Comparison of the Concentration of Encephalitozoon cuniculi Genotypes I and III in Inflammatory Foci Under Experimental Conditions

Introduction

Microsporidia are unicellular, intracellular, spore-forming, ubiquitous parasites belonging to the phylum Microsporidia found to infect a wide range of hosts, including humans. The best described microsporidian is Encephalitozoon cuniculi, which infect a wide range of vertebrates and invertebrates.¹ Currently, four genotypes of E. cuniculi have been determined based on molecular differences in the internal transcribed spacer (ITS) region of the rRNA genes.^2,3 While E. cuniculi genotype I has rarely been recorded in asymptomatic birds, rodents, rabbits, dogs, horses, boars, buffalos, great apes, and seldom in humans^4–18 and E. cuniculi genotype III has been identified only in lemmings, rabbits, birds, swine, dogs, blue foxes, snow leopard, tamarins and squirrel monkeys, and also in humans, causing a variety of pathologies ranging from asymptomatic to fatal infections,^4,19–34 Encephalitozoon cuniculi genotype II has been found in many different species of hosts causing a wide spectrum of pathogenesis.^35–40 The recently described genotype IV has been detected in cats, dogs, marmots, and in immunosuppressed humans,^3,41,42 causing disseminated encephalitozoonosis in the latter species. However, contrary to the first three genotypes, genotype IV is not available for in vitro culture and thus for laboratory experiments.

Encephalitozoon species infect several cell types in mammalian hosts, including epithelial and endothelial cells, fibroblasts, macrophages, and astrocytes⁴³ mainly in the gastrointestinal tract, kidney, brain, eyes, sinuses, and muscles.^{35,36,44–46} Recently, a possible involvement of immune cells connected with pro-inflammatory immune responses serving as vehicles transporting E. cuniculi genotype II purposefully across the whole host body towards inflammation was suggested, based on the occurrence of microsporidia in inflamed tissues,⁴⁷ and experimentally verified.⁴⁸

The present study was designed to compare the progress of concentration of Encephalitozoon cuniculi genotypes I and III in sites of inflammation with previously published data on E. cuniculi genotype II⁴⁸ and to answer the question of whether individual genotypes of E. cuniculi differ in dispersal within the host body following experimentally induced inflammation.

Materials and Methods

Ethics Statement

All experimental procedures complied with the law of the Czech Republic (Act No 246/1992 Coll., on the protection of animals against cruelty). The study design was approved by ethical committees at the Biology Centre of CAS, the State Veterinary Administration, and the Central Commission for Animal Welfare under protocol no. 35/2020.

Mice

Eight-week-old Severe Combined ImmunoDeficient (SCID) and immunocompetent (BALB/c) mice originally obtained from Charles River (Sulzfeld, Germany) were bred in plastic cages with sterilized wood-chip bedding situated in flexible film isolators (BEM Znojmo, Czech Republic) with HEPA filters. Mice were supplied with a sterilized diet (TOP-VELAZ Praha, Czech Republic) and sterilized water ad libitum.

Parasites

The spores of E. cuniculi genotype I (lagomorph subtype, ATCC^®-50503TM) originally isolated from rabbit, OH, USA in 1978 (ATCC^® Product Sheet), and spores of E. cuniculi genotype III originally isolated from pet rodents, Steppe lemmings (Lagurus lagurus), CZE in 2014³² were grown in vitro in Green monkey kidney cells (VERO, line E6) maintained in RPMI-1640 medium (Sigma-Aldrich, St. Louis, MO) supplemented with 2.5% heat inactivated fetal bovine serum. Spores were harvested and purified from cells by centrifugation over 50% Percoll (Sigma-Aldrich) at 1100×g for 30 min and washed three times in deionized water before storing in deionized water at 4°C supplemented with antibiotics (Sigma-Aldrich, 100 U/mL penicillin, 100 μg/mL streptomycin, and 2.5 μg/mL amphotericin B). The spores were washed with deionized water before use.

Experimental Protocol

Fifteen SCID and 27 BALB/c mice were per orally (P.O.) infected with the dose of 10⁷ either E. cuniculi genotype I or III spores in 0.2 mL of deionized water using intragastric gavage. The inflammation in experimental SCID and BALB/c was induced by the usage of 50 μL of Freund’s Incomplete Adjuvant (Sigma-Aldrich) inoculated in the acute phase of infection⁴⁹ to the hind limb muscle. Mice of both strains injected intramuscularly with 50 μL of sterile phosphate buffered solution (PBS) at the same intervals were used as negative controls. The animal’s health, mortality, and morbidity were recorded at 12-hour intervals.

Three mice from each group were euthanized by cervical dislocation every seventh day post-infection (DPI), and sterile samples for molecular detection and histology were prepared from limb muscles and liver by the usage of a different pair of sterile dissection tools.

DNA Isolation

Tissue samples were homogenized by bead disruption using a FastPrep^®–24 Instrument (MP Biomedicals, CA) and 0.5 mm glass beads (Biospec Products, Inc., Bartlesville, OK) at a speed of 5.5 m/s for 1 min. Total DNA was extracted using a commercially available column based isolation kit, DNeasy Blood & Tissue Kit (QIAGEN, Hilden, Germany), according to the manufacturer’s instructions. Extracted DNA was stored at −20°C.

PCR Amplification

Nested PCR protocols were used for amplification of a partial sequence of 16S rRNA, using microsporidia-specific primers.^50,51 DNA obtained from spores of E. cuniculi genotype II and ultrapure water was used as a positive and negative control, respectively. PCR products were evaluated by gel electrophoresis.

qRT-PCR

DNA samples from limb muscles and liver were processed using a quantitative RT-PCR protocol amplifying a 268-bp region of the 16S rRNA gene of E. cuniculi⁵² and a 137-bp region of murine β-actin as a housekeeping gene.⁵³ Each run included negative controls consisting of unspiked specimens and diluent blanks. The total amount of spores in 1 g of individual tissue samples was calculated based on the number of β-actin copies in the tissue sample based on a method routinely established in the laboratory.^54,55

Statistical Analysis

Differences in microsporidia presence in inflammatory foci compared to the non-induced site were analyzed by Student’s t-tests and the conformity of the variances of the tested groups was verified by an F-test at a confidence level of α≤0.05. All computations were made using Statistica 6.0 software (StatSoft CR, Praha, Czech Republic).

Results

The application of Freund’s Incomplete Adjuvant elicited a local inflammatory immune response in all animals characterized by the formation of a purulent inflammatory focus in the hind limb muscle. On the other hand, no macroscopic changes were observed in the PBS-inoculated mice.

Induction of Inflammation in SCID Mice

In immunodeficient SCID mice, E. cuniculi genotypes I and III were detected in all screened tissue already prior to induction of inflammation (Figure 1A and B) with a higher spore burden in E. cuniculi genotype III infected mice, reaching up to 5.2×10² spores per gram of tissue.

Figure 1 Maximum value of E. cuniculi spores per 1 gram of tissue (number in white fields), frequency and spore burden of E. cuniculi spores per gram of tissue (colored squares; each square represents one mouse), and infection trend illustration (red fields) in immunodeficient SCID mice infected either with E. cuniculi genotype I (A) or E. cuniculi genotype III (B) and induced in the acute phase of infection. aPeroral infection 107 spores of E. cuniculi in 200 µL dH2O and intramuscular injection of PBS; bperoral infection 107 spores of E. cuniculi in 200 µL dH2O and intramuscular injection of Freund’s Incomplete Adjuvant; grey column, intramuscular injection of all animals in the group into the right thigh muscle; *significant difference between right and left leg; #significant difference between control and experimental group; colored squares, positive capture of microsporidia in the sample according to the quantity scale.

In the control group of immunodeficient SCID mice infected with E. cuniculi genotype I (Figure 1A) and with no inflammation induction (injected with PBS only), no effect was observed and the number of spores increased continuously till the end of experiment 35 days post-infection (DPI), reaching up to 1.7×10⁴ spores per gram of tissue (Figure 1A).

After induction of inflammation of 7 DPI in the experimental group, in the non-induced leg, the spore burden was similar to the control group; however, a significant (p<0.05) increase in the number of spores in the induced right hind leg was observed, starting at 14 DPI and reaching up to 2.4×10⁵ spores per gram of tissue at 35 DPI, compared to the left hind leg, containing maximally 3.1×10⁴ spores per gram of tissue at 35 DPI. In addition, there was a significant difference (p<0.05) in spore burden in induced animals compared to the control mice, and the infection trend was more intense in induced animals (Figure 1A).

A similar situation was observed in SCID mice infected with E. cuniculi genotype III. There was no difference within control groups (Figure 1B), except that the spore burden in individual tissues was higher compared to E. cuniculi genotype I, reaching up to 4.7×10⁶ spores per gram of tissue at 35 DPI.

In the experimental group infected with E. cuniculi genotype III with induced inflammation, there was a one order shift toward the induced site starting at 14 DPI reaching 2.7×10², 2.1×10⁴, 2.1×10⁶, and 1.0×10⁷ spores per gram of tissue, compared to 1.5×10¹, 2.2×10³, 1.5×10⁵, and 1.3×10⁶ spores per gram of tissue in the control group (Figure 1B). Surprisingly, a significant decrease in spore burden was observed in the non-induced site at 14 DPI (eg, 1.5×10⁶ spores per gram of tissue). Moreover, there was a significant difference (p<0.05) in spore burden in induced animals compared to the control mice starting from 21 DPI (Figure 1B). The infection trend was significantly more intense and reached a higher spore burden in the induced site compared to SCID mice infected with E. cuniculi genotype I (Figure 1A).

Induction of Inflammation in BALB/c Mice

Also in immunocompetent BALB/c mice were E. cuniculi genotypes I and III, detected prior to induction of inflammation in all screened tissue samples at least in one animal per group (Figure 2A and B).

Figure 2 Maximum value of E. cuniculi spores per 1 gram of tissue (number in white fields), frequency and spore burden of E. cuniculi spores per gram of tissue (colored squares; each square represents one mouse), and infection trend illustration (red fields) in immunocompetent BALB/c mice infected either with E. cuniculi genotype I (A) or E. cuniculi genotype III (B) and induced in the acute phase of infection. aPeroral infection 107 spores of E. cuniculi in 200 µL dH2O and intramuscular injection of PBS; bperoral infection 107 spores of E. cuniculi in 200 µL dH2O and intramuscular injection of Freund’s Incomplete Adjuvant; grey column, intramuscular injection of all animals in the group into the right thigh muscle; *significant difference between right and left leg; #significant difference between control and experimental group; colored squares, positive capture of microsporidia in the sample according to the quantity scale.

In the control group of immunocompetent BALB/c mice infected with E. cuniculi genotypes I and III, no differences in microsporidia distribution were observed and the number of spores was comparable in both legs, increased at the beginning of the infection, reaching up to 2.3×10² and 9.9×10¹ spores per 1 g of tissue, respectively, and followed by continuous disappearance and clearance of microsporidia in latter weeks of infection from muscles, namely 42 DPI (E. cuniculi genotype I) and 49 DPI (E. cuniculi genotype III) (Figure 2A and B).

The induction of inflammation in immunocompetent BALB/c mice infected with E. cuniculi genotype I during the acute phase of infection (7 DPI) resulted in the disappearance of microsporidia from the non-induced left hind leg and an increase of their concentration toward the induced right hind leg in all experimental mice (p<0.05) starting from 14 DPI (Figure 2A), reaching up to 5.3×10² spores per 1 g of tissue reported 2 weeks post-inflammation induction. Moreover, the inflammation induction resulted in a significant increase (p<0.05) in concentration of spores in the induced site in the experimental group 35 DPI and prolonged persistence of microsporidia till 56 DPI in the induced leg compared to the control group (Figure 2A).

One order shift towards the induced site was also reported in immunocompetent BALB/c mice infected with E. cuniculi genotype III with induced inflammation, starting at 14 DPI and reaching the maximum of 7.9×10¹ spores per gram of tissue at 14 DPI. The persistence of microsporidia was prolonged for 2 weeks in the induced right hind leg compared to the control group (Figure 2B).

Discussion

Of the several species of microsporidia that infect humans, E. cuniculi belongs to the most common.³⁹ Horizontally transmitted environmentally resistant life cycle stage spores access host tissues by germinating in the gut lumen, where infections are typically initiated in gut epithelial cells, and sometimes muscle cells. However, the target tissues – those in which microsporidia develop to the infectious stage, represent a species-specific interaction with its host. As E. cuniculi lacks the organ/cell specificity, it can infect a wide spectrum of host cells including epithelial cells, vascular endothelial cells, kidney tubule cells, and possibly other cell types, and can be found in most tissues, with a predilection, in mammals, for the brain and kidney.^49,55–58 However, microsporidian spores are non-motile and short distance host dissemination is limited to a unique mechanism of host cell invasion using 10–50 μm long polar tubes that deliver this organism into the host cell. As Encephalitozoon species are able to survive and replicate in a variety of immune cells, including resident and migratory macrophages and other phagocytic cells such as neutrophils, and eosinophils, monocytes, and dendritic cells, which may contribute to the dissemination of E. cuniculi throughout the host organism,^59,60 suggesting that induction of chemokines for inducing innate immune inflammation may also promote recruitment of host cells for continued infection and dissemination.^61–63 Recently, a role for the involvement of such immune cells connected with pro-inflammatory immune responses (eg, trafficking macrophages and other phagocytic cells such as neutrophils, monocytes, dendritic cells, and eosinophils) in the expansion of E. cuniculi infection in host tissues has been suggested based on the occurrence of microsporidia in inflamed tissues⁴⁷ and the targeted migration seen following experimental induction of inflammation towards inflammation foci.⁴⁸

Although four genotypes of E. cuniculi have been identified to date, human infections are mostly associated with E. cuniculi genotypes I and II,⁶⁴ which are found primarily in HIV-infected patients,^65–67 and in transplant recipients.^3,31,68,69

In animals, E. cuniculi genotype III seems to be more aggressive and infections caused by this strain can lead to the death of immunocompetent hosts,^28,32,33 whereas E. cuniculi genotypes I and II mostly represent a parasite that occurs in host species in a latent asymptomatic form.^{4,6–8,10–12,17,70}

As the particular genotypes of E. cuniculi differ not only in infection onset speed and extent of infection, but also in chemotherapy sensitivity,^{49,55,57,58,71,72} thus host survival and progression of the infection depend not only on the immune status of the host but also on the E. cuniculi genotype causing the infection. Under experimental conditions, the infection extent and albendazole sensitivity of E. cuniculi genotype I was comparable to E. cuniculi genotype II and the infection onset speed and mortality rate were similar to E. cuniculi genotype III.⁷²

As the results of the present study show, the migration potential of E. cuniculi genotypes and their concentration in induced inflammatory foci varied in both immunocompetent and immunodeficient mice.

In SCID mice the concentration of microsporidia was the most intense in E. cuniculi genotype III, reaching 1.0×10⁷ spores per gram of tissue at 35 DPI, compared to 2.4×10⁵ and 5.0×10⁵ reported in genotypes I and II, respectively (present study),⁴⁸ (Figure 3A). This correspond to the data published by Sak et al^55,58,72 and Kotková et al,^49,71 ie, the infection intensity of E. cuniculi genotypes I and II was comparable, whereas E. cuniculi genotype III reached up to two orders higher spore burden.

Figure 3 Maximum value of E. cuniculi spores per 1 gram of tissue (number in white fields), frequency and spore burden of E. cuniculi spores per gram of tissue (colored squares; each square represents one mouse), and infection trend illustration (red fields) in immunodeficient SCID mice (A) and immunocompetent BALB/c mice (B) infected with E. cuniculi genotype II (modified from Brdíčková et al48). aPeroral infection 107 spores of E. cuniculi in 200 µL dH2O and intramuscular injection of PBS; bperoral infection 107 spores of E. cuniculi in 200 µL dH2O and intramuscular injection of Freund’s Incomplete Adjuvant; grey column, intramuscular injection of all animals in the group into the right thigh muscle; *significant difference between right and left leg; #significant difference between control and experimental group; colored squares, positive capture of microsporidia in the sample according to the quantity scale.

Induction of inflammation in the experimental group of SCID mice infected with E. cuniculi genotype I or III resulted in a significant increase of spore burden at the site of inflammation, while, in the control group, the amount of spores per gram of tissue was approximately identical in both legs. Moreover, the significant concentration of microsporidia in the inflammation-induced site was comparable in the case of E. cuniculi genotypes I and III and was reported from 14 DPI, whereas the only significant difference between induced and non-induced site was described in SCID mice infected E. cuniculi genotype II at 21 DPI⁴⁸ (Figure 3A). The elevated concentration of spores in muscle tissues at the end of the experiment regardless of inflammation induction could be attributed to the systemic spread of E. cuniculi in the terminal phase infection, as described already by Kotková et al.⁴⁹

Also, in control immunocompetent BALB/c mice, the course of infection caused by E. cuniculi I was similar to E. cuniculi genotype II with a random occurrence of microsporidia in the limbs and infection clearance starting from 42 DPI, compared to E. cuniculi genotype III with the infection clearance at 49 DPI (present study),⁴⁸ (Figure 3B), suggesting that, during the experiment, microsporidia may have been eliminated by the host immune system or microsporidia remaining in the body in undetectable amounts. The peak of infection in muscles was reported earlier in E. cuniculi genotypes II and III (21 DPI) than in infection with E. cuniculi genotype II (28 DPI) (present study),⁴⁸ (Figure 3B).

The induction of inflammation in experimental BALB/c mice led to the disappearance of all genotypes of E. cuniculi from the non-induced leg starting either from 14 DPI (genotypes II and III) or 21 DPI (genotype I) (present study),⁴⁸ (Figure 3B). Moreover, a significant increase of microsporidia towards the sites of inflammation in the induced leg was observed from 14 DPI in all genotypes, reaching a peak of E. cuniculi concentration either at 14 DPI (genotype III) or 21 DPI (genotypes I and III). On the other hand, no difference was observed between the occurrence frequencies and spore burden of microsporidia in induced and non-induced legs in the particular control groups (present study),⁴⁸ (Figure 3B).

Conclusions

Based on our results we can conclude that the induction of inflammation had caused increased concentrations of E. cuniculi genotypes I and III in inflammatory foci, as previously reported for E. cuniculi genotype II. The frequency of microsporidia occurrence was not random, but shifted in favor of the site of induced inflammation compared to control groups, where spores were accidental or their specific DNA was not detected at all.

Moreover, our study confirmed differences among E. cuniculi genotypes reported previously based on the infection onset, infection extent, and host mortality rate. Based on the present study, we can conclude that particular genotypes of E. cuniculi also differ in dispersal rate within host bodies after experimentally induced inflammation and E. cuniculi genotype I possesses a combination of features reported in genotypes II and III.

As the mechanisms of microsporidia dispersal as well as the potential vehicle cells remain speculative, detailed specification is needed. The understanding of the connection of microsporidia dispersal with pro-inflammatory immune responses, addressing the tissue specificity and the mechanisms of transfer to target tissues, represents an important challenge with potential consequences for human health and brings novel insight into microsporidia problems and developing reliable preventive and therapeutic strategies.