Spillover

by David Quammen

  One community of those chimps had been followed and studied for thirteen years by a Swiss biologist named Christophe Boesch. During the 1992 episode, Boesch and his colleagues noticed a sudden drop in the population—some chimps died, others disappeared—but the scientists didn’t detect a cause. Then, in late 1994, eight more carcasses turned up over a short period of time, and again other animals went missing. Two of the chimp bodies, only moderately decayed, were cut open and examined by researchers at Taï. One of those proved to be teeming with an Ebola-like agent, though that wasn’t apparent at the time. During the necropsy, a thirty-four-year-old female Swiss graduate student, wearing gloves but no gown, no mask, became infected. Infected how? There wasn’t any obvious moment of fateful exposure, no slip of the scalpel, no needlestick mishap. Probably she got chimp blood onto a broken patch of skin—a small scratch?—or caught a gentle splash of droplets in the face. Eight days later, the woman started shivering.

  She took a dose of malaria medicine. That didn’t help. She was moved to a clinic in Abidjan, Côte d’Ivoire’s capital, and there treated again for malaria. Her fever continued. On day 5 came vomiting and diarrhea, plus a rash that spread over her whole body. On day 7 she was carried aboard an ambulance jet and flown to Switzerland. Now she was wearing a mask, and so were the doctor and the nurse in attendance. But no one knew what ailed her. Dengue fever, hantavirus infection, and typhoid were being considered, and malaria still hadn’t been ruled out. (Ebola wasn’t at the top of the list because it had never been seen in Côte d’Ivoire.) In Switzerland, hospitalized within a double-door isolation room with negative air pressure, she was tested for a whole menu of nasty things, including Lassa fever, Crimean-Congo hemorrhagic fever, chikungunya, yellow fever, Marburg virus disease, and now, yes, Ebola virus disease. The last of those possibilities was investigated using three kinds of assays, each one specific: for Ebola virus, for Sudan virus, for Reston virus. No positive results. The antibodies in those assays didn’t recognize the virus, whatever it was, in her blood.

  The laboratory sleuths persisted, designing a fourth assay that was more generalized—comprehensive for the whole group of ebolaviruses. Applied to her serum, that one glowed, a positive, announcing the presence of antibodies to an ebolavirus of some sort. So the Swiss woman was the world’s first identified victim of what became known as Taï Forest virus. The chimpanzee she had necropsied, its tissues tested later, was the second victim, recognized posthumously.

  Unlike the chimp, she survived. After another week, she left the hospital. She had lost thirteen pounds and her hair later fell out, but otherwise she was okay. Besides being the initial case of Taï Forest virus infection, the Swiss woman holds one other distinction: She is the first person known to have carried an ebolavirus infection off the African continent. There is no reason to assume that she will be the last.

  13

  Ebolavirus spillovers continued throughout the 1990s and into the twenty-first century, sporadic and scattered enough to make field research difficult, frequent enough to keep some scientists focused and some public health officials worried. In 1995, soon after the Côte d’Ivoire episode, it was Ebola virus in Kikwit, about which you’ve read. Six months after that outbreak, as you’ll also recall, the new one began at Mayibout 2. What I haven’t yet mentioned about Mayibout 2 is that, though the village lies in Gabon, the virus was Ebola as known originally from Zaire, which seems to be the most broadly distributed of the group. At the timber camp near Booué, Gabon, it was Ebola virus.

  Also that year, 1996, Reston virus reentered the United States by way of another shipment of Philippine macaques. Sent from the same export house near Manila that had shipped the original sick monkeys to Reston, Virginia, these went to a commercial quarantine facility in Alice, Texas, near Corpus Christi. One animal died and, after it tested positive for Reston virus, forty-nine others housed in the same room were euthanized as a precaution. (Most of those, tested posthumously, were negative.) Ten employees who had helped unload and handle the monkeys were also screened for infection, and they also tested negative, but none of them were euthanized.

  Uganda became the next known locus of the virus in Africa, with an outbreak of Sudan virus that began near the northern town of Gulu in August 2000. Northern Uganda shares a border with what in those days was southern Sudan, and it wasn’t surprising that Sudan virus might cross or straddle that border. Cross it how, straddle it how? By way of the individual movements or the collective distribution of the reservoir species, identity unknown. This is a pointed example of why solving the reservoir mystery is important: If you know which animal harbors a certain virus and where that animal lives—and conversely, where it doesn’t live—you know where the virus may next spill over, and where it probably won’t. That provides some basis for focusing your vigilance. If the reservoir is a rodent that lives in the forests of southwestern Sudan but not in the deserts of Niger, the goat herders of Niger can relax. They have other things to worry about.

  In Uganda, unfortunately, the 2000 spillover led to an epidemic of Sudan virus infections that spread from village to village, from hospital to hospital, from the north of the country to the southwest, killing 224 people.

  The case fatality rate was again “only” 53 percent, exactly what it had been in the first Sudanese outbreak, back in 1976. This precise coincidence seems to reflect a significant difference in virulence between Sudan and Ebola viruses. Their difference, in turn, might reflect different evolutionary adjustments to humans as a secondary host (though random happenstance is also a possible explanation). Many factors contribute to the case fatality rate during an outbreak, including diet, economic conditions, public health in general, and the medical care available in the location where an outbreak occurs. It’s hard to isolate the inherent ferocity of a virus from those contextual factors. What can be said, though, is that Ebola virus appears to be the meanest of the four ebolaviruses you’ve heard about, as gauged by its effect on human populations. Taï Forest virus can’t reliably be placed on that spectrum at all, not yet—for lack of evidence. Having infected just one known human (or possibly two, counting an unconfirmed later case) and killed none, Taï Forest virus may be less prone to spillover. It may or may not be less lethal; one case, like one roll of the dice, proves nothing about what’s likely to emerge as numbers grow larger. Then again, Taï Forest virus might also be spilling more frequently but inconsequentially—infecting people yet not causing notable illness. No one has screened the populace of Côte d’Ivoire to exclude that possibility.

  The role of evolution in making Taï Forest virus (or any virus) less virulent in humans is a complicated matter, not easily deduced from simple comparison of case fatality rates. Sheer lethality may be irrelevant to the virus’s reproductive success and long-term survival, the measures by which evolution keeps score. Remember, the human body isn’t the primary habitat of ebolaviruses. The reservoir host is.

  Like other zoonotic viruses, ebolaviruses have probably adapted to living tranquilly within their reservoir (or reservoirs), replicating steadily but not abundantly and causing little or no trouble. Spilling over into humans, they encounter a new environment, a new set of circumstances, often causing fatal devastation. And one human can infect another, through direct contact with bodily fluids or other sources of virus. But the chain of ebolavirus infection, at least so far, has never continued through many successive cases, great distances, or long stretches of time. Some scientists use the term “dead-end host,” as distinct from “reservoir host,” to describe humanity’s role in the lives and adventures of ebolaviruses. What the term implies is this: Outbreaks have been contained and terminated; in each situation the virus has come to a dead end, leaving no offspring. Not the virus in toto throughout its range, of course, but that lineage of virus, the one that has spilled over, betting everything on this gambit—it’s gone, kaput. It’s an evolutionary loser. It hasn’t caught hold to become an endemic disease within human populations. It hasn’t
caused a huge epidemic. Ebolaviruses, judged by experience so far, fit that pattern. Careful medical procedures (such as barrier nursing by way of isolation wards, latex gloves, gowns, masks, and disposable needles and syringes) usually stop them. Sometimes simpler methods can bring a local spillover to a dead end too. This has probably happened more times than we’ll ever know. Advisory: If your husband catches an ebolavirus, give him food and water and love and maybe prayers but keep your distance, wait patiently, hope for the best—and, if he dies, don’t clean out his bowels by hand. Better to step back, blow a kiss, and burn the hut.

  This business about dead-end hosts is the conventional wisdom. It applies to the ordinary course of events. But there’s another perspective to consider. Zoonoses by definition involve events beyond the ordinary, and the scope of their consequences can be extraordinary too. Every spillover is like a sweepstakes ticket, bought by the pathogen, for the prize of a new and more grandiose existence. It’s a long-shot chance to transcend the dead end. To go where it hasn’t gone and be what it hasn’t been. Sometimes the bettor wins big. Think of HIV.

  14

  In late 2007 a fifth ebolavirus emerged, this time in western Uganda.

  On November 5, 2007, the Ugandan Ministry of Health received a report of twenty mysterious deaths in Bundibugyo, a remote district along the mountainous border with the Democratic Republic of the Congo (the new name, as of 1997, for what had been Zaire). An acute infection of some unknown sort had killed those twenty people, abruptly, and put others at risk. Was it a rickettsial bacterium, such as the one that causes typhus? An ebolavirus was another possibility, but considered less likely at first, because few of the patients hemorrhaged. Blood samples were gathered quickly, flown to the CDC in Atlanta, and tested there, using both a generalized assay that might detect any form of ebolavirus and specific assays for each of the known four. Although the specific tests were all negative, the general test rang up some positives. So on November 28, the CDC informed Ugandan officials: It’s an ebolavirus, all right, but not one we’ve ever seen.

  Further laboratory work established that this new virus was at least 32 percent different genetically from any of the other four. It became Bundibugyo virus. Soon a CDC field team arrived in Uganda to help respond to the outbreak. As usual in such situations, their efforts along with those of the national health authorities involved three tasks: caring for patients, trying to prevent further spread, and investigating the nature of the disease. The eventual tally was 116 people infected, of whom 39 died.

  Also as usual, the scientific team later published a journal article, in this case announcing the discovery of a new ebolavirus. First author on the paper was Jonathan S. Towner, a molecular virologist at the CDC with field experience in the search for reservoirs. Besides guiding the lab work, he went to Uganda and did a stint with the response team. The Towner paper contained a very interesting statement, as an aside, concerning the five ebolaviruses: “Viruses of each species have genomes that are at least 30–40% divergent from one another, a level of diversity that presumably reflects differences in the ecologic niche they occupy and in their evolutionary history.” Towner and company suggested that some of the crucial differences between one ebolavirus and another—including the differences in lethality—might be related to where and how they live, where and how they have lived, within their reservoir hosts.

  The events in Bundibugyo left many Ugandans uneasy. And they were entitled to their uneasiness: Uganda now held a sorry distinction as the only country on Earth that had suffered outbreaks of two different ebolaviruses (Sudan virus at Gulu in 2000, Bundibugyo virus in 2007), as well as outbreaks of both Ebola virus disease and Marburg virus disease, caused by another filovirus, within a single year. (The creepy circumstances of the Marburg spillover, at a gold mine called Kitaka in June 2007, are part of a story I’ll come to in its turn.) Given such national ill fortune, it’s not surprising that there were rumors, stories, and anxieties circulating among Ugandans, in late 2007, that made tracing genuine ebolavirus leads all the more difficult.

  A pregnant woman, showing signs of hemorrhagic fever, delivered her baby and then died. The baby, left in the care of a grandmother, soon died too. That was sad but not peculiar; orphaned infants often die in the hard conditions of a village. More notable was that the grandmother also died. An ape (chimp or gorilla?) reportedly bit a domestic goat, infecting it; the goat was slaughtered in due course, skinned by a thirteen-year-old boy, and then the boy’s family began falling ill. No, a dead monkey was eaten. No, bats were eaten. Mostly these tales couldn’t be substantiated, but their currency and their general themes reflected a widespread, intuitive comprehension of zoonoses: Relations between humans and other animals, wild or domestic, must somehow lie at the root of the disease troubles. In early December, and then again in January 2008, came reports of suspicious animal deaths (monkeys and pigs) in outlying regions of the country. One of those reports also involved dogs that died after being bitten by the sickened monkeys. Was it an epidemic of rabies? Was it Ebola? The Ministry of Health sent people to collect specimens and investigate.

  “Then there was a new epidemic—of fear,” said Dr. Sam Okware, Commissioner of Health Services, when I visited him in Kampala a month later. Among Dr. Okware’s other duties, he served as chairman of the national Ebola virus task force. “That was the most difficult to contain,” he said. “There was a new epidemic—of panic.”

  These are remote places, he explained. Villages, settlements, small towns surrounded by forest. The people feed themselves mostly on wildlife. During the Bundibugyo outbreak, residents of that area were shunned. Their economy froze. Outsiders wouldn’t accept their money, scared that it carried infection. Population drained from the major town. The bank closed. When patients recovered (if they were lucky enough to recover) and went home from the hospital, “again they were shunned. Their houses were burned.” Dr. Okware was a thin, middle-aged man with a trim mustache and long, gesticulant hands that moved through the air as he spoke of Uganda’s traumatic year. The Bundibugyo outbreak, he said, was “insidious” more than dramatic, smoldering ambiguously while health officials struggled to comprehend it. There were still five questions pending, he said, and he began to list them: (1) Why were only half of the members of each household affected? (2) Why were so few hospital workers affected, compared to other Ebola outbreaks? (3) Why did the disease strike so spottily within the Bundibugyo district, hitting some villages but not others? (4) Was the infection transmitted by sexual contact? After those four he paused, momentarily unable to recall his fifth pending question.

  “The reservoir?” I suggested. Yes, that’s it, he said: What’s the reservoir?

  Bundibugyo virus in Uganda, 2007, completes the outline sketch of ebolavirus classification and distribution as presently known. Four different ebolaviruses are scattered variously across Central Africa and have emerged from their reservoir hosts to cause human disease (as well as gorilla and chimpanzee deaths) in six different countries: South Sudan, Gabon, Uganda, Côte d’Ivoire, the Republic of the Congo, and the Democratic Republic of the Congo. A fifth ebolavirus seems to be endemic to the Philippines, and to have traveled from there several times to the United States in infected macaques. But how did it get to the Philippines, if the ancestral origin of ebolaviruses is equatorial Africa? Could it have arrived there in one soaring leap, leaving no traces in between? From southwestern Sudan to Manila is almost seven thousand miles as the bat flies. But no bat can fly that far without roosting. Are ebolaviruses more broadly distributed than we suspect? Should scientists start looking for them in India, Thailand, and Vietnam? Or did Reston virus get to the Philippines the same way Taï Forest virus got to Switzerland and Johannesburg—by airplane?

  If you contemplate all this from the perspective of biogeography (the study of which creatures live where on planet Earth) and phylogeny (the study of evolving lineages), one thing becomes evident: The current scientific understanding of ebolaviruses const
itutes pinpricks of light against a dark background.

  15

  People in the villages where Ebola struck—the survivors, the bereaved, the scared but lucky ones not directly affected—had their own ways of understanding this phenomenon, and one way was in terms of malevolent spirits. In a single word, which loosely encompasses the variety of beliefs and practices seen among different ethnic and language groups and is often used to explain rapid death of adults: sorcery.

  The village of Mékouka, on the upper Ivindo River in northeastern Gabon, offers an instance. Mékouka was one of the gold camps in which the outbreak of 1994 got its start. Three years later, a medical anthropologist named Barry Hewlett, an American, visited there to learn from the villagers themselves how they had thought about and responded to the outbreak. Many local people told him, using a term from their Bakola language, that this Ebola thing was ezanga, meaning some sort of vampirism or evil spirit. Asked to elaborate, one villager explained that ezanga are “bad human-like spirits that cause illness in people” as retribution for accumulating material goods and not sharing. (This wouldn’t seem to apply to that man on the upper Ivindo, in 1994, who reportedly shared his tainted gorilla meat before he died.) Ezanga could even be summoned and targeted at a victim, like casting a hex. Neighbors or acquaintances, envious of the wealth or power someone has amassed, could send ezanga to gnaw at the person’s internal organs, making him sick unto death. That’s why gold miners and timber-company employees suffered such high risk of Ebola, Hewlett was told. They were envied and they didn’t share.