DURHAM, N.C. - Researchers have noticed a curious phenomenon
in sub-Saharan Africa: many female prostitutes who come in
contact with men infected with HIV do not themselves become
ill.

Further study of these women shows that the mucous lining
their vaginas contains HIV-specific secretory immunoglobulin A
(IgA) antibodies that may be able to stop the virus and keep it
from entering the bloodstream. It is possible these women have
become "naturally" immunized against future HIV infections.

A research team led by Duke University Medical Center is
hoping to use this knowledge to create a vaccine that could
stimulate similar responses in the uninfected. Additionally,
they plan to employ a novel delivery system that can be easily
used in underdeveloped nations.

Project leader Dr. Bart Haynes, chairman of the department
of medicine at Duke, believes that the team will develop a
vaccine that can be delivered nasally and would stimulate
immune responses in the user at both the mucosal and systemic
levels.

"I think we will be able to start testing our next
generation of candidate vaccines in humans in a couple of
years," said Haynes -- a time-span that seems to be just around
corner considering he has spent 15 years studying the virus.
"Making a vaccine has turned out to be more difficult than we
ever believed. The virus is more complex than we thought and is
very cunning in its ability to subvert our immune systems.
However, there will be a vaccine."

To prove the merit of this approach, the National Institutes
of Health is supporting the research in the amount of $5.5
million over the next five years.

According to a recent report by the U.S. surgeon general, of
the estimated 33.4 million people worldwide who are infected
with HIV, 22.5 million live in sub-Saharan Africa, 6.7 million
in South and Southeast Asia, and 1.4 million in Latin America.
By comparison, there are about 665,000 infected in the United
States

More than 2.4 million have died worldwide in the past
year.

In the under-developed nations, the main route of virus
transmission is sexual, the researchers said. That's where the
observations about the prostitutes come in. In the vaginal
mucosal lining of these women, researchers have detected
secretory IgA (S-IgA) that specifically binds to HIV. S-IgA is
a type of antibody that is produced and secreted into mucosal
secretions to protect the body against bacteria and viruses
that attack the body mucosal tissues.

"These women were probably exposed to a low dose of the
virus, then developed a local infection that was cleared up by
the body's defenses, inducing local S-IgA mucosal immunity,"
explained Duke immunologist and co-investigator Herman Staats,
Ph.D. "Over time, it seems, despite multiple exposures to the
virus, these women don't become infected.

"With HIV, traditional injected vaccines haven't worked
well," Staats said. "A vaccine targeted to mucosal linings,
however, may be able to stimulate an immune response that
protects people by blocking the ability of HIV to infect at
mucosal tissues, thereby preventing a full-blown
infection."

Researchers have known that if a mucosal surface at one part
of the body is challenged -- for example, the lining of the
nose -- any S-IgA immune response that is generated may also
appear in other mucosal surfaces, such as the mouth, upper
respiratory tract, GI tract or reproductive organs.

"One benefit of a nasally introduced vaccine is that it is
non-invasive, so presumably it would be more feasible to use on
a widespread basis throughout the world, especially in the more
under-developed areas," Staats said.

The task ahead -- and it is a daunting one, the researchers
say -- is coming up with a potent method for stimulating the
production of immune system cells that recognize HIV as an
invader and destroy it.

In the Duke project, the researchers hope to develop a
vaccine with a one-two punch. One would stimulate immunity in
the form of S-IgA in the mucous to act as a first line of
defense, while the second would come in the form of another
type of immune response known as IgG antibodies. IgG
antibodies, which are formed to defend the body against
specific invaders, circulate throughout the bloodstream to
neutralize any virus that gets through the first line of
defense.

"This combination is important because either one alone is
unlikely to be able to completely eliminate the virus," said
David Montefiori, Ph.D., Duke co-investigator who is focusing
on neutralizing antibodies. "There are some areas, or
compartments, in the body where the virus hides that the immune
system can't penetrate.

"With this in mind, the trick is to produce an immune
response that constantly puts pressure on the virus and never
lets it gain strength," Montefiori said.

In order for an antibody to kill, or neutralize a virus, it
must first "latch" onto the invader. What makes HIV such a
formidable foe for researchers is that since it is constantly
mutating, finding a site on the surface that is the same for
all variants of the virus is difficult.

The Duke researchers are focusing on a protein called gp120
that resides on the surface of the virus and is essential for
viral infectivity.

When gp120 comes into contact with and latches onto its
cellular receptor protein, CD4, the virus can enter the target
cell and initiate a new round of infection. Neutralizing
antibodies can block the ability of gp120 to attach to the
target cells, thereby preventing the virus from gaining
entry.

The Duke candidate vaccine is a mixture of synthetic
peptides taken from the gp120 regions of four significantly
different strains of HIV. This vaccine, developed by Haynes and
colleagues, has been shown to be strong enough to stimulate
neutralizing antibodies, and since it is based on synthetic
peptides, there are no fears of the vaccine actually causing
infection.

In general, the vaccines produced to date have not been
sufficiently effective in generating strong and lasting immune
responses.

To "boost" the ability of their vaccine to induce an immune
response, the Duke researchers plan to add something -- called
an adjuvant -- to increase the vaccine's effectiveness and
life-span. By interacting with the vaccine, the adjuvant serves
to strengthen the antigenicity of the vaccine.

The researchers are evaluating a variety of proteins known
as cytokines, as well as mutant cholera toxin, to act as
adjuvants for their nasal vaccines.

"With peptide vaccines, an adjuvant is required to get
potent, long-lived antibody and cellular responses to HIV,"
Staats said.

Once candidate vaccines and adjuvants have been developed,
they will first be tested in animals. Later this year,
researchers hope to begin testing the vaccine in small mammals,
and if all goes well, into non-human primates by next year
before moving rapidly on to humans, Haynes said.

These experiments will be handled by project co-investigator
Dr. Norman Letvin, an immunologist and virologist at Harvard
Medical School's Beth Israel Deaconess Medical Center in
Boston.

"Our role in this project is to use our non-human primate
models to explore novel vaccine strategies for preventing HIV
infection," Letvin said. "Any successes in these monkey studies
should rapidly translate into success in humans. The power of
the models we have developed is enormous.

"The virus (SIV) that we use to infect monkeys is a very
close relative of HIV - it can infect monkeys and cause AIDS,"
he said. "We can look at new vaccines in a monkey and can
measure antibodies and T cell responses, but the challenge
remains in seeing how effective the vaccine is in blunting
infection."

One major limitation of the SIV model is that the outer
membrane of SIV is different from that of HIV, so if the goal
is creating a vaccine based on a surface protein, the SIV model
has limitations. So the researchers developed a SHIV animal
model by combining human and monkey AIDS viruses to create a
new virus that expresses a HIV envelope on the 'backbone' of
SIV. Both the SIV and SHIV models will be used in the current
project.