lable at ScienceDirect

Microbial Pathogenesis 65 (2013) 82e88

Contents lists avai

Microbial Pathogenesis

journal homepage: www.elsevier .com/locate/micpath

Tenacibaculum maritimum infection: Pathology andimmunohistochemistry in experimentally challenged turbot(Psetta maxima L.)

Luis Daniel Faílde a,*, Ana Paula Losada a, Roberto Bermúdez b, Ysabel Santos c,María Isabel Quiroga a

aDepartamento de Ciencias Clínicas Veterinarias, Facultad de Veterinaria, Universidad de Santiago de Compostela, 27002 Lugo, SpainbDepartamento de Anatomía y Producción Animal, Facultad de Veterinaria, Universidad de Santiago de Compostela, 27002 Lugo, SpaincDepartamento de Microbiología y Parasitología, Facultad de Biología, Universidad de Santiago de Compostela, 15782 Santiago de Compostela, Spain

a r t i c l e i n f o

Article history:Received 25 January 2013Received in revised form10 September 2013Accepted 13 September 2013Available online 1 October 2013

Keywords:Tenacibaculum maritimumPathologyImmunohistochemistryExperimental infectionPathogenesis

* Corresponding author. Tel.: þ34 982 820 000 223E-mail addresses: luisdaniel.failde@usc.es (L.D. Fa

(M.I. Quiroga).

0882-4010/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.micpath.2013.09.003

a b s t r a c t

Tenacibaculum maritimum is the causative agent of tenacibaculosis, a bacterial disease with a worldwidedistribution, which causes important losses in the turbot aquaculture. Despite the importance of thisbacterium, little is known about pathogenesis of the tenacibaculosis, pattern of lesions and the portal ofentry of T. maritimum. Turbots (Psetta maxima) were experimentally infected with T. maritimum usingsubcutaneous and intraperitoneal routes of inoculation and samples of skin and internal organs weretaken throughout the assay. Fish inoculated by both infection routes suffered a septicaemia but only thesubcutaneous inoculation reproduces the disease signs described in natural outbreaks. Bacterial antigenwas detected by immunohistochemistry in the internal organs 3 h after infection in fish inoculated bysubcutaneous route and 6 h after the inoculation of fish by intraperitoneal route. In summary, bothroutes of inoculation are able to cause an infection and bacteraemia in the fish. However, subcutaneousinoculation route reproduces the disease in a faster and more reliable way than the intraperitoneal route.Moreover, bacterium spreads along the internal organs easily, but needs a gateway to penetrate in theorganism and this portal of entry could be skin.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Tenacibaculosis is an important bacterial disease that affects alarge number of marine fish species, causing heavy losses for theaquaculture industry worldwide [1,2]. The disease is caused byTenacibaculum maritimum a Gram-negative filamentous bacteria[1]. Although this disease has a great impact in aquaculture, rela-tively little is known about its pathogenesis and routes of infection[3]. External lesions of the disease are very different depending onaffected fish species, for this reason the disease has been named inmany ways according to the macroscopic lesions [1,2,4e15]. Thepathology of tenacibaculosis has been associated with character-istic gross lesions on the body surface of fish such as ulcers, ne-crosis, eroded mouth, frayed fins and tail rots, and sometimesnecrosis on the gills and eyes, but the histopathological studies arevery scarce [1]. Recently, histopathological lesions of a severe

04.ílde), misabel.quiroga@usc.es

All rights reserved.

ulcerative disease outbreak associated with T. maritimum in culti-vated sole have been reported [15]. These authors describedextensive necrosis of the muscle layers with loss of epidermis anddermis and a mild to moderate inflammatory response [15].

It is known that fish are constantly exposed to waterbornepathogens. In order to cause disease pathogens need to gain accessto the host and penetrate its primary barriers [10]. It is generallyaccepted that the major routes of bacterial entry are through themucosal surfaces of skin, gastrointestinal tract and gills [11e15].

Animal testing is an essential tool for the study of infectiousdiseases in order to clarify the pathogenesis of the disease, tocharacterize the role of factors of the host and microorganisms inthe development of the pathologic process, or to test newdrugs andvaccines [16].

With regard to tenacibaculosis, different models of infectionhave been attempted to reproduce the disease. Wakabayashi et al.[17] and Baxa et al. [18] demonstrated that bath challengewas not areliable method of inducing the disease unless the skin was pre-viously scarified or abraded [19]. However, other authors havereproduced the disease in bath using prolonged immersion of fish[20] or high inoculum doses [21]. Other authors reported that

L.D. Faílde et al. / Microbial Pathogenesis 65 (2013) 82e88 83

intramuscular and intraperitoneal routes were not effective inreproducing tenacibaculosis in experimentally inoculated fish[7,17,20,22].

The aim of this research was evaluate subcutaneous and intra-peritoneal routes of entry of T. maritimum with the goal of gaininsight into the pathogenesis of this important disease to improvetreatment and prophylactic strategies and minimize economiclosses.

2. Materials and methods

2.1. Fish

One hundred fifty healthy turbot with an average weight of62.11�27.32gwere employed. Fishwere reared in500 l tanks (length115 � width 96 � depth 65 cm), with aerated and sand-filteredseawater (32& salinity) at 18 �C and fed with a commercial diet un-til the infection experiments started after the quarantine period.

2.2. Experimental procedure

For challenge, bacterial cultures of the strain LL01.8.3.8 (sero-type O1), grown for 48e72 h in FMM agar plates, were harvestedwith saline solution, washed two times by centrifugation (10,000�g for 10 min) and suspended in saline solution at a concentration of109 cells/ml (Tube 3, McFarland scale). Cell viability of bacterialsuspension was verified by the plate dilution method using FMMplates and counting the bacterial colonies produced.

Thirty fish were challenged by subcutaneous (SC) injection inthe dorsal median sinus [23,24] with 0.1 ml of the bacterial sus-pension (108 cfu/fish) of the strain LL01.8.3.8 of T. maritimum. Forthe intraperitoneal (IP) challenge, one group of 30 fish was inocu-lated with 0.1 ml of a high dose (HD) of the bacterial suspension(109 cfu/fish) and 30 fish were inoculated with 0.1 ml of a low dose

Fig. 1. SC experiment. A. Clusters of bacteria located in the subcutaneous region in skin anvessel (arrow). HeE. Bar 100 mm. C. Extensive necrosis affecting muscular, hypodermis andInset. Inflammatory cells in the damaged areas. PAS. Bar 10 mm. D. Ulcerative dermatitis wflammatory cells affecting superficial and deep muscles (asterisk). HeE. Bar 200 mm.

(LD) of the bacterial suspension (108 cfu/fish). Thirty control fishreceived 0.1 ml of saline sterile solution.

Samples from challenged and control fish were taken at 3, 6, 24,48, 72 h post-inoculation (hpi) and 7 days post-inoculation (dpi).Fish were sacrificed by overexposure to Tricaine methane sulpho-nate (MS222) (SigmaeAldrich, Germany) before sampling.

2.3. Bacteriological isolation

In each sample point, a small piece of spleen and kidney fromfive control and five experimentally infected fish were seeded onthe surface of plates containing Marine Agar (MA; Cultimed, Pan-reac Química Sau, Barcelona, Spain) and FMM agar [25] and incu-bated at 25 �C for 24e72 h for bacteriological analysis. Pure culturesof the isolates obtained in MA and FMM agar plates were identifiedusing morphological, physiological and biochemical tests and APIZYM (bioMérieux) [26,27] and were confirmed by PCR and sero-logical methods as previously described [28].

2.4. Histological and immunohistochemistry studies

In each sample point tissue samples from five control and fiveexperimentally infected fish were taken. Samples from skin andunderlying muscle, spleen, kidney, gastrointestinal tract, liver, gillsand heart were fixed in Bouin’s solution for 18e24 h and embeddedin paraffin wax. Sections 2e3 mm in thickness were collected onsilane coated slides, allowed to dry overnight, and then they weredewaxed and hydrated. The sections were stained with haema-toxylineeosin (HeE) and periodic acid Schiff (PAS) or used toimmunohistochemistry (IHC) technique.

For the IHC the samples of all fish used in the experiment weretested. All incubations were performed at room temperature in ahumid chamber and all washing procedures consisted of three suc-cessive 5 min immersions in phosphate-buffered saline (PBS; 8 mM

d superficial muscles (asterisk). HeE. Bar 200 mm. B. Inflammatory cells around blooddermis layers with an inflammatory response associated (asterisk). HeE. Bar 100 mm.ith loss of epidermis and dermis (arrows) and necrosis in the muscular layer with in-

Table 1Immunohistochemical detection of T. maritimum in different organs of subcutane-ously inoculated turbot at each sampling point.

Subcutaneously inoculated turbot

Skin Spleen Kidney Liver Heart GI tract Gills

3 hpi þ þ þ � � � �6 hpi þ þ þ � � � �24 hpi þ þ þ � � � �48 hpi þ þ þ � � � �72 hpi þ þ þ � � � �7 dpi þ þ þ � � � �

L.D. Faílde et al. / Microbial Pathogenesis 65 (2013) 82e8884

Na2HPO4, 3mMNaH2PO4,150mMNaCl, 0.5% (v/v) Tween20, pH7.4).Endogenous peroxidase activity was quenched by incubating inperoxidase blocking reagent (Dako, Denmark) for 30 min. The sec-tions were then washed again and incubated with a 1:400 (concen-trations of 10e0.01 mg/mL) working dilution of the affinity purified(HiTrap� Protein A HP, GE Healthcare) rabbit anti-T. maritimumLL01.8.3.8 immunoadsorbed antibody (anti-Tm) for 2 h, washed andincubated for 30 minwith an anti-rabbit EnVisionþ System LabelledPolymer-HRP (Dako). After rinsing, the sections were finally devel-oped using diaminobenzidine (Dako) as chromogen, washed in wa-ter, counterstained with haematoxylin, dehydrated and mounted.Negative controlswere carried out substituting the primary antibodyor the secondary antibody for PBS or an irrelevant polyclonal anti-body. All the immunostaining series were performed using anautostainer instrument (Autostainer Universal Staining System,Dako), in order to standardize the technique avoiding variations intimes of incubation, development and counterstaining.

3. Results

3.1. Bacterial isolation

T. maritimumwas isolated from all spleen and kidney samples offish inoculated both by SC and IP injection. No bacteria wererecovered or PCR detected from internal organs from fish of thecontrol group.

3.2. Clinical signs and gross lesions

In both groups of fish inoculated with T. maritimum bacterialsuspensions, anorexia and lethargy were observed from 3 hpi to theend of the experiment in all fish.

Fig. 2. Immunohistochemistry against T. maritimum antigen in SC experiment. A. Strong inopositivity in the cytoplasm of phagocyte cells in the point of inoculation in the hypodermiimmunostaining in the cytoplasm of phagocyte cells at 24 hpi. Bar 200 mm. D. Ulcer withcytoplasm at 7 dpi. Bar 100 mm. E. Muscular fibres displayed evident degeneration and necrDetail of phagocyte cells detected in the damaged areas with the cytoplasm strongly staine

In the fish inoculated subcutaneously, the first gross lesionswere detected at 24 hpi, consisting in slight depigmentation of theskin at the point of inoculation over the dorsal median sinus. Thedepigmentation area progressed along the skin over the dorsalmedian sinus at 48 hpi. The inoculation point was swollen anddiscrete skin ulceration was observed. At 72 hpi, in the inoculationpoint the lesions displayed a circular appearance with a slighthyperaemic halo around. Finally, at 7 dpi, fish showed a moderateulcerative dermatitis and hyperaemia in the area of inoculationwith loss of the epidermis and dermis. Besides, some fish exhibiteddiffuse hyperaemia or haemorrhages in the base and tip of ventraland dorsal fins and in the interradial tissue of pectoral and caudalfins in the last samplings.

In fish inoculated intraperitoneally, no macroscopical lesionswere observed along the experiment in any fish.

3.3. Histopathological examination and immunohistochemistry

3.3.1. SC experimentThe study of all fish inoculated subcutaneously gives an evolu-

tion in the lesions since fish showed different degrees of

mmunostaining in the point of inoculation (asterisk) at 3 hpi. Bar 500 mm. B. Immu-s (asterisk) at 24 hpi. Bar 200 mm. C. Severe necrosis in the muscles (asterisk) and strongmuscles exposed and a high number of phagocyte cells with positive staining in theosis with strong immunostaining in the inflammatory cells at 7 dpi. Bar 100 mm. Inset.d. Bar 10 mm.

L.D. Faílde et al. / Microbial Pathogenesis 65 (2013) 82e88 85

histological damage along the experiment. At 3 hpi, in the point ofinjection, clusters of bacteria among connective tissue of the hy-podermis and muscular bundles could be seen. Fragments ofdegenerated muscular tissue were observed without inflammatoryresponse associated (Fig. 1A).

At 6 hpi, an early inflammatory response consisting of gran-ulocytes was seen around blood vessels as well as between themuscle bundles, spreading into the damaged area (Fig. 1B).

At 24, 48, 72 hpi and 7 dpi degeneration of muscle fibres showeddifferent degrees of severity. Some fibres displayed shrinkage andseparation of muscular epimysium, loss of striation, and finallyareas of necrosis surrounded by inflammatory cells, mainly PASpositive granulocytes and macrophages, that extended through theconnective tissue of hypodermis reaching and affecting othermuscular packets (Fig. 1C and C inset). The epidermis and dermiswere detached or absent, and scattered inflammatory cellsextending to the ulcerated area were observed (Fig. 1D). In thesamples of internal organs, no evidence of histopathologicalchanges was present in any specimen along the experiment.

Immunopositivity was mainly detected in skin and muscle fi-bres, spleen and kidney (Table 1).

Fig. 3. Immunohistochemistry against T. maritimum antigen in SC experiment. A. Distributioin spleen surrounded by round cells (arrow) with a strong positivity in the cytoplasm. Bar 20endothelial cells with positive cytoplasm (arrow). Bar 20 mm. D. Distribution of bacterialperitubular capillaries with cytoplasm positive against the antibody (arrowhead) and distribcells in sinusoids of kidney with cytoplasm staining with the antibody (arrowhead). Bar 20

In samples of skin and muscles, evolution of lesions was clearlydefined. At 3 hpi, in the point of injection, a great amount of pos-itive bacterial antigen was detected, affecting hypodermis and su-perficial muscular packages, but no inflammatory response wasassociated (Fig. 2A). At 24 hpi, immunoreactivity into the cytoplasmof the granulocytes and macrophages was detected, mainly in theinoculation site and spreading through connective tissue of hypo-dermis (Fig. 2B) and muscles (Fig. 2C).

The bacterial antigen was also observed in areas with severeulcerative dermatitis (Fig. 2D and E). In these locations, phagocytecells showed the cytoplasm with a strong immunostaining againstthe bacterial antigen (Fig. 2 inset).

In spleen, the immunoreactivity was distributed mainly in thecytoplasm of macrophages located around the ellipsoids from 3 hpito the end of the experiment (Fig. 3A and B). In kidney a largenumber of positive macrophages randomly distributed in the renalinterstitial tissue and within and around peritubular capillaries wasdetected from 3 hpi to the end of the experiment (Fig. 3D and E).

Moreover, bacterial antigenwas detectedwithin the sinusoids inthe cytoplasm of macrophage cells and in the cytoplasm of endo-thelial cells in both organs (Fig. 3C and F).

n of bacterial antigen around the ellipsoids in spleen. Bar 100 mm. B. Capillary ellipsoidmm. C. Sinusoid in the spleen with positive round cells in bloodstream (arrowhead) andantigen in the peritubular capillaries of kidney. Bar 100 mm. E. Macrophages into theuted between the cells of parenchyma of the kidney (arrow). Bar 20 mm. F. Endothelialmm.

Table 2Immunohistochemical detection of T. maritimum in different organs of intraperito-neally inoculated turbot at each sampling point.

Intraperitoneally inoculated turbot

Skin Spleen Kidney Liver Heart GI tract Gills

LD HD LD HD LD HD LD HD LD HD LD HD LD HD

3 hpi � � � � � � � � � � � � � �6 hpi � � þ þ þ þ � þ � þ � þ � þ24 hpi � � þ þ þ þ � þ � þ � þ � þ48 hpi � � � þ � þ � þ � þ � þ � þ72 hpi � � � þ � þ � � � � � � � �7 dpi � � � þ � þ � � � � � � � �

Fig. 4. IP experiment. A. Spleen with a necrotic zone (asterisk) in the serosa and extending into the pancreas at 7 dpi. HeE. Bar 100 mm. B. Kidney showing haemorrhage, necrosis(asterisk) and depletion of haematopoietic tissue at 7 dpi. HeE. Bar 100 mm. C. Liver with necrotic area in the parenchyma (asterisk) at 7 dpi. HeE. Bar 100 mm. D. Intestinal tractshowed inflammation and necrosis (arrow) that extends from serous layer to muscular layer at 6 hpi. HeE. Bar 200 mm. E. Sample of stomach with loss of epithelium and presence ofmild inflammatory response at 6 hpi. HeE. Bar 200 mm.

L.D. Faílde et al. / Microbial Pathogenesis 65 (2013) 82e8886

3.3.2. IP experimentNeither microscopical lesions nor reactivity with the antibody

anti-Tmwere detected along the experiment in the skin of turbot IPinjected.

In the animals inoculated with HD, the spleen showed inflam-matory cells and necrosis in the serosa, extending to the pancreas(Fig. 4A). Extensive haemorrhage was observed in the parenchymaof kidney, with necrosis and depletion of haematopoietic tissue(Fig. 4B). The liver showed multifocal areas of necrosis with slightinflammatory response and haemorrhage (Fig. 4C). In the gastro-intestinal tract, the inflammatory cells were observed extendingfrom the serosa to the muscular layer along with haemorrhage andnecrosis (Fig. 4D). Moreover, in some areas of the mucosa it couldbe seen a slight epithelial desquamation and presence of inflam-matory cells (Fig. 4E). In general, in the affected organs, the damagewas very extensive and spreading from the serous layer of the or-gans to the central parenchyma.

The spleen and kidney from HD inoculated turbot showedimmunopositivity against the bacterial antigen from 6 hpi to theend of the study (Table 2). In the LD group, the bacterial antigenwasdetected only at 6 hpi and at 24 hpi in spleen and kidney. In bothgroups, the immunoreactivity displayed a similar distribution thatin subcutaneously injected fish.

Occasionally, in liver from turbot of HD experiment, immuno-positive cells were detected in serosa and in necrotic areas in theparenchyma (Fig. 5A and Table 2).

In addition, in some fish of HD experiment, the gastrointestinaltract displayed positive bacterial antigen in serosa and in someareas, the immunostaining spreaded to the muscular layer (Fig. 5B)reaching the mucosal surface, where there was also a strong andextensive positive staining in necrotic areas and vessels (Fig. 5Cand D).

In turbot from the HD experiment, gills displayed strong positivestaining into vessels especially on the top of lamellas (Fig. 5E). In

the heart, immunostaining was seen into the cytoplasm ofmacrophage cells located into the sinus venous (Fig. 5F).

4. Discussion

The present study provides the first description of morphopa-thological and immunohistochemical changes in the tissues ofturbot experimentally infected with T. maritimum by subcutaneousand intraperitoneal inoculation. The experiments showed that in-jection by both routes is able to reproduce the disease and is reli-able to study the pathogenesis, although by the IP injection thecutaneous lesions were not reproduced. In addition, in the IPexperiment a higher concentration of bacteria was necessary tocause damage in the internal organs and to detect the bacterialantigen in a large number of organs. Besides, the subcutaneousinoculation in the dorsal median sinus, reproduces tenacibaculosisin similar way to natural infection, according to gross and micro-scopic lesions previously described [7,15,21,29e31].

Histopathological descriptions about tenacibaculosis are scantand mainly focused in gills and ulcerative lesions in skin

Fig. 5. Immunohistochemistry against T. maritimum antigen in IP experiment. A. Liver with a necrotic area in the parenchyma and a marked immunostaining. Bar 100 mm. B.Intestinal tract showed necrosis with positive staining extending from serous layer (arrowhead) to muscular layer (arrow). Bar 200 mm. C. Immunopositivity in muscular layers(arrow), vessels (arrowhead) and necrotic area (asterisk) in the mucosa of gastrointestinal tract. Bar 200 mm. D. Intestinal surface showing evident immunostaining (arrow). Bar200 mm. E. Gill with strong positive staining into vessels on the top of lamellas. Bar 100 mm. F. Immunostaining in heart into the sinus venous (arrow). IHC. Bar 200 mm.

L.D. Faílde et al. / Microbial Pathogenesis 65 (2013) 82e88 87

[8,21,30,32,33] with scarce references on distribution of bacterialantigen in skin and internal organs.

In the SC inoculation, an inflammatory response in the point ofinoculation was observed at 6 hpi to the end of the experiment. Inthe muscles, histopathological changes started with mild degen-erative changes and finally extensive areas of necrosis wereobserved. Similar pathological changes have been observed in sal-monids [30] and sole [15]. In the fish IP inoculated withT. maritimum, no lesions were observed in skin and muscularsamples along the entire study. However, organs in coelomic cavityshowed an inflammatory response in different serosas and necrosisin spleen, kidney, liver and gastrointestinal tract.

Immunohistochemical analysis was able to identifyT. maritimum antigen in both experiments showing differentpattern of distribution according to the route of inoculation.Interestingly, SC route was faster than IP route with regard to thedistribution of T. maritimum in fish tissues since bacterial antigenwas observed at 3 hpi and throughout the experiment in kidneyand spleen; and later at 6 hpi in spleen and kidney of turbotsintraperitoneally inoculated with the HD bacterial suspension.

In our SC study, a strong immunopositive reactionwas observedin the hypodermis in the point of inoculation, associated with

bacterial antigen detection within blood vessels. This finding coulddemonstrate that T. maritimum invades the bloodstream andrapidly establish a septicaemia. Similar positive reaction was re-ported by Lovoll [10] in Atlantic salmon infected by Moritella vis-cosa. These authors also reported that extracellular products couldbe part of a strategy of invasion of the bacteria that can use thebloodstream to disseminate and facilitate a systemic infection. Inthis sense, it has been reported that extracellular products fromT. maritimum possess several virulence factors like toxins and en-zymes [30,34e36] which have high proteolytic and haemolyticactivity [35] and cause necrosis [36]. For this reason, it is suggestedin the current study that T. maritimum could follow the samestrategy to damage the tissue and reach the bloodstream reportedto M. viscosa [10]. Besides, in the cutaneous lesions granulocytesand macrophages are the main inflammatory cells observed.Therefore, fish phagocytes would not only be able to respond topathogen-derived molecules [37] but also to endogenous mole-cules, such as collagen derived proteolytic fragments, that wouldsignal the presence of damage [38]. In fact, recently it has beenproved that collagen proteolytic fragments regulate the activationof phagocytic cells in gilthead seabream and exacerbate the im-mune response [39]. This finding would indicate that inflammatory

L.D. Faílde et al. / Microbial Pathogenesis 65 (2013) 82e8888

response, mainly composed by phagocytes, would be intenser inareas where the tissular damage was more evident and the releaseof inflammatory mediators from these cells might contribute to thedamage.

In the current study, gastrointestinal tract displayed positivebacterial antigen in serosa, spreading to the muscular layer andreaching the mucosal surface, where the bacterial antigen wasdetected above the epithelium. This finding supports the hypoth-esis that the intestine of infected fish could be a reservoir of thebacteria and is similar to the results obtained our group in tenaci-baculosis outbreaks in sole (data not published).

Moreover, erosion, necrosis and bacterial filamentous mats ingills have been reported associated with T. maritimum in salmonids[40,41]. However, in the present study, the presence of T. maritimumantigen in blood vessels of gills as well as the absence of histologicallesions may point to the arrival of the pathogen at this organ bybloodstream, although future studies should be performed toelucidate the role of the gills in the pathogenesis of the disease.

5. Conclusions

In summary, both routes of inoculation are able to cause diseaseand septicaemia in the fish. However, subcutaneous route re-produces the disease faster and in a more similar way than intra-peritoneal route. Macroscopic and microscopic findings indicatethat T. maritimum causes a great local damage in the point of entryand triggers an inflammatory response. Finally, bacterium spreadsalong the internal organs easily, but needs a gateway to penetrate inthe organism and this portal of entry could be skin.

References

[1] Santos Y, Pazos F, Barja JL. Flexibacter maritimus, causal agent of flexibacter-iosis in marine fish. In: Oliver G, editor. ICES identification leaflets for diseasesand parasites of fish and shellfish, vol. 55. Copenhagen, Denmark: Interna-tional Council for the Exploration of the Sea; 1999. p. 1e6.

[2] Toranzo AE, Magariños B, Romalde JL. A review of the main bacterial fishdiseases in mariculture systems. Aquaculture 2005;246:37e61.

[3] Avendaño-Herrera R, Irgang R, Magariños B, Romalde JL, Toranzo AE. Use ofmicrocosms to determine the survival of the fish pathogen Tenacibaculummaritimum in seawater. Environ Microbiol 2006;8:921e8.

[4] Baxa DV, Kawai K, Kusuda R. Characteristics of gliding bacteria isolated fromdiseased cultured flounder, Paralichthys olivaceous. Fish Pathol 1986;21:251e8.

[5] Wakabayashi H, Hikida M, Masumura K. Flexibacter maritimus sp. nov., apathogen of marine fishes. Int J Syst Bacteriol 1986;36:396e8.

[6] Bernardet JF, Campbell AC, Buswell JA. Flexibacter maritimus is the agent of“black patch necrosis” in Dover sole in Scotland. Dis Aquat Org 1990;8:233e7.

[7] Alsina M, Blanch AR. First isolation of Flexibacter maritimus from cultivatedturbot (Scophthalmus maximus). Bull Eur Assoc Fish Pathol 1993;13:157e60.

[8] Chen ME, Henry-Ford D, Groff JM. Isolation and characterization of Flexibactermaritimus frommarine fishes of California. J Aquat Anim Health 1995;7:318e26.

[9] Ostland VE, LaTrace C, Morrison D, Ferguson HW. Flexibacter maritimusassociated with a bacterial stomatitis in Atlantic salmon smolts reared in net-pens in British Columbia. J Aquat Anim Health 1999;11:35e44.

[10] Lovoll M, Wiik-Nielsen CR, Tunsjo HS, Colquhoun D, Lunder T, Sorum H, et al.Atlantic salmon bath challenged with Moritella viscosa e pathogen invasionand host response. Fish Shellfish Immunol 2009;26:877e84.

[11] Smith PA, Pizarro P, Ojeda P, Contreras J, Oyanedel S, Larenas J. Routes of entryof Piscirickettsia salmonis in rainbow trout Oncorhynchus mykiss. Dis Aquat Org1999;37:165e72.

[12] Ling SHM, Wang XH, Lim TM, Leung KY. Green fluorescent protein-taggedEdwardsiella tarda reveals portal of entry in fish. FEMS Microbiol Lett2001;194:239e43.

[13] Ringø E, Jutfelt F, Kanapathippillai P, Bakken Y, Sundell K, Glette J, et al.Damaging effect of the fish pathogen Aeromonas salmonicida ssp. salmonicidaon intestinal enterocytes of Atlantic salmon (Salmo salar L.). Cell Tissue Res2004;318:305e11.

[14] Ringø E, Myklebust R, Mayhew TM, Olsen RE. Bacterial translocation andpathogenesis in the digestive tract of larvae and fry. Aquaculture 2007;268:251e64.

[15] Vilar P, Faílde LD, Bermúdez R, Vigliano F, Riaza A, Silva R, et al. Morphopa-thological features of a severe ulcerative disease outbreak associated withTenacibaculum maritimum in cultivated sole, Solea senegalensis (L.). J Fish Dis2012;35:437e45.

[16] Chiavolini D, Pozzi G, Ricci S. Animal models of Streptococcus pneumoniaedisease. Clin Microbiol Rev 2008;21:666e85.

[17] Wakabayashi H, Hikida M, Masumura K. Flexibacter infection in culturedmarine fish in Japan. Helgol Meeresunters 1984;37:587e93.

[18] Baxa DV, Kawai K, Kusuda R. Experimental infection of Flexibacter maritimusin black sea bream (Acanthopagrus schlegeli) fry. Fish Pathol 1987;22:105e9.

[19] Bernardet JK, Kerouault B, Michel C. Comparative study on Flexibacter mar-itimus strains isolated from farmed sea bass (Dicentrarchus labrax) in France.Fish Pathol 1994;29:105e11.

[20] Avendaño-Herrera R, Toranzo AE, Magariños B. A challenge model for Tena-cibaculum maritimum infection in turbot, Scophthalmus maximus (L.). J Fish Dis2006;29:371e4.

[21] van Gelderen R, Carson J, Nowak B. Experimentally induced marine flex-ibacteriosis in Atlantic salmon smolts Salmo salar. II. Pathology. Dis Aquat Org2011;95:125e35.

[22] Pepin JF, Emery E. Marine Cytophaga-like bacteria (CLB) isolated from diseasedreared sea bass (Dicentrarchus labrax L.) from French Mediterranean coast.Bull Eur Assoc Fish Pathol 1993;13:165e7.

[23] Treasurer J, Cox C. Intraperitoneal and dorsal median sinus vaccination effectson growth, immune response, and reproductive potential in farmed Atlanticsalmon Salmo salar. Aquaculture 2008;275:51e7.

[24] LaPatra SE, Plant KP, Alcorn S, Ostland V, Winton J. An experimental vaccineagainst Aeromonas hydrophila can induce protection in rainbow trout Onco-rhynchus mykiss (Walbaum). J Fish Dis 2010;33:143e51.

[25] Pazos F, Santos Y, Macías AR, Núñez S, Toranzo AE. Evaluation of media for thesuccessful culture of Flexibacter maritimus. J Fish Dis 1996;19:193e7.

[26] Santos Y, Romalde JL, Bandín I, Magariños B, Núñez S, Barja JL, et al. Usefulnessof the API-20E system for the identification of bacterial fish pathogens.Aquaculture 1993;116:111e20.

[27] Bernardet JF, Nakagawa Y, Holmes B. Proposed minimal standards fordescribing new taxa of the family Flavobacteriaceae and emended descriptionof the family. Int J Syst Evol Microbiol 2002;52:1049e70.

[28] Cepeda C, García-Márquez S, Santos Y. Detection of Flexibacter maritimus infish tissues using nested PCR amplification. J Fish Dis 2003;26:65e70.

[29] Pazos F, Santos Y, Núñez S, Toranzo AE. Increasing occurrence of Flexibactermaritimus in marine aquaculture of Spain. Fish Health Sec Asian Fish SocNewsl 1993;21:1e2.

[30] Handlinger J, Soltani M, Percival S. The pathology of Flexibacter maritimus inaquaculture species in Tasmania, Australia. J Fish Dis 1997;20:159e68.

[31] Cepeda C, Santos Y. First isolation of Flexibacter maritimus from farmed Sen-egalese sole (Solea senegalensis, Kaup) in Spain. Bull Eur Assoc Fish Pathol2002;22:388e92.

[32] Powell M, Carson J, van Gelderen R. Experimental induction of gill disease inAtlantic salmon Salmo salar smolts with Tenacibaculum maritimum. Dis AquatOrg 2004;61:179e85.

[33] Olsen AB, Nilsen H, Sandlund N, Mikkelsen H, Sørum H, Colquhoun DJ.Tenacibaculum sp. associated with winter ulcers in sea-reared Atlantic salmonSalmo salar. Dis Aquat Org 2011;94:189e99.

[34] Baxa DV, Kawai K, Kusuda R. In vitro and in vivo activities of Flexibactermaritimus toxins. Rep USA Mar Biol Inst Kochi Univ 1988;10:1e8.

[35] Pazos F. “Flexibacter maritimus”: estudio fenotípico, inmunológico y molec-ular. Tesis Doctoral. Spain: Universidad de Santiago de Compostela, Santiagode Compostela; 1997.

[36] van Gelderen R, Carson J, Nowak B. Effect of extracellular products of Tena-cibaculum maritimum in Atlantic salmon, Salmo salar L. J Fish Dis 2009;32:727e31.

[37] Sepulcre MP, López-Castejón G, Meseguer J, Mulero V. The activation of gilt-head seabream professional phagocytes by different PAMPs underlines thebehavioural diversity of the main innate immune cells of bony fish. MolImmunol 2007;44:2009e16.

[38] Oppenheim JJ, Yang D. Alarmins: chemotactic activators of immune re-sponses. Curr Opin Immunol 2005;17:359e65.

[39] Castillo-Briceño P, Sepulcre MP, Chaves-Pozo E, Meseguer J, García-Ayala A,Mulero V. Collagen regulates the activation of professional phagocytes of theteleost fish gilthead seabream. Mol Immunol 2009;46:1409e15.

[40] van Gelderen R, Carson J, Nowak B. Experimentally induced marine flex-ibacteriosis in Atlantic salmon smolts Salmo salar. I. Pathogenicity. Dis AquatOrg 2010;91:121e8.

[41] Mitchell SO, Rodger HD. A review of infectious gill disease in marine salmonidfish. J Fish Dis 2011;34:411e32.