January 2007, updated August 2007

Despite overwhelming evidence attesting to their innocence, last month six medical workers were sentenced to death in a Libyan trial. The crime with which the five Bulgarian nurses and one Palestinian doctor are charged is indeed horrifying. After an outbreak of HIV at the Al-Fateh hospital, the Libyan government accused the defendants of committing an act of bioterrorism by deliberately injecting 426 hospitalized children with HIV-tainted blood. The HIV strain is particularly virulent and has already contributed to the deaths of more than 50 of the infected children. The families of both the victims and the accused are enraged, and the controversy is escalating: the case has already seen one retrial and further appeals are likely. But what about science? What does the scientific evidence have to say?

Where's the evolution?

Many independent lines of evidence suggest that the accused nurses and doctor are innocent. The history of the outbreak, co-infection of the HIV-positive children with various hepatitis strains (suggesting repeated infection via dirty needles), and other indicators of unsafe medical practices all imply that the real culprit in this tragedy is the hospital itself. As a case in point, at least one child was infected with HIV at the hospital after the medics were arrested, when the medics could not have played a role in the infection. Now, even more evidence exculpating the accused has come to light — and that evidence is evolutionary.

HIV viruses evolve — just as bacteria, fruit flies, flowering plants, hominids, and other living things do. However, the timescale of evolutionary change in these different organisms varies widely. Because of their higher mutation rates and shorter generation times, viruses evolve at lightning speed compared to, for example, fruit flies. So while, a lineage of fruit flies might evolve into five distinct species over hundreds of thousands of years as it invades five tropical islands, an HIV virus can evolve into five distinct lineages in less than a year as it infects five different victims.

And whether we are thousands of years or mere months too late to see it in action, we rarely get the chance to observe evolutionary diversification. We can, however, reconstruct that history through genetic clues left behind. Over time, all lineages (especially viruses) accumulate mutations in their genomes — some beneficial, some slightly deleterious, and most with no discernable effect at all. Whatever their impact, those mutations record events in the lineage’s history, turning the genome into something of a marked up road map — a spotty, but informative, account of where the lineage has been at different points in its evolutionary past. And in the case of an HIV virus, that road map may reveal where the virus came from, how old it is, and potentially, how it entered its victim in the first place.

This month, just a week before the death penalty trial ended, biologists reported the results of their investigation into the evolutionary history of the HIV strain infecting the Libyan children. The biologists assembled the RNA sequences of the Libyan viruses, along with many other HIV strains, and used the mutations scattered throughout those sequences to reconstruct their evolutionary tree — or phylogeny. That tree revealed that the children’s viruses form a tight-knit family — as would be expected if they stemmed from a single introduction of HIV to the hospital, which was then passed to many different patients.

The biologists also used the amount of genetic change in the viruses to estimate when the different viral lineages split off from one another — and what they found convincingly showed that the medics could not have caused the infection cluster. The medics arrived at the hospital in March of 1998. If the medics had intentionally infected the children once they arrived, then the first victims would have been infected after the medics’ arrival — and hence, the victims’ viruses would have begun to diverge from other viral lineages after March of 1998 (as shown in Scenario 1 below). However, when the biologists reconstructed the family tree of the children’s viruses, that’s not what they found. Instead they discovered a phylogeny resembling that shown in Scenario 2 below. The children had been infected and their viruses had begun diverging several years before the medics even arrived on the scene!

The phylogeny not only verifies the innocence of the medics, but also points to a possible source of the virus. The phylogenetic analysis revealed that the children’s viruses are closely related to HIV strains common in West Africa. Many West African migrants come to Libya seeking work, and so it seems likely that the virus was accidentally introduced to the hospital when an infected worker or his or her child was treated there.

So the evolutionary history of the HIV outbreak, along with many other lines of evidence (e.g., indicators of unsafe medical practices at the hospital), point to the innocence of the accused doctor and nurses; the Libyan HIV cluster seems to have evolved from a West African strain that was introduced to the hospital and began spreading and diversifying long before the accused even began work there. Despite the strength of this evidence, the medical workers have been sentenced to death — but stay tuned! The defense lawyers have already announced plans to appeal, and many hold out hope that political negotiation will prompt Libyan judges to reconsider the evidence.