A team of researchers, led by Enrique Marcos and F. Xavier Gomis Rüth, from the Spanish Council for Scientific Research at the Barcelona Institute of Molecular Biology (IBMB-CSIC), based in the Barcelona Science Park, and David Baker, director of the Institute for Protein Design at the University of Washington, have designed and created proteins that do not exist in nature and whose structure mimics the folded immunoglobulins of antibodies. The results of the work, published in Nature Communications, could serve as a basis for the development of more affordable monoclonal antibody-based drugs. 

Current drugs based on monoclonal antibodies rely on modifying a small region of the antibodies to specifically recognise and attack specific targets, such as cells or virus, among others. These are currently the most promising and breakthrough drugs in the pharmaceutical industry, especially for the treatment of different types of cancer, autoimmune diseases and, more recently, viral infections.

However, these drugs are still costly therapies as there are limitations that slow down their progress, such as low stability, large size and difficult large-scale production, among others. That’s why they are highly expensive drugs to develop, produce and, because they are very unstable, also difficult to distribute, as they require appropriate storage and refrigeration conditions.

The part of the antibodies that is modified is a very specific one. “All antibodies have a very similar structure, but at their ends they differ in a small variable region that allows each antibody to specifically recognise a target,” reveals Enrique Marcos. This variable region is a scaffold with folded immunoglobulins, which are suited for anchoring antigen-binding hypervariable loops to interact and recognise specific pathogens. The research that has just been published describes a computational strategy to design “small immunoglobulins like those of antibodies with customised structures, high stability and with the capacity for anchoring flexible areas with the capacity to bind to the desired target,” explain the scientists.

The scientific team has followed this computational strategy to generate the new molecules. Afterwards, they checked by crystallography that the structures obtained were those predicted in the models, which means that “we can design the proteins with high accuracy,” adds Marcos.

Design model for a de novo immunoglobulin-like domain anchoring a ligand-binding loop (IBMB/CSIC).

The work opens the door to the design of antibody-like proteins with tailor-made structures and better biophysical properties than current monoclonal antibodies, which would be a breakthrough for the development of more accessible drugs and enable new mechanisms of action.

» Reference article: Tamuka M. Chidyausiku, Soraia R. Mendes, Jason C. Klima, Marta Nadal, Ulrich Eckhard, Jorge Roel-Touris, Scott Houliston, Tibisay Guevara, Hugh K. Haddox, Adam Moyer, Cheryl H. Arrowsmith, Xavier Gomis-Rüth , David Baker & Enrique Marcos. De novo design of immunoglobulin-like domains, Nature Communications volume 13, Article number: 5661 (2022) https://doi.org/10.1038/s41467-022-33004-6