Genomic analysis of human oligodendrocyte differentiation
MetadataShow full item record
Oligodendrocytes (OL) responsible for making the axonal myelin are lost in demyelinating diseases. This leads to loss of saltatory conduction of signals within axons and eventually axonal death, which OL progenitor cells (OPCs) have the potential rescue. However, the molecular mechanisms that underlie human OL lineage commitment and OL differentiation are poorly understood. The optimal source of appropriate human cells for transplantation-based therapy has yet to be determined. Direct reprogramming of fibroblasts into specialized cells such as neurons and cardiomyocytes suggest that reprogramming may provide a novel source of cells. In the first aim for this study, we present the design of a strategy for direct reprogramming human fibroblast cells into OL progenitor cells (OPCs). We hypothesized that over-expression of native OPC transcription factors (TF) would reprogram human fibroblasts. Using a combination of traditional differential gene-expression analysis and Weighted Gene Co-expression Network Analysis (WGCNA), we identified 10TF which best correlated with OL development. To identify OPC reprograming, we developed a GFP reporter lentivirus under the control of an OL lineage-specific SOX10 enhancer MCS5. Fetal O4 presenting OLs expressed SOX10:MCS5-GFP 4.5 fold higher relative to other primary human cells and importantly GFP was not detected in GFAP+ astrocytes or Tuj1+ neurons. Using human fibroblasts, we found that the 10TF induced high levels of GFP in 8.35% of infected cells (n=3). We confirmed the OL identity of these cells using immunostaining for OPC-specific markers, NG2 and O4. To refine the pool of TF which were necessary and sufficient to reprogram fibroblast, we infected fibroblasts with each combination of 9 TFs, each missing a single TF. By this method, we identified a minimal combination of 4 TF sufficient for high GFP-expression in fibroblasts. Future studies will determine whether GFP-expressing fibroblasts are capable of OL maturation and myelination using in vitro and in vivo models. Furthermore, for the second goal of this study we aimed to identify signals that regulate human OPC differentiation as they were applicable in providing novel means to induce remyelination. WGCNA was applied to define human OPC differentiation-specific transcriptional profiles. Interestingly, lentiviral over-expression of OL differentiation network member GNB4, could significantly promote OPC differentiation in vitro. We sought to validate the functions of GNB4 in vivo. As such, we transplanted lentiviral infected hOPCs over-expressing GNB4 or mCherry, the control, into the corpus callosum of neonatal shiverer/rag2 hypo-myelinating mice. At 8 weeks post-implantation we perfused, sectioned, and stained their brains via immunohistochemistry using markers for proliferation (Ki67), and differentiation as astrocyte (GFAP) and OL lineages (CC1 and MBP). At 8 weeks, GNB4-infected hOPC transplanted animals exhibited more than a 3 fold increase in MBP within the corpus callosum (unpaired t-test p < 0.05, n=4). These results confirm the roles of GNB4 in hOPC differentiation GNB4 could represent future pharmacological target for remyelination therapy. In contrast, we demonstrate that knock down of OPC non-differentiation module member SULF2 in NG2 positive OPCs, during remyelination in lysolecithin demyelinated lesions significantly increased the density of mature Plp1 positive OLs in mice spinal cord white matter. Furthermore, we noted a sulfatase inhibitor was capable of significantly inhibiting BMP and WNT signaling in a dose dependent manner. This places the sulfatase inhibitor in a unique position to block two major OPC differentiation inhibitory pathways. Thus inhibition of sulfatase which was another member of WGCNA generated modules promises to be a novel target for modulating OPC mediated remyelination.