Biology subjects

Ozbun, L.

Publications and source records attributed to Ozbun, L..

4 recordsLinked to original sources

bioRxiv2026

Re-establishment of TAD boundary organization during DNA replication

A ubiquitous feature of higher order genome organization is the presence of topologically associating domains (TADs). The chromatin architectural proteins cohesin and CTCF are known critical organizers of TADs, but the mechanisms of TAD establishment and maintenance, including their accurate duplication during genome replication, are not well characterized. To address this gap, we used high-throughput imaging-based CRISPR/Cas9 knock-out screening to discover chromatin factors involved in maintenance and establishment of TADs. Among the cellular factors that affect TAD organization, we found enrichment for cell cycle proteins, especially components of the DNA replication machinery. Accordingly, we demonstrate that TADs undergo temporary unfolding during S-phase DNA replication and that interference with progression through replication impedes restoration of normal TAD structure. Mechanistically, inhibition of the RPA complex prevents cohesin and CTCF binding, delays post-replication TAD re-folding and affects TAD folding in non-cycling cells. These results provide novel insights into how TAD structures are re-established during genome duplication.

cell biology↗

bioRxiv2026

Ubiquitin-dependent recruitment of SLFN11 to chromatin is regulated by deubiquitinase and RNF168

The molecular mechanisms driving SLFN11 chromatin recruitment remain partially elucidated. Using high-throughput imaging of 162 oncology-focused compounds in U2OS cells with inducible SLFN11 expression, we discovered that deubiquitinase (DUB) inhibitors drive massive SLFN11 recruitment to chromatin, preferentially at promoter regions while concurrently suppressing transcription. DUB inhibitors such as VLX-1570 promote ubiquitin-dependent enrichment of SLFN11 without detectable DNA damage, distinct from the camptothecin-induced RPA-associated SLFN11 foci formed at stressed replication forks. Yet, SLFN11 chromatin recruitment both by DUB inhibitors and DNA damage are suppressed by TAK243 demonstrating their ubiquitylation dependency. RNF168 is required for SLFN11 ubiquitylation and its subsequent chromatin association, and ubiquitylation within SLFN11s middle linker domain (lysines 390, 391, and 429) with K27-linked polyubiquitin chains is essential for the chromatin recruitment of SLFN11. These findings suggest the importance of SLFN11 ubiquitylation by RNF168 for SLFN11 chromatin recruitment and SLFN11 transcriptional regulatory role at promoter regions.

cell biology↗

bioRxiv2026

Kinetic proofreading as a mechanism for transcriptional specificity in living human cells

How target genes selectively respond to their specific transcription factors (TF) amid the vast excess of non-specific TFs in the nucleus remains a fundamental question in gene regulation. Here, we develop an integrated single-molecule imaging framework to quantitatively link TF dynamics with nascent transcription kinetics at endogenous gene loci to address the transcription factor specificity. Using endogenously Halo-tagged glucocorticoid receptor (GR), we show that ligand activation rapidly increases GR chromatin binding and residence times without substantially altering search kinetics. Live-cell nascent RNA imaging reveals that GR is essential for Dex induction of ERRFI1 transcription via increased burst frequency. Gene locus-specific, dual-color tracking demonstrates that GR display longer residence times in general near its target gene ERRFI1 compared to the non-target MYH9 locus, consistent with a kinetic proofreading model. A high-throughput imaging-based CRISPR screen identifies ATP-dependent processes, including neddylation and chromatin remodeling, and similarly, acute inhibition of the TFIIH XPB selectively impairs ERRFI1 transcription while sparing MYH9, implicating ATPase activity in GR target discrimination. In all, these findings establish that promoters function as dwell-time rather than occupancy detectors, discriminating specific from non-specific TF interactions through energy-dependent mechanisms.

biophysics↗

bioRxiv2020

A Deep Learning Pipeline for Nucleus Segmentation

Deep learning is rapidly becoming the technique of choice for automated segmentation of nuclei in biological image analysis workflows. In order to evaluate the feasibility of training nuclear segmentation models on small, custom annotated image datasets that have been augmented, we have designed a computational pipeline to systematically compare different nuclear segmentation model architectures and model training strategies. Using this approach, we demonstrate that transfer learning and tuning of training parameters, such as the composition, size and pre-processing of the training image dataset, can lead to robust nuclear segmentation models, which match, and often exceed, the performance of existing, off-the-shelf deep learning models pre-trained on large image datasets. We envision a practical scenario where deep learning nuclear segmentation models trained in this way can be shared across a laboratory, facility, or institution, and continuously improved by training them on progressively larger and varied image datasets. Our work provides computational tools and a practical framework for deep learning-based biological image segmentation using small annotated image datasets.

cell biology↗