Nature News · Feb 11, 2026 · Collected from RSS
MainThe recent emergence of MPXV lineages characterized by human-to-human transmission through sexual networks has led the World Health Organization (WHO) to declare public health emergencies of international concern in 2022 (ref. 1) and 2024 (ref. 2). Efforts were immediately scaled up in endemic African countries to reinforce mpox surveillance systems. Concurrently, sustained human-to-human transmission was shown to leave a distinct signature in MPXV genomes identifiable as APOBEC3-induced mutations, providing a tool to determine how much MPXV evolution happened in humans12. Building on these advances, recent genomic surveillance data from the Democratic Republic of Congo (DRC), the Republic of Congo, Nigeria and Cameroon clearly showed that MPXV diversity mostly reflects numerous, independent zoonotic spillovers13,14,15. Importantly, epidemiological data from the DRC suggest that these spillovers increased in frequency from 2010 to 2024, a period during which the national surveillance system has been relatively stable3.Identifying the animal(s) that serve as reservoir(s) for MPXV may help in managing the risk of spillover to humans and preventing subsequent outbreaks fuelled by human-to-human transmission. Following others16, we define a reservoir as a natural host in which the virus can circulate permanently and from which transmission to another host—humans in the case of zoonoses—is possible and documented. According to this definition, no MPXV reservoir has yet been identified. Nevertheless, extensive information has accumulated over the five decades since the virus discovery, pointing to several potentially involved species.MPXV was first isolated from wildlife in 1985, when a Thomas’s rope squirrel (Funisciurus anerythrus) captured in the DRC tested positive7. This finding supported the idea that this and other African rodents are natural hosts, a notion previously supported by sero-epidemiological studies indicating orthopoxvirus (OPV) circulation in this group of mammals. PCR analyses of museum specimens later detected MPXV DNA in five rope squirrel species, including F. anerythrus (45 of 362; 12.4%) and Funisciurus pyrropus (8 of 201; 4%)9. Most recently, a large screening of small mammals in the DRC reported near-complete genomes of MPXV derived from a Thomas’s rope squirrel, another squirrel (Paraxerus sp.) and one soft-furred mouse (Praomys jacksoni)8. Furthermore, spatial overlap analyses between ecological niches of 99 African mammal species and human mpox index cases similarly pointed to squirrels as the most likely source of human infections5 (Extended Data Fig. 1 shows rope squirrel species distribution). These repeated and independent findings suggest that (rope) squirrels may be among the natural hosts of MPXV. However, direct evidence of transmission from these animals (or any other potential natural host) to humans or other hosts is still lacking.By contrast, a plausible chain of events involving African rodents was reconstructed during an MPXV outbreak affecting humans outside the African continent, in the USA in 2003 (ref. 4). All human cases were linked to pet prairie dogs (Cynomys sp.) sold by a single distributor, who had previously housed them with several African rodent species imported through a single shipment from Ghana. Among these, at least three species (giant pouched rats—Cricetomys sp., rope squirrels—Funisciurus sp. and dormice—Graphiurus sp.) tested positive for MPXV6. Subsequent investigations in Ghana found individuals from the same genera either seropositive or PCR-positive for OPV17. Notably, this detailed reconstruction identified only the prairie dogs—an incidental host—as the direct source of human infection, under very specific circumstances that provide little insight into zoonotic transmission in endemic areas. Furthermore, it was not possible to determine the relative contribution of each African rodent species to virus transmission, either among themselves or to prairie dogs. Although these investigations strengthened the suspicion that some African rodents may serve as MPXV reservoirs, none could be formally confirmed.Captive non-human primates (NHP) were associated with the discovery of MPXV18, which resulted in the misleading naming of the virus. More recently, long-term health monitoring at the Taï Chimpanzee Project19 (TCP) in Taï National Park (TNP), Côte d’Ivoire, showed that MPXV also affects wild NHP, opening a window into the ecology of this virus in its original sylvatic environment. We detected MPXV in a dead sooty mangabey (hereafter used interchangably with ‘mangabey’; Cercocebus atys) in 2012 (ref. 20) and later identified three independent MPXV outbreaks in 2017 and 2018 that hit distinct groups of western chimpanzees (Pan troglodytes verus) living in the same forest21. These studies provided insights into different clinical presentations and described viral genomic diversity in this area, but the source of these outbreaks remains unknown. In late January 2023, we started observing clinical signs compatible with MPXV infection in several infants from the habituated mangabey group, which consisted of 80 individuals at that time (Fig. 1). Sooty mangabeys form stable, cohesive, matrilocal social groups and nearly 30% of the study group developed skin lesions over the following 12 weeks. Making use of the longitudinal non-invasive sample collection started in the early 2010s22, along with the systematic carcass monitoring23,24 and rodent sampling efforts in and around TNP, we conducted an outbreak investigation that included an extensive search for a potential source of infection.Fig. 1: Fire-footed rope squirrel range and map of the sooty mangabey territory in TNP.a, Distribution of the fire-footed rope squirrel in Africa (striped red) according to IUCN redlist32 and the location of TNP in Côte d’Ivoire for reference (green). b, Home range of the habituated sooty mangabey group at TCP, approximated as the 95% minimum convex polygon (MCP) (dark blue) line of the mangabey movements tracked over the period of a year (light blue). The squirrel icon represents the location of the necropsy of the fire-footed rope squirrel which was found positive for MPXV. Fire-footed rope squirrels are territorial, with typical home ranges of a few hectares. Range sizes33: male F. pyrropus 5.2 ha, subadult female 2.3 ha, lactating female 1 ha. This squirrel home range would not have overlapped with that of the sooty mangabey study group. Map in a adapted from OpenStreetMap (https://www.openstreetmap.org), reproduced under a Creative Commons CC BY-SA 2.0 licence. The dead squirrel icon in b was created using Inkscape.Full size imageOn 27 January 2023, an infant mangabey developed red macular lesions on the forehead, back of the head, chest and legs (Fig. 2a), accompanied by the onset of lethargy and anorexia. Lesions quickly spread to the entire body and the individual died within 48 h, on 29 January. By early March, five other infants developed similar lesions alongside lethargy, anorexia and lymphadenopathy. Macular skin lesions progressed to papulopustular stages and three of these infants died. A milder form of the disease, consisting of either a diffuse rash with only about 5–20 skin lesions (Fig. 2b), or fewer isolated lesions appearing in a single part of the body (for example, face, limbs or tail; Fig. 2c), affected 20 other mangabeys of all age groups (Extended Data Table 1). In all affected animals, papulopustular lesions evolved to crusts and ultimately scabs (Extended Data Fig. 2). Overall, the disease swept through the group until the end of April 2023, resulting in 26 out of 80 (32.5%) mangabeys developing at least one visible skin lesion and four deaths. Trained veterinarians wearing a complete set of personal protective equipment and following strict biosafety protocols25 performed on-site necropsies on three of the four infants. The body of the fourth infant was never found.Fig. 2: Different degrees of maculopustular rash associated with MPXV infection in three sooty mangabeys.a, Severe maculopustular rash (more than 20 lesions) spread all over the body. b, Moderate maculopustular rash (5–20 lesions) localized on the head. c, Mild maculopustular rash (1–5 lesions) localized on the neck. White arrows point at the observed skin lesions. Photo by Taï Chimpanzee Project/Carme Riutord-Fe.Full size imageTo confirm infection with MPXV, we first tested necropsy samples from the three infants and identified viral DNA in all main organs (Supplementary Table 1). We then performed a group-wide outbreak investigation by analysing 170 faecal samples collected from the mangabeys during the outbreak window, defined as the period in which clinical signs were visible in the group (Extended Data Table 2 and Supplementary Table 2). We detected MPXV DNA in 36 faecal samples collected from 19 individuals (7 symptomatic and 12 asymptomatic; Fig. 3). Of these 19, 14 were mothers of symptomatic infants and only 6 of them developed lesions (Supplementary Table 3a,b). We did not detect MPXV in 89 faecal samples collected after clinical signs resolved. These findings show that MPXV caused disease in a large proportion of this group and may have infected an even larger pool of individuals subclinically, consistent with earlier observations in chimpanzees from TNP21.Fig. 3: Detection of MPXV DNA and of fire-footed rope squirrel DNA in sooty mangabey faeces.Each line represents a different individual. Filled circles represent virus-positive samples, empty circles represent virus-negative samples. The solid red circle indicates the detection of squirrel DNA in a sample that was also MPXV positive and the unfilled red circle represents the detection of squirrel DNA in a sample that was MPXV negative. Vertical grey shadowing indicates the period in which clinical signs were observed in the mangabey group. For each individual, the blue horizontal shadowing indicates the period during