Cryptosporidium spp., Giardia duodenalis, Enterocytozoon bieneusi and Other Intestinal Parasites in Young Children in Lobata Province, Democratic Republic of São Tomé and Principe

Rare systemic studies concerning prevalence of intestinal parasites in children have been conducted in the second smallest country in Africa, the Democratic Republic of São Tomé and Príncipe. Fecal specimens from 348 children (214 in-hospital attending the Aires de Menezes Hospital and 134 from Agostinho Neto village) in São Tome Island were studied by parasitological and molecular methods. Of the 134 children from Agostinho Neto, 52.2% presented intestinal parasites. 32.1% and 20.2% of these children had monoparasitism and polyparasitism, respectively. Ascaris lumbricoides (27.6%), G. duodenalis (7.5%), T. trichiura (4.5%) and Entamoeba coli (10.5%) were the more frequent species identified in the children of this village. Giardia duodenalis (7.5%) and E. bieneusi (5.2%) were identified by PCR. Nested-PCR targeting G. duodenalis TPI identified Assemblage A (60%) and Assemblage B (40%). The E. bieneusi ITS-based sequence identified genotypes K (57.1%), KIN1 (28.6%) and KIN3 (14.3%). Among the 214 in-hospital children, 29.4% presented intestinal parasites. In 22.4% and 7.0% of the parasitized children, respectively, one or more species were concurrently detected. By microscopy, A. lumbricoides (10.3%) and Trichiuris trichiura (6.5%) were the most prevalent species among these children, and Cryptosporidium was detected by PCR in 8.9% of children. GP60 locus analysis identified 6.5% of C. hominis (subtypes IaA27R3 [35.7%], IaA23R3 [14.3%], IeA11G3T3 [28.6%] and IeA11G3T3R1 [21.4%]) and 2.3% of C. parvum (subtypes IIaA16G2R1 [20.0%], IIaA15G2R1 [20.0%], IIdA26G1 [40.0%] and IIdA21G1a [20.0%]). G. duodenalis and E. bieneusi were identified in 0.5% and 8.9% of the in-hospital children, respectively. G. duodenalis Assemblage B was characterized. The E. bieneusi genotypes K (52.6%), D (26.4%), A (10.5%) and KIN1 (10.5%) were identified. Although further studies are required to clarify the epidemiology of these infectious diseases in this endemic region the significance of the present results highlights that it is crucial to strength surveillance on intestinal pathogens.


Introduction
Intestinal parasites are an important cause of morbidity and mortality worldwide, especially in low-middle income countries in tropical and sub-tropical regions. The hot and humid climate, high population density, poor conditions of hygiene and the presence of insects as vectors or merely as mechanic carriers of parasites, limited economic resources, and some social-cultural habits (food and others), promote undoubtedly the parasite transmission in these regions [1]. It is estimated that worldwide, about two billion people are affected by soil transmitted helminths such as Ascaris lumbricoides, Ancylostoma/Necator spp. and Trichuris trichiura, 50 million by Entamoeba histolytica and 2.8 million by Giardia duodenalis (synonymous G. lamblia and G. intestinalis) [2][3][4]. Another protozoan, Cryptosporidium, is considered an important agent of intestinal disease in humans worldwide [5,6]. More recently, the emerging microsporidia, especially Enterocytozoon bieneusi species has often been described as a frequent human pathogenic microorganism causing gastrointestinal infections and/or disseminated pathology, according to the species involved [7]. In addition to infecting humans, G. duodenalis, Cryptosporidium spp, and microsporidia are found in a wide range of animals including livestock, companion animals, and wildlife worldwide.
Despite their wide occurrence, giardiasis and cryptosporidiosis are considered neglected diseases by the World Health Organization, largely due to lack of studies in low-middle income countries [8]. The epidemiology of microsporidiosis is less clear in these deprived regions where few studies have been carried out.
The transmission to humans of some helminthes, protozoa and microsporidia is via the fecal-oral route and can occur through direct contact with infected persons (anthroponotic transmission) or animals (zoonotic transmission), or by ingestion of contaminated food (foodborne transmission) or water (waterborne transmission) [9]. The fecal-oral transmission route is facilitated by the contamination of water and soil due to the absence of adequate sanitation and hygiene, especially in rural areas of developing countries. When the water/soil is contaminated, the resilient infective forms (eggs, cysts, spores) of the pathogenic organisms can be transported to vegetables, fruit, hands, tools, handles doors, currency, etc. and be easily swallowed accidentally by humans [10].
The use of molecular approaches has allowed inter-and intraspecific genetic characterization of G. duodenalis, Cryptosporidium spp., and E. bieneusi, thus facilitates the identification of infection source(s) of these pathogens and improve understanding of their epidemiology.
With a population of about 183,176 [11], the Democratic Republic of São Tomé and Príncipe (DRSTP) is the second smallest country in Africa, and lies approximately 180 miles from Gabon on the West African coast. The two main islands (São Tomé and Principe) form part of a chain of extinct volcanoes and both are mountainous. São Tomé and Principe islands feature a tropical wet and dry climate with a relatively lengthy wet season (October to May) and a short dry season (June-September). The country is one of the smallest economies in Africa. Economic growth is estimated to be 4.5% in 2012, marking a gradual recovery. Nevertheless, poverty remains rampant affecting more than 50% of the population [12]. Enteric diseases in low-income countries are an important public health problem that may be manifested as impaired mental health, impaired growth and poor academic performance, especially among children and adolescents, which are more at risk of infections than the adult population [13]. São Tomé and Principe islands have all environmental, economic and social conditions that favor the occurrence and transmission of enteric pathogens. However, limited systemic studies concerning prevalence of intestinal parasites in children have been conducted to elucidate the epidemiology of these parasites in DRSTP [14,15]. The present study was conducted to assess the prevalence of helminthes, protozoan and the microsporidia Enterocytozoon bieneusi in children in DRSTP.

Ethical Statement
This study was approved by the Ministry of Health and Social Affairs of The Democratic Republic of São Tomé e Príncipe and Dr. Ayres de Menezes Hospital (S. Tomé). Informed written consent was obtained from the parents or guardians on the behalf of the children involved in the study.

Patient Population and Specimen Collection and Processing
Fecal specimens were collected from 348 children in São Tomé Island from July 2009 to September 2009. Among them, 214 inhospital children attending the Aires de Menezes Hospital (0u21'23"N 6u43'20"E, integrated in an urban area of São Tomé Island) and all children (134) living at Agostinho Neto village (0u22'4"N 6u38'38"E), a typical African rural community with more than 1,600 acres, north District of Lobata (São Tomé Island) were enrolled in the study. Multiple specimens (two to three) were collected from both groups of children. A structured questionnaire was used to collect socio-demographic characteristics (age, gender), clinical information (symptoms: diarrhea, fever and gastrointestinal complaints) and other potential risk factors (source of drinking water, sanitation, feeding habits, food preservation and domestic animals). Diarrhea was defined as the emission of three or more unformed stools within a 24-h period. All samples were stored at 4uC in 2.5% (w/v) potassium dichromate solution without preservatives prior to being used in parasitological and molecular characterizations.
The stool samples were treated by a modified water-ether sedimentation method. For detection of parasitic agents, a wet mount of each fresh concentrate fecal sample was prepared in a saline and iodine solution and observed under light microscopy for eggs, cysts and trophozoites of various parasites. Also, a permanent slide was prepared for each sample. Smears were examined using modified Ziehl-Neelsen staining for the presence of Cryptosporidium spp., Cyclospora sp. and Cystoisospora belli.

Microscopy and PCR methods
PCR was performed on all collected samples. Genomic DNA was extracted from feces using the FastDNA SPIN kit for soil [16]. DNA preparations were analyzed by a small subunit (SSU) rRNA-based polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) technique that detects and differentiates common Cryptosporidium species in humans [17]. Cryptosporidium hominis and Cryptosporidium parvum present were subtyped by a PCR-sequencing tool targeting a ,400-bp fragment of the 60 kDa glycoprotein (gp60) gene [18]. The recommended subtype nomenclature was used in naming C. hominis and C. parvum subtypes [9].
E. bieneusi was detected by nested-PCR analysis of the entire internal transcribed spacer (ITS) region of the rRNA gene. E. bieneusi genotyping was carried out by sequence analysis of the ITS rRNA gene [7].
Three PCR replicates per sample using 3 ml of extracted DNA per PCR, were performed to increase the accuracy of the methods. PCR products were analyzed by 1.5% agarose gel electrophoresis and ethidium bromide-staining.
Samples that gave positive results with at least one method used were considered positive for this study.

DNA sequence analysis
The secondary PCR products were purified using a Jetquick kit and sequenced in both directions on an ABI3100 automated sequencer (Applied Biosystem, Foster City, CA.). The accuracy of the nucleotide sequence was confirmed by sequencing two separate PCR products from the same sample. The sequences obtained were analyzed together with the reference sequences from the GenBank database using the BLASTN (www.ncbi.nlm. nih.gov) and ClustalX (ftp://ftp-igbmc.u-strasbg.fr/pub/ ClustalX/) programs to determine Cryptosporidium, G. duodenalis and E. bieneusi genetic variability.

Data analysis
The Chi-square test or Fisher's exact test was used to compare prevalence of parasites among age groups, gender, symptoms (diarrhea, fever and gastrointestinal complaints) and to evaluate potential risk factors (source of drinking water: piped water, well; sanitation: public sewer, septic tank; indoor toilet; feeding habits: breastfeeding, bottlefeeding, solids, breastfeeding + solid feed, bottle feeding + solid feed, breast + bottle + solids; food preservation: refrigerators; domestic animals). All statistical analyses were performed using SPSS version 17.0. A P value of ,0.05 was considered statistically significant.  (Tables 2 and 3).
Among the children positive for helminthes, protozoa and E. bieneusi the majority reported no symptoms. Out of a total of 45 helminth-positives, 25 G. duodenalis-positive and seven E. bieneusipositive children, only four, two and one child, respectively, reported to have diarrhea. Fourteen children with helminth infection had gastrointestinal complaints and four presented fever. Only seven of the G. duodenalis-positive children reported fever.
1.2 Intestinal parasites and risk factors. No statistically significant differences were observed between the presence of the identified parasites in fecal samples among children and the following parameters evaluated: age, gender, symptoms, presence of public sewer, indoor toilet, refrigerators and domestic animals.
The identification of G. duodenalis by TPI based nested-PCR was more common in children (17.9%; 5/28) who reported having a well next to home than in children (4.7%; 5/106) without well (Fisher's exact test = 5,538, P = 0.033). In addition, the presence of G. duodenalis was more frequent in children having septic tank (18.6%; 8/43) near the house than in children who did not have septic tank near home (2.2%) (Fisher's exact test = 11,383, P = 0.002).
Among children who were positive for intestinal parasites by microscopy, 60.7% (68/112) consumed solid food (exclusively or on had bottle feeding) and 4.6% (1/22) were breastfed. The association between the presence of intestinal parasites and children on solid food diet (exclusively or had combined feeding) was highly significant (x2 = 23,226, P#0.001).
In the group of breastfed, all children had more than 2 years old and only in one child (4.6%) was detected the presence of intestinal parasites by microscopy. On the other hand in the group of children on solid food diet (the majority eat solid food exclusively and one had bottle feeding) there were similar percentages of positives among children over 2 years in comparison with children under 2 years (59.8% vs 64%).
2. In-hospital children from an urban area (Aires de Menezes Hospital).
G. duodenalis and E. bieneusi were also identified by PCR in 0.5% (1/214) (the sample was also positive by microscopy) and 8.9% (19/214) of the in-hospital children, respectively (Table 3). Sequence analysis of TPI locus identified Assemblage B in the only G. duodenalis positive sample. Based on PCR analysis of the E.  Table 3. Distribution of intestinal parasites (microscopy and/or PCR) and genetic characterization of Cryptosporidium spp., Giardia duodenalis and Enterocytozoon bieneusi among children from the two communities analyzed.
Note: a Subtypes, Assemblages and genotypes are indicated for C. hominis and C. parvum, G. duodenalis and E. bieneusi, respectively; b Samples identified by PCR were also positive by microscopy; c Despite several efforts is was not possible to identified A sub-assemblages; ND -Not determined. Among the children found to be positive for helminthes, protozoa and E. bieneusi more than 50% reported diarrhea and/or gastrointestinal complaints.
2.2. Intestinal parasites and risk factors. No significant association between the variables analyzed (age, gender, symptoms, source of drinking water, sanitation, food preservation and domestic animals) and the presence of intestinal parasites was observed in children from the Aires de Menezes Hospital. However, a higher number of children infected with intestinal parasites were observed among those who were on solid food diet (exclusively or had combined feeding) (21.2%; 38/179) than those who were breastfed (5.7%; 2/35) (x2 = 4,637, P = 0.031).
As noticed for the children of the village, in the group of breastfed (exclusively) all children were more than 2 years old and in only two children (5,7%) the occurrence of intestinal parasites was detected by microscopy. In the group of children on solid food diet (exclusively or had combined feeding) were detected a higher percentage of intestinal parasite-positives (43,4%) among children over 2 years old in comparison with children under 2 years (11,9%). Moreover, this latter association is statistically significant (x2 = 23.125, P#0.001)
Children from Agostinho Neto village showed significantly higher probability to be infected by at least one species (

Discussion
Intestinal parasites once considered to be controllable in developed countries remain a major cause of morbidity and mortality worldwide [20]. Poor water supply and sanitation are two of the most critical factors associated with the spread of intestinal diseases in low-income countries. Most of the population living in these regions is comprised of people without sustainable access to safe drinking water and basic sanitation. The public health importance of intestinal pathogens as a major concern in most developing countries has been exacerbated by the cooccurrence of malnutrition, malaria and HIV/AIDS. In addition, reported infection rates may be underestimated, because most of the health institutions lack suitable diagnostic methods to detect low levels of parasites. Some diagnostic tools for specific intestinal parasites, especially for the newly emerging opportunistic pathogen E. bieneusi species, are not available in peripheral health institutions [20].
DRSTP is a low-income tropical country, where climatic and living conditions facilitate the transmission of many intestinal pathogens including helminthes, protozoa, and the emerging E. bieneusi, particularly in children. Children are naturally more susceptible to infections and generally have poorer personal hygiene habits, thus with a potential higher exposure to intestinal organisms.
The present study has shown that the prevalence of intestinal parasites is high among children in Agostinho Neto village (52.2%), a typical rural African community, and in-hospital pediatric population (29.4%) in São Tomé Island. Marked differences on monoparasitism and polyparasitism rates were observed between the two populations. Children from the rural community were significant more likely to be infected with single and multiple species of parasites than the in-hospital population. The higher frequency of single or multiple-pathogen infection in children from Agostinho Neto village may possibly be linked to specific characteristics of this population. Children living in the rural community can be subject to a higher degree of environmental exposure to potential sources of infection (e.g. contaminated water, farm animals and wildlife), may have distinct behavioral habits (e.g. the present population included older children with more autonomy) and poorer hygiene practices and sanitation that may increase the risk of infection in comparison with the hospitalized children. Polyparasitism is likely the accumulated outcome of infection routes, host exposures, and susceptibility, as well as behavioral, sociological, and economic factors. Some studies have reported that polyparasitism may contribute more to morbidity than single-species infections. Also, multiple species infections may increase susceptibility to other infections, such as malaria. Consequently, efforts have to be made to better understand the consequences of the co-existence of parasites within the same host on the immunological responses to each species and, more importantly, whether such interactions affect susceptibility or clinical outcome of other pathogens [21,22].
The overall frequency of helminthes (33.6% vs 12.6%) and protozoa (29.1% vs 11.2%) were significantly higher in Agostinho Neto village's children than in-hospital children. The difference observed between the two populations is probably also associated with the antiparasitic therapy (albendazole) campaign targeting children from the urban community in the Aires de Menezes Hospital, which occurred 2-3 months before this study. However, this may only explain the lower occurrence of helminthes and Giardia since the antiparasitic drugs used are almost ineffective on the other protozoa.
Although several common helminthes were identified in fecal samples from both groups, A. lumbricoides (27.6% vs 10.3%) and T. trichiura (6.5% vs 4.5%) were the most prevalent. The infection rate of these parasites is within the range described in other studies. Nevertheless, caution should be taken when interpreting differences in infection rates using data that could have been obtained using different methodologies and/or from specific groups of population. Pampiglione et al. in 1987 reported higher prevalence of A. lumbricoides (87.7%) and T. trichiura (64.3%) in children and adults from São Tomé and Principe. The same authors also identified 4.6% of Ancylostomidae, 6.8% of S. stercoralis and 0.2% of Hymenolepis [15]. More recently Belo et al. in 2005 identified A. lumbricoides, T. trichiura and Ancylostomidae in 70.8%, 68.5% and 4.6%, respectively in schoolchildren, from São Tomé [23]. The overall prevalence of protozoan G. duodenalis and Cryptosporidium spp. was statistically different between the two populations analyzed: G. duodenalis was detected more often in the rural region (7.5% vs 1.9%) and Cryptosporidium spp. were detected more often in-hospital children (8.9% vs 0%) ( Table 3). The prevalence of these parasites in both populations is lower than the prevalence described in some studies on African pediatric populations. Epidemiological studies reported a high incidence of parasite infections, including Cryptosporidium spp. and G. duodenalis during the rainy season as compared to the hot and dry season [24][25][26][27][28]. As the fecal samples collected in this study were collected during the dry season (July-August), this could be the reason for the low presence and/or absence of these intestinal parasites among the two children populations analyzed. Both Cryptosporidium spp. and G. duodenalis are waterborne and therefore a high prevalence is expected in the wet season.
The significant association between Cryptosporidium infection and in-pediatric hospital population may be linked to: a) children from the hospital have higher physical debility due to disease responsible for hospitalization. In addition, in this study the symptoms (fever, diarrhea and gastrointestinal complaints) were more commonly reported among in-patients group; b) lower median age observed among in-patients (1 year old) in comparison with the median age of children from the rural area (4 years old). Children younger than 2 years of age are frequently infected in community and hospital settings in developing countries [25,29]. Both reasons stated may predispose in-hospital population to be more vulnerable to the infection by an opportunistic parasite as Cryptosporidium spp.
The significant association between G. duodenalis infection and children from the rural region may be related to any behavioral habit and/or source of environmental infection among this area. For instance, consumption of unsafe water clearly represents a significant risk for giardiasis. However, the contamination of such water supplies may result from humans, farm animals, and wildlife. [30]. More studies are needed to track G. duodenalis source(s) of infections among this region. The majority of children from Agostinho Neto village seemed to be asymptomatic. Out of a total of 25 G. duodenalis positive-cases only two children reported diarrhea, and seven had gastrointestinal complaints (but were also co-infected with other intestinal organisms). Heresi and Cleary suggested that Giardia were part of the natural intestinal flora of people in tropical countries [31]; however, they are likely to be present as a result of poor environmental sanitation. Although it was not possible to assess the implications of G. duodenalis infection in the health of these children (e.g. evaluate their nutritional status), chronic infections are frequently associated with impaired child growth emphasizing the importance of these results to implement monitoring and treatment of giardiasis among children from this endemic area.
In this study, E. bieneusi carriage was higher in-hospital children (8.9%) than in children from the rural community (5.2%). Both rates determined are similar to those found in previous studies on African populations (children and/or adults): Democratic Republic of Congo (1.7%) [32], Gabon (3.0%) [33] Cameroon (5.2%) [34], Uganda (6.6%) [35], Nigeria (9.3%) [36], Niger (10.5%) [37]. E. bieneusi were more common among in-hospital population (of the total of 19 E. bieneusi-positives, 16 ranged from 0-2 years) than in children from Agostinho Neto village (the seven E. bieneusipositives age ranged from 2-10 years) but this difference was not statistically significant. However, the reasons specified above to justify the higher number of cases of Cryptosporidium spp. infection among the in-hospital population can also be applied for E. bieneusi since this species is also considered an opportunistic organism and may affect younger and sick children than the apparently healthy children from the rural community.
Cryptosporidium hominis and C. parvum were the only species identified among in-hospital children. The presence of a higher rate of C. hominis (6.5%) than C. parvum (2.3%) is in accordance with previous studies in Africa and other developing countries, where 79-90% of infections are caused by C. hominis. [38]. Variations in the distribution of Cryptosporidium species in humans are considered an indication of differences in infection sources [18]. The predominance of C. hominis has also been reported in other paediatric populations in Africa, such as in South Africa [39][40][41] Malawi [42], Kenya [43], Uganda [35] and Nigeria [44], indicating that anthroponotic transmission may play a major role in the epidemiology of Cryptosporidium in this continent.
Both C. hominis Ia and Ie subtype families characterized in this study are included in the group of the four most commons subtype families (Ia, Ib, Id and Ie) that are usually observed in children in developing countries [9,41,44]. Infections with the Ia and Ie subtype families were more likely asymptomatic [17]. However, in the present study, 10 of the 14 C. hominis-positive in-hospital children had diarrhea, although the diarrhea could have been caused by other concomitant pathogens.
The C. hominis Ia subtypes identified, IaA27R3 (35.7%) and IaA23R3 (14.3%), have been previously described in GenBank Database. The C. hominis Ia subtype family observed in this study has been reported as one of the predominant subtype families in humans [45], including Nigeria [36,44,46,47]. The C. hominis Ie subtypes identified, IeA11G3T3 (28.6%) and IeA11G3T3R1 (21.4%) have been reported in human infections from other african low-middle income countries, such as Nigeria [44,46] and South Africa [41]. Similarly, to our results, in most low-middle income countries, humans with the subtype family Ie are mostly infected with subtype IeA11G3T3, with few exceptions for Jamaica and China where IeA12G3T3 has been detected [48,49] Although the anthroponotic IIc subtypes of C. parvum are responsible for most human infections in low-middle income countries [38,39,42,50] only the zoonotic IIa and IId subtypes, IIaA16G2R1 (20.0%), IIaA15G2R1 (20.0%), IIdA26G1 (40.0%) and IIdA21G1a (20.0%) were identified in the study. In the African continent, Molloy et al., reported IIa C. parvum subtypes (IIaA15G2R1and IIaA16G1R1) in two children from rural Nigeria [44], and Adamu et al. reported IIa C. parvum subtypes (IIaA15G2R1, IIaA16G2R1 and IIaA16G1R1) in 12 children in Ethiopia [51]. The IIa and, to a less degree, IId are also common C. parvum subtype families in European countries [9,52], and the subtype IIa has been detected in both humans and ruminants, for instance in Iran and other Mideast countries [53].The finding of IIa and IId subtypes families in this study may suggest that C. parvum in children in São Tomé island is likely to be of zoonotic origin. However the C. parvum IIa subtype family usually described as zoonotic was reported previously as being associated to anthroponotic transmission in developing countries [9]. Similarly to the present study, iIqbal et al. reported Ia and Ie C. hominis and IIa and IId C. parvum as the more frequent subtype families among children from Kuwait [54].
Sequence analysis of TPI locus identified Assemblages A (60%) and B (40%) of G. duodenalis among children from Agostinho Neto village, and Assemblage B (100%) in-pediatric hospital population. Studies on genetic characterization of Giardia in humans in the sub-Saharan Africa are scarce [55,56]. East Africa is considered to be the most endemic regions for G. duodenalis [57]. One of these studies was performed in rural western Uganda and G. duodenalis typing showed that the distribution between assemblage A and B was 53% and 47%, respectively [56]. More recently, Ankarklev et al. reported a predominance of Assemblage B, followed by assemblage AII and by mixed assemblage infections of G. duodenalis among young children in Uganda [58]. In Rwanda, 85.9% of G. duodenalis-positive samples isolated from children were assemblage B and 12.7% were assemblage A2, in addition to one assemblage A1 and one mixed assemblage A+B [59]. In rural and in-hospital children in São Tomé island, anthroponotic assemblages A and B predominate, indicating person-to-person transmission plays an important role in giardiasis epidemiology. This is supported also by the absence of animal-specific assemblages of G. duodenalis, such as C and D, which are common in dogs, and E, F, G, which are common in livestock, cats, and rodents, respectively [30].
The transmission of E. bieneusi in São Tomé is less clear due to the lack of data concerning this organism in the country. The majority of E. bieneusi found in this study belong to the group that infects both humans and animals. Within this group, genotype Type IV was the dominant one (in both groups of the children studied). D and A genotypes were found only in-pediatric hospital population. Both Type IV and D are well known zoonotic E. bieneusi genotypes, having been reported in humans, macaque, baboons, cattle, pigs, dogs, and some wild mammals. In contrast, genotype A is mostly an anthroponotic genotype. Both genotypes KIN1 (identified among children from Agostinho Neto Village and in-hospital population) and KIN3 (in children from Agostinho Neto Village) are within the ITS E. bieneusi genotypes group identified exclusively in humans until now. Studies carried in African countries as Uganda (Type IV) [60], Cameroon (A, D and Type IV) [33,34], Niger (A, D and Type IV) [37], Gabon (A, D and Type IV) [33], Malawi (D and Type IV) [61], Democratic Republic of Congo (D, KIN1 and KIN3) [32] and Nigeria (D and Type IV) [36,62] reported a common occurrence of these genotypes in humans. These results suggest that both antroponotic and zoonotic transmission might be responsible for E. bieneusi infection in the children studied. However, more systematic collection of epidemiological data and sampling of animals, drinking water, and fresh produce are needed to clarify the source of E. bieneusi in DRSTP children.
Both the overall detection of intestinal parasites and the cases of monoparasitism and/or polyparasitism identified by microscopy among children from Agostinho Neto Village were significantly associated with the presence of septic tank next to home. Also, significant associations between G. duodenalis PCR-positive and children that reported having well or septic tank near the house were observed. Thus, risk of exposure to intestinal pathogens identified as a result of inadequate water and sanitation is high among this rural population. Reducing the risks to which children from this region are exposed through unsafe water or adequate sanitation is particularly important given the large disease burden attributable to these issues, especially in children.
The significant association between the presence of intestinal microorganisms detected by microscopy and children that consume solid food (exclusively or have combined feeding) from both populations analyzed suggest that those parasites can have foodborne origin. However, it must be taken into account that the start of intake of solid food is associated to an increased mobility of the child, changes on behavioral and hygienic habits and therefore to an increased contact with the environment and other potential risk factors. Although no definite conclusions can be drawn as seen in other studies [63,64], data from this study suggest that breastfeeding is highly protective against intestinal diseases. For instance, protection from G. duodenalis infection by breast milk has previously been reported [65]. Taking into account the high endemicity of G. duodenalis in the study areas it seems plausible that protective antibodies [66] are transmitted by the breast-milk of seropositive mothers. In addition, lactoferrin and leukocytes in breast milk could be involved in protection of the younger children under breastfeeding [67].
Clinical features of these intestinal parasites are indistinguishable from one another, ranging from asymptomatic, to severe watery diarrhea, fever, weight loss, and possible extraintestinal manifestations. Although the presence of fever, diarrhea and gastrointestinal complaints were more commonly reported among in-pediatric children no statistically significant association were observed between the presence of symptoms and any of intestinal pathogens identified. Thus, the majority of children analyzed seem to be asymptomatic in accordance with reports by others. Nevertheless, asymptomatic carriage may cause severe long-term consequences for the child's health. Beyond that these children play a crucial role in the transmission of theses parasites to other susceptible hosts, working as continuous reservoir for these parasites.
This is an important study because it provides updated data on the occurrence of several intestinal pathogens in two different communities in São Tome Island. To our knowledge, this is the first record on genetic characterization of Cryptosporidium spp., G. duodenalis and E. bieneusi in children from this African country. Data from the present study have suggested a major role of anthroponotic transmission in the epidemiology of cryptosporidiosis and giardiasis. Both anthroponotic and zoonotic transmission were probably involved in microsporidiosis. Further molecular epidemiological studies enrolling domestic and wild animals, water and other potential sources of environmental infection in welldefined endemic locations are required to clarify de epidemiology of these infectious diseases in this tropical country.
While improving sustainable access to safe drinking water and basic sanitation facilities is a key target, the impact of both personal and domestic hygiene behaviours should not be neglected. Educational campaigns (e.g., promote handwashing, lectures on measures of prevention and diseases control and exclusively breastfeeding) among children and their caretakers have been shown to be an effective tool in reducing the risk of infection and reinfection in children improving children's health from these communities [63,[68][69][70][71]. These campaigns should be implemented for routine in this endemic region. Changing the population behavior is a long-term process, since so many times it is linked to socio-cultural habits but with targeted training meaningful positive improvement can be achieved.
Almost half of DRSTP population may be considered vulnerable to infections, namely by parasites, since nearly 44% has between 0 and 14 years old (2012), and is estimated that each year about 13.1% of children under the age of 5 years (2009) is underweight due to malnutrition [11]. Moreover, a large majority do not have access to sustainable drinking water and proper sanitation. In summary, the specific scenario observed in this endemic region highlights the significance of the results obtained in the present study and suggests that it is crucial to strengthen surveillance on intestinal pathogens. The results of this study were reported to the health authorities of S. Tomé e Príncipe in order to implement the appropriate treatment measures among this population and the authors expect that this data have contributed to the improvement of children's health from these communities.
The synergistic effect of policies targeting both improvements of hygiene and health conditions of the population by the competent authorities of DRSTP, will enhance the successful outcomes with future programs for prevention and eradication of intestinal infections, one of the major health risks to children.