It Eradicated Smallpox, But How?

Researchers are finally collecting clues about the life-long protection afforded by the smallpox vaccine, the gold standard of vaccines

By Andreas von Bubnoff

One riddle currently confronting AIDS vaccine researchers is identifying the immune correlates of protection against HIV. But they are far from alone on this type of quest. For many vaccines the correlates of protection elude researchers even after they have been used for decades. Once a vaccine works, there is little interest in figuring out why, even though there are benefits to understanding just how an effective vaccine affords protection. “We should really know how the things that work, work,” says Shane Crotty, an associate professor for vaccine discovery at the La Jolla Institute of Allergy and Immunology.

Smallpox, caused by variola virus, was eradicated in the late 1970s with a vaccine that was a live preparation of the genetically related vaccinia virus. Crotty refers to the smallpox vaccine as the gold standard because it is the only vaccine that has ever led to the eradication of a disease. Yet, for several reasons, the correlates of protection for this vaccine are still unknown. When smallpox was eradicated, many of the modern methods to measure immune responses weren’t yet available. At that time, researchers could not measure T-cell responses, says Mark Slifka, associate professor at the Vaccine & Gene Therapy Institute at Oregon Health & Science University. The human data on how the smallpox vaccine works are mostly from observational studies, and because there are no naturally occurring smallpox infections anymore, it would be impossible to do a randomized clinical trial today of a smallpox vaccine to study the correlates of protection. It would also be unethical to perform a placebo-controlled clinical trial because of the existence of an effective vaccine, Slifka says.

Renewed interest in trying to understand the smallpox vaccine is in part driven by the need to develop a new vaccine with fewer side effects, which could be used to guard against a potential bioterrorism attack with the remaining stockpiles of smallpox, Crotty says.

The vaccinia smallpox vaccine, called Dryvax, would be considered too dangerous to use for some because it can cause serious side effects in immunocompromised people, including people with AIDS, according to D. Huw Davies, a project scientist at the University of California in Irvine. “The side effects mean that [Dryvax] wouldn’t be approved today,” Davies says. Another smallpox vaccine called ACAM2000, which was recently approved, is a safer, cloned version of Dryvax manufactured according to more stringent modern Good Manufacturing Practice standards. But this vaccine still causes side effects in immunocompromised people, adds Davies. Researchers are now working on safer smallpox vaccines such as those based on modified vaccinia Ankara (MVA), which is derived from vaccinia but is unable to replicate in human cells and also lacks genes that suppress the host immune system.

Identifying the correlates of protection for the smallpox vaccine in humans may never be possible, but researchers are starting to collect clues about the way it protects by studying vaccinated individuals, people who have survived an infection, and by using animal models. They have found that the original smallpox vaccine primarily works on the basis of neutralizing antibodies, the way most vaccines are thought to protect, and that the antibody responses are surprisingly variable and redundant. At the same time, they are trying to identify certain markers in the antibody response that they hope to use to predict whether a safer, alternative vaccine will be protective.

Searching for the correlates

Progress in understanding how the smallpox vaccine works has been slow in coming. “It’s been thought for quite some time that the smallpox vaccine does work on the basis of neutralizing antibodies, but it was really just [a few] years ago that that was directly shown,” says Crotty, referring to a 2005 study by Genoveffa Franchini’s group at the National Cancer Institute that provided evidence in animal experiments that antibodies are required for protection (1). “That experiment nailed it,” Crotty says.

In the study, researchers vaccinated monkeys with the human smallpox vaccine and then inhibited either the humoral or the cellular immune responses to determine which part of the vaccine-induced immune response was required for protection against intravenous monkeypox challenge. They found that only inhibiting the antibody response eliminated the protection afforded by the vaccine. “They were able to show in a relevant challenge model that antibody-mediated protection is the main component for protection,” Slifka says. The study also showed that transferred human antibodies could protect unvaccinated animals against the challenge virus, he adds, suggesting that antibodies are both necessary and sufficient for protection.

This confirms observations in humans, Slifka says, referring to a 1941 study that showed that transfer of serum from smallpox survivors could protect infected people from death (2). In another study, antibody alone protected the majority of children with a genetic defect in their T-cell responses from dying from vaccinia infection after a smallpox vaccination (3). Slifka says the fact that in the absence of cellular immunity the immune serum from vaccinated people protected most of the children suggests that while T cells play a role, antibody alone may almost be completely sufficient for protection against smallpox.

While antibodies are clearly important for protection, it’s unclear which antigens they need to be directed against or which concentration of antibodies is required. According to Slifka, some studies suggest that high antibody titers appear to be a marker of protective immunity (4; 5). However, these are observational studies that don’t include a control group. T-cell responses, which may also have contributed to protection, were also not measured in these studies.

To learn more, Slifka is now studying the antibody and cellular responses in a cohort of smallpox survivors and people who received a smallpox vaccination to see if the vaccine induces an immune response similar to that in natural infection.

Others like Davies and Crotty have started to systematically study the antibody response to smallpox vaccine using microarray chips that contain most of the approximately 200 smallpox proteins. In a typical person vaccinated with Dryvax, they have identified antibodies to 20-30 proteins, only about a dozen of which are surface proteins (6; 7). “We get immunoreactivity to membrane proteins as well as proteins that are not in the membrane of the virus, and even proteins that don’t end up in the virus at all,” says Philip Felgner, a coauthor of these studies and the director of the applied proteomics research laboratory at the University of California in Irvine. Antibodies to non-surface antigens are probably directed to proteins released from necrotic infected cells, according to Davies.

These studies show that antibody responses to the vaccinia vaccine are surprisingly variable—only half of the antibody responses in two people are typically directed toward the same smallpox proteins. Also, the dominant response in two vaccinated people will likely be to a different smallpox protein.

“There is no single protein that a person always has to have a response to in order to get protection,” says Felgner. The cellular immune responses to the vaccine are probably even more variable in humans than antibodies, says Felgner, whose lab used the same preparations of the complete set of smallpox proteins to study T-cell responses to the vaccine by stimulating peripheral blood cells from vaccinated people.

The antibody responses to smallpox also appear to be redundant. Even if antibodies which in and of themselves are sufficient for protection are removed from serum from vaccinated people, this serum can still protect against infection (7). This suggests that there is not a single mechanism for protection.

It seems that as long as an antibody covers the surface of the virus, it will protect, Crotty says. And it doesn’t matter which of the smallpox proteins on the surface an antibody binds to. “What you really need is an antibody that covers the surface of the pathogen sufficiently so that the pathogen can’t bind to the target,” he adds. “[It’s like] throwing a net over the virus.” Large complement proteins might also play a role, assisting antibodies in binding to each other to cover the virus.

Predicting protection

While there doesn’t seem to be a single, clearly defined immune response induced by the smallpox vaccine, Felgner has used the smallpox protein microarray in animal experiments to identify markers in the antibody response that might predict protection. His group compared the immune responses of vaccinated rabbits that were protected from challenge with ones that weren’t protected despite vaccination and identified three markers that were associated with protection. Felgner is currently also analyzing data from non-human primate studies.

Researchers will use these markers to evaluate samples from a Phase I clinical trial of the MVA vaccine against smallpox to see if it can protect as well as Dryvax, says Felgner. The US Food and Drug Administration (FDA) may then consider using this data, along with the animal data, as the basis for licensure of the MVA vaccine. If approved, it would add MVA as a safer alternative to Dryvax and ACAM2000. In cases like this, where there are no humans infected with a given pathogen, the FDA has a “two animal rule,” Felgner says, which means that evidence from two different animal models is sufficient for vaccine licensure.

Now that the immune response elicited by the smallpox vaccine has been rather clearly described, Crotty says, the next big question about the original smallpox vaccine is how it can give such long-lasting protection. “Why is it that you can give one immunization with this vaccine and you get a fantastic protective antibody response and it lasts for life?” he asks.

The principles learned from a vaccine that protects against smallpox are unlikely going to apply directly to development of an AIDS vaccine. There are many differences between smallpox and HIV. Unlike HIV with its one surface protein, smallpox virus is very large, with about 200 genes and dozens of surface proteins. And also unlike HIV, smallpox doesn’t mutate much.

Given these differences, smallpox may not be the best example to guide development of an AIDS vaccine. “We have been applying the rules of conventional vaccinology to HIV since it emerged in 1983,” says D. Huw Davies, a project scientist at the University of California in Irvine, “but this has largely failed us.” While antibodies are likely important for protection to both smallpox and HIV, something very different from conventional vaccines needs to be developed against the rapidly evolving HIV, Davies adds.

Still, there are some general lessons. If there is anything to be learned from understanding the smallpox vaccine, “it’s that neutralizing antibodies are so key for the protection,” says Shane Crotty, an associate professor for vaccine discovery at the La Jolla Institute of Allergy and Immunology. “It’s yet another piece of information that suggests that you probably need to be able to make neutralizing antibodies.” What’s more, the success of the smallpox vaccine shows that in principle, it is possible to develop a vaccine that can induce life-long immunity and long-lasting T-cell and antibody memory, says Mark Slifka, associate professor at the Vaccine & Gene Therapy Institute at Oregon Health & Science University.

AIDS vaccine researchers are also generating candidates with vectors derived from vaccinia virus. Some are using

replication-incompetent strains such as modified vaccinia Ankara (MVA), while others use replication-competent strains. “Most workers are using replication-deficient strains because of safety,” says Bernard Moss, chief of the laboratory of viral diseases at NIAID (see Go forth and multiply, IAVI Report, May-June 2008).

But worldwide use of the smallpox vaccine has taught researchers enough about the benefits and risks to be able to use replication-competent vaccinia virus, says Julia Hurwitz, a member in the department of infectious diseases at St. Jude Children’s Research Hospital. “Less is known about the newer, non-replicating vaccine vectors,” she adds. Her group has conducted a Phase I safety trial with a replicating vaccinia virus vector based on the smallpox vaccine (8).

Zhiwei Chen, an associate professor and director of the AIDS Institute at the University of Hong Kong, is developing an attenuated but replication-competent vector derived from a vaccinia strain called Tiantan, which was used for the eradication of smallpox in China. This vector carrying the gene for the spike protein of the severe acute respiratory syndrome (SARS) virus induces between 20- to 100-fold higher levels of neutralizing antibodies to the spike protein than the replication-incompetent MVA strain after intranasal or oral administration in mice. “[A] replicating vaccine may offer better immunogenicity and induce better memory response,” says Chen, who presented the findings at a recent conference on mucosal vaccines in Porto, Portugal (see Mucosal Vaccines: Insights from different fields, IAVI Report, Nov.-Dec. 2008). He eventually plans to use this vector to develop an AIDS vaccine candidate. —AvB


1. Nat. Med. 11, 740, 2005
2. Bulletin de l’Institut d’hygiene du Maroc 1, 59, 1941
3. Pediatrics 18, 109, 1956
4. Am. J. Trop. Med. Hyg. 21, 214, 1972
5. Bull. WHO 52, 307, 1975
6. Proc. Natl. Acad. Sci. 102, 547, 2005
7. J. Virol. 82, 3751, 2008
8. Eur. J. Clin. Microbiol. Infect. Dis. 23, 106, 2004