K13 blocks KSHV lytic replication and deregulates vIL6 and hIL6 expression: a model of lytic replication induced clonal selection in viral oncogenesis

Abstract

Background: Accumulating evidence suggests that dysregulated expression of lytic genes plays an important role in KSHV (Kaposi's sarcoma associated herpesvirus) tumorigenesis. However, the molecular events leading to the dysregulation of KSHV lytic gene expression program are incompletely understood.

Methodology/principal findings: We have studied the effect of KSHV-encoded latent protein vFLIP K13, a potent activator of the NF-kappaB pathway, on lytic reactivation of the virus. We demonstrate that K13 antagonizes RTA, the KSHV lytic-regulator, and effectively blocks the expression of lytic proteins, production of infectious virions and death of the infected cells. Induction of lytic replication selects for clones with increased K13 expression and NF-kappaB activity, while siRNA-mediated silencing of K13 induces the expression of lytic genes. However, the suppressive effect of K13 on RTA-induced lytic genes is not uniform and it fails to block RTA-induced viral IL6 secretion and cooperates with RTA to enhance cellular IL-6 production, thereby dysregulating the lytic gene expression program.

Conclusions/significance: Our results support a model in which ongoing KSHV lytic replication selects for clones with progressively higher levels of K13 expression and NF-kappaB activity, which in turn drive KSHV tumorigenesis by not only directly stimulating cellular survival and proliferation, but also indirectly by dysregulating the viral lytic gene program and allowing non-lytic production of growth-promoting viral and cellular genes. Lytic Replication-Induced Clonal Selection (LyRICS) may represent a general mechanism in viral oncogenesis.

Figure 1. K13 blocks lytic replication in BCBL1-TREx-RTA cells.

A. Expression of K13-ER^TAM in BCBL1-TREx-RTA cells as determined by immunoblotting with a Flag antibody. B–C. Treatment with 4OHT induces nuclear translocation (B) and DNA-binding (C) of p65 in BCBL1-TREx-RTA cells expressing K13-ER^TAM but is without effect in the control cells. Nuclear translocation was measured by indirect immunofluorescence analysis using a p65/RelA primary antibody (Santa Cruz Biotechnology). D. Inhibition of TPA-induced K8.1 and ORF59 expression by K13. BCBL1-TREx-RTA cells expressing an empty vector and K13-ER^TAM, respectively, were left untreated or pretreated with 4OHT for 18 h and then induced with TPA for 96 h. K8.1 and ORF59 expression was detected by indirect immunofluorescence analysis with the indicated antibodies and revealed by Alexa-488-conjugated secondary antibodies. Nuclei were counterstained with Hoechst 33342. Cells were imaged with an Olympus Fluorescent microscope equipped with a SPOT camera. A representative of two independent experiments is shown.

Figure 2. K13 blocks RTA-induced KSHV lytic reactivation.

A. 4OHT treatment blocks RTA-induced K8.1 and ORF59 expression in K13-ER^TAM cells. The experiments were performed essentially as described for Figure 1D with the exception that RTA expression was induced by treatment with doxycycline (10 ng/ml). A representative of two independent experiments is shown. B. Flow cytometry analysis showing inhibition of RTA-induced K8.1 expression by 4OHT pretreatment in K13-ER^TAM cells. C. Equivalent induction of RTA upon doxycycline treatment in the vector- and K13-ER^TAM -expressing BCBL1-TREx-RTA cells in the absence or presence of prior treatment with 4OHT. Cells were treated with the indicated doses of doxycycline for 72 h prior to immunoblotting. D. K13 blocks RTA/doxycycline-induced production of infectious virions. 293PAN-Luc cells were infected in triplicate in a 24 well plate with 200 µl of cell-free supernatant collected from cells described in 2A. 72 h post-infection, luciferase activity was measured in cell lysates. The values (Mean±SEM) shown are from a representative of three independent experiments performed in triplicate. E. A semi-quantitative PCR assay showing inhibition of RTA/doxycycline-induced production of infectious virions by K13 in the cellular supernatants collected in 2D.

Figure 3. Role of the NF-κB pathway in the inhibition of KSHV lytic reactivation by K13.

A. Status of the NF-κB pathway in BCBL1-TREx-RTA cells expressing wild-type K13 and its NF-κB-defective mutants, K13-58AAA and K13-67AAA, respectively, as measured by an ELISA-based DNA-binding assay. B. Wild-type K13 blocks doxycycline-induced K8.1 and ORF59 expression, while the NF-κB-defective mutants fail to do so. Cells were treated with doxycycline (10 ng/ml) and immunofluorescence analysis performed as described for Figure 1D.

Figure 4. Down-regulation of K13 induces lytic gene expression.

A–B. Cells were transfected with a control siRNA or a siRNA against K13 (Table S1) using oligofectamine (Invitrogen). Approximately 96 h post-transfection, down-regulation of K13 (A), vCyclin (B) and LANA-1 (C), and induction of RTA/ORF50 (D) gene expression was demonstrated by qRT-PCR. Real-time PCR reactions were performed in triplicate and the data presented as fold change in target gene expression (Mean±S.E.). E. Indirect immunofluorescence analysis showing up-regulation of ORF59 expression in cells transfected with K13 siRNA. Cells were analyzed 96 h post-transfection.

Figure 5. K13 protects cells against lytic replication-induced cell death.

A. BCBL1-TREx-RTA-MSCV and K13 cells were treated with doxycycline (20 ng/ml) for 4 days to induce lytic replication. Cells were washed and incubated in doxycycline-free medium for 1 week. Cell viability was measured using MTS assay as described previously and plotted relative to untreated cells. The values (Mean±S.E.) shown are from a representative of three independent experiments performed in triplicate. B. Equivalent induction of RTA upon doxycycline treatment in the vector- and K13-expressing BCBL1-TREx-RTA cells. Cells were treated with the indicated doses of doxycycline for 72 h prior to immunoblotting. C–D. Induction of lytic replication with TPA in BCBL1-TREx-RTA-K13 cells leads to the emergence of cells (K13/TPA) with increased K13 expression (C) and NF-κB activity (D), as measured by immunoblotting and an ELISA-based NF-κB DNA-binding assay, respectively. BCBL1-TREx-RTA-K13 cells were left untreated or treated with TPA (20 ng/ml) for 4 days followed by recovery in drug-free medium for 4 weeks prior to analyses. E. Induction of lytic replication in BCBL1-TREx-RTA-MSCV cells with TPA leads to emergence of cells with increase in endogenous K13 expression as measured by qRT-PCR. Treatment with TPA, followed by growth in drug-free media, was carried out essentially as in 4D. Real-time PCR reactions were performed in triplicate and the data presented as fold change in target gene expression (Mean±S.E.).

Figure 6. K13 differentially modulates the expression of KSHV genes following TPA treatment.

A, B. 4OHT pretreatment effectively blocks TPA-induced RTA up-regulation in BCBL1-TREx-RTA-K13- K13-ER^TAM cells as measured by immunoblotting with an RTA polyclonal antibody (A) and RT-PCR analysis (B), respectively. G3PDH serves as a normalization control. C. Differential effect of K13 on the expression of TPA-target lytic genes and LANA-1. BCBL1-TREx-RTA-K13-ER^TAM cells were treated with TPA for 96 h with and without prior treatment with 4OHT and expression of the indicated genes measured by real-time RT-PCR analysis and normalized relative to GNB2L1 (housekeeping control). Real-time PCR reactions were performed in triplicate and the data presented as fold change in target gene expression (Mean±S.E.).

Figure 7. Effect of K13 on lytic replication-induced vIL6 and hIL-6.

A. K13 activity fails to block TPA- and doxycycline-induced vIL6 induction. BCBL1-TREx-RTA-MSCV and K13-ER^TAM cells were treated with TPA (20 ng/ml) and doxycycline (10 ng/ml) for 96 h with and without prior treatment with 4OHT, followed by immunostaining with a vIL6 antibody. Nuclei were counterstained with Hoechst 33342. A representative of two independent experiments is shown. B. BCBL1-TREx-RTA cells expressing an empty vector, wild-type K13 or K13 mutants defective in NF-κB activation were treated with doxycycline for 96 h followed by immunofluorescence staining with an antibody against vIL6 as described in 7A. C. K13 fails to block RTA-induced vIL6 secretion. Culture of BAF-130 cells in the absence of WEHI-3B-conditioned medium (WH) led to a dramatic loss of cell viability (lane 1) which was rescued by the addition of supernatants (S.N.) from doxycycline-treated BCBL1-TREx-RTA-MSCV and K13-ER^TAM cells (lanes 5 and 9). Induction of K13 activity via 4OHT pretreatment had no inhibitory effect on vIL6 production (lane 10). Addition of a rabbit polyclonal antibody to vIL6 (300 ng/ml) effectively reversed the effect of doxycycline-treated supernatants on BAF-130 survival whereas a control antibody was without effect, thereby confirming the contribution of vIL6 to the observed effects. The values (Mean±S.E.) shown are from a representative of at least two independent experiments performed in triplicate. UT, untreated. D. K13 promotes RTA-induced hIL6 secretion. Secretion of hIL6 was measured using an ELISA kit (BD Biosciences, San Diego) in cell supernatants (10 µl) collected from BCBL1-TREx-RTA-MSCV and K13-ER^TAM cells that had been treated with doxycycline in the absence and presence of 4OHT, as described in Figure 7A. The values (Mean±S.E.) shown are from a representative of two independent experiments performed in triplicate.

Figure 8. A speculative model of Lytic Replication-Induced Clonal Selection (LyRICS) in KSHV tumorigenesis.

Infection with KSHV leads to a population of latently-infected cells with varying levels of K13 expression. Cells with low K13 expression are eliminated by apoptosis and lytic replication, while those with intermediate to high expression are stimulated to proliferate through the stimulatory effect of the NF-κB pathway on cell cycle and through the secretion of growth-promoting cytokines (e.g. hIL6). Elevated K13 expression also blocks the production of key viral proteins needed for lytic replication, thereby protecting cells from lytic replication-induced cell death. However, K13 has a permissive effect on RTA-induced vIL6 production and cooperates with it to stimulate hIL6 secretion further, thereby dysregulating the lytic gene expression program. The dysregulated expression of viral and cellular genes leads to further deregulation of signaling pathways controlling cellular survival, proliferation, immune response, and angiogenesis, initially leading to polyclonal expansion and, subsequently, through the accumulation of additional genetic and epigenetic alterations, to monoclonal cellular proliferation. Increased K13 expression, either alone or in combination with other viral (e.g. vCyclin, LANA-1, vGPCR and K1) and cellular proteins (e.g. proinflammatory cytokines) may lead to dysregulated expression of additional viral and cellular proteins in a cell-type and context-dependent manner, which may contribute to the pathogenesis of different cancers associated with KSHV infection. It is conceivable that acquisition of secondary genetic and epigenetic changes in the later stages of the disease may reduce or obviate the need for continuous K13 expression and NF-κB activity in some cases (not depicted).