The main research goal of NCCR co-Director Christian Heinis and his team is the development of therapeutics based on the so-called bicyclic peptides, and its recent variation, to combine favorable properties of two major classes of therapeutics, the monoclonal antibodies and the small molecule drugs. In the interview, Christian Heinis clearly explains the reason behind this research interest, the innovative methodology used to create these peptides and the therapeutic potential behind some leads.
Your lab aims at developing drugs. How can chemical biology help with this?
Chemical biology has become an important discipline for the pharmaceutic industry, providing a range of powerful tools. It is for example used to study the molecular basis of diseases, to generate new drug leads or to identify the targets of drugs. In my laboratory, we are applying chemistry and biology for generating ligands that bind and modulate disease targets with the goal of developing new therapeutics. There are many diseases with limited or no treatment options. I am thinking for example of certain types of cancer, neurodegenerative disorders such as Alzheimer’s and many rare diseases.
What are the challenges in developing drugs?
The challenges are manifold and vary from one disease to another. In some cases, the molecular basis of the disease is not yet understood, and scientists simply do not know how to interfere with the medical condition. For other diseases, the mechanisms and potential drug targets are known, but it is difficult to develop molecules that efficiently bind and modulate them. Yet other challenges, including drug toxicity and safety, are faced in the pre-clinical and clinical phases of the drug development process. My laboratory is addressing the challenge of developing molecules to “difficult” disease targets that are not tractable to established drug modalities such as small molecule drugs.
Why is it difficult to generate ligands against certain disease targets?
Small molecule ligands can be developed most easily against proteins containing clefts or pockets into which small molecule drugs can nicely fit. In contrast, it is more challenging to target proteins with flat, featureless surfaces with small molecules, as they cannot form enough molecular contacts to reach strong binding affinities. This problem has been overcome in recent years with the establishment of monoclonal antibodies as a therapeutic class. Monoclonal antibodies have a large interaction surface and it is possible to develop antibodies against virtually any protein target. Last year, seven out of the ten best-selling drugs were monoclonal antibodies or protein-antibody Fc fusion proteins. Unfortunately, antibodies also have their limitations. For example, they need to be administered by injection, they diffuse only slowly into tissues, they have limited stability as they can unfold or precipitate and they cannot be chemically synthesized, which complicates the production and hinders the modification with alternative chemical building blocks.
How could the limitations of monoclonal antibodies be overcome?
All limitations of monoclonal antibodies I named are linked to their large size, being 150,000 kDa and thus nearly 1000-fold larger than small molecule drugs. Scientists have tried to make antibodies smaller by dissecting them into fragments or by designing peptides mimicking the target-binding face, the paratope. Since the rational design of such peptidic antibody mimetics was challenging, I had proposed a strategy based on phage display, in which peptide mimetics are developed by in vitro evolution. The peptides named “bicyclic peptides” contain two peptide loops that can bind to protein targets in a similar way as antibodies surface loops, the complementarity determining regions (CDRs). I had developed this strategy together with Sir Greg Winter at the LMB in Cambridge, UK, and we were able to obtain high-affinity ligands to different protein targets.
Can you explain how the bicyclic peptides are developed?
The method relies on phage display, a technology developed in the ‘80s for the generation of peptides that bind to targets of interest. In phage display, billions of random peptides are genetically encoded into phage, thus allowing the identification of relevant peptide sequences through isolation from large libraries by affinity selection followed by DNA sequencing. In order to generate bicyclic peptides, the phage-encoded linear peptides are cyclized in a chemical reaction prior to the affinity selection. Bicyclic peptides turned out to have excellent binding properties such as high affinity and good target selectivity, similar to antibodies. At the same time, bicyclic peptides have a good stability. They can be synthesized easily and produced at a large scale. In contrast to antibodies, non-natural amino acids can be incorporated into their sequence and they can easily be modified with cytotoxic drugs, fluorophores or tags.
Have you developed bicyclic peptides against drug targets?
Yes, we have developed bicyclic peptides to a range of disease targets. Bicyclic peptides typically bind their target with nanomolar affinity and can be improved to achieve picomolar dissociation constants, which is more than enough for a drug. We have tested several bicyclic peptides in vivo, such as a picomolar bicyclic peptide inhibitor of the coagulation factor XIIa, a target that contributes to thrombosis as well as to coagulation and inflammation in heart-lung machines. We have demonstrated its therapeutic effects in a range of disease models. More bicyclic peptides to various targets are developed by Bicycle Therapeutics, a spin-off that is commercially exploiting the bicyclic peptide phage display technology. Two bicyclic peptides have entered clinical studies this year, which is a very exciting development.
What are the future challenges and plans of your lab?
Two limitations of bicyclic peptides are that they cannot, in their current form, be applied orally and that they do not efficiently cross membranes, thus preventing them from reaching intracellular targets. We plan to overcome both of these limitations in the future. Regarding oral delivery, we have recently made an important step by developing a new peptide format termed “double-bridged peptides” that resists proteases in the gastrointestinal tract. This work was supported by the NCCR Chemical Biology. For the development of cell permeable ligands, my laboratory is currently developing new methods for the generation and screening of macrocycles that have a smaller molecular weight than bicyclic peptides and are less polar, both key properties to enable passive membrane crossing. The NCCR Chemical Biology is playing a very important role in the development of these methods.
Christian Heinis studied Biochemistry at the Swiss Federal Institute of Technology in Zurich (ETH). After a PhD in the group of Prof. Dr. Dario Neri at ETH, he did two postdocs, with Prof. Kai Johnsson (EPFL) and Sir Gregory Winter (LMB, UK). In 2008 he started as Assistant Professor at EPFL (supported with an SNSF professorship) and was promoted in 2015 to Associate Professor. His main research interest is the development of peptide macrocycle therapeutics. An important activity of his laboratory is the generation of antagonists based on bicyclic peptides by phage display. Christian is a scientific co-founder of the start-up company Bicycle Therapeutics.
Cover picture: Designed by Topntp26 (Freepik)