Washington Post

Wednesday, February 2, 2000

AIDS Debate: Where and When Did It Start? Immunologist-Mathematician Tries to Debunk Journalist's Position That Virus Is Tied to

By David Brown

SAN FRANCISCO, Feb. 1 -- The origin of the AIDS virus is an abiding mystery. The oldest known sample of the virus dates from 1959. It was found in blood collected for a long-finished medical study in the country that used to be known as the Belgian Congo. The source was an "adult Bantu male." His name was never recorded.

Most experts in the rarefied world of evolutionary virology doubt AIDS is that young. Some think it may have been holed up in isolated African populations for more than 100 years. A few, however, think that 1959 is very, very close to ground zero.

The most meticulously researched and closely argued case for that hypothesis appeared last fall in a book called "The River," written by the British journalist Edward Hooper. In it he argues that HIV was probably introduced to human beings during the world's first mass vaccinations against polio, which occurred in Central Africa in 1958. The virus, Hooper hypothesizes, was a contaminant from primate kidney cells, in which the virus was grown to produce a "live" vaccine.

Despite its 851 pages of text (and 174 pages of end notes), "The River" offers no hard evidence for the theory that the AIDS epidemic was unwittingly begun by the hand of man--an "iatrogenic" illness of unprecedented magnitude. His case is based on plausibility, on the idea that it could have happened in the way he describes.

The essential condition of his hypothesis--namely, that chimpanzees, and not other primates, were the source of the kidney cells--has been declared untrue by several aging scientists who worked on the vaccine nearly 50 years ago. Documentary evidence for many of the events upon which Hooper's hypothesis depends is long gone, if it ever existed. Unless an unadulterated sample of 1958 vaccine is found to contain the AIDS virus--a possible, but unlikely discovery--the theory will never be absolutely proved or disproved.

At a scientific conference here Monday, however, a virologist offered her own evidence--also indirect and unprovable--that the contaminated polio vaccine theory of AIDS's origin isn't right. She says that the strains of human immunodeficiency virus (HIV) now infecting 34 million people worldwide probably descended from a strain of simian immunodeficiency virus (SIV) introduced into human beings no later than the 1930s.

Bette Korber, an immunologist and mathematician at Los Alamos National Laboratory, used the laboratory's database of global HIV gene sequences to infer when the virus strains now circulating had a common "ancestor."

HIV is a microbe constantly undergoing mutation, producing progeny that are slightly different from their progenitors. This produces groups and sub-groups that are reasonably distinct from one another in their DNA sequences. Group M is the HIV family responsible for virtually all the cases of HIV infection on earth. M's many sub-groups--more than 10 have been described to date--presumably all descended from a single SIV transmitted through a bite or a cut from a chimpanzee. Chimpanzees, and not other primates, are assumed to be the source because the strain of SIV they carry today is most closely related to HIV.

Korber's analysis required the construction of about 100 possible "family trees" of HIV in which the contemporary strains have a common ancestor. To capture as many scenarios as possible, she and her colleagues assumed in one model that HIV evolved at fixed rate over time, and in another that its rate of mutation varied over time. This allowed them to back-date HIV to its "founder" virus. Each tree took about eight hours to construct, using the world's fastest computer, the ACL Nirvana, which does 1 trillion calculations per second.

Their calculations came up with 1930, plus or minus 4.5 years, as the time when HIV took off.

It could have entered human beings before then and sputtered for years in a small population before breaking out. But 1930 is the latest it could have first infected a person in order for there to have been enough time for HIV to diversify to the extent it has today.

In the 100 or so family trees constructed, the time of HIV origin varied between about 1910 and about 1950. None, however, showed an introduction of the virus as late as 1958, the year of the Congo polio campaign. For HIV to have arrived that late, there would had to have been multiple chimpanzee-to-man transfers of slightly different viruses--perhaps as many as 10.

"This seems to me to be quite an unlikely scenario," Korber told a gathering of AIDS researchers at the Seventh Conference on Retroviruses and Opportunistic Infections.

Hooper's hypothesis requires more than a late introduction of SIV from chimpanzees. It also requires that the virus have survived the manufacturing process, and also not to have infected any of the relatively small number of Europeans who also got it in the 1950s. (As many as 1 million people may have gotten it in the Congo.)

The vaccine in question was called CHAT, a name derived from that of the child polio victim from which the strain was first isolated for use in a vaccine. It was made by Hilary Koprowski, a man who ran the Wistar Institute, a research center affiliated with the University of Pennsylvania in Philadelphia.

In the 1950s, Koprowski and his colleagues were in a neck-and-neck competition with Albert Sabin to develop the first safe and effective oral vaccine. (The injected Salk vaccine existed, but wasn't practical for use in developing countries.) Sabin eventually won the race.

At the conference yesterday, Stanley Plotkin, a 67-year-old virologist who was one of Koprowski's closest associates in developing CHAT, said: "I will reiterate--no chimpanzee tissues were used to make oral polio vaccine." He told Hooper that the virus used for the Wistar vaccine was grown in kidney cells from rhesus monkeys, which do not carry SIV in the wild. Documents recounting CHAT's step-by-step production, however, no longer exist.

Hooper believes there are two reasons chimpanzee tissue might have been used. First, vaccinologists of that time were experimenting with numerous types of cells. Second, there's evidence--as unbelievable as it may seem--that at least six shipments of chimpanzee kidneys (minced and in a biological solution) were sent unrefrigerated from the Congo to Philadelphia. There, they were used, apparently successfully, by University of Pennsylvania hepatitis researchers.

The kidneys came from animals housed at a jungle camp maintained by Koprowski and Belgian colonial doctors outside Stanleyville (now Kisangani). Koprowski, however, said he used the animals only for testing the safety of his vaccine, not for making it.

Among the other pieces of evidence Hooper uses to support his theory is the striking geographic overlap of the sites where CHAT was given and the oldest cases of AIDS have been found. Critics of the theory say, however, the coincidence is not surprising, as those were the populous and well-traveled parts of the Congo.

But the theory of a pre-1958 introduction is not without its own coincidences.

The 1930 date offered by Korber accounts only for Group M virus. Group O and Group N--two very rare variants--probably were introduced separately. In addition, HIV-2, a microbe related to the AIDS virus (which is known formally as HIV-1) has four different subtypes, each of which appear to have been independently passed to human beings in West Africa. That means that although chimpanzees and people have lived closely in that continent for thousands of years, there were at least seven unrelated transfers of virus from the former to the latter, probably this century.

The Wistar Institute has one vial of CHAT vaccine left over from the 1958 Congo campaign. Plotkin said yesterday it has agreed to distribute the sample to three independent laboratories, which will test for HIV, and also for chimpanzee DNA. The one other sample known to exist, held by the Karolinska Institute in Sweden, was tested for HIV several years ago. None was found.

A conference has been scheduled at the Royal Society in London in May for AIDS researchers to formally address Hooper's hypothesis. Plotkin was asked what he would think if HIV was found in the old samples.

"As a scientist, I will have to accept the evidence," he said. "But I believe such an outcome is very unlikely."