Introduction

Microsporidia are a ubiquitous group of obligate intracellular parasites that infect all major animal groups (both vertebrates and invertebrates). Transmission occurs mainly through fecal-oral routes, with sources of infection including other infected humans and animals [1], contaminated water [2], [3], [4], and food [1]. Thus far, at least 14 microsporidian species are considered to be pathogenic for humans, Enterocytozoon bieneusi being the most prevalent species in humans worldwide followed by Encephalitozoon intestinalis [1], [5], [6], [7]. Gastrointestinal tract is the main localization of these parasites with chronic diarrhea as the most frequent clinical manifestation.

Several microsporidia species have been confirmed in a broad range of animals [8], [9], suggesting a zoonotic potential. Although the epidemiology of microsporidian infections still remains unclear, application of molecular techniques based on PCR and sequence analysis, both for species identification and within species genotyping, has increased the available information of microsporidia genetic diversity and it has also proven to be useful in the identification of route(s) and source(s) of transmission of these parasites. In the case of E. bieneusi, sequence analysis of the internal transcribed spacer (ITS) region has revealed a considerable genetic variation within E. bieneusi isolates of human and animal origins, with more than 100 genotypes described so far [7], [9], [10], [11]. These studies have identified E. bieneusi genotypes host-associated to several animals and also some genotypes with no host specificity, which are considered zoonotic [12], [13], [14]. Henriques-Gil et al. [15] identified 4 groups (I-IV) of genotypes strongly differentiated from each other. Group I (zoonotic group) included most of the published sequences (94%), with genotypes that are associated with nearly all human infections [15]. The occasional sequences included in the divergent groups II, III, and IV have so far been restricted to specific hosts such as raccoons (III) and cats and dogs (IV).

In Spain, microsporidia have been confirmed in humans, animals and environmental samples, indicating that these parasites could be more frequent than previously thought and that they should be considered as a potential public health problem. Both Immunocompromised and immunocompetent populations are affected with this infection, with data on microsporidia presence not only in HIV/AIDS patients [16], [17], [18] but also in HIV-negative patients including travelers [19], the elderly [20], organ transplant recipients [21], Crohn’s disease patients [22] and the immunocompetent population [23]. Similar to data described worldwide, E. bieneusi is the species most frequently identified. Animals hosts including dogs, goats and rabbits [24], [25], pigeons [26], [27] and soil and fecal samples (presumably from cats and dogs ) from urban parks [28] have also been described for microsporidia in several regions of Spain, with E. bieneusi and Encephalitozoon hellem as the species identified. All these data suggest that human pathogenic microsporidia circulate in the environment in Spain and support the idea that the most frequent microsporidia associated with human infection are of zoonotic origin.

In this study, we investigate the presence of microsporidia spores in fecal samples from domestic and wild animals from Spain using microscopic and molecular methods. Additionally, taking into account that there is little data on molecular characterization of microsporidia in animal samples in the country, we also studied the genetic diversity of E. bieneusi from animal sources including domestic, farm and wild animals, through the sequence analysis of the ITS region. Our results should contribute to the knowledge of the molecular epidemiology of this microsporidia in Spain.

Materials and Methods

Animal stool samples and staining method

A total of 159 fecal samples from domestic (cats and dogs), farm (rabbits, pigs and ostriches) and wild (foxes) animals from different autonomous communities of Spain were evaluated for the presence of microsporidia (Figure 1). All samples were apparently normal with no signs of intestinal disorders. Fecal samples were collected by using a sterile spatula and gloves, and placed into labeled sterile container. The samples were stained with Weber’s chromotrope stain [29]. Microscopic analysis was performed with a magnification of 1000 X and positive criteria included spores with a bright pinkish red stain and either a clear vacuole-like polar end or a belt-like stripe in the middle of the spore.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Geographic distribution of the samples included in the study.

Animal samples from six different autonomous communities of Spain were evaluated.

https://doi.org/10.1371/journal.pone.0092289.g001

Results and Discussion

Eighty-five samples of 159 exhibited structures that were compatible with microsporidia spores by Webeŕs stain with 37 of them being confirmed by PCR. Microsporidia distribution data obtained by both techniques showed that dogs and pigs had the highest number of positive samples for these parasites (Table 1). The differences between the stain and PCR results could be explained by: 1- the presence of PCR inhibitors in the feces; this is a difficult sample matrix which has been shown to contain abundant inhibitors frequently co-purified with fecal DNA and which can, therefore, cause false negative detection results. 2-the presence of spores belonging to other microsporidia that are not detected with the generally group-specific PCR used in our study; this should not be surprising since there have been more than 150 genera described so far, with strong genetic divergence among them.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Microsporidia species and E. bieneusi genotypes in animals from different regions of Spain.

https://doi.org/10.1371/journal.pone.0092289.t001

Enterocytozoon bieneusi was the most commonly detected species (20 out of 37 samples), followed by E. intestinalis (4 out of 37 samples) and A. algerae (1 out of 37 samples). Undetermined species of microsporidia were detected in 14 out of the 37 samples indicating that additional species not associated with human disease were also identified in the animals examined here. Although most of the animals were infected by a single species, two pigs exhibited a co-infection. The high frequency of E. bieneusi is not surprising since a large variety of vertebrate hosts are known to be infected by this species, including humans, other mammals and birds. Consequently, any of them may act as potential reservoirs that facilitate the environmental pollution and continuous transmission of this microsporidia [9]. These findings are in agreement with several studies on microsporidia presence in animals from different regions of Spain, with most of them describing E. bieneusi as the main species identified in animals including dogs and goats [25]; pigeons [26] and dogs and rabbits [24]. In the case of E. intestinalis, despite being one of the most common microsporidian species found in humans worldwide, it has been less frequently identified in host animals compared to E. bieneusi [8]. In our study, even though the number of animals evaluated and positive samples for E. intestinalis was lower, the data obtained confirm its presence in domestic animals such as cats and pigs, and contribute to the current knowledge of the spectrum of animal hosts for this parasite, increasing the number of potential sources of infection for humans. It is also important to highlight that, to our knowledge; this is the first diagnosis of E. intestinalis and E. bieneusi in ostriches. There are only two previous reports of microsporidia presence in these animals, with E. hellem as the only microsporidia species identified [39], [40]. A. algerae primarily infects moquitoes, but has been reported in humans [41], [42], [43]. We believe this is the first identification of A. algerae in another mammal, namely pigs.

Enterocytozoon bieneusi is a complex species with multiple genotypes and diverse host range and pathogenicity [7], [9]. The genetic data on E. bieneusi diversity rely almost exclusively on the analysis of the ITS region, which provides valuable information about the transmission and pathogenic potential of this parasite. In our study, the sequence analysis of this region showed 4 distinct genotypes (Table 1). All of the genotypes belonged to Group I as previously defined by Henriques-Gil et al. [15]. In the case of dogs, all E. bieneusi positive samples (7) had genotype A (GenBank accession no. AF101197), which has long been known as a human specific genotype [9], [11], with only one report of its presence in birds from the Czech Republic [44]. This finding could suggest either that dogs can also be infected from a human source, or, that both, humans and dogs could be infected from the same environmental source. Indeed, other studies have revealed a diverse group of genotypes for E. bieneusi isolated from dogs, cats and other pets occasionally analyzed. Zoonotic transmission between dogs and humans is possible for genotypes D, Peru6, WL11, type IV [9,_ENREF_40 45] and, according to our results, also for the genotype A. However, more samples from these animals should be genotyped to confirm the zoonotic transmission of genotype A. By contrast, a dog-specific genotype, PtEb IX [9], [11] corresponding to the group II of sequences that are divergent to the vast majority of E. bieneusi ITS genotypes [15], could have an exclusive dog-to-dog transmission.

Regarding E. bieneusi diversity in the other animals included in our study, only one sample could be successfully genotyped from each animal species (pigs, rabbits, ostriches and foxes). Genotype I identified in swine has commonly been reported in livestock animals (cattle) [7], [9], [46] and more recently in humans (symptomatic children) [46], but so far there are no reports of its presence in pigs [9]. More studies should be performed in order to confirm that this genotype could be considered relevant in pigs. Henriques-Gil et al. [15] showed that genotypes obtained from swine or cattle are significantly segregated, though some of them may appear in these hosts, plus humans and other domestic animals.

Genotype D (GenBank accession no. AF023245) was detected in both fox and rabbits. This is the genotype that has the widest distribution, and has been found in virtually all host species so far analyzed, including humans, domestic and wild animals, such as foxes [7], [9], [10], [14]. This is the first report in rabbits, and should not be surprising as genotype D probably represents one of the most ancient ITS genotypes in E. bieneusi [15]. In the case of the ostriches, we identified the type IV genotype in the only sample that was positive for this microsporidia. Like the previous one, this genotype also occupies a central position in the haplotype network described by Henriques-Gil et al. [15], and likewise it has been found in animals and humans, suggesting that these birds can be included in the extensive network of potential reciprocal E. bieneusi infections.

Conclusion

This study shows that human pathogenic microsporidia species are present in domestic, farm and wild animals in Spain, corroborating their potential role as a source of human infection and environmental contamination. However, larger studies are needed to confirm that animal contact is a high risk for human infection. Our findings demonstrate that E. bieneusi genotypes of zoonotic importance are circulating in animals that are in close contact with humans, such as dogs and pigs, and therefore could participate in the transmission of this microsporidia in different regions of Spain. Furthermore, the identification of human classical E. bieneusi genotype (genotype A in dogs) suggests the absence of a microsporidia transmission barrier between humans and these animals. Additionally, since E. bieneusi is a cause of significant human disease in immunocompromised patients (HIV-infected patients and solid-organ transplant recipients), children and the elderly, among others, our results must be considered in order to establish prophylactic measures that could reduce its transmission.

Acknowledgments

The authors are indebted to Sergio Llorens for his valuable technical assistance and to Brian Crilly for helpful revision of the manuscript.