Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 86 (2), 1021-33

Cooperation Between Viral Interferon Regulatory Factor 4 and RTA to Activate a Subset of Kaposi's Sarcoma-Associated Herpesvirus Lytic Promoters

Affiliations

Cooperation Between Viral Interferon Regulatory Factor 4 and RTA to Activate a Subset of Kaposi's Sarcoma-Associated Herpesvirus Lytic Promoters

Xiangmei Xi et al. J Virol.

Abstract

The four Kaposi's sarcoma-associated herpesvirus (KSHV)-encoded interferon (IFN) regulatory factor homologues (vIRF1 to vIRF4) are used to counter innate immune defenses and suppress p53. The vIRF genes are arranged in tandem but differ in function and expression. In KSHV-infected effusion lymphoma lines, K10.5/vIRF3 and K11/vIRF2 mRNAs are readily detected during latency, whereas K9/vIRF1 and K10/vIRF4 mRNAs are upregulated during reactivation. Here we show that the K10/vIRF4 promoter responds to the lytic switch protein RTA in KSHV-infected cells but is essentially unresponsive in uninfected cells. Coexpression of RTA with vIRF4 is sufficient to restore regulation, a property not shared by other vIRFs. The K9/vIRF1 promoter behaves similarly, and production of infectious virus is enhanced by the presence of vIRF4. Synergy requires the DNA-binding domain (DBD) and C-terminal IRF homology regions of vIRF4. Mutations of arginine residues within the putative DNA recognition helix of vIRF4 or the invariant cysteines of the adjacent CxxC motif abolish cooperation with RTA, in the latter case by preventing self-association. The oligomerization and transactivation functions of RTA are also essential for synergy. The K10/vIRF4 promoter contains two transcription start sites (TSSs), and a 105-bp fragment containing the proximal promoter is responsive to vIRF4/RTA. Binding of a cellular factor(s) to this fragment is altered when both viral proteins are present, suggesting a possible mechanism for transcriptional synergy. Reliance on coregulators encoded by either the host or viral genome provides an elegant strategy for expanding the regulatory potential of a master regulator, such as RTA.

Figures

Fig 1
Fig 1
vIRF4 and RTA cooperate to activate the K10/vIRF4 and K9/vIRF1 promoters in uninfected cells. (A) Schematic showing the vIRF gene cluster and surrounding open reading frames. The four vIRFs (K9/vIRF1, K10/vIRF4, K10.5/vIRF3, and K11/vIRF2) are arranged in tandem and are flanked by ORF57 (posttranscriptional regulator Mta) and ORF58 (EBV BMRF2 homologue). The TSSs (arrows) of the genes in this region have been mapped by 5′ rapid amplification of cDNA ends or primer extension (12, 31, 50, 51). Promoter fragments extending from immediately upstream of the initiation codon and incorporating the TSSs upstream of ORF57, K9/vIRF1, and K10/vIRF4 are shown as black bars below the map. (B) The K10/vIRF4 (vIRF4-luc) and Mta/ORF57 (ORF57-luc) reporters were transiently transfected into KSHV-positive BC3 and JSC-1 cells (open bars) or KSHV-negative HeLa and SLK cells (filled bars) together with an expression vector encoding RTA or an empty vector. The constitutively active LTc promoter was assayed in parallel in order to normalize transfection efficiency. The relative activity of the reporter in the presence or absence of RTA was calculated for each cell line after normalization and plotted as the fold induction, with standard deviations. (C) Response of the K9/vIRF1 reporter (vIRF1-luc) to RTA in KSHV-positive BC3 and JSC-1 cells (open bars) and KSHV-negative HeLa and SLK cells (filled bars). Values were calculated as described for panel B. (D) HeLa cells were transfected with the indicated reporters (250 ng/well of a 24-well plate) and expression plasmids encoding vIRF4 (200 ng) and/or RTA (25 ng). Fold induction was calculated relative to the reporter alone (mock). Values represent the means and standard errors of the means of three independent transfections.
Fig 2
Fig 2
Synergy with RTA is a unique property of vIRF4. (A) Schematic comparing the general structure of the four KSHV vIRFs with cellular IRF7. Two regions of sequence homology are indicated: the putative DNA-binding domain at the N terminus (black) and part of the C-terminal transactivation/dimerization domain (shaded). (B) HeLa cells were transfected with 250 ng vIRF4-luc and expression plasmids carrying RTA (25 ng) and/or vIRF1, vIRF2, vIRF3, vIRF4, and IRF7 (200 ng). Fold induction was calculated relative to the reporter alone (mock). Values represent the means and standard errors of the means of three independent transfections. (C) Assay using vIRF1-luc cotransfected with plasmids expressing RTA (25 ng) and/or vIRF1 and vIRF4 (200 ng).
Fig 3
Fig 3
vIRF4 contributes to KSHV reactivation. (A) KSHV latently infected iSLK.219 cells were either mock treated or transfected with empty vector or vector expressing full-length vIRF4. The next day, the transfected cultures were induced with 100 ng/ml doxycycline (DOX) to drive expression of a DOX-regulated RTA cDNA stably integrated into the cell genome and maintained for 72 h. Culture medium was collected, filtered to remove debris, including cells, and used to infect 293-PAN-Luc reporter cells. After 48 h, lysates were prepared and assayed for luciferase activity. Values represent the means and standard errors of the means of three independent transfections. (B) iSLK.219 cells were transfected with dsiRNAs against EGFP or K10/vIRF4 mRNA sequences. Cultures were induced to reactivate by using doxycycline, and RNA was harvested 72 h later and analyzed by quantitative RT-PCR using primers to detect K9/vIRF1 (open bars) and K10/vIRF4 (filled bars). (C) Results of an experiment similar to that shown in panel B, except that the culture medium was collected from induced and uninduced iSLK.219 cells and assayed for infectious KSHV virions by using 293-PAN-Luc cells. Values are expressed relative to the averages of the uninduced samples.
Fig 4
Fig 4
Analysis of the K10/vIRF4 promoter. (A) Schematic showing K10/vIRF4 and K10.5/vIRF3 and the intergenic region. Transcription of K10/vIRF4 initiates at two discrete sites (thin arrows). The 0.9-kb fragment used in the vIRF4-luc reporter is shown below, with the three additional truncations (Δ1, Δ2, and Δ3). Genomic coordinates of each promoter fragment are given on the right. (B) Activity assay results in HeLa cells transfected with full-length (FL) and truncated versions of vIRF4-luc (250 ng), vIRF4 (200 ng), and RTA (37.5 ng). Values are plotted relative to the mean activity of FL vIRF4-luc (set at 100%). Data are derived from three independent transfections. (C) Response of vIRF4Δ3-luc to RTA (37.5 ng) and/or vIRF4 (200 ng). Activity is plotted relative to the reporter in the presence of RTA only. (D) Nucleotide sequence of the K10/vIRF4 promoter region (vIRF4-luc) and the first codons of exon 1 (corresponding to nucleotides 88,902 to 89,815 of the BC-1 reference genome). Distal and proximal TPA-inducible transcription start sites (+1) (12) are shown in bold lowercase letters. Putative TATA boxes located 30 to 31 nucleotides upstream of each TSS are highlighted. Matches to the CSL and C/EBP binding site consensus sequences are also indicated. The minimal vIRF4/RTA-responsive fragment (vIRF4Δ3-luc) is underlined.
Fig 5
Fig 5
N- and C-terminal domains of vIRF4 are required for RTA synergy. (A) Schematic showing limits of the N- and C-terminal vIRF4 truncations. Expression levels were assessed by immunoblotting using anti-vIRF4 antiserum (kind gift of Jae Jung and Hye-Ra Lee). (B) Activity assay in HeLa cells transfected with vIRF4-luc reporter (250 ng) and expression plasmids carrying full-length or truncated vIRF4 (200 ng) in the presence or absence of RTA (37.5 ng). Activity (fold induction) was calculated relative to that of the reporter with full-length vIRF4 only. Values represent the means and standard errors of the means of three independent transfections. (C) vIRF4 and RTA independently localize to the nucleus. Indirect immunofluorescence analysis of HeLa cells transfected with plasmids expressing T7-epitope-tagged RTA and Flag-epitope-tagged vIRF4 proteins (200 ng each). After fixation, coverslips were probed with a mix of anti-T7 (rabbit polyclonal) and anti-Flag (mouse monoclonal) antibodies followed by a mixture of anti-mouse and anti-rabbit fluorescence-coupled secondary antibodies.
Fig 6
Fig 6
Mutations in helix 3 of vIRF4 disrupt synergy with RTA. (A) Primary sequence of the vIRF4 N terminus (residues 1 to 98), showing the locations of the predicted α-helices (filled cylinders) and β-strands (horizontal arrows). Helix 3 is also shown as a helical wheel projection with hydrophobic residues (F, L, V, and Y) clustered on one face (lower right). The four arginine residues (R83, R86, R89, and R90) that were changed in turn to alanine are indicated with an asterisk. (B) Immunoblot (with anti-Flag antibody) of lysates prepared from cells transfected with each vIRF4 plasmid. The species corresponding to full-length vIRF4 is indicated. Equal loading was demonstrated by blotting for α-tubulin. (C) Activity assay using HeLa cells transfected with vIRF-luc (250 ng) and each vIRF4 derivative (200 ng) in the presence of RTA (37.5 ng). Fold induction was calculated relative to the reporter alone. Values represent the means and standard errors of the means of three independent transfections.
Fig 7
Fig 7
The signature cysteine motif in vIRF4 is essential for the transactivation function. (A) Schematic of vIRF4. The inset shows the sequence immediately downstream of helix 3, including the CxxC motif common to all KSHV vIRFs (and RRV vIRFs [data not shown]). Corresponding sequences of KSHV vIRF1, vIRF2, and vIRF3 are shown using the invariant arginine (R89 in vIRF4) to anchor the alignment. Cysteine-120 and cysteine-123 were individually changed to alanine. (B) Immunoblot (with anti-Flag antibody) of lysates prepared from cells transfected with each vIRF4 plasmid. The species corresponding to full-length vIRF4 is indicated. Equal loading was demonstrated by blotting for α-tubulin. (C) Activity assay using HeLa cells transfected with vIRF4-luc (250 ng) and each vIRF4 derivative (200 ng) in the presence of RTA (25 ng). Fold induction was calculated relative to the reporter alone. Values represent the means and standard errors of the means of three independent transfections. (D) vIRF4 is capable of self-association. Lysates were prepared from HeLa cells transfected with plasmids encoding combinations of T7 and Flag-tagged proteins and subjected to immunoprecipitation using Flag-coupled beads. Precipitates were resolved by SDS-PAGE and blotted with anti-T7 antibody. Expression was confirmed by blotting lysates directly. (E) Self-association is disrupted by mutation of the signature cysteine residues. Coimmunoprecipitation analysis was conducted as described for panel D, using wild-type (WT) and mutant (C120A and C123A) versions of vIRF4.
Fig 8
Fig 8
Transactivation and DNA-binding functions of RTA contribute to synergy. (A) Schematic showing known functional domains in RTA. DNA binding is mediated by the N terminus, which includes a region of positively charged residues (basic) and a leucine-rich region (LR) that is critical for oligomerization. The C terminus is required for transcriptional activation mediated by a serine/threonine-rich region (ST) and a separate activation domain (AD). Derivatives lacking the transactivation region (RTAΔSTAD) or leucine-rich region (RTAΔLR) are shown below. Mutation of R166 within the basic region disrupts DNA binding. (B) Immunoblot (with anti-Flag antibody) of lysates prepared from HeLa cells transfected with each RTA plasmid. Detection of Rho-GDI showed equal loading. (C) Activity assay using HeLa cells transfected with PAN-luc or LTi-luc (250 ng) and each RTA expression plasmid (25 ng). Fold induction was calculated relative to the reporter alone. Values represent the means and standard errors of the means of three independent transfections. (D) Activity assay results with HeLa cells transfected with vIRF1-luc or vIRF4-luc (250 ng) and each RTA expression plasmid (25 ng) in the presence of wild-type vIRF4 (200 ng).
Fig 9
Fig 9
vIRF4 and RTA antagonize binding of a cellular factor(s) to the K10/vIRF4 proximal promoter. (A) Reticulocyte lysates were programmed with T7 RNA polymerase-transcribed plasmids carrying full-length vIRF4 and RTA in the presence of [35S]methionine. In vitro translation products were resolved by 8% SDS-PAGE and visualized by autoradiography. Relative molecular masses (in kDa) were determined using prestained protein standards run concurrently. The position of the K10/vIRF4 promoter Δ3 probe fragment encompassing the proximal promoter is shown. (B) Gel mobility shift analysis results with a 32P-labeled K10/vIRF4 promoter Δ3 fragment probe incubated with rabbit reticulocyte lysates programmed with empty vector (lane 2) or plasmids encoding RTA or vIRF4 (lanes 3 to 8). Nonspecific complexes observed with multiple probes are marked with an asterisk. Arrows mark a low-mobility complex that incorporated RTA and an uncharacterized complex (X) that was sensitive to the presence of vIRF4 and RTA. (C) Schematic of a working model for activation of the K10/vIRF4 promoter by RTA and vIRF4. RTA and a cellular factor(s) (X) associate with the proximal promoter but do not activate transcription. Inclusion of vIRF4 leads to reprogramming of the complex, either by displacement of X or by incorporation of vIRF4 into a larger complex that is not resolved on the gel. Cellular coactivators recruited by RTA and/or other components of the complex are then able to activate transcription.

Similar articles

See all similar articles

Cited by 15 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms

Substances

LinkOut - more resources

Feedback