Morphological description and multilocus genotyping of Onchocerca spp. in red deer (Cervus elaphus) in Switzerland

Onchocercosis is a parasitic disease caused by over 30 Onchocerca spp. (Nematoda: Filarioidea) and predominantly affecting ungulates. Four Onchocerca spp. have been described in the European red deer (Cervus elaphus). Onchocerca flexuosa and Onchocerca jakutensis form subcutaneous nodules in the back region. The other two species, Onchocerca skrjabini and the lesser-known Onchocerca garmsi, are found freely in the subcutaneous tissue of carpal and tarsal joints, and the sternal region, respectively. The presence of Onchocerca spp. in eight red deer shot in the hunting season during September 2020 in the Grisons region, Switzerland, was investigated by analysing nodules and free worms in the subcutaneous tissue. The obtained worms were morphologically and genetically identified as O. jakutensis, O. flexuosa and O. skrjabini. The latter two are first reports from Switzerland, and morphological redescriptions of these two species are presented. Onchocerca skrjabini and O. jakutensis are newly described from the sternal region of deer. One female of O. jakutensis was found free in the subcutaneous tissue of the sternal region, an atypical presentation for this species. Phylogenetic analyses were based on four mitochondrial and one nuclear loci, revealing that O. jakutensis belongs to a clade which so far only included non-cervid Onchocerca spp. Analysis of sequences from this study and GenBank entries revealed two distinct subpopulations of O. skrjabini: one from European red deer and another from Japanese serow and sika deer. Morphological identification can be challenging, also because worm location in the host is less strictly determined than previously described. Genetic identification is straightforward for O. flexuosa, O. jakutensis and O. skrjabini for which complete data of five loci are now available whereas genetic data of O. garmsi is still lacking.


Introduction
Onchocercosis is a filarial disease caused by over 30 described species of the genus Onchocerca (Filaroidea: Onchocercidae) which are transmitted by insect vectors (Anderson, 2000). Onchocerca spp. mainly affect ungulates (Bain et al., 2014;Uni et al., 2015), with two exceptions: O. lupi causes ocular infections in carnivores, and O. volvulus is the agent of river blindness in humans (Bain et al., 2014) which affects people mainly in tropical Africa and America (World Health Organization, 2022). In addition to O. volvulus, humans can suffer from zoonotic infections of different Onchocerca species, among others O. jakutensis (Koehsler et al., 2007;Wesołowska et al., 2020), which is one of the four species that infect European red deer (Cervus elaphus). The adult nematodes of O. jakutensis (syn. O. tubingensis) are found on the posterior part of the back, the pelvis area and the thighs (Schulz-Key, 1975b;Demiaszkiewicz, 1993;Bosch et al., 2016). Microfilariae are found in the skin overwhelmingly from the abdomen to the sternum and in the base of the ear (Schulz-Key, 1975b). Onchocerca flexuosa adults have overlapping predilection sites with O. jakutensis, their nodules being found in subcutaneous nodules in the back between the shoulder blades to the lumbar region, in the thigh region and rarely in the front legs (Schulz-Key, 1975b;Plenge-Bönig et al., 1995). Their microfilariae occur in large numbers in the skin of the caudal abdomen and the thighs (Schulz-Key, 1975b). Onchocerca skrjabini adults are free in the subcutaneous tissue of the tight skin around the carpal and tarsal joints. Microfilariae are abundant around the site of the adult worms but also on the nose and the base of the ears. The highest concentrations are found in the cutaneous tissue on the ears, especially on the outside (Schulz-Key, 1975b). The fourth species described in European red deer, O. garmsi, lives freely in the subcutaneous tissue above the sternum, and its microfilariae are mostly found around the ventral parts of the animal and the base of the ears, very similar to O. jakutensis (Schulz-Key et al., 1975).
In Switzerland, so far only adults of O. jakutensis have been recorded, with a prevalence of 24% in red deer in the Grisons region (Bosch et al., 2016). All four species have, however, been described in red deer from neighbouring Southern Germany (Schulz-Key et al., 1975;Schulz-Key, 1975b), and, therefore, it might be assumed that they also occur in Switzerland. The aim of this study was to further investigate the presence of Onchocerca spp. in Swiss red deer. In addition, two species are morphologically redescribed, and three species are genotyped using five loci. In view of the vast range of distribution of the two species O. flexuosa and O. skrjabini and the different hosts parasitised in European countries and in Japan, a more detailed morphological (drawings, photographs) and genetic characterisation of individuals was expedient. This is the first comprehensive morphological description of these two species, provided in English.

Origin of samples, worm preparation and morphological identification
The skins and carcasses of eight wild red deer (Cervus elaphus) were selected for investigation by the evident presence of Onchocerca nodules under the skin or in the fascia of the dorsal muscles. They were among the many deer shot during the official hunting season in September 2020 by certified hunters and were processed in the slaughterhouse in Bonaduz (Switzerland). All animals lived in the northern mountains of the Alps, in the catchment area of the river Rhine in the Grisons region, Switzerland.
Worms for further morphological and/or genetic characterisation were collected either from nodules or from the subcutaneous tissue of the skin of the carpal or sternal region. Samples of ca. 8 cm by 10 cm were cut off from the skin removed from deer. The presence of O. skrjabini was investigated by adding methylene blue, to stain the parasites, and by using the dissecting microscope (Fig. 1A). Reliable results were obtained after digesting the shaved skin tissue (see below) until only the nematodes remained (Fig. 1B).
Digestion of tissue and isolation of adult worms: The nodules and skin samples were incubated at 58 • C in a subtilisin-enzyme solution (10% v/v) (Enzyrim OSA, Bauer Handels GmbH, Switzerland), buffered at pH 8.0 (109888 Titrisol, Merck, Germany) and 2 drops of Mollescal-C (Bauer Handels GmbH) per 10 ml. Samples were incubated between 2 h (small nodules) to 12 h (thick skin samples). After lysis, specimens were fixed and stored in 70% ethanol and examined with a light microscope with differential interference contrast optics. Drawings were done using an optical Leitz drawing tube mounted on the microscope. The Onchocerca specimens were cleared in chloro-lactophenol if necessary. Identification was done according to the keys of Bain (1981) and other relevant descriptions (Bain and Schulz-Key, 1974;Demiaszkiewicz, 1989). No histopathology analyses were done.

DNA isolation
Additional nodules used for genetic characterisation only were dissected with scissors and forceps. A piece of about 2 cm or several smaller pieces of individual worms were put in 50 μl of PBS (phosphatebuffered saline) and homogenized for 30-60 s using a motor-driven pestle. DNA was then isolated with the Qiamp DNA mini kit (Qiagen, Germany) according to the protocol for DNA purification from tissues (elution volume 200 μl).

PCR and sequencing
PCRs targeting the mitochondrial NADH-dehydrogenase 5 (nd5), 12S rDNA, 16S rDNA, cytochrome c oxidase subunit 1 (cox1), and the nuclear 5S intergenic spacer sequence (5S IGS) were done with the primers and conditions listed in supp. Table 1. Amplifications were carried out in a C1000 touch thermal cycler (BioRad, Switzerland). A total volume of 50 μl per reaction was used, consisting of 19 μl ddH 2 O, 25 μl Multiplex PCR master mix (Qiagen), each 1 μl of the forward and reverse primer  Table 1. In cases of weak amplicon production, PCR was repeated with double the amount of DNA or by running up to 50 cycles. Amplicons were analysed on a 1.5% agarose gel stained with GelRed (Biotium, USA) and purified using the Minelute PCR purification kit (Qiagen). The 5S IGS amplicons of about 400 bp were excised from the agarose gels and extracted using the Minelute Gel extraction kit (Qiagen). Amplicon concentrations were measured using the NanoDrop ONEc (Thermo Scientific, USA), and the samples were prepared according to the sequencing company's information for economy run sequencing (Microsynth, Switzerland). Sequences containing ambiguous bases were additionally sequenced with the reverse primer. Sequences were submitted to GenBank under the accession numbers ON854615-ON854658 (Table 1).

Data analysis
Mega 11 (Tamura et al., 2021) was used for assessment of the electropherograms, alignments and phylogenetic analyses. A test for Best-Fit Substitution Model was run, and the substitution model with the lowest BIC scores (Bayesian Information Criterion) were selected to generate phylogenetic trees using the Maximum Likelihood mechanism with 500 bootstraps. GenBank accession numbers of additional sequences of filarial species used in this study are presented in supp. Table 2.

Results
A total of 86 Onchocerca specimens were morphologically identified from eight deer.
They belonged to Onchocerca jakutensis, a species already known to occur in Switzerland (Bosch et al., 2016), to O. flexuosa and O. skrjabini, which are both new records for Switzerland. The latter two species are therefore described in detail, both morphologically (see 3.1) and genetically (see 3.2).
Eight Onchocerca individuals were used for detailed morphological description. The same eight individuals and five others were subjected to detailed genetic analysis (Table 2). An additional 72 worms were retrieved from other nodules, putatively also suitable for antigen isolation in the frame of another project. All these worms were analysed at the nd5 locus, and a more detailed genetic analysis was done on 7 specimens ( Table 2; for features of these worms with regard to sex, etc., see supp. Table 3). Onchocerca jakutensis was identified in all eight deer with evident nodules, O. flexuosa in 5 out of 8. Onchocerca skrjabini was retrieved from the skin of the carpal region of 4 deer and also from the sternal region in 2 of them. Double and triple infections were recorded as shown in Table 2.
a This sequence is identical to that of another isolate and was therefore not separately submitted to GenBank. b In the alignments, this sequence is identical to the trimmed one of another isolate (indicated in brackets) but the whole sequence was submitted separately to the GenBank. c No sequence could be obtained.
+ = species verified morphologically; number of specimens (total = 85) identified at the nd5 genetic locus provided in parentheses; neg. = no infection with this species; n.i. = not investigated.
Nodules of O. jakutensis were found adhering to the skin from the lower back and femoral region embedded in the connective tissue of the subcutis. Most O. flexuosa specimens were isolated from nodules within the fascia of dorsal muscles along the back from the scapular region to the sacrum, and a few in the femoral region. The nodules of this species remained on the carcass after the skin was removed, whereas the nodules of O. jakutensis adhered to the removed skin. Females were obtained after digestion of the nodules allowing to collect unharmed individuals. Males were found inside or in the outer membranes around the nodules of O. flexuosa and O. jakutensis.
In contrast to these two species, which live convoluted in nodules, O. skrjabini was found outstretched and in large loops in the connective tissue from the hair follicles down to the periost. Only fragments could be collected of the relatively long females, most likely as a consequence of removing the skin from the carcass. Assigning the fragments to individual specimens becomes questionable, particularly if the worm burden is high (Fig. 1B). In addition, fewer fragments with tail/head endings were found amongst the numerous midbody parts. The extremities tend to get lost in the tissue remaining on the host animal. Fig. 2 documents the high worm burden in the sternal skin sample of deer D (Table 2). In this sample, the fragments with taxonomic elements were identified as O. skrjabini, but one fragment was identified as O. jakutensis using genetic analysis (Table 1, isolate OJ_E).

Morphological description
All worms used for morphological description were genetically confirmed and are marked in Table 1. In addition to the specimen used for drawings, the measurements of one more male and female specimen of O. flexuosa and two more male and female specimens of O. skrjabini are added in parentheses.

Onchocerca flexuosa
Adult worms are white-brownish, opalescent. The intestinal tract of some specimens is brown coloured, in other specimens, all internal organs are brown (Figs. 4 and 9). Mouth opening minute in both sexes, smaller than 2 μm in diameter. Oesophagus clearly divided into anterior muscular and wider and longer posterior glandular part. 3.1.1.2. Female. The enzymatically freed female was found wound up in many loops forming one lump. Body curled (Fig. 6), fragmented in 19 pieces, with total length of 378 mm. Cuticle of females heteromorphous over full length of body, in contrast to cuticles of males.

Table 3
Obtained sequences and their length and identity with sequences of Onchocerca species from GenBank. In parentheses: best matches with species with low identity (indicating lacking corresponding sequences in GenBank).  Surface of front and back end with fine rounded transversal annulations over a clear, transparent deeper cuticular layer. Surface of cuticle close to cloaca of one female forming Onchocerca-typical ridges with striae in deeper layer with ratio 1 : 4 respectively. Adjacent to midbody ridges of upper cuticle transformed to transversal rounded cushions, with striae underneath still visible (Fig. 7A). At midbody cuticle with rudimentary ridges on smooth surface, but basis of cuticle forming transversal ridges (Fig. 7B), changing over to undulation of entire cuticle (Fig. 7C). In some posterior parts cuticle transformed to transversally repeated pattern of one bigger swelling followed by two smaller swellings (Figs. 3G and 7D). Oesophagus 8.5 mm (6.9 mm) long, with slightly thicker and much longer glandular part. Nerve ring at 220 μm (200 μm), vulva at 370 μm (350 μm) from anterior end. The anterior 2 mm of body tapering slowly, then more rapidly from just behind the vulva to the blunt head end. Width at nerve ring 130 μm (120 μm), at vulva 180 μm (150 μm), at posterior end of oesophagus and midbody 200 μm (220 μm) (Fig. 3D). Ovejector enlarged to more than half of diameter of body ( Fig. 3E). Intestine and uteri forming loops inside spacious female body ( Fig. 8). Tail 310 μm (550 μm) long, conical, end rounded, with two prominent phasmids and two protuberances at tip (Fig. 3F). In other, degenerating females, the uterine tubes appeared very stretched inside a tightly undulated body, which becomes very fragile (Fig. 9).

Onchocerca skrjabini
Adult worms are whitish, opalescent very thin and difficult to detect by the naked eye in situ (Fig. 1). Mouth opening minute in both sexes, smaller than 2 μm in diameter. Oesophagus of one male and of two female specimens with well distinguished muscular and glandular part. In all other specimens the transition from the muscular to the glandular   μm (110 μm, 105 μm) long, both with well-developed heads. Left spicule with distal half flattened with margins bend ventrad to form a closed tube with clearly visible proximal and distal opening. Right spicule forming a ventral groove in distal half, the distal end enlarged dorsally with a subterminal rounded knob ( Fig. 10D and E). Papillae in a pericloacal and a distal group. An unpaired precloacal papilla, 4 pairs of papillae laterally to cloaca and one small pair medially behind the cloacal opening. Posterior group with 3 pairs of papillae: One terminal pair close to tip of tail and 2 subterminal pairs. Distribution slightly varying between specimens ( Fig. 10B and C). Pair of phasmids opening in a spine between the terminal and subterminal pair of papillae.
Cuticle with Onchocerca-typical two layers: Outer cuticle with marked transversal annulation, which fade out over the lateral line without touching each other, without or with only rudimental bifurcations (Fig. 13). The internal medullar part of cuticle transversely striated, ratio of annular ridges: striae is mostly 1 : 4 (Figs. 10G and 11). Towards head and tail end the annulation is smoothing out, becoming smaller and narrower (Fig. 10H).
Body diameter tapers rapidly towards anus. Tail with slight clubshape 170 μm (140 μm, 150 μm) in length, with diameter of 70 μm in all three specimens. Two pointed subterminal phasmids and a fine terminal knob at tail tip ( Fig. 10H and I).

Genetic analyses
Of the 85 worms identified at the nd5 locus, 20 specimens were chosen for additional analysis at three mitochondrial loci and one genomic locus (Table 1).
Obtained sequence lengths and their identity with Onchocerca spp. from the GenBank are given in Table 3. For all five loci of O. flexuosa (n = 6), high sequence identities with corresponding entries in GenBank were observed. Onchocerca jakutensis (n = 5) could be identified at four of the five loci. A novel sequence for this species was obtained for the 5S IGS (highest identity of 93.1% with sequence of O. gutturosa). Isolates OS_A to OS_I (n = 9) had high identities with sequences of O. skrjabini from Japanese sika deer (AM779804, AM749271) at two loci (12S rDNA: 97.1-97.7, cox1: 97.6-98.1%). No corresponding sequences were available at the other loci, highest identities were 92-93.1% (nd5), 95.5-95.7% (16S rDNA) and 81.8% (5S IGS) with sequences of different Onchocerca species. At the nd5 locus, there was an additional high identity (98.6-98.9%) with a sequence derived from uncharacterized microfilariae (KU886068) from red deer shot in the same region (Bosch et al., 2016). Morphologically, fragments of these specimens were identified as O. skrjabini after Bain and Schulz-Key (1974) (7 specimens) or, following Demiaszkiewicz (1989), as O. garmsi (2 specimens).
The 5S IGS PCR produced amplicons of around 450 bp, but sequencing often failed, and the obtained sequences of around 150-370 bp harboured multiple ambiguous bases indicating intra-individual polymorphism. For comparison with corresponding GenBank entries, a sequence of about 160 bp length was selected   Fig. 9. Detail of Onchocerca flexuosa female: Uterine tubes (arrow) straightened inside curly body. ON854655-ON854658). The O. skrjabini sequence harbours a 13 bp deletion which is unique compared to the other Onchocerca spp. sequences.
The sequences of nd5 and cox1 were translated into their corresponding protein sequences. The cox1 protein sequences were all identical between the three species. The nd5 protein sequences showed interspecies variation (13/133 amino acids). They were conserved within the species, showing no variation among the O. jakutensis sequences. One polymorphism each was evident among the sequences of O. flexuosa (V to I) and O. skrjabini (L to F), in both cases affecting amino acids belonging to groups of strongly similar properties.
For alignments, sequences were trimmed to lengths of corresponding GenBank entries, exemptions being indels.
A phylogenetic analysis of the concatenated sequences of the partial cox1 gene and 12S rDNA, including eleven other Onchocerca spp., placed O. jakutensis in a cluster with O. gutturosa and O. lienalis, a clade also including O. volvulus and O. ochengi (Fig. 14). This placement of O. jakutensis close to O. volvulus is supported by analyses, with fewer available sequences, of the concatenated partial nd5 gene, cox1 gene, 12S rDNA and 16S rDNA (supp. Fig. 1) and of the 5S IGS (supp. Fig. 2).
Sequence variability at the cox1 (Fig. 15) and 12S rDNA loci (supp. Fig. 3) of Onchocerca specimens from different countries was obvious in O. skrjabini from Japan and Switzerland as well as in O. flexuosa originating from Japan and several European countries.

Morphology
All specimens of O. flexuosa are well characterised and contrasted from the other Onchocerca spp. from the same host, considering length of oesophagus, location of vulva, form of male tail with spicule length and form, and distribution of the papillae. The morphological data given by Bain and Schulz-Key (1974) are confirmed with the description in this work. Additional typical characteristics found in this investigation are the spacious body of the female, the conspicuously swollen ovejector, the various forms of the heteromorphous cuticle and the brown colouration of aging specimens. Further, not mentioned by these authors is the curly form of the body of females (Fig. 6A). This form can also be considered as typical for O. flexuosa. The close up reveals stressed uteri with undulating wide integument around (Fig. 9). Schulz-Key (1975a) considered this aspect an artefact from stretching during preparation. In the present survey, stretched uteri were found regularly after enzymatic digestion of the nodules without prior handling the specimen. This was neither observed in O. skrjabini nor O. jakutensis after using the same digestion method (Fig. 6B). Schulz-Key (1975a) described the growth and decay of nodules in the host over time and the morphological changes of the females living inside. He noted that the female bodies change to the curly form in connection with the swelling of the cuticula. This might result in the barely visible or even lacking cuticular annulation (= ridges) in females noted above, also mentioned by Bain and Schulz-Key (1974) and Hidalgo et al. (2015). The illustrations given of the very different aspects of the heteromorphous cuticle in the female can be helpful in identifying pieces of worms removed from typical or atypical hosts. The aging process is accompanied by an increasing brown staining of the internal organs without involving the cuticle and terminated by multiple breakages of the body, until only crumbs of females are found. No calcification inside the parasite was seen, in contrast to O. jakutensis (Bosch et al., 2016).
The morphology of the O. skrjabini specimens investigated include most of the original description of Bain and Schulz-Key (1974) and the key in Bain (1981). However, one remarkable characteristic of this species not explicitly mentioned before is the very thin and long anterior part of the female body, which extends over at least 120 mm as measured in one specimen (Fig. 1B, arrows). The diameter increases posteriad steadily to 250-270 μm, to more than double the anterior size    . 11). The unpaired uterus inside the thin anterior part fills the available space, which provides a very typical aspect of tightness for O. skrjabini females (Figs. 10G and 12). Another species affecting cervids, O. eberhardi, was also reported with a narrow anterior body (Uni et al., 2007), though of only 3 mm in length. Bain and Schulz-Key (1974) and Bain (1981) described an unclear differentiation between the muscular and the glandular part of the oesophagus. However, the present work revealed one male and two female specimens with a clear differentiation, marked by an increase of diameter by 25% and by a massive presence of glandular cells. A clearly differentiated glandular oesophagus was also depicted by Yagi et al. (1994) for females of O. skrjabini retrieved from dwarf goats (Capra hircus) from Japan. They determined this as a slight morphological difference between Japanese and European specimens. It is interesting that such differentiated individuals were also found in deer from Switzerland.
The very long and thick microfilariae distinguish O. skrjabini from the other three Onchocerca spp. in the same host. Adults of O. skrjabini are easily distinguishable from O. flexuosa as described above. The differences to O. jakutensis and especially to O. garmsi are smaller. All these three Onchocerca spp. have a distance between head-end to vulva of more than four times the distance from head-end to nerve ring. Onchocerca skrjabini differs by its much longer oesophagus (>1000 μm) from O. jakutensis.
The differences of O. skrjabini to O. garmsi were described, but based on the fragments of one female only available at the time . Following these authors, the female of O. skrjabini is slightly smaller at midbody, but the measurements overlap with those of O. garmsi. Onchocerca skrjabini has a slightly smaller oesophagus and longer microfilariae. Onchocerca garmsi was described having a conical tail in the female, which distinguishes it from O. skrjabini and O. jakutensis. However, Demiaszkiewicz (1989) also found a fragment of a female with a cylindrical tail in his material from sternal skin and added it to his redescription of O. garmsi. Demiaszkiewicz (1989) gave the first description of the male of O. garmsi. The characteristic precloacal papillae of the male tail depicted allows to differentiate it from O. skrjabini. In contrast, he also found a fragment with a male-tail without this difference to O. skrjabini in the sternal skin and added it to the description of O. garmsi. The inclusion of these additional fragments obscures the differentiation between O. garmsi and O. skrjabini, especially when only fragments are available for identification.
Following the literature, the species O. garmsi could be found in the sternal skin Demiaszkiewicz, 1989). Considering the high worm burden found at this site (Fig. 2) and the here described occurrence of O. skrjabini in sternal skin raises the question whether the material of O. garmsi described by (Demiaszkiewicz, 1989) possibly was heterogenous.
In our investigation of the sternal skin, all fragments containing characteristics with taxonomic value (e.g. microfilariae, head parts, male tails) were identified as O. skrjabini after Bain and Schulz-Key (1974), confirmed by genetic analysis. The specimens of O. skrjabini in this investigation originate from the carpal skin and, in two hosts, also from the skin over the sternum. This is a new site reported for O. skrjabini in the deer.

Genetic identification
Genetic identifications at the nd5 locus were in good agreement with the morphological ones. A fragment of a female tail, found free in the digested sternal skin (OJ_E , Table 1), could not be identified morphologically (lack of characteristic features) and was genetically identified as O. jakutensis. Thus, besides the described O. garmsi and O. skrjabini, a third species can be found in the sternal skin. A further remarkable finding is the free-living female of O. jakutensis found in the sternal region; this is a new presentation for this species.
One male worm in this study was genetically assigned to O. flexuosa in a deer from which otherwise morphologically only O. jakutensis was identified, and no nodules were found that matched O. flexuosa (animal D, Table 2).
There are very few studies describing the occurrence of O. garmsi, and no reference sequences are deposited in GenBank. The availability of DNA sequences for clear identification would be an asset. Thus, we  (Nei and Kumar, 2000). The tree with the highest log likelihood (− 4758.48) is shown. Bootstrap values over 50 are shown next to the branches. Sequences newly generated in this study are in bold. requested a piece of stored O. garmsi specimens from the two European institutions that have published on this species (W. Stefański Institute of Parasitology, Warsaw, Poland; Muséum National d'Histoire Naturelle (MNHN), Paris, France), but unfortunately both had to decline our request, because the specimens were either no longer available or present the species holotype. Thus, the morphological description of O. garmsi correlated with genetic data awaits further investigation.

Genetic analyses
Genotyping provides a relatively simple way to determine the species of a located worm specimen. Initially, only the nd5 gene was analysed in this study. While O. flexuosa and O. jakutensis could both be accounted for with high conformity, the presumed O. skrjabini sequences had a high identity with the sequence of an undetermined species, but no significant match (max. 93%) with other entries in GenBank. Therefore, the analysis was extended to a number of genetic loci used previously for species identification and phylogenetic analysis of Onchocerca (Casiraghi et al., 2001;Krueger et al., 2007).
The aim of this study was not to provide a full-scale phylogenetic analysis but to provide a basic understanding on the relationship between the species O. flexuosa, O. jakutensis and O. skrjabini from this study. Lefoulon et al. (2017) used the concatenated sequences of seven loci (12S rDNA, cox1 and five others not addressed in the present study) from thirteen Onchocerca spp., including four of the seven known cervid Onchocerca, to derive a phylogenetic tree. They described three clades, with the cervid Onchocerca spp. belonging to clade I (O. cervipedis) or clade II (O. eberhardi, O. flexuosa, O. skrjabini). Onchocerca jakutensis was not included in that analysis. Our work places it in clade three, as the only species with a cervid host. Thus, the conclusion of Lefoulon et al. (2017) that a host switch among Bovidae, Canidae and humans occurred in this clade needs reconsideration.
While O. skrjabini have been described in European red deer, sequences found in GenBank all stem from specimens obtained in Japan from sika deer (Cervus nippon) or Japanese serow (Capricornis crispus) (Ferri et al., 2009;Lefoulon et al., 2015) and only from two mitochondrial genes (12S rDNA, cox1). A phylogenetic analysis including the sequences from our study revealed two subpopulations, one from European red deer and one from Japanese serow and sika deer (Fig. 15,  supp. Fig. 3). Lefoulon et al. (2017) showed the mean intraspecific nucleotide distance of the cox1 gene in most of the studied species to be lower than 2%. The difference between the O. skrjabini from Switzerland and those from Japan, with 2.48%, is slightly higher. Slight morphological differences of these Japanese isolates as compared to O. skrjabini from European red deer (Bain and Schulz-Key, 1974) were also described (Yagi et al., 1994), raising the question on their taxonomic status.
Genetic variability between O. flexuosa from two continents (Europe, Asia) at the cox1 locus exists (Fig. 15) but only a single sequence is available from Asia (Japan) (Abd-Ellatieff et al., 2022). 12S rDNA sequences only exist from European O. flexuosa isolates, without a geographic clustering.  (Tamura and Nei, 1993). The tree with the highest log likelihood (− 2902.13) is shown. Bootstrap values over 50 are shown next to the branches. Sequences newly generated in this study are in bold.
For O. jakutensis, sequences and descriptions exist only from European red deer, and no zoogeographic variability is obvious.
A few technical problems appeared during genotyping. The identity of the ninth worm morphologically described in this study in detail, morphologically O. skrjabini, was confirmed by sequencing but the quality of these sequences was low, and they were therefore not considered for extensive genetic analysis. The 16S rDNA locus was difficult to amplify with the O. skrjabini specimens, and the amount of DNA and number of replication cycles had to be increased to obtain at least faint bands in gel electrophoresis. 12S rDNA sequences of O. skrjabini generated with the forward primer were very short or of low quality. The reverse primer worked better, although the obtained sequences were shorter (median 372 bp) than those of the other two species (median 435 bp). The reason for this might be a poly T stretch in O. skrjabini around 90 bp after the binding site of the forward primer, possibly leading to polymerase slippage during sequencing. Ferri et al. (2009), the authors of the O. skrjabini sequences in GenBank, do not mention any problems with sequencing.
Gel electrophoresis of the 5S IGS PCRs showed multiple bands. Corresponding sequences have previously been generated by cloning the PCR products (Xie et al., 1994) or by excising the band of the expected size (Masatani et al., 2021) as also done in this study. Still, sequencing was unsatisfactory. Although both forward and reverse primers were applied, consensus sequences varied considerably in length, with copious ambiguous bases, and the amplification did not work at all in some cases. Showing intra-individual polymorphism, as noted by Krueger et al. (2007), sequencing with excised bands seems not to be an appropriate method.

Prevalence and vectors
With the selection procedure of deer in this study, no conclusions relating the prevalence of Onchocerca in the red deer population in this region is possible. The prevalence of O. flexuosa in Europe varies from 30.9% in Denmark (Nielsen et al., 2022), to 96.3% in Germany, where other Onchocerca spp. where also found with prevalences of 82% (O. skrjabini), 22.9% (O. jakutensis) and 2% (O. garmsi) (Schulz-Key, 1975b). In Switzerland, the prevalence of O. jakutensis was found to be comparable, with 24% (Bosch et al., 2016).
Two taxa, Simuliidae (black flies) and Ceratopogonidae (biting midges, genus Culicoides), are the incriminated vectors of Onchocerca (Post et al., 2007;Wang et al., 2020). Upon ingestion by the vector, microfilaria move to the insect thorax which can be a very rapid step as shown with O. cervicalis in its Culicoides vector (15 min; Mellor, 1975). There, they develop into infectious L3 which move to the insect mouth parts and are deposited on the skin during the insect feeding on the vertebrate host. Morphologic discrimination of the larval stages between species is difficult. There are only few studies on genetic identification of Onchocerca other than O. volvulus in their natural vectors. For example, PCR primers specific for O. fasciata were designed to identify C. puncticollis as vector (Wang et al., 2020). For the three Onchocerca spp. infecting red deer from this study, only studies based on morphology exist. Two species of Simuliidae were demonstrated to harbour L3 of O. skrjabini (Schulz-Key and Wenk, 1981) and O. flexuosa (Frank et al., 1968). A re-examination of the material of the latter study also revealed the presence of O. skrjabini (Schulz-Key and Wenk, 1981) which was not known at the time of the study. For Ceratopogonidae, the data on vector competence for the three Onchocerca spp. are scant. After feeding on a deer naturally infected with the three species, microfilaria of O. flexuosa and O. jakutensis were identified in the thorax, but developing larvae were only identified in one insect, in which case the Onchocerca sp. could not be determined, and no infective L3 were detected (Schulz-Key and Wenk, 1981). Further investigations with genetic identification of L3 recovered from biting midges are needed to clarify their vector role.
Several of the examined deer showed double or triple infections with Onchocerca spp. Bosch et al. (2016), examining deer from the same region as in this study, found only O. jakutensis. They only examined the deer skins and thus possibly missed both O. flexuosa which remains on the carcass after skinning, and O. skrjabini that does not form nodules. Taken together, the present work, in addition to Bosch et al. (2016), provides an advanced overview of the Onchocerca spp. found in red deer in the Grisons region in Switzerland, as morphological redescriptions and correlated multilocus genotyping.

Declaration of competing interest
The authors declare that there is no conflict of interest.