Identification of Entamoeba histolytica and E . dispar infection in Maceió , Alagoas State , northeast Brazil

Introduction: A cross-sectional study was carried out to assess the prevalence of E. histolytica and E. dispar by examining stool samples obtained from 1,003 students of public schools in Maceió, Alagoas, Brazil. Methodology: All stool samples were processed using the spontaneous sedimentation technique and examined microscopically for the presence of Entamoeba species. In order to distinguish infections caused by E. histolytica, fecal samples presenting cysts of Entamoeba were subjected to specific enzyme-linked immunosorbent assay (ELISA) and polymerase chain reaction (PCR). Results: The analysis of the fecal specimens by microscopy identified 6.4% (64/1,003) students positive for E. histolytica/E. dispar/E. moshkovskii cysts. The prevalence of E. histolytica detected by ELISA was 3.0% (30/1,003) and by PCR 2.8% (28/1,003), but the difference is not statistically significant (p > 0.05). The prevalence of E. dispar in schoolchildren was 5.0% (50/1,003). Mixed infections with E. histolytica and E. dispar were also detected by PCR. Even though immunological and molecular methods have shown similar results for identification of E. histolytica, ELISA is advantageous over the PCR since it is relatively cheaper and easier to perform. Conclusions: Our study demonstrated the occurrence of E. histolytica in Maceió and highlights the need to introduce a specific diagnostic test to detect amoebiasis cases in public laboratories.


Introduction
Amoebiasis is a severe infection caused by Entamoeba histolytica, and it is considered an important cause of morbidity and mortality in the world [1,2].Higher prevalences are found in tropical areas of South and Central America, Africa, the Asian subcontinent, and Pacific countries, mainly because of unsanitary living conditions.The similarities between these regions include low socioeconomic development and poor environmental sanitation, both of which favor parasite transmission [2][3][4].E. dispar and E. moshkovskii are considered commensal non-pathogenic species of amoeba, morphologically indistinguishable from E. histolytica, the causative agent of invasive amoebiasis.A differential diagnosis between these species is critical because, according to the World Health Organization, treatment should only be administered to patients who have been diagnosed as having E. histolytica [1].The most commonly used diagnostic methods to specifically separate these species are enzyme-linked immunosorbent assay (ELISA) for the detection of E. histolytica antigen, and polymerase chain reaction (PCR) to amplify amoebic DNA [1].
In Brazil, the frequency of Entamoeba infection displays regional variations.The highest levels of amoebiasis prevalence rates are observed in the northern region of the country.In Belém and Manaus municipalities, both in the Amazon region, the prevalence of E. histolytica infections recorded were 29% and 6.8%, respectively [5][6][7].In the northeast region, the prevalence for amoebiasis in Fortaleza (Ceará State) is 14.9% [3], whereas in Pernambuco and Bahia States, no case of infection due to E. histolytica has been reported in the literature [8][9][10][11].
The present study aimed to assess the specific occurrence of infection by E. histolytica and E. dispar in Maceió, Alagoas State in northeast Brazil.

Design of the study and sample calculation
A cross-sectional epidemiological study was conducted in the urban area of Maceió, state of Alagoas, Brazil (Figure 1), from March 2008 to December 2010.The reference population comprises 67,738 students 4-15 years of age selected from the four education coordination units (CES) representing all geographical areas of the city.This age group is considered to be at increased risk for Entamoeba infection [2].In each CES, public schools were stratified according to the overall average of students (645 students/school), resulting in two groups of schools, one with more and other with fewer than this average number of students.This criterion was adopted because of the possibility that students from larger schools have characteristics from those who attend smaller schools.Sample size calculation using an expected frequency of 11% of Entamoeba infection [12] with an acceptable error of 20% and a confidence level of 95% returned a minimum sample size of 768 subjects.The study protocol was approved by the ethics committee of the Universidade Federal de Alagoas (reference number 011577/2005-19).

Microscopic examination
Fresh unpreserved stool samples were collected and processed using the spontaneous sedimentation technique [13].Four slides of each sample stained with Lugol's iodine solution were prepared and examined microscopically by four different technicians.Samples positive for E. histolytica/E.dispar/E.moshkoviskii cysts were further characterized for distinction of the species using immunological and molecular techniques.

Antigen detection
Microscopically positive fecal specimens for Entamoeba were tested by ELISA for E. histolytica lectin-specific antigen.The commercially available kit E. histolytica II (Techlab, Blacksburg, USA) was used according to the manufacturer's instructions.This kit uses monoclonal antibodies against the Gal/GalNAcspecific lectin (adhesin molecule) of E. histolytica.A sample was considered positive if the optical density (OD), measured at 450 nm, was ≥ 0.05 after subtracting the negative control.

DNA amplification
The same fecal specimen sediments used for antigen identification were used for DNA extraction.DNA was extracted from 200-400 mg of each stool sample with the PSP Spin Stool DNA Kit (Invitek, Hayward, USA), following the manufacturer's instructions.
The amplification was done as described Rivera et al. [15], with minor modifications.The PCR reagents comprised 20 mM Tris-HCl, 50 mM KCl, 200 M dNTP, 1.5mM MgCl2, 20 pM of the specific pair of primers (E.histolytica or E. dispar), 2U Taq Polymerase Platinum (Invitrogen, Carlsbad, USA), 400 ng of bovine serum albumin, 2 L of DNA and ultrapure H2O to a final volume of 25 L.Amplification conditions consisted of 35 cycles of 60 seconds at 94°C, 60 seconds at 59°C, and 60 seconds at 72°C (PCR Express Thermo Hybaid, Ashford UK).PCR products were subjected to electrophoresis in 2% agarose, staining with ethidium bromide (10 mg/mL) and visualization in a UV transilluminator (Vilber Lourmat TFX-20.M, Eberhardzell, Germany).The speciesspecific product size for E. histolytica and E. dispar was 100 and 101 bp, respectively.
DNA extracted from E. histolytica and E. dispar cultures were used as positive controls in each PCR run.DNA extraction of cultured parasites was done as described by Sambrook et al. [16].

Statistical analysis
The data was analyzed in Epi-Info version 3.5.1 computer software (Centers for Disease Control and Prevention).Chi-Square ( 2 ) and student t test were used to compare proportions.Statistically significant difference was defined as p < 0.05.Risks were estimated by odds ratio (OR).

Results
Stool samples were collected from 1,003 schoolchildren (54% males and 46% females) with a mean age of 9.7 ± 2.6 years, attending any of the 16 chosen public schools located in different districts of the city.Students from 40 out of the 50 districts of Maceió were selected.
A total of 64 (6.4%) of all 1,003 examined fecal specimens were identified as having Entamoeba cysts, either singly or in combination with other intestinal parasites.E. histolytica coproantigen by ELISA was detected in 30 (3.0%) of these microscopically positive samples.The average age of infected children was 10 ± 6.8 years, and no difference was observed in relation to the infection by age group (p > 0.05) (Table 1).
The PCR detected mono-infection with E. histolytica in 21.9% (14/64) and with E. dispar in 56.3% (36/64) of the stool specimens microscopically found with Entamoeba cysts.Mixed infection with both E. histolytica and E. species were detected in 21.9% (14/64) of stool samples identified as positive for Entamoeba spp.
Table 2 summarizes the ELISA and PCR results of the 64 fecal specimens microscopically positive for Entamoeba infection.Though two samples positive by ELISA showed negative results by PCR (Table 2), there was no difference in the positivity between the two techniques (p > 0.05).When ELISA was used as a reference standard, the PCR sensitivity and specificity were 93.3% and 100%, respectively, with a Kappa correlation index of 0.96.
Stool samples from 48 students presenting no parasite cysts by microscopy were tested for E. histolytica and E. dispar by PCR, with negative results.Considering that microscopy screened all negative samples for E. histolytica/E.dispar cysts, and regarding mono-and co-infection cases, the overall prevalence for E. histolytica was 2.8% (28/1,003), and 5.0% (50/1,003) for E. dispar.The difference between the proportions of each species was statistically significant (p < 0.05).Thus, in the studied area, the non-pathogenic parasite E. dispar was 1.8 times more common than E. histolytica infection (OR: 1.8 [95% CI: 1.14-20.93],p < 0.05;  2 : 6.46).

Discussion
The present results are consistent with epidemiological studies carried out in different regions of Brazil and the world, in which the two species of Entamoeba were found infecting the population [1,5,6,17].In Brazil, there are only few locations where no cases of amoebiasis been found, such as Pernambuco and Bahia states [8][9][10][11].
In general, there is a higher number of E. dispar cases in relation to the E. histolytica infections, as reported in other studies, including in Brazil, as well as the occurrence of mixed infections with E. histolytica and E. dispar [18][19][20][21][22].
Antigen detection by ELISA and DNA amplification through PCR showed similar results and good correlation for specific amoebiasis diagnosis, as observed by other authors [23].This agreement between ELISA and PCR indicates that either one is suitable to the specific detection of E. histolytica in a stool specimen.The choice of the method will depend on the budget and the objective of the study [23][24][25].The two samples found to be ELISA positive but PCR negative may be due the presence of DNA amplification inhibitors in the fecal samples.Bacterial proteases, nucleases, cell debris, and other factors may interfere in the PCR, preventing DNA amplification [26].Although no specific test was performed to detect E. moshkovskii, most infections caused by this amoeba worldwide are commonly reported as a mixed infection [27,28].In Brazil, there is a lack of data concerning the prevalence of this Entamoeba species.

Conclusions
In Brazil, particularly in the northeast, there are few surveys indicating the occurrence of amoebiasis cases.The present study contributes to the knowledge of the epidemiological situation of this morbidity in the country.The results indicate the need for specific diagnostic techniques to detect E. histolytica in the evaluated population.

Figure 1 .
Figure 1.Map showing the survey location.

Table 1 .
Distribution based on age group of microscopy-positive Entamoeba histolytica/E.dispar stool samples among school children from Maceió, Alagoas State, northeast Brazil.
a Detection by microscopy of E. histolytica/E.dispar/E.moshkoviskii complex cysts.

Table 2 .
Comparison of results obtained by enzyme-linked immunosorbent assay (ELISA) and polymerase chain reaction (PCR) performed on 64 stool specimens microscopically positive for Entamoeba histolytica/E.dispar complex.
a ELISA test detected E. histolytica-specific antigen in stool specimen; b PCR amplified specific DNA fragment of E. histolytica.