Introduction

Some of the most relevant human infectious diseases are vector-borne parasitosis such as the filarial worms responsible for lymphatic filariasis, like Wuchereria bancrofti Wilson, 1822 and Brugia malayi Brug, 1927. Many vector-borne parasites are also responsible for disease in domestic and wild animals. Filariods (superfamily Filarioidea) are nematode parasites that can be found in different organs of the definitive host, with a biological cycle characterized by the release of microfilariae into the blood vessels, which are ingested and transmitted to another primary host by an intermediate hematophagous host, an arthropod. Filaroids are frequently reported in Europe where some species are endemic1. These species have different pathogenicity for man and animals, representing a growing concern1,2,3.

While, for example, Dirofilaria immitis threatens dogs and cats, causing a severe and often fatal cardiocirculatory disease referred to as ‘heartworm disease’, the zoonotic Dirofilaria repens Railliet & Henry, 1911 induces a non-pathogenic subcutaneous or connective muscular fasciae infestation in dogs but is more frequently found in man4.

Until now only a few species of filarial worms were described in hares, namely from Dirofilaria and Micipsella genera. The Micipsella genus includes three different species, namely M. numidica Seurat, 1917, first described in rodents5 but also reported in 2016 in the European hare (Lepus europaeus)6, M. indica Rao, 1938 described in the Indian hare (Lepus nigricollis) on India7, and M. brevicaudata Lyons et Hansen, 1961 described in the American desert hare (Lepus californicus)8. A revision of the most relevant features of all Onchocercidae described so far on lagomorphs, including several hare species and cottontail rabbits, is summarised in Table 1. Presently, there are no molecular studies in the Iberian Peninsula that have unequivocally identified the species of parasites that have been found. In onchocercids (member of Superfamily Filariodea, Family Onchocercidae), also known as filarial worms or filarioids, the vulva is posterior to the nerve ring, contrarily to filariids (members of Family Filariidae), where the vulva is anterior to the nerve ring.

Table 1 Information on relevant filarial species found in rabbits and hares.

Although fundamental, the classic morphology-based identification has proved insufficient for nematode identification, as well as to understand the existent phenotypic diversity, mainly due to the sparse morphological variations found among closely related taxa. Given the high degree of morphological intraspecific variability and the interspecific similarities, taxonomical classification is sometimes extremely complicated and controversial. For instance, regarding genus Lernaea, morphology is not a reliable taxonomic tool as molecular data and experimental infection reveal that L. cyprinacea Linnaeus, 1758 and L. cruciata Le Sueur, 1824 are conspecific9.

Different molecular methods, ranging from fingerprint to sequencing analyses, but also protein-based information, have been used to complement morphology-based data and circumvent their limitations. Reclassification of organisms is often based on genetic data, rather than on the phenotype or other biological features10. Partial sequences of mitochondrial and nuclear genes have been largely used to classify species of eukaryotic parasites and revolutionized our understanding of the distribution and evolutionary history of several parasites occurring worldwide11. In these studies, partial sequences of mitochondrial ((12S rRNA and coxI (cytochrome oxidase subunit I)) and nuclear genes (18S rDNA (that encodes the 18S rRNA), myoHC (that encodes the myosin heavy chain), hsp70 (that encodes the 70 kilodalton heat shock protein), rbp1(that encodes the RNA polymerase II large subunit)), were used for species identification and comparison12,13,14. Furthermore, the combined analyses of nuclear and mitochondrial markers have proved successful in species discrimination or to explore available molecular data15. Recent studies have described new parasite species by concatenating several genes sequences, in order to increase the discriminatory power of the phylogenetic analyses when comparing onchocercid species13,16,17.

In this study, we report the presence of filarial parasites in two adult Iberian hares, carrying out an in-depth investigation regarding their taxonomic classification. The results support the occurrence of a new species of Micipsella in Iberian hares. The hares in which the filarial worms were observed were received by the National Reference Laboratory for Animal Diseases, within the scope of a wild leporid populations health status assessment carried out in Portugal between August 2017 and June 2020 by the Project +Coelho (Dispatch 4757/17, 31th may). During this period, dozens of dead Iberian hares were submitted to necropsy and pathological examination.

Materials and methods

Sample origin

The animals used in this study were collected dead from the field. No animals were handled or euthanized for the purposes of this study.

An adult male Iberian hare (hare-1, 25456PT19) with a poor body condition (1.7 kg), was collected in Beja district, Portugal, on the 22nd of August of 2019. A second specimen, an adult male (hare-2, 35468PT19) with a good body condition (2.2 kg) was collected in Portalegre district, Portugal, on the 10th of November 2019. While the first animal arrived fresh at the Instituto Nacional de Investigação Agrária e Veterinária (INIAV, I.P.), the second was received frozen. Several organ samples from both hares collected during the necropsy were fixated in 10% neutral buffered formalin (w/v), routinely paraffin-embedded, sectioned at 4 µm, and stained with Hematoxylin and Eosin (H&E). During necropsy, three filarial parasites were observed in hare-1 and four parasites in hare-2. Those specimens were collected and preserved in 70% ethanol for morphological identification.

Morphological characterization

Morphometric analysis was based on one adult filaria from hare-1, and three adults from hare-2. The length of the filariae was measured in a plastic tray before a small middle section was cut for molecular analysis. For microscopic visualization, the anterior and posterior parts of the parasites were mounted on glass slides using lactophenol, for clarification. All observations and measurements were carried out on a Leica DM IL LED Inverted Microscope (Leica, Wetzlar, Germany) and photographed using a LEICA EC3 photography system (Leica, Wetzlar, Germany). Macroscopic drawings were performed manually, using an Olympus BX51 microscope with an Olympus™ Drawing Attachment Tube (Olympus, Hamburg, Germany).

DNA extraction

For nucleic acid extraction, 5 mm fragments of the middle section of six parasites were incubated with 20 μL proteinase K (600 mAU/mL) in 200 μl PBS (w/v) and submitted to extraction using the MagAttract 96 cador Pathogen Kit (Qiagen, Hilden, Germany) in a BioSprint 96 nucleic acid extractor (Qiagen, Hilden, Germany), according to the manufacturer’s protocol. DNA concentration was determined by A260 measurement (Qubit 4 Fluorimeter by Invitrogen, California, USA). Nucleic acids were preserved at − 20 °C until use.

Molecular characterization

Polymerase chain reactions (PCR) targeting mitochondrial and nuclear genes were performed in a final volume of 25 μl under the following final conditions: 1x buffer including 1.5 mM MgCl2, 0.2 mM of each deoxynucleotide triphosphate (dNTP), 1 mM each of forward and reverse primers, 1 unit of the DNA polymerase (High Fidelity DNA Polymerase, Roche, Basel, Switzerland) and 20 ng of genomic DNA. To assess the specificity of the reactions, DNA extracted from Dirofilaria immitis Leidy, 1856, was included in each PCR, as a positive control. Several pairs of primers were used to amplify the partial sequences of two mitochondrial genes (12S rDNA (~450 bp) and coxI (~600 bp)), and four nuclear genes (18S rDNA (~740 bp), myoHC (~785 bp), rbp1 (~ 640 bp) and hsp70 (~610 bp)), as described by Lefoulon et al.13.

Amplicons from two parasites (one from each hare) were purified using the NZYGelpure kit (NzyTech, Lisbon, Portugal) following the manufacturer’s instructions and then directly sequenced using the ABI Prism BigDye Terminator v3.1 Cycle sequencing kit on a 3130 Genetic Analyser (Applied Biosystems, Foster City, CA, U.S.A). The obtained nucleotide sequences (Table 2) were analysed and assembled into consensus sequences using the BioEdit version 7.2.5 software and submitted to GenBank (Genbank access numbers are provided as Supplementary data, Table S1). Nucleotide sequences were translated using MegaX10.1 software.

Table 2 Percentage of nucleotide similarity between M. iberica genes obtained from hare-1 and hare-2 and other filarioid genes sequences from public databases.

Phylogenetic analysis

The partial sequences of each gene were aligned separately using Mega X software and manually assembled. For phylogenetic analysis based on peptide and nucleotide alignments, doubtful residues, and the nucleotides in the same position, were removed, in all sequences. The 12S rDNA, 18S rDNA, coxI, myoHC, hsp70 and rbp1 partial nucleotide sequences of the filaroids understudy were concatenated to obtain a more robust phylogenetic analysis, according to Lefoulon et al.13. The two concatenated sequences obtained separately from hare-1 and hare-2 filariae, were aligned with the homologous concatemers constructed from sequences available in the database. To allow comparison with previous works, most of the sequences used by Lefoulon et al.13, were also included in our phylogenetic analysis.

The phylogenetic trees were obtained in Mega X, by Maximum Likelihood using the General Time Reversible model (GTR), with gamma distribution and/or allowance of invariant sites and 10 Gamma Categories (Gamma with invariant sites option). Evolutionary analyses were conducted in MEGA X. Pairwise identities and the heat map for nucleotide identity matrix were calculated using SDT v1.229.

Declaration

This study did not use live animals and was carried out within the scope of a National Plan for the Control of Rabbit Haemorrhagic Disease Virus 2 in rabbits (Dispatch no. 4757/2017 of 31 May), with the legal authorisations from the National Authority-the Institute for Nature Conservation and Forests (Instituto da Conservação da Natureza e das Florestas, I.P. ICNF).

Results

Necropsy and histopathology

At necropsy, three alive filarial worms were collected from the posterior vena cava of hare-1 and four dead specimens were obtained from the thoracic cavity of hare-2, more specifically from the interior of the vessels next to the diaphragm (Figs. 1 and 2). The worms were preserved in 70% ethanol for morphological and molecular characterization. Severe microfilaremia was found during the histopathological analysis of hare-1, mainly in the liver (Fig. 3), lungs, kidneys (Fig. 4), and spleen, but also in other organs such as the testis.

Figure 1
figure 1

Necropsy from hare-1. Incision of posterior vena cava, in its passage through the diaphragm, exposed several specimens (black arrows) of parasites, still with mobility.

Figure 2
figure 2

Necropsy from hare-1. One of the filariae observed is held by tweezers and scissors.

Figure 3
figure 3

Liver from hare-1. Presence of a microfilaria inside a sinusoid. ×400, H&E.

Figure 4
figure 4

Kidney from hare-1. Presence of a microfilaria in an interstitial tissue vessel. ×400, H&E.

Morphological characterization

Two female filariae and one fragmented specimen of undetermined gender were found in hare-1. A total of three female filariae and one fragmented of undetermined gender were observed in hare-2. Regarding the four complete mature specimens (n = 4 females), the mean measurements included a 157 mm (153–165) length, a 652.5 μm (550–780) width, a 740 μm (720–750) oesophagus and distance ranging from the vulva to the anterior extremity of 684 μm (620–800). The nematodes presented a filiform body morphology and tapered ends (Figures 5 and 6). Their extremity was elongated, digitiform, and rounded, and showed a rectilinear oesophagus. In the blood smears, the mean length of the microfilaria was 107.94±3.32 μm.

Figure 5
figure 5

(A) Anterior extremity of a female specimen from hare-1 (×100), (B) Joint zone of the oesophagus with the intestine (arrow) and vulva (arrow head), female, hare-1 (×200), (C) Posterior extremity of a female specimen, hare-1. (×100).

Figure 6
figure 6

Schematic view of anterior and posterior extremities of a female specimen. (A) Posterior part, immature female, ventral view; (B) anterior part, ventral view; (C) posterior part, lateral view; (D) anterior part, ventral view; (E) anterior part, lateral view. Abbreviations: an-anus, oe-oesophagus, vu-vulva, nr-nerve ring, in-intestine, pap-papilla.

Molecular characterization

Amplicons with the expected sizes were obtained by conventional PCR for all genes of the two filariae’ DNA preparations. For all genes investigated, the nucleotide sequences from one filaroid obtained from hare-1 and another from hare-2, showed 100% identity between each other, exception made for the 18S rRNA gene, where a 2-nucleotide insertion was observed in the filariae from hare-1, when compared to that from hare-2.

The concatenated sequences for 12S rDNA, 18S rDNA, coxI, myoHC, hsp70 and rbp1 genes from hare-1 were used to generate a ML tree (not shown). The identified filarial worms in this study showed a closer phylogenetic relation with Rumenfilaria andersoni Lankester & Snider, 1982. The unavailability in public databases of 12S rDNA, myoHC, hsp70 and rbp1 sequences for Micipsella species prevented us from including these sequences in our phylogenetic analysis. For this reason, a second concatenated alignment, including only the 18S rDNA and coxI, was constructed to compare the filariae sequences from hare-1 and hare-2 with those publicly available from Micipsella numidica.

Figure 7 shows the heat map of pairwise identity using the five gene concatomer. In accordance with the phylogenetic inference, the higher similarity was observed with Rumenfilaria andersoni (~94%).

The phylogenetic analysis using concatenated partial sequences from coxI and 12S rDNA allowed to include a sequence from M. numidica (KR232089) previously described in other species of hares namely European hare6. The unrooted circular phylogenetic tree is shown in Figure 8.

Similarly, the heat map of pairwise identity of 12S rDNA and coxI concatemer using a muscle alignment showed that the studied specimens have the highest similarities (~88–90%) with M. numidica, Chanderella quiscali Linstow, 1904 and Dirofilaria repens Railliet & Henry, 1911 (Figure 9).

Figure 7
figure 7

Heat map of nucleotide identity using concatenated partial DNA sequences from 18S rDNA, 12S rDNA, coxI, myoHC and hsp70 genes.

Figure 8
figure 8

Phylogenetic tree using concatenated partial DNA sequences from 18S rDNA and coxI. Ancestral states were inferred by Maximum Likelihood using the General Time Reversible (GTR) model. The tree shows a set of possible nucleotides (states) at each ancestral node based on their inferred likelihood at site 1. This analysis involved 27 nucleotide sequences. There was a total of 777 positions in the final dataset. Evolutionary analyses were conducted in MEGA X. Genbank access numbers are provided as Supplementary data, Table S1.

Figure 9
figure 9

Heat map of nucleotide identity using concatenated partial DNA sequences from 12S rDNA and coxI.

Discussion

In 2019, the International Conservation Union (IUCN) attributed the status of “minor concern” to Lepus granatensis in the Red Book of Threatened Species, although a declining trend was recognized30. Adding to this continuous reduction, new diseases were recently identified in the species, impacting severely in the survival many local wild populations30,31,32. One of the major parasites threats to Iberian hare is Cysticercus pisiformis, the larval stage of Taenia pisiformis. Interestingly, during a 35-month survey carried out within the scope of a national surveillance programme in continental Portugal, the incidence of internal parasitism in Iberian hares was lower than in wild rabbits33. In this manuscript, we describe the presence of filarial parasites in two adult Iberian hares from South Portugal. Being must less abundant than wild rabbits, the restrictions inherent to the opportunistic sampling of Iberian hares limited the number of filarial worms available for the analysis. Even though no male filarioid specimens were collected from the two hares, the morphological and genetic characteristics based in 12S rDNA and coxI partial sequences, showed that the filarioid species found resemble Micipsella species.

The prevalence of this parasite in the wild populations and its real pathogenic potential is unknown. However, given the adults’ dimensions and their location in the blood vessels, the risk of vascular thrombosis is unavoidable.

Currently, only a few studies are available on onchocercid from hares and rabbits. Known filarioid species from lagomorphs were included in five different genera, namely Micipsella, Loaina, Cercopithifilaria, Brugia and Dirofilaria.Genus Micipsella (Onchocercidae) comprises three recognized species, namely M. numidica (Seurat, 1917), M. indica (Rao, 1938) and M. brevicaudata8. Genus Loaina20 was created to include Dirofilaria species of rabbits from North America and comprises two species, Loaina scapiceps (Leidy, 1886) and Loaina uniformis24, the latter considered as the type species. This genus is morphologically distinct from Dirofilaria and other Dirofilariinae20. Loaina uniformis has been reported from subcutaneous tissues of several rabbit species (Sylvilagus floridanus, S. palustris, S. aquaticus) in different USA states. Loaina scapiceps, infecting the tarsal bursa of the hind feet, has also been reported in the mentioned above rabbit species as well as in different hare species (Lepus americanus, L. campestris, L. washingtonii). One Brugia species has been described on the abdominal lymphatics and subcutaneous tissues of rabbits (Sylvilagus aquaticus, S. floridanus)28. Brugia leporis20 was reported in rabbits in Louisiana, USA. Cercopithifilaria leporinus26 was described as a new species from the subcutaneous tissues of the trunk of snowshoe hares (Lepus americanus) in Canada26. Another species, Dirofilaria timidi25, was regarded as a species inquirenda. Only partially described, its proper taxonomic group proved difficult to be determined since it does not belong in Dirofilaria genus.

Micipsella numidica lives in the peritoneal cavity and circulatory system of hares being more rarely found in rabbits. Available information refers that M. numidica does not have a very strict location within the host having been found also in the abdominal cavity, between the intestinal mesenteries, in the circulatory system, particularly in the portal vein, the supra-hepatic vein or the capillaries of the greater omentum. Since the portal vein and its direct tributaries are not systematically opened at necropsy, the number of parasitised animals is certainly underestimated and this parasite is probably more widespread than thought. In other hare species, the white-tailed jackrabbits (Lepus townsendi) from the USA and the Indian hare (Lepus nigricollis), two additional species of Micipsella are currently known, namely M. brevicaudata8, a parasite of the peritoneal cavity and M. indica7 found in the circulatory system (heart and portal vein), respectively.

Micipsella numidica was first described in Algeria in 19175 under the name Filaria numidica but in 1921 Seurat created the genus Micipsella for this nematode. The first specimens were collected from the abdominal cavity of Desert hares (Lepus capensis). Since then, the parasite has been identified several times in different regions of the world, being reported in Armenia34, Mongolia35, Hungary18 and in Italy6.

Micipsella numidica is a thread-like nematode, tapering at both ends. The tail, in both sexes, is elongated, digitiform and rounded. The anterior end is thinned and forms a hemispherical cap with a mouth at the top, followed by an undivided, rectilinear oesophagus, which is followed by an elongated intestine that widens considerably at its origin. The nerve ring is situated in the anterior third of the oesophagus. The thick cuticle is generally devoid of striation.

Mature female parasites from our study have a length ranging from 153 mm to 165 mm, which are greater than those described by Seurat5 (70–140 mm) and Ivashkin (60–93 mm). A similar width and distance of oesophagus from the cephalic end was found between mature females from the three studies (quais?). The vulva opens at 620-800 μm from the cephalic end, similarly to the specimens described by Seurat. Since no male specimens were collected in the current study, no information could be obtained regarding the male morphological distinctive features.

The microfilariae of M. numidica measure 95-189 μm long by 3.55-4.4 μm wide. They have no sheath and the posterior end is rounded. Micipsella indica is a large Filaria (male, 70 to 100 mm; female, 120 to 140 mm).

The morphology of nematodes is remarkably constrained36, exemplified by the challenge that Caenorhabditis elegans Maupas, 1900 and C. briggsae Briggs, 1944 morphological discrimination pose to most trained nematode taxonomists, besides an estimated date of divergence of 80–110 Mya37, long before the segregation between the mouse and human lineages. Since cryptic species must abound in the phylum Nematoda, molecular-based techniques are the only practical approach to recognize and differentiate36.

Misidentifications based on morphological approaches resulted in huge economic losses around the world38. Nowadays, molecular methods allow the recognition of many new taxa, some based on sequence information alone39. Per these authors, the high resolution of sequencing analysis overcomes the limited capacity of morphological, image-based and protein-based methods. Agreeing with this, other authors consider that molecular data is better than morphological data to support phylogenetic studies40,41,42.

The strategy of sequencing different genes and performing a concatenated phylogenetic analysis is being increasingly used worldwide, notably using these aforementioned genes and others such as myoHC and hsp70, questioning the previously existing classifications based on smaller sequences or on morphology13. In this study, we combined mitochondrial DNA barcodes with nuclear data to circumvent the downsides linked to maternal inheritance. For this reason, we used partial sequencing data of set of mitochondrial and nuclear genes to obtain a phylogenetic classification for the two filariae specimens found. The targeted genes included coxI, 12S rDNA, and 18S rDNA, but also hsp70, myoHC and rbp1.

Cytochrome c oxidase subunit 1 mitochondrial gene (coxI or COI) is a standardized molecular marker for the comparison and classification of animal species43 discriminating vertebrates and invertebrates44,45,46, demonstrating its power as a marker for identification from other closely related animal species. The coxI gene appears to provide a better phylogenetic signal than the other mitochondrial genes (e.g. 12S rDNA, 16S rDNA)47,48. This is thought to be the result from the rapid evolution of coxI gene, which allows discriminating between closely related species and to investigate intraspecific diversity43. Besides the identification of already known as well as new species, sequencing of coxI gene was suggested as a standard for cryptic taxa discovery, an association of different life stages of the same species and wildlife conservation genetics45.

The mitochondrial 12S rDNA is a genetic marker useful to study the molecular systematics of nematodes and to reveal intra-phyla relationships49. It is often concatenad to other mitochondrial genes such as coxI13 and 16S DNA49, to potentiate the discriminatory power of the nucleotide variability.

The small subunit 18S rDNA gene is the most frequently used marker for taxonomic identification in eukaryotes, phylogeny and evolution investigations50,51.

The ribosomal internal transcribed spacer (ITS) region has shown insufficient discriminatory power for nematode classification. However, the nucleotide sequences of a fragment within the small subunit nuclear ribosomal DNA (18S rDNA or SSU), provides adequate information to identify genera of nematodes and may even distinguish between species52.

BLAST analysis of the 12S rDNA and coxI partial sequences of the filariae under study revealed the highest similarities with M. numidica (Splendidofilariinae), 90.06% and 91.35%, respectively (Table 2). As refered before, no sequences are presently available for Micipsella genus regarding genes myoHC, hsp70 and rbp1 and 18S rDNA. Furthermore, given the lower resolution at lower taxonomic levels of nuclear 18S rDNA compared with mitochondrial 12S rDNA49, the 18S rDNA sequence of the specimens being described showed 99.15% similarity both with Rumenfilaria andersoni (KP760163) (Spendidofilariinae) and Filarioid sp. (EF081340) (Filarioidea). Also, regarding the partial sequences of myoHC, hsp70 and rbp1 genes obtained from the filaroids reported here, the higher similarities were observed with Rumenfilaria andersoni (96.97%)(KP760258), Rumenfilaria andersoni (92.52%)(KP760456) and Rumenfilaria andersoni (96.23%)(KP760309), respectively. The genetic distances (Figs. 7 and 9) corroborated the phylogenetic findings.

The pairwise identities based in coxI and 12S rDNA concatenated sequences showed the higher values between species of the same genus, namely ~90% between Dirofilaria immitis (Dirofilariinae) and D. repens, and 88% between Brugia malai (Onchocercinae) and B. timori. The similarity between the studied specimens and M. numidica of the same order of magnitude (91%), expose the remarkable genetic proximity with M. numidica with which they may share the same genus. The coxI gene contains a higher level of sequence diversity, particularly in the variable regions, making this region ideal for resolution at lower taxonomic levels39.

In conclusion, using molecular data, the parasites were identified as phylogenetically closer to Micipsella numidica despite differing from it in 8.65% and 9.94% nucleotide similarity (Table 2). Particularly, the relatively low genetic identity in the coxI gene from the two filariods is incompatible with a same species. Along with the morphological differences registered, namely larger females, these findings support that Micipsella described in this study and M. numidica are not the same species exhibiting morphological intraspecific variability, but instead, two different species. Since the report of M. numidica in Iberian hares (Segovia, 2014) was not accompanied by molecular and morphometric data, there is no certainty that the specimens reported at that time in Spain correspond in fact to the species M. numidica, and not to the one here described. The possibility that several subspecies may exist within the species M. numidica, with the two specimens under analysis being subspecies within this taxon, cannot be further investigated given the scarcity of genetic information available in the public databases to compare with.