Spillover

by David Quammen

  The new virus, SIV as found in African green monkeys, became the closest known relative of HIV. But it wasn’t that close; many differences distinguished the two at the level of genetic coding. The resemblance, according to Essex and Kanki, was “not close enough to make it likely that SIV was an immediate precursor of HIV in people.” More likely, those two viruses represented neighboring twigs on a single phylogenetic branch, separated by lots of evolutionary time and probably some extant intermediate forms. Where might the missing cousins be? “Perhaps, we thought, one could find such a virus—an intermediate between SIV and HIV—in human beings.” They decided to look in West Africa.

  With help from an international team of collaborators, Kanki and Essex gathered blood samples from Senegal and elsewhere. The samples arrived with coded labeling, for blind testing in the laboratory, so that Kanki herself didn’t know their country of origin, nor even whether they derived from humans or monkeys. She screened them using tests for both SIV and HIV. Despite one possible misstep involving a lab contamination, her team found what they had thought they might: a virus intermediate between HIV and SIV. With the code unblinded, Kanki learned that the positive results came from Senegalese prostitutes. In retrospect it made sense. Prostitutes are at high risk for any sexually transmitted virus, including a new one recently spilled into humans. And the density of the rural human population in Senegal, where African green monkeys are native, makes monkey-human interactions (crop-raiding by monkeys, hunting by humans) relatively frequent.

  Furthermore, the new bug from Senegalese prostitutes wasn’t just halfway between HIV and SIV. It more closely resembled SIV strains from African green monkeys than it did the Montagnier-Gallo version of HIV. That was important but puzzling. Were there two distinct kinds of HIV?

  Luc Montagnier now reenters the story. Having tussled with Gallo over the first HIV discovery, he converged more amicably with Essex and Kanki on this one. Using assay tools provided by the Harvard group, Montagnier and his colleagues screened the blood of a twenty-nine-year-old man from Guinea-Bissau, a tiny country, formerly a Portuguese colony, along the south border of Senegal. This man showed symptoms of AIDS (diarrhea, weight loss, swollen lymph nodes) but tested negative for HIV. He was hospitalized in Portugal, and his blood sample hand-delivered to Montagnier by a visiting Portuguese biologist. In Montagnier’s lab, the man’s serum again tested negative for antibodies to HIV. But from a culture of his white blood cells Montagnier’s group isolated a new human retrovirus, which looked very similar to what Essex and Kanki had found. In another patient, hospitalized in Paris but originally from Cape Verde, an island nation off the west coast of Senegal, the French team found more virus of the same type. Montagnier called the new thing LAV-2. Eventually, when all parties embraced the label HIV instead, it would be HIV-2. The original became HIV-1.

  The paths of discovery may be sinuous, the labels may seem many, and maybe you can’t tell the players without a scorecard; but these details aren’t trivial. The difference between HIV-2 and HIV-1 is the difference between a nasty little West African disease and a global pandemic.

  88

  During the late 1980s, as Kanki and Essex and other scientists studied HIV-2, a flurry of uncertainty arose about its provenance. Some challenged the idea that it was closely related to (and recently derived from) a retrovirus that infects African monkeys. An alternative view was that such a retrovirus had been present in the human lineage for as long as—or longer than—human time. Possibly it was already with us, a passenger riding the slow channels of evolution, when we diverged from our primate cousins. But that view left an unresolved conundrum: If the virus was an ancient parasite upon humans, unnoticed for millennia, how had it suddenly become so pathogenic?

  Recent spillover seemed more likely. Still, the case against that idea got a boost in 1988, when a group of Japanese researchers sequenced the complete genome of SIV from an African green monkey. The animal came from Kenya. The nucleotide sequence of its retrovirus proved to be substantially different from the sequence for HIV-1, and different in roughly the same degree from HIV-2. So the monkey virus seemed no more closely related to the one human virus than to the other. That contradicted the notion that HIV-2 had lately emerged from an African green monkey. A commentary in the journal Nature, published to accompany the Japanese paper, celebrated this finding beneath a dogmatic headline: HUMAN AIDS VIRUS NOT FROM MONKEYS. But the headline was misleading to the point of falsity. Not from monkeys? Well, don’t be so sure. It turned out that researchers were just looking at the wrong kind of monkey.

  Confusion came from two sources. For starters, the label “African green monkey” is a little vague. It encompasses a diversity of forms, sometimes also known as savannah monkeys, that occupy adjacent geographical ranges spread out across sub-Saharan Africa, from Senegal in the west to Ethiopia in the east and down into South Africa. At one time those forms were considered a “superspecies” under the name Cercopithecus aethiops. Nowadays, their differences having been more acutely gauged, they are classified into six distinct species within the genus Chlorocebus. The “African green monkey” sampled by the Japanese team, because it was “of Kenyan origin,” probably belonged to the species Chlorocebus pygerythrus. The species native to Senegal, on the other hand, is Chlorocebus sabaeus. Now that you’ve seen those two names you can forget them. The difference between one African green monkey and another is not what accounts for the genetic disjunction between SIV and HIV-2.

  The trail backward from HIV-2 led to another monkey entirely: the sooty mangabey. This is not one of the six Chlorocebus species, not even close. It belongs to a different genus.

  The sooty mangabey (Cercocebus atys) is a smoky-gray creature with a dark face and hands, white eyebrows, and flaring white muttonchops, not nearly so decorative as many monkeys on the continent but arresting in its way, like an elderly chimney sweep of dapper tonsorial habits. It lives in coastal West Africa, from Senegal to Ghana, favoring swamps and palm forests, where it eats fruit, nuts, seeds, leaves, shoots, and roots—an eclectic vegetarian—and spends much of its time on the ground, moving quadrupedally in search of fallen tidbits. Sometimes it ventures out of the bottomlands to raid farms and rice paddies. The sooty mangabey is hard to hunt within the swampy forests but, because of its terrestrial foraging habits and its taste for crops, easy to trap. Local people treat it as an annoying but edible sort of vermin. Sometimes also, if they’re not too hungry, they adopt an orphan juvenile as a pet.

  What brought the sooty mangabey to the attention of AIDS researchers was coincidence and an experiment on leprosy. It was an instance of the old scientific verity that sometimes you find much more than you’re looking for.

  Back in September 1979, scientists at a primate research center in New Iberia, Louisiana, south of Lafayette, noticed a leprosy-like infection in one of their captive monkeys. This seemed odd, because leprosy is a human disease caused by a bacterium (Mycobacterium leprae) not known to be transmissible from people to other primates. But here was a leprous monkey. The animal in question, a sooty mangabey, female, about five years old, had been imported from West Africa. The researchers called her Louise. Apart from her skin condition, Louise was healthy. She hadn’t, so far as the records showed, yet been subjected to any experimental infection. They were using her in a study of diet and cholesterol. The New Iberia facility didn’t happen to work on leprosy infections, so once Louise’s condition had been recognized she was transferred to a place, also in Louisiana, that did: the Delta Regional Primate Research Center, north of Lake Pontchartrain. The researchers at Delta were glad to get her, for one very practical reason. If Louise had acquired her leprosy naturally, then (contrary to previous suppositions) the disease might be transmissible in populations of sooty mangabey. And if that were true, then the sooty mangabey could prove valuable as an experimental model for studies of human leprosy.

  So the Delta team injected some infectious material from Louise into another sooty mangabey. T
his one was a male. Unlike Louise, he’s nameless in the scientific record, remembered only by a code: A022. He became the first in a chain of experimentally infected monkeys that turned out to carry more than leprosy. The scientists at Delta had no idea, not at first, that A022 was SIV-positive.

  The leprosy from Louise took hold easily in A022, which was notable, given that earlier attempts to infect monkeys with human leprosy had failed. Was this strain of Mycobacterium leprae a peculiarly monkey-adapted variant? If so, might it succeed in rhesus macaques too? That would be convenient for experimental purposes, because rhesus macaques were far cheaper and more available, in the medical-research chain of supply, than sooty mangabeys. So the Delta team injected four rhesus macaques with infectious gunk from A022. All four developed leprosy. For three of the four, that proved to be the least of their troubles. The unlucky three also developed simian AIDS. Suffering chronic diarrhea and weight loss, they wasted away and died.

  Screening for virus, the researchers found SIV. How had their three macaques become SIV-positive? Evidently by way of the leprous inoculum from the sooty mangabey, A022. Was he unique? No. Tests of other sooty mangabeys at Delta revealed that the virus was “endemic” among them. Other investigators soon found it too, not just among captive sooty mangabeys but also in the wild. Yet the sooty mangabeys (native to Africa), unlike the rhesus macaques (native to Asia), showed no symptoms of simian AIDS. They were infected but healthy, which suggested that the virus had a long history in their kind. The same virus made the macaques sick, presumably because it was new to them.

  The roster of simian immunodeficiency viruses was growing more crowded and complex. Now there were three known variants: one from African green monkeys, one from rhesus macaques (which they probably acquired in captivity), and one from sooty mangabeys. Needing a way to identify and distinguish them, someone hit upon the expedient of adding tiny subscripts to the acronym. Simian immunodeficiency virus as found in sooty mangabeys became SIVsm. The other two were labeled SIVagm (for African green monkeys) and SIVmac (for Asian macaques). This little convention may seem esoteric, not to mention hard on the eyes, but it will be essential and luminous when I discuss the fateful significance of a variant that came to be known as SIVcpz.

  For now it’s enough to note the upshot of the leprosy experiment in Louisiana. One scientist from the Delta team, a woman named Michael Anne Murphey-Corb, collaborated with molecular biologists from other institutions to scrutinize the genomes of SIVs from sooty mangabeys and rhesus macaques, and to create a provisional family tree. Their work, published in 1989 with Vanessa M. Hirsch as first author, revealed that SIVsm is closely related to HIV-2. So is SIVmac. “These results suggest that SIVsm has infected macaques in captivity and humans in West Africa,” the group wrote, placing the onus of origination on sooty mangabeys, “and evolved as SIVmac and HIV-2, respectively.” In fact, those three strains were very similar, suggesting divergence fairly recently from a common ancestor.

  “A plausible interpretation of these data,” Hirsch and her coauthors added, to make the point plainly, “is that in the past 30–40 years SIV from a West African sooty mangabey (or closely related species) successfully infected a human and evolved as HIV-2.” It was official: HIV-2 is a zoonosis.

  89

  But what about HIV-1? Where did the great killer come from? That larger mystery took somewhat longer to solve. The logical inference was that HIV-1 must be zoonotic in origin also. But what animal was its reservoir? When, where, and how had spillover occurred? Why had the consequences been so much more dire?

  HIV-2 is both less transmissible and less virulent than HIV-1. The molecular bases for those fateful differences are still secrets embedded in the genomes, but the ecological and medical ramifications are clear and stark. HIV-2 is confined mostly to West African countries such as Senegal and Guinea-Bissau (the latter of which, during colonial times, was Portuguese Guinea), and to other areas connected socially and economically within the former Portuguese empire, including Portugal itself and southwestern India. People infected with HIV-2 tend to carry lower levels of virus in their blood, to infect fewer of their sexual contacts, and to suffer less severe or longer-delayed forms of immune deficiency. Many of them don’t seem to progress to AIDS at all. And mothers who carry HIV-2 are less likely to pass it to their infants. The virus is bad, but not nearly so bad as it could be. HIV-1 provides the comparison. HIV-1 is the thing that afflicts tens of millions of people throughout the world. HIV-1 is the pandemic scourge. To understand how the AIDS catastrophe has happened to humanity, scientists had to trace HIV-1 to its source.

  This takes us circling back to the city of Franceville, in southeastern Gabon, and its Centre International de Recherches Médicales (CIRMF), the same research institute at which Eric Leroy would later base his studies of Ebola. At the end of the 1980s, a young Belgian woman named Martine Peeters worked as a research assistant at CIRMF for a year or so, during the period between getting her diploma in tropical medicine and going on for a doctorate. The CIRMF facility maintained a compound of captive primates, including three dozen chimpanzees, and Peeters along with several associates was tasked with testing the captive animals for antibodies to HIV-1 and HIV-2. Almost all of the chimps tested negative—all except two. Both the exceptions were very young females, recently captured from the wild. Such baby chimps, like other orphan primates, are sometimes kept or sold off as pets after the killing and eating of their mothers. One of these animals, a two-year-old suffering from gunshot wounds, had been brought to CIRMF for medical treatment. She died of the wounds, but not before surrendering a blood sample. The other was an infant, maybe six months old, who survived. Blood serum from each of them reacted strongly when tested against HIV-1, less strongly when tested against HIV-2. That much was notable but slightly ambiguous. Antibody testing is an indirect gauge of infection, relatively convenient and quick, but imprecise. Greater precision comes with detecting fragments of viral RNA or, better still, isolating a virus—catching the thing in its wholeness and growing it in quantity—from which a confident identification can be made. Martine Peeters and her co-workers succeeded in isolating a virus from the baby chimp. Twenty years later, when I called on her at her office at an institute in southern France, Peeters remembered vividly how that virus showed up in a series of molecular tests.

  “It was especially surprising,” she said, “because it was so close to HIV-1.”

  Had there been any previous hints?

  “Yes. At that time we knew already that HIV-2 most likely came from primates in West Africa,” she said, alluding to the sooty mangabey work. “But there was no virus close to HIV-1 already detected in primates. And until now, it’s still the only virus close to HIV-1.” Her group had published a paper, in 1989, announcing the new virus and calling it SIVcpz. They did not crow about having found the reservoir of HIV-1. Their conclusion from the data was more modest: “It has been suggested that human AIDS retroviruses originated from monkeys in Africa. However, this study and other previous studies on SIV do not support this suggestion.” Left implicit: Chimpanzees, not monkeys, might be the source of the pandemic bug.

  By the time I met her, Martine Peeters was director of research at the Institut de Recherche pour le Développement, in Montpellier, a handsome old city just off the Mediterranean coast. She was a small, blonde woman in a black sweater and silver necklace, concise and judicious in conversation. What sort of response had met this discovery? I asked.

  “HIV-2, people accepted it readily.” They accepted, she meant, the notion of simian origins. “But HIV-1, people had more difficulties to accept it.”

  Why the resistance? “I don’t know why,” she said. “Maybe because we were young scientists.”

  The 1989 paper got little attention, which seems peculiar in retrospect, given the novelty and gravity of what it implied. In 1992 Peeters published another, describing a third case of SIVcpz, this one in a captive chimpanzee that had been shipped to Brussels from Zaire.
All three of her SIV-positive results had been in “wild-born” chimpanzees taken captive (as distinct from animals bred in captivity) but that still left a gap in the chain of evidence. What about chimps still in the wild?

  With only such tools of molecular biology as available in the early 1990s, the screening of wild chimps was difficult (and unacceptable to most chimp researchers), because the diagnostic tests required blood sampling. Lack of evidence from wild populations, in turn, contributed to skepticism in the AIDS-research community about the link between HIV-1 and chimps. After all, if Asian macaques had become infected with HIV-2 in their cages, from contact with African monkeys, might not SIV-positive chimpanzees simply reflect cage-contact infections too? Another reason for skepticism was the fact that, by the end of the 1990s, roughly a thousand captive chimpanzees had been tested but, apart from Peeters’s three, not a single one had yielded traces of SIVcpz. These two factors—the absence of evidence from wild populations and the extreme rarity of SIV in captive chimps—left open the possibility that both HIV-1 and SIVcpz derived directly from a common ancestral virus in some other primate. In other words, maybe those three lonely chimps had gotten their infections from some still-unidentified monkey, and maybe the same unidentified monkey had given HIV-1 to humans. With that possibility dangling, the origin of HIV-1 remained uncertain for much of the decade.

  In the meantime, researchers investigated not just the source of HIV but also its diversity in humans, discovering three major lineages of HIV-1. “Groups” became the preferred term for these lineages. Each group was a cluster of strains that was genetically discrete from the other clusters; there was variation within each group, since HIV is always evolving, but the differences between groups were far larger. This pattern of groups had some dark implications that dawned on scientists only gradually and still haven’t been absorbed in the popular understanding of AIDS. I’ll get to them shortly, but first let’s consider the pattern itself.