A mouse-like primate threatened with extinction has provided the "missing link" in the evolutionary history of the HIV virus, promising to transform the scientific understanding of the family of viruses to which HIV belongs.

Research into the Madagascan grey mouse lemur published by a team from Stanford University school of medicine, California, suggests that far from being a relatively recent phenomenon, the family of primate lentiviruses to which HIV belongs may be scores of millions of years old.

The study also suggests the endangered lemurs, only found on the Indian Ocean island, may have survived a prehistoric "Aids-like" epidemic before developing an immunity to the disease, promising important insights into how the human epidemic might unfold.

It had been believed that the two strains of HIV found in humans had existed in primates for 1m years at most, with some scientists believing that they may have only been around for hundreds of thousands of years. Aids itself is thought to have existed for just over 100 years.

The study, described in the Proceedings of the National Academy of Sciences in the US, also raises the possibility that similar viruses, believed to be confined at first to African primates, may have been more widely distributed at one time.

The Stanford research suggests that a virus closely related to HIV may have been present in the grey mouse lemur population for at least 14m years, when the last land bridges between Madagascar and the African continent disappeared. Researchers believe it could even be as much as 85m years old, which would make it the oldest ancestor of HIV ever discovered.

Robert Shafer, one of the two lead researchers on the project and an expert in drug resistance to HIV at Stanford medical school, described the work as filling "one of the most important missing links" in the history of how the viruses evolved.

He said yesterday: "The more understanding there is of the links between different lentiviruses and the genomes of the different primates affected, the more we can understand how the proteins that block infection in some primates work. If we understand how hosts have controlled infection over millennia, then that opens the way to developing new drugs or to other ways of encouraging innate resistance."

The discovery of the virus in the saucer-eyed nocturnal animal, described in the Proceedings of the National Academy of Sciences, has been the equivalent of a piece of genetic archaeology – piecing together fragments of the virus's DNA from the animal's genome.

The significance of the discovery for potential research into a cure for human HIV is to be found in the question of how primates infected with the simian version of the HIV virus developed protection from Aids through several genes that slow or block retroviral reproduction.

Over 25 million people across the world have died from Aids-related illnesses since the virus was first identified in the US 27 years ago. Two-thirds of the 33 million people who are at present infected with the virus are in sub-Saharan Africa.

Until the Stanford research was disclosed it had been believed that the lentivirus family had emerged too recently to have been part of this evolutionary development of a resistance to retroviruses. But if the Stanford researchers are correct and lentiviruses are many millions of years old, it could change the understanding of the evolution of immune defences against retroviruses, with implications for HIV treatments or vaccines.

But that is where the problem lies. Lemurs are rapidly dwindling in numbers, and time is fast running out.

Dr Welkin Johnson an expert in primate retrovirology at the Havard medical school, said: "What is surprising is how nicely this fits into the puzzle. It would be a palaeontologist's dream to find data like this that is so unambiguous. It is exciting. What it provides is a distant mirror into how Aids works and the impact it may have on human society in a few thousand years' time. It opens the door to the long-term understanding of how viruses evolve."