First detection of Gongylonema species in Geotrupes mutator in Europe

In October 2017, a sampling of beetles captured by 27 pitfall traps was carried out according to a standardized sampling procedure (Lobo et al., 1988). The collections were carried out in three localities of western Spain, in the provinces of Salamanca and Zamora. Three types of excrements (pig, bovine, canine) were used as bait or attractant at each location. The captures of each trap were collected 65−80 h after trap placements and immediately deposited in 10 × 10 × 10 cm plastic containers containing soil and stored at 15°C during transport to the laboratory. The collected adult specimens with infective larvae were identified using morphological identification keys provided by Martín-Piera and López-Colón (2000). The beetles were dissected under a stereoscope for the detection of nematode larvae. In case of infection, larvae were placed in 70% ethanol until molecular analyses could be performed, with the exception of one worm selected for optical (× 100−200; Nikon eclipse 80i and digital photomicrography system, DXM1200F) and scanning electron (SEM) microscopy. For SEM, specimens were dehydrated through a graded series of ethanol from 70 to 100%, freeze-dried with liquid CO₂ according to the critical point method, and mounted onto stubs. Mounting of the sample on the sample holder was done using double-sided carbon adhesive tape. Mounted specimens were then sputter-coated with a gold layer (20 nm ± 5%) and observed with an FE-SEM Hitachi S-4800 (FE by Field Emission, with Field Emission Electron Gun).

Genomic DNA (gDNA) was extracted from individual worms found in beetles using an NZY tissue gDNA kit (NZYTech) following the manufacturer’s instructions. The cox1 genes were partially amplified using the primer set NTF and NTR following a PCR protocol previously described for the analysis of Spirurida (Casiraghi et al., 2001), and the product was sent for sequencing to STABvida (https://www.stabvida.com/es). Sequences were edited in Chromas Lite 2.1.1 (Technelysium Pty Ltd) and consensus sequences for each forward/reverse pair were created in BioEdit Sequence Alignment Editor (version 7.2.5, Carlsbad, CA. USA). Finally, the nucleotide sequences obtained (> 652 pb in length) were deposited at the DNA Data Bank of Japan (DDBJ: accession numbers LC577505 and LC612845). In addition to the specimens collected in the course of this study, a DNA extract prepared using a formerly unidentified Gongylonema specimen, obtained from Otus scops, Spain, and analyzed in a previous study (Esperón et al., 2013), was provided by its authors in order to obtain its corresponding sequence for the cox1 barcode section (DDBJ: accession numbers LC620542).

Phylogenetic analyses were carried out based on the construction of multiple alignments of nucleotide (nt) sequences either comprising a wide range of nematodes (including Cylicospirura, Dirofilaria, Gongylonema, Mastophorus, Onchocerca, Physaloptera, Setaria, Spirocerca, and Thelazia; n = 31) or restricted to the Gongylonema genus (n = 18). These were constructed using the iterative G-INS-I method as implemented in MAFFT vs. 7 (https://mafft.cbrc.jp/alignment/server/) followed by their edition using GBlocks (http://molevol.cmima.csic.es/castresana/Gblocks_server.html). Phylogenetic trees were built using the Maximum Likelihood (ML) optimization criterium and the Mega X software (Kumar et al., 2018), and their topological stability was assessed by bootstrapping with 1,000 re-samplings of the aligned sequence data. Alternatively, a Bayesian approach was also used for phylogenetic reconstruction, taking advantage of the BEASTv1.10.4 software (Suchard et al., 2018). For both situations, the best-fitting evolutionary model used was GTR + Γ+ I. Bayesian analyses included two independent Markov chain Monte-Carlo (MCMC) runs carried out assuming a lognormal-relaxed molecular clock (as suggested by Mega X Molecular Clock test), and a constant population coalescent prior. These runs were executed until 1 × 10⁸ states were sampled (10% of which were discarded as burn-in), and the final tree sample was summarized as a maximum clade credibility (MCC) tree.

Intra- and inter-group genetic distances were calculated using the Tamura/Nei model with gamma-distributed rates (α = 0.65), as suggested by Mega X. Sequence similarity searches were carried out using BLASTn (MegaBlast option; https://blast.ncbi.nlm.nih.gov/Blast.cgi) and Barcode of Life Data Sytems-v4 (http://www.boldsystems.org/). Finally, a principal coordinate analysis was carried out using PCOORD (http://www.hiv.lanl.gov/content/sequence/PCOORD/PCOORD.html)to visualize the genetic relationships between the cox1 sequences of the different Gongylonema species.

A total of 551 dung beetle specimens were analyzed and three nematode infective larvae were detected in two specimens of Geotrupes mutator (Marsham, 1802), indicating a prevalence of 0.36% over the total of beetle specimens sampled. One of the beetles hosted a single larva, the other dung beetle specimen hosted two Gongylonema larvae. They were captured in Campillo de Azaba (Salamanca province; Lat: 40.510, Long: –6.689), in a valley area and a contiguous holm oak forest area, respectively. Both specimens were collected in pitfall traps baited with pig dung.

Members of the genus Gongylonema are easily recognized because their cuticle is covered by large verruciform thickenings, especially prominent on the anterior part of the body. By optical microscope, the morphological features of one specimen clearly indicated the presence of a third larval stage of Gongylonema. The total length of the specimen was 1.55 mm, width at mid-body of 0.075 mm, and a distance from the anus to the posterior end of the body of 0.085 mm (Fig. 1.1). The cephalic structure showed the typical mouthparts of this genus (Fig. 1.2).

Figure 1:

The detailed observation by SEM showed the classic elevation peri buccal collar present in its internal face (Fig. 1.4), as well as an inner circle of labial papillae constituted by six well-developed labial papillae arranged in two groups of three papillae located laterally. In the external face of the peri buccal collar, there was an outer circle of labial papillae formed by four small papillae (Fig. 1.4). More externally, a circle of cephalic papillae constituted by four large papillae could also be observed (Fig. 1.5). Two large amphids were located lateral to the mouth and two lateral depressions. Two lateral alae extend throughout the body length (Fig. 1.6). Finally, the caudal region of the worm was adorned with a tuft of a dozen cuticular fingerings (Fig. 1.7).

Regarding molecular analyses, two partials nematode cox1 sequences were obtained (accession numbers LC577505 and LC612845) from specimens collected in the course of this study. The data coming from an additional specimen previously described from Madrid (Esperón et al., 2013) (accession numbers LC620542) was also included in the genetic analyses performed in this study. A search for similar sequences using Blastn unambiguously identified (E values = 0) several cox1 homologues among Gongylonema nematodes, indicating high similarity with G. nepalensis and G. pulchrum. However, while the matches covered > 97% of our sequences, the identity percentage for the best-fit ones averaged only 88%. On the other hand, a Blast-2seq analysis revealed that the cox1 LC620542 sequence (from Madrid) shared < 87% identity with both LC577505 and LC612845. Moreover, an attempted automatic molecular identification of the Spanish sequences in question using the BOLDSYSTEMS cox1-based molecular identification engine failed to produce any best match.

The Bayesian phylogenetic tree including many genera of nematodes (Cylicospirura, Dirofilaria, Gongylonema, Mastophorus, Onchocerca, Physaloptera, Setaria, Spirocerca, and Thelazia) (Fig. 2a) indicated that the cox1 sequences of the collected specimens were associated with those corresponding with to the Gongylonema genus. Furthermore, the collected specimens unambiguously segregated away from all the monophyletic clusters that defined the different species in the Gongylonema radiation (G. aegypti, G. neoplasticum, G. nepalensis, and G. pulchrum) (Fig. 2a). Interestingly, the two specimens hosted by G. mutator are grouped together, but the third one obtained from Otus scops in Madrid was segregated in a sister branch, suggesting that we could be dealing with two different Gongylonema species.

Figure 2:

The analysis of the genetic distances supported the suggestion that the three considered specimens belong to 2 different Gongylomena species since the genetic distances between them are similar to those shared by the other reference species (G. aegypti, G. neoplasticum, G. nepalensis, and G. pulchrum) (Fig. 2b). This separation was further confirmed by the Principal Coordinate Analysis (PCOORD) (Fig. S1A). Finally, as expected, when a phylogenetic tree only including Gongylomena cox1 sequences was obtained (Fig. S1B), the sequences here reported branched away from all the other species-defining branches. Although we cannot confirm the specific taxonomic status of the collected Gongylonema specimens, the analysis of the obtained cox1 sequences improve our understanding of the interspecific genetic divergences of this little-studied genus.

Figure S1:

In Spain, seven species of genus Gongylonema have been reported (Cordero et al., 1994; Fernandez-De-Mera et al., 2003; da Costa Cordeiro et al., 2018). Although we have not been able to find out which specie is our specimens, must a priori rule out G. pulchrum and G. neoplasticum present in Spain based on genetic analysis. Knowing that adults of Gongylonema (Progongylonema) are distinguished mainly from those of the subgenera Gongylonema (Gongylonema) and Gongylonema (Gongylonemoides) by the absence of cuticular bosses, their inclusion in this subgenus would be probable since this morphological character has not been observed in the specimen here analyzed. However, it is possible that our specimen may not yet have all the developed characteristics, as other authors observed for the different stages in other species (Alicata, 1935). The only known member of this subgenus in Spain is G. (Progongylonema) pacoi which was described for the first time in 1992 parasitizing the mucosa under the tongue of Corvids from the north of Cordoba province (Southern Spain) (Hernandez-Rodriguez and Gutiérrez-Palomino, 1992). Another Gongylonema that was detected in birds in Spain, was found in Madrid in various specimens of Otus scops (Esperón et al., 2013). Although the taxonomic identity of this species could not be established initially, our results have suggested the do not correspond to any of the Gongylonema species with available public genetic data. This would support the existence of a higher diversity of this genus in Spain. On the other hand, the infesting larva of G. soricis is similar to our nematode due to its close size (1545 µm-1580 µm) and its caudal ornamentation adorned with a tuft of cuticular fingerings (Quentin and Gunn, 1981). G. soricis has been rarely collected in Spain, except for records in Valencia in 1996 (Portoles et al., 1996). Taking into account these data and that its definitive host is grey shrew (Crocidura russula) which are highly distributed in Spain, G. soricis could be a good candidate for the identity of our specimen analysed in this study. Other potential candidate species include G. (Gongylonematidae) pithyusensis n. sp. and G. mucronatum. The first was described for the first time in 1977 parasitizing the oesophagus mucosa of the dormouse Eliomys quercinus (Mas Coma, 1977) in the Balearic Islands, Spain. Since then, it has only been detected once in Murcia (southeast) (Esteban et al., 2004), while the second was reported from Atelerix algirus (North African hedgehog) in North Africa and Spain (on the Balearic islands and Valencia: see Ferrer Màrius Vicent et al. (2000). However, as the distribution of A. algirus is limited to the east of the Iberian Peninsula, it is unlikely that our specimens belong to this species. The last species of Gongylonema inhabiting the Iberian peninsula is G. musculi, which had been recorded occasionally on several Mediterranean islands (Milazzo et al., 2003). The limited information about the morphology of the infecting larvae of this species prevent us from making any comparison. Therefore, the lack of morphological and genetic information on Iberian Gongylonema species momentarily makes a more specific definitive determination impossible.

While knowledge on the presence of nematodes in beetles from Spain could be regarded as helpful, it is either still scarce, or in the case of Gongylonema, non-existent. Although species of the genera Ascarops, Diplocapter, Diplogaster, Eudiplogaster, Holodiplogaster, Physocephalus, Pelodera, Paroigolaimella, Rhabditis, and Spirocerca have been detected in G. mutator, only G. pulchrum had been reported (Poinar, 1975). Therefore, this is the first report of other Gongylonema sp. in G. mutator naturally infected, at least in Europe. G. mutator is a widely distributed species in Europe from the south of Sweden to the Caucasus and the Iberian Peninsula, that also inhabit some Asian regions of Iran Kazakhstan, and Turkey (Löbl and Smetana, 2016). In the Iberian Peninsula this dung beetle species is frequent in the northern half part (Martín-Piera and López-Colón, 2000).

The only infected beetles with the analysed larvae were collected in traps baited with pig excrements. In general, Geotrupinae species are frequently associated with dung of mammalian herbivores (Martín-Piera and López-Colón, 2000). G. mutator digs several tunnels immediately below, or to the side, of the cow and horse dung pats (Teichert, 1955), using them for feeding or nesting. Cows and horses are thus the most probable candidates to host this parasite, although the general low trophic specificity of dung beetles (Martin-Piera and Lobo, 1996) has not yet made this point any clearer.

According to the morphologic and genetic analysis on the taxonomic identity of the Gongylonema species from G. mutator, there are the following possible interpretations:

The lack of genetic data of other Gongylonema species makes it impossible to relate this Gongylonema species with another already described only using morphological characters of adult specimens. However, based on the data of a comparative morphological and morphometric study, it makes us think that it could be preliminarily identified as G. soricis.

Much remains to be learned about the systematics of Gongylonema. Different studies (Makouloutou et al., 2013a, 2013b; Setsuda et al., 2016, 2018; Varcasia et al., 2017) have provided molecular data about Gongylonema genus, in line with our work. However, our data together with those of Esperón et al. (2013) indicate a possible greater diversity of Gongylonema species in Spain, and an evident scarcity of genetic information to support their genetic analyses. Future molecular analyses are needed for better understanding the ecobiology of Gongylonema in other regions and careful searching for more specimens in a European framework.