Foamy viruses (FVs) are members of the Spumavirus genus of the Retroviridae family [1]. These exogenous complex retroviruses are highly prevalent in several animal species, including primates, felines, bovines and equines where they cause persistent infections [2–7]. Simian foamy viral (SFV) infection has also been reported in 1 to 4 % of persons occupationally exposed to non-human primates in zoos, primate centers and laboratories, mainly in Northern America but also in Europe [8–12]. Very recently, naturally acquired SFV infections have been described in few hunters living in Cameroon, central Africa [13] (and Calattini et al., in preparation) and in one person with frequent contacts with Macaca fascicularis in a temple in Bali, Indonesia [14].

Foamy viruses are considered as non-pathogenic in naturally or experimentally infected animals [15, 16]. Furthermore, they do not seem to cause any disease in the very few humans who were accidentally infected, and who have then beneficiated of a long-term medical and biological follow-up [9, 11, 12, 17]. This lack of pathogenicity contrasts strongly with the cytopathic effect that is seen in vitro in infected cell cultures, with the appearance of "foamy-like" syncitia [15, 18, 19].

In contrast to the HIV/SIV lentiviruses, foamy viruses exhibit a very low genetic drift in vivo [2, 20–22]. Phylogenetic analyses have also demonstrated a species-specific distribution of foamy viruses. This indicates a long-term co-evolution of such retroviruses with their natural hosts [23]. Recently, Switzer et al. demonstrated that FVs might have co-speciated with Old World primates for at least 30 million years [24]. Such features could explain their possible lack of pathogenicity that is observed in vivo and the long-life persistence of the infection [4, 20, 21]. Worth noting is that the great majority of the viral strains yet characterized concerns African monkeys and Apes. Indeed, relatively few data are known on the variability of FVs in Asian monkeys, despite an important biodiversity of such animals, especially within the macaques species [8, 24, 25].

While the molecular features of foamy viruses have been extensively studied in vitro [15, 18, 19, 26], only few data are available on the characteristics of FVs in vivo, including epidemiological determinants [3, 4, 16, 20–22]. As an example, the timing and modes of primary infection are not well known.

The few published epidemiological studies indicate that among captive non human primate populations, antibodies seroprevalence to SFVs can reach up to 75–100% in adults [4, 16, 20]. Furthermore, there is only one recent study reporting the SFV seroprevalence in a free-ranging group of non-human primates (NHPs) [14]. This study concerns a group of 38 macaques living in Bali, Indonesia. However, most studies are cross-sectional works in captive animals and no long-term follow-up searching specifically for time and mode of seroconversion had been performed. Regarding the modes of infection, some studies have shown that SFVs are present at a high concentration in the saliva of infected animals [26–28] Throat mucosa has been shown to be an important site for viral replication in African green monkeys [27], and a very recent study demonstrated high levels of viral RNA in oral tissues of macaques [28]. All together, this suggests that bites, scratches and mucosal splashes can be mechanisms of transmission, at least in some animals. Other studies in captive colonies of baboons have suggested that sexual and/or mother to offspring transmission through saliva contacts can occurred [2, 20].

We have conducted a study in a free-breeding colony of Macaca tonkeana housed in the Strasbourg Primatology Center in France. This colony was followed for more than 24 years for behavioral investigations including the study of social relationships and reproductive behaviors [29–34]. The goals of our current study were: 1) to determine the prevalence of SFV infection in this colony, 2) to characterize the viruses that infect these animals and to study their genetic variability overtime through a long-term follow-up, 3) to try to get new insights concerning the timing and modes of foamy viruses primary infection in these monkeys by combining serology and molecular means as well as studies of familial structures and long-term behavioral observations.

Our findings on the FV prevalence confirmed that captive colonies of non-human primates are often highly endemic for foamy viruses [4, 16, 20]. In fact, in our study, nearly half of the animals (and 89% of the adults) are infected by FVs. Furthermore, for the first time, we extended herein to M. tonkeana species the presence of such high levels of FVs infection.

It is important to note that our study took place in a quite large, free-breeding colony, which is not the case in most of the few other studies performed for FVs in NHP captive colonies [2, 4, 8, 20, 22]. Moreover, some of the published studies have been performed in colonies comprising different species of monkeys. For instance, in a captive US colony of 254 baboons, including 150 adults and 104 juvenile, 88% (132/150) of the adults were FVs seropositive. However, this colony comprised at least 3 different subspecies of baboon (Papio ursinus, P. anubis and P. cynocephalus) of different origins, with some animals having been recently captured in East Africa, while most of the other baboons were long time residents of the captive colony (5 to >15 years). Furthermore, in this group, most of the young animals were removed from the breeding harem at 6–9 months of age, therefore reducing the opportunities of being infected by FVs [2, 28]. In another US colony of baboons, all the 38 adults, housed in gang cages, were found FVs seropositive while all but one of the 10 juveniles, were found seronegative. However, in this work too, the juveniles were not housed in most of the cases with their mothers, having been removed from them shortly after birth [20].

Regarding specifically Asian monkeys, a survey of a colony of M. fascicularis, held at Health Canada (Ottawa), and all bred from wild-caught animals, indicated that 80% of the 395 animals were infected by SFVs [8]. Verschoor et al., found also that 69.4% of 108 orangutan blood samples originating from a reintroduction center in East Kalimatan were found seropositive for FVs [37]. Lastly, only one study recently published has been performed in a free-ranging colony of monkeys, i. e. a group of 38 macaques (mostly adults) living in a temple in Central Bali, Indonesia. In this case, the seroprevalence for FVs was of 89.5%, reaching 93% in adults [14].

In our study, we have also analyzed the presence of FVs proviral DNA in the peripheral blood buffy-coat DNA of most of the studied animals. The PCR negativity obtained in some animals with a clear positive WB, was probably linked to a very low viral load in the peripheral blood buffy-coat (< than 1–10 copies in 500 ng DNA, i.e. 75 000 cells). Moreover, we can strongly suggest here that the negativity of PCR is not due to the presence of divergent foamy viruses, but to a low viral load. Indeed, in two animals, T6 and TG1, for which the integrase PCR was negative in the uncultured PBMCs, we were able to amplify the same integrase fragment on cultured cells. Up to know, little is known about the FVs proviral load in naturally infected NHPs [28]. In a study of African green monkeys, a very low proviral load was detected in most tissues and very recently, a report described that SFV DNA was present at a low copy number in PBMCs and tissue from macaques [27, 28]. Previously, we found, using a similar semi-quantitative technique, a low proviral load ranging from 1 to 100 copies for 500 ng of peripheral blood buffy-coat DNA, in a series of wild-caught chimpanzees [38]. Furthermore, such lack of detection of FVs sequences has also been reported in the PBMCs DNA of several hunters, living in remote villages of South Cameroon, who where found to exhibit a clear FV WB seroreactivity with the presence of the gag doublet [13].

Analysis of the foamy viral sequences found in the 17 monkeys, for which the integrase gene could be amplified, indicates clearly the presence of 6 different groups of FVs strains in this colony of M. Tonkeana. This colony has been carefully followed for more than 24 years for behavioral investigations. Furthermore, these animals have never been in contact with any other monkeys since their arrival in France in 1972 and they have neither been used in any biomedical experiments, nor injected with any biological materials [39]. Keepers did not manipulate animals from different species of the center with the same gloves. Cross-species transmission in the primatology center was thus very unlikely, but of course, it cannot be ruled out with 100% certainty. Thus, the viruses currently present in this colony should have been present originally in some of the founders of the colony. As all the members of this troop originated from only 5 different animals, this means that some of these monkeys (or at least one) should have been, at a given time, infected by more than one foamy viral strain. These different viruses were subsequently disseminated by natural means in other animals of the colony.

Concerning the possibility that the Tonkean macaques may have become infected, prior to arrival at the primate center, with SFVs from other primates species which they are sympatric with, within Sulawesi, it is important to note that the seven different taxa (species or subspecies, according to the different current classifications) of macaques present in Sulawesi are allopatric, which exclude any transmission between two different species in a given region [40]. However, the exact geographical origin of some of the founders is not known with great precision within Sulawesi Island (central or peripheral region of the distribution area of the M. Tonkeana). Thus, we cannot totally exclude a contamination in the wild (prior to the arrival in the primatology center) of some of the founders with a foamy viral strain originating from a related species or subspecies.

By using classical nested PCR methods for the integrase gene, as previously described, we did not find in any of the studied animals, a clear evidence for multiple infections by different foamy viruses. This is based on the following arguments: 1) We always found the same strain in the molecular follow-up of a specific given animal overtime and this even after 12 years of in vivo evolution. 2) In two animals, we sequenced ten clones of a PCR experiment and only one strain was amplified for a given animal. 3) The viruses isolated after cultures of PBMCs were identical to that found in the uncultured PBMCs of the monkeys. However, search for multiple infections in these animals are ongoing, using very sensitive PCR methods as previously described [41].

The presence of different strains of FVs in a colony, as found in the M. tonkeana troop, is not without precedent. Indeed, Schweizer et al. found in a troop of 19 African green monkeys (originating from Kenya), and living together in a monkey house, four different FVs clusters with high homologies (>95%) in the envelope surface domain gene [22]. Between the clusters, the divergences ranged from 3 to 25%, indicating thus that four different strains or subtypes of simian FVs were prevalent in this colony. In another study, Blewett et al., have shown the presence of 2 different FVs (based on pol and LTR sequences) in a colony of baboons [2]. However, these two distinct clades consisted of isolates from yellow and olive baboon and isolates from chacma baboons respectively. Very recently, Jones-Engel et al., found in M. tonkeana the presence of at least 4 different strains of FVs [14]. This observation was based on the analysis of a small fragment of the LTR and some of the observed clades were not clearly supported phylogenetically (low bootstrap values).

The exact modes and timing of SFV transmission in monkeys is unknown although both sexual and oral transmission have been suspected [20, 27]. Furthermore, it is rather difficult to compare our results with those of the very few other published studies with a follow-up as they were performed, as seen above, in different kinds of colony with sometimes the immatures being removed from their mother, either shortly after birth or after few months [2]. Moreover, modes and timing of infection for a specific virus can vary according to the different behaviors of different monkey or apes species studied, as well demonstrated in the case of STLV-1 [42–44]. Here, the possibility to study a free-breeding colony of monkeys with long-term follow-up with both plasma and DNA sequential samples and behavioral investigations provide us a unique work opportunity.

We provided here serological and molecular data arguing against a significant transmission of simian foamy viruses from mother or father to child as well as between siblings.

Concerning a possible sexual acquisition of SFV, it is interesting to note that the seroconversion timing of SFVs strongly contrasted with that found for the herpes B virus, whose primary mode of infection is sexual contact [45]. In this colony of M. tonkeana, most males become positive for herpes B between 2.5 and 6 years of age because they may early start to mount adult females (outside their fertility period), whereas females seroconverted only after puberty, i.e. from 5 years of age (unpublished data from Strasbourg Primatology Center). With regards to foamy viruses, a majority of individuals remained negative until 7 years of age making thus improbable, mounts as a possible way of transmission.

Most cases of seroconversion for foamy viruses occurred when individuals reached adulthood, a period of life that entails an increased likelihood of biting. After 7 years of age, for instance, males entered in competition for access to oestrous females and they occasionally received wounds from their rivals [40]. Indeed, it is clear that in this colony, SFVs seroconversion followed the first important recorded episode of severe bite (with a dermal wound) in 7 out of 10 animals. Concerning the molecular point of view, it is very difficult in case of a severe wound to know exactly who is the animal responsible for the bite. However, in our colony, we can demonstrate the direct transmission of a specific FV strain from a positive to a seronegative animal after an episode of severe bites. All together, these data suggest strongly natural transmission of SFVs via severe bites with contact of saliva from the infected animal to the blood strain of the recipient. However, as viral loads have been shown to be very important determinants for transmission of other primate retroviruses, more conclusive evidence for SFVs transmission routes in primates will require determination of viral loads in different body fluids such as saliva, semen, vaginal lavages and breast milk.

Our findings fit very well with studies demonstrating that SFVs are present in the saliva of infected macaques and baboons [2, 28] and that oral tissues are important site for FV replication in African green monkeys and macaques [27]. Furthermore, it is interesting to note that most of the SFV infections, reported in persons occupationally exposed to non-human primates in zoos or primates centers, have probably been acquired through bites [8–12, 14, 46]. Very recently, natural acquired SFV infections have also been found in few hunters in Cameroon, central Africa [13] and in an ongoing study, we could demonstrated that bites from a monkey or an ape is, in central Africa, a major risk factor for acquiring such SFV infection ([47] and Calattini et al., in preparation).