Masked Vector Rides Past Immune Defense

Tested in Animals, Chimeric Vaccine Carrier Heads Toward Clinical Trials

Used as vectors for carrying viral genes, adenoviruses have become one of the most promising vaccine strategies because they are able to provoke a strong immune response without causing disease. HIV vaccines that use adenovirus vectors have shown promising results in animal trials, but there is a potential hurdle to their effectiveness: they make use of one of the most common adenovirus serotypes, one that many people have developed immunity to. And if the vector itself is blocked by the immune system, it has little chance of delivering its cargo.

Photo by Graham Ramsay

Dan Barouch and Diane Roberts gave a common adenovirus vector the mask of a much less common form of the virus. In this disguise, the vector passes by the immune system unnoticed.

A study led by Dan Barouch, HMS assistant professor of medicine at Beth Israel Deaconess Medical Center, offers a strategy that allows researchers to circumvent pre-existing immunity to this common serotype, adenovirus type 5 (Ad5), without abandoning the vector altogether. The strategy replaces small portions of the virus’s outer shell that are recognized by antibodies with the analogous portions of a much rarer serotype. This chimeric vector leaves the Ad5 vector essentially intact, but it looks like a new virus to the immune system.

In an editorial accompanying the study in the May 11 issue of Nature, John Mascola, deputy director of the Vaccine Research Center at the NIH, writes that the “potential of this technology is considerable,” and calls the results “a tribute to the application of modern immunology and structural biology to vaccine design.”

Strategies for HIV
Ad5 vector–based vaccines for HIV have proved promising because they can provoke T cells to action. Previous viral vaccines worked by spurring the immune system to produce antibodies against the virus. But in the case of HIV, no known antigen can stimulate antibodies that are broad enough to fight the constantly mutating virus. As a result, most current efforts to develop an HIV vaccine have been tied to cellular immunity—the ability of killer T cells to destroy infected cells. It is still thought that antibodies may be needed to prevent infection completely but that T cell–stimulating vaccines could reduce the risk of infection or control levels of the virus in people already infected.

Vaccines based on the Ad5 vector make use of replication-defective adenoviruses to deliver HIV genes into cells, thus fooling the body into thinking the cells are infected. Two current vaccine candidates, produced by Merck and the NIH Vaccine Research Center, use Ad5 vectors and are in late-stage clinical trials. AIDS researchers are looking to the results of these trials as a proof-of-principle of cellular immunity’s role in fighting HIV.

But a potential hurdle is that most people have already been exposed to Ad5. “People are worried about the effects of pre-existing immunity,” said Raphael Dolin, HMS dean for academic and clinical programs, who leads the Harvard HIV Vaccine Unit that is conducting trials of HIV vaccines with Ad5 vectors. Perhaps because it is so effective at infecting human cells, Ad5 is also one of the most prevalent adenovirus serotypes, and Dolin notes that the prevalence of Ad5 immunity is highest in developing regions such as sub-Saharan Africa, where an HIV vaccine is most needed.

Hexon Mobile
Last year, Barouch was awarded a major NIH grant to lead an effort to develop novel candidate HIV vaccines, including devising alternatives to Ad5, such as chimeric vectors and those using different serotypes of adenovirus. Although other serotypes show potential, “it makes a lot of sense to modify Ad5,” Barouch said. It is the best studied vector, and it has already passed substantial regulatory hurdles and has been subject to a good deal of manufacturing groundwork.

Image courtesy of Dan Barouch

A shell game. The hexon protein of adenovirus-5 (Ad5) is shown in a ribbon diagram (left) and space-filling model (right), from two angles. Seven regions that vary greatly between Ad5 and other serotypes of adenovirus are modeled in red. Hexons like this pack together to form a viral shell; the variable regions are exposed on the shell’s surface and are recognized by antibodies.

In previous research, Barouch’s lab found that both T cells and antibodies contribute to Ad5 immunity but that antibodies are the driving force. The outer shell of an adenovirus particle is made mostly of proteins called hexons, as well as pentons that form a base from which fiber spikes protrude. Previously, Barouch and his colleagues found that antibodies against Ad5 vectors primarily targeted the hexons. His group and others have tried to replace whole hexons with those of other serotypes, but because they are essential to viral structure, these viruses failed to form.

Barouch’s team wanted to see if they could isolate just the part of the protein that elicits antibodies. The hexon protein contains seven short regions that differ widely among serotypes. Because the crystal structure of the hexon proteins is known, the team could map the location of these hypervariable regions in a model of the hexon. The regions accounted for much of the protein’s exposed surface. Putting together the information about sequence and structure, Barouch’s team hypothesized that these regions might be the key targets of Ad5-specific antibodies.

After three years of challenging technical work, the team managed the tricky business of performing a hypervariable-region transplant, replacing just these short sections of the viral genome. The corresponding donor regions came from a strain of adenovirus, Ad48, that is highly uncommon.

The chimeric vector was able to grow efficiently in cultured cells. Moreover, in both mice and monkeys, a chimeric vector carrying the simian immunodeficiency virus (SIV) protein Gag was able to provoke immune responses similar to those generated by an Ad5 vector. And the immune responses elicited by the chimeric vector were not suppressed in animals with preexisting Ad5 immunity.

The team is partnering with the Dutch pharmaceutical company Crucell to bring the new vectors into clinical trials. The extent to which preexisting immunity will truly stymie the current vaccines continues to be debated. However, Barouch notes, even if Ad5 vaccines are effective, a booster shot using the same vector might not be. And if a successful Ad5 vaccine was developed for infections like malaria or TB, then the two vaccines might counteract each other. “A vector that is used for any one indication would likely not be successful for other pathogens or as boosters,” he said, so the best strategy is to devise multiple vectors that work.

—Courtney Humphries

top