Epstein-Barr virus microRNAs repress BCL6 expression in diffuse large B-cell lymphoma

Article metrics

Diffuse large B-cell lymphoma (DLBCL) is the most common form of lymphoma, accounting for 30–40% of all newly diagnosed lymphomas. DLBCL is considered a heterogeneous disease, with some specific clinicopathological variants of DLBCLs being associated with the presence of the EBV.1 EBV is a lymphotropic virus that has been implicated in the development of several lymphoid malignancies, mainly Burkitt lymphoma (BL) and Hodgkin's lymphoma and with low prevalence in DLBCL.1 BCL6 is a key transcriptional repressor during normal B-cell differentiation that has been shown to repress NF-kB in some DLBCLs.2 In some B-cell lymphomas, BCL6 expression was inversely correlated with LMP1 expression, and some evidences suggest that LMP1 can cause downregulation of BCL6(ref. 3), but other possible mechanisms have not been studied. We have found a strong inverse correlation between BCL6 protein expression and EBV infection (P<0.001, Figures 1a and b) in a series of 149 DLBCL samples, where only one out of 34 EBV-positive cases (2.94%) expressed BCL6, although 87 out of 115 EBV-negative cases expressed BCL6 (75.65%). However, this correlation was independent of LMP1 because 54% of EBER-positive samples were LMP1-negative (Spearman, P-value: 0.18). Little is known about the mechanisms that cause the absence of BCL6 in EBV-positive DLBCL; however, the possibility that EBV-encoded miRNAs could contribute to BCL6 repression has never been explored.

Figure 1
figure1

(a) Immunohistochemical staining of two cases of DLBCL showing an opposite pattern of BCL6 expression and EBV presence. (b) Graphic representation and statistical analysis by χ2-test of the immunohistochemical results in a series of 149 DLBCL patients. (c) Gene clustering reveals the pattern of expression of EBV-encoded miRNAs in EBV-positive DLBCL patients. (d) RT–PCR was carried out on 40 DLBCL samples using probes for five miRNAs. Black bars indicate standard deviation. (e) RT–PCR analysis of five EBV-encoded miRNAs in DLBCL and BL cell lines. Expression data is represented in a logarithmic scale and has been normalized using the Ramos EBV-negative cell line as the background signal. (f) Levels of BCL6 protein expression in the same cell lines show an inverse correlation with the levels of EBV-encoded miRNAs.

We have studied the miRNA expression profile in 36 DLBCL samples using miRNA microarrays. Prevalence of EBV in this series (33 EBV-negative and 3 EBV-positive) was close to normal prevalence in DLBCL.1 Twenty-two out of thirty-two analysed viral miRNAs were significantly upregulated in EBV-positive samples (>two-fold change; t-test corrected P-value<0.05, Figure 1c, Supplementary table 2 and Supplementary table 3). None of the miRNAs belonged to the BHRF1 cluster, a finding that confirms previous results suggesting that the highest levels of these miRNAs are reached during the lytic cycle.4 To test the hypothesis that some of these miRNAs can potentially downregulate BCL6, we made a bioinformatic prediction using the miRanda algorithm. This prediction indentified 21 binding sites in the 3′ UTR of BCL6 for 18 EBV-encoded miRNAs (Supplementary table 1). More than 70% of these predictions were also found with other algorithms (Supplementary Figure 1). Ten of these miRNAs were also differentially expressed between EBV-positive and EBV-negative DLBCL cases. This high proportion of miRNAs potentially targeting BCL6 suggests a physiological mechanism of the virus to reduce the amount of BCL6. Consequently, we selected four EBV-encoded miRNAs for further validation. These miRNAs were selected according to the following parameters: 1) attainment of a high score from miRanda and 2) differential expression in the EBV-positive DLBCL cases. In this way, we selected ebv-miR-BART3, ebv-miR-BART9 and ebv-miR-BART17-5p. We also selected ebv-miR-BART7 because, although it does not achieve one of the highest scores from miRanda, the program predicts two binding sites for the miRNA in the 3′ UTR of BCL6, increasing the probability of it being a bona fide regulator of BCL6 (Supplementary Figure 2). To analyse these miRNAs in a larger series of cases, we selected 40 cases (15 EBV-positive and 25 EBV-negative) from the 149-patients series to check their expression by RT–PCR. The mean of the viral miRNAs expression among EBER-positive patients was 10- to 100-fold higher compared with EBER-negative patients (Figure 1d). We also analysed EBV-encoded miRNA levels by RT–PCR in BL and DLBCL-derived cell lines and observed a correlation between BCL6 protein and miRNA levels (Figures 1e and f). These results indicate that a relatively low level of expression of these miRNAs could enable the high level of expression of BCL6 observed in EBV-positive BL,5 thus explaining the differences between BL and DLBCL.

Therefore, to validate the putative BCL6-regulatory role of these miRNAs, we transfected synthetic miRNAs and measured the luciferase activity in a reporter system in which we cloned the 3′UTR of BCL6 (Supplementary Materials & methods). With this approach, we noted a significant reduction in luciferase signal for three of the four viral miRNAs assayed (Figure 2a) when compared with either the reporter transfected with a miRNA mimic negative control sequence or the vector without the 3′ UTR of BCL6. Ebv-miR-BART3, ebv-miR-BART9 and ebv-miR-BART17-5p reduced the luciferase signal by at least 60%. The effect of these miRNAs on the endogenous BCL6 protein was investigated in lymphoid EBV-negative BCL6-expressing cell lines. We used three DLBCL-derived cell lines (DB, SU-DHL-4 and SU-DHL-6) and one BL-derived cell line (Ramos) to test the effect of these miRNAs in different lymphoid models. In DLBCL-derived cell lines transfection of the viral miRNAs ebv-miR-BART3, ebv-miR-BART9 and ebv-miR-BART17-5p led to a reduction in the levels of BCL6 protein ranging from 25 to 72%, 48 h after transfection (Figure 2b). In contrast, ebv-miR-BART7 was able to reduce BCL6 protein expression only in SU-DHL-4 and SU-DHL-6 whereas it had little or no effect in DB. In Ramos cell line, transfection of the miRNAs led to a significant reduction of BCL6 protein levels in the case of ebv-miR-BART7, ebv-miR-BART9 and ebv-miR-BART17-5p, whereas the cells showed no response to ebv-miR-BART3. This may reflect a cell context-dependent phenomenon that could be related to target-site accessibility. Moreover, we used EBV-positive DLBCL-derived cell lines to allow inhibition of endogenous EBV-encoded miRNAs by miRNA inhibitors. Inhibition of ebv-miR-BART9 and ebv-miR-BART17-5p, led to a moderate increase in the expression of BCL6 in the Farage cell line but not in the DoHH-2 cell line, which expressed the highest levels of EBV-encoded miRNAs (Figure 2c).

Figure 2
figure2

(a) Some predicted miRNAs can downregulate BCL6. Co-transfection of several miRNAs with a luciferase reporter with (pGL3-BCL6) or without (pGL3-Co) the 3′UTR of BCL6 led to specific inhibition for miRNAs ebv-miR-BART3, ebv-miR-BART9 and ebv-miR-BART17-5p when compared with the effect of a miRNA mimic negative control sequence. (b) Transfection of the EBV-negative DLBCL and BL cell lines with several miRNAs led to decreased BCL6 expression. BCL6 signal was normalized using tubulin. (c) Transfection of miRNA inhibitors in EBV-positive cell lines led to a moderate upregulation of BCL6 in the Farage cell line when ebv-miR-BART9 and ebv-miR-BART17-5p were inhibited. An equimolar mixture of the four anti-miRNAs (200 nM final concentration) was also able to increase BCL6 levels. DoHH-2 cell line showed no response to miRNA inhibition. BCL6 signal was normalized using tubulin.

EBV is more frequently found in tumors with a plasmablastic phenotype in which other factors (for example, expression of BLIMP1) could be repressing BCL6 at the transcriptional level.6 Our data support a role of EBV microRNAs that could help to diminish BCL6 expression facilitating BCL6 transcriptional repression by BCL6 targets such as BLIMP1.7 However, the reason by which EBV downregulates BCL6 in DLBCL remains unknown. It is known that several EBV-driven lymphomas rely on the activation of the NF-kB pathway8 as well as ABC-type DLBCL.9 Viral proteins such as LMP1 and LMP2A can activate the NF-κB pathway directly or indirectly to promote the survival of the host cell.8, 10, 11, 12 Because BCL6 represses NF-κB under normal and pathogenic conditions,2 BCL6 downregulation may be necessary to promote survival of the EBV-positive neoplastic cells in some DLBCLs. EBV does not repress BCL6 in Burkitt lymphoma; however, in this tumor type, activation of the NF-κB pathway has been shown to induce apoptosis,13 and so the role of the virus might be different in this disease.

References

  1. 1

    Swerdlow SHCE, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW (eds). WHO Classification: Pathology and Genetics of Tumors of Haematopoietic and Lymphoid Tissues. IARC Press: Lyon, 2008.

  2. 2

    Perez-Rosado A, Artiga M, Vargiu P, Sanchez-Aguilera A, Alvarez-Barrientos A, Piris M . BCL6 represses NFkappaB activity in diffuse large B-cell lymphomas. J Pathol 2008; 214: 498–507.

  3. 3

    Carbone A, Gaidano G, Gloghini A, Pastore C, Saglio G, Tirelli U et al. BCL-6 protein expression in AIDS-related non-Hodgkin's lymphomas: inverse relationship with Epstein-Barr virus-encoded latent membrane protein-1 expression. Am J Pathol 1997; 150: 155–165.

  4. 4

    Pfeffer S, Zavolan M, Grasser FA, Chien M, Russo JJ, Ju J et al. Identification of virus-encoded microRNAs. Science 2004; 304: 734–736.

  5. 5

    Ferry JA . Burkitt's lymphoma: clinicopathologic features and differential diagnosis. Oncologist 2006; 11: 375–383.

  6. 6

    Shaffer AL, Lin KI, Kuo TC, Yu X, Hurt EM, Rosenwald A et al. Blimp-1 orchestrates plasma cell differentiation by extinguishing the mature B cell gene expression program. Immunity 2002; 17: 51–62.

  7. 7

    Vasanwala FH, Kusam S, Toney LM, Dent AL . Repression of AP-1 function: a mechanism for the regulation of Blimp-1 expression and B lymphocyte differentiation by the B cell lymphoma-6 protooncogene. J Immunol 2002; 169: 1922–1929.

  8. 8

    Keller SA, Hernandez-Hopkins D, Vider J, Ponomarev V, Hyjek E, Schattner EJ et al. NF-kappaB is essential for the progression of KSHV- and EBV-infected lymphomas in vivo. Blood 2006; 107: 3295–3302.

  9. 9

    Davis RE, Brown KD, Siebenlist U, Staudt LM . Constitutive nuclear factor kappaB activity is required for survival of activated B cell-like diffuse large B cell lymphoma cells. J Exp Med 2001; 194: 1861–1874.

  10. 10

    Mosialos G, Birkenbach M, Yalamanchili R, VanArsdale T, Ware C, Kieff E . The Epstein-Barr virus transforming protein LMP1 engages signaling proteins for the tumor necrosis factor receptor family. Cell 1995; 80: 389–399.

  11. 11

    Cahir McFarland ED, Izumi KM, Mosialos G . Epstein-barr virus transformation: involvement of latent membrane protein 1-mediated activation of NF-kappaB. Oncogene 1999; 18: 6959–6964.

  12. 12

    Cahir-McFarland ED, Carter K, Rosenwald A, Giltnane JM, Henrickson SE, Staudt LM et al. Role of NF-kappa B in cell survival and transcription of latent membrane protein 1-expressing or Epstein-Barr virus latency III-infected cells. J Virol 2004; 78: 4108–4119.

  13. 13

    Klapproth K, Sander S, Marinkovic D, Baumann B, Wirth T . The IKK2/NF-{appa}B pathway suppresses MYC-induced lymphomagenesis. Blood 2009; 114: 2448–2458.

Download references

Acknowledgements

We would like to acknowledge all members of CNIOs Lymphoma Group for their thoughtful discussion. This study was funded with the following grants: from Ministerio de Educación y Ciencia, Spain SAF2007-65957-C02-02, SAF2008-03871; from the Fondo de Investigaciones Sanitarias, RTICC RD06/0020/0107 and PI10/00621; from Fundación Científica de la Asociación Española Contra el Cáncer (AECC). LDL is supported by a fellowship from Ministerio de Sanidad y Consumo (FI08/00038); NM is supported by a contract from Ministerio de Sanidad y Consumo (CP06/00002).

Author information

Correspondence to M A Piris.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Leukemia website

Supplementary information

Supplementary Figure S1 (PPT 177 kb)

Supplementary Figure S2 (PPT 83 kb)

Supplementary Figure Legends (DOC 19 kb)

Supplementary Table S1 (XLS 31 kb)

Supplementary Table S2 (XLS 20 kb)

Supplementary Table S3 (XLS 19 kb)

Supplementary Materials and Method (DOC 37 kb)

Rights and permissions

Reprints and Permissions

About this article

Further reading