluor 488-conjugated phalloidin at a final concentration of 165 nM in PBS for 7 minutes to visualize cortical actin at the cell periphery.

To examine genome delivery of superinfecting virus, infected cells were superinfected at 6 hpi with HSV-2 186 WT EdC labelled virus at an MOI of 3 for 1 hour at 37°C. After inoculation, the inoculum was removed, and cells were incubated in low pH citrate buffer (40mM Na citrate pH 3.68, 10mM KCl, 0.8% NaCl) for 3 minutes at RT to inactivate extracellular virus. Cells were washed three times with complete medium and incubated for 3 h after the superinfection (10 hpi after the initial infection) at which time cells were fixed and EdC detected by click-chemistry using AlexaFlour 488 picolyl azide according to the manufacturer’s instructions.

Statistical analysis

All statistical analyses were performed using GraphPad Prism version 9.1.2.