lable at ScienceDirect

Fish & Shellfish Immunology 39 (2014) 42e50

Contents lists avai

Fish & Shellfish Immunology

journal homepage: www.elsevier .com/locate/ fs i

Full length article

Histopathology and stress biomarkers in the clam Venerupisphilippinarum from the Venice Lagoon (Italy)

Michele Boscolo Papo a, Daniela Bertotto a, Francesco Quaglio a, Marta Vascellari b,Francesco Pascoli b, Elena Negrato a, Giovanni Binato b, Giuseppe Radaelli a,*aDipartimento di Biomedicina Comparata e Alimentazione, Università degli Studi di Padova, Agripolis, Viale dell’Università 16 e 35020 Legnaro (PD), Italyb Istituto Zooprofilattico Sperimentale delle Venezie, Viale dell’Università 10 e 35020 Legnaro (PD), Italy

a r t i c l e i n f o

Article history:Received 29 January 2014Received in revised form9 March 2014Accepted 23 April 2014Available online 2 May 2014

Keywords:ClamHistologyPerkinsosisHaemocytesStress biomarkers

* Corresponding author.E-mail address: giuseppe.radaelli@unipd.it (G. Rad

http://dx.doi.org/10.1016/j.fsi.2014.04.0161050-4648/� 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

The aim of this study was to evaluate the histomorphology and the stress response in the bivalve Ven-erupis philippinarum sampled in four differently polluted sites of the Venice Lagoon (Palude del Monte,Marghera, Ca’ Roman and Val di Brenta). This species is often used as bioindicator of environmentalpollution since it can bioaccumulate a large variety of pollutants because of its filter feeding. Chemicalanalyses for heavy metals (Cd, Cu, Hg and Pb) and polycyclic aromatic hydrocarbons (PAHs) were per-formed on whole soft tissues of V. philippinarum. The histological evaluation of clams revealed thepresence of Perkinsus sp. infection in animals from all sites, although a very high prevalence of parasiteswas evidenced in clams from Ca’ Roman. Perkinsus sp. were systemically distributed in the mantle, in theintestine and digestive gland, in gonads and gills. The trophozoites of Perkinsus sp. were found isolated orin cluster surrounded by a heavy hemocitical response. Haemocytes always exhibited an immunoposi-tivity to cytochrome P4501A (CYP1A), heat shock protein 70 (HSP70), 4-hydroxy-2-nonenal (HNE) andnitrotyrosine (NT) antibodies. The digestive gland of animals from Palude del Monte showed the highestmalondialdehyde (MDA) concentration, whereas clams from Ca’ Roman exhibited the highest quantity ofmetallothioneins.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The Venice Lagoon is a transitional environment on the Adriaticcoast influenced by such human activities as agriculture, industryand tourism. The sludge of Venice and the rivers from the hinter-land pour into the lagoon, where sediments trap pollution.

For this reason, in recent years numerous national and inter-national projects have been carried out in order to evaluate thequality status of the lagoon environment. Biomonitoring pro-grammes usually involve the use of biomarkers, which representbiochemical, physiological or behavioural variations measured intissues, biological fluids or whole organisms [1,2]. Several verte-brate and invertebrate species are employed in marine monitoringprogrammes as sentinel models to evaluate environmental quality.Among invertebrates, the Manila clam Venerupis philippinarum haslargely been used to investigate the water/sediment pollution incoastal lagoon ecosystems since it is a filter-feeding bivalve living in

aelli).

soft bottoms [3e8]. Moreover, it represents an important economicresource for fisheries in North Adriatic lagoons, where this speciesis fished and farmed.

At the cellular level, the metabolism of environmental stressorsfrequently results in the formation of reactive oxygen species (ROS)[9]. They are produced naturally during metabolism and their toxiceffects are usually prevented by antioxidants, both molecular andenzymatic ones. During oxidative stress conditions, the productionof ROS is greater than the ability of cells to remove them, leading tolipid peroxidation, protein carbonils formation and cell death[10,11]. Lipid peroxides are unstable indicators of oxidative stress incells that decompose to form more complex and reactive com-pounds such as malondialdehyde (MDA), which is a naturalbyproducts of lipid peroxidation [12]. MDA originates from theoxidative degradation of PUFAs and represents a highly toxicaldehyde with a specific affinity to proteins and DNA [12].

4-hydroxy-2-nonenal (HNE), the most abundant and toxic a,b-unsaturated aldehyde, originates from the b-cleavage of hydro-peroxides from u-6 PUFAs and is mainly involved in the inhibitionof protein and DNA synthesis, in the inactivation of enzymes, and isalso a potent mutagen agent [12]. Moreover, one of the most

Fig. 1. Map of the Venice Lagoon indicating the location of sampling station: Paludedel Monte [45.28.59 N; 12.21.15 E], Marghera [45.25.55 N; 12.16.09 E], Ca’ Roman[45.14.28 N; 12.16.55 E] and Val di Brenta [45.11.50 N; 12.15.38 E].

M. Boscolo Papo et al. / Fish & Shellfish Immunology 39 (2014) 42e50 43

important ROS is the superoxide radical, which reacts with nitricoxide giving rise to peroxynitrite, a potent oxidant that may oxidizeproteins, lipids and DNA [13]. Nitrotyrosine (NT) is a relativelystable marker for peroxynitrite production [13].

Heat shock proteins (HSPs), also called stress proteins, are afamily of highly conserved cellular proteins that are present in allcells in all life forms [14e16]. An evident biomarker role was shownto be played by Heath Shock Protein 70 (HSP70) that protects cellsagainst harmful conditions by binding and refolding damagedproteins. There are constitutive members (HSC70) of the heat shockproteins, which play important chaperoning role in unstressedcells, and inducible (HSP70) forms, which are expressed indetectable levels after acute stressor insults [14,17]. In aquaticspecies, the expression of HSP70 has been studied in fish afterexposure to heat shock, pesticides, virus, metals and other toxiccompounds [18e21]. In mussel, an increased expression of HSP70has been detected after exposure to contaminants [22,23]. More-over, an upper regulation of HSP70 has been observed in the clamVenerupis decussata upon Perkinsus olseni infection [24].

The cytochrome P4501A (CYP1A) subfamily is involved in thebiotransformation of a variety of contaminants like poly-chlorodibenzo dioxins (PCDDs), polycyclic aromatic hydrocarbons(PAHs), polyhalogenated aromatic hydrocarbons (PHAHs) andpolychlotobiphenils (PCBs). Its induction plays a central role intransforming pesticides in aquatic organisms [25].

Metallothioneins (MTs) are ubiquitary proteins that have beenfound in a very wide range of organisms, including vertebrates,invertebrates, plants and bacteria [26]. MTs exhibit high affinity formetals, giving rise to their important role in metal metabolism,detoxification of heavy metals, immune response and the antioxi-dant process [27,28]. The use of (MTs) seems to be recognized as themost valid method to indicate metal exposure. In fact MTs areinduced by metals residues and measurements of their levels are atpresent used in both vertebrates and invertebrates.

The aim of this study was to evaluate the histomorphology ofthe different tissues and organs of the bivalve V. philippinarumsampled in four differently polluted sites of the Venice Lagoon:Palude del Monte, Marghera, Ca’ Roman and Val di Brenta [29]. Wealso used an immunohistochemical approach to detect the locali-zation of the following oxidative stress biomarkers: CYP1A, HSP70,HNE and NT. Chemical analyses for heavy metals (Cd, Cu, Hg andPb) and PAHs were performed on whole soft tissues of clams.Moreover, to test the amount of lipid peroxidation, we evaluatedthe MDA concentration in digestive gland using the thiobarbituricacid reactive substances (TBARS), a well-established assay forscreening and monitoring lipid peroxidation. The amount of met-allothioneins, as a biomarker of heavy metal exposure, was esti-mated in digestive gland by a spectrophotometric method.

2. Materials and methods

2.1. Organisms

The clams V. philippinarumwere collected at four sites of VeniceLagoon (northern terminus of the Adriatic Sea) characterized bydifferent pollution levels. All the clamswere adult organismswith ashell size of 4.03 � 0.3 cm. The four sites were: Palude del Monte[45.28.59 N; 12.21.15 E], Marghera [45.25.55 N; 12.16.09 E], Ca’Roman [45.14.28 N; 12.16.55 E] and Val di Brenta [45.11.50 N;12.15.38 E] (Fig. 1). For each site, 50 animals were caught, imme-diately transferred in portable fridge at 4 �C to the laboratory andprocessed for the analysis. For histology and immunohistochem-istry, the whole body from 40 animals (10 clams/site) was fixed in4% paraformaldehyde prepared in phosphate-buffered saline (PBS,0.1 M, pH 7.4) at 4 �C overnight, washed in PBS, dehydrated through

a graded series of ethanol and embedded in paraffin. Consecutivesections were cut at a thickness of 4 mm using a microtome.

For the chemical analysis the whole soft body from 40 animals(10 clams/site) was immediately frozen in liquid nitrogen andstored at �80 �C until analysis.

For the TBARS assay and metallothionein determination,digestive glands were grouped in three pools of 10 clams per siteand stored at �80 �C.

2.2. Chemical analysis

2.2.1. PAH determinationChemical analysis was performed on whole soft tissues of V.

philippinarum. Extraction was carried out using SampliQ BufferedQuEChERS AOAC Extraction kit (Agilent, CA, USA). An amount of 3 gof homogenized clam sample was weighed in a 50 ml centrifugetube; before extraction an Agilent Ceramic Bar Homogenizer wasadded. To sample 10 ml deionized water and 12 ml acetonitrilewere added; after each solvent addition the sample was shaked for15 min. After addition of salts (6 g anhydrous MgSO4 and 1.5 ganhydrous NaOAc) tubes were vigorously shaked for 1 min, andthen centrifuged at 4000 rpm for 10 min. Clean-up was performedby means of Agilent Dispersive 15 ml SPE Fatty Sample AOAC kit. A4 ml aliquot of the previous upper organic extraction layer wastransferred into a dispersive SPE 15 ml tube containing salts(400 mg PSA/400 mg C18 EC/1200 mg anhydrous MgSO4). Subse-quently tubes were vigorously shaked for 1 min and then centri-fuged at 4000 rpm for 10 min. After filtration with disposablesyringe filter (Millipore) the extract is ready for analyticaldetermination.

HPLC-FLD analyses were carried out by means of an AllianceEmpower HPLC system equipped with a 2475 Multi l fluorimetricdetector a sample manager and a quaternary solvent manager(Waters, MA, USA). The column was a Supelcosil LC-PAH150 � 3 mm 5 mm HPLC column with a Supelguard LC-1820 � 3mm guard column (Supelco, PA, USA), kept at 30 �C. Theflow rate was 0.5 ml min�1 with an injection volume of 30 ml. Thesample manager was maintained at 30 �C.

M. Boscolo Papo et al. / Fish & Shellfish Immunology 39 (2014) 42e5044

Two spiked (2 mg kg�1) blank samples and one blank samplewere included in each analytical batch. The quantification of ana-lytes was carried out using a matrix-matched calibration curve inthe concentration range of 0.225e7.5 mg L�1.

2.2.2. Metal determination (Cd, Cu, Hg and Pb)The concentration of selected metal elements was determined

in the whole soft tissues of clams. Following homogenization andacid digestion (HNO3) in Teflon liners by means of a CEM (Mat-thews, NC, USA) Mars Xpress microwave oven, the metal elementswere determined by Atomic Absorption Spectrometry (AAS) withflame atomisation (F-AAS) for Cu (wavelength 324.8 nm) andelectrothermal atomisation (graphite furnace GF-AAS) for Cd(wavelength 228.8 nm) and Pb (wavelength 283.3 nm). Analyseswere performed by means of a Thermo Electron M6 mkII AtomicAbsorption Spectrometer, with D2 and Zeeman backgroundcorrection, equipped with flame burner and GF95 Graphite Furnaceatomiser (Thermo Scientific, Waltham, MA, USA).

For Hg determination an Altec (Prague, Czech Republic, EU) TDAAAS direct analyser AMA254 was employed at the followinganalytical conditions: wavelength 253.6 nm, accumulation time200 s, drying time 150 s and decomposition time 45 s.

Limits of quantification (LOQs) of the employed analyticalmethods are hereby reported: Pb 0.03 mg/kg; Cd 0.01 mg/kg; Cu0.10 mg/kg and Hg 0.002 mg/kg. Trueness of analytical data wasverified by means of Certified Reference Materials (BCR185R, NRCCDORM3) analysed concurrently with samples in each analyticalbatch.

2.3. Histology

Dewaxed sections including all internal organs from each clamwere stained with Haematoxylin and Eosin (H&E) and Periodic AcidShiff (PAS). Clams histological sections were examined by lightmicroscopy for general morphology, parasite presence and histo-logical alteration.

2.4. Immunohistochemistry

Immunohistochemistry was carried out by an automatedimmunostainer (Autostainer link 48 Dako, Italy). Sections weredeparaffinized in xylene, rehydrated in graded ethanol and rinsedin distilled water. Heat-induced antigen retrieval was performed in10 mM citrate buffer (pH ¼ 6.0) at 97 �C for 15 min. Endogenousperoxidases were neutralized by incubating the sections with theEnVision FLEX Peroxidase-Blocking Reagent (SM801, Dako), for10 min at RT; serial sections were incubated overnight at þ4 �Cwith the following primary antibodies: mouse polyclonal HSP70antiserum, dilution 1:200 (Stressgen Biotechnologies, USA), rabbitpolyclonal CYP1A antiserum, dilution 1:500 (Biosense laboratories,Norway), mouse monoclonal NT antiserum, dilution 1:1000 (Gen-eTex, Inc, USA), mouse monoclonal HNE antiserum, dilution 1:100(Abcam, UK). Sections were then incubated with the detectionsystem EnVision FLEX/HRP (Dako), whereas the EnVision FLEXSubstrate Buffer EnVision FLEX DAB (Dako) was used as chromogen.The sections were then counterstained with the EnVision FLEXHaematoxylin (Dako). The specificity of the immunostaining wasverified by incubating the sections with PBS instead of the specificprimary antibody.

2.5. TBARS assay

MDA content was evaluated in three pools of 10 digestive glandsof V. philippinarum. MDA is the most abundant reactive carbonylcompounds derived from polyunsaturated fatty acid peroxidation.

The amount of lipid peroxidation in the digestive gland of the clamswas assessed by measuring thiobarbituric acid-reactive substances(TBARS) according to Yoshida et al. (2005) [30]. Thiobarbituric acidreaction was carried out by mixing 0.2 ml sodium dodecyl sulphatesolution (8.1%, w/v), 1.5 ml acetic acid buffer (20%, v/v, pH 3.5),1.5 ml thiobarbituric acid (1%, v/v), 0.775 ml water and 0.05 mlethanol containing butylated hydroxytoluene (0.8 wt.%, w/v) with200 mg digestive gland. The reaction mixture was incubated at100 �C for 60 min and then cooled in ice followed by mixingvigorously with 1ml water and 5ml of n-butyl alcohol and pyridine(15/1, by volume). Then, the mixture was centrifuged (1400 g, 0 �C)for 10 min and then supernatant was measured spectrophoto-metrically at 535 nm. Tetramethoxypropanewas used as a standardto estimate TBARS formation as nanomoles of malondialdehyde(MDA) equivalents per g of digestive gland.

2.6. Metallothionein

The amount of metallothioneinwas estimated in digestive glandhomogenates, according to a spectrophotometric method [31]. Thepools of 10 digestive glands were homogenized in Tris Buffer (Tris20 mM, 0.5 M sucrose and pH 8,6) containing anti-proteoliticsubstances, then centrifugated at 15,000 g for 30 min at 4 �C. Thesupernatant was treated with ethanol and chloroform to obtain apellet with the metallothionein. The pellet was washed with thehomogenizing buffer adding chloroform and ethanol, centrifugedat 6000 g for 10 min and dried using nitrogen gas stream. MTconcentration was quantified spectrophotometrically at 412 nmusing the Ellman’s reagent [32] by evaluating the SH residue con-tent. The GSH (Sigma G4251) was used as standard and the resultsexpressed as mg/g of tissue.

2.7. Statistical analysis

Statistical analysis was carried out with STATISTICA 9.1 (StatSoft,Inc.). Differences among sites on pollutants concentrations wereassessed by ANOVA and Tuckey HSD post hoc. Differences amongsites on MDA and metallothionein were assessed by non-parametric ANOVA. In all analyses a p < 0.05 value was acceptedas significant. All data are reported as mean � standard error of themean (SEM).

3. Results

3.1. Chemical analysis

Heavy metals and PAH quantity estimated in the whole body ofthe clams are presented in Table 1. Maximum concentration of allthe heavy metals and PAH was found in animals sampled at Mar-ghera. In the same animals, concentrations of Pb (0.57 � 0.14 mg/kg) and PAH (6.17 � 0.77 mg/kg) were significantly higherthan those observed in animals sampled at the other sites of theLagoon.

3.2. Histology

The examination of the histological sections revealed theoccurrence of the parasite Perkinsus sp. (Fig. 2AeF) which wasdetected in 100% of the clams sampled at Ca’ Roman site, 90% of theclams sampled at Val di Brenta and Marghera and 80% of the clamssampled at Palude del Monte. Animals from Ca’ Roman alsoexhibited the highest number of trophozoites. Perkinsus sp. weresystemically found in a variety of host tissues including mantle,intestine, digestive gland, gills, kidney, heart, testis and ovary(Fig. 2AeF). The presence of parasite in the different tissues and a

Table 1Top of the table: heavy metals and polycyclic aromatic hydrocarbons (PAHs) concentrations in Venerupis philippinarum sampled in different sites of Venice Lagoon. Values areexpressed as means� SEM. n¼ 10. Significant differences are indicated by different letters. Bottom of the table: Perkinsus sp. presence in the various organs of V. philippinarumcollected at different areas of Venice Lagoon. Scale of parasite infestation: - absence; þ low;þþmedium; þþþ high. n ¼ 10. Sites: 1 Palude del Monte; 2Marghera, 3 Ca’ Romanand 4 Val di Brenta.

Sites 1 2 3 4

Heavy metals and PAH concentrations Cd mg/kg 0.16 � 0.03a 0.19 � 0.05a 0.09 � 0.02a 0.17 � 0.02a

Cu mg/kg 1.25 � 0.14a 2.06 � 0.30a 1.54 � 0.11a 1.53 � 0.11a

Hg mg/kg 0.05 � 0.02a 0.07 � 0.01a 0.05 � 0.01a 0.05 � 0.01a

Pb mg/kg 0.29 � 0.09a 0.57 � 0.14b 0.11 � 0.02a 0.18 � 0.03a

PAHc mg/kg <2.5a 6.17 � 0.77b <2.5a <2.5a

Perkinsus sp. presence Gills þþ þþ þþþ þMantle þ þþ þþ þIntestine þ þþ þþþ þDigestive gland þ þþ þþ þGonads þ þþ þþþ þKidney þ þ þþ þHeart - þþ þþþ -

c PAH concentrations were determined as the sum of benzo(a)antracene, benzo(b)fluoranthene, benzo(a)pyrene and crisene.

M. Boscolo Papo et al. / Fish & Shellfish Immunology 39 (2014) 42e50 45

scale of parasite infestation are reported in Table 1. The trophozo-ites of Perkinsus sp. were found isolated or in cluster of trophozoitessurrounded by a heavy hemocitical response (Fig. 2B, C, E and F).The hemocitical response was composed of granular eosinophilichaemocytes with PAS-positive granules (inset Fig. 2B and D), hya-line (agranular) haemocytes and rare brown cells. A few number ofencapsulated trophozoites appeared degraded.

Fig. 2. Histological identification of Perkinsus sp. in different tissues of Venerupis philippinarudel Monte. A) Numerous trophozoites (arrows) are detectable in the mantle (H&E staining).Massive inflammatory reaction with granular eosinophilic haemocytes (asterisks) is visible aPerkinsus sp. (arrows) with massive cellular infiltration (asterisks) in the epithelium and ctrophozoites (arrow) in the gills, which show sloughing of filament epithelial cells (PAS reactthe parasites (arrow) (H&E). F) Ovary showing parasites (arrow) surrounded by a haemocytic

In the mantle, the cellular infiltration around the parasitespresented various degrees of intensity independently of theamount of trophozoites (Fig. 2A). Occasionally, the haemocyticalreaction was surrounded by an eosinophilic, strongly PAS-positive,amorphous matrix without parasites.

In the digestive gland, trophozoites were observed in theinterstitial space between the tubules of the gland (Fig. 2B) and

m. A, C and E: animals sampled at Ca’ Roman site; B, D and F: animals sampled at PaludeB) Trophozoites of Perkinsus sp. (arrow) in the interstitial space of the digestive gland.round the parasites (H&E staining). Insert in B shows PAS positivity of trophozoites. C)onnective tissue of the intestinal tract (arrow) (H&E staining). D) Detection of singleion). E) Testis parenchyma showing a massive haemocytic infiltration (asterisks) aroundinfiltration (asterisks) (H&E). Scale bars: A, C, E and F 40 mm; B, insert in B and D 20 mm.

M. Boscolo Papo et al. / Fish & Shellfish Immunology 39 (2014) 42e5046

rarely in the glandular epithelium. Degeneration and necrosis werenoticed in digestive diverticula in case of severe infestation byPerkinsus sp.

Encapsulations were observed in the digestive tract, particularlyin the connective tissue, in the muscular wall and less frequently inthe intestinal epithelium (Fig. 2C). The connective layer and themuscular wall resulted thicker and with abundant inflammatoryresponse.

Numerous clusters of trophozoites or single parasites werefound distributed in the connective tissues of the gills surroundedby a granulocytic reaction at different level of intensity (Fig. 2D).The highly infected gills showed erosion, or sloughing off of thesurface epithelial layer.

In the gonads of both genders the haemocytic infiltrationaround the parasites invaded the interfollicular connective tissueand the follicular walls reducing the reproductive area (Fig. 2E andF). Perkinsus sp. were focally distributed in the parenchyma ofkidney and heart although the intensity of infestation was lowerthan that observed in the other tissues and organs.

Rare clams with a severe Perkinsus sp. infestation exhibitedisolated trophozoites in the adductor muscles, foot and siphones.

Colonies of intracellular prokaryotic rickettsia-like microor-ganisms (Fig. 3A) were observed within the epithelial cells of gills,mantle and digestive gland tubules and muscular fibres of themantle and intestine in 3% of animals from Val di Brenta and 2%fromMarghera. Trichodinid ectoparasites (Fig. 3B) and other sessileperitrich ciliates (Fig. 3C) were occasionally found on the surface ofthe mantle without host response. Unidentified helminths (Fig. 3D)were observed with low prevalence into the lumen of intestine.

3.3. Immunohistochemistry

All tested antibodies gave an immunoreactivity at the level ofthe mantle, intestine, digestive gland and gills of animals from all

Fig. 3. Histological identification of other parasites in different tissues of Venerupis philippanimals sampled at Ca’ Roman. A) Colonies of intracellular prokaryotic rickettsia-like micrperitrich ciliates located on the surface of the mantle. D) Unidentified helminth into the lu

sites (Fig. 4AeF). The reactivity was localized in the cytoplasm ofhaemocytes, which were often organized in cluster of cellsdistributed throughout the epithelium and the underlying con-nective tissue. No differences in terms of intensity and distributionof immunoreactivity were detected among sites.

3.4. TBARS assay

Malondialdehyde (MDA) concentration varied from 13.95�1.68to 36.80 � 5.98 nmol/g (Fig. 5). MDAwas significantly higher in thedigestive gland of clams sampled at Palude del Monte than in thosecollected at the other sites (Marghera 24.46 � 0.92 nmol/g and Valdi Brenta 22.14 � 0.87 nmol/g). Animals from Ca’ Roman showedthe lowest quantity of MDA.

3.5. Metallothionein content

Fig. 6 illustrates the metallothionein (MT) content in thedigestive gland of V. philippinarum at four sites of the Lagoon. Theclams from Ca’ Roman exhibited the highest quantity of MT(76.77 � 3.42 mg/g). The lowest MT levels were detected in animalsfrom Val di Brenta (41.43�5.58 mg/g). Marghera exhibited a con-centration of 54.39 � 12.19 mg/g whereas Palude del Monte aquantity of 67.66 � 10.68 mg/g.

4. Discussion

The study of the biological responses of organisms to differentenvironmental conditions and the quantitative evaluation of theirphysiological status is being considered as a successful approach forthe assessment of environmental quality [33e35]. Among in-vertebrates, bivalve molluscs are widely used as sentinel models inmonitoring programmes due to their wide distribution, sedentarylifestyle and their ability to bioaccumulate several chemicals in

inarum. All sections are stained with H&E. AeC: animals sampled at Val di Brenta; D:oorganisms (arrows) in the gills. B) Trichodinid ectoparasite in the mantle. C) Sessilemen of intestine. Scale bars: A and C 20 mm; B 10 mm; D 40 mm.

Fig. 4. Immunohistochemical localization of CYP1A, NT, HNE and HSP70 in different tissues of Vernerupis philippinarum. All sections are counterstained with haematoxylin. A)Digestive gland of a clam from Val di Brenta site showing an immunoreactivity to HSP70 antibody in the cytoplasm of haemocytes. B) Gills of a clam from Ca’ Roman site showingthe presence of Perkinsus sp. (arrow). The parasite is surrounded by haemocytes with an intense immunopositivity to CYP1A antibody (asterisks). C) Intestine of a clam from Paludedel Monte showing an intense immunopositivity to NT antiserum in haemocytes distributed in the epithelium as well as connective tissue. D) Haemocytes immunopositive to NT(arrows) distributed in the gills of a clam from Val di Brenta site. E) Intestine of a clam from Val di Brenta site showing haemocytes immunopositive to HNE antiserum (arrows). F)Gills of a clam from Marghera site. Arrow indicates a trophozoite surround by haemocytes immunopositive to HNE antiserum. Scale bars: AeF 20 mm.

M. Boscolo Papo et al. / Fish & Shellfish Immunology 39 (2014) 42e50 47

their tissues, thus reflecting the contamination state of their habitat[36e39].

The aim of this study was to evaluate the general morphologyand the oxidative stress response in the bivalve V. philippinarumsampled in four different sites of the Venice Lagoon: Palude delMonte, Marghera, Ca’ Roman and Val di Brenta. The oxidativestress response was evaluated both by determining the levels ofMDA and metallothioneins in the digestive gland and by animmunohistochemical analysis aimed to detect the localization in

Fig. 5. Malondialdehyde content (nmol/g) in digestive gland of clams (Venerupisphilippinarum). Data are presented as means � SEM. Different letters indicate signifi-cant differences. n ¼ 3 pools of 10 animals.

different tissues of some oxidative stress biomarkers: CYP1A,HSP70, HNE and NT. Chemical analyses performed to detectheavy metals and PAHs on whole soft tissues of the clamssampled at the different sites evidenced the maximum concen-trations in animals from Marghera. This site is one of the mostimportant chemical industrial area in Italy and our resultsconfirm that its industrial activities affect the surrounding envi-ronment by contamination of soil groundwater and inner tidalcanals.

Fig. 6. Metallothioneins quantity, expressed as mg/g, in digestive glands of Venerupisphilippinarum collected at four stations (means � SEM). Letters indicate significantdifference. n ¼ 3 pools of 10 animals.

M. Boscolo Papo et al. / Fish & Shellfish Immunology 39 (2014) 42e5048

Since many environmental pollutants such as heavy metals andPAHs are known to be strongly pro-oxidant, we focused ourresearch on the study of biomarkers reflecting oxidative stress.Bivalves have evolved an extensive battery of antioxidant defences[10], including heat shock proteins [40,41], components of the cy-tochrome P450 (CYP)-dependent monooxygenase system [42],metallothioneins [43,44] and variations in these biomarkers werefound to be useful in environmental monitoring. In our study, animmunopositivity to CYP1A, HSP70, HNE and NT antibodies wasdetected by immunohistochemistry in various tissues: mantle, in-testine, digestive gland and gills. No differences in terms of in-tensity and distribution of immunoreactivity were detected amongsites. The reactivity was localized in the cytoplasm of haemocytes,which were often organized in cluster of cells distributedthroughout the epithelium and the underlying connective tissue. Inbivalves, haemocytes constitute the primary line of defence againstmaterials recognized as non-self [45]. The presence of non-selfmaterials in tissues initiates a complex molecular signallingcascade to stimulate cell-mediated immune responses, mainlyinvolving phagocytosis or encapsulation of foreign materials, andthe production of reactive oxygen species (ROS) [46,47].

In the present work, oxidative stress status was also evaluated indigestive gland by the measure of markers of lipid peroxidation:the thiobarbituric acid reactive substances (TBARS). Although thespecificity of TBARS towards compounds other than MDA has beencontroversial, the assay continues to be the most widely employedformat for monitoring lipid peroxidation [29,48e50]. In this study,MDAwas significantly higher in the digestive gland of the molluscssampled at Palude del Monte than in those collected at the othersites, whereas animals from Ca’ Roman showed the lowest quantityof MDA. In accordance with our results, a high concentration ofMDA was observed in the mussel Mytilus galloprovincialis sampledin an area closed to Palude del Monte [51].

Although the clams from Ca’ Roman exhibited the highestquantity of MT, the results is contrast with the highest concentra-tion of all the heavy metals detected in animals sampled at Mar-ghera. In a previous work a high concentration of MT was detectedin the mussel M. galloprovincialis sampled at the site of Ca’ Roman[51].

The general morphology of clams revealed a massive presenceof Perkinsus species in animals from all examined sites. Its presencehas been previously described in clams sampled in the NorthAdriatic Sea by Abollo et al. (2006) [52] who classified it as P. olseniby a species-specific polymerase chain reaction-restriction frag-ment length polymorphism (PCR-RFLP) assay. Perkinsus sp. infes-tation is known to cause a reduction of growth, gametogenesisinhibition, defence system reduction and it is correlated to mor-tality of clams as well as reduction of semen in nature [53e56]. Inthe present study, the level of infestationwas particularly intense inclams from Ca’ Roman and Val di Brenta, in accordance with pre-vious results [57,58]. It is noteworthy that the annual temperaturecycle and salinity determine a seasonal pattern of variation ofprevalence and intensity [57,59]. Most epizootiological studies onperkinsosis have focused on the geographical range of the diseaseand the effects of salinity on parasite and disease dynamics; prev-alence and intensity of perkinsosis in the natural environment in-crease with increasing salinity [59]. Among the examined lagoonsites, Ca’ Roman is the nearest to the sea which exhibits a salinityhigher than that of lagoon environment. Animals from Ca’ Romanalso exhibited the highest number of trophozoites if comparedwithclams from other sites. The parasite was detected in 90% of theclams sampled at Val di Brenta and Marghera and 80% of the clamssampled at Palude del Monte. Since Perkinsus is eliminated by thefecal emission [60], the level of the infestation could be related tothe animal population density in the sampled areas. Moreover,

Perkinsus sp. were detected in all the examined tissues with theexception of the nervous tissue. The presence of Perkinsus sp.induced a strong host response, with an infiltration of numeroushaemocytes into the surrounding parasitized tissues.

In the present study, trichodinid ectoparasites and other sessileperitrich ciliates were occasionally found on the surface of themantle without host response. Peritrich ciliates remain a poorlystudied group in molluscs and only a few publications had reportedthe occurrence of these ectoparasites in marine bivalves [61e63].There was no evidence of pathology by Trichodynia species as re-ported by Bower et al. (1992) [64]. In the samples examined, thepresence of protozoa epibionts was not correlated to lesions ofdisease and these ciliates could be considered ectocommensals.

In conclusion, our histopathological examination evidencedthat clam populations from the Venice lagoon were affected byserious perkinsosis, which under stressful conditions could inducesevere mortality. Although perkinsosis has been observed to in-crease with exposure to heavy metals and PAHs, with the hy-pothesis that pollutants impair the host defence capacities [65],we described the parasite presence of parasitism in all examinedsites. There is evidence that immunological functions areextremely sensitive to anthropogenic chemicals at exposure con-centrations that are not toxic [66e69]. Although immune responseis affected by contaminants it is difficult to identify the degree ofpollution that can reduce the immune system defence. All oursampling sites exhibit a basal level of pollution; this conditioncould influence the presence of Perkinsus and suggests that otherfactors, such as salinity and temperature, could modulate repli-cation and development of Perkinsus sp. [70]. Interestingly, para-sites were surrounded by a severe haemocitical response, whichalways showed a strong immunopositivity to CYP1A, HSP70, HNEand NT, indicating an active defence response of the host incombating parasitic infection.

Acknowledgements

Authors wish to thank Giovanni Caporale and Erica Melchiottifor technical assistance. Special thanks to Laguna Project snc teamthat provided the clams. This research was supported by grantsfrom the University of Padua (Progetto ex 60%) and Istituto Zoo-profilattico delle Venezie (Ricerca Finalizzata Sanitaria 2006, Drg. n.3094, 3 October 2006) to G. Radaelli and D. Bertotto. Moreover, thework of Michele Boscolo Papo was financially supported by a grantfrom Veneto Agricoltura (GABASDOT11).

References

[1] Depledge MH, Fossi MC. The role of biomarker in environmental assessment(2). Invertebrates. Ecotoxicology 1994;3:161e72.

[2] Cajaraville MP, Bebianno MJ, Blasco J, Porte C, Sarasquete C, Viarengo A. Theuse of biomarkers to assess the impact of pollution in coastal environments ofthe Iberian Peninsula: a practical approach. Sci Total Environ 2000;247:295e311.

[3] Matozzo V, Tomei A, Marin MG. Acetylcholinesterase as a biomarker ofexposure to neurotoxic compounds in the clam Tapes philippinarum from theLagoon of Venice. Mar Pollut Bull 2005;50:1686e93.

[4] Matozzo V, Marin MG. First evidence of altered vitellogenin-like protein levelsin clam Tapes philippinarum and in cockle Cerastoderma glaucum from theLagoon of Venice. Mar Pollut Bull 2007;55:494e504.

[5] Matozzo V, Monari M, Foschi J, Cattani O, Serrazanetti GP, Marin MG. Firstevidence of altered immune responses and resistance to air exposure in theclam Chamelea gallina exposed to benzo(a)pyrene. Arch Environ ContamToxicol 2007;56:479e88.

[6] Matozzo V, Binelli A, Parolini M, Locatello L, Marin MG. Biomarker responsesand contamination levels in the clam Ruditapes philippinarum for bio-monitoring the Lagoon of Venice (Italy). J Environ Monit 2010;12:776e86.

[7] Matozzo V, Binelli A, Parolini M, Previato M, Masiero L, Finos L, et al.Biomarker responses in the clam Ruditapes philippinarum and contaminationlevels in sediments from seaward and landward sites in the Lagoon of Venice.Ecol Indicat 2012;19:191e205.

M. Boscolo Papo et al. / Fish & Shellfish Immunology 39 (2014) 42e50 49

[8] Boscolo Papo M, Maccatrozzo L, Bertotto D, Pascoli F, Negrato E, Poltronieri C,et al. Expression of CYP4 and GSTr genes in Venerupis philippinarumexposed to benzo(a)pyrene. Ann Anat; 2013. http://dx.doi.org/10.1016/j.aanat.2013.11.003 [in press].

[9] Halliwell B, Gutteridge JMC. Free radicals in biology and medicine. 4th ed.Oxford: Oxford University Press; 2007.

[10] Winston GW, Di Giulio RT. Prooxidant and antioxidant mechanisms in aquaticorganisms. Aquat Toxicol 1991;19:137e61.

[11] Davies KJA. Oxidative stress, the paradox of aerobic life. In: Rice Evans C,Halliwell B, Land GG, editors. Free radical and oxidative stress: environment,drugs and food additives. London: Portland Press Ltd; 1995. pp. 1e31.

[12] Aldini G, Dalle-Donne I, Facino RM, Milzani A, Carini M. Intervention strategiesto inhibit protein carbonylation by lipoxidation-derived reactive carbonyls.Med Res Rev 2007;27:817e68.

[13] Ischiropoulos H, Zhu L, Chen J, Tsai M, Martin JC, Smith CD, et al. Peroxynitrite-mediated tyrosine nitration catalyzed by superoxide dismutase. Arch BiochemBiophys 1992;298:431e7.

[14] Morimoto RI, Tissieres A, Georgopoulos C. Stress proteins in biology andmedicine. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press;1990.

[15] Iwama GK, Thomas PT, Forsyth RB, Vijayan MM. Heat shock protein expres-sion in fish. Rev Fish Biol Fish 1998;8:35e56.

[16] Feder ME, Hofmann GE. Heat-shock proteins, molecular chaperones, and thestress response: evolutionary and ecological physiology. Annu Rev Physiol1999;61:243e82.

[17] Yamashita M, Hirayoshi K, Nagata K. Characterization of multiple members ofthe HSP70 family in platyfish culture cells: molecular evolution of stressprotein HSP70 in vertebrates. Gene 2004;336:207e18.

[18] Eder KJ, Köhler HR, Werner J. Pesticide and pathogen: heat shock proteinexpression and acetylcholinesterase inhibition in juvenile Chinook salmon inresponse to multiple stressors. Environ Toxicol Chem 2007;26:1233e42.

[19] Hansen BH, Garmo OA, Olsvik PA, Andersen RA. Gill metal binding and stressgene transcription in brown trout (Salmo trutta) exposed to metal environ-ments: the effect of pre-exposure in natural populations. Environ ToxicolChem 2007;26:944e53.

[20] Maradonna F, Carnevali O. Vitellogenin, zona radiata protein, cathepsin D andheat shock protein 70 as biomarkers of exposure to xenobiotics. Biomarkers2007;12:240e55.

[21] Bertotto D, Poltronieri C, Negrato E, Richard J, Pascoli F, Simontacchi C, et al.Whole body cortisol and expression of HSP70, IGF-I and MSTN in earlydevelopment of sea bass subjected to heat shock. Gen Comp Endocrinol2011;174:44e50.

[22] Porte C, Biosca X, Solé M, Borghi V, Martinez M, Chamorro J, et al. Chemical,biochemical and cellular responses in the digestive gland of the musselMytilus galloprovincialis from the Spanish Mediterranean coast. Biomarkers2001;6:335e50.

[23] Rola RC, Monteiro Mda C, Reis SR, Sandrini JZ. Molecular and biochemicalbiomarkers responses in the mussel Mytilus edulis collected from SouthernBrazil coast. Mar Pollut Bull 2012;64:766e71.

[24] Simão MF, Leite RB, Rocha C, Cancela ML. Changes in bioturbation of ironbiogeochemistry and in molecular response of the clam Ruditapes decussatesupon Perkinsus olseni infection. Arch Environ Contam Toxicol 2010;59:433e43.

[25] Stegeman JJ, Hahn ME. Biochemistry and molecular biology of mono-oxygenases: current perspective on forms, functions, and regulation of cyto-chrome P450 in aquatic species. In: Malins DC, Ostrander GK, editors. Aquatictoxicology: molecular, biochemical and cellular perspectives. Boca Raton:Aquatic Publisher; 1995. pp. 87e206.

[26] Kagi JHR. In: Suzuki K, Imura N, Kimura M, editors. Metallothionein III: bio-logical roles and medical implications. Basel: Birkhauser-Verlag; 1993.pp. 29e55.

[27] Davis SR, Cousins RJ. Metallothionein expression in animals: a physiologicalperspective on function. J Nutr 2000;130:1085e8.

[28] Serafim A, Bebianno MJ. Effect of a polymetallic mixture on metal accumu-lation and metallothionein response in the clam Ruditapes decussatus. AquatToxicol 2010;99:370e8.

[29] Pascoli F, Negrato E, Di Giancamillo A, Bertotto D, Domeneghini C,Simontacchi C, et al. Evaluation of oxidative stress biomarkers in Zosterisessorophiocephalus from the Venice Lagoon, Italy. Aquat Toxicol 2011;101:512e20.

[30] Yoshida Y, Itoh N, Hayakawa M, Piga R, Cynshi O, Jishage K, et al. Lipid per-oxidation induced by carbon tetrachloride and its inhibition by antioxidant asevaluated by an oxidative stress marker, HODE. Toxicol Appl Pharmacol2005;208:87e97.

[31] Viarengo A, Ponzano E, Dondero F, Fabbri R. A simple spectrophotometricmethod for metallothionein evaluation in marine organisms: an application toMediterranean and Antarctic molluscs. Mar Environ Res 1997;44:69e84.

[32] Ellman GL. A colorimetric method for determining low concentrations ofmercaptans. Arch Biochem Biophys 1958;74:443e50.

[33] Akcha F, Izuel C, Venier P, Budzinsk H, Burgeot T, Narbonne J. Enzymaticbiomarker measurement and study of DNA adduct formation in benzo [a]pyrene-contaminated mussels, Mytilus galloprovincialis. Aquat Toxicol2000;49:269e87.

[34] Banni M, Negri A, Rebelo M, Rapallo F, Boussetta H, Viarengo A, et al.Expression analysis of the molluscan p53 protein family mRNA in mussels

(Mytilus spp.) exposed to organic contaminants. Comp Biochem Physiol CToxicol Pharmacol 2009;149:414e8.

[35] Banni M, Negri A, Dagnino A, Jebali J, Ameur S, Boussetta H. Acute effects ofbenzo[a]pyrene on digestive gland enzymatic biomarkers and DNA damageon mussel Mytilus galloprovincialis. Ecotoxicol Environ Saf 2010;73:842e8.

[36] Viarengo A, Burlando B, Dondero F, Marro A, Fabbri R. Metallothionein as atool in biomonitoring programmes. Biomarkers 1999;4:455e66.

[37] Banni M, Jebali J, Daubeze M, Clerandau C, Guerbej H, Narbonne JF, et al.Monitoring pollution in Tunisian coasts: application of a classification scalebased on biochemical markers. Biomarkers 2005;10:105e16.

[38] Narbonne JF, Aarab N, Clérandeau C, Daubèze M, Narbonne J, Champeau O,et al. Scale of classification based on biochemical markers in mussels: appli-cation to pollution monitoring in Mediterranean coasts and temporal trends.Biomarkers 2005;10:58e71.

[39] Widdows J, Donkin P, Staff FJ, Matthiessen P, Law RJ, Allen YT, et al. Mea-surement of stress effects (scope for growth) and contaminant levels inmussels (Mytilus edulis) collected from the Irish Sea. Mar Environ Res2002;53:327e56.

[40] Sanders BM. Stress proteins in aquatic organisms: an environmentalperspective. Crit Rev Toxicol 1993;23:49e75.

[41] Lyons C, Dowling V, Tedengren M, Gardeström J, Hartl MG, O’Brien N, et al.Variability of heat shock proteins and glutathione S-transferase in gill anddigestive gland of blue mussel, Mytilus edulis. Mar Environ Res 2003;56:585e97.

[42] Solé M, Livingstone DR. Components of the cytochrome P450-dependentmonooxygenase system and ’NADPH-independent benzo[a]pyrene hydroxy-lase’ activity in a wide range of marine invertebrate species. Comp BiochemPhysiol C Toxicol Pharmacol 2005;141:20e31.

[43] Vergani L, Grattarola M, Borghi C, Dondero F, Viarengo A. Fish and molluscanmetallothioneins. FEBS J 2005;272:6014e23.

[44] Lehtonen KK, Leiniö S, Schneider R, Leivuori M. Biomarkers of pollution effectsin the bivalvesMytilus edulis andMacoma balthica collected from the southerncoast of Finland (Baltic Sea). Mar Ecol Prog Ser 2006;322:155e68.

[45] Cheng TC, Bui MN, Howland KH, Schoenberg DA, Sullivan JT. Effect of pre-injection of Crassostrea virginica with bacteria on subsequent chemotacticresponse by its hemocytes. J Invertebr Pathol 1981;38:122e6.

[46] Pipe RK. Generation of reactive oxygen metabolites by the hemocytes of themussel Mytilus edulis. Dev Comp Immunol 1992;16:111e22.

[47] Anderson RS. Hemocyte-derived reactive oxygen intermediate production infour bivalve mollusks. Dev Comp Immunol 1994;18:89e96.

[48] Armstrong D, Browne R. The analysis of free radicals, lipid peroxides, anti-oxidant enzymes and compounds related to oxidative stress as applied to theclinical chemistry laboratory. Adv Exp Med Biol 1994;366:43e58.

[49] Duran A, Talas ZS. Biochemical changes and sensory assessment on tissues ofcarp (Cyprinus carpio, Linnaeus 1758) during sale conditions. Fish PhysiolBiochem 2009;35:709e14.

[50] El-Sharaky AS, Wahby MM, Bader El-Dein MM, Fawzy RA, El-Shahawy IN.Mutual anti-oxidative effect of gossypol acetic acid and gossypol-iron com-plex on hepatic lipid peroxidation in male rats. Food Chem Toxicol 2009;47:2735e41.

[51] Pampanin DM, Camus L, Gomiero A, Marangon I, Volpato E, Nasci C. Suscep-tibility to oxidative stress of mussels (Mytilus galloprovincialis) in the VeniceLagoon (Italy). Mar Pollut Bull 2005;50:1548e57.

[52] Abollo E, Casas SM, Ceschia G, Villalba A. Differential diagnosis of Perkinsusspecies by polymerase chain reaction-restriction fragment length poly-morphism assay. Mol Cell Probes 2006;20:323e9.

[53] Da Ros L, Canzonier WJ. Perkinsus, a protistan threat to bivalve culture in theMediterranean basin. Bull Eur Assoc Fish Pathol 1985;5:23e5.

[54] Villalba A, Casas SM, López C, Carballal MJ. Study of perkinsosis in the carpetshell clam Tapes decussatus in Galicia (NW Spain). II. Temporal pattern ofdisease dynamics and association with clam mortality. Dis Aquat Organ2005;65:257e67.

[55] Ceschia G, Sgro L, Ramirez Tafur R, Sello M, Zentilin A. Relationship betweenpresence of Perkinsus olseni and mortality in manila clam (Ruditapes philip-pinarum) cultivated in Marano Lagoon (Italy). Ittiopatologia 2007;4:215e25.

[56] Gomez-Leon J, Aranguren R, Gestal C, Novoa B, Figueras A. Temporal distribu-tion of potentially pathogenic agents detected on carpet-shell clam, Ruditapesdecussatus cultured in Galicia (NWSpain). Aquat Living Resour 2007;20:185e9.

[57] Ceschia G, Zanchetta S, Cestaro M, Zambon M, Sello M. Health aspects offarming manila clam (Tapes philippinarum) in the Veneto Region. Ittiopato-logia 2004;1:49e58.

[58] Meneghetti F, Moschino V, Da Ros L. Gametogenic cycle and variations inoocyte size of Tapes philippinarum from the Lagoon of Venice. Aquaculture2004;240:473e88.

[59] Villalba A, Reece KS, Ordas MC, Casas SM, Figueras A. Perkinsosis in molluscs:a review. Aquat Living Resour 2004;17:411e32.

[60] Park KI, Yang HS, Kang HS, Cho M, Park KJ, Choi KS. Isolation and identificationof Perkinsus olseni from feces and marine sediment using immunological andmolecular techniques. J Invertebr Pathol 2010;105:261e9.

[61] Uzmann JR, Stickney AP. Trichodina myicola n. sp., a peritrichous ciliate fromthe marine bivalve Mya arenaria L. J Protozool 1954;1:149e55.

[62] Xu K, Song W. Two trichodinid ectoparasites from marine molluscs in theYellow Sea, off China, with the description of Trichodina caecellae n. sp.(Protozoa: Ciliophora: Peritrichia). Syst Parasitol 2008;69:1e11.

M. Boscolo Papo et al. / Fish & Shellfish Immunology 39 (2014) 42e5050

[63] Xu K, Lei Y, Al-Rasheid KAS, Song W. Two new ectoparasitic ciliates, Sphe-nophrya solinis sp. nov. and Planeticovorticella paradoxa sp. nov. (Protozoa:Ciliophora), from marine molluscs. J Mar Biol Assoc U.K. 2011;91:265e74.

[64] Bower SM, Blackbourn J, Meyer GR. Parasite and symbiont fauna of Japaneselittlenecks, Tapes philippinarum (Adams and Reeve, 1850), in British Columbia.J Shellfish Res 1992;11:13e9.

[65] Chu FLE, Volety AK, Hale RC, Huang Y. Cellular responses and diseaseexpression in oysters (Crassostrea virginica) exposed to suspended fieldcontaminated sediments. Mar Environ Res 2002;53:17e35.

[66] Duchemin MB, Auffret M, Wessel N, Fortier M, Morin Y, Pellerin J, et al.Multiple experimental approaches of immunotoxic effects of mercury chlo-ride in the blue mussel, Mytilus edulis, through in vivo, in tubo and in vitroexposures. Environ Pollut 2008;153:416e23.

[67] Galloway TS, Depledge MH. Immunotoxicity in invertebrates: measurementand ecotoxicological relevance. Ecotoxicology 2001;10:5e23.

[68] Höher N, Köhler A, Strand J, Broeg K. Effects of various pollutant mixtures onimmune responses of the blue mussel (Mytilus edulis) collected at a salinitygradient in Danish coastal waters. Mar Environ Res 2012;75:35e44.

[69] Liu J, Pan LQ, Zhang L, Miao J, Wang J. Immune responses, ROS generation andthe haemocyte damage of scallop Chlamys farreri exposed to Aroclor 1254.Fish Shellfish Immunol 2009;26:422e8.

[70] Soudant P, Chu FLE, Volety A. Hosteparasite interactions: marine bivalvemolluscs and protozoan parasites, Perkinsus species. J Invertebr Pathol2013;114:196e216.