Proteins designed from the ground up are usually too regular and continuous to be useful in vaccine design, which must grapple with irregular and discontinuous epitopes, that is, the complex antigenic features that are recognized by the immune system’s antibodies. And so, these proteins, which are called de novo proteins, have been limited to mimicking the simplest epitopes.
Impatient with the sorts of de novo proteins that have been designed with existing computational systems, scientists based at École Polytechnique Fédérale de Lausanne (EPFL) built a computational system of their own, one capable of designing artificial proteins that rival the complexity of real proteins. The system, called TopoBuilder, designs scaffolds for irregular and discontinuous neutralization epitopes.
The EPFL scientists, led by Bruno Correia, a professor at the Laboratory of Protein Design and Immunoengineering (LPDI), described TopoBuilder in an article that appeared May 15 in Science, in an article entitled, “De novo protein design enables the precise induction of RSV-neutralizing antibodies.” The scientists decided to train TopoBuilder on respiratory syncytial virus (RSV), which causes serious lung infections and is a leading cause of hospitalization in infants and the elderly, because it seemed a suitable test object.
“Despite several decades of research,” noted Correia, “there is still no vaccine or cure for respiratory syncytial virus.”
“When a vaccine doesn’t work well,” he continued, “we tend to think that it’s because the antibodies produced are not protective. It’s usually because our immune system is simply making the wrong type of antibodies.”
To demonstrate TopoBuilder’s abilities, Correia and colleagues generated three epitope-focused immunogens designed to induce a targeted neutralizing antibody response to RSV.
“In both mice and nonhuman primates, cocktails of three de novo–designed immunogens induced robust neutralizing responses against the respiratory syncytial virus,” the article’s authors wrote. “Furthermore, the immunogens refocused preexisting antibody responses toward defined neutralization epitopes.”
The artificial proteins created with TopoBuilder “don’t exist in nature,” stressed Che Yang, a PhD student and a co-leading author of the study. “TopoBuilder lets you construct proteins virtually as if you were putting Lego bricks together, added Fabian Sesterhenn, a PhD student and co-leading author. “Assembling artificial proteins that have novel functions is absolutely fascinating.”
The artificial proteins designed by TopoBuilder were realized in the laboratory and then tested in animal models, and they triggered the immune system to produce specific antibodies against weak spots in RSV. “Our findings are encouraging because they indicate that one day, we will be able to design vaccines that target specific viruses more effectively, by prompting the immune system to generate those particular antibodies,” asserted Correia. “We still have a lot of work ahead to make the vaccine we developed more effective—this study is a first step in that direction.”
Methods for creating de novo proteins have applications well beyond immunology—they can also be used in various branches of biotechnology to expand the structural and functional range of natural proteins. “We can now use the protein design tools to create proteins for other biomedical applications such as protein-based drugs or functionalized biomaterials,” concluded Sesterhenn.