Where did HIV come from, and when? And does it matter?

In the view of Paul Sharp of Nottingham University, the available evidence from phylogenetic studies shows close agreement using different methodologies. Presentations yesterday by Betty Korber of Los Alamos National Laboratory in the USA and Anne-Mieke Vandamme of the Rega Institute in Belgium both dated the last common ancestor for all HIV group M sub-types at around 1930, and if this ancestor was a human immunodeficiency virus, this would be the very latest date at which transfer might have occurred. The Rega Institute group estimate the transfer could have occurred anywhere between 1590 and 1760, with 1675 the most likely date.

The technique used by both groups involved estimating the speed of molecular change within chosen sequences, by comparing a variety of sequences from different dates. The Rega group removed from their calculations all sites within a sequence which did not conform to the molecular clock, a dating approach already verified in individuals with known hepatitis C infection dates. They found that the more non-conforming sites they removed, the more accurate their dating. The Los Alamos group did not assume a stable rate of evolution, and used the earliest known HIV isolate (collected in Kinshasa in 1959) in comparison with other later isolates to verify their method. Looking at the V3/V5 regions of the HIV envelope from 197 sequences gathered in Zaire in April 1997, they looked at prior sequences from the same region to establish evolution rates.

If HIV was introduced to the human population before the Second World War, why didn't it cause recognisable AIDS cases before the 1970s?

Kevin de Cock pointed out in his Monday presentation that the small number of HIV case reports come predominantly from academic centres, and that unusual diseases were most likely to be noticed in Europeans with access to care at academic medical centres. Unusual cases of cryptococcal meningitis went un-noted until the late 1970s, although polio researcher Stanley Plotkin said that in his experience, "you don't recognise things unless you are prepared for them, especially if they are low incidence illnesses".

This pattern prior to the early 1980s suggests that HIV was causing a low incidence of AIDS. If it had been causing a high incidence of illness, argues tropical medicine specialist Alan Fleming, the illness would have been recognised by local people without the aid of epidemiologists. That is what happened in Uganda, where a new disease was named Slim by local people before any epidemiological investigation was conducted.

Three groups of HIV-1 exist today - group M, group O and group N. While group M is diverse, groups O and N are not, and remain highly restricted in one location. HIV-2 has one clear group with a separate path of descent, from the form of SIV found in sooty mangabeys in West Africa. This genetic tree suggests four separate transfers to humans, a puzzlingly rare chain of events given the potentially frequent contact with chimps in Central Africa, and the dating of chimp to human virus transfer back to the 18th century or even earlier.

In the opinion of Daniel Law-Beer, an epidemiologist from the University of Oxford, a star-like geography of infection would need to have been established early on in the epidemic, otherwise chance extinction events in one village could have eliminated the new virus from the human population quickly. He estimates that at least 60% of SIV transfers were dead-end infections of this sort which may have flared and died in remote locations.

Sir Robert May of Oxford University also suggested that it was quite plausible for many viral transfers to have remained confined in locations where local custom did not encourage mixing between populations in different villages.

By his calculations, if the virus spread equally within a village and outside a village, this would result in the extinction of the virus within a village, and a ripple effect which would result in lower and lower prevalence as the virus travelled from its original 'hearth'. The virus would establish a foothold in a new village only to die out because of low rates of partner change, and it would be transferred out to other villages if inhabitants took partners in other places, and so on.

This effect would result in an initial peak followed by a slow decline over several decades, before the virus moved into geographically concentrated populations with high rates of sexual partner change - the urban conditions of Kinshasa in the 1970s for example. Sir Robert May's group estimate the delay at between 30 and 40 years if the reproductive rate (the number of individuals infected by each case) was only slightly above 1.

If SIV crossed in several places and then began to spread slowly, where are the most likely places for this to have occurred?

A review of sub-type diversity by Martine Peeters found that the greatest diversity (indicative of the greatest duration of opportunities for recombination with other viruses) lies in the region of Kinshasa, and that Zaire generally shows the greatest degree of sub-type diversity of any region in Africa. But does this necessarily make Zaire the ancestral hearth of HIV, or just the crossroads at which a number of viruses were able to come together?

At the moment this question is still highly debatable. As sequence information accumulates and phylogenetic dating becomes more sophisticated it may become easier to answer, but at present any assertions on the location of first transfer are highly speculative.

Although the oral polio vaccine hypothesis may be fatally wounded in the view of Preston Marx of Tulane University Primate Centre, that doesn't exclude other iatrogenic possibilities. Marx presented his research into the potential role of unsterile injecting equipment not only in spreading HIV, but in turning it into an aggressive human immunodeficiency virus in the first place.

Marx began to be interested in this possibility after looking for human HIV-2 infection in West Africa that might be associated with sooty mangabey contact. These primates are the hosts of SIVsm, the ancestor of HIV-2, and Marx was surprised to find little evidence of HIV-2 infection. Only 2 individuals out of 9509 in Sierra Leone were HIV-2-positive, despite no difficulties in finding SIVsm-positive mangabeys kept as household pets. Might HIV-2 be a relatively non-pathogenic infection, also difficult to transmit onward due to extremely low viral titre? Cohort evidence certainly suggests this is so, and that HIV-2 can incubate for upwards of 40 years.

Had something occurred to amplify the virulence of HIV-1 after its transfer from chimpanzees? It is known that SIV becomes more aggressive if passaged through a new host, especially if the index case is experiencing primary infection. Might HIV-1 not do the same? And what would provide the most efficient means of passaging HIV through multiple hosts?

Marx points to the massive increase in the use of antibiotics and injections in Africa during the 1950s. Before the Second World War, syringe use was limited because they were expensive and made out glass. Plastic syringes were introduced in 1959, when their cost dropped up to one hundred fold, and the use of penicillin and choloroquine became more common in Africa. Re-use of syringes was so commonplace as to be unremarkable, and continues, and Marx points out that "if we are correct about this, ...we will continue to get new strains. It could wreck vaccine research."

Charles Gilks of Liverpool University School of Tropical Medicine points out that the extremely high prevalence of hepatitis C in Egypt is directly attributable to injectable treatments used to combat schistosomiasis in mass campaigns during the 1950s. He also points out that human/chimpanzee contact accelerated hugely in the 20th century driven by the imperatives of medical and scientific research. Chimps were used in malaria experiments because it was noted that they carried three out of four of the malaria parasite species which affected humans. Doctors are known to have injected themselves with blood from malarial chimps in the 1930s, and subsequent experiments included the injection of chimp blood into 20 individuals in Antwerp and prisoners in the USA. It was feared at the time that if malaria was eradicated by measures such as DDT spraying in the developing world, chimps might serve as a reservoir for renewed human exposure, in just the same way as a rat recently infected a small girl in the Netherlands with a pox virus that caused severe illness - in the very year smallpox is expected to be eradicated.

Animals will always provide a reservoir for viruses which could threaten human populations, no matter how conceited we become about our ability to manage the world of microbes and their genetic code. As organiser Robin Weiss pointed out, "There are lots of instances of transfer of other retroviruses from animals to humans". But humans too have the potential to wreak terrible havoc, and the Royal Society meeting was a welcome opportunity to review concerns about the potential role of medical science in causing or amplifying the HIV epidemic. "There is good veterinary and human evidence of iatrogenic transfer of retroviruses", Robin Weiss told the meeting.

Does it matter where HIV came from? "Unless we understand where HIV came from we run the risk of new emergencies, and unless we understand the ecology that allowed it to spread, we will be unable to control newly identified diseases", said Preston Marx on the opening day of the meeting, in response to criticims that the meeting was irrelevant to the needs of the developing world. Rarely has an HIV science meeting asked more relevant questions. As Albert Osterhaus noted in his scene setting presentation that opened the meeting, "HIV is the champion zoonosis of recent years".