Previous Article | Next Article ![]()
Journal of Virology, October 2005, p. 13180-13185, Vol. 79, No. 20
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.20.13180-13185.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
The Wistar Institute, Philadelphia, Pennsylvania 19104
Received 25 February 2005/ Accepted 14 July 2005
|
|
|---|
|
|
|---|
Reactivation of KSHV latency can be initiated by various environmental and chemical triggers, including hypoxia (13, 23), phorbol esters, and sodium butyrate (10, 14, 32, 33, 37, 46). Additionally, superinfection with other viruses, including cytomegalovirus (CMV) and human immunodeficiency virus, can induce lytic cycle gene expression from latent KSHV (31, 44, 45). All of these agents are thought to act, directly or indirectly, by stimulating transcription activation of the KSHV ORF50 immediate-early (IE) promoter. Ectopic expression of the ORF50 protein (RTA) induces lytic cycle gene expression and replication (20, 29, 42). RTA can autostimulate ORF50p (14) through a mechanism that may involve interaction with several cellular DNA binding factors (22, 27, 38). Nucleosome positioning and histone tail modifications can also influence ORF50p transcription activity (28). Several other activities that activate and repress ORF50 have been identified (22, 23, 38, 46, 47), and these may contribute to complex regulatory properties and the different activity levels in various cell types.
Immediate-early promoters from alpha- and betaherpesviruses can be activated by a virus-encoded virion factor (24). To determine whether ORF50p was subject to KSHV virion-mediated transcription activation, we purified KSHV virions from BCBL1 cells induced with sodium butyrate and tetradecanoyl phorbol acetate (TPA). For comparison, we also purified Epstein-Barr virus (EBV) virions from D98/HR1 cells and mock-treated virions from uninfected DG75 cells treated with sodium butyrate and TPA. To produce KSHV virions, BCBL1 cells (0.5 x 106/ml) were induced with 1 mM sodium butyrate and 50 ng/ml TPA for 72 h. The cell-free medium was collected by two rounds of centrifugation at 1,600 x g for 15 min and passaged through a 0.45-µm filter. Virions were pelleted from the filtrate by centrifugation at 100,000 x g for 1 h through 5 ml of a 5% sucrose cushion in a Beckman SW28 rotor. The virion pellets were resuspended in Dulbecco's modified Eagle's medium with 10% fetal bovine serum for infection or washed and resuspended in phosphate-buffered saline for Western blotting and colloidal blue staining (Invitrogen). Virions prepared from 109 cells were typically used for infection of 107 293 or BCBL1 cells, as indicated below.
The various virion preparations were analyzed by colloidal blue staining of sodium dodecyl sulfate-polyacrylamide gels (Fig. 1A) and by Western blotting with antibody specific for the tegument proteins ORF45 (Fig. 1B) and ORF50 (Fig. 1C). As expected, ORF45 was detected in KSHV virions only, and ORF50 was detected only in BCBL1 cells and not virions (Fig. 1C). We next assayed the ability of KSHV virions to stimulate ORF50p-Luc in transfected 293 cells (Fig. 1D). KSHV has been shown to infect and initiate early lytic gene expression in 293 cells (4, 17, 49). We found that KSHV virions specifically stimulated ORF50p-Luc by
50-fold but did not activate the control pGL3 luciferase vector. In contrast, EBV or mock-treated virions did not activate ORF50p-Luc or pGL3. We also found that ORF50p (290 to +1) was stimulated
20-fold by KSHV virions, suggesting that most of the responsive elements were within the first 290 nucleotides of the ORF50 ATG.
![]() View larger version (28K): [in a new window] |
FIG. 1. KSHV virions stimulate ORF50p-Luc transcription in 293 cells. (A) The protein profile of mock-treated, EBV, and KSHV virions were analyzed by colloidal blue staining of sodium dodecyl sulfate-polyacrylamide gels. Lane M, molecular weight markers. (B) Western blot analysis of virions shown in panel A detected with anti-KSHV ORF45-specific antibodies. (C) ORF50-specific Western blot of virions (left panel) or BCBL1 cells induced with sodium butyrate and phorbol ester (NaB/TPA) (right panel). (D) Luciferase assay of 293 cells transfected with pGL3, pORF50p-Luc, or pORF50(290/+1)-Luc, followed by no treatments or incubation with either mock-treated, EBV, or KSHV virions, as indicated.
|
10-fold, relative to
18.5-fold for untreated virions. This slight decrease in transcription activation most likely reflects the effects of virus-encoded proteins, like RTA, that can weakly autoactivate ORF50p (14, 38). UV-irradiated or untreated virions were then assayed for their ability to stimulate endogenous ORF50p in latently infected BCBL1 cells (Fig. 2C). Mock-treated, KSHV, and UV-inactivated KSHV virions were incubated with BCBL1 cells and assayed by RT-PCR for their ability to alter ORF50 or LANA mRNA levels. We found that KSHV and UV-irradiated virions stimulated ORF50 mRNA significantly above background levels (Fig. 2C). In contrast, LANA mRNA levels did not increase significantly when BCBL1 cells were incubated with KSHV and UV-irradiated KSHV virions. None of these treatments had significant effects on cellular GAPDH (glyceraldehyde-3-phosphate dehydrogenase) control levels, and all amplified products were dependent upon the addition of reverse transcriptase, indicating that RNA, and not DNA, was being amplified. Essentially identical results were observed when cRNA levels were quantitated by real-time PCR (Fig. 2D). To determine if protein synthesis was required for virion activation of ORF50 transcription, cycloheximide was added to BCBL1 cells 4 h prior to incubation with mock-treated, KSHV, or UV-inactivated KSHV virions (Fig. 2E and F). We found that ORF50 mRNA was weakly (
1.5-fold) stimulated by KSHV and UV-inactivated KSHV in the presence of cycloheximide (Fig. 2E). This weak activation may reflect a requirement for some protein synthesis or may reflect the cytotoxicity of cycloheximide treatment on BCBL1 cells. The complex effects of cycloheximide were further revealed by the reduction in mRNA levels for K12 and PAN, two viral genes responsive to RTA transcription activation (11). To eliminate the complex effects of cycloheximide on cellular ß-actin, we compared ORF50 and K12 mRNA levels relative to KSHV-encoded PAN mRNA levels (Fig. 2F). Using this analysis, we found that UV-inactivated KSHV stimulated ORF50 mRNA approximately twofold but had no effect on K12 mRNA (Fig. 2F). Furthermore, cycloheximide treatment did not eliminate the ability of KSHV to stimulate ORF50p-luciferase mRNA in 293 cells (Fig. 2G). These findings suggest that a virion component specifically activates ORF50 relative to other Rta-responsive genes, like K12 and PAN.
![]() View larger version (40K): [in a new window] |
FIG. 2. Stimulation of endogenous ORF50 by KSHV virions is independent of protein synthesis. (A) mRNA isolated from 293 cells infected with mock-treated, KSHV, or UV-irradiated KSHV virions was analyzed by RT-PCR with RT (left) or without RT (right). RT-PCR products were analyzed with primers specific for KSHV ORF50, KSHV LANA, or cellular GAPDH. (B) 293 cells were transfected with pGL3 or OR50p-Luc and assayed for luciferase activity. Transfected cells were infected with mock-treated or KSHV virions or with KSHV virions inactivated by UV treatment. (C) BCBL1 cells were incubated with mock-treated, KSHV, or UV-inactivated KSHV virions and then analyzed by RT-PCR with primers specific for ORF50, LANA, and GAPDH mRNA as indicated. Reactions without reverse transcriptase are shown at the right. (D) Same as in panel C, except that mRNA was quantitated by real-time PCR and shown graphically as the mRNA abundance relative to that of ß-actin. (E and F) BCBL1 cells treated with 20 µg/ml cycloheximide pre- and postinfection were incubated with mock-treated, KSHV, or UV-inactivated KSHV and assayed for mRNA levels of ORF50, PAN, and K12 relative to ß-actin mRNA (E) or relative to KSHV PAN mRNA (F) by quantitative RT-PCR. (G) 293 cells transfected with ORF50-Luc were infected with mock-treated virions or KSHV in the presence of 20 µg/ml cycloheximide and assayed for luciferase mRNA relative to cellular ß-actin by quantitative RT-PCR. RLU, relative light units.
|
![]() View larger version (23K): [in a new window] |
FIG. 3. Promoter sequence requirements for KSHV virion activation. (A) 293 cells were transfected with luciferase constructs containing EBV-Zp, K8-IE, K8-DE, ORF50p, or a pGL3 control and assayed for luciferase activity after incubation with mock-treated or KSHV virions. (B) Deletion mutants of ORF50p containing sequences (1038 to +1), (290 to +1), (157 to +1), (100 to +1), or pGL3 were compared in luciferase activity assays in 293 cells incubated with mock-treated or KSHV virions. C) Substitution mutations in ORF50p (157 to +1) were generated at positions 148, 135, 128, 120, 114, 98, 93, and 84 and were assayed for luciferase activity in 293 cells after incubation with mock-treated or KSHV virions. RLU, relative light units.
|
Substitution mutagenesis of ORF50p revealed that several regions contribute to virion-mediated transcription activation. In particular, we found that mutagenesis of sequences at positions 84 and 98 reduced the virion response to background levels. Substitutions at 148, 128, 114, and 93 reduced virion activation almost twofold, suggesting that multiple regions of the ORF50 minimal promoter constitute the KSHV virion response element (Fig. 3C).
The above results suggest that KSHV virions may alter some general activity that associates with the ORF50 immediate-early gene promoter. To examine this possibility, nuclear extracts from mock- and KSHV virion-infected 293 cells were assayed for ORF50 DNA binding activity in electrophoretic mobility shift assays (EMSA) (Fig. 4A). Mock-infected 293 cells contained at least three major ORF50p DNA binding activities (Fig. 4A, lane 2). The fastest-migrating activity was most likely not specific for ORF50p since a species with a similar mobility was detected in ORF73 DNA binding (Fig. 4A, lane 5). Interestingly, this nonspecific binding activity disappeared in KSHV-infected cell nuclear extracts (Fig. 4A, lanes 3 and 6). KSHV infection did not alter the slower-migrating and more-specific ORF50p DNA binding activities (Fig. 4A, lane 3). These findings suggest that KSHV infection eliminates a nonspecific DNA binding activity that strongly associates with ORF50p and perhaps other viral promoters.
![]() View larger version (43K): [in a new window] |
FIG. 4. Changes in protein composition at ORF50p in vitro and in vivo. (A) ORF50p (157 to +1) or ORF73 DNA probes were incubated with nuclear extracts derived from mock-treated or KSHV-infected 293 cells and assayed by EMSA. (B) 293 cells transfected with ORF50p-Luc were assayed by chromatin immunoprecipitation with antibody to acetyl histone H3 (AcH3), AcH4, or control IgG. Chromatin immunoprecipitation DNA was amplified with primers specific for ORF50p or GAPDH in mock ()- or KSHV (+)-treated cells. Ten percent of the total input DNA is indicated (Input).
|
We have shown here that KSHV virions induced transcription activation of ORF50p in transfected 293 cells (Fig. 1 and 2) and from endogenous viral genomes in latently infected PEL cells (Fig. 2C and D). The minimal ORF50p nucleotide sequence required for KSHV virion activation was mapped to a region within 157 to + 1 of the initiation ATG, and several substitution mutations within this region eliminated virion activation (Fig. 3). KSHV virions induced changes in the nucleoprotein complex on ORF50p DNA in EMSA and stimulated histone H3 and H4 acetylation on transfected ORF50p plasmid DNA (Fig. 4). Together, these results suggest that KSHV virions possess an activity that alters cellular protein interactions with ORF50p. EMSA did not detect a novel protein association with ORF50p, suggesting that a KSHV-encoded VP16-like activity does not assemble at ORF50p. Alternatively, KSHV is known to bind to cell surface integrins and initiate phosphoinositide 3-kinase-mediated signal transduction (3, 35). Initiation of integrin signaling by KSHV stimulates phosphoinositide 3-kinase, and protein kinase C pathways may have indirect effects on stress kinase pathways and indirectly stimulate ORF50p transcription (35).
Other herpesviruses encode virion proteins that can affect viral gene expression and influence lytic cycle progression. The herpes simplex virus (HSV) virion protein VP16 can assemble with cellular factors on the HSV immediate-early promoter TAATGARAT sites to activate lytic cycle gene expression (reviewed in reference 48). HSV virion protein 22 (VP22) can inhibit nucleosome assembly that blocks lytic cycle progression (43). The CMV UL82-encoded virion protein pp71 can activate CMV major immediate-early promoter-dependent transcription and is required for lytic replication (7, 30, 41). Although KSHV does not have an obvious homologue of these virion regulatory proteins, our data suggest that KSHV virions possess a trans-inducing function that is similar in principle to those characterized for HSV and CMV.
This work was supported by the NIH (CA085678) and by funds from the Pennsylvania Department of Health to P.M.L.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»