by Pepin
Of course there are many sources of error: inaccuracies in the censuses, lower exposure to chimps in the 5% of the population that was truly urban, no exposure in parts of Moyen-Congo where there were no P.t. troglodytes, exposures among adolescents, recall biases in the estimate of the proportion exposed to chimpanzee blood, etc. However, this provides us with an order of magnitude of the number of individuals potentially exposed to SIVcpz. This number must now be multiplied by the percentage of chimpanzees infected with SIVcpz and by the probability of transmission during each exposure, if the source chimpanzee was indeed infected with SIVcpz.
We will assume that the SIVcpz prevalence in 1921 among P.t. troglodytes communities of central Africa was similar to what it is today. In Cameroon, 5.9% of wild P.t. troglodytes are infected with SIVcpz. Thus, of the 1,350 adults exposed to chimpanzee blood, about 80 were exposed to a SIVcpz-infected ape. Then how many of these 80 individuals did acquire SIVcpz? This depends on the probability of transmission during each exposure.20
This probability must have varied according to the degree of viraemia, i.e. the quantity of SIVcpz present in the ape‘s blood. The higher the viraemia, the more infectious a given amount of blood was. In humans, the degree of HIV-1 viraemia increases markedly as the disease progresses towards AIDS. Although it was initially assumed that SIVcpz is not pathogenic to chimpanzees, there is now much evidence supporting the opposite view. Experimental HIV-1 infection of chimpanzees can lead to rapid loss of their CD4 lymphocytes and full-blown AIDS with opportunistic infections. More recently, it was proven that SIVcpz was pathogenic in wild P.t. schweinfurthii. In the Gombe reserve, ninety-four chimpanzees habituated to human contact were followed for nine years. Analyses of their stools showed which chimpanzees were infected with SIVcpz. Over this period, the mortality rate in the seventeen chimpanzees infected with SIVcpz was ten times higher than that among the non-infected animals. Although based on a limited number of observations, this relative mortality was similar to what happens in humans infected with HIV-1. An autopsy, done on three of the infected chimpanzees, revealed a reduction in the number of their CD4 lymphocytes. In one case, less than three years had elapsed between the infection with SIVcpz and death. This suggested that SIVcpz infection in chimpanzees could lead to immunodeficiency and high viraemia, which would make these animals more infectious, not just to other chimps, but to humans as well. It remains unknown whether SIVcpz is as pathogenic in P.t. troglodytes as in their cousin P.t. schweinfurthii.21–28
So how much virus is there in their blood? In an early study using insensitive methods, our friend Noah, naturally infected with SIVcpz-ant, was found to be viremic at a low level. Using the blood of Noah injected IV, a second chimp, Ch-Ni, was experimentally infected with SIVcpz-ant. Very much like humans who develop acute HIV infection, Ch-Ni developed very high viraemia (5 × 106 viral copies/ml), which declined markedly in the ensuing months, eventually reaching levels similar to those of Noah. A few years later, using modern viral quantification methods, Noah was shown to have persistent and rather high viraemia (105 copies/ml). Ch-Ni maintained stable but lower viraemia (104 copies/ml). In humans, such levels are found at an advanced stage of HIV-1 infection, with a diagnosis of AIDS. Higher HIV-1 levels (105–107 copies/ml) are seen for a few weeks during acute infection and much later as a pre-terminal condition in untreated patients or in those whose virus has become resistant to all available therapies. To summarise, we can assume that in chimpanzees the degree of viraemia, and thus the potential for blood-borne transmission, is similar to that in humans.23,29–31
To estimate the frequency of SIVcpz transmission when someone is exposed to chimpanzee blood, we can extrapolate from studies among healthcare workers exposed to blood from an HIV-1-infected patient, which correspond fairly well to what we are interested in here: the risk of transmission after a single accidental exposure through the skin. Since the mid-1990s, healthcare workers exposed to HIV-1 have been given preventive antiretroviral drugs, which decrease the probability of transmission by 80%. Before such prophylactic methods became widely used, the overall risk of transmission was 0.3% per exposure, but was modulated by factors reflecting the quantity of viruses inoculated during the injury. Transmission was much less common (0.03%) for exposures that involved only non-penetrating contact (say, a splash) between blood and mucous membranes (the mouth or eyes) or non-intact skin. The risk when blood came into contact with intact skin was near zero. Conversely, with deeper injuries, or accidents during which a larger quantity of blood was inoculated, the risk of transmission could be as high as 25%, especially when the incident involved a patient with terminal illness.32,33
When a hunter or cook was exposed to chimpanzee blood containing SIVcpz in the bush of central Africa, some exposures would have been similar to those of healthcare workers, while others were far worse, for example a deep injury with a knife used for butchering or a traumatic wound inflicted by a struggling chimpanzee. Overall, the risk of transmission was probably higher than among healthcare workers sustaining needle pricks, because the injuries were more severe. As an educated guess, let us say that it was between 1 and 3%, that is, up to ten times more frequent than in occupationally exposed healthcare workers.
We had calculated that eighty adults living in central Africa in 1921 had been exposed to blood containing SIVcpz while handling chimpanzee carcasses or hunting. With 1% transmission, the result is one human infected from chimps, and two or three if the risk of transmission per exposure was closer to 3%. Obviously, there are several sources of errors so these estimates cannot be taken at face value, but the bottom line is that the number of persons infected naturally with SIVcpz around 1921 must have been small, and almost certainly less than ten.
We know that over several decades a cross-species transmission occurred at least four times for HIV-1, i.e. once for each of groups M, N, O and P, each of which is thought to reflect distinct cross-species transmission events rather than evolution within humans. It occurred at least eight times for each of the different groups of SIVsmm, which became HIV-2, as we will see later. Thus, the epidemic of HIV-1 group M was triggered not because a lot of humans were infected directly from chimpanzees but because a rare case of infection managed to spread and multiply, something which all the others that preceded it had not managed to do. In the following chapters, we will try to understand why this time the subsequent human-to-human transmission was so effective. But apart from the occupational infections of hunters or cooks, could other modes of transmission have been responsible for the first case(s) of SIVcpz transmission from chimpanzee to man? We will now review three such hypotheses.
The river
Oral poliovirus vaccines (OPV) distributed in the 1950s and early 1960s were massively contaminated with a simian virus (simian vacuolating virus 40), which originated from the macaque cells used to grow the vaccinal virus. Fortunately, this virus was not pathogenic for humans. In an article published in Rolling Stone in 1992, journalist Tom Curtis proposed the theory that HIV-1 came from the contamination of OPV with SIV from African green monkeys. Of course, it was later shown that African green monkeys were not the source of HIV-1, as their own SIVs were too different from HIV-1. Then in 1999, Edward Hooper published a book entitled The river. A journey back to the source of HIV and AIDS, which focused on the theory that chimpanzee cells had been used to produce an experimental oral polio vaccine called CHAT, developed by Hilary Koprowski at the Wistar Institute in Philadelphia. Koprowski had collaborated with Belgian scientists from the Stanleyville public health laboratory in the Belgian Congo. Clinical trials of the CHAT vaccine were conducted between 1957 and 1960 in the vicinity of Stanleyville, in the capital Léopoldville, as well as in the Ruzizi valley of Ruanda–Urundi. Eventually, the Koprowski vaccine proved inferior to the other OPV developed by his competitor Albert Sabin, and CHAT was never commercialised.34
The Stanleyville laboratory set up a colony of chimpanzees (presumably, all P.t. schweinfurthii or Pan paniscus) ne
arby, at a place called Camp Lindi, primarily to verify experimentally not just the efficacy but also the neurological virulence of the CHAT vaccine. Experimental oral polio vaccines were based on live viruses which had been empirically attenuated through repeated passages over various cell cultures, and scientists were worried that these viruses could eventually regain their original virulence, in which case the vaccine would cause the disease rather than prevent it. Camp Lindi chimps were given the CHAT vaccine, then the wild polio virus, and monitored clinically for neurological deficits. Others were sacrificed for their spinal cord to be examined after intraspinal injection of the CHAT virus. In addition, about thirty chimpanzees were used for hepatitis research, in an attempt to identify the agents of viral hepatitis through inoculation of stool suspensions from human patients.
The main theory in The river was that the Stanleyville laboratory had supplied the Wistar Institute with chimpanzee tissues obtained from sacrificed animals, which had then been used in Philadelphia to produce the CHAT vaccine so that SIVcpz had been introduced into some vaccine lots, which were then sent back to the Congo where their oral administration in the late 1950s started the pandemic. A number of elements described in the book itself made this theory implausible, most importantly the fact that Koprowski had conducted very large trials of the same CHAT vaccine in his native Poland, where HIV-1 did not emerge. It would have been an extraordinary coincidence for SIVcpz-contaminated vaccines to be re-exported from Philadelphia only to the very country where, allegedly, the SIVcpz-infected chimpanzee tissues had been obtained. Then Hooper changed his hypothesis somewhat, and suggested that the Stanleyville laboratory had produced batches of the vaccine locally, using cells from locally procured chimpanzees, which led to contamination with SIVcpz in Stanleyville rather than in Philadelphia.
Many circumstantial elements argued against either version of the hypothesis. There is no documentary evidence that chimpanzee cells were ever used, anywhere in the world, to produce OPV. Scientists had easy access to small monkeys of the Macaca genus, which were abundant in Asia, cheaper and easier to handle than chimpanzees, raised fewer ethical issues and worked well in cell culture systems. In 1955, up to 200,000 rhesus monkeys were imported into the US for medical research. Furthermore, as noted by the late Dr Paul Osterrieth, a scientist who worked at the Stanleyville laboratory for a few years, it was technically impossible for this rather basic facility to produce any kind of novel viral vaccine. The laboratory just did not have the human and material resources for such an endeavour. The annual reports of the Stanleyville laboratory for the crucial years never mentioned any local production of OPV, something that, had they achieved it, the laboratory workers would certainly have been proud of. Neither did the reports of the colony‘s health system mention any local production of OPV – while the same reports provided much information about other vaccines produced by the network of public health laboratories of the Belgian Congo. Hooper seems to have confused conditionnement, which meant local dilution of concentrated frozen vaccine stock or its distribution from a large container to smaller containers, with local production or amplification of the vaccine strain.35–37
Nevertheless, to test this hypothesis once and for all, old vials containing the CHAT vaccine, which had been kept frozen for decades, were located. Some came from the Wistar Institute, where CHAT had been developed and produced, and conspiracy theorists could argue that this institution had a vested interest in supplying vials which they already knew were not contaminated with HIV-1. But other CHAT vials were fished out of freezers at the CDC in Atlanta and at Britain‘s National Institute for Biological Standards and Control (NIBSC). Some vials contained the very batches (10A-11 and 13) that had been used in the Belgian colonies. Samples were tested by several institutions, including the NIBSC, the Institut Pasteur, the Max Planck Institute, the Karolinska Institute, the New York University School of Medicine and Roche Molecular Systems. All reached the same conclusions: there was no HIV or SIV nucleic acids in these vials; there was no chimpanzee DNA, only DNA from rhesus (Macaca mulatta) or cynomolgus (Macaca fascicularis) monkeys and, in one batch which had been grown on human diploid cells, Homo sapiens DNA; and there was poliovirus, which meant that there had not been extensive degradation of viral nucleic acids over the very long storage period.38–40
Furthermore, 131 faecal samples were collected from chimps in the forested areas around Kisangani (ex-Stanleyville), where the laboratory and chimpanzee colony had been located: only one was SIV-positive. This virus, obtained from a P.t. schweinfurthii, was called SIVcpz-DRC1. In phylogenetic trees, it was close to the P.t. schweinfurthii viruses obtained from Uganda and Tanzania and with Noah‘s SIVcpz-ant, and clearly distinct from all P.t. troglodytes SIVs.41
Therefore, there is no evidence that chimpanzee cells were used in the Stanleyville laboratory to prepare batches of OPV vaccines, or that any OPV was produced there. And even if chimpanzee cells had been used in Stanleyville, they could not have contained the P.t. troglodytes SIVcpz that triggered the pandemic. There is no evidence either of any retrovirus having been present in the old batches of CHAT available for testing. Furthermore, even when taking into account the margin of error on these estimates, SIVcpz emerged in human populations at least twenty-five years before the CHAT trials. This theory can be firmly rejected.
Bold experiments
Because of their similarity to humans, chimpanzees have been used for almost a century as animal models of many infectious diseases, to prove that a putative pathogen is the cause of a given disease, to evaluate new vaccines, etc. In the course of these experiments, chimpanzees have been injected with various amounts of human blood or other types of human specimens, containing a wide diversity of infectious agents: viruses (HIV-1 of course, poliomyelitis virus, hepatitis B virus (HBV), hepatitis C virus (HCV) and yellow fever virus), prions (proteins causing kuru, Creutzfeldt-Jakob disease and scrapie), bacteria (the aetiological agents of tuberculosis, leprosy, gonorrhoea and trachoma) and parasites (causing malaria and the Guinea worm). There was even a chimpanzee model of alcoholism and addiction to narcotics!
However, it is rather extraordinary that, on several occasions, the reverse was done: the IV injections of chimpanzee blood in humans. The first such experiment was performed in Sierra Leone by Donald Blacklock and Saul Adler, who injected two Europeans (presumably, themselves) subcutaneously (SC) and intravenously with small quantities of blood from a chimpanzee infected with malaria parasites. They did not develop malaria. We do not know what happened next but they certainly got worried when the donor chimpanzee died a few days later, with the autopsy showing a disseminated Strongyloides infection, with this intestinal worm being present in the chimp‘s bloodstream! Shortly thereafter, at the Institut Pasteur in Paris, a man was injected IV with 40 cc of chimpanzee blood as part of a comparative study of blood groups of chimpanzees and humans. The volunteer apparently tolerated the procedure well. With great foresight, the Pasteur scientist noted that such experiments should be avoided in the future, ‘to keep off potentially transmitting to humans hitherto unknown infectious pathogens of chimpanzees‘!42–45
This advice seems to have fallen on deaf ears. Renowned parasitologist Jérme Rodhain, director of the Tropical Medicine Institute in Brussels and then in Antwerp, studied whether malaria parasites of primates were transmissible to man. As a secondary interest, or perhaps as a moral justification, he also investigated whether fever induced by malaria could have a beneficial effect on patients with late-stage syphilis. Such work would be unthinkable today but in those days there were no ethics committees and each scientist could decide whether an experiment was morally acceptable. Similar investigations were conducted in other parts of the world, especially India, with inoculation to humans of blood obtained from Asian apes and monkeys.46
Rodhain carried out a series of experiments in which chimpanzee blood (5–10 cc) was injected IV in humans, most of them patients with syphilitic dementia. The chimpanzees (Thomas, Suzanne
, Simone . . .) originated from the Belgian Congo; although Rodhain alluded to them as P.t. verus, in retrospect they were probably P.t. schweinfurthii. Between 1938 and 1940, Rodhain injected chimpanzee blood to twenty-six patients. He did prove that some of these malaria parasites were infectious to humans, and that the parasite named by others Plasmodium rodhaini, in his honour, was in fact Plasmodium malariae. We should give him a lot of credit for being humble since he himself killed the chance of his name going down in history as the name of a species. Rodhain conducted similar experimentations with other species of malaria parasites, most notably Plasmodium reichenowi, a parasite of chimpanzees and gorillas. In 1954–5, he injected chimpanzee blood in four more psychiatric patients, and managed to transmit Plasmodium schwetzi.47–52
These experiments were done in hospitals in Antwerp and clearly could not have led to the emergence of HIV-1 in Africa. Furthermore, as the apes originated from the Belgian Congo, they were all presumably Pan troglodytes schweinfurthii (Rodhain would not have confused P.t. verus with Pan paniscus, the other ape potentially available from his Belgian collaborators). It seems unlikely that other scientists carried out similar experiments for malaria research, since Rodhain would have mentioned this in the detailed reviews of the subject that he wrote in the discussion section of his own papers. Of course, one could hypothesise that similar experiments were carried out in the nascent research institutions in Africa and the results never disseminated, but this would be pure speculation. There is no evidence of similar experiments being conducted in the Léopoldville research laboratory, where Rodhain had worked during the early part of his career and which he continued to visit intermittently from his academic posts in Belgium.
