The finding is the latest to link our health to the microbes that share our bodies, what's called the microbiome. And while it highlights another difficulty in developing vaginal "microbicides" to block HIV infection, it also offers the prospect of one day identifying women who are particularly vulnerable.
"This is an important study," said microbiologist Sharon Hillier of the University of Pittsburgh, who wasn't involved in Thursday's report but leads the Microbicide Trials Network that tests potentially protective products. "It does tell us that this is another factor we have to consider."
Women make up about half of the nearly 37 million people worldwide living with HIV. They're particularly at risk in hard-hit Africa. Scientists have long sought unobtrusive ways for women to protect themselves when their partners won't use a condom — especially in poor countries where another option, a daily anti-HIV pill, isn't widely available.
The research re-examined an early vaginal gel containing the AIDS drug tenofovir that had seemed partially protective in one study only to fail in another. Puzzled scientists thought maybe some women simply didn't use the gel properly.
Samples saved from 688 South African women who tested that gel tell a different tale. Some of those women had a reproductive tract dominated by species of "friendly" bacteria from the Lactobacillus family. Others harbored less healthy species that lead to bacterial vaginosis, inflammation that can increase risk of certain health problems.
The tenofovir gel reduced HIV infection by 61 percent in women who harbored the mostly healthy lactobacilli — but by only 18 percent in women with the less healthy vaginal bacteria, researchers reported in the journal Science.
Why? The team mixed tenofovir with different types of bacteria in lab dishes. Tenofovir stayed around longer when mixed with healthy lactobacilli, while concentrations plummeted rapidly when mixed with a particularly bad bug named Gardnerella vaginalis. The unhealthy bacteria were breaking down the drug before it could do its job, the study found.
The ramifications? Today, other types of microbicides are being developed such as a vaginal ring that emits a very different anti-HIV drug. Those other drugs must be tested to tell whether vaginal bacteria might undermine them, too, Hillier said. As for tenofovir, she questioned if a higher dose might solve the problem.
"We have to look at biological variability in each person," said microbiologist Nichole Klatt of the University of Washington, who led Thursday's research with a team of U.S. and Canadian scientists. Microbicides aside, scientists have linked bacterial vaginosis to a higher risk of HIV infection — begging the question of whether it ever will be possible to make vaginal bacteria healthier to lower that risk in the first place.
For now, Klatt said simple, relatively inexpensive pH tests can indicate if a woman lacks protective vaginal bacteria. "What's really important is understanding this moving forward and being able to advise women more," Klatt said. Some women might be told, "this prevention might work for you because you have lots of lactobacillus" while others "might need a back-up."