Tackling the Undruggable Cancer Proteome with Covalent Macrocycles
Lead Research Organisation:
University of Bath
Abstract
Each cancer patient has a unique combination of proteins that promotes the growth of their tumour. In stratified medicine, these proteins are identified, and a drug that inhibits their function is administered to the patient. This personalised approach is often highly effective and results in less side effects compared to standard chemotherapy. Unfortunately, this approach only works for 10% of patients because inhibitors for most cancer proteins are lacking. Many of these proteins are deemed ‘undruggable’ as they function through protein-protein interactions (PPIs), which are difficult to disrupt with small molecules as interfaces are typically void of binding pockets.
However, undruggable proteins possess reactive nucleophilic, or ‘ligandable’, hotspots that can be targeted with molecules called electrophiles. Electrophiles permanently inhibit the function of an undruggable protein by forming a covalent bond at a ligandable hotspot. Identifying a covalent inhibitor that selectively targets an undruggable cancer protein while leaving proteins present in our healthy tissues unmodified is a difficult process. If a drug lacks selectivity it can cause side effects for patients.
In this proposal, we will develop a phage display technology platform to enable billions of covalent peptide inhibitors (also known as targeted covalent macrocycles (TCMs)) to be rapidly generated and screened against undruggable protein targets. These molecules combine the properties of a peptide and an irreversible inhibitor, binding to shallow PPI interfaces with high affinity and selectivity by forming interactions over a large surface area and achieving permanent target engagement by covalent modification of a proximate ligandable hotspot. Crucially, we will incorporate a range of tyrosine-targeting electrophiles into our libraries as tyrosine residues are highly enriched at PPI interfaces.
To prototype our platform, we will identify a covalent peptide that modifies tyrosine 82 on Ras-like protein (RAL) and blocks interaction with guanine exchange factors (GEFs). RAL-GEF PPIs promote pancreatic cancer progression through multiple mechanisms but are considered undruggable as they lack binding pockets for small molecule engagement. Consequently, a RAL-selective drug could be highly beneficial to a patient population that suffers from limited treatment options.
Ultimately, this work will enable selective inhibitors to be rapidly identified for a variety of undruggable proteins, which in the long term will allow more cancer patients to benefit from personalised treatments.
However, undruggable proteins possess reactive nucleophilic, or ‘ligandable’, hotspots that can be targeted with molecules called electrophiles. Electrophiles permanently inhibit the function of an undruggable protein by forming a covalent bond at a ligandable hotspot. Identifying a covalent inhibitor that selectively targets an undruggable cancer protein while leaving proteins present in our healthy tissues unmodified is a difficult process. If a drug lacks selectivity it can cause side effects for patients.
In this proposal, we will develop a phage display technology platform to enable billions of covalent peptide inhibitors (also known as targeted covalent macrocycles (TCMs)) to be rapidly generated and screened against undruggable protein targets. These molecules combine the properties of a peptide and an irreversible inhibitor, binding to shallow PPI interfaces with high affinity and selectivity by forming interactions over a large surface area and achieving permanent target engagement by covalent modification of a proximate ligandable hotspot. Crucially, we will incorporate a range of tyrosine-targeting electrophiles into our libraries as tyrosine residues are highly enriched at PPI interfaces.
To prototype our platform, we will identify a covalent peptide that modifies tyrosine 82 on Ras-like protein (RAL) and blocks interaction with guanine exchange factors (GEFs). RAL-GEF PPIs promote pancreatic cancer progression through multiple mechanisms but are considered undruggable as they lack binding pockets for small molecule engagement. Consequently, a RAL-selective drug could be highly beneficial to a patient population that suffers from limited treatment options.
Ultimately, this work will enable selective inhibitors to be rapidly identified for a variety of undruggable proteins, which in the long term will allow more cancer patients to benefit from personalised treatments.