Investigating molecular interaction networks of nuclear factor one transcription factors

Abstract

Gene expression is a tightly regulated process involving multiple steps, with transcription factors (TFs) serving as key regulators. TFs bind to specific DNA sequences, interact with other proteins, and modulate the expression of target genes. Among these, the Nuclear Factor I (NFI) family plays a significant role in development and cancer. NFI transcription factors are key regulators of gene expression, influencing cellular proliferation, differentiation, and specialization across various tissues. In vertebrates, the NFI family includes four members: NFIA, NFIB, NFIC, and NFIX. Alternative splicing of NFI genes gives rise to multiple isoforms. These proteins contain a highly conserved N-terminal DNA-binding and dimerization domain, enabling them to recognize and bind a consensus sequence (TTGGC(N5)GCCAA) as either homo- or heterodimers. While performing their regulatory functions, NFI family members interact with other proteins and the chromatin. However, aside from a few specific interactions detected in certain contexts and the interactions of NFIs with other transcription factors, the interactomes of NFIs, their dynamics, and their functions remain largely unexplored. In this thesis, we employed complementary omics-level techniques, i.e., interactomics (affinity purification mass spectrometry (AP-MS) and proximity-dependent biotinylation (BioID)), and chromatin immunoprecipitation sequencing (ChIP-Seq), to obtain a comprehensive view of the NFI proteins and their interactions. The MAC-tag approach was utilized, enabling combined AP-MS and BioID analyses. The HA epitope within the MAC3-tag, along with the BioID (UltraID) sequence, allows detection of tagged protein expression by immunoblotting and immunofluorescence, and the use of the same Flp-In™ T-REx™ 293 cell lines in ChIP-Seq for studying protein-DNA interactions. In addition to the MAC-tag approach in isogenic inducible Flp-In™ T-REx™ 293 cell lines, we utilized lentivirus transduction and electroporation methods to achieve forced expression in neuroblastoma and glioblastoma cell lines, which are generally difficult to transfect using standard methods. Our experimental approach encompassed all four predominant isoforms of NFI family members and the atypical isoform NFIB4, which lacks the DNA-binding domain and was initially linked to megakaryocyte differentiation. Interaction partners were identified through proximity labeling with varying biotinylation times, while stable interactors were captured using AP-MS analysis. In parallel, ChIP-seq experiments were performed to uncover potential NFI target genes. We also investigated NFIA's regulatory impact on one of its interacting partners, the transcription factor SOX2. Additionally, we used AlphaFold 3 to predict DNA binding, modeling interactions between NFI proteins and their ChIP-seq target DNA sequences, enabling us to identify key regions of the NFI proteins involved in DNA binding. We observed that, despite exhibiting some redundancy, each family member had unique high-confidence interactors and target genes, highlighting distinct roles within the transcriptional regulatory networks. Time-dependent interactome mapping revealed the dynamic nature of NFI interactions with chromatin remodelling and transcriptional regulator complexes. UltraID-based proximity labeling identified extensive, evolving interactions with chromatin remodeling and transcriptional complexes, particularly SWI/SNF and Mediator. AP-MS confirmed stable associations with SWI/SNF complex proteins, and most of the interactors are validated by co-expression and pull down experiments, underscoring the collaborative roles of NFIs and SWI/SNF in chromatin remodeling. While previous studies associated NFIA and NFIB with the Mediator complex in neural stem cells, our findings reveal that NFIC also interacts with Mediator complex; however, these associations appear transient or context-dependent and are only detected by proximity labeling. The interactome of the atypical short isoform, NFIB4, is enriched with proteins involved in mRNA regulation, suggesting that NFIs may have roles beyond traditional DNA binding and transcriptional modulation. Meanwhile, structural analysis using AlphaFold3 revealed four conserved binding regions—loop, helix1, helix2, and segment—that are critical for DNA interaction. Additionally, predictions with ChIP-seq results showed that NFIs can bind to motifs with variable spacer lengths, indicating flexibility in motif recognition. By integrating ChIP-seq with RNA interference (RNAi) experiments, our study identified SOX2 as an important target of NFIA. NFIA knockdown led to an altered genomic binding profile of SOX2, accompanied by notable shifts in pathway enrichment among SOX2 target genes. Specifically, pathway analysis of potential SOX2 targets post-NFIA silencing revealed enrichment in the cAMP signaling pathway. Our findings reveal that NFIs interact with a diverse array of chromatin remodelers, transcriptional regulators, and other proteins, highlighting their extensive roles in regulating gene expression. Time-resolved interactomics demonstrated dynamic and context-dependent interactions with key complexes, such as SWI/SNF and Mediator, suggesting that NFIs actively participate in coordinating transcriptional activity. Additionally, integration of ChIP-seq data with RNA interference experiments showed that NFIA modulates the activity of other transcription factors, including SOX2, altering its genomic binding profile and influencing pathway enrichment among its target genes. These observations, coupled with the isoform-specific interaction profiles and their distinct sets of target genes, position NFIs as central regulators capable of modulating complex transcriptional networks and influencing cellular processes across diverse contexts. Therefore, we propose that NFIs function as master regulators, modulating other transcription factors and interacting with various proteins across multiple cellular processes. While NFIs may exhibit pioneering capabilities, this requires validation through direct evidence of their ability to bind closed chromatin. Future studies could determine whether NFIs access nucleosome-occupied or otherwise inaccessible chromatin, helping to confirm their potential role as pioneering transcription factors. In conclusion, this thesis advances our understanding of the NFI family's roles in gene regulation, revealing both established and novel interactions that shape cellular function across various contexts. By employing complementary omics-level approaches, we identified distinct interaction profiles and target genes for each NFI member, underscoring their individual roles within transcriptional regulatory networks. The unique interactome of the NFIB4 isoform, particularly its association with mRNA regulatory proteins, suggests a broader functional scope for NFIs that may extend beyond traditional DNA binding to influence post-transcriptional processes. Additionally, our findings on NFIA's regulatory effects over SOX2 binding and pathway enrichment point to NFIs as master regulators with the potential to guide cellular fate decisions and differentiation. These insights lay the groundwork for future research aimed at exploring the mechanisms of NFI function in chromatin remodeling and gene expression, as well as their therapeutic implications in diseases such as cancer and neurodevelopmental disorders.