Advances on Dientamoeba fragilis Infections

Dientamoeba fragilis is an enteric protozoan parasite that remains neglected, probably due to the misconception that it is uncommon and non-pathogenic. As more information be‐ came available and antimicrobial agents were developed with activity against this para‐ site, it became clear that D. fragilis is responsible of an active infection, associated with symptoms such as abdominal pain and diarrhea. The clinical presentation of dientamoe‐ biasis varies from asymptomatic carriage to symptoms ranging from altered bowel mo‐ tions, abdominal discomfort, nausea and diarrhea with associated eosinophilia reported in up to 50% of paediatric and 10% of adult patients. Moreover, controversy exists over the protective role of the parasite in priming the immune system in a beneficial way such as in selecting beneficial bacteria, keeping potential harmful microbial intruders at bay or producing metabolites beneficial to the host. Thus, a number of ambiguities and obscuri‐ ties surrounding D. fragilis infections exist. Moreover, the means by which this parasite is transmitted has not been fully defined. The diagnostic recognition of this parasite in fecal examinations requires specific processing and expertise; thus, it is possible that many in‐ fections with D. fragilis may go undiagnosed. A number of studies conducted on small numbers of case reports have demonstrated parasite clearance, as well as resolution of clinical symptoms following treatment with various antiparasitic compounds such as pa‐ romomycin, hydroxyquinolines and the 5-nitroimidazoles, including metronidazole and tinidazole. In addition there is very little in vitro susceptibility data available for the or‐ ganism making some current treatment options questionable. This chapter reviews the scientific literature relating to Dientamoeba's life cycle, prevalence, diagnosis and patho‐ genicity.


Introduction
Dientamoeba fragilis is an enteric protozoan parasite that remains obscure and neglected. While many infections remain asymptomatic, it is now generally accepted that D. fragilis is account-able for an active infection, concomitant with abdominal symptoms, nausea, and diarrhea. Moreover, controversy exists over the protective role of the parasite in priming the immune system in a beneficial way such as in selecting beneficial bacteria, keeping potential harmful microbial intruders at bay or producing metabolites beneficial to the host. Furthermore, the parasite's transmission mode remains a mystery. The microscopic identification and diagnoses of D. fragilis in stool requires skill and expertise; consequently, it is likely that many infections may go unidentified. Numerous studies have reported the effectiveness of treatment regimens using compounds such as paromomycin, hydroxyquinolines, and 5-nitroimidazoles, including metronidazole and tinidazole in the parasite eradication and the resolution of clinical symptoms. In addition, there is very little in vitro susceptibility data available for this parasite, making some current treatment options questionable. This chapter reviews the scientific literature relating to Dientamoeba's life cycle, prevalence, diagnosis, and pathogenicity.

Recognition D. fragilis as a pathogen
D. fragilis is a ubiquitous protozoan parasite found in the gastrointestinal tract of humans. Electron microscopy [1] and molecular phylogenetic studies of the SSU rRNA gene [2,3] have recently confirmed the relationship of this parasite to trichomonads (lacking flagella). Although its pathogenic potential is still controversial, Jepps and Dobell in1918 were the first to report its pathogenicity when it was found to be the only agent detected in three patients with gastrointestinal clinical symptoms [5]. Since then, many investigators have shown that patients infected with D. fragilis generally presented with bowel disorders with symptoms such as diarrhea, loose stools, and epigastric abdominal pains [6][7][8][9][10][11][12]. Furthermore, mounting evidence is accumulating reinforcing the pathogenic potential of D. fragilis [13][14][15][16][17][18][19][20]. Lately, irritable bowel syndrome (IBS) has been linked to D. fragilis infections as a possible cause [21,22], further underscoring its role in the causation of disease. A great deal of controversy exists on the mode of transmission of D. fragilis, and while Enterobius vermicularis nematode has been accepted to play a role in its transmission, more recently a report described the discovery of a new cyst stage in its life cycle [23].
Globally, the prevalence rates of D. fragilis infections vary depending on the identification tool used [6,24,25,26]. Using the traditional light microscope, the rates of infections oscillate between 0.4% and 52% [26]. Nevertheless, using indirect immunoflourescent assay, Chan et al. (1996) reported a prevalence rate of 91% [27]. The application of more sensitive identification tools such as PCR and culture has the extra advantage of providing accurate prevalence data [28]. Considered as a pathogen by several researchers, numerous reports have revealed that D. fragilis elimination with parasitic drugs normally relieves the clinical symptoms in the absence of other pathogens. However, there is currently no consensus as to the ideal treatment regimen [20,29,30]. The aim of this chapter is to review the recent developments and advances made on this frequently overlooked parasite and the disease dientamoebiasis.

Biology and life cycle of D. fragilis
Ranging in size from 5 to 15 μm in diameter, D. fragilis is a single-celled pleiomorphic trophozoite containing up to four nuclei [32,20]. A large proportion of D. fragilis trophozoites are typically binucleated with a large, fragmented, central karyosome without peripheral chromatin differentiated clearly in stained fecal smears [32,10]. Banik et al. (2012) have recently extensively described the surface structures and ultrasrtuctural details of D. fragilis populations grown in xenic culture [31]. Using the scanning electron microscope, the group reported the existence of two different trophozoite populations-smooth and ruffled cells. Whether this represents a significant difference biologically or even in terms of the parasite's pathogenicity remains to be elucidated. Using the transmission electron microscope, neither mitochondria nor peroxisomes were reported [33,34]. Nevertheless, a conspicuous organelle detected was the hydrogenosomes. Like many other organisms living in oxygen-deprived or anaerobic environments, these hydrogenosomes most probably represent the site of anaerobic respiration and energy production [35][36][37][38][39]. Different activities such as amoeboid movement, phagocytosis, and bacterial adhesion to trophozoite surfaces were also reported by Banik and others (2012) [31]. Like many other parasitic protozoa such as Trichomonas vaginalis [40,41,33], Giardia [42], and Leishmania [43], virus-like particles (VLPs) have also been reported to be seen in D. fragilis trophozoites. Many groups have reported an association between the presence of VLPs within T. vaginalis and variations in protozoa phenotypes, virulence factors, and disease pathogenesis [44][45][46]. More details of the ultrastructure of D. fragilis are available in a review authored by Banik et al. (2012) [31].
The complete life cycle and the mode of transmission of D. fragilis remain ambiguous and equivocal. The only known stage thus far is the trophozoite (Fig. 2). Dobell (1940) was the first to predict E. vermicularis egg to act as a vector for the transmission of D. fragilis [47]. Recently, Roser et al. (2013) have detected D. fragilis DNA inside E. vermicularis eggs agreeing with the prediction of Dobell in 1940 [48]. While many reports of a higher than anticipated rate of coinfection between D. fragilis and E. vermicularis led researchers to postulate E. vermicularis as the probable vector responsible for its transmission [48,49], other groups have proved no coinfections with D. fragilis and other worms, suggesting fecal-oral transmission as the possible mechanism of transmission of D. fragilis [9,10] [50]. The role of animals and zoonotic transmission of the parasite is still ambiguous despite a recent study reporting pigs and sheep as natural hosts of dientamoebiasis [51]. The reader is invited to read an excellent review on the topic written by Clark et al. (2014) [52].

Epidemiology of dientamoebiasis and its occurrence
Since its hypothetical association with IBS and other bowel disorders, probable pathogenicity, and the existence of gaps in its life cycle and mode of transmission, many investigators have become increasingly aware of the importance of D. fragilis. This has led to the development of more sensitive diagnostic techniques for its proper identification and determination of its accurate prevalence. It is now recognized as being more prevalent than Giardia [7,11,25,27,29,[53][54][55][56][57][58][59][60][61][62][63][64][65]. Table 1 shows the prevalence rates of D. fragilis ranging between 0.3% and 90% in many countries worldwide. With the exception of few studies, light microscopy was the tool used in those studies. The use of more sensitive techniques such as PCR or cultivation may result in different and more accurate prevalence rates [10,28,66]. Unlike many parasitic infections, D. fragilis has been shown to have high infection rates in developed countries than in underprivileged countries [25,67].  [24]. In contrast, other investigators reported a higher incidence rate in children and in less than 20 years old [8,16,27,29,63,74,95,100]. In a recent study by Al-Hindi and Abu Shammala (2013) in the Gaza strip regarding age, children less than 5 years of age were reported to have a prevalence of 11.3%, while the age-group 20-26 years had 15.4% [4]. This is in contrast to findings by Girginkardesler et al. (2003), who reported that D. fragilis infection was higher among children than adult [29]. No plausible explanation to these variations in age distribution of D. fragilis incidences is proposed. Nevertheless, hygiene and modest sanitation have been suggested as likely to prejudice groups to infections with D. fragilis and other intestinal protozoa irrespective of age making the fecal oral route as the probable route of transmission [12,24,53,101]. With respect to the association between gender and D. fragilis infections, numerous studies report dissimilar trends. While several investigators report more infection incidences in females than males [16,24,55,57,86,95], other studies describe a drift towards males in certain age-groups [4,8,60]. For more details on the subject, the reader is advised to consult the review by Barratt et al. (2011) [12].

Pathogenicity and clinical symptoms of dientamoebiasis
Originally proposed as a pathogen in 1936 by Hakansson, there still remains some reluctance by many investigators accepting D. fragilis as a pathogen [102,103] [55,57]. Another report corroborating the pathogenic potential of the organism is an Australian study in which 5.4% (35/650) of patients with bowel disorders were reported to have Dientamoeba in their stools, with 83.3% of them suffering from diarrhea [10]. Furthermore, Dientamoeba has been linked it with irritable bowel syndrome (IBS) [22,97,105]. Patients carrying Dientamoeba may also experience eosinophilia [10,63,64,103,106,107].

Treatment of D. fragilis infections
While still not recognized as a pathogen, the ability to resolve associated symptoms by eradicating D. fragilis using different drugs provides some proof for its possible pathogenic nature [6,16,20,69,102,103,[108][109][110]. There is still no agreement as to the best regimen for the complete elimination of the organism. Stark et al. (2010b) and Preiss et al. (1990) reported a treatment ineffectiveness and/or relapse of dientamoebiasis following the use of metronidazole only [11,64]. In a recent Danish randomized trial, 96 children in Denmark with D. fragilis infection and chronic gastrointestinal symptoms were treated with a 10-day course of metronidazole or placebo [111]. Change in gastrointestinal symptoms following treatment did not differ significantly between the groups. Eradication of D. fragilis was significantly greater in the metronidazole group as assessed by PCR 2 weeks after completion of therapy, although PCR positivity rebounded by 8 weeks after completion of therapy to levels comparable with those seen in placebo recipients. The eradication of D. fragilis was significantly greater in the metronidazole group, although it declined rapidly from 62.5% 2 weeks after end of treatment to 24.9% 8 weeks after end of treatment. The findings of the study did not provide evidence to support routine metronidazole treatment of D. fragilis-positive children with chronic gastrointestinal symptoms. However, the complete resolution of symptoms and elimination of the organism were noted following the administration of either iodoquinol, paromomycin, or a combination of the two [11,107]. Most recently, Halkjaer et al. (2015) described a case history of a 16-year-old Danish patient who had suffered severe abdominal discomfort and flatulence through his lifetime following infection with D. fragilis. The patient was treated initially with a high dose of metronidazole, which eradicated the parasite and kept him without symptoms for 1 year [112]. However, recurrence of the symptoms and recurrence of the D. fragilis infection were thereafter treated successfully with paromomycin [112]. Other drugs that are also reported to effectively eradicate the parasite leading to clinical cure included oxytetracycline, doxycycline, tinidazole, secnidazole, ornidazole, and erythromycin [29,30,64,102,113]. Despite the lack of randomized controlled trial data, the literature suggests paromomycin is a more efficacious agent than metronidazole [6,11,114]. New potential therapeutic compounds are constantly being screened for by investigators. More recently,  have shown that there is no therapeutic response against dientamoebiasis with benzimidazoles (such as albendazole and mebendazole) [115].

Role of genetic characteristics of the infecting strains in the pathogenesis of dientamoebiasis
The outcome of an infection may depend on several factors, among which the genetic characteristics of the specific pathogen have been identified as an important one. The virulence and disease outcome has been linked to the genotypes of few parasites such as Entamoeba histolytica and Giardia lamblia [116][117][118][119][120][121][122][123][124]. Despite its inability to ascertain correlation between genotype and disease outcome, evidence emerged using the ssu rRNA gene of at least two genetically distinct variants (genotypes 1 and 2) of D. fragilis are in existence [6,9,20,125,126]. Thus, in the case of D. fragilis infections, the ssu rRNA gene demonstrated inferiority as a tool for molecular epidemiological studies [127]. Accordingly, new molecular tools were employed to demonstrate the association between variants and clinical disease outcome. One such tool is the use of C-profiling in which the cysteine nucleotide pattern is compared between samples for evidence of genetic variation on the internal transcribed spacer regions (ITS1 and ITS2) of the ribosomal gene [128]. These regions are noncoding sequences reported to be suitable for molecular characterization of phylogenetically related organisms [129]. Bart et al. (2008) and Stark and his group (2009) documented the occurrence in variations of ITS1 in D. fragilis isolates [21,130]. Furthermore, a correlation between certain ITS1 variants and disease outcome was reported [130]. Recently, Barratt et al. (2010) found that the growth of Dientamoeba in certain media formulations varied between different isolates, and while all Dientamoeba isolates described by Barratt and colleagues were from patients with gastrointestinal complaints, this work indicates that phenotypic diversity exists in Dientamoeba and that the variation noted is likely to have a genetic basis. Nevertheless, it is still unclear whether the two genotypes differ in their pathogenicity [131].

Diagnosis of dientamoebiasis
While it is difficult to identify the trophozoites of D. fragilis morphologically, the only diagnostic tool used to detect D. fraglis is microscopy using permanent stained smears. A large variety of stains have been used for the microscopic examination of E. fragilis. However, the most commonly used that also give much clearer images of the parasites are the trichrome and the iron-hematoxylin stains. The sample should be fixed immediately after staining to avoid degeneration of the trophozoites and staining should also occur sooner [107]. Trophozoites range in size from 5 to 15 μm in length, from 9 to 12 μm in width, normally with 1-2 fragmented nuclei with visible holes seen through the nucleus center. Smears may also contain trophozoites with the typical four nucleated form. No cyst stage has been recovered yet from humans despite being observed in mice [23]. Even under ideal conditions, with prompt preservation of stool and evaluation of permanent stained smears by experienced microscopists, Stark et al.
(2010a and 2011) reported a sensitivity of 34% and 38%, respectively, compared to PCR (realtime and multiplex tandem-PCR) [10,132]. Despite numerous studies reporting common occurrence of D. fragilis infection, no clinical antigen-based, molecular, or serologic diagnostics have been commercially developed to aid with laboratory identification to date, although current molecular based methods are used for research [133]. The culture of D. fragilis has been reported and is done in similar conditions as that of E. histolytica. Liquid or diphasic media is used that can be in xenic or axenic conditions [10]. Another diphasic medium based on the Loeffler's slope has also been demonstrated, and Earle's balanced salt solution (EBSS) has been successfully used for the growth of D. fragilis [23].
Molecular diagnostic methods have been very instrumental for the improvement of our understanding of different infections. There has been a significant gain in the development of molecular methods for the detection of D. fragilis, although compared to other organisms, this improvement has been much slower [32]. Several PCR protocols have been developed for the detection of this organism mainly for research laboratories. These protocols vary from conventional PCR to real-time PCR with increased sensitivity and specificity. Primers based on the small ribosomal RNA gene have been developed for this purpose [9]. Verweij and colleagues have developed a real-time PCR protocol using the 5.8S ribosomal RNA gene and they showed that this method was both specific and sensitive [28]. A variation of PCR based on the amplification of the internal transcribed spacer 1 region of D. fragilis has also been used for the molecular characterization of the parasite [130]. The actin gene has also been used as a target for the molecular characterization of this parasite [128]. Generally, the detection and/or the molecular characterization of the parasite begin with DNA purification, which is a very important and critical step in the amplification of the organism. Following DNA purification, the PCR master mixed is prepared depending on the procedure to be used. In the case of detection, the PCR protocol is generally sufficient. However, the molecular characterization often requires a sequencing step with or without the purification of the PCR amplicons. Other methods that have been used so far for the molecular characterization of D. fragilis include high-resolution melt curve analysis (HRM) and restriction fragment length polymorphism after amplification by PCR [9,22].
Using HRM, Hussein and colleagues found 4 genetic profiles of which the first and most common profile and the last profile (Profile 4) were more associated with diarrhea compared to the two middle profiles [22]. However, the ITS showed two major genotypes although there were subgenotypes among those main categories. In another study, the ITS-1-5.8S rRNA gene-ITS-2 region of D. fragilis was found to be highly variable and pyrosequencing method identified 11 different alleles of the ITS-1 sequence showing the limitation of this gene in the molecular characterization of the parasite [130]. Briefly, the use of molecular methods has increased our knowledge on these organisms; much still remains to be discovered for the better understanding of issues related to pathogenicity, diagnosis, and prognosis.

Conclusion
Known for almost a hundred years now, D. fragilis still remains a mysterious organism although much has been learned. The use of molecular biology has clarified its classification not as an amoeba but as a trichomonad. However, its pathogenicity as well as its genetic diversity still remains to be clarified. Diagnosis particularly in developing areas of the world where it could be common remains difficult because microscopy is not sensitive. This is made to be even more difficult because of the uncertainty of the existence of a cyst stage, which so far has only been demonstrated in very limited studies. Real-time PCR has been proven to be more sensitive compared to all the other diagnostic methods, including conventional PCR, microscopy, and culture. Further studies are needed, and collaboration between developing and developed countries will help boost the research capacity on this infection and improve our understanding of its distribution, pathogenicity, and immunology.