The Latency-associated Nuclear Antigen of Kaposi’s Sarcoma-associated Herpesvirus Transactivates the Telomerase Reverse Transcriptase Promoter*

,

Telomeric DNA, found at the distal ends of chromosomes, consists of a 6-base pair repeat of the sequence TTAGGG (1). Telomeres have an average length of 5-15 kilobases and function to prevent chromosome degradation, end-to-end chromosome fusions, and chromosome loss (2,3). Additionally, telomeres can induce cellular senescence when they are critically shortened (4,5). The telomeres of most human somatic cells shorten with each cycle of chromosome replication because DNA polymerase is unable to completely replicate the lagging strand (6). Telomerase is a ribonucleoprotein enzyme complex that functions to stabilize telomere length via the addition of new repeat sequence (2). Telomerase is active in cells with high regenerative capacity such as lymphocytes, hematopoietic progenitor cells, keratinocytes, and uterine endometrial cells (7)(8)(9). Moreover, telomerase is typically active in tumor-derived cell lines as well as in malignant tissues (10 -12). The activation of telomerase is a common and perhaps necessary step as cells move beyond replicative crisis to immortality associated with cancer cells (11).
Three subunits of telomerase have been identified. Two components are expressed in virtually all human cells with no correlation to telomerase activity. The first is an RNA component, hTR, that provides the template from which additional repeat sequence is constructed (13,14). The second is an integral protein component, TP1, the human homologue of the Tetrahymena telomerase P80 gene (15,16). Neither is downregulated during cellular differentiation when decreases in telomerase activity are observed (15,(17)(18)(19). The third component, hTERT, 1 is responsible for the enzymatic activity of telomerase (20,21). In contrast to hTR and TP1, the expression of hTERT correlates with telomerase activity, and hTERT mRNA is down-regulated during cell differentiation and upregulated during cell immortalization (9,10,22,23). The expression of hTERT is likely regulated through the action of transcription factors targeting the hTERT promoter. An active hTERT promoter is observed in telomerase-positive cell lines and cancer cells, whereas the promoter is repressed in telomerase-negative cells (24,25).
The products of known oncogenes and tumor suppressors have been shown to activate and repress, respectively, the hTERT promoter. Following the initial identification of Myc as an activator (25)(26)(27), further effects were observed in studies with the transactivator Sp1 and with the repressors WT1, Mad1, p53, and MZF-2, all shown to modulate the activity of the hTERT promoter (28 -32). Although the E6 gene product of human papillomavirus type 16 has been shown to activate telomerase by functional assays (33), to date no tumor virus gene product has been shown to specifically activate the hTERT promoter.
Prior to cases related to the AIDS epidemic of the early 1980s, KS was a relatively rare skin neoplasm typically afflicting elderly men of Mediterranean descent, immunosuppressed individuals such as transplant recipients, and young men of sub-Saharan African countries (34,35). The high prevalence of KS among HIV-infected homosexual men and the low prevalence among HIV-infected hemophiliacs and intravenous drug users strongly implicated transmission of an infectious agent through sexual activity (36,37). KSHV, also referred to as human herpesvirus 8, was subsequently identified by polymerase chain reaction-based studies and designated a ␥-herpesvirus placing it in the same family as the primate Rhadinovirus herpesvirus saimiri and the human Lymphocryptovirus Epstein-Barr virus (38,39). KSHV is strongly linked to Kaposi's sarcoma with infected individuals converting to seropositivity prior to expressing a disease phenotype (40). There is a greater than 90% correlation between virus nucleic acid detection and disease (41). KSHV also targets the endothelial-derived spindle cell, which is the primary tumor cell associated with KS (42,43). KSHV has since been associated with other malignancies including multifocal Castleman's disease and a rare B cell lymphoma, the body cavity-based lymphoma or primary effusion lymphoma (44 -46). Body cavity-based lymphoma-derived cell lines are the only KSHV-infected cells easily amenable to study in culture, and KSHV infection in these cell lines is predominantly latent (47). More recently, human-derived microvascular endothelial cells have been shown to support latent infection with KSHV. These cells undergo morphological changes resulting in a phenotype that is similar to that seen in the formation of spindle cells in KS (48). Additionally, KSHV has been shown to infect a variety of cultured cells, including HEK 293, by reverse transcription-polymerase chain reaction detection of a spliced late mRNA (49). KSHV has also been detected by polymerase chain reaction in infected HEK 293 cells but not uninfected ones during serial passage (50).
LANA is a highly immunogenic protein encoded by open reading frame 73 of the KSHV genome (51). The 1162-amino acid LANA protein has definable domains that include an amino-terminal proline-rich region, an ϳ100-amino acid acidic domain, a large glutamine-rich repetitive domain, and a leucine zipper motif (see Fig. 1; Refs. 39 and 52). Additionally, LANA possesses potential nuclear localization signals and numerous phosphorylation sites recognized by several common kinases (53,54). LANA is constitutively expressed during viral latency and is important for maintenance of the viral episome during chromosome replication (55,56). The proliferative nature of KS and other KSHV-associated diseases as well as structural motifs of LANA that potentially interact with a variety of cellular factors suggest a role for LANA in mediating viral oncogenesis through transcriptional regulation. Recently, it was shown that LANA antagonizes the tumor suppressor p53, thereby protecting against cell death (57). LANA has also been shown to target the retinoblastoma protein regulating E2F responsive promoters (58). Here, we show that LANA likely contributes to cell immortalization by transactivation of the hTERT promoter.

EXPERIMENTAL PROCEDURES
Plasmids, Cell Lines, and Culture Conditions-The preparation of pGL3B-hTERT luciferase reporter constructs and the pA3M-LANA expression construct have been described previously (25,56). The pGL3B-hTERT constructs were obtained from J. Carl Barrett. HEK 293 cells are human embryonic kidney cells transformed by adenovirus type 5 DNA; the HEK 293T cell line is an HEK 293-derived line that stably expresses the SV40 T-antigen (59). BJAB is a B cell derived from a patient with Epstein-Barr virus-negative African Burkitt's lymphoma (60). BJAB and Rat-1 cell lines were obtained from Elliott Kieff. SUSM-1 cells maintain their telomeres by a telomerase-independent mechanism and are consequently negative for telomerase activity (61). The SUSM-1 cell line was prepared by mutagen treatment of fetal human diploid fibroblasts (62). SUSM-1 cells were provided by Masayoshi Namba. BC-3 is a KSHV-positive body cavity-based lymphomaderived cell line obtained from the American Type Culture Collection (63). HEK 293, HEK 293T, Rat-1, and SUSM-1 cells were grown in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 10% fetal bovine serum, 2 mM glutamine, 25 units/ml penicillin, 25 g/ml streptomycin, and 10 g/ml gentamicin. BJAB cells were grown in RPMI medium 1640 (Life Technologies, Inc.) supplemented as for Dulbecco's modified Eagle's medium. BC-3 cells were grown in RPMI medium 1640 supplemented as for Dulbecco's modified Eagle's medium but with 20% fetal bovine serum. Cells were grown at 37°C in a humidified environment supplemented with 5% CO 2 .
Transfection and Luciferase Assay-HEK 293, HEK 293T, Rat-1, and SUSM-1 cells were collected at 70% confluency by trypsinization with trypsin-EDTA (Life Technologies, Inc.). 10 million cells were resuspended, along with plasmid DNA (generally 5 g of pGL3-Basic reporter and 10 g of pA3M expression construct), in 400 l of Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. The cells were transfected by electroporation with the Bio-Rad Gene Pulser II at 210 V and 975 microfarads. Transfected cells were transferred to 100-mm plates in 10 ml of Dulbecco's modified Eagle's medium with 10% fetal bovine serum. The plates were incubated at 37°C in a humidified environment supplemented with 5% CO 2 for 20 h. BJAB cells were collected at 5 ϫ 10 5 cells/ml. 10 million cells were transfected at 220 V and 975 microfarads.
At 20 h, cells were harvested and washed once with phosphatebuffered saline (Life Technologies, Inc.). Luciferase activity was determined as per manufacturer's instructions using the luciferase assay system with reporter lysis buffer (Promega). Briefly, 200 l of lysates were prepared by freeze/thaw in reporter lysis buffer. The lysates were subsequently diluted 1:10 in reporter lysis buffer, and 40 l of the dilution was mixed with 100 l of luciferase assay reagent. Luminescence was measured for 10 s by the Opticomp I luminometer (MGM Instruments, Inc.). It should be emphasized that lysates were always diluted to ensure that luciferase activity was within the linear range of the assay. The results shown represent experiments performed in triplicate. The experiments were repeated multiple times to ensure that the observed trends were reproducible.
Western Blot Analysis-The same lysates prepared for luciferase assay were aliquoted and mixed 1:1 with 2ϫ SDS gel-loading buffer (100 mM Tris⅐Cl, pH 6.8, 200 mM dithiothreitol, 4% SDS, 0.2% bromphenol blue, and 20% glycerol), placed at 95°C for 5 min, and spun to remove cell debris. Soluble proteins were fractionated by electrophoresis on a 6% SDS-polyacrylamide gel. The fractionated protein was then transferred to a 0.45-m nitrocellulose membrane. The membrane was blocked with 5% dry milk in PBS and was subsequently blotted with human serum, adsorbed with B cell extracts, and previously shown to be reactive to LANA (56, 64) at a 1:50 dilution followed by horseradish peroxidase-linked protein A (Amersham Pharmacia Biotech) at a 1:5000 dilution. The blot was visualized by a standard chemiluminescence detection protocol (Amersham Pharmacia Biotech).
Immunofluorescence-BJAB cells were transfected with either pA3M or pA3M-LANA and incubated as described above. Cells were harvested and washed with PBS. Slides of transfected BJAB and nontransfected BC-3 were prepared and fixed with 1:1 acetone:methanol. The slides were incubated with 20% goat serum for 30 min at room temperature in a humidity chamber. After washing with PBS, slides were incubated with specific antibody, either human serum reactive to LANA at 1:100 dilution or Myc mouse monoclonal IgG at 1:500 dilution, for 2 h. After washing thoroughly with PBS, slides were incubated with either FITCconjugated goat anti-human antibody or FITC-conjugated goat antimouse antibody at 1:1000 dilution for 1 h. Slides were washed thoroughly with PBS, and proteins were visualized on an Olympus BX60 fluorescent microscope. The photographs were captured using an Olympus digital camera and the Esprit program version 1.2.
In Vitro DNA Binding-Sp1 DNA probes were prepared by annealing complementary oligonucleotides containing the GC-rich Sp1 DNA binding site (Life Technologies, Inc. custom primers). The wild-type probe sequence was taken from Ϫ119 to Ϫ98 of the hTERT promoter (GCGCGGACCCCGCCCCGTCCCG). The mutant probe sequence was GCGCGGACCCCGAACCGTCCCG. Probes were end-labeled with terminal transferase and [␣-32 P]dGTP. Sp1 mouse monoclonal IgG was purchased from Santa Cruz Biotechnology. In vitro translated LANA was made with the TNT quick-coupled in vitro translation kit (Promega) according to the manufacturer's recommendations. Nonspecific competitor DNA was TLBR1, an unrelated 31-base pair probe. BJAB nuclear extract was prepared as previously detailed (65). Electrophoretic mobility shift assay reactions were performed as described previously (66).

RESULTS
LANA Activates Transcription of the Full-length hTERT Promoter in HEK 293 Cells-The proliferative nature of cells latently infected with KSHV as well as structural domains of LANA having the potential to mediate protein-protein and protein-DNA interactions suggest a role for LANA in regulating cellular gene expression (Fig. 1). Because expression of hTERT is primarily regulated at the transcriptional level, the effect of LANA on hTERT expression was investigated by testing the ability of LANA to regulate expression from luciferase reporter constructs containing the hTERT promoter in HEK 293 cells. The HEK 293 cell line has been employed in previous studies of KSHV infectivity and propagation. In one study KSHV was cytotoxic to 293 cells and was detected by polymerase chain reaction in infected cells but not uninfected ones during serial passage (50). In another study 293 cells were susceptible to KSHV infection by detection of a spliced late mRNA (49). These data support the use of the HEK 293 cell line for preliminary examination of potential LANA effects on gene expression. The region Ϫ1665 to ϩ5 of the hTERT promoter was cloned into the pGL3-Basic luciferase reporter construct such that initiation of transcription via the hTERT promoter would drive the transcription and ultimately the translation of the luciferase gene (25). This reporter, pGL3B-TRTP, was transfected into HEK 293 cells along with pA3M-LANA, a Myc-tagged LANA expression vector. Initial cotransfection of LANA expression vector with the hTERT reporter construct consistently resulted in 4.5-fold activation relative to reporter construct alone. Increasing the concentration of the LANA expression construct consistently augmented the activation to 7-fold. These data indicate that the transactivation of the hTERT promoter in the HEK 293 cell line is directly proportional to the quantity of LANA expressed as demonstrated by the observed dose-response relationship (Fig. 2). Further increasing the amounts of LANA resulted in abrogation of this activity and increased cell death (data not shown).
The hTERT Promoter Element Is Endogenously Activated in Human Cell Lines-To assess the possibility that the above response was due to either the total amount of DNA transfected or the particular cell line employed, similar assays were performed in five cell lines, HEK 293 and 293T, BJAB, Rat-1 fibroblasts, and SUSM-1 fibroblasts. HEK 293, HEK 293T, BJAB, and Rat-1 fibroblasts have significant endogenous telomerase activity. In contrast, SUSM-1 fibroblasts maintain their telomeres by a telomerase-independent mechanism and were previously shown to have little or no telomerase activity (61). In HEK 293T and 293 cell lines, a 15-fold increase in luciferase activity was observed when the telomerase promoter was cloned upstream of the luciferase gene. Similarly, the BJAB cell line showed a 20-fold increase in activation (Fig. 3A). It should be noted that pGL3-Basic background activity differed in these three cell lines and that fold activation of the hTERT promoter was always calculated relative to the appropriate pGL3-Basic/pA3M vector alone control.
To demonstrate that expression of LANA further augments transcription at the hTERT promoter, the aforementioned cell lines were transfected with pGL3B-TRTP luciferase reporter and either LANA expression plasmid or vector control. In HEK 293T and BJAB cell lines, expression of LANA consistently resulted in an additional 1.5-2-fold activation of the hTERT promoter over background activity. Moreover, in HEK 293 cells expression of LANA resulted in 5-7-fold activation relative to background (Fig. 3B). These data indicate that LANA can activate the hTERT promoter over endogenous hTERT promoter activity in these telomerase-positive cell lines and that this effect is not due to the total amount of DNA transfected. As discussed in more detail below, the modest activation observed in HEK 293T and BJAB cell lines was expected because these lines express potential activators of the hTERT promoter, SV40 T-antigen and Myc, respectively. Just as expression of the T-antigen overwhelms LANA-mediated activation of the hTERT promoter in the HEK 293 background, Myc expression likely limits potential fold activation of the hTERT promoter in the BJAB cell line.
In comparison, Rat-1 fibroblasts and the telomerase-negative fibroblast cell line SUSM-1 were examined for LANAmediated activation of the hTERT promoter. No significant activation over endogenous hTERT promoter activity was observed in either fibroblast cell line (Fig. 3B). It should be noted that although SUSM-1 demonstrated less endogenous hTERT promoter activity than the other cell lines examined (Fig. 3A), in our experiments the activation seen was greater than expected based on previous reports (25,61). Additionally, it was observed that as the SUSM-1 cell line was expanded in culture for the transfection experiment, the cells acquired a more quickly growing phenotype. This change in phenotype is perhaps attributable to changes in cellular gene expression that  (39,52). LANA possesses putative nuclear localization signals (NLS) at both the amino and carboxyl termini (53,54). Potential phosphorylation, glycosylation, myristoylation, and amidation sites are also present but are too numerous to be displayed here.
partially activated the hTERT promoter explaining the significant endogenous activation observed by luciferase assay.
Transiently Transfected LANA Is Expressed in Human Cell Lines-To demonstrate that the LANA expression vector employed here, pA3M-LANA, effectively expresses LANA protein in our transient reporter assay, cell lysates from a number of cell lines were examined by immunoblot and immunofluorescence. The results demonstrated a series of bands above the 215-kDa marker in pA3M-LANA-transfected HEK 293T and 293 cells; no bands were present in pA3M control lanes (Fig.  4A, lanes 2 and 4 and lanes 1 and 3, respectively). Protein corresponding to 1.5 million HEK 293 cells was necessary to mimic the LANA intensity of 0.5 million HEK 293T cells, indicating that LANA expression was enhanced in HEK 293T relative to HEK 293, probably because of the presence of the SV40 T-antigen (Fig. 4A, lanes 2 and 4).
To compare the level of expression in transiently transfected cells with KSHV-infected cells, immunofluorescence analysis was performed. Aliquots of BJAB cells were transfected as for hTERT luciferase assays, and slides were prepared for both transfected BJAB and BC-3, a KSHV-positive body cavitybased lymphoma-derived cell line (63). In BC-3 and LANAtransfected cells, FITC-mediated fluorescence was localized to the nucleus (Fig. 4B, compare left and right panels). To demonstrate that this fluorescence was specific for the Myc-tagged LANA expressed here, Myc monoclonal IgG was used for both pA3M and pA3M-LANA-transfected BJAB. Nuclear localization of the fluorescence signal was observed for pA3M-LANAtransfected BJAB but not for the pA3M-negative control (Fig.  4C, compare left and right panels).
LANA Activates Transcription of Truncated hTERT Promoter Constructs-Since the recent cloning of the hTERT promoter, the products of known oncogenes and tumor suppressors have been shown to both activate and repress the hTERT promoter via specific protein-DNA interactions (25, 28 -32). To examine which region or regions of the hTERT promoter are responsible for LANA-mediated activation, reporter plasmids, with serial truncations of the hTERT promoter cloned upstream of the luciferase gene (25), were cotransfected with LANA expression plasmid in HEK 293 cells (Fig. 5A). In all cases, both endogenous activity and LANA-enhanced activity were normalized as fold activation relative to pGL3-Basic/ pA3M vector control as shown in the left column in panels B-F of Fig. 5. The endogenous or LANA-independent activation of these hTERT truncations increased significantly with the removal of 5Ј sequence to position Ϫ408, perhaps attributable to elimination of repressor binding (Fig. 5, compare B and C, middle columns). Subsequent removal of 5Ј sequence to Ϫ149 depressed endogenous activation ϳ2-fold (Fig. 5, compare D and E, middle columns). This is consistent with previous reports that the elimination of the region from Ϫ208 to Ϫ149 and specifically the E-box that it contains depresses endogenous hTERT activation (25). Despite variations in the endogenous activation of the hTERT promoter with serial 5Ј deletion, LANA consistently activated transcription relative to background activity in all constructs examined. To demonstrate that LANA does not similarly activate transcription of the luciferase gene in the context of a pGL3-Basic vector with no hTERT promoter sequence inserted, HEK 293 cells were transfected with pGL3-Basic and either LANA expression plasmid or vector as was done for pGL3B-hTERT promoter constructs. Fold activation in the presence of LANA was then calculated relative to endogenous activation of the appropriate pGL3 reporter (Fig. 6). These data demonstrate two important effects. First, although endogenous or LANA-independent activation of the hTERT promoter differs for each hTERT promoter truncation (Fig. 5), LANA constitutively activates transcription ϳ3-4-fold relative to endogenous promoter activity for all hTERT promoter truncation constructs tested (Fig. 6). Second, LANA does not activate transcription of the luciferase gene in the context of a pGL3-Basic vector with no hTERT sequence in- FIG. 3. LANA transactivates the full-length hTERT promoter relative to vector control in 293T, 293, and BJAB cell lines. A, 10 million HEK 293T, HEK 293, BJAB, Rat-1, or SUSM-1 cells were transfected either with 5 g of pGL3B-TRTP or with pGL3-Basic control and 10 g of pA3M vector control. The promoter activity for each combination of plasmids was expressed as the fold activation relative to pGL3-Basic/pA3M control. The means and standard deviations from three independent transfections are shown. B, cell lines as above were transfected with 5 g of pGL3B-TRTP and 10 g of either pA3M-LANA or pA3M vector control. The promoter activity for each combination of plasmids was expressed as the fold activation relative to pGL3B-TRTP/ pA3M control. The means and standard deviations from three independent transfections are shown. serted (Fig. 6). These two observations suggest that the smallest hTERT promoter truncation examined, the Ϫ130 to ϩ5 fragment, is sufficient to mediate LANA activation of the hTERT promoter relative to endogenous activation.
LANA Targets the Sp1-DNA Interaction in the Context of the hTERT Promoter-The Ϫ130 to ϩ5 region of the hTERT promoter is significant in that it contains five GC-rich boxes (25). Previous electrophoretic mobility shift assay experiments have demonstrated binding of transcription factor Sp1 to these five sites (24); further, it has been suggested that Myc, via the aforementioned E-box, and Sp1 are the major determinants of endogenous hTERT expression (28). To ask the question of whether LANA activates the hTERT promoter via a GC-boxmediated interaction, a probe was designed encompassing a single GC-box. The sequence for the double-stranded DNA probe was taken from Ϫ119 to Ϫ98 of the hTERT promoter and was chosen because this GC-box matches exactly the consensus Sp1 binding sequence. By electrophoretic mobility shift assay, the probe interacted with BJAB nuclear extracts to yield at least two GC-box-specific bands (Fig. 7, arrows on left). These bands were designated as specific shifts because they both disappeared with GC-box mutant probe (Fig. 7, right panel). Additionally, these two specific complexes were competed by the addition of 200ϫ specific competitor; however, these bands were not competed by nonspecific DNA competitor (Fig. 7, left  panel). Furthermore, one of the specific bands was supershifted with Sp1 monoclonal antibody (Fig. 7, asterisk). In the presence of in vitro translated LANA, this Sp1-specific signal was significantly ablated along with the other GC-box-specific bands (Fig. 7, arrows). This ablation was expected because a new LANA complex would likely be too large to migrate significantly on the gel. It should be noted that these GC-box-specific signals were not affected by the addition of nonspecific competitor protein. The small inset at the left of Fig. 7 shows the LANA-specific signal from the in vitro translated LANA used in this assay. Taken as a whole, these data suggest that the FIG. 4. LANA is expressed in HEK 293T, HEK 293, and BJAB cells. A, from Fig. 3, selected lysates were fractionated by electrophoresis on a 6% SDS-polyacrylamide gel. Volumes corresponding to 0.5 million HEK 293T cells and 1.5 million HEK 293 cells were used. The protein was transferred to a nitrocellulose membrane; the membrane was blotted with human serum reactive to LANA from a KS patient followed by horseradish peroxidase-linked protein A. The blot was visualized by a standard chemiluminescence detection protocol. B, BJAB transfected with pA3M-LANA and KSHV-infected BC-3 were fixed in 1:1 methanol:acetone and incubated with KS patient serum reactive to LANA. The secondary antibody used for detection was FITC-conjugated goat anti-human antibody. C, BJAB transfected with either pA3M-LANA or pA3M were fixed in 1:1 methanol:acetone and incubated with monoclonal mouse anti-Myc antibody to detect Myc-tagged LANA. The secondary antibody used for detection was FITC-conjugated mouse anti-human antibody. endogenous DNA-protein complex formed at the GC-box and including Sp1 is targeted by LANA. DISCUSSION Latent infection with KSHV is believed to play a causal role in several proliferative lesions, namely KS, body cavity-based lymphoma, and multicentric Castleman's disease (40, 44 -46).
KSHV also transforms primary human endothelial cells in vitro, and by telomeric repeat amplification protocol assay it was demonstrated that these KSHV-transformed cells had enhanced telomerase activity; telomerase activity was not detected in uninfected cells (43). Although the E6 gene product of human papillomavirus type 16 activates telomerase by telomeric repeat amplification protocol assay (33), the effect of other viruses and their gene products on the regulation of telomerase activity remains relatively unexplored. Here we propose that LANA, a constitutively expressed protein detected in latent KSHV infection, potentially contributes to primary cell transformation and the proliferative nature of KSHV-infected lesions by activating the hTERT promoter.
Just as KSHV does not mediate transformation of all cell types infected, LANA-mediated activation of the hTERT promoter was not equally efficacious in all cell lines examined here. The most dramatic activation was observed in the HEK 293 cell line as LANA activated the wild-type hTERT promoter 5-7-fold. In contrast, activation was ϳ2-fold in HEK 293T and BJAB cell lines. This discrepancy was expected and can likely be attributed to differences in endogenous Myc expression and subsequent activation of the hTERT promoter in these cell lines. BJAB is a B cell line derived from a patient with Epstein-Barr virus-negative African Burkitt's lymphoma (60). African Burkitt's lymphoma is notable for translocation of the Myc gene into the immunoglobulin heavy chain locus resulting in constitutive high level Myc expression (67). Myc expression is known to strongly activate the hTERT promoter (25). Consequently, in the BJAB cell line, LANA-mediated fold activation was likely less dramatic because of this already strong endogenous activation mediated by Myc. Similarly, HEK 293T expresses the SV40 T-antigen, a promiscuous transactivator that has been shown to activate the human Myc promoter (68). In fact, endogenous activation of the hTERT promoter was 10-fold higher in HEK 293T relative to HEK 293 by transient transfection luciferase assay, suggesting that endogenous hTERT promoter activity was indeed up-regulated in the HEK 293T FIG. 6. The specificity of LANA activation of the hTERT promoter is conferred by an element or elements between ؊130 and ؉5. The fold activation was calculated with the data presented for HEK 293 cells in Fig. 5. The luciferase activity for cells transfected with 5 g of luciferase reporter and 10 g of pA3M-LANA was divided by the luciferase activity for cells transfected with 5 g of the same luciferase reporter and 10 g of pA3M vector control. The error bars represent the summation of relative error for each of the above values. cell line (data not shown). Although the cell lines employed here provide a good initial screen of LANA-mediated hTERT expression effects, we are currently planning further experiments to examine these effects in endothelial cells and primary B cells, cell lines particularly relevant to latent KSHV infection.
Ideally, the cell lines employed in these studies would have undetectable telomerase activity and, consequently, minimal endogenous activation of the hTERT promoter; however, because up-regulation of telomerase activity is a common step in cell immortalization, most cell lines amenable to the assays utilized here have significant detectable telomerase activity. An exception is the SUSM-1 cell line. This line was derived by mutagen treatment of fetal human diploid fibroblasts and maintains telomere length by a telomerase-independent mechanism (61). However, transient transfection experiments performed to assay for LANA-mediated hTERT promoter activation indicated that neither SUSM-1 nor Rat-1 fibroblast cell lines facilitated activation. This suggests that cell-specific factors in the human embryonic kidney and BJAB cell lines necessary for LANA-mediated activation were absent or sequestered in these fibroblast cell lines. As was mentioned previously, all cell lines examined, including the telomerasenegative cell line SUSM-1 (61), showed significant endogenous activation of the hTERT promoter (Fig. 3A). It would be of interest to further examine LANA-mediated activation of the hTERT promoter in a human embryonic or B cell line lacking endogenous hTERT promoter activity.
Promoter truncation studies implicated the Ϫ130 to ϩ5 region of the hTERT promoter in LANA-mediated activation. This region was important in that it contained five GC-rich boxes, the consensus binding sites for transcription factor Sp1. Electrophoretic mobility shift assays were performed to assess the interaction of LANA using a probe containing a selected GC-box. The data presented here suggest that LANA alters a specific interaction between Sp1 and the GC-box. However, neither a LANA-specific shift nor a supershifted complex in the presence of LANA was observed. Although Sp1, a 558-amino acid protein, was amenable to gel shift assays resulting in specific shifted complexes and supershifted complexes, LANA, a 1162-amino acid protein, would likely be difficult to assay by this method. One explanation for our data is that LANA interacts with the Sp1-DNA complex, creating a new complex that is too large to migrate significantly on the gel. Alternatively, LANA might be recruited to the GC-box displacing the Sp1 complex, creating new complexes and releasing the bound DNA. It should, however, be noted there was no specific interaction between LANA and the GC-box in the absence of BJAB nuclear extract. Overall, these data suggest that the endogenous DNA-protein complex formed at the GC-box is targeted by LANA, potentially implicating the GC-box in LANA-mediated activation of the hTERT promoter.