If true, the hypothesis would explain why chimps, which share more than 98 percent of their DNA with humans, don't develop AIDS.
The theory stems from a study of DNA in 35 chimps conducted by the Biomedical Primate Research Center in the Netherlands. The chimps in the study were found to share an unusually uniform cluster of genes in the area that controls their immune systems' defenses against disease.
"Chimps show more genetic variation than humans in all areas — with this one exception, which is seriously condensed," said Dr. Ronald Bontrop, who led a Dutch team that worked with statisticians from the University of California.
The findings will be published in the coming issue of the Proceedings of the National Academy of the Sciences of the United States. An online version was already available on the academy's Web site.
Bontrop told The Associated Press that the chimps' lack of genetic diversity, found in genes related to the immune system's defense against disease, suggests that a lethal sickness attacked chimps in the distant past.
This unknown disease would have wiped out all or almost all chimps that didn't have the right immune system genes to fend it off, leaving the survivors with a uniform set.
This, combined with the knowledge that modern chimps are largely immune to the AIDS virus and its simian variants, pointed toward an AIDS-like disease as the culprit.
Scientists believe that HIV, the virus that causes AIDS, originated in apes or monkeys and was transferred or mutated its way into the human population about 50 years ago.
Dr. Luis Montaner, an associate professor at the Wistar Institute in Philadelphia who studies HIV and wasn't involved in the Dutch study, said the findings were intriguing.
"They justify looking at wild populations of chimpanzees to see if they show the same reductions (in genetic variability) as the chimps in the study," he said.
If the findings hold true for all chimpanzee populations, he said, an AIDS-like epidemic in the past would be a plausible explanation — but not the only one. It could have been caused by a sickness unrelated to AIDS, or an epidemic, which occurred more recently.
The theory put forward by the Dutch researchers notes that chimps split into four subspecies around 1.5 million years ago. Since all subspecies represented in the study share the same genetic reduction, the researchers estimated the epidemic happened before that split.
Additional evidence from other studies suggested that bonobo apes also have a genetic reduction similar to that of the chimps, pushing the time of the epidemic back to 2 million years ago, when bonobos and chimpanzees shared a common ancestor.
But Montaner, the independent researcher, said a more modern disease could have spread among the various subspecies, even though they are separated by wide distances.
He also said there is no definitive proof linking specific genes with resistance to AIDS in either chimpanzees or humans, although researchers have found some evidence and are looking for more.
Bontrop said chimpanzee immune systems appear to defeat HIV by targeting part of the virus's proteins that don't mutate. A similar defense mechanism may be at work in humans who have been exposed to HIV repeatedly but don't get sick, he said, suggesting an area for further study.
"It's important to understand mechanisms of disease and resistance in order to help develop vaccines," Bontrop said.
Both Bontrop and Montaner said it was impossible to estimate how many chimps died — but that it was possibly as much as 90 percent or more.
If the theory of an ancient chimp epidemic would hold true for humans, he said, "the implications are pretty scary."