|
|
||||||||
Department of Tumor Virology, Institute for Genetic Medicine, Hokkaido University, Sapporo, Japan
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Various reagents have been found to induce virus activation in latently EBV-infected cells. They include halogenated pyrimidines (2, 3), phorbol esters (4), anti-Ig Abs (5, 6), and sodium butyrate (7). Although we do not know the physiological stimuli that control activation of the productive cycle and the switch from latency in vivo, B cell Ag receptor (BCR) signaling, which is mimicked by anti-Ig treatment, would serve as a more physiologically relevant activator.
BCR signaling is initiated upon binding of an Ag to membrane Ig, which induces receptor aggregation and subsequent activation of the src family tyrosine kinases such as Lyn, Fyn, and Syk. Activation of these tyrosine kinases results in activation of phospholipase C-
2 (PLC-
2). PLC-
2 hydrolyzes phosphatidylinositol diphosphate (PIP2) to diacylglycerol and inositol triphosphate (IP3), which in turn induce activation of protein kinase C (PKC) and increase intracellular Ca2+, respectively. Another key pathway activated after BCR cross-linking involves phosphatidylinositol 3-kinase (PI3K) and mitogen-activated protein kinases (MAPKs). These signals are central to B cell development and the response to the Ag (8, 9).
Anti-Ig Abs efficiently induce lytic virus replication in some BL cell lines, including Akata (6, 10, 11), which expresses a limited number of EBV latent gene products, including EBV-determined nuclear Ag (EBNA) 1, two EBV-encoded small RNAs known as EBV-encoded small RNA1 and EBV-encoded small RNA2, the rightward transcripts from the BamHI A region, and a very small amount of latent membrane protein 2A (LMP2A) (termed type I latency) (12, 13). LMP2A, which is expressed in a high amount in EBV-immortalized lymphoblastoid cell lines (LCLs), is known to act as a dominant-negative regulator of Lyn and Fyn kinase activities, and interferes with EBV activation after BCR cross-linking (14, 15). Therefore, low LMP2A expression is important for efficient EBV activation in anti-Ig-treated cells (16). However, many BL-derived cell lines with type I latency are nonresponsive or low-responsive to EBV activation by BCR cross-linking despite no or low LMP2A expression.
The present study aimed to identify signaling pathways that determine the sensitivity to BCR-mediated EBV activation. The results indicate that the absence of PI3K impedes the progression of signals through the BCR and becomes a determinant of unresponsiveness to BCR-mediated EBV activation.
| Materials and Methods |
|---|
|
|
|---|
All cells were maintained in RPMI 1640 medium (Sigma-Aldrich, St. Louis, MO) supplemented with 10% FBS (Invitrogen, Groningen, The Netherlands), penicillin (40 U/ml), and streptomycin (50 µg/ml). Akata (11), Mutu-I (17), Daudi (18), Sav-I, Kem-I, and Oku-I are EBV-positive BL cell lines. EBV-negative Daudi cell clones were isolated from parental Daudi cells by the limiting dilution method (19).
Reagents
F(ab')2 fragments of rabbit anti-human IgM and IgG polyclonal Abs (DAKO, Copenhagen, Denmark) were used for BCR cross-linking. Abs usedfor immunoblot analysis of BCR signaling were polyclonal rabbit Abs against phospho-extracellular signal-regulated kinase (ERK) MAPK, phospho-p38 MAPK (BioSource International, Camarillo, CA), ERK, Akt, and phospho-Akt (Cell Signaling Technology, Beverly, MA), and mouse mAbs against p38 MAPK (BD Transduction Laboratories, Lexington, KY), and phospho-tyrosine (p-Tyr-100) (Cell Signaling Technology). Mouse mAbs against
-actin (Sigma-Aldrich) and EBV BZLF1 protein (DAKO) were used for analysis of EBV activation. A specific inhibitor of PI3K, wortmannin (Sigma-Aldrich), and recombinant human insulin-like growth factor 1 (rIGF-1) (R&D Systems, Minneapolis, MN) were used for inhibition and activation of PI3K, respectively. FITC-conjugated anti-human CD19 mouse mAb (Beckman Coulter, Brea, CA) was used for flow cytometry.
Immunoblot analysis
Cells (1.5 x 106 cells per sample) were treated with anti-IgG or IgM Abs (7.5 µg/ml) for 24 h. Then they were lysed in 200 µl lysis buffer (50 mM Tris-HCL (pH 7.5), 150 mM NaCl, 50 mM NaF, 5 mM EDTA, 1% (v/v) Triton X-100, 10% glycerol, 1 mM Na3VO4, 1 mM PMSF, 2 µg/ml pepstatin, 2 µg/ml aprotinin, and 2 µg/ml leupeptin). Lysates were subsequently denatured, resolved by 810% SDS-PAGE, and electrotransferred to nitrocellulose membranes. Membranes were reacted with the primary Abs; anti-phospho ERK and anti-phospho-p38 (0.5 µg/ml), anti-ERK (1:2,000), anti-p38 MAPK (1:2,500), anti-Akt and anti-phospho-Akt (1:1,000), and anti-BZLF1 (1:40), followed by treatment with a HRP-conjugated anti-rabbit or anti-mouse IgG Ab (1:2,0001:10,000) (Amersham Pharmacia Biotech, Piscataway, NJ). Membranes were visualized with an ECL kit (Amersham Pharmacia Biotech).
Immunofluorescence assay
Cells were smeared on a glass slide and fixed in acetone for 10 min. Fixed cells were incubated with an anti-BHRF1 mouse mAb (C8441) (dilution at 1:20) at 37°C for 1h. After washing with PBS, cells were stained with an FITC-conjugated anti-mouse IgG Ab (1:200) (DAKO) for 30 min followed by washing with PBS.
RT-PCR analysis
RT-PCR was conducted as described previously (19). Primer pairs used for detection of B cell adaptor for phosphatydylinositol 3-kinase (BCAP) were 5'-CAACATGCTCAATTCCCGATC-3' and 5'-CTGCCACCTTTGTTGAGATGC-3'.
Reporter plasmid construction
The promoter region of the BZLF1 gene from -222 to + 8 was amplified by PCR using the primers; 5'-CGAACGCGTGCCATGCATATTTCAAC-3' and 5'-GGAGAGTCTGGTGCAATGTTTAGTG-3'. The PCR product was digested with MluI and BglII and then subcloned upstream of the luciferase gene of PGV-basic vector 2 (PGV-B2; Wako, Osaka, Japan).
Transfections and promoter analysis
Transfections were performed using lipofectamine reagent (Invitrogen) according to the manufacturers protocol. Transfected cells were cultured in 6-well plates for 24 h and further incubated for 24 h with addition of anti-Ig Abs and/or other reagents. Luciferase activities were determined using the dual-luciferase reporter assay system (Promega). To normalize luciferase activities, activities of the PGV reporter plasmid were divided by those of the pRL vector driven by the thymidine kinase gene promoter of HSV, which was cotransfected as an internal control. Reproducibility of results was confirmed by three independent transfections and each transfection was done in duplicate. Values were expressed as the means ± SE of three experiments.
| Results |
|---|
|
|
|---|
We studied six type I BL cell lines, Akata, Daudi, Mutu-I, Sav-I, Kem-I, and Oku-I. First, we examined the expression of membrane Ig on these cells by flow cytometry. The results indicated that the Akata cell line was positive for IgG, and other cell lines were positive for IgM. The level of Ig expression was not different among cell lines (data not shown). Then cells were incubated in a medium containing 7.5 µg/ml of anti-IgG (for Akata cells) or IgM (for Daudi, Mutu-I, Sav-I, Kem-I, and Oku-I cells) Abs for 24 h, and EBV activation was assessed by expression of an EBV early protein, BHRF1, by the indirect immunofluorescence assay. As shown in Table I and Fig. 1, anti-Ig treatment induced BHRF1 in 66% of Akata cells, in
4% of Mutu-I and Sav-I cells, and in none of the Daudi, Kem-I, and Oku-I cells. Spontaneous BHRF1 expression without anti-Ig treatment was < 0.1% in all cell lines. Thus, we selected Akata, Mutu-I, and Daudi cells to study biochemical differences that determine the response to BCR-mediated EBV activation.
|
|
We examined activation of tyrosine kinases in anti-Ig-treated Akata and Daudi cells. Immunoblot analysis with phosphorylated tyrosine-specific Abs demonstrated the rapid appearance of phosphorylated tyrosines, peaking 5 min after addition of anti-Ig, to a similar degree in both cell lines (Fig. 2). A similar activation of tyrosine kinases was also observed in anti-Ig-treated Mutu-I cells (data not shown).
|
The three main signal transduction pathways that have been established from the BCR in response to cross-linking with anti-Ig Abs are illustrated in Fig. 3A (8, 9). We studied whether these three pathways were activated after BCR cross-linking. As shown in Fig. 3B, immunoblot analysis with phosphorylated protein-specific Abs for Akt, ERK, and p38 MAPK, which are all known targets of BCR signal transduction, demonstrated the appearance of phosphorylated forms in anti-Ig-treated Akata and Mutu-I cells. In contrast, in anti-Ig-treated Daudi cells, phospho-Akt was not detected, phospho-ERK was faintly detected, and p38 MAPK was constitutively phosphorylated irrespective of anti-Ig treatment.
|
Because BCR-mediated signaling did not induce the PI3K pathway in Daudi cells, we attempted to activate the PI3K pathway by IGF-1, which activated the PI3K pathway independent of signaling through BCR (20). As shown in Fig. 4A, IGF-1 alone had a substantial effect on the phosphorylation of Akt, and simultaneous treatment with IGF-1 and anti-Ig Abs greatly increased the phosphorylation of Akt. The immunofluorescence assay indicated that IGF-1 alone could not induce BHRF1, but the simultaneous treatment with IGF-1 and anti-Ig Abs induced BHRF1 in 12% of Daudi cells (Table I, Fig. 4B). EBV activation by the simultaneous treatment was further confirmed by detection of an EBV lytic protein, BZLF1 (13), by immunoblot analysis (Fig. 4C).
|
To identify the BCR-associated PI3K signaling defect in Daudi cells, possible defects in the PI3K adaptors, CD19 and BCAP, were examined. As shown in Fig. 5, CD19 and BCAP were equally expressed in both Akata and Daudi cells.
|
Although three main pathways involved in BCR-mediated signal transduction were impaired in Daudi cells, restoration of the PI3K pathway by IGF-1 made Daudi cells responsive to BCR-mediated EBV activation. Therefore, we studied whether the other two signaling pathways, ERK and p38 MAPK, were also restored by IGF-1 treatment in Daudi cells. As shown in Fig. 6, IGF-1 alone had a little or no effect on the phosphorylation of ERK and p38 MAPK, but the simultaneous treatment with IGF-1 and anti-Ig Abs greatly increased their phosphorylation in Daudi cells, suggesting that PI3K influenced the activation of ERK and p38 MAPK pathways.
|
Studies on Daudi cells suggested that PI3K was critical for BCR-mediated EBV activation. Therefore, we examined the effect of a PI3K inhibitor on BCR-mediated EBV activation in Akata and Mutu-I cells. The cells were treated with anti-Ig Abs in the presence of wortmannin (0.2 µM), a PI3K inhibitor, for 24 h. Wortmannin treatment did not give any effect on cell viability. Immunoblot analysis indicated that wortmannin treatment inhibited the phosphorylation of Akt by anti-Ig Abs (Fig. 7A). Further studies on EBV expression indicated that expression of BHRF1 and BZLF1 was also inhibited (Fig. 7, B and C), suggesting the importance of PI3K in BCR-mediated EBV activation.
|
BZLF1 is the switch for EBV activation from a latent infection and the first-transcribed EBV gene after BCR cross-linking (21, 22). We studied whether PI3K activates the transcription from the BZLF1 promoter in the absence of other EBV products. A BZLF1 promoter-luciferase gene plasmid was transfected into EBV negative-Daudi cells, and cells were then treated with IGF-1 and anti-Ig Abs for 24 h. As shown in Fig. 8, anti-Ig Abs alone slightly up-regulated luciferase expression from the BZLF1 promoter, and dual treatment with IGF-1 and anti-Ig Abs greatly increased the luciferase activity. This increase by the dual treatment was greatly reduced by addition of wortmannin. These results suggested that the BZLF promoter was a target of BCR signaling, and that PI3K was important for BCR-mediated BZLF1 activation.
|
| Discussion |
|---|
|
|
|---|
Although BL cells with type I latency are free from inhibition of signal transduction by LMP2A, many BL cell lines are not susceptible for EBV activation by anti-Ig Abs. The present study was aimed at defining the molecular mechanisms that determine the response to BCR-mediated EBV activation. The comparison of Akata, Mutu-I, and Daudi cells, which are representative responders and nonresponders to BCR-mediated EBV activation, respectively, indicated that three main signaling pathways, PI3K, ERK, and p38 MAPK, were impaired in Daudi cells, and that restoration of PI3K restored ERK and p38 MAPK pathways, which was accompanied by EBV activation. Furthermore, a specific inhibitor for PI3K, wortmannin, inhibited EBV activation in anti-Ig-treated Akata and Mutu-I cells. These results indicated that the absence of PI3K impedes the progression of signals through the BCR, acting as a determinant of unresponsiveness to BCR-mediated EBV activation.
However, the degree of EBV activation after restoration of PI3K was not striking in all BL cell lines examined as compared with that in Akata cells. In particular, Oku-I cells were completely unresponsive to EBV activation after dual treatment with IGF-1 and anti-Ig Abs. These results suggested the existence of other cellular or viral factors influencing BCR-mediated EBV activation.
The present results also indicated the existence of cross-talk among the three signaling pathways, PI3K, ERK, and p38 MAPK, because restoration of PI3K by IGF-1, which activated PI3K independent of BCR-mediated signaling, was accompanied by restoration of ERK and p38 MAPK activation in anti-Ig-treated Daudi cells. This is consistent with recent reports that PI3K activates the phosphorylation of ERK and p38 MAPK (28, 29).
We have previously shown that a specific inhibitor for PKC and a calmodulin antagonist inhibit BCR-mediated EBV activation in Akata cells (30), suggesting the requirement of PKC and Ca2+/calmodulin pathways for BCR-mediated EBV activation. Other reports indicated that inhibitors for MAPK (ERK and p38) blocked BCR-mediated EBV activation in Akata cells (31, 32). More recently, Darr et al. (33) reported that an inhibitor of PI3K blocked anti-Ig-induced EBV activation in Akata cells. These results suggest that all signaling pathways, PKC, Ca2+/calmodulin, PI3K, and MAPK (ERK and p38), must be activated for efficient EBV activation by BCR cross-linking.
Another aspect that may influence the response to BCR cross-linking is the Ig class; Akata cells express membrane IgG, and other cells express IgM. A recent report has described that IgG-containing BCR transmits a signal distinct from that of IgM- or IgD-containing BCR, although all three use the same signal-transduction component (34).
In conclusion, we demonstrated that PI3K is important for BCR-mediated EBV activation in BL cells that may represent in vivo latency as discussed above. Clarification of the mechanism, which regulates PI3K activity, would provide an insight into the molecular basis of maintenance and disruption of EBV latency in B-lymphocytes.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Kenzo Takada, Department of Tumor Virology, Institute for Genetic Medicine, N15 W7, Kita-ku, Sapporo 060-0815, Japan. E-mail address: kentaka{at}med.hokudai.ac.jp ![]()
3 Abbreviations used in this paper: BL, Burkitts lymphoma; BCR, B cell Ag receptor; PLC-
2, phospholipase C-
2; PKC, protein kinase C; PI3K, phosphatidylinositol 3-kinase; MAPK, mitogen-activated protein kinase; EBNA, EBV-determined nuclear antigen; LMP2A, latent membrane protein 2A; LCL, lymphoblastoid cell line; ERK, extracellular signal-regulated kinase; IGF-1, insulin-like growth factor 1; BCAP, B-cell adaptor for phosphatydylinositol 3-kinase. ![]()
Received for publication May 2, 2003. Accepted for publication November 11, 2003.
| References |
|---|
|
|
|---|
specificity on a Burkitt lymphoma cell in vivo and in derived culture lines. Cancer Res. 28:1300.
-induced apoptosis in Burkitts lymphoma. EMBO J. 21:954.[Medline]
This article has been cited by other articles:
![]() |
N. Dussault, E. Ducas, C. Racine, A. Jacques, I. Pare, S. Cote, and S. Neron Immunomodulation of human B cells following treatment with intravenous immunoglobulins involves increased phosphorylation of extracellular signal-regulated kinases 1 and 2 Int. Immunol., August 8, 2008; (2008) dxn090v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Wen, D. Iwakiri, K. Yamamoto, S. Maruo, T. Kanda, and K. Takada Epstein-Barr Virus BZLF1 Gene, a Switch from Latency to Lytic Infection, Is Expressed as an Immediate-Early Gene after Primary Infection of B Lymphocytes J. Virol., January 15, 2007; 81(2): 1037 - 1042. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Emura, D. Saito, T. Kakizoe, and M. J. Blaser Infection, Cancer and Prevention: Report of the 19th International Symposium of the Foundation for Promotion of Cancer Research Jpn. J. Clin. Oncol., November 1, 2006; 36(11): 745 - 755. [Full Text] [PDF] |
||||
![]() |
J. Yuan, E. Cahir-McFarland, B. Zhao, and E. Kieff Virus and Cell RNAs Expressed during Epstein-Barr Virus Replication J. Virol., March 1, 2006; 80(5): 2548 - 2565. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Huang, G. Liao, H. Chen, F. Y. Wu, L. Hutt-Fletcher, G. S. Hayward, and S. D. Hayward Contribution of C/EBP Proteins to Epstein-Barr Virus Lytic Gene Expression and Replication in Epithelial Cells J. Virol., February 1, 2006; 80(3): 1098 - 1109. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Daibata, K. Bandobashi, M. Kuroda, S. Imai, I. Miyoshi, and H. Taguchi Induction of Lytic Epstein-Barr Virus (EBV) Infection by Synergistic Action of Rituximab and Dexamethasone Renders EBV-Positive Lymphoma Cells More Susceptible to Ganciclovir Cytotoxicity In Vitro and In Vivo J. Virol., May 1, 2005; 79(9): 5875 - 5879. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |