Distinct Biology of Kaposi’s Sarcoma-Associated Herpesvirus from Primary Lesions and Body Cavity Lymphomas

Abstract

The DNA sequence for Kaposi’s sarcoma-associated herpesvirus was originally detected in Kaposi’s sarcoma biopsy specimens. Since its discovery, it has been possible to detect virus in cell lines established from AIDS-associated body cavity-based B-cell lymphoma and to propagate virus from primary Kaposi’s sarcoma lesions in a human renal embryonic cell line, 293. In this study, we analyzed the infectivity of Kaposi’s sarcoma-associated herpesvirus produced from these two sources. Viral isolates from cultured cutaneous primary KS cells was transmitted to an Epstein-Barr virus-negative Burkitt’s B-lymphoma cell line, Louckes, and compared to virus induced from a body cavity-based B-cell lymphoma cell line. While propagation of body cavity-based B-cell lymphoma-derived virus was not observed in 293 cell cultures, infection with viral isolates obtained from primary Kaposi’s sarcoma lesions induced injury in 293 cells typical of herpesvirus infection and was associated with apoptotic cell death. Interestingly, transient overexpression of the Kaposi’s sarcoma-associated herpesvirus v-Bcl-2 homolog delayed the process of apoptosis and prolonged the survival of infected 293 cells. In contrast, the broad-spectrum caspase inhibitors Z-VAD-fmk and Z-DEVD-fmk failed to protect infected cell cultures, suggesting that Kaposi’s sarcoma-associated herpesvirus-induced apoptosis occurs through a Bcl-2-dependent pathway. Kaposi’s sarcoma-associated herpesvirus isolates from primary Kaposi’s sarcoma lesions and body cavity-based lymphomas therefore may differ and are likely to have distinct contributions to the pathophysiology of Kaposi’s sarcoma.

Kaposi’s sarcoma (KS), a recurrent multifocal angioproliferative disorder, is the most common neoplasm in patients with AIDS (4, 31). The association of unique viral DNA sequences in KS lesions was first shown by representational difference analysis (10). These sequences showed similarity to several oncogenic herpesviruses, most notably Epstein-Barr virus (EBV) and herpesvirus saimiri, and the virus was proposed as a novel member of the gammaherpesvirus family, termed KS-associated herpesvirus (KSHV) or human herpesvirus 8 (HHV-8). More than 95% of tumor tissues obtained from patients with AIDS harbor KSHV DNA sequences. Since then, several investigators have reported the presence of these sequences in virtually all epidemiologic forms of KS and the less common AIDS-associated B-cell lymphoproliferative disorders, such as body cavity-based B lymphomas (BCBL) and multicentric Castleman’s disease (8, 33). In addition, KSHV DNA sequences have been frequently detected in circulating B cells and peripheral blood mononuclear cells of AIDS-KS patients (18, 37). These cumulative observations have stimulated extensive investigations into the properties of this putative new herpesvirus and its relation to KS pathogenesis.

KS lesions have the features of both hyperplastic proliferation and neoplastic growth. They often arise synchronously over widely dispersed areas in immunosuppressed individuals and have been proposed to originate from the clonal outgrowth of a circulating progenitor cell (26). Interestingly, seroepidemiological studies indicate a linear increase in KSHV seropositivity months and even years prior to the onset of KS (17, 20). Although highly sensitive in situ techniques or electron microscopic analyses indicate the presence of KSHV in fresh KS biopsy specimens, viral replication is seldom prominent (7, 34). Viral DNA sequences appear to be contained in the nuclei of cells within large circular episomal structures, a genomic form characteristic of latent herpesvirus (13); however, KS spindle-shaped cells, the hallmark of KS lesions, tend to lose these sequences after in vitro passage. Though viral production is limited, KS biopsy specimens contain KSHV, and progeny virions can be propagated serially in the human renal epithelial cell line 293 (15). Thus, it is not known whether KSHV DNA sequences found in tumor tissues represent mature virions in a small fraction of infected cells or cryptic latent infections.

On the other hand, the availability of B-lymphoid cell lines established from AIDS-associated BCBL has provided fundamental insights into the molecular biology of KSHV. These cell lines were shown to be latently infected with KSHV and, in many instances, with EBV strains that usually infect and transform B lymphocytes (3, 9, 29). Episomal KSHV DNA copy numbers appear to be 50 to 100 times greater in BCBL-derived cell lines than in KS biopsy specimens. In addition, upon induction with phorbol esters (e.g., 12-O-tetradecanoylphorbol-13-acetate [TPA]), EBV-negative BCBL-derived cell lines can substantially increase virus production (29). Most of the KSHV genome has now been sequenced from BCBL cell lines and KS specimens (25, 30), and interestingly, recent reports have suggested that different strains of KSHV may exist, based on sequence divergence between the two viral DNA sources (40). In this report, we analyze the transmission of KSHV in vitro from primary KS specimens to the EBV-negative Burkitt’s B-lymphoma cell line Louckes and describe phenotypic differences in virus from KS lesions grown in B-cell lines compared to those derived from a chronically infected BCBL.

MATERIALS AND METHODS

Cell lines.

BCBL-1 is a B-cell line derived from an EBV-negative body cavity-based lymphoma that is latently infected with KSHV (NIH AIDS Research and Reference Reagent Program, Rockville, Md.). Louckes is an EBV-negative Burkitt’s B-lymphoma cell line. Namalwa is an EBV-positive Burkitt’s B-lymphoma cell line (kindly provided by Erle Robertson). All cell lines were grown in RPMI 1640 medium supplemented with 10% fetal calf serum (FCS), 200 mM l-glutamine, and antibiotics (penicillin [100 U/ml] and streptomycin [100 μg/ml]). The human embryonic kidney 293 cells were maintained as previously described (15). Biopsy specimens from advanced KS skin lesions (tumor nodule) were obtained from human immunodeficiency virus (HIV)-infected patients (University of Michigan, Ann Arbor). These lesions were shown to be DNA PCR positive for KSHV and negative for other known herpesviruses (herpes simplex virus types 1 and 2, EBV, and cytomegalovirus) and HIV.

Cocultivation conditions.

To generate BKS-1 cell lines, Louckes cells were seeded at a density of 2 × 10⁵ cells per 0.1 ml of RPMI 1640 medium supplemented with 10% FCS, 200 mM l-glutamine, and antibiotics into 96-well microplates and incubated for 2 h at 37°C. Suspension cells were covered with eight-well Anopore tissue culture inserts of 0.2-μm pore size (Nunc Inc.), and early-passage primary KS cells found to be KSHV positive were seeded at a density of 2 × 10⁴ cells per well. Four days following cocultivation, Louckes cells were transferred to 24-well plates without KS cells and then propagated at low density (0.5 × 10⁶ cells/ml) into 25-cm² flasks.

Chemical induction.

Viable cells, seeded at a density of 10⁶ cells per ml of fresh RPMI 1640 medium supplemented with 10% FCS, 200 mM l-glutamine, and antibiotics, were cultured in presence or absence of inducing agent for 12 to 96 h. TPA (Sigma) was added to a final concentration of 20 ng/ml.

DNA extractions and sources of KSHV.

To prepare total genomic DNA, cells were harvested, washed once in phosphate-buffered saline (PBS), resuspended in 5 mM Tris-HCl (pH 7.4) containing 0.5% sodium dodecyl sulfate (SDS), 2 mM EDTA, and 0.5 mg of proteinase K per ml, and incubated for 2 h at 56°C. DNA was extracted with phenol-chloroform, precipitated with ethanol, dried, and resuspended in TE (10 mM Tris-HCL [pH 8], 1 mM EDTA). Nuclear fractions were prepared by resuspending the cells in 20 mM Tris (pH 7.9)–3 mM MgCl₂–2 mM CaCl₂. Samples were incubated for 20 min on ice, and Nonidet P-40 (NP-40) was added to a final concentration of 0.5%. Nuclei were then separated from the cytoplasmic fraction by centrifugation at 1,500 × g for 15 min. Pellets were incubated for 24 h at 37°C in presence of proteinase K, and DNA was subsequently extracted with phenol-chloroform as described above.

To release cell-associated viral particles, cell-free lysates were prepared by three cycles of freezing and thawing. To isolate extracellular viral particles, culture media from infected cells (10⁶ cells/ml) were subjected to centrifugation at 3,000 rpm for 10 min and filtered through a 0.45-μm-pore-size membrane. Cleared supernatants were further centrifuged for 2 h at 25,000 × g at 4°C (Beckman SW28 rotor), and pelleted viral particles were resuspended in Dulbecco’s modified Eagle’s medium (DMEM) and snap-frozen. To confirm the presence of encapsidated KSHV DNA in pelleted materials, nuclease resistance assays were performed as previously described (15, 29).

PCR amplification and Southern blot analyses.

Genomic DNA was analyzed by PCR for KSHV sequences as previously described (10). Total RNA was isolated from 5 × 10⁶ to 10 × 10⁶ infected cells by using TRIzol (GIBCO BRL), and contaminating genomic DNA was eliminated with RNase-free DNase I followed by heat-inactivation of the enzyme. Reverse transcriptase (RT)-mediated PCR (RT-PCR) was performed with 1 μg of RNA by using the Superscript RT system (GIBCO BRL). Control reactions were performed with omission of RT. PCR conditions used for the amplification and/or hybridization of both KSHV DNA and RNA sequences were as previously described (15). PCR primers and probes for the KSHV open reading frames (ORFs) were as follow: ORF K7 (T1.1), 5′-GCTTGAGTCAGTTTA GCACTGGGAC-3′ (sense) and 5′-GGAGATTGAATCCAATGCAATAACC-3′ (antisense); ORF16 (v-Bcl-2), 5′-GCTCAGCCCTATTAAGCTATACATCAC-3′ (sense) and 5′-TCATACGCATATACAGGTAAAACGG-3′ (antisense); ORF21 (thymidine kinase [TK]), 5′-CGCCTGTGACCGTGGACTACAGGAATGTTT-3′ (sense) and 5′-GCGGGTCTCCCGCCTTACCAGACTTCATCA-3′ (antisense); ORF22 (glycoprotein H [gH]), 5′-CCGCCGAAGTCGCCGAGGACCTCAGGGTAA-3′ (sense) and 5′-AACAACGAGTCCGGCGTAGCGCTCTATGGA-3′ (antisense); ORF25 (major capsid protein [MCP]), 5′-AAGTCATCCAGACAACCCATAATCAAG-3′ (sense) and 5′-TTCTCCAGGTGCAGTAGAATATCATCC-3′ (antisense); and ORF26 (minor capsid protein [mCP]), 5′-AGCCGAAAGGATTCCACCATTGTGCTC-3′ (sense) and 5′-TCCGTGTTGTCTACGTCCAGACGATAT-3′ (antisense). PCR primers for the EBNA-2 gene of EBV type 1 or 2 were as previously described (2). A primer set specific for the human β-actin cDNA was used as a quantitative control. Southern blot analysis of genomic DNA was performed with a ³²P-labeled 233-bp KSHV-specific probe derived from the MCP gene as previously described (15).

ISH analysis.

Cells were washed twice with PBS and resuspended at a density of 10⁶ cells/100 μl of PBS. Cell suspensions were cytospun onto slides and fixed in 4% paraformaldehyde (pH 7.4) for 20 min at room temperature. In situ hybridization (ISH) was performed as previously described (16). A 2.9-kb (encompassing bp 36,688 to 39,594) and a 9.5-kb (encompassing bp 37,676 to 47,183) PCR fragment cloned from the BCBL-1 cell line were labeled with digoxigenin-dUTP by nick translation to generate an HHV-8-specific probe. The hybridized probes were visualized by staining with alkaline phosphatase-conjugated antidigoxigenin antibodies (Boehringer Mannheim) according to the manufacturer’s instructions.

In situ transmission electron microscopic (TEM) analysis.

BKS-1 and BCBL-1 cells, untreated or treated with TPA, were harvested by centrifugation at 200 × g for 5 min. Cell pellets were resuspended in 2.5% glutaraldehyde in PBS (pH 7.4) and incubated for 5 min at room temperature. The cells were then centrifuged at 200 × g for 5 min and incubated for 2 h at 4°C. The samples were rinsed, dehydrated, and embedded in Epon. The specimens were sectioned, stained with uranyl acetate and lead citrate, and examined with a Zeiss 109 transmission electron microscope.

Plasmids and transient transfection assay.

Full-length KSHV v-Bcl-2 cDNAs from BCBL-1 and BKS-1 cells were amplified by PCR and subcloned into the mammalian expression vector pCR3.1 (Invitrogen) according to the manufacturer’s instructions. Plasmids containing v-bcl-2 in the reverse orientation relative to the cytomegalovirus promoter-enhancer in pCR3.1 were also obtained. In vitro translation of the different constructs (1 μg) was carried out with the TNT T7 quick-coupled transcription/translation system (Promega) in the presence of [³⁵S]methionine, and the reaction mixtures were subjected to SDS-polyacrylamide gel electrophoresis. For transient expression of v-Bcl-2, 293 cells seeded at a density of 0.5 × 10⁶ cells per well into six-well Costar plates were transfected by the calcium phosphate method.

Viral infection.

293 cells were seeded at a density of 10⁶ cells per well into six-well Costar plates and cultured in DMEM supplemented with 10% FCS, 200 mM l-glutamine, and antibiotics for 24 h. Culture medium was replaced with 0.8 ml of fresh DMEM, and cells were inoculated with 0.2 ml of DNase-treated pelleted viral particles harvested from supernatants of uninduced or TPA-induced BKS-1 or BCBL-1 cells. Following adsorption for 1 h, cells were washed once with PBS, and DMEM supplemented with 10% FCS, 200 mM l-glutamine, and antibiotics was added. Cells were incubated at 37°C in the presence or absence of 100 μM Z-VAD-fmk or Z-DEVD-fmk (Enzyme System Products, Dublin, Calif.). Cell viability and growth in the cultures were monitored daily by the trypan blue exclusion method. Live and dead cells were scored in duplicate by microscopic examination. To delineate apoptotic and necrotic cell populations in the infected cultures, aliquots of cells were stained with fluorescein isothiocyanate (FITC)-conjugated annexin-V (Boehringer Mannheim) and propidium iodide (PI), and preparations were analyzed by flow cytometry according to the manufacturer’s instructions.

RESULTS

Transmission of KSHV DNA sequence in Louckes cells.

We first analyzed the transmission of KSHV in vitro from primary KS lesions to an EBV-negative Burkitt’s B-lymphoma cell line, Louckes, which has proven invaluable in studies of the EBV life cycle (36, 39). Primary cell lines established from two AIDS-associated cutaneous KS biopsy specimens were cocultured with Louckes cells separated by Anopore inserts of 0.2-μm pore size. Four days after cocultivation, the majority of cocultured Louckes cells showed altered morphology and reduced cell growth compared to uninfected cells, and the KS spindle cells, deprived of scatter factor which is required for their growth (15), were no longer detectable in the cell culture.

PCR analysis of genomic DNA from cocultured Louckes cells revealed the presence of several KSHV-encoded genes but no EBV DNA (Fig. 1A and B). The intensity of the KSHV DNA signal in these cells was typically less than that of a single-copy cellular gene, encoding β-actin (Fig. 1B, lower panel). The KSHV DNA signal was much higher in the chronically infected lymphoma BCBL-1 (Fig. 1A, upper panel). Indeed, quantitative PCR analysis using the TaqMan PCR detection system (Applied Biosystems) indicated ∼80 episomal DNA copies per cell in BCBL-1, compared to <1 copy per cell in Louckes cells infected with KSHV (Fig. 1C), herein termed BKS-1 cells. In these analyses, an exponential increase in viral DNA copy number was noted in BKS-1 cells, which appeared more prominent upon stimulation with TPA (Fig. 1C, day 11). It is important to note that the multiplicity of infection in Louckes cells was estimated to be ≤0.01, which partly explains the lower initial copy number per cell and subsequent increase with time after exposure to virus (Fig. 1C). These observations were consistent with previous reports showing detection of higher genomic copy numbers in BCBL cell lines than in primary KS lesions (29).

FIG. 1.

Transmission of KSHV DNA sequences from primary KS lesions into Louckes cells. PCR analysis was performed with 1 μg of total genomic DNA extracted from BCBL-1 and Louckes cells following cocultivation with primary KS lesions (BKS-1) as described in the text. PCR products were separated by electrophoresis on an ethidium bromide-stained agarose gel (2%). Numerous functional and structural KSHV genes were amplified from both cell lines (A), whereas EBV DNA was undetectable (B). EBV standard was generated from DNA isolated from Namalwa cells which contain two integrated EBV genomes per cell, and cellular β-actin was used as a control for PCRs. (C) Quantitative PCR analysis of KSHV DNA sequences was performed with the TaqMan PCR detection system (Applied Biosystems) according to the manufacturer’s instructions. Total genomic DNA was extracted from BKS-1 cells (10⁶) at the indicated times following cocultivation, and PCR analyses were performed in triplicate, using a custom fluorescent probe and primers for ORF26 (mCP) of KSHV. At day 9, cells were cultured in fresh medium containing or lacking TPA (20 ng/ml) for 48 h prior to DNA isolation. (D) Sequence variability of ORF K1 among KSHV DNA samples from different sources. In this summary of the results of PCR sequencing analysis of two segments of the ORF K1 obtained from a cutaneous KS lesion and BCBL-1, only nucleotides present at variable positions are given, with positions that differ from the prototype KSHV sequence (GeneBank accession no. 1102887) enclosed in black boxes. Primers used and sizes of the PCR products are as follows: 5′-GTACAATCAAGATGTTCCTGTATG-3′ (sense) and 5′-ACAAGTGACTGTGTTTGATGG-3′ (antisense), giving a 337-bp product; and 5′-CCGTGTCACAAACTAAATACT-3′ (sense) and 5′-TATCTTACCTGAATGTCAGTACCA-3′ (antisense), giving a 478-bp product.

Sequence divergence at the extreme ends of the KSHV genome was recently reported for different KSHV isolates (21, 40). To determine whether there were genetic differences between lesion- and lymphoma-derived viruses, sequence analysis of ORF K1, located at the 5′ region of the viral genome, was performed. Several nucleotide changes were noted in DNA from primary KS lesions compared to the published BCBL-1-derived KSHV sequence (Fig. 1D). Nucleotide changes that resulted in amino acid substitutions included those at positions 137 (methionine to leucine), 141 (arginine to leucine), 182 (methionine to leucine), 683 (phenylalanine to serine), 741 (leucine to valine), 778 (threonine to isoleucine), 886 (proline to asparagine), and 893 (proline to threonine). There was a silent substitution at base 698 (T changed to C). These findings documented that there were genetic differences between KSHV isolates from primary KS lesions and B lymphomas.

KSHV DNA replication in BKS-1 cells.

To assess the levels of KSHV transmission in the cell cultures, several analyses were performed. ISH using digoxigenin-labeled DNA probes specific for KSHV sequence was performed 4 weeks following initial passage. For probe synthesis in these experiments, a 2.9-kb PCR fragment containing the lytic gene ORF22 of KSHV (gH) was derived from the BCBL-1 genome, and equal periods of staining with alkaline phosphate-conjugated antidigoxigenin were used for all slide preparations. In contrast to negative controls, KSHV-specific signals were identified in the nucleus of a small but consistent percentage (1 to 2%) of BKS-1 cells, with various amounts of cytoplasmic staining (Fig. 2A, panels i and iii). A similar pattern of staining was observed in latently infected BCBL-1 cells (Fig. 2A, panel ii). No signal was observed in uninfected Louckes cells (data not shown). Under identical conditions, we performed ISH analysis of BKS-1 cells treated for 48 h with medium containing or lacking TPA (20 ng/ml). The intensity of KSHV-specific signals increased in BKS-1 cells after this stimulation, presumably reflecting induction of the lytic/productive phase of KSHV (Fig. 2A; compare panels iii and iv). Approximately 20% of the TPA-induced cells were positive for KSHV. Moreover, significant cytotoxicity was noted in BKS-1 cell cultures by light microscopy over the ensuing 48 h of treatment with TPA compared to uninfected Louckes cells exposed to TPA or uninduced BKS-1 cells (data not shown). Finally, to document viral DNA replication in BKS-1 cells, we performed Southern blot analysis of genomic DNA isolated at various times following TPA treatment. Viral DNA was detectable as early as 12 h after TPA stimulation, and signals increased substantially by 48 h (Fig. 2B, lanes 2 to 4). These findings suggested transmission and replication of KSHV from KS lesions to Louckes cells, although the virus did not grow to high titer in this cell type.

FIG. 2.

Viral DNA synthesis in BKS-1 cells. (A) Transmission of the KSHV genome demonstrated in infected Louckes cells by ISH using specific KSHV probes. In each experiment, viable cells (10⁶ cell/ml) were washed and cultured in fresh medium containing or lacking TPA (20 ng/ml) for up to 48 h prior to the preparation of slides as described in Materials and Methods. (i) No signals were detected in BKS-1 cells after hybridization with an EBV-specific probe. (ii) In contrast, nuclei with various amounts of cytoplasmic staining were identified in the majority of latently infected BCBL-1 cells with a 2.9-kb KSHV-specific probe described in Materials and Methods. (iii) Under identical conditions of hybridization with the KSHV-specific probe, a similar pattern of staining was noted in BKS-1 cells. (iv) The percentage of KSHV-positive BKS-1 cells (∼20%) and intensity of signals increased following treatment with TPA. (B) Southern blot analysis of genomic DNA (10 μg) isolated from uninduced BKS-1 cells or at various times after TPA induction. Viral DNA sequences were detected with a ³²P-labeled 233-bp KSHV DNA probe after digestion with BamHI. The arrow indicates the 330-bp product expected after digestion with BamHI.

Induction of lytic replication in BKS-1 cells.

To examine the pattern of KSHV gene expression in BKS-1 cells, we analyzed viral transcripts by RT-PCR. Total RNA was extracted from uninduced or TPA-induced BKS-1 cells 24 h after treatment, and equal amounts of DNase-treated RNA were subjected to RT-PCR. Both TK and gH KSHV mRNAs were detected in infected, but not uninfected, Louckes cells (Fig. 3A). These transcripts were more abundant after TPA induction, even though untreated and treated cells showed roughly comparable amounts of cellular β-actin RNA control (Fig. 3B). No PCR signal was detected without the addition of RT under these conditions (data not shown), ruling out potential contamination in the reactions.

FIG. 3.

Productive lytic infection in BKS-1 cells. (A) Expression of KSHV lytic genes encoding TK and gH was examined by RT-PCR in uninduced or TPA-induced BKS-1 cells. Gene expression was detected in RNA obtained from uninduced cells (lane 2); levels of expression in BKS-1 cells increased following TPA induction (lane 3); no expression of KSHV transcripts was observed in uninfected Louckes cells (lane 1). (B) RT-PCR for the β-actin gene used as a quantitative control revealed comparable amounts of cDNA amplified from each sample (lanes 2, 4, and 6). No PCR signal was detected in the samples without the addition of RT (lanes 1, 3, and 5). (C) Nuclease sensitivity assay for extracellular enveloped particles derived from BKS-1 cell cultures. Culture fluids from equivalent amounts of infected cells were concentrated and incubated with (+) or without (−) NP-40. Each preparation was then exposed to pronase and digested with DNase before nucleic acid extraction. The isolated DNA underwent PCR amplification, and specific hybridization of PCR products to a ³²P-labeled 233-bp KSHV DNA probe was performed as described in Materials and Methods. Nuclease assay of the BKS-1 cell culture 1 week (left) and 4 weeks (right) after initial passage. Cultures fluids of parallel BCBL-1 and BKS-1 cell cultures were analyzed 48 h after addition of fresh medium containing or lacking TPA (20 ng/ml).

To determine whether TPA treatment increased the release of mature KSHV particles into the media of BKS-1 cells, supernatants from unstimulated or TPA-induced BKS-1 cell cultures were ultracentrifuged and subjected to nuclease digestion (15, 29). These fractions were treated with pronase and DNase I in the presence or absence of the nonionic detergent NP-40, and virus-associated DNA was analyzed by PCR and Southern blotting with a radiolabeled KSHV probe. If the particles were enveloped, the lipid bilayer would protect viral DNA sequences from the combined action of exogenous protease and nuclease. Indeed, under these conditions, digestion of encapsidated KSHV DNA extracted from the concentrated BKS-1 supernatant fraction was observed only in the presence of NP-40, confirming the presence of enveloped mature virions in the preparation (Fig. 3C, left). To investigate TPA induction of KSHV replication in BKS-1 cells, similar experiments were performed. For comparison, supernatants of parallel BCBL-1 and BKS-1 cell cultures (equivalent of 5 × 10⁶ cells) were harvested 48 h after addition of fresh medium containing or lacking TPA. PCR and Southern blotting analyses of the KSHV genome revealed that viral DNA replication occurred in both cultures, and there was a marked increase in the release of mature particles after TPA treatment (Fig. 3C, right).

Finally, the presence of virus particles in TPA-induced BKS-1 cell cultures was examined by TEM after serial passages. Typically, condensation and margination of host chromatin were readily apparent in the majority of BKS-1 cells (Fig. 4). These cells also showed irregular nuclear membranes and edematous cytoplasm. Although such changes are consistent with the cell injury associated with viral infection, only 1% of the cell nuclei were found to contain herpesvirus-like capsids, suggesting possible bystander killing or cytotoxicity induced by expression of early viral genes. These nucleocapsids contained electron-dense central cores surrounded by a thick tegument (∼100-nm diameter). By contrast, BCBL-1 cells contained numerous empty capsids in addition to electron-dense nucleocapsids previously described (29). We estimate that ∼5% of BCBL-1 which contained viral particles underwent necrosis. The nuclei of BCBL-1 cells were better preserved than those of BKS-1 cells, and the cells were nearly completely viable. Mature enveloped virions (140 to 160 nm in diameter) were detected in both cell lines within cytoplasmic cisternae, in vesicles, and extracellularly. Although the electron microscopic observations alone do not establish replication competence of virus, these data, in combination with the viral DNA replication and transcriptional analyses, indicate that Louckes cells are susceptible to KSHV infection with viral isolates from primary KS lesions.

FIG. 4.

Visualization of KSHV in BKS-1 by TEM. (A and B) High (×82,400) magnification of a typical nucleocapsid with an electron-dense core seen in the nuclei of BKS-1 (A) and BCBL-1 (B) cells. (C) An enveloped particle morphologically mature in the extracellular space of a BKS-1 cell (magnification, ×25,900). (D) General view of a TPA-treated BCBL-1 cell. Numerous empty capsids or capsids with an electron-dense core (dark arrows) are present within the nucleus (Nu) of the cell. Electron-opaque bodies showing sites of assembly of capsids within the nucleus (*) and a cytoplasmic (Cyt) vesicle containing an enveloped mature particle (arrow) can also be seen (magnification, ×25,900).

KSHV from lesions, but not lymphomas, induce apoptosis in 293 cells.

We previously showed that KSHV could be isolated from cutaneous KS lesions and serially propagated in the human embryonic kidney cell line 293 (15). In this cell system, KS-derived viruses had the propensity to induce cytotoxicity in infected 293 cell cultures. To determine whether this cytotoxicity was associated with apoptosis and whether a similar effect was induced by BCBL-1-derived virus, 293 cells were infected with DNase-treated, pelleted viruses from supernatants of TPA-induced BCBL-1 or BKS-1 cells. To detect the KSHV genome in cell culture, DNA from nuclear fractions was isolated from uninfected or infected 293 cells at different times after infection and analyzed by PCR. Although 293 cells were susceptible to infection with both isolates (Fig. 5A), lytic growth appeared more prominent with BKS-1-derived virus. Viral DNA synthesis in infected cells peaked at day 2 in both infected cell cultures (Fig. 5A, lanes 3 and 8); however, by day 4 viral DNA sequences were lost in cultures inoculated with virus derived from BCBL-1, suggesting an abortive infection from the lymphoma-derived virus (Fig. 5A, lanes 11 to 12). Pretreatment of BKS-1-derived virus by heat inactivation (65°C for 10 min) or UV irradiation of pelleted viruses prior to adsorption completely eliminated the detection of viral DNA sequences in 293 cells and its cytotoxic effects (data not shown).

FIG. 5.

KSHV lytic infection induces apoptosis in 293 cells. (A) Replication of viral DNA in 293 cells infected with viruses derived from TPA-treated BKS-1 or BCBL-1 cells. Genomic DNA from infected cells was isolated at the indicated time postinfection and amplified by PCR using primers for KSHV mCP. PCR products were subjected to electrophoresis on 2% agarose gels and transfer to nylon membranes. Hybridization to the same KSHV capsid gene probe revealed bands of 233 bp. (B) Cytotoxicity of KSHV in 293 cells. Cell viability in the infected cultures was monitored by staining with trypan blue exclusion. Data shown are the mean of three independent experiments ± standard deviation, and the percentage of cell viability was defined as the relative number of viable infected cells versus uninfected cells. (C) Detection of KSHV-induced apoptosis in 293 cells by flow cytometric analysis. At the indicated time postinfection, aliquots of cells (10⁶) infected with virus derived from TPA-treated BKS-1 or BCBL-1 (open curves) were collected and stained with FITC-conjugated annexin-V and PI. Uninfected cells served as control (solid curve). Representative histograms of several independent experiments are shown.

Concurrent with KSHV DNA synthesis, cell rounding and a loss of adherence became readily apparent in 293 cell cultures infected with BKS-1-derived virus. Cell death in these cultures substantially increased as early as day 2 after infection (Fig. 5B). By contrast, no cytotoxicity was noted in BCBL-1-infected 293 cell cultures, even though cell growth was slightly reduced. To determine whether apoptosis contributed to KSHV-induced cell death, aliquots of infected 293 cells were stained with FITC-conjugated annexin-V and PI and analyzed by flow cytometry. Annexin-V binds specifically to phosphatidylserine, an integral component of the inner plasma membrane of healthy cells that is exposed during the initial steps of apoptosis. This assay allows quantitation and delineation of viable, apoptotic, and necrotic cell populations. Flow cytometric analyses showed an increase in annexin-V immunostaining beginning 1 day after infection (Fig. 5C, upper panel). After 3 days, 82% of the cells were either undergoing apoptosis (annexin-V positive) or scored as dead cells (PI positive) when infected with BKS-1-derived virus, in contrast to the relatively low levels of cell death in cultures infected with BCBL-1-derived virus (Fig. 5C, lower panel, ∼17%).

Expression of KSHV v-Bcl-2 inhibits apoptosis in 293 cells.

The KSHV genome was recently shown to encode a functional v-Bcl-2 homolog (ORF16) expressed both in KS specimens and BCBL cell lines (11, 32). Therefore, we investigated the functional role of v-Bcl-2 in regulating apoptosis and prolonging cell survival after infection by KSHV. Full-length coding sequences of v-Bcl-2 from BCBL-1 and BKS-1 were amplified by PCR and subcloned into the mammalian expression vector PCR3.1. The proteins were expressed, and bands of ∼19.0 kDa were detected on an SDS-polyacrylamide gel (Fig. 6A). 293 cells were then transfected with increasing amounts of v-Bcl-2 construct and challenged with BKS-1-derived virus. Within 3 days after infection, v-Bcl-2 increased cell viability (∼50%) in the culture compared to a culture transfected with the negative control plasmid (∼4%). Interestingly, inclusion of Z-VAD-fmk or Z-DEVD-fmk, both inhibitors of interleukin-1β-converting enzyme (ICE)-like proteases, had no effect on cell viability in infected cultures (Fig. 6B). Moreover, the percentage of annexin-V-staining cells was reduced with increasing amounts of v-Bcl-2, confirming its ability to protect cells against apoptosis induced by the virus (Fig. 6C). Similar data were obtained with both v-Bcl-2 constructs. This effect was not due to an alteration of viral replication since viral DNA sequences were detected in 293 cells by PCR and Southern blotting analyses (data not shown) with prolonged incubation, and cytotoxic effects were eventually observed, suggesting that late gene expression can overcome the protective effects of v-Bcl-2.

FIG. 6.

Transient expression of v-Bcl-2 inhibits KSHV-induced cytotoxicity in 293 cells. (A) Full-length v-Bcl-2 cDNAs amplified from BCBL-1 or BKS-1 cells were subcloned into the expression vector pCR3.1. The proteins were in vitro translated in the presence of [³⁵S]methionine, separated on an SDS–12% polyacrylamide gel, and analyzed by autoradiography. No expression was observed with a construct containing v-bcl-2 in the opposite orientation (antisense). Positions of molecular weight markers (MW) are indicated in kilodaltons. (B) 293 cells were transfected with increasing amounts of plasmids encoding v-Bcl-2 or v-Bcl-2 in the reverse orientation (5 μg) by the calcium phosphate method. At 24 h posttransfection, the cells were infected with viruses derived from TPA-treated BKS-1 cells. Infected cells were cultured in the presence or absence of 100 μM Z-VAD-fmk or Z-DEVD-fmk. Mock-transfected cell cultures either uninfected (mock) or inoculated with a viral preparation pretreated by heat inactivation at 65°C for 10 min (mock-HI) served as controls. At day 3 postinfection, cell viability in the cultures was monitored by the trypan blue exclusion method. Data shown are the mean of three independent experiments ± standard deviation, and the percentage of viable cells was determined as the relative number of viable cells versus the total number of cells. (C) In parallel, an aliquot of infected cells (open curves) was stained with FITC-conjugated annexin-V and analyzed by flow cytometry. Mock-transfected cells (solid curve) served as control. Representative histograms of several independent experiments are shown.

DISCUSSION

Limitations in systems which allow productive viral replication have hampered the characterization of the KSHV life cycle and further definition of its role in the development of KS lesions. In AIDS-associated KS lesions, most cells are latently infected and few (1 to 5%) support productive infection (34). Furthermore, KS spindle cells established in culture tend to lose the viral genome in early passages (1). In contrast, several lines of evidence suggest that KSHV replicates more efficiently in other cell types, including B cells and macrophages (5, 22). Accordingly, KSHV lytic growth can be induced from the latently infected BCBL cell lines (3, 29). In addition, the human renal epithelial cell line 293 was shown to be susceptible to viral infection. In recent studies, serial propagation of KSHV was achieved in 293 cells from KS biopsy specimens and from saliva of individuals with past or current KS (15, 35). Herein, we have established transmission conditions of KSHV from early-passage cell lines derived from cutaneous KS lesions to an EBV-negative Burkitt’s B-lymphoma cell line, Louckes. Following serial passage, a low level of viral DNA replication and transcription was detected and further induced by TPA treatment in presumably latently infected Louckes cells, suggesting that KSHV lytic growth has occurred in this population (Fig. 1C and 2). Furthermore, the presence of mature progeny virions was demonstrated by nuclease sensitivity assays and TEM analysis (Fig. 3 and 4). The Louckes cell system therefore provides an alternative system for the study of viral gene expression and characterization of the virus life cycle in B-cell lines.

Previous experiments have shown the transmission of KSHV DNA sequences from BC-1, a cell line that is dually coinfected with EBV, to an EBV-positive Burkitt’s B-lymphoma cell line, Raji; however, the KSHV genome was lost with serial passages and productive lytic replication was marginal (24), possibly because the presence of EBV interfered with KSHV induction or genetic alterations after prolonged tissue culture occurred in the BC-1 cell line (23, 30). Size measurements of viral DNA by pulse-field gel electrophoresis analyses using EBV-positive BCBL-derived cell lines suggested a KSHV genome in the range of 210 to 270 kb (22, 24). In contrast, the genome size based on purified encapsidated DNA from BCBL-1 cells was estimated to be approximately 165 kb, whereas estimates from various KS lesions indicate a genome larger than 172 kb (28). Reminiscent of EBV, discrepancies among KSHV DNA sources might be explained by strain variability or genetic alterations such as rearrangements, mutations, or deletions. Accordingly, the existence of different KSHV strains has been suggested, based on DNA sequence heterogeneity between BCBL and KS specimens (40). In this study, we performed sequencing analysis of the KSHV ORF K1, located at the extreme 5′ end of the genome, and showed nucleotide divergence between cutaneous KS lesions and previously reported BCBL viral sequences. Thus, strain variability exists among different KSHV isolates, particularly those derived from lesions or lymphomas.

In this regard, we previously reported that KSHV isolates obtained from primary KS lesions were highly cytopathic upon infection of 293 cells (15). In the present study, 293 cells exhibited similar characteristic cytopathic changes associated with infection with virus-derived from KSHV-positive Louckes cells. Indeed, these cells were susceptible to apoptosis, as determined by flow cytometry following annexin-V staining (Fig. 4C). Although apoptosis was mediated by KSHV infection, it appeared to occur in many apparently uninfected cells, as noted previously (15). Such effects occurred in 293 cells when virus structures were not detected by TEM in morphologically apoptotic cells. Similar findings have been described recently for infection with HHV-6 and HIV type 1 (14, 19). It is possible that viral gene products or cytokines elicited during infection rendered uninfected bystander cells susceptible to apoptosis. Nevertheless, entry of virions and transcription of viral DNA was required for this effect since heat-inactivated viral preparations and UV-irradiated virions did not induce cell death. In contrast, virus-derived from TPA-induced BCBL-1 cells did not induce apoptosis in 293 cells. This observation may be the consequence of an abortive infection since the KSHV genome was lost from these infected cell cultures and serial passage in 293 cells was ineffective (data not shown). Alternately, we cannot exclude the possibilities that the two sources differ in viral gene expression and that genomic divergences contribute to their respective phenotypes. Renne et al. (27) recently reported that although 293 cells are highly susceptible to KSHV infection by virus from induced BCBL-1 cells, replication is not productive, consistent with the findings of this study. Regardless of the precise mechanism, viral DNA replication per se correlates with KSHV-induced apoptosis.

Recently, the KSHV genome was shown to encode a functional v-bcl-2 gene homolog (ORF16) in both KS specimens and BCBL cell lines (11, 32). Viral replication studies using recombinant Sindbis viruses harboring KSHV v-Bcl-2 indicated that this potent cell death suppressor could inhibit Sindbis virus-induced apoptosis (11). The data reported here document the antiapoptotic property of v-Bcl-2 in the KSHV life cycle. Transient transfection experiments using infected 293 cell cultures showed that expression of this gene product prevented early apoptotic cell death. In contrast, two broad-spectrum caspase inhibitors, Z-VAD-fmk and Z-DEVD-fmk, failed to prolong survival in infected cell cultures. These factors are potent, cell-permeating inhibitors of various forms of apoptosis, including cell death mediated by Fas/APO-1, dexamethasone, and HIV (6, 12). Conversely, Z-VAD-fmk does not prevent nonapoptotic cytolysis (i.e., necrosis) provoked by overexpression of the cellular death agonist Bak (38). Consistent with this observation, v-Bcl-2 neither homodimerizes nor heterodimerizes with Bax and Bak and may have evolved to escape their proapoptotic effects (11). The terminal effector of KSHV-induced cell death thus has yet to be defined; however, it is likely a downstream target of Bcl-2-related proteins. Taken together, these studies strongly suggest that the KSHV isolates obtained from KS lesions differ biologically from lymphoma-derived viruses, particularly with respect to the cytotoxic effects of the virus from primary lesions. Further characterization of viruses from these alternative sources will help to clarify their contributions to the pathogenesis of KS lesions and AIDS-associated lymphomas.