Myxobolus opsaridiumi sp. nov. (Cnidaria: Myxosporea) infecting different tissues of an ornamental fish, Opsaridium ubangiensis (Pellegrin, 1901), in Cameroon: morphological and molecular characterization

We report a new myxozoan, Myxobolus opsaridiumi sp. nov., infecting the ornamental fish Opsaridium ubangiensis (Pellegrin, 1901) collected from the Anga River near the city of Yaounde, Cameroon. Plasmodia were found in the skin, muscles and spleen. The overall prevalence of infection was 54.7% (288 parasitized fish out of 526 examined). The myxospores were ovoid to subspherical in frontal view and lenticular in lateral view. The valves were symmetrical and relatively thick, without edge markings. The myxospore measurements were 10.7 ± 0.14 (10–11.5) μm long, 9 ± 0.15 (8–10) μm wide and 6.2 ± 0.7 (5.6–7.2) μm thick. The two ovoid polar capsules were equal in size, converging and opening together at the anterior end, measuring 5 ± 0.07 (4.3–6.0) μm long and 2.7 ± 0.07 (2.2–3.0) μm wide. Polar filaments were coiled from 5 to 7 turns. Histopathological analysis revealed no inflammatory reaction associated with the infection. A BLAST search found that the newly obtained 18 rDNA sequence had a low sequence similarity with available sequences for Myxobolus on GenBank. A phylogenetical analysis based on ribosomal DNA partial sequences showed that M. opsaridiumi sp. nov. is closely associated with several species of Myxobolus infecting cyprinid fish. LEKEUFACK-FOLEFACK G.B. et al., Myxobolus opsaridiumi sp. nov. from Opsaridium ubangiensis 57


Introduction
Myxozoa are cosmopolitan microscopic cnidarians that live as endoparasites of invertebrate and vertebrate hosts, primarily in aquatic environments (Kent et al. 2001;Lom & Dyková 2006;Jiménez-Guri et al. 2007). More than 2400 species and 64 genera are recognized Atkinson et al. 2018;Liu et al. 2019;Mathews et al. 2020). They are characterized by morphological simplicity and life-cycle complexity (Fiala et al. 2015). These parasites are identifiable at the spore stage, particularly as actinospores in invertebrate hosts and myxospores in vertebrates. The myxospore is a multicellular structure composed of one to 15 nematocyst-like polar capsules, each containing coiled, extrusible filaments, and at least one ameboid infective germ or sporoplasm (Kent et al. 2001). Infection can be coelozoic, most often in the gallbladder and urinary bladder, or histozoic, targeting skin, muscles, gills or the digestive system. Myxosporean parasites are significant pathogens of fish in both wild and cultured stocks throughout the world, and they are responsible for production losses in farmed fish (Lom & Dyková 2006). Within the class Myxosporea, the genus Myxobolus includes approximately 900 described species (Eiras et al. , 2014Kaur & Singh 2012). Most species infect specific organs, but some have been reported to infect several organs in the same host species (Lekeufack & Fomena 2013;Eiras et al. 2014).
Until recently, myxosporean species were identified only by the morphological and metric characteristics of the myxospore, organ specificity and tissue tropism (Molnár 1994;Dyková & Lom 2007). However, classical zoological methods make it difficult to distinguish morphologically similar myxosporean species that can infect identical tissues and develop in closely related host species (Molnár et al. 2002(Molnár et al. , 2009Lisnerová et al. 2020). Molecular methods have become increasingly popular tools in parasitological studies to identify myxozoan species (Kent et al. 2001;Mansour et al. 2014;Liu et al. 2016). A combination of morphology and molecular classification, considering host range and tissue specificity, provides a more precise approach to distinguishing valid myxosporeans species from identified taxa (Kent et al. 2001;Fiala 2006;Molnár et al. 2009).
Opsaridium ubangiensis Pellegrin, 1901 is a benthopelagic cypriniform of the family Cyprinidae. It is the single species of its genus present in Lower Guinea, is widely distributed in coastal basins from Cameroon to Congo and can reach more than 120 mm in length (Stiassny et al. 2007). In addition to its high-quality meat and rapid growth, species of the genus Opsaridium play an important role in national and international pet trades as ornamental (aquarium) fish. Until now, more than 270 species of Myxosporea have been reported to infect African fishes (Deli et al. 2017) and none were known to infect O. ubangiensis.
In the laboratory, specimens were first examined with the naked eye (eyes, fins, operculum, scales, skin) and then with an Olympus BO61 binocular lens to detect the presence of cysts. After dissection of the hosts, internal organs (gills, heart, liver, kidneys, spleen, gallbladder, gonads and intestines) were removed and examined individually. Some of the recorded cyst contents were identified using an optical microscope with a 100 × objective lens.

Myxospore and histological examinations
Permanent smears of spores were fixed in methanol and stained with May Grünwald-Giemsa. Drawings of fresh spores were carried out using a wild M20 microscope equipped with a camera lucida. Measurements were made on at least 50 spores as proposed by Lom & Arthur (1989). Microphotographs of fresh and stained spores were taken using an Olympus BH-2 microscope equipped with a microphotograph.
For histological studies, tissue samples from infected skin, muscle and spleen were fixed in 10% neutral buffered formalin. The fixed tissue samples were embedded in paraffin wax and then sliced into sections 3 µm thick. Sections were stained with hematoxylin and eosin (H&E) and finally examined and photographed using an Axio Imager Z2 microscope (Carl Zeiss Inc., Berlin, Germany) equipped with a digital camera (Axiocam MRC5, Zeiss). Plasmodia from skin, muscles and spleen were preserved in absolute ethanol for molecular analysis.

Molecular characterization
Molecular analyses were performed in the Department of Zoology at King Saud University. Genomic DNA was extracted using the DNeasy Blood & Tissue Kit (Qiagen, Valencia, CA, USA) following the manufacturer's instructions. Partial SSU rDNA gene was amplified using the primers MyxF144 and MyxR1944 (Mansour et al. 2013). Polymerase chain reaction (PCR) amplifications were performed in a T100 TM Thermal Cycler (Bio-Rad, Singapore). The PCR program consisted of an initial denaturation step for 5 min at 94°C, followed by 35 cycles of 30 s at 94°C, 30 s at 56°C and 120 s at 72°C, and a final extension of 72°C for 10 min. All PCR reactions were performed in a final volume of 30 μl, containing 1 × iProof High Fidelity Master Mix (Bio-Rad, Hercules, CA), 0.2 μM of each primer and 50-100 ng of genomic DNA. The PCR products were sequenced by a commercial sequencing company (Macrogen Inc., Seoul, South Korea) using the same primers used for PCR amplification. A consensus sequence was obtained from both sense and anti-sense strands of samples originating from three different PCR products. This consensus sequence was used to query similar sequences of Myxobolus using BLAST in GenBank (Altschul et al. 1997). Sequences were aligned by applying the default parameters in ClustalX ver. 2.1.0.12 (Larkin et al. 2007). Phylogenetic analyses were performed by Bayesian inference (BI) and maximum likelihood (ML) methods. The BI analyses employed MrBayes ver. 3.2 (Ronquist et al. 2012) and the Markov Chain Monte Carlo (MCMC) method for 1 500 000 generations, with two independent runs of four simultaneous MCMC chains (nchains = 4). Trees were saved every 100 generations (sample freq = 100). ML analyses were based on the General Time Reversible model with a gamma-distributed rate and invariant sites (GTR + G + I) using PhyML ver. 3.0 (Guindon et al. 2010). The analysis involved 22 nucleotide sequences. Positions containing gaps and missing data were removed. Cnidaria Hatschek, 1888Unranked Subphylum Myxozoa Grassé, 1970Class Myxosporea Bütschli, 1881Order Bivalvulida Schulman, 1959Suborder Variisporina Lom & Noble, 1984 Family Myxobolidae Thélohan, 1892 Genus Myxobolus Bütschli, 1882

Type material
CAMEROON • infected skin, muscle and spleen of Opsaridium ubangiensis with plasmodia; Centre Region, Anga River, Yaounde; deposited in parasitological collection of the Zoology Department Museum, College of Science, King Saud University, Saudi Arabia; Myxospsar/12/2018.

Infected tissues
Skin, muscles and spleen.

Vegetative stages
Ovoid, spherical or ellipsoid plasmodia, variable in size, measuring from 0.3 mm to 2.5 mm in length and 0.2 mm to 1.5 mm in width.

Clinical finding and histopathology
Based solely on gross observation of the fish, no signs of disease were observed. Parasitized fish harbored cysts on skin, muscles and spleen. On skin, white cysts up to 2 mm long were collected from the body flanks of some fish ( Fig. 2A). Sections revealed that plasmodia developed in the connective tissue of the dermis beneath the underside of scales (Fig. 2B). Plasmodia were flattened and surrounded by a thin membrane and an internal endoplasm comprising a loosely defined matrix containing developed spores (Fig. 2B).
Some plasmodia were spotted within muscle cells (Fig. 2C). Plasmodia were spindle-shaped, centrally located in the cell and not surrounded by a visible wall. No evidence of inflammation or immune-cell recruitment was seen. The integrity of myofibrils within the infected fibers showed some degree of lysis, with partial loss of myofibrillar details and striations (Fig. 2C-D). These lesions were observed close to plasmodia. Mature spores were scattered in the cytoplasm of infected cells (Fig. 2D).
Infected spleens had plasmodia of up to 2.5 × 1.5 mm (Fig. 3A). They were white, isolated or clustered (Fig. 3B). Some infected spleens were heavily infected and plasmodia were randomly distributed in the   whole organ. In these cases, abnormal enlargement of the spleen was evident (Fig. 3C). Histological sections revealed that, for moderately infected spleens, cysts were either fixed to the external region of the organ (Fig. 3D) or completely implanted within it (Fig. 3E). Development of cysts in the spleen was asynchronous (Fig. 3D-F). Atrophy of the adjacent splenic cells surrounding the cyst was likely due to mechanical compression (Fig. 3F-G). Each plasmodium was surrounded by a wall of a monolayer of flat cells (Fig. 3F-G). The central part of the plasmodium was occupied by fully mature spores, with initial stages of development visible in the periphery (Fig. 3G).

Phylogenetic position
Partial SSU rDNA sequences obtained from different organs were 100% identical. The consensus sequence of 1667 base pairs was submitted to GenBank with the accession number MN497413. This sequence did not match any publicly available myxozoan sequence. The phylogenetic position of the newly sequenced species was analyzed with maximum likelihood and Bayesian inference methods. Both methods produced an identical topology. Myxobolus opsaridiumi sp. nov. occurs in a large clade that includes species infecting cyprinids (Fig. 4)

Discussion
Based on morphological features of the myxospores infecting the skin, muscle and spleen of Opsaridium ubangiensis, the parasite was assigned to the family Myxobolidae Thelohan, 1892 and the genus Myxobolus Bütschli, 1882. To date, no Myxosporidia had been reported to infect O. ubangiensis. This is therefore the first myxobolid infection encountered in this cyprinid fish. Using robust morphological, histological, ecological and molecular data, the described Myxozoa was considered a new species, designated as Myxobolus opsaridiumi sp. nov.
In Egypt, Ali et al. (2002) reported M. caudatus in the caudal fin of Barbus bynni (Forsskål, 1775) (Cyprinidae). Myxobolus caudatus differs from our species in that it has a larger myxospore and more developed polar capsules containing more coils of the polar filament (8-9 vs 5-7).
Myxobolus clarii forms cysts in the testis of Clarias gariepinus Burchell, 1822 (synonym: Clarias lazera Valenciennes, 1840) (Clariidae) in Egypt. The myxospores of this species, although ovoid and of comparable size, have a slightly pointed anterior and a rounded posterior. Its polar capsules are relatively small.
Myzobolus gariepinus, a parasite of the ovaries of Clarias gariepinus (Clariidae), which is found in Botswana, forms spores with a small and blunt point at the anterior end and dimensions that exceed the maximum ranges of the spore of the parasite being described, and with more developed polar capsules.
The polar capsules of M. gariepinus occupy only approximately a third of the spore cavity, while the spore of the species described here has two polar capsules that occupy close to half the space of the spore cavity. Boungou et al. (2006) described two species of myxosporidia from Burkina Faso with a general shape comparable to that of our parasite, M. heterotisi and M. sourouensis, which are gill parasites of Heterotis niloticus (Cuvier, 1829) (Osteoglossidae). Myxobolus heterotisi differs from our species in that it has larger spores and more developed polar capsules (6.41 × 3.53 μm vs 5 × 2.7 μm) that contain a polar filament with more coils (10 vs 5-7). Myxobolus sourouensis has longer spores (11-14 μm vs 10-11.5 µm) which are slightly nippled at the anterior end.
In Chad, Synodontis schall (Bloch & Schneider, 1801) (Mochokidae) harbors M. synodontisi in its gills. Although of comparable shape, the myxospores of this species are longer. They contain more developed polar capsules, which cover the anterior two-thirds of the myxospore cavity, and are sometimes slightly asymmetrical, with a greater number of coils of the filament (13-15 vs 5-7).
Although of comparable size (10.6 × 9 vs 10.7 × 9 μm), the spores of this species have an intercapsular process. In addition, its polar capsules are longer (6.6 vs 5 μm) compared to those of the species being described.
Among all compared species, the 18S rDNA sequence of only M. caudatus is available in GenBank. This species exhibits 90% sequence similarity with our present species. The newly obtained sequence diverges substantially from that of all previously sequenced species of Myxobolus. The maximum similarity was obtained with M. haichengensis, which reportedly infects the gill filaments of Abbottina rivularis in China. The single species of Myxobolus sequenced from Cyprinidae collected in Cameroon was M. dibombensis (Lekeufack et al. 2019), which exhibited only 83% similarity. The phylogenetic tree did not show strict clustering of selected species of Myxobolus according to geographic locality or tissue tropism. It is evident that species infecting cyprinids from around the world and different tissues clustered together in the same clade and can be separated from species infecting other hosts from different families and in some cases by tissue tropism. This data is consistent with previous reports, suggesting that phylogenetic affinity of the fish host provided the strongest co-evolutionary congruence (Fiala 2006;Molnar et al. 2011;Zhang et al. 2014;Karlsbakk et al. 2017;Zhao et al. 2020).
In our investigation we found plasmodia of M. opsaridiumi sp. nov. in three different organs (skin, muscle and spleen). Studies on the site selection of fish myxosporeans infecting the gills, fins, kidneys and muscle suggest that most myxosporean species have a strict tissue tropism determined by the site of the infected organ used for species identification (Molnár 1994;Molnár & Székely 2014). Myxozoan parasites are considered as one of the most striking examples of parasite radiation, which could explain their high diversity . This great diversity would be acquired by their success in exploiting new hosts and new geographic distributions (Fiala et al. 2015). Adaptation to a new host would be facilitated when the newly exploited species was closely related to the original host within a similar tissue. Another means of radiation is the ability of the same species to explore multiple habitats within the same host and to have the ability to switch in other hosts as well. The selective pressure of the new tissue would be stronger than that of a new related host and is usually associated with evolutionary advantages (Patra et al. 2018). This explains the high specificity of many myxozoan genera and species to the infected tissues, as in the cases of species of Ceratomyxa Thelohan, 1892 infecting the gallbladder and Sphaerospora spp. mostly being parasites of the urinary system.
Myxosporidia infections on the skin are rare but easily observable. The presence of cysts of M. opsaridiumi sp. nov. on the skin did not cause an inflammatory reaction in host fish, with proliferation a direct consequence. According to Lekeufack & Fomena (2013), the presence and proliferation of cysts of Myxosporidia on the skin are due to the direct contact of this organ with the immediate aquatic environment of the fish, providing an attachment surface to the actinospores.
The histopathological analysis of the skeletal muscle revealed no signs of inflammation or recruitment of phagocyte cells. The integrity of myofibrils within the infected myofibers exhibited some degree of degeneration, with a partial loss of myofibrillar details and striations. These changes were observed in regions close to the plasmodia. Such histopathological observations in the muscle were similar to those reported by Manrique et al. (2015) (Longshaw et al., 2003); skeletal muscle infection by spores of M. cyprini Doflein, 1898 in Cyprinus carpio Linnaeus, 1758 andM. pseudodispar Gorbunova, 1936 in R. rutilus reported by Baska (1987). As a result of M. artus Akhmerov, 1960 infection in the muscle, Ogawa et al. (1992) observed damaged growth in C. carpio. Most of these pathological changes are responsible for myoliquefaction and the production of meat unfit for consumption (Gilman & Eiras 1998), with the possibility of food poisoning when the meat is consumed raw or undercooked.
Because teleost fish have no medullary cavity in their bones, the spleen and kidney serve as the primary haemopoietic organs (Agius & Roberts 2003), and as fish have no lymph nodes, the spleen plays an essential role in antigen trapping (Press 1998). Although some species of Myxobolus reportedly infect the spleen of teleost fish (Eiras et al. , 2014, no histological observations have been made in affected spleens. It appears that M. opsaridiumi sp. nov. can be responsible for damage in the spleen of Opsaridium ubangiensis, as the development of the plasmodia reduces the vessel lumen and in some cases completely obstructs the lumen of the spleen arterioles. A high parasite load could therefore compromise blood circulation and undermine spleen function. Specific patterns of development and the final site of sporogenesis can vary among species of Myxozoa. Without observation of the developing stages it is difficult to determine whether the spores developed at the site of the infection or were carried there by the host's macrophages. This observation led Dyková (1984) to conclude that myxospores are generally transported by macrophages homing in on macrophage centers of hematopoietic organs such as the kidney, spleen and hepatopancreas, where they are destroyed. However, in the present study only vegetative stages of M. opsaridiumi sp. nov. were observed in the spleen and no free spores were seen. The spleen is therefore a normal site for M. opsaridiumi sp. nov. sporogenesis.