First Identification and Genotyping of Enterocytozoon bieneusi and Prevalence of Encephalitozoon intestinalis in Patients with Acute Diarrhea in the Republic of Korea

Encephalitozoon intestinalis and Enterocytozoon bieneusi can cause diarrhea in humans, especially severe diarrhea in immunocompromised patients. However, there have been few studies on Enc. intestinalis and Ent. bieneusi in patients with acute diarrhea in the Republic of Korea (ROK). In this study, fecal samples were collected from 1241 patients with acute diarrhea in 2020. Among these, 24 cases of Enc. intestinalis and one case of Ent. bieneusi were detected via PCR amplification of small subunit ribosomal RNA. Genotyping of the internal transcribed spacer region sequence revealed that the detected Ent. bieneusi genotype was in Group 1. This study provides the first evidence that Ent. bieneusi exists in humans in addition to animals in the ROK. To identify the causative agent, continuous monitoring of Enc. intestinalis and Ent. bieneusi is necessary for patients with acute diarrhea in the ROK.


Introduction
Microsporidia, which include more than 170 genera and 1300 species, are opportunistic protozoan pathogens that infect a variety of vertebrate and invertebrate hosts [1]. Enterocytozoon, Pleistophora, Encephalitozoon, Vittaforma, Trachipleistophora, Brachiola, Nosema, and Microsporidium have been reported to infect humans. Encephalitozoon intestinalis and Enterocytozoon bieneusi are known to be the most common cause of intestinal diseases [2,3]. Microsporidia are present everywhere and spread through various routes [4]. Microsporidia that infect humans have been identified in animals and water sources [5], and species infecting humans also infect animals, raising the concern for zoonotic transmission [6]. In particular, a direct evidence for zoonotic transmission of microsporidiosis has been reported in a child after close contact with Enc. cuniculi-infected pups [7]. In addition, spores present in stools of infected individuals can be transmitted horizontally through fecal-oral transmission or ingestion of contaminated food and water [6,8].

Sequence Analysis of Enc. intestinalis and Ent. bieneusi Using Small Subunit Ribosomal RNA
Enc. intestinalis and Ent. bieneusi DNA fragments were sequenced using both the forward and reverse primers Mic C and Mic D. All 24 cases of Enc. intestinalis (KDCA  showed 100% similarity with the small subunit ribosomal RNA (SSU rRNA) gene and were clustered with previously reported isolates (Accession numbers: KM058742, DQ453122, and JF932507), based on their molecular phylogenies. The SSU rRNA sequence of the single Ent. bieneusi case (KDCA 25) was also clustered with previously reported isolates (Accession numbers: MG976584, MH027470, and KJ019869) ( Figure 1).
Enc. intestinalis and Ent. bieneusi DNA fragments were sequenced using both the forward and reverse primers Mic C and Mic D. All 24 cases of Enc. intestinalis  showed 100% similarity with the small subunit ribosomal RNA (SSU rRNA) gene and were clustered with previously reported isolates (Accession numbers: KM058742, DQ453122, and JF932507), based on their molecular phylogenies. The SSU rRNA sequence of the single Ent. bieneusi case (KDCA 25) was also clustered with previously reported isolates (Accession numbers: MG976584, MH027470, and KJ019869) ( Figure 1).

Figure 1.
Phylogenetic tree of Encephalitozoon spp. and Enterocytozoon bieneusi isolates and selected accessions from Gen-Bank, based on small subunit ribosomal RNA gene fragment sequences. The phylogenetic tree was constructed using nucleotide sequence alignments, with the Kimura two-parameter algorithm as the distance method and neighbor-joining as the tree composition method. The black squares indicate the known genotypes identified in this study. The sequence of Nosema bombycis was used as the outgroup.

Genotype of Ent. bieneusi
Based on the sequencing analysis of the internal transcriptional spacer (ITS) gene, we detected one Ent. bieneusi genotype, Korea-WL2. The KDCA25 Ent. bieneusi isolate obtained in the present study was identical to LC436502 from a raccoon dog and LC436503 from a Korean water deer in the ROK [20]. KDCA25 showed one nucleotide substitution at position 230 (T/G) compared to existing genotype D sequences ( Figure 2). Phylogenetic analysis was carried out to understand the genetic relationships among the Ent. bieneusi genotypes. A neighbor-joining tree was constructed using the Ent. bieneusi ITS nucleotide sequences from humans and domestic animals. KDCA25 was clustered into Group 1, the human pathogenic group (Figure 3).

Figure 1.
Phylogenetic tree of Encephalitozoon spp. and Enterocytozoon bieneusi isolates and selected accessions from GenBank, based on small subunit ribosomal RNA gene fragment sequences. The phylogenetic tree was constructed using nucleotide sequence alignments, with the Kimura two-parameter algorithm as the distance method and neighbor-joining as the tree composition method. The black squares indicate the known genotypes identified in this study. The sequence of Nosema bombycis was used as the outgroup.

Genotype of Ent. bieneusi
Based on the sequencing analysis of the internal transcriptional spacer (ITS) gene, we detected one Ent. bieneusi genotype, Korea-WL2. The KDCA25 Ent. bieneusi isolate obtained in the present study was identical to LC436502 from a raccoon dog and LC436503 from a Korean water deer in the ROK [20]. KDCA25 showed one nucleotide substitution at position 230 (T/G) compared to existing genotype D sequences (Figure 2). Phylogenetic analysis was carried out to understand the genetic relationships among the Ent. bieneusi genotypes. A neighbor-joining tree was constructed using the Ent. bieneusi ITS nucleotide sequences from humans and domestic animals. KDCA25 was clustered into Group 1, the human pathogenic group (Figure 3).  (1). Each sequence is identified by its genotype designation. Sequence differences of ITS sequences obtained in this study are shown. "." indicates an identical nucleotide to MK696083 (D).  (1). Each sequence is identified by its genotype designation. Sequence differences of ITS sequences obtained in this study are shown. "." indicates an identical nucleotide to MK696083 (D). Pathogens 2021, 10, x FOR PEER REVIEW 5 of 9 Figure 3. Phylogenetic tree based on neighbor-joining analysis of internal transcriptional spacer (ITS) sequences. The phylogenetic relationships between the Enterocytozoon bieneusi genotype determined in this study and previously reported genotypes from GenBank were inferred by neighborjoining analysis of ITS sequences, based on Kimura two-parameter model genetic distances. The numbers on the branches are percentage bootstrapping values from 1000 replicates. Each sequence is identified by its accession number, host, origin, and genotype designation. The black square indicates the known genotype identified in this study.

Discussion
Ent. bieneusi and Enc. intestinalis are both common microsporidia species responsible for gastrointestinal diseases in humans [24]. Ent. bieneusi is the most common microsporidial cause of intestinal diseases [25]. However, Enc. intestinalis showed a higher infection rate than Ent. bieneusi in the ROK. This could have two possible explanations. First, the detection rate depends on patient condition. Ent. bieneusi infection accounts for 30-51% of all cases of diarrhea in immunocompromised patients [26]. Previous studies on Enc. intestinalis and Ent. bieneusi focused on immunocompromised patients. Although this study focused on immunocompetent patients with diarrhea, it is necessary to include immunocompromised patients as a control group in future studies. Second, developed countries have reported gradual decreases in the prevalence and occurrence of Ent. bieneusi in immunocompromised patients due to the use of antiretroviral therapies and improved hygiene [27]. sequences. The phylogenetic relationships between the Enterocytozoon bieneusi genotype determined in this study and previously reported genotypes from GenBank were inferred by neighbor-joining analysis of ITS sequences, based on Kimura two-parameter model genetic distances. The numbers on the branches are percentage bootstrapping values from 1000 replicates. Each sequence is identified by its accession number, host, origin, and genotype designation. The black square indicates the known genotype identified in this study.

Discussion
Ent. bieneusi and Enc. intestinalis are both common microsporidia species responsible for gastrointestinal diseases in humans [24]. Ent. bieneusi is the most common microsporidial cause of intestinal diseases [25]. However, Enc. intestinalis showed a higher infection rate than Ent. bieneusi in the ROK. This could have two possible explanations. First, the detection rate depends on patient condition. Ent. bieneusi infection accounts for 30-51% of all cases of diarrhea in immunocompromised patients [26]. Previous studies on Enc. intestinalis and Ent. bieneusi focused on immunocompromised patients. Although this study focused on immunocompetent patients with diarrhea, it is necessary to include immunocompromised patients as a control group in future studies. Second, developed countries have reported gradual decreases in the prevalence and occurrence of Ent. bieneusi in immunocompromised patients due to the use of antiretroviral therapies and improved hygiene [27].
In 2015, Kim et al. [19] reported a 5% (7/139) prevalence of Enc. intestinalis in patients with diarhhea in the ROK; however, we found a prevalence of 2% (24/1241). In this study, the infection rate was high in the 20-29 (2/43, 4.7%) and 30-39 (3/46, 6.5%) age groups. However, in previous studies, the infection rates were high in the 11-20 (4/15, 26.7%) and 31-40 (1/5, 20%) age groups. In addition, no infection rate was reported for the 21-30 age group by Kim et al. [19] as it contained only one sample, while 20-29 group had the second-highest prevalence in this study. We believe that sampling differences could have affected the low prevalence in this study. As no data related to the infection sources or other clinical information are available for the positive cases in the previous study, it is difficult to explain the differences between our results and those of the previous study.
In this study, the Ent. bieneusi isolate (KDCA 25) belonged to Group 1. The sequence matched with those previously reported from domestic wildlife [19], suggesting a possible infection through fecal-oral route via contaminated water, the environment, or food from domestic wildlife [35]. There have been no previously reported cases of Ent. bieneusi in the ROK, except for domestic and wild animals. Previous studies on Ent. bieneusi infection have shown a total infection rate of 45.2% (71/157) in domestic animals [3] and 18% (53/314) in native calves [21]. Other studies have reported infection rates of 16% (38/237) in pigs with diarrhea [22] and 14.9% (80/538) in cattle [23], 1.9% (4/210) in bat feces [30], 2.6% (15/502) in wild boars [31], and 8.3% (15/180) in milk specimens from cows [32]. As previously reported, the possibility of human infection is suggested by the continuous reports of Ent. bieneusi infections in domestic wildlife and livestock. Microsporidian spores remain viable in water after desiccation following incubation at various temperatures, suggesting that indirect zoonotic transmission of microsporidia between animals and humans could occur through exposure to contaminated water, food, or aerosols [6,[40][41][42].
In this study, we detected Ent. bieneusi and Enc. intestinalis in domestic patients with diarrhea, providing the first evidence that Ent. bieneusi exists in humans in addition to animals in the ROK. Therefore, continuous monitoring for Enc. intestinalis and Ent. bieneusi is necessary for patients with acute diarrhea. In additon, epidemiologic analysis is needed to understand the infection pathways of Enc. intestinalis and Ent. bieneusi.

Fecal Sample Collection and DNA Extraction
A total of 1241 stool samples from patients with diarrhea were collected through the Enteric Pathogens Active Surveillance Network (Enter-Net) of the Korea Disease Control and Prevention Agency (KDCA) in 2020. The samples were examined for bacteria, viruses, and parasitic protozoa to determine the cause of diarrhea.

Nucleotide Sequencing and Phylogenetic Analysis
The amplified Enc. intestinalis and Ent. bieneusi DNA fragments were sequenced using both forward and reverse primers ( Table 2). The resulting nucleotide sequences were subjected to BLAST searches (https://blast.ncbi.nlm.nih.gov/Blast.cgi accessed on 13 October 2021) using the data available on GenBank (http://www.ncbinlm.nih.gov/ genbank/ accessed on 13 October 2021). Multiple alignments were conducted by ClustalW using BioEdit version 7.2.5 (Ibis Therapeutics Inc., Carlsbad, CA, USA). To determine the similarity and difference rates between the sequences, the MegaAlign program (DNASTAR, Madison, WI, USA) was employed. Finally, a phylogenetic analysis was performed with MEGA software (version 5.02) using the maximum parsimony algorithm with a Kimura two-parameter model assessed using bootstrap analysis with 1000 replications. Each sequence was identified by its accession number, host, origin, and genotype designation.

Statistical Analysis
Fisher's exact test was used to assess the association between Enc. intestinalis test positivity and factors such as sex and age group. Odds ratios and 95% confidence intervals were used to measure univariate associations. In this study, p-values < 0.05 were considered statistically significant. All statistical analyses were performed using SPSS Statistics 23.0 (International Business Machines Corporation, New York, NY, USA).