Magda Bienko

In eukaryotic cells, ~50% of the nucleus is filled with proteins and RNAs, many of which are closely associated with DNA. In exciting preliminary work, we discovered a new class of nuclear RNAs that form genome-wide, long-range contacts with DNA and mainly consist of intronic sequences. During neurodifferentiation, these trans-contacting intronic RNAs (TIRs) are produced from very long, neuronally expressed, protein-coding genes enriched in risk loci for neurodevelopmental disorders (NDDs). TIRs form an intricate localization pattern in the nucleus, which we hypothesize is critical for shaping the local and global 3D genome structure, as well as for coordinating gene expression in brain cells. Here, we tackle this hypothesis by first charting TIRs in neurons/glial cells differentiated from hESCs/iPSCs either from healthy donors or NDD subjects (Aim1). We then pioneer MARINDA for simultaneous all-way profiling of DNA and RNA contacts in single cells and apply it to reconstruct the 3D DNA-DNA, RNA-DNA, and RNA-RNA connectome in mouse brain cells (Aim2). Finally, we combine antisense oligonucleotide and genetic engineering techniques to silence selected TIRs or re-localize them within the nucleus and then measure the effects of these perturbations at the genetic, epigenetic, and phenotypic level in neurons (Aim 3). This project has the potential to establish a new fundamental role of nuclear RNAs in neurobiology and uncover new pathomechanisms for NDDs.

In exciting preliminary experiments leveraging our GPSeq method we discovered that the genome of mammalian cells in interphase folds into a steep radial gradient of guanine and cytosine (GC) density, which seems to persist at the level of individual mitotic chromosomes. However, we still lack a fundamental understanding of how this higher-order 3D genome architecture is established and what its functional implications are. Here, I go beyond the state-of-the-art and propose that the observed steep radial GC-gradient is a universal design principle of the radial arrangement of the genome in the nucleus—which I call the radiality principle—that provides a biophysical framework for spatially orchestrating key nuclear processes, beyond gene expression regulation. To test this hypothesis, in this project I pursue five Objectives: (1) First, we develop a novel approach (GP-C) for high-resolution single-cell 3D genome reconstructions to study whether the radial GC-gradient is indeed a universal property of nuclei across different species and cell types. (2) Next, we apply GP-C together with RNA-seq to monitor genome radiality and concurrent gene expression changes as cells undergo karyotype rewiring or significant epigenetic perturbations. (3) In parallel, we develop innovative approaches to probe the internal structure of mitotic chromosomes and model how genome radiality is reorganized as cells traverse mitosis. (4) We then expand our preliminary finding of large-scale DNA-RNA contact hubs that seem to shape cell-type specific radiality landscapes by opposing the radial GC-gradient. (5) Finally, we pioneer methods for profiling nuclear proteins radially and apply them to test the bold hypothesis that the radiality principle provides a blueprint for organizing numerous nuclear processes. This project aims at conclusively addressing long-standing questions in the field of 3D genome biology and proposes novel mechanisms of nuclear function regulation.