Studies on genetic and epigenetic regulation of gene expression dynamics

The information required to build an organism is contained in its genome and the first biochemical process that activates the genetic information stored in DNA is transcription. Cell type specific gene expression shapes cellular functional diversity and dysregulation of transcription is a central tenet of human disease. Therefore, understanding transcriptional regulation is central to understanding biology in health and disease. Transcription is a dynamic process, occurring in discrete bursts of activity that can be characterized by two kinetic parameters; burst frequency describing how often genes burst and burst size describing how many transcripts are generated in each burst. Genes are under strict regulatory control by distinct sequences in the genome as well as epigenetic modifications. To properly study how genetic and epigenetic factors affect transcription, it needs to be treated as the dynamic cellular process it is. In this thesis, I present the development of methods that allow identification of newly induced gene expression over short timescales, as well as inference of kinetic parameters describing how frequently genes burst and how many transcripts each burst give rise to. The work is presented through four papers:

In paper I, I describe the development of a novel method for profiling newly transcribed RNA molecules. We use this method to show that therapeutic compounds affecting different epigenetic enzymes elicit distinct, compound specific responses mediated by different sets of transcription factors already after one hour of treatment that can only be detected when measuring newly transcribed RNA.

The goal of paper II is to determine how genetic variation shapes transcriptional bursting. To this end, we infer transcriptome-wide burst kinetics parameters from genetically distinct donors and find variation that selectively affects burst sizes and frequencies.

Paper III describes a method for inferring transcriptional kinetics transcriptome-wide using single-cell RNA-sequencing. We use this method to describe how the regulation of transcriptional bursting is encoded in the genome. Our findings show that gene specific burst sizes are dependent on core promoter architecture and that enhancers affect burst frequencies. Furthermore, cell type specific differential gene expression is regulated by cell type specific burst frequencies.

Lastly, Paper IV shows how transcription shapes cell types. We collect data on cellular morphologies, electrophysiological characteristics, and measure gene expression in the same neurons collected from the mouse motor cortex. Our findings show that cells belonging to the same, distinct transcriptomic families have distinct and non-overlapping morpho-electric characteristics. Within families, there is continuous and correlated variation in all modalities, challenging the notion of cell types as discrete entities.

List of scientific papers

I. Leonard Hartmanis, Gert-Jan Hendriks, Daniel Ramsköld, Per Johnsson, Christoph Ziegenhain, Anton Larsson, Salomé Hahne and Rickard Sandberg. Deciphering direct transcriptional effects of epigenetic compounds by large-scale new RNA profiling. [Manuscript]

II. Leonard Hartmanis*, Christoph Ziegenhain*, Daniel Ramsköld, Michael Hagemann-Jensen and Rickard Sandberg. Inference of bursting QTLS to decipher the regulation of transcriptional bursting. *Equal contribution. [Manuscript]

III. Anton JM Larsson, Per Johnsson, Michael Hagemann-Jensen, Leonard Hartmanis, Omid Faridani, Björn Reinius, Åsa Segerstolpe, Chloe Rivera, Bing Ren and Rickard Sandberg. Genomic encoding of transcriptional burst kinetics. Nature. 2019, 565, 251-254.
https://doi.org/10.1038/s41586-018-0836-1

IV. Federico Scala, Dmitry Kobak, Matteo Bernabucci, Yves Bernaerts, Cathryn René Cadwell, Jesus Ramon Castro, Leonard Hartmanis, Xiaolong Jiang, Sophie Laturnus, Elanine Miranda, Shalaka Mulherkar, Zheng Huan Tan, Zizhen Yao, Hongkui Zeng, Rickard Sandberg, Philipp Berens and Andreas S. Tolias. Phenotypic variation of transcriptomic cell types in mouse motor cortex. Nature. 2021, 598, 144-150.
https://doi.org/10.1038/s41586-020-2907-3