Gastrointestinal parasite assemblages in wild mammals from two forest localities in Gabon

Résumé fourni par la source

Gastrointestinal parasite assemblages are influenced by host compatibility, exposure, behavior, and environmental conditions; however, baseline data remain limited in Central Africa. We characterized gastrointestinal parasite morphotypes in 279 fecal samples collected from five wild mammal taxa at Lopé National Park and Makatamangoye, Gabon. Parasites were identified using flotation and sedimentation techniques and classified to the lowest morphologically defensible taxonomic level. The analysis included overall infection rates, infracommunity richness, sample-based rarefaction, and presence–absence community composition within localities and between host populations. After excluding unidentified structures, an ectoparasite, and a plant-parasitic nematode, 140/279 samples were tested positive (50.2%; 95% CI 44.2–56.2), resulting in the retention of 25 parasite taxa or morphotypes. Strongylid-type eggs (65 detections), Trichuris spp. (46), Bertiella spp. (24), Ascaris spp. (21) and Ternidens spp. (20) were the most frequently observed. Component-community richness ranged from four taxa in Pan troglodytes to 19 in Potamochoerus porcus . Differences among host populations within localities were apparent for both overall infection and infracommunity richness. Jaccard-based PERMANOVA revealed compositional differences among infected host samples at Lopé (pseudo-F = 2.40, p < 0.001) and Makatamangoye (pseudo-F = 2.24, p = 0.001), although caution is warranted due to heterogeneous dispersion at Lopé. Sørensen dissimilarity between localities within host taxa ranged from 0.33 to 0.73, often driven by nestedness-resultant differences. These findings provide a morphology-based inventory of gastrointestinal parasite assemblages, however, unequal sampling and limited taxonomic resolution prevent causal attribution to habitat disturbance or confirmation of zoonotic transmission. Further molecular characterization and replicated longitudinal sampling are needed.