Native parasite affecting an introduced host in aquaculture: cardiac henneguyosis in the red seabream Pagrus major Temminck & Schlegel (Perciformes: Sparidae) caused by Henneguya aegea n. sp. (Myxosporea: Myxobolidae)

Background Henneguya Thélohan, 1892 (Myxobolidae) is one of the most species-rich genera of myxosporean parasites infecting fish. Although common in nature, there are few reports of these parasites causing important disease in aquaculture. In this paper, we describe a new species of Henneguya infecting Pagrus major (Temminck & Schlegel), a fish host introduced to the Mediterranean Sea from Japan in the late 1980s. Results Large plasmodia of the parasite were found in the bulbus arteriosus and in the ventricle of the infected fish. Spores were found mainly in the kidney and heart and were accompanied by melanized macrophages or vascular intimal proliferation mixed with a mild non-suppurative response, respectively. Comparisons of morphometric data for spore and polar capsule length and width, suggest a unique combination of features in the newly described species. Molecular analysis, based on 18S rDNA sequence of the parasite, followed by phylogenetic analysis, indicated that the parasite described here is a novel species of Henneguya, clustered with the marine congeneric species. Conclusions Henneguya aegea n. sp. infects in aquaculture P. major, a host introduced as eggs to the Mediterranean from Japan. Despite the high host specificity of the myxobolid parasites, H. aegea n. sp. seems to be able to use P. major as a host and propagate successfully, causing morbidity and mortality. This could result in spillback of the new species from high density cultured non-native P. major to native fish hosts.


Background
Since its inception, Mediterranean marine aquaculture has been mainly reliant on the farming of two fish species, the gilthead seabream, Sparus aurata Linnaeus, and European seabass, Dicentrarchus labrax (Linnaeus). Diversification, with the introduction of new species, is considered as the most appropriate way to ensure continued economic viability and sustainable growth of the industry in order to meet the increasing demand of consumers for high quality protein. Amongst the species that have long been considered as promising alternatives is the red porgy, Pagrus pagrus (Linnaeus), which is a highly valued, indigenous sparid [1]. Despite its good performance as an aquaculture species at all stages of production, the skin color of farmed red porgies darkens soon

Methods
Twenty apparently healthy fish (weight: 53.7-361.7 g, total length: 15.3-25.5 cm) were randomly collected in 2016 from the affected farm, which has been reporting persistent morbidity and mortality of cultured red seabream since 2010. The affected fish were reared in net pens in the sea. Usually, mortalities were in the range of 1-5% in each mortality episode and affected mostly larger fish. The fish were examined macroscopically for external and internal lesions or abnormalities. Samples from gills, kidney, liver and heart were fixed in 10% buffered formalin for routine histology. Infected hearts were dissected, and cysts of parasites were removed and fixed for electron microscopy in 2.5% glutaraldehyde in cacodylate buffer or examined fresh using a light microscope equipped with a digital camera.

Spore morphology
Henneguya spores were obtained from freshly ruptured cysts and heart scrapings from the sampled fish and measured with the aid of a light microscope equipped with an image analysis system, calibrated with a micrometric scale, following the guidelines of Lom & Arthur [16]. Measurements were based on 55 fresh spores and are presented in micrometres as the range followed by the mean ± standard deviation, SD.

Scanning electron microscopy (SEM)
Samples for SEM were washed with sodium cacodylate buffer, post-fixed with 1% OsO 4 and dehydrated in an ascending alcohol series, mounted on stubs, sputtercoated with gold palladium and examined using a JEOL JSM-6390LV scanning electronic microscope at 15 kV at the Electron Microscopy Laboratory of the University of Crete.

Histology
Formalin-fixed tissues were dehydrated in a 70-96% ethanol series and embedded in glycol methacrylate resin (Technovit 7100, Heraeus Kulzer, Germany). Serial sections (3-5 µm thick) were obtained using a microtome (Leica RM2245, Germany) with disposable blades. After drying, slides were stained with methylene blue/azure II/ basic fuchsin (Polychrome) [17]. Another set of samples was also processed for paraffin sectioning; these tissues were fixed for 72 h in 10% buffered formalin, dehydrated in an ascending ethanol series (70-100%), embedded in paraffin and cut in 2-3 µm thin sections. They were stained using a standard protocol with haematoxylin and eosin (H&E).

Micro-CT
Two infected and one uninfected heart were also examined with micro-CT. These samples were fixed in 4% phosphate-buffered formalin and dehydrated to 70% ethanol for 3 days before scanning. Subsequently, two different staining agents were used in each infected sample in order to enhance the contrast between the soft tissues. One affected and one control heart were stained with 0.3% phosphotungstic acid (PTA) in 70% ethanol while the second affected heart was stained with 1% iodine in 96% ethanol according to the protocol of Metscher [18]. The micro-CT scans of the hearts were performed at the Hellenic Centre for Marine Research (HCMR) using the SkyScan 1172 micro-CT scanner (SkyScan, Bruker, Belgium). This scanner uses a tungsten X-ray source with an anode voltage ranging from 20 to 100 kV, 11 MP CCD camera (4000 × 2672 pixel) and a maximal resolution of < 0.8 μm/pixel. The hearts #1 and the healthy control were scanned at a voltage of 67 kV and 150 μΑ, while the heart #2 was scanned at a voltage of 80 kV and 124 μΑ. All scans were performed with an aluminum filter for a full rotation of 360° at the highest camera resolution. The projection images acquired during the scanning procedure were reconstructed into cross-section images using the SkyScan's NRecon software (NRecon, Skyscan, Bruker, Belgium) which implements a modified Feldkamp's back-projection algorithm. Furthermore, 3D volume renderings of the scanned specimen were created using the CTVox software (CTVox, Skyscan, Bruker, Belgium) to visually investigate the anatomy of the internal and external 3D structures of the sample.

DNA isolation and sequencing
Infected arterial bulbs preserved in 95% ethanol were used for DNA extraction. Samples were centrifuged at 8000×g for 10 min and the ethanol supernatant was removed. DNA was extracted from the dried pellet using QIAgen DNeasy kit (Qiagen, Valencia, California, USA) according to manufacturer's instructions. Small subunit ribosomal DNA (SSU rDNA) was amplified by the polymerase chain reaction (PCR) technique using Taq PCR Master Mix kit (Qiagen) and the primers Myxo-specF (5′-TTC TGC CGT ATC AAC TWG TTG-3′) [19] and 18R (5′-CTA CGG AAA CCT TGT TAC G-3′) [20]. PCR reactions were performed in a BIO-RAD MJ Mini Personal Thermal Cycler with an initial denaturation at 94 °C for 3 min, followed by 35 cycles of denaturation at 94 °C for 1 min, annealing at 56 °C for 1 min and extension at 72 °C for 90 s and a final extension step at 72 °C for 5 min. PCR products were visualized in an ethidium bromide-stained 1% agarose gel and then purified with PURELINK PCR purification kit (Invitrogen -Thermo Fisher Scientific Inc, Waltham MA, USA). PCR products were sequenced using ABI3730xl sequencer (Applied Biosystems) according to the protocol BigDye Terminators 3.1 (Applied Biosystems, ThermoFisher Scientific Inc, Waltham MA, USA).

Phylogenetic analysis
All available SSU rDNA sequences of Henneguya species infecting fish were downloaded from GenBank and used in the present study; Ceratomyxa diplodae Lubat, Radujkovic, Marques & Bouix, 1989 (KX099691.1) was used as the outgroup. The novel sequences were edited by eye, trimmed to 1830 bp with Geneious 9.1 and aligned with the GenBank sequences using ClustalW in Geneious 9.1 with default settings. Phylogenetic analysis was done by Neighbour-Joining method in MEGA 7 [21] using Tamura-Nei model and pairwise deletion option. Genetic distances were calculated using MEGA 7 and the "compute pairwise distances" option of the Tamura-Nei model.

Gross pathology
Sporadic incidents of diagnosed red sea bream henneguyosis from Leros Island date back to 2010. In these cases, either persisting low morbidity and daily mortality (≤ 1%), or sudden deaths of apparently healthy-looking fish were reported with occasional minor superficial skin lesions. Mortality was low and not associated with a particular size range of fish or season of year but was of concern to the farmers when large marketable fish were lost or when candidate broodstock fish under quarantine were affected. Most of the fish examined (~ 90%) appeared normal externally with no hemorrhages on skin or erosion of fins, but sometimes the skin was dull with superficial inflammation, minor erosions and scale loss. Gills were moderately inflamed with excessive mucus secretions.
At gross examination, infected hearts had enlarged arterial bulbs as well as obvious areas of myocardium degeneration (Figs. 1a, b, 2a). When dissected, these hearts revealed numerous large, white to cream-colored, 1 to 2 mm in diameter, irregularly shaped plasmodia in the ventricle (Fig. 2b) containing large numbers of developing Henneguya sp. spores (Fig. 2c). Abundant free mature spores were found in microscopic examination of the atrium and bulbus arteriosus (Fig. 2d). Spores were apparently disseminated via blood circulation to all body parts and organs. Numerous mature spores were found in the kidney, eliciting a host response characterized by proliferation of the melanomacrophage centers (Fig. 2e).
Examination of the macrophage aggregations in fresh squash preparations of the kidney by light microscopy at higher magnification revealed that they encompassed free parasitic spores (Fig. 2f ). A few spores were also present in the ventricle, gills, digestive mucosa, liver parenchyma ( Fig. 1d) and gall-bladder wall.
Livers at necropsy appeared swollen and pale due to vacuolar hepatic lipidosis, sometimes with petechial hemorrhages and areas with a greenish tinge apparently due to post-mortem bile imbibition (Fig. 1a-c). Cohesive hepatic parenchyma (squash) observed under the microscope showed diffuse fatty degeneration (vacuolar hepatopathy) and an abundance of Henneguya spores (Fig. 1d).

Histopathology
Multiple cysts (plasmodia) of the myxozoan parasite with a broad eosinophilic, hyaline cyst walls measuring 10 μm in thickness were observed in the lumen of the ventricle and bulbus arteriosus of the infected hearts ( Fig. 3a, b, d). Within the cysts, numerous developing spores were visible (Fig. 3a, c). Several ellipsoid clear areas directed towards the arterial lumen of the bulbus were observed in the cyst wall; these were previously described as pinocytotic channels [22]. The cysts observed were often closely aligned to the endothelium but without real attachment. The endothelium of the endocardium and numerous large vessels showed papilliform proliferations into the vessel lumens consisting of the endothelial cells (Fig. 3e). Within the myocardium, there were only very small foci of degeneration consisting of loss of the myofibers and slight infiltrations of macrophages and lymphocytes. Similar proliferations of the intima of larger blood vessels were also present especially in the spleen and liver. Here, the lumens of the vessels were almost completely obstructed by the intima proliferations (arteriosclerosis). This lesion was characterized more by proliferative and degenerative changes than inflammation within the vessel wall. Few inflammatory cells were intermingled within the proliferations (Fig. 3f ). Gills were mostly normal with very mild fusion of some secondary lamellae with focal lymphocytes and histiocytic infiltrates.

Micro-CT
In the first sample (heart #1), the micro-CT scans revealed numerous large plasmodia located in the ventricle ( Fig. 4a-c). A membrane surrounding the plasmodia was visible in the images. In this sample, no plasmodia were found in the bulbus arteriosus. In the second sample (heart #2), three large plasmodia were identified in the ventricle and one in the basis of the bulbus arteriosus (Fig. 4d). Following scanning, the sample was dissected in the area where the plasmodia were seen in micro-CT in order to validate the tomography findings (Fig. 4e). The plasmodia were removed, and squash preparations confirmed that these were plasmodia of the Henneguya sp. Using the CTAnalyzer software (CTAN, Skyscan, Bruker, Belgium), we measured the volume of the three plasmodia and the ventricle. The three plasmodia occupied approximately 2% of the total volume of the ventricle. A video of the scanning of the two hearts is provided in Additional file 1: Video S1.

Differential diagnosis
A comparation of taxonomic characters with congeneric species of the genus ( Taking into account the high intraspecific size variability of parasites in the same sample and the possibility that the length of the caudal processes is no longer considered a systematic character, other infection parameters and/ or molecular data should be taken into consideration to determine species distinct status. To our knowledge, Henneguya aegea n. sp. is the first member of the genus with plasmodia developing not only in the bulbus arteriosus but also in the ventricle (tissue tropism).

Scanning electron microscopy (SEM)
Observations with SEM showed that the spores had a smooth surface and two unequal valves which were joined forming a conspicuous sutural line. At the front part of the spore, two polar tubule discharge pores were observed, each situated in either valve connected to the sutural line. The long caudal processes were separated but were in contact for at least the first half of their length (Fig. 6).

Molecular analysis and phylogeny
A small subunit rRNA (SSU rRNA) gene sequence of 1727 nt in length was obtained from H. aegea n. sp. All SSU rRNA gene sequences (n = 52) from Henneguya spp. available on GenBank (August 2018) were retrieved and aligned with H. aegea n. sp. sequence using ClustalW in Geneious 9.1. Following trimming which resulted in an alignment with a total of 1830 characters including alignment gaps, a phylogenetic tree using the Neighbor-Joining method with 1000 bootstraps was constructed with  (Fig. 7). The highest similarity expressed as % identity (% of bases/ residues which are identical) was with H. mauritaniensis (89.8%) followed by H. pagri (89.5%).

Discussion
As marked in the differential diagnosis, the spore measurements of the new species resemble those of Henneguya sp. reported by Caffara et al. [14] in Italy, but differs in other parameters such as the length and width of the median polar capsule and the length of the caudal processes. The caudal processes of H. aegea n. sp. are much longer than those of Henneguya sp. found in the  Ganapati, 1941 and H. lateolabracis. However, it has been suggested that the caudal processes of myxobolids might not be a valid taxonomic character. The fact that this character is intermixed within the phylogeny of myxobolids suggests that the genetic capacity to develop spore tails exists broadly within this family, but only certain lineages express it. The irregular occurrence of caudal processes throughout the myxozoan tree of life is emphasized further by the appearance of this trait in the spores of some species of Myxobolus Bütschli, 1882 as reported for M. turpisrotundus Zhang, 2009 [24]. The close affinity between Myxobolus and Henneguya has been revealed in both morphological and phylogenetic studies [6,[24][25][26].
Identification of myxosporean parasites can be greatly assisted by their tissue/organ preference since apart from being highly host-specific, many myxosporeans also show high tissue tropism [27]. The plasmodial stages of H. aegea n. sp. were located in the heart tissue, in the bulbus arteriosus and the ventricle, although many of the spores were observed in the liver and few in the gills and kidney, likely as a result of dispersion of the parasite through the blood flow. Several myxosporean species have been found to infect fish hearts (  [22,29] and one unidentified Henneguya sp. from the gilthead seabream [14]. The plasmodia of the majority of the heart-infecting Henneguya spp. are located exclusively in the bulbus arteriosus. The plasmodia of H. aegea n. sp. were found both in the bulbus arteriosus but also in the ventricle of its host which differentiates this species from the other described Henneguya spp. To our knowledge, the only other parasite of this genus that has its plasmodia in the ventricle is Myxobolus hearti Chen, 1998 from the Prussian carp, Carassius gibelio (Bloch) [28], which, however, does not have plasmodia in the bulbus arteriosus.
With the growing molecular data that have become available in recent years, we examined the phylogenetic position of the new species, using the SSU rRNA gene as a phylogenetic marker against all sequences from congeneric species available on GenBank. In the phylogenetic tree, H. aegea n. sp. grouped with the other Henneguya spp. infecting marine species and formed a separate clade with H. cynoscioni, H. tunisiensis, H. pagri and H. mauritaniensis. The phylogeny obtained is in agreement with previously published analyses for Henneguya spp. [22,30].
Based on morphology/morphometry data, tissue tropism and comparison with other available genetic sequences (molecular phylogenetic analysis), we concluded that the species isolated in the present study is a novel species, H. aegea n. sp. One of the most interesting aspects of our study is related to the fact that the host, P. major is not native to the Mediterranean but rather introduced in the late 1980s [31]. This, together with the fact that myxobolid parasites are generally highly host-specific [32] suggests that the cultured P. major has contracted the parasite from a wild host, which remains unknown. It can be hypothesized that the original host is genetically close to P. major, more likely a member of the family Sparidae since myxobolid parasites usually infect a single host or a limited number of closely related hosts. The possibility that the novel parasite was brought to the Mediterranean Sea via the introduced host is highly unlikely. Only fertilized eggs of P. major were imported to Europe from Japan and thus H. aegea n. sp. could not be brought together since myxosporeans are not transmitted vertically. Generally, only a small fraction of the parasite species accompanies their host during its introduction to a new area [33]. Torchin et al. [33] suggested that "introduced populations are often derived from relatively small subsets of native populations (and sometimes from uninfected life-history stages), and this reduces the probability of introducing parasites along with a host species. Another potential limitation for the establishment of introduced parasites is that many parasites have complex life-cycles requiring more than one host". Both elements of this hypothesis are valid in this case since imported eggs can be seen as uninfected stages of the host and the life-cycle of Henneguya spp. involves an actinosporean alternative stage that develops in an oligochaete [34].
Despite the fact that the parasite was probably not introduced with the host, there is a clear risk that has emerged with the importation of P. major into the Mediterranean. If the native/local parasite is highly hostspecific and the introduced host is competent for that particular parasite, there is a tendency for amplification of the disease with a "spillback" onto native hosts [35]. These concepts, however, have been assessed and examined in respect to invasive or introduced host species that may interact freely in the new locality with native hosts. The situation in aquaculture differs significantly, since introduced fish species are confined in a limited space and may act as a reservoir for the parasites. An example comes from outbreaks of Kudoa amamiensis Egusa & Nakajima, 1980 in farmed Japanese amberjacks, Seriola quinqueradiata Temminck & Schlegel, introduced from northern to southern Japan, where amberjacks were assumed to be accidental hosts of the parasite that was found in several native wild fishes [36]. Future studies should focus on the identification of the natural hosts (fish and possible non-fish intermediate hosts) in the specific geographical location in order to evaluate the likelihood of the spillback hypothesis.
The spores of H. lateolabracis and H. pagri are the cause of lesions in the heart, spleen and gill tissues of the affected fish as already described by other authors [13,37]. Arteriolosclerosis is common in the animal kingdom, as is the well-known atherosclerosis in humans, and is defined as a chronic arterial change with loss of elasticity and narrowing of the lumen by proliferation of the arterial wall with degeneration [38]. It differs from atherosclerosis lesions commonly seen in humans which is characterized by inflammation, fatty degeneration and mineralization of the vessel wall and is rather a degenerative disease. The narrowing of the vessel lamina by the intima proliferations can lead to tissue infarctions and degeneration of the surrounding tissue in any organ, especially under conditions of increased stress and activity. A similar result was found in a case of the Prussian carp infected with M. hearti [28]. Although arteriosclerosis in animals leads less often to infarction than atherosclerosis in humans, and is often regarded as a minor finding in animals, we believe that the severity of the lesions in the tissues examined with complete occlusion of the arterial lumen, might be the cause or a major contributor to the increased mortalities of the diseased fish.
Micro-CT technology provides fascinating insights regarding the morphology and the location of the plasmodia in a non-destructive way. The same samples can be later used for histology or dissection in order to confirm and further study the presence of the plasmodia as in this case or in other lesions. In our investigation, we have also noticed significant differences between the healthy and the infected hearts in regard to the density of the tissue (data not shown). In a following study, we intend to study these differences in combination with histopathology.

Conclusions
This study showed that even highly host-specific parasites may affect exotic species causing disease and morbidity. The novel myxobolid parasite, H. aegea n. sp. caused lesions in the blood vessels of cultured P. major which is an introduced species in the Mediterranean. The natural host of the parasite is not known; however, based on the host specificity of Henneguya spp. and the type of pathology observed, it is expected that the presence of a novel competent host raised in aquaculture at high densities could result in spillback of H. aegea n. sp. to native fish host populations.
Additional file 1: Video S1. 3D-video rendering of the two micro-CT scans of the two infected hearts showing, in different section planes, the internal location of spore-containing plasmodia (red arrows).