Research Briefs

CMV-based Vaccine Candidate Protects Macaques from Systemic Infection

An experimental AIDS vaccine consisting of a replicating rhesus cytomegalovirus (CMV) vector carrying several SIV genes protected four of 12 rhesus macaques from systemic infection after repeated low-dose rectal challenge with SIVmac239, a recent study has found (1). The study suggests that the vaccine induced effector memory T cells directly in the mucosal tissues, which protected the animals from the challenge virus by keeping it from spreading systemically.

Louis Picker, a professor at Oregon Health & Science University and the lead author of the study, says the premise for this study was that replicating vectors like CMV continuously express antigens and, as a result, induce effector memory cells right at the mucosal sites where the pathogen infects. Within minutes or hours after being presented with an antigen from the challenge virus on the surface of an infected cell, effector memory cells can become CD4+ or CD8+ Tcells and either make cytokines or kill the target cells, Picker says. In contrast, prime-boost vaccine candidates that use non-replicating vectors such as MRKAd5, which was tested in the STEP trial, induce central memory Tcells. These cells retreat to inductive sites such as the lymph nodes some time after the antigen goes away. Upon challenge, they have to first migrate to the mucosal tissues, which can take as long as a week—likely too long to keep an infection like SIV or HIV from spreading systemically.

"If you want to have a vaccine that contributes to controlling infection at the very outset of the virus crossing the genital or rectal epithelium, [it has to induce] effector memory cells," Picker says, adding that to "keep effector memory cells specific for a particular antigen in these sites, [we need to have] antigen around all the time."

In the study, Picker and his colleagues vaccinated 12 rhesus macaques sequentially with three rhesus CMVs, carrying the SIV genes gagrev-tat-nef, and env, respectively. Half of the vaccinated macaques also received a boost with a combination of all three vectors. These vaccinated and 16 unvaccinated monkeys then received weekly low-dose intrarectal challenges with SIVmac239.

After 12 challenges, all of the unvaccinated macaques were infected. In contrast, four of the vaccinated macaques were protected from progressive infection. Two of them did not show any measurable virus in plasma even after 13 challenges, and two showed low transient viral levels after the first challenge, but not later. All four are still virus free, Picker says, almost a year after the first challenge.

The four protected macaques showed CD8+ T-cell responses specific to two SIV proteins that were not included in the vaccine, showing that they were indeed exposed to the challenge virus. However, CD8+ Tcell depletion did not result in elevated virus levels, suggesting that CD8+ Tcells were not what kept their virus levels in check. Also, Picker says, while the study was too small to make a definitive conclusion, the protection status of the vaccinated animals did not correlate with MHC class I variants Mamu-B*08 or B*17, which have been associated with CD8+ Tcell-mediated control of viral replication—one of four protected animals and five of the eight unprotected animals had either B*08 or B*17. "[The infection] is either controlled by another mechanism or [it] was cleared and the infection is no longer there," he concludes.

One explanation for how the animals could have cleared the virus is SIV-specific effector memory T cells, which were observed in blood and broncheo alveolar lavage of the vaccinated macaques. Picker says the study did not measure these cells in the rectal mucosa because it would have interfered with the experiment, but he suspects they were likely there at the time of challenge, keeping the virus replication below a threshold at which it could sustain itself.

Next, Picker plans to study correlates of local protection in the rectal mucosa. In collaboration with IAVI, he will also be comparing this CMV regimen with combination regimens of CMV/Ad5 and DNA/Ad5. Picker says there are also discussions at IAVI about developing a CMV vector that could be used in humans.

"This is a promising finding, which of course ultimately will have to be confirmed in human clinical trials," says Stanley Plotkin, a consultant at Sanofi Pasteur. Still, Plotkin says getting FDA approval for a trial with CMV could be a challenge. "They tend to be very risk-averse."

When it comes to possible human trials, one concern is preexisting immunity, because, according to Picker, virtually everybody in developing countries and half the people in developed countries have been exposed to CMV. However, this study suggests that preexisting immunity may not be an issue. Another concern is whether a replicating vector like CMV will be safe in humans, especially in immune-compromised people. It almost never causes problems in people unless they are immune compromised, Picker says, adding that a human CMV vector would be engineered to have even less pathogenicity.

Replicating vectors are not necessarily the only way to keep antigens around at mucosal sites to induce effector memory cells, Picker says. Another way to achieve this could be an annual shot of perhaps even non-persistent vectors. "The mindset has been you do a prime, you do a boost, and that’s it—the person is supposed to be protected for life," he says. "But the protection they are going to need I believe is effector memory protection and that’s going to require antigen there frequently, if not all the time." —Andreas von Bubnoff

1. Nat. Med. 15, 293, 2009

Microbicide Inhibits Innate Immune Response

A microbicide containing the surfactant glycerol monolaurate (GML) protected rhesus macaques from repeat high-dose vaginal challenge with SIVmac239, at least in part by inhibiting the early innate immune response, according to a study led by Ashley Haase, a professor at the University of Minnesota (1).

In a previous study, Haase and colleagues treated rhesus macaques once daily for months with K-Y warming gel, which is used as a personal lubricant in humans, either alone or containing GML, and found that neither formulation harmed the vaginal epithelium (2).

The researchers then vaginally challenged two GML-treated and two K-Y only treated animals from this safety study with two high-dose challenges of 100,000 infectious doses of SIVmac239, separated by four hours. This challenge leads to infection in at least 90% of animals, Haase says. The gel was applied twice, one hour before each challenge. Two weeks later, one of the two K-Y only animals was infected, while both GML-treated animals were not.

When the same challenge was given to three additional GML-treated and three K-Y only treated control animals from the safety study, one of the control animals was infected by two weeks, and none of the GML-treated animals were infected. After an additional challenge a few weeks later, the remaining two control animals became infected, while all three GML-treated animals remained uninfected. However, Haase says eventually one of the GML-treated animals did become infected five months after the second challenge.

The candidate microbicide PRO 2000 was recently found to reduce the risk of HIV infection by 30% in a Phase IIb trial (see Canvassing CROI, IAVI Report, Jan.-Feb. 2009). While the results were not statistically significant, they generated excitement among microbicide researchers. PRO 2000 is thought to work by inhibiting the virus itself, but Haase’s study suggests that GML works, at least in part, by inhibiting a chemokine produced by the mucosal epithelium called MIP-3α, part of the innate immune response. Haase says production of MIP-3α leads to an influx of more CD4+ Tcells to the site of infection, providing more target cells for the virus. "Everybody, myself included, thinks of these innate and inflammatory responses as the host responses to prevent and contain infection," he says. "But on balance they do just the opposite—they bring in the fuel."

Within a day after vaginal challenge in Haase’s study, researchers observed production of MIP-3α by the mucosal epithelium. This attracts plasmacytoid dendritic cells, which produce chemokines that in turn attract CD4+ Tcells. GML-treated monkeys had less MIP-3α in their vaginal fluid than untreated animals, Haase says, adding that microarray analysis showed a downregulation of MIP-3α expression in these animals.

GML, found naturally in breast milk, has been widely used by the food and cosmetics industry as an emulsifier, Haase says. Because of its antimicrobial properties, it has also been used in tampons to prevent toxic shock syndrome.

"Only time will tell whether this is a major breakthrough," says Robin Shattock, a professor of cellular and molecular infection at St. George’s, University of London, who was not involved in the study. He says it is unclear how often the compound might have to be applied to block infection, and suggests that GML may also have directly inactivated the virus. Haase acknowledges that GML may also work by inhibiting the virus, but says that preliminary observations suggest much lower GML levels in the animals than those at which one would expect to see viral inhibition. —Andreas von Bubnoff

1.Nature 458, 1034, 2009
2. Antimicrob. Agents Chemother. 52, 4448, 2008

Researchers Catch HIV on Film

Using high-speed three-dimensional imaging equipment and an infectious clone of HIV embedded with green fluorescent protein (GFP), researchers were recently able to track and film in real-time the movement of HIV Gag in live CD4+ Tcells. These movies show what happens when HIV-infected cells collide with uninfected CD4+ Tcells and convey how rapidly the viral material—with the help of adhesive contacts called virological synapses that are formed at the juncture of CD4+ Tcells—passes from infected cell to uninfected cell (1).

Together, virologists at Mount Sinai School of Medicine in New York City, who created the fluorescent HIV clone known as HIV Gag-iGFP, and physicists at the University of California-Davis, who supplied the expertise in high-speed imaging, produced 12 movies. Some depict just a few seconds in the life cycle of the virus, while others—with the help of time-lapsed photography—span several days. Although these short films may not become Hollywood blockbusters, after a week on YouTube (www.youtube.com/GreenVSLab), one had more than 150,000 hits.

Benjamin Chen, the Mount Sinai virologist who created HIV Gag-iGFP, says a fast video microscope capable of taking three-dimensional images of infected cells every second or so, showed HIV Gag quickly congregates at the virological synapse, forming a button shape, once an infected cell touches an uninfected cell. The footage then shows the viral proteins being ushered into a target cell’s endosome, a membrane-bound compartment that many other viruses use to gain entry into cells but which HIV was not thought to favor much. When the HIV Gag-iGFP was compared to an infectious HIV clone without the Env protein, researchers found that this protein is critical for formation of synapses.

The role of cell-associated virus in HIV transmission has long been a mystery. Chen says future vaccine strategies should perhaps look at unique cell-surface Env epitopes that block cell-associated virus from spreading. —Regina McEnery

1. Science 323, 1743, 2009