Drug and antibody research often center around complex cell membrane proteins. These proteins, embedded in cell membranes, are challenging to study due to their insolubility in water-based solutions. Researchers have now developed a method to create soluble analogues of these proteins, making them more accessible for drug discovery.
“By redesigning these membrane proteins into stable, soluble analogues, we have made them easier to work with,” explains Casper Goverde, a PhD student at the Laboratory of Protein Design and Immunoengineering (LPDI) in the School of Engineering.
Goverde and his team used deep learning to design synthetic soluble versions of membrane proteins typically used in pharmaceutical research. This innovative approach eliminates the need to extract these proteins from cells directly. Instead, they can design soluble analogues using a computational method and then produce them in bulk using bacteria, allowing for more straightforward study and interaction with drug candidates.
“Producing soluble protein analogues using E. coli is estimated to be around 10 times cheaper than using mammalian cells,” adds PhD student Nicolas Goldbach.
The team’s research findings have been published in the journal Nature.
Changing Protein Design Strategies
While artificial intelligence has previously been used to design new protein structures, this study focused on creating more accessible versions of existing protein folds. By using deep learning networks, the researchers were able to predict amino acid sequences for soluble versions of key membrane proteins based on their natural 3D structures.
To achieve this, the team utilized the AlphaFold2 structure prediction platform from Google DeepMind to generate amino acid sequences, which were then optimized using a second deep learning network, ProteinMPNN, for functional and soluble proteins. This approach proved successful in creating soluble proteins with native functionality, including complex folds that had been challenging to design previously.
Advancing Biochemistry Research
One significant achievement of the study was the successful design of a soluble analogue of the G-protein coupled receptor (GPCR), a crucial pharmaceutical target found in human cell membranes. Redesigning the GPCR shape as a stable soluble analogue opens doors for faster drug screening.
Additionally, the researchers believe this approach can be applied to vaccine development and cancer therapeutics. By designing soluble analogues of proteins such as claudins, known for their role in tumor resistance, the study demonstrated the potential of this method in creating new targets for drug development.
