Atypical E2F Dependent Dysregulation Of Cell Cycling By Microrna-142 Regulates T-Cell Responses and Experimental Graft-Versus-Host Disease

Noncoding microRNAs (miRs) have recently been linked to immune system function. We investigated the role of miR-142, a hematopoietic specific miR, in regulating T cell responses. To understand its specific function in T cell immunity we utilized homologous recombination technology and generated mutant mice bearing a targeted deletion of the miR-142 gene on the B6 background. The homozygous miR-142 knockout (KO) animals were viable, fertile and showed no apparent developmental anomalies. Thymic analyses of the miR142^-/- animals demonstrated no significant differences when compared with WT littermate controls in total thymocytes, early thymic progenitors, DP and DN cells. Bone marrow studies demonstrated similar numbers of LSK+ HSCs while analyses of secondary lymphoid organs (spleen and popliteal lymph nodes) demonstrated similar absolute numbers of naïve T cells (CD44^low62L⁺), memory like T cells (CD44^hi62L⁺CCR7^- and CD44^hi62L^-CCR7^- cells), CD4⁺25⁺Foxp3⁺ cells, CD69⁺VLA4⁺CD3⁺ cells and weekly peripheral blood examination demonstrated similar lymphocyte counts as the WT littermates.

Functional studies, however, demonstrated that when compared with WT T-cells, the KO T-cells showed significantly slower rate of proliferation by CFSE analyses upon stimulation anti-CD3⁺ and 28⁺ antibodies (P<0.003). They showed lower IL-2, IFNγ and IL-17 but greater amount of IL-6 production (P<0.001) and demonstrated greater apoptosis (P<0.02). Cell cycling analyses with flow cytometry demonstrated that a significantly greater percent of the miR142^-/- T-cells were in the S and G2 phase (P<0.01) when compared with WT T-cells suggesting altered cell cycling. Similar reduction in proliferation, cytokine secretion by the miR142^-/- T-cells was also observed upon in vitro stimulation with allogeneic BALB/c DCs.

To determine the in vivo relevance of miR142 deficiency in T cells, we next utilized MHC mismatched B6àBALB/c model of allogeneic BMT. BALB/c animals were lethally irradiated (9Gy) and transplanted on day 0 with 5x10⁶ BM from WT B6 animals along with 2x10⁶ splenic CD90⁺T cells from the WT or miR142^-/- donors. The allogeneic animals that received KO T-cells showed significantly less severe clinical, histopathological GVHD (GI tract on days 7 and 21) and mortality (P<0.02). Analyses of donor T cells on day 7 post-BMT demonstrated reduced expansion and IFNγ secretion (P<0.04) but showed no significant differences in the ratio of Treg:conventional T-cells between the WT and KO T-cell allogeneic recipients. To further confirm the specific role of miR142 deficiency, we next treated the WT animals with miR-142 anatgomir (days 1, 3 and 7) and found that it significantly reduced GVHD mortality (P<0.003). The KO T-cells also reduced GVHD mortality in a MHC matched minor mismatched B6→C3H.sw BMT model demonstrating strain independent effects.

To further determine the miR142 specific molecular mechanisms we performed extensive bioinformatic analyses. In light of a defect in T cell cycling in miR142^-/- T-cells, we focused on the putative miR142 targets that are known to regulate cell cycling. Two of the three bioinformatic programs suggested the following known regulators of cell cycling, EGR2, DAG, all eight E2F transcription factor family (the typical E2F1-6 and atypical E2F7-8) members as putative targets. We next performed DNA microarray analyses to determine differential gene expression patterns in miR142^-/- and WT T cells, which demonstrated an increase in the expression (>15x) of only the atypical E2Fs, namely E2F7 and E2F8, but not in any of the other above predicted cell cycle regulating molecules. The increase in the expression of the atypical E2Fs in the miR142^-/- T-cells was next confirmed by PCR analyses at baseline (unstimulated) and also sequentially at 6, 12, 24 and 48 hours following in vitro stimulation. Knockdown of E2F7 and E2F8 in miR142^-/- T cells with sh-RNA rescued their proliferative responses and corrected the cell cycling defects to the levels comparable to WT T-cells, thus demonstrating that the atypical E2F7 and 8 are critical for miR-142 mediated regulation of T cells.

Thus our data show (a) generation of a novel miR142 knockout mouse (b) demonstrate that miR142 regulates T cell responses in vitro and in vivo by targeting atypical E2Fs and (c) suggest that targeting miR142 in vivo with its antagomir might be a novel therapeutic strategy for regulating GVHD.