In the more than 50 years since the rise of modern molecular biology, researchers have developed vaccines against a range of pathogens, from SARS-CoV-2 (the virus that causes COVID-19) to Streptococcus pneumoniae (the bacterium responsible for pneumococcal disease). However, despite years of effort, vaccines against a handful of bacterial pathogens—namely Pseudomonas aeruginosa, Staphylococcus aureus, and Neisseria gonorrhoeae—remain just out of reach.

These bacteria wreak havoc on the body in various ways. P. aeruginosa preys primarily on people in healthcare settings with certain underlying conditions (e.g., cystic fibrosis) or who have wounds from surgery or burns; it infects everything from the lungs to the blood. S. aureus, another healthcare-associated pathogen that causes sepsis, endocarditis, pneumonia and more, is similarly broad in its infectious scope. While N. gonorrhoeae’s claim to fame is gonorrhea, a sexually transmitted infection with a global annual incidence of nearly 87 million adults, the pathogen can also cause invasive infections in the mouth, eyes and throat.

In the past, infections caused by P. aeruginosa, S. aureus and N. gonorrhoeae could be treated with antibiotics. However, all 3 bacterial species have become increasingly resistant to antibiotics—a 2019 report from the Centers for Disease Control and Prevention (CDC) estimated that, in 2017 alone, multi-drug resistant P. aeruginosa and methicillin-resistant S. aureus (MRSA) caused over 32,600 and 323,700 infections among hospitalized U.S. patients, respectively. New tools, like vaccines, will be critical for controlling these pesky pathogens.

But why is designing vaccines for this microbial trio so difficult? Though each pathogen possesses unique traits and structures, they have remained vaccine-free for several reasons.

The Bugs are Antigenically Loaded

A critical step in vaccine development is determining what structure or molecule on, or secreted by, a pathogen the vaccine should target. Ideally, the chosen molecule, or antigen, plays a central role in the colonization/pathogenesis of the microbe. For example, vaccines against S. pneumoniae target its capsule, the sticky polysaccharide coating that surrounds S. pneumoniae cells. Capsule is considered the most important S. pneumoniae virulence factor. It helps the bacteria adhere to host cells and protects them from immune defenses and therefore makes an excellent target for vaccines. 

For pathogens like P. aeruginosa, S. aureus and N. gonorrhoeae, it’s more complicated. These bacteria come loaded with surface structures and secreted factors that facilitate colonization and infection. The membrane of P. aeruginosa, for example, is decorated with long strands of lipopolysaccharide (LPS), lipoproteins, pilli, 5 different protein secretion systems and over 2 dozen porin proteins involved in everything from molecular transport to antibiotic resistance and surface binding. P. aeruginosa also secretes factors, including a potent exotoxin and proteolytic enzymes, that modulate its pathogenesis. S. aureus and N. gonorrhoeae are similarly well-endowed.

“Any of these immunogenic factors could be considered protective agents,” Dr. Joanna Goldberg, a professor in the Department of Pediatrics at Emory University School of Medicine studying P. aeruginosa vaccine targets, said in a talk given at ASM Microbe in June 2022. Because of this, “it is not clear which antigen should be included in a vaccine.”

In other words, finding the best vaccine target for such antigenically loaded bacteria is like playing a game of darts with a moving board—you might hit something, but it likely won’t be the bullseye.

Moreover, during infection, “[virulence] factors can be gained or lost, modified or diverge,” Goldberg explained.

N. gonorrhoeae, for instance, reversibly switches proteins on and off via phase variation—each cell in a population can flash different proteins simultaneously. An antigen that is present 1 minute then gone the next is not an ideal vaccine target.

Additionally, the virulence factors each pathogen deploys in 1 stage or location of infection may not be relevant in another. Indeed, S. aureus alters expression of its virulence factors as it transitions from living harmlessly in the nose (around 33% of people harbor S. aureus in their nostrils) to becoming an invasive, sepsis-causing pathogen. Because of this versatility, a vaccine “needs efficacy at different [body] sites,” said Dr. Sanjay Ram, a professor at the UMass Chan Medical School studying N. gonorrhoeae.

Goldberg, the P. aeruginosa researcher, noted that “antigen selection may depend on the type of infection” and suggested that, rather than a general-purpose vaccine, using different vaccine components for different infection manifestations (e.g. an acute burn wound versus chronic lung infection) could be a more fruitful route. Along these lines, combining multiple antigens into a single vaccine may increase clinical efficacy in diverse contexts. This concept is similar to multivalent vaccines that protect against different strains of a pathogen, such as quadrivalent meningococcal conjugate vaccines that prevent infection by 4 different serogroups of N. meningitidis, the cause of meningococcal disease.

Protective Immunity (?)

The purpose of a vaccine is to generate a protective immune response against microbial invaders. The problem is, researchers are unclear what protective immune responses against S. aureus, P. aeruginosa and. N gonorrhoeae researchers actually look like.

During infection, pathogens are detected by innate immune cells, which activate B and T cells (adaptive immune cells). B cells produce antibodies that flag a pathogen for elimination by other immune cells, such as neutrophils. T cells support antibody production, recruit other cells to the site of infection and, depending on the type, may inactivate or eliminate pathogens directly. Vaccines trigger these responses on a smaller scale, etching the pathogen into the immune system’s memory so it can deal with future attacks. Antibodies are often the primary focus of vaccine research because of their integral role in pathogen recognition and memory responses. However, for antigenically diverse microbes wielding an arsenal of immune evasion mechanisms, this may not be enough. For instance, in terms of S. aureus vaccines, “everyone was [initially] targeting antibody formation against Staph,” said Dr. Jean Lee, an associate professor of Medicine at Brigham and Women’s Hospital. “Then it shifted to T cell immunity. But, in reality, we need both.”

Goldberg agreed. “One immune mechanism may not be sufficient,” she said, pointing out that effective P. aeruginosa vaccines may need to kickstart antibody-mediated killing while also inhibiting toxins produced by the microbe. Roping in T cells may not be a bad idea either. In some cases, like for N. gonorrhoeae, it’s not a matter of what constitutes a protective immune response, but whether such a thing exists.

“Is there protective immunity following natural [N. gonorrhoeae] infection? The short answer is no,” Ram said. Given vaccination is meant to mimic natural infection to induce similar immune responses, this is problematic. Ram emphasized that the “the humoral [i.e., antibody] response to N. gonorrhoeae is complicated” and little is known about T cell immunity to N. gonorrhoeae, which, like for S. aureus and P. aeruginosa, might be useful.

Still, for each of these pathogens, context is important. All 3 bacteria weaponize different virulence factors depending on infection location and severity. The immune responses likely mirror those changes, suggesting that what is a “good” immune response changes from one situation to the next.

Good Animal Models Can be Hard to Find 

Preclinical studies in animal models allow investigators to determine if, and how, a vaccine works before making the leap into humans. Mice are the most widely used models in biomedical research. Unfortunately, sometimes pathogens don’t infect mice the same way they infect humans. Goldberg, Lee and Ram all agreed: a lack of robust animal models is one of the key barriers to developing vaccines for P. aeruginosa, S. aureus and N. gonorrhoeae.

There are several ways mouse models have fallen short. For instance, Goldberg highlighted that P. aeruginosa researchers “would like to have models where the lifespan of the animals is longer.” The pathogen commonly infects elderly individuals—the short lifespan of mice doesn’t lend itself to studying vaccine efficacy for older populations or those suffering from chronic infection.

In addition, as Lee pointed out for S. aureus, “there’re a lot of virulence determinants and immune evasion molecules that are human-specific and don’t even affect mice—many of the toxins have receptors that are missing in mice.” As such, any role these factors play in human infection (and vaccine efficacy) are missed in animal models. This is even more of an issue for N. gonorrhoeae, which only naturally infects humans.

It's not only a matter of if colonization of animals by a given pathogen is possible, but also when such colonization takes place. “Most studies [have been conducted] in mice that are naïve and have never seen S. aureus before,” Lee said. This contrasts with the human population where, as previously mentioned, nearly 1/3 of people are already colonized with the S. aureus. Pre-exposure shapes immune responses to the pathogen and can influence if, and how, a protective response is generated, which has implications for vaccine development. Indeed, a recent study revealed that in mice naïve to S. aureus, a vaccine targeting the bacterial surface protein, IsdB, elicited a protective antibody response against infection. However, pre-exposure to S. aureus triggered production of non-protective antibodies; vaccination merely boosts this non-protective (i.e., ineffective) response. These findings may shed light on why an IsdB vaccine failed in clinical trials in humans.

What's Next for Vaccine Development? 

What needs to happen to transform vaccines against S. aureus, N. gonorrhoeae and P. aeruginosa from zeros to heroes? The pathogenic plasticity of these organisms, coupled with researchers’ relatively hazy picture of what a protective immune response looks like, necessitates a deeper understanding of host-pathogen interactions in diverse infection contexts. Such knowledge is critical for creating viable vaccine candidates, as well as determining who gets them, and how and when they should be used.

Better models are also needed. Luckily, advancements in this area have been made. For instance, humanized transgenic mice (i.e., mice carrying human genes or cells) have made it possible for N. gonorrhoeae researchers to study interactions between microbial vaccine targets and human receptors. Continuing to improve existing models, and develop new ones, will be an important step toward developing vaccine candidates that survive the jump from the lab to the clinic.

Moreover, vaccine developers may not have to start from scratch. It was recently shown that vaccination against N. meningitis serotype B, which causes meningitis B and shares much of its genome with N. gonorrhoeae, is cross-protective against gonorrhea. As such, it may be possible to use meningitis B vaccines to protect against both bacteria.

Today, P. aeruginosa and S. aureus researchers are better equipped than they were in the past. Drawing from a rich history of vaccine “don’ts,” can inform the creation of vaccines whose futures are brighter than their predecessors’.

Will it be another 50 years before vaccines against these pathogens hit the market? Time will tell. If anything, the fact it’s taken so long prompts a little admiration for the elusiveness, adaptability and antigenic diversity of our microbial foes.
 

---

Research in this article was presented at ASM Microbe, the annual meeting of the American Society for Microbiology, held June 9-13, 2022, in Washington, D.C.