Contact networks of small mammals highlight potential transmission foci of Lassa mammarenavirus

Introduction Lassa fever, caused by Mammarenavirus lassaense (LASV), is an endemic zoonosis in several West African countries. Human infections primarily arise from rodent-to-human transmission, with Mastomys natalensis, a synanthropic species, serving as the principal reservoir. In Sierra Leone, small-mammal communities vary across land use gradients, potentially shaping LASV transmission risk in human populations. However, the extent to which anthropogenic environments facilitate small-mammal interactions conducive to viral transmission remains poorly understood. Methods We conducted small-mammal sampling over 43,266 trap nights, detecting 684 rodents and shrews in the LASV-endemic Eastern Province of Sierra Leone. To assess potential for within- and between-species transmission, we constructed space-sharing networks based on co-trapping within species-specific radii informed by home range estimates. These networks approximated shared space use rather than direct interaction, allowing comparison of potential encounter patterns across habitats. Results Small-mammal communities in agricultural settings had greater species richness and were more connected than those in villages and forests, although overall space-sharing rates did not differ substantially. Network topology varied by land use, with greater modularity in village networks. Notably, space sharing among M. natalensis individuals was more frequent in agricultural settings than in villages, suggesting that land use may modulate opportunities for intra-specific viral transmission. LASV seroprevalence across the small-mammal community was 5.7%, with antibodies detected in nine species. We found no major differences in seroprevalence by land use or network complexity. Conclusion These findings underscore the importance of cross-habitat surveillance to understand host ecology and LASV transmission. Accounting for species-specific space use and habitat-dependent interaction structures is crucial for identifying key hosts and spillover settings.