Kuru experiments triggered the emergence of pathogenic SIVmac

Cross-species transmission of the simian immunodeficiency virus (SIVsm) that naturally infects sooty mangabeys (SM) to different species of macaques resulted in the emergence of the reference strain SIVmac and of the related viruses [1–6]. Unlike the vast majority of naturally infected African non-human primates (NHP), SIVmac infection of Asian macaques induces an AIDS-like disease in virtually all infected animals [1,2,5]. For these reasons, SIVmac infection of macaques is the most commonly used animal model for studies of AIDS pathogenesis and vaccines. One should note however that the SIVsm family (which gave rise to SIVmac) is the only SIV shown to be pathogenic in rhesus macaques (Rh). All other SIV when experimentally inoculated into Rh have failed to induce immunesuppression and were cleared by the Rh host [7–10]. This is also supported by our recent observation that Rh infections with primary SIVsm isolates may not be as pathogenic as previously believed. Surprisingly, in some cases Rh cleared primary SIVsm infection, with no detectable viral loads for over 1 year [11,12] (C. Apetrei, unpublished data). In this context, several questions remain to be addressed to understand the mechanism(s) of emergence of pathogenic viruses following cross-species transmission: Why is SIVmac239 highly pathogenic in Rh? What were the circumstances of the emergence of this pathogenic virus? What are the requirements for a cross-transmitted virus to become pathogenic in a new host? The answers to these questions are not easy to obtain because the emergence of pathogenic SIVmac occurred in the 1970s and it was not until the mid 1980s that the virus was discovered, by which point it had already reached its full pathogenic potential. However, we have collected and reassembled data that strongly suggest that highly invasive experimental manipulations, rather than casual contact, inadvertently gave rise to the highly pathogenic SIVmac isolate, used in AIDS research from the 1980s to the present time. The first NHP equivalent of HIV, SIVmac, was discovered in 1985 in Rh at the New England Primate Research Center (NEPRC) and traced back to the California National Primate Research Center (CNPRC) [1,13]. SIVstm, another early strain identified [3], was found in 1992 in stump-tailed macaques (STM) at the Yerkes National Primate Research Center (YNPRC) [3]. Similarly, SIVstm was also traced back to the CNPRC [3,14]. At the CNPRC, both viruses generated outbreaks of simian AIDS in the 1970s [3,14]. Retrospective phylogenetic analysis of these outbreaks suggests that the emergence of both SIVmac and SIVstm followed cross-species transmission of SIVsm to these species of macaques [14,15]. These cross-species transmissions were postulated to result from simple contact transmission between the different monkey species in captivity [14]. However, the high frequency of cross-species transmission of SIV at the CNPRC is highly unusual and remains unexplained. In 50 years of SM colonies in Primate Centers in the US the only other documented SIVsm transmission from mangabeys to Rh occurred 10 years later, at the Tulane National Primate Research Center (TNPRC) and was experimentally triggered [5]. During our molecular epidemiology study of SIVsm strain diversity and circulation over the last 30 years, we identified the source SM strains for both SIVmac and SIVstm[15]. Also, we have characterized SIVsm diversity in all available samples collected over the last 30 years from SM in all four colonies in the US. These findings allowed us to identify a strong connection between the emergence of SIVmac and SIVstm and the experiments conducted by C. Gajdusek in numerous macaques and SM to study the pathogenesis of kuru in NHP. Kuru experiments to develop an animal model for prion disease were carried out extensively in different species of primates at the CNPRC and New Iberia Research Center (NIRC) from the 1960s to 1986 [16–20]. Among the species used were SM at both centers and macaques at the CNPRC. SM were included in kuru research at the CNPRC until 1977, then the SM colony was discontinued. Originally, the CNPRC SM colony was established with animals directly imported from Africa. In 1977, SM from the kuru studies at the CNPRC were sent to the NIRC where kuru research by Gajdusek and his colleagues was continued until 1986 [16]. At the NIRC more SM imported from Africa were added to the ongoing kuru experiments. We were able to trace the movement and usage of some of these SM and have characterized the SIVsm infection status through serology and sequencing [15]. We uncovered three independent pieces of evidence supporting the hypothesis of a role for kuru research in the emergence of SIVmac. (i) The first line of evidence is the geographical and temporal coincidence between the emergence of the two SIV (i.e., SIVmac and SIVstm) and the kuru experiments. At the CNPRC, kuru experiments involved STM and Rh, as well as SM [16–19]. In both Rh and STM, kuru experiments were carried out several years prior to the AIDS outbreaks. In fact, this time interval between the kuru experiments and the AIDS outbreak is equivalent to the incubation period of simian AIDS [18,19]. Moreover, there is species-specific temporal coincidence between kuru experiments and emergence of pathogenic SIV in macaques: kuru studies on Rh predated those on STM by over 5 years (1963 versus 1968) [18,19] and the epizootic of AIDS in Rh occurred 7 years earlier than in STM (1968 versus 1975) [14]. The incubation period of AIDS in STM is known to be longer than in Rh [21]. (ii) The second line of evidence is that the limited SIVsm diversity in SM at the NIRC is directly correlated to the animals' used in kuru experiments (Fig. 1). Animals in the kuru study group (which includes those transferred from CNPRC) were all infected with lineage 7 SIVsm strains, whereas those in the control group were all infected with lineage 4 strains (Fig. 1). In contrast, SIVsm shows an extremely high variability in wild monkeys [15,22,23]. The genetic distances between strains originating from the same troop were between 15 and 25% in Gag [15,22,23]. Thus far, only two SIVsm strains out of the 27 characterized in wild SM by different groups showed genetic divergence comparable to the 2% in the kuru experiments (2%) [23]. These two strains have been confirmed to be related through vertical transmissison. Monkeys in the kuru study group were not epidemiologically related prior to the experiments. We were able to prove that numerous SIVsm genetic lineages were imported to the US through SM importation [15]. SIV was not known at that time, which precluded a selection of animals for the kuru experiments based on SIVsm lineage. Therefore, it is extremely improbable that monkeys from the kuru control group (mostly imported from Africa) were all naturally infected with the same SIVsm strain. As such, this observation supports the hypothesis that SIVsm transmission in these two groups of SM occurred by inoculation during kuru experiments. (iii) Finally, the third line of evidence derives from our serological analysis on serial serum samples collected over a 7-year-period (1979–1986) from SM housed at the NIRC. This analysis revealed a seroconversion in one of the mangabeys in the kuru study group, confirming that kuru experiments resulted in SIV transmission between SM (Fig. 2).Fig. 1: Molecular epidemiologic evidence of SIVsm serial transmission during kuru experiments. SIVsm lineages segregate according to experimental and control group in kuru research. Lineages (color-coded: red, purple, turquoise, blue, orange, light green, dark violet, pink and rust, for lineages 1 through 9, respectively) are clusters of SIV that are highly related and branch together. Reference strains from macaques are shown in black. HIV-2 strains (light gray) were added to show that different SIVsm lineages roughly display the same divergence pattern as HIV-2 groups. SIVsm strains from wild-caught SM originating from the same are of Sierra Leone (light blue) are as divergent as the different lineages, supporting the origin of these lineages through SM importation from Africa. Lineages 8 and 9 are strains that were cross-species transmitted to macaques at the California National Primate Research Center (boxed). The tree is based on env fragments (405 base pairs after gapstripping) and was constructed by neighbor-joining on nucleotide sequences. Only relevant bootstrap values are shown. The scale bar indicates nucleotide substitutions per site. Strain nomenclature includes the assigned lineage, the Primate Center of origin, the year of sample collection, SM identification and the macaque species for macaque strains.Fig. 2: Dynamics of anti-SIVsm gp41 antibodies in SMs from the kuru experiments at the New Iberia Research Center (NIRC). When transferred to New Iberia, the SM had been previously used in kuru experiments. SIVsm transmission may have therefore occurred prior to transfer of most animals. One seroconversion was documented in SM #CFU212: serum from this monkey was sampled in 1979 and lacked anti-SIVsm antibodies when it arrived in NIRC. Serial specimens from CFU212 showed increased anti-SIVsm reactivities, supporting recent seroconversion. CFU212 is shown in black. The remaining monkeys in the kuru study group are shown in dark gray; monkeys in the kuru control group are shown in light gray. ELISA cut-off level (0.20) showed with a dashed line.These three independent lines of evidence implicate kuru studies in the emergence of two pathogenic SIV strains, SIVmac and SIVstm, in macaques, now used in studies of simian AIDS. This kuru connection also explains the very high frequency of SIVsm cross-species transmissions at the CNPRC. Moreover, serial passage of SIV through kuru experiments is a plausible mechanism for adaptation of SIVsm to macaques following the initial cross-species transmission. The only other occurrence of cross-species transmission of SIVsm to macaques occurred after Rh inoculation with blood from SM that were known to be naturally infected with Mycobacterium leprae in an attempt to transmit leprosy [5,24]. These animals were also infected with SIVsm but this was not known at the time of the leprosy experiments. This experimentally triggered cross-species transmission of SIVsm resulted in simian AIDS in Rh and the subsequent discovery of the SIVsmB670 [5]. Moreover, during leprosy experiments, the serial passage of SIVsm was shown to be involved in the selection of highly pathogenic SIVsmB670[5]. Serial passage of SIV in macaques during kuru experiments would thus explain the highly pathogenic potential of SIVmac strains. One should also note that serial passage was proven to be the main mechanism behind increasing pathogenic potential in the case of SHIV [25,26]. Definitive proof is hampered by the lack of archival records and samples from the 1970s. Published literature is also incomplete, because kuru experiments in the early 1970s involved more than 1500 monkeys and apes in attempts to serially passage the disease, but only about 300 cases were published. In one of the Gajdusek's papers it is stated that “attempts to serially propagate kuru and C-J disease in the aforementioned species of animals (which included Rh, stump-tailed macaques, cynomolgous macaque and mangabeys) have already been initiated” [27]. However, no follow-up to the mentioned paper was found. Therefore, we believe that the confirmation of our hypothesis lies in the high number of animals that were used in these extensive studies but were never accounted for in subsequent publications. Our analysis thus far presents a highly plausible chain of events whereby SIV was transmitted from SM to macaques through highly invasive experimental manipulations, and not simply casual contact. The fact that many of these experiments involved serial passages from animal to animal would have facilitated the development of the highly pathogenic SIVmac in common use today. Acknowledgements We thank Michael Metzger, Nora Dillon, Rajeev Gautam and Nathalia Katz for their excellent technical assistance. This work was supported by funds from Grants RO1 AI-19301, RO1 AI-44596, RO1 AI149809, RO1 AI064066, P20 RR020159 and P51 RR000164 from the National Institute of Health and from NINDS contract N02-NS-4-2358 to New Iberia Research Center.