Tularemia is an emerging zoonosis caused by the Gram-negative bacterium Francisella tularensis, which is able to infect a range of animal species and humans. Human infections occur through contact with animals, ingestion of food, insect bites or exposure to aerosols or water, and may lead to serious disease. F. tularensis may persist in aquatic reservoirs. In the Netherland, no human tularemia cases were notified for over 60 years until in 2011 an endemic patient was diagnosed, followed by 17 cases in the 6 years since. The re-emergence of tularemia could be caused by changes in reservoirs or transmission routes. We performed environmental surveillance of F. tularensis in surface waters in the Netherlands by using two approaches. Firstly, 339 samples were obtained from routine monitoring -not related to tularemia- at 127 locations that were visited between 1 and 8 times in 2015 and 2016. Secondly, sampling efforts were performed after reported tularemia cases (n = 8) among hares or humans in the period 2013-2017. F. tularensis DNA was detected at 17% of randomly selected surface water locations from different parts of the country. At most of these positive locations, DNA was not detected at each time point and levels were very low, but at two locations contamination was clearly higher. From 7 out of the 8 investigated tularemia cases, F. tularensis DNA was detected in at least one surface water sample collected after the case. By using a protocol tailored for amplification of low amounts of environmental DNA, 10 gene targets were sequenced. Presence of F. tularensis subspecies holarctica was confirmed in 4 samples, and in 2 of these, clades B.12 and B.6 were identified. This study shows that for tularemia, information regarding the spatial and temporal distribution of its causative agent could be derived from environmental surveillance of surface waters. Tracking a particular strain in the environment as source of infection is feasible and could be substantiated by genotyping, which was achieved in water samples with only low levels of F. tularemia present. These techniques allow the establishment of a link between tularemia cases and environmental samples without the need for cultivation.

To characterize the inter-epidemic ecology of Francisella tularensis, we surveyed vertebrates and invertebrates for the abundance, spatial distribution, and status of infection at a site in northern California that had evidence of endemic type B tularemia. We collected 2910 mosquitoes, 77 biting flies, 704 ticks, 115 mammals, and 1911 aquatic invertebrates in 2013-2014. Real-time PCR on all mosquitoes, 40 biting flies, 113 aquatic invertebrates, and 650 ticks did not detect F. tularensis DNA. Indirect enzyme linked immunosorbent assay (ELISA) on 109 mammals revealed 2 (of 2, 100%) seropositive feral cats, 1 (of 24, 4.5%) seropositive black rat, and 5 (of 10, 50%) seropositive Virginia opossums. A riparian reserve, ∼1 km from the primate research center, had the highest seroprevalence in mammals and the highest capture success for invertebrate vectors whereas opossums, cats, and ground squirrels in close proximity to the primate center had high seroprevalence and abundant fleas. Well-vegetated regions with standing water appeared to be ideal habitats for biotic components of tularemia enzootic persistence. Mesocarnivores may facilitate the spread of F. tularensis, and high densities of rodents and their fleas may be a mechanism for amplification and spillover.