The cryptogenic parasite Haplosporidium pinnae invades the Aegean Sea and causes the collapse of Pinna nobilis populations

The cryptogenic parasite Haplosporidium pinnae has caused mass mortality of the protected endemic Mediterranean bivalve Pinna nobilis in the western Mediterranean, since the autumn of 2016. Herein, we confirm the spread of the parasite in the eastern Mediterranean, and report a mass mortality event, with > 93% average mortalities, in the coastal waters of Lesvos Island (Greece, Aegean Sea). Histopathological study of collected specimens revealed the presence of a haplosporian-like protozoon in different life cycle stages, mainly within the digestive gland of the infected Pinna nobilis, with many uniand bi-nucleate parasite cells, plasmodia and sporocysts in the wide lumen of digestive tubules causing the collapse of epithelial cells, and apparently low host haemocyte reaction. The parasite was identified as H. pinnae by molecular methods (PCR amplification and sequencing of a part of small subunit ribosomal DNA gene, and comparison with available records in Genbank). In many sites, 100% mortality was recorded, whereas in a single site (among 13 surveyed sites) mortality was relatively low (36%), successful recruitment was observed and the parasite was not detected. The latter observation stresses the importance of possible parasite-free refugia sites. We call for continuous monitoring of the spread of the parasite and its impacts, and for urgent targeted research and actions to identify the factors affecting the parasite’s virulence, investigate biotic and abiotic conditions that characterize refugia sites, and strictly protect the remaining P. nobilis populations to increase the chances for the survival of the species.


Introduction
The fan mussel Pinna nobilis Linnaeus, 1758 is an emblematic Mediterranean endemic species and ranks as one of the largest bivalves in the world.It can reach an anterio-posterior length of 120 cm (Zavodnik et al. 1991) and a lifespan of 45 years (Rouanet et al. 2015).It occurs at depths < 60 m, with the apex of the shell anchored in the soft substrate by byssus threads.It typically thrives in seagrass meadows, such as Posidonia oceanica, Cymodocea nodosa, Zostera marina, and Z. noltii (Prado et al. 2014) but also in macroalgal beds (Katsanevakis and Thessalou-Legaki 2009), unvegetated sandy habitats (Katsanevakis 2006), estuarine areas (Addis et al. 2009), and even in small sandy patches within predominantly rocky habitats (Tsatiris et al. 2018).It plays a key ecological role by filtering water and retaining large amounts of organic matter from suspended detritus, thus reducing turbidity (Trigos et al. 2014), providing hard substrate colonized by numerous epibionts in soft-bottom habitats (Addis et al. 2009;Rabaoui et al. 2009), and, at high densities, acting as an ecosystem engineer able to create biogenic reefs (Katsanevakis 2016).
P. nobilis has been exploited since antiquity for its meat and byssus from which sea silk was produced (Maeder 2008).However, during recent decades its populations have suffered substantial declines (Basso et al. 2015), and the species has been assigned a strict protection status under the EU Habitats Directive (92/43/EEC, Annex IV), the Protocol for Specially Protected Areas and Biological Diversity in the Mediterranean of the Barcelona Convention (Annex II), and the national legislation of most Mediterranean countries.Despite its protection, in many Mediterranean countries it is still illegally exploited (Katsanevakis 2007) and even served in restaurants (Katsanevakis et al. 2011).In addition to human exploitation, the species is threatened by many cumulative stressors, such as the loss of seagrass beds, marine pollution, boat anchorage, destructive fishing practices, invasive species, and climate change (Basso et al. 2015).
Since autumn 2016, a new threat has been added to this long list of cumulative stressors.A new parasite of unknown origin, named Haplosporidium pinnae by Catanese et al. (2018), has caused a mass mortality event (MME) in the western Mediterranean, apparently due to heavy inflammatory host response and severe general dysfunction.By June 2017, about 90% of the Spanish P. nobilis populations were lost, and mortality rates reached 100% in the south and central Mediterranean coasts of the Iberian Peninsula and the Balearic Islands (Vázquez-Luis et al. 2017).Thereafter, severe mortality in fan mussel populations was also observed in the Italian and French coasts (Catanese et al. 2018).As a response to this MME, fan mussel has been declared as critically endangered in Spain.H. pinnae seems to be highly species-specific, as the congeneric species P. rudis was not affected in the impacted areas (Vázquez-Luis et al. 2017).
The protist order Haplosporida includes over 50 described species classified in the genera Haplosporidium, Minchinia, Bonamia and Urosporidium (Arzul and Carnegie 2015;Azevedo and Hine 2017).These species parasitize aquatic invertebrates and can be highly pathogenic.For example, the haplosporidans Haplosporidium nelson, Bonamia ostreae, and B. exitiosa have been reported to cause mass mortalities to various oyster species (Ford and Tripp 1996;Engelsma et al. 2014).
The molecular and phylogenetic analysis conducted by Catanese et al. (2018) revealed that the parasite infecting P. nobilis is clearly included in the Haplosporida group and is closely related to another haplosporidan parasite causing high mortality in the cultured shrimp Penaeus vannamei in the Caribbean Sea (Nunan et al. 2007) and Indonesia (Utari et al. 2012).Nevertheless, they identified many differences between the two parasites, which appeared to be a separate group from other Haplosporidium, Minchinia and Bonamia species, and suggested that these two species may represent a distinct new genus within the order Haplosporida.
It is quite probable that this parasite is an alien species in the Mediterranean Sea, introduced in 2016 or shortly before.According to Catanese et al. (2018) "the scarce variability detected in the SSU rDNA sequence of H. pinnae could suggest a recent arrival of this parasite in the Spanish Mediterranean coast, which would be consistent with the unprecedented recent MME".However, in the absence of evidence on the native range of the species, definite assessment of its biogeographic status is de facto impossible, and thus the species should be classified as cryptogenic (Essl et al. 2018).
In 2016-2017, there was no record of unusual fan mussel mortality in the eastern Mediterranean.But by mid-2018, anecdotal information of mass mortalities in the Aegean Sea alarmed marine scientists.This study is the first attempt to assess the status of fan mussel populations in a case study area in the Aegean Sea by means of underwater visual surveys, and to investigate through histological and molecular analyses of sampled fan mussels whether the parasite H. pinnae has expanded its range in the eastern Mediterranean, infecting P. nobilis populations.

Study area
The study was conducted in Lesvos Island, northern Aegean Sea.Lesvos Island is the third largest island of the Aegean Archipelago, with a surface area of 1633 km 2 .Two elongated, shallow and semi-enclosed bays characterize the shape of the island, the Kalloni and Gera gulfs (Figure 1).These gulfs are highly productive marine ecosystems with rich biodiversity, and for this reason they are included in the Natura 2000 European network of protected areas (area code for Kalloni Gulf: SCI GR4110004, and for Gera Gulf SCI GR4110013).Both gulfs have supported large P. nobilis populations (Tsatiris et al. 2018, for Gera;unpublished data, for Kalloni).

Assessment of P. nobilis mortality rates
A total of thirteen sites were investigated through underwater visual surveys along the coastal waters of Lesvos Island between August and October 2018 (Figure 1, Table 1), to verify anecdotal reports of extensive   mortalities of P. nobilis populations.The sites were selected on the basis of existing information on the presence of P. nobilis populations.Three of these sites were located in Gera Gulf at locations where previous research (in the summer of 2016) reported very high population densities (Tsatiris et al. 2018), and two in Kalloni Gulf, where P. nobilis used to have very high densities as well (unpublished data).Two sites (sites 4 and 6 in Table 1) were surveyed again three months after the initial sampling (i.e. in December 2018); these were the sites in which specimens were collected for histopathology and molecular analyses (see below).
Surveys to assess mortality rates involved the counting of live and dead individuals.Live individuals were further assessed as "presumed healthy" (apparently unaffected) and "presumed moribund".Each survey lasted 45 to 60 minutes.Presumed moribund individuals were discriminated from healthy ones based on a delayed valve-closing reflex (Catanese et al. 2018;Vázquez-Luis et al. 2017).Some of these possibly infected individuals were observed not to be able to completely seal their bivalve shell.Only freshly dead individuals were counted; empty shells that presumably died more than a year ago (as evaluated on the spot by the level of fouling) and shells that were not anchored were not counted as "dead".The selected depth range of the surveys was between 1-10 m at soft bottom habitats, seagrass meadows, or a mixture of these habitats (Table 1).

Collection of specimens for histopathology and molecular analysis
Surveys confirmed the occurrence of a MME of fan mussel populations in Lesvos Island, and thus it was decided to investigate the possible expansion of the parasite H. pinnae in the Aegean Sea.For this reason, a limited number of presumed moribund and healthy individuals were collected, transported to the lab and prepared for histopathological and molecular analyses.P. nobilis is a strictly protected species, so, in order to collect specimens for laboratory analysis, a permission license for research sampling was requested and acquired from local authorities.One presumed healthy and two presumed moribund individuals were collected from each of two of the surveyed sites (Figure 1), one with high mortality rates in Gera Gulf (site 4; Table 1), and one with a high percentage of live individuals in Kalloni Gulf (site 6; Table 1).Collected specimens were placed in a cooler container and were immediately transferred to the laboratory for dissection.

Histopathological study
Samples for histopathology were collected from the following tissues of both presumed healthy and infected specimens from both collection sites: muscle, byssus gland, digestive gland, gills, mantle and gonads.Samples were approximately 2 × 2 cm and after dissection they were immediately placed in 10% phosphate buffered formalin for fixation.Formalin-fixed tissues were processed routinely, following the procedure of Drury and Wallington (1980).Briefly, samples were dehydrated in a series of alcohol solutions starting from 70% to 100% and then tissues were embedded in paraffin.Paraffin-embedded tissues were cut in slices 3-5 μm thick using a Leica RM2125 RT microtome.Tissue sections were placed on glass slides and after rehydration were stained with Haematoxylin-Eosin and Giemsa methods (Roberts 1989).Tissue slides were examined with a Nikon phase contrast microscope coupled with a Microcomp integrated digital imaging analysis system (Nikon Eclipse 80i, Nikon Co., Tokyo, Japan), used also for photography.

Molecular analyses
Molecular tests for Haplosporidium pinnae detection were conducted with PCR amplification and sequencing of a part of small subunit ribosomal DNA (SSU rDNA) gene, which has been used by Catanese et al. (2018) for molecular characterization of this new parasite species.Duplicated samples from specimens and tissues that had been dissected for histological analyses were kept in pure ethanol at −20 °C for molecular analyses (Supplementary material Table S1).In order to avoid "sampling effect" and subsequent false negative results, each tissue sample (30-350 mg) was cut into replicate pieces which were then processed independently (Table S1).Initially, replicates were chopped with sterile scissors and washed out (incubation for 15 min, centrifugation at 13000 g for 2 min, and discarding of supernatant) two times with 800 μL distilled sterile water and three times with 600 μL lysis buffer (0.5 M Tris, 0.1 M EDTA, 2% SDS, pH 8.8) according to Darriba (2017).After the last wash, the tissue was treated with 10 μL proteinase K (20 mg/ml) in 600 μL lysis buffer and incubated at 55 °C overnight.Total DNA was extracted following a standard isopropanolammonium acetate (5M) precipitation protocol (Sambrook et al. 1989).DNA pellet was dissolved in 50-200 μL pure water (depending on the size of pellet) and concentration of extracted total DNA was estimated with NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific).Two PCR amplifications were performed for each DNA sample using the forward primer HPNF3 (Catanese et al. 2018) paired with reverse primers HPNR3 (Catanese et al. 2018) and 18S9R (Medlin et al. 1988) respectively.The 20 μL reaction mix included 4 μL KAPA HiFi Fidelity buffer (5x), 0.8 μL KAPA dNTP Mix (10 mM), 1 μL from forward primer (10 μM), 1 μL from reverse primer (10 μM), 1 μL KAPA HiFi HotStart DNA polymerase (1 U/μL) and 50-80 ng DNA.The PCR cycling protocol consisted of an initial denaturation step of 5 min at 98 °C, followed by 35 cycles of 20 s at 98 °C, 30 s at 55 °C (for HPNF3/ HPNR3) or 57 °C (for HPNF3/18S9R) and 30 s at 72 °C, followed by a final extension step of 5 min at 72 °C.Negative controls were used in each PCR in order to check for possible contamination.PCR products were purified with a standard ethanol precipitation protocol and sequenced using the HPNF3 primer and the BigDye Terminator version 3.1 Cycle Sequencing Kit (Applied Biosystems Inc.) on an ABI 3730 capillary sequencer.The retrieved sequences were edited with MEGA 6.06 (Tamura et al. 2013) and were re-checked by visual inspection of raw fluorogram data.The Blastn suite (https://blast.ncbi.nlm.nih.gov/Blast.cgi) was used in order to check for similarity of our sequences with the publicly available haplotypes of H. pinnae or other parasites in GenBank.

Mortality rates
During the visual surveys in Lesvos Island, between August and October 2018, a total of 950 P. nobilis individuals were recorded, of which 886 were dead, and 64 live (Figure 1; Table 1).These numbers indicate an average mortality rate of 93% in the surveyed sites of Lesvos Island.
The present assessment revealed the collapse of the previously thriving population of Gera Gulf (estimated at ~ 215,000 individuals in 2016; Tsatiris et al. 2018), with mortality rates ranging between 91% and 100%.The northern (inner) site of Kalloni Gulf (site 6; Table 1) was the only site where the ratio between live and dead P. nobilis individuals appeared to be not very far from normal, with a 36% mortality rate.However, at the southern part of Kalloni, a dense population thriving in shallow waters up to 1.5 m deep had a 99% mortality rate.At the remaining sites, which were situated outside the two semi-enclosed gulfs, 100% mortality was recorded in seven of them, while in one site three moribund specimens were found showing an extremely slow valve-closing reflex.
In early December 2018, mortality rates in site 6 were estimated at 20% (94 live and 23 dead individuals were detected), substantially lower than three months earlier.P. nobilis shell fragments of various sizes were observed at the upper infralittoral zone limit (at depths of 0.1-0.2m) or on the beach, probably mainly dead individuals stranded by storms, as dead individuals lack the anchorage of the byssus threads and can be uprooted more easily than live individuals.Thus, the lower estimates of mortality rates are probably related to the stranding of dead individuals.A prominent number of juvenile individuals (ten individuals of age class 0+) was recorded among the live fan mussels, with only one recorded dead juvenile.At site 4 in Gera Gulf no live individuals were detected in the December survey (among 130 detected dead individuals), indicating that the MME was still in progress in early September when the samples were collected.

Necropsy observations
The length and width of collected specimens varied between 26-34 cm and 9-11 cm respectively.In all individuals, the symbiotic decapod Nepinnotheres pinnotheres was found in the shell cavity.Macroscopic observation of presumed infected specimens from Gera Gulf showed emaciation and softness of the visceral mass that had a watery appearance.Characteristic was the dark coloration of the digestive gland of infected specimens in comparison to the presumed healthy individuals.Gills showed a normal appearance, but the mantle was retracted and folds were black.

Histopathological study
Histopathological study revealed the presence of a haplosporidan-like protozoon in different life cycle stages within the digestive gland of the two infected Pinna nobilis individuals from Gera Gulf, whereas no visual observation of parasites was evident in the samples from Kalloni Gulf.Many uni-and bi-nucleate parasite cells, plasmodia and sporocysts were found in the wide lumen of digestive tubules causing the collapse of epithelial cells (Figure 2A, B).The presence of protozoan cell stages in the connective tissue of the digestive gland was less severe.Host haemocyte reaction appeared to be moderate, little or absent.Moreover, few sporocysts were found in intramuscular connective tissue (Figure 3) of an individual Pinna nobilis with a very heavy infection in its digestive gland.
All presumed healthy Pinna nobilis specimens (controls) had a normal histological structure of their organs and tissues without any visual identification of protozoan or metazoan parasites.

Molecular analysis
All individuals from Gera Gulf (one healthy and two moribund) provided amplicons for both primer pairs.With the exception of DNA extracted from the gonads of the healthy individual, DNA samples from every other tissue gave positive PCR results (Table S1).For each positive sample we aligned the retrieved amplicon sequences and obtained final concatenated sequences of good quality that ranged in length from 494 to 707 bp (Supplementary material file "Sequences_alignment.fas").All sequences were found to be identical (100% similarity after check with Blastn) with the only available record of H. pinnae SSU rDNA gene (Accession number: LC338065.1)submitted in Genbank by Catanese et al. (2018).These results provide a direct evidence that the haplosporidan-like protozoon which had been observed in histological analyses is actually H. pinnae.In Kalloni Gulf, no PCR amplification was achieved in any of the tissue samples, neither from the healthy nor from the presumed moribund individuals.

Discussion
The results of this work provide direct evidence on the collapse of the P. nobilis populations in Lesvos Island due to the spread of the infectious parasite H. pinnae.The high mortality levels observed, ranging between 90-100% in most sites, rank this episode as an MME (Fey et al. 2015), which has affected all surveyed populations at the Island of Lesvos, including two of the largest populations recorded in Greece (in Gera Gulf, Tsatiris et al. 2018, and Kalloni Gulf, unpublished data).Although our survey was restricted to Lesvos, anecdotal reports in social media suggest that MMEs have occurred in most parts of the Aegean Sea, as well as other parts of the eastern Mediterranean (i.e., Turkey and Cyprus), probably due to the spread of H. pinnae.Such large scale MMEs pose a major threat to the conservation of the species both at a local/national and at a pan-Mediterranean scale.
Gross observations during necropsy of the infected individuals were characteristic of this parasitic disease, mainly concerned the digestive gland and the mantle, and are in agreement with the findings of Darriba (2017) in the western Mediterranean.Microscopic observation of tissue samples revealed heavy infestation especially of the digestive gland and its digestive tubules.Lower numbers of parasites were observed in the connective tissue of the digestive gland and in intramuscular connective tissue.Such findings are in agreement with reports by both Darriba (2017) and Catanese et al. (2018).An alarming finding of our histopathological study confirming the observations of the latter authors is the very low or in some cases absent reaction of host heamocytes, which may be responsible for the MME recorded all over the Mediterranean.
Molecular identification of H. pinnae in specimens from at least one site in Lesvos Island, confirms the appearance of this infectious parasite in the Aegean Sea (and the eastern Mediterranean) for the first time.Positive PCR amplification in individuals or tissues that were identified as "parasite-free" by histological analysis, points out that the more sensitive PCR assay and subsequent sequencing offer a better chance of detecting low intensity infections than do traditional techniques (Stokes et al. 1995).Negative results of molecular analysis in Kalloni Gulf are in accordance with histological examination and the mortality assessment survey results, and may be indicative of absence of infection in this specific location or the specific sampled individuals.
The scarcity of baseline data on the density and population structure of P. nobilis in the Greek waters of the Ionian and Aegean seas render it relatively difficult to assess the magnitude of the problem at a national level.However, considering the various threats posed to the survival of the species (Katsanevakis 2007;Katsanevakis et al. 2011;Basso et al. 2015), along with the alarming rate of this mass mortality outbreak (Vázquez-Luis et al. 2017; this study; and other unpublished records), an emergency response that will include several precautionary measures for parasite-free areas, is urgently required.As initially suggested for the Spanish populations by Vázquez-Luis et al. (2017), P. nobilis should be given a protection status of "Critically Endangered", both at a national and at a Mediterranean level, at least until the parasite infections subside and populations show signs of recovery.
Systematic surveys and monitoring are needed to define the spatiotemporal progress of the parasite impact, and to assess the factors that trigger and/or enhance H. pinnae infestation.Previous studies on the occurrence of MMEs on various species in the Mediterranean Sea suggest that there is a link between these events with atypically high seawater temperatures and prolonged stratification of the water column during the late summer months (e.g.Cerrano et al. 2000;Garrabou et al. 2009).These conditions induce extended physiological stress, which reduces the resistance of the host organisms to proliferating pathogens (Cerrano et al. 2000).Similarly, high seawater temperatures during the summer months may have lowered the physiological capacity of P. nobilis to resist parasite infections due to a high respiratory demand at increased temperatures (Trigos et al. 2015), which subsequently enhanced parasite manifestation.Unfortunately, the environmental cues that have triggered H. pinnae proliferation and virulence are currently unknown, and cannot be easily inferred from closely related parasite species, as haplosporidans flourish under contrasting environmental conditions (Arzul and Carnegie 2015).Undoubtedly, further investigation is needed to identify and, if possible, control the factors that enhance H. pinnae survival, dispersion and virulence.Species-specific experiments which test the physiological responses of host and parasite organisms to combinations of biotic and/or abiotic factors will provide valuable information to support mitigation actions.
Additional field surveys in different areas and depth zones may also reveal the existence of natural refugia, i.e. areas where healthy, unaffected populations of P. nobilis are still thriving.Our results indicate that such a refuge may be available at the site located at the inner part of the Kalloni Gulf.In this site, mortality rates were relatively low, successful recruitment was observed, and no H.pinnae parasites were detected in the P. nobilis collected specimens.The lack of the parasite could be indicative of the dominance of some site-specific conditions, which may either control parasite survival and prevalence (Arzul and Carnegie 2015) or increase host health and vigour thus making fan mussels more resistant to infections.Salinity and temperature in Kalloni follow a longitudinal gradient from the sea to the inner zone, which is positive in the hot season and negative in the cold season, i.e. the inner sites of the gulf have higher temperatures and salinities than the seaward sites during summer, and lower temperatures and salinities than the seaward sites during winter (Millet and Lamy 2002).Hence, the annual variability of temperature and salinity in the inner sites is much higher than that of the seaward sites.Salinity in site 6 has been reported to range between 36 psu in spring and 45 psu in autumn, substantially affected by the neighbouring extended solar saltworks (Evagelopoulos and Koutsoubas 2008).It is possible that these special environmental conditions of site 6 have contributed to the reduced virulence of the parasite, which has caused nearly complete mortality in the seaward areas of the gulf.
However, no solid conclusions can be based on a single site of high survival.Clearly, a more dedicated large-scale study is needed to assess the existence of parasite-refugia in Kalloni Gulf and elsewhere.Nevertheless, it is important to highlight that such areas are vital to the survival of the species, as they represent pockets of parasite-resistant populations, and should be strictly protected as soon as possible.Besides, these refugia will play a significant role for the restoration of fan mussel populations by acting as sources either of recruits to other areas or of healthy adult individuals for transplantation management actions (Katsanevakis 2016) into suitable areas of predicted low parasite virulence.However, a better understanding of the environmental and genetic connectivity between different populations is of utmost importance for future transplantation efforts, in order to ensure that sufficient genetic variability is maintained to increase the chances of survival of the transplanted populations, as well as of the P. nobilis species as a whole (González-Wangüemert et al. 2018).Comparisons between parasite-infected and parasite-free areas may also provide additional information on the environmental factors that trigger or delay the outbreak, and help clarify the past and present relation between H. pinnae and its P. nobilis host (i.e.symbiotic versus infectious parasitism; Darriba 2017).
Overall, it is important to note that in all collected individuals the distinction between healthy (presumed to be unaffected) and moribund (at an advanced stage of infection) individuals by experienced diving scientists during in situ visual inspection has proved to be problematic, as H. pinnae was not detected in the presumed infected individuals from Kalloni Gulf, whereas it was detected in the presumed healthy individual from Gera Gulf.Similar to other bivalve molluscs, the fan mussel's speed of response to external stimuli can be related to a number of factors that affect bivalve behaviour and physiology, other than parasite infections (e.g.El-Shenawy 2004;García-March et al. 2008).Hence, reports regarding visual observations of moribund individuals should be treated with caution and be substantiated with auxiliary information.For example, supplementary data on approximate number of individuals present within a given area, the corresponding percent mortality, and the existence of other potential causes of mortality (e.g.uprooted or broken shells due to mechanical damage), will be useful to define the present conditions of a given site, especially when relevant information is provided by non-experts / citizen scientists.Nevertheless, the engagement of citizen scientists in the data collection process should be promoted, as it has been proven to be very useful in the detection of MMEs over large spatial scales (e.g.Vázquez-Luis et al. 2017;Jones et al. 2018).Once appropriately trained to follow simplified procedures in a systematic manner that can be easily validated, citizen scientists can be a vital asset in all steps of the MME-response process; from the initial detection of the problem to the participation in mitigation and monitoring actions (Foster-Smith and Evans 2003;Cerrano et al. 2017).
Finally, the control of illegal exploitation and marketing of fan mussels must be intensified and reinforced by respective awareness raising campaigns that will inform stakeholders and consumers about the magnitude of the problem and the additional risks involved when collecting individuals from affected and unaffected populations, or when transferring parasite-infected individuals to healthy areas and zones.This is of particular importance for regions such as the Aegean Sea, where historically dense populations of P. nobilis have suffered large declines due to poaching, while legislative measures that prohibit the illegal trading and selling of fan mussels are not being effectively implemented (Katsanevakis et al. 2008(Katsanevakis et al. , 2011)).

Figure 1 .
Figure 1.Mortality assessment of Pinna nobilis population, at thirteen sites at Lesvos Island.The size of the pie charts represents the number of recorded individuals of fan mussels.Red and green color of the pie charts, show the proportion of dead and live individuals, respectively.Yellow asterisks indicate sites where specimens were collected for histological and molecular analysis.

Table 1 .
Number (N) of dead and live (presumed healthy or moribund) Pinna nobilis individuals, along with site-specific information of the thirteen sites investigated in Lesvos Island.Sites where P. nobilis individuals were collected for further analyses are marked with an asterisk (*)."Mixed habitat" refers to a mixture of Posidonia oceanica tufts and soft substrate patches.

Figure 2 .
Figure 2. Histopathological sections through the digestive gland of Pinna nobilis infected with the haplosporidan-like protozoon.A: Α digestive tubule fully occupied by abundant stages of the protozoon and complete destruction of its epithelium (Haematoxylin-Eosin stain, x400); B: A digestive tubule with severe occupation by protozoan cell stages, resulting in a dilated lumen and collapsed epithelial cells (Giemsa stain, x400).Photomicrographs by Dr D. Doukas.