Unexpected Echinococcus multilocularis infections in shepherd dogs and wolves in south-western Italian Alps: A new endemic area?

The European distribution of Echinococcus multilocularis has been reported to be expanding east and north, whereas its southern limits are deemed stable. During a study on Echinococcus granulosus s.l. infections in wolves and dogs in the Italian Maritime Alps, we unexpectedly detected the presence of E. multilocularis eggs in four fecal samples from at least two shepherd dogs, and in five wolf fecal samples. This finding, in an area about 130 km south of the southernmost E. multilocularis report in the Alps, may suggest a rapid expansion southward. While infections in foxes are currently being investigated, these data seem to indicate the potential for a new E. multilocularis endemic area. If this will be confirmed, the implementation of surveillance programs in wild and domestic canids and preventative measures will become a priority.

Introduction

Echinococcus multilocularis, a tapeworm with an indirect life cycle involving mammalian predators as definitive hosts (primarily wild or domestic canids; DH hereafter) and mammalian prey species as intermediate hosts (primarily small rodents; IH hereafter; reviewed in (Romig et al., 2017)), is the etiological agent of alveolar echinococcosis (AE hereafter), the third most relevant human foodborne parasitic disease worldwide (FAO/WHO, 2014). Once mature, the adult worms, attached to the intestinal tract of the DHs, release in the host intestinal lumen their last proglottids with hundreds of embryonated eggs that will be later excreted in the environment within the host feces. Once ingested by a competent IH, or by people (as dead-end host), the eggs hatch and release larvae (oncospheres) in the intestinal tract of the host that pass through the intestinal wall into the blood vessels, and from there to the target organs (almost exclusively the liver). There the oncospheres will develop into metacestodes that will replicate asexually producing thousands of infectious larvae (protoscoleces). When the infectious IH is preyed upon by a competent DH, the protoscoleces evaginate, attach to the wall of the small intestine and develop into adults. The asexual multiplication phase of the metacestode creates very severe clinical conditions, due to an infiltrating tumor-like process that damage the tissues and will eventually invade other organs through a metastasis-like process (Kern et al., 2017). Human AE is estimated to affect more than 18,000 new patients worldwide every year (Torgerson et al., 2010), with about 150–200/year expected in central Europe alone, and an incremental trend in the last decades (Vuitton et al., 2015). Human AE is deadly if untreated, and still highly pathogenic with a fatality of 16% despite recent advances in treatment (Kern et al., 2017). In Europe the parasite is distributed mostly in its central and eastern part, extending into northern France, southern Scandinavia (Denmark, southern Sweden) and parts of the Balkan peninsula (Combes et al., 2012; Deplazes et al., 2017; EFSA, 2015; Oksanen et al., 2016; Umhang et al., 2016).

At present, the parasite is deemed to be expanding north and east, but its southern limits, at least in the Alpine region, are described as stable (Conraths and Deplazes, 2015) and some authors suggested that the limited distribution of Microtus arvalis in southern slopes of the Alps may play a role in limiting E. multilocularis distribution in Switzerland (Guerra et al., 2014). Recent findings of E. multilocularis infections in foxes from Southern France in the Hautes Alpes Department (05) seem to indicate a possible expansion southward (Combes et al., 2012; Umhang et al., 2016).

In Italy, the only reports of E. multilocularis infections have been in the eastern Alps in the Trentino-Alto Adige Region (EFSA, 2015) where multi-locus microsatellite analysis suggested an autochthonous focus of the parasite (Casulli et al., 2009), and where a recent meta-analysis based on 26 papers indicated a prevalence in foxes between 0.5 and 2.9 (Oksanen et al., 2016).

Material and methods

From June to November 2017, we conducted a fecal survey of taeniid cestodes in wild and domestic canids in a protected area (Parco Regionale delle Alpi Liguri) of the Southern Italian Alps in the Imperia Province (Liguria region), about 25 km from the Mediterranean coast in Italy (Fig. 1).

Fig. 1

Locations of wolf (blue dots) and dog (orange dots) fecal samples positive to Echinococcus multilocularis collected during a survey on Echinococcus spp. carried out from June to November 2017 in a mountainous area in the Alps of the Imperia Province, Italy. In the map are also reported the southernmost reports of Echinococcus multilocularis (E. multilocularis) to date in Europe (France, Drs. Boué and Umhang, pers. communication; North-Eastern Italian Alps, Croatia, as in (Beck et al., 2018)). Two dog fecal samples were collected from the same pasture and are represented by a single dot (noted as 2×). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

The area is located in north-western Italy and connects the Apennines with the Maritime Alps (44°10′N 8°05′E; 6041 ha). The elevation ranges from 0 to 2200 m a.s.l. Chestnut and beech woods are the predominant forests. Meso-fauna includes wolf Canis lupus, fox Vulpes vulpes, and dogs Canis lupus familiaris as predators, and roe deer Capreolus capreolus, chamois Rupicapra rupicapra, and the wild boar Sus scrofa. Livestock amounts to more than 3200 cattle, 1200 sheep and goats, and ca 50 horses. Potential IH species for E. multilocularis includes Arvicola scherman, Chionomys nivalis, Microtus multiplex, Microtus savii, Myodes glareolus, Apodemus alpicola, Apodemus flavicollis, Apodemus sylvaticus, Mus musculus along with Rattus norvegicus and Rattus rattus.

We identified 10 standard pathways (4.3 ± 0.27 km) that were walked on a monthly basis to locate and collect feces of wolves and domestic dogs, particularly those owned by local shepherds.

In order to correctly discriminate wolf feces to those of dogs, molecular host identification was performed purifying the genomic DNA with the QIAamp DNA Stool Mini Kit (QIAGEN, Hilden, Germany) and analyzing the 281 bp sequence of the mtDNA control region (Sindičić et al., 2011).

Samples were stored at −80 °C for at least 3 days to deactivate the eggs (Veit et al., 1995), then kept at −20 °C until subsequent analyses.

The samples were first screened for taeniid eggs using ZnCl2 flotation followed by sieving (Mathis et al., 1996). Single eggs were separated and molecularly characterized using partial sequences of the nad1 mtDNA gene as marker (Hüttner et al., 2008) (Table 1). Egg samples which yielded no positive result after nad1 PCR were further analyzed targeting sequences of the cob gene (Hüttner et al., 2008) (Table 1). Readable sequences of partial nad1 and cob were aligned using MEGA7 (Kumar et al., 2016) and compared to GenBank retrieved homologous sequences of Echinococcus multilocularis, E. granulosus, E. ortleppi, E. canadensis, Taenia hydatigena, T. krabbei and T. ovis, according to availability.

Table 1

Primer pairs used for PCR to molecularly characterize Echinococcus multilocularis in canid fecal samples from the Imperia area in Liguria, Italy, in 2017.

Target genes	Primers for the first PCR (5′-3′)a	Primers for the second PCR (5′-3′)
nad1 (mtDNA)b	F: TGTTTTTGAGATCAGTTCGGTGTG R: CATAATCAAACGGAGTACGATTAG	F: CAGTTCGGTGTGCTTTTGGGTCTG R: GAGTACGATTAGTCTCACACAGCA
cob (mtDNA)b	F:TTATGCTATACTTCGGTGTATTA R: ATAAGGATACTCCGGATGACAAC	F: TCGGTGTATTAATTCGAAGATTG R: GATGACAACCACCCAAATAAGTC

F, forward primer; R, reverse primer.

Primers designed by (Hüttner et al., 2008).

Results and discussion

We collected a total of 183 fecal samples. Definitive host species identification by sequencing part of the mtDNA control region confirmed field observations for all the E. multilocularis-positive dog fecal samples, while one wolf fecal sample (IM WLF 79) out of six E. multilocularis-positives turned out to be of a dog so we ended up with 151 feces of wolves and 32 of dogs. We then could screen 152/183 feces (wolf, 120; dog, 32; the rest were discarded from the analyses as they were either too small to allow for both diet analysis and parasitology, or too dry and old) and detected Taeniidae eggs in 11.66% of wolf feces (14/120) and 15.62% of dog feces (5/32).

We isolated 10 to 20 single taeniid eggs from 14 wolf fecal samples and from 5 dog fecal samples. Four out of five dog fecal samples contained E. multilocularis eggs along with other Taeniidae infections (Taenia hydatigena, T. krabbei, T. ovis; Table 2), whereas only 5 out of the 14 wolf fecal samples contained E. multilocularis eggs, and showed coinfection with Taenia spp. and E. ortleppi in two samples (see sequences alignments for positive specimens of E. multilocularis at partial cob Fig. 2a and b; and for positive specimens of E. ortleppi at partial nad1, Fig. 2c).

Table 2

Molecular analysis results conducted on eggs harvested from fecal samples of wolves (WLF; upper table) and dogs (DOG; lower table) collected from June to November 2017 in a mountainous area in the Alps of the Imperia Province, Italy. In table are reported the total number of eggs harvested in each sample, the number of eggs positive (+ve) to nad1 and cob and the corresponding sequencing results.

ID	N Taeniid Eggs extracted	nad1		cob
		N +ve eggs	sequencing results	N +ve eggs	sequencing results
IM WLF 42	7	0	NA	2	42.4Echinococcus multilocularis
					42.7 Taenia hydatigena
IM WLF 88	10	6	L 88.2 Echinococcus ortleppi	0	NA
			L 88.4 Echinococcus multilocularis
			L 88.5 Echinococcus ortleppi
			L 88.7 Taenia krabbei
			L 88.8 Echinococcus ortleppi
			L 88.10 Taenia krabbei
IM WLF 94	10	6	94.1 Taenia hydatigena	1	94.10 Echinococcus multilocularis
			94.2 Taenia hydatigena
			94.3 Taenia hydatigena
			94.4 Taenia ovis
			94.5 Taenia hydatigena
			94.8 Taenia hydatigena
IM WLF 107	10	0	NA	3	107.1 Taenia krabbei
					107.2 Teania hydatigena
					107.4Echinococcus multilocularis
IM WLF 143	10	8	143.1 Taenia hydatigena	0	NA
			143.2 Taenia hydatigena
			143.3Echinococcus multilocularis
			143.4 Taenia hydatigena
			143.5 Taenia hydatigena
			143.7 Echinococcus ortleppi
			143.8 Taenia hydatigena
			143.9 Echinococcus ortleppi
IM WLF 79 (DOGa)	7	3	79.1 Taenia ovis	4	79.3 Taenia ovis
			79.2 Taenia ovis		79.4 Echinococcus multilocularis
					79.5 Echinococcus multilocularis
			79.6 Taenia ovis		79.7 Taenia ovis
DOG 11	9	3	C 11.2 Taenia hydatigenaC 11.7 Echinococcus multilocularisC 11.8 Taenia hydatigena	0	NA
DOG 14	10	4	14.6 Taenia hydatigena	1	14.8 Echinococcus multilocularis
			14.7 Echinococcus multilocularis
			14.9 Taenia hydatigena
			14.10 Taenia hydatigena
DOG 30	10	2	30.3 Taenia hydatigena	0	NA
			30.9 Echinococcus multilocularis

Initially classified as wolf.

Fig. 2

Multiple alignment of partial mitochondrial cob (124bp) from three specimens identified as Echinococcus multilocularis analyzed in the present paper (the first three input sequences) with: (a) six E. multilocularis sequences retrieved from GenBank after comparison by Local Alignment Search Tool BLAST; (b) five sequences referred to Echinococcus granulosus (Eg, EgG1), E. ortleppi (EgG5), E. canadensis (EgG6-7) and Taenia hydatigena () retrieved from GenBank. (c) Multiple alignment of partial mitochondrial nad1 (139bp) from three specimens identified as Echinococcus ortleppi analyzed in the present paper (the first three input sequences) with sequences retrieved from GenBank belonging to other representatives of E. ortleppi, E. granulosus, E. canadensis, E. multilocularis, T. krabbei, T. ovis and T. hydatigena. Dots indicate identity with nucleotide of the first sequences listed.

Importantly, some fecal samples of both wolf and dog were from the same individuals, but we are sure that the dog samples came from at least two different individuals belonging to two different shepherds, and not allowed to roam at night. No travel history that may justify allochthonous infections was reported for these shepherd dogs.

Although we cannot exclude that some of the wolf fecal samples were from the same animal, previous studies have indicated the presence of at least two or more distinct wolf packs in the areas we investigated (Imbert et al., 2016; Marucco and Avanzinelli, 2017), so we believe that the positive wolf fecal samples are from more than one individual. Individual wolf genotyping from our fecal samples was not conducted because of lack of sufficient fecal matter after the repeated molecular testing.

Whereas the Taenia spp. infections were known to be common in these areas (Gori et al., 2015), the detection of E. ortleppi (G5) is unexpected. This is the first report of G5 in Italy in wolves, and in general in DHs in this country. The cycle of this parasite is mainly cattle–dog, but this species has also been identified in other IHs such as buffaloes, pigs, sheep and humans as dead-end hosts (Dinkel et al., 2004; Romig et al., 2017). Echinococcus ortleppi was reported for the first time in Italy by Casulli and colleagues (Rinaldi et al., 2008) in one bovid, and infections without fertile cysts of the cattle strain were reported in Campania (Rinaldi et al., 2008). Definitely, considering the potential for zoonotic infections, further investigation in dogs and wolves is to be implemented in this area to understand which wild and domestic IH species are involved in the maintenance of the parasite in the environment.

Finally, but more importantly, the finding of E. multilocularis in both dogs and wolves was not only unprecedented, but also surprising and remarkable, given its high zoonotic potential. In the last years, a screening of E. multilocularis infections in fox carcasses was carried out (Magi et al., 2015), along with a screening of gastrointestinal parasites in wolves (Gori et al., 2015), but no E. multilocularis infections were detected.

In our study area, there are several species of potentially competent IHs, and both dogs and wolves are known to host E. multilocularis (Romig et al., 2017). A previous study on wolf diet indicated predation upon small mammals with a frequency up to 7.6% in the same area (Imbert et al., 2016). Similarly, free ranging shepherd dogs are likely to be predators of these small mammals as well, although we do not expect that to be frequent. Despite that, it is not very likely that a stable E. multilocularis population could be maintained without the involvement of a DH species with a diet more focused on IHs such as the red fox. With one exception (China, Tibetan plateau (Moss et al., 2013; Vaniscotte et al., 2011)), there is no evidence that wolves and dogs may sustain the cycle.

For this reason, we have already started a collection of fox feces and organizing an opportunistic fox carcass interception in coordination with the local animal health agencies, but more in-depth studies are needed to assess the actual extent of the current distribution of E. multilocularis in this region and the possible origin of infections. Likely, a larger screening of foxes along the Maritime Alps on both sides of the border (France and Italy) will be required. Moreover, since E. multilocularis positive dogs are expected to be present only in high endemic areas, our finding poses a potential threat for public health, and a screening of Echinococcus infections in dogs is strongly recommended.

This is the southern-most report of E. multilocularis in Italy, and the southern-most in Europe beside the most recent findings in Croatia and Serbia (Beck et al., 2018; Lalošević et al., 2016), and an older record from the European part of Turkey (Deplazes et al., 2017). This infection is likely due to a south-eastward expansion of the current E. multilocularis range in France.

Being a reportable disease, we recommend to start an epidemiological surveillance program to monitor the distribution of E. multilocularis in the Maritime Alps, along with an alert to the public health and veterinary professionals to warn about the possibility of AE in people and animals in the area.

The study of marginal populations of E. multilocularis offers a unique opportunity to explore the emergence of parasites with complex life cycles at the edge of their distribution, and explore the role of the host community in the establishment of new endemic areas. Moreover, monitoring these marginal populations is also of paramount importance to detect parasite range expansion and develop proper evidence-based surveillance strategies and preventative measures.