Medicine October 2000

The Hunt for the Origin of AIDS

The notion that AIDS arose from a polio vaccine made with contaminated chimpanzee cells—the thesis of the best-selling book The River—is far from the only theory about how the epidemic started, and it is hotly disputed. The quest for the source of the epidemic is intensifying, as researchers scour the jungle for clues and try to "walk back" the disease genetically with the help of the world's most powerful computers

EViral family treetraordinary claims require extraordinary proof. The OPV/AIDS claim is, without question, extraordinary. The proof to date is not. And the arguments that have been mounted against the theory are compelling, especially when viewed in the context of other instances in which scientists have posited links between a medicine-related intervention and a disease.

The River opens with a brief description of one of those instances, involving the British anesthesiologist John Snow's famous conclusion that the pump water on Broad Street in London was responsible for an outbreak of cholera in that city in 1854. At Snow's urging, the authorities removed the pump's handle on September 8 of that year, shutting off the flow of what was indeed cholera-generating dirty water. "One wonders whether a similar date, denoting a 'Eureka moment' for AIDS, will ever be written into the medical textbooks," Hooper writes.

But in the retelling of the Snow story Hooper and many others leave out an inconvenient detail: by the time the pump was disabled, the epidemic had already begun to wane. The cause-and-effect relationship, then, is fuzzier than the lore would have it. And in epidemiology, which relies heavily on correlations, fuzziness often prevails.

Consider the following case. From 1955 to 1963 nearly 100 million Americans received polio vaccines that were contaminated with a simian virus known as SV40. The contamination occurred by the same mechanism that Hooper blames for AIDS: the monkey kidney cells used to make the vaccine harbored the as-yet-undiscovered virus. Scientists began looking into the matter in the early 1960s. Those investigators, unlike Hooper in his investigations into OPV and AIDS, did not have to rely on old vaccine samples or on hypothetical infections of particular primates in order to identify the contaminant: the vaccine was still being made, and monkeys infected with SV40 were easy to find.

Still, researchers cannot reach a consensus about whether SV40 harms human beings. An Italian scientist, Michele Carbone, has demonstrated a link between SV40 and a cancer known as mesothelioma, which causes chest and lung tumors; he has even matched viral particles in patients' tumors with particles in old samples of the vaccine. But mesothelioma takes a decade or more to develop, and, as in the case of AIDS, this complicates attempts to understand cause and effect. Some large epidemiological studies have shown no correlation between the virus and the disease, whereas others do show a correlation. And so the debate continues, the issue unresolved. (For an in-depth look at Carbone's theory, see "The Virus and the Vaccine," by Debbie Bookchin and Jim Schumacher, in the February, 2000, Atlantic.)

This is not to say that resourceful researchers cannot make persuasive arguments that medical accidents long in the past (of precisely the sort Hooper suggests) have caused epidemics. In 1942 some 50,000 U.S. servicemen developed acute hepatitis, in one of the largest hepatitis B epidemics ever documented. Investigators at the time discovered that the men had been vaccinated against yellow fever three months earlier. Scientists subsequently came to suspect that the vaccine was contaminated with hepatitis B, an unknown agent in 1942. The suspicion was sound, but for many years no one could demonstrate the connection.

Finally, in a study published in The New England Journal of Medicine in 1987, a group of researchers made a convincing case for what the connection is likely to have been. Unable to find any stored blood or vaccine samples, they turned to old hospital and payroll records instead. They located 221 servicemen who had had symptoms of hepatitis in 1942, and 171 men who were asymptomatic but whose vaccines had come from one of the lots most strongly linked to the hepatitis outbreak. For a control group they selected 205 men who had joined the Army after a new, safe yellow-fever vaccine went into use. The researchers took blood samples from all the men. They found hepatitis B antibodies in 97.7 percent of those in the first group and in 77 percent of those in the second group--but in only 13 percent of those in the control group. Most scientists now believe that the yellow-fever vaccine caused the hepatitis epidemic.

Last March scientists published a paper in the British journal The Lancet that links the staggeringly high prevalence of hepatitis C in Egypt (about 20 percent of the country's population is infected, the highest percentage in the world) to dirty needles used in a public-health campaign. Records show that beginning in the 1920s the Egyptian government gave people infected with schistosomiasis (a parasitic infection acquired primarily by swimming in contaminated water) multiple injections of tartar emetic—a highly toxic drug that offered only limited relief.

The researchers describe how clinicians, if they "sterilized" their needles at all before re-use, boiled them only briefly. "The skilful doctor began injecting at 9:20 a.m. and completed 504 injections of men, women and children by 10:10 a.m.," according to one contemporaneous account they cite. "Allowing for a 10-min rest, the time taken for each injection was thus just under 5 [seconds].... This remarkable performance is being repeated at various tempos all over Egypt." Clearly, the needles constituted a likely means of transmission.

Scouring public-health records, the researchers then found that from 1964 to 1982—when an effective drug became available—an average of about 250,000 Egyptians a year received nine injected treatments. They analyzed blood samples from 8,499 Egyptians, breaking the group down according to age and geography (northern Egypt has less schistosomiasis, and so a smaller proportion of the population there received injections). More than half the forty- to fifty-year-olds in southern Egypt had hepatitis C antibodies, whereas only 35 to 40 percent in northern Egypt did; younger Egyptians showed no clear pattern of infection. Although the researchers offer no direct evidence, these figures make a compelling case that dirty needles used in the tartar-emetic campaign caused Egypt's hepatitis C epidemic.

Compared with the theories set forth in the New England Journal and Lancet papers, the OPV/HIV connection posited in The River has a much less plausible mechanism of transmission, a more confusing timeline, and weaker geographic and age correlations. Hooper's documentation of cases is far sketchier, and his theory involves many more speculations. His theory is, of course, still possible—it's just not probable. And science, after all, is the business of sifting the probable from the possible.

The Second Strain

SHORTLY after I arrived at the Franceville research station last March, Paul Telfer drove me from his home on the sprawling campus to the facility's primate colony, which houses 151 mandrills, seventy-four chimpanzees, twelve sun-tailed monkeys, eight gorillas, and an assortment of about seventy-five other monkeys. Outside was a loading dock heaped with bananas. Inside I was greeted by a French researcher with a diapered baby chimpanzee in her lap.

A local hunter had brought the baby, Za-Za, to the center after killing her mother. I peered into her sorrowful eyes—the one feature of chimps that looks wholly human—and instinctively began stroking her hand. The researchers told me that after Za-Za arrived, scientists found SIVcpz antibodies in her blood.

Despite the flood of attention that SIVcpz infection has received, the data on this subject have come from remarkably few chimpanzees—which probably reflects only the fact that few chimps have been sampled to date. At the time of Za-Za's arrival at the research station, scientists had isolated evidence of the virus in only seven chimps. Two were in the Franceville station, and three were in Cameroon. One was a chimp named Noah who had been imported to an Antwerp zoo from Zaire. One had died after several years in a primate colony in the United States; she had been captured in the wild in Africa, but no one knew precisely where.

Za-Za, it turns out, isn't really No. 8: researchers found antibodies to the virus but not the virus itself in her blood. By the time I met her, the antibodies had begun to disappear, which typically happens when a baby acquires antibodies during gestation or through breast milk but escapes infection.

Later that evening, after our trip to Bakoumba to look for mandrills, I had a beer with Telfer at his home, a wood-trimmed concrete-block townhouse whose architecture, including a loft bedroom, seemed oddly hip for this remote part of Africa. Telfer told me stories of his fieldwork in Sierra Leone from 1991 until July of 1994, when civil war forced him to flee. With government soldiers riding on the roof of his car and driving armored personnel carriers, he raced in a convoy from his rural lab site to the capital, Freetown, from which he eventually flew out of the country. His boss then and now, the American virologist Preston Marx, soon began collaborating with the scientists at Franceville. Telfer moved to the research station in the fall of 1994 and has lived there off and on ever since.

Telfer booted up his laptop to show me a digital picture taken on a previous trip to Bakoumba. If we were lucky, he said, we might find something similar on the outing planned for the next day. Hunters often cannot eat all their kill, so they display the extra meat in front of their shacks. Driving past one shack, Telfer had noticed a pair of bloody mandrill arms hung up for sale. He pulled over to speak with the woman who lived there and asked what had happened to the mandrill's head. He had photographed the answer.

The gruesome image shows the woman holding the head of a male mandrill. The animal has big, crooked teeth; one eye is open in a death glare, and blood courses across the monkey's face and down its beard. The head partly blocks the message on the woman's T-shirt, so that it reads IS BEAUTIFUL. Close inspection reveals that mandrill blood is dripping onto two of the woman's fingers. "If people asked me how HIV could have entered humans, this is what I would show them," Telfer said.

As Hooper writes in the postscript to the paperback edition of The River, he has become "increasingly persuaded" that natural transfer can occur. But he still does not believe that this process was responsible for the AIDS epidemic. Hooper argues that natural transfers of SIVcpz simply "do not result in well-adapted, viable human viruses." Preston Marx, who has appointments at the Tulane Regional Primate Research Center, in Covington, Louisiana, and at the prominent Aaron Diamond AIDS Research Center, in Manhattan, and who has pursued the origin question as persistently as anyone, more or less shares that perspective, and also agrees with Hooper that the epidemic arose in the 1950s in the region of the Congo. He differs dramatically from Hooper, however, when it comes to explaining how the epidemic began.

When assessing causality, scientists use a gauge that they call "necessary and sufficient." HIV, for example, is both necessary and sufficient to cause AIDS: you can't get the disease without the virus, and the virus in itself can cause the disease. Hooper thinks that natural transfer is neither necessary nor sufficient to explain the origin of the AIDS epidemic. Marx, who finds the OPV theory implausible, thinks that natural transfer is necessary but not sufficient.

The OPV theory has several flaws, Marx contends. Not even the discovery of SIVcpz in the CHAT samples would prove that the vaccination campaign caused the epidemic, although it would certainly bolster Hooper's case. The amount of virus in the vaccine that was administered may have been infinitesimally small, or it may not have been virulent enough to establish an infection. Without a positive result from the CHAT samples, the theory that some Lindi chimps were infected with SIVcpz cannot be proved: there are no stored blood or tissue samples from the chimps. And chimps apparently do not become ill from the virus, so records of the health of the Lindi chimps (which, oddly, Hooper cites) are meaningless. In any event, Marx questions whether spraying the vaccine into people's mouths—the method by which it was administered—would transmit many viable SIVcpz infections. "Simply dropping virus in the throat is not enough to cause the epidemic," he says. "If pure mucosal exposure to SIV could induce an AIDS epidemic, then we would have seen AIDS a long time ago. Mucosal exposure has been going on for tens of thousands of years in places where they hunt mangabeys and chimps."

Marx brings up mangabeys—a species of large monkey—because they relate to an entirely separate and frequently ignored aspect of the epidemic. When people refer to HIV, they generally mean HIV-1, the most common type of the virus. But a second, weaker, type exists, HIV-2, which is found mainly in West African countries such as Guinea-Bissau, Côte d'Ivoire, the Gambia, and Sierra Leone. "Everyone thinking about this has to deal with the fact that these two diseases emerged within twenty-five years of each other, if not ten years of each other," Marx says. "Some event had to spark that. You can't deal with just one."

As of last January researchers had found at least eighteen distinct SIVs in twenty African primate species. They have known for more than a decade that one of these viruses has strong similarities to HIV-2. It was first isolated in captive sooty mangabeys, and is referred to as SIVsm.

In the early 1990s Telfer and Marx found SIVs in several wild sooty mangabeys in Sierra Leone. They then screened the blood of 9,309 people in the area to see if any were infected with a similar virus. Two people had HIV-2 infections. Although in one case the virus was a common strain, in the other the strain was unique, found in no other people, and it was closely related to an SIVsm isolate. This constitutes "absolute proof," Marx says, that a natural transfer of SIVsm to human beings had occurred.

Marx finds it telling that so few people in their population sample had HIV-2 infections; to him, this fact indicates that SIVsm rarely establishes infections in people. He has found another unique HIV-2 strain, in neighboring Liberia. The existence of unique strains supports the idea that those strains are not easily passed from one person to another. Taken together, these observations suggest that SIVsm and its cousin SIVcpz needed help in order to adapt to human beings and cause the HIV-2 and HIV-1 epidemics. Marx believes that the influx of needles to Africa after World War II, and their improper use there, are to blame.

To test his theory, Marx teamed up with Ernest Drucker, an epidemiologist at the Montefiore Medical Center, in New York. "What we found, looking at the history of injections since World War Two, is that there was tremendous demand for injectable chloroquine in the 1950s, then smallpox vaccine, then antibiotics," Marx explains. "People in Africa learned that injections could often cure a serious illness, and they came to demand them. But at the same time, the medical infrastructure in many parts of Africa was degrading." Marx acknowledges the limitations of this epidemiological argument. "We're dealing in correlations, not proof," he says.

Hooper has attempted to explain the HIV-2 epidemic by introducing another contaminated-polio-vaccine hypothesis. The vaccine he points to here, one developed by Pierre Lépine, of the Pasteur Institute, in Paris, has an even sketchier history than CHAT. Hooper has pieced together much of the history with the help of Simon Wain-Hobson, a leading AIDS researcher who now works at the Pasteur. Hooper posits a number of scenarios in which SIVsm could have contaminated some of Lépine's vaccine and been transmitted to people in West Africa. However, there is no evidence that Lépine used sooty-mangabey kidneys to make the vaccine, or that it was given in the areas that subsequently developed the worst HIV-2 epidemics. "The arguments about how SIVcpz and SIVsm found their way into vaccines are highly contrived," Marx told me. "Anybody can sit here and make up a story."

Inspector Javert?

HOOPER gets on my nerves," Beatrice Hahn, the University of Alabama AIDS researcher, told me recently. Hahn objects not so much to Hooper's research, although she does disagree with his theory, as to his manner and tactics, which she refers to as "annoying bullshit." I heard similar sentiments from several of the researchers I spoke with.

Part of the reason that so much controversy and emotion attend the origin-of-AIDS debate has to do with Hooper himself. He has a brilliant command of facts and the tenacity of a prosecuting attorney, but he is argumentative to a fault and deeply suspicious of anyone who disagrees with his views. "Try to rise above the defensiveness that surrounds the issue of origin, especially over in the States," he e-mailed me at one point. "We still have much to learn—on that I do agree with you. But The River has demonstrated that there is, at the least, a case to answer on the OPV/AIDS theory. Let's address it full on, not ignore it. So be fair. Be honest. Be even-handed. Please." Beatrice Hahn and other natural-transfer advocates, he wrote, "may be unconsciously driven by a desire to arrive at a 'politically correct' explanation for AIDS." Nature magazine has "a deep-seated bias against the hypothesis."

The fact that few AIDS researchers put stock in the OPV/AIDS thesis rankles Hooper. "This theory has been dealt with less than fairly by the scientific community," he charged in another e-mail. "People are trying to dismiss this theory by resorting to the worst type of emotional diatribes."

To some extent Hooper is right: scientists and scientific institutions have been vitriolic in their attacks on OPV/AIDS proponents, and they have dragged their feet regarding experiments that could help to resolve the issue. The Rolling Stone article that introduced the theory drew a libel suit from Hilary Koprowski, as did a subsequent piece by an Associated Press reporter (both were settled out of court). Koprowski's lawyer also sent Hooper a letter, putting him "on notice ... that any publication that is scientifically unsound and therefore obviously defamatory in nature will be promptly pursued in the appropriate courts." Several researchers discouraged Hooper from pursuing his hypothesis. Meanwhile, the Wistar Institute failed to follow through on a 1992 recommendation by an outside panel of experts that it test the one relevant vaccine sample still on hand. It finally sent portions of the sample to three independent labs last spring.

And Hooper's analysis of why so many scientists are so agitated about his theory makes a certain amount of sense. "Different groups are threatened by the arguments; that's the heart of it," he says. The idea that CHAT sparked the AIDS epidemic, he contends, scares Beatrice Hahn and other phylogeneticists because it undermines their work. Vaccine makers and advocates fear that the public could lose faith in vaccines, and the governments that supported the CHAT trials are loath to be held responsible for the epidemic. As for scientists who might be tempted to pursue the theory, he says, they worry that their colleagues might ostracize them for stepping out of line and that their funding could suffer.

In what seems an attempt to convey that he is not a fanatical Inspector Javert, Hooper asserts that he does not intend to pursue the OPV/AIDS hypothesis forever. "To be frank, I can hardly wait for the day when I can, in clear conscience, leave the investigation of this business to the scientists, the day when I can withdraw, publicly, from the debate," he told me in an e-mail. "But I won't be doing this until I am certain that Science is really going to investigate the issue properly, fully and honestly. Not much sign of that at present, I fear—bar the [Royal Society] meeting."

In the end, though, Hooper's dire portrait of capital-S science is overdrawn. A group of devoted scientists, albeit a small one, has been addressing the issue properly, fully, and honestly. That Hooper finds their work wanting reveals more about him than about the scientists and their efforts.

Calamities in the Wild

IN January of last year thousands of researchers from around the world gathered in Chicago for the 1999 Conference on Retroviruses and Opportunistic Infections—the largest annual AIDS meeting held in the United States. Beatrice Hahn stole the show with new data about the origin of HIV and AIDS.

In 1994 a researcher at the National Cancer Institute had cleaned out his freezer, where he had stored the frozen liver, spleen, and brain of a chimpanzee named Marilyn. He called Hahn, because nearly a decade earlier Marilyn had had an odd blood-test result. She had been caught in an unspecified African country and shipped to Holloman Air Force Base, in New Mexico, in 1963; she spent the rest of her life in various primate colonies in the United States. In 1985 researchers looking for HIV antibodies in primates drew blood from Marilyn and ninety-seven other chimps living in a colony in Alamogordo, New Mexico. Marilyn had the only positive test result. She soon gave birth to stillborn twins and then died herself; her body parts ended up in the NCI freezer.

With the technology available at the time, scientists had been unable to isolate SIVcpz from Marilyn's tissues. Hahn extracted pieces of the virus from the samples; the genetic sequence of the strain looked similar to the sequences of the two SIVcpzs that had been found in Gabon at Franceville. On a viral family tree all three of the chimps' viruses are clustered near various strains of HIV-1; this suggested to Hahn that they had all branched off from a common ancestor relatively recently. However, they had distinct differences from the virus in Noah, the chimp in the Antwerp zoo (the infected Cameroonian chimps had not yet been discovered).

At the meeting in Chicago, Hahn proposed a solution to this puzzle, one that would appear later that week in a paper in Nature. The combination of the high-profile conference, the publication in a prestigious journal, and Hahn's decision to use her data to assail the African trade in bush meat (generally, meat from chimps and other primates) led to front-page stories in newspapers in the United States and elsewhere. The New York Times story opened with the overblown assertion that "the riddle of the origin of the AIDS virus has apparently been solved."

Although Hahn's study did significantly bolster the case that chimps were the "natural reservoir" of HIV-1 (Marilyn, with the fourth SIVcpz isolate, increased the evidence by 33 percent), her real message became lost or garbled in much of the press coverage. Hahn had analyzed the mitochondrial DNA of all four chimps known to be infected. She suggested that Marilyn and the two Gabonese animals belonged to one of four subspecies, Pan troglodytes troglodytes, whereas Noah was a Pan troglodytes schweinfurthii. Because chimps do not naturally cross rivers, most scientists at the time believed that each subspecies had a restricted range, with P.t. troglodytes found from Cameroon to Gabon and P.t. schweinfurthii found in the Congo, including in the area where Camp Lindi had been. Hahn therefore asserted that the HIV-1 epidemic could not have begun in the Congo—a conclusion that makes the OPV theory that much more unlikely. (Hahn's thinking about subspecies would later change, but her conclusion would not.)

Hooper lost no time in mounting a response. In June of last year, hoping to learn more about the Camp Lindi chimps, he traveled to Kisangani. He was accompanied by the pre-eminent Oxford University evolutionary biologist William Hamilton, who wrote the foreword to The River and gave Hooper the equivalent of $3,000 toward his research. Once in Kisangani, Hooper and Hamilton split up the tasks: Hooper interviewed former workers from Lindi, while Hamilton collected fecal samples from pet chimps in the area. Hooper hoped that analyses of chimp and viral DNA found in the feces would prove that schweinfurthii chimps carry an SIVcpz similar to the most common strains of HIV-1 in human beings: such a finding, he believed, would refute Hahn's theory and bolster the OPV case. And he knew that another species, Pan paniscus, also lived in the area and had been used at Lindi; perhaps it, too, could provide clues.

Last January, Hamilton set out for Kisangani again, this time to collect fecal and urine samples from wild chimps. Not only would doing so increase the sheer number of samples, but also, he reasoned, it would increase the proportion of samples likely to contain the virus: pet chimps are typically caught as infants, and it is believed that SIV infections rarely pass from mother to baby—thus infections are more commonly found in adults. Hamilton was accompanied by an Oxford graduate student named Michael Worobey and by one of Worobey's friends; both had been fire fighters and were experienced at climbing trees—a skill that would be helpful in reaching chimpanzees' nests.

Although the group has not revealed the results as of this writing, it did collect several dozen fecal and urine samples—at an extremely high cost. On the first day in the forest Worobey spiked his hand on a palm tree. Four days later, his hand severely swollen, he left the others and walked twelve miles to the nearest village, where he got a bush taxi to Kisangani. By the time he reached medical help, he had developed blood poisoning.

Hamilton returned from the forest two days later. A few days after that he began to feel sick, and he spent most of one day in bed. "The next morning he informed us that he had the sweats," Worobey recalls. Unlike his traveling companions, Hamilton had not taken anti-malarial pills; he had contracted the disease years before, in the Amazon, and believed that his immune system could defeat a new infection. He tried a malaria treatment from a pharmacy, but the group decided to head back to London right away. "By the time we got to Heathrow, Bill was as good as we'd seen him," Worobey says. The next day, January 30, Hamilton felt ill again and went to a hospital, where he collapsed and slipped into a coma, his internal organs hemorrhaging. He died five weeks later.

Presented by

Jon Cohen is a correspondent with Science and the author of Almost Chimpanzee (Times Books/Henry Holt & Company, 2010), from which his October 2010 Atlantic story was adapted.

The Horrors of Rat Hole Mining

"The river was our source of water. Now, the people won't touch it. They are repulsed by it."

Join the Discussion

After you comment, click Post. If you’re not already logged in you will be asked to log in or register with Disqus.

Please note that The Atlantic's account system is separate from our commenting system. To log in or register with The Atlantic, use the Sign In button at the top of every page.

blog comments powered by Disqus


The Horrors of Rat Hole Mining

"The river was our source of water. Now, the people won't touch it."


What's Your Favorite Slang Word?

From "swag" to "on fleek," tweens choose.


Cryotherapy's Dubious Appeal

James Hamblin tries a questionable medical treatment.


Confessions of Moms Around the World

In Europe, mothers get maternity leave, discounted daycare, and flexible working hours.


How Do Trees Know When It's Spring?

The science behind beautiful seasonal blooming

More in Technology

More back issues, Sept 1995 to present.

Just In