Efficient and reproducible experimental infections of rats with Blastocystis spp.

Although Blastocystis spp. infect probably more than 1 billion people worldwide, their clinical significance is still controversial and their pathophysiology remains poorly understood. In this study, we describe a protocol for an efficient and reproducible model of chronic infection in rats, laying the groundwork for future work to evaluate the pathogenic potential of this parasite. In our experimental conditions, we were unable to infect rats using vacuolar forms of an axenically cultivated ST4 isolate, but we successfully established chronic infections of 4 week-old rats after oral administration of both ST3 and ST4 purified cysts isolated from human stool samples. The infection protocol was also applied to 4 week-old C57BL/9, BALB/C and C3H mice, but any mouse was found to be infected by Blastocystis. Minimal cyst inoculum required for rat infection was higher with ST3 (105) than with ST4 (102). These results were confirmed by co-housing experiments highlighting a higher contagious potential of ST4 in rats compared to ST3. Finally, experiments mimicking fecal microbiota transfer from infected to healthy animals showed that Blastocystis spp. could easily infect a new host, even though its intestinal microbiota is not disturbed. In conclusion, our results provide a well-documented and robust rat model of Blastocystis chronic infection, reproducing “natural” infection. This model will be of great interest to study host parasite interactions and to better evaluate clinical significance of Blastocystis.


Introduction
Blastocystis spp. are anaerobic enteric protist found in the intestinal tract of a wide range of animals, and are probably the most prevalent human parasites [1]. Four morphological stages of Blastocystis spp. have been described including amoeboid, granular, vacuolar and cystic stages. Both vacuolar and cystic forms are commonly found in human fecal samples and cyst is a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 considered to be the infectious stage [1][2][3]. Blastocystis spp. have been classified into 17 subtypes (ST) based on nuclear small subunit (SSU) ribosomal RNA-encoding gene, the ST1 to ST9 and ST12 being recovered from human [4][5][6][7]. ST3 is the most frequent subtype in human, followed by ST1, ST2 and ST4 [5]. However, this distribution depends on geographic areas, and a higher prevalence of ST4 was reported from the north of Europe [8][9][10].
Blastocystis spp. infection occurs in both immunocompetent and immunocompromised hosts and recent studies highlighted higher prevalence of these parasites in patients with Irritable Bowel Syndrome (IBS) and colorectal cancer [11][12][13][14]. However, the involvement of Blastocystis spp. in human diseases is highly debated and well-designed epidemiological studies still need to be performed. Animal experimental models for Blastocystis infection have been purposed for proving Koch's postulates, but few animal-based studies have been published so far, suggesting difficulties in the establishment and reproducibility of these models [15]. To date various animal models were used for Blastocystis infection as rodents (rats or mice), guinea pigs or chickens. Moreover, studies have shown that both vacuolar and cystic stages of different STs can efficiently infect animals [16][17][18][19]. The heterogeneity of infection methods with various animal models and/or different STs inoculation confirm that the establishment of a standardized model is necessary.
In this study, we evaluated several methods to inoculate Blastocystis spp. into laboratory animals (mice and rats) using different parasitic stages (vacuolar and cystic forms) isolated from in vitro cultures or from human or animal feces. Interestingly, we succeeded in the development of a reproducible model of chronic infection in juvenile and adult Wistar rats using purified cysts isolated from human stools.

Animals
All animals were housed in animal biosafety level 2 (21-22˚C, 12:12 h light-dark cycle) with access to food and water ad libitum. Three-and ten-week-old specific-pathogen-free (SPF) Wistar male rats and three-week-old SPF C57BL/9, BALB/C and C3H mice were purchased from Charles River Lab (Saint Germain Nuelles, France). All experiments were performed according to the ethical guidelines set out in the Guide for the Care and Use of Laboratory Animals and with approval of the "Comité d'Ethique pour l'Expérimentation Animale Auvergne" (C2E2A), the local ethics committee (Reference number: EU0116-3003). After experimental infections with Blastocystis, fecal samples were collected every week until sacrifice.

Blastocystis ST4 axenic cultures
Blastocystis ST4 WR1 strain isolated from Wistar rats and axenized by Chen et al. [20] was cultured anaerobically in Iscove's Modified Dulbecco's Medium (IMDM) (Gibco, Life Technologies, Carlsbad, CA, USA) with 10% horse serum (PAA, Pasching, Austria) and 1% penicillin/ streptomycin (Gibco, Life Technologies, Carlsbad, CA, USA) at 37˚C. Vacuolar forms were collected from 48h-old cultures by centrifugation at 1000 g for 10 min and suspended in sterile PBS to reach a concentration of 10 7 parasites per milliliter and were used for experimental infections.

ST4 strain
In a first experiment, four-week-old Wistar male rats were orally inoculated with 10 7 ST4 vacuolar forms per rat from in vitro axenic culture (n = 14). Eight rats were used to evaluate kinetic of ST4 route through the gastrointestinal tract and six rats were kept to monitor the progression of infection over time. To reduce gastric acidity, rats received 3 oral administrations every 2 hours of cimetidine (50 mg/kg, Mylan, Canonsburg, PA, USA) (n = 4) or 0.2 M sodium bicarbonate (Cooper, Melun, France) (n = 4) before oral inoculation with 10 7 vacuolar forms of ST4 in a second experiment. Similarly as before, four rats were used to evaluate kinetic of ST4 route through the gastrointestinal tract and four rats were kept to monitor the progression of infection over time.

Kinetic of ST4 route through the gastrointestinal tract of experimentallyinfected rats
After oral administration of ST4, rats were euthanized at 3 (n = 2), 6 (n = 2), 12 (n = 2) or 24 h (n = 2) post-inoculation (PI). Stomach content was collected at 3 h PI, stomach and duodenum contents at 6 h PI, ileum and caecum contents at 12 h PI, and caecum, proximal and distal colon contents at 24 h PI.
For the experiments with animals pretreated with cimetidine or sodium bicarbonate, euthanasia of rats was done at 3 or 6 h (n = 2 for each time and condition). Stomach contents were collected at 3 h PI and caecum contents at 6 h PI.

Blastocystis isolates collected from human stools
Four asymptomatic human carriers of Blastocystis ST2, ST3, ST4 and ST7 were identified at the medical laboratory of Parasitology of the Clermont-Ferrand teaching hospital (France). Both Blastocystis cysts and vacuolar forms were observed in stools for patients with ST2, ST3 and ST4, whereas only vacuolar forms were observed for the ST7 carrier. Fresh stools were collected for animal infections (approved by the research ethics committees of the Clermont-Ferrand Hospital, "Comité de Protection des Personnes Sud-Est 6", France, agreement ref: 2014/ CE29). Subtyping was performed by SSU rDNA sequencing as previously described [10,21]. Sequences were analyzed using the Basic Local Alignment Search Tool (BLAST; http://blast. ncbi.nlm.nih.gov/). washes in distilled water were performed by centrifugation at 1700 g for 10 min. The pellet was suspended in 0.5 ml of distilled water and mixed with 7 ml Percoll solution (d = 1.13) (MP Biomedicals, Illkirch, France). Then, 6 ml of distilled water were added, followed by a centrifugation at 1200 g for 10 min. The interface between distilled water and Percoll solution was collected, adjusted to 15 ml with distilled water and centrifuged at 1700 g for 10 minutes. This step using Percoll gradients was performed twice. Pellets containing purified cysts were suspended in 0.5 ml of an antibiotic cocktail containing 0.01 g/L Vancomycin (Sandoz, Levallois-Perret, France), 0.1 g/L Amoxicillin-clavulanic acid (Sandoz, Levallois-Perret, France) and 0.1 g/L Cefotaxim (SteriMax, Oakville, ON, Canada) diluted in sterile PBS (Gibco, Life Technologies, Carlsbad, CA, USA) and incubated for 6 h at 4˚C. Cysts were washed twice in sterile PBS by centrifugation at 300 g for 5 min and finally suspended in sterile PBS. Ten microliters of the suspension were plated on LB media and incubated at 37˚C for 24 h in order to quantify the number of Colony Forming Units (CFU). Cysts were counted in Malassez chamber and suspension was diluted in sterile PBS to obtain concentrations of 10 2 , 10 3 ,10 4 or 10 5 cysts per milliliter.

Oral administration of purified cysts isolated from human stools
Wistar male rats (four or twelve-week-old) were orally inoculated with 10 2 /rat to 10 5 /rat ST2, ST3 or ST4 purified cysts isolated from human stools (n = 5 for each condition). Blastocystis isolated from rat feces after infection were subtyped by sequencing the SSU rRNA encoding gene as described above. For mice infections, 10 5 ST4 purified cysts per animal were orally inoculated to four-week-old C57BL/6 (n = 3), BALB/C (n = 3) or C3H mice (n = 3).

Co-housing experiments
Four-week-old Wistar male rats were orally inoculated with 10 5 ST3 (n = 2) or ST4 (n = 2) purified cysts per animal, as described above. Four-weeks PI, these animals were co-housed for 6 weeks with eight-week-old non-infected rats (n = 4 per ST).

Oral administration of rat feces containing Blastocystis cysts
Fresh feces from experimentally-infected rats, containing ST3 or ST4 cysts, were collected. Almost 300 mg of feces samples were pooled and suspended in 3 ml of sterile PBS. Then, eight-week-old rats (n = 3, for each ST) were orally inoculated with 10 5 cysts. Method of cyst purification from human stools. About 50 g of stools were suspended in 200 ml of distilled water and filtered (1mm) to remove larger particles. After washing by centrifugation in distilled water (1700 g for 10 min), cysts were concentrated by sucrose gradient (d = 1.20) followed by two Percoll gradients (d = 1.13). Then, cysts were incubated with an antibiotic cocktail composed of Vancomycin, Amoxicilin-clavunalic acid and Cefotaxim to eliminate bacteria. After washing to remove antibiotics, purified cysts were suspended in sterile PBS and quantified. This method was adapted from Yoshikawa's protocol [18].

Fecal microbiota transplantation (FMT) using Blastocystis-positive human stools
Fresh stools from infected humans, containing ST2 or ST3 or ST4 cysts, or ST7 vacuolar forms, were collected. Almost 30 g of each fecal samples were suspended in 90 ml of sterile PBS. Moreover, almost 30 g of ST4 fecal sample were cryopreserved in 90 ml of PBS/glycerol 10% and stored at -80˚C for two months. Four-week-old Wistar male rats were orally inoculated with 1 ml (corresponding to 300 mg of Human stools) of the suspension containing either ST2 (n = 3), ST3 (n = 3), ST4 with or without cryopreservation (n = 3 for each condition) cysts or ST7 (n = 3) vacuolar forms.

Xenic cultures of Blastocystis-positive animal or human feces
Human or rat feces were cultured in Jone's medium at 37˚C in anaerobic chambers as previously described [22]. After 48 h of incubation, the culture was observed by standard light microscopy (x 400). Identification of vacuolar forms revealed the presence of viable parasites.

Immunofluorescence staining of intestinal sections
Four weeks PI, ST4-or ST3-infected rats were euthanized and the intestine was prepared as "Swiss rolls" of 4 cm and fixed by incubating in PFA 4% during 24 h. Then, the rolls were transferred in sucrose 30% during 24-48 h at 4˚C. Finally, the rolls were included in OCT compound (CellPath, Newton, United Kingdom). Sections of OCT-embedded intestine were incubated with 5% milk in PBS for 1 h, and then washed in 0.1% Triton-X100 in PBS. Anti-Blastocystis mouse polyclonal antibodies (diluted 1:200) were added for 1 h at RT, washed, and incubated further 1 h at RT with AlexaFluor 546-conjugated secondary antibody (diluted 1:1000) (antimouse IgG, Thermo Fisher Scientific, Waltham, MS, USA). Sections were washed and then incubated with fluorescein phalloidin (Thermo Fisher Scientific, Waltham, MS, USA) diluted 1:100 in PBS during 20 min. Finally, sections were washed and stained with DAPI. For each ST, six sections of duodenum, jejunum, ileum, caecum and colon from three rats were observed.

Transmission electron microscopy (TEM)
Samples were washed in 0.2 M Sodium Cacodylate buffer and fixed overnight at 4˚C with a mixture of 2% glutaraldehyde and 0,5% PAF in 0,2 M Sodium Cacodylate buffer. Specimens were then washed three-times in 0.2 M Sodium Cacodylate buffer, post-fixed 1 h with 1% OsO4 in 0.2 M Sodium Cacodylate buffer, and washed three-times (10 min) in 0.2 M Sodium Cacodylate buffer. Specimens were then dehydrated in a graded ethanol and acetone solution. Subsequently, they were infiltrated with acetone and EPON resin mixture (2:1) for 1 h, with acetone and EPON resin mixture (1:1) for 1 h, and with acetone and EPON resin mixture (1:2) for 1 h. Specimens were embedded in resin overnight at RT, and cured for 2 days in a 60˚C oven. Thin sections (70 nm) were cut using a UC6 ultramicrotome (Leica, Wetzlar, Germany) and stained with uranyl acetate and lead citrate. Carbone was evaporated using CE6500 unit. Specimens sections were observed at 80 kV with a Hitachi H-7650 TEM and a camera Hamamatsu AMT40.
Electron microscopy preparations were all performed by the "Centre d'Imagerie Cellulaire Santé" (Clermont-Ferrand, France). All chemical products were from Electron Microscopy Science, and distributed in France by Delta Microscopies.

Inoculation of rats with vacuolar forms of in vitro-cultivated Blastocystis ST4
In all our tested conditions, oral infections with vacuolar forms of Blastocystis ST4 axenic cultures failed (Table 1) even in animals pretreated with molecules which increase intragastric pH (cimetidine 50 mg/kg or sodium bicarbonate 0.2 M). To confirm the absence of parasites, we performed a follow-up of the parasite course through the intestinal tract of rats from 3 h to 24 h post-infection (PI) by xenic culture and qPCR. Even though high parasite loads were inoculated (10 7 per animal), xenic culure and qPCR were all negative suggesting that vacuolar forms are not able to pass through the stomach alive (Table 2). After treatment with cimetidine or sodium bicarbonate, qPCR analysis revealed the presence of Blastocystis DNA in the caecum of rats 6 h PI but no viable parasitic form was observed after xenic cultures from collected caecum contents ( Table 2). To avoid stomach passage, ST4 vacuolar forms were directly inoculated into the caecum of rats, but here again all animals remained uninfected (Table 1). These data suggest that vacuolar forms are not able to infect rats in our experimental conditions.

Infection of mice and rats with purified cysts isolated from human stool samples
We optimized a purification protocol described previously by adding another Percoll gradient step and a treatment with an antibiotic cocktail (Fig 1). This protocol decreased drastically the number of bacteria below 250 CFUs inoculated per animal. From 50 g of human stool about 4 million cysts were purified for each subtype (Fig 2A). The size of purified cysts ranged from 2 to 7 μm and cysts were characterized by the presence of two or four nuclei (Fig 2B). Both juvenile (four-week-old) and adult (twelve-week-old) rats were successfully infected by oral inoculation of 10 5 cysts of Blastocystis ST2, ST3 and ST4 per animal ( Table 1). Cultures of feces collected 8 weeks after oral inoculation of cysts were positive for Blastocystis, confirming a chronic infection by the parasite.
We then evaluated the minimal inoculum for both ST3 and ST4 by using infectious dose ranging from 10 2 to 10 5 purified cysts per animal. Interestingly, rats were chronically infected with the lowest cyst inoculum of ST4 (10 2 ), whereas higher inoculum of ST3 (10 5 ) was required for the establishment of infection (Table 1).
During all experiments, only cystic forms were observed in feces from infected rats for both Blastocystis subtypes. Indeed excreted cysts in rat feces were characterized by light microscopy and immunofluorescence labeling using polyclonal antibodies. Cysts from ST3 and ST4 infected rats were similar in size to those isolated from human stools and were also characterized by the presence of two to four nuclei (S1 Fig).
The same infection protocol was further applied to 3 different genetic mice backgrounds, C57BL/6, BALB/c and C3H mice (Table 1). However, we were not able to obtain any infection, suggesting that immunocompetent mice could be resistant to infection by these ST4 Blastocystis strains.

Transmission between animals by co-housing experiments or oral inoculation
Eight-week-old ST3 (n = 2) or ST4 (n = 2) infected rats were co-housed with eight-week-old naive rats (n = 4 per subtype) during six weeks. Infections were monitored by microscopy  (Table 1). Naive eight-week-old rats (n = 3 for each subtype) were also orally inoculated with feces from ST3 or ST4 infected rats. Only ST4-inoculated rats were infected (Table 1).

Fecal microbiota transplantation from human to rats
Four-week-old rats were orally inoculated with fresh human stools containing Blastocystis ST2, ST3, ST4 or ST7, mimicking fecal microbiota transplantation. Both cystic and vacuolar forms were present in ST2, ST3 and ST4 human positive stools. In contrast, only vacuolar forms were observed in ST7-positive stools. Infection monitoring was done following FMT by xenic cultures of rat feces. Vacuolar forms were detected in culture feces from ST2, ST3 and ST4 infected rats but no viable parasite was identified in ST7-inoculated rat feces (Table 1). Interestingly, we also performed FMT from Human to rat after cryopreservation of Blastocystis ST4 stools, by following the recommended protocol for Human to Human FMT [23]. After 2 months at -80˚C, stools were thaw out and transferred to naive rats. All animals were infected, suggesting that the procedure of cryopreservation use for Human to Human FMT also maintain Blastocystis alive.

Colonization of the intestinal tract by Blastocystis
Both ST3-and ST4-infected rats by purified cysts from human stools were sacrificed four weeks PI. Full gastrointestinal tract was collected and "Swiss rolls" were performed on duodenum, jejunum, ileum, caecum, proximal and distal colon. Blastocystis vacuolar forms were observed from duodenum to distal colon for both ST3 and ST4 (S2 Fig). Immunofluorescence labeling revealed that parasites were mainly localized in the intestine lumen (Fig 3A), and sometime in close contact with epithelial cells (Fig 3B). No invasive form was observed. Transmission electron microscopy performed on colonic sections confirmed the presence of parasites in the lumen (Fig 3C) and in close contact with epithelial cells (Fig 3D).
In the present study, we aimed to provide a robust and well-described protocol to obtain efficient and reproducible experimental infection with Blastocystis spp. Four different human Blastocystis STs were used in our experiments, including ST3, the most common ST found in human, followed by ST2 and ST4, the latter reported to be highly prevalent in Europe [8][9][10], and ST7 whose whole genome was the first sequenced [33,34]. ST4 was also used because it is the most prevalent ST found in rodents and its whole genome was recently sequenced [35]. Parasites used in our study originated from axenic cultures (ST4-WR1, first isolated from a laboratory rat) [20] or were isolated from human stools by a highly efficient purification process (ST2, ST3 and ST4) adapted from Yoshikawa et al [18].
Based on the literature and the easiest models to handle in an animal facility, we focused on common rodent models with rats and mice. We were not able to infect rats with the axenic ST4 WR1 strain. In our in vitro culture conditions that strain only produced vacuolar forms.
Then, failures in animal infections could be explained by sensitivity of vacuolar forms to gastric pH, but also to oxygen exposure during inoculum preparation [1,36]. Indeed, studies have shown that rat stomach pH ranged from 3.2-3.9 [37]. However, we also failed to infect animals with ST4 WR1 by increasing the intragastric pH or after intracaecal injection of parasites. Then, we hypothesized that long-term culture and/or axenization of vacuolar forms may have attenuated the infectious potential of the ST4 WR1 strain. Indeed, the maintenance of the parasites as Trypanosoma cruzi, Entamoeba histolytica or Leishamania infantum in laboratory cultures led to gene expression changes and decreased infectivity [38][39][40][41]. However, previous studies have reported successful infections of rats using axenic Blastocystis strains [16,17,19]. We cannot exclude that some axenic strains may conserve their infectious potential, or may produce cysts in axenic culture conditions. Indeed, cyst is considered to be the main infectious stage, responsible for feco-oral contamination between hosts [18,24,25,27,28,42]. Different protocols have been described for cyst isolation from human stool samples [17,18]. We optimized Yoshikawa's protocol in order to decrease cyst-associated bacteria [18]. Then, an additional Percoll step was performed and purified cysts were incubated with a wide spectrum antibiotic cocktail targeting both aerobic and anaerobic bacteria. Moreover, successful infections following this treatment demonstrated low impact of this antibiotic treatment on Blastocystis cyst viability.
We were able to infect rats with ST2, ST3 and ST4 cysts by using high parasite load (10 5 cysts/animal). Infected animals were followed as long as 8 weeks post-infection. Animals excreted Blastocystis cysts until the end of experiments, confirming the establishment of a chronic infection in our model.
Our results reinforced that ST4 infection failures were not related to a resistance mechanism, but more likely to the stage used or long-term in vitro cultivation. The susceptibility of animals to Blastocystis was reported to be age-dependent by Moe et al., but these authors didn't mentioned the Blastocystis ST used in their experiments [17]. In our study, we were also able to infect twelve-old-week rats with Blastocystis ST4 cysts, suggesting that at least for this ST, age is not a limiting factor. However, the effects of aging on susceptibility to ST3 infection remain to be investigated. Interestingly, rats were infected with as low as 10 2 ST4 cysts per animal, whereas the minimal inoculum dose required for infection with ST3 reached 10 5 cysts suggesting host adaptation. Transmission between animal by co-housing experiments and oral inoculations confirm our hypothesis. Transmission between animal by co-housing experiments and oral inoculations confirm our hypothesis. These results may explain why ST4 is the more prevalent than ST3 in rodents [43]. However, oral inoculation with less than 10 2 ST4 cysts per rat remains to be investigated, but a previous work has shown that infection efficiency with 10 cysts of ST4 RN94-9 or ST4 NIH:1295:1 strains varies between 20-100% [18].
Our results suggest that even though host barrier is not critical for Blastocystis infection, some STs are more adapted to particular hosts. Then, we applied our purification protocol to infect 3 strains of mice. The first one was juvenile BALB/c mice that have been shown to be more susceptible to parasite infections [44,45] such as Blastocystis [17,19]. We also used C3H mice that have been described to be susceptible to Entamoeba histolytica [46] and Giardia intestinalis [47]. Finally, C57BL/6 mice, being the most used genetic background for transgenic mice, were also used. However, we were not able to infect mice even with high ST4 cyst inoculum (10 5 /animal), suggesting that mice would be resistant to the ST4 strain used in our study. A recent study has shown that the Dextran Sodium Sulfate (DSS) treatment result in biophysical changes in mucus layer (increased penetrability to microorganism), increasing susceptibility of mice to Blastocystis ST7 colonization for at least 3 days after intra-ceacal injection [48]. These results suggest that mucus layer play in important role in the resistance of mice for Blastocystis persistence.
Immunofluorescence labelling revealed the presence of the parasites (ST4 or ST3) all along the intestinal tract of infected animals. Blastocystis were detected in the intestinal lumen and in close contact with epithelial cells. Granular forms presenting empty vacuoles or vacuoles containing electron-dense granules as described previously by Tan [1], were observed in direct contact with epithelial cells.
Moreover, FMT experiments highlighted the capacity of Blastocystis to overcome barrier microflora and infect a new host, even though its microbiota is not altered. We also demonstrated that cryopreservation procedure used for the stool storage before Human to Human FMT keep Blastocystis alive and able to infect a new host [23]. These data support the recent recommendations for the screening of fecal donors by confirming the ability of Blastocystis to be directly transmitted through stools [23].
In conclusion, our work provides a well-documented and reproducible animal model of Blastocystis chronic infection, reproducing "natural" infection. This model will be of great interest to decipher the host-parasite-microbiota interactions, and to better evaluate clinical significance of Blastocystis.
Supporting information S1 Fig. Blastocystis ST4 cysts from experimentally infected rats. (A) Cysts (red arrow) from rats observed by light microscopy. As purified cysts from human stools, size ranged from 2 to 7 μm. (B) Cysts from rats observed by immunofluorescence after labeling with mouse polyclonal anti-Blastocystis ST4 antibodies (Green) and DAPI staining (Blue). Cysts contained two to four nuclei (white arrows, two on this picture). (TIF)