Characterization of a Murine Model for Encephalitozoon hellem Infection after Dexamethasone Immunosuppression

Background: Encephalitozoon hellem (E. hellem) belongs to a group of opportunistic pathogens called microsporidia. Microsporidia infection symptoms vary and include diarrhea, ocular disorders and systemic inflammations. Traditionally, immunodeficient animals were used to study microsporidia infection. To overcome the difficulties in maintenance and operation using immunodeficient mice, and to better mimic natural occurring microsporidia infection, this study aims to develop a pharmacologically immunosuppressed murine model of E. hellem infection. Methods: Wild-type C57BL/6 mice were immunosuppressed with dexamethasone (Dex) and then E. hellem spores were inoculated into the mice intraperitoneally. Control groups were the Dex-immunosuppressed but noninoculated mice, and the Dex-immunosuppressed then lipopolysaccharide (LPS)-treated mice. Mice body weights were monitored and all animals were sacrificed at the 15th day after inoculation. Tissue fragments and immune cells were collected and processed. Results: Histopathological analysis demonstrated that E. hellem inoculation resulted in a disseminated nonlethal infection. Interestingly, E. hellem infection desensitized the innate immunity of the host, as shown by cytokine expressions and dendritic cell maturation. We also found that E. hellem infection greatly altered the composition of host gut microbiota. Conclusions: Dex-immunosuppressed mice provide a useful tool for study microsporidiosis and the interactions between microsporidia and host immunity.


Introduction
Microsporidia are a group of intracellular pathogens causing opportunistic infections [1,2]. The most common species for mammals are Encephalitozoon hellem (E. hellem), Encephalitozoon cuniculi (E. cuniculi), Enterocytozoon bieneusi (E. bieneusi), and Encephalitozoon intestinalis (E. intestinalis) [3][4][5]. The infection outcomes and clinical symptoms are associated with pathogen species and hosts immune-state. Usually, in immunocompromised hosts the symptoms are more obvious including persistent diarrhea and even lethal outcomes [4,6,7]. In immunocompetent hosts the infections are usually asymptomatic, yet in immunocompetent rabbits, the clinical signs are shown after E. cuniculi infection [8][9][10]. So far, most of what is known about the host immune responses against microsporidia are from severe combined

Pathogen and Cell Cultures
E. hellem strain (ATCC 50504/50451) is gift of Prof. Louis Weiss (Albert Einstein College of Medicine, New York, NY, USA), and are reproduced in rabbit kidney cells (RK13, ATCC CCL-37). RK13 cells were cultured in 10% fetal bovine serum (FBS) (ThermoFisher, Waltham, MA, USA) containing Minimum Essential Medium Eagle (MEM) with penicillin-streptomycin at 5% CO 2 until confluent and then infected with E. hellem. The spores were collected from culture media, purified by passing them through 5 µm size filter (Millipore, Billerica, MA, USA) to remove host cells, concentrated by centrifugation, and stored in sterile distilled water at 4 • C [31]. Spores used in these experiments were counted with a hemocytometer (three times/sample and averaged).

Animals and Experimental Design
Groups of C57BL/6 mice (six weeks in age, female) were reared in animal care facility according to the Southwest-University-approved animal protocol (SYXK-2017-0019). Dexamethasone-induced immnosuppression was administered to all mice, and achieved by intraperitoneal injection of dexamethasone (Aladin, Cas 2392-39-4, Shanghai, China) at the dose of 5 mg/kg/day for six days, method adopted from previous studies [26,28,32]. The experimental group of mice were then infected with 1 × 10 7 E. hellem spores/day for two days; while the control groups of mice were then injected either with 1XPBS or lipopolysaccharide (LPS, 5 mg/kg). After these treatments, all mice were then housed for another 15 days and body weights were monitored, starting on the last day of dexamethasone-injection, then measured again on the day of mice sacrifice (end of experiment) for a total 18 days period of time. At the endpoint of the experiment, all mice were sacrificed by CO 2 inhalation. Samples of blood, feces and various organs were collected from all mice for further investigations. E. hellem-infected mice and uninfected mice organs (colon, spleen and liver) were collected and frozen sectioned respectively. All sections were then stained by hematoxylin and eosin (H&E staining). Plasma samples were prepared from the peripheral blood of three groups of mice.

Detection of E. hellem Spores
Colonization of E. hellem spores within host organ tissues were detected by fluorescence staining with calcofluor white (CFW). The colonization of E. hellem in host were further confirmed by PCR assay to show the existence in immune cells, the dendritic cells, using primers 5 -TGAGAAGTAAGATGTTTAGCA-3 and 5 -GTAAAAAGACTCTCACACTCA-3 [33].

Cytokines Expressions
Peripheral blood samples were collected from mice of all three groups, the uninfected control, E. hellem-infected and LPS-treated. Sodium citrate was used as anticoagulant, and blood samples were centrifuged at 1000× g for 10 min to enrich plasma. All plasma samples were then subjected to cytokine expression assay, using mouse Interleukin 12 ELISA kit and Mouse IL-6 ELISA Kit (ThermoFisher Scientific, Waltham, MA, USA).

Dendritic Cells (DCs) Isolation
Spleens were collected from mice of all three groups, the uninfected control, E. hellem-infected and LPS treated. Isolation of DCs starts with mince the spleen into homogenous paste using commercially available Spleen Dissociation Medium (STEMCELL Technologies, Vancouver, Canada) in tissue culture dish, followed by gently passing several times through a 16 Gauge Blunt-End Needle attached to a 3 cc Syringe and then through a primed 70 µm nylon mesh filter. The filtered single cell suspension was then subjected to the EasySep Mouse Pan-DC Enrichment Kit (STEMCELL Technologies, Vancouver, Canada) to isolate dendritic cells only. This kit targets non-DCs for removal with biotinylated antibodies recognizing specific cell surface marker on unwanted cells in solution. Briefly, Enrichment Cocktail and Biotin Selection Cocktail, combinations of monoclonal antibodies, were added to the sample. Next, the samples were incubated with Streptavidin-bound magnetic particles to solution and a magnet was used to immunomagnetic negative selection of the dendritic cells. The purified DCs were then counted and cultured in RPMI Medium 1640 (supplemented with 10% FBS, 50 ng/mL GM-CFS (granulocyte-macrophage colony-stimulating factor) and penicillin/streptomycin) (Gibco, Waltham, MA, USA) in a 37 • C, 5% CO 2 incubator.

Flow Cytometry Analysis
The isolated dendritic cells, either from uninfected control mice, E. hellem infected mice or LPS treated mice were all subjected to flow cytometer and analyzed for cell maturation. In brief, the DCs were firstly washed with 1× PBS/0.3% BSA, and then stained with FITC-conjugated anti-CD40 and PE-conjugated CD86 (BioLegend, San Diego, CA, USA) for 30 min at 4 • C. CD40+/CD86+ DCs were considered as more matured DCs. Flow cytometry analysis was performed by a FACSCanto II flow cytometer (BD Biosciences) and data were collected and analyzed with FACSDiva software (v6.1.2).
Taxonomy-based analyses were conducted using the Ribosomal Database Project Naive Bayes classifier, with a confidence of 95%. The remaining sequences with more than 97% similarity threshold were clustered into OTUs [37]. Each read was classified to phylum, class, order, family, and genus with the SILVA reference database. The relative abundance of species was calculated by dividing the number of sequences of each taxonomic level by the total number of sequences per sample [38].Microbiome diversity was estimated by alpha diversity (within communities) and beta diversity (between microbial communities) analyses. Alpha diversity evaluated species richness with ACE index and Chao 1 index, diversity with Shannon and Simpson indexes [39].

Statistical Analysis
Each experiment includes various numbers of samples (n), and each experiment has been repeated at least three times. Statistical analysis of results was conducted by using one-way ANOVA and Student's t-test to identify the differences between two groups, with p < 0.05 being considered a significant difference.

Colonization of E. hellem in Dexamethasone-Immunosuppressed Mice
Histological examination of mouse organs, such as liver and colon showed colonization of E. hellem spores (Figure 1), detected by CFW staining (Figure 1a). The following examination of isolated dendritic cells, by PCR using E. hellem specific primers, also confirmed the successful colonization of E. hellem within host body and immune cells ( Figure 1b). Together, these results showed that dexamethasone-immunosuppression facilitates the successful infection and colonization of E. hellem within the host. and internal transcribed spacer (primers 5′-CTTGGTCATTTAGAGGAAGTAA-3′ and 5′-TCCTCCGCTTATTGATATGC-3′) [34]. PCR reactions were performed on an ABI GeneAmp 9700 PCR system (Applied Biosystems, Waltham, MA, USA). The amplicons were purified and trimmed barcodes, the remaining sequential reads of each sample were linked by FLASH V1.2.7 [35,36]. Taxonomy-based analyses were conducted using the Ribosomal Database Project Naive Bayes classifier, with a confidence of 95%. The remaining sequences with more than 97% similarity threshold were clustered into OTUs [37]. Each read was classified to phylum, class, order, family, and genus with the SILVA reference database. The relative abundance of species was calculated by dividing the number of sequences of each taxonomic level by the total number of sequences per sample [38].Microbiome diversity was estimated by alpha diversity (within communities) and beta diversity (between microbial communities) analyses. Alpha diversity evaluated species richness with ACE index and Chao 1 index, diversity with Shannon and Simpson indexes [39].

Statistical Analysis
Each experiment includes various numbers of samples (n), and each experiment has been repeated at least three times. Statistical analysis of results was conducted by using one-way ANOVA and Student's t-test to identify the differences between two groups, with p < 0.05 being considered a significant difference.

Colonization of E. hellem in Dexamethasone-Immunosuppressed Mice
Histological examination of mouse organs, such as liver and colon showed colonization of E. hellem spores (Figure 1), detected by CFW staining (Figure 1a). The following examination of isolated dendritic cells, by PCR using E. hellem specific primers, also confirmed the successful colonization of E. hellem within host body and immune cells ( Figure 1b). Together, these results showed that dexamethasone-immunosuppression facilitates the successful infection and colonization of E. hellem within the host.

Effects of E. hellem Infection on Body Weight
The body weights of all three groups of mice were monitored. Compared to the LPS-treated mice, the uninfected control and E. hellem infected mice did not lose any weight ( Figure 2).

Histopathological Analysis of E. hellem-Infected Mice
Consistently with our previous results, no significant signs of inflammation or organ damages were observed in the E. hellem-infected group. Yet in few cases, around three in all eight E. hellem-infected mice, histopathological changes were seen (Figure 3). In the colon, enlarged microvilli were observed, indicating swelling or edema caused by local inflammation. In spleen and liver, lymphoepithelioid granulomas were seen showing signs of interstitial nephritis or infiltrating lymphocytes.

Cytokines Expressions Affected by E. hellem Infection
Inflammatory cytokines IL-6 and Il-12 expression levels were quantified using specific ELISA kits. The results showed that E. hellem-infection did not upregulate either IL-6 or IL-12 expression, as shown in Figure 4. These findings further demonstrate that our model mimics latent microsporidia infection which is consistent with our previous results.

Cytokines Expressions Affected by E. hellem Infection
Inflammatory cytokines IL-6 and Il-12 expression levels were quantified using specific ELISA kits. The results showed that E. hellem-infection did not upregulate either IL-6 or IL-12 expression, as shown in Figure 4. These findings further demonstrate that our model mimics latent microsporidia infection which is consistent with our previous results.

Cytokines Expressions Affected by E. hellem Infection
Inflammatory cytokines IL-6 and Il-12 expression levels were quantified using specific ELISA kits. The results showed that E. hellem-infection did not upregulate either IL-6 or IL-12 expression, as shown in Figure 4. These findings further demonstrate that our model mimics latent microsporidia infection which is consistent with our previous results.   confirmed that E. hellem infection did not upregulate IL-12 expression, although no statistical differences among groups (n.s = No significance, n = 8 for each group in one experiment, and the experiments have been repeated for three times).

Dendritic Cells Maturation Affected by E. hellem Infection
The matured DCs were determined by flow cytometry using fluorescent labeled anti-CD40 and anti-CD86 antibodies, and LPS-treated group were assigned as positive control. The amounts of CD40+/CD86+ double positive DCs were recorded and the ratios were calculated. As shown in Figure 5, E. hellem infection did not promote DCs maturation as compared to LPS treatment. This result again confirmed our previous data, that our model mimics latent microsporidia infection and this kind of infection is more likely to happen in nature.
erences among groups (n.s = No significance, n = 8 for each group in one experiment, and eriments have been repeated for three times).
ndritic Cells Maturation Affected by E. hellem Infection matured DCs were determined by flow cytometry using fluorescent labeled anti-CD 6 antibodies, and LPS-treated group were assigned as positive control. The amo D86+ double positive DCs were recorded and the ratios were calculated. As sh , E. hellem infection did not promote DCs maturation as compared to LPS treatme ain confirmed our previous data, that our model mimics latent microsporidia infect of infection is more likely to happen in nature. ure 5. Flow cytometry analysis of DCs maturation. DCs isolated from uninfected hellem-infected mice and LPS-treated mice were subjected to flow cytometry analy of CD40 + /CD86 + double positive DCs were recorded, and the ratios of uninfe ted group and E. hellem-infected to LPS-treated group were calculated and prese s shown, DCs from either E. hellem-infected or uninfected group have significan of matured cells. (* = p < 0.05, n = 8 for each group in one experiment, and the expe eated for three times).
ce Gut Microbiota Affected by E. hellem Infection ative abundance of microbiota of each group was compared on phylum and genus l n Figure 6. Our findings are as follows: although Shannon index showed no sig es between the E. hellem-infected and uninfected groups; the microorganisms' comp y the bacteria changed a lot after E. hellem infection. It's shown in the figure t Figure 5. Flow cytometry analysis of DCs maturation. DCs isolated from uninfected control mice, E. hellem-infected mice and LPS-treated mice were subjected to flow cytometry analysis. The amounts of CD40 + /CD86 + double positive DCs were recorded, and the ratios of uninfected to LPS-treated group and E. hellem-infected to LPS-treated group were calculated and presented in figure. As shown, DCs from either E. hellem-infected or uninfected group have significant lower numbers of matured cells. (* = p < 0.05, n = 8 for each group in one experiment, and the experiments have repeated for three times).

Mice Gut Microbiota Affected by E. hellem Infection
Relative abundance of microbiota of each group was compared on phylum and genus level, as shown in Figure 6. Our findings are as follows: although Shannon index showed no significant differences between the E. hellem-infected and uninfected groups; the microorganisms' composition, especially the bacteria changed a lot after E. hellem infection. It's shown in the figure that the abundance of Firmicutes and Verrucomicrobiota phylum turn to be more abundant after E. hellem infection. The abundances of Bacteroidota and Bacteroidetes phylum decrease after E. hellem infection. On the genus level, Akkermansia and Lactobacillus rise after E. hellem infection, while Turicibacter and Parasutterella decrease after E. hellem infection. Together, E. hellem infection greatly disturbs the host microbiota composition.

Discussion
In this study, we established and evaluated a murine model of microsporidia infection. E. hellem is the one of the most commonly found human-infecting microsporidia pathogens but relatively less studied on murine models [4]. Here in this study, we firstly administered dexamethasone into wild-type mice to facilitate microsporidia invasion, then E. hellem was inoculated and successfully colonized within various organ and immune cells. These results showed that dexamethasone treatment is suitable in establishing the murine model of E. hellem infection. Our results also showed that E. hellem infection is systemic yet nonlethal, and evokes limited levels of host immune response. We also found that E. hellem infection altered host gut microbiota. These results confirmed that E. hellem infection of the dexamethasone-immunosuppressed mice mimics the natural occurring or latent microsporidia infection, and provides a valuable murine model for future studies.
Traditionally, microsporidia is considered as opportunistic pathogens and microsporidia infects immunodeficient hosts [40,41]. However, accumulating evidence pointed out the potential of microsporidia infecting host naturally and even in immune-competent individuals [7,22]. These infections could be latent or symptomatic depends on the species and virulence of pathogens [9,10]. The 'living-together' strategy reflects a balanced pathogen-host relationship [42]. This inference is supported by our findings that E. hellem could successfully colonize within host but could not evoke significant organ damage or systemic inflammation. We believe current study mimics the latent infection model of E. hellem. It is interesting to notice the fact that latent microsporidiosis could develop to reactivation and redissemination [21,43]. Our study proves this inference in a certain way, as we showed that E. hellem infection may desensitize the host immunity, leads to less proinflammatory cytokines expressions and delayed DCs maturation. We also investigated other functionalities of DCs after E. hellem infection, including the surface marker expressions and signaling pathway changes (data now shown). Dexamethasone is a widely used anti-inflammatory drug, and may also alter the host gut microbiota composition and facilitate opportunist infections [44,45]. We used dexamethasone only to facilitate the successful infection or colonization of E. hellem in mice. As a matter of fact, we designed our study so that all mice, including the 'E. hellem-infected',

Discussion
In this study, we established and evaluated a murine model of microsporidia infection. E. hellem is the one of the most commonly found human-infecting microsporidia pathogens but relatively less studied on murine models [4]. Here in this study, we firstly administered dexamethasone into wild-type mice to facilitate microsporidia invasion, then E. hellem was inoculated and successfully colonized within various organ and immune cells. These results showed that dexamethasone treatment is suitable in establishing the murine model of E. hellem infection. Our results also showed that E. hellem infection is systemic yet nonlethal, and evokes limited levels of host immune response. We also found that E. hellem infection altered host gut microbiota. These results confirmed that E. hellem infection of the dexamethasone-immunosuppressed mice mimics the natural occurring or latent microsporidia infection, and provides a valuable murine model for future studies. Traditionally, microsporidia is considered as opportunistic pathogens and microsporidia infects immunodeficient hosts [40,41]. However, accumulating evidence pointed out the potential of microsporidia infecting host naturally and even in immune-competent individuals [7,22]. These infections could be latent or symptomatic depends on the species and virulence of pathogens [9,10]. The 'living-together' strategy reflects a balanced pathogen-host relationship [42]. This inference is supported by our findings that E. hellem could successfully colonize within host but could not evoke significant organ damage or systemic inflammation. We believe current study mimics the latent infection model of E. hellem. It is interesting to notice the fact that latent microsporidiosis could develop to reactivation and redissemination [21,43]. Our study proves this inference in a certain way, as we showed that E. hellem infection may desensitize the host immunity, leads to less proinflammatory cytokines expressions and delayed DCs maturation. We also investigated other functionalities of DCs after E. hellem infection, including the surface marker expressions and signaling pathway changes (data now shown). Dexamethasone is a widely used anti-inflammatory drug, and may also alter the host gut microbiota composition and facilitate opportunist infections [44,45]. We used dexamethasone only to facilitate the successful infection or colonization of E. hellem in mice. As a matter of fact, we designed our study so that all mice, including the 'E. hellem-infected', 'uninfected' and 'LPS-treated' mice, were all dexamethasone pretreated. In this way, we could exclude the effects of dexamethasone on host immune state, and only compare the differences before and after E. hellem infection and to better investigate the roles of E. hellem on the host. It has been reported that microsporidia infection of immunocompetent humans may cause a short acute diarrheal phase followed by asymptomatic infection, or chronic malabsorbtive diarrhea may develop [46]. In our study we found that E. hellem infection altered host gut microbiota, and these findings may help to explain the above phenomena. We found that after E. hellem infection, the relative abundances of Firmicutes and Bacteroidetes phyla changed significantly with the former increase while the later decrease. It is known that the intestinal microbiota consists of over 1000 various bacterial species, mainly belonging to Firmicutes and Bacteroidetes phyla, containing both beneficial and pathogenic microbes [47]. Thus the changes of these two phyla after E. hellem infection would significantly modify the host metabolism and immunity. In addition, we found that the relative abundance of Blautia phylum increased after E. hellem infection. A recent study pointed out the association between Blautia and Crohn's disease [48]. These findings may help to explain the acute or persistent diarrheal and malabosorbtion in microsporidia infected individuals. Our findings on genus levels also revealed multiple changed groups and the results are worthy of further investigation. For example, we found the relative abundance of genus Akkermansia increase after E. hellem infection. Akkermansia muciniphila is a rising star in the control of obesity and diabetes [49], thus it will be reasonable to infer that E. hellem infection altered the metabolism of host. In addition, the abundances of genus Parasutterella and Turicibacter were downregulated. The former is found to be core component of gut microbiota, and later one is important for host to make serotonin, an important neurotransmitter and chemical to reduce inflammation [50,51]. E. hellem infection caused decrease of these two components, may help to explain the dysbiosis of host microbiota and the desensitized immunity.

Conclusions
In conclusion, dexamethasone-induced immunosuppression facilitates E. hellem infection on wild-type mice. This murine infection model better mimics the naturally occurring and latent microsporidia infection. Our findings showed that this kind of infection is mostly asymptomatic yet E. hellem can successfully reproduce within the host body. We further demonstrated that E. hellem infection may desensitize host immunity and disturbed host gut microbiota. Together, our model provides a great tool to study microsporidia and host interactions.