Detection of Toxoplasma gondii in wild bivalves from the Kerguelen and Galapagos archipelagos: influence of proximity to cat populations, exposure to marine currents and kelp density

[Display omitted] •Toxoplasma gondii DNA was detected in bivalves from the Kerguelen and Galapagos archipelagos.•Its occurrence differed between sampling sites in Kerguelen but not in Galapagos.•In both archipelagos, quantitative PCR-positive bivalves were found on islands without cats.•Our finding demonstrated the waterborne transport of oocysts over at least 20 km.•Exposure to currents and giant kelp density affect T. gondii detection in bivalves. Oocysts of the protozoan Toxoplasma gondii are found in felid feces and can be washed into coastal waters, where they persist for months, attaching to algae and accumulating in invertebrates. We used wild bivalves to assess contamination of coastal waters of the Kerguelen and Galapagos archipelagos by this zoonotic parasite. Additionally, we leveraged the contrasting situations of these archipelagos to identify some potential drivers of contamination. In the Galapagos, with a cat density reaching 142 per km2, 15.38% of the sampled oysters (Saccostrea palmula) tested positive for T. gondii by quantitative real-time PCR (qPCR) (n = 260), and positive samples were found in all eight sampling sites. In Kerguelen, with 1-3 cats per km2, 40.83% of 120 tested mussels (Mytilus edulis platensis) were positive, and positive samples were found in four out of the five sampling sites. These findings provide evidence of T. gondii contamination in the coastal waters of these archipelagos. Furthermore, T. gondii-positive bivalves were found on islands located 20 km away (Galapagos) and 5 km away (Kerguelen) from the nearest cat population, indicating that T. gondii oocysts can disperse through waterborne mechanisms over several kilometers from their initial deposition site. In the Galapagos, where runoff is infrequent and all sites are exposed to currents, the prevalence of qPCR-positive bivalves did not show significant variations between sites (p = 0.107). In Kerguelen where runoff is frequent and site exposure variable, the prevalence varied significantly (p