Mechanisms of gene expression and recombination

National Institute Of Arthritis And Musculoskeletal And Skin Diseases

Investigators

Linked publications, trials & patents

Abstract

Transcription is regulated by DNA elements known as enhancers, where transcription factors bind to specific DNA motifs. Before genes are turned on, enhancers become accessible; understanding the controlled accessibility and generation of active enhancers has been a longstanding interest of this lab. Nipped-B-like protein (NIPBL), associated with MAU2, is required for the association of the cohesin complex with DNA and higher order chromatin and loop extrusion.Using CRISPR-Cas9-based gene editing, NIPBL IDR deletion (NIPBL-IDRdel) mutants were generated in CH12 B cells. We found that the NIPBL intrinsically disordered region (IDR) is likely responsible for tuning transcriptional regulation but not overall chromatin organization. To dissect the basis of the transcriptional alterations, analysis of NIPBL protein partners was determined via mass spectroscopy. Preliminary data reveal that while NIPBL interacts with key members of transcription machinery such as transcription factors (CNOT1, CNOT2, RUNX3, NFAT2, NFKB1), RNA Polymerase II subunits and Mediator complex subunits, deletion of the IDR selectively impairs this interaction. This year we also examined the impact of cohesin on DNA repair. We found that cohesin rings are recruited to DNA breaks through phosphorylation, leading to enhanced loop extrusion and increased insulation of the broken DNA. We determined that phosphorylation deficient cells were delayed in DNA repair with improper end joining, resulting in translocations and cis deletions An important feature of transcription regulation is the combinatorial action of transcription factors at active regulatory DNA. This fiscal year, we explored transcription factor cooperativity in mammals by analyzing 500 mouse and human primary cells, combining an atlas of transcription factor motifs, footprints, binding (ChIP-seq), transcriptomes (RNA-seq), and accessibility (ATAC-seq). We uncovered two transcription factor groups that colocalized with most expressed factors, forming stripes in hierarchical clustering maps. The first group included lineage-determining factors that occupied DNA elements broadly, consistent with their key role in tissue-specific transcription. The second one, dubbed universal stripe factors (USFs), comprised 30 SP, KLF, EGR, and ZBTB family members that recognized overlapping GC-rich sequences in all tissues analyzed. Knockouts and single-molecule tracking revealed that USFs imparted accessibility to colocalized partners and increased their residence time. Transcriptional regulation is the means whereby cells orchestrate expression of individual genes or group of genes. Recent studies with single cells have shown that, for the most part, transcription is not a continuous activity but it occurs in bursts. The amount of RNA produced for each gene is directly proportional to the amplitude and frequency of such bursts. How these two parameters are controlled is not totally clear, but it appears that DNA sequence information at promoters determine the amplitude, while burst frequencies rely on sequence information at enhancers. Much has been learned on transcriptional bursting through the study of individual genes. However such experiments are time consuming and do not provide a global view of gene expression in cell populations or tissues. To address this deficiency, this fiscal year we have developed a computer model to analyze single cell transcriptomes. This approach provides for the first time a comprehensive view of bursting in a cell population without the need to label individual genes with fluorescent molecules. Our goal moving forward is to use this strategy with cells deficient for proteins known to play key roles in transcription to determine whether they regulate bursting. A clinically important aspect of studying gene expression is to better understand gene regulatory networks occurring in malignant cells, such as B cell lymphoma. This will allow us to identify novel targets that can lead to innovative therapeutic interventions. Diffuse large B cell lymphoma (DLBCL) is the most common type of aggressive lymphoma in the US. Based on a DLBCL derived cell line, we established reporter cell lines for 47 key B cell genes and performed genome-wide CRISPR/Cas9 screening coupled to fluorescence-activated cell sorting to identify regulators of gene expression. Another important feature of transcription is its control of recombination in maturing B cells. In the bone marrow for instance, transcription facilitates the recombination of antibody variable genes to D and J segments to produce the V(D)J portion of antibodies. This fiscal year, our laboratory has continue exploring how transcription of these genes facilitate the physical pairing of V, D and J segments during recombination. In this project, we have developed a genome-wide method and software suite to infer transcriptional kinetics from single-cell RNA sequencing data. We applied this approach across multiple cell types and perturbations and found that: Regulatory genes burst more frequently than housekeeping genes, the frequency of bursting is regulated by the mediator (MED26), burst size is governed by MYC, and chromatin architecture proteins (cohesin and CTCF) as well as BRD4 can modulate both parameters. Cohesin, BRD4 and MED26 act at different stages of the transcription process; notably, MED26 functions downstream of both chromatin spatial architecture and the preinitiation complex, working with BRD4, to initiate transcriptional burst.