Skip to main content

An experimental and bioinformatic toolbox for functional epigenomics and its application to epigenetically making and breaking a cancer cell

Periodic Reporting for period 3 - EpigenomeProgramming (An experimental and bioinformatic toolbox for functional epigenomics and its application to epigenetically making and breaking a cancer cell)

Reporting period: 2019-12-01 to 2021-05-31

The emergence of a complex organism requires an impressive degree of coordination and cooperation between cells. The human body is made up of about 30 to 40 trillion cells that work together to maintain bodily functions. These cells follow a genetic program of specialization and cooperation that provides for controlled growth as well as the planned death of cells that are no longer needed. This collaboration is only possible because almost all the cells of a body carry roughly the same genetic material. Darwinian selection therefore is suspended for an evolutionary moment: for the survival of common genes it no longer matters whether a cell propagates itself or its sister cells take on this task. Therefore, the cells of the body do not compete for scarce resources, but complement and support each other.

But how do cells specialize when they all share the same genetic material? Evolution has chosen the second and third dimension: while the letters of the DNA remain unchanged, the DNA molecules in the cell are elaborately packaged, twisted, and wound up. This way, it not only becomes possible to fit two meters of DNA into the microscopically small cells, but the packaging also provides a way to control which genes can be used by a specific cell in the body. For example, an insulin-producing cell in the pancreas can activate its insulin gene at any time because it is freely accessible in the middle of the cell nucleus. However, it cannot readily activate brain-specific genes; they are rolled up and locked away on the inner wall of the cell nucleus. This intricate level of regulation prevents the wrong genes from being inadvertently activated and thereby confusing the regulatory state of the cells. The additional layer of gene regulation in two and three dimensions is often referred to as "epigenetics" or "epigenomics".

Epigenetic gene regulation appears to be ubiquitous in cancer, in the sense that all cancers that have yet been studied in detail show widespread and cancer-specific epigenetic alterations: As a result of environmental influences or simply by chance, cells are sometimes epigenetically deprived of access to important genes. Normally the affected cells die and do no further damage. But sometimes it hits important genes that regulate the growth of the cell. Then it can happen that a single cell stops its cooperation with the organism and begins with unregulated growth - a tumor arises. In my laboratory at the CeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, we investigate the role of epigenetic changes in the development of cancer. In the context of my ERC project, we develop and apply wet-lab and computational methods that help us understand which of the many epigenetic alterations observed in cancer have a functional role.
We have performed extensive work on the role of epigenetics in cancer, and we have achieved four main results:

1. Development and application to chronic lymphocytic leukemia of a combined wet-lab and computational method for identifying epigenetic changes that carry functional relevance in cancer (Rendeiro et al. in revision, preprint available on bioRxiv).

Summary: Chronic lymphocytic leukemia (CLL) is a heterogeneous disease. Nevertheless, the BTK inhibitor ibrutinib provides effective treatment for the vast majority of CLL patients. To define the underlining regulatory program, we analyzed high-resolution time courses of ibrutinib treatment in closely monitored patients, combining cellular phenotyping (flow cytometry), single-cell transcriptome profiling (scRNA-seq), and chromatin mapping (ATAC-seq). Our study describes the cellular, molecular, and regulatory effects of therapeutic B cell receptor inhibition in CLL at high temporal resolution, and it establishes a broadly applicable method for epigenome/transcriptome-based treatment monitoring.

2. Development and validation of the KPNN method for interpretable deep learning on gene-regulatory networks, with the goal of inferring causal epigenetic and gene-regulatory mechanisms in cancer (Fortelny et al. submitted).

Summary: Deep learning has emerged as a powerful methodology for predicting a variety of complex biological phenomena. Here we demonstrate deep learning on biological networks, where every node and every edge has a direct molecular equivalent and interpretation. To enable interpretable deep learning on KPNNs, we introduce three methodological advances in the learning algorithm. We demonstrate the power of our approach on two single-cell RNA-seq datasets. KPNNs thus combine the predictive power of deep learning with the interpretability of biological networks.

3. Development and initial application of CROP-seq, a breakthrough technology for evaluating the cellular response to thousands of genetic/epigenetic perturbations in parallel using CRISPR single-cell sequencing (Datlinger et al. 2017 Nature Methods).

Summary: CRISPR-based genetic screens are accelerating biological discovery, but current methods have inherent limitations. We combined pooled CRISPR screening with single-cell RNA sequencing into a broadly applicable workflow, directly linking guide RNA expression to transcriptome responses in thousands of individual cells. Our method for CRISPR droplet sequencing (CROP-seq) enables pooled CRISPR screens with single-cell transcriptome resolution, which will facilitate high-throughput functional dissection of complex regulatory mechanisms and heterogeneous cell populations.

4. Development and application to chronic lymphocytic leukemia of a combined wet-lab and computational approach for prioritizing drug combinations for leukemia therapy using epigenome mapping and chemosensitivity profiling (Schmidl et al. 2019 Nature Chemical Biology)

Summary: The BTK inhibitor ibrutinib has substantially improved therapeutic options for chronic lymphocytic leukemia (CLL). Although ibrutinib is not curative, it has a profound effect on CLL cells and may create new pharmacologically exploitable vulnerabilities. To identify such vulnerabilities, we developed a systematic approach that combines epigenome profiling (charting the gene-regulatory basis of cell state) with single-cell chemosensitivity profiling (quantifying cell-type-specific drug response) and bioinformatic data integration. Here we identified ibrutinib-induced changes as a starting point for rational design of ibrutinib combination therapies, and we establish a broadly applicable method for investigating treatment-specific vulnerabilities by integrating epigenetic cell states and phenotypic drug responses.
During the first half of this ERC project, important methodological advances have been made, allowing us to identify with increased precision which of the many epigenetic alterations observed in cancer may carry functional relevance in cancer.
During the second half of the project, the main focus will shift toward the development of methods that make it possible to induce epigenetic changes in a targeted and predictable manner, and on the application of these methods to cancer.

In summary, this ERC project has already achieved substantial progress beyond the state of the art, mainly through the development of powerful and widely applicable methods that combine elements of wet-lab and computational research, and it is expected to provide important further advances, results, and insights with direct relevance to cancer epigenetics.
CROP-seq - a breakthrough method for CRISPR single-cell sequencing
DNA methylation - an epigenetic mark closely linked to cancer