Introduction

Microsporidia are obligate intracellular parasites with the capacity to infect almost all animals, from protozoa to humans. Human microsporidial infection depends on a compromised immune system, but the incidence of this opportunistic infection has increased greatly since the 1980s with the spread of acquired immune deficiency syndrome (AIDS). The primary symptoms of microsporidiosis result from invasion and severe damage to skeletal muscle. In addition, infection induces cutaneous ulcers and malignant diseases of the respiratory system and uropoietic system, further aggravating the suffering of AIDS patients and prompting more intensive research into the pathogenesis of microsporidiosis (Weber et al. 1994; Lom 2002; Cali and Takvorian 2003; Cali et al. 2005).

Fish are one of the principal hosts of microsporidia in the wild. More than 158 microsporidian species in 17 genera have been described that infect fish (Casal et al. 2008). Microsporidia invade internal organs, muscle, and other tissues, and form numerous sporophorocysts that affect fish growth, development, and survival (Leiro et al. 1996; Lom et al. 1999; Azevedo and Matos 2002; Terry et al. 2003; Baquero et al. 2005). Microsporidial infection can greatly reduce the quantity and quality of fish products, and has resulted in severe economic losses to the aquaculture industry. For example, infection by Heterosporis anguillarum in the cultivated Japanese eel Anguilla japonica and Loma salmonae infection of the Pacific salmon Oncorhynchus spp. are important concerns facing modern aquaculture (Lom et al. 2000; Shaw et al. 2000; Lom 2002).

Pleistophora is the most prevalent fish microsporidia. Since the first species, Pleistophora typicalis, was described in Myoxocephalus scorpius, more than 30 additional fish Pleistophora have been discovered (Lom and Corliss 1967; Lom 2002; Abdel-Ghaffar et al. 2009). However, the phylogenetic classification of Pleistophora is still controversial. Many analyses have divided this genus into multiple groups, but the position of many species remains uncertain (Canning and Hazard 1982; Cheney et al. 2000; Nilsen 2000). In recent years, the new genus Ovipleistophora has been recognized and other species added to Heterosporis (Lom et al. 2000; Pekkarinen et al. 2002).

Pleistophora hyphessobryconis is a prevalent infectious species in this genus. It was first found in the neon tetra Paracheirodon innesi (a species once known as Hyphessobrycon innesi but later moved to the genus Paracheirodon) and so is also called “neon tetra disease.” P. hyphessobryconis mainly infects ornamental fish with a broad host range, including species from the orders or families Characidae, Cyprinidae, Cyprinodontidae, Poecilidae, and Cichlidae (Lom and Corliss 1967; Lom 2002). To date, however, the details of P. hyphessobryconis infection in most fish species have not been described. Specifically, there is little data concerning the life cycle and ultrastructure of these microsporidia, leading to diagnostic uncertainty. More recently, P. hyphessobryconis has aroused further concern. Sanders et al. (2010) found zebrafish Danio rerio infected by P. hyphessobryconis from three different laboratories and determined that the infectious agent can spread through the feeding system. It is not known if or how this infection has impacted basic research, particularly in developmental biology, that relies on the zebrafish model. Therefore, it has been suggested that P. hyphessobryconis be added to the list of monitored laboratory pathogens.

The tiger barb Puntius tetrazona is a small sized fish originating in Malaysia, Indonesia, and surrounding areas. Its bright color and easy breeding have made it popular with ornamental fish enthusiasts. We described the first microsporidian infection in tiger barb. Furthermore, we demonstrated that these microsporidia, identified as P. hyphessobryconis, can proliferate in transparent mutants of tiger barb, allowing direct visualization and characterization of microsporidian infection in vivo. The life cycle and ultrastructure of microsporidium within tiger barb were investigated, and the small subunit (SSU) rRNA gene was sequenced to confirm that the infectious agent was P. hyphessobryconis. The morphological characteristics and taxonomic position of P. hyphessobryconis are also revaluated by comparing rRNA sequences of different microsporidian species.

Materials and methods

Collection of specimens

Several batches of fish with suspected microsporidial infections were obtained from different ornamental fish culture farms in Liwan District, Guangzhou, Guangdong, China, and kept for observation in 40 L aquaria equipped with filtration systems and heaters. Water temperature was maintained at 24−26 °C and a quarter of the water volume was changed every 2 days. The fish were fed daily with commercial dehydrated diet. Fish showing symptoms of microsporidial infection were euthanized with an overdose of tricaine methanesulfonate (MS-222). The whitish and opaque muscular tissue were selected, spread on glass microscope slides, and pressed under cover slips for microscopic observation.

Artificial infection by microsporidia

All fish were maintained and treated humanely. The use of the fish in this study was approved by the Aquatic Animal Research Committee at Pearl River Fishery Research Institute.

Two tiger barb mutants, albino and transparent, are maintained in our laboratory. Groups of 30 5-day-old mutant fish were reared in small aquaria (1 L) at 26 °C. The fresh microsporidian spores were isolated from the tissues of infected tiger barbs. After counting the spores using a hemocytometer, about 80,000 isolated spores were added to the aquarium water to infect the tiger barbs. A second dose of spores was added three days later. These tiger barb were fed newly hatched Artemia twice a day. After 15 days’ exposure to the spores, the fish were moved into a larger aquarium (12 L) and fed commercial dry feed. The fish were regularly monitored for changes in muscle appearance indicative of microsporidial invasion. Fish judged to be infected by changes in muscle transparency were removed and examined for microsporidial infection under light microscopy.

Histological preparation

Infected muscle tissue was fixed in Bouin’s fixative, rinsed, dehydrated through a graded ethanol series, embedded in paraffin, cut at 5 μm, and stained with hematoxylin and eosin (H&E).

Transmission electron microscopy

For ultrastructural studies of the infectious agent, small pieces of infected muscle were fixed for 48 h in 0.1 M phosphate buffer containing 2.5 % glutaraldehyde at 4 °C, and then post-fixed in 1 % osmium tetroxide in 0.1 M phosphate buffer at 4 °C. After dehydration in a graded ethanol series, the samples were embedded in Spurr’s resin. Ultrathin sections were stained with uranyl acetate and lead citrate for observation under a CM 10 Philips electron microscope at 40 kV.

DNA sequencing

Primers V1f (5′-CACCAG GTT GAT TCT GCC TGA C-3′) and 1492r (5′-GTT ACC TTG TTA CGA CTT-3′) were designed based on the gene sequence of the small subunit (SSU) rRNA of microsporidia and used for PCR amplification. The thermocycle protocol consisted of an initial denaturation step of 94 °C for 4 min, followed by 30 cycles of 94 °C for 40 s, 55 °C for 40 s, 72 °C for 1.5 min, and a final single extension at 72 °C for 8 min. The PCR products were purified using a small quantity DNA gel extraction kit (Qiagen) and analyzed by 1 % agarose gel electrophoresis. The purified product was cloned into the PMD18-T vector (TaKaRa), and the positive clones were identified and sent to Shanghai Biological Engineering Co., Ltd. for sequencing.

Distance and phylogenetic analysis

Sequencing results were compared to SSU rRNA genes of microsporidia from Pleistophora and other closely related genera in Genbank using ClustalW and MEGA4.0 software to construct a phylogenetic tree. The accession numbers of the microsporidian SSU rRNA sequences retrieved from GenBank were P. hyphessobryconis (GU126672), Pleistophora hippoglossoideos (AJ252953), P. typicalis (AF044387), Ovipleistophora ovariae (AJ252955), Ovipleistophora mirandellae (AF356223), Loma embiotocia (AF320310), Loma sp. (AF104081), Ichthyosporidium sp. (L39110), Pseudoloma neurophilia (AF322654), Glugea sp. (AY090038), Glugea anomala (AF044391), Glugea atherinae (U15987), Loma acerinae (AJ252951), Microsporidium sp. STF (AY140647), Trachipleistophora hominis (AJ002605), Vavraia culicis (AJ252961), Pleistophora sp.2 (AF044389), H. anguillarum (AF387331), Microgemma sp. (AJ252952), and Myosporidium merluccius (AY530532).

Results

Clinical signs and pathogen

Tiger barbs with signs of P. hyphessobryconis infection avoided the main groups, swam slowly, and were darker in color (Fig. 1). Some fish presented with necrotic lesions of the skin and fin ray due to secondary infections by bacteria or water mold. The infection rate in some cultivation groups reached 50 %. In wet mount preparations, sporophorocysts were diffusely distributed among skeletal muscular fibers under light microscopy (Fig. 2). Sporophorocysts were generally round (Fig. 3), while a few were irregularly shaped. The sporophorocysts were packaged within a membrane and varied in size and structure according to devolvement stage. Spores within some of the sporophorocysts could be clearly observed (Fig. 2). Fresh spores were mostly ovoid or pyriform, slightly acute in the anterior extremity, and blunt in the posterior extremity. The average size of fresh spores was 6.9 ± 0.5 × 3.9 ± 0.4 μm. Mature spores were more translucent, with obvious light refraction in the posterior aspect, suggesting the development of a posterior vacuole (Fig. 4).

Figs. 1–6
figure 1

Symptoms of the tiger barb with microsporidial infection and light microscopic images of the microsporidia. 1 Dark coloration of a tiger barb parasitized by microsporidia. 2 Microsporidial sporophorocysts in different developmental stages are shown in wet mount preparations of muscle. Meront (M) and sporophorous vesicle (arrow) with spores (Sp) are showed. 3 Most sporophorocyst (arrow) are round. 4 The posterior aspect of the mature spore shows significant light refraction indicative of a vacuole. 5 An albino tiger barb artificially infected with microsporidia showing obvious symptoms of microsporidiosis. 6 Microsporidial infection is easily observed in a transparent mutant tiger barb, and the spatial distribution of infected muscle (arrowhead) is highly visible. Scale bars = 5 mm (1, 5, and 6) and 10 μm (2, 3, and 4)

Full size image

Microsporidial infection of mutant tiger barb

By the time the fish were moved into larger aquaria, only 29 albino tiger barbs and 26 transparent tiger barbs remained. Sixty days post-infection, these fish were all alive, but 11 of 29 albino fish were infected, and symptoms of infection were found in 15 of 26 transparent mutant fish. It was difficult to observe the early symptoms of infection in albino tiger barb. Whitish and turbid muscle was observed over the course of the disease, however, and became more obvious in the later stages. Eventually, much of the body appeared turbid (Fig. 5). In contrast, it was far easier to observe the early symptoms of infection in the transparent mutant tiger barb. First, punctate areas of turbidity were faintly discernible by the naked eye. With the proliferation of microsporidium, the contrast between the white opaque tissue and transparent muscle became obvious and gradually extended so that the spatial distribution of the lesions was highly visible (Fig. 6). Infected muscle tissues showing these pathological changes were harvested and observed in wet mount preparations. Many microsporidia were observed under the light microscope. Thus, the transparent tiger barb mutant allowed for the in vivo observation of microsporidial infection and spread in real time.

Histological studies

Microsporidia parasitized the muscle bundle and stained cardinal red by hematoxylin and eosin (H&E). Most were round and surrounded by a membrane, and so could be easily separated from muscle tissue. At the initial infection site, there was always a great quantity of microsporidia. Microsporidia were observed in almost every muscle bundle, and several were often found in cross-sections of a bundle (Fig. 7). Under the light microscope, different merogonial stages were observed, from uninucleate and binucleate to multinucleate meronts (Fig. 8). Sporonts, sporoblasts, and mature spores were found in large numbers within the sporophorous vesicles (Figs. 9 and 10). Parasitism by microsporidia could cause liquefaction necrosis and swelling in parts of the muscle fiber (Figs. 8, 9, and 10), resulting in major injury to the fish.

Figs. 7–10
figure 2

Histological sections of tiger barb muscle infected with microsporidia. 7 Microsporidia (arrow) parasitizing muscle tissue. Microsporidia can be observed throughout the muscle bundle when the infection is severe. 8 Microsporidia during the different developmental stages. Uninucleate, binucleate, and multinucleate meronts can be detected in the same muscle bundle. Infection leads to liquefaction of muscle tissue (white arrowhead). Uninucleate meront (UM), binucleate meront (BM), multinucleate meront (MM). 9 Sporoblasts (Sb) within the sporophorous vesicles (SV) in muscle tissue. 10 Spores (Sp) abound in sporophorous vesicles. In this specimen, liquefaction of muscle tissue is severe (white arrowhead). Scale bars = 100 μm (1) and 10 μm (2, 3, and 4)

Full size image

Ultrastructural observations

Throughout the entire life cycle, microsporidia remain nearly round with clearly observable nuclei. During merogonial stages, the surface was enveloped by a simple membrane with slight wrinkling. In the earliest developmental stage, uninucleate meronts had one well-defined orbicular nuclei and clear karyolemma (Fig. 11). The cytoplasm contained many regions of endoplasmic reticulum and scattered ribosomes. Granular material formed by dead host cells surrounded many meronts. Uninucleate meronts divided quickly to form binucleate meronts (Fig. 12). After many divisions, nucleus with irregular shapes were seen in multinucleate meronts (Fig. 13). As the microsporidium entered sporogony, the outer membrane thickened (Figs. 15 and 16), and the multinucleated merogonial plasmodia exhibited several cisternae of endoplasmic reticulum surrounding the nucleus (Fig. 14). Sporogony was also associated with a conspicuous change in the structure of the membrane, as two layers were formed and gradually separated. The outer layer was much thicker and became the sporophorous vesicle surrounding the sporont, while the inner layer retracted away from the vesicle to form the thinner plasmalemma. Sporogonial plasmodia were divided into many clusters (Fig. 17). Each cluster contained many nuclei, displayed a variety of small vesicles, free ribosomes, stacks of endoplasmic reticula, and finally dissociated into several uninuclear sporoblasts (Fig. 18).

Figs. 11–20
figure 3

Ultrastructures of the different developmental stages of the microsporidian parasitizing tiger barb. 11, 12 Meronts of microsporidia with one (uninucleate meront) or two (binucleate meront) nuclei (N), enveloped by a simple membrane. 13 Multinucleate meronts with several nucleus (N). 14 As microsporidia develop into sporogony, the membrane thickens, and endoplasmic reticulum (arrow) is visible in the plasmodia. 15, 16 Compared to the merogonial stage, the sporogonial membrane is obviously thicker (arrowhead). 17 Sporogony gives rise to sporoblasts. 18 A sporophorous vesicle with developing sporoblasts (Sb). 19 The development of individual sporoblasts appears synchronized, and the nuclei migrate to one pole. 20 The sporoblasts become regular, and the polar filaments (PF) are recognizable. Scale bars = 1 μm

Full size image

Within the same sporophorous vesicle, the development of sporoblasts appeared synchronized. The morphology of sporoblasts was irregular in the early stage, and the electron-dense cytoplasm was packed with membranous material. Sporoplasm then became dense and the nuclei migrated to one pole of the parasite (Fig. 19). Concomitant with these changes, the individual sporoblasts became more regular in appearance, and various organelles were clearly recognizable (Fig. 20), including polar filaments, endospores, and exospores, as the sporoblast finally developed into mature spores.

Spores were generally pyriform, and the wall was composed of a distinct electron-dense exospore and an electron-lucent endospore (Figs. 21 and 22). Mature spores were mononuclear, with a conspicuous clear vacuole occupying a large volume of the posterior pole. The anchoring disc was found at the anterior end of the spore (Figs. 21, 22 and 23). Polar filaments were mainly located in the middle and lower part of the spore, and appeared as inclined long and thin tubes. One end was connected to the anchoring disc, thinning slightly as it extended to the posterior extremity of the spore, then curling into a spiral shape. In some sections, the polar filaments showed 2−3 layers with 36−42 turns surrounding the posterior vacuole (Fig. 24). The polaroplast folded around the polar filament in the apical region of the mature spore (Figs. 25 and 26).

Figs. 21–26
figure 4

Ultrastructure of microsporidian spores in tiger barb. 21 Longitudinal section of a mature spore, showing the spore wall (Wa), polar filament (PF), and posterior vacuole (Va). 22 Details of the apical region of a mature spore, showing anchoring disc (AD) in close contact with the wall. 23 Mature spore showing the nuclei (N), polar filaments, and posterior vacuole. 24 Polar filaments with external membrane (arrowhead) and central dense mass (arrow). 25 Details of the apical region of a mature spore showing the polaroplast (Pp). 26 The lamellar region showing the polaroplast and polar filament. Scale bars = 0.5 μm

Full size image

Molecular analysis

Total genomic DNA was extracted from infected fish for PCR amplification and sequencing of the small subunit (SSU) rRNA gene. A 1,359 bp sequence was obtained with 52.39 % GC content. The sequence has been submitted to GenBank with accession number JN575482. Blast comparison revealed little difference compared to the SSU rRNA gene of P. hyphessobryconis from neon tetra, with 99.5 % homology and no insertions or deletions.

From the phylogenetic tree constructed from the SSU rRNA gene sequences of other related microsporidians, different branches of the genus Pleistophora were revealed. The SSU rRNA gene sequence of the microsporidium of tiger barb was close to that of P. hyphessobryconis, and clustered together with O. ovariae and O. mirandellae. The H. anguillarum that infects Japanese eel is also located in this clade (Fig. 27). The sequence obtained by our group has 95.4 % homology with O. mirandellae and O. ovariae, and 93.3 % homology with H. anguillarum. The archetype species in genus Pleistophora, P. typicalis, is located in another clade, and clusters together with P. hippoglossoideos. It has 89.2 % homology with the sequence obtained by our group.

Fig. 27
figure 5

Phylogenetic tree of SSU rRNA sequences constructed to compare the Pleistophora sp. parasitizing tiger barb with other microsporidia. Transcripts of 1,312 bp were used to reconstruct the tree

Full size image

Discussion

The life cycle of microsporidium in tiger barb was systematically studied by light microscopy and EM ultrastructural analysis. Some characteristics of this microsporidium were consistent with other Pleistophora species: (1) nuclei were isolated at all stages of the life cycle; (2) merogonial plasmodia were round and bounded by a membrane; (3) in the early stage of sporogony, the outer layer of the thickened bilayer membrane became the sporophorous vesicle, while the inner layer differentiated into the plasma membrane; (4) a great number of spores was produced within each sporophorous vesicle, and each spore contained a similarly structured anchoring disc and polar filament; (5) microsporidia specifically infected skeletal muscle, with no xenoma formation (Canning and Nicholas 1980; Faye et al. 1990; Leiro et al. 1996; Terry et al. 2003; Abdel-Ghaffar et al. 2009).

Skeletal muscle is the principal tissue infected by microsporidia of the genera Pleistophora. P. typicalis, Pleistophora carangoidi, and Pleistophora finisterrensis also parasitize fish muscle (Canning and Nicholas 1980; Leiro et al. 1996; Lom et al. 1999). During the latter stages of infection, muscular tissues of tiger barb appeared cloudy and edematous. Histological sections revealed that the microsporidium parasitizing muscular tissues were in different stages of development, including merogony, sporogony, and mature spore stage. Parasitic invasion led to necrosis of muscular tissue and a series of inflammatory reactions. By analysis of microsporidial microstructure, infection characteristics, and the pathological changes induced by infection, we surmised that this microsporidium was similar to the P. hyphessobryconis isolated from infected zebrafish. Indeed, the fresh microsporidia were similar in size and shape to those isolated from zebrafish (Sanders et al. 2010). BLAST comparison indicated that the SSU rRNA of this tiger barb microsporidium was 99.5 % homologous to that of P. hyphessobryconi, with no insertions or deletions. Combined morphological and molecular analyses indicated that the parasite infecting tiger barb was P. hyphessobryconis, a microsporidium with a broad host range.

Life cycle and ultrastructural features are used for microsporidial classification. Thus, electron microscopy provides a wealth of valuable information for the phylogenetic classification of microsporidia (Lom and Nilsen 2003). Lom and Corliss (1967) studied the ultrastructure of P. hyphessobryconis parasitizing neon tetra, especially the morphology of the spores, and provided strong evidence that P. hyphessobryconis belongs to the genus Pleistophora. In our study, we described the morphology of microsporidium in infected tiger barb during different developmental stages, including merogony, sporogony, and spore, providing a detailed record of the life cycle. In neon tetra and tiger barb, many morphological characteristics of P. hyphessobryconis are similar, such as the morphology of the exospore, endospore, nuclei, anchoring disc, and polar filaments. Furthermore, the developmental cycle is similar, at least in the earlier stages. However, there are some differences: both the nuclei and plasmodia of merogonial and sporogonial P. hyphessobryconis are oval-shaped in neon tetra but rounded in tiger barb. In neon tetra, the microsporidial membrane is smooth and compact, and surrounds the plasmodium, while in tiger barb, wrinkles were observed in the membrane, and these creases were more obvious in sporophorous vesicles. Therefore, there are differences in P. hyphessobryconis morphology during development in different hosts. Recent studies also demonstrated differences in the morphological characteristics of microsporidia (including differences in size, shape, and number of nuclei) both between and within species (Dunn and Judith 2001). Further research is required, however, to determine the reasons and pathological significance of these host-specific traits.

Fish microsporidia usually show host specificity, infecting only one species or closely related species from the same genus. P. hyphessobryconis, however, has an unusually broad host range. It was reported that nearly 20 species of fish from different orders or families can be infected by P. hyphessobryconis (Lom 2002), and many of these species are kept as ornamental fish. The feeding modes of ornamental fish may increase the chance of close contact with other species of different geographic regions, leading to transmission of the pathogen. However, the broad host specificity of P. hyphessobryconis is still unexplained and has drawn considerable attention from both fish culturists and researchers. P. hyphessobryconis is an important representative microsporidia and prevalent fish pathogen, so development processes and morphological characteristics in different hosts may provide a better understanding of the infection mechanism in different species.

The sequence analysis of the SSU rRNA gene has proven to be a valuable tool for the classification and phylogenetic analysis of microsporidia. Previous studies found that traditional classification methods for microsporidia were inconsistent with findings from molecular biology (Nilsen et al. 1998; Nilsen 2000; Bell et al. 2001). Baker et al. (1995) performed a phylogenetic analysis of microsporidia of the genera Endoreticulatus and Nosema, and found differences between classification systems established by molecular data and those based on morphology or ultrastructure. At present, differences in morphology and life cycle are the features most commonly used for classification of microsporidia. However, combined molecular phylogeny and morphology may allow for more accurate classification.

The molecular phylogeny results reported in our study are in general agreement with previous studies (Nilsen et al. 1998; Nilsen 2000; Pekkarinen et al. 2002). P. hyphessobryconis is most closely related to O. ovariae, O. mirandellae, and H. anguillarum. The archetype species of the genus Pleistophora, P. typicalis, is located in another clade together with P. hippoglossoideos. P. hyphessobryconis parasitizes fish muscle, and its morphology is also similar to Pleistophora. Canning and Nicholas (1980) classified P. hyphessobryconis as the representative species of this genus based on microscopic and ultrastructural features. P. hyphessobryconis and O. ovariae have the typical morphological features characteristic of Pleistophora. However, P. hyphessobryconis lacks some of the characteristicsthat distinguish Ovipleistophora from Pleistophora: (1) during meront stages, the Ovipleistophora meront envelope is ruptured and releases uni- and multinucleate cells; (2) Ovipleistophora spores are dimorphic, with both macro- and microspores; (3) Ovipleistophora typically parasitize ovaries. P. hyphessobryconis and O. ovariae share several morphological characteristics and developmental patterns, such as a round merogonial plasmodium and synchronous development of sporoblasts in sporophorous vesicles, in addition to broad host ranges (Pekkarinen et al. 2002). Microsporidia have been studied for more than a century, and species classification methods are continually evolving. There is no widely accepted classification system, however, and the characteristics of some genera and species remain unclear (Lom 2002; Lom and Nilsen 2003). Our research indicates that the methods of analysis have greatly affected the phylogenetic classification of P. hyphessobryconis. Therefore, the characteristics of Pleistophora and Ovipleistophora require further description and analysis.

Fish microsporidia form a unique group within the microsporidian phylum, with distinct physiological adaptations to allow for broad host environments. Some research also indicates that microsporidia that parasitize fish may share a common origin (Nilsen 2000; Bell et al. 2001; Dunn and Judith 2001). However, the mechanisms for host adaptation remain unexplained. For instance, T. hominis parasitizes immunologically deficit patients is closely related to the fish microsporidium Pleistophora spp. (Cheney et al. 2000). The fish microsporidium, G. atherinae, shares a common epitope with Encephalitozoon cuniculi, a microsporidium that causes opportunistic infection in AIDS patients (Delbac et al. 1998), so the relationship between these two species warrants additional study. Some microsporidia infecting humans also have a broad host range, such as E. cuniculi, T. hominis, and Nosema algerae (Weber et al. 1994; Lom and Nilsen 2003; Cali et al. 2005). Microsporidia are opportunistic pathogens in humans, but the incidence of infection is significantly higher today with the spread of AIDS, suggesting that microsporidia infection in humans may originate in non-human animals. P. hyphessobryconis is the fish microsporidia with the broadest host range, and small differences in morphology can be observed in different hosts, while these differences may relate to host-specific adaptation mechanisms. Phylogenetic analysis also demonstrated that P. hyphessobryconis is closely related to T. hominis, which also infects humans. Thus, P. hyphessobryconis infection in fish may be a robust model in which to study the general mechanisms of microsporidial infection and host adaptation.

Transparent mutants of tiger barb can be artificially infected by P. hyphessobryconis and showed the typical symptoms of infection, so the microsporidium can be obtained prepensely to provide a source of P. hyphessobryconis for research. In transparent mutants, the muscle damage induced by P. hyphessobryconis was directly visible, and the dynamic progression and spatial distribution of pathogenic invasion could be observed in real time. Therefore, the transparent mutant of tiger barb is a valuable experimental model to study the pathogenesis of P. hyphessobryconis infection, with significant applications for future research in the treatment and control of microsporidiosis.