Spirocerca lupi is a parasitic nematode of mainly wild and domestic canids belonging to the order Spirurida, family Spirocercidae (Nascimento Gomes et al., 2021). This parasite has an indirect life cycle with coprophagous beetles as intermediate hosts, canids as definitive hosts, and other animals such as rats, lizards, or birds that can act as paratenic hosts (Rojas et al., 2020b). Canids become infected by eating beetles or paratenic hosts with the infective L3 larvae, which then migrate through the stomach, penetrate the gut wall, and travel through the gastric artery walls until they reach the aorta and the esophagus (Rojas et al., 2020b). In this way, S. lupi induces a disease known as spirocercosis, which usually presents as esophageal nodules that can progress from fibrous and inflammatory lesions to neoplastic lesions in the esophagus (Rojas et al., 2020b).

This parasite is distributed in the tropics and subtropics of the world, and extensive reports from Israel (Aroch et al., 2015), Southeast Asia (Aroch et al., 2015; Hoa et al., 2021), Africa (Greeff et al., 2018), and different regions of Europe (Giannelli et al., 2014; Psáder et al., 2017) have been published. In the Americas, S. lupi has been detected in Southern United States of America (USA) in coyotes (Canis latrans), gray foxes (Urocyon cinereoargenteus), red foxes (Vulpes vulpes) and bobcats (Lynx rufus) (Pence and Stone, 1978; Morrison and Gier, 1979), in domestic dogs from USA (Pence and Stone, 1978), Mexico (Rodríguez-Vivas et al., 2019), Costa Rica (De Aguiar et al., 2021), and Brazil (Costa Santos et al., 2012), in maned wolves (Chrysocyon brachyurus) from Brazil (Blume et al., 2014), Andean foxes (Lycalopex culpaeus) from Peru (Gomez-Puerta et al., 2018) and Chile (Di Cataldo et al., 2023), dogs in Colombia (Santisteban-Arenas and Piedrahita, 2016), bush dogs (Speothos venaticus) in Argentina (Rinas et al., 2009), and dogs in Grenada (Chikweto et al., 2012). Most reports have been done with the observation of eggs in feces or post-mortem macroscopic observation of adult worms in lesions at necropsy, but molecular analyses have been implemented for species confirmation only is some of these reports (Gomez-Puerta et al., 2018). The importance of running both microscopic and molecular methods was demonstrated with the description of Spirocerca vulpis, a sister species of S. lupi that infects red foxes in Europe (Gama et al., 2020; Rojas et al., 2020a). S. vulpis is macroscopically undistinguishable from S. lupi but shows 30 um teeth in the buccal capsule and molecular divergence of 0.19-0.25% in the 18S rDNA and 9.34% in the cytochrome oxidase subunit 1 (cox1) when compared to S. lupi (Rojas et al., 2018b). In this setting, cryptic diversity within species can also be detected by using morphometric comparisons and mitochondrial and ribosomal markers (Chaves-Gonzalez et al., 2022). Importantly, S. vulpis has not been reported in the Americas, despite the introduction and range expansion of red foxes to North America in modern times (Hoffman and Sillero-Zubiri, 2021) and its finding in Portugal (Gama et al., 2020), Switzerland (Rojas et al., 2020a) and France (Medkour et al., 2020).

Molecular variation among S. lupi isolates has been observed in the internal transcribed spacer 1 (ITS1) loci and the cox1 gene, which has been useful to trace the species variability (Rojas et al., 2018a). Furthermore, cox1 sequences are easier to work with S. lupi than the highly variable ITS1, due to the great intraindividual variation observed in the ITS1 with up to six different copies of the loci in a single specimen (Rojas et al., 2018a). Accordingly, a phylogenetic study using the ITS1, cox1 and 18S rDNA loci, concluded that there is a high genetic variation among S. lupi specimens circulating in Africa, Asia, and Europe (Rojas et al., 2018a). The findings in the three loci led to the identification of two genotypes of S. lupi: the first genotype included specimens from Australia, India, Israel, and South Africa, and the second genotype with specimens from Hungary and Italy (Rojas et al., 2018a). Accordingly, a 0.14%, 8.06% and 6.48% variation in the 18S rDNA, ITS1 and cox1, respectively, were found between genotypes. However, the phylogenetic relationships of S. lupi in the Americas is not yet elucidated, and cox1 sequences from Andean foxes from Peru and Chile (Gomez-Puerta et al., 2018; Di Cataldo et al., 2023) confirmed the high genetic variability observed in geographical locations in the Americas. The present study analyzed cox1 and 18S rDNA gene fragments of specimens from Costa Rica, Mexico, and the United States and compared them with the specimens from the Old World.

The evolutionary history of the Spirocerca spp. has been speculated since the recent description of the S. vulpis and the characterization of different S. lupi genotypes. Phylogenetic analysis of S. lupi in the Americas has been limited, but our results shed light on the speciation process of this parasite. The data obtained in this study revealed that specimens from Costa Rica, Mexico, the United States, and Peru were closely related among each other, and to those from Israel according to cox1 analysis. Interestingly, our data at the phylogenetic and phylogeographic levels shows that S. vulpis from Europe was the common ancestor species from which S. lupi genotypes 1 and 2 derived and geographically expanded across Asia and Africa. According to the available data, it was shown that S. lupi from the Americas might have originated from Western Asia later in time. In fact, canids in the American continent, may have migrated across the Panama land bridge from North to South America (Chavez et al., 2022). Extant species in South America are monophyletic and have diversified throughout the years after being exposed to different environments, but the similarity in their diets and behaviors may have aided in the spread of S. lupi all over the continent (Chavez et al., 2022). Ribosomal 18S rDNA sequence analysis showed great similarities between S. lupi specimens from India, Israel, South Africa, and Costa Rica, as shown in the phylogenetic and haplotype network analyses. The 18S rDNA gene is a highly conserved and functional region, therefore, it evolves slower and is useful for phylogenetic reconstruction in higher taxonomic levels, but does not always discriminate among specimens from the same species (Kobayashi, 2011; Chan et al., 2021). Therefore, the 18S rDNA Bayesian phylogenetic tree showed that the Spirocerca spp. belong to a group that is separated from other members of the family Spiruridae as previously observed (Rojas et al., 2018b). This group includes two S. vulpis sequences and several S. lupi sequences, including the ones from Costa Rica. Since two S. lupi genotypes have been determined, mitochondrial genes were used in this study to attain a larger intraspecies resolution than that obtained for the 18S rDNA gene.

Mitochondrial DNA markers, such as the cox1, are suitable for species identification and establishing relationships between members of a population since they evolve fast in nematodes and, therefore, generate a higher degree of genetic variability than other molecular markers (Blouin, 2002; Chan et al., 2021). As pointed out before, molecular yardsticks for classifying cryptic diversity or separating species should adapt to each species, since each group has its own intrinsic variability (Chaves-Gonzalez et al., 2022). Despite the great variability in the ITS1 loci, cox1 sequences mirrors the clustering of the ITS1 (Rojas et al., 2018a), therefore, the latter is more practical to analyze. Bayesian inference phylogenetic tree, nucleotide distance PcoA, haplotype network and phylogeographic analyses for the S. lupi cox1 sequences confirmed the separation within the previously defined genotype 1 (Rojas et al., 2018a) into genotype 1A with sequences from specimens from the Americas and Israel and genotype 1B with sequences from South African, Indian and Chinese S. lupi. In addition, the results shown herein maintain the S. lupi genotype 2 with sequences of worms from Hungary (Rojas et al., 2018a). When comparing sequences from the Americas to the other sequences within genotype 1, the nucleotide distance heatmap showed a higher pairwise identity among the American vs Israeli specimens, than with Indian, Chinese, and South African specimens. Thus, verifying that the American sequences belong in the genotype 1A along with the Israeli sequences.

For the cox1 analysis of S. lupi it has been determined that the optimal number of k clusters is 3. Therefore, this analysis has been useful to categorize individuals into different populations based on objective features, such as genetic data, and not just on traits such as geographical location or hosts the specimens were found in (Pritchard et al., 2000). All specimens from the Americas, including the ones from Peru generated in another study, were categorized in Group 1 with the Israeli specimens in Genotype 1A. Group 2 contained worms from South Africa, India and China and Group 3 clustered with specimens of Bosnia and Herzegovina, Spanish, and Hungary. The results were very similar when using k=4 and k=5 clusters, however, one single individual can be a part of multiple clusters (Rosenberg et al., 2002), which was another interesting finding in this study with the South African and Mexican specimens. In k=4, the Mexican sequence showed a small percentage of divergence from Group 1 and a genetic similarity to Group 3, which mainly involves the Hungarian specimens ( Figure 3B ). In k=5, the Mexican sequences showed a genetic similarity to Group 2, which clusters the South African specimens ( Figure 3C ). Moreover, the PcoA and phylogeographic analyses confirmed the previous findings and clearly showed the separation of Genotype 1 into 1A and 1B. With these results, it can be inferred, once again, that the American specimens are more similar to the Israeli ones.

The Peruvian specimens, unlike the rest of the American specimens, were collected from Andean foxes, not domestic dogs, and the authors in this study found less than 5% nucleotide difference between the Peruvian and African, Asian, and European sequences (Gomez-Puerta et al., 2018). The results obtained there were fairly similar to ours, when comparing the short sequences of the cox1 (Gomez-Puerta et al., 2018). These sequences also showed a 98-99% similarity to the other sequences from the Americas. This high similarity might be explained by the fact that S. lupi infected a myriad of canid species, including Andean foxes due to their phylogenetic proximity, and lack of host specificity. Andean foxes might have become infected with S. lupi as a spillover effect from infected dogs or other unrecognized or extinct infected canid hosts in the area (Guzmán-Sandoval et al., 2007; Gomez-Puerta et al., 2018). However, there is still much we don’t know about S. lupi host-specificity in the Americas. For example, a study in South Africa found that black-backed jackals (Canis mesomelas) are less susceptible than dogs to the parasite (Bumby et al., 2017), but the role of Andean foxes and other canids in the distribution and cycle of S. lupi in the Americas is not well elucidated yet. Therefore, to further understand how S. lupi circulates in this region, more epidemiological and molecular studies are needed.

One of the most interesting questions that derive from this study is the geographic expansion route of S. lupi between the continents, especially if the parasite migrated from Israel to America as suggested by the phylogeographic approach. The migration from one continent to another is not unique to S. lupi. Small et al. (Small et al., 2019), studied the spread of the human filarioid Wuchereria bancrofti through the tropics by using whole genome DNA sequences, finding that this species probably arrived to Haiti (New World) from Africa along with people during the transatlantic slave trade. In addition, the authors proposed that the genetic diversity among these parasite populations could be due to the different environments, vectors and variations in the immune response of the different hosts (Small et al., 2019). Similarly, Laidoudi et al. (2022), also highlighted the importance of the epidemiological context when discussing genetic diversity among vector-borne helminths of canids, as well as the difference in the vectors and hosts involved (Laidoudi et al., 2022). The results obtained in the W. bancrofti analysis are highly robust due to the study of the whole set of genes of several isolates. On the other hand, herein we could analyze a single mitochondrial locus from more than 20 different specimens since the genome of S. lupi has not been sequenced yet. Therefore, conclusions drawn from the present work should be interpreted cautiously. Nevertheless, to further explain the migration of S. lupi, when it spread through the Americas and the genetic diversity established, more epidemiological studies with more comprehensive, geographic and host sampling, molecular characterization, and more robust molecular datasets is needed.

Phylogeographic tools employed herein have been designed for the analysis of the evolutionary processes of rapidly changing viruses and bacteria (Bielejec et al., 2016). Importantly, these methods have been used before in the analysis of other organisms like the protozoa Hepatozoon spp. (Vasquez-Aguilar et al., 2021) or Northern Birch mice (Andersen et al., 2022) using longer sequences such as mitochondrial DNA or a higher number of specimens. Therefore, a major limitation of this study relies in the analysis of a lower number of S. lupi sequences due to the difficulty in obtaining whole specimens from dog lesions. Moreover, similar phylogeographic algorithms designed specifically for helminths are scarce or do not exist. The presented results must be interpreted with caution since mutational rates and generational times vary with genome sizes and will vary in viruses and bacteria (Lynch, 2010), and in this case, in nematodes. The herein analyses highlight the use and limitations of bioinformatic applications and the adaptation of these tools to different biological groups, such as helminths.

Spirocerca lupi collected from three geographic areas of the Americas belongs to genotype 1 according to the analysis of a cox1 fragment and is closely related to specimens from Israel, as inferred using a combination of phylogenetic, haplotype and phylogeographic analyses. Additionally, it is suggested that the Spirocerca spp. evolutionary history began with S. vulpis in Eurasia, leading to its divergence into S. lupi which later spread to Africa, Asia, and South America, along with canid hosts. This study analyzed a single locus of 32 S. lupi and S. vulpis specimens from 11 different geographical regions of the world, and thus, these results should be interpreted with caution since phylogeographic algorithms are designed for bacterial or viral genomes. This in turn, highlights the need for bioinformatic tools tailored to reflect the evolutionary history of helminths.

The original contributions presented in the study are included in the article/Supplementary files, further inquiries can be directed to the corresponding author.

The manuscript presents research on animals that do not require ethical approval for their study.

PA-S, JR-Q and AR conceived the study. VM-H, GV and RR-V provided S. lupi specimens. PA-S, JR-Q, JM-M, GB, and AR analyzed the data and generated the figures. PA-S and AR wrote the first draft of the manuscript. All authors critically reviewed and edited the manuscript and approved its final version.