By David N. Leff

Live rabies virus in a vaccine against AIDS?

This may sound like a biowarfare weapon, but in fact it's working well in mice. What's more, it jump-started not one, but both arms of the animals' immune systems - the B cells that churn out antibodies, and the cytotoxic T lymphocytes (killer T cells) that attack their HIV-1 target cells.

Most successful vaccines against viral diseases rely on killed or attenuated (weakened) viral particles to tweak antibodies into action. Killer T cells rarely make it to the front lines.

A task-force of vaccine-designers at Jefferson Medical College in Philadelphia reported their results so far in the Proceedings of the National Academy of Sciences (PNAS), dated March 28, 2000, under the title: "Recombinant rabies virus as potential live-virus vaccines for HIV-1." The paper's senior author, Roger Pomerantz, directs the Center for Human Virology at the medical college, a branch of Thomas Jefferson University.

"Development of a protective HIV-1 vaccine," he told BioWorld Today, "still remains a major goal in halting the HIV-1 pandemic." Describing the vaccine approach as "novel," Pomerantz added: "The rabies virus is the vehicle to get an HIV envelope protein, gp160, to express itself, thus getting the attention of the immune system."

The co-authors enlisted as their vaccine vector a strain of live rabies virus (RV) that does not harmfully infect a wide range of animal species. Among its advantages, their paper points out: "In contrast to most other viral vectors, only a negligible seropositivity to RV exists in the human population, so immunization with a RV-based vector against HIV-1 would not interfere with immunity against the vector itself.

"Vaccines that are licensed now," Pomerantz pointed out, "are either killed or attenuated. Using a viral vector for a vaccine has never been approved in humans. This would be a new approach - using another virus as a vector to immunize against a second virus."

HIV's outer envelope consists of the antigenic glycoprotein gp160. The co-authors cloned its coding region so that the resulting recombinant RVs expressed HIV-1 gp160, along with the other RV proteins.

Rabies Virus Makes Nice In Cytoplasm

"We chose rabies," Pomerantz explained, "even though it makes people nervous, because rabies virus when it is attenuated does not kill cells as its close relative, vesicular stomatitis virus, does. So RVs are actually less cytopathic, which is what we want. What's nice about them is that they are RNA viruses, but they don't integrate into the cell nucleus. All of their expression is in the cytoplasm. We want low levels of ongoing replication, not killing these cells, just presenting antigens."

Into their weakened, safe, live rabies virus, the co-authors inserted the HIV-1 gp160 envelope protein gene, thus creating a recombinant RV construct. They then injected this package into the footpads of three mouse groups:

¿ Animals that received the RV vector alone;

¿ Those that got the vector plus its HIV gp160 gene;

¿ Controls receiving nothing.

"Then," Pomerantz recounted, "we looked at the results of this in vivo vaccine trial: Did it make antibodies to gp160? We showed that it did. Did the antibodies work? After a booster shot of the gp160 antigen, the mice had developed neutralizing antibodies to the RV strain with which they were infected.

"Finally, besides antibodies, the CTLs reacted with gp120 on virally infected cells. They were maintained for many months even without boosts. The mice all lived. They had no adverse clinical reactions. None developed rabies."

Pomerantz made the point that HIV vaccine designers "are trying to aim at immune reactions from both humoral-arm B-cell antibodies and cellular-arm T-lymphocyte killer cells, with varied success. What we found - and we haven't published this new data except for a few lines in the PNAS article - is that CTLs are easily expressed with this RV-HIV vaccine. They last for at least four months in mice, so they're stable.

"And as for the humoral arm," he went on, "injection of the rabies live-viral vector, followed several weeks later by the boost with recombinant gp120, gave us very good neutralizing antibodies. So like many vaccines, to get both arms of the immune system, we would need a prime booster, something like what a lot of childhood vaccines require."

Monkeys Now. Humans Next Year?

"So now at least in mice," Pomerantz observed, "we have both arms of the immune system with what looks like protective immunity. Since one can't challenge mice with HIV, our next step is to do it in macaque monkeys with SIV - simian immunodeficiency virus. And that's what we've just begun doing right now.

"We're collaborating with Chris Miller, veterinary virologist at the University of California/Davis primate center," he said."Those monkeys were inoculated in their footpads and intramuscularly just within the last couple of weeks. I would imagine we'll have preliminary data within six months."

The co-authors chose to move from mice to macaques because, Pomerantz pointed out, "chimps just don't get HIV disease very commonly. What we want to do is not only look for protection against infection, but also make sure that we have a disease-causing model. So more work on mice continues while the macaques go on. If the monkey data looks good, then we would look at a Phase I human trial. Not sooner than next year; certainly not this year."

Next on the group's list of putative RV-vector vaccines is hepatitis C virus. "We've already made some constructs," Pomerantz revealed, "but no preliminary data yet. We're looking at a variety of hepatitis C antigens," he added, "in particular the E1 protein, which has the most data on cytotoxic lymphocytes and neutralizing antibodies."

Pomerantz and the PNAS paper's lead author, biochemist Matthias Schnell, are joint inventors of a pending patent, claiming "all negative-stranded RNA viruses to use as a vaccine approach against HIV and other diseases."