April 14, 2020 -- A research collaboration between scientists at the University of Leeds and AstraZeneca resulted in the development of a technique that can significantly reduce the discovery time of possible new antibody-based therapies. The approach, published in Nature Communications on April 14, shows how protein fragments can be screened for susceptibility to aggregation in early drug discovery.
Antibodies are proteins produced by the immune system in response to pathogens. Recombinant antibodies (synthesized in a laboratory) have recently emerged as highly potent therapies. These protein biopharmaceutical products can aggregate during manufacturing and storage. Aggregation compromises the quality, stability, and even safety of a drug product.
"Antibody therapeutics have revolutionized medicine. They can be designed to bind to almost any target and are highly specific," said senior author David Brockwell, PhD, associate professor in the Astbury Centre for Structural Molecular Biology at the University of Leeds, in a statement. "But a significant problem has been the failure rate of candidates upon manufacturing at industrial scale. This often only emerges at a very late stage in the development process -- these drugs are failing at the last hurdle."
When designing recombinant antibodies, researchers are not limited to a single protein sequence. Rather, they can select from an array of similar antibodies with an ability to bind tightly to a disease-causing agent. This gives researchers a range of proteins to screen to determine which are more likely to be successful in the development process.
Screening for aggregates
In the current study, the researchers developed a tripartite β-lactamase enzyme assay (TPBLA) that allows for the identification and ranking of aggregation-prone peptides. Here, the test protein is fused in frame between the two domains of the E. coli periplasmic enzyme β-lactamase. This enables scientists to directly link antibiotic resistance of the bacteria to how aggregation-prone the antibody fragment is.
A readout on agar-containing antibiotics gives an indication of whether the antibody aggregates and therefore if it will survive the manufacturing process. If the antibody clumps or unfolds, the antibiotic will kill the bacteria. Conversely, if the antibody is stable, the bacteria thrive and will display antibacterial resistance in the presence of the antibiotic. Importantly, the assay does not use arbitrary methods to destabilize proteins, like heat or chemical denaturation, that may not reflect the inherent properties of the test protein during manufacturing or disease.
Researchers can harvest the bacteria that have successfully grown in the presence of the antibiotic and clone the protein sequence for the test antibody. These antibodies would move forward in the development pipeline.
Directed evolution of screened antibodies
But the research team thought that they could take the test a step further, with a process they call directed evolution.
Using the idea of natural selection -- in which mutations sometimes make proteins more stable -- directed evolution generates sequences that may be better at protecting against aggregation and disease than others. To achieve this, protein sequences hosted in bacterial cells that have shown antibiotic resistance are harvested and scored.
By identifying where mutations occur in the relative location of the protein, the cause of decreased or increased aggregation can be putatively assigned. After checking to ensure that new antibody sequences still retain binding capabilities, they can be taken forward in the development process as "manufacturable" candidates.
Overall, gaining a better understanding of the relationship between sequence, solubility, and aggregation will help researchers decipher the developability of promising biologic candidates and the prediction of mutations that may cause protein aggregation during disease.
"We will be putting the sequence information we gather into a database," said co-author David Lowe, PhD, research lead at AstraZeneca. "As the database gets bigger, it may well be possible with artificial intelligence and machine learning to be able to identify the patterns in protein sequences that tell us that a protein can be scaled up for pharmaceutical production without needing any experiments. That is our next challenge and one we are tackling right now."
Do you have a unique perspective on your research related to molecular biology or drug discovery? Contact the editor today to learn more.