Introduction
The Taeniidae family comprises tapeworms that are causative agents of significant zoonotic diseases in humans, such as echinococcosis and taeniasis. Echinococcosis is caused by the larval stage of species in the genus Echinococcus. Domestic and wild carnivores act as definitive hosts (DHs), whereas the intermediate hosts (IHs) comprise a wide range of mammals, including ungulates, rodents and lagomorphs, as well as humans as accidental hosts. Adult worms reside within the intestinal tract of the DHs and lay eggs in feces, thereby contaminating the environment. The IHs are infected after accidental ingestion of these eggs, and the DHs after ingestion of infected organs of IHs (Thompson, Reference Thompson2017; Romig and Wassermann, Reference Romig and Wassermann2024). Echinococcus granulosus sensu lato (s.l.) and E. multilocularis, which cause human cystic echinococcosis (CE) and alveolar echinococcosis (AE), respectively, are among the most important zoonotic cestodes worldwide (Deplazes et al., Reference Deplazes, Rinaldi, Alvarez Rojas, Torgerson, Harandi, Romig, Antolova, Schurer, Lahmar, Cringoli, Magambo, Thompson and Jenkins2017; Lundström-Stadelmann et al., Reference Lundström-Stadelmann, Rostami, Frey, Torgerson, Riahi, Bagheri, Kaethner, Lachenmayer, Beldi, Gasser and Hemphill2025; Robertson et al., Reference Robertson, Minato, Devleesschauwer, Di Bari, Vaes, Keddy, Sripa, Fernandez, Pires, Majowicz, Lake, Roberts, Nane, Kretzschmar, Mughini-Gras, Rostami, Carabin, Budke, Donadeu and Torgerson2026). Many other taeniid species have a similar life cycle, based on DH infection through ingestion of infected IH organs (Yang et al., Reference Yang, Li, Zhang, Zhao, Sun and Wang2024).
Carnivore-transmitted taeniid species other than Echinococcus spp., such as Taenia hydatigena, T. multiceps and T. ovis, are associated with livestock, particularly sheep. DHs gain access to the infected internal organs of IHs, either naturally or through post-slaughter offals, thereby maintaining the domestic life cycle. Lagomorphs serve as IHs for T. pisiformis and T. serialis, whereas rodents serve as IHs for T. crassiceps and Mesocestoides spp. (Deplazes et al., Reference Deplazes, Eichenberger and Grimm2019; Yang et al., Reference Yang, Li, Zhang, Zhao, Sun and Wang2024). Nevertheless, the life cycle of Mesocestoides spp. differs from that of most other cestodes and is more complex. In this cycle, oribatid mites are the first IH and rodents, amphibians or reptiles are the second. Carnivorous DHs become infected by preying on these second IHs (Deplazes et al., Reference Deplazes, Eckert, Mathis, Samson-Himmelstjerna and Zahner2016; Saari et al., Reference Saari, Näreaho and Nikander2019).
Echinococcus granulosus s.l. is a complex of parasite species whose life cycle is predominantly associated with domestic transmission involving livestock (Romig et al., Reference Romig, Deplazes, Jenkins, Giraudoux, Massolo, Craig, Wassermann, Takahashi and de la Rue2017; Thompson, Reference Thompson2017). The species of the complex differ in their main IHs, with E. granulosus sensu stricto (s.s.) in sheep, E. equinus in equids, E. ortleppi in cattle and E. canadensis in camels and pigs (G6/G7) or in cervids (G8/G10) (Vuitton et al., Reference Vuitton, McManus, Rogan, Romig, Gottstein, Naidich, Tuxun, Wen and Menezes da Silva2020; Massolo et al., Reference Massolo, Simoncini and Romig2022). The life cycle of the parasite E. multilocularis is based on the predator-prey relationship between foxes and small rodents. In some areas, such as Asia, dogs may also play an important role as DHs (Lempereur et al., Reference Lempereur, Nijsse, Losson, Marechal, De Volder, Schoormans, Martinelle, Danlois and Claerebout2020; Toews et al., Reference Toews, Musiani, Checkley, Visscher and Massolo2021). The microcyst stage, which causes AE, develops mainly in the liver of small mammals (Vuitton et al., Reference Vuitton, McManus, Rogan, Romig, Gottstein, Naidich, Tuxun, Wen and Menezes da Silva2020).
The E. granulosus s.l. complex is essentially cosmopolitan, whereas E. multilocularis is restricted to the Holarctic in the northern hemisphere, but its known range has expanded over the past decades through spreading in Europe and improved surveillance, so that the parasite is now found in regions previously thought free of infection (Baumann et al., Reference Baumann, Shi, Liu, Bao, Schmidberger, Kratzer, Li, Barth, Baumann, Bloehdorn, Fischer, Graeter, Graf, Gruener, Henne-Bruns, Hillenbrand, Kaltenbach, Kern, Kern, Klein, Kratzer, Ehteshami, Schlingeloff, Schmidberger, Shi, Staehelin, Theis, Verbitskiy and Zarour2019; Umhang et al., Reference Umhang, Bastid, Avcioglu, Bagrade, Bujanić, Bjelić Čabrilo, Casulli, Dorny, van der Giessen, Guven, Harna, Karamon, Kharchenko, Knapp, Kolarova, Konyaev, Laurimaa, Losch, Miljević, Miterpáková, Moks, Romig, Saarma, Šnábel, Sreter, Valdmann and Boué2021; Romig and Wassermann, Reference Romig and Wassermann2024). Humans can act as dead-end hosts for both parasites. Human infection follows the ingestion of water or food contaminated by Echinococcus eggs, or hand-to-mouth contamination, notably after contact with infected DHs or with a contaminated environment (Tamarozzi et al., Reference Tamarozzi, Legnardi, Fittipaldo, Drigo and Cassini2020; Torgerson et al., Reference Torgerson, Robertson, Enemark, Foehr, van der Giessen, Kapel, Klun and Trevisan2020; Casulli and Tamarozzi, Reference Casulli and Tamarozzi2021). CE is a slowly progressive disease where cysts can cause organ compression, rupture and anaphylaxis, while AE behaves as a tumour-like hepatic lesion that is fatal without treatment. More than a million people worldwide live with these diseases. Echinococcosis causes about 19 300 deaths a year and a total loss of 871 000 disability-adjusted life years, 183 573 due to CE and 687 823 to AE (Torgerson et al., Reference Torgerson, Devleesschauwer, Praet, Speybroeck, Willingham, Kasuga, Rokni, Zhou, Fèvre, Sripa, Gargouri, Fürst, Budke, Carabin, Kirk, Angulo, Havelaar and de Silva2015). The economic burden of CE is estimated at US$3 billion per year (Torgerson et al., Reference Torgerson, Keller, Magnotta and Ragland2010; WHO, 2017), and the most recent World Health Organization WHO synthesis indicates that the global burden of foodborne parasitic diseases, including echinococcosis, has not substantially decreased over the last 2 decades (Robertson et al., Reference Robertson, Minato, Devleesschauwer, Di Bari, Vaes, Keddy, Sripa, Fernandez, Pires, Majowicz, Lake, Roberts, Nane, Kretzschmar, Mughini-Gras, Rostami, Carabin, Budke, Donadeu and Torgerson2026). CE is endemic in more than 100 countries and is particularly prevalent in pastoral and rural areas of low- and middle-income countries, whereas AE predominates in Asia, particularly China, as well as in eastern and central Europe, Türkiye, Iran, Russia, northern Japan and the Arctic tundra of Alaska (Altintas, Reference Altintas2008; Baumann et al., Reference Baumann, Shi, Liu, Bao, Schmidberger, Kratzer, Li, Barth, Baumann, Bloehdorn, Fischer, Graeter, Graf, Gruener, Henne-Bruns, Hillenbrand, Kaltenbach, Kern, Kern, Klein, Kratzer, Ehteshami, Schlingeloff, Schmidberger, Shi, Staehelin, Theis, Verbitskiy and Zarour2019; Lundström-Stadelmann et al., Reference Lundström-Stadelmann, Rostami, Frey, Torgerson, Riahi, Bagheri, Kaethner, Lachenmayer, Beldi, Gasser and Hemphill2025).
Human CE and AE have a high prevalence in Türkiye. The number of CE cases rose from 408 in 2008 to 1867 by the end of 2019, while for AE, 641-918 confirmed cases were reported between 1939 and 2018 (Torgerson et al., Reference Torgerson, Keller, Magnotta and Ragland2010; Deplazes et al., Reference Deplazes, Rinaldi, Alvarez Rojas, Torgerson, Harandi, Romig, Antolova, Schurer, Lahmar, Cringoli, Magambo, Thompson and Jenkins2017; Altintas et al., Reference Altintas, Karamil, Turkum, Akil, Sakarya, Bozkaya, Šoltys and Altintas2020; Lundström-Stadelmann et al., Reference Lundström-Stadelmann, Rostami, Frey, Torgerson, Riahi, Bagheri, Kaethner, Lachenmayer, Beldi, Gasser and Hemphill2025). Most human cases, particularly concerning AE, occur in the north-eastern regions of the country (Altintas, Reference Altintas2008; Gürler et al., Reference Gürler, Bölükbaş, Açıcı and Umur2019). Molecular epidemiological studies have increased on animal hosts in the country. These studies report CE in various IHs, mostly due to E. granulosus s.s. (Simsek et al., Reference Simsek, Balkaya and Koroglu2010; Simsek and Cevik, Reference Simsek and Cevik2014; Kesik et al., Reference Kesik, Simsek, Kilinc and Koroglu2019; Avcioglu et al., Reference Avcioglu, Bia, Balkaya, Kirman, Akyuz and Guven2022; Selcuk et al., Reference Selcuk, Aslan Celik, Celik, Celik, Ercan, Uslug, Tekin and Simsek2024). A series of studies conducted on free-roaming dogs between 2000 and 2020 reported E. granulosus s.l. positivity rates ranging from 0.8 to 14% (Kuru et al., Reference Kuru, Aypak and Aysul2013; Öge et al., Reference Öge, Öge, Gönenç, Sarımehmetoğlu and Özbakış2017; Oge et al., Reference Oge, Oge, Ozbakis and Gurcan2017; Acıöz et al., Reference Acıöz, Çeliksöz, Özçelik, Değerli, Öztop and Malatyalı2018; Oguz et al., Reference Oguz, Ozdal, Kilinc and Serdar Deger2018). These figures are not directly comparable, since Guzel et al. (Reference Guzel, Yaman, Koltas, Demirkazik and Aktas2008) used a copro-ELISA and Öter et al. (Reference Öter, Bilgin, Tınar and Tüzer2011) used a conventional coprological examination, neither of which distinguishes between co-circulating Echinococcus and Taenia species, whereas the remaining studies used PCR-based methods allowing species-level differentiation. Recently, a positivity rate of 14.1% (63/446) was reported from the north-eastern province of Erzurum. The identified species comprise E. granulosus s.s., E. equinus, E. ortleppi, E. canadensis (G6/G7) and E. multilocularis (Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Akyuz, Bia, Gulbeyen and Yaya2021b). Moreover, E. multilocularis was reported in the region in red foxes (Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Akyuz, Bia, Gulbeyen and Yaya2021a), rodents (Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Bia, Gülbeyen, Kurt, Yaya and Demirtaş2017), a Eurasian lynx (Avcioglu et al., Reference Avcioglu, Guven, Balkaya and Kirman2018) and a grey wolf (Akyuz et al., Reference Akyuz, Kirman, Guven, Balkaya and Avcioglu2024). The regional prevalence of E. multilocularis in the north-eastern region was recently estimated at 8.7% in free-roaming dogs using a sieving/flotation method followed by classical PCR (S/F-cPCR) (Akyuz and Avcioglu, Reference Akyuz and Avcioglu2025).
Apart from the surveys mentioned above, Echinococcus infection in free-roaming dogs acting as DHs has not been further investigated in the north-eastern region of Türkiye. In this region, a considerable proportion of dogs roam freely within urban areas, are unowned and have uncontrolled movements, and they can also come into contact with wildlife in suburban areas; red foxes, in turn, access urban areas in search of human-related food and are able to adapt to human settlements (Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Akyuz, Bia, Gulbeyen and Yaya2021a). Where the environment is contaminated by taeniid eggs shed by both foxes and dogs, the potential role of free-roaming dogs in human infection deserves particular consideration, due to their closer contact with humans (Toews et al., Reference Toews, Musiani, Checkley, Visscher and Massolo2021).
High-sensitivity molecular methods are vital for accurate diagnosis, designing control and elimination programmes and assessing the risk of echinococcosis in endemic regions (Hunt, Reference Hunt2011; Pomari et al., Reference Pomari, Piubelli, Perandin and Bisoffi2019; Bagó et al., Reference Bagó, Hoelzl, Knauer, Kübber-Heiss and Smith2021). Copro-DNA methods are valuable in co-endemic regions, detecting infection throughout the prepatent and patent periods without relying solely on eggs, and identifying minimal DNA (around 1 copy μL−1) for E. granulosus s.l. (Maksimov et al., Reference Maksimov, Bergmann, Wassermann, Romig, Gottstein, Casulli and Conraths2020). The probe-based approach enhances the reliability of species-specific detection (Isaksson et al., Reference Isaksson, Hagström, Armua-Fernandez, Wahlström, Ågren, Miller, Holmberg, Lukacs, Casulli, Deplazes and Juremalm2014).
The present study had 2 objectives. The first was to evaluate the occurrence of E. granulosus s.l., E. multilocularis, Taenia spp. and Mesocestoides spp. in free-roaming dogs across 7 provinces of north-eastern Türkiye, using probe-based copro-real-time PCR (copro-qPCR) for the Echinococcus species and classical PCR performed directly on copro-DNA (copro-cPCR) followed by sequencing for Taenia and Mesocestoides species, in order to clarify the role of the DH in the transmission of these parasites. The second was to compare, on the same samples, the detection rates and the level of agreement obtained with different PCR-based approaches: copro-qPCR vs S/F-cPCR (taeniid type egg DNA) for E. multilocularis, and copro-qPCR vs copro-cPCR for Taenia spp. Because no independent reference standard could be applied to these animals, this comparison is deliberately restricted to detection rates and agreement, and no estimate of diagnostic sensitivity is derived. These findings aim to advance the understanding of Echinococcus transmission in north-eastern Türkiye, while identifying reliable fecal-based diagnostic tools where AE and CE are co-endemic.
Materials and methods
Study area
This study was conducted in 7 north-eastern provinces of Türkiye: Ağrı, Ardahan, Bayburt, Erzincan, Erzurum, Iğdır and Kars (Figure 1). This region is located between 2000 and 2200 meters above sea level, has a continental climate and is the highest region in Türkiye (Çelik et al., Reference Çelik, Kopar and Bayram2018). This region has the largest agricultural population of the country, where livestock farming is the main source of income (OECD, 2016). Although the 7 provinces are located in the same region, there are differences in their landscape, climatic conditions and livestock production capacity (Supplementary Table S1). The values in Supplementary Table S1 are long-term provincial normals from the Turkish State Meteorological Service (Turkish State Meteorological Service, 2025), and livestock numbers are those reported by the Republic of Türkiye Ministry of Agriculture and Forestry (Ministry of Agriculture and Forestry of the Republic of Türkiye, 2025). Farm animal density (animals per km2) was calculated by the authors as the total number of cattle, sheep and goats per province divided by the provincial surface area.
The study area, north-eastern region of Türkiye.
There is a large population of free-roaming dogs in this study area. In the present study, the term ‘free-roaming dog’ refers to ownerless dogs living without confinement in urban, suburban and rural areas. These dogs have adapted to life on the streets and feed on food found in refuse, by hunting small mammals, or by scavenging the carcasses of dead animals. The region is served by at least 1 animal shelter in every province and the dogs are collected from the city centre and suburbs at regular intervals by municipal authorities. The animals sampled in this study were therefore not permanently housed shelter dogs, but dogs caught while roaming and sampled shortly after capture, so that their parasitological status reflects exposure acquired while roaming freely rather than conditions within the shelter. A proportion of veterinary procedures (e.g. sterilization, anti-parasitic treatment, vaccination) are performed on sheltered dogs. However, the vast majority of dogs are not part of this cohort in this region due to the limited capacity of the shelters, the number of staff, and the uncontrolled mobility of dogs in the region. Each province contains at least 1 slaughterhouse, typically situated in close proximity to the city centre. These establishments are constructed in a manner that is not isolated from their immediate surroundings, and it is possible for a dog to roam around the facilities. The dogs in the region are known to roam freely in the city centre, its suburbs and rural areas, with uncontrolled reproduction and movement.
The red foxes (Vulpes vulpes) are widely distributed across a large geographical area in Türkiye. Limited data are available on population density, with estimates ranging from 0.07 to 4.0 individuals per km2 depending on the region (Albayrak et al., Reference Albayrak, Giannatos and Kabasakal2012; Ambarlı et al., Reference Ambarlı, Ertürk and Soyumert2016). Despite the absence of precise demographic data for the north-eastern region, Erzurum is documented as the most endemic area for E. multilocularis in foxes within Türkiye (Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Akyuz, Bia, Gulbeyen and Yaya2021a).
Sampling
The fecal samples of free-roaming dogs had previously been collected by Akyuz and Avcioglu (Reference Akyuz and Avcioglu2025) between June 2019 and September 2020 from municipal animal shelters in the provinces of the north-eastern region; each sample was collected from an individually caged dog. A total of 494 of the 1069 samples available from the previous study were selected, corresponding to approximately half of the samples from each province, as permitted by the available budget. Samples were drawn using random numbers, independently of the earlier taeniid egg and E. multilocularis results. Available and selected sample numbers per province are shown in Table 2. Among the 494 samples, 133 (26.9%) had previously tested positive for taeniid eggs by sieving/flotation and 49 (9.9%) for E. multilocularis by S/F-cPCR.
Extraction of nucleic acids
Approximately 300 mg of each dog fecal sample was used for the extraction of copro-DNA. Each fecal sample was mixed in 2 mL Matrix E tubes (MP Biomedicals, LLC, Santa Ana, CA, USA) containing 1.4 mm ceramic spheres (ZrO2 64%, SiO2 33%, specific gravity 4), 0.1 mm silica spheres (specific gravity 2.5), one 4 mm glass bead (SiO2, specific gravity 2.5), and then 1 mL of CTAB buffer solution (Promega, Madison, WI, USA) before being incubated for 5 min at 95 °C in a heater block.
The samples were then bead-beaten 3 times for 60 sec each time at a maximum speed of 5.0 m s−1 using a FastPrep homogenizer (MP Biomedicals, Santa Ana, CA, USA). Subsequently, 40 µL of proteinase K (20 mg mL−1) was added to each sample, vortexed for 15 sec, and incubated at 70 °C for 10 min. The tubes were then centrifuged at 10 000 × g without braking for 5 min to collect the supernatant. The DNA extraction was performed via the Maxwell RSC Pure Food Pathogen Kit (Promega, Madison, WI, USA) in combination with a Maxwell RSC 48 automated instrument (Promega, Madison, WI, USA), according to the kit manual. This kit is routinely used in the Anses laboratory for copro-DNA extraction from carnivore feces and is included in an accredited method (NF 47-600-2) for detection of E. multilocularis DNA from carnivore feces. Adaptation to fecal material was achieved by the upstream lysis steps described above, and purification followed the manufacturer’s protocol without modification. Copro-DNA extracts were stored at −20 °C until analysis.
During every DNA extraction run, a negative control, consisting of only lysis buffer, was included to verify the absence of contamination throughout the entire process. Additionally, a positive control based on a homogenous fox fecal matrix obtained from a fox experimentally infected by E. multilocularis was used to validate each step of the extraction procedure.
Molecular analyses and sequencing
The detection of E. granulosus s.l. DNA was realized by copro-qPCR analyses based on assays previously published (Maksimov et al., Reference Maksimov, Bergmann, Wassermann, Romig, Gottstein, Casulli and Conraths2020), under standardized conditions in a multiplex format for the simultaneous detection of E. granulosus s.s. and E. canadensis, as well as an internal control using a QuantStudio™ 5 Real-Time PCR (Thermo Fisher Scientific, Waltham, MA, USA) machine. As there is cross-reactivity of E. ortleppi DNA with the E. canadensis assay, all E. canadensis-positive samples were retested with the species-specific E. ortleppi (G5) probe-based assay (Maksimov et al., Reference Maksimov, Bergmann, Wassermann, Romig, Gottstein, Casulli and Conraths2020), and were considered positive for E. canadensis only when this second assay was negative. The detection of E. multilocularis DNA by qPCR was performed in accordance with the methodology of Isaksson et al. (Reference Isaksson, Hagström, Armua-Fernandez, Wahlström, Ågren, Miller, Holmberg, Lukacs, Casulli, Deplazes and Juremalm2014) and under standardized conditions, combined with the detection of an internal control for every single sample using a Rotor-Gene Q (Qiagen, Hilden, Germany) machine. The qPCR was performed for detection of Taenia spp. (Oberli et al., Reference Oberli, Furrer, Skoko, Müller, Gottstein and Bittel2023) by selecting a subsample of 141 among the 494 copro-DNA obtained using a Rotor-Gene Q (Qiagen, Hilden, Germany) machine, corresponding to all samples from 3 provinces (Ağrı, Bayburt, Kars). These 3 provinces were pre-selected for method comparison only, due to cost and reagent availability considerations.
Each qPCR run incorporated a negative control (sterile water) and a positive control (quantified DNA of targeted species), as well as the positive and negative controls from the extraction step. Samples were considered inhibited when the internal control was not detected or when its Cq value exceeded 37 (a Cq value of 35 being expected). The copro-DNA was then diluted ten-fold, or one hundred-fold if necessary, in order to remove inhibition and obtain an interpretable result for the target parasitic DNA. The composition and conditions of the qPCR analyses are presented in Table 1. All qPCR analyses were performed in duplicate for each copro-DNA sample. When a Cq value associated with a characteristic amplification curve was obtained for only 1 of the duplicates, the assay was repeated, with at least 1 Cq value being required for a detection of the parasitic species to be considered positive.
TaqMan® qPCR reaction conditions used for the detection of E. granulosus s.l., E. multilocularis and Taenia spp
Table 1 Long description
Reagents, volumes or concentrations, cycling conditions, and references are listed for four qPCR assays: E. granulosus s.s. and E. canadensis, qPCR E. ortleppi, qPCR E. multilocularis, and qPCR Taenia spp. For E. granulosus s.s. and E. canadensis, Luna Universal master mix is 10 microliters with initial activation at 95 degrees Celsius for 1 minute, then 45 cycles of 95 degrees Celsius for 15 seconds and 60 degrees Celsius for 30 seconds; primers G1 and G3 forward and reverse are 0.4 micromolar, TQ G1 and G3 FAM probe is 0.2 micromolar, primers G6 and G10 forward and reverse are 0.4 micromolar, TQ G6 and G10 HEX probe is 0.2 micromolar, TQ Cl Cy5 probe is 0.1 micromolar, and IAC G1 and G3 2 G at 100 copies per microliter is 1 microliter. For qPCR E. ortleppi, Maxima probe master mix is 10 microliters with initial activation at 95 degrees Celsius for 10 minutes, then 45 cycles of 95 degrees Celsius for 15 seconds and 60 degrees Celsius for 60 seconds; ROX dye is 300 nanomolar, G5 forward and reverse are 1.5 micromolar, TQ G5 HEX probe is 0.2 micromolar, and TQ Cl Cy5 probe is 0.1 micromolar. For qPCR E. multilocularis, Maxima probe master mix is 10 microliters with initial activation at 95 degrees Celsius for 10 minutes, then 45 cycles of 95 degrees Celsius for 15 seconds and 60 degrees Celsius for 60 seconds; qEM MGB forward and reverse are 1.9 micromolar, TQ Em MGB FAM probe is 0.1 micromolar, TQ Cl Cy5 probe is 0.1 micromolar, and IAC Em MGB at 100 copies per microliter is 1 microliter. For qPCR Taenia spp., Maxima master mix is 10 microliters with initial activation at 95 degrees Celsius for 10 minutes, then 45 cycles of 95 degrees Celsius for 15 seconds, 53 degrees Celsius for 30 seconds for annealing, and 72 degrees Celsius for 30 seconds for extension; Cest F and Cest R are 1.5 micromolar, water is 1.8 microliters, and MGB Probe is 0.2 micromolar.
IAC, internal amplification control; TQ, TaqMan®; FAM, HEX, Cy5, reporter dyes.
Taenia genus-specific Cest4-for and Cest5-rev primers (Trachsel et al., Reference Trachsel, Deplazes and Mathis2007), which amplify 267 base pairs (bp) of the small subunit of ribosomal RNA (rrnS), were used with classical PCR performed directly on the copro-DNA extracted in this study. The DNA of Taenia spp. previously confirmed by molecular analysis as a positive control and distilled water as a negative control, were included in each PCR run. The PCR products were analysed by 2% agarose gel electrophoresis, stained with SYBR Safe (Invitrogen, Carlsbad, CA, USA), and visualized using UV transillumination (Vilber Lourmat, Quantum ST4 1100/20M, Collegien, France). The parasite species were identified and confirmed following both copro-cPCR and qPCR analysis, with Sanger sequencing performed by a private company (Eurofins Genomics GmbH, Ebersberg, Germany). Sequences obtained were analysed using the Basic Local Alignment Search Tool (BLASTn) (National Center for Biotechnology Information, www.ncbi.nlm.nih.gov/BLAST). Sequencing revealed that, in addition to Taenia species, both assays also amplified DNA from other cestodes, specifically Mesocestoides spp. and Dipylidium caninum (see Results).
Statistical analyses
The occurrence data of E. granulosus s.s. and E. multilocularis in the free-roaming dog fecal samples were subjected to a chi-square (χ2) test to assess the association between positivity and province, at P < 0.05. As the 2 diagnostic methods were applied to the same samples, the difference between the paired detection rates was assessed with McNemar’s test with continuity correction: copro-qPCR vs S/F-cPCR for E. multilocularis, and copro-qPCR vs copro-cPCR for Taenia species. In order to evaluate the level of agreement between the methods (copro-qPCR and S/F-cPCR) for E. multilocularis, and for Taenia species (copro-cPCR and copro-qPCR), Cohen’s Kappa analysis was employed. The resulting Kappa coefficients were interpreted according to the criteria proposed by Landis and Koch: 0.00–0.20 as slight, 0.21–0.40 as fair, 0.41–0.60 as moderate, 0.61–0.80 as substantial, and 0.81–1.00 as almost perfect agreement (Landis and Koch, Reference Landis and Koch1977). As no independent reference standard was available, no estimate of diagnostic sensitivity or specificity could be calculated, and the results of this study are expressed as detection rates and agreement between methods. All statistical analyses were conducted using R software (v4.3.2) via the RStudio interface (R Core Team, 2025).
Results
Epidemiological results
Echinococcus granulosus s.s. DNA was detected in 99 of the 494 selected fecal samples (20.0%) from all the 7 provinces in the north-eastern region. Furthermore, 2 samples (0.4%) from Erzurum province tested positive for E. canadensis (G6/G7) DNA, whereas none of the samples tested positive for E. ortleppi. Echinococcus multilocularis DNA was detected using the copro-qPCR approach in all the 7 provinces, with an overall prevalence of 5.7% (28/494). The other cestode species detected in the samples were subsequently identified as follows: T. hydatigena (96 samples, 19.4%), Mesocestoides spp. (44, 8.9%), T. multiceps (28, 5.7%), T. ovis (5, 1%), T. pisiformis (3, 0.6%), T. crassiceps (3, 0.6%) and T. serialis (2, 0.4%) (Table 2). Detection of the 2 most common Taenia species (i.e. T. hydatigena and T. multiceps) and Mesocestoides spp. was obtained in all the 7 provinces, with the exception of the absence of T. multiceps in Erzincan.
Copro-real-time PCR analysis results for the detection of E. multilocularis, E. granulosus s.l. and other cestode species in fecal samples collected from free-roaming dogs in north-eastern Türkiye
Table 2 Long description
Copro real time PCR results are listed by province for dog fecal samples, with counts for collected samples and selected samples, plus detections of E. granulosus s.l. as E. granulosus s.s. and E. canadensis, E. multilocularis by qPCR and by S or F cPCR, and other cestode species. Selected samples total 494 out of 1069 collected. E. granulosus s.s. detections by province are: Agri 14, Ardahan 5, Bayburt 21, Kars 5, Igdir 9, Erzincan 3, and Erzurum 42, for a total of 99 with a confidence interval shown in the total row. E. canadensis appears only in Erzurum with 2 detections, and the total row lists 2 with a confidence interval. E. multilocularis detections by qPCR are: Agri 3, Ardahan 3, Bayburt 1, Kars 5, Igdir 1, Erzincan 2, and Erzurum 13, totaling 28 with a confidence interval; S or F cPCR detections are: Agri 4, Ardahan 1, Bayburt 6, Kars 9, Igdir 6, Erzincan 4, and Erzurum 19, totaling 49. Other cestode species detections include T. hydatigena totaling 96, Mesocestoides spp. totaling 44, T. ovis totaling 28, T. serialis totaling 5, T. multiceps totaling 28, T. crassiceps totaling 3, and T. pisiformis totaling 3, with provincial counts shown in each row and several provinces reporting dashes for some species.
S/F (±), sieving/flotation method analyses tested results; cPCR, classical PCR.
a The results were previously obtained by Akyuz and Avcioglu (Reference Akyuz and Avcioglu2025) on the same subsamples used in this study using the sieving/flotation method, followed by classical PCR. The total proportions are presented with a 95% confidence interval.
Co-infection by E. granulosus s.l. and E. multilocularis was detected in 7 (1.4%) out of the samples examined (Table 3). Detection of E. granulosus s.s. and other cestode species was obtained in 65 dogs (13.2%). Among these samples, 45 tested positive for Taenia species with livestock as IHs. Co-infection of other cestode species and E. multilocularis was detected in 18 (3.6%) samples, of which 2 were identified as T. crassiceps and 4 as Mesocestoides spp., both involving small rodents as IHs, while 1 was due to T. serialis and 5 involved T. hydatigena.
Results by province for co-infection of E. multilocularis, E. granulosus s.l. and Taenia species in fecal samples. Values are the number of positive samples and, in brackets, the percentage of the positive samples analysed in that province. The total number and proportions of positive cases from all 7 provinces are presented within a 95% confidence interval
Table 3 Long description
Counts of positive fecal samples are listed by province for co-infections involving E. granulosus s.s., E. multilocularis, and named Taenia and Mesocestoides taxa, with percentages shown for each province and an overall total with a confidence interval. Samples analysed were 48 in Agri, 54 in Ardahan, 44 in Bayburt, 49 in Kars, 43 in Igdir, 47 in Erzincan, and 209 in Erzurum, for 494 total. For, E. granulosus s.s. plus E. multilocularis, positives were 1 in Agri, 1 in Ardahan, 1 in Kars, and 4 in Erzurum, totaling 7. For, E. granulosus s.s. plus other cestode species, positives were 9 in Agri, 3 in Ardahan, 16 in Bayburt, 4 in Kars, 5 in Igdir, 2 in Erzincan, and 26 in Erzurum, totaling 65. Under, E. granulosus s.s. plus cestode from livestock, T. hydatigena totals 37 with provincial counts 7 Agri, 9 Bayburt, 3 Kars, 3 Igdir, and 15 Erzurum; T. multiceps totals 5 with 2 Bayburt, 1 Igdir, and 2 Erzurum; T. ovis totals 3 with 1 each in Igdir, Erzincan, and Erzurum. Under, E. granulosus s.s plus others, T. serialis appears once in Erzurum, T. crassiceps appears once in Ardahan, and Mesocestoides spp. totals 14 with 2 Agri, 1 Ardahan, 1 Bayburt, 1 Kars, and 9 Erzurum. For, E. multilocularis plus other cestode species, positives were 2 Agri, 1 Ardahan, 1 Bayburt, 3 Kars, 1 Igdir, 1 Erzincan, and 9 Erzurum, totaling 18. Under, E. multilocularis plus cestode from small rodents, T. crassiceps totals 2 with 1 Bayburt and 1 Erzurum, and Mesocestoides spp. totals 4 with 2 Kars, 1 Erzincan, and 1 Erzurum. Under, E. multilocularis plus others, T. serialis appears once in Erzurum and T. hydatigena totals 5 with 1 Agri and 4 Erzurum.
–, not detected.
The prevalence of E. granulosus s.s. varied significantly among provinces. The highest positivity was detected in Bayburt (47.7%, 21/44), followed by Ağrı (29.2%, 14/48), Iğdır (20.9%, 9/43), Erzurum (20.1%, 42/209), Kars (10.2%, 5/49) and Ardahan (9.3%, 5/54), while the lowest prevalence was found in Erzincan (6.4%, 3/47) (P < 0.05), and the regional detection rate was 20.0% (99/494). The highest E. multilocularis prevalence was found in Kars (10.2%, 5/49), followed by Ağrı (6.3%, 3/48), Erzurum (6.2%, 13/209), Ardahan (5.6%, 3/54) and Erzincan (4.3%, 2/47), while the lowest prevalence was found in Bayburt (2.3%, 1/44) and Iğdır (2%, 1/43) (P > 0.05) and the overall regional positivity rate was 5.7% (28/494).
Method comparisons
The S/F-cPCR method, which was previously used to detect E. multilocularis from these fecal samples, was compared with the copro-qPCR results obtained here from the same samples (Table 4). The detection rate was higher with S/F-cPCR (9.9%, 49/494) than with copro-qPCR (5.7%, 28/494), and the difference between these paired proportions was significant (McNemar’s χ2 = 7.55, df = 1, P = 0.006). Twelve samples were positive by both methods, 37 were positive only by S/F-cPCR and 16 only by copro-qPCR. The level of agreement between the 2 diagnostic methods was fair [n: 494, κ: 0.26 (95% confidence interval [CI]: 0.12–0.40)]. All 37 samples positive only by S/F-cPCR belonged to the subset in which taeniid eggs had been observed, whereas 8 of the 16 samples positive only by copro-qPCR contained no detectable taeniid eggs (Table 4). The number of taeniid eggs did not differ significantly between the 12 samples positive by both methods (median 4, range 1 to >20) and the 37 samples positive only by S/F-cPCR (median 5, range 1 to >20; P = 0.093); samples containing more than 20 eggs were present in both groups.
Comparative results of S/F-cPCR and copro-qPCR for E. multilocularis detection from free-roaming dogs’ fecal samples. The comparison was made by matching the same samples previously examined by Akyuz and Avcioglu (Reference Akyuz and Avcioglu2025). Concordant results indicate samples positive (+/+) or negative (−/−) by both S/F-cPCR and copro-qPCR methods. Discordant results mean that the same samples are positive with 1 method but negative with the other, with copro-qPCR in the first and S/F-cPCR in the second position
Table 4 Long description
Counts and percentages compare copro-qPCR and S/F cPCR results for E. multilocularis across fecal samples grouped by presence or absence of taeniid eggs. For presence of taeniid eggs, N was 133, with 20 copro-qPCR positive and 49 S/F cPCR positive; concordant positive was 12 and concordant negative was 76, with 8 discordant plus or minus and 37 discordant minus or plus. For absence of taeniid eggs, N was 361, with 8 copro-qPCR positive and 0 S/F cPCR positive; concordant positive was 0 and concordant negative was 353, with 8 discordant plus or minus and 0 discordant minus or plus. Totals were N 494, with 28 copro-qPCR positive and 49 S/F cPCR positive; concordant positive was 12 and concordant negative was 429, with 16 discordant plus or minus and 37 discordant minus or plus.
S/F-cPCR, sieving/flotation method followed by classical PCR; qPCR, quantitative PCR.
The concordance between the copro-cPCR followed by sequencing and copro-qPCR methods for detection of Taenia spp. was evaluated in 141 samples (Table 5). The presence of species T. hydatigena, T. multiceps, T. pisiformis and Mesocestoides spp. was detected by both methods. However, T. ovis (n = 1) and T. crassiceps (n = 1) were detected exclusively by copro-cPCR, while Dipylidium caninum was detected solely by copro-qPCR. Regarding the overall results for the presence of Taenia spp. solely, the diagnostic agreement between the copro-qPCR and copro-cPCR methods was moderate [n: 141, κ: 0.56 (95% CI: 0.42–0.70)]. The proportions of samples for Taenia species were 32.6 and 24.8% for copro-cPCR and copro-qPCR, respectively. This difference between paired proportions was not significant (McNemar’s χ2 = 0.13, df = 1, P = 0.72).
Comparative results of classical copro-PCR (copro-cPCR) and real-time copro-PCR (copro-qPCR) for detection of Taenia and other cestode species after sequencing. Concordant refers to samples positive (+/+) or negative (−/−) by both copro-cPCR and copro-qPCR methods. Discordant results mean that same samples are positive with 1 method but negative with the other, with copro-cPCR given first
Table 5 Long description
Counts and percentages are reported for 141 selected samples, comparing copro cPCR with copro qPCR for seven taxa, plus a summary of concordant and discordant outcomes. For T. hydatigena, copro cPCR reports 30 samples at 21.3 percent and copro qPCR reports 22 at 15.6 percent. For T. multiceps, copro cPCR reports 11 at 7.8 percent and copro qPCR reports 7 at 5 percent. For T. pisiformis, copro cPCR reports 3 at 2.1 percent and copro qPCR reports 6 at 4.3 percent. For Mesocestoides spp., copro cPCR reports 13 at 9.2 percent and copro qPCR reports 6 at 4.3 percent. For T. ovis and T. crassiceps, copro cPCR reports 1 sample at 0.7 percent for each, while copro qPCR reports 0 for both. For D. caninum, copro cPCR reports 0 and copro qPCR reports 28 at 19.9 percent. In the outcome summary, positive by both methods is 55 at 39 percent and negative by both methods is 55 at 39 percent, for 110 total concordant results at 78 percent. Discordant results include 17 at 12.1 percent in the category labeled plus or minus, and 14 at 9.9 percent in the category labeled minus slash plus, for 31 total discordant results at 22 percent.
In order to evaluate the ability to detect copro-DNA of taeniid species in the absence or presence of taeniid eggs, only fecal samples containing a single species among the Echinococcus and Taenia genera were selected, in order to ensure that the eggs previously observed corresponded to the species detected (Table 6). With regard to the 6 parasitic species detected without co-infection, 124 fecal samples were considered, including 43 with the presence of taeniid eggs. While the detection of E. multilocularis by copro-qPCR was more frequently obtained in the presence of taeniid eggs (6 out of 10 cases), the opposite was observed for E. granulosus s.l. (6 out of 29 cases). Regarding Taenia species, the detection of copro-DNA was obtained mainly in the absence of taeniid eggs (54 out of 85 cases).
Species-level identification of taeniid eggs in fecal samples previously found positive by sieving-flotation (Akyuz and Avcioglu, Reference Akyuz and Avcioglu2025), using copro-qPCR for E. multilocularis and E. granulosus s.l. and copro-cPCR with sequencing for Taenia spp. Only single-species cases were included. Values are the number of samples and, in brackets, the percentage of the column total
Table 6 Long description
Counts and column percentages are given for taeniid egg presence and absence across species identified by molecular tests. In the Total column, Presence is 43 samples at 34.7 percent and Absence is 81 samples at 65.3 percent, for 124 samples total. For E. multilocularis, Presence is 6 samples at 60 percent and Absence is 4 samples at 40 percent, total 10. For E. granulosus s.l., Presence is 6 samples at 20.7 percent and Absence is 23 samples at 79.3 percent, total 29. For Total Echinococcus spp., Presence is 12 samples at 30.8 percent and Absence is 27 samples at 69.2 percent, total 39. For T. hydatigena, Presence is 19 samples at 33.3 percent and Absence is 38 samples at 66.7 percent, total 57; for T. multiceps, Presence is 11 samples at 50 percent and Absence is 11 samples at 50 percent, total 22. For T. pisiformis, Presence is 1 sample at 33.3 percent and Absence is 2 samples at 66.7 percent, total 3; for T. ovis, Presence is not detected and Absence is 3 samples at 100 percent, total 3. For Total Taenia spp., Presence is 31 samples at 36.5 percent and Absence is 54 samples at 63.5 percent, total 85.
–, not detected.
Discussion
In Türkiye, research on echinococcosis began to attract attention after 2000. However, there is still limited information about the status of the disease, especially in the north-eastern region, where the highest number of human AE and CE cases are reported (Gürler et al., Reference Gürler, Bölükbaş, Açıcı and Umur2019; Altintas et al., Reference Altintas, Karamil, Turkum, Akil, Sakarya, Bozkaya, Šoltys and Altintas2020; Kurt et al., Reference Kurt, Avcioglu, Guven, Balkaya, Oral, Kirman, Bia and Akyuz2020; Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Akyuz, Bia, Gulbeyen and Yaya2021a). Free-roaming dogs can be found in this region in overlapping domestic, wild and peridomestic environments (Toews et al., Reference Toews, Musiani, Checkley, Visscher and Massolo2021, Reference Toews, Musiani, Smith, Checkley, Visscher and Massolo2024). The assessment of their prevalence for E. multilocularis and E. granulosus s.l. but also other cestode species sharing the same lifecycle, is very useful for a better understanding of their role in transmission. The recent development of new DNA extraction methods from fecal samples and specific qPCR assays should facilitate the production of epidemiological data compared to previous classical methods based on the presence of taeniid eggs (Isaksson et al., Reference Isaksson, Hagström, Armua-Fernandez, Wahlström, Ågren, Miller, Holmberg, Lukacs, Casulli, Deplazes and Juremalm2014; Maksimov et al., Reference Maksimov, Bergmann, Wassermann, Romig, Gottstein, Casulli and Conraths2020; Kunisch et al., Reference Kunisch, Wassermann, Peters, Feldmann, Prinzenberg, Klink, Bergmann, Wernike, Maksimov and Schares2025).
The higher detection rate obtained with S/F-cPCR than with copro-qPCR (9.9 and 5.7%, respectively) is at first sight counter-intuitive, since copro-qPCR detects parasite DNA irrespective of egg excretion. The same pattern has been reported in experimentally infected foxes, in which PCR on eggs recovered by flotation and sieving outperformed PCR on DNA extracted directly from feces during the low patent period (Al-Sabi’ et al., Reference Al-Sabi’, Kapel, Deplazes and Mathis2007). In the absence of an independent reference standard, the higher positivity of 1 method documents a higher detection rate and not a higher diagnostic sensitivity.
Firstly, copro-qPCR was performed on a fixed 300 mg aliquot, whereas flotation concentrates eggs from a much larger amount of feces, and egg intensity affects PCR outcome in fecal samples (Klein et al., Reference Klein, Liccioli and Massolo2014). The number of eggs, however, does not support a simple burden effect, since samples positive by S/F-cPCR alone did not carry fewer taeniid eggs than concordant positives; the comparison is limited, since taeniid eggs are morphologically indistinguishable and mixed infections were frequent, so the total count is not a proxy for the number of E. multilocularis eggs. Secondly, the copro-qPCR was performed after prolonged storage at −20 °C, and free parasite DNA is more exposed to nuclease activity and freeze-thaw damage than DNA enclosed within the egg (Klein et al., Reference Klein, Liccioli and Massolo2014), so a differential loss of the non-egg fraction is plausible. Cross-contamination during the processing of egg-derived DNA, or cross-reaction with E. granulosus s.l. or Taenia spp., cannot be excluded (Maksimov et al., Reference Maksimov, Bergmann, Wassermann, Romig, Gottstein, Casulli and Conraths2020). Conversely, half of the samples positive for E. multilocularis by copro-qPCR alone contained no taeniid eggs, illustrating the capacity of copro-DNA approaches to detect DNA released from attached adult worms or from detached worms passed in the feces.
Published comparisons do not point in a single direction; the copro-qPCR has reached sensitivities comparable to, or exceeding, those of parasitological methods when assessed against intestinal examination (Isaksson et al., Reference Isaksson, Hagström, Armua-Fernandez, Wahlström, Ågren, Miller, Holmberg, Lukacs, Casulli, Deplazes and Juremalm2014; Maas et al., Reference Maas, van Roon, Dam-Deisz, Opsteegh, Massolo, Deksne, Teunis and van der Giessen2016; Kunisch et al., Reference Kunisch, Wassermann, Peters, Feldmann, Prinzenberg, Klink, Bergmann, Wernike, Maksimov and Schares2025), yet some samples are detected only by egg-based protocols (Yasur-Landau et al., Reference Yasur-Landau, Genad, Salant, Dvir, Mazuz and Baneth2023) and flotation followed by a species-specific PCR performs at a level close to intestinal scraping (74 vs 78%) (Széll et al., Reference Széll, Sréter-Lancz and Sréter2014). The direction of the discordance thus may depend on host and parasite species, storage history and protocol details rather than on any intrinsic superiority of 1 approach, since the 2 protocols measure different biological events. This is also apparent in the single-species infections. Among the 124 samples concerned, 81 (65.3%) contained no taeniid eggs; against this background, E. granulosus s.l. was detected in the absence of eggs in 23 of 29 cases (79.3%), which is consistent with prepatent infection, discontinuous egg excretion and the frequent unhooking of worms, whereas E. multilocularis was detected in egg-positive samples in 6 of 10 cases (60.0%). Taenia species were detected in the absence of eggs in 54 of 85 cases (63.5%), close to the background value, even though their biology, in which few worms release numerous proglottids and thus large numbers of eggs, would rather predict detection in egg-positive samples (Alvarez Rojas et al., Reference Alvarez Rojas, Mathis and Deplazes2018); the numbers involved are, however, small and these proportions should be regarded as descriptive. These findings must be cautiously interpreted, as a positive copro-PCR only demonstrates the presence of parasite DNA. Neither the presence of viable or infective eggs nor the patency of the infection at the time of sampling can be inferred from it. Investigations of both Echinococcus and Taenia species under controlled experimental conditions would provide a more accurate understanding of the significance of non-egg parasite DNA for the detection of infection in both prepatent and patent periods, especially in regard to the discontinuous excretion of eggs.
For Taenia spp., the moderate agreement between copro-cPCR and copro-qPCR (κ = 0.56) suggests that primer and probe design influence the diagnostic outcome, although the difference between the 2 detection rates was not significant (P = 0.72) and the 2 assays should therefore be regarded as differing mainly in the range of species they detect rather than in overall performance. While both assays were initially targeted at Taenia species, both also amplified Mesocestoides spp., which was previously reported by Trachsel et al. (Reference Trachsel, Deplazes and Mathis2007). Nevertheless, no information about cross-reaction was reported by Oberli et al. (Reference Oberli, Furrer, Skoko, Müller, Gottstein and Bittel2023), while amplification of D. caninum was obtained. This cestode is of little interest when focusing on cestodes with a lifecycle similar to that of Echinococcus or Taenia species, as it involves additional time and cost for sequencing. The absence of detection of some Taenia species by the qPCR assay compared with cPCR may be explained by its original design, which targeted the zoonotic species relevant to human diagnosis (Oberli et al., Reference Oberli, Furrer, Skoko, Müller, Gottstein and Bittel2023). In the absence of another qPCR assay for the detection of Taenia species, the cPCR from Trachsel et al. (Reference Trachsel, Deplazes and Mathis2007) still constitutes the most relevant approach, even though it was initially published for DNA extracted from taeniid eggs after flotation and not directly from copro-DNA. Nevertheless, the absence of an internal control, which is very useful for copro-DNA, may be problematic and may lead to potential underestimation of Taenia species.
In the present study, E. granulosus s.s. (20.0%) was identified as the dominant Echinococcus species, which is consistent with previous molecular studies conducted in Türkiye and particularly in the north-eastern region (Kesik et al., Reference Kesik, Simsek, Kilinc and Koroglu2019; Avcioglu et al., Reference Avcioglu, Bia, Balkaya, Kirman, Akyuz and Guven2022; Selcuk et al., Reference Selcuk, Aslan Celik, Celik, Celik, Ercan, Uslug, Tekin and Simsek2024). The prevalence of E. granulosus s.l. in Erzurum was found to be 21.1% in this study, a notable increase from the 10.8% previously reported (Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Akyuz, Bia, Gulbeyen and Yaya2021b). While it is conceivable that the infection burden has increased over time, this higher prevalence may also be due to the use of more sensitive qPCR methods capable of detecting very low DNA amounts during both prepatent and patent periods (Maksimov et al., Reference Maksimov, Bergmann, Wassermann, Romig, Gottstein, Casulli and Conraths2020). The high prevalence of E. granulosus s.s. across the provinces, such as 20.1% in Erzurum and 20.9% in Iğdır, correlates with the high CE positivity rates reported in sheep from the same provinces (Avcioglu et al., Reference Avcioglu, Bia, Balkaya, Kirman, Akyuz and Guven2022). This situation can be explained by the socio-economic structure in the region, where livestock farming is the main livelihood and urbanization is underdeveloped. Traditional and faith-based (Holy Sacrifice) slaughtering procedures remain widespread, offering dogs frequent access to infected offal. This active domestic lifecycle between dogs and livestock is further supported by the detection of other livestock-sourced parasites found in the same dogs, including T. hydatigena (19.4%), T. multiceps (5.7%) and T. ovis (1%). Bayburt, which had the highest E. granulosus s.s. detection rate, has the lowest farm animal density of the 7 provinces (34.7 animals per km2), so that livestock density alone does not explain the provincial pattern; the frequency of uncontrolled home slaughtering and of dog access to offal, which were not measured here, are likely to be more relevant.
The detection of E. canadensis is in accordance with previous reports of this species in the country, more specifically the genotype G7, which can be strongly supposed to be involved also in the present study. The absence of camel and pig farming in the region suggests a parasitic lifecycle between dogs and likely wild boars and/or goats, as previously described, even if cases in sheep and cattle were also reported (Snábel et al., Reference Snábel, Altintas, D’Amelio, Nakao, Romig, Yolasigmaz, Gunes, Turk, Busi, Hüttner, Sevcová, Ito, Altintas and Dubinský2009; Kaya Kesik et al., Reference Kaya Kesik, Çelik, Günyaktı Kılınç, Karabulut, Çevik and Şimşek2021; Gunyakti Kilinc et al., Reference Gunyakti Kilinc, Celik, Kesik and Simsek2024; Selcuk et al., Reference Selcuk, Aslan Celik, Celik, Celik, Ercan, Uslug, Tekin and Simsek2024). Although the prevalence of E. canadensis was lower than that of E. granulosus s.s., the zoonotic risk associated with this species should not be overlooked, particularly given the reported human cases in eastern Türkiye (Bakal et al., Reference Bakal, Çelik, Simsek, Kesik and Günyaktı Kılınç2021).
The distribution of E. multilocularis depends on the interplay between parasite characteristics and climatic conditions (Kapel et al., Reference Kapel, Torgerson, Thompson and Deplazes2006). Sustainable lifecycles are established in the cool, humid ecological niches of north-eastern Türkiye, where high-altitude valleys support dense populations of red foxes (Vulpes vulpes) and voles (Microtus spp.) (Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Bia, Gülbeyen, Kurt, Yaya and Demirtaş2017, Reference Avcioglu, Guven, Balkaya, Kirman, Akyuz, Bia, Gulbeyen and Yaya2021a; Gürler et al., Reference Gürler, Demirtaş, Bölükbaş, Gençay, Barılı, Karaca, Akçay, Açıcı, Umur and Deplazes2023). In this study, E. multilocularis copro-DNA was detected in all provinces with a regional prevalence of 5.7%. In Erzurum, the prevalence was 6.2% (13/209), nearly double the 3.58% previously reported (Avcioglu et al., Reference Avcioglu, Guven, Balkaya, Kirman, Akyuz, Bia, Gulbeyen and Yaya2021b). The provincial figures did not differ significantly and did not follow the climatic gradient of the region: the lowest detection rates were recorded in Iğdır, the driest and warmest province of the study area, but also in Bayburt, so that the present data do not allow any association between local climate and E. multilocularis occurrence to be established. Free-roaming dogs in this region serve as a critical link between domestic and sylvatic lifecycles, as already reported elsewhere for domestic dogs (Toews et al., Reference Toews, Musiani, Smith, Checkley, Visscher and Massolo2024). The high density of foxes, their synanthropic feeding habits and the expansion of human settlements into rural areas contribute to dogs becoming involved in the sylvatic cycle. This ‘bridge’ role is supported by the identification of other rodent-associated parasites in the free-roaming dogs investigated, such as Mesocestoides spp. (8.9%), and T. crassiceps (0.6%). Notably, this study presents the first report of T. crassiceps in dogs in Türkiye, a parasite previously identified only in foxes (Öter et al., Reference Öter, Bilgin, Tınar and Tüzer2011).
The co-detection of E. granulosus s.l. and E. multilocularis (1.4%), together with both livestock-associated and rodent-associated cestode species, provides further insight into the transmission dynamics in the region. The concurrent presence of E. granulosus s.s. with livestock-related Taenia species, and E. multilocularis with rodent-associated parasites such as Mesocestoides spp. and T. crassiceps, suggests the existence of overlapping domestic and sylvatic transmission cycles. These findings indicate that some free-roaming dogs may have access to both infected offal and preyed small mammals, thereby becoming exposed to parasites maintained in distinct ecological cycles. Consequently, in this context, free-roaming dogs act as epidemiological links between domestic and wildlife transmission systems. Taken together, these findings indicate complex and partially overlapping transmission dynamics of echinococcosis in north-eastern Türkiye. Integrated approaches, supported by sensitive molecular tools such as copro-qPCR, may improve surveillance and understanding of parasite circulation in co-endemic areas.
This study has several limitations. The samples were collected in 2019-2020, so the prevalences reported here represent a baseline rather than a current estimate. Without an independent reference standard, only detection rates and between-method agreement could be assessed, and the sensitivity and specificity of each assay remain unknown. Finally, the comparison of methods for Taenia species was based on a subset of samples from 3 provinces and cannot be extrapolated to the whole study area.
Conclusion
The findings indicate that free-roaming dogs may play a significant role in the lifecycle of E. granulosus s.s. and E. multilocularis but also other Taenia species in the north-eastern region of Türkiye. The sympatric presence of CE and AE increases the zoonotic risk, and control and diagnostic strategies for echinococcosis should therefore be designed for both domestic and wild environments. Applied to the same 494 samples, the 2 protocols yielded 9.9% and 5.7% for E. multilocularis (κ = 0.26): prevalence figures from egg-based and copro-DNA-based protocols are therefore not interchangeable, and the method used must be reported in surveillance programmes. Implementing holistic prevention and control programmes, including dog population management and movement control, regular antiparasitic treatment, monitoring of home slaughtering practices, and public health education, is of critical importance.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0031182026103035.
Data availability statement
The data supporting the findings of this study are available from the corresponding author upon reasonable request.
Acknowledgements
The dog illustration used in the graphical abstract was partially generated using Gemini 1.5 Pro (Google LLC). The graphical abstract was subsequently redesigned with the assistance of Claude (Anthropic, Claude Opus 5) and edited and reviewed by the authors. Use of artificial intelligence. Generative artificial intelligence tools were used in the preparation of this manuscript as follows. Gemini 1.5 Pro (Google LLC) was used to generate the dog illustration of the graphical abstract. Claude (Anthropic) was used to redesign the layout of the graphical abstract and to assist with language editing of the manuscript. No generative artificial intelligence tool was used to generate, analyse or interpret the data, or to draw the scientific conclusions, and none is listed as an author. The authors reviewed all artificial intelligence-assisted output and take full responsibility for the content of the publication.
Author contributions
M.A., H.A. and G.U. conceived and designed the study. M.A., R.K., E.G., H.A. and G.U. conducted data gathering and performed investigations. M.A., R.K., E.G., H.A. and G.U. performed statistical and formal analyses. M.A., E.G., H.A. and G.U. performed data validation. M.A., H.A. and G.U. wrote the original draft and revised the manuscript. G.U. coordinated the project administration.
Financial support
This study was funded by the Scientific and Technological Research Council of Turkey (TÜBİTAK, Project No. 2219-1059B192402138).
Competing interests
The authors declare that there are no conflicts of interest.
Ethical standards
Ethical approval was obtained from the Atatürk University Animal Research Local Ethics Committee (Approval number: 2019/09).
Open access
