Abstract
Echinococcus multilocularis is a zoonotic cestode that can infect wildlife, domestic animals, and humans. In humans, infection with the larval stage of the parasite causes the disease alveolar echinococcosis, which can be fatal if left untreated. Surveillance for the parasite in New York State occurred during the 2021–2022 coyote (Canis latrans) hunting season. Fecal samples and the gastrointestinal tracts (GIT) from 43 coyote carcasses were collected from hunters and trappers across 8 counties. Fecal samples were screened for E. multilocularis DNA using a multiplex PCR. Three samples tested positive for E. multilocularis DNA. Subsequently, adult cestodes were collected from GIT samples using the sedimentation, filtration, and counting technique. Phylogenetic analysis of DNA sequences from the nad2 and cob genes from individual worms indicated these New York sequences cluster with E. multilocularis sequences from Europe. This is the first report of adult E. multilocularis cestodes in New York State, as well as the first detection of the European haplotype of E. multilocularis in wildlife in the northeastern United States.
Echinococcosis is a World Health Organization (WHO) designated neglected zoonotic disease caused by infection with Echinococcus species metacestode parasites (WHO, 2011). Echinococcus species are obligate parasites of mammals, infecting wild canids as definitive hosts with a variety of species-specific intermediate hosts. Two species within this genus are of veterinary and public health concern, Echinococcus granulosus and Echinococcus multilocularis, for their role as the etiologic agents for cystic echinococcosis and alveolar echinococcosis, respectively. Alveolar echinococcosis is the more severe disease of the two, with a fatality rate of over 90% if left untreated due to the growth and metastasis of the larval cysts (Eckert and Deplazes, 2004).
In North America, a history of underreporting and misdiagnosis of the disease has resulted in knowledge gaps around the epidemiology and transmission of the parasites between wildlife, domestic animals, and humans (Budke et al., 2009). The recorded range of E. multilocularis in North America was divided into 2 geographic regions; the Northern Tundra Zone (NTZ), which consisted of coastal Alaska and the Canadian Arctic, and the North Central Region (NCR), which spans central Canada and into the U.S. Midwest (Eckert and Deplazes, 2004; Davidson et al., 2012). Limited surveillance for the parasite in wildlife in these endemic regions and few human echinococcosis cases outside of Alaska contributed to a lack of awareness and research of E. multilocularis for decades.
Recently, surveillance for E. multilocularis has increased following reports of the parasite in domestic and wild canids in novel areas in Canada, revealing a large range expansion, as well as changes in the parasite’s prevalence and epidemiology in endemic and newly colonized areas. Echinococcus multilocularis was detected in wolves in Canada outside the NTZ and NCR boundaries, indicating that the parasite’s range is no longer restricted to the 2 discrete geographic locations (Schurer et al., 2014, 2016). Increased prevalence of E. multilocularis in and outside the NCR has been reported in urban coyotes, wild canid populations, and domestic animals in Alberta, Saskatchewan, and British Columbia, Canada (Catalano et al., 2012; Jenkins et al., 2012; Gesy et al., 2014; Liccioli et al., 2014). Eastward expansion in Canada has been well documented in Ontario, where high prevalence has been seen in urban coyotes, and alveolar echinococcosis cases have been reported in domestic animals and humans for the first time (Skelding et al., 2014; Peregrine, 2015; Kotwa et al., 2019; Massolo et al., 2019).
The introduction of the European haplotype of E. multilocularis may be impacting the observed range expansion in North America, as evidenced by the continued detection of the non-native haplotype in novel locations in the United States and Canada. There are 2 native mitochondrial haplotypes in North America, N1 and N2, each correlating to one of the 2 endemic regions, the NTZ and NCR (Nakao et al., 2009). The North American haplotypes are genetically distinct from the E. multilocularis haplotypes in Europe (E1–E5) and are suspected to be less pathogenic than the European variant of the parasite (Nakao et al., 2009; Jenkins et al., 2012). The first detection of the European haplotype of E. multilocularis in North America occurred in British Columbia, Canada, in 2009 when a domestic dog was treated for infection with the larval stage of the parasite (Jenkins et al., 2012; Peregrine et al., 2012). The European variant has since been documented in wildlife and domestic animals across Canada (Gesy et al., 2014; Kotwa et al., 2019; Santa et al., 2021, 2023). The establishment and expansion of the European haplotype in North America has already resulted in increased cases of alveolar echinococcosis in human patients in Canada and the United States, suggesting a resurgence of this disease in North America (Massolo et al., 2019; Polish et al., 2021; Schurer et al., 2021).
Recent studies of prevalence and distribution of E. multilocularis in wildlife in Ontario exemplify the changing epidemiology of the parasite. Ontario was considered free of E. multilocularis until 2012 when alveolar echinococcosis was diagnosed in several domestic and exotic animals (Kotwa et al., 2019). Since then, E. multilocularis prevalence in wild canids was found to be 23%, even in areas of dense human population, indicating a risk for spillover to humans (Kotwa et al., 2019). This prevalence was comparable to highly endemic regions for the parasite in Canada, despite the relatively recent introduction of E. multilocularis to this area (Catalano et al., 2012). Range expansion of E. multilocularis into eastern North America is likely driven by the movement of infected coyotes, which is supported by the increasing commonality of coyotes as definitive hosts in newly affected areas. Echinococcus multilocularis was discovered in the northeastern United States for the first time in 2018, during a passive surveillance study of coyote (Canis latrans) parasites in Fort Drum, New York (C. M. Whipps, unpubl. data). The migration of infected coyotes from Ontario into northern New York could have initiated the establishment of the parasite in wildlife. The spread of human alveolar echinococcosis cases in Canada followed the spread of the European variant of E. multilocularis, and for this reason, we aimed to identify the distribution and genetic character of E. multilocularis across New York State to better inform public health and disease control measures.
In this study, hunters and trappers provided carcasses of coyotes harvested in New York for non-research purposes over the winter of 2021–2022. The collection of the carcass was dependent on receiving geographic data on where the animal was harvested from the submitter. A fecal sample and the entire gastrointestinal tract (GIT) were collected from each carcass and frozen at −80 C for 3–5 days per WHO safety guidelines (https://apps.who.int/iris/handle/10665/42427) to deactivate any parasite eggs, then kept at −20 C until analysis (Eckert et al., 2001). We collected fecal and GIT samples from 43 coyotes across 8 counties (Ontario, Sullivan, Dutchess, Ulster, Rensselaer, Delaware, Orange, Montgomery) in New York State.
We extracted DNA from fecal samples (n = 43) using Zymo ZR Fecal DNA Miniprep Protocol (Zymo Research Corporation, Irvine, California). To screen for Echinococcus species DNA in each fecal sample, we performed a multiplex PCR test described by Trachsel et al. (2007), which amplifies and differentiates Echinococcus granulosus, Echinococcus multilocularis, and Taenia spp. Fecal samples positive for E. multilocularis DNA had the corresponding GIT sample thawed for parasite collection. Three of the 43 fecal samples (6.98%) tested positive for E. multilocularis. Adult E. multilocularis cestodes were collected from each positive GIT using the sedimentation, filtration, and counting technique (Gesy et al., 2013). All cestodes in each GIT were preserved in 30 ml 70–100% ethyl alcohol for population estimation and genetic analysis. Parasite abundance was estimated from approximately 10% of the worm sample from each GIT by counting the worms in 3 ml of each 30 ml sample. Parasite abundance estimates ranged from 810 to 6,410 adult cestodes in the GIT samples.
To confirm the results of the multiplex PCR test, PCR products were sequenced from any E. multilocularis–positive samples using the multiplex primers above, which target the E. multilocularis mitochondrial NADH dehydrogenase subunit 1 (nad1) gene. Amplified products were purified using the E.Z.N.A.® Cycle Pure Kit (Omega Bio-Tek, Norcross, Georgia) and sequenced using the Cest1 forward primer at the DNA Analysis Facility on Science Hill at Yale University (New Haven, Connecticut). Reactions were performed using BigDye™ Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Thermo Fisher Scientific, Inc., Waltham, Massachusetts), with dye terminator removal using Performa® DTR Ultra 96-Well Plates (Edge Biosystems, San Jose, California). Capillary electrophoresis was performed on a 3730xl DNA Analyzer (Applied Biosystems, Thermo Fischer Scientific, Inc.) with a 96-capillary 50 cm array, using default instrument protocols.
To further genetically characterize the cestodes for phylogenetic analyses, we extracted DNA from 6 worms collected from each GIT using the Qiagen DNeasy Blood and Tissue Kit (Qiagen Inc., Valencia, California). Only a subset of worms from each GIT was sequenced due to the high degree of genetic similarity of the sequenced fecal DNA between all positive samples. We also evaluated additional mtDNA genes. Here PCR was performed using primers designed by Nakao et al. (2009) for the NADH dehydrogenase subunit 2 (nad2) gene and Gesy and Jenkins (2015) for the cytochrome b (cob) gene. PCR products were sequenced as above, using corresponding forward primers for each gene. A 527 bp region from the nad2 gene and 551 bp region from cob gene were aligned and concatenated for each worm that was sequenced for haplotype comparison using Clustal Omega and Unipro UGENE software (Okonechnikov et al., 2012; Madeira et al., 2022). The resulting sequences were compared to aligned and concatenated sequences of the nad2 and cob genes from described haplotypes from Asia, Europe, and North America published on GenBank (Table I). Maximum likelihood phylogenetic analyses of these concatenated sequences were conducted using IQ-Tree (Nguyen et al., 2015; Hoang et al., 2018). Taenia saginata was used as an outgroup. The optimal substitution model for our analysis was determined within IQ-Tree as HKY+F+I. Bootstrap support values were calculated with 1,000 replicates.
For our 3 PCR positive fecal samples, the resulting 290 bp sequences of the nad1 gene (GenBank OQ606770–OQ606772) were 100% identical to E. multilocularis sequences from eastern Europe (GenBank MN444805) and European strains found in North America (GenBank KC848475). The DNA sequences from the nad2 (GenBank OP596325–OP596327) gene from the New York samples were identical to one another, as were the New York cob gene sequences (GenBank OP596328–OP596330). We used representative sequences of the individual worms collected from GIT sample (E316, E320, E321) for our phylogenetic analysis. Our analysis placed the DNA sequences from our New York samples in a well-supported lineage containing DNA sequences from samples from Poland (GenBank KY205705 and KY205675), Slovakia (GenBank AB461405 and AB461397), and European haplotypes detected in Virginia (GenBank OK268249 and OK268251; Fig. 1). Our sequences for both the nad2 and cob genes had 100% similarity to the respective genes from confirmed European haplotype sequences.
Detection of the European haplotype of E. multilocularis in New York State supports the growing evidence for a resurgence of this parasite within North America and a novel expansion in the northeastern United States. The coyotes that were infected with E. multilocularis were harvested in Dutchess and Montgomery County, in southern and central New York State, respectively. Given that E. multilocularis was first detected in northern Jefferson County in 2018, it is likely that the range of E. multilocularis now extends across the state. The recent identification of the European haplotype of E. multilocularis in wildlife in Virginia (Polish et al., 2022) suggests that the parasite may be expanding its range across the eastern United States.
The rising prevalence of coyotes as definitive hosts for E. multilocularis increases the risk of parasite spillover from wildlife to domestic animals and humans. Coyotes efficiently utilize urban and semi-urban spaces, which can facilitate more frequent contact with humans and domestic animals. Infected coyotes can contaminate the environment with E. multilocularis eggs, which can result in an established parasite population in a new area, in the presence of a suitable rodent intermediate host, as well as provide an avenue for spillover. New York has a large and widespread coyote population, as well as an active statewide community of hunters and trappers that regularly interact with this species throughout the 6-mo-long hunting season. The extended hunting season, lack of bag limits on coyotes, and culture of coyote hunting competitions in the state increase the exposure of humans to potentially infected coyotes.
The introduction of this zoonotic parasite into a previously unaffected region presents serious health risks to wildlife, domestic animals, and humans. This risk will be exacerbated by the public and clinical lack of knowledge about the parasite’s biology, distribution, and risk factors for infection. As the complete range of E. multilocularis within the northeast is currently unknown, widespread public health outreach to hunters, trappers, wildlife rehabilitators, and veterinary workers represents the best course of action to raise awareness for the parasite and safety procedures that can prevent spillover from wildlife to humans.
We would like to thank the SUNY Center for Applied Microbiology for the partial funding of this work. We would also like to thank M. Stevens, S. Meisner, and M. Jackling for their invaluable support and assistance in sample collection. Thank you to the New York hunters and trappers for their help in sample collection.