Introduction

The mink (Neovison vison) is a semiaquatic species belonging to the family Mustelidae, which was originally native to North America, but its range has now expanded to many parts of Europe and South America. It is highly prised for its fur and was introduced to Danish fur farms in the 1930s (Anon 2017a). Currently, Denmark is the world’s largest producer of mink skins, with an annual production of around 17 million skins (Anon 2017b).

Eimeria (Coccidia:Eimeriidae) is a genus of apicomplexan parasites that cause the widespread enteric disease, coccidiosis, in a large range of hosts (Tenter et al. 2002). Clinical signs of coccidiosis in mustelids can include diarrhoea, dehydration, weakness, lethargy and weight loss, resulting in high morbidity and also mortalities (Sledge et al. 2011). The burden of Eimeria infections in farmed mink is unknown, although the prevalence in Danish farms is relatively high.

To date, a total of 16 Eimeria species from the family Mustelidae have been identified and recorded in the coccidian database (Duszynski et al. 2000). Of these, four species have been recorded from mink: E. furonis (Hoare 1927), E. hiepei (Gräfner et al. 1967), E. mustelae (Kingscote 1934) and E. vison (McTaggart 1960; Hindsbo et al. 1995). Eimeria furonis has been molecularly characterised from ferret and mink (Abe et al. 2008), but the remaining three Eimeria species were identified on the basis of their oocyst morphological features only. In the present study, we report the morphological and molecular characterisation of an Eimeria sp. across 30 Danish mink farms.

Materials and methods

Sample collection and examination

A survey was conducted on 30 Danish mink farms to determine the prevalence and species of coccidian parasites in Danish farmed American mink. Faecal samples were collected every 7–30th day from April to October 2016, from five female breeders and pooled faecal samples from kits of each female breeder (n = 315 mink), in total 4140 faecal samples. Hence, each mink was repeatedly tested with a mean of 13 samples per animal. Fifteen female breeders died during the study period and were replaced by others; hence, these were sampled fewer timers. The farms received commercially produced feed from three producers located geographically in distinct areas: 10 farms from Southern Jutland (farms A–J), 10 from Zealand (farms K–T) and 10 from Northern Jutland (farm U–AB). Samples were stored at 4 °C until parasitological examination. Prior to examination, all faecal samples were scored 1–5 according to consistency (faecal score 1: hard; 2: soft and moist, solid form; 3: soft and moist, with texture, without solid form; 4: soft with texture, without solid form; 5: watery, without texture). The scores 4 and 5 are considered diarrhoea, whilst the scores 1–3 are non-diarrhoetic (Hansen et al. 2016). All faecal samples were analysed for unsporulated oocysts per gram of faeces (OPG) with a modified McMaster technique (Roepstorff and Nansen 1998), using saturated NaCl with 50% glucose monohydrate as the flotation fluid (specific gravity = 1.27 g/mL). The three samples with the highest number of oocysts per gram of faeces (8166 OPG, 15,600 OPG and 96,000 OPG) were subsequently stored at − 18 °C until DNA extraction. A pool of faeces from these three samples was placed in 2% (w/v) potassium dichromate solution (K2Cr2O7), mixed well and kept at room temperature in the dark to facilitate sporulation. Sporulated oocysts were observed using a Leica DMR light microscope, and images were taken using differential interference contrast (DIC) microscopy with a ×100 oil immersion objective (Fig. 1).

Fig. 1
figure 1

Nomarski interference-contrast photomicrographs of Eimeia vison-like oocysts. Scale bar = 20 μm

Full size image

DNA isolation

Total DNA was extracted from 200 mg of faeces from three samples using a Power Soil DNA Kit (Mo Bio Laboratories, Carlsbad, CA) with some modifications. Briefly, the faeces for DNA extraction were subjected to four cycles of freeze/thaw (liquid nitrogen followed by boiling water) to ensure efficient lysis of oocysts before being processed using the manufacturer’s protocol.

PCR amplification and sequencing

A partial 18S fragment (1221 bp) was generated from three positives with the EiGTF1 and EiGTR1 primers for the external amplification and the primers EiGTF2 and EiGTR2 for the internal reaction (Yang et al. 2012). Similarly, a partial cytochrome c oxidase subunit I (COI) gene sequence (718 bp) was amplified by using a nested PCR as described by Yang et al. (2016).

The amplicons from the internal PCRs were gel purified using an in house filter tip method as previously described (Yang et al. 2013). All PCR products were sequenced using the internal forward and reverse primers in duplicate, using amplicons from different PCR runs. An ABI Prism™ Dye Terminator Cycle Sequencing Kit (Applied Biosystems, Foster City, CA) was used for Sanger sequencing according to the manufacturer’s instructions. The raw sequence data were edited and analysed using the CLC genomics workbench (version 10).

Phylogenetic analysis

Phylogenetic trees were generated for Eimeria spp. at the 18S and COI loci with additional isolates from GenBank. Parsimony, distance and maximum likelihood (ML) analyses were conducted using Molecular Evolutionary Genetics Analysis software (MEGA), version 7 (Arizona State University, Tempe, AZ, USA).

Statistical analyses

All samples were scored as either positive or negative for E. vison-like oocyst excretion. If at least one sample per mink was positive for E. vison-like oocysts during the study period, the mink was considered positive. Chi-squared tests were used to calculate the prevalence for breeding females and kits, the prevalence for each farm and to compare faecal consistency between positive and negative samples. The 95% confidence interval (CI) for the prevalence was calculated on the mean. A p value < 0.05 was required for significance. All statistical analyses were done in SAS studio, version 3.6 (SAS Institute Inc., Cary, NC, USA).

Results

Prevalence and morphological analysis

A total of 108/4140 (2.6%, 2.1–3.1 95% CI) mink samples were positive for E vison-like oocysts, with 78/315 (24.8%) mink being positive at least once during the study period. Of these, 41/152 (27.0%) were kits and 37/163 (22.7%) were female breeders, with a mean of 1.41 and 1.35 positive samples per positive mink, respectively. Across the 30 farms, the prevalence of Eimeria vision-like positive samples ranged from 0.0 to 8.3% (Table 1), whilst the prevalence of Eimeria vision-like positive mink ranged from 0.0 to 70.0% (Table 1). On farms located on Zealand (farms K–T), 53.8% (20.0–70.0%) of the mink excreted E. vison-like oocysts at least once during the study period. In compassion, 14.4% (0.0–30.0%) and 5.7% (0.0–18.2%) of the mink on farms from South Jutland (farms A–J) and North Jutland (farms U–AB), respectively, excreted E. vison-like oocysts at least once during the study period. Analysis of correlation between being positive for oocysts and faecal score revealed that the probability of presenting diarrhoea was unrelated to oocyst excretion (p = 0.396, χ2) (data not shown).

Table 1 Comparison of the prevalence of Eimeria vison-like in mink (Neovison vison) on 30 farms in Denmark as determined by microscopy
Full size table

Phylogenetic analysis at the 18S locus

A 1221-bp 18S rRNA sequence was obtained from all three isolates and all produced identical sequences. Unfortunately, no sequences from E. vison were available in Genbank at the 18S locus. Phylogenetic analyses at this locus using parsimony, ML and NJ analyses produced similar results (Fig. 2a; ML tree shown). Sequences from the E. vison-like genotype grouped in a separate clade and exhibited 99.6% genetic similarity to E. sp. ex Apodemus agrarius (JQ993655), from a striped field mouse (A. agrarius) from the Czech Republic (Kvičerová and Hypša 2013). The mink sequences shared 99.4% similarity with two Isospora spp.: I. neochmiae (KT224380) from a captive-bred red-browed finch (Neochmia temporalis) in Australia (Yang et al. 2016) and I. sp. MAH-2013a (KF648870) from a Superb Glossy Starling (Lamprotornis superbus) in Canada. Analysis of a shorter region of 18S (531 bp) revealed that the E. vison-like genotype sequences sat close to the clade formed by E. furonis (AB329724) and E. sp. ex A. agrarius (JQ993655) sharing 97.7 and 98.1% similarity, respectively (Fig. 2b).

Fig. 2
figure 2

Evolutionary relationships of the E. vision-like genotype inferred by ML analysis of 18S rRNA sequences (1221 bp). Percentage support (> 50%) from 1000 pseudoreplicates from bootstrap analysis

Full size image

Phylogenetic at the COI locus

Unfortunately, no sequences from E. furonis or E. vison were available in GenBank at the COI locus. Phylogenetic analysis of the 718-bp COI sequence from the three mink isolates placed them in a clade with E. cf. ictidea (KT203399) (99.4% similarity at nucleotide level and 100% identity at the amino acid level), from a black footed ferret (Mustela nigripes) from Canada (Fig. 3). The sequence also exhibited 90.3% similarity with E. mephitidis (KT203398), from a striped skunk (Mephitis mephitis) from Canada. Compared with other Eimeria COI sequences, the similarities between non-mink derived sequences were all below 90%.

Fig. 3
figure 3

Evolutionary relationships of the E. vision-like genotype inferred by ML analysis of COI sequences (718 bp). Percentage support (> 50%) from 1000 pseudoreplicates from bootstrap analysis

Full size image

Species description

Only one morphotype of sporulated oocyst was observed. Sporulated oocysts are ellipsoidal in shape, with a smooth 1.4-μm-thick (1.3–1.6 μm) bi-layered oocyst wall. Oocysts measured 21.0 × 13.8 μm (19.9–22.9 × 13.0–14.4 μm), with an oocyst length/width (L/W) ratio of 1.5 (1.0–1.6). Oocyst micropyle and residuum was absent, and polar granule was present. The oocysts contained two roughly ovoid sporocysts measuring 8.3 × 6.4 μm (6.6–9.9 × 5.5–6.9 μm) with a sporocyst L/W ratio of 1.3 (1.0–1.6). Each sporocyst contained two sporozoites. Sporocyst residuum was present, composed of numerous granules in an ovoid mass.

Host: Mink (Neovison vison).

Locality: Denmark.

Prevalence: 2.6% (108 /4141).

Other hosts: Unknown.

Prepatent period: Unknown.

Patent period: Unknown.

Site of infection: Unknown.

Sporulation time: 5–6 days.

Material deposited: DNA sequences have been deposited in GenBank under accession numbers MG011726 and MG011727 for the 18S and COI loci, respectively.

Etymology: Morphologically, this new Eimeria isolate was similar to E. vison, which was originally identified from farmed American mink, but as no molecular data is available for E. vison, this new isolate is named E. vison-like.

Discussion

This is the first longitudinal study to report E. vison-like oocyst excretion in farmed mink across several Danish farms. In total, 2.6% of faecal samples from 315 mink were positive for E. vison-like oocysts, with farm prevalence ranging from 0.0 to 8.3%. Only a few previous studies have examined the prevalence of Eimeria spp. in Danish farmed mink (Hindsbo et al. 1995; Hammer et al. 2004; Chriél et al. 2013), and to our knowledge, only Hindsbo et al. (1995) have reported the prevalence (6.4%) of E. vison in Danish farmed mink from only one farm. Both female breeders and kits excreted oocysts. The prevalence was noticeably higher on farms located on Zealand compared to those located in Jutland. Lastly, there was no relationship between E. vison-like positive mink and diarrhoea.

Sporulated oocysts of E. vison-like were similar to, however, slightly smaller than the dimension of oocysts from E. vison, which were also ellipsoidal and measured 22.8 × 15.4 μm, with an oocyst L/W ratio of 1.5 (Levine 1948; Table 2). Unfortunately, no molecular characterisation has been carried out for E. vison. We are therefore unable to fully compare the isolate from Danish mink with E. vison. The other three Eimeria species described from mink are morphologically different from E. vison-like (Table 2).

Table 2 Comparison of the morphology of Eimeria spp. previously described from mink (Neovison vison) with the oocyst dimensions of E. vison-like from Danish mink farms
Full size table

Molecular characterisation of E. vison-like isolates at 18S locus revealed that it was most similar to E. sp. ex A. agrarius from the striped field mouse from the Czech Republic (JQ993655) based on 18S sequence phylogenetic analysis of 1221 bp. Analysis of shorter region of 18S (531 bp) grouped it next to the clade formed by E. furonis, which was originally identified from mink in Denmark and E. sp. ex A. agrarius. At the COI locus, E. vison-like was most closely related to E. ictidea isolates from the black footed ferret from Canada (KT203399) (99.4% genetic similarity). There are however morphological differences between E. ictidea and E. vison-like isolates. For example, oocysts of E. ictidea measure 23.0 × 17.7 μm (Hoare 1935), which is larger than E. vison-like (Table 2 ). Moreover, E. ictidea oocysts have been reported to possess an oocyst residuum (Jolley et al. 1994), which is absent from E. vison-like (Table 2 ).

There were no COI sequences available in GenBank from Eimeria sp. from minks.

In summary, this is the first report of the morphological and molecular characterisation of an Eimeria species in farmed mink (Neovision vison) from Denmark. Further characterisation of Eimeria species in American mink is necessary, using a combination of morphological, biological and molecular techniques.