Here I propose to use small molecules to degrade proteins specifically around sites of DNA damage by using the damage itself as a homing signal. The approach will create new ways to study DNA damage, but will also offer translational possibilities in cancer. Cancer cells are often acutely sensitive to DNA damage because they have one or more faulty DNA damage response pathways - a feature that makes them highly dependent on their remaining DNA repair systems. We will pioneer two novel and related chemical approaches for selectively degrading proteins by modulating DNA damage response pathways with bifunctional DNA damaging molecules. We will do this by reprogramming E3 ligases. E3 ligases are multi-protein complexes that catalyse the formation of polyubiquitin chains on its substrates, leading to their degradation in the protein recycling station known as the proteasome. A recent revolutionary advance in chemical biology is to use small molecules to change the specificity of E3 ligases, leading to the degradation of user-defined proteins. By degrading proteins instead of inhibiting them, these small molecules achieve levels of functional modulation typically only possible with genetic techniques. We are inspired by this new protein degradation technology, but will take it in a new direction. Chemical damage of DNA recruits E3 ligases as well as critical DNA damage response proteins in preparation for DNA repair. We will invent a new generation of small molecule protein degradation catalysts and reagents by repurposing these natural responses to DNA damage. We will accomplish our goal with three aims: Aim 1: Use DNA damage as a homing signal for induced protein degradation Aim 2: Use direct repair of DNA damage by the repair protein methylguanine methyltransferase (MGMT) to promote the degradation of other proteins Aim 3: Promote pleiotropic protein degradation by recruiting broadly acting E3 ligases to sites of DNA damage I propose an ambitious project that will create conceptually novel ways to study the DNA damage response and potentially build new medicines.

RNA therapeutics suffer from stability, immunogenicity, and delivery problems. We will take inspiration from two recent discoveries in biology to help us build better RNA therapeutics: 1. Methylated RNAs are substrates for direct reversal repair enzymes and their methylation state controls their function; 2. Exosomes are natural delivery systems for RNAs of all sizes. The fact that Nature uses alkyl groups to control the function of RNAs suggests that chemists could too. I will study which alkyl groups on RNA get repaired, and which do not, facilitating the design of RNAs with alkyl groups that can be permanent or temporary. This information will help us graft drug-like properties into RNA by judicious installation of specific alkyl groups. For RNA therapies to be successful they need to stay out of the liver and reach their targets. In another major objective we will use a reaction developed in my group to functionalize the surface of exosome nanoparticles isolated from cells and use these as vehicles to deliver our synthetic RNAs.

Despite great advances over the past decade, the field of bioconjugation remains deficient in some key areas. For example, reactions with high rate constants in biological media are important but rare. Since the rate of chemical coupling reactions decrease quadratically with concentration, high-rate constant reactions are essential to monitor biological processes at relevant concentrations. My group has developed a series of high-rate constant coupling reactions and in the present proposal we outline how these reactions could be exploited in chemical biology.

We will build a general and broadly-applicable platform technology for the targeted delivery of drugs. In contrast to current approaches our targeting motifs will derive from libraries of modular small molecule macrocyclic scaffolds, opening the door to a variety of previously inaccessible targets. Our guided drugs will be designed to be released at the target through the development of smart linkers that respond to a disease microenvironment or cell receptor. The drugs themselves will derive from known compounds complemented by designs created in the Schneider group. Many pharmaceutical agents are unselective, causing toxicity to normal organs and preventing dose escalation to therapeutically active regimens. Selectively delivering and activating drugs at the site of disease holds great promise for dramatically increasing the therapeutic index of bioactive molecules, thus providing substantial benefits to patients. A successful general platform that exploits small ligands for pharmacodelivery applications would represent a transformative innovation for both academia and industry. Current strategies for targeted drug delivery rely mainly on antibodies as "vehicles", but recent disappointments in Phase III clinical trials (>5 high-profile clinical failures) with antibody-drug conjugates call into question their real value. Moreover, because of their large size and unacceptably high cost, antibodies can only be used for delivering ultra-potent drugs, hindering development opportunities with conventional pharmaceutical agents and in areas outside oncology. A handful of naturally-occurring small organic ligands for common receptors (folate receptor, carbonic anhydrase IX) have been proposed as an alternative to antibodies for pharmacodelivery strategies, but a general approach that avoids large biomolecules has not been described due to limitations in ligand discovery and to the lack of suitable strategies for drug release at the site of disease. The ability to create "smart drugs" will emerge from advances made in each of the subgroups: Specifically, we will construct encoded chemical libraries of unprecedented size and quality (dozens of millions of compounds) and screen these libraries for ligands against at least ten validated accessible markers of cancer and chronic inflammation. In addition, we will develop and implement innovative technologies for the smart release and activation of bioactive payloads at the site of disease. The linker/release triggers for therapeutic payloads will be tailored to respond to the disease microenvironment and to selectively act on the target cells of choice. In addition to employing well-known cytotoxic agents like tubulin inhibitors (DM1, MMAE) as payloads, we will primarily explore the use of membrane-disruptive peptides. These designed cytotoxic peptides have the unique advantage of (i) directly addressing the plasma membrane of cancer cells as target without the involvement of proteins, which minimizes the risk of cancer cell escape by mutation or development of resistances; (ii) cancer cell-selective membrane targeting as a result of our molecular design aiming at custom-tailored highly potent cytotoxic peptide payloads; and (iii) rapid plasma clearance so that the peptides, once released from the carrier, will act only locally at the site of release, thereby minimizing unwanted off-target and side-effects. The proposal is bold, requiring expertise in chemical, biochemical, and computational technology - a combination beyond the reach of any one research group. The synergy created through the participation of three complementary teams is essential. Prof. Neri (ETH Zürich) will be responsible for the production of target proteins, for the construction and screening of DNA-encoded libraries of chemically-modified cyclic peptides, for the development of small molecule-drug conjugates and for in vivo testing of the most promising compounds in mouse models of cancer and of rheumatoid arthritis. Prof. Gillingham (University of Basel) will be responsible for creating macrocycle libraries, importing them into the DNA encoded format, as well as developing new redox-responsive cleavable linkers. Prof. Schneider (ETH Zürich) will be responsible for the computational design, chemical synthesis, biochemical testing and engineering of novel peptide cytotoxins that kill target cells from the outside by direct cancer cell disruption.

Nature uses reversible methylation of DNA and RNA to regulate genes, and many organisms have the capacity to repair alkylative damage to DNA and RNA. In this project we will explore whether native RNA repair enzymes can be co-opted to control the release of chemically engineered RNAs in vitro and in vivo.

Understanding and controlling the biology of nucleic acids is partially a challenge in synthetic chemistry: How to precisely manipulate a single subunit in an ocean of near-identical units. Addressing this challenge in chemo- and site-selectivity is vital since tinkering with a molecules structure offers one of the best ways to learn how it works. Tinkering also offers a pathway to refine and even create new function. Evolution is exactly this sort of iterative variation to craft function. Accordingly, exploring and expanding the role of a particular DNA or RNA would be facilitated by synthetic access variants consisting of diverse structural perturbations. This task is especially urgent for RNA since every year new layers in its rich chemistry and biology are unveiled. Enzymes are Nature's solution to whole host of selectivity issues including functional group compatibility, chemoselectivity, and site-selectivity. Their outstanding selectivity stems not only from the primary active-site interactions that promote catalysis, but also from the secondary interactions that guide substrate selection and binding. Drawing inspiration from Nature's two-pronged approach, we will identify catalysts for the modification of nucleic acids and then append these catalysts to guiding sequences that precisely define the sites of alkylataion through Watson-Crick base-pairing. New chemical methods to selectively modify nucleic acids are important because of their central role in biology and medicine.