Miniproteins are a revolutionary
class of therapeutics, typically defined as proteins under 100 amino
acids that maintain a stable, defined 3D structure. They occupy a
crucial space between traditional small-molecule drugs and large antibody
biologics. Their compact structure offers significant pharmaceutical
advantages:
- Superior Stability: Easier to store, manufacture and potentially
formulate into non-injectable forms, like oral drugs.
- Enhanced Penetration: Their small size allows them to efficiently penetrate tissues and cells, reaching previously inaccessible intracellular disease targets—a major limitation for antibodies.
The rapid acceleration of
miniprotein drug development is fundamentally driven by Artificial Intelligence (AI) and
computational biology tools, such as AlphaFold. Researchers use software
(like Rosetta, Amber and principles pioneered by AlphaFold) to design entirely de
novo (from scratch) miniproteins, greatly accelerating the discovery process by
predicting the most stable and effective folds. This computational
approach allows for rapid, atomic-level precision in modelling, design. Ongoing research highlights
miniproteins’ role in treating difficult diseases like:
- Infectious Diseases: Demonstrated effectiveness as potent inhibitors
against viral targets, including components of the SARS-CoV-2 spike
protein.
- Inflammatory Conditions: Being developed to inhibit key
inflammatory proteins, such as Interleukin 23 (IL-23) and
Interleukin 17 (IL-17), offering new treatments for chronic
autoimmune diseases.
The future of miniproteins is expansive, leveraging computational advancements to explore their use as targeted delivery vehicles to carry drugs or imaging agents directly to specific disease sites (e.g. within tumors). And as AI and high-throughput systems mature, the automated design of miniproteins will open new doors for personalized therapeutics, allowing for rapid drug creation tailored to an individual patient's unique disease markers.
https://doi.org/10.1016/j.jmb.2021.167160
https://pubs.acs.org/doi/10.1021/acs.accounts.7b00186