Research Projects

Our group is trying to better understand how transcriptional droplets form, which proteins they contain and how they are used to control transcriptional programs. To this end, we are using a multi-disciplinary approach that involves biochemistry, cellular biology and close collaborations with biophysicists

Transcriptional Condensates: From Formation to Function

This PhD project explores how pioneer transcription factors (PTFs) use biomolecular condensation to remodel chromatin structure. While PTFs are known to access DNA within compacted chromatin, the role of their condensates in facilitating chromatin opening remains unclear. To address this, chromatin is reconstituted in vitro using fluorescently labeled histones and defined DNA templates, and its interaction with PTFs is studied using single-molecule techniques. These include optical tweezers to detect nucleosome unwrapping through force spectroscopy, and the carpet assay to visualize condensate formation on surface immobilized chromatin via TIRF microscopy. Together, these approaches allow to dissect how PTF condensation influences chromatin accessibility at the single-molecule level.

PTF Condensates in Chromatin Remodeling

The Role of Biomolecular Condensates in the Pancreatic Transcription Factor Network

This project aims to understand whether phase separation among pancreatic transcription factors (TFs) plays a role in endocrine lineage commitment. The transition from pancreatic progenitor to endocrine progenitor is driven by a hierarchically organised TF network whose sequential activation must be tightly controlled to achieve switch-like precision. While TFs are classically understood to act through DNA binding and co-factor recruitment, recent models highlight the role of intrinsically disordered regions (IDRs) in the formation of condensates to regulate gene expression. Many core TFs in this network carry extended IDRs with sequence features predictive of phase separation, suggesting that condensation may be part of the machinery driving endocrine cell fate commitment. To investigate this, we use recombinant proteins to characterise how individual TFs and their partners form condensates in vitro, and how DNA shapes this behaviour. We then use inducible cellular systems to ask whether condensate formation changes a TF's transcriptional output and chromatin accessibility, and finally track condensation during differentiation to test whether it is required for lineage commitment. Together, these approaches allow us to dissect whether condensation is a functional part of how pancreatic TFs control cell fate.

Figure: Carpet assay: DNA is immobilised on a functionalised cover slip. Top: TIRFM images of 300 nM (blue) and 1500 nM (red) NHA9-Alexa594 bound to l-phage DNA. Bottom: Intensity profile showing that at 300 nM, NHA9 adsorbs to DNA; at 1500 nM, distinct high-intensity foci form. Quantification of TIRFM data reveals more NHA9 molecules per focus at 1500 nM than can physically adsorb to the DNA, confirming condensate formation. At 300 nM, the spots contain far fewer molecules. TIRFM image of plasmid DNA (Sytox Orange, red) stained with catalytically inactive (or dead Cas9) dCas9–sgRNA complexes (green) shows successful targeting of different genomic regions on the same DNA molecule.

This research, aims to understand how transcriptional condensates (also known as transcriptional droplets) form, which macromolecules they contain, and how they help regulate gene expression.

To study this, we use purified proteins and test their ability to form condensates in the laboratory, both with and without DNA. Using advanced single-molecule assays, we can observe these condensates in real time and measure their physical properties. A key approach in our work is the use of single-molecule in vitro assays, which allow us to reconstitute and analyse transcriptional condensates under highly controlled conditions. In our "carpet assay", individual DNA molecules are immobilised on a glass surface and visualised using total internal reflection fluorescence microscopy (TIRFM). This platform enables high-throughput testing of different experimental conditions and has already been used to demonstrate the formation of NUP98-HOXA9 condensates on DNA.

We are currently adapting this workflow to investigate additional transcription factors. While earlier studies relied on lambda phage DNA, we have developed a custom 35 kb DNA construct that can incorporate specific transcription factor binding sites. Combined with fluorescently labelled dCas9 and guide RNAs, this system allows us to mark precise locations on the DNA and directly observe the recruitment of transcription factors to their target sites. By introducing specific mutations into proteins, we can identify which regions are important for condensate formation and function. We then introduce these mutations into human cells and use next-generation sequencing technologies to determine how they affect gene expression.

Klf4 as a Molecular Key: Unlocking Facultative Heterochromatin to Drive Cellular Identity Changes

This project investigates the reprogramming potential of Klf4, a pioneer transcription factor and one of the four canonical Yamanaka factors critical for cellular identity transitions. By introducing Klf4 into a Klf4-negative “blank slate” human cell line, the study aims to delineate its capacity to access and remodel facultative heterochromatin—regions of the genome that are developmentally poised yet remain transcriptionally silent. Through engagement with these compacted chromatin domains, Klf4 may unlock gene expression programs essential for cell fate determination.

To achieve a comprehensive understanding of Klf4’s reprogramming dynamics, the approach combines an inducible expression system with single-cell RNA sequencing, enabling precise temporal control over Klf4 activation and high-resolution mapping of transcriptional responses at the single-cell level.

A central focus of the project is to assess the impact of both native Klf4 and MBP-tagged Klf4 on the formation and regulation of transcriptional condensates—phase-separated nuclear compartments thought to organize and coordinate gene expression. Comparative transcriptomic profiling will be performed to determine how the MBP tag influences Klf4-mediated condensate formation and alters global gene expression patterns. This strategy will reveal mechanistic insights into the role of Klf4 in chromatin remodeling and transcriptional regulation, as well as potential artefacts introduced by common tagging approaches.