ARID3B: a Novel Regulator of the Kaposi's Sarcoma-Associated Herpesvirus Lytic Cycle

Abstract

Kaposi's sarcoma-associated herpesvirus (KSHV) is the causative agent of commonly fatal malignancies of immunocompromised individuals, including primary effusion lymphoma (PEL) and Kaposi's sarcoma (KS). A hallmark of all herpesviruses is their biphasic life cycle-viral latency and the productive lytic cycle-and it is well established that reactivation of the KSHV lytic cycle is associated with KS pathogenesis. Therefore, a thorough appreciation of the mechanisms that govern reactivation is required to better understand disease progression. The viral protein replication and transcription activator (RTA) is the KSHV lytic switch protein due to its ability to drive the expression of various lytic genes, leading to reactivation of the entire lytic cycle. While the mechanisms for activating lytic gene expression have received much attention, how RTA impacts cellular function is less well understood. To address this, we developed a cell line with doxycycline-inducible RTA expression and applied stable isotope labeling of amino acids in cell culture (SILAC)-based quantitative proteomics. Using this methodology, we have identified a novel cellular protein (AT-rich interacting domain containing 3B [ARID3B]) whose expression was enhanced by RTA and that relocalized to replication compartments upon lytic reactivation. We also show that small interfering RNA (siRNA) knockdown or overexpression of ARID3B led to an enhancement or inhibition of lytic reactivation, respectively. Furthermore, DNA affinity and chromatin immunoprecipitation assays demonstrated that ARID3B specifically interacts with A/T-rich elements in the KSHV origin of lytic replication (oriLyt), and this was dependent on lytic cycle reactivation. Therefore, we have identified a novel cellular protein whose expression is enhanced by KSHV RTA with the ability to inhibit KSHV reactivation.

Importance: Kaposi's sarcoma-associated herpesvirus (KSHV) is the causative agent of fatal malignancies of immunocompromised individuals, including Kaposi's sarcoma (KS). Herpesviruses are able to establish a latent infection, in which they escape immune detection by restricting viral gene expression. Importantly, however, reactivation of productive viral replication (the lytic cycle) is necessary for the pathogenesis of KS. Therefore, it is important that we comprehensively understand the mechanisms that govern lytic reactivation, to better understand disease progression. In this study, we have identified a novel cellular protein (AT-rich interacting domain protein 3B [ARID3B]) that we show is able to temper lytic reactivation. We showed that the master lytic switch protein, RTA, enhanced ARID3B levels, which then interacted with viral DNA in a lytic cycle-dependent manner. Therefore, we have added a new factor to the list of cellular proteins that regulate the KSHV lytic cycle, which has implications for our understanding of KSHV biology.

FIG 1

Analysis via SILAC coupled to LC-MS/MS of nuclear proteome changes in response to KSHV-RTA expression. Flp-In-293 cells were used to develop a cell line with doxycycline (Dox)-inducible FLAG-tagged RTA expression (iRTA-293). (A) Cells were labeled with R10K8 (heavy)- or R0K0 (light)-labeled SILAC medium for 2 weeks. FLAG-RTA expression was induced in heavy-labeled cells for 12 h, and these were fractionated into nuclear and cytoplasmic compartments. Fractionation success was validated by immunoblot analysis using the nuclear (Lamin-B1) and cytoplasmic (GAPDH) markers. As expected, RTA expression was largely restricted to the nuclear compartment and was found only in doxycycline-treated cells. (B) FLAG-RTA expression was analyzed by confocal immunofluorescence that revealed that all doxycycline-treated cells were positive for FLAG-RTA and untreated cells (-Dox; R0K0-labeled cells) were negative for RTA expression, demonstrating tight regulation of FLAG-RTA. Bars, 10 μm. (C) KEGG pathway analysis of nuclear proteins that were upregulated 2-fold or more due to RTA expression, with at least 2 independent peptides revealing various cellular pathways associated with RTA expression. GO term, gene ontology term.

FIG 2

RTA enhances ARID3B expression. (A) Doxycycline-induced expression in iRTA-293 cells for 12 h and reactivation of the KSHV lytic cycle by the addition of doxycycline in TREx-BCBL1-RTA cells increased ARID3B expression as measured by qRT-PCR. Error bars represent standard deviations from the mean from three independent biological replicates, and quantification was normalized to GAPDH using the ΔΔC_T method. Student's t test was used to determine statistical significance. Three independent experiments were performed for each cell line. (B) Reactivation of the lytic cycle from latently infected iSLK-BAC16 and TREx-BCBL1-RTA cells led to an increase in ARID3B protein expression. In this experiment, separate immunoblots were used for the detection of RTA (denoted by a line). Dox., doxycycline. (C) Quantification of protein expression in panel B (see Materials and Methods). Data are derived from two independent experiments, and error bars represent standard deviations of the mean (calculated from technical replicates within independent experiments). *, P < 0.05 (Student's t test).

FIG 3

ARID3B relocalizes to KSHV replication compartments upon reactivation of the lytic cycle. (A) Confocal immunofluorescence analysis demonstrated that ARID3B relocalized to discrete, RTA-positive foci that resemble replication compartments (white arrows) upon doxycycline (Dox)-induced reactivation of the KSHV lytic cycle in TREx-BCBL1-RTA cells. (B) Further evidence suggesting that ARID3B relocalized to replication compartments, demonstrated by its colocalization with RTA and EdU-positive foci (white arrows). EdU is used as a marker of newly replicated virus (57). Bar, 10 μm.

FIG 4

ARID3B inhibits KSHV lytic reactivation. (A) Latently infected iSLK-BAC16 cells were independently transfected with two separate siRNA pools targeting ARID3B (or a nontargeting scramble control siRNA) for 72 h and then reactivated by addition of doxycycline (a further 24 h; see Materials and Methods). RT-qPCR was used to quantify ARID3B, ORF57, and gB levels, and the fold difference in expression was compared to scramble control using GAPDH levels to normalize between samples. Error bars represent standard deviations from the mean from two independent experiments, with mean values calculated from the technical replicates of each experiment. Statistical analyses demonstrated a significant increase in lytic gene expression in ARID3B siRNA-treated cells compared to scramble control. **, P < 0.0001 (Student's t test). (B) Knockdown of ARID3B led to an enhancement of lytic protein expression. (C) Quantification of panel B (see Materials and Methods). (D) Knockdown of ARID3B enhanced viral genome replication. ARID3B siRNA or scramble control siRNA transfection of iSLK-BAC16 cells followed by reactivation of the lytic cycle for 72 h led to an increase in viral genome replication, as quantified by qPCR of the KSHV ORF57 gene. Cellular GAPDH was used to normalize between samples, and error bars represent standard deviations from the mean from three independent biological replicates (*, P < 0.05). (E) Knockdown of ARID3B enhances virion production. Media from iSLK-BAC16 cells that had been transfected with ARID3B siRNA or scramble control siRNAs were used to infect KSHV-negative HEK293T cells. As a measure of infection, RT-qPCR analysis of the KSHV lytic gene ORF57 was performed 24 h following infection. Data represent three independent infections in a single experiment, and Student's t test was used to determine statistical significance (*, P < 0.05). (F) Overexpression of FLAG-ARID3B inhibits reactivation of lytic cycle-associated protein expression. Expression vector containing FLAG-tagged ARID3B (FLAG-ARID3B) or FLAG-SRAG (used as a negative control) was transfected into iSLK-BAC16 cells for 24 h, followed by doxycycline (Dox.)-induced reactivation of the KSHV lytic cycle. Cell lysates were harvested 24 h later and subjected to immunoblot analysis of lytic proteins ORF57 and mCP (minor capsid protein).

FIG 5

RTA and ARID3B do not directly interact. Immunoprecipitation assay showing that RTA interacts with known binding partner ORF59 but not ARID3B. The lytic cycle was induced in ARID3B-transfected iSLK-BAC16 cells. Proteins were immunoprecipitated with the indicated antibodies, followed by immunoblot (IB) analysis.

FIG 6

ARID3B interacts with the KSHV genome in a lytic reactivation-dependent manner. (A) DNA affinity assays performed according to the method in reference ; biotinylated PCR products spanning oriLyt left of KSHV (GenBank accession number NC_009333.1) and control DNA amplified from the RTA coding region were bound to streptavidin-Dynabeads and incubated with lysates from reactivated 293T rKSHV.219 cells expressing FLAG-ARID3B. This suggested that ARID3B bound these sequences with preference for the A/T-rich region. IB, immunoblotting. (B) TREx-BCBL1-RTA cells were treated with doxycycline for 18 h (w/Dox) to reactivate the lytic cycle or were left untreated (w/o Dox). Samples were subjected to ChIP with an antibody to ARID3B or normal rabbit IgG and primers specific for sequences in the A/T-rich region of oriLyt or the RTA coding region. The percentage of output versus input DNA was calculated and is presented relative to normal rabbit IgG values from unreactivated (w/o Dox) cells (set to 1). Data represent two biological replicates (i.e., independent ChIPs) and two technical replicates per ChIP from a single experiment, and values are given as the mean ± standard deviation. Student t tests were used to determine statistical significance.

FIG 7

A model of how ARID3B inhibits lytic reactivation. Periodic reactivation of KSHV involves replication of the viral genome and the production of new virions, a necessary step for the maintenance of latency and the pathogenesis of KS (2). Viral DNA replication initiates from the cis-acting oriLyt where proteins essential for this process are recruited. These include viral proteins RTA and K8 and the six core proteins (DNA polymerase, DNA processivity factor, etc.) (80) in addition to various cellular trans-acting proteins (25). We hypothesize that during reactivation RTA is expressed, leading to the enhanced expression of ARID3B and its recruitment to viral genomes, specifically AT-rich elements such as those found in oriLyt. ARID3B may compete with factors required for reactivation for binding with the KSHV genome in order to modulate the levels of reactivation or for the establishment of latency.