[Much new scientific evidence about the origins of AIDS has emerged during the last year. The following essay is based on one that was posted on Brian Martin’s web-page in February 2004. This new version of January 2005 is considerably enlarged and updated. E.H.]
Introduction: Some brief background to the origins controversy
Given that it presented as such a deadly and dramatically “new” disease when it was first recognised among gay men in the USA in 1981, it is little surprising that colourful rumours about the origins of the AIDS epidemic should have flourished from the start.
The debate about origins acquired more relevance and factual underpinning in 1987 with the appearance of the first essays written and posted to individual scientists by that extraordinary and enigmatic armchair detective, Louis Pascal. It gained further prominence with Tom Curtis’s well-researched “Origins of AIDS” article, published in Rolling Stone in 1992.
However, it has only been since the publication of The River in 1999 that the question of how AIDS began has become a topic of heated and ongoing controversy.
Essentially there are two major theories of origin that seek to explain how man acquired the pandemic AIDS virus (HIV-1) from the common chimpanzee (Pan troglodytes), the animal which is host to the immediate viral ancestor, SIVcpz (the SIV, or simian immunodeficiency virus, of the chimpanzee). These two theories are the bushmeat theory (which has various subsidiary versions) and the OPV theory.
The bushmeat (or “cut hunter”) theory proposes that HIV-1 crossed from a chimpanzee to a human in the course of the hunting, butchery and eating of chimpanzees as bushmeat. (There are several inherent problems with this scenario, most notably its failure to explain why the four documented outbreaks of AIDS occurred only in the twentieth century, and not before.)
The OPV theory proposes that HIV-1 crossed via an oral polio vaccine (OPV) called CHAT that was prepared in chimpanzee cells and administered to nearly one million “volunteers” in central Africa (the Belgian Congo and Ruanda-Urundi) in the late 1950s. The main problem here is that the leading scientists who prepared CHAT vaccine and who helped organise the African vaccination trials (Hilary Koprowski, Stanley Plotkin and Paul Osterrieth) deny that they used chimp cells to prepare the vaccine. (Koprowski and Plotkin were based at the Wistar Institute in Philadelphia, while Osterrieth headed the virology lab at the Laboratoire Medical de Stanleyville in the Belgian Congo, the headquarters of the African CHAT operation. All three men also conducted research at the nearby chimpanzee camp at Lindi, where up to 600 chimps were housed, nearly 400 of which were “used” during the polio research.)
However, the recent MFP/Galafilm television documentary, “The Origins of AIDS”, shows that the denials by the vaccine-makers are far from compelling. There are substantial gaps and inconsistencies in their accounts of what they did and did not do, and these accounts are directly contradicted by numerous other witnesses.
The scientific evidence for and against the OPV theory
Quite apart from the historical evidence about the usage and testing of CHAT vaccine in Africa there is the scientific evidence pro and anti the OPV theory. The line that has been taken by many prominent members of the medico-scientific community (including Koprowski, Plotkin and Osterrieth) is that the bushmeat theory has won the day, and that the OPV theory has been “destroyed” under the cool, clear light of the scientific microscope. However, a closer analysis of the available evidence might lead one to very different conclusions.
“The Origins of AIDS” film touches briefly, and relevantly, on two of the scientific issues that are relevant to this debate (the early epidemiology of AIDS as compared to the venues of CHAT vaccination; and the testing of old CHAT samples). These apart, however, it makes no serious attempt to examine the science.
I regret this. Indeed, I urged the film-makers on several occasions to tackle at least two of the other issues that lie at the core of the scientific debate: the genetic “age” of HIV-1, and the likely primate source of the human virus. But they decided not to do this, perhaps because they wanted to avoid complicating their film. This was understandable from a cinematographic point-of-view (and indeed, their judgement was vindicated, for “The Origins of AIDS” went on to win several prizes, and was short-listed for an International Emmy). However, it was regrettable from a scientific perspective. A real chance had been lost to move the debate forward.
I believe that the so-called “disproofs” or “refutations” of the OPV theory which have been put forward by certain influential scientists (including those mentioned above) and broadcast in journals such as Nature and Science are – without exception – based on inaccurate information and biased analysis. Poor science, in short.
In reality, as becomes increasingly clear with each new piece of information that comes to light, the latest scientific evidence on the origins of AIDS offers very strong support to the OPV theory.
The alleged “disproofs” of the OPV theory
Professor Robin Weiss is said by many to be Britain’s leading retrovirologist, and it was he who, together with another respected retrovirologist, Simon Wain-Hobson, organised the Royal Society meeting on “The Origins of HIV and the AIDS epidemic” in September 2000, a meeting that was convened at unusually short notice in order to respond to the information and analysis presented in The River. A third organiser, the celebrated evolutionary biologist Bill Hamilton (who was also the leading scientific supporter of the OPV theory), died unexpectedly in March 2000, leaving Weiss and Wain-Hobson to stage the meeting on their own.
There was a storm of controversy about whether the meeting should go ahead at all: two American scientists (Beatrice Hahn and Bette Korber) declined their invitations to speak, and Plotkin and Koprowski hinted they might do the same.
In the end, the surviving organisers saved the meeting. They rejigged it in a way that made it acceptable to the dissenting scientists, who came back on board. They postponed it from May to September 2000, allegedly to allow more tests to be completed. And they decided that Professor Weiss would deliver the “Concluding Remarks”.
The closing speech is traditionally the moment at which a senior scientist, supposedly a neutral arbiter, attempts to pull together all the evidence that has been presented at a scientific meeting, and to arrive at some balanced conclusions. Weiss, however, gave an extremely one-sided analysis. He even admitted that his opinions were just his “plain personal prejudice”, although these words did not appear in the final published version of the speech. What did appear in the published version was the following judgement: “I am not saying that the OPV hypothesis is conclusively disproved; but there is such a burden of doubt that the arguments needed to sustain the hypothesis become increasingly distorted.”1
In interviews given after the meeting, Weiss was less circumspect. He implied that the theory had been fatally wounded, and perhaps not unexpectedly, much of the lay and scientific press followed his lead.
Seven months after the Royal Society meeting, Nature and Science published simultaneously on what, it was claimed, was new evidence indicating that the OPV theory had been “disproved”. This “new evidence” consisted of test results from a sealed phial of CHAT vaccine that had originated from the Wistar Institute in 1958, and an allegedly new argument that was, however, again based on the principles of phylogenetic dating [see (vi), below]. In reality, this was nothing more than a repackaging of arguments already advanced at the Royal Society meeting.
Despite this, there was blanket coverage in the news pages and on the web-sites of the world’s two leading scientific journals, and its tone was valedictory. There were commentaries with titles such as: “Polio researcher innocent of HIV pandemic”, and “Polio vaccines exonerated” (as if the safety of all polio vaccines had been questioned, rather than that of just one experimental OPV used in Africa in the 50s). Robin Weiss had written the commentary in Nature, and he claimed that the newly-discovered phial of CHAT came from the same batch that had been used in Koprowski’s campaigns in Africa, and concluded that because it had proved negative for HIV, SIV and chimpanzee DNA, this settled the debate. He ended his commentary with the memorable line: “Some beautiful facts have destroyed an ugly theory”.2
Science headed its commentary “Disputed AIDS theory dies its final death”, and its author, Jon Cohen, was as dismissive as Robin Weiss.3 “The theory is not only merely dead, it’s really most sincerely dead” was his jocular verdict, apparently inspired by the Munchkins from The Wizard of Oz.
What, however, was noticeable to many OPV proponents was that neither Nature nor Science had ever devoted any space to discussing the theory itself. In fact, during the preceding 13 years, both journals had repeatedly rejected submissions about the theory, even when they were written by scientists as eminent as Bill Hamilton. During that period, however, they had devoted a steadily increasing amount of coverage to alleged disproofs of the theory.
In this particular instance, the two journals had clearly co-ordinated their responses in order to present a united front. It seems that their intention was to bury the inconvenient theory forever (and perhaps, at the same time, to “rid [themselves] of this turbulent priest”, in other words to shut me up, at least in figurative terms). Even if they did not succeed in either of these aims, their blanket rejection of the theory did have an enormous impact. As I write this (at the start of 2005) most neutral scientists and lay persons seem still to believe that the debate is settled, and that the OPV theory has been disproved.
But on what grounds have the OPV sceptics reached their conclusions? At various times, they have come up with seven different arguments against the OPV theory:
- that a sample from the very same version of CHAT vaccine that was given in Africa has been tested, and found free of HIV, SIV, or chimpanzee DNA;
- that local chimps (for instance the chimps at Lindi camp) were not infected with SIV;
- that the Lindi chimps were not infected with “the right SIV”;
- that chimpanzee tissues were never used to make the Koprowski vaccine;
- that even if contaminated chimp kidneys had been used, an SIV would not have survived through to the final vaccine;
- that phylogenetic dating analysis demonstrates that HIV-1 predated the OPV trials; and
- that there is anyway no meaningful correlation between the Koprowski vaccinations in Africa and the first appearances of HIV-1 and AIDS.
I shall respond to each of these alleged disproofs in turn.
As I shall demonstrate, not one of these arguments offers anything approaching a disproof or refutation of the OPV theory. Some of them can be readily shown to be incorrect. The remainder are inherently flawed, usually because the claims they make are exaggerated, or are not supported by the available data.
“Disproof” 1: The claim that a sample from the very same version of CHAT vaccine that was given in Africa has been tested, and found free of HIV, SIV, and chimpanzee DNA
The sealed phial of vaccine that was sent by Hilary Koprowski (by then director of the Wistar Institute in Philadelphia) to Stockholm in 1958, and from Stockholm to London in the 1980s (and which was finally opened and tested for HIV, SIV and chimpanzee DNA in 2000), did come from the same vaccine pool, CHAT 10A-11, that had been administered in the Ruzizi Valley, on the borders of the Belgian Congo and Ruanda-Urundi in Africa.
However, the pool numbers are completely irrelevant to this argument. What I repeatedly emphasised in The River, and at the Royal Society conference, is that different labs prepared different batches of Koprowski’s vaccine pools at different times, and in the cells of different primate species.
Some brief background is needed. A “pool” and a “batch” of oral polio vaccine are two very different things. Essentially, OPV consists of living poliovirus that has been attenuated, or weakened, to the point where it is no longer dangerous for humans, but can still immunise a human subject against poliomyelitis. The pool number (eg 10A-11) of an OPV such as CHAT identifies vaccine that has been attenuated to a certain specified level, and then frozen. Samples from this stored vaccine pool may then be used as seed material to make individual batches of vaccine.
A batch of OPV means vaccine that has been prepared in a single production run. Thus every sample from a vaccine batch contains homogeneous (or identical) material. However, the same cannot be said about every sample from a vaccine pool. It all depends which batch of that pool is being analysed.
Nowadays, the OPV-making process would begin by preparing a seed virus (a pool of vaccine which is tested and characterised, and from which all future pools are then prepared). This is the system that Albert Sabin (who developed the sugar lump OPV that has been given to the majority of the world’s population) used, even back in the mid-1950s. However, as Stanley Plotkin has admitted, “No seed system was used” for CHAT, which means that each pool was prepared in sequence from a sample of the previous pool. Thus any contaminating virus that was present in one pool would also be present in all subsequent pools.
When no seed system is used, it is equally possible to prepare new batches of vaccine from samples of previous batches. The technique for growing a new pool or a new batch is exactly the same. The virologist takes a small sample of the vaccine and grows it in a good substrate, which for polio vaccine means a tissue culture (a sheet of cells prepared in the laboratory) made from the kidneys of a primate: a monkey or ape.
This is very similar to putting a teaspoon of yogurt into a substrate of warm milk, in order to produce a fresh batch of yogurt. With yogurt, the process takes 12 hours or so. With poliovirus, it takes two to three days.
When I interviewed Hilary Koprowski in 1993, he repeatedly emphasised the fact that OPV is “living material”, and that there is nothing to prevent a scientist who has received a phial of OPV from preparing further batches of vaccine from that phial. (This, of course, could be done either with or without the donor’s permission.)
What all this means is that it is crucial to identify and test the vaccine batches that were given to people in Africa, rather than the vaccine pools.
It is now apparent that most of the CHAT vaccine that was administered to “volunteers” in Africa consisted of batches that were locally prepared in Stanleyville. So although pool 10A-11 of CHAT was used in the Ruzizi Valley, this was not the same vaccine material as that in the phial sent to Stockholm, which later proved to be free of HIV, SIV and chimpanzee DNA.
More evidence about the local preparation of CHAT is coming in all the time, and because the vaccine-makers and their agents are making such a determined effort to persuade witnesses to “modify” the key aspects of their testimony, this is not the right moment to detail who has said what (although some of this information is already in the public domain). But I can summarise as follows. To date, I and/or others have interviewed seven people who say that the vaccine used in Africa (or at least some of it) was prepared locally, in Stanleyville. Moreover, I have interviewed five people who state that this vaccine was being made in chimpanzee cells, and three people who state that it was specifically made in the cells of the chimps from Lindi camp. Three of the aforementioned persons are primary witnesses, who were directly involved with events at the time, and the rest are secondary witnesses. There is a great deal of additional testimonial and documentary evidence to support these claims.
Also noteworthy is the fact that the accounts of events provided by many of the vaccinators, but especially by the head of virology at the Laboratoire Medical de Stanleyville, Paul Osterrieth, have fundamentally changed on several key issues. The most striking example of this relates to the speech made by Dr Osterrieth at the Royal Society meeting in 2000, when he stated unequivocally that “The polio vaccine sent from Philadelphia was never conditioned or handled in my laboratory, to which only I had the key”. This claim has been contradicted by a wide range of other witnesses (including, interestingly, Koprowski himself), all of whom say that it was Osterrieth who was in charge of (and handled) the polio vaccine at the LMS.
But now, buried deep in his latest account of events (published in “Vaccine” [2004; 22; 1831-1835], a journal which has Stanley Plotkin as one of its editors), Dr Osterrieth concedes that “dilution of the concentrated stock [of the vaccine] was necessary before administration, and I probably did do some of that work, to help Dr Courtois and Dr Ninane.”
Dr Osterrieth’s backtracking is crucial – albeit unsurprising. Three years ago, I forecast that the next fall-back position of the vaccine-makers would be to claim that all they were doing was diluting the vaccine that came from abroad.
Yet this is far from a persuasive explanation. Several contemporary articles and archives (including those by the head of the Stanleyville lab, Ghislain Courtois) describe the dilution of the Koprowski vaccines in Africa, in a measured amount of saline solution, as a process that was conducted in the field at the start of each day’s vaccinations. Even on those rare occasions when dilution might have been conducted in the lab (as could, for instance, have happened when the vaccination was to be in Stanleyville itself), this was work that required only a few seconds, provided the glassware was sterile. (I have an archival film clip which shows a small quantity of oral polio vaccine being pipetted from a phial, and then transferred to a larger flask of what is presumably saline solution. The whole dilution process is recorded, and it takes about fifteen seconds.) This brief process of “dilution” bears little resemblance to the accounts given in April 2001 by Osterrieth’s assistant, Jacques Kanyama, who portrayed the vaccine-making as a lengthy procedure which Osterrieth conducted alone, and often after-hours, in his laboratory.
Dr Osterrieth’s most recent version of events is a belated attempt to rescue himself from an untenable position. But taken together, his two accounts show that while professing frankness, he is in fact repeatedly evasive, and quite unable to give a frank account of what he was actually doing with the vaccine. One can only conclude that this is because if he did so, he would incriminate himself. (In much the same way, Dr Osterrieth portrays himself as having been only peripherally involved with the procedures at Lindi. By contrast, the African workers at the Camp recall that he was the doctor who was most frequently involved with the sacrificing of chimps, and the obtaining of blood and organs.)
So it is the individual vaccine batches that were used to vaccinate in Africa that need to be tested. However, in The Origins of AIDS film, Robin Weiss states that these batches of African vaccine don’t exist any more. He may or may not be right. I personally suspect that they probably do still exist in certain freezers, but that the institutions holding them would not be keen to release them for independent testing.
As the commentary in The Origins of AIDS rather neatly pointed out, the newly-discovered “Stockholm phial” of vaccine had never been opened between the time it left the Wistar Institute in 1958 and the time of its testing in 2000. Furthermore, it had never been anywhere near Africa. The same applies to the six other phials of CHAT that were tested, all of which were samples obtained from the Wistar Institute freezers. None of these phials had been near Africa, and none of them therefore is relevant to the OPV theory, or can provide a disproof of the theory.
Robin Weiss’s claim in Nature that the Stockholm vaccine sample was from the very same batch that had been used in Africa is incorrect, and his claim that the OPV theory has been “destroyed” is therefore based on false information. One has to ask why such an eminent virologist (with decades of experience dealing with pools and batches of vaccine) would make such an elementary blunder, by confusing the two terms.
To sum up, the most frequently-quoted “disproof” of the OPV theory is actually not a disproof at all. There is substantial evidence that most of the CHAT vaccine used in Africa was locally made, and no independent scientist has had the opportunity to test this locally-made vaccine for HIV, SIV and chimpanzee DNA.
“Disproof” 2: the claim that local chimps (for instance the chimps at Lindi camp) were not infected with SIV
First of all, some background is necessary. Simon Wain-Hobson, who in 1999 agreed to collaborate with Bill Hamilton and myself by testing samples of chimp faeces and urine that Bill was to collect on his two trips to Kisangani (formerly Stanleyville) in the Democratic Republic of Congo (DRC), has never officially reported the details or results of his SIV testing of these samples.
At the Royal Society meeting, Wain-Hobson uttered just one sentence from the floor about the Hamilton chimp samples, which came in response to an apparently pre-planned question from Koprowski’s former right-hand man, Stanley Plotkin. As I recall, Wain-Hobson stated that he had found no evidence of SIV infection in the samples, but he provided no details of how many samples (if any) he had tested, or of how he had tested them.
Since Bill Hamilton’s death in March 2000, I have sent Wain-Hobson five e-mails requesting feedback on the testing he had carried out on these samples. In the later e-mails, I asked him whether (if he no longer wished to collaborate) he would release the samples so that others could do the work.
He made no response to the first four requests. In the fifth e-mail, I requested a prompt reply, or at the very least information about when I could expect to hear from him. This time he sent a response, which was: “You know about blue moons? So there we are.” It can be safely assumed that this semi-cryptic answer meant “Never”.
Similarly, in either late 2000 or early 2001, Beatrice Hahn was given aliquots of some of Hamilton’s chimp samples by Michael Worobey (who had accompanied Hamilton on his second trip to the DRC) and before the end of 2001 (as I later discovered) her lab was repeatedly finding protein bands indicative of SIV in one of the urine samples.4 These findings suggested that chimps from the heart of the DRC rain forest, from an area to the north of the road between Kisangani and Wanie-Rukula (which is close to where many of the Lindi chimps were collected in the 1950s), are infected with SIV. Yet nothing was reported in the medical literature.
The situation only changed in April 2004, because by that time Worobey had gone back to Africa to collect further samples of chimpanzee faeces from a forest some 130 kilometres from Kisangani, to the east of Wanie-Rukula. One of these faecal samples turned out to contain SIV, and Worobey and Hahn obtained its genetic sequence. They published a brief communication in Nature which reported the new sequence, and which included a brief mention of the original SIV-positive urine samples.5 However, the findings were presented in a heavily biased manner, with the commentary claiming, incorrectly, that the new data “refuted” the OPV theory. [As explained in the next section, there is nothing in their data to support this conclusion.]
The most impressive study of SIV infection among wild chimpanzees is that published by Santiago (with Hahn and Sharp as co-authors) in 2003, which reported that 13% of a single wild troupe of this same chimpanzee sub-species (Pan troglodytes schweinfurthii) in Gombe, Tanzania, were SIV-infected.6 If that percentage applied to the 400-odd chimps used during the Lindi polio research, then approximately 50 would have been naturally SIV-infected before arriving at the camp. Moreover, the co-caging and group-caging that was routine at Lindi would have encouraged onward spread of SIV from chimp to chimp.
In summary, the claim that the Lindi chimps were not SIV-infected is almost certainly wrong.
“Disproof” 3: that the Lindi chimps were not infected with “the right SIV”
Beatrice Hahn and Paul Sharp (and more recently, Michael Worobey) have also argued that the Lindi chimps are of the “wrong subspecies”, claiming that the closest relative and true ancestor of the pandemic AIDS virus (HIV-1 Group M) is the simian immunodeficiency virus (SIV) of Pan troglodytes troglodytes chimpanzees originating from Cameroon and Gabon in west central Africa.
As far as the available data goes, this conclusion is supportable. But the claim is far from being proven, and several important riders need to be added as regards its relevance to the origins debate.
For one thing, primatologists such as Pascal Gagneux have questioned whether Pan troglodytes troglodytes and Pan troglodytes schweinfurthii (the chimpanzee found in the DRC, the former Belgian Congo) should be defined as two separate subspecies, or whether they should be redesignated as belonging to a single subspecies.7
SIVs are found naturally only in African primates, and to date SIVs have been identified among some thirty different African primate species. The P. t. troglodytes and P. t. schweinfurthii SIVs are the only two SIVs that are closely related to the AIDS pandemic virus, HIV-1(M).
To date, some nine chimpanzee SIV isolates have been sequenced or part-sequenced, and there is wide divergence between the different sequences. Hahn and colleagues are correct when they state that up to now the Pan troglodytes troglodytes SIVs from Cameroon and Gabon seem typically to bear about 80% homology with HIV-1 Group M, whereas the Pan troglodytes schweinfurthii SIVs from DRC and Tanzania seem typically to bear only about 70% homology. 8 (The precise degree of homology is far less precise than that, and varies from gene to gene.)
However, where Hahn and Worobey are not correct is when they claim that their findings “disprove” or “refute” the OPV theory. This is not a simple argument, but there are at least five reasons why their claims are wrong:
- The only chimp SIV that Hahn and Worobey have personally sequenced from the DRC came from an animal that originated from the Parisi Forest, some 80 miles south-east of Kisangani. (The one other reported chimp SIV sequence from the DRC came from Noah, an animal that was detected by officials at Antwerp airport, and whose precise geographical origin within the DRC remains unknown.) In their letter reporting this data, Hahn and Worobey write that “The ‘OPV/AIDS theory’ claims that chimpanzees from the vicinity of Stanleyville – now Kisangani in the Democratic Republic of Congo [DRC] – were the source of a simian immunodeficiency virus (SIVcpz) that was transmitted to humans when chimpanzee tissues were allegedly used in the preparation of OPV.” They go on to imply that their Parisi chimp came from an area where the Lindi chimps were collected, and to assert that because it was infected with a typical schweinfurthii-type SIV sequence, this proves that the Lindi chimps were infected with the “wrong” virus, and therefore refutes the OPV theory. There are many flaws in this claim. Firstly, the records that exist do not indicate, or even suggest, that any chimps from the Parisi area were ever brought to Lindi.
- More importantly, however, Hahn and Worobey’s assertion that the OPV theory claims that chimps from around Stanleyville were the source of the AIDS pandemic is, quite simply, false. In fact, as I have reported several times, the Lindi chimps were collected from a huge swathe of rain forest covering some 300,000 square miles – from Zapai in the north to Wanie Rukula in the south, and from Mambasa in the east to Mbandaka in the west. If Hahn and Worobey want to claim that they have sampled chimp SIVs from the areas that supplied chimps to Lindi, they would need first to collect chimp samples from right across the DRC rain forest and the savannah belt to the north. (And even then, as Pascal Gagneux has pointed out, it could be that the chimp group which provided the SIV strain immediately ancestral to HIV-1(M) has since died out.)
- Moreover, there is still much to learn about the schweinfurthii chimps, for they are not all alike. I have more than two dozen photos of the Lindi chimps that were taken by the French hunter Gilbert Rollais, who collected the chimps for Lindi camp in the 1950s. Apart from the photos of pygmy chimps or bonobos (Pan paniscus), the great majority of these animals have the classic appearance of Pan troglodytes schweinfurthii, with dense, dark fur that covers the brow. However, two of these Lindi chimps are clearly different: they both have the pale face and early balding that is typical of Pan troglodytes troglodytes. [See Jonathan Kingdon, The Kingdon Field Guide to African Mammals, page 10.] The caption provided by Rollais to one of these two photos indicates that the animal was captured in the north of the Belgian Congo, almost certainly from Ango or Bondo territoires, to the north of the River Uele. (Chimps from this area have long been rumoured to be “genetically interesting”, and they may turn out to be more closely related to Pan troglodytes troglodytes than previously thought. About 60 to 80 of the Lindi chimps are believed to have come from this area.)
- The second balding chimp in the Lindi photos is not sourced geographically, and the animal may therefore have been an “official” Pan troglodytes troglodytes originating from the western part of central Africa (Congo Brazzaville, Gabon, Equatorial Guinea or Cameroon). There is much to support this claim. For instance, one of the 54 Lindi chimps for which there are records is documented as coming from Coquilhatville (now Mbandaka) territoire, which lies across the River Congo from the ranges of both schweinfurthii and troglodytes. This animal was one of the first to be brought to Lindi camp in early 1957, and it survived there for at least two years, during which time it must have been co-caged and group-caged, just like the other Lindi apes. Clearly, therefore, this one chimp alone could have introduced a troglodytes-like SIV to other Lindi chimpanzees. Furthermore, I have interviewed someone who lived in Stanleyville throughout the 1950s, and another who visited there (Kisangani) twice in the 1980s, having travelled there on one of the Congo river barges from Kinshasa. Both witnesses reported that chimps were regularly brought upriver on these river barges for sale in Stanleyville/Kisangani. Some of these animals originated from Coquilhatville (now Mbandaka), and some from even further west; they therefore in all likelihood included some Pan troglodytes troglodytes chimps. The 1950s witness adds that some of these “imported” chimps are likely to have been brought to Stanleyville specifically because the word had gone out that the doctors there were purchasing chimpanzees. Further information on this subject is still coming in. But to summarise, there is now substantial evidence indicating that at least some of the chimps at Lindi were Pan troglodytes troglodytes from the west of the continent.
- However, given the evidence that has emerged, especially during the last four years, about the propensity of SIVs to recombine [see (vi), below], the geographical sources of the Lindi chimps may finally prove to be a moot point. HIV-1 shares the same set of genes with both of the chimp SIVs (troglodytes and schweinfurthii), but not with any of the other African primate SIVs that have been discovered. But there is still at least a 20% genetic “gap” between HIV-1(M) and even the most HIV-like of these chimp SIVs. At present a recombination event between two or more chimp SIVs (whether troglodytes or schweinfurthii, or a mixture of the two) in tissue culture looks to be the most parsimonious explanation for bridging that gap between the chimp and human viruses. (This, of course, is exactly what I believe happened in the chimpanzee tissue cultures that are reported as having been used to prepare CHAT vaccine in Stanleyville.)
In summary, the “wrong sub-species” claim is unproven, and in any case is extremely unlikely to represent a refutation of the OPV theory, not least because Pan troglodytes troglodytes chimps appear to have also been present at Lindi camp.
“Disproof” 4: that chimpanzee tissues were never used to make the vaccine
The allegedly crucial polio trials conducted in the Lindi chimps were extremely poorly reported, with only brief (and seemingly deliberately vague) details being provided in the annual reports of the Laboratoire Medical de Stanleyville (LMS) and in scattered articles and published conference proceedings down the years. But these scattered sources do concur on one thing: that the official reason for Lindi camp was (a) to test the efficacy of the Koprowski vaccines in chimps through vaccination and subsequent challenge with virulent poliovirus; and (b) to test the safety of the Koprowski vaccines in chimps by intraspinal inoculation.
Recently, further details have come to light which demonstrate the full extent of the official polio research that was staged at Lindi. The results are very similar to those that I tentatively proposed in The River (pages 708-714). Just over 360 chimpanzees were “used” in the polio trials. Even if one allows for a natural death rate in captivity of 25% (90 chimps), and subtracts the 60-odd chimps that were sacrificed during the official experiments, this still leaves some 210 chimps unaccounted for. [Further details will be published presently.]
It is now apparent that neither of the official experiments conducted on the Lindi chimps as part of Koprowski’s polio trials produced scientific results of any value. This is why no proper article about the polio trials at Lindi was ever published (though on several occasions such articles were described as being “in preparation”).
In reality, the Lindi chimpanzees were also used for other purposes related to the polio vaccine, as I discovered during a second visit to Kisangani in April 2001, and in further research conducted since.
During my 2001 visit to Kisangani, I conducted further interviews with the surviving Lindi “caretakers”, including Joseph Limbaya Mwenge, the camp nurse, who was in charge of the other Africans working at the camp. These witnesses confirmed that with the exception of four or five favourites (the first chimps to be brought to the camp, which were treated almost as pets), and the sixty-odd chimps that were transferred to a new station sited immediately behind the LMS when Lindi was closed down in January 1960, all the other Lindi chimpanzees either died or were sacrificed. (The grand total of chimpanzees that passed through Lindi between 1956 and 1960 was between 500 and 600.)
Crucially, some of these witnesses report that blood and organs were being obtained from anaesthetised chimps just before they were sacrificed. Such organ removals from living chimpanzees bear no relation to the “official” polio efficacy and safety experiments. Furthermore, they are entirely different from the liver biopsies that Stanley Plotkin has disingenuously insisted the African workers must have been describing. (These biopsies were carried out as part of the hepatitis research conducted at Lindi in 1958-9). The reality (as Dr Plotkin knows) is that it is normal practice to remove whole organs (especially kidneys) from living, anaesthetised animals in order to prepare tissue culture, for instance for vaccine cultivation.
I and others also interviewed several Africans who had worked at the Laboratoire Medical de Stanleyville (LMS) during the 1950s. These witnesses reported that tissue culture had indeed been prepared at the LMS, that it had been prepared “mainly from chimpanzees”, and that moreover the head of the virology department, Paul Osterrieth, had been “making the polio vaccines” in his lab.
Since returning from Africa, I have obtained confirmation of the key elements of these testimonies from various Belgian sources, including one eminent doctor who stated that the purpose of the Lindi chimpanzees was, quite simply, for “the preparation of the vaccine”.
This unique aspect of the Congo CHAT trials (making fresh batches of vaccine in chimpanzee cells, and doing this locally, in Stanleyville) is the key detail which is missing from The River, albeit largely because of the denials of the major protagonists: the Belgian and American vaccine-makers. Over the last three years, every stage of the local preparation process has been multiply confirmed by different sources. 9 The vaccine-makers continue to issue strenuous denials, but these are often self-contradictory, and their attempts to explain away the counter-evidence are increasingly implausible.
Further research has revealed that it was routine practice in the 1950s to propagate OPV locally (to give live polio vaccines one final “passage” in locally available cells) in labs close to where vaccination trials were being staged. For one thing, it was a way of making sure that the attenuated poliovirus was still alive after its long journey by air. This procedure happened with the vaccines of Sabin and Lepine, as well as Koprowski, and in places as far apart as Switzerland, the USSR and South Africa, as well as the Belgian Congo.
On the basis of the accumulated evidence, some of which has only emerged recently and which I shall report in due course, I can state with confidence that new batches of vaccine were being made at the LMS by cultivating vaccine material from the Wistar in locally-prepared chimpanzee tissue culture, thus boosting both the titre (concentration) and quantity of the vaccine. 10
In summary, the claim that chimpanzee tissue culture was never used to propagate the Koprowski vaccines in the Belgian Congo is untrue.
“Disproof” 5: that even if contaminated chimp kidneys had been used, an SIV would not have survived through to the final vaccine
At the Royal Society conference, a short paper by doctors Beale and Horaud proposed that even if SIV-contaminated chimp kidneys had been used to prepare CHAT vaccine, the chances of SIV virions surviving through to the final vaccine would be trillions-to-one against. 11 This back-of-the-envelope calculation had been arrived at by multiplying together six different sets of odds, which represented the doctors’ estimations of the chances against SIV surviving six individual procedures (such as the addition of the enzyme, trypsin) which would presumably have been involved if professional vaccine preparation techniques had been employed from first to last.
Because there are no surviving protocols or records revealing how the vaccine was handled in Stanleyville (or indeed how any of the early pools of CHAT were made), we have no idea whether any of the foregoing procedures apply in this case.
However, there is (as previously mentioned) strong evidence that the vaccine was passaged further in chimp cells in Stanleyville. Moreover, there is evidence that pooled chimpanzee serum was used as growth medium (rather than the more usual medium of foetal calf serum). The available evidence suggests that the Stanleyville vaccine was prepared in a rough-and-ready fashion using materials that were to hand, meaning that five of the six factors mooted by Beale and Horaud would not have applied. The one factor that still might well be relevant (washing the kidney pieces) would lead merely to a ten-fold loss of SIV, according to the doctors’ estimates. This is very different from a trillion-fold SIV loss.
The weakness of the doctors’ argument has recently been highlighted by an SIV expert, previously no friend of the OPV theory, who read my paper, “Dephlogistication”, and sent an unsolicited e-mail of encouragement, as follows: “An exploded kidney contains macrophages. HIV and SIV reside in macrophages, can be produced by macrophages and can replicate in macrophages….[W]hether you use trypsin or not, there will be macrophages in there that could be the source of SIV production. The plausibility of whether or not OPV could have been the source [of AIDS] boils down to whether chimps were used on site for production of [vaccine] lots used in the human testing.” Later, he added: “And if chimpanzee serum was indeed used for the production of OPV lots used in the human trials, this is an unambiguous source of SIV/HIV, since serum from a positive animal will contain about 1,000 to 1,000,000 SIV/HIV particles per ml.” [See below for evidence of chimp serum usage. In this instance, the term “lot” can be equated with “batch”.]
In summary, the claim that SIV would not survive through to the final vaccine is highly conjectural, and not one piece of evidence has come to light to support any of the conjecture. By contrast, a leading retrovirologist has revealed that if infected chimps were used to make the vaccine, then it is plausible to suppose that SIV would have made it through to the final preparation that was given to humans.
“Disproof” 6: that phylogenetic dating analysis demonstrates that HIV-1 predated the OPV trials
Geneticists such as Paul Sharp, Bette Korber, Eddie Holmes, Ann-Mieke Vandamme and Michael Worobey have come up with a different “disproof” of the OPV theory. These scientists all favour the concept of “phylogenetic dating” for HIV-1, and argue that the most recent common ancestor (MRCA) of all the AIDS viruses seen today must have existed in or around 1931, plus or minus roughly 15 years. The dates postulated by the different scientists vary by a few years here or there, but all insist that HIV must have existed before the OPV trials in the 1950s.
Phylogenetic dating (calculating the age of a virus by assuming it has a constant mutation rate) is an impressively accurate technique when applied to DNA viruses (conventional viruses, such as smallpox or yellow fever), but it is an entirely inappropriate technique for retroviruses such as HIV-1 and chimpanzee SIV. This is because retroviruses evolve predominantly through recombination, rather than through mutation (point substitution).
To put it another way, the calculations of the phylogenetic daters are based on a constant molecular clock, which has no way of allowing for recombination. Yet recent studies report evidence of “massive” levels of recombination in HIV-1, and indicate that the intrinsic recombination rate of this virus is some ten times greater than its point-substitution rate.
Such studies suggest that for retroviruses, and the HIVs in particular, phylogenetic dating is an attempt to measure the “wrong” evolutionary process. In any case, these studies also show that ignoring recombination (as the phylogenetic daters effectively do) would lead one to place the MRCA too far back in time. 12
The Danish geneticist, Mikkel Schierup, has demonstrated that even datasets that phylogeneticists such as Bette Korber claim to have weeded clean of recombinant sequences are in reality still full of recombinants. Schierup has highlighted the evidence not only for substantial recombination in HIV-1, but also (crucially) for substantial early recombination, before the virus diversified into subtypes. Such early recombination would render the concept of phylogenetic dating for HIV-1 even less valid. 13
The evidence provided by Gerry Myers (former head of the HIV/AIDS Sequence Database in Los Alamos) is also significant here. He too found that there was something strange about HIV’s phylogenetic tree: that with one exception (subtype B), the various HIV-1 subtypes were almost equidistant from each other, suggesting that they had branched from the ancestral virus at roughly the same time. This effect, which Myers called a “starburst”, suggested to him that “something extraordinary [had] happened”,14 and that an unnatural, “punctuated event” might have been involved in the birth of HIV.15 In other words, Myers found evidence suggesting that multiple and near-simultaneous transfers of SIV to humans had occurred, as distinct from the single chimp-to-human event that the phylogenetic daters insist was the source of the pandemic.
Because of poor health, Professor Myers was unable to travel to the Royal Society meeting to deliver his paper about the punctuated event; it was instead read by Tom Burr, who had provided the statistical analysis. However, Myers did make it clear to me before the meeting that he considered that the 28 known African trials of CHAT in the late fifties provided the most plausible explanation for the starburst phylogeny. This analysis would suggest that several different variants of SIV transferred to humans at roughly the same time, which is of course exactly what the OPV theory postulates. To both Dr Myers and myself, this seems like the most plausible source of the different HIV-1(M) subtypes.
It is important to bear in mind that the chimpanzee tissue explants that are documented as having been prepared at the LMS in 1958 also contained “isologous serum” (serum from other chimpanzees) as a nutrient medium.16 This suggests that these chimp tissue explants (which were actually primitive chimp tissue cultures) were effectively pooled, which further increases the likelihood of SIV recombination in vitro. Here, surely, is the explanation of the recombination that lies at the very beginnings of HIV-1(M).
Interestingly, Donald Burke, the former head of the US military’s AIDS research programme, also felt that recombination between different strains must have occurred early in the AIDS epidemic in Africa, producing strains that were viable enough to spread. Burke postulated that this must have happened around the time of urbanisation in Africa, which allowed the sexual coming together of groups that had previously lived apart.17
This is fascinating, because Burke’s scenario again points to the middle of the twentieth century as the time when the subtypes formed. What Beatrice Hahn and her supporters therefore need to explain is why (and indeed how) significant recombination would have occurred in humans at that very early stage of the epidemic, when only a very few individuals were infected.
Hahn tends to brush aside such questions, arguing that it is not her job to propose a scenario to explain her theory. I think this is nonsense, and that her refusal to “get specific” is based on the fact that if she did so, the flaws and shortcomings in her analysis would be revealed. Let me pose just two of the questions that are begged by her theory. (a) Why did a hunter or bushmeat seller from the Pan troglodytes troglodytes area (Congo Brazzaville, Gabon, Equatorial Guinea and Cameroon) get infected with a troglodytes SIV, and not transmit this virus to anyone locally, but only to someone several hundred miles away in the Belgian Congo, where the HIV-1(M) epicentre is clearly situated? And (b), why did different HIV-1 strains emanating from that single source suddenly start recombining in mid-twentieth century, to create the many different subtypes which we recognise today, and which have survived almost unchanged in the half century since? I believe that the scenario required by the bushmeat theorists, when examined more closely, becomes highly implausible.
By contrast, if one returns to Myers’ scenario, there was indeed a potential punctuated event in the 1950s, and it was this: SIV recombination occurring in vitro in the Stanleyville chimpanzee tissue cultures, allowing new and unique SIV variants to grow in the live polio vaccines that were administered in at least 28 separate vaccination campaigns across central Africa between 1957 and 1960.
In summary, the claims that have been made by the phylogenetic daters (and later extensively quoted by people such as Stanley Plotkin and Hilary Koprowski) are based on intrinsically flawed science. Their conclusions (that HIV-1 predates the OPV trials) are unproven, and are based on little more than wishful thinking.
“Disproof” 7: that there is anyway no meaningful correlation between the Koprowski vaccinations in Africa and the first appearances of HIV-1 and AIDS
Finally, there is the epidemiological argument. In 2002 I was approached by a biological modeller and statistician whose papers have been published in Nature and other leading journals, who had followed the origins debate from afar, and who felt that some of the scientific arguments that had been put up against the OPV theory were not convincing. This man offered to take a look at the raw data on the mooted CHAT/AIDS connection, and to do some number-crunching. The early results were sufficiently intriguing to prompt further research.
In the end, we decided to write a paper on the work, and to analyse not just the OPV theory, but also the other major hypothesis of origin, the “cut hunter” or “bushmeat” theory (and the various sub-hypotheses of the bushmeat theory, such as the urbanisation hypothesis, and Preston Marx’s dirty needles hypothesis).
The analysis detected a significant link in only one instance: when the early foci of HIV-1 infection were compared to places where Koprowski’s vaccines had been administered in Africa in the late 1950s. The correlations are “highly significant” (meaning that there is less than a 1 in 1,000 chance that they could be caused by coincidence alone). Furthermore, they applied both on a macrocosmic scale, across central Africa, and on a microcosmic scale, in the small country of Burundi.18
This study substantially undermines the one full epidemiological paper that Robin Weiss allowed to be presented at the Royal Society meeting. This was written by Kevin De Cock, a Belgian-American epidemiologist based at the CDC, who had previously collaborated with Hahn and Sharp on some of their papers proposing Pan troglodytes troglodytes as the true source of HIV-1. De Cock’s paper ignored the CHAT vaccinations in Rwanda and Burundi (over half of the African total), and then concluded that there was no significant association between CHAT and AIDS.19 At the end of the London meeting, I asked Dr De Cock why he had omitted the data from Rwanda and Burundi, but he was unable to give a meaningful answer.
To sum up: the claim that there is no link between CHAT vaccination in Africa and early HIV infections and AIDS is not supported by the data. In reality, there is an extremely dramatic correlation between the places where Koprowski’s vaccines were fed in central Africa, and the first appearances of HIV-1(M) and AIDS some few years later.
The latest historical evidence (especially about local vaccine preparation in Stanleyville) is revelatory, and demonstrates beyond reasonable doubt that live polio vaccines prepared in chimpanzee tissue cultures (at least some of which were almost certainly contaminated with chimpanzee SIV) were administered to up to one million African “volunteers” in the 1957-1960 period. The specific populations where the vaccine was given were the first in the world to experience HIV-1 infections and AIDS some five to twenty years later.
This latest scientific evidence is also strongly supportive of the OPV theory. Significantly, it is now apparent that every one of the alleged “disproofs” of the OPV theory, which were so loudly trumpeted at the Royal Society meeting and in the pages of Nature and Science, is either false or inherently flawed.
This raises questions about the integrity and motivation of some of the scientists who have so eagerly announced such claims. One of these questions has to be: are senior members of the scientific establishment engaged in a deliberate cover-up in order to protect the good name of vaccination (as well as certain professional and financial vested interests)?
If so, let them be reassured on one score, at least. The safety of vaccination per se is not in question here, for there is little doubt that vaccination has saved more lives than any other public health measure ever introduced.
It is the safety of one particular polio vaccine (the African version of CHAT) which is being questioned, together with the activities (both in the fifties and more recently) of the scientists who prepared and tested it.
Edward Hooper. Original article completed: February 29th, 2004. This updated version completed: January 27th, 2005.
1 See papers in published proceedings of the Royal Society conference: R.A. Weiss, S. Wain-Hobson (editors), “Origins of HIV and the AIDS Epidemic”; Phil. Trans. Roy. Soc. Lond. B; 2001; 256; 771-977.
6 M.L. Santiago, P.M. Sharp, G.M. Shaw, B.H. Hahn et al., “Foci of endemic simian immunodeficiency virus infection in wild-living eastern chimpanzees (Pan troglodytes schweinfurthii)”; J. Virol.; 2003; 77; 7545-7562.
7 P. Gagneux, M.K. Gonder, T.L. Goldberg and P. A. Morin, “Gene flow in wild chimpanzee populations: what genetic data tell us about chimpanzee movement over space and time”; Phil. Trans. Roy. Soc. Lond. B; 2001; 256; 889-897.
10 E. Hooper, “Dephlogistication, imperial display, apes, angels, and the return of Monsieur Emile Zola. New developments in the origins of AIDS controversy, including some observations about ways in which the scientific establishment may seek to limit open debate and flow of information on ‘difficult’ issues”; Atti dei Convegni Lincei; 2003; 187; 27-230.
12 S. Wain-Hobson, J-P Vartanian, A. Meyerhans et al., “Network analysis of human and simian immunodeficiency virus sequence sets reveals massive recombination resulting in shorter pathways”; J. Gen Virol.; 2003; 84; 885-895.