scorecardresearch Skip to main content

Harvard lab delves into Ebola outbreak

Early this summer, tubes of inactivated Ebola virus from Sierra Leone began arriving at Harvard University, packed in dry ice. Scientists on the Cambridge campus worked in round-the-clock shifts to analyze the contents of the vials.

Their work was urgent — and deadly serious. The researchers oscillated between a rush to unravel the genetic fingerprint of the lethal virus and the futile wish that time would slow down, as they learned that the same scientists and medical workers in Africa who had collaborated with them were now being felled by the disease.

"We wake up, we learn some news, we bawl our eyes out, and we try and figure out what we can do," said Pardis Sabeti, a computational biologist at Harvard who supervised the work.


On Thursday in the journal Science, the team of scientists — including five African colleagues who died from Ebola before the research could appear — published the richest and most detailed portrait yet of the virus that is ravaging West Africa.

The researchers' analysis of Ebola samples taken from 78 patients in Sierra Leone reveals myriad small ways the pathogen has changed, accumulating 341 mutations that set it apart from past outbreaks.

The study also tracks how the virus spread into Sierra Leone in late May, when 14 women attended the funeral of a healer who had been working in Guinea. Researchers analyzed samples taken from a dozen of those women and found they carried back from the funeral two slightly different strains of the virus, the researchers concluded.

As the scientists made their discoveries, those findings were made available nearly in real time. Ebola samples would arrive in Cambridge, and within two weeks highly specialized equipment at the Broad Institute would generate genomic blueprints — a remarkable timeline, outside researchers said.


A decade ago, "if we had outbreaks, whatever the outbreak was — dengue or chikungunya — two years after the epidemic you would say scientists working for the past two years have now shown this original strain originated in Indonesia.

"Whereas now, in real time, as the epidemic is going on, we're able to do that pinpointing in the most elegant, specific, and sensitive manner," said Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases. "From a scientific standpoint, that is phenomenal."

Because the findings are being generated and made public at such lightning speed, scientists have had little time to draw many conclusions. Keeping the genomes private so that Sabeti and her lab could be the first to examine them fully — as is traditionally done in science — wasn't an option. They began posting the genome sequences of the virus online in late June, hoping that crowd-sourcing the problems to laboratories around the world could more rapidly yield insights to help halt the spread of the outbreak, believed to be responsible for more than 1,500 deaths.

"There are two sides of it," said Sabeti, who led the research with a scientist from Tulane University, Robert Garry, and the late Dr. Sheik Humarr Khan of Sierra Leone, a specialist in treating viral hemorrhagic fevers who was infected with the virus and died in July. "My lab has never been this sad and never been this motivated since I started it."

Outside scientists said the genomes will provide clues that could help check whether tests used to detect the disease are keeping up with the mutating virus and allow scientists to trace the lineage of the virus, which they believe diverged from a Middle African strain of Ebola in 2004.


The data also show how rapidly the virus is mutating. Past outbreaks have burned out quickly and, therefore, have not given the virus as much of an opportunity to change in ways that make it more of a public health threat.

The durability of the ongoing outbreak means it is vital to keep an eye on how it is changing, Fauci said. So far, researchers have not seen evidence that the mutations are making the virus more dangerous.

But the researchers have already discovered that five standard tests used to diagnose Ebola in blood samples are not a perfect match for this strain. Now, they are testing whether those differences would have an effect on the test being able to certify someone has the virus.

"A small degree of variation can have a significant public health impact, depending on where it is and what it causes in the virus," said Dr. Daniel Bausch, an associate professor in the department of tropical medicine at Tulane School of Public Health. He was not involved in the work.

Bausch compared the diagnostic tests and the virus to a lock and key that need to match up in order to work. If there's a nick in the part of the key that goes into the lock, the door may not open, but if the nick resides on the handle of the key, it will still work. Right now, it's not clear whether the genetic changes the researchers have detected are in the handle or the body of the key and, thus, whether they would affect the tests.


Thomas Geisbert, a professor of microbiology and immunology at the University of Texas Medical Branch in Galveston, said he didn't immediately see any mutations that would affect the ability of ZMapp, an experimental therapy that has been used in several patients, to work.

The collaboration was made possible by relationships that developed for entirely different reasons. Years ago, Sabeti combed the human genome looking for genes that had been shaped by evolution. One of the genes highlighted in that study was known to be involved in infection with Lassa fever, a deadly viral hemorrhagic disease present in West Africa.

She began forging partnerships and setting up the laboratory infrastructure and training to study that virus on the ground in Sierra Leone and Nigeria, with an interest in how the virus and people had co-evolved. Over the years and after many trips to the Kenema Government Hospital in Sierra Leone, that academic interest developed into a rich collaboration and a network of friendships.

When Sabeti and colleagues got word that Ebola had been detected in Guinea, two scientists from her laboratory traveled to the hospital to bring with them primers, short strands of genetic material that would allow for early detection of the disease.


Stephen Gire, a research scientist in Sabeti's laboratory, returned with a team in July to streamline lab procedures.

"The burden of cases was overwhelming," Gire said, especially in the laboratory, where the workload had increased tenfold. He helped adjust procedures to ensure that workers would have time to drink water and take breaks, because working 18- to 20-hour days in protective coveralls was taking a toll on researchers.

Members of Sabeti's laboratory have not been able to travel back to Sierra Leone since July because of restrictions, but have been doing what they can to support friends and collaborators from afar.

"We can honor those who have died by making sure that we do everything we can to contribute to stopping this outbreak and understanding more about it for the future," Gire wrote in an e-mail. "And so that is why we have pushed and pushed to get this data out as quickly as possible."

Carolyn Y. Johnson can be reached at
Follow her on Twitter @carolynyjohnson.