Investigation of NFİB function and regulation of its putative target genes in human neural stem cell and SH-SY5Y neuroblastoma cell lines

Abstract

The central nervous system comprises numerous neuronal and glial subpopulations that have unique identities. Molecular mechanisms that underlie the generation of this cellular diversity have been under investigation. During development, the formation of cell subclasses with particular features is determined by tissue-specific transcription factors (TF). TFs sequence-specifically bind to DNA, interact with other proteins, and affect the expression of target genes. One of the key TFs in the developing brain is the Nuclear Factor I (NFI) family. There are four members (A, B, C, and X) in vertebrates. NFI proteins comprise a highly conserved N-terminal DNA binding and dimerization domain and bind to a TTGGC(N5)GCCAA consensus sequence as homo or heterodimers. However, they have a less conserved C-terminal transcription modulation domain which may lead to differential transcriptional regulation of target genes. In the developing mouse, Nfia, Nfib, and Nfix are expressed in an overlapping but distinct expression pattern in different regions of the embryonic brain, while their expression is restricted to stem cell niches in the adult. In the central nervous system, deletion of each member leads to delayed glial and neuronal differentiation, aberrant cell migration and increased proliferation. In the developing hindbrain, only the absence of Nfib leads to delayed development of several precerebellar nuclei, indicating that Nfib may play a unique role in this system. Dysregulated expression of NFIs have also been linked to tumor growth and progression, however, with opposing effects. For example, NFIB is oncogenic and promotes metastasis in colorectal cancer, melanoma, gastric cancer, estrogen receptor (ER)-negative breast cancer, and small cell lung cancer; while it has a tumorsuppressive function in non-small cell lung cancer, glioblastoma, osteosarcoma, and cutaneous cell carcinoma. NFIs perform their context dependent, cell-type and tissue specific functions, via regulation of specific set of downstream transcriptional targets. Despite the fact that NFI binding motifs have been found in the promoter and upstream enhancer regions of many genes, only a few of them have been so far investigated as direct NFI targets. Further identification and characterization of downstream targets of NFIs in various tissues will help elucidate molecular mechanisms that regulate embryonic development and related diseases, as well as cancer pathologies. In an attempt to investigate how NFIB regulates neurogenesis in developing precebellar nuclei, differentially expressed genes in E14 Nfib knock-out mouse precerebellar neuroepithelium have been analyzed. The RNA profiling analysis revealed putative candidates for further research. Of these putative NFIB targets, we selected Cdon (Cell adhesion molecule related, down regulated by oncogenes) and Fgf15 (Fibroblast growth factor 15), since these genes have been implicated in neural development of the cortex. We examined NFIB-mediated transcriptional regulation mechanisms of CDON and FGF19, the human ortholog of mouse Fgf15, in human neural stem cells (hNSCs derived from H9 ESC, Gibco). Neural stem cell culture systems provide an in vitro model of human neural development. Since NFIs have been reported to regulate neuron production in diverse parts of the developing brain, they may have comparable functions in vitro. Understanding these processes and the underlying molecular mechanisms in vitro will also help understand how the brain develops in vivo as well as failures in this process. Thus, in the first chapter of the thesis, we set out to examine NFI function and regulation mechanisms of potential NFIB targets, CDON and FGF19, in neuronal differentiation of hNSCs in vitro. RT-qPCR analyses revealed that mRNA expression of NFIB, NFIC, and NFIX is downregulated, whereas NFIA is upregulated in differentiating hNSCs. Since NFIA levels are quite low in these cells, overall NFI expression levels decrease during neuronal differentiation in hNSCs. These cells express NFIB at much higher levels compared to the other NFI members. Therefore, this study focuses on NFIB's role in hNSCs. We analyzed cell proliferation and differentiation by BrdU incorporation assays and immunofluorescence staining of neural stem and neuronal marker proteins. However, NFIB overexpression or knockdown did not affect the proliferation or neuronal differentiation potential of hNSCs. Nevertheless, these data cannot preclude NFIB's potential role in differentiation and/or self-renewal of hNSCs since NFIB could be silenced only by 30–50% in these cells and analyses were performed in whole cell populations that might mask possible changes induced by NFIB loss. Moreover, in NFIB overexpression experiments, we may need other proteins acting as cofactors that are not supplied along with NFIB. This study identifies FGF19 as a novel downstream target of NFIB in hNSCs. Human FGF19 is preferentially expressed in the fetal brain, among other tissues. Recombinant human FGF19 treatment has been shown to enhance neuronal differentiation in mouse neuroepithelial and cortical cells. In accordance with these data, FGF19 expression increases in differentiating hNSCs. Moreover, FGF19 expression increases in NFIB silenced hNSCs while it is reduced in NFIB overexpressing cells, indicating that NFIB regulates FGF19 transcription in hNSCs. Indeed, NFIs directly repress FGF19 promoter-driven luciferase activity, confirming that NFIs transcriptionally target FGF19. Moreover, chromatin immunoprecipitation (ChIP) assays showed that NFI proteins occupy −777 (relative to the transcription start site) in hNSCs, indicating NFI interaction with the FGF19 promoter in vivo. Since NFIB expression decreases upon neuronal differentiation, while FGF19 increases and NFIB directly represses FGF19 in hNSCs, future studies are required to address functional relevance of NFIBmediated FGF19 repression in the control of self-renewal and neural differentiation of these cells. In the absence of NFI, Cdon, a cell surface glycoprotein of the immunoglobulin (Ig) superfamily, is upregulated in the developing mouse brain. CDON is expressed in various tissues, primarily in the brain, muscle, and endocrine tissues during human and murine embryogenesis. Moreover, CDON is implicated in proliferation and differentiation control as it promotes myogenesis and neurogenesis in vitro and is essential for proper brain and skeletal-muscle development. However, in this study, CDON expression decreased in differentiating hNSCs and it did not change in NFIB overexpressed or silenced hNSCs, analyzed by RT-qPCR. These data indicate that CDON is not an NFIB target in this system. Recently, CDON has been described as a dependence receptor that induces apoptosis in the absence of its ligand SHH. During cancer progression, in an environment with limited SHH, tumorigenic tissue may downregulate CDON to eliminate its apoptotic activity. Indeed, CDON expression decreases in colon, lung, and neuroblastoma tumors, implicating a tumor suppressor role for CDON. As NFIs are involved in progression of various cancers, we examined whether NFIs regulate CDON transcription in SH-SY5Y human neuroblastoma cells. ChIP assays showed that NFIs bind to human CDON gene regulatory regions, -8 and -941 (relative to the transcription start site), in hNSCs and SH-SY5Y cells. Moreover, NFIs repress CDON promoter-driven luciferase expression via interacting with those NFI sites. Finally, CDON is upregulated in NFIB silenced SH-SY5Y cells, suggesting that the NFIB-CDON axis may be involved in neuroblastoma biology. On the other hand, silencing NFIB in SH-SY5Y cells decreases cell viability and proliferation, suggesting an oncogenic role for NFIB in neuroblastoma. Next, we tested the hypothesis that NFIB may affect SH-SY5Y cell survival by suppressing expression and thereby, pro-apoptotic activity of CDON. However, downregulation of CDON on its own could not rescue the phenotype induced by NFIB silencing, most likely because other NFIB downstream targets, which may include p21, are also involved. Further studies are required to investigate the functional consequences of NFIB mediated CDON repression in other developmental systems and disease models. NFIB's oncogenic effects in SH-SY5Y cells may involve inhibition of apoptosis and/or regulation of cell cycle components. Moreover, NFIB might promote differentiation of SH-SY5Y cells and/or contribute to the aggressive state of neuroblastoma tumorigenesis. However, these and underlying mechanisms need to be further investigated.