A Simple Genotyping Method for Rapid Differentiation of Blastocystis Subtypes and Subtype Distribution of Blastocystis spp. in Thailand

Blastocystis spp. is one of the most common protozoa of humans and animals worldwide. The genetic diversity of Blastocystis spp. might be associated with a wide range of symptoms. However, the prevalence of each subtype is different in each country. Until now, there is no standard method for subtyping of Blastocystis spp. We developed a sequential restriction fragment length polymorphism (RFLP) analysis for the rapid differentiation of human Blastocystis subtypes. A large-scale study was also conducted to determine the subtype distribution of Blastocystis spp. in Thailand. Stool samples were collected from 1025 school-age students in four regions of Thailand. Blastocystis infections were identified by direct smear, formalin ethyl-acetate concentration technique (FECT), Boeck and Drbohlav’s Locke-Egg-Serum (LES) medium culture, and polymerase chain reaction (PCR) of small-subunit ribosomal DNA (SSU rDNA). Subtypes of Blastocystis spp. were determined by RFLP. Phylogenetic tree of partial SSU rDNA sequences of Blastocystis spp. was constructed using the Maximum Likelihood (ML) method. Out of 1025 students, 416 (40.6%) were positive for Blastocystis spp. Using two steps of RFLP reactions, we could determine subtype one–three among these students. Subtype 3 was the most common subtype (58.72%) in Thai students, followed by subtype 1 (31.2%), and subtype 2 (10.1%). Blastocystis subtype 3 was the most prevalent in all regions of Thailand. The subtype distribution of Blastocystis spp. in Thailand was different from other countries.


Introduction
Blastocystis spp. is a common enteric protozoan in human and animals worldwide. Little is currently known about the biology of Blastocystis spp. Until now, taxonomy, life cycle, mode of transmission, as well as treatment management of the Blastocystis infection remains unclear. Moreover, conflicting studies or reports concerning the pathogenicity of the protozoan continue to be published. In the past, Blastocystis spp. was thought to be pathogenic only in immunocompromised patients [1]. The public health significance of Blastocystis infection has increased in the last decade. Blastocystis infection causes various symptoms in immunocompetent individuals in the absence of any parasitic co-infection [2,3]. Prevalence of Blastocystis infection is about 1%-23% in developed countries [4,5], Therefore, the subtype distribution of Blastocystis spp. in Thailand was also studied among children in all regions in Thailand in this study.

PCR-RFPL Analysis for Differentiation of Blastocystis Subtypes
To determine genetic diversity (subtype distribution) of Blastocystis spp., 480 bp of SSU rDNA fragments were successfully amplified with the Blast 505-532 (BH-SSU-F 5 GGAGGTAGTGACAATAAATC 3 ) and Blast 998-1017 (BH-SSU-R 5 TGCTTTCGCACTTGTTCATC 3 ) [24] primers from 1025 stool specimens of school-age students from 4 provinces, including Ang Thong, Nan, Nakhon Ratchasima, and Kanchanaburi which located in central, Northern, Northeastern and Western of Thailand, respectively. About 109 stool specimens were positive (26.2%) with this pair of primers. Non-specific PCR products were not found. The RFLP analysis was then used to differentiate subtypes of Blastocystis spp. The full diagram of PCR-RFLP analysis was shown in Figure 1. The PCR amplicons were separately digested with Mfe I and Ase I enzyme. From 109 PCR positive samples, 45 samples showed pattern A (suspected subtype 1 or subtype 2) in step 1 digestion. In pattern A, digestion with Mfe I enzyme showed DNA fragments sized 435 bp and 45 bp (not clearly seen) (Figure 2), while the PCR amplicons were uncut by Ase I enzyme ( Figure 3). Out of 109 PCR positive samples, 64 samples showed pattern B (suspected subtype 3 or subtype 5) in step 1 digestion. In pattern B, digestion with Mfe I enzyme showed DNA fragments sized 395 bp, 45 bp, and 40 bp (not clearly seen) (Figure 2), while digestion with Ase I enzyme showed DNA fragments sized 255 bp, and 225 bp (Figures 2 and 3). Unfortunately, pattern C, D, and E were not found in this study.
From pattern A, the 480 bp-PCR amplicons were digested with BsrGI enzyme in step 2 digestion to differentiate between subtype 1 and subtype 2 ( Figure 1). Out of 45 samples with pattern A, 11 samples were uncut with BsrGI enzyme (Figure 4). These samples were identified as subtype 2. Out of 45 samples digested with BsrGI enzyme, 34 samples showed the DNA fragment sized about 415 bp ( Figure 4). These samples were identified as subtype 1.
From pattern B, the 480 bp-PCR amplicons were digested with BspHI in step 2 digestion to differentiate between subtype 3 and subtype 5. All of the 64 samples with pattern B in step 1 digestion showed the same pattern of RFLP analysis with BspHI enzyme. DNA fragments sized 360 bp and 90 bp were detected ( Figure 5). These samples were identified as subtype 3.

PCR-RFPL analysis for differentiation of Blastocystis subtypes
To determine genetic diversity (subtype distribution) of Blastocystis spp., 480 bp of SSU rDNA fragments were successfully amplified with the Blast 505-532 (BH-SSU-F 5ʹ GGAGGTAGTGACAATAAATC 3ʹ) and Blast 998-1017 (BH-SSU-R 5ʹ TGCTTTCGCACTTGTTCATC 3ʹ) [24] primers from 1025 stool specimens of school-age students from 4 provinces, including Ang Thong, Nan, Nakhon Ratchasima, and Kanchanaburi which located in central, Northern, Northeastern and Western of Thailand, respectively. About 109 stool specimens were positive (26.2%) with this pair of primers. Non-specific PCR products were not found. The RFLP analysis was then used to differentiate subtypes of Blastocystis spp. The full diagram of PCR-RFLP analysis was shown in Figure 1. The PCR amplicons were separately digested with Mfe I and Ase I enzyme. From 109 PCR positive samples, 45 samples showed pattern A (suspected subtype 1 or subtype 2) in step 1 digestion. In pattern A, digestion with Mfe I enzyme showed DNA fragments sized 435 bp and 45 bp (not clearly seen) (Figure 2), while the PCR amplicons were uncut by Ase I enzyme ( Figure 3). Out of 109 PCR positive samples, 64 samples showed pattern B (suspected subtype 3 or subtype 5) in step 1 digestion. In pattern B, digestion with Mfe I enzyme showed DNA fragments sized 395 bp, 45 bp, and 40 bp (not clearly seen) (Figure 2), while digestion with Ase I enzyme showed DNA fragments sized 255 bp, and 225 bp ( Figure 2 and Figure 3). Unfortunately, pattern C, D, and E were not found in this study. Step 1 Step 2 Step 3  From pattern A, the 480 bp-PCR amplicons were digested with BsrGI enzyme in step 2 digestion to differentiate between subtype 1 and subtype 2 ( Figure 1). Out of 45 samples with pattern A, 11 samples were uncut with BsrGI enzyme (Figure 4). These samples were identified as subtype 2. Out of 45 samples digested with BsrGI enzyme, 34 samples showed the DNA fragment sized about 415 bp ( Figure 4). These samples were identified as subtype 1.
From pattern B, the 480 bp-PCR amplicons were digested with BspHI in step 2 digestion to differentiate between subtype 3 and subtype 5. All of the 64 samples with pattern B in step 1 digestion showed the same pattern of RFLP analysis with BspHI enzyme. DNA fragments sized 360 bp and 90 bp were detected ( Figure 5). These samples were identified as subtype 3.   From pattern A, the 480 bp-PCR amplicons were digested with BsrGI enzyme in step 2 digestion to differentiate between subtype 1 and subtype 2 ( Figure 1). Out of 45 samples with pattern A, 11 samples were uncut with BsrGI enzyme (Figure 4). These samples were identified as subtype 2. Out of 45 samples digested with BsrGI enzyme, 34 samples showed the DNA fragment sized about 415 bp ( Figure 4). These samples were identified as subtype 1.
From pattern B, the 480 bp-PCR amplicons were digested with BspHI in step 2 digestion to differentiate between subtype 3 and subtype 5. All of the 64 samples with pattern B in step 1 digestion showed the same pattern of RFLP analysis with BspHI enzyme. DNA fragments sized 360 bp and 90 bp were detected ( Figure 5). These samples were identified as subtype 3.    . RFLP analysis of SSU rDNA by step 2 digestion with BsrGI enzyme to differentiate between ST1 and ST2. Lane M: 1 kb DNA ladder; Lane 1-3: digestion with BsrGI obtained fragments sized 415 bp which were identified as ST1; Lane 4-6: digestion with BsrGI showed uncut bands which were identified as ST2.

Phylogenetic analysis
After subtype differentiation with PCR-RFLP analysis, we randomly selected 37 samples with a different pattern of RFLP analysis from Ang Thong Province to perform DNA sequencing analysis. The ML tree analysis was constructed to characterize Blastocystis isolates in this study compared with the published sequence in the GenBank database ( Table 1). The phylogenetic tree showed three distinct subtypes among Blastocystis spp. isolated from students in the Ang Thong province ( Figure  6). These results confirmed the results of the PCR-RFLP analysis.

Phylogenetic Analysis
After subtype differentiation with PCR-RFLP analysis, we randomly selected 37 samples with a different pattern of RFLP analysis from Ang Thong Province to perform DNA sequencing analysis. The ML tree analysis was constructed to characterize Blastocystis isolates in this study compared with the published sequence in the GenBank database ( Table 1). The phylogenetic tree showed three distinct subtypes among Blastocystis spp. isolated from students in the Ang Thong province ( Figure 6). These results confirmed the results of the PCR-RFLP analysis.

Comparative Study of the Different Techniques for Blastocystis Detections
Out of 1025 stool samples collected from students in 4 regions of Thailand (Table 2). 416 students (40.6% prevalence) were positive for Blastocystis spp. by at least one method of detection (Table 3). While the highest prevalence of Blastocystis spp. was found among students in Kanchanaburi province (21.5%), the lowest prevalence was found among students in Nakhon Ratchasima province (0.6%) (p < 0.001). The prevalence in Ang Thong, and Nan province were 8.3%, and 10.2%, respectively (Table 3). A comparison between the direct smear, formalin-ethyl acetate concentration (FECT), Boeck and Drbohlav's Locke-Egg-Serum (LES culture), and PCR showed a significant difference (p < 0.001) ( Table 3). Out of 416 infected students, 105 students were positive for Blastocystis spp. by direct smear (25.2%), 122 students were detected by FECT (29.3%) (p > 0.05). Interestingly, Blastocystis spp. was detected in 277 samples by LES culture (66.6%). The detection rate of LES culture was higher than direct smear with statistical difference (p < 0.001).

Distribution of Blastocystis Subtypes in Thai Students
Using the 3-step RFLP analysis of SSU rDNA, only subtype 1-3 were found among student in Thailand. Other subtypes were not found in this population. The most common subtype was subtype 3 (64 of 109; 58.7%), followed by subtype 1 (34 of 109; 31.2%), and subtype 2 (11 of 109; 10.1%), respectively (Table 3). Subtype 3 was found with the highest prevalence in all of the province we studied (46%-100%) (Figure 7). The distribution pattern of Blastocystis spp. in each province was similar (subtype 3, followed by subtype 1 and subtype 2, respectively). Moreover, the proportion of each subtype in each province was not significantly different. Subtype 3 was found with the highest proportion (61%) in Nan province. Subtype 1 and subtype 2 were almost found with the highest proportion in Kanchanaburi province (39% and 15%, respectively) Among other provinces ( Figure 7). Unexpectedly, we found only subtype 3 among students in Nakhon Ratchasima province, which is located in the Northeast region of Thailand ( Figure 7). However, there were five positive samples in the Nakhon Ratchasima province. Therefore, this prevalence might be underestimated and did not provide an exact distribution of subtypes.

Discussion
In this study, we reported a high prevalence of Blastocystis spp. (40.6%) in school-age students ( Table 3) from 4 provinces, which located in central, Northern, Northeastern, and Western Thailand. Our data were consistent with several studies, which show that Blastocystis infections are common among children in Thailand. This prevalence was higher than the prevalence in adults (14.5%-37.2%) [10][11][12] and in children in Bangkok (13.6%-31.9%) [38,52], but slightly lower than in children in orphanage (45.2%) [53]. This may be due to the overcrowded environment and poor sanitation in the orphanage. Moreover, we found that the infection rates of Blastocystis spp. in each region of Thailand were significantly different. The reasons for this difference may be directly related to the specific geographic characteristics, as well as to ecological, sanitary, socioeconomic, and cultural factors [54].
Our study showed that the highest prevalence of Blastocystis infection was found in

Discussion
In this study, we reported a high prevalence of Blastocystis spp. (40.6%) in school-age students (Table 3) from 4 provinces, which located in central, Northern, Northeastern, and Western Thailand. Our data were consistent with several studies, which show that Blastocystis infections are common among children in Thailand. This prevalence was higher than the prevalence in adults (14.5%-37.2%) [10][11][12] and in children in Bangkok (13.6%-31.9%) [38,52], but slightly lower than in children in orphanage (45.2%) [53]. This may be due to the overcrowded environment and poor sanitation in the orphanage. Moreover, we found that the infection rates of Blastocystis spp. in each region of Thailand were significantly different. The reasons for this difference may be directly related to the specific geographic characteristics, as well as to ecological, sanitary, socioeconomic, and cultural factors [54].
Our study showed that the highest prevalence of Blastocystis infection was found in Kanchanaburi (21.5%) province located in the western region of Thailand. Similar to the previous reports [55,56], the source of the infections could be drinking water contaminated with the parasites. The majority of the infected students were Karen and Myanmar migrants. Their basic health education and sanitation are poor; for examples, some families still use the cesspool latrine, while waste from the house and farm are drained directly into the nearest natural canal. They also usually drink mountain water (data from the questionnaire and observation). In this study, we found a lower prevalence rate of Blastocystis infection (0.49%) in the Northeastern region compared to other studies (1.6%-5.6%) [57,58]. The reasons for this difference may be due to the difference of sanitary and socioeconomic factors. The schools we surveyed in Nakhon Ratchasima province are in the urban area. The environment in schools is clean. From the questionnaire, we observed that the family status, such as occupation and estimated monthly incomes of the parents, of all students in the Nakhon Ratchasima province is quite well. About 73.1% of student's families in Nakhon Ratchasima have mid to high monthly income (>10,000 THB; >1500 USD), while 85.8% of student's families in Kanchanaburi have low monthly income (<10,000 THB; <1500 USD). The major occupation of parents in Nakhon Ratchasima is government officer (48.1%), while parents in Kanchanaburi is labor (70.8%).
In the comparative study, we found that FECT could not significantly increase the detection rate of Blastocystis detections over direct smear (29.3%, and 25.2%, respectively) ( Table 3). This might be due to the fact that some protozoa could be destroyed or distorted during the process of FECT. The highest detection rate for the detection of Blastocystis spp. was found by LES medium culture (66.6%). Our results showed that 37.4% of the Blastocystis infected patients were diagnosed only by LES medium culture, but not by direct smear and FECT. This study confirmed that detections of Blastocystis spp. by direct smear and FECT may not be sensitive enough for the diagnosis of the infections. PCR and real-time PCR are highly sensitive methods for the diagnosis of many infectious diseases. The sensitivity of PCR technique is 2-fold higher than the microscopic method using a modified iron-hematoxylin stain for Blastocystis detections [59]. However, we did not detect Blastocystis infection in all infected students by PCR for SSU rDNA. This could be attributed to the fact that there are many PCR inhibitors; such as bile salts, bilirubin, heme, and carbohydrates, contaminated in stool samples [60,61]. False negative results are also reported in samples with low levels of parasites [60]. Other possibilities included the limitation during the transportation of stool samples from rural areas.
The associations between the subtypes and clinical symptoms, as well as zoonosis of Blastocystis spp., are controversially reported. Moreover, the prevalence of each subtype reported in each country is different. Subtype 3 (41.7%-92.3%) is the most common subtype reported in most countries, followed by subtype 1 and subtype 4 [35,56]. Alternatively, the most common subtype found in Spain is subtype 4 (94.1%) [31]. Subtype 1 (51.1%) is the dominant subtype in China [62], in Libya [63], and in some regions in Thailand [10,38,55]. Moreover, the distribution pattern and the proportion of each subtype reported from each study are also different. These differences might be due to the different subtyping method of the parasites. Until now, there has been no standard genotyping method for Blastocystis spp. A simple molecular method for subtyping of Blastocystis spp. might be useful for the better understanding of the clinical association and zoonotic transmission by the parasite.
In this study, we successfully developed a simple genotyping method to identify the subtypes of Blastocystis spp. in humans. We also studied the subtype distribution of Blastocystis spp. in Thailand. Our PCR-RFLP method is a simple, rapid, and cost-effective method, which allow us to differentiate the human Blastocystis subtype 1-9. However, even when we determined the subtypes of Blastocystis isolated from 109 stool samples collected from different regions in Thailand, we identified only subtype 1-3 among these students. The most common subtype found in all regions was subtype 3 (46%-100%) ( Table 3 and Figure 7). Our results were consistent with the previous studies in Thailand that reported subtype 3 as the dominant subtype (53.2%-80%) in different populations [11,46,52,64,65]. However, our results were different to some studies that report subtype 1 as the major subtype (77.9%-100%) [10,38,55].
The different proportion and different pattern of subtype distribution in each region of Thailand are also reported in previous studies in Thailand. A study among students in Chachoengsao province reported that the dominant subtype is subtype 1 (77.9%), followed by subtype 2 (22.1%) [55]. Blastocystis spp. found in military personnel and dogs is also identified as subtype 1 [10]. Interestingly, there is no subtype 3 found among these students and military staffs. Similarly, a study in a home for girls in Bangkok, Thailand reported subtype 1 as the major subtype (94.8%), followed by subtype 6 (3.5%) and subtype 2 (1.7%) [38], however, subtype 3 was not detected in these girls. On the other hand, a study in Nonthaburi province, a province near Bangkok, Thailand reported only subtype 3 (76%), subtype 1 (16%-20%) and unidentified subtypes (4%-8%) [64]. Moreover, some studies in Thailand reported other subtypes that we did not detect in this study. Only one study reported subtype 4 (0.5%) among villagers at the Thai-Myanmar border [11]. Subtype 6 (3.5%-10%) and subtype 7 (10%-17.9%) are detected in a home for girls in Bangkok, and in a hospital in the Northeast region of Thailand [38,46,65]. These subtypes are reported as avian subtypes [23], but can transmit to humans by zoonosis [66]. Detections of zoonotic subtypes in Thailand might be due to contact with pets and farm animals [38]. Large-scaled molecular epidemiological studies, both in humans and in animals, will be useful to improve the understanding of the transmission of Blastocystis spp. in Thailand. Moreover, the clinical significance of Blastocystis in Thai patients, as well as the study of immunological aspects, should be further investigated for a better understanding of the Blastocystis spp. pathogenicity in humans.

Population and Stool Samples
This study was approved by the Ethics Committee of the Faculty of Medicine at Chulalongkorn University, Bangkok, Thailand (IRB No. 464/55). A total of 1025 stool samples were collected from students from 4 provinces including Ang Thong, Nan, Nakhon Ratchasima, and Kanchanaburi, which are located in central, Northern, Northeastern, and Western of Thailand, respectively. Stool collection containers were distributed to all students. To avoid contamination, all students were well informed about specific precaution and guideline for specimen collection. Written informed consent was obtained from each individual or their parents/guardians who participated in the study.

Stool Examination
The stool samples were kept under room temperature during transportation to the Department of Parasitology, Faculty of Medicine, Chulalongkorn University. To diagnose Blastocystis spp. stool examinations were performed. First, stool samples were examined microscopically by direct smear using in normal saline and iodine preparation. The remaining stool samples were examined by FECT, and LES medium, as previously described [9]. The LES medium culture was performed for protozoa isolation. The presence of intestinal parasite eggs, larvae, or cysts was determined microscopically. All of stool samples were independently examined by two examiners. The remaining stool samples of each individual were preserved at -20 • C until DNA extraction.

DNA Extraction
Genomic DNA of Blastocystis spp. was extracted from approximately 250 mg of stool samples or from suspension collected from LES cultures using E.Z.N.A. ® Stool DNA Kit (OMEGA Bio-tek Inc., Norcross, GA, USA) according to the manufacturer's protocol. DNA samples were stored at −20 • C until used.

Polymerase Chain Reaction (PCR)
The small subunit ribosomal DNA (SSU rDNA) was amplified using the universal forward primer Blast 505-532 (BH-SSU-F 5 GGAGGTAGTGACAATAAATC 3 ) and the universal reverse primer Blast 998-1017 (BH-SSU-R 5 TGCTTTCGCACTTGTTCATC 3 ), as previously described [24]. Based on conserved regions of all available published nucleotide SSU rDNA sequences of Blastocystis spp. from GenBank, these primers could amplify all human Blastocystis subtypes. These primers produced an approximately 480 bp PCR product containing a highly variable region that allows the subtyping of Blastocystis spp.
The PCR was carried out in a 50 µl reaction containing 1× PCR buffer, 3 mM of MgCl 2 , 0.

Restriction Fragment Length Polymerase (RFLP)
To differentiate between the subtypes of Blastocystis spp., RFLP analysis was developed. Based on all available published nucleotide SSU rDNA sequences of Blastocystis subtypes from GenBank, the restriction map was predicted for all subtype using the BioEdit program version 7.0.5.2. The 3-step digestion with 5 restriction enzymes (MfeI, AseI, BsrGI, BspHI, and DraI) was designed to be used for RFLP analysis of SSU rDNA of subtype 1-9 ( Figure 1). In step 1 digestion, the 480 bp-PCR amplicons were digested separately with MfeI and AseI enzyme (New England BioLabs, Pickering, ON, Canada). The digestion produced 5 RFLP patterns (Pattern A-E) (Figure 1). Pattern A was suspected to be subtype 1 or subtype 2. Pattern B was suspected to be subtype 3 or subtype 5. Pattern C was suspected to be subtype 7. Pattern D was suspected to be subtype 9. Pattern E was suspected to be subtype 4, or subtype 6, or subtype 8 ( Figure 1). To differentiate between subtype 1 and subtype 2 (Pattern A), the 480 bp-PCR amplicons was digested with BsrGI in step 2 digestion. To differentiate between subtype 3 and subtype 5 (Pattern B), the 480 bp-PCR amplicons was digested with BspHI. To differentiate subtype 4 from subtype 6 and 8 (Pattern E), the 480 bp-PCR amplicons was digested with BsrGI in step 2 digestion, and then digested with DraI in step 3 digestion to differentiate between subtype 6 and subtype 6 ( Figure 1).
The reaction mixture was about 6 µl of PCR product, 2 units of restriction enzyme, 1 µl of appropriate buffer and dH 2 O to final 10 µl. The RFLP mixture was incubated at 37 • C for 2 h and reactivated the reaction at 65 • C for 20 min. The RFLP patterns were detected on 10% acrylamide gel. The DNA fragments were stained with ethidium bromide and visualized under UV light.

DNA Sequencing and Phylogenetic Analysis
To confirm the subtypes of Blastocystis, the 20 µl of PCR products were purified by ExoSAP-IT clean-up kit (Affymetrix, High Wycombe, UK). The DNA sequencing was performed in both directions (1st Base Laboratories Sdn. Bhd., Selangor, Malaysia). All nucleotide sequences were analyzed and edited by the BioEdit Sequence Alignment Editor version 7.0.5.2. The multiple alignments with the published SSU rRNA gene available in the GenBank Database (Table 1) were performed by ClustalW multiple alignment options in the BioEdit software (Mega4 software version 4.0, Tempe, AZ, USA). Phylogenetic analysis of the aligned nucleotide sequences was also performed by MEGA software version 6. Phylogenetic trees were constructed using the Maximum Likelihood (ML) method base on the Tamura 3-parameter (T92) using a discrete Gamma distribution (+G) Tamura 3-parameter