CHARACTERIZATION OF LESIONS INDUCED BY SPIROCERCA VULPIS (SPIRURIDAE: SPIROCERCIDAE) IN RED FOXES (VULPES VULPES)

Spirocerca lupi (Rudolphi 1809) is a parasitic helminth mainly associated with the domestic dog (Canis lupus familiaris). It induces the formation of esophageal nodules that can transform into osteosarcoma and fibrosarcoma in chronic infections (Lobetti 2000; Ranen et al. 2004, 2008; van der Merwe et al. 2008). Up to 25% of these nodules may undergo neoplastic transformation (Dvir et al. 2001). The parasite has an indirect life cycle involving several hosts, including coprophagous beetles as the intermediate host, with birds, reptiles, amphibians, and small mammals as paratenic hosts, and the definitive hosts (DHs) are carnivores such as domestic dogs (Canis lupus familiaris; Bailey 1972) and possibly the red fox (Vulpes vulpes L.), the Iberian lynx (Lynx pardinus L.), and the Iberian wolf (Canis lupus signatus; Rojas et al. 2018a; Valcárcel et al. 2018). A DH is typically infected by ingesting an intermediate host or a paratenic host containing infective larvae (L3). Once in the DH's stomach, the L3 hatch and penetrate the gastric wall, reaching the aorta approximately 3 wk after ingestion, then requiring a further 10–12 wk for maturation. Mature nematodes then leave the aorta and migrate to the esophagus (Lora-Michiels et al. 2003; van der Merwe et al. 2008).

In dogs, Spirocerca-induced lesions in natural infestations have been analyzed (Dvir et al. 2001, 2010). According to the degree of pathogensis, two stages were established in nonneoplastic lesions (Dvir et al. 2010): early inflammation characterized by fibrocytes and abundant collagen, and a preneoplastic stage characterized by activated fibroblasts and reduced collagen. The nodule ultimately evolves into a malignant sarcoma (Dvir et al. 2011).

The red fox is one of the most widespread European wild canids and is considered a key species under the One Health framework, acting as a major reservoir for extraintestinal nematodes transmissible to other animals and to humans. Recent studies (e.g., Rojas et al. 2018b; Martín-Pérez et al. 2020) indicate that infection in the red fox is due to a newly described species of Spirocerca (S. vulpis). Reanalysis of S. lupi–like organisms obtained from red foxes in previous reports may reveal they are S. vulpis, as Rojas et al. (2020) confirmed in nematodes from Switzerland identified after being stored for more than 20 years. In western Spain, our group observed that 6% of the red foxes analyzed had S. vulpis, using molecular and morphological analysis, and no S. lupi was found (Martín-Pérez et al. 2020).

In red foxes, Spirocerca-associated lesions typically occur as gastric wall nodules (Segovia et al. 2001; Ferrantelli et al. 2010; Diakou et al. 2012; Al-Sabi et al. 2014). The histopathological lesion is characterized by granulomatous cellular infiltrates surrounding nematodes. The parasites are found in a central cavitary space surrounded by an eosinophilic and granular exudate (Rojas et al. 2018b). Neoplastic forms have not yet been described in foxes, and there is a lack of studies regarding the evolution of Spirocerca-associated lesions. Our study aimed to determine the nature of the lesions induced by S. vulpis in red foxes from Spain.

The study was carried out in five provinces (Salamanca, Ávila, Zamora, Valladolid, and Cáceres) in western Spain (41°13′46″N, 5°29′48″W). Red foxes are hunted for predator population control under licenses authorized by the regional governments of the autonomous communities of Castile and León and of Extremadura (Law 9/ 2019, of 28 March, amending Law 4/1996, of 12 July, on Hunting in the Autonomous Community of Castile and León; Law 14/2010, of 9 December, on Hunting in Extremadura). A total of 1936 legally hunted or road-killed red foxes were harvested from 2016 to 2018. Thus, the animals were not euthanized specifically for this study, and no ethics committee approval was necessary. The approximate age of the individuals was determined by analysis of the dental formula (Saenz de Buruaga et al. 1991), differences in harvest date from the most probable date of birth (foxes in the Mediterranean area produce cubs during the spring; Voigt and Macdonald 1984; Zapata et al. 1997), and the external appearance of the fox based on researcher experience. Animals were classified as young (under 1 yr old), subadults (between 1 and 2 yr old), and adults (more than 2 yr old).

Foxes were necropsied in the field and systematically examined to assess visible lesions of Spirocerca spp. in gastric walls and abdominal and thoracic organs. A total of 163 had Spirocerca spp.–like compatible lesions. Of these, 71 animals without autolysis were selected, with one nodule being examined per fox.

Nodules (n=71) were measured with a caliper to determine their macroscopic size. For a good understanding and statistical evaluation, three ranges were established for nodule sizes: <1.2 cm, >1.2 and <3 cm, and >3 cm.

Genomic DNA from individual specimens was isolated using an NZY tissue gDNA kit (NZYTech, Lisbon, Portugal). Amplification of DNA was performed as described for Spirurida spp. (Casiraghi et al. 2001). From each sample, PCR products amplified were purified and sequenced by STABvida (Monte da Caparica, Setúbal, Portugal) in forward and reverse senses, using the same primers as for the PCR. Sequences were edited in Chromas Lite 2.1.1 (Technelysium, Queensland, Australia), and consensus sequences for each forward-reverse pair were created in BioEdit, using CLUSTAL-W version 2.0 (University College Dublin, Dublin, Ireland). Obtained sequences with more than 600 base pair (n=20) were subjected to BLASTn analyses (National Center for Biotechnology Information 2020). The identity at the species level was based on sequence homology both taking into consideration the higher similarity in the BOLD Systems identification tool (BOLD Systems 2020) and with the BLASTn tool of the sequences deposited in GenBank (Benson et al. 2005).

Nodules (n=71) were fixed in neutral buffered formalin (3.5%, 0.1 M, and pH 7.2), routinely processed, and embedded in paraffin wax. Five-micron sections were stained with H&E to assess the microscopic changes (necrosis, hemorrhages, and collagen) and to count eosinophils and neutrophils. Microphotographs were taken using a microscope (Eclipse 80i, Nikon, Tokyo, Japan) with a digital camera (DXMI200F, Nikon).

Samples were analyzed by two different pathologists using a light microscope (20× magnification). Lesion classification was performed according to cell distribution. Lesions were classified as diffuse when a widespread pattern was observed. Lesions were classified as patchy if the cell distribution was focal or multifocal. When both patterns (patchy and diffuse) were observed in the same sample, it was classified as a mixed lesion.

Ten samples of each type of lesion (patchy and diffuse) were selected for immunohistochemical analysis as well as 10 control samples taken from the gastric walls of noninfected foxes. Only two samples had mixed lesions; these were not analyzed by IHC, due to extensive autolysis.

Avidin-biotin complex (ABC Vector Elite, Vector Laboratories, Burlingame, California, USA) was used for immunolabeling. All samples were dewaxed in an oven, rehydrated, and then treated in 3% hydrogen peroxide in methanol for 15 min to quench the peroxidase activity. Samples were washed with tris-buffered saline (TBS) 0.01 M and pH 7.2. Antigen retrieval methods included enzymatic digestion with trypsin/alphachymotrypsin (0.5% trypsin and 0.5% alphachymotrypsin, Sigma-Aldrich, Gillingham, Dorset, UK) at 37 C for 10 min or heat-treated by microwave using tris-ethylenediaminetetraacetic acid pH 9.0; 20 min, and 700 W. Samples were then mounted on a Sequenza immunostaining center (Shandon Scientific, Runcorn, UK). Primary antibody cross-reactivity with tissue constituents was prevented by using 1.5% normal serum block, which matched the host species of the link antibody that was applied to the sections for 20 min. Details of the primary antibodies used, their specificity, concentration, and incubation time are summarized in Table 1. Sections were washed in TBS and then incubated for 30 min with the appropriate biotinylated secondary link antibody (Vector Laboratories), previously washed twice in TBS. After 30 min of incubation at room temperature with avidin-biotin complex (Vector Elite Kit, Vector Laboratories), the signal was detected using 3.30-diaminobenzidine tetrahydrochloride (DAB, Sigma-Aldrich) and lightly counterstained with Mayer's hematoxylin (Surgipath, Peterborough, UK) for 5 min.

To establish a semiquantitative score, micro-photographs were taken in 20 nonoverlapping, randomly selected fields. Necrosis, hemorrhages, and collagen in the fields were scored from 0 to 3 (0=absent; 1=<33% presence; 2=33–66% presence; 3=>66% presence). The number of neutrophils and eosinophils (by H&E) was scored from 0 to 3 (0=absent; 1=minimal number of cells [<33%]; 2=moderate number of cells [between 33% and 66%]; 3=high number of cells [>66%]). Finally, the infiltrate intensity of immunolabeled cells (T and B lymphocytes and macrophages or monocytes) was scored according to Dvir et al. (2011): 0=scant or absent; 1=positive cells evident but not in all fields; 2=positive cells present in all fields but markedly fewer in number than other cells; 3=positive cells in all fields and more predominant than other cells.

Differences in prevalence and distribution of the cell types was tested using the chi-square test. The differences between the scores of the different types of infiltrate were analyzed using a Kruskal-Wallis test. P values <0.05 were considered significant.

In the 163 Spirocerca-positive red foxes, the main location for grossly visible nodules was the stomach (96.9%; Fig. 1A). Gastric wall nodules were present in 118 foxes (72.4%). Nodules in both the gastric wall and gastric omentum were found in 40 animals (24.5%), while five foxes had nodules only in the small intestine wall (3.1%). No other location was noted. Nematode larvae were present in 96.3% of the nodules. Spirocerca-positive nodules were significantly more abundant in young animals (50.3%) than in subadults (34.4%) or adults (15.3%; P<0.05).

Following PCR, using BLAST analysis, identity with S. vulpis oscillated between 99.5% and 100% with the cox1 sequences of Spirocerca sp. from a naturally infected red fox in 2014 (e.g., GenBank accession nos. KJ605489.1 and KJ605487.1) and between 99.3% and 100% with the newly described species of S. vulpis in 2018 (e.g., GenBank nos. MH634016.1 and MH633993.1). The identities with the reference sequence for S. lupi (GenBank no. MF403001.1) were always below 93%. When submitted to the BOLD System identification tool, all sequences presented more than 99% similarity with S. vulpis sequences. Thus, all our samples are considered S. vulpis.

The mean size of the nodules was 2.21 cm. The percentiles of the data collected were as follows: 25% 1.19 cm, 50% 2.56 cm, and 75% 3.24 cm; 29.7% of nodules had a diameter less than 1.2 cm, 26.1% were between 1.2 and 3 cm, and 44.2% were larger than 3 cm. A t-test for paired samples was conducted to compare nodule size and the observed number of adult worms. More nematodes were seen in the largest nodules compared to the smaller ones (P<0.05).

Mitosis was insignificant in all observed fields; similarly, giant cells were very rare. Overall, 21.1% of the nodules had patchy distribution (Fig. 1B), 76.1% had diffuse distribution (Fig. 1C), and 2.8% had mixed forms. Significant differences were not found between the size of the nodule and the type of lesion (P=0.57).

A lymphoplasmacytic cell infiltrate occurred in diffuse lesions. Hemorrhages were scarce (score of 0 in 54% of fields), and no significant differences were found in diffuse lesions versus patchy lesions (Table 2). In diffuse lesions, the presence of collagen was low (score of 0–1 in 71% of fields), and necrosis was observed in a low-moderate proportion (score of 1–2 in >61% of fields; Table 2). Neutrophils were evident (score of 1 in 80% of fields) in diffuse lesions but significantly fewer than in the patchy lesions (P=0.017). Eosinophils were present in low-moderate numbers and few macrophages were present (Table 3).

Patches lesions were characterized by a focal or multifocal presentation of lymphoplasmacytic cell infiltrate. An infiltrate of neutrophils (score of 2–3 in 80% of fields) and macrophages was observed, mostly around the nematodes. Eosinophils were present in moderate number (score 2 in 70% of fields) in the patchy lesions; this was not significantly different from diffuse lesions (Table 3). Collagen and fibrocytes occurred in many samples (score of 3 in 60% of fields) with significant differences in collagen versus diffuse lesions (P<0.05). No active or binucleated fibroblasts were noted. Binucleate cells were seen, most of which were lymphocytes (Fig. 2A) but also macrophages (Fig. 2B). In addition, the presence of necrosis was found in a low proportion of lesions (score of 0–1 in 93% of fields), located around the parasite. Hemorrhages were very rare (score of 0 in 80% of fields; Table 2).

With respect to the CD3+ marker, significant differences could be appreciated between the types of lesions (P<0.05). In the patchy group, this marker was reported as the predominant positive cell type (score of 3 in 50% of fields). However, its presence was minimal and not predominant in the diffuse lesions (Table 3 and Fig. 3A).

Positive CD79α cells (B lymphocytes) were present but not in all fields. A significant difference (P=0.028) was observed in the number of CD79 alpha cells between groups, being more numerous in the patchy lesions (score of 1 in 80% of fields) than in the diffuse lesions (70%, score of 0; Table 3 and Fig. 3B).

The CD68 macrophage marker represented the lowest percentage of stained cells in all samples. No significant differences were recorded for the CD68 macrophage marker between the two types of lesions, but it was more evident in the patchy group (score of 2 in 50% of fields; Table 3 and Fig. 3C).

Nodules of S. vulpis were primarily found in the gastric area in the red fox, differing from the main tropism of S. lupi in dogs, with adults found in nodules in the esophageal wall (Mazaki-Tovi et al. 2002). Our results are partially consistent with other studies (e.g., Ferrantelli et al. 2010; Al-Sabi et al. 2014; Rojas et al. 2018b; Valcárcel et al. 2018). Previous studies did not find nodules in the gastric omentum or in the intestinal wall; however, we did not find lesions in the pericardium, aorta (Morandi et al. 2014; Rojas et al. 2018b), or lymph nodes (Reina et al. 1994). In jackals (Canis mesomelas), nodules were also found in the thoracic aorta, while gastric wall lesions were uncommon (Bumby et al. 2017). In dogs, aberrant migrations of S. lupi L3 occur through the mesenteric arteries, leading to small or large intestinal infarction (Brenner et al. 2020), and aberrant larval migration towards the central nervous system may cause serious neurological manifestation (Rojas et al. 2019). Spirocerca vulpis L3 also follow the route of the mesenteric arteries, modifying the typical life cycle, until reaching locations in the omentum, intestine, or gastric wall. In our study, we did not observe mesenteric multifocal necrotizing eosinophilic arteritis in these places. We found the greatest number of S. vulpis worms in the gastric wall, forming a nodule communicating with the gastric lumen through an operculum, unlike aberrant migration to the intestine (Brenner et al. 2020).

Al-Sabi et al. (2014) hypothesized that the atypical locations could relate to genetic differences between several haplotypes of S. lupi. The species S. vulpis seems better adapted to the red fox, resulting in development of nodules located in the gastric wall (Rojas et al. 2018b). However, additional studies are needed to understand the relationship between different Spirocerca spp. subspecies and the lesions found in foxes.

The size of the nodules found in our studies is similar to those described previously (Sanchís-Monsonis et al. 2019; Rojas et al. 2018b; Valcárcel et al. 2018) and is associated with the number of confined adults.

Recent studies described the presence of S. lupi (Brenner et al. 2020) in dogs from the age of 6 mo. We observed S. vulpis infestation in different ages of foxes, but the highest incidence occurred in young animals, despite the 140–161 day cycle after infection.

In the absence of mitosis and giant cells, all the lesions were classified as nonneoplastic, unlike those in spirocercosis in dogs (van der Merwe et al. 2008; Dvir et al. 2011). Preneoplastic lesions in dogs had moderate to severe lymphoplasmacytic infiltrate and collagen (Dvir et al. 2010); Rojas et al. (2019) suggest that the nodule progresses from an inflammatory fibrocytic lesion to a preneoplastic nodule characterized by the presence of active fibroblasts that may eventually undergo neoplastic transformation to sarcoma.

In our study, the predominant cells were lymphocytes, neutrophils, and, to a lesser extent, plasma cells and macrophages. Studies in dogs have highlighted the presence of leukocytosis as a significant finding (Mylonakis et al. 2001), and a severe infiltrate of eosinophils has been reported (Brenner et al. 2020). In diffuse lesions, other cell types prevailed over eosinophils. The presence of eosinophils is associated with parasitic reactions, along with the capacity or capability for vasopermeability, facilitating the arrival of other cells to the inflammation foci.

We found a high presence of neutrophils in the nodules. Dvir et al. (2010) found that in 40% of nonneoplastic cases the inflammatory infiltrate was predominantly lymphoplasmacytic, in 24% of cases lymphocytes and neutrophils were mixed, and in 21% of cases neutrophils predominated, compared to 25%, 5%, and 70%, respectively, in the neoplastic cases. Myeloid cells and especially neutrophils play a major role in the innate local inflammatory response in the spirocercosis-induced nodule in dogs (Dvir el al. 2010). Neutrophils mediate tissue damage through the release of cytokines, proteases, and other factors contained in their cytoplasmic granules and by regulating the activity of the adaptive immune response, including both T- and B-cell activation (Appelberg 2007; Kobayashi 2008; de Oliveira et al. 2016).

Necrosis was present in a low to moderate proportion in the diffuse lesions and a smaller presence in the patchy forms. The destruction of tissue that occurs in the invasive phase of large parasites, as described by Jubb et al. (2015), could be associated with necrosis in the initial lesions during invasion of S. vulpis in the red fox. In the diffuse lesions, the hemorrhages occur in the esophagus of dogs due to the rupture of blood vessels (Head et al. 2002). Both necrosis and hemorrhages are slightly more evident in diffuse lesions than in patchy lesions in foxes, but no significant differences were found in the statistical analysis.

Spirocerca lesions had not previously been characterized using immunolabeling in foxes. We observed an increase in the number of T cells over B cells. However, it was not possible to differentiate between CD4 and CD8 T cells, as this requires frozen samples, which we could not obtain. Immunolabeled T cells in the patchy group demonstrated their important role in the development of more homogeneous lesions associated with a coordinated immune response between different cell populations (Ulrichs et al. 2004; García-Jiménez et al. 2013). More studies are necessary to determine the implications of the regulatory T lymphocytes in the development of the lesion in foxes, as in dogs (Dvir et al. 2010). Such cells, along with their mediators, could play an important role in the development of the Spirocerca-associated lesions (Dvir et al. 2010).

The low presence of B lymphocytes in both patchy and diffuse lesions indicate early stages of lesion development. These cells are required for the development of chronic stages, including the possible induction of carcinogenesis (Cain et al. 2009).

The importance of inflammation of macrophages in the development of S. lupi lesions was described by Dvir et al. (2011). We observed these cells in patchy lesions but not in all fields. Macrophages may take part in the remodeling of the lesion. Lower numbers of these cells is associated with a greater presence of collagen. The CD68 marker was used as a macrophage marker, although a specific marker such as MAC387 could have been contrasted. For economic reasons, only the CD68 was used since, in our group's experience, we have found a high similarity with MAC387 and with other macrophage markers.

Neoplastic lesions have been seen in dogs (Dvir et al. 2010, 2011) and in mice (xenograft model; Stettner et al. 2005). However, we did not observe these. These foxes are subjected to high hunting pressure, thus it is possible that these animals are harvested before tumor development.

Similar studies (Dvir et al. 2011) indicated the need for molecular research (e.g., interleukins and chemokines) to analyze the development and evolution of the lesion caused by this parasite in foxes. Experimental studies are needed to categorize the possible evolution (or nonevolution) toward the typical tumoral forms induced in various anatomical regions. The lesional pattern based on implied cells points to an initial stage with a diffuse disposition and a later patchy evolution, which may help the animal control the inflammatory process.

In summary, the pathology of S. vulpis differs from that observed in S. lupi–infected dogs. It appears, based on changes in inflammatory cells and the presence of collagen, that S. vulpis lesions in the red fox progress from a diffuse to a patchy distribution, possibly through the control of the parasite infestation by the host.

This research was partially supported by grant GR18148 funded by the Regional Ministry of Economy and Infrastructure, Extremadura Government, and the European Regional Development Fund “A way to make Europe” and by the grant “Ayudas a grupos de la Universidad de Extremadura.”