Vaccine manufacturers need to be ever vigilant lest their products become unwittingly contaminated by microbes lurking in the source materials used to make vaccines1. It was with such a possibility in mind that in 1999 Edward Hooper2 promulgated a startling theory about how the human immunodeficiency virus type 1 (HIV-1) first came to infect humans. Hooper proposed that early batches of live oral polio vaccine (OPV), supposedly grown in chimpanzee kidney cell cultures at the Wistar Institute, Philadelphia, and tested in Africa, became contaminated with a chimpanzee virus that was able to infect its new host — with lethal consequences.

Three brief papers3,4,5, beginning on page 1045 of this issue, bring in a 'not guilty' verdict on OPV. Tests by Blancou et al.3 and Berry et al.4 on early OPV stocks prepared at the Wistar Institute in the 1950s show that they were propagated in the kidney cells of rhesus monkeys, and that they lack nucleic-acid sequences related to either HIV or chimpanzees. And Rambaut and colleagues' evolutionary analysis5 of HIV-1 subtypes from the Democratic Republic of Congo (Zaire) indicates that the common ancestor to HIV-1 group M, which gave rise to the pandemic strains, was present in a human host long before the first OPV field trials were conducted in the Congo during the late 1950s.

The idea that HIV might have been introduced to humans through contaminated polio vaccines has been around a long time. I first learned about it from an antivivisectionist tract circulated in 1987. The theory became notorious through Tom Curtis's article6 in Rolling Stone in 1992, which proposed that HIV came from a simian immunodeficiency virus of African green monkeys (SIVAGM). By 1963, production of OPV had switched from using cells from Asian macaques to those from African green monkeys, following the discovery of another virus — simian vacuolating virus 40 — as a contaminant in macaque tissue1,7. Twenty-five years later, many African green monkeys were found to harbour SIVAGM, necessitating screening for this virus or use of human cell substrates in vaccine preparation.

But SIVAGM is only distantly related to HIV-1, and cannot have been its precursor. The closest known animal virus to HIV-1 is SIVCPZ of chimpanzees; in fact, the three main groups of HIV-1, named M, N and O, differ from each other as much as they differ from SIVCPZ, suggesting that there were at least three separate instances of cross-species transmission8. So if the OPV hypothesis on the origin of HIV were to remain plausible, the use of chimpanzee kidneys for OPV propagation had to be invoked2. Yet there is no documentation that chimpanzees were ever used for this purpose and, along with a paper simultaneously published in Science9, these new reports3,4,5 dismiss what reasonable conjecture existed.

From the sensitive and specific forensic DNA tests based on the polymerase chain reaction, it is clear that the cellular substrate for the early OPV batches was macaque tissue. This has been determined by three independent laboratories3,9 in samples made available by the Wistar Institute, which included the OPV batch (designated CHAT 13) used in Léopoldville (Kinshasa) between 1957 and 1959.

The OPV batch that Hooper2 considered to be under most suspicion, however, was CHAT 10A-11, which was also tested in Belgium's African colonies in the late 1950s. This batch was not available from the Wistar, but an original vial of the batch was found at Britain's National Institute for Biological Standards and Control (Fig. 1), and the new tests show that it was prepared from rhesus-macaque cells4. Further tests showed that these batches contained type-1 poliovirus only, as originally recorded for CHAT vaccines. There is no reason to doubt their authenticity.

Figure 1: An original vial of CHAT 10A-11 oral polio vaccine (OPV).
figure 1

The vial was found at the National Institute for Biological Standards and Control, Potters Bar, UK. New tests show that the vaccine was prepared from the tissue of rhesus macaques, not of chimpanzees as required by the OPV-transmission hypothesis. (Photo courtesy of Harvey Holmes4.)

Full size image

Last year, Korber and colleagues10 extrapolated the timing of the origin of HIV-1 group M back to a single viral ancestor in 1931, give or take about 12 years for 95% confidence limits. Because this calendar of events obviously pre-dated the OPV trials, in the revised version of his book2 Hooper suggested that group M first began to diverge in chimpanzees, and that there were then several independent transfers of virus to humans via OPV. In that case, several OPV batches should bear evidence of their production in chimpanzee tissue, yet no such evidence has been found3,4,9. In any case, the data of Rambaut et al.5, and those described in a forthcoming paper by Sharp et al.11, indicate that Hooper's interpretation is vanishingly unlikely. Rambaut et al. re-analysed sequence data, reported last year12, on the complexity of Congolese HIV-1 group M isolates. The diversity of HIV-1 in the Congo appears to pre-date the evolutionary radiation or 'starburst' of the major virus subtypes that have since spread across the world. But the genetic sequences conform to a single origin, presumably 'patient zero', the first human to become infected.

Defenders of the OPV hypothesis hold that the alternative route of cross-species virus transfer, an accidental infection from a chimpanzee, seems equally unlikely. Most of today's human infections, however, have come from various animal sources within the past 10,000 years or so13, a split-second on the evolutionary timescale. So SIVCPZ may have infected humans on numerous occasions in human history without becoming firmly established in our species, rather like the Lassa fever and Ebola viruses. What may have helped HIV-1 group M to take off epidemically could have been the use of non-sterile needles and syringes in Africa in the mid-twentieth century14. This mode of transmission is currently behind the explosive spread of HIV-1 in parts of Russia and China. So one can envisage the involvement of an unintended medical factor in the adaptation of HIV to humans after it had crossed species from chimpanzees.

The new data3,4,5,9 may not convince the hardened conspiracy theorist who thinks that contamination of OPV by chimpanzee virus was subsequently and deliberately covered up. But those of us who were formerly willing to give some credence to the OPV hypothesis will now consider that the matter has been laid to rest. One may expect to hear an argument that chimpanzee kidney cells could have been used locally in Africa to amplify the batches of OPV prepared at the Wistar Institute in macaque cells, and that it was these vaccine samples that became contaminated. But the facilities for this type of cell culture did not exist in the Congo in 1957–59; besides, the evolutionary analysis of HIV-1 group M contradicts such a view.

The irony is that these new studies would almost certainly not have been undertaken if Hooper2 had not called for the analysis of DNA in stored OPV stocks, on the suggestion of the late Bill Hamilton. So we owe Hooper and Hamilton a debt of gratitude for pressing the case for those tests. When the preliminary results of the investigations were announced at the Royal Society in London last September, Hooper nonetheless dismissed them as “irrelevant to the OPV hypothesis”. But that simply isn't so — on the contrary, some beautiful facts have destroyed an ugly theory.