Prevalence, Infection Intensity and Molecular Diagnosis of Mixed Infections with Metastrongylus spp. (Metastrongylidae) in Wild Boars in Uzbekistan

The aim of the present study was to characterize the diversity of Metastrongylus spp. in wild boars and the earthworm intermediate host species contributing to the maintenance of the life cycle. Here, wild boars were subjected to parasitological necropsies, and lungworm species were identified morphologically, followed by confirmation using ITS-2 sequencing and a phylogenetic analysis. Earthworms were collected from wild boar habitats and investigated for the presence of larvae. The prevalence of Metastrongylus spp. in wild boars was 78.8%, and many individuals were positive for all three detected species, Metastrongylus pudendotectus, Metastrongylus salmi and Metastrongylus elongatus. The phylogenetic analysis did not clearly resolve all species, except for M. pudendotectus. Age group and season had no influence on prevalence, while intensity was significantly higher in autumn than in spring and summer (Kruskal–Wallis followed by Dunn’s test). Three out of six investigated earthworm species were positive for metastrongyloid larvae (prevalence of 10.4–16.7%), but neither their phylogenetic relationship nor ecological microhabitats were able to explain these differences. Further sequence data should be used to improve the resolution in phylogenetic trees to determine potential cryptic species in the genus, while the application of deep sequencing approaches might provide insights into species-specific epidemiology and pathology.

Wild boars live in small groups, constantly migrating from one place to another. They are omnivorous, feeding on vegetative underground (roots, rhizomes, tubers and bulbs) and the above-ground parts of plants, as well as various fruits, berries and seeds. Animals play a significant role in the diet of the wild boar, which is dominated by earthworms, mollusks, the larvae and adults of soil insects, small reptiles, amphibians and even small rodents, and they also scavenge on carrion and dead fish. Usually, wild boars obtain food by digging up the top layers of soil [14].
The Metastrongylus species were differentiated based on morphological features under an ML 2000 series microscope (Meiji Techno, Saitama, Japan). Adult nematodes were identified based on the morphology of the female's posterior body, a pair of massive trilobed lips and long filiform spicules and on the male's atypical bursa. The female has a relatively long vagina. The females and males of the different species were identified based on small differences in the morphology of the vulva and the spicules, respectively [1,4,16].

Collection of Third-Stage Larval Lungworms from Earthworms
More than 500 specimens of earthworms were studied regarding their potential role as intermediate hosts of metastrongylid nematodes. The collection of earthworms was carried out in juniper forests near springs, tugays, reed thickets and grasslands, which are all well-known places for rookeries and the feeding of wild boars in the Tashkent, Jizzakh and Bukhara regions of Uzbekistan [44]. For a quantitative analysis of the earthworms in places and on paths visited by the wild boars, the method of the manual sorting of soil samples was used, including plots with an area of 25 cm × 25 cm [44,45]. Initially, the litter layer was examined at the designated site, and then layer-by-layer excavations were carried out as follows: from the surface to a depth of 10 cm, from 10 to 20 cm and from 20 to 30 cm. Adult and juvenile earthworms were selected from all soil layers. The collected lumbricides, with labels indicating the sample number, date, place of sampling and layer depth, were placed in containers with soil. Under laboratory conditions, the soil was washed off the worms, and, taking into account morphological features, they were divided into two groups: the first was fixed in a 2% formalin solution to determine the species, and the second was used to isolate metastrongylid larvae.
The species identification of earthworms (Lumbricidae) was established in accordance with the guide developed by Perel [46]. Furthermore, the species identification of the collected Lumbricidae was confirmed at the Department of Zoology of the Karshi State University. The infection of earthworms with the larvae of metastrongylids was determined using generally accepted methods [47]. For this, to detect, identify and count the nematode larvae, compressors and stereomicroscopes (MBS-10) were used. The collected earthworms were pre-killed with a 1% formalin solution. Then, the cuticle was cut with scissors, and the esophagus, goiter and muscular stomach with the blood vessels surrounding them were separated. These organs were examined using the compressor method under a microscope for the presence of metastrongylus larvae. In addition, earthworms were investigated using the method of digestion in artificial gastric juice. The number of earthworms found to be infected by larvae was counted. The size of the invasive larvae of Metastrongylus is 0.570 mm × 0.03 mm, and the posterior end ends in a tip (younger larvae at the caudal end of the button); not far from the caudal end, there is a small cuticular spine; the anterior end is blunt and cut off.

DNA Extraction, PCR Amplification and Sequencing
DNA was extracted using a DNA Purification kit (Qiagen, New Dehli, India) and eluted twice with 100 µL of the AE buffer provided in the kit. PCR amplification used 0.25 µM of each of the primers NC1 (ACGTCTGGTTCAGGGTTGTT) and NC2 (TTAGTTTCTTTTCCTCCGCT) [

Phylogenetic Analysis
The sequences of both strands were compared and edited using BioEdit [48]. Similar sequences in GenBank were identified using BLASTn searches [49]. All sequences of the genus Metastrongylus and two sequences of Angiostrongylus vasorum, one Aleurostrongylus abstrusus sequence and one Protostrongylus hobmaieri sequence, the latter to be used as an outgroup, were downloaded. The sequences were aligned using MAFFT 6.5 [50] on an online server [51] using the Q-INS-I option to consider RNA structure information for alignment. It was decided to use the "Leave gappy regions" option in order to avoid the artificial alignment of the non-homologous parts of the ITS-2 sequences. A maximum-likelihood phylogenetic tree was calculated on the IQ-TREE server using version 1.6.12 [52,53]. Modelfinder [54] was used to identify the optimal nucleic acid substitution model, including FreeRate heterogeneity models with four rate categories based on the lowest Bayesian information criterion. Node support was calculated using ultrafast bootstrapping [55], the Shimodaira-Hasegawa (SH)like approximate likelihood ratio test [56] and an approximate Bayes test [57]. The trees were visualized using FigTree.v.1.4.4 (Andrew Rambout, Edinburgh, UK).

Statistical Analysis
Statistical analyses were either conducted in GraphPad Prism 5.02 or in R 4.1.1. Prevalence with 95% confidence intervals and significant differences in prevalence were calculated with the functions binom.wilson and tab2by2.test from the epitools package 0.5-10.1. Kruskal-Wallis tests, followed by Dunn's post hoc tests and Mann-Whitney U tests, were conducted in GraphPad. If p values were corrected for multiple testing, the Bonferroni method was used.

Prevalence, Intensity and Species Composition
As a result of the parasitological examination of the wild boars, sexually mature nematodes of three species from the genus Metastrongylus were found: M. elongatus, M. pudendotectus and M. salmi. Quantitative data on the prevalence and intensity of infection (number of worms in infected animals) for the three regions of Uzbekistan are shown in Table 1. Only seven of the pigs had no lungworms at all. The majority, 20 out of 33 pigs (60.6%), were infected by all three lungworm species. Five pigs were positive for M. pudendotectus and either M. salmi (n = 3) or M. elongatus (n = 2). One pig was only positive for M. pudendotectus, which was also the species with the highest prevalence, but the prevalence of the different species was neither significant for the complete dataset nor for the individual regions. Moreover, the differences between the regions were not significant. N, total number of animals; n, number of positive animals; 95% CI, 95% confidence interval. There were no significant differences in prevalence between seasons using the mid-p exact test.
The intensity was significantly higher in Jizzakh than in Tashkent ( Figure 1) for M. podendotectus and M. salmi, as well as for all Metastrongylus spp. together. Moreover, worm counts were significantly higher in M. podendotectus than in M. salmi and M. elongatus. N, total number of animals; n, number of positive animals; 95% CI, 95% confidence interval. There were no significant differences in prevalence between seasons using the mid-p exact test.

Molecular Characterization of Metastrongylus spp. from Uzbekistan
In order to confirm the morphological species identification, the ITS-2 regions were successfully amplified and sequenced from adult lung nematode specimen samples, resulting in sequences of approximately 495 bp for all three species. One ITS-2 sequence for each species was deposited in GenBank ( Table 2). The obtained nucleotide sequences were analyzed together with the available GenBank data by constructing a maximum-likelihood phylogenetic tree ( Figure 2).

Molecular Characterization of Metastrongylus spp. from Uzbekistan
In order to confirm the morphological species identification, the ITS-2 regions were successfully amplified and sequenced from adult lung nematode specimen samples, resulting in sequences of approximately 495 bp for all three species. One ITS-2 sequence for each species was deposited in GenBank ( Table 2). The obtained nucleotide sequences were analyzed together with the available GenBank data by constructing a maximum-likelihood phylogenetic tree ( Figure 2). The obtained nucleotide sequences were analyzed together with the available GenBank data, resulting in the phylogram shown in Figure 2. In this tree, M. pudendotectus formed a very homogenous, moderately well-supported cluster clearly separated from all other species. The other species M. elongatus, M. salmi, M. confuses and M. asymetricus formed a highly supported supracluster (indicated in Figure 2), but the differences among the species within this supracluster were relatively small. There was one very poorly supported cluster 1 (0% SH-LRT support) that contained three subclusters with much higher support. These are the two clusters of M. elongatus (syn. M. apri) named in Figure 2 as the M. elongatus genotype group (GG) I and II. Between the two M. elongatus GGs, two sequences were located that only differed by a two-base-pair IN/DEL, i.e., Y08009 and Y08007. While the former is labeled as M. salmi in GenBank, the latter is M. confuses. The M. salmi sequence must have been obtained from a morphological misidentified specimen since it is the only M. salmi labeled sequence within this cluster. Thus, these sequences most likely represent M. confuses. A second, poorly supported large cluster 2 (43.1% SH-LRT support) was dominated by sequences labeled as M. salmi. This cluster 2 contained three well to highly supported subclusters with different M. salmi genotypes (M. salmi GGI-III in Figure 2). Within the M. salmi GGIII cluster, a single M. asymetricus sequence is located, connecting with a relatively long branch ( Figure 2).

Seasonality of Infection
The prevalence of infection of animals in different seasons of the year ranged from 50 to 92.3%, with the highest prevalence occurring in autumn and the lowest prevalence occurring in summer, but the differences were not significant due to the small number of animals in some of the seasons, particularly in summer ( Table 3). The median intensity of infection was between 75 and 477 worms, again with the highest intensity in autumn and the lowest in summer (Figure 3). Intensity was significantly higher in autumn than in summer and winter, while spring was similar to autumn, although the differences were not significant for any of the other seasons ( Figure 3).

Age Dynamics of Infections with Metastrongylids
The prevalence and intensity of the infections of the animals w between juvenile and adult wild boars. In the six juveniles included alence was slightly lower than in the 27 adults, but the difference w ble 4). Infection intensity was also slightly lower in juveniles than 301 and 397, respectively) ( Figure 4), but, again, this difference was Table 4. Prevalence of metastrongylid infection in different age groups of

Age Dynamics of Infections with Metastrongylids
The prevalence and intensity of the infections of the animals were further compared between juvenile and adult wild boars. In the six juveniles included in the study, the prevalence was slightly lower than in the 27 adults, but the difference was not significant (Table 4). Infection intensity was also slightly lower in juveniles than in adults (medians of 301 and 397, respectively) ( Figure 4), but, again, this difference was not significant. N, total number of animals; n, number of positive animals; 95% CI, 95% confidence interval. There were no significant differences in prevalence between seasons using the mid-p exact test. N, total number of animals; n, number of positive animals; 95% C were no significant differences in prevalence between seasons usin  Horizontal lines indicate the medians of each dataset. Datasets were compared using a Man-Whitney U test, but no significant differences were found.

Prevalence of Metastrongylid larvae in Different Oligochaete Species
In total, six different species of earthworms were identified, and the sample size of each species was considerably high (n = 46-96). Metastrongylid larvae were found in three species, Aporrectodea caliginosa trapezoides, Eisenia veneta and Octolasium lacteum, all with a prevalence between 10 and 15%, while the species Aporrectodea jassyensis, Eisenia veneta and Dendrobaena byblica were consistently negative (Table 5). Almost all comparisons between the prevalence of positive and negative species were significant with only one exception, A. jasiniensis vs. A. jassyensis. Positive species were found to have an endogeic ecological niche, i.e., species burrowing extensive horizontal tunnel systems and feeding predominantly on soil, or to be at least partially endogeic (Eisenia fetida, epiendogeic). However, species negative for metastrongylid larvae were also sometimes endogeic (A. jassyensis) or epigeic, i.e., species dwelling on the ground and in the litter layer and feeding predominantly from plant litter [58].

Discussion
The genus Metastrongylus contains several parasites of pigs that are rarely investigated. Due to the obligate intermediate host, these parasites are not relevant in industrial pig production [59], and their clinical relevance has also been questioned [60]. Since wild boars are the most frequently infected hosts, these parasites have mostly been neglected by researchers in the recent past.
Metastrongylus spp. have been reported worldwide with variable frequencies [10,16,[61][62][63][64][65][66][67][68]. In a previous study conducted by Kuchboev et al. [25], a similarly high prevalence of 84.6-92.2% was observed for the same three species of Metastrongylus. In this study, M. pudendotectus was slightly more prevalent than the other two species. Another study from Central Asia involved only 10 wild boras from two different regions of Kazakhstan. Again, the same three species were found. However, the prevalence was considerably lower, with 42.8% for M. pudendotectus and M. elongatus, but the authors did not report the prevalence of M. salmi [14]. In Eastern Europe, a prevalence of 100% was found in the Ryazan region of Russia, but the authors only found M. pudendotestus and M. elongatus [69]. In a national park close to Moscow, two out of five wild boars were positive for the three species M. elongatus, M. pudendotectus and M. salmi [70]. In Belarus, 98.4% of adult and 100% of juvenile wild boars were positive for Metastrongylus spp., with M. pudendotectus being the most prevalent species, followed by M. elongatus and M. salmi [71]. The same three species were also detected in Bulgaria [72]. In contrast, all five Metastrongylus species were found in Spain [16], Poland [73] and Switzerland [74]. In Switzerland, the overall prevalence was 77.4%, with M. pudendotectus being found most frequently, followed by M. salmi, M. confusus and M. apri, with M. asymetricus being the least frequently found [74]. In Poland, the order was M. pudendotectus, M. salmi, M. asymmetricus, M. elongatus and M. confusus. When looking into geographical regions other than Eurasia, a high prevalence of 84.5% for Metastrongylus spp. was described for wild boars in Morocco. Again, M. pudendotectus was the most prevalent species (84.5%), but with a prevalence of 72.7%, M. confusus was the second most frequently found species, followed by M. salmi (51.5%), whereas M. elongatus was not found. In Japan, wild boars showed 100% prevalence, and 64.3% were infected with all four species that were detected. The order of prevalence was M. asymmetricus, M. salmi, M. pudendotectus and M. elongatus. In feral pigs in Florida, M. apri, M. salmi and M. pudendotectus were collected during necropsies, with a prevalence of 94%, 76% and 65%, respectively [9]. Apparently, in Eastern European and Central Asian populations of wild boars, the presence of M. pudendotectus, M. elongatus and M. salmi (in the order of their typical prevalence) is widely observed, and the results from Uzbekistan are in line with many previous data from this region.
In Uzbekistan, the majority of wild boars were infected with multiple Metastrongylus species. This was also the case for all other studies cited above regarding the prevalence and species composition of pig lungworm. The high number of Metastrongylus species in combination with the very high overall prevalence that was described in all the studies make it clear that co-infections must be the rule rather than the exception. However, none of the studies, including the present one, has attempted to statistically analyze whether co-infections occur more frequently than one would expect according to the prevalence of each individual Metastrongylus species. The simple reason for this is that the number of animals included in these studies ranged from 5 to less than 50, and, thus, this is far too low for meaningful analyses of such dependencies.
Earthworms are a major part of the feed of wild boars, and, thus, Metastrongylus spp. turned out to be highly prevalent (63.6-92.3%), with infections reaching very high intensities in the present study. While the prevalence of the nematodes was not significantly influenced by age group or the season of the year, there was a significant effect of the season on intensity, with autumn showing a higher intensity than summer and winter. However, these results should be considered with caution since the very low intensities in summer were due to only three infected wild boars from this season. The difference between summer and autumn was based on much higher animal numbers, but it was also only moderate. In contrast to the results shown here, the infestation of earthworms in spring, summer and autumn has been described to be approximately the same in Belarus [30]. This suggests that differences in infection burden in pigs over the year might be due to differences in the role of earthworms in their diet.
Earthworms of the family Lumbricidae that serve as intermediate hosts can have a very high prevalence of almost 100%, with an intensity of tens to a few thousand larvae in one worm as described for species such as Aporrectodea caliginosa, Dendrobaena octaedra, Eisenia fetida and Lumbricus terrestis [28]. The infection intensity of wild boars with Metastrongylus spp. has been shown to be the major factor influencing infections in earthworms [75]. Earthworms that become infected by metastrongylid larvae remain infected lifelong [15], which can lead to the accumulation of larvae to high infection burdens. Most frequently and intensively infected were earthworm species that live in the upper layer of soil, plant debris and humus on the surface of the earth, i.e., epigeic species. The deeper the habitat of a worm species in soil, the lower the prevalence of metastrongyloids in these worms [28]. However, the results of the present study are contradictory to this view since all entirely epigeic species with their habitat in the upper leaf litter were negative. All positive species were either endogeic or epi to epiendogeic. However, the ecological category alone was also not sufficient to explain the pattern of infection observed in the present study in earthworms since the 46 specimens of the endogeic species A. jassyensis were all negative.
The interaction of Metastrongylus spp. with wild boars and relevant lumbricid earthworm species has rarely been investigated. The earthworms collected in the present study came from typical wild boar habitats. The fact that very closely related earthworm species, such as the two Aporectodea and the two Eisenia species, differed so strongly in the prevalence of metastrongylid larvae suggests that the members of these species pairs differ in their ecological macro-or micro-habitats. Since both are considered to be endogeic, different macrohabitats with different wild boar densities or activities might be important variables. The collection of different earthworm species with sampling sites that are georeferenced on a very small scale and with reference to the surrounding vegetation and wild boar activity might help to provide explanations for the unusual distribution pattern of metastrongyloid larvae observed here. However, such investigations are very labor-intensive since the sample site should be considered the statistical unit in many of the methods needed to analyze such data.
Based on both morphologic and molecular data, all isolated lungworms from the wild boars in Uzbekistan were M. elongatus, M. pudendotectus and M. salmi. As discussed above, many of the wild boars were infected with multiple Metastrongylus species simultaneously. This makes it difficult to determine whether there are differences in the epidemiology (e.g., preference of intermediate host species) or pathogenicity of the different species. Molecular tools have the potential to improve the available data on individual Metastrongylus spp., e.g., tools based on non-invasive fecal samples, in the future.
The literature data on the comparative study of the ITS-2 of other representatives of lungworms show that there are small but stable species differences that allow for the information about the structure of this site to be used as a rather effective tool for resolving the controversial issues of the taxonomy of nematodes of this group [6,76]. For metastrongylids, Conole et al. [37] used sequencing and a single-strand conformation polymorphism analysis (SSCP) of the ITS-2 to identify the species M. elongatus, M. pudendotectus and M. salmi in wild boars, which allowed for the direct display of sequence variation within and among individuals representing each species.
In the phylogenetic tree, M. pudendotectus is clearly separated from the other species. was also only poor. Alternative interpretations of the tree could be that all sequences in the supracluster belong to a single, genetically quite variable species, or that each of the genotype groups represents a different species. This would mean that M. elongatus would be split into two species, and M. salmi would be split into three species. Additional sequence data are required to determine which of these alternatives is the preferable hypothesis and to also better understand the phylogenetic history of M. asymetricus in the genus. In previous studies, it was shown that combined analyses of nuclear (e.g., ITS-2, ITS-1 and β-tubulin isotype 1) and mitochondrial (e.g., cytochrome oxidase 1, 12S and 16S subunit mitochondrial rRNAs) genes can help to obtain both a reliable phylogenetic tree and enough resolution to discriminate between closely related, sometimes cryptic, species [41,[77][78][79]. However, such analyses in the future will rely on the availability of material for rarer species, such as M. confusus and M. asymetricus.
PCR-based approaches have become an important tool for investigating parasite communities and populations but also host-parasite interactions, and they will probably also help in livestock management in the future [80]. The ability to monitor individual hosts rapidly over time makes it possible to investigate (i) the contribution of different helminth species to total parasite burden, (ii) the ecological relationships between helminth species and the magnitude and direction of genetic correlations between resistances to different nematode species in host populations and (iii) the occurrence of anthelmintic resistance in certain parasite populations. While this is already quite advanced today for the gastrointestinal parasitic nematodes of livestock, including massively parallel sequencing of amplicons, such as ITS-2 or the isotype 1 β-tubulin, to describe the parasite community or the resistance status of multiple species [81], it has not yet been described for pulmonary parasites. Such methods would allow non-invasive investigations of lungworm communities in wild boars but can also be extended to other species, including protected wildlife.  Institutional Review Board Statement: Ethical review and approval were waived for this study due to the fact that the wild boars were not shot for the purpose of the study. The investigators received only lungs of wild boars shot for meat production.

Informed Consent Statement: Not applicable.
Data Availability Statement: Data are contained within the article. Sequence data were deposited in GenBank.