MDSC Polarization and Helicobacter-Induced Gastric Metaplasia

摘要

Abstract Chronic Helicobacter infection and the resulting inflammation induces gastric metaplasia in the corpus of WT mice but not in mice null for Gli1, demonstrating that this pre-neoplastic lesion requires the induction of canonical Hedgehog signaling. Moreover, we reported that a subset of myeloid cells expressing surface markers and T cell suppressor function indicative of myeloid-derived suppressor cells (MDSCs) express Schlafen4 (Slfn4), a direct target of the Gli1 transcription factor and a known myeloid differentiation factor. Since MDSCs are immature cells, collectively our studies demonstrate that maturation of this myeloid cell subpopulation requires Gli1 and produces proinflammatory cytokines creating a gastric microenvironment favorable for metaplasia and neoplastic transformation. More recently, we have analyzed the human homologs of Slfn4, which include SLFN5 and SLFN12L. We reported that peak expression of SLFN5 in gastric tissue occurred in human subjects with intestinal metaplasia who about a decade later developed gastric cancer. We therefore considered that the polarization of myeloid cells to MDSCs prior to neoplastic transformation might predict who is more likely to develop gastric cancer and as such could provide a therapeutic target to prevent future transformation. We used RNA-Seq and Nanostring microarrays to identify transcripts and microRNAs expressed in Slfn4+-MDSCs from the stomachs of a Helicobacter-infected mouse and found that miR130b co-localized with Slfn4+ cells in the metaplastic mouse stomach. A similar result was observed for SLFN12L in the metaplastic human stomach suggesting that miR130b might identify patients with gastric metaplasia. Our preliminary results demonstrated that Slfn4 and miR130b are required to exert T-cell suppression. In addition to type 1 interferon-regulated genes identified by RNA-Seq, these Slfn4+-MDSCs also expressed several tumor necrosis factor superfamily ligands (TNFsf) and the alarmin IL-1a. Therefore we will test the hypothesis that debris from damaged gastric epithelial cells activates Damage-activated molecular pattern (DAMP) signaling, production of IFNa and polarization to MDSCs, which contribute to a metaplastic phenotype. Aim 1, we will define how DAMP signals induce polarization of Slfn4+-MDSC. In Aim 2, we will define how Slfn4 contributes to MDSC function. In Aim 3, we will define the contribution of Slfn4+-MDSCs to Helicobacter-induced metaplasia. In Aim 1, we will use a combination of flow cytometry, T cell suppression assays and transfection studies to identify the cell populations producing IFNa and the gene targets responding to this DAMP-activated cytokine. In Aim 2, we will used mass spectrometry and pull down assays to identify Slfn4 and SLFN12L interacting proteins. In Aim 3, we will conditionally delete Slfn4 from Gli1- expressing cells to define their contribution to the metaplastic changes observed after Helicobacter infection. Completion of these aims will result in a better understanding of this MDSC subpopulation that can potentially be used as a biomarker and target of small molecule inhibitors.