Cytochrome c oxidase subunit 1 gene reveals species composition and phylogenetic relationships of Oesophagostomum spp. infecting pigs in northeastern Brazil

Abstract Helminths of the genus Oesophagostomum cause enteric diseases and affect domestic animals such as pigs. The aim of this study was to explore the species composition and genetic diversity of Oesophagostomum spp. infecting pigs in close contact with humans in the state of Piauí, Brazil. Eighty-seven fecal samples were collected for parasitological tests and molecular analysis. Through microscopy, the overall positivity rate for strongyliform eggs was 81.6% among the pigs studied. Forty-two strongyliform egg samples were subjected to PCR and six cox1 sequences (637 bp) were identified for the genus Oesophagostomum. The sequences were identified as Oesophagostomum dentatum, O. quadrispinulatum and O. columbianum. In the phylogenetic tree and haplotype network, 89 sequences were separated into seven clusters, which also included reference sequences from GenBank. Oesophagostomum dentatum and O. quadrispinulatum were seen to be closely related species and formed a monophyletic group related to O. aculeatum. Oesophagostomum columbianum showed similarity with sequences from parasites infecting small ruminants and the clade was positioned closer to O. bifurcum. High interspecific diversity was found and intraspecific diversity varied according to the species. This was the first study to characterize Oesophagostomum DNA sequences obtained from pigs in Brazil.


Introduction
The genus Oesophagostomum (family Chabertiidae, order Strongylida) includes worms ranging in length from 6.5 to 24 mm. The anterior end of the body presents a shallow oral cavity, a dilated cuticle forming a cephalic vesicle and a radiated crown present with varied structure (Taylor et al., 2010). Species within this genus are hosted by a wide diversity of mammals.
Considering only the species with the greatest impact on public health and veterinary medicine, pigs can be infected with the species O. dentatum and O. quadrispinulatum (Lin et al., 2014). Oesophagostomum bifurcum, O. stephanostomum, O. brumpti and O. aculeatum have previously been described in human and non-human primates (Glen & Brooks, 1985);and O. columbianum, O. venulosum and O. asperum in sheep and goats (Gaddam et al., 2017). Despite the first human case of 'O. stephanostomum' infection was recorded from Brazil, the actual systematic position of this worm has not been settled (Railliet & Henry, 1910;Thomas, 1910).
Oesophagostomiasis causes economic losses in pig farming, affecting both large and small producers (Li et al., 2017). Despite differences in size, Strongylida eggs are similar in morphology, which makes it difficult to identify species with parasitological examinations. Although obtaining larvae through stool cultures helps in making genusspecific diagnoses, species identification remains limited (Benavides et al., 2007).
To overcome these limitations, molecular taxonomy through DNA sequencing is widely applied to diagnoses, to enable phylogenetic reconstructions and evolutionary deductions of infectious agents (De & Bandyopadhyay, 2008). The objective of the present study is to explore the species composition and genetic diversity of Oesophagostomum infecting pigs in rural communities in northeastern Brazil.

Material and Methods
The fieldwork was performed from 2014 to 2017 in two regions in the state of Piauí, northeastern Brazil: Nossa Senhora de Nazaré (NSN) and Teresina (TER), in periurban and rural low-resource communities ( Figure 1). Fecal samples from pigs (n = 78 in NSN and 9 in TER) were collected after spontaneous defecation, on the ground. The pigs were in rudimentary shelter or in peridomestic or domestic areas during the field visits. These samples were individually stored in properly identified plastic bags, placed in a container with ice and sent to the field laboratory for parasitological examinations. The feces were processed using the Ritchie method (centrifugal sedimentation with ethyl acetate) and sucrose flotation. Samples that contained helminth eggs were then frozen until DNA extraction.
Genomic DNA was extracted from 42 fecal samples positive for strongyliform eggs using the DNeasy Blood & Tissue kit (Qiagen, Hilden, Germany), in accordance with the manufacturer's instructions. The partial cytochrome c oxidase subunit 1 (cox1) gene was amplified using the Platinum Taq DNA polymerase (Invitrogen, Waltham, MA, USA) with a final volume of 50 µL. A cocktail of three primer pairs that recover cox1 barcodes from diverse nematode lineages parasitic on vertebrates, including members of three orders and eight families was used (Prosser et al., 2013). The PCR conditions were as follows: initial denaturation at 94 ºC for 5 min, followed by 35 cycles of 94 ºC for 40 s, 55 ºC for 40 s, 72 ºC for 1 min and a final extension at 72 ºC for 5 min. The PCR products were purified using the DNA Illustra GFX PCR and gel band purification kit (GE HealthCare, Pittsburgh, USA) and were subjected to sequencing of both strands of DNA using the BigDye Terminator v. 3.1 cycle sequencing kit (Thermo Fisher Scientific, Foster City, USA) using the following primers: M13F 5'-TGT AAA ACG ACG GCC AGT -3' (forward) and M13R 5'-CAG GAA ACA GCT ATG AC -3' (reverse) (Messing, 1993). Capillary electrophoresis was performed using an ABI 3730 automated DNA sequencer (Applied Biosystems). Sequences showing overlapping peaks on electropherograms were cloned using the pGEM®-T Easy vector system (Promega, Madison, WI, USA) with Escherichia coli DH5-alpha cells in brain heart infusion broth (Sigma-Aldrich, St. Louis, MO, USA) on disposable plates. After cloning, the PCR and sequencing 3/13 Phylogenetic inferences of Oesophagostomum spp.
The Bioedit software v.7.0.4 (Hall, 1999) was used to edit the nucleotide sequences (401 bp). The Basic Local Alignment Search Tool (BLAST) (NCBI, 2022) was used to verify the similarity of the nucleotides with sequences of nematodes from GenBank. Orthologous sequences (n = 83) were retrieved from GenBank ( Table 1) and sequences with degenerate bases were not included. The sequences obtained in the study (n = 6) were deposited in GenBank under accession numbers MK282837 to 42.
The relationships in the haplotype network were inferred through median joining using the Network v.10.2.0 and DnaSP v.6 software (Rozas et al., 2017). Free vector images were used in the figures of the phylogenetic tree and haplotype network, to represent the hosts (Flaticon, 2022;SVG SILH, 2022). The genetic diversity indexes of Oesophagostomum populations were calculated using the Arlequin v.5.2.2 software (Excofer & Lischer, 2010).
The Fst fixation index was determined on all populations, using the Arlequin v.5.2.2 software to estimate the genetic differentiation among populations, with a significance of 1,000 permutations (Excofer & Lischer, 2010). This study was

Results
Eighty-seven fecal samples from pigs were analyzed and the positivity rate for strongyliform eggs through microscopy was 81.6% (71/87). Six sequences (637 bp) were identified as belonging to the genus Oesophagostomum. From NSN, four samples were characterized as O. quadrispinulatum and one as O. columbianum. In TER, one sample was characterized as O. dentatum. Three cox1 sequences showed overlapping peaks, thus indicating the presence of more than one species or genotype. Cloning of the fragments enabled identification of the species O. quadrispinulatum and O. dentatum. Fecal samples positive for strongyliform eggs which were negative for Oesophagostomum in the molecular analysis allowed the identification of Trichostrongylus sp. (n=1) and Metastrongylus spp. (n=10) through cox1 sequencing. Therefore, no Oesophagostomum species previously characterized in humans was found in the studied swine populations.
Alignment of the sequences of the present study in relation to 83 Oesophagostomum spp. reference sequences (401 bp cox1 sequences) from GenBank ( Table 1) was performed. The ML phylogenetic tree (Figure 2) showed that the Oesophagostomum sequences were organized into three main groups. These groups included seven clades: In the study area, O. quadrispinulatum was the predominant species. The sequences of O. quadrispinulatum and O. dentatum were grouped in the same clade, in which there were only sequences from pigs (showing 99% similarity with the reference sequences), and were more closely related to O. aculeatum (cluster A). The O. columbianum sequence in the present study was grouped into the same clade as parasites obtained from small ruminants. Interestingly, three sequences that clustered close to O. columbianum (Figure 2) were located in another arm of the tree (showing 99% similarity with the reference sequence).
The haplotype network (Figure 3) based on the cox1 locus showed topology similar to the phylogenetic tree. The 89 sequences used in the phylogenetic analyses were distributed into 86 haplotypes ( Table 2). Six novel haplotypes of Oesophagostomum were identified in the present study. The species O. aculeatum showed a star shape, with a central haplotype, which has been identified in Asia. In general, the groups had long arms due to the number of polymorphisms identified among the species. Genetic diversity indices revealed high interspecific diversity in the genus Oesophagostomum, with H ± SD = 0.9992 ± 0.0018 and 129 polymorphic sites ( Table 2). The intraspecific diversity varied according to the species. Oesophagostomum columbianum showed the lowest intraspecific variability with H ± SD = 0.6667 ± 0.3143 and 4 polymorphic sites. The genetic divergence (Fst) results were similar to the genetic diversity analyses ( Table 3). The intraspecific divergence among specimens of O. columbianum was greater than the interspecific divergence of the samples analyzed (O. columbianum Asia Fst = 1; O. bifurcum Fst = 0.55). Furthermore, the Fst value between O. quadrispinulatum and O. dentatum was high: Fst = 1.

Discussion
In the present study, the proportion of fecal samples from pigs that were positive for strongyliform eggs was higher than was found in Colombia (12.9%; 36/279) (Pinilla et al., 2020), India (19.9%; 74/371) (Yadav et al., 2021) and other region of Brazil, where it reached 46.6% (41/88) (Barbosa et al., 2015). Considering that the rearing systems reported by these authors were also extensive, our results may be explained by the handling and hygiene conditions of the pigs in the areas studied. Factors such as type of rearing, non-disinfection of drinking fountains and non-deworming are related to high frequencies of enteric helminths in pigs (Nansen & Roepstorff, 1999).
Making diagnoses based only on observation of eggs can lead to erroneous results due to similarities in morphology. Strongyliform eggs may belong to several species of pig parasites, including Oesophagostomum spp., hookworms, Trichostrongylus spp., Hyostrongylus rubidus and Metastrongylus salmi.
The present study used DNA barcoding to access species composition and genetic diversity. Primers for the mitochondrial target cox1 are "eclectic" due to high levels of intraspecific conservation and moderate interspecific variability, thereby providing identification of haplotypes and species in biological material (Hebert et al., 2004). It was possible to identify three distinct Oesophagostomum species in the area studied: O. dentatum, O. quadrispinulatum and O. columbianum. Interestingly, all sequences obtained in the present study were from different and undescribed haplotypes. None of these species are recognized as having zoonotic potential. Despite this, we cannot rule out the possibility of transmission of Strongylida parasites from pigs to human hosts. A previous study by our research  group in NSN demonstrated that the strongyliform eggs found in human samples belong to the genus Necator, but not the species N. americanus (Monteiro et al., 2019).
Although O. dentatum and O. quadrispinulatum are parasites normally found in pigs, O. columbianum usually infects small ruminants. In the communities studied, pigs, goats and sheep are raised in close contact with each other, thus indicating the possibility of cross-host transmission. This type of management facilitates ingestion of goat and sheep feces by pigs, given that they are coprophagous, and enables passage of O. columbianum eggs or larvae through the digestive tract and presence of their DNA in fecal samples (pseudoparasitism), or even cross host transmission. In the present study, feces were collected fresh after defecation in the environment and, even though appropriate measures were taken at the time of collection, occurrence of contamination from the environment cannot be ruled out.
The phylogenetic tree and the haplotype network were structured based on mitochondrial DNA sequences and, therefore, were based on matrilineal inheritance. Within this perspective, O. dentatum and O. quadrispinulatum are closely related and formed a monophyletic group with two distinct clades. Similarly, phylogenetic analyses on O. dentatum and O. quadrispinulatum recovered from pigs in different regions of China generated a cluster with two clades that formed a monophyletic group (Lin et al., 2012a). The last authors used ribosomal DNA targets, which resulted in similar formation of a monophyletic group, with different groupings in the species O. quadrispinulatum, thereby indicating the presence of distinct genotypes or subspecies (Lin et al., 2014). Four distinct and novel haplotypes were identified in our O. quadrispinulatum sequences.
Pigs (Sus scrofa domesticus) are not autochthonous species from the Americas, having been introduced during the colonization process by Europeans. More recently, the introduction of the wild boar (Sus scrofa scrofa) in the 1990s led to their conversion into wild animals, as an exotic species, which population growing is uncontrolled in Brazil. Piauí is one of the states where its presence has not yet been registered. Oesophagostomum dentatum and O. quadrispinulatum infect domestic pigs in Brazil, Europe and Asia, with O. dentatum being identified in wild boar in Brazil as well (Silva & Müller, 2013;Li et al., 2017). It can be deduced that the process of introduction of pig farming in Brazil and its expansion also enabled the expansion of these helminth species.
Inclusion of sequences from other Oesophagostomum species in the phylogenetic analysis demonstrated the existence of seven distinct clades for this genus and that the Oesophagostomum species in pigs are closer to the nonhuman primate species O. aculeatum (which parasitizes monkeys and orangutans in Asia) and O. stephanostomum (which infects chimpanzees and gorillas in Africa). Oesophagostomum columbianum was found in pigs in the present study. A nucleotide BLAST (BLASTn) analysis in GenBank showed a similarity of 99% with O. columbianum in sheep from China (Zhao et al., 2013), and with Oesophagostomum sp. in goats and sheep from Brazil (Monteiro et al., unpublished data). This is the first study exploring nucleotide sequences of Oesophagostomum in Brazil. These findings highlight the usefulness of molecular tools for investigating the taxonomy of strongyliform eggs observed in parasitological examinations, monitoring the presence of infection in herds ante-mortem, guiding control measures and providing data for studies on resistance to anthelmintics.