Human DNA contains two types of biologically instructive information: the canonical nucleobases A, G, T, C, and the epigenetic nucleobases mC, hmC, fC, and caC. Canonical nucleobases encode the molecular identity (the sequence) of all RNAs and proteins that are synthesized by a cell, whereas epigenetic nucleobases dynamically regulate this synthesis.
This regulation shapes the phenotype of cells during differentiation and development, and its perturbation is a key trigger of cancer.
Canonical nucleobases can be decoded in a programmable manner by nucleic acids and their analogs via Watson-Crick-base pairing, and the simplicity of this recognition has enabled revolutionary developments in the biological sciences.
In contrast, molecular scaffolds with programmable recognition of epigenetic nucleobases do not exist, which represents a roadblock for developments in epigenetics basic research and cancer diagnostics.
Likewise, the reversible writing, reading and erasing of epigenetic nucleobases (DNA modifications) during cell differentiation, embryonic development and cancer development is highly dynamic, but studying this dynamics is limited by the lack of methodology to write, read, and erase the nucleobases with high spatiotemporal resolution.
Cancer represents a leading cause of death in the human population accounting for ~10 Mio. deaths per year (2020), and the increasing life-expectation further intensifies this problem. Epigenetic nucleobases play key roles in cancer development and serve as important biomarkers for cancer research and diagnostics.
A striking feature of aberrant, cancer-causing nucelobase patterns are their early occurrence during malignant transformation, making them highly valuable biomarkers for early-stage detection. Early detection in turn drastically increases the treatment success rate, and has been identified by the WHO as key factor for reducing cancer deaths. Novel approaches that reveal the actual chemical signals at play thus bear a high potential for biomarker discovery and the development of diagnostics assays. EPICODE dealed with the design and application of novel designer proteins that address fundamental problems in the field of DNA recognition.
These represent basic research tools with potential to be applied in different study settings, including in vitro DNA analytics/Diagnostics, imaging approaches and functional studies.
Specifically, we developed probes that can be programmed to recognize epigenetic DNA nucleobases in user-defined DNA sequences, as well as recognize specific combinations of two epigenetic DNA nucleobases in the two strands of DNA duplexes. Such molecules can serve as versatile probes in vitro and in vivo, i.e. to enrich DNA molecules from complex, genomic DNA backgrounds for targeted and genome-wide analyses, and for direct cellular imaging. We further designed optochemical tools that allow to precisely control the writer and eraser enzymes as well as the reader proteins of epigenetic DNA nucleobases with spatiotemporal resolution by light directly in cells. This allowed us to precisely study the kinetics of nucleobase editing and the triggered downstream chromatin changes in states associated to normal and malignant physiology, and to edit the local epigenetic nucleobase landscape of user-defined genomic loci.
Given the central role of epigenetic nucleobases in cancer and the universality of this approach, EPICODE thus provided enabling and broadly applicable methodology for cancer epigenetics research and diagnosis.
