Evolution and function of DNA elements involved in cardiovascular inflammation

Transcription factors bind to tens of thousands of sites in the genome, but only some of these binding events actually change gene expression. Separating the functional sites from the rest is one of the central problems in human gene regulation. To tackle it, we use a "Zoo-ChIP" approach: we map where a transcription factor binds in the same cell type across several species, then ask which binding sites have been conserved and which are species-specific.

We apply this approach to the inflammatory response of vascular endothelial cells, the cells that line blood vessels. When these cells sense inflammatory signals, the transcription factor NF-κB rapidly switches on hundreds of genes, and chronic activation of this program contributes to atherosclerosis and other cardiovascular diseases. Comparing NF-κB binding across mammals has helped us find the elements that control this response. It also revealed that NF-κB often binds DNA that is still wrapped around nucleosomes, and that this feature is conserved at the large "super-enhancer" regions around key inflammatory genes (Alizada et al. 2021).

We are also interested in where new regulatory elements come from. Many NF-κB binding sites lie within transposable elements, mobile DNA sequences that have copied themselves throughout the genome over evolutionary time. Some of these were co-opted long ago and are shared across mammals, while others are recent and specific to individual lineages, giving each species its own variations on the inflammatory response (Wang et al. 2025).

We work closely with the labs of Dr. Jason Fish, Dr. Jennifer Mitchell and Dr. Sushmita Roy on these projects.

Evolution and function of DNA elements involved in early heart development

The heart is the first organ to form in the embryo, and the transcription factors that build it are remarkably similar from fish to humans. Much less is known about the regulatory elements that tell these factors which genes to switch on, in which cells, and at what time. Our goal is to find these elements and use them to understand the gene regulatory networks that first set cells on the path to becoming heart.

Zebrafish are ideal for this work: their embryos are transparent and develop outside the mother, so we can watch reporter genes switch on in living embryos. When we profiled open chromatin in zebrafish during gastrulation, the stage when the body plan is first laid out, we were surprised to find roughly 6,000 regions whose matching sequences are also open in the human genome. These accessible conserved non-coding elements (aCNEs) sit near genes that control development, many of which are also targets of Polycomb repressive complexes. Among them, we found a set of ancient elements that are especially active in the cells that will go on to form the heart (Yuan et al. 2018). Following these elements led us to a conserved network of GATA transcription factors that helps early mesoderm cells choose between becoming heart or pharyngeal tissue (Song, Yuan et al. 2022).

Only a small number of aCNEs have been tested so far. We are now testing many more in zebrafish and beginning to ask how human genetic variants that fall within these deeply conserved elements affect their activity.

This project is carried out together with the lab of Dr. Ian Scott.

Deciphering gene regulatory networks involved in childhood development

The timing of puberty varies widely between children and is strongly influenced by genetics. Large human studies have linked hundreds of genomic regions to pubertal timing, but for most of them it is unclear which genes are involved or where and when they act. To help connect these genetic signals to biology, we have mapped how gene expression changes across postnatal development and the pubertal transition in mice, with attention to differences between sexes and between cell types in the hypothalamus and pituitary, the brain regions that control puberty (Hou et al. 2022; Sokolowski et al. 2024).

We are now studying an animal that takes pubertal control to an extreme. Naked mole-rats live in large colonies in which only a single breeding female and a few males reproduce; the other members can remain pre-pubertal for years, as long as they stay in their colony, but can become reproductively active when removed from it. Understanding how this suppression works at the level of genes and their regulation required a high-quality genome, so we produced a new chromosome-level assembly and annotation of the naked mole-rat genome. It is now available on Ensembl and serves as a reference for the research community (Sokolowski et al. 2024).

We work closely with the labs of Dr. Mark Palmert and Dr. Melissa Holmes on these projects.

Understanding extreme adaptations in rodents using genome sequencing and comparative epigenomics

Rodents make up about 40% of all mammal species and live in nearly every environment on Earth, from deserts to the Arctic. This diversity makes them a natural experiment for asking how genomes change as animals adapt to new conditions. As part of the Rodent 2K consortium, which aims to sequence the genomes of a broad range of rodent species, we are combining genome sequencing with comparative epigenomics to find the genetic and regulatory changes behind unusual rodent traits.

One focus is the spiny mice (Acomys). We are sequencing the genomes of several Acomys species, including the golden spiny mouse, Acomys russatus, which lives in some of the driest deserts of the Middle East and can survive on a high-salt diet with very little water. How its kidneys and other tissues manage salt and water under these conditions is not well understood. By comparing its genome and gene regulation with those of related species that live in less extreme environments, we hope to learn how this balance is achieved, with potential lessons for human kidney and cardiovascular health.

This work is done in collaboration with many labs, including those of Dr. Noga Kronfeld-Schor, Dr. Mathieu Lemaire and Dr. Jared Simpson

Genome biology of topoisomerase II beta

DNA inside the nucleus is long, tightly packed and constantly being unwound to be read and copied. This creates tangles and twisting strain that must be released. Type II topoisomerases solve the problem by cutting both strands of the DNA double helix, passing another stretch of DNA through the gap, and sealing the break again. One of these enzymes, topoisomerase II beta (TOP2B), is particularly active near genes that are being transcribed.

We found that TOP2B binds the genome together with CTCF and cohesin, proteins that fold DNA into loops and domains (Uusküla-Reimand, Hou et al. 2016). Because TOP2B briefly breaks DNA as part of its normal work, the places where it acts can become weak spots. When these breaks are not repaired properly, they can lead to mutations and to chromosome rearrangements, including those seen in some leukemias that arise after chemotherapy. We are now studying how TOP2B activity shapes the pattern of mutations found in cancer genomes (Uusküla-Reimand et al. 2024), and more broadly how it links transcription, genome folding and genome stability.

This work has been done with several collaborating labs, including those of Dr. Jüri Reimand, Dr. Neal Sondheimer, Dr. Anne-Claude Gingras and Dr. Caroline Austin.

Translational genomics

Many of the questions we study in the lab have direct relevance to patients, and much of our work is done alongside clinicians at SickKids and elsewhere. Our trainees often work on these projects, which bring the genomic and gene regulatory approaches we use in basic research to bear on rare diseases.

Genome sequencing now finds a diagnosis for many children with suspected genetic conditions, but many families still receive no answer. Often a variant has been found but its effect is unclear, or the cause lies in a part of the genome that is hard to interpret. Measuring RNA can help, because it shows directly whether a gene is expressed at the wrong level or spliced incorrectly. Working with the Department of Paediatric Laboratory Medicine (DPLM) and The Centre for Applied Genomics (TCAG) at SickKids, we helped bring RNA sequencing from blood into clinical practice and studied how much it adds for children who remained undiagnosed after genome sequencing (Hou et al. 2025).

We also study rare diseases caused by changes to regulatory DNA itself. Quebec platelet disorder is an inherited bleeding disorder caused by a duplication in the genome. By combining epigenomic profiling of patient cells with cross-species comparisons and enhancer tests in human cells and zebrafish, we showed that the duplication places a conserved enhancer active in megakaryocytes, the cells that make platelets, next to a gene it does not normally control. This "enhancer hijacking" causes the gene to be switched on in the wrong place (Liang et al. 2020).