Over 15 genera of protozoa parasitizing humans are known, some of these are considered natural commensal and others are related to intestinal infections responsible for generating symptoms in the infected hosts. The most frequent protozoans with cosmopolitan distribution are Giardia, Entamoeba, and Blastocystis. In Colombia, as reported in the latest national survey of intestinal parasitism in school population for the years 2012–2014, the most prevalent intestinal protozoan pathogen was Blastocystis with 57.7%, the second was Entamoeba histolytica/dispar/moshkovskii complex (Entamoeba complex) with 17% and the third was Giardia with a national prevalence of 15.4% (Ministerio de Salud y Protección Social, 2015). Additionally, Cryptosporidium has been reported with a frequency of 0.5% in the country. Currently, these are considered emerging opportunistic parasites of importance in public health at international level (Yoder et al., 2012). Factors associated with these parasitic infections are usually fecal contamination of soil and food, access to drinking water, wastewater use, lack of sanitation and vulnerable socio-economic conditions. Thus, maintaining the prevalence of intestinal parasitic infections in populations with these characteristics generating a latent risk to endure a dynamic transmission among its inhabitants (Ortiz et al., 2012).

One of the most common parasitic diseases in different geographical areas worldwide is giardiasis. It is estimated that 200 million people have symptoms and approximately 500,000 new cases per year are reported (Casero et al., 2015). In a retrospective study sought to describe the incidence of giardiasis in Colombia, for the years 2009–2013 revealed a report of 15,851 cases, of these, 50.3% were women; 58.4% were under 10 years old and 14.8% were 10–19 years old. Some survey data indicate that in industrialized countries, the prevalence ranges between 2 and 5% and the rate for developing countries ranges from 20 to 30% (Rodríguez-Morales et al., 2016). This disease is caused by Giardia intestinalis, an enteric pathogen that parasitize humans, domestic animals and wildlife. Studies in Giardia isolates have allowed to characterize the genetic diversity of the parasite, based on molecular markers such as the small subunit rRNA, glutamate dehydrogenase (gdh), triose phosphate isomerase (tpi) and β-giardin (Ryan and Caccio, 2013), allowing to identify eight assemblages where A and B are related to infections in humans and animals and further subdivided into sub-assemblages: AI, AII, BIII, and BIV with no strict associations based on clinical presentation (Mayrhofer et al., 1995; Thompson, 2000). Assemblages C and D infecting dogs, assemblage E infecting ruminants and pigs; Assemblages F infecting cats; G infecting rodents and assemblage H infecting seals and gulls (Monis et al., 2009). These parasitic subpopulations are of epidemiological significance as a potential zoonosis and wide distribution in domestic animals that generate different transmission routes (Ramírez et al., 2015). This re-emerging parasite is a protozoan that produces symptomatic conditions such as diarrhea, abdominal pain and malabsorption syndrome (Stensvold, 2013; Roberts et al., 2014).

Likewise, the blastocystosis is an intestinal colonization caused by a stramenopile known as Blastocystis which is often observed in fecal samples from humans and animals; with a worldwide distribution frequency reaching 100% in developing countries and exceeding 56% in developed countries (Scanlan et al., 2014). Blastocystis pathogenicity remains controversial, some authors suggest that the presence of infection might cause diarrhea, flatulence, bloating, urticaria and irritable bowel syndrome (IBS) (Casero et al., 2015). However, recent microbiome studies suggest that Blastocystis colonization is usually associated with a healthy gut microbiota, rather than with gut dysbiosis generally observed in metabolic or infectious inflammatory diseases of the lower gastrointestinal tract (Audebert et al., 2016). Also, a metagenomics approach showed that individuals with intestinal microbiota dominated by Bacteroides were much less prone to having Blastocystis-positive stool than individuals with Ruminococcus and Prevotella-driven enterotypes showing that the presence of Blastocystis might be benefitial for the human health (Andersen et al., 2015). Blastocystis has been divided into subtypes (STs) based on 18S rDNA polymorphisms and so far, 17 STs have been identified. The STs related to infections in humans and animals are 1-9 and 12, and the STs 10, 11, 13–17 have only been detected in animals (Scanlan et al., 2014). However, the pathogenicity of this organism is under strong debate, mainly due to a high rate of asymptomatic carriers, the differences in host susceptibility, intestinal microbiota and/or different pathogenic potential of different genetic STs (Stensvold et al., 2012; Ramírez et al., 2014). In the case of Colombia, the reported studies suggest the existance of STs 1, 2, 3, 4, 6, and 7 (Ramírez et al., 2016).

Lastly, Cryptosporidium is a cause of cryptosporidiosis. A diarrheal disease affecting especially in children under 5 years and patients with immune deficiencies (Kotloff et al., 2013). In the developed countries, diarrhea is the most common reason for missing work, while in the developing world, it is a leading cause of death. Internationally, the mortality rate is 5–10 million deaths each year (Nemes, 2009). In this scenario, Cryptosporidium is a major cause of diarrheal disease, globally (Shirley et al., 2012). Unlike many common causes of infectious enteritis, control and treatment of this infection are still problematic. Different reports indicated that cryptosporidiosis causes more than 99,000 deaths and 8.3 million disabilities (adjusted by years of life) by 2010; Most of these deaths and disabilities were common in developing countries (Ghenghesh et al., 2016). The course of infection may be intra and extra intestinal, symptomatic and asymptomatic in some individuals causing secretory diarrhea and malabsorption syndrome (Davies and Chalmers, 2009).

Molecular tools have been used to discriminate and subtype Cryptosporidium species. One of the popular subtyping tools is the DNA sequence analysis of the 60 kDa glycoprotein (gp60, also called gp40/15); Gp60 is found on the surface of the apical region of invasive phases of the parasite, and one of the dominant targets for neutralizing antibody responses in humans (Wanyiri et al., 2007). Currently, nearly 20 Cryptosporidium species and genotypes have been reported in humans, including C. hominis, C. parvum, C. meleagridis, C. felis, C. canis, C. cuniculus, C. ubiquitum, C. viatorum, C. muris, C. suis, C. fayeri, C. andersoni, C. bovis, C. scrofarum, C. tyzzeri, C. erinacei, and Cryptosporidium I genotypes from horse, skunk, and chipmunk (Xiao, 2010; Liu et al., 2014; Ryan and Hijjawi, 2015). Within each subtype family, subtypes differ from each other mostly in the number of trinucleotide repeats (TCA, TCG, or TCT microsatellite; Xiao, 2010).

The incidence and prevalence of intestinal parasites in children is most likely due to lack of natural or acquired resistance and differences in behavior and habits closely related to environmental and socioeconomic determinants. In Colombia, there is information of the most common intestinal parasites but still lacks information about parasitic subpopulations (genotypes, subtypes, or assemblages) circulating in the population. In the case of the Colombian Amazon, it is not known in indigenous communities the frequency of different intestinal parasites that may be present in these ethnic groups, which are at risk of infection because of living in inappropriate conditions, associated with fecal contamination of soil and food, inadequate drinking water, wastewater use, lack of sanitation and vulnerable socioeconomic conditions. Thus, the prevalence of intestinal parasitic infections in populations with these sociodemographic characteristics is maintained, creating a risk for transmission dynamics among its inhabitants and a public health problem in the country (Rinne et al., 2005). Therefore, the aim of this study was to perform molecular diagnosis and genotyping from stool samples positive for Giardia, Blastocystis, and Cryptosporidium in children under 15 years of four indigenous communities in the Colombian Amazon.

The diagnosis of gastrointestinal parasites has traditionally been based on feces microscopy and the use of conventional techniques based on concentration for the subsequent visualization of parasitic stages. This leads to obtain a variability in sensitivity and diagnostic specificity, being totally dependent of the observer's expertise (Mejia et al., 2013). The object then is to improve and/or strengthen the diagnosis of parasitic infections, for which the use of molecular tools arises due to its high sensitivity allowing specific identification of various pathogens, i.e., species identification and subpopulations as is the case of Giardia assemblages, Blastocystis STs and Cryptosporidium species. This translates into specific and timely treatment to the organism and better clinical management of the patients. One of the applied/used strategies by different research groups is the molecular epidemiology. It has proven to be of great importance as it helps to understand the transmission dynamics and genetic determinants of parasitic infectious diseases (Ramírez et al., 2012). For example, recent reports have shown that giardiasis in childhood is associated with malnutrition, stunting and impaired cognitive development, regardless of the presence of diarrhea, prevalent among children two to five years old, according to several reports from developing countries (Montresor, 2002). It is clear that the incidence of intestinal parasites is higher in children than in adults because of the lack of natural resistance and differences in their behavior and habits (World Health Organization, 2012). The above closely related to the environmental determinants and socio-economic characteristics of a population, which generate a higher risk of infection. The World Health Organization (WHO) in its action plan raises early protection of children and vulnerable communities; detection, treatment and control of intestinal parasites are subject to follow, because this is one of the leading causes of morbidity and mortality in children worldwide (Haque et al., 2005; Ministerio de Salud y Protección Social, 2015).

An immune response, which can be individually variable and influenced by nutritional status, genetic factors, and repeated exposure to G. intestinalis suggesting that, contributes to low detection rates of Giardia seen in older children as well as an explanation to the asymptomatic infections (Forsell et al., 2016). Our results are in accordance with these premises, where we observed high rates of infection for Giardia (23.7% by microscopy and 64.8% by qPCR) and Blastocystis (35.2% by microscopy and 88.7% by qPCR) and also for other parasites such as Entamoeba, Ascaris, Trichuris, and hookworms. Also, the frequencies of infection for Giardia and Blastocystis are in accordance to other reports in Colombia showing that in the Amazon region Giardia has a frequency of 37.3% and Blastocystis has a frequency of 40.7%, comparable to the frequencies of the country being for Giardia 15.4% and Blastocystis 57.7% (Ministerio de Salud y Protección Social, 2015). The differences are striking in terms of the sensitivity of the techniques which are also supporting the fact of using molecular methods instead of conventional microscopy as has been clearly stated by different authors (Bertrand et al., 2005; Asher et al., 2012; Mejia et al., 2013; Boadi et al., 2014; Ramírez et al., 2014). These can be explained by the intermittent excretion of cysts in Giardia and also the expertise of the operator who implements microscopy.

The geographical distribution of the G. intestinalis assemblages is intriguing since there is a lack of geographical structuring across the globe (in the case of the human-infective assemblages A and B). In central and south-America, there exist areas with differential distribution of assemblages. Some reports from Mexico, Brazil and Colombia identifies higher frequencies of assemblage A (Eligio-Garcia et al., 2008; Kohli et al., 2008) while in Nicaragua, Argentina and also Colombia some authors report the predominance of assemblage B (Lebbad et al., 2008; Minvielle et al., 2008; Arroyo-Salgado et al., 2014; Ramírez et al., 2015). This suggests that the distribution is not geographically associated and is more linked to the socioeco-epidemiological factors of the population studied. This is corroborated by our findings where the most frequent assemblage was AI (61%), contrasting the different findings in three sympatric regions of the country. For example in the Caribbean and central region the most frequent assemblage was B (Arroyo-Salgado et al., 2014; Ramírez et al., 2015) but in the Andean region assemblage A was more frequent (Rodríguez et al., 2014). Intriguingly, our results showed that the most frequent sub-assemblage was AI (61%) that has been more associated to animals than humans. This could be attributed to the lack of water treatment of the population studied and a close contact with wild animals as is frequent in the amazon region. The presence of AI could be also possible due to contamination of public water with raw sewage from animal and human sources (Volotão et al., 2007; Helmy et al., 2009). It is clearly that the indigenous populations studied lack the access to potable water that is highlighted in our findings.

Our results also showed that after AI, assemblage BIII was the second most frequent. Curiously, this assemblage is highly associated to infections in dogs and cats which might be suggesting a strong zoonotic transmission in the community of Villa Andrea with 35.5% positivity in children under 5 years (Table 1; Lebbad et al., 2008; Ryan and Caccio, 2013). Regarding the genetic diversity indexes is observed that San Juan del Socó is the community with the highest genetic diversity and the second community with the highest rate of Giardia infection (65%), and assemblage AI was the most frequent (40.4%). Curiously, this community is the farthest from the municipality of Puerto Nariño suggesting that active enzootic transmission might be occurring representing a hot-spot of diversity where distinct genotypes from animals might be transmitted to humans.

The existence of many subtypes within families of C. hominis and C. parvum supports the complexity of Cryptosporidium transmission and its zoonotic potential in developing countries (Leav et al., 2002; Gatei et al., 2007). The subgenotypes considered anthroponotic (IIc and IIe) are the most common, along with the IId zoonotic subgenotype and the subgenotype IIc present in human infections (Akiyoshi et al., 2006; Sharma et al., 2013). These data supports our findings of IdA19 and IaA12R8 in C. hominis and IIcA5G3c in C. parvum. Herein, we show the first report of C. viatorum in Colombia; the data collected in different studies indicate the presence of C. viatorum in Asia (Nepal, Bangladesh, India, Pakistan, and potentially Dubai), Africa (Nigeria, Ethiopia and Kenya) and Central America and the Caribbean (Guatemala and Barbados) (Stensvold et al., 2015). C. viatorum is currently the only species of Cryptosporidium that has been reported only in humans but the data are still insufficient to ensure that this species is restricted to human infections; also presumed presents a global distribution (Elwin et al., 2012). Previous studies indicate that the C. viatorum species has a genetic variation which may be related to differences in the geographical origin of the infection in different regions where it has been found (Elwin et al., 2012).

On the other hand, a high frequency of Blastocystis was observed in our study (88.7%; Table 1), which agrees with previous reports of infection by Blastocystis in Colombia (Arias et al., 2010; Ramírez et al., 2014). In addition, this frequency is very similar to other countries in South America ranging from 40 to 70% in humans (Jiménez et al., 2012; Ramírez et al., 2016). Regarding the distribution of STs in humans, a study in South America showed a vast diversity occurrence of STs (1, 2, 3, 4, 5, 6, 7, 8, and 12). In the case of Colombia have been reported STs 1, 2, 3, 4, 6, and 7 in humans (Malheiros et al., 2011; Ramírez et al., 2016). Our results show the predominance of STs 1, 2, and 3 which is consistent with previous reports even in indigenous communities such as Mato Grosso, Brazil (Malheiros et al., 2011). Herein, we show several important findings which are the description of ST4 in indigenous communities since this ST is more common in Europe and less frequent in South America. In addition, ST6 that is more common in birds and may be an evidence of zoonotic transmission across the population studied (Stensvold et al., 2007). When the results of the 18S alleles were retrieved, we observed that allele 4 from ST1 was frequent across the four communities; this allele has already been described in Colombia and with a frequency of 9% in all South America. Also, this allele has been reported in Didelphis marsupialis from Colombia supporting again a plausible zoonotic transmission (Ramírez et al., 2014, 2016). In the case of the ST2, alleles 12 and 15 were observed in all the communities as has already been described in South America with frequencies of 14 and 4% respectively, but intriguingly San Juan del Socó showed a higher number of ST2 alleles (6 in total) and in accordance with the genetic diversity indexes already described (Table 3; Figure 3). Also, allele 9 was found in this community and previously described in dogs and rats suggesting again a possible zoonotic profile (Ramírez et al., 2014). Regarding the ST3, the most frequent allele was 34, which has been reported in Colombia and with a frequency of 12% in South America (Ramírez et al., 2016). Interestingly, this allele has been associated to urticaria in Argentinean patients and also to cattle in Colombia (Ramírez et al., 2014; Casero et al., 2015). For ST4, we found alleles 42 and 90, the allele 42 has been described worldwide but allele 90; is novel in our country because dogs and cats have been found infected with alleles 42 and 133 (Alfellani et al., 2013; Stensvold and Clark, 2016). Finally, for ST6 we identified allele 134 in Puerto Nariño which is a novel description for the country because there is only a report of allele 122 (Ramírez et al., 2016).

According to the results obtained with mixed infections between Blastocystis and G. intestinalis (Figure 4) once again puts in evidence the need to employ techniques based on the use of different molecular markers to conduct a real identification of different subpopulations present in a specific area, and achieve associations regarding risk factors (Ryan and Caccio, 2013). Additionally, in the case of mixed infections, it is highlighted the frequency of infection with ST5, its presence in Latin America is only described in Bolivia 20% (n = 8; Ramírez et al., 2016), previously it had been reported in the UK, Australia and Pakistan (Yakoob et al., 2010; Wang et al., 2014). The importance of this ST is the zoonotic potential, as had previously been described in infections in pigs and cattle (Yan et al., 2007), reaffirming the importance in direct relation to human activities and possible infections that can occur between different epidemiological scenarios. In short, it is worth noting the fact of finding a high rate of mixed infections 13.7%, this being an indicator of active transmission by generating new MLGs as observed in Trypanosoma cruzi (Ramírez et al., 2012).

The Amazon basin is the largest in the world and is the fifth freshwater reserve on the planet. Contradictorily, people living in these areas have poor quality of life, which favors the infection of diseases of fecal-oral transmission. Therefore, proper control of giardiasis, blastocystosis and cryptosporidiosis particularly in the Amazon region requires improving the quality of drinking water and reducing environmental contamination by feces, and control of the close contact with wildlife; despite of its high prevalence and its impact on health worldwide, enteric protozoa infections emerge as neglected because conditions are not taken into account as the controls used for infections caused by helminths. In different studies have shown that people living near a water body like a river, are at greater risk of becoming infected with some intestinal parasite as in the case of G. intestinalis and Blastocystis. In a recent study from the Brazilian amazon, the authors found a predominance of assemblage A which is in accordance with our findings (Coronato Nunes et al., 2016). This highlights the importance of our findings, and shows the need to implement strategies for control and prevention in the communities studied because they are considered to be vulnerable indigenous communities.

In conclusion, we employed molecular methods and high-resolution genotyping to unravel the transmission dynamics of Giardia, Blastocystis, and Cryptosporidium across four indigenous communities of the Colombian Amazon basin. We observed a high profile of zoonotic transmission regarding the Giardia assemblages and Blastocystis STs/alleles. Also, we highlight the elevated frequency of infection by these two protozoans suggesting an active transmission in the area. San Juan del Socó was the community with the highest genetic diversity for both pathogens and the closest to the jungle space. Our findings reinforces the need to deploy better epidemiological surveillance systems for enteric pathogens in the area and the use of molecular epidemiology for a vast comprehension of the transmission dynamics including the zoonotic potential of these species.

JR analyzed the data, revised the final version and supervised the project. NG, JT, LS, ASal, CR, and MR analyzed the samples by microscopy. ASan wrote the manuscript, analyzed the data and conducted the molecular biology experiments. MM conducted the statistical analyses. PR and ML revised the manuscript and analyzed the data. LX and YQ conducted the typing of Cryptosporidium.

We thank Dirección de Investigación Universidad Nacional de Colombia grant number 22946 for funding this project.