Previous Article | Next Article ![]()
Journal of Virology, May 2007, p. 4848-4857, Vol. 81, No. 9
0022-538X/07/$08.00+0 doi:10.1128/JVI.02530-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Ginette Dambrine,1
Denis Soubieux,1
Emmanuel Kut,1 and
Denis Rasschaert1,2*
Equipe Télomérase et Lymphome Viro-induit, Centre INRA de Tours, Unité IASP 213, 37380 Nouzilly, France,1 Université François-Rabelais Sciences et Techniques, Parc de Grandmont, 37200 Tours, France2
Received 16 November 2006/ Accepted 13 February 2007
|
|
|---|
|
|
|---|
The detailed molecular mechanisms by which MDV transforms T cells are still unknown. The rapid onset of MDV-induced lymphoma formation, however, suggests that one or more virus-encoded oncogenes are involved in the transformation process. The MDV-encoded 339-amino-acid protein Meq, a basic leucine zipper protein (bZIP), is the only member of the Jun/Fos oncoprotein family present in herpesviruses. This protein has previously been reported to be involved in MDV-induced tumorigenesis (2, 21). Through homodimerization or heterodimerization with c-Jun, JunB, or Fos, Meq can bind to promoters containing AP-1 or to so-called Meq-responsive element (MERE) sites, thereby regulating the transcription of other viral and cellular genes (20). Another viral gene encoding the viral RNA subunit of telomerase (vTR) has been reported to have a major impact on the malignancy of MDV-induced lymphoma (30). We previously identified the vTR gene in the repeat regions flanking the UL (TRL and IRL) region. Thus, as for the gene encoding Meq, the vTR gene is present as two copies in the MDV genome (10). The deletion of both copies of the vTR gene from an infectious clone of the highly oncogenic MDV strain RB-1B reduced the tumor incidence by 60% in infected chickens compared with that of parental virus, with no effect on lytic MDV replication or the establishment of latency (30). This study also showed that the constitutive expression of vTR in the chicken fibroblast line DF-1 resulted in a phenotype consistent with transformation similar to that of DF-1 cells transformed by the MDV oncoprotein Meq.
The most straightforward interpretation of the tumor-promoting properties of vTR is that this subunit enhances telomerase activity, leading to the stabilization of telomeres, the nucleoprotein structures constituting the ends of eukaryotic chromosomes. Indeed, telomerase is a ribonucleoprotein composed of TERT, the telomerase reverse transcriptase, and TR, the telomerase RNA subunit encoding the template sequence reverse transcribed by the telomerase enzyme to telomeres. Telomerase upregulation in human cancer plays a key role in cell immortalization, and abnormally high levels of telomerase activity are detectable in over 85% of human cancers (17, 1). A role for vTR in the T-cell immortalization process would be due to the capacity of this subunit to enhance telomerase activity during MDV infection. Indeed, a comparative functional study has shown that vTR yields higher levels of telomerase activity than its chicken ortholog, chTR, when combined with recombinant chicken TERT protein in vitro (10). This greater efficiency of vTR over chTR seems to be due to mutations affecting the pseudoknot core domain, which presumably stabilizes the pseudoknot P2 helix (11).
Human TR (hTR) seems to be constitutively expressed in a broad range of human cells, but hTR levels are generally higher in cancer cells, and there is evidence that the upregulation of hTR is an early event during multistage tumorigenesis (35). TERT expression in humans is otherwise correlated with telomerase activity. Several transcription factors regulating human TERT (hTERT) transcription have been identified. These factors include c-Myc and Sp1, which have been described as the principal activators of hTERT transcription, with responsive elements located within the proximal core hTERT promoter (19, 33). Human cancers caused by infections with oncogenic viruses, e.g., human papilloma virus 16 (HPV16), Kaposi's sarcoma-associated herpesvirus (KSHV), and Epstein-Barr virus (EBV), have been associated with the upregulation of telomerase activity through a virus-mediated upregulation of hTERT expression (18, 15). Thus, all these viruses increase cellular telomerase activity, but MDV is currently the only one known to encode a functional telomerase subunit in its genome.
We provide the first demonstration that telomerase activity is strongly upregulated in the peripheral blood mononuclear cells (PBMC) of chickens infected with the highly oncogenic MDV strain RB-1B. The observed increase in telomerase activity was positively correlated with high levels of vTR expression. We also identified a TATA-like box as the main element controlling the basal transcriptional activity of promoters vTR and chTR. Finally, we carried out a functional analysis of the vTR promoter-specific cis elements. We found that vTR expression was specifically regulated by an E-box located two nucleotides downstream from the transcriptional initiation site in two different transformed avian cell lines. Chromatin immunoprecipitation (ChIP) assays demonstrated that the vTR promoter region encompassing this E-box binds the oncoprotein c-Myc. We also found that the Ets binding site (EBS) was specifically involved in regulating vTR transcription in the MDV-transformed lymphoblastoid cell line MSB-1.
|
|
|---|
Experimental in vivo assay and PBMC isolation. We used White Leghorn specific pathogen-free B13B13 chickens highly susceptible to Marek's disease virus strain. These animals were bred, raised, and housed in isolated accommodations. Chickens were injected intramuscularly at the age of 8 weeks with 1,000 PFU of the highly oncogenic MDV strain RB-1B, using a chicken embryo fibroblast suspension as previously described (8). PBMC were collected as previously described (9) at various times after inoculation to measurement telomerase activity and gene expression.
All the surviving chickens were killed 45 days after inoculation, and necropsies were performed. Mortality and macroscopic tumors were recorded. All experimental procedures were conducted in compliance with approved protocols for the use of animals in research.
Statistical analysis was carried out with the nonparametric Kruskal-Wallis test, as implemented in Simstat version 2.04 (Provalis Research). We compared the telomerase activity and the relative gene expression levels of PBMC from infected and from control chickens. We considered P values of
0.045 to be statistically significant.
TRAP assay. We quantified the telomerase activity of 600 ng of protein extracted from PBMC, using the semiquantitative fluorescence-based telomeric repeat amplification protocol (TRAP) assay, as previously described (10). Briefly, the PCR steps were carried out using the 6-carboxytetramethylrhodamine-labeled forward primers TS and CX-ext as the reverse primers. An internal amplification standard was also added to the PCR mixture. PCR products were analyzed by capillary electrophoresis (ABI Prism 310; PerkinElmer Life Sciences). The telomerase activity level of each protein extract was estimated by adding the integrated value of each telomerase elongation product normalized to the integrated value of the internal amplification standard.
Reverse transcription-PCR analysis. Levels of chTERT, chTR, and vTR expression were quantified by fluorescence-based reverse transcription (RT)-PCR, as previously described (10). The primers used for reverse transcription were an oligodT(15) primer for chTERT and the DS4 primer (5'-GCCCGCTGAAAGTCAGCGAGTA-3') for chTR/vTR. The resulting cDNA was amplified by PCR using specific tetrachlorofluorescein phosphoramidite-labeled forward primers and specific reverse primers (Table 1). An internal control (IC) was included in the PCR for each sample (sizes of the IC amplification products are shown in Table 1). The PCR protocol consisted of 35 cycles of 45 s at 94°C, 45 s at 55°C, and 45 s at 72°C. Amplification products were analyzed with an automated ABI Prism 310 fragment analyzer (Perkin-Elmer Corp., MA). Gene expression was calculated as the ratio of the integrated value for the gene product normalized to the integrated IC value.
|
View this table: [in a new window] |
TABLE 1. Primers used for RT-PCR and size of PCR products
|
AP1, pvTR
E1, pvTR
E2, pvTR
CAATx, pvTR
CAATy, pchTR
CAATx, and pchTR
CAATy. The primers and templates used in this approach are shown in Table 2 and 3. Distal mutations were introduced into the promoter directly via the primer sequence used for amplification. This method was used to generate the pvTR
E3 construct (Table 2 and 4). The same method was used to generate the pvTR
TATA, pvS-TR
Sp1, pvTR
EBS, and pchTR
TATA constructs, using two successive PCR amplifications (Table 2 and 4). The pvTR
E2
E3 and pvTR
EBS
E3 constructs were generated by amplification by PCR from pvTR
EBS and pvTR
E2 (Table 2 and 4). All these modified promoters were inserted into the pGL3-Basic vector, as described above.
![]() View larger version (15K): [in a new window] |
FIG. 2. Comparison of the promoter regions of the viral and the chicken TR genes. Potential transcription factor binding sites are indicated in the promoter sequences. The transcriptional start site (+1) is indicated on each sequence by an arrow. Sites present in viral TR promoter sequences but absent from the chicken TR promoter include AP-1, E-boxes, and the EBS. Four Sp1 motifs are found in identical locations in the vTR and chTR promoter sequences. The vTR promoter contained four Sp1 sites in addition to these four sites, whereas the chTR promoter contained three additional Sp1 sites.
|
|
View this table: [in a new window] |
TABLE 2. Primers used to perform mutated TR promoters
|
|
View this table: [in a new window] |
TABLE 3. Production of mutant constructs: internal mutations
|
|
View this table: [in a new window] |
TABLE 4. Production of mutant constructs: distal mutations
|
All of the intermediate and final constructs were checked by sequencing with appropriate primers.
Luciferase assay. For the luciferase assay, Renilla luciferase plasmids (pRL) were used for cotransfection to standardize transfection efficiency. PCDNAMLuc, which carries the firefly luciferase gene downstream from the CMV promoter, and the pGL3-Basic vector were also used for transfection as positive and negative controls, respectively.
LMH cells were seeded on 96-well-plates (5 x 104 cells/well). They were cultured overnight and cotransfected with 300 ng of luciferase reporter plasmids and 3.75 ng of pRL plasmids, using Lipofectamine 2000 transfection reagent (Invitrogen-Life Technologies, Cergy Pontoise, France). For c-Myc overexpression assays, cells were cotransfected with 150 ng of c-Myc expression vector, the reporter plasmids, and pRL.
MSB-1 cells were electroporated using an Equibio "EasyjecT Plus" electroporator (single pulse, 400 V, 1500 µF) and aluminum electrodes (4-mm cuvette;Eurogentec). For all assays, 8 x 106 cells were electroporated in the presence of 40 µg of luciferase reporter plasmids and 500 ng of pRL diluted in RPMI 1640 medium. After electroporation, cells were plated in 6-well plates containing 2.5 ml of RPMI 1640 medium supplemented with 10% fetal bovine serum. The plates were then incubated at 41°C.
Luciferase assays were performed 24 h after transfection, using a Dual-Luciferase reporter assay system (Promega, Madison, WI). Firefly luciferase activities were normalized with respect to Renilla luciferase activity. Each transfection reaction was carried out at least three times. Mean relative luciferase activity is presented. The significance of differences between promoter constructs was assessed with the nonparametric Kruskal-Wallis test implemented in Simstat version 2.04 (Provalis Research). We compared the luciferase activity for each mutated promoter with that for the corresponding wild-type promoter. We considered P values of
0.045 to be statistically significant.
Western blot analysis. The whole LMH cell extracts used for Western blotting to assess c-Myc levels were prepared by osmotic shock followed by sonication. Proteins (40 µg) were subjected to electrophoresis in 10% polyacrylamide gels containing SDS and electrotransferred onto nitrocellulose membranes (Macherey-Nagel, Düren, Germany). The c-Myc antibody (Ab) (sc-42X; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was used as the primary antibody. Proteins were visualized using a BCIP/NBT substrate kit (Zymed Laboratories, CA).
ChIP assays.
ChIP assays were carried out as previously described (24). Briefly, chromatin from 107 MSB-1 cells was cross-linked, washed, resuspended in lysis buffer, and sonicated 12 times for 6 s each, with an 18-W pulse (Vibra Cell 75455; Bioblock Scientific). A fraction of total chromatin was taken as the total input DNA control. The protein/DNA complexes were immunoprecipitated by incubation at 4°C overnight with a mixture of anti-c-Myc Abs (5 µg and 5 µg; sc-42X and sc-764X; Santa Cruz Biotechnology, Santa Cruz, CA) or with mouse immunoglobulin G1
(IgG1
) Abs (2 µg; Sigma-Aldrich, Lyon, France) as a negative control. Immunoprecipitated complexes were collected with protein A/protein G beads (1:1). Immunoprecipitation products were washed with low-salt washing buffer, high-salt washing buffer, lithium chloride washing buffer, and finally with TE (Tris-EDTA) buffer. Immunocomplexes were extracted, and the cross-linking was reversed by incubation at 65°C overnight. Chromatin fragments purified with a QIAquick PCR purification kit (QIAGEN, Courtaboeuf, France) were eluted in 50 µl of TE buffer. PCR was initiated by incubation at 94°C for 3 min, followed by 22 cycles of 94°C for 45 s, 58°C for 45 s, 72°C for 45 s, and 72°C for 10 min. PCRs targeting the vTR and chTR promoters were carried out in 1.1x ReddyMix PCR Master Mix (Abgene, Courtaboeuf, France) with primer pairs 804/729 and M180/M166, respectively (Table 2).
|
|
|---|
![]() View larger version (20K): [in a new window] |
FIG. 1. Telomerase activity and relative expression of telomerase subunit genes during MDV strain RB-1B (MDV-RB-1B) infection. Eight-week-old chickens were inoculated intramuscularly with 1,000 PFU of highly oncogenic MDV strain RB-1B. Telomerase activity and gene expression were analyzed on days 1, 7, 14, 21, and 27 postinfection, using PBMC from infected and control chickens. Means and standard deviations are given. Values for infected chickens significantly different from those for control chickens are indicated by an asterisk.
|
Basal transcriptional activities of viral and chicken TR promoters are essentially driven by the TATA-like box element. As vTR was the only telomerase subunit displaying high levels of expression during MDV tumorigenesis in chickens, we compared the transcriptional regulation of vTR with that of chTR. We identified several transcription factor binding sites common to the chicken and the viral TR promoter sequences and other cis elements specific to the viral promoter sequence (Fig. 2; (10). Most of the common cis regulatory elements were elements generally thought to control basal transcription, such as the TATA-like box (at 30), two CCAAT boxes (at 65 and 95), and several GC boxes that serve as consensus binding sites for Sp1 (at 42, 49, 168, and 197).
We analyzed the role of the TATA-like box and the two CCAAT boxes (CCAATx and CCAATy) in the two promoters, using site-directed mutagenesis to generate a set of mutations within the vTR and the chTR promoters. The TATA-like box (30 to 27) was replaced by a thymine hexamer in the pvTR
TATA and the pchTR
TATA constructs. Similarly, the CCAATx box (95 to 91) and the CCAATy box (65 to 61) were replaced by thymine pentamers in the pvTR
CCAATx, pchTR
CCAATx, pvTR
CCAATy, and pchTR
CCAATy constructs. We then used a luciferase assay to assess the ability of each promoter construct to drive the firefly luciferase reporter gene expression by transiently transfecting LMH and MSB-1 avian cells with the mutated constructs (Fig. 3A and B).
![]() View larger version (22K): [in a new window] |
FIG. 3. Luciferase promoter assay for vTR (A) and chTR (B) in LMH and MSB-1 cells. Wild-type and mutant vTR/chTR promoter fragments were inserted upstream from the firefly luciferase reporter gene in the pGL3-Basic vector to generate pvTR and pchTR. Schematic diagrams of pvTR and pchTR are shown at the top. Black rectangles indicate Sp1 sites. An arrow indicates the transcription start sites of the vTR and chTR genes. Firefly luciferase activity was normalized with respect to the Renilla luciferase activity obtained from the cotransfected pRL plasmid. The luciferase activities of the pvTR and pchTR wild-type plasmids were normalized to 100%. The means and standard deviations of at least three independent experiments are shown. Values for the mutated promoter significantly different from those for the corresponding wild-type promoter are indicated by an asterisk.
|
TATA and the pchTR
TATA constructs was 12% and 10%, respectively, of the activity levels observed for pvTR and pchTR in LMH cells and 16% and 24%, respectively, of the levels observed for pvTR and pchTR in MSB-1 cells. Mutations of the CCAATx and CCAATy boxes of the vTR promoter reduced promoter activity levels to 57% and 43%, respectively, of that in LMH cells and to 44% and 47%, respectively, of that in MSB-1 cells. An identical pattern was observed for the chTR promoter (Fig. 3B). The promoter activity level of the pchTR
CCAATx and pchTR
CCAATy constructs was 40% and 60%, respectively, lower than that of the wild-type chTR promoter in LMH cells and 49% and 53%, respectively, lower than that of the wild-type chTR promoter in MSB-1 cells. These results suggest that the CCAATx box and the CCAATy box are involved in vTR and chTR promoter regulation and that the TATA-like box element is critical for the basal transcription activity associated with the chicken and viral TR promoters.
E-box 3 is essential for vTR promoter activity in LMH and MSB-1 cells.
We compared the transcriptional activities of the chicken and viral TR promoter regions in LMH and MSB-1 cells, with a luciferase assay (Fig. 4A). The vTR promoter was 2.8-fold and 4.5-fold more active than the chTR promoter in LMH and MSB-1 cells, respectively. These results suggest that some cis elements present only on the viral promoter sequence may be responsible for the promoter's higher transcriptional efficiency. Several cis elements, one AP-1 site (at 527), three E-boxes (at 389, 184 and +2), and one EBS (at 38), were found only in the vTR promoter. As outlined above, AP-1 sites are potential binding sites for Meq, E-boxes are involved in the transcriptional regulation of hTERT (19), and Ets binding sites are involved in the transcriptional regulation of cell proliferation (28). These cis elements are therefore potential candidates for a major role in regulating the malignancy-promoting expression of vTR. We therefore generated a set of mutations within the vTR promoter sequence, using site-directed mutagenesis as described above (each targeted site was replaced by a thymine polymer), to enable us to identify the responsive elements involved in increasing activity. We constructed five mutated promoters from pvTR as follows: pvTR
AP1, pvTR
E1, pvTR
E2, pvTR
E3, and pvTR
EBS, with mutations in the AP-1 site (527 to 521), in E-box 1 (389 to 383), E-box 2 (184 to 178), and E-box 3 (+2 to + 7), and in the EBS (38 to 32), respectively. These mutated plasmids were used to transfect LMH and MSB-1 cells, and a luciferase assay was performed (Fig. 4B). Mutations of the AP-1 site and E-box 1 had no effect on promoter activity, whereas the mutation of E-box 3 decreased transcriptional activity by 75% and 72% with respect to the activity of pvTR in LMH and MSB-1 cells, respectively. E-box 3 is, thus, a critical cis-acting element in the transcriptional activation of vTR in both cell lines.
![]() View larger version (29K): [in a new window] |
FIG. 4. Characterization of responsive elements involved in vTR transcription. (A) Comparison of vTR and chTR promoter activities in LMH and MSB-1 cells. The luciferase activity of the pvTR plasmid is normalized to 100%. The means and standard deviations from at least three independent experiments are shown. All the values obtained for pchTR are significantly different from those obtained for pvTR (indicated by an asterisk), and the P values for nonparametric Kruskal-Wallis tests are shown. (B) Identification of the DNA elements responsible for vTR transcriptional regulation in LMH and MSB-1 cells. Schematic diagram of pvTR is shown. Five mutations were generated within the vTR promoter sequence by site-directed mutagenesis. The promoter activity of each construct was measured by the luciferase assay, normalized with respect to Renilla luciferase activity, and expressed as a value relative to pvTR (wild-type) activity. The means and standard deviations of at least three independent experiments are shown. Values for mutated promoters significantly different from those for the vTR wild-type promoter are indicated by an asterisk.
|
EBS
E3 and pvTR
E2
E3, the first with mutations within the EBS and E-box 3 and the second with mutations within E-box 2 and E-box 3, respectively (both sites were replaced by a thymine polymer), to determine whether the EBS and E-box 2 act in synergy with E-box 3 in MSB-1 cells. These plasmids were used in luciferase assays with the MSB-1 cell line (Fig. 4B). The pvTR
EBS
E3 and pvTR
E2
E3 constructs reduced transcriptional activity levels by 93% and 84%, respectively, whereas mutations of the EBS, E-box 2, or E-box 3 alone led to 82%, 76%, and 72% decreases in activity levels, respectively. These results suggest that the EBS and E-box 2 cooperate with E-box 3 to regulate the promoter activity of vTR in the MSB-1 cell line.
Sp1 binding sites are involved in the transcriptional regulation of vTR.
Our results show that the four cis elements of the vTR promoter functional in the two cell types tested (the CCAATx and CCAATy boxes, the TATA-like box, and E-box 3) are located close to the transcriptional initiation site (between 95 and +7). Moreover, the EBS, which is specifically functional in MSB-1 cells, is also located in this promoter region. Three of the eight Sp1 sites identified in the vTR promoter sequence are also located in this region. We investigated whether these Sp1 sites were involved in the transcriptional regulation of vTR by studying this proximal promoter region using a short vTR promoter comprising the 104 to +20 region of the promoter sequence (pvS-TR). The transcriptional efficiency of pvS-TR was then compared with that of an equivalent promoter region in which the three Sp1 sites had been replaced by a thymine polymer (pvS-TR
Sp1). The transcriptional activity of the truncated pvS-TR construct was not significantly different from that of the pvTR construct in LMH cells (P = 0.1432) (Fig. 5A). The 56% lower level of transcriptional activity observed for the truncated promoter, pvS-TR, than for the entire promoter, pvTR, in MSB-1 cells seems to be due to the absence of E-box 2 from this construct. In both types of cells, mutation of the three Sp1 sites led to levels of transcriptional activities at least 81% lower than those for pvS-TR (Fig. 5B). This result suggests that the three Sp1 sites, located at 42, 49, and 102, are involved in the transcriptional regulation of vTR.
![]() View larger version (21K): [in a new window] |
FIG. 5. Functional study of Sp1 sites located close to the transcriptional start site in LMH and MSB-1 cells. (A) Comparison of pvTR and pvS-TR activities in LMH and MSB-1 cells. Schematic diagram of pvS-TR is shown. The promoter activity of the pvS-TR construct was assessed with the luciferase assay, normalized with respect to Renilla luciferase activity, and expressed relative to the activity of pvTR (the entire vTR promoter). The means and standard deviations of at least three independent experiments are shown. Values for pvS-TR significantly different from those for pvTR (indicated by an asterisk) and P values for nonparametric Kruskal-Wallis tests are shown. (B) Involvement of Sp1 sites in vTR transcriptional regulation. The promoter activity of pvS-TR Sp1 was assessed with the luciferase assay, normalized with respect to Renilla luciferase activity, and expressed relative to the activity of pvS-TR (wild type). The means and standard deviations of at least three independent experiments are shown. Values for mutated promoters significantly different from those for the short vTR wild-type promoter are indicated by an asterisk.
|
E3, or the pchTR plasmid carrying the wild-type vTR promoter sequence, the corresponding E-box 3 mutant, and the E-box-free chTR promoter sequence, respectively. Western blotting analysis confirmed that there was an effective ectopic expression of c-Myc protein in c-Myc-cotransfected cells (Fig. 6A). The overexpression of c-Myc protein increased wild-type vTR promoter activity by a factor of 6.5 in LMH cells (Fig. 6B). Reporter plasmids carrying the E-box 3-mutated vTR promoter and pchTR responded similarly to c-Myc overexpression, with a response that was significantly different from that obtained for the wild-type vTR promoter (P = 0.0027 and P = 0.0201, respectively). These results indicate that c-Myc transactivates vTR and that E-box 3 is involved in this activity.
![]() View larger version (19K): [in a new window] |
FIG. 6. Effect of c-Myc overexpression on vTR promoter activity in LMH cells. (A) Effective overexpression of c-Myc in LMH cells was checked by Western blotting. (B) Effect of c-Myc overexpression on the transcriptional efficiency of the vTR promoter. LMH cells were cotransfected with the c-Myc expression vector (pCDNA3Myc) and reporter plasmids (pvTR, pvTR E3 or pchTR) using Lipofectamine 2000. Each transfection was carried out at least in triplicate and standard deviation bars are shown. The promoter activity of each construct was compared with that of the wild-type vTR promoter, which was defined as 100. Assays with a significantly different value for cotransfection with pvTR and pCDNA3Myc are indicated by an asterisk, and the corresponding P values from nonparametric Kruskal-Wallis tests are shown.
|
E3 displayed minimal responses to c-Myc overexpression, we assessed the specific and direct association of c-Myc with the vTR promoter by carrying out ChIP experiments with MSB-1 cells. We analyzed immunoprecipitated DNA by PCR, using primer pairs for the vTR and chTR promoters. We found that c-Myc was recruited to the vTR promoter in MSB-1 cells, whereas this factor did not bind to the chTR promoter, which contains no E-box elements (Fig. 7). These results suggest that the endogenously expressed c-Myc protein mediates the transcriptional regulation of vTR by interacting with the region encompassing the functional cis elements.
![]() View larger version (28K): [in a new window] |
FIG. 7. Direct association of c-Myc with the vTR promoter. Schematic diagrams of the vTR promoter (left) and the chTR promoter (right) are shown at the top. The pairs of primers used for the PCR amplification are indicated on the schematic diagrams of the two promoters. Electrophoretic analysis of the PCR amplification products, using immunoprecipitated DNA (IP) or total input DNA (Input), is shown below. Mouse IgG1 antibodies (right column) were used in ChIP assays as a negative control. The column on the left shows the PCR amplification products from DNA immunoprecipitated with c-Myc antibodies. vTR promoter-specific primers amplified a 189-bp product, whereas chTR promoter-specific primers amplified a 279-bp product.
|
|
|
|---|
We found that vTR expression was significant on day 7 postinfection and that it increased during the latent period and lymphoma development (Fig. 1). Consistent with the results of a previous study (30), an analysis of endogenous vTR expression in cells revealed that vTR was expressed in MDV-infected primary chicken embryo cells and in MDV-transformed lymphoblastoid cell lines (MSB-1, MDCC-PA9, and MDCC-PA5) (data not shown). The vTR gene therefore seems to be both a lytic and a latent gene, as it is expressed during both of these phases of MDV infection. However, semiquantitative RT-PCR analysis revealed that vTR levels were up to 17-fold higher in T cells sustaining latent infection (MSB-1, MDCC-PA9, and MDCC-PA5) than in permissive cells undergoing viral replication (MDV-infected chicken embryo cells) (data not shown). These results are of particular interest given the functionality of vTR. Indeed, Trapp et al. (30) have shown that the tumor-promoting effects of vTR expression cannot be attributed to a functional role in MDV replication or in the establishment or reactivation from latency. Instead, these effects seem to involve events downstream from the primary establishment of infection. Thus, vTR seems to display its oncogenic properties only during latent infection in lymphocytes, which sustain high levels of vTR expression.
In this study, we identified several cis elements specifically involved in vTR transcriptional regulation. Comparative analysis of the vTR and chTR promoters showed that the transcriptional efficiency of the viral promoter is up to fourfold higher than that of the chTR promoter. Our analysis of the functional significance of the TATA-like box and the two CCAAT boxes, using luciferase assays in two avian cell lines, showed that basal transcriptional regulation was similar for the two promoters. Indeed, the two CCAAT boxes had equivalent functions, and the TATA-like box was found to be essential for the basal transcriptional activity of both promoters (Fig. 3). Thus, as previously described for the hTR promoter (36, 37), the vTR and chTR promoter sequences have cis regulatory elements in common, and these elements are involved in maintaining basal TR expression levels. However, consistent with our comparative analysis of the activities of the vTR and the chTR promoters, quantification of the endogenous levels of vTR and chTR expression in MSB-1 cells by semiquantitative RT-PCR demonstrated that vTR expression was 14-fold stronger than chTR expression in these MDV-transformed T cells (data not shown). The functionality of E-box 3, located 2 bp downstream from the transcription start site, in both cell types studied suggests that this cis element is the principal element responsible for the higher levels of activity observed for the vTR promoter. A similar canonical E-box element, located 22 bp downstream from the transcription initiation site, is involved in hTERT transcriptional regulation (16, 29). Moreover, as described for the hTERT promoter (19), Sp1 binding sites located near E-box 3 are involved in vTR transcriptional activation. Thus, the vTR and hTERT promoter sequences include similar cis regulatory elements, and these elements are the main elements controlling vTR and hTERT expression in cells. Our findings indicate that vTR transcription is regulated by responsive elements involved in the regulation of basal transcription, as is the case for the chTR and hTR promoters, and that the E-box 3 cis element is required for high levels of transcription, as previously reported for hTERT. The mutagenesis assays carried out with LMH and MSB-1 cell lines identified no negative cis elements in the vTR promoter sequence. Thus, the vTR promoter seems to be the homolog of the original avian cellular chTR promoter but with certain responsive elements related to oncogenesis, particularly the E-box located in the transcribed region of vTR, introduced into the viral genome during the course of evolution, resulting in high levels of vTR expression during infection.
E-boxes serve as binding sites for proteins of the bHLH-Zip transcription factor superfamily, including the Myc/Mad/Max family of transcription factors (http://www.myccancergene.org/) (7). bHLH-Zip proteins and their E-box recognition sites act as crucial positive regulators of diverse biological processes, including cell proliferation and death (12). Previous studies have shown that c-Myc can induce telomerase activity through the transcriptional activation of hTERT (13, 32). Our overexpression assays, in which the reporter plasmid carrying the wild-type vTR promoter sequence was used for cotransfection with pCDNA3Myc, showed that c-Myc activates transcription of the vTR gene. Moreover, our ChIP assays showed that endogenously expressed c-Myc binds to the vTR promoter sequence in an MDV-transformed cell line. Thus, the transactivation of vTR expression induced by the interaction of c-Myc with the E-boxes located within the vTR promoter sequence is involved in the high levels of vTR expression observed during MDV-induced lymphomagenesis. The c-Myc oncoprotein is also involved in the increase in telomerase activity due to HPV16 infection. Indeed, the E6 oncoprotein encoded by HPV16 has been shown to interact directly with c-Myc, and c-Myc/E6 complexes have been shown to activate hTERT expression, increasing telomerase activity in infected tumor cells (31). The c-Myc oncoprotein therefore seems to be a common element of the pathways leading to higher levels of telomerase activity in MDV- and HPV16-transformed tumor cells.
In this study, we have shown that E-box 2 and the EBS, located 184 bp and 38 bp upstream, respectively, from the transcription initiation site of vTR are specifically functional in MSB-1 cells but not in LMH cells (Fig. 4B). This finding is of particular interest because the MSB-1 cell line is a T-lymphoblastoid line transformed by MDV, whereas the epithelioid LMH cell line represents a cell population which is not susceptible to transformation during MDV infection. This result therefore provides evidence that vTR is specifically regulated in transformed T cells. A previous study describing the crystal structures of c-Myc-Max heterodimers selectively bound to duplex oligonucleotides containing E-boxes revealed that two c-Myc-Max heterodimers tightly associate to form a bivalent heterotetramer, providing a substantial platform for the assembly of an additional protein factor (25). As the EBS and E-box 2 act together with E-box 3 to regulate vTR expression (Fig. 4B), an interaction between the EBS-bound factor and a Myc-Max heterotetramer bound to E-boxes 2 and 3 may be involved in the specific transcriptional regulation of vTR observed for MSB-1 cells. Moreover, as the functional EBS is located between the E-boxes, the formation of a specific chromatin structure should involve the specific location of those cis elements on the promoter sequence. Thereby, the assembly of a large nuclear complex on vTR promoter, particularly in MDV-transformed lymphoblastoid cells, may account for the restricted, cell type-dependent activity of the vTR promoter. Tess software (www.cbil.upenn.edu/cgi-bin/tess/tess) analysis of the vTR promoter sequence revealed that the EBS contains specific binding domains for three Ets proteins, PU.1, Ets-1, and PEA3. ChIP assays using a rabbit polyclonal antibody against the chicken PU.1 protein (Santa Cruz Biotechnology, CA) (23) failed to precipitate the vTR promoter sequence (data not shown). A previous study showed that PEA3 is primarily expressed in epithelial cells and not in hematopoietic cells (34) and that high levels of Ets-1 expression are restricted to lymphoid tissues in adults. Ets-1 may therefore be the factor responsible for driving vTR transcriptional regulation through the EBS located within its promoter sequence. Future studies should characterize the interaction between bHLH-Zip transcription factors and Ets proteins leading to the specific coregulation of vTR in MDV-transformed lymphoblastoid cells.
Finally, our study suggests that the viral oncoprotein Meq has no direct effect on vTR transcriptional regulation. Meq has a structure similar to that of the Jun/Fos family oncogenes and is a transactivator that can heterodimerize with c-Jun and bind AP1 sites (20). In our luciferase assays, mutations in the AP-1 site located in the vTR promoter sequence had no effect on the transcriptional activity of the vTR promoter in Meq-expressing MSB-1 cells (Fig. 4B). This result is consistent with a previous study in which a ChIP-based approach scanning the entire MDV genome for Meq binding sites failed to identify the vTR promoter sequence as a target binding site for Meq (20). Thus, although vTR and Meq may cooperate during lymphoma formation and dissemination, the oncogenic properties of Meq required for MDV-induced lymphomagenesis do not seem to involve the direct regulation of vTR transcription by Meq. Our results therefore identify EBS factors and the oncoprotein c-Myc as the main elements involved in the high levels of vTR expression specific to the MDV-transformed T-lymphoblastoid cell line. Indeed, the binding of a factor to the EBS in the vTR promoter specifically in lymphoblastoid cells may induce Myc-Max heterotetramer complex formation, leading to high levels of vTR expression. Thus, high levels of vTR expression in these cells, due to cooperation between factors bound to the TATA-like box, E-box, and the EBS, may be an essential prerequisite for the full expression of the oncogenic properties of vTR. Consistent with this idea, the constitutive expression of vTR in the nontransformed chicken cell line DF-1 resulted in a phenotype consistent with transformation (30). Thus, the tumor-promoting effects of vTR must involve its specific overexpression in lymphoblastoid cells during the latent period and the transcriptional regulation of vTR seems to be essential for the expression of its oncogenic properties.
This work was supported by grants from the Ligue Nationale Contre Le Cancer-Comité du Cher and from the Région Centre.
Published ahead of print on 21 February 2007. ![]()
Present address: Department of Medicine, Division of Hematology, and Cancer Biology Program, Stanford School of Medicine, Stanford, CA 94305. ![]()
|
|
|---|
-mediated viral inhibition. Virology 277:58-65.[CrossRef][Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»