Impact of Enterobius vermicularis infection and mebendazole treatment on intestinal microbiota and host immune response

Background Previous studies on the association of enterobiasis and chronic inflammatory diseases have revealed contradictory results. The interaction of Enterobius vermicularis infection in particular with gut microbiota and induced immune responses has never been thoroughly examined. Methodology/Findings In order to answer the question of whether exposure to pinworm and mebendazole can shift the intestinal microbial composition and immune responses, we recruited 109 (30 pinworm-negative, 79 pinworm-infected) first and fourth grade primary school children in Taichung, Taiwan, for a gut microbiome study and an intestinal cytokine and SIgA analysis. In the pinworm-infected individuals, fecal samples were collected again at 2 weeks after administration of 100 mg mebendazole. Gut microbiota diversity increased after Enterobius infection, and it peaked after administration of mebendazole. At the phylum level, pinworm infection and mebendazole deworming were associated with a decreased relative abundance of Fusobacteria and an increased proportion of Actinobacteria. At the genus level, the relative abundance of the probiotic Bifidobacterium increased after enterobiasis and mebendazole treatment. The intestinal SIgA level was found to be lower in the pinworm-infected group, and was elevated in half of the mebendazole-treated group. A higher proportion of pre-treatment Salmonella spp. was associated with a non-increase in SIgA after mebendazole deworming treatment. Conclusions/Significance Childhood exposure to pinworm plus mebendazole is associated with increased bacterial diversity, an increased abundance of Actinobacteria including the probiotic Bifidobacterium, and a decreased proportion of Fusobacteria. The gut SIgA level was lower in the pinworm-infected group, and was increased in half of the individuals after mebendazole deworming treatment.


Methodology/Findings
In order to answer the question of whether exposure to pinworm and mebendazole can shift the intestinal microbial composition and immune responses, we recruited 109 (30 pinwormnegative, 79 pinworm-infected) first and fourth grade primary school children in Taichung, Taiwan, for a gut microbiome study and an intestinal cytokine and SIgA analysis. In the pinworm-infected individuals, fecal samples were collected again at 2 weeks after administration of 100 mg mebendazole. Gut microbiota diversity increased after Enterobius infection, and it peaked after administration of mebendazole. At the phylum level, pinworm infection and mebendazole deworming were associated with a decreased relative abundance of Fusobacteria and an increased proportion of Actinobacteria. At the genus level, the relative abundance of the probiotic Bifidobacterium increased after enterobiasis and mebendazole treatment. The intestinal SIgA level was found to be lower in the pinworm-infected group, and was elevated in half of the mebendazole-treated group. A higher proportion of pre-treatment Salmonella spp. was associated with a non-increase in SIgA after mebendazole deworming treatment. PLOS

Introduction
The inverse epidemiology data of parasitosis and autoimmune diseases has led to the hypothesis that childhood exposure to parasites might have protective effects against the development of allergies and autoimmunity [1]. The immunomodulatory roles of helminth have been well studied, and several helminth-derived components might regulate the immune system [2,3]. Enterobius vermicularis (human pinworm) is the most common parasite encountered in developed countries, and it has been suggested as a good candidate for testing the link between the "hygiene hypothesis" and disease [4]. In Taiwan, enterobiasis is found in about 0.6-3% of primary school children [5,6]. A previous study using peri-anal tape tests and questionnaires with Taipei primary school children suggested a negative correlation between pinworm infection and allergic airway diseases [7]. Similarly, enterobiasis has been found to be associated with a decreased risk of allergic wheezing in Turkish school-aged children [8]. However, a large population cohort study collecting data of mebendazole prescriptions and chronic inflammatory diseases in Denmark showed that enterobiasis does not reduce the risk for asthma, type 1 diabetes (type 1 DM), arthritis, or inflammatory bowel disease (IBD) [9]. These abovementioned reports lacked mechanistic studies and did not examine the interaction between pinworm and intestinal microbiota. The imbalance of pro-inflammatory and anti-inflammatory gut bacteria, or dysbiosis, is associated with autoimmune diseases such as type 1 DM, IBD, rheumatoid arthritis (RA), along with pro-inflammatory conditions, such as obesity, atherosclerosis and colon cancer [10][11][12]. Since the parasites and intestinal commensal bacteria reside in the same environment, interaction between these two micro-organisms is conceivable. In humans, it has been reported that helminth infections may increase intestinal bacterial diversity, and alter the composition of microbiota [13][14][15]. Gut microbiota is suspected as causing T helper type 1 (Th1) responses in Trichuris muris infections in mice, and Schistosoma mansoni has been suggested to cause Th1-mediated inflammation and granuloma formation via alteration of microbiota [16,17]. Furthermore, the protective mucosal immune response against Toxoplasma gondii has been reported to be provided by gut microflora that stimulate dendritic cells [18]. Recently, Ramanan et al. reported that helminth infections may restore the number of goblet cells via suppression of an intestinal pro-inflammatory Bacteroides species, and thus protect genetically susceptible mice from the development of Crohn's disease [19]. Therefore, the net immunomodulatory effect of pinworm on an individual may be dependent on its interaction with that individual's intestinal microbiota.
In this study, we examined the impact of Enterobius exposure on the composition of gut microflora, and we investigated the interactions among pinworm, microbiota, and host immune responses in a prospective cohort of 109 primary school-aged children. Through our observations of differences in probiotic bacteria abundance and changes in gut levels of the protective secretory IgA (SIgA), we hypothesized the possible correlations of pinworm and mebendazole exposure with the inflammation status of the gut.

Ethics statement
The study cohort consisted of 109 primary school children (1 st and 4 th grades) who had undergone pinworm screening in 2015, Taichung, Taiwan. The study was approved by the Research Ethics Committee of China Medical University Hospital (CMUH104-REC1-115). Written informed consent was obtained from parents.

Study design
A total of 30 children were tested negative for enterobiasis, while 79 were tested positive by anal tape screening. In children with positive pinworm anal tape results, additional stool samples were collected in tipped ova concentration tubes and were stained and fixed with freshly prepared merthiolate-iodine formaldehyde (MIF). After further mixing with ethyl acetate and centrifugation at 1500 rpm for 5 minutes, the sediments were examined carefully under light microscope to detect the presence of co-infected helminth eggs or protozoans as described previously [20]. Stool specimens were collected again from 65 pinworm-infected individuals 2 weeks following 100 mg mebendazole treatment, which underwent MIF-microscopic examination and 16s rRNA gene sequencing. We did not detect co-infection with other helminths or protozoans by MIF-concentration-sedimentation method in pinworm (+) samples before and after mebendazole treatment ( Table 1). As shown in the flow diagram (S1 Fig), metagenomics analysis was performed on 30 pinworm (-), 65 paired pinworm (+) mebendazole (-) and pinworm (+) mebendazole (+) samples.

DNA extraction and gene sequencing
Stool specimens were collected at home and transported to our laboratory within 3 hours in ice, and were fixed in Transwab tubes (Sigma, Dorset, UK). DNA extraction was performed using the QIAamp DNA Stool Mini Kit (Qiagen, California, USA). PCR primers F515 (5'-GTGCCAGCMGCCGCGGTAA-3') and R806 (5'-GGACTACHVGGGTWTCTAAT-3'), were designed to amplify the V4 domain of bacterial 16S ribosomal RNA gene as described previously [21]. The Nextera adapter sequence (Illumina, California, USA) was added to the 5'-end of the primer set for library preparation. PCR using 50~150 ng DNA was performed with 1 cycle of 98˚C for 30 sec, 30 cycles of 98˚C for 10 sec, 60˚C for 30 sec, 72˚C for 30 sec, and a final extension of 72˚C for 5 min. Amplicons were purified using the AMPure XP Beads (Beckman Coulter, Indianapolis, USA), and quantified using Nanophotometer (IMPLEN, München, Germany). The Illumina Nextera Index Primer kit was used to create the library. The qualities and quantities of purified libraries were checked by 2% agarose gel electrophoresis, Qubit (Thermo Fisher Scientific, Massachusetts, USA) and qPCR methods. Finally, libraries were normalized to the same concentration and sequenced by Illumina Miseq sequencer.
Bioinformatic analysis of 16S rRNA gene sequencing data FASTX-Toolkit (http://hannonlab.cshl.edu/fastx_toolkit) was used to process the raw read data files. Sequence qualities were checked in 3 steps: (i) The minimal acceptable Phred quality score of sequences was 20 (having over 70% of the sequence bases ! 20). (ii) Following quality trimming from the sequence tail, the sequences over 100 bp and those with an acceptable Phred quality score of 20 were retained. (iii) Both forward and reverse sequencing reads which met the first and second requirements were retained for subsequent analyses.
UPARSE [22] was used to create operational taxonomic unit (OTU) clustering. Bowtie2 [23] was then used to align OTUs with 16S rRNA gene sequences of bacteria. These sequences were taken from the SILVA ribosomal RNA sequence database (version 115). Following sequence data collection, sequences were extracted by V4 forward primer and reverse primer. To prevent repetitive sequence assignments, V4 sequences from SILVA were then grouped into several clusters by 97% similarity using UCLUST. A standard of 97% similarity with the database was applied. Intestinal SIgA and cytokine detection Fecal samples were weighed before adding equal amounts of sterile PBS together with Pierce proteinase inhibitor (Thermo Fisher Scientific). After thorough mixing and centrifugation at 10000 g for 10 minutes, the fecal supernatants were stored at -80˚C until analysis. Stool secretory IgA (SIgA) was analyzed using Immundiagnostik ELISA kit (Bensheim, Germany), IL-1ß, and IL-4 levels were measured using Quantikine ELISA kits (R&D Systems, Minneapolis, USA) according to manufacturer's instructions. We further grouped the samples according to levels of SIgA, IL-1ß, and IL-4. Based on median levels detected, SIgA was considered high at >150 μg/ml, and low at <80 μg/ml; IL-1ß was high at >10 pg/ml, and low at <0.5 pg/ml; and IL-4 was high at >10 pg/ml, and low at <5 pg/ml. In paired samples (collected before and after mebendazole treatment), SIgA level was considered to be elevated given a greater than 1.1-fold increase.

Statistical analysis
A rarefaction process was applied to normalize the operational taxonomic unit (OTU) table following taxonomy assignment in the bioinformatic analyses. Alpha diversity (Shannon index, inverse Simpson index and richness) was calculated. Beta diversity using weighted UniFrac distance metrics [24], principal coordinate analysis (PCoA) and unsupervised clustering were performed. Multiple response permutation procedure (MRPP) in an R package "vegan" (https://cran.r-project.org/web/package=vegan) was performed to assess community difference in PCoA. Wilcoxon rank sum test were used to compare non-paired variables (e.g., P-v.s. P+M-), and Wilcoxon signed rank test were used to compare paired variables (e.g., P+M-v.s. P+M+). The differentially expressed bacteria were filtered by the following criteria: (i) P value < 0.05 (ii) Fold change > 1.40 or < 0.71 (iii) At least one group achieved an average relative abundance of 0.5%. Sex and pair factors were adjusted. For multiple group comparisons the false discovery rate (FDR) was controlled by using Benjamini-Hochberg (BH) FDR multiple test correction. Pathway enrichment analysis was performed using an R package 'Tax4Fun' [25]. ANOVA test was used to calculate the enrichment difference. Mann-Whitney U tests using GraphPad Prism version 5 were performed to compare fecal cytokine and SIgA levels between groups. Wilcoxon signed rank test was used to analyse the paired stool SIgA data before and after mebendazole treatment.

Study population and characteristics of intestinal microbiota
We analyzed the influence of pinworm infection on gut microbiome in a cohort of 109 children in the first or fourth grade of primary school. Grade and sex effects were both insignificant among the groups (p = 0.650, p = 0.403, respectively, Table 1). In the pinworm (+) group, additional stool specimens were collected to perform MIF concentration sedimentation procedure on to detect co-infection of other parasites. As shown in Table 1, no co-infection was detected. In the pinworm (-) group, possible confounding factors including a mostly-meat diet, recent (within these 2 months) gastroenteritis, recent respiratory tract infection with oral medication, and recent confirmed use of antibiotics were recorded, and diversity analysis of the gut microbiota showed that recent respiratory tract infection with oral medication might decrease the intestinal microbial diversity (Table 1). Further differential abundance analysis at the phylum level showed that children with recent respiratory tract infection and oral medication (8 of 10 had possible antibiotics usage) had a trend to correlate with relatively less abundance of Fusobacteria (0.01%± 0.18% vs. 1.65%±4.62%, p = 0.03, FDR = 0.17). Information about the above confounding factors was not available in the pinworm (+) group.
At the species level, pinworm infection was associated with a trend of increased proportions of Faecalibacterium prausnitzii, Ruminococcus flavefaciens, Alistipes purtredinis, Bifidobacterium longum and uncultured Oscillospira sp. (Fig 3a, percentages and p values are shown in S1 Table), as well as a trend of decreased relative abundance of Acidaminococcus intestine, Megasphaera elsdenii, Veillonella dispar and Fusobacterium varium (Fig 3b and S1 Table). The relative abundance of Faecalibacterium prausnitzii and Ruminococcus flavefaciens were lower, while the proportion of Bifidobacterium longum and uncultured Oscillospira sp. were higher after mebendazole deworming (Fig 3a, S1 Table). Mebendazole deworming was not associated with an increase in the relative abundance of Acidaminococcus intestine, Megasphaera elsdenii, Veillonella dispar and Fusobacterium varium as compared with the pinworm (+) mebedazole (-) group (Fig 3b).

Association of fecal SIgA and cytokine levels with pinworm infection and intestinal bacterial abundance
To analyze the impact of Enterobius exposure on a host's intestinal immune response, stool samples from pinworm-uninfected, pinworm-infected and untreated, and pinworm-infected and treated groups were collected and measured for their SIgA, IL-1ß and IL-4 levels. Pinworm infection was found to be associated with a significant decrease in gut SIgA levels (median level of uninfected vs. pinworm (+) mebendazole (-) group: 125.59 μg/ml vs. 109.56 μg/ml, p<0.01, Fig 4a). The amount of fecal IL-1ß and IL-4 were similar before and after pinworm infection (Fig 4b and 4c). Furthermore, the fecal levels of SIgA and cytokines were grouped into low, medium and high as described in the Methods Section. In the pinworm (-) group, possible confounding factors such as recent gastroenteritis and respiratory tract infection with oral medication were collected; and in this group, correlation studies of SIgA and cytokine levels with intestinal microbial taxa revealed an association of a higher Prevotella proportion with a decreased amount of gut SIgA (relative abundance of Prevotella in SIgA medium-high vs. SIgA low, 6.22% ± 2.97% vs. 26.46% ± 11.52%, p = 0.006 (corrected for respiratory and gastrointestinal infection factors) , Fig 4d), and association of a higher Collinsella abundance with a decreased amount of gut IL-4 (relative abundance of Collinsella in IL-4 medium-high vs. IL-4 low, 0.36% ± 0.21% vs. 1.69% ± 0.84%, p = 0.043 (corrected for respiratory and gastrointestinal infection factors) , Fig 4d).
At the species level, a higher proportion of Salmonella enterica was noted in pre-treatment samples of the SIgA-non-increased group (fold = 7.81, p = 0.012), and it decreased after mebendazole deworming (fold = 0.08, p = 0.010) (S3a and S3b Fig). The association of the increased relative abundance of the probiotic bacteria Streptococcus thermophilus and

Discussion
We examined the impact of Enterobius vermicularis infection and the effect of mebendazole deworming on intestinal microbial composition and mucosal immune responses in 109 primary school children. Both enterobiasis and mebendazole deworming were associated with altered intestinal microbiome.
Consistent with a previous study on helminth-infected microbiota [14], pinworm infection in our study was associated with increased intestinal bacterial diversity. Furthermore, it has been reported that hookworm infection in human subjects with celiac disease could not only increase gut microbial richness, but also regulate gluten-induced inflammation [26]. We did not conduct a pinworm infection trial on human with chronic inflammatory diseases. However, in our study we found that Enterobiasis was associated with an increased relative abundance of Faecalibacterium prausnitzii and Alistipes species. Faecalibacterium prausnitzii has been shown to be underrepresented in the gut of patients with IBD, type 2 DM, and obesity; while Alistipes spp. has been reported to be overrepresented in irritable bowel syndrome (IBS) patients reporting abdominal pain, and in depressive individuals, suggesting a possible role in disturbing the intestinal serotonergic system [10,[27][28][29][30][31]. The pro-inflammatory role of Alistipes spp. remains speculative. As for the bacterial taxa which showed lower percentages after pinworm infection, both Veillonella spp. and Fusobacterium spp. have been suggested to be correlated with pro-inflammatory conditions such as ulcerative colitis and colon cancer, and Veillonella dispar and Fusobacterium varium have been detected in colon carcinoma in adenoma [32][33][34]. Whether bacterial species altered by pinworm favors an anti-inflammatory profile requires further investigations. Anthony et al. (2007) noted that the major immune response raised against helminth infection is the Th2-type response, consisting of an expansion of Th2 helper T cells, eosinophils, mast cells, basophils, elevated IgE, IL-4 and other cytokines, including IL-5 and IL-13 [2]. A previous study in children in central Greece suggested a Th2-type oriented response to pinworms based on elevated serum levels of IgE and eosinophil cationic protein (ECP) [35]. However, we found no differences in the fecal IL-4 levels among the pinworm (-), pinworm (+) mebendazole (-), and pinworm (+) mebendazole (+) groups. In our study, mebendazole deworming was found to be associated with an increased proportion of Collinsella. Interestingly, in the pinworm (-) group, after correction for possible confounding factors including recent gastroenteritis and respiratory infection with oral medication (and possible antibiotics usage), an inverse correlation between Collinsella abundance and gut IL-4 level was detected. Additional research on germ-free animals is needed to evaluate the effect of pinworm and mebendazole on gut Collinsella and IL-4 levels.
IL-1ß is another cytokine that could be altered by parasites, and its over-activation is associated with chronic inflammatory diseases [36]. To establish the chronicity of infection, the murine helminth Heligmosomoides polygyrus bakeri (Hp) has been observed to downregulate the host's IL-4 response by promoting IL-1ß production [37]. In contrast, the parasite Fasciola hepatica has been seen to directly inhibit host IL-1ß secretion [38]. We found that pinworm infection alone did not significantly alter the fecal IL-1ß.
The principle immunoglobulin involved in combating intestinal microbial infection and maintaining mucosal homeostasis with commensal bacteria is SIgA, which mediates antiinflammatory functions via multiple mechanisms [39]. A lack of SIgA can also cause inflammatory diseases [40,41]. Bacteroides thetaiotaomicron colonization in mice has been reported to elevate SIgA levels via an influx of IgA-producing B cells and an increase of polymeric immunoglobulin receptor (pIgR) that mediates the transport of IgA across epithelia [42]. In this study, pinworm infection was found to be correlated with lower gut SIgA level. Furthermore, the amount of intestinal SIgA was found to be negatively associated with the relative abundance of Prevotella. Whether this pinworm-microbial interaction influences gut SIgA production is unclear. Of note, after mebendazole deworming treatment, SIgA levels increased in half of the pinworm-infected subjects. We observed that the intestinal pathogen Salmonella enterica was overrepresented in the SIgA non-increased group, when compared with the SIgAincreased group. Although the relative proportion of Salmonella enterica was lower after mebendazole deworming in the SIgA-non-increased samples, the increase of the probiotic species Streptococcus thermophilus and Bifidobacterium longum following mebendazole treatment was only observed in the SIgA-increased group. Our results suggest that the relative abundance of Salmonella might have a negative effect on the mebendazole deworming -associated increase in the amount of SIgA and probiotic species in the gut.
Mebendazole is a classic anti-helminth drug, which is well-tolerated [43], and is routinely given to pinworm-positive school aged children. A Danish study on pinworm infection and risk of chronic inflammatory diseases even used mebendazole treatment history as a surrogate for enterobiasis diagnosis [9]. The results of our study show that increased percentages of the known probiotic species, Streptococcus thermophilus, and another anti-inflammatory bacterium, Collinsella aerofaciens [44], could be associated with mebendazole deworming, but not with pinworm infection alone. Our study is limited in that we did not use anal tape, a more sensitive method for detection of pinworm eggs than MIF concentration sedimentation, to evaluate the efficacy of mebendazole deworming. However, Wang CC et al. reported in J Microbiol Immunol Infect. 2009 that the efficacy of mebendazole treatment on eradicating pinworm in primary school children in Taichung, Taiwan, was 96% [6]. Mebendazole was found to have anti-inflammatory, anti-angiogenesis and oncogene-suppressing activities in a mouse model of colon cancer initiation [45]. The direct effect of mebendazole on gut microbiota composition remains to be investigated. An enrichment pathway analysis of the microbiome in our study showed that the increase in the percentages of microbes involved in the metabolism of fatty acid elongation and caffeine after pinworm infection only became significant when a comparison was made between the pinworm (+) mebendazole (+) group and the pinworm (-) group. Further metabolomics studies are needed to evaluate if pinworm and mebendazole treatment could alter the metabolism of commensal bacteria and subsequently influence a host's immune system.
In our study, the change in microbial composition detected two weeks after administration of mebendazole on pinworm-infected children could be confounded by late onset effects of enterobiasis. A larger prospective cohort study with a longer follow-up on gut microbiomes will help to determine more exactly the duration and dynamics of the change in microbiota and in SIgA levels after mebendazole deworming treatment. Host genetics and diet are confounding factors for long-term follow up. Variations in the human genome have been found to favor the colonization of different gut microbiota [46]. A high fat and low fiber diet has been shown to be associated with reduced beneficial microbes producing short chain fatty acids (SCFAs) and thus such a diet increases the risk of inflammatory and autoimmune diseases [47]. In addition, our results revealed that recent respiratory tract infection with oral medication usage in the study population may be inversely correlated with intestinal microbial diversity and a decreased relative abundance of Fusobacterium and Acidaminococcus, which might interfere with the effect of pinworm infection. Thus, differential host genetics, lifestyles, respiratory tract infection rates and medication usage can all contribute to the inconsistent association of enterobiasis and the risk of inflammatory diseases observed in pediatric cohorts of various countries.
In conclusion, Enterobius vermicularis infections are associated with increased intestinal microbial diversity, and decreased gut SIgA levels. Several bacterial taxa exhibited differential abundance in pinworm (-), pinworm (+) mebendazole (-), and pinworm (+) mebendazole (+) groups. Mebendazole deworming was correlated with increased intestinal SIgA level and a higher proportion of probiotic bacteria in half of the infected subjects. To better understand the causal relationships of pinworm infection and mebendazole treatment on gut microbial composition and hosts' immune responses, more experiments including animal studies are needed.