A Cross-Sectional Study of Entamoeba histolytica/dispar/moshkovskii Complex in Salvador, Bahia, Brazil

Epidemiological studies on species-specific Entamoeba infections are scarce due to the morphological similarity of pathogenic Entamoeba histolytica and nonpathogenic E. dispar and E. moshkovskii. The diagnosis of E. histolytica is frequently based on coproantigen (E. histolytica-Gal/GalNAc lectin specific) detection by immunoassays. However, specific E. histolytica-lectin is not expressed in cysts, which are eliminated by asymptomatic individuals leading to false-negative results and an underestimation of amebiasis prevalence. Molecular techniques based on the amplification of parasite DNA have been shown to be a highly sensitive and specific method that allows the detection of different Entamoeba species. This study aimed to assess the frequency of the species from E. histolytica/dispar/moshkovskii complex by molecular and immunological techniques in individuals attended at a public health system in Salvador-Bahia, Brazil. A cross-sectional study involving 55,218 individuals was carried out. The diagnosis was based on microscopy revealing E. histolytica/dispar/moshkovskii complex. The species differentiation was performed by E. histolytica-specific antigen, serological evaluation and by molecular technique. The overall prevalence of E. histolytica/dispar/moshkovskii complex determined by microscopy was approximately 0.49% (273/55,218). E. histolytica-specific antigen detection and molecular characterization returned 100% negativity for E. histolytica. However, serological evaluation returned an 8.9% positivity (8/90). In the stool samples analysed by PCR, it was not possible to identify E. histolytica and E. moshkovskii, although circulating IgG anti-E. histolytica has been detected.


Introduction
Entamoeba histolytica is a common pathogenic protozoan widely distributed throughout the world. It is responsible for amoebic dysentery and amoebic liver abscess, resulting in human suffering and death. Amebiasis is significantly associated with food and drinking water supplies contaminated with human faeces and affects approximately 10% of the world's population, with 50,000-100,000 deaths reported annually, making this a significant cause of death due to protozoan parasites [1]. The E. histolytica prevalence is overestimated due to its epidemiological overlap with other morphologically identical species, i.e., Entamoeba dispar and Entamoeba moshkovskii, currently composing the E. histolytica/E. dispar/E. moshkovskii complex [2][3][4][5]. Besides, the cysts of Entamoeba hartmanni, another nonpathogenic amoeba, even smaller in size (> 10 m), can be confused with E. histolytica cysts in the parasitological examination.
The prevalence of each species from this complex is not well characterized in many geographic regions. It is wellestablished that only E. histolytica leads to invasive disease in humans. Although some reports suggest a potential role of both E. dispar and E. moshkovskii in provoking disease in humans, they are still considered to be nonpathogenic and free-living amoebas, respectively [6][7][8][9][10]. Nevertheless, the differentiation of all Entamoeba species is substantial 2 BioMed Research International for preventing unnecessary and indiscriminate treatment with anti-amoebic chemotherapy, which could lead to drug resistance [11]. The World Health Organization advises that cases determined to have E. histolytica should be treated, whether or not clinical symptoms are present [12]. Additionally, control measures could be more efficiently applied in geographical areas with well-known epidemiological settings.
Regarding the low viability of trophozoites in diarrheal specimens and irregular faecal cyst excretion in asymptomatic hosts, the use of different diagnostic methods is required to increase the sensitivity of parasite identification in faecal samples. Immunoassays for coproantigen detection of specific E. histolytica-Gal/GalNAc lectin have been used as alternative methods for the diagnosis of amebiasis [13]. Molecular techniques based on the amplification of parasite DNA have been shown to be a highly sensitive and specific method that allows the detection of different Entamoeba species [14]. However, a negative result does not rule out the presence of the parasite due to interference from PCR inhibitors present in faeces that may hamper DNA amplification.
In Brazil, due to regional differences in sanitation and socioeconomic conditions, E. histolytica/dispar/moshkovskii complex distribution is irregular, with 2.5-11% in the South and Southeast, 19% in the North and the Amazon region, and approximately 10% in the Northeast and Central West [15]. In the city of Salvador, the capital of the Brazilian state of Bahia, the presence of the complex was shown in some studies conducted by our group with a prevalence of 5% and 3.2% in individuals attended at public [16] and the private health system respectively. Moreover, 15% (262/1,788) of positive stool samples for E. histolytica/E. dispar complex was analysed by PCR and no DNA amplified for E. histolytica. DNA amplified for E. dispar was observed in 86.6% of samples (227/262) [17]. However, in a previous study also conducted by our group, 4.6% (7/153) of hospitalized diarrheal children from Salvador tested positive for specific lectin of E. histolytica-Gal/GalNAc [18].
These findings further support evidence that E. histolytica may be infecting humans in Salvador. Given the importance of the actual prevalence of E. histolytica, E. dispar, and E. moshkovskii infections, this study aimed to assess the frequency of each species by molecular and immunological techniques in individuals attended at the public health system in Salvador-Bahia, Brazil.

Ethical Considerations.
Approval was granted by the Institutional Review Board (IRB) for Human Research at the Gonçalo Moniz Institute, Oswaldo Cruz Foundation (FIOCRUZ), Salvador, Bahia-Brazil, registration number 100/2006. All procedures were conducted in strict adherence to the principles laid out in the Declaration of Helsinki. Informed consent was obtained from participants who agreed to participate of the study. Samples were anonymously coded to protect each participant's identity.

Study Design and Population.
A cross-sectional study was performed from February 2010 to June 2014 involving 55,218 individuals who attended to the Clinical Laboratory of the Pharmacy College (LACTFAR; Federal University of Bahia). This population is mainly characterized by lowincome individuals who are dependent upon the public health system. Stool examination was performed by laboratory staff. A flowchart illustrating the study design is provided in Figure 1.

Sample Selection.
Two hundred and seventy-three patients diagnosed with amoeba cysts in their stool were contacted and invited to participate in the second phase of the study. A single fresh faecal specimen and 5 ml sample of blood for sera collection were obtained from 90 individuals who agreed to participate in the study. Coproantigen and IgG-specific antibodies detection and DNA amplification by PCR were used to diagnose the species of the complex.

Microscopic Examination.
Approximately 3 g of faecal material was processed using the formalin-ethyl-acetate centrifugation method. Morphological analysis was conducted to detect the presence of tetranucleated cysts to confirm the diagnosis previously established by the spontaneous sedimentation technique. A 50 l portion of the concentrated sample was mixed with iodine, spotted on a glass slide and covered with a coverslip (24 × 24 mm). Slides were viewed at 40X magnification, and results were expressed as a number of cysts per gram of faeces.

Parasite Cell
Culturing. Entamoeba trophozoites were obtained under xenic conditions in modified Pavlova's medium with bacterial flora, for coproantigen testing. Briefly, 1 gram of fresh stool was washed five times with distilled water, and 500 l of sediment was incubated at 37 ∘ C under 5% CO 2 in 10 ml of Pavlova's medium in round-bottom threaded culture glass tubes (15 × 1.5 cm). Amoebas were maintained with thrice-weekly subcultures to ensure viability through the exponential growth phase. The transformation of cysts into trophozoites, as well as trophozoite development and viability, was monitored daily for five days.

E. histolytica-Specific Antigen Detection.
Positive stool samples were preserved without fixative and stored at −20 ∘ C for posterior coproantigen testing (galactose adhesin) which was performed using the E. histolytica II assay (TechLab, Blacksburg, VA, USA). ELISA testing was carried out in accordance with manufacturer's directions. Absorbance at 450 nm was measured using a Bio-Rad Model 3550-UV Microplate Reader (Bio-Rad, CA, USA).

Serological Evaluation. The RIDASCREEN5
Entamoeba test (R-Biopharm AG, Darmstadt, Germany) was used to detect specific anti-Entamoeba histolytica IgG in human sera. The method utilizes microtiter plates coated with purified antigens. Briefly, serum samples were loaded at 1:50 in sample diluent and incubated at RT for 15 min. Following incubation, the microplates were washed with washing buffer to remove   , and 2.5 l of DNA sample. An initial DNA amplification step was performed using the E-1/E-2 primer set in a MyCycler6 Thermal Cycler (Bio-Rad, CA, USA). The first cycle consisting of 2 min at 96 ∘ C was followed by 30 cycles of denaturation for 1 min at 92 ∘ C. Primers were annealed for 1 min at 56 ∘ C and extended for 1.5 min at 72 ∘ C. An additional extension step was performed at 72 ∘ C for 7 min. For nested amplification, 2.5 l of amplicon from the first reaction and primer sets EH-1/EH-2, ED-1/ED-2, and Mos-1/Mos-2 were used under identical conditions as those described above, with the exception of annealing temperature of 48 ∘ C. PCR products of tested samples and internal controls (DNA of E. histolytica and E. dispar) were analysed by gel electrophoresis. DNA fragments were separated on a 1.0% (w/v) agarose gel (Invitrogen Life Technologies, CA, USA) containing 0.5 g ethidium bromide/ml. Gels were photographed under ultraviolet illumination (Loccus Biotecnologia, SP, Brazil).

Data
Analysis. Data were analysed using scatter plot graphing software (GraphPad Prism v.7, CA, USA). Descriptive statistics are presented as means or medians ± standard deviation, or percentages, to describe the characteristics of the studied population, including the prevalence of E. histolytica, E. dispar, and E. moshkovskii. The Shapiro-Wilk test was employed to assess data normality, and Levene's test was used to evaluate the homogeneity of variance. When these two assumptions were confirmed, ANOVA was used for sample comparisons; otherwise, Fisher's exact test was employed. All analyses were two-tailed and a p-value less than 5% was considered significant (p < 0.05).

Analysis of Stool Samples.
The routine parasitological analysis of 55,218 stool samples by microscopic examination identified E. histolytica-like cysts in 273 (∼0.49%). The prevalence was significantly higher in adults (≥ 18 years). Similarly, there was a significant difference in the prevalence of infection between male and female subjects (Table 1). From the 273 positive individuals for the E. histolytica/dispar/moshkovskii complex, 90 (33%) agreed to participate in the survey (Figure 1).

E. histolytica-Specific Antigen
Detection. Antigen detection by the E. histolytica II assay was employed to identify positive specimens for E. histolytica. All 90-stool samples tested were negative for E. histolytica, which was defined as an optical density value lower than 0.05 after subtraction of negative control value Additionally, it was verified if the negative antigen detection was due to the absence of trophozoites in the well-formed stool. Thirty-three samples were randomly selected and added to Pavlova's medium for growth of Entamoeba species. Incubation led to the growth of E. histolytica/dispar/moshkovskii in 28 (84.9%) cultures. Following the antigen detection of E. histolytica in the trophozoites transformed in culture, all specimens presented negative results.    [16,17]. This decrease could be explained in part by the positive impact of government programs, which have improved sanitation infrastructure, thus reducing the occurrence of intestinal parasites among residents of the Salvador Metropolitan Area [20]. Furthermore, the recent expansion of health education and health care services also may have contributed to the early identification of E. histolytica new cases and prompt treatment. Additionally, it appears that E. histolytica is more common in North and rare in other regions, including Salvador [17,21,22].

Serological Evaluation
The PCR technique presented a sensitivity of 80% and specificity of 100%. Our data are in agreement with our previous report [17], in which we found a sensitivity value of 86.6%. In fact, 18 DNA samples did not amplify even after successive dilutions. It suggests either nondiagnosed E. hartmanni infections, which could be excluded by morphometric analysis and/or PCR [17,23], or nonspecific substances present in the faecal sediment that inhibited DNA amplification [24]. Previous data from our group conducted in Salvador-Bahia showed that non-amplified DNA samples were not associated with E. hartmanni [17]. E. histolytica coproantigen detection revealed negative results for all assayed samples, indicating no E. histolytica heterodimer galactose/N-acetyl-galactosamine (Gal/GalNac) lectin in faecal specimens. Significant variability in performance for E. histolytica antigen detection assays in both nonendemic [25,26] and endemic areas [27,28] has been reported. Conversely, a serological assay produced eight positive cases (8.9%), suggesting the previous exposure to the parasite. All the eight serologically positive samples rendered negative results for E. histolytica by PCR but positive for E. dispar. Antibodies remain detectable for years after successful treatment of E. histolytica, so it is difficult to distinguish between active and past infection reliably. Moreover, E. histolytica infection confirmed by lectin antigen can occur without antibody detection since the antigens search is more sensitive for E. histolytica diagnosis [29].
It is important to note that E. dispar has the same transmission path as other pathogenic protozoa, such as E. histolytica, indicating exposure to faecal contamination. Possibly, the production of IgG anti-E. histolytica is associated with immunological memory indicating the parasite circulation in Salvador, Bahia Brazil, in accordance with the previous data of our group [18]. Although some reports suggest a potential role of E. dispar in provoking intestinal and extra-intestinal symptoms in humans, its pathogenicity remains unclear [30]. The prevalence of E. dispar is 10 times higher than that of E. histolytica worldwide, and the attending physician must decide if treatment is necessary based more on clinical evidence [1].

Conclusions
In this study, the prevalence rate of E. histolytica/dispar/ moshkovskii complex based on faecal examination by optical microscopy was around 0.49%. In the analysed samples by coproantigen and PCR, it was not possible to prove the presence of E. histolytica and E. moshkovskii. Only E. dispar was diagnosed by PCR, although the presence of circulating IgG anti-E. histolytica has been detected.

Data Availability
The data supporting the conclusions of this article are provided within the article. The datasets generated and analysed during the current study are available from the corresponding author upon reasonable request.

Conflicts of Interest
The authors declare that they have no conflicts of interest.