Scientists have created a high-throughput, fast method to design and generate thousands of different, miniature, stable proteins from scratch that can bind to specific therapeutic targets. A professor of biochemistry at the University of Washington School of Medicine and his colleagues reported their findings in an article published in Nature on September 27th.
The prevention and treatment of infectious diseases (such as influenza) and the development of antidotes against neurotoxins are the two goals of this research. This method can rapidly synthesize thousands of new candidate drugs, namely small molecule proteins. These computer-designed proteins, which previously did not exist in nature, combine the stability and bioavailability of small molecule drugs with the specificity and efficacy of larger biological agents. These miniature protein adhesives have the potential to become a new batch of drugs, bridging the gap between small molecule drugs and biologics. They can be designed to bind molecules with high selectivity to targets, but they are more stable and easier to bind to where they are needed.
Researchers at the University of Washington utilized a computer modeling platform called Rosetta to design thousands of miniature proteins, approximately 40 amino acids in length. These proteins that do not exist in nature are predicted by Rosetta modeling software to tightly bind to molecular targets, thereby inhibiting the normal function of target proteins. Due to their small size, these short proteins are often very stable and can be stored without refrigeration. They are also easier to enter the body than large protein drugs such as monoclonal antibodies.
Previously, these micro proteins were usually derivatives of naturally occurring proteins. However, these derived proteins are not more effective than monoclonal antibodies. Due to the de novo design of the micro proteins in this study, it is possible to more freely customize the desired functions and make modifications and adjustments more easily. In this study, researchers attempted to design two sets of micro proteins: one that can prevent influenza virus from invading cells; Another set of neurotoxins that can bind and neutralize deadly botulism is considered a potential biological weapon.
Computer modeling has determined the amino acid sequences of thousands of short proteins suitable for and combined with influenza and botulinum targets. Researchers have created a large number of very short DNA fragments that can encode the required micro proteins to be produced in yeast cells, and then observe their binding to targets. The target is influenza H1 hemagglutinin and botulinum toxin B. In summary, this method allows them to design and test 22660 proteins in just a few months, and identify the best micro proteins from them.
When evaluating the optimal micro protein, it was found that both the anti influenza protein and the redesigned and synthesized micro protein could neutralize the influenza virus and prevent botulinum toxin from entering brain cells. The researchers reported that nasal sprays containing a custom designed protein, which were treated within 72 hours before or after influenza virus exposure, could protect mice from infection. The treatment effect is better than existing influenza antibodies. Protein property tests show that they are very stable and, unlike antibodies, do not become inactive in high temperature environments. Small proteins also rarely trigger or have no immune response, making them a more promising new type of antiviral drug.
Original source：Aaron Chevalier, Daniel-Adriano Silva, Gabriel J. Rocklin et al. Massively parallel de novo protein design for targeted therapeutics. Nature, 2017; DOI: 10.1038/nature23912
