Establishment of recombinant KSHV with varying numbers of TR copies

To assess the transcription function of TR, we first generated mutant KSHV with a reduced number of TR copies using BAC16 recombination [23]. A DNA fragment containing the kanamycin expression cassette and an I-SceI restriction enzyme site was introduced into TR fragments with a red recombination [24]. The number of insertions and position of the kanamycin cassette within TR may vary in each bacterial colony due to multiple identical TR sequence units [2]. Recombination within TR fragments was subsequently induced by the expression of the red recombinase in E. coli after induction of I-SceI with L-Arabinose (Fig 1A). Recombination generated various numbers of TR copies randomly in the KSHV BAC16. After isolating BAC DNA from bacterial colonies, we identified the bacterial clone that harbors 5, 7, 10, or 21 (BAC16 wild type [25]) copies of TR (Fig 1B). Each BAC DNA was transfected in iSLK cells, and the stable iSLK cells were established with hygromycin selection (Fig 1C). While we attempted multiple times, we failed to establish BAC-stable cells with two copies of TR (TR2). We also noticed that GFP signal intensity was significantly lower in TR deletion mutants, although nearly 100% of iSLK cells showing GFP expression (Fig 1D and 1E). Next, we confirmed the circularization of the KSHV genome in TR5, TR21-infected cells with BCBL-1 cells as a positive control. For this assay, we used Plasmid-Safe ATP-dependent DNase (PSAD), which selectively degrades linear double-stranded DNA (dsDNA) while leaving circular DNA (episomes) intact. We determined that KSHV genomes were successfully circularized in infected cells and the KSHV episome numbers were found to be 5.76, 8.08, and 27.49 per cell in TR5, TR21, and BCBL-1 cells, respectively (Fig 1F). The results were also consistent with the fact that we could recover the infectious recombinant KSHV from latently infected cells.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Generation of recombinant KSHV BAC with different numbers of TR copies.

(A). Schematic diagram of KSHV BAC16 with different copy numbers of TR. The kanamycin cassette with I-SceI recognition sequence, along with the TR homologous sequence, was generated by PCR with pEP-Kan plasmid as a template. This DNA fragment was cloned into KSHV BAC16 by homologous recombination. The kanamycin cassette was deleted by recombination with induction of I-SceI in bacteria by incubation with L-arabinose. Correct insertion of the different copy numbers of TR was confirmed by PstI digestions. (B). Confirmation of TR copies and BAC integrity. Purified KSHV BAC16 was digested by PstI and subjected to electrophoresis. Agarose gel was stained with Ethidium bromide (EtBr) (left panel). The TR-containing fragments were probed with Southern blotting with a fluorescently labelled TR probe (right panel). The numbers in the Southern blotting represent the length (base pair) of the TR fragment after PstI digestion. The 1 kb DNA ladder is shown. (C). Fluorescence images of iSLK cells with different copy numbers of TR. KSHV BAC16 with different copy numbers of TR were transfected into iSLK cells and selected with hygromycin (500µg/ml). The mock indicates iSLK cells without KSHV BAC16. Images were taken after successful establishment of stable cells. Scales: 300μm. (D). Proportion of GFP positive cells in iSLK cells with different copy numbers of TR. iSLK cells harboring KSHV BAC16 with different copy numbers of TR were cultured under hygromycin (500µg/ml), and the proportion of GFP positive cells was calculated using flow cytometry. (E). Mean fluorescence intensity (MFI) of iSLK cells with different copy numbers of TR. MFI in the FITC channel was determined using FlowJo v10 software. (F) KSHV episome copy number. Episomal DNA was extracted from TR5-, TR21-infected cells and BCBL-1 cells using the Hirt extraction method, followed by overnight treatment with Plasmid-Safe ATP-dependent DNase (PSAD). KSHV genome copy number was determined by quantitative PCR using a standard curve generated from serial dilutions of KSHV BAC16 DNA. Data are presented as mean copy number per cell. Error bars represent standard deviation (SD).

https://doi.org/10.1371/journal.ppat.1013061.g001

TR is important for viral gene expression

To examine the transcription function of TR in KSHV infection, we first isolated infectious KSHV virions from iSLK cells that were stably transfected with KSHV BAC16 DNAs. The amount of KSHV virion copy was adjusted and infected to iSLK cells at the viral genomic copy/cell ratio of 10:1. We then examined the effect of TR copies in viral gene expression during de novo infection. No statistically significant difference was observed in the relative KSHV genome copies in iSLK cells at 24 hours post-infection (Fig 2A), indicating that reduced TR copy number does not affect the efficiency of virus infection. Total RNAs were harvested 24 hours post-infection, and RT-qPCR was performed for the latent and inducible viral gene expression. As shown in Fig 2B, both KSHV latent and inducible viral genes (LANA, PAN RNA, vIL-6, and K8.1) expressed significantly more in KSHV-TR21 infected cells compared to KSHV-TR5 infected cells. Recombinant KSHV-infected cells were then selected with hygromycin, and KSHV-TR21 or KSHV-TR5 latently infected cells were established. Like the BAC transfection, KSHV-TR21 infected iSLK cells showed higher GFP intensity than KSHV-TR5 infected iSLK cells (Fig 2C). To examine whether the higher GFP expression is at the transcription level, we next measured GFP transcripts per KSHV genome. The GFP gene is encoded downstream of the EF1α promoter in the KSHV BAC16 backbone and is constitutively expressed in the recombinant KSHV-infected cells [23]. As shown in Fig 2D, the GFP was more actively transcribed in the KSHV-TR21 latent chromatin than KSHV-TR5 latent chromatin. We considered that EF1α promoter regulation is independent of KSHV latent protein expression; therefore, this result may suggest that KSHV-TR21 establishes a local nuclear microenvironment that is more favorable for gene transcription.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. KSHV inducible gene expression is positively regulated in a TR-dependent manner.

(A). Relative viral DNA levels in KSHV-infected iSLK cells. iSLK cells were infected with TR5-KSHV or TR21-KSHV for 24 hours and viral DNA levels were determined by quantitative PCR (qPCR). GAPDH expression was used for internal control. Relative viral DNA levels in TR5-KSHV iSLK cells was set as 1. Data was analyzed using a two-sided unpaired Student’s t-test and shown as mean ± SD. (B). KSHV gene expression in iSLK cells during de novo infection. Relative RNA levels of the indicated viral genes were determined at 24 hours by Reverse Transcription qPCR (RT-qPCR).18S rRNA expression was used for internal control. Relative gene expression in TR5-KSHV iSLK cells was set as 1. Data was analyzed using two-sided unpaired Student’s t test and shown as mean ± SD. (C). Fluorescence and brightfield images of iSLK cells infected with TR5-KSHV or TR21-KSHV. The iSLK cells were infected with the equal amount of TR5-KSHV and TR21-KSHV followed by selection with hygromycin (500µg/ml). Images were taken after successful establishment of stable cells. Scales: 300μm. (D). Relative GFP expression in iSLK cells infected with TR5-KSHV or TR21-KSHV. Total RNAs were harvested from iSLK cells infected with the TR5-KSHV or TR21-KSHV. 18S rRNA expression was used for internal control, and relative GFP expression was normalized by the viral genome. Relative GFP expression in TR21-KSHV-infected iSLK cells was set as 1. Data was analyzed using two-sided unpaired Student’s t test and shown as mean ± SD. (E). Relative KSHV gene expression in iSLK cells infected with TR5-KSHV or TR21-KSHV. Each iSLK cell was reactivated by sodium butyrate (1 mM) and doxycycline (1 μg/ml) for 48 hours. Total RNAs were purified two days after reactivation. 18S rRNA was used for internal control, and relative gene expression in TR21 iSLK cells after reactivation was set as 1. The gene expression was normalized to KSHV genome. Data was analyzed using two-sided unpaired Student’s t-test and shown as mean ± SD. (F). Immunoblotting of KSHV latent protein (LANA) and lytic proteins (Rta, vIL-6, K8α, and Mta) in TR5-KSHV or TR21-KSHV-infected iSLK cells. iSLK cells were reactivated by sodium butyrate (1 mM) and doxycycline (1 μg/ml) for 48 hours and total cell lysates were prepared two days after reactivation and subjected to immunoblotting with specific antibodies. β-actin was used as a loading control. Quantification of band intensities were calculated by ImageJ software and shown under the images. Each relative intensity was normalized by β-actin. (G). KSHV virion production after reactivation of TR5-KSHV or TR21-KSHV-infected iSLK cells. Virions in the supernatant four days after reactivation by sodium butyrate (1 mM) and doxycycline (1 μg/ml) were treated with DNaseI, followed by DNA purification. Results are shown as viral copies per milliliter. ORF6 was used to quantify the viral DNA. The data was analyzed using a two-sided unpaired Student’s t-test and shown as mean ± SD.

https://doi.org/10.1371/journal.ppat.1013061.g002

Next, we examined the overall efficacies of KSHV reactivation with RT-qPCR. We reactivated KSHV-TR21 and KSHV-TR5 infected iSLK cells with doxycycline and sodium butyrate and examined viral gene expression and KSHV replication. The results showed that the viral gene expression was more robust in KSHV-TR21 genome after stimulation (Fig 2E), resulting in increased viral protein expression (Fig 2F) and virion production in the supernatant (Fig 2G). These findings suggest that a higher number of TR copies support gene transcription more efficiently regardless of viral or exogenous genes encoded in the episome.

BRD4 at the TR region is important for KSHV inducible gene transcription

The results described above led us to hypothesize that cellular and viral proteins at the TR should be important for the KSHV reactivation. This is analogous to the cellular enhancers, which harbor arrays of transcription factor binding sites to maintain transcription-related proteins at the genomic loci [26]. Previous studies showed that LANA interacts with BRD4 [21,27,28] and is localized at TR [5,28]. BRD4 is known to be a localized cellular enhancer [29] and regulates transcription elongation [30]. Accordingly, we focused on whether the shortened TR compromised the recruitment of BRD4 to the viral promoters. Our idea was that the TR recruits BRD4 via sequence-specific LANA DNA binding and that a higher number of TR copies is advantageous for recruiting more BRD4 to the TR through LANA/BRD4 protein interaction; this mechanism established local nuclear environment favorable for transcription activation in the large 3D nuclear space.

First, we examined differences in LANA accumulation on the KSHV episome, using the size of LANA dots as an indicator. An immunofluorescence assay with an anti-LANA monoclonal antibody showed that LANA dots became smaller when we shortened the TR (Fig 3A and 3B). Proximity ligation assays suggested that larger LANA dots interacted more frequently with BRD4, suggesting that TR21 provides a better platform to recruit BRD4 (Fig 3C and 3D). We also examined the BRD4 occupancies and histone H3 lysine 27 acetylation (H3K27Ac) in KSHV-TR21 and TR5-infected iSLK cells. As shown in Fig 3E, BRD4, and H3K27Ac were more frequently occupied at the TR21 latent chromatin, consistent with the frequency of LANA interactions with BRD4 at LANA nuclear bodies and promoter activity (Fig 3D).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. BRD4 regulates KSHV inducible gene expression in the TR dependent manner.

(A). Immunofluorescence image of the LANA protein in TR5-KSHV or TR21-KSHV-infected iSLK cells. Anti-LANA rat monoclonal antibody and alexa-647 (secondary) goat anti-rat antibody were used for staining LANA in TR5- or TR21-KSHV-infected iSLK cells and Cy5 channel was used for its detection. LANA dots (upper panels) and DAPI (lower panels) are shown. Scales: 50 μm. (B). Perimeter of each LANA dots. The size of each LANA dot in TR5- or TR21-KSHV-infected iSLK cells was calculated by imageJ (ver.1.53). Data was analyzed using a two-sided unpaired Student’s t-test and shown as mean ± SD. (C). Representative images of proximity ligation assay (PLA) for BRD4 and LANA interaction in TR5- or TR21-KSHV-infected iSLK cells. TR5- or TR21-KSHV-infected iSLK cells were subjected to PLA using antibodies against BRD4 (1:100 dilution) and LANA (1:100 dilution). Red dots indicate BRD4-LANA interactions. Scales: 20 μm. (D). Quantification of PLA dots. Each PLA dot in TR5- or TR21-KSHV-infected iSLK cells was calculated by imageJ (ver.1.53). (E). Enrichment of BRD4 and H3K27Ac in TR5- or TR21-KSHV-infected iSLK cells. TR5 or TR21-KSHV-infected iSLK cells were subjected to CUT&RUN analysis with anti-BRD4 (1:100 dilution) or anti-H3K27Ac (1:100 dilution) antibodies to examine their occupancy on the specific promoter regions in the KSHV genome. The control region was the coding region of the ORF23, which has been shown not to accumulate BRD4. Normal Rabbit IgG antibody was used as a negative control. BRD4 and H3K27Ac recruitment was calculated as percentage to input DNA. Data was analyzed using a two-sided unpaired Student’s t-test and shown as mean ± SD. (F). Immunoblotting assay of BRD4 and KSHV proteins. r.219 iSLK cells were reactivated by doxycycline (1 μg/ml) and sodium butyrate (1 mM) 24 hours after MZ1 (100 nM) treatment. Two days after reactivation, total cell lysates were collected for immunoblotting to detect BRD4 and KSHV proteins (vIL-6 and K-Rta). β-actin was used as a loading control. (G). Virion production in r.219 iSLK cells with or without MZ1 treatment. Virions in the supernatant four days after r.219 iSLk cells reactivated by sodium butyrate (1 mM) and doxycycline (1 μg/ml) were treated with DNaseI, followed by DNA purification. MZ1 or DMSO were added one day before reactivation. Results are shown as viral copies per microliter. Data was analyzed using a two-sided unpaired Student’s t-test. (H). Proportion of GFP positive cells after r.219 KSHV infection. r.219 iSLK cells were reactivated by sodium butyrate (1 mM) and doxycycline (1 μg/ml) 24 hours after MZ1 or DMSO treatment. Four days after reactivation, an equal amount of supernatant was added to iSLK cells, and the proportion of GFP-positive cells was determined by flow cytometry one day after infection. (I). Experimental flow of LANA degradation. mAID-LANA KSHV BAC16 iSLK cells were treated with indole-3-acetic acid (IAA) (5 μM) for four hours, followed by reactivation with sodium butyrate (1 mM) and doxycycline (1 μg/ml) for one day to examine the effect of LANA on BRD4 accumulation on the KSHV genome. (J). Immunoblotting assay of KSHV gene expression in LANA AID iSLK cells with or without IAA treatment. One day after reactivation, total cell lysates were collected for immunoblotting to detect KSHV proteins (LANA and K-Rta). β-actin was used as a loading control. (K). BRD4 occupancy in mAID-LANA KSHV BAC16 iSLK cells. One day after reactivation. BRD4 enrichment on the Ori-RNA promoter and TR region was determined using CUT&RUN. Normal Rabbit IgG antibody was used as a negative control. BRD4 recruitment was calculated as percentage to input DNA. Data was analyzed using a two-sided unpaired Student’s t-test and shown as mean ± SD.

https://doi.org/10.1371/journal.ppat.1013061.g003

To assess the contribution of BRD4 to inducible viral gene transcription, we used MZ1, a selective BRD4 inhibitor [31], and lytic gene expression and viral production were measured. We selected MZ1, a proteolysis targeting chimera (PROTAC) that degrades BRD4, instead of a BET inhibitor. BET inhibitor, acting as an acetylated histone mimetic competitor, would increase the non-chromatin bound BRD4, which may help viral transactivators to form transcriptionally active protein complexes at target gene promoters, and induce KSHV reactivation effectively [32–34]. Importantly, our data in Fig 3F demonstrate that MZ1 treatment altered vIL-6 expression and viral DNA production without significantly affecting the expression of K-Rta or the housekeeping gene β-actin. These findings suggest that the effects of MZ1 are not solely due to a global transcriptional shutdown. The immunoblotting and qPCR showed that KSHV inducible protein expression and infectious KSHV virion production were inhibited by the BRD4 degradation in r.219 iSLK cells (Fig 3F–3H).

Next, we examined the role of LANA in BRD4 recruitment on the KSHV latent chromatin. We used mAID-LANA KSHV BAC16 (TR21) iSLK cells [35] to examine whether rapid LANA degradation changes the frequency of BRD4 recruitment. We first incubated the cells with IAA to induce LANA protein degradation for 4 hours and stimulated KSHV reactivation with doxycycline and sodium butyrate (Fig 3I and 3J). The results showed that BRD4 occupancy at the Ori-RNA promoter and TR was significantly decreased by LANA degradation, suggesting that LANA enhances the recruitment of BRD4 to the KSHV episome during latency (Fig 3K). As expected, when we triggered KSHV reactivation, BRD4 recruitment was significantly increased. To our surprise, BRD4 recruitment to the Ori-RNA promoter was not affected by LANA degradation after reactivation. On the other hand, BRD4 recruitment to the TR was reduced with LANA degradation during reactivation. The results indicated that BRD4 recruitment is differently regulated between TR and lytic inducible promoters. It is important to note that BRD4 recruitment at TR is a magnitude higher than that of the Ori RNA promoter region, and LANA is known to bind TR sequence directly.

Increased inducible gene expression from a chromatin template with higher TR copies

The experiments above were performed with a cell population in a dish, including non-responder and responding cells with reactivation stimuli. While an increased TR copy number demonstrated increased viral gene expression, it remains unclear whether this outcome is due to an increased number of reactivating cells in the dish, increased viral gene transcription from an episome template, or a combination of both. To examine whether an increased TR copy number increases the frequency of promoter activation, as a gene regulatory domain, we established a new recombinant KSHV, which encodes chicken Bu-1 (cBu-1) extra cell surface domain under the control of the vIL-6 promoter. The vIL-6 coding sequence was then fused with the cBu-1 with the GSG-P2A peptide sequence. The vIL-6 promoter was selected because vIL-6 is highly transcribed among lytic inducible genes during reactivation [36–38] and is one of the direct K-Rta target genes [39,40]. The GSG-P2A peptide allows two proteins to be translated from the same mRNA by causing the ribosome to fail at making a peptide bond [41–43] (Fig 4A). Human cells do not express the chicken protein on their cell surface. Using cBu-1 monoclonal antibody increases the detection of vIL-6 transcribing cells by amplifying signals with fluorescent-labeled secondary antibodies. This strategy identifies and conveniently isolates reactivating living cells with magnetic beads. Fig 4B demonstrates the co-expression of cBu-1 molecule and vIL-6 protein in a cell. Using protein A/G magnetic beads (Fig 4C), we could isolate the vIL-6 transcribing cell population from the dish, which increases the resolution of downstream analyses approximately 10 times (Fig 4D and 4E).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Establishment of cBu-1 BAC16 for the isolation of reactivated cell population.

(A). Schematic model of the isolation of reactivated cells using chicken Bu-1 (cBu-1) BAC16 system. cBu-1 external domain sequence along with GSG-P2A peptide sequence was inserted as an N-terminal fusion with vIL-6 protein. After reactivation by doxycycline (1 μg/ml) and sodium butyrate (1 mM), cBu-1 is expressed on the cell surface only in the vIL-6 expressed cells, and anti-cBu-1 antibody can bind to cBu-1 protein for isolation. (B). Representative FACS profiles of iSLK cells with cBu-1 BAC (cBu-1 iSLK cells). Two days after reactivation by doxycycline (1 μg/ml) and sodium butyrate (1 mM), cells were fixed by 2% paraformaldehyde and permeabilized by 0.1% Triton X-100. Cells were then stained by vIL-6 antibody and cBu-1 antibody for two hours at room temperature. After washing with PBS, cells were incubated with secondary antibodies (APC for vIL-6 and PE for cBu-1). (C). Brightfield image of cBu-1-expressed cells and non-expressed cells. Two days after reactivation, cells were incubated with a biotin-conjugated anti-cBu-1 antibody (10 μg/ml) for one hour at room temperature. Cells were washed with PBS twice, followed by incubation with streptavidin magnetic beads for one hour at room temperature. Cells were then observed on the cover glass. Scales: 20 μm. (D). Immunoblotting of KSHV lytic proteins (ORF59 and vIL-6) in cBu-1 iSLK cells with or without cBu-1 enrichment after reactivation. cBu-1 iSLK cells were reactivated by doxycycline (1 μg/ml) and sodium butyrate (1 mM) for two days, and cBu-1 expressed cells (IP) were then enriched with cBu1 monoclonal antibody (10 μg/ml). The cell population that did not express cBu-1 was also harvested as flow thorough (FL) at the same time. β-actin was used as a loading control. (E). vIL-6 expression of cBu-1 expressed cells (IP) and non-expressed cells (FL) after isolation by cBu-1 antibody. 18S rRNA expression was used for internal control. Relative gene expression in FL population was set as 1. Data was analyzed using a two-sided unpaired Student’s t-test and shown as mean ± SD. (F). Experimental design of the isolation of reactivated cells by cBu-1 system for the following analysis. cBu-1 virus with different copy numbers of TR (cBu-1 TR21 or cBu-1 TR6) were infected with iSLK cells and cultured with hygromycin. One day after reactivation by doxycycline (1 μg/ml) and sodium butyrate (1 mM), cBu-1 expressed cells (IP), and cBu-1 non-expressed cells (FL) were sorted by magnetic beads for the following analysis. (G). Relative KSHV gene expression and viral genome in iSLK cells after cBu-1 enrichment. cBu-1 TR21- or cBu-1 TR6-KSHV-infected iSLK cells were reactivated by doxycycline (1 μg/ml) and sodium butyrate (1 mM). One day after reactivation, reactivated cells were isolated according to cBu-1 expression, and total RNAs and DNAs were purified to determine the viral gene expression. 18S rRNA and actin was used as an internal standard for normalization for relative gene expression and viral genome respectively. ORF6 was used for quantifying the viral DNA. Relative gene expression in cBu-1 TR6-KSHV-infetced iSLK cells was set as 1. Data was analyzed using two-sided unpaired Student’s t test and shown as mean ± SD. (H). BRD4 and H3K27Ac occupancy in cBu-1 TR21- or cBu-1 TR6-KSHV-infected iSLK cells after cBu-1 enrichment. cBu-1 TR21- or cBu-1 TR6-KSHV-infected iSLK cells were reactivated by doxycycline (1 μg/ml) and sodium butyrate (1 mM). One day after reactivation, reactivated cells were isolated according to cBu-1 expression, and BRD4 or H3K27Ac enrichment of the Ori-RNA promoter TR, and vIL-6 promoter region was determined. Normal Rabbit IgG antibody was used as a negative control. Data was analyzed using a two-sided unpaired Student’s t-test and shown as mean ± SD.

https://doi.org/10.1371/journal.ppat.1013061.g004

With the cBu1-vIL-6 recombinant BAC in our hand, we again generated recombinant KSHV with reduced TR copies. While we screened over 50 colonies, we could not find the KSHV BAC possessing TR5, and the closest one was TR6. Accordingly, we used the TR6 clone in the following experiment (cBu-1 TR6-KSHV-infected iSLK cells). The recombinant KSHVs were isolated from BAC stably transfected cells, and the recombinant virus infected iSLK cells were used for the following studies.

As shown in Fig 4F, we isolated reactivating cells from cBu-1 TR6 or cBu-1 TR21-KSHV-infected iSLK cells; total RNA was extracted from these cells, which were attached to the magnetic beads. Similarly, CUT&RUNs were performed on the isolated reactivated cell population on the magnetic beads. The relative abundance of BRD4 recruitment per genome copy during active transcription was compared between TR21 and TR6. The robustness of inducible viral genes in the reactivating cells with stimuli was also measured with RT-qPCR. Consistent with the results shown above, TR6-KSHV-infected iSLK cells demonstrated decreased inducible gene expression per genome in the reactivated cells compared to TR21 KSHV-infected iSLK cells (Fig 4G). Decreased gene expression was accompanied by the decreased frequency or duration of BRD4 recruitment at the inducible promoter (Fig 4H). However, when we directly compared actively transcribing gene (vIL-6, i.e. cBu-1) in isolated positive cell population, we found much smaller differences in transcription rates (Fig 4G). The recruitment of BRD4 and H3K27Ac modification at the vIL-6 promoter also did not follow a similar pattern with other promoters (Fig 4H, the most right panel). The results suggested that increased transcription at PAN RNA or Ori-RNAs in TR21 may be due to a increased frequency of interactions between TR (enhancer) and inducible promoters. Increased interactions due to larger coverage with larger TR copies may lead a greater number of actively transcribing episomes in the infected cell or greater coverage of promoter regions within the episome, resulting in an overall increase in inducible viral transcription in the cBu-1(+) cell population.

Dynamic LANA NBs regulation by transcription activation and the formation of transcription memory

Previous reports showed that KSHV reactivation is a highly heterogeneous process, with only a selected cell population supporting KSHV reactivation in the dish [27,44,45]. To understand the underlying heterogenic responses, we utilized cBu-1 antibody to isolate iSLK cell populations that supported KSHV reactivation and established two cell populations originating from the same dish (Fig 5A). This step allowed us to synchronize the cell culture history and generates comparable two cell cultures from a dish. The isolated cells were cultured for an additional 10 days, and the size of LANA nuclear bodies (NBs) (enhancer sizes) was examined. Inducible gene transcription was then analyzed using RT-qPCR (promoter activation). Interestingly, the size of LANA NBs was significantly larger in the IP cell population (previously reactivated cell population) than flow through (FL) cells (non-responder with prior stimuli) (Fig 5B and 5C). The number of LANA NBs was comparable between the two cell populations (Fig 5B). The RT-qPCR showed that the reactivated cell population (IP) supported inducible gene expression approximately 4–10 times than the non-responded cell population (FL) (Fig 5D). The results suggest that (i) size of LANA NBs are regulated by prior transcription activation (i.e., enhancers can be activated by stimuli) and (ii) they form a transcription memory, which is similar to dormant enhancer activation during cell differentiation [46]. The increased size of LANA NBs by the previous activation also suggested that not all of the 21 TR unit may be equally activated in the episome.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. LANA NBs functions as transcription memory.

(A). Experimental workflow of the isolation of reactivated cells by cBu-1 system for the following analysis. Two days after reactivation by doxycycline (1 μg/ml) and sodium butyrate (1 mM), cBu-1 expressed cells (IP) and cBu-1 non-expressed cells (FL) were sorted by magnetic beads. Cells were then cultured for ten days for the following analysis. (B). Immunofluorescence analysis showing LANA dots in IP and FL population. Anti-LANA rat monoclonal antibody and alexa-647 (secondary) goat anti-rat antibody were used for staining LANA in IP and FL population from cBu-1 TR21-KSHV-infected iSLK cells. Cy5 channel was used for its detection. LANA dots (upper panels) and DAPI (lower panels) are shown. Scales: 20 μm. (C). Quantification of LANA dots. The size of each LANA dot in IP and FL population was determined and calculated by imageJ (ver.1.53). (D). Relative KSHV gene expression in IP and FL population. IP and FL population were cultured for ten days, followed by reactivation by doxycycline (1 μg/ml) and sodium butyrate (1 mM). Two days after reactivation, total RNAs were purified to determine the viral gene expression. 18S rRNA was used as an internal standard for normalization, and relative gene expression was normalized by the viral genome. Data was analyzed using two-sided unpaired Student’s t test and shown as mean ± SD. (E) Transcriptional memory of KSHV lytic gene expression in ECFCs infected with cBu1-vIL-6 virus. Endothelial colony-forming cells (ECFCs) were infected with cBu1-vIL-6 KSHV and selected with hygromycin (50 μg/ml). Infected cells were reactivated with sodium butyrate (1 mM) and subjected to immunoprecipitation using a cBu1-specific antibody (10 μg/ml) to isolate the reactivated population (IP). The non-reactivated (flow-through, FL) and reactivated (IP) populations were cultured separately for 10 days, followed by a second round of viral reactivation with sodium butyrate (1 mM). Expression levels of PAN RNA, K8.1, vIL-6 and ORF57 were measured by real-time PCR. The gene expression are shown as relative to non-reactivated FL. Error bars represent standard deviation (SD).

https://doi.org/10.1371/journal.ppat.1013061.g005

To confirm the transcriptional memory in more physiologically relevant cells, we also used Endothelial colony forming cells (ECFC). The lymphatic ECFCs were preapred from IPSCs [47], infected with cBu1-vIL-6 KSHV virus, and selected with hygromycin. The infected cells were reactivated with sodium butyrate, followed by isolating with anti-cBu1 antibody conjugated magnetic beads. The reactivated (IP) and non-reactivate cells (FL) were then cultured for 10 days, after which KSHV virus was reactivated again. Consistent with the studies with iSLK cells, the results showed that prior reactivated cell population (IP) supported inducible gene expression more effectively than non-responder population(FL) (Fig 5E).

Using a strategy to sync/isolate similar classes of cell populations, we next examined the association between TR length and the magnitude of reactivation (e.g., the extent of transcription memory formation after resting for 10 days) (Fig 6A). Similar to the results above, cBu-1 TR6-KSHV-infected iSLK cells showed lower inducible gene expression per genome compared with cBu-1 TR21-KSHV-infected iSLK cells (Fig 6B). Decreased gene expression was accompanied by a reduced frequency or duration of BRD4 recruitment at the inducible promoter during reactivation, although similar degree of H3K27Ac were present at the site (Fig 6C). These results suggest that higher TR copy numbers anchor more LANA through DNA sequence-specific interactions (Fig 3A) and that the LANA/TR complex can recruit BRD4 to the latent episome (Fig 3C). An increased local concentration of BRD4 near the latent episome within the 3D nucleus facilitates BRD4 delivery to individual KSHV inducible promoters, possibly through interactions with transcription factors such as K-Rta.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6.  Higher TR copy numbers in KSHV increase local BRD4 and H3K27Ac concentration, resulting in stronger transcriptional memory.

(A) Experimental workflow of the isolation and the comparison of reactivated cBu-1 TR21- and TR6-KSHV infected iSLK cells. cBu-1 TR21 or cBu-1 TR6 virus were infected with iSLK cells and preapred stably infected cells were cultured under hygromycin selection. Two days after reactivation, cBu-1 expressed cells were sorted by magnetic beads. Cells were then cultured for ten days for the following analysis. (B). Relative KSHV gene expression in cBu-1 TR21- and cBu-1 TR6-KSHV infected iSLK cells. cBu-1 expressed population in cBu-1 TR21- and TR6-KSHV infected iSLK cells were cultured for ten days after reactivation. Cells were then reactivated again, and total RNAs were purified to determine the viral gene expression. 18S rRNA was used as an internal standard for normalization, and relative gene expression was normalized by the viral genome. Data was analyzed using two-sided unpaired Student’s t test and shown as mean ± SD. (C). BRD4 and H3K27Ac occupancy in cBu-1 TR21- and cBu-1 TR6-KSHV-infected iSLK cells. BRD4 or H3K27Ac enrichment on the Ori-RNA promoter and TR region was determined using CUT&RUN. Normal Rabbit IgG antibody was used as a negative control. Data was analyzed using a two-sided unpaired Student’s t-test and shown as mean ± SD.

https://doi.org/10.1371/journal.ppat.1013061.g006

LANA and TR forms liquid-liquid phase-separated droplets with BRD4

BRD4 localizes at enhancers, and its intrinsically disordered regions (IDRs) form liquid-liquid phase separated (LLPS) droplets that concentrate transcription-related enzymes [16]. LANA condensates are also reported to be partially mediated by LLPS [48] and colocalize with BRD4 [5,32,49,50]. Accordingly, we next examined whether LANA can form LLPS with or without BRD4 and evaluated the contribution of TR DNA fragments to LLPS formation in vitro. To directly visualize their interactions and droplet formation, full-length EGFP-LANA and mCherry-BRD4 were expressed with a recombinant baculovirus and purified from insect cells (Fig 7A, upper and 7B, left). As a control, we also purified EGFP or mCherry tag alone (Fig 7A, lower and 7B right). The in vitro phase separation studies showed that EGFP-LANA did not form droplets by itself under the conditions we used (10% PEG6,000) (Fig 7C). However, LANA rapidly formed LLPS in the presence of mCherry-BRD4, as demonstrated by the colocalization of green and red signals in the same droplet (Fig 7C). The droplets were disrupted in the presence of high salt concentration or 1,6 HD (Fig 7D), suggesting that BRD4/LANA droplets had liquid-like properties, which is consistent with a previous report on BRD4 [16]. Notably, incubation with TR21 DNA fragments isolated from BAC16 DNA significantly increased the size of BRD4/LANA LLPS droplets, whereas a similar size of non-TR containing KSHV genome fragment isolated from the same BAC16 DNA did not (Fig 7E and, 7F). These results suggest that KSHV TR units serve as a platform for the formation of a LANA-mediated genomic enhancer. Taken together, we conclude that KSHV TR functions as a cleverly designed KSHV enhancer to maintain epigenetically active latent chromatin (Fig 8). Our findings support previous studies that the interaction of LANA with BRD4 is vital and central mechanism for the KSHV inducible gene regulation.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. BRD4 and LANA protein form LLPS.

(A). Representative image of fluorescent proteins. Indicated proteins were purified with Flag-beads and imaged under UV light. From left to right (up): mCherry-BRD4 and EGFP-LANA, (bottom) mCherry and EGFP proteins. (B). SDS-PAGE analysis of purified proteins. Molecular weight markers (kDa) are indicated on the left. (C). Representative images of LANA LLPS with or without BRD4. EGFP-LANA (1 μM) was mixed with mCherry-BRD4 protein (1 μM) or mCherry protein (1 μM). Images were taken five minutes after the mixture of the proteins. Scales: 20 μm. (D). Representative images of LANA and BRD4 LLPS in the presence or absence of NaCl or 1,6 HD. EGFP-LANA (1 μM) and mCherry-BRD4 (1 μM) were mixed with sodium chloride (NaCl) (1 M) or 1,6-hexanediol (1,6HD) (10%) for ten minutes, and images were taken under fluorescence microscopy. Scales: 10 μm. (E). Representative images of LANA and BRD4 LLPS in the presence or absence of the purified TR fragment. EGFP-LANA (1 μM) were incubated with purified TR21 DNA fragment (50 ng) or KSHV genome lacking TR (≃ 16kbp, 50 ng) for ten minutes at room temperature. EGFP-LANA were then mixed with mCherry-BRD4 protein (1 μM) for five minutes at room temperature. Scales: 20 μm. (F) Quantitative measurements of EGFP-LANA and mCherry-BRD4 LLPS droplet. Droplets were determined and calculated with roundness > 0.5 by using ImageJ (ver.1.53). Data was analyzed using two-sided unpaired Student’s t test and shown as mean ± SD.

https://doi.org/10.1371/journal.ppat.1013061.g007

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. Schematic model of TR functions as an inducible enhancer for KSHV lytic gene transcription.

KSHV LANA binds TR sequence specific manner and LANA recruits BRD4 with protein-protein interactions. The LANA forms LLPS in presence of BRD4, TR DNA fragment serves as a platform to recruit multiple copies of LANA to facilitate larger BRD4 containing LLPS formation. The larger LLPS (enhancer) increases a chance to interact with inducible promoters tightly packed in unique region.

https://doi.org/10.1371/journal.ppat.1013061.g008

Materials and Methods

Ethics Statement

Experiment involving human blood was approved under the Office of Ethics Review Committee at Fujita Health University approval (HG19–018). PBMCs were isolated from a blood sample obtained from a healthy donor. Formal written consent was not obtained from donor due to anonymity.

Chemicals, reagents, and antibodies

Dulbecco’s modified minimal essential medium (DMEM), RPMI 1640 medium, fetal bovine serum (FBS), phosphate-buffered saline (PBS), Trypsin-EDTA solution, and 100 X penicillin-streptomycin–L-glutamine solution were purchased from Thermo Fisher. Puromycin and G418 solution were obtained from InvivoGen. Hygromycin B solution was purchased from Enzo Life Science. Anti-LANA rat monoclonal antibody (clone LN53) was purchased from Millipore Sigma and anti-H3K27Ac rabbit antibody (clone D5E4) was purchased from Cell Signaling Technology. For proximity labeling assay, anti-LANA rabbit monoclonal antibody (clone 5L10, Millipore Sigma) and anti-BRD4 mouse antibody (E4X7E, Cell Signaling Technology) were purchased. Anti-cBu-1 mouse monoclonal antibody (biotin-conjugated) was purchased from BioVenic. Alexa 647-conjugated secondary antibody, SlowFade Gold anti-fade reagent, Lipofectamine 2000 reagent, and high-capacity cDNA reverse transcription kit were purchased from Thermo Fisher. The Quick-RNA Miniprep kit was purchased from Zymo Research, and the QIAamp DNA mini kit was purchased from QIAGEN.

Cell Culture

iSLK, iSLK.r219 [65] and BCBL-1 cells [66] were obtained from Dr. Don Ganem. iSLK and iSLK.r219 cells were cultured in DMEM supplemented with 10% FBS and 1% penicillin-streptomycin-L-glutamine solution, and maintained at 37 °C in a humidified 5% CO₂ incubator. For iSLK parental cells, puromycin (1 µg/mL) was included to maintain selection. The de novo infection to the iSLK cells was performed at viral genomic copy/cell ratio of 10:1. For iSLK.r219 cells, the medium additionally contained hygromycin B (200 µg/mL) and G418 (250 µg/mL) to ensure retention of the latent KSHV.219 genome. BCBL-1 cells were maintained in RPMI-1640 medium supplemented with 10% FBS and 1% penicillin-streptomycin-L-glutamine solution at 37 °C with 5% CO₂.

Induced pluripotent stem cells (iPSCs) were established from peripheral blood mononuclear cells (PBMCs) through the introduction of episomal plasmids encoding the Yamanaka reprogramming factors (OCT4, SOX2, KLF4, and c-MYC). Following transfection, cells were maintained in StemFit medium (Ajinomoto, Japan) on iMatrix-511–coated culture plates under hypoxic conditions (5% O₂) at 37°C. Colonies exhibiting characteristic iPSC morphology were manually isolated and subsequently expanded.

For endothelial induction, iPSCs were pre-cultured for two days in StemFit medium, then treated with activin A (10 ng/mL) together with fibroblast growth factor (FGF)-2, vascular endothelial growth factor (VEGF)₁₆₅, and bone morphogenetic protein 4 (BMP4) (10 ng/mL each) for 24 hours. The next day, cytokine-containing medium was replaced with Stemline II complete medium (Sigma) supplemented with FGF-2, VEGF₁₆₅, and BMP4 (10 ng/mL each) to facilitate endothelial lineage specification and proliferation. The differentiation medium was refreshed on days 3, 5, 7, and 9. On day 12, cultures were switched to a mixture of three parts EGM-2 endothelial growth medium and one part Stemline II complete medium. Endothelial colony-forming cell (ECFC)-like clusters appeared as tightly adherent and characterized as described previously [47]. ECFC cells were infected with cBu1-vIL-6 KSHV at viral genomic copy/cell ratio of 10:1.

Virus preparation

For virus production, iSLK.r219 cell lines were stimulated with 1 mM sodium butylate and 1 μg/ml doxycycline. Two days after chemical induction, sodium butylate, and doxycycline were removed, and cells were incubated with fresh DMEM for another 2 or 3 days. Recombinant KSHV particles were collected from the culture supernatant after centrifugation (2000 rpm, 3 min) and stored at -80°C until use. To quantify the encapsidated viral DNA copies, viral supernatants were first treated with MgSO₄ (100 mM) and DNase I (12 μg/mL) for 15 min at room temperature to degrade un-capsidated DNA. This reaction was stopped by the addition of EDTA to 5 mM, followed by heating at 70°C for 15 min. Viral genomic DNA was purified using the QIAamp DNA Mini Kit according to the manufacturer’s protocol. Elute solutions were used for real-time qPCR to determine viral copy number, as described previously [67].

Preparation of recombinant KSHV BACs

BAC16 was a generous gift from Dr. Jae U. Jung [Cleveland Clinic Learner Center [23]]. To test the role of TR, the number of TR copies was manipulated with BAC recombination. Long primer pairs that encode a partial sequence of TR were used to prepare transfer DNA fragments for recombinant by amplifying the Kanamycin cassette from the pEP-Kan plasmid [24]. The amplified DNA fragment was transformed into BAC16 KSHV containing E.coli for recombination. The DNA fragments were randomly integrated into TR, which is a direct repeat of 801 bp fragments. With the I-SceI induction with L-Arabinose, we induced double-strand breaks at direct repeat regions, and subsequent induction of red recombinase with increased temperature induces random recombination within TRs. The resulting BAC clones contain different numbers of TR copies due to recombination at homologous recombination within TRs. The Kanamycin-sensitive E.coli clones were cultured in 10 mL of LBs, and purified BAC DNAs were digested with PstI or KpnI to estimate the number of TR copies. Based on the agarose gel electrophoresis and EtBr staining, we selected several E.coli clones containing different numbers of TR copies. Similarly, the GSG-P2A chicken Bu1 extracellular domain was inserted in the N-terminal region of vIL-6 (K2) as a fusion. For that, the codon-optimized cBu1 fragment was synthesized and cloned into a pBS vector (Table 1). Kanamycin expression cassette with I-SceI restriction enzyme fragment with 50 bp homology arm was amplified and cloned into PstI restriction enzyme site in the cBu1 fragment. Long primer pairs with homology arm with K2 N-terminal region were used to amplify the cBu1-Kan fragment, and recombination was performed by transforming purified PCR fragment into BAC containing E.coli as described previously. The Kanamycin cassette was removed with red-recombination, and the surrounding sequence was confirmed by sequencing. We confirmed junctions and the cBu1-vIL-6 coding sequence. Integrity of the BAC was confirmed by restriction enzyme digestion of the purified BACs. The resulting cBu-1-vIL-6 BAC was used as a backbone and generated TR deletion BACs, as described above.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Primer and synthesized DNA fragments used in this study.

https://doi.org/10.1371/journal.ppat.1013061.t001

KSHV genome copy number

KSHV episomal DNA was isolated from TR5, TR21 and BCBL-1 cells using the Hirt extraction method, which selectively enriches for low molecular weight DNA. Briefly, 1 × 10⁶ cells were lysed in Hirt lysis buffer (10 mM Tris-HCl pH 8.0, 0.6% SDS, 10 mM EDTA) and incubated with 1 M NaCl (final concentration) overnight at 4°C to precipitate high molecular weight chromosomal DNA. The supernatant containing episomal DNA was collected following centrifugation at 15,000 × g for 30 minutes and subjected to phenol-chloroform extraction and ethanol precipitation. To eliminate contaminating linear DNA, the extracted samples were treated with Plasmid-Safe ATP-dependent DNase (PSAD; Biosearch Technologies) overnight at 37°C, according to the manufacturer’s protocol. This enzymatic treatment selectively degrades linear double-stranded DNA, preserving circular episomal DNA. Quantification of KSHV episome copy number was performed by quantitative PCR (qPCR) using ORF21 primers. A standard curve was generated using serial dilutions of a known copy number of KSHV BAC16 DNA and used to calculate the absolute copy number in the samples. The episome copy number was normalized to the cell number.

Preparation of fluorescence tagged recombinant protein

pFAST-BAC1 plasmid was first engineered to have an N-terminal Flag tag and fluorescence proteins, EGFP or mCherry. The restriction enzyme site, the CpoI site, was included as a cDNA cloning site to make a fusion protein with EGFP or mCherry. The cDNAs of our protein of interest were cloned into the CpoI sites, and recombinant baculoviruses were prepared with the Bac to Bac system as previously described [67,68]. The transfer plasmid, pFAST-BAC1 vectors encoding Flag-EGFP-LANA, Flag-mCherry-BRD4, Flag-EGFP alone, or Flag-mCherry tag alone was recombined with Bac to Bac system. Recombinant baculovirus bacmid DNA isolated from E.coli was transfected into Sf9 cells by using polyethylenimine (Sigma), and recombinant viruses were subsequently amplified twice. Expression of recombinant proteins was confirmed by immunoblotting with anti-Flag monoclonal antibody (Sigma). Large-scale cultures of Sf9 cells (50 ml) were infected with recombinant baculovirus at a multiplicity of infection (MOI) of 0.1 to 1.0, and cells were harvested 48 hrs after infection. Recombinant proteins were purified as described previously [67]. The purity and amount of protein were measured by BCA assay (Invitrogen) and further confirmed by SDS-PAGE and coomassie blue staining using bovine serum albumin (BSA) as a standard.

Immunoblotting

Cells were washed with PBS, lysed in lysis buffer (50 mM Tris-HCl [pH 6.8], 2% SDS, 10% glycerol), and boiled for 3 min. The protein concentrations of the lysates were quantified with a BCA Protein Assay Kit (Thermo Fisher). Protein samples were separated by SDS-PAGE using 10% agarose gel and transferred to transfer membranes (Millipore-Sigma, St. Louis, MO, USA), which were incubated in 5% non-fat milk at room temperature for 2 hours. The membrane was incubated with the primary antibody at 4°C overnight or at room temperature for 2 hours. The membrane was then incubated with horseradish-peroxidase-conjugated secondary antibody or Alexa-647-conjugated secondary antibody at 25°C for 1 hour. For cell fractionation, cells were suspended with hypotonic buffer (20 mM Tris-HCl (pH 7.4), 10 mM NaCl, 3 mM MgCl2, 0.5 mM DTT, and proteinase inhibitor cocktail) for 15 min on ice and 0.5% NP-40 was added. Cells were then centrifuged for 10 min at 3000 rpm to collect the supernatants (cytoplasm fraction). Pellets were washed with PBS 2 times and suspended with protein lysis buffer (nuclear fraction).

RT-qPCR

Cells were washed with PBS, and total RNA was extracted using the Quick-RNA miniprep kit (Zymo Research, Irvine, CA, USA). Purified RNA was incubated with DNase I for 15 min and reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher, Waltham, MA USA). The resulting cDNA was used for qPCR. SYBR Green Universal master mix (Bio-Rad) was used for qPCR according to the manufacturer’s instructions. Each sample was normalized to β-actin RNA, and the duct fold change method was used to calculate relative quantification. Primers used for qPCR is listed in Table 1. All reactions were run in triplicate using specific primers designed by PrimerQuest (Integrated DNA Technologies).

Cleavage Under Targets and Release Using Nuclease (CUT&RUN)

CUT & RUN [69] was performed by following the online protocol developed by Dr. Henikoff’s lab with a few modifications to fit our needs. Cells were washed with PBS and wash buffer [20 mM HEPES-KOH pH 7.5, 150 mM NaCl, 0.5 mM Spermidine (Sigma, S2626), and proteinase inhibitor (Roche)]. After removing the wash buffer, cells were captured on magnetic concanavalin A (ConA) beads (Polysciences, PA, USA) in the presence of CaCl₂. Beads/cells complexes were washed three times with digitonin wash buffer (0.02% digitonin, 20 mM HEPES-KOH pH 7.5, 150 mM NaCl, 0.5 mM Spermidine and 1x proteinase inhibitor), aliquoted, and incubated with specific antibodies (1:50) in 250 μL volume at 4°C overnight. After incubation, the unbound antibody was removed with digitonin wash buffer three times. Beads were then incubated with recombinant Protein A/G–Micrococcal Nuclease (pAG-MNase), which was purified from E.coli in 250 μl digitonin wash buffer at 1.0 μg/mL final concentration for 1 h at 4 °C with rotation. Unbound pAG-MNase was removed by washing with digitonin wash buffer three times. Pre-chilled digitonin wash buffer containing 2 mM CaCl₂ (200 μL) was added to the beads and incubated on ice for 30 min. The pAG-MNase digestion was halted by the addition of 200 μl 2 × STOP solution (340 mM NaCl, 20 mM EDTA, 4 mM EGTA, 50 μg/ml RNase A, 50 μg/ml glycogen). The beads were incubated with shaking at 37 °C for 10 min in a tube shaker at 300 rpm to release digested DNA fragments from the insoluble nuclear chromatin. The supernatant was then collected by removing the magnetic beads. DNA in the supernatant was purified using the NucleoSpin Gel & PCR kit (Takara Bio, Kusatsu, Shiga, Japan). qPCR was used to examine enrichment at selected genomic regions. CUT&RUN-qPCR data were normalized to input DNA.

Immunofluorescence staining

Cells were cultured on coverslips and fixed with 2% PFA for 10 min at room temperature. Cells were incubated in PBS with 0.1% Triton X-100 for permeabilization for 20 min at room temperature and blocked (1% BSA, 0.05% Tween-20 in PBS) for 60–120 min at room temperature. The cells were incubated with anti-LANA rat monoclonal antibody (clone LN53, Millipore Sigma) diluted in blocking buffer (1:100) for overnight at 4 °C. After staining overnight at 4 °C, cells were washed by permeabilization buffer three times. The cells were incubated with Alexa Fluor 647 goat anti-rat IgG (Invitrogen, 1:1,000) for 1 hour in the dark at room temperature. Cells were counterstained with 4,6-diamidino-2-phenylindole (DAPI) and examined under All-in-One Fluorescence Microscope (BZ-X710, KEYENCE)

Proximity ligation assay

The Proximity Ligation Assay (PLA) was performed by Duolink Proximity Ligation Assay (Millipore Sigma) according to the manufacturer’s instructions. Briefly, cells were seeded on glass coverslips in a 6-well plate at a density of 3 × 10⁵ cells per well and cultured overnight in DMEM medium. Cells were fixed with 2% paraformaldehyde in PBS for 10 min at room temperature, permeabilized with 0.1% Triton X-100 in PBS for 10 min, and blocked with the 1X blocking buffer in the kit. Primary antibodies, including anti-LANA rabbit antibody and anti-BRD4 rabbit antibody, were diluted (1:100) in the blocking buffer and incubated with the samples overnight at 4°C. After washing, the samples were incubated with PLA probes (anti-mouse MINUS and anti-rabbit PLUS) for 1 hour at 37°C. Ligation was performed by incubating the samples with ligase solution for 30 min at 37°C, followed by amplification with polymerase solution for 100 min at 37°C. Coverslips were mounted onto slides using 10 μl Duolink In Situ Mounting Medium with DAPI. Images were taken by All-in-One Fluorescence Microscope (BZ-X710, KEYENCE)

In vitro phase separation assay

To induce phase separation, purified proteins were diluted to a final concentration of 10 µM in a buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 1 mM DTT and 4% polyethylene glycol. The mixture was incubated at room temperature for 10 min. Phase separation was visually confirmed by fluorescence microscopy (BZ-X710, KEYENCE). The protein solution was loaded on a glass slide with a coverslip attached by two parallel strips of tape.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 9.4.1 software. Results are shown as mean ± SD with dots representing individual measurements. Statistical significance was determined by Student’s t-test, ratio paired t-test, or one-way ANOVA with Tukey’s multiple comparison test, and correction for false discovery rate (FDR) as described in each figure legend. FDR corrected p < 0.05 was considered statistically significant.