ORIGINAL ARTICLE

Gene Expression Profiles of Spleen, Liver, and Head Kidneyin Turbot (Scophthalmus maximus) Along the InfectionProcess with Philasterides dicentrarchi Using an Immune-Enriched Oligo-Microarray

Belén G. Pardo & Adrián Millán & Antonio Gómez-Tato &

Carlos Fernández & Carmen Bouza &

José Antonio Alvarez-Dios & Santiago Cabaleiro &

Jesús Lamas & José M. Leiro & Paulino Martínez

Received: 15 December 2011 /Accepted: 22 January 2012 /Published online: 26 February 2012# Springer Science+Business Media, LLC 2012

Abstract We evaluated the expression profiles of turbot inspleen, liver, and head kidney across five temporal points ofthe Philasterides dicentrarchi infection process using an8x15K Agilent oligo-microarray. The microarray included2,176 different fivefold replicated gene probes designed froma turbot 3′ sequenced EST database. We were able to identify221 differentially expressed (DE) genes (8.1% of the wholemicroarray), 113 in spleen, 83 in liver, and 90 in head kidney,in at least 1 of the 5 temporal points sampled for each organ.Most of these genes could be annotated (83.0%) and function-ally categorized using GO terms (69.1%) after the additionalsequencing of DE genes from the 5′ end.Many DE genes wererelated to innate and acquired immune functions. A highproportion of DE genes were organ-specific (70.6%), althoughtheir associated GO functions showed notable similarities in

the three organs. The most striking difference in functionaldistribution was observed between the up- and downregulatedgene groups. Upregulated genes were mostly associated toimmune functions, while downregulated onesmainly involvedmetabolism-related genes. Genetic response appeared clus-tered in a few groups of genes with similar expression profilesalong the temporal series. The information obtained willaid to understand the turbot immune response and willspecifically be valuable to develop strategies of defenseto P. dicentrarchi to achieve more resistant broodstocks forturbot industry.

Keywords Turbot . Scophthalmus maximus .Philasteridesdicentrarchi . Oligo-microarray . Immune response .

Functional classification

Electronic supplementary material The online version of this article(doi:10.1007/s10126-012-9440-9) contains supplementary material,which is available to authorized users.

B. G. Pardo :A. Millán :C. Fernández : C. Bouza :P. Martínez (*)Departamento de Genética, Facultad de Veterinaria,Universidad de Santiago de Compostela, Campus de Lugo,27002 Lugo, Spaine-mail: paulino.martinez@usc.es

A. Gómez-TatoDepartamento de Geometría y Topología, Facultadde Matemáticas, Universidad de Santiago de Compostela,15782 Santiago de Compostela, Spain

J. A. Alvarez-DiosDepartamento de Matemática Aplicada, Facultad de Matemáticas,Universidad de Santiago de Compostela,15782 Santiago de Compostela, Spain

S. CabaleiroCluster de la Acuicultura de Galicia (CETGA),Punta de Couso s/n,15965 Aguiño (A Coruña), Spain

J. LamasDepartamento de Biología Celular y Ecología,Universidad de Santiago de Compostela,15782 Santiago de Compostela, Spain

J. M. LeiroDepartamento de Microbiología y Parasitología, Institutode Investigación y Análisis Alimentarios, Universidad de Santiagode Compostela,15782 Santiago de Compostela, Spain

Mar Biotechnol (2012) 14:570–582DOI 10.1007/s10126-012-9440-9

Introduction

The turbot (Scophthalmus maximus) is one of the mostpromising species of European aquaculture. Production inEurope reached 9,142 tons in 2009, and it is predicted todouble up in size in 2014. Pathologies constitute one of themain problems of turbot culture. Among these, scuticocilia-tosis, caused by the parasite Philasterides dicentrarchi, hasproduced important losses to turbot industry (Sterud et al.2000; Iglesias et al. 2001). The disease, described for thefirst time in cultured turbot in NW Spain (Dyková andFigueras 1994), has been later recognized as an importantproblem in other flatfish species from Asia, likeParalichthys olivaceus (Ototake and Matsusato 1986; Kimet al. 2004). The disease shows as a severe systemic infec-tion which usually leads to animal death. The parasitespreads through blood and affects brain, gills, liver, andintestine, and sometimes to spleen, kidney, and muscle.Dying fish show cutaneous ulcerae, darkening of skin,swimming alterations, severe exophthalmia, and abdominaldistension due to the accumulation of ascitic fluid, whichsuggest an important inflammatory response (Iglesias et al.2001; Moustafa et al. 2010). Several studies have suggestedan important role of phospholipases as a virulence factor inparasite multiplication because they disrupt host defenses byaffecting intracellular signal transduction in immuneresponses (Seo et al. 2005; Lee and Kim 2010; Salinas etal. 2011). Also, cysteine proteinases of P. dicentrarchi mayinduce host leucocyte apoptosis and led to a significantdecrease in the killing activity of serum as a mechanism ofpathogenesis and evasion of the turbot innate immuneresponse (Paramá et al. 2007; Piazzon et al. 2011a).Finally, different reports showed that antioxidative enzymes,like superoxide dismutase and catalase, and proteases areactivated for protecting the parasite against host phagocyte-mediated oxidative damage (Kwon et al. 2003; Lee et al.2004; Piazzon et al. 2011b).

The parasite ability to infect host depends on its viru-lence, but also on its interaction with the host immunesystem. Significant changes in a variety of components ofthe host immune system have been detected during P. dicen-trarchi infection. Thus, some serum components exhibited asignificant chemoattractant activity towards P. dicentrarchi,suggesting their implication in host detection during parasiteinfection (Paramá et al. 2004a). Expression of lysozyme andserum complement, powerful antimicrobial moleculesinvolved in innate response, and antibodies, crucial for adap-tive response, were observed to increase in challenged turbotas infection progresses (Paramá et al. 2004a; Sitjà-Bobadilla etal. 2008; Palenzuela et al. 2009). Particularly, Leiro et al.(2008) demonstrated that the antibody-mediated classicalcomplement pathway would explain the parasite killing capa-bility of immunized-fish serum.

Recently, the capacity of some chemotherapeutic agentsto control scuticociliatosis has been evaluated. It has beenshown that P. dicentrarchi is sensitive in vitro to commercialor newly synthesized antiprotozoals (Iglesias et al. 2002;Paramá et al. 2004b; Quintela et al. 2003). Certain naturalpolyphenols have anti-parasitic properties (Leiro et al.2004a; Morais et al. 2009; Budiño et al. 2011a) and alsoexerted important modulatory effects on inflammatory hostresponse (Castro et al. 2008). Parasites can be rapidlydestroyed when they are in sea water by applying inperoxide baths chemicals such formaldehyde or glutar-aldehyde, hydrogen, or Jenoclean (Iglesias et al. 2002;Paramá et al. 2005). However, when located as endo-parasites in fish, no systemic chemotherapeutic treat-ments have yet proven effective (Iglesias et al. 2002;Song et al. 2009). In consequence, a considerable efforthas been directed at the development of a vaccine as anattractive alternative to chemotherapeutic treatments foreffective prevention of this disease (Lamas et al. 2008;Palenzuela et al. 2009). Surface antigens of the parasiteproved their capability for switching host agglutinationreactions, but, as observed in other parasites (Simon andSchmidt 2007), this reaction induces a rapid antigenreplacement to overcome host defenses (Iglesias et al.2003a). Additionally, the remarkable antigen differencesobserved between isolates and the usual coexistence ofdifferent strains at farm facilities complicates the achievementof a universal vaccine and the control of scuticociliatosis usingthis option (Song et al. 2009; Piazzon et al. 2008; Budiño et al.2011b, c).

Obtaining more resistant broodstocks is an appealingsolution to control diseases, although emphasis should beput in obtaining more robust fish focusing in genes involvedin general immune response to pathogens (Ødegård et al.2011). The economic cost of vaccines or treatments wouldbe saved and the resistance would be transmitted to off-spring, depending on the genetic architecture and interac-tions of the genes involved. To our knowledge, no studieson the genetic component of the variation associated withresistance to P. dicentrarchi or other parasites have beenreported to date in turbot. However, low to moderate heri-tability estimates have been obtained for resistance to para-sites in livestock animals (Mackinnon et al. 1991; Douch etal. 1995; Bisset and Morris 1996), and in Atlantic salmon,moderate to high heritabilities have been reported forresistance to different parasites (Ødegård et al. 2011).Additionally, significant genomic associations for resistanceto parasites have been reported in rainbow trout (Nichols et al.2003; Baerwald et al. 2010) and Atlantic salmon (Gilbey et al.2006), and we were able to specifically identify up to sevenconsistent QTL (quantitative trait loci) in turbot for resistance/survival to P. dicentrarchi (Rodríguez-Ramilo et al., unpub-lished data). This information suggests a genetic basis for

Mar Biotechnol (2012) 14:570–582 571

tolerance to scuticociliatosis and supports its study for futureapplication in turbot genetic breeding programs.

Improving vaccines or developing breeding strategies forcontrolling scuticociliatosis requires the deepest possibleknowledge of the host–pathogen interactions underlyingturbot response. In this sense, functional studies on candi-date genes using quantitative PCR (Q-PCR), developingcDNA or subtractive libraries, or applying the more power-ful microarray technology are essential to understand host–parasite interaction. Lee and Kim (2011) showed the activa-tion of specific P. dicentrarchi genes related to signal trans-duction, cell proliferation, membrane transportation, proteintranslocation, and transcription regulation by using suppres-sion subtractive hybridization in infected P. olivaceus indi-viduals. Although few studies have been performed tounderstand the genetic basis of scuticociliatosis from thisperspective, this methodology is being increasingly appliedin aquaculture species for the purpose of rendering usefulinformation (Tsoi et al. 2004; Wynne et al. 2008; Skugor etal. 2009; Fleury et al. 2010). Microarrays constitute a pow-erful technology suitable to dissect complex biological func-tions and as such they have been used to address differentbiological issues in cultured fish (Rise et al. 2004; Lo et al.2005; Jorgensen et al. 2008; Wynne et al. 2008; Park et al.2009). Recently, an oligo-microarray with 2,716 specificprobes enriched in immune system genes representing arobust tool for analyzing gene expression profiles has beendescribed in turbot (Millán et al. 2010). This microarray wassuccessfully applied to study gene expression of immune-organs in response to Aeromonas salmonicida, a bacteriumresponsible of furunculosis (Millán et al. 2011).

In this study, we analyzed gene expression profiles ofimmune-related organs (spleen, liver and head kidney) ofturbot fry injected intracelomically with P. dicentrarchiacross a temporal series involving the main episodes of theinfection process using the turbot oligo-microarray. Themain goal was to identify genes and functions regulated inresponse to this parasite and to ascertain the main geneticroutes and functional categories associated to the responsewithin each organ. The comparison of gene expressionprofiles among organs and against pathogens previouslystudied contributed to understand genetic response to P.dicentrarchi in an integrated way, and to identify bothspecific and general genes involved in the response to differ-ent pathogens.

Materials and Methods

Biological Samples

One hundred and fifty individuals of ∼30 g from 20 equallyrepresented families from a turbot company (Alrogal SA)

were transported to Centro Tecnológico Gallego de Acuicultura(NW Spain) facilities, where the experiment was per-formed. Prior to challenge, fry were maintained during10 days at standard culture conditions to cut down ontransport stress and to check if they were free from diseases. P.dicentrarchi (I1 isolate, Budiño et al. 2011b) were isolated,under aseptic conditions, from ascitic fluid of infected turbotand cultured at 18°C in L-15 Leibovitz medium, supple-mented with 5% fetal bovine serum (FBS), lipids (lecithinand Tween 80), nucleosides, glucose, and antibiotic antimy-cotic solution (100 units/ml of penicillin G, 0.1 mg of strep-tomycin sulfate, and 0.25 mg/ml of amphotericin B), aspreviously described (Iglesias et al. 2003b).

Experimental Design

We followed a very similar experimental design to P.dicentrarchi to that applied previously to identify DE genes inresponse to Aeromonas salmonicida (Millán et al. 2011), buttaking into account the differences derived from the specificinfection processes of both pathogens. We were interested inanalyzing the response of turbot as species to P. dicentrarchiin three of the main immune organs (spleen, liver, and headkidney) along a temporal series representing the main epi-sodes of infection (1, 3, 7, 15, and 25 days). Accordingly,individual samples were pooled at each sampling point toaverage interindividual variation. The high similarity of DEgenes observed between biological replicates in response to A.salmonicida infections (around 90%; Millán et al. 2010) sup-ported the idea of emphasizing more organs and times alonginfection rather than the number of biological replicates. Asingle control point (time 0; non-injected) was used since avery slight effect of injection on gene expression was previ-ously reported in these three organs in turbot (Millán et al.2010).

Five fish were sacrificed on day 0 to be used as theunique control in the experiment and 150 were challengedwith a highly virulent strain of P. dicentrarchi as previouslydescribed (Paramá et al. 2003). Briefly, turbot were injectedintracelomically (i.p.) with 0.1 ml of a parasite suspensionof 2.5×106 parasites/ml. The presence of P. dicentrarchi ininfected fish was checked by plating drops of the asciticfluid and blood that were impregnated with ammoniacalpyridinated silver carbonate, and biometric measurementsof taxonomically important characters were made under anoptical microscope (Budiño et al. 2011b). Groups of 5 fishwere sacrificed at 1, 3, 7, 15, and 25 days post-challenging.All fish used in the experiment were sacrificed by decapita-tion. Equal amounts of spleen, liver, and head kidney weresampled from each fish, pooled per organ at each samplingpoint and immediately stored in liquid nitrogen. And sothen, 18 samples, 15 from infected fish (3 organs×5 samplingpoints) and 3 from controls (3 organs) were collected for gene

572 Mar Biotechnol (2012) 14:570–582

expression evaluation. Pooled tissues were stored at −80°Cuntil used for RNA extraction.

All experiments were carried out in accordance withEuropean regulations on animal protection (Directive 86/609), outlined in the Declaration of Helsinki. All experi-mental protocols were approved by the Institutional AnimalCare and Use Committee of the University of Santiago deCompostela (Spain). For all the procedures, the fish wereanesthetized with benzocaine (50 mg/ml).

RNA Extractions and Microarray Hybridizations

Pooled tissues were ground to a fine powder in a mortar andpestled with liquid nitrogen. Total RNA was extracted frompooled tissues using TRIZOL Reagent (Life Technologies)according to manufacturer’s recommendations. All extrac-tions were performed by the same researcher. RNA qualityand quantification were evaluated in a Bioanalyzer (BonsaiTechnologies) and in a NanoDrop® ND-1000 spectropho-tometer (NanoDrop® Technologies Inc), respectively, priorto labeling and hybridization.

An 8x15K Agilent oligo-microarray (2,716 probes) includ-ing 5 replicates of each probe (gene) per microarray and8 microarrays per slide (Millán et al. 2010) was used toevaluate gene expression profiles after challenging with P.dicentrarchi. Three slides were used in the experiment, oneper organ, following a 1-color labeling approach. In eachslide, 6 microarrays (1 control+5 temporal samples) wereused. Hybridizations were performed at the Universidad deSantiago de Compostela Functional Genomics Platform usingAgilent Gene Expression Analysis. All work was carried outon the same day and by the same researcher.

Microarray Analysis

Microarray hybridization and filtering of data were per-formed as described by Millán et al. (2010). Since normalityof the log (log-normality) microarray signal is assumed, thelog2 transformation of the ratios treatment/control was usedin the statistical analysis. Normalization within each micro-array was carried out using the LOESS method, whichassumes that most genes in microarrays are not differentiallyexpressed versus the control. The mean log2 ratio for each ofthe 2,176 genes in the microarray was obtained by averag-ing the five extant replicates per microarray. A doublesimultaneous criterion was used to identify with high con-fidence DE genes: (a) genes with log2 ratios ≥2 or ≤−2 forup- and downregulated genes, respectively, and (b) geneswhich deviated from the null hypothesis (mean log2 ratio00)using t tests atP<0.05 after Bonferroni correction. Microarraydata from this study are available from Gene ExpressionOmnibus at www.ncbi.nlm.nih.gov/geo/ under accessionnumbers GSE35184.

DE genes showed a moderate annotation success bothdue to the short EST lengths (mostly around 500 bp) andbecause they were sequenced from the 3′ end to constructthe turbot database (Pardo et al. 2008). To improve annota-tion, we additionally sequenced all non-annotated DE genesfrom the 5′ end following Pardo et al. (2008). The 5′ and 3′end sequences of each gene were aligned using ClustalXv.2.0 (Larkin et al. 2007) and then searched using Blastn,Blastx in Uniref, and nr NCBI databases, and further ana-lyzed with AutoFACT. The corresponding outputs weresubsequently parsed using BioPerl for fish-relevant hits(e value<10−5) and significant UniGene information. GO,KEGG, and COG terms were extracted from AutoFACToutput.

DE genes were classified into functional categories usingthe Gene Ontology (GO) terms taking as reference theprevious classification of immune/defense-related genes inturbot reported by Park et al. (2009). Starting from annota-tions in the turbot EST database (Pardo et al. 2008), GOcategories were assigned to each gene using QUICKGO(http://www.ebi.ac.uk/QuickGO/) and AmiGO (http://amigo.geneontology.org/cgi-bin/amigo/go.cgi). Due to the relevanceof immune functions for the study, immune/defense-relatedcategories were prioritized in our classification. Also, morethan one category was assigned to specific genes when theywere considered relevant for the study.

Q-PCR Microarray Validation

To validate microarray results by Q-PCR, we used a modifica-tion of the random stratified procedure proposed byMiron et al.(2006) as described by Millán et al. (2011). A set of 14 genescovering the fold change (FC) range variation across the differ-ent conditions in our experiment were selected (between −4.17for trypsin in liver 25 days and 8.82 for apoliprotein A1 in headkidney 25 days; Table S2). FCs of these genes in the 15microarrays were ordered by FC value (14 genes×15microarrays0210 cases) and stratified considering boththe width and the abundance of values within eachstratum. Sets of genes were selected randomly withineach stratum completing 64 cases out of the 210 possi-ble ones (abundance between 2 and 16 cases per stra-tum). Ubiquitin A-52 was used as housekeeping genefor Q-PCR analysis according to our previous results(Millán et al. 2011). RNA (1 μg) was reverse transcribed intocDNA using AffinityScript Multiple Temperature cDNASynthesis kit according to the supplier ’s protocol(AGILENT TECHNOLOGIES). The Q-PCR analysis wascarried out in a MX3005P thermocycler (STRATAGENE)using 1 μl of cDNA in a 20-μl reaction following theBrilliant III Ultra-Fast SYBR® Green QPCR Master Mix(AGILENT TECHNOLOGIES) as described by Millán et al.(2011).

Mar Biotechnol (2012) 14:570–582 573

Statistical Procedures

Normalized data were organized into .mev files using ahomemade R program for statistical analysis with theMultiExperiment Viewer (MeV) program (Saeed et al.2006). The Self Organizing Tree Algorithm (SOTA) program(Dopazo and Carazo 1997) was used to identify the mainexpression gene profile groups across the temporal serieswithin each organ. These groups, holding similar (non-signif-icantly different) gene profiles, could provide evidences on themain routes of the immune response of turbot against P.dicentrarchi. To detect significant gene profiles, we usedthe Pearson correlation and adjusted cell variability(non-significantly different gene profiles within group)to P>0.05.

Results and Discussion

Quality of Microarray Data and Validation by Q-PCR

RNA extracted from tissues showed high quality parame-ters, RIN (RNA Integrity Number) values ranging between8.9 and 10, mostly close to 10 (average 9.8). A total of 64cases from the following 14 randomly selected genes cov-ering most FC range in our experiment were used to validatemicroarray data (Table S1; Fig. 1): Drtp-1 (FE950116),inter-alpha trypsin inhibitor (FE944298), heavy chain 3UDP-glucose pyrophosphorylase 2 (FE944840), comple-ment C3 (FE948765), apolipoprotein C-I precursor(FE950936), TBT-binding protein (FE951052), peroxisomeproliferator-activated receptor gamma (FE944839), 1 CC

chemokine SCYA118 (FE951183), alcohol dehydrogenase8 (FE952304), antifreeze polypeptide (AFP) precursor(FE947215), hepcidin (FE947613), and three other non-annotated genes (FE945068, FE943625. FE949894). Thenumber of cases analyzed per gene in different microarraysranged from 4 to 6 due to the random selection approachused. All microarrays were validated using at least 3 genes(spleen, 7 and 15 days; liver, 15 days) up to a maximum of 9genes (head kidney, 25 days) with an average of 5 genes permicroarray. Ubiquitin A-52, the selected housekeepinggene, showed very low variation across the 15 experimentalconditions tested (mean0−0.01; SD00.11) and no signifi-cant differences were detected when contrasting the FC valuesof the 15 experimental conditions against the null hypothesis(FC00) (mean±error, 0.014±0.029; t0−0.480ns). Q-PCR andmicroarray expression values showed the same FC sense atmost cases (87.1%) and discrepancies were minor and onlyobserved within the ±1 FC range (Fig. 1) as reported in otherstudies (Morey et al. 2006; Tingaud-Sequeira et al. 2009;Ferraresso et al. 2010). Absolute differences between bothtechniques (│FC microarray−FC Q-PCR│) were higher forupregulated than for downregulated genes (mean differences,0.67 vs 0.35). The highest difference was detected for hepci-din in head kidney at 25 days post-challenge (3.09); but in80.1% cases, the divergence was <1. Both data sets werehighly correlated (Pearson ρ coefficient00.952; P00) andno significant differences were detected between them afterapplying a t test (t0−0.750; P00.450). These values aresimilar to those observed in other studies (Patterson et al.2006; Ferraresso et al. 2010; Fleury et al. 2010) and supportthe consistency of our microarray data.

DE Genes in Response to P. dicentrarchi

Cumulative mortality along the experiment after challengeswith P. dicentrarchi was 54%. Mortality sharply increasedfrom day 7, reached the highest figure at day 13, and thendecreased until the end of the experiment on 25 days. Only 1out of 103 individuals finally dissected did not show thepresence of the ciliate either in the ascitic fluid or in theblood and was discarded for further analyses. A total of 221genes (8.1% of the whole microarray) resulted DE in at least1 of the 15 conditions tested in the experiment (3 organs×5times; Tables S1 and S2). A higher proportion of DE genesresulted up- (68.3%) rather than downregulated (26.2%), theremaining ones (5.4%) being up- or downregulated at dif-ferent organs or points of the temporal series. A total of 189DE genes (85.5%) could be successfully annotated usingBlastn and Blastx searches, while in 10.4% of genes noinformation could be obtained, the remaining 4.1% beinghomologous to cDNA clones of other fish species. Mostannotated genes could be functionally classified and associ-ated to GO terms in our study (74.7%; Table S2). For a more

Fig. 1 Comparison of Q-PCR and microarray fold change (FC) in allexperimental conditions using 64 cases covering most FC range from14 randomly selected genes of turbot after challenge with Philasteridesdicentrarchi, following a modified stratified approach by Miron et al.(2006)

574 Mar Biotechnol (2012) 14:570–582

manageable classification, we grouped low frequency cate-gories into more general ones as in our previous study(Millán et al. 2011). According to these criteria, the mostrepresentative categories were metabolic process (41.8%),cellular processes (15.7%), immune response (9.1%), apo-ptosis (6.1%), and regulation of biological process (5.4%).Different unique sequences in the database showed the sameannotation despite their being apart as a result of the bio-informatic pipeline (Tables S1 and S2). Some of these pairslike fatty acid binding protein (ID (identifier code number inthe turbot database): 2738, 842), inter-alpha trypsin inhibitorheavy chain (ID: 127, 2537), trypsinogen 1 precursor (ID:682, 728), and warm temperature acclimation related65 kDa protein (ID: 816, 828) showed sharply differentexpression profiles appearing activated in different organsalong the temporal series, which suggests their paralogouscondition within the same gene family. Conversely, pair ofsequences of ceruloplasmin (ID: 768, 1401), complementC9 (ID: 182, 2855), and hyaluronic acid binding protein 2(ID: 779, 1031) showed nearly identical profiles suggestingthat a single gene sequence was wrongly split in severalchunks by the bioinformatic pipeline.

Many DE genes and gene families in our study appearedassociated to essential functions related to innate and acquiredimmunity. Among the most representative ones, we couldidentify several related to chemokines, chemotaxins, comple-ment, immunoglobulins, major histocompatibility complex,interferon, lectins, cytochrome P450, and lysozyme. Someof these genes had been previously claimed to be involvedin the response to P. dicentrarchi. So, a lysozyme precursorand up to nine complement-related proteins, which had beenreported to increase their expression along infection with P.dicentrarchi (Paramá et al. 2004a; Sitjà-Bobadilla et al. 2008;Palenzuela et al. 2009), appeared activated in our study mostlyin spleen and head kidney. Also, several antibody-relatedgenes (immunoglobulin light chain L2, immunoglobulinheavy chain, immunoglobulin M) and proteins associatedwith antigen processing and presentation (MHC class Ia andIIa, minor histocompatibility antigen H13, CD209 antigen-like protein D), crucial for adaptive response, were detectedmostly upregulated in our study as previously reported (Sitjà-Bobadilla et al. 2008; Leiro et al. 2008). It has been suggestedthat cysteine proteinases of P. dicentrarchi may induce hostleucocyte programmed cell death (Paramá et al. 2007).However, most apoptosis-related genes detected in liver inour study were also identified in response to A. salmonicida(Millán et al. 2011). Only the minor histocompatibility antigenH13 and interferon induced with helicase C domain 1, asso-ciated to apoptosis GO terms, were specifically regulated inresponse to scuticociliatosis. It has been reported that theparasite activates antioxidative enzymes to prevent oxidativedamage mediated by host phagocytes (Kwon et al. 2003; Leeet al. 2004; Piazzon et al. 2011b). Killing of parasites by

phagocytes may be oxygen-dependent (respiratory burst) oroxygen-independent (Verhoef 1991; Clark 1999). The respi-ratory burst, induced by chemotactic stimulation or phagocy-tosis, involves release of reactive oxygen intermediates (ROS)that kill parasites, but that may also cause tissue damage andinflammation (Baggiolini and Wymann 1990). ROS arepotent inducer of the pro-inflammatory stress response typi-fied by pro-inflammatory cytokines, prostaglandins, throm-boxanes, leukotrienes, leukocyte adhesion molecules, andchemokine synthesis (Tse et al. 2004). We demonstrated thatinfection of turbot withP. dicentrarchimodulates complementactivation and intracellular ROS production, and increasesscavenging of extracellular ROS (Leiro et al. 2004b). In ourstudy, we could detect a significant upregulation of glutathi-one peroxidase in host liver in contrast to A. salmonicidaresponse (Millán et al. 2011), which could reflect higheroxidative activity to counterbalance parasite defenses. Also,several chemokine and chemotaxins appeared upregulated inspleen and head kidney. Many DE genes identified inour study had been previously detected in response toA. salmonicida (Millán et al. 2011) and nodavirus (Parket al. 2009), suggesting a general role in response to differentpathologies.

DE Genes and Functions: Comparison Among Organs

Spleen, liver, and head kidney showed notable differences ingene expression in accordance with their specific functionalroles in the immune response (Danneving et al. 1994;Brattgjerd and Evensen 1996; Zapata et al. 1996). Amongthe 221 DE genes detected in at least 1 of the 5 temporal pointssampled at each organ, 113 were detected in spleen, 83 inliver, and 90 in head kidney. Up- and downregulation profileswere markedly different between organs, the liver showing alarge amount of downregulated genes while in head kidneyand, especially, spleen dominating upregulation profiles(Fig. 2). The amount of DE genes was significantly lower(49.9%) than that detected in our previous study in response toA. salmonicida (Millán et al. 2011), evidencing a less intenseimmune response to P. dicentrarchi. Unlike our previouswork, spleen (from 47.3% to 51.1%) and, especially, headkidney (from 26.5% to 40.7%) assumed a more relevant role,to the detriment of liver (from 52.2% to 37.6%). Also, a highproportion of DE genes were organ-specific (70.6%), the livershowing the most differentiated response with 81.9% exclu-sive DE genes and head kidney the lowest one (34.4%).Again, the highest proportion of DE genes was shared byspleen and head kidney (35.3% among upregulated). Finally,most DE genes resulted downregulated in the liver, whileupregulated in head kidney and, especially, the spleen.

The evolution of the response to P. dicentrarchi in eachorgan along the temporal series is presented in Fig. 3, wherethe absolute values of controls (abscissae) are represented

Mar Biotechnol (2012) 14:570–582 575

against the log2 ratios of DE genes (ordinates). FC valuesranged between 8.28 (apolipoprotein A-1, head kidney) and−4.17 (trypsin, liver). Several genes representing theextremes of FC ranges in the three organs could not beannotated (spleen, −3.37, ID: 2676; liver, 6.61, ID: 1529;head kidney, −3.68, ID:2950; Table S1), highlighting theimportance of including non-annotated genes in the micro-array as a way for identifying relevant genes or func-tions in response to the experimental conditions of thestudy. These genes will require further attention in thefuture. Downregulated genes showed absolute average

values higher than upregulated ones, which is explainedby the higher expression values of metabolic-associatedgenes, mostly downregulated, as previously reported(Millán et al. 2011). Response roughly decreased overtime in the three organs, particularly in the spleen wherethe strongest response was detected in days 1 and 3 andnearly no DE genes were detected from day 7 onwards.However, an acute response was detected again atday 25 in the liver and head kidney, which showed alarge amount of downregulated and upregulated genes,respectively, suggesting that the parasite may have been

Down-regulated Up-regulatedFig. 2 Venn diagrams of up-and downregulated differential-ly expressed (DE) genes in theturbot immune-related organs inresponse to Philasteridesdicentrarchi

Fig. 3 Differentially expressed (DE) genes across the temporal series in the immune-related organs of turbot in response to Philasterides dicentrarchi.Log2 values of controls (abscissae) are represented against the log2 ratios of DE genes (ordinates)

576 Mar Biotechnol (2012) 14:570–582

dormant avoiding the immune system of the host andlater on reactivated the infection. The fact that P. dicen-trarchi can act as an endoparasite that divides rapidly inmany organs and feeds on cells and other tissue com-ponents may explain this observation (Iglesias et al.2001). Remarkably, a large proportion of DE genesactivated in head kidney at 25 days were also upregu-lated in spleen at 1 and 3 days (59.7% out of 77 genes;Table S1), including many genes related to acute phaseresponse like α-2-macroglobulin, several apolipoproteins(3), complement-related proteins (5), fibrinogen (α, β,γ), IGF-related proteins (2), hepcidin, and lysozyme,among others. A similar phenomenon was observed inthe liver, although not so accentuated, and 36.7% ofgenes blocked at 25 days was also downregulated at 1and 3 days (Table S1), mostly including metabolismrelated enzymes as 6-phosphofructo-2-kinase, phospholi-pase A2, carboxypeptidasae A2, trypsin, and chymotryp-sinogen. A technical replicate microarray was hybridizedconfirming the results obtained, which should be furtheranalyzed at individual level in the future.

Grouping genes according to GO categories greatlyreduces the interpretative challenge posed by a long list ofDE genes, putting the focus on the analysis of most repre-sentative functions (Allison et al. 2006). Figure 4 presentsthe functional distribution of DE genes in the three organssplit into up- and downregulated genes using spider dia-grams. Although highly specific DE gene sets were detectedat each organ, the functions involved both in up- and down-regulated genes showed notable similarities, as previouslyreported by Millán et al. (2011) in response to A. salmoni-cida. Thus, metabolic process, regulation of biological pro-cess, immune response, cell process, transport, and responseto stress were upregulated in a similar proportion in the threeorgans. We cannot discard that the GO hierarchical level 2used for a manageable classification and comparison be-tween organs may be too general and this could explainthe between-organ similarity observed. In fact, specific

upregulated functions were also detected in spleen (responseto stimulus), liver (apoptosis and biosynthetic process), andhead kidney (signal transduction) evidencing functionalorgan specificity. Among downregulated functions, met-abolic process, transport, and regulation of biologicalprocess were the most salient ones in the three organs.Spleen showed a rather different profile including spe-cific downregulated functions like transcription factor,biosynthetic process, and response to stress. Theseresults are quite similar to those obtained in our studyof response toA. salmonicida (Millán et al. 2011), especially ifwe look at liver functions, while spleen showed the highestdiscrepancies. The small amount of downregulated genes inspleen may explain the different functional distributionobserved between both studies.

In accordance with this general picture, relevant genesinvolved in the immune and defense responses were mostlyupregulated, indicating an activation of the immuneresponses in infected fish. Among these, several chemo-kines, necessary for inducing leukocyte extravasation andchemotaxis towards the affected tissues (Charo andRansohoff 2006); interferon-related genes, a family of cyto-kines with antiviral activity but also with many functions inthe immune responses (Dinarello 2000); several complement-related genes; perforin, a protein relevant in cytotoxicity thatforms pores in cell membranes (Toda et al. 2011); and someMHC genes, related to antigen presentation of great relevancein the adaptive immune responses against infection(Trowsdale and Parham 2004; Ramachandra et al. 2009).Among upregulated genes, those related to complement seemto be of special relevance, as complement activation is criticalin the defense against P. dicentrarchi (Leiro et al.2008). Their activation is probably related to comple-ment consumption occurred during infection and thenecessity of increase complement-related proteins tocontrol the disease. Downregulation of metabolic pro-cess may be related to the energy savings associated tothe lower activity and food intake of ill animals, as

Fig. 4 Spider diagramsshowing functionalclassification of up- and down-regulated genes in the immune-related organs of turbot inresponse to Philasteridesdicentrarchi. Numbers repre-sent percentages of the genesinvolved in each class withinorgan

Mar Biotechnol (2012) 14:570–582 577

previously reported (Millán et al. 2011). Several endo-parasites interact with the host in the special livingenvironment, competing for energy and nutrients fromthe host cells by manipulating the host metabolism (Xuet al. 2010).

Main Genetic Routes of Response to P. dicentrarchi

Immune response involves a large variety of genes andfunctions, which are organized in a limited number of routestriggered in a coordinate manner. Thus, similar expressionprofiles along the temporal series of the infection processwere detected in large gene sets by Millán et al. (2011). Weused the SOTA (Dopazo and Carazo 1997) algorithm toidentify coherent gene expression profiles among DE genesin the three organs. DE genes were clustered at 9 groups inspleen, 10 in liver, and 8 in head kidney according to thestatistical criteria used (cell variability, P<0.05; Table S2,Fig. S1). These groups ranged from 1 to 65 genes in spleen,1 to 29 genes in liver and 1 to 62 genes in head kidney. Mostgenes were clustered in two main groups in each organ(Fig. 5), representing 80% of DE genes of spleen (groups3 and 4), 60.2% of liver (groups 9 and 10), and 82.2% ofhead kidney (groups 5 and 8). Remarkably, these groups,though split by SOTA, showed very similar profiles alongthe temporal series within each organ. Most relevant genes

within these groups are presented in Table 1 classifiedaccording to function.

Groups S3 and S4 from spleen showed very similarexpression profiles and were analyzed together (Fig. 5).These groups included only upregulated genes and wererelated to complement, hemostasis, immunity, stress response,proteases, apolipoproteins, cell adhesion and growth, andiron metabolism and transport. Group S4 included allapolipoproteins and group S3 was predominant in mostother groups. Interestingly, most genes of S3 and S4groups (80.9%) were also detected following identicalprofiles in spleen in response to A. salmonicida byMillán et al. (2011; Table 1 in bold characters). Thesegenes likely represent the initial acute response to everypathogenic process driven from spleen and follow acommon expression profile.

Groups L9 and L10 were the highest relevant ones withinliver and showed downregulated profiles mostly (Table S2,Fig. 5). Group L10 included metabolic related genes mostly,while group L9 showed a more diverse functional spectruminvolving response to stress (upregulated), response to stim-ulus and defense response related genes, and some metabol-ic enzymes. Remarkably, the first five liver groups (L1–L5),although not too large, showed nearly mirror profiles to themain groups L9 and L10 (Table S2; Fig. S1). These groupsincluded highly relevant genes related to immune response

Spleen

1d 3d 7d 15d 25d

Liver

1d 3d 7d 15d 25d 1d 3d 7d 15d 25d

Head Kidney

Group HK8 62 genes

Group HK5 12 genes

Group L10 29 genes

Group L9 21 genes

Group S3 65 genes

Group S4 25 genes

Fig. 5 Two main significant differentially expressed gene profilesobtained with SOTA algorithm (Dopazo and Carazo 1997) in theimmune-related organs of turbot in response to Philasterides dicen-trarchi. Ordinates: log2 ratio infected/control; abscissae: sampling

times along the infection process (12 h and 1, 3, 7, and 21 days).The red line represents the centroid of the group, obtained as theaverage of all expression profiles of each group

578 Mar Biotechnol (2012) 14:570–582

like MHC (Ia and IIa), minor histocompatibility antigen H13,endoplasmin, and some mitochondrial-related genes (cyto-chrome c oxidase subunit II, mitochondrial ribosomal proteinS12), all upregulated.

All groups detected in head kidney showed rather similarprofiles with a first activation at 7 d and a strong reactivation at

25 d (Fig. S1). Group HK8was the most important including 62genes (68.9%), all upregulated, and involved a large variety offunctions with many immune-related genes (Table S2; Fig. 5).Among adaptive immune response, several chemokine, chemo-taxin, and immunoglobulin related genes were detected. Theremaining genes were also mostly activated in spleen at 1 and

Table 1 Relevant differentially expressed (DE) genes of the main expression profile groups identified within spleen, liver and head kidney usingthe SOTA algorithm (Dopazo and Carazo 1997) in turbot after challenge with Philasterides dicentrarchi.

Complement-related Response to stress Haemostasis ImmunityComplement C1-inhibitor Catechol-O-methyltransferase Anticoagulant protein C 3 Alpha-2-macroglobulin 3

negonirbiFB-2C/BtnemelpmoC 3IIInibmorhtitnAniahc CD209 antigen-like protein D 3

negonirbiF2C/FBtnemelpmoC chain Coagulation factor IX prec. 3 Chemotaxin 3

Complement C3 Fibrinogen Coagulation factor VIIb precursor (4) Immunoglobulin heavy chain 3Complement C8B )4(1tnemgarfnibolgotpaHnegonimsalP Lysozyme C precursor 3

Complement C9 Warm temperature accl. Prot. (3, 4) Heparin cofactor II 3 Peptidoglycan recognition protein II 3

Comp. Binding prot Heparin cofactor II 3

Up Comp. C1q-like adipose specific protein (4) Apolipoproteins Thrombin B chain 3 Cell adhesionComplement regulatory plasma protein Apolipoprotein (4) Kallikrein B, plasma 1 3 Fibronectin 1 3

Apolipoprotein A1 Kininogen 1 3 Vitronectin (4)

Proteases-related proteins Apolipoprotein B-100 precursor (4)Alpha-1-antitrypsin Apolipoprotein C-I precursor (4) Iron metabolism and transporters Cell growth

Inter-alpha trypsin inhibitor heavy chain 3 Apolipoprotein C-II (4) Ceruloplasmin 3 IGF binding protein-related protein 1 3

Serine protease inhibitor Serotransferrin (4) Insulin-like growth factor binding prot. 2 E (4)

Trypsinogen 1 precursor Transferrin 3

Metabolism Cell motility Defense response Metabolic processrosrucerp4esatsalEesanik-2-otcurfohpsohp-6 Galectin 3-binding protein Carboxylesterase

Apolipoprotein A-IV3 Collagenase (up) Interferon induced with helicase C domain 1 TrypsinDown Apolipoprotein domain containing protein

Carboxyl ester lipase Response to stimulusCarboxypeptidasae A2 Interferon inducible protein

Chymotrypsinogen 1 Transmembrane protein 7

Na/K ATPase alpha subunit isoform 3 VHSV-induced protein-5

NADH dehydrogenase 1 subcomplex 9

Phospholipase A2

Trypsinogen-like serine protease

Response to stress Cell adhesionUp Glutathione peroxidase 1 FXYD domain cont. ion transport reg. 5a

Serum/glucocorticoid regulated kinase 1

SPLEEN

Group L9 (21 genes)

Groups S3 and S4

LIVERGroup L10 (29 genes)

Group HK5 (12 genes)Apolipoproteins Cell growth Haemostasis Response to stress

Apolipoprotein A1 ietorpezeerfitnAytsrihtdoolB2nietorpgnidnibFGI n type IV Apolipoprotein C-I precursor IGF binding protein 1 Coagulation factor X precursor nixartneP

Apolipoprotein A-IV3 Thrombin B chainUp Cell adhesion Angiotensinogen (down) Protease-related

Immnune response Claudin 25b Kininogen 1 Chymotrypsin B precursorPeptidoglycan recognition protein II Fibronectin 1 Elastase 1 precursor

Alpha-2-macroglobulin Myelin prot. zero-like prot. 2 precursor Elastase 2 precursorCC chemokine SCYA118 Vitronectin Response to stress Trypsinogen 1

Hepcidin precursor Fibrinogen alpha chain Trypsinogen 3Hepcidin Iron transporters Fibrinogen beta Antifreeze polypeptide (AFP) precursor

Chemotaxin Transferrin Fibrinogen gamma Carboxypeptidasae A2 Immunoglobulin heavy chain Serotransferrin

Complement-related Protease-related

Complement factor B/C2-B Serine (or cysteine) proteinase inhibitorComplement C1 inhibitor Trypsinogen 1 precursor

Complement C3 Chymotrypsinogen 1Complement C9

Complement regulatory plasma protein

Group HK8 (62 genes)Head Kidney

Mar Biotechnol (2012) 14:570–582 579

3 days as outlined before, and may represent an acute responseagainst disease reactivation, but in this case driven by headkidney. Group HK5 involved only upregulated genes includingmostly protease-related genes. Two antifreeze protein, two ela-stases, and two trypsinogen genes were present in this group.The remaining HK groups included also relevant immune geneslike lysozyme, chemokine, warm temperature acclimation andimmunoglobulin, and two insulin growth factor related genes.

In summary, the turbot oligo-microarray demonstrated to be auseful and consistent tool to identify immune-related genes inP. dicentrarchi infected fish. That being said, relevant immune-related genes surely remain to be incorporated into this oligo-microarray through the increase of EST resources, valuableinformation was obtained to understand the turbot response tothis pathogen. Thus, (a) response involved specific immune-genes and functions that, although some of them had beenpreviously reported in response to this parasite in turbot, manyothers have been reported for the first time in this study; (b) thethree organs analyzed, though including a rather similar functionspectrum, showed specific sets of DE genes in response to thispathogen; (c) upregulated genes were mainly associated toimmunity, while downregulated ones were associated to metab-olism; and (d) an intense reactivation of disease was detected at25 days according to head kidney and liver profiles suggestingthat the parasite could elicit immune response. Results of thisstudy will be useful to enhance defense strategies against scuti-cociliatosis for achieving more resistant broodstock and satisfythe demands of turbot industry.

Acknowledgments This study was supported by a Consellería dePesca e Asuntos Marítimos and the Dirección Xeral de I+D—Xunta deGalicia project (2004/CP480) and by the Spanish Government (Con-solider Ingenio Aquagenomics: CSD200700002) project. Authors wishto thank Lucía Insua for technical assistance. Belén G. Pardo wassupported by an Isidro Parga Pondal research fellowship from Xuntade Galicia (Spain).

References

Allison DB, Cui XQ, Page GP, Sabripour M (2006) Microarray dataanalysis: from disarray to consolidation and consensus. Nat RevGenet 7:55–65

Baerwald MR, Petersen JL, Hedrick RP, Schisler GJ, May B (2010) Amajor effect quantitative trait locus for whirling disease resistanceidentified in rainbow trout (Oncorhynchus mykiss). Heredity106:920–926

Baggiolini M, Wymann MP (1990) Turning on the respiratory burst.Trends Biochem Sci 15:69–72

Bisset SA, Morris CA (1996) Feasibility and implications of breedingsheep for resilience to nematode challenge. Int J Parasitol 26:857–868

Brattgjerd S, Evensen O (1996) A sequential light microscopic andultrastructural study on the uptake and handling of Vibrio salmo-nicida in the head kidney phagocytes of experimentally infectedAtlantic salmon, Salmo salar L. Vet Pathol 33:55–65

Budiño B, Pata MP, Leiro J, Lamas J (2011a) Differences in the in vitrosusceptibility to resveratrol and other chemical compoundsamong several Philasterides dicentrarchi isolates from turbot.Parasitol Res. doi:10.1007/s00436-011-2664-1

Budiño B, Lamas J, Pata MP, Arranz JA, Sanmartín ML, Leiro J(2011b) Intraspecific variability in several isolates ofPhilasterides dicentrarchi (syn. Miamiensis avidus), a scuticoci-liate parasite of farmed turbot. Vet Parasitol 175:260–272

Budiño B, Lamas J, Arranz JA, González A, Pata MP, Devesa S, Leiro J(2011c) Coexistence of several Philasterides dicentrarchi strains in aturbot fish farm. Aquaculture. doi:10.1016/j.aquaculture.2011.09.039

Castro R, Lamas J, Morais P, Sanmartín ML, Orallo F, Leiro J (2008)Resveratrol modulates innate and inflammatory responses in fishleucocytes. Vet Immunol Immunopathol 126:9–19

Charo IF, Ransohoff RM (2006) The many roles of chemokines andchemokine receptors in inflammation. N Engl J Med 354:610–621

Clark RA (1999) Activation of the neutrophil respiratory burst oxidase.J Infect Dis 179:309–317

Danneving BH, Lauve A, Press C, Landsverk T (1994) Receptor-mediated endocytosis and phagocytosis by rainbow trout headkidney sinusoidal cells. Fish Shellfish Immunol 4:3–18

Dinarello CA (2000) Proinflammatory cytokines. Chest 118:503–508Dopazo J, Carazo JM (1997) Phylogenetic reconstruction using a

growing neural network that adopts the topology of a phyloge-netic tree. J Mol Evol 44:226–233

Douch PGC, Green RS, Morris CA, Bisset SA, Vlassoff A, Baker RL,Watson TG, Hurford AP, Wheeler M (1995) Genetic and pheno-typic relationships among anti-Trichostrongylus colubriformis anti-body level, faecal egg count and body weight traits in grazingRomney sheep. Livest Prod Sci 41:121–132

Dyková I, Figueras A (1994) Histopathological changes in turbotScophthalmus maximus due to a histophagous ciliate. Dis AquatOrg 18:5–9

Ferraresso S, Milan M, Pellizzari C, Vitulo N, Reinhardt R, CanarioAVM, Patarnello T, Bargelloni L (2010) Development of an oligoDNA microarray for the European sea bass and its application toexpression profiling of jaw deformity. BMC Genomics 11:354

Fleury E, Moal J, Boulo V, Daniel JY, Mazurais D, Hénaut A,Corporeau C, Boudry P, Favrel P, Huvet A (2010) Microarray-based identification of gonad transcripts differentially expressedbetween lines of Pacific oyster selected to be resistant or suscep-tible to summer mortality. Mar Biotechnol 12:326–339

Gilbey J, Verspoor E, Mo TA, Sterud E, Olstad K, Hytterød S, Jones C,Noble L (2006) Identification of genetic markers associated withGyrodactylus salaris resistance in Atlantic salmon Salmo salar.Dis Aquat Organ 71:119–129

Iglesias R, Paramá A, Alvarez MF, Leiro J, Fernández J, Sanmartín ML(2001) Philasterides dicentrarchi (Ciliophora, Scuticociliatida) asthe causative agent of scuticociliatosis in farmed turbotScophthalmus maximus in Galicia (NW Spain). Dis AquatOrgan 46:47–55

Iglesias R, Paramá A, Alvarez MF, Leiro J, Sanmartín ML (2002)Antiprotozoals effective in vitro against the scuticociliate fish path-ogen Philasterides dicentrarchi. Dis Aquat Organ 49:191–197

Iglesias R, Paramá A, Alvarez MF, Leiro J, Ubeira FM, Sanmartin ML(2003a) Philasterides dicentrarchi (Cilliophora: Scuticociliatida)expresses surface immobilization antigens that probably induceprotective immune responses in turbot. Parasitology 126:125–134

Iglesias R, Paramá A, Alvarez MF, Leiro J, Aja C, Sanmartín ML(2003b) In vitro growth requirements for the fish pathogenPhilasterides dicentrarchi (Ciliophora, Scuticociliatida). VetParasitol 111:19–30

Jorgensen SM, Afanasyev S, Krasnov A (2008) Gene expressionanalyses in Atlantic salmon challenged with infectious salmonanemia virus reveal differences between individuals with early,intermediate and late mortality. BMC Genomics 9:179

580 Mar Biotechnol (2012) 14:570–582

Kim SM, Cho JB, Kim SK, Nam YK, Kim KH (2004) Occurrence ofscuticociliatosis in olive flounder Paralichthys olivaceus byPhiasterides dicentrarchi (Ciliophora: scuticociliatida). DisAquat Organ 62:233–238

Kwon SR, Kim CS, Kim KH (2003) Differences between short- andlong-term cultures of Uronema marinum (Ciliophora:Scuticociliatida) in chemiluminescence inhibitory activity, antiox-idative enzyme and protease activity. Aquaculture 221:107–114

Lamas J, Sanmartín ML, Paramá AI, Castro R, Cabaleiro S, Ruiz deOcenda MV, Barja JL, Leiro J (2008) Optimization of an inacti-vated vaccine against a scuticociliate parasite of turbot: Effect ofantigen, formalin and adjuvant concentration on antibody re-sponse and protection against the pathogen. Aquaculture278:22–26

Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA,McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R,Thompson JD, Gibson TJ, Higgins DG (2007) Clustal W andclustal X version 2.0. Bioinformatics 23:2947–2948

Lee EH, Kim KH (2010) Molecular cloning and functional character-ization of protein phosphatase 2C of two scuticociliates—Uronema marinum and Miamiensis avidus (Ciliophora:Scuticociliatia). Acta Protozoologica 49:281–288

Lee EH, Kim KH (2011) Identification of differentially expressedgenes in parasitic phase Miamiensis avidus (Ciliophora:Scuticociliatia) using suppression subtractive hybridization. DisAquat Organ 94:135–142

Lee EH, Kim SM, Kwon SR, Kim SK, Nam YK, Kim KH (2004)Comparison of toxic effects of nitric oxide and peroxynitrite onUronema marinum (Ciliata: Scuticociliatida). Dis Aquat Organ58:255–260

Leiro J, Arranz JA, Paramá A, Alvarez MF, Sanmartín ML (2004a) Invitro effects of the polyphenols resveratrol, mangiferin and (-)-epigallocatechin-3-gallate on the scuticociliate fish pathogenPhilasterides dicentrarchi. Dis Aquat Organ 59:171–174

Leiro J, Arranz JA, Iglesias R, Ubeira FM, Sanmartín ML (2004b)Effects of the histiophagous ciliate Philasterides dicentrarchi onturbot phagocyte responses. Fish Shellfish Immunol 17:27–39

Leiro J, Piazzón MC, Budiño B, Sanmartín ML, Lamas J (2008)Complement-mediated killing of Philasterides dicentrarchi(Ciliophora) by turbot serum: relative importance of alternativeand classical pathways. Parasite Immunol 30:535–543

Lo J, Lee S, Xu M, Liu F, Ruan H, Eun A, He Y, Ma W, Wang W, WenZ, Peng J (2005) 15,000 Unique zebrafish EST clusters and theirfuture use in microarray for profiling gene expression patternsduring embryogenesis. Genome Res 13:455–466

Mackinnon MJ, Corbet NJ, Meyer K, Burrow HM, Bryan RP, HetzelDJS (1991) Genetic parameters for testosterone response toGnRH stimulation and scrotal circumference in tropical beefbulls. Livest Prod Sci 29:297–309

Millán A, Gómez-Tato A, Fernández C, Pardo BG, Álvarez-Dios JA,Calaza M, Bouza C, Vázquez M, Cabaleiro S, Martínez P (2010)Design and performance of a turbot (Scophthalmus maximus)oligo-microarray based on ESTs from immune tissues. MarBiotechnol 12:452–465

Millán A, Gómez-Tato A, Pardo BG, Fernández C, Bouza C, Vera M,Alvarez-Dios JA, Cabaleiro S, Lamas J, Lemos ML, Martínez P(2011) Gene expression profiles of spleen, liver and head kidneyin turbot (Scophthalmus maximus) along the infection processwith Aeromonas salmonicida using an immune-enriched oligo-microarray. Mar Biotechnol 13:1099–1114

Miron M, Woody OZ, Marcil A, Murie C, Sladek R, Nadon R (2006)A methodology for global validation of microarray experiment.BMC Bioinforma 7:333

Morais P, Lamas J, Sanmartín ML, Orallo F, Leiro J (2009) Resveratrolinduces mitochondrial alterations, autophagy and a cryptobiosis-like state in scuticociliates. Protist 160:552–564

Morey JS, Ryan JC, VanDolah FM (2006) Microarray validation:factors influencing correlation between oligonucleotide microar-rays and real-time PCR. Biol Proced Online 8:175–193

Moustafa EM, Naota M, Morita T, Tange N, Shimada A (2010)Pathological study on the scuticociliatosis affecting farmedJapanese flounder (Paralichthys olivaceus) in Japan. J Vet MedSci 72:1359–1362

Nichols KM,YoungWP, Danzmann RG, Robison BD, Rexroad C, NoakesM, Phillips RB, Bentzen P, Spies I, Knudsen K, Allendorf FW,Cunningham BM, Brunelli J, Zhang H, Ristow S, Drew R, BrownKH, Wheeler PA, Thorgaard GH (2003) A consolidated linkage mapfor rainbow trout (Oncorhynchus mykiss). Anim Genet 34:102–115

Ødegård J, Baranski M, Gjerde B, Gjedrem T (2011) Methodology forgenetic evaluation of disease resistance in aquaculture species:challenges and future prospects. Aquacult Res 42:103–114

Ototake M, Matsusato T (1986) Notes on scuticociliata infection ofcultured juvenile flounder Paralichthys olivaceus. Bull Natl ResInst Aquacult 9:65–68

Palenzuela O, Sitja-Bobadilla A, Riaza A, Silva R, Aran J, Alvarez-Pellitero P (2009) Antibody responses of turbot Psetta maximaagainst various antigen formulations of scuticociliates Ciliophora.Dis Aquat Org 86:123–134

Paramá A, Iglesias R, Álvarez MF, Leiro J, Aja C, Sanmartín ML(2003) Philasterides dicentrarchi (Ciliophora: Scuticociliatida):experimental infection and posible routes of entry in farmedturbot (Scophhalmus maximus). Aquaculture 217:73–80

Paramá A, Iglesias R, Alvarez MF, Sanmartín ML, Leiro J (2004a)Chemotactic responses of the fish-parasitic scuticociliatePhilasterides dicentrarchi to blood and blood components of theturbot Scophthalmus maximus, evaluated using a new microplatemultiassay. J Microbiol Methods 58:361–366

Paramá A, Iglesias R, Álvarez MF, Leiro JM, Quintela JM, Peinador C,González L, Riguera R, Sanmartín ML (2004b) In vitro efficacy ofnew antiprotozoals against Philasterides dicentrarchi (Ciliophora,Scuticociliatida). Dis Aquat Org 62:97–102

Paramá A, Luzardo A, Blanco-Méndez J, Sanmartín ML, Leiro J(2005) In vitro efficacy of glutaraldehyde-crosslinked chitosanmicrospheres against the fish-pathogenic ciliate Philasteridesdicentrarchi. Dis Aquat Org 64:151–158

Paramá A, Castro R, Lamas J, Sanmartín ML, Santamarina MT, Leiro J(2007) Scuticociliate proteinases may modulate turbot immuneresponse by inducing apoptosis in pronephric leucocytes. Int JParasitol 37:87–95

Pardo BG, Fernández C, Millán A, Bouza C, Vázquez-López A, Vera M,Alvarez-Dios JA, Calaza M, Gómez-Tato A, Vázquez M, CabaleiroS,Magariños B, LemosML, Leiro JM,Martínez P (2008) Expressedsequence tags (ESTs) from immune tissues of turbot (Scophthalmusmaximus) challenged with pathogens. BMC Vet Res 4:37

Park KC, Osborne JA, Montes A, Dios S, Nerland AH, Novoa B,Figueras A, Brown LL, Johnson SC (2009) Immunologicalresponses of turbot (Psetta maxima) to nodavirus infection orpolyriboinosinic polyribocytidylic acid (pIC) stimulation, usingexpressed sequence tags (ESTs) analysis and cDNA microarrays.Fish Shellfish Immunol 26:91–108

Patterson TA, Lobenhofer EK, Fulmer-Smentek SB, Collins PJ, ChuTM, Bao W, Fang H, Kawasaki ES, Hager J, Tikhonova IR,Walker SJ, Zhang L, Hurban P, de Longueville F, Fuscoe JC,Tong W, Shi L, Wolfinger RD (2006) Performance comparison ofone-color and two-color platforms within the MicroArray QualityControl (MAQC) project. Nat Biotechnol 24:1140–1150

Piazzon C, Lamas J, Castro R, Budiño B, Cabaleiro S, Sanmartín ML,Leiro J (2008) Antigenic and cross-protection studies on twoturbot scuticociliate isolates. Fish Shellfish Immunol 25:417–424

Piazzon C, Lamas J, Leiro J (2011a) Role of scuticociliate proteinasesin infection success in turbot, Psetta maxima (L.). ParasiteImmunol 33:535–544

Mar Biotechnol (2012) 14:570–582 581

Piazzon MC, Wiegertjes GF, Leiro J, Lamas J (2011b) Turbot resis-tance to Philasterides dicentrarchi is more dependent on humoralthan on cellular immune responses. Fish Shellfish Immunol30:1339–1347

Quintela JM, Peinador C, González L, Iglesias R, Paramá A, ÁlvarezMF, Sanmartín ML, Riguera R (2003) Piperazine N-substitutednaphthyridines, pyrodithienopyrimidines and pyridothienotria-zines: new antiprotozoals active against Philasterides dicen-trarchi. Eur J Med Chem 38:265–275

Ramachandra L, Simmons D, Harding CV (2009) MHC molecules andmicrobial antigen processing in phagosomes. Curr Opin Immunol21:98–104

Rise ML, Jones SRM, Brown GD, von Schalburg KR, Davidson WS,Koop BF (2004)Microarray analyses identify molecular biomarkersof Atlantic salmon macrophage and hematopoietic kidney responseto Piscirickettsia salmonis infection. Physiol Genomics 20:21–35

Saeed AI, Bhagabati NK, Braisted JC, Liang W, Sharov V, Howe EA,Li J, Thiagarajan M, White JA, Quackenbush J (2006) TM4microarray software suite. Methods Enzymol 411:134–193

Salinas I, Maas EW, Muñoz P (2011) Characterization of acid phos-phatases from marine scuticociliate parasites and their activationby host’s factors. Parasitology 138:836–847

Seo JS, Kim MS, Lee SH, Kim KH, Lee HH, Jeong HD, Chung JK(2005) Uronema marinum: identification and biochemical char-acterization of phosphatidylcholine-hydrolyzing phospholipase C.Exp Parasitol 110:22–29

Simon MC, Schmidt HJ (2007) Antigenic variation in ciliates: antigenstructure, function, expression. J Eukaryot Microbiol 54:1–7

Sitjà-Bobadilla A, Palenzuela O, Alvarez-Pellitero P (2008) Immuneresponse of turbot, Psetta maxima (L.) (Pisces: Teleostei), toformalin-killed scuticociliates (Ciliophora) and adjuvanted formu-lations. Fish Shellfish Immunol 24:1–10

Skugor S, Jørgensen SM, Gjerde B, Krasnov A (2009) Hepatic geneexpression profiling reveals protective responses in Atlanticsalmon vaccinated against furunculosis. BMC Genomics 10:503

Song JY, Sasaki K, Okada T, Sakashita M, Kawakami H, Matsuoka S,Kang HS, Nakayama K, Jung SJ, Oh MJ, Kitamura SI (2009)

Antigenic differences of the scuticociliate Miamiensis avidusfrom Japan. J Fish Dis 32:1027–1034

Sterud E, Hansen MK, Mo TA (2000) Systemic infection withUronema-like ciliates in farmed turbot, Scophthalmus maximus(L.). J Fish Dis 23:33–37

Tingaud-Sequeira A, Chauvigne F, Lozano J, Agulleiro MJ, Asensio E,Cerda J (2009) New insights into molecular pathways associatedwith flatfish ovarian development and atresia revealed by tran-scriptional analysis. BMC Genomics 10:434

Toda H, Araki K, Moritomo T, Nakanishi T (2011) Perforin-dependentcytotoxic mechanism in killing by CD8 positive T cells in ginbunacrucian carp, Carassius auratus langsdorfii. Dev Comp Immunol35:88–93

Trowsdale J, Parham P (2004) Defence strategies and immunity-relatedgenes. Eur J Immunol 34:7–17

Tse HM, Milton MJ, Piganelli JD (2004) Mechanistic analysis of theimmunomdulatory effects of a catalytic antioxidant on antigen-presenting cells: implication for their use in targeting oxidation-reduction reactions in innate immunity. Free Radic Biol Med36:233–247

Tsoi SC, Ewart KV, Penny S, Melville K, Liebscher RS, Brown LL,Douglas SE (2004) Identification of immune-relevant genes fromAtlantic salmon using suppression subtractive hybridization. MarBiotechnol 6:199–214

Verhoef J (1991) The phagocytic process and the role of complementin host defense. J Chemother 3(Suppl 1):93–97

Wynne JW, O’Sullivan MG, Cook MT, Stone G, Nowak BF, LovellDR, Elliott NG (2008) Transcriptome analyses of amoebic gilldisease-affected Atlantic salmon (Salmo salar) tissues reveal lo-calized host gene suppression. Mar Biotechnol 10:388–403

Xu T, Ping J, Yu Y, Yu F, Yu Y, Hao P, Li X (2010) Revealing parasiteinfluence in metabolic pathways in Apicomplexa infectedpatients. BMC Bioinforma 11(11):S13

Zapata, A.G., Chibá, A., and Varas, A. (1996). Cells and tissues of theimmune system of fish. In: The Fish Immune System. Organism,Pathogen and Environment (Iwama, G., and Nakanishi, T., eds.).Academic Press. pp 1–62.

582 Mar Biotechnol (2012) 14:570–582