AI Identifies Potential Gonorrhea Vaccine Targets

Researchers have used artificial intelligence (AI) to identify promising target proteins with the potential for development into a vaccine against gonorrhea. The study, which shows the proteins can induce immune responses in vitro and in animal studies, is published in mBio.

Limited treatment options

Gonorrhea is a sexually transmitted infection caused by the bacterium Neisseria gonorrhoeae. It affects over 80 million people around the world each year and is becoming especially difficult to treat as it has developed resistance to nearly all known antibiotics.

Lack of treatment options can have serious health implications if the infection goes untreated, and it can even increase the risk of contracting HIV.

Vaccines against gonorrhea are promising prospects to prevent infection, but there is no approved vaccine to date. With the advance of AI approaches, the search for a suitable vaccine could be on the horizon.

Now, a new study, borne of a collaboration between academics and Evaxion, an AI immunology startup, has described the use of an AI model to identify two novel antigens with the potential to be developed into an experimental vaccine against gonorrhea.

Antigens identified with AI

The AI model, named Efficacy Discriminative Educative Network (EDEN), was designed to identify vaccine targets in infectious microbes, and in the current study was used to predict how well different combinations of antigens could reduce populations of N. gonorrhoeae.

They applied the model to 10 clinically important strains of N. gonorrhoeae to predict which bacterial proteins could potentially be recognized by the immune system in a vaccine to help combat the infection.

“To the best of our knowledge, this correlation has not been shown before,” said Sanjay Ram, senior author of the study and a professor at the University of Massachusetts Chan Medical School.

The researchers next tested and validated the identified antigens in both laboratory and animal models, revealing two proteins involved in cell division as promising candidates. Interestingly, neither of these two proteins was previously known to be exposed on the cell surface.

In line with EDEN’s predictions, blood samples taken from mice immunized with the two proteins were able to kill bacteria from multiple strains of N. gonorrhoeae in laboratory tests. Additionally, mice that were infected after immunization showed a decreased bacterial burden.

“That really was a surprise,” Ram said. “Nobody would have predicted that these two proteins that were believed not to be surface exposed would work in vaccines, and other researchers reacted with skepticism.”

Next, given the success of the two proteins in tests, the team then combined them into a single protein – known as a “chimeric protein” – which in turn showed similar efficacy in inducing immune responses in both laboratory and animal models.

The investigations also revealed a mechanism the vaccine candidate uses to clear the bacterial infections – however, further studies are required to determine if this mechanism is also present in humans.

Potential for vaccine development

Aside from N. gonorrhoeae, the researchers are also using EDEN to investigate other proteins with potential as vaccine candidates against other microbial diseases.

The project’s next steps include progressing beyond in vitro experiments and animal studies to investigate if these proteins could also provide protection for humans. Recently, the research team has also forged a partnership with a South African biotechnology company for the development of an mRNA vaccine based on the identified antigens.

Reference: Gulati S, Mattsson AH, Schussek S, et al. Preclinical efficacy of a cell division protein candidate gonococcal vaccine identified by artificial intelligence. mBio. 2023;0(0):e02500-23. doi: 10.1128/mbio.02500-23

This article is a rework of a press release issued by the American Society for Microbiology. Material has been edited for length and content.