Abstract
The germinal center (GC) reaction includes two critical events: somatic hypermutation (SHM) and class switch DNA recombination (CSR). These processes are tightly controlled, thus preventing excessive DNA injury which could lead to loss of normal B lymphocytes as well as the survival of cells with oncogenic DNA lesions. MicroRNA-155 (miR-155) plays an important role in immune cell biology; miR-155 knock-out (KO) mice have a defective mature B cell development characterized by a decreased number of GC B cells, whereas the Eμ-miR-155 transgenic mouse model develops and oligoclonal proliferation which evolves to B cell lymphoma. These observations suggest that miR-155 may regulate B cell sensitivity and response to DNA damage, which could mechanistically explain the phenotypes observed in the gain and loss-of-function animal models.
To examine this concept, we purified mature B cells from multiple pairs of miR-155 KO and WT littermates, and stimulated them with LPS and IL4 thus recapitulating the GC reaction in vitro. Next, we used immunohistochemistry to quantify γH2AX at the DNA double-strand breaks (DSBs) foci, western blot to evaluate the phospho-p53 (Ser15) levels, and real-time RT-PCR to quantify the expression of p53 target genes (p21, GADD45a, cdc25c, PCNA). In these assays, we found a significantly higher γH2AX staining in B cells null for miR-155 than in WT controls (number of foci per nucleus: 4.1±0.9 vs. 1.4±0.2, p<0.01), and a markedly elevated p53 activity, defined by its phospho-level and target genes expression. To confirm these observations in a more physiologic setting, we immunized miR-155 WT and KO mice with NP-CGG, collected spleens and purified mature B-cells. Quantification of subpopulation by FACS confirmed that miR-155 mice developed fewer GC B cells, and the examination of DSBs foci and p53 target genes expression supported our in vitro observation of a heightened sensitivity to DNA damage and p53 engagement in miR-155 null cells. We concluded that miR-155 deficiency in B cells associates with excessive DNA damage and p53 activity. To further isolate define the role of miR-155 in these events, we used a retrovirus system to rescue miR-155 expression in B cells derived from the KO mouse, and defined the pattern of DNA damage response. In these assays, cells transduced with a MSCV-miR-155 construct formed fewer DSBs foci than their control counterparts transduced with an empty vector (9.0±2.1 vs. 4.9±1.8, p<0.05) and showed significantly lower p53 activity, defined by target gene expression. To test if miR-155 controls the DNA damage induced by broad genotoxic stresses, we exposed miR-155 KO or WT B cells, thymocytes and bone marrow cells to 5Gy of ionizing radiation (IR) or etoposide (4µM). In these assays, albeit a robust induction of γH2AX foci formation and p53 activation were detected, there was no significant difference between WT and miR-155 KO mice, in any of the cell types analyzed. We concluded that the heightened sensitivity to genotoxic stress in miR-155 KO mice is specific to B cells undergoing the GC reaction. This observation suggests a potential role in this process for activation-induced cytidine deaminase (AID), a key regulator of the DNA damage inducing SHM and CSR, which is also a miR-155 target gene. We confirmed that AID expression is higher in miR-155 KO than in WT B cells, and to test its contribution to the excessive DSB and p53 activity found miR-155 null cells we used an RNAi approach. We found that the inhibition of AID levels in miR-155 KO B cells significantly reduced γH2AX foci formation and expression of p53 target genes.
Together, our data highlight a hitherto unappreciated interaction between miR-155 and the p53 pathway, involving DNA lesions that are germane to the GC reaction and the control of AID expression/activity. The excessive engagement of p53 is this setting may explain, at least in part, the loss of normal GC B cells found in the miR-155 KO mice. Conversely, it is possible that cells expressing abnormally high miR-155 levels show a subpar activation of the DNA damage response thus leading to the accumulation of oncogenic mutations. This paradigm would provide a mechanistic explanation for the initial poly/oligoclonal proliferation reported in the Eμ-miR-155 mice, which eventually evolves into a B-cell lymphoma
Disclosures:
No relevant conflicts of interest to declare.
Maximal killing of lymphoma cells by DNA damage–inducing therapy requires not only the p53 targets Puma and Noxa, but also Bim
