LMP1 Upregulates IL10 and Constitutively Activates Stat3 To identify cytokines that may contribute to the increased survival and growth of lymphomas, the expression levels of a panel of cytokines were screened on CD19+ MACS-purified B cells, using an RPA probe set for IL4, IL5, IL10, IL13, IL15, IL9, IL2, IL6, and IFNγ. Expression levels were quantified with a phosphorimager and normalized to the ribosomal housekeeping gene L32. None of the tested cytokines were detected in wild-type lymphocytes, therefore cytokine:L32 ratios were set to 1 in the mouse B cell lymphoma line 967. Transcription of IL10, IL15, and IFNγ were reproducibly detected in LMP1 transgenic lymphocytes and lymphoma cells and was higher than in the B cell lymphoma cell lines 967 and K46μ (Figure 5A). There was no significant difference in the expression of IL15 and IFNγ between LMP1 transgenic lymphocytes and lymphoma cells, suggesting that upregulation of IL15 and IFNγ is induced by LMP1 expression in healthy lymphocytes but is not a unique property of malignant lymphocytes. Strikingly, IL10, a B lymphocyte stimulatory cytokine, was increased 1.5- to 5-fold in the wild-type and LMP1 transgenic lymphoma cells compared to LMP1 transgenic lymphocytes (Figure 5A). Production of IL15 and IFNγ has been associated with induction of cytotoxic effector responses in cells latently infected with EBV [33,34]. However, transformation and growth properties induced by EBV are associated with the upregulation of IL10 [35–38]; hence, the effects of IL10 upregulation on the growth properties of the lymphoma cells were further examined. Immunoblot analysis indicated that LMP1 transgenic lymphocytes and wild-type and LMP1 transgenic lymphoma cells had corresponding increased levels of phosphorylated α and β isoforms of activated Stat3, a target of the IL10 receptor (Figure 5B). However, when comparing the same lymphomas, there was no correlation between the levels of IL10 induction and the levels of Stat3 activation. This suggests that the activation of Stat3 is not solely induced by IL10 or that Stat3 activation may be constitutive. Additionally, there was no correlation between the levels of LMP1 expression and the levels of IL10 induction (Figures 1A and 5A). This indicates that the induction of IL10 is a general property associated with enhanced survival and may only be indirectly affected by LMP1. Neutralizing antibodies to IL10 did not affect the survival of lymphoma cells as determined by the MTS assay (unpublished data), suggesting constitutive activation of Stat3. This was confirmed by immunoblot analysis such that in the presence of anti-IL10 neutralizing antibodies, pStat3 levels remained activated in lymphoma cells isolated from wild-type and LMP1 transgenic lymphomas (Figure 5C). Exogenous addition of IL10 enhanced pStat3 activation above constitutive levels, indicating that lymphoma cells are responsive to IL10 treatment (Figure 5C). This means that although the lymphoma cells have constitutive Stat3 activation, it may be further enhanced by IL10 induction. The neutralizing effect of the anti-IL10 antibody was confirmed by pre-incubation of IL10 with anti-IL10 antibody compared to a rat IgG1 isotype control (Figure 5C). Figure 5 LMP1 Upregulates IL10 Expression and Constitutively Activates Stat3 (A) Relative expression of IL10, IL15, and IFNγ mRNA in WT and LMP1 transgenic B cells (CD19+), as detected with an Rnase protection assay. Mouse lymphoma cell lines 967 and K46μ were used as controls. Expression levels were quantified with a phosphorimager and values were normalized to the ribosomal housekeeping gene L32. The cytokine:L32 ratio was set to 1 in the mouse B cell lymphoma line 967. (B and C) Immunoblot analysis of activated pStat3 in purified B cells (CD19+) from WT and LMP1 transgenic mice (B) at the time of harvest, and (C) 4 h after culture with or without IL10, a neutralizing antibody to IL10, or a rat IgG1 isotype control. (C) Shown are the results for WT lymphoma 1 and LMP1 transgenic lymphoma 1. Arrows indicate the positions of the α and β isoforms of Stat3. Actin was used as a loading control. (D) Immunohistochemistry detection of activated nuclear pStat3 in the spleens of WT and LMP1 transgenic mice. Scale bar, 20 μm. Nuclear translocation of pStat3 is a consequence of activation, and nuclear pStat3 was not detected by immunohistochemistry staining of spleen sections from control mice. However, nuclear pStat3 was detectable in LMP1 transgenic mice and wild-type lymphomas and was detected more homogeneously in LMP1 transgenic lymphomas (Figure 5D). The constitutive activation of pStat3 and abundant nuclear Stat3 suggests that Stat3 signaling contributes to LMP1-mediated lymphoma development.