Taxonomic composition of bacteria associated with cultivated mollusks Crassostrea lugubris and Perna viridis and with the water of the Gulf of Nha Trang Lagoon, Vietnam

One hundred and four strains of heterotrophic bacteria have been isolated and characterized from two species of bivalve mollusks cultivated in the Gulf of Nha Trang (Vietnam) and from the water of a mariculture farm. The isolates have been identified on the basis of morphological, physiological, biochemical, and chemotaxonomic properties, as well as by the content of G+C bases in DNA. In the microflora of mollusks, Vibrio alginolyticus was predominant; the pathogenic species V. harveyi and V. splendidus were found as well. Staphylococci and bacilli occupied the second place in abundance after vibrios. In addition, coryneforms and enterobacteria, as well as Pseudomonas spp. and Pseudoalteromonas spp., were revealed. The composition of the water microflora was more diverse as compared with the microflora of mollusks. In the water, Bacillus spp., Vibrio spp., and Pseudomonas spp. were predominant. Brevibacterium spp. and other coryneform bacteria, as well as enterobacteria, occurred in significant amounts. In addition, Pseudoalteromonas spp., Marinococcus sp., Halobacillus sp., Shewanella sp., Sulfitobacter sp., and bacteria of the CFB cluster were noticed. The presence of pathogenic and conditionally pathogenic bacterial species in the water and mollusks is probably the reason for the high death rate of cultivated animals at the mariculture farm.     

Immunohistochemistry of great scallop Pecten maximus larvae experimentally challenged with pathogenic bacteria

Three challenge experiments were carried out on larvae of the great scallop Pecten maximus. Larvae were bath-challenged with Vibrio pectenicida and 5 strains resembling Vibrio splendidus and one Pseudoalteromonas sp. Unchallenged larvae were used as negative controls. The challenge protocol was based on the use of a multidish system, where the scallop larvae (10, 13 and 15 d post-hatching in the 3 experiments, respectively) were distributed to 2 ml wells with stagnant seawater and exposed to the bacterial cultures by bath challenge. Presence of the challenge bacteria in the wells was verified by polymerase chain reaction (PCR). A significantly increased mortality was found between 24 and 48 h in most groups challenged with V. pectenicida or V. splendidus-like strains. The exception was found in larval groups challenged with a Pseudoalteromonas sp. LT 13, in which the mortality rate fell in 2 of the 3 challenge experiments. Larvae from the challenge experiments were studied by immunohistochemistry protocol. Examinations of larval groups challenged with V. pectenicida revealed no bacterial cells, despite a high degree of positive immunostaining. In contrast, intact bacterial cells were found in larvae challenged with V. splendidus. In the case of larvae challenged with the Pseudoalteromonas sp., positive immuno-staining was limited to visible bacteria inside the digestive area and cells of the mucosa. The experiments confirm that V. splendidus and V. pectenicida are pathogenic to scallop larvae, and that the Pseudoalteromonas strain is probably not a primary pathogen, although it cannot be ruled out as a secondary pathogen.     

Vibrio splendidus biotype 1 as a cause of mortalities in hatchery-reared larval turbot, Scophthalmus maximus (L.)

AIMS: To characterize bacteria associated with turbot larvae feeding on Artemia and identify pathogens causing mortalities in larvae. METHODS AND RESULTS: To identify bacteria associated with mortalities in larval turbot rearing, bacteria were isolated from homogenates of Artemia or from several batches of well-performing or poorly performing turbot larvae. Samples were plated onto marine agar and were characterized using biochemical tests and BIOLOG GN plates. Total culturable aerobic bacteria ranged from 1.9 x 10(5) to 1.8 x 10(6) CFU per larva and >96% of bacteria identified were vibrios. Almost all bacteria were haemolytic and clustered into two phenons represented by Vibrio alginolyticus and Vibrio splendidus. The bacterial flora of Artemia was almost entirely V. alginolyticus, whereas V. splendidus biotype 1 dominated the larval turbot gut flora (69/115 isolates in seven experiments) and formed four different groups based on BIOLOG GN reactions. Of 16 isolates tested for virulence towards turbot larvae, four of the 11 V. splendidus biotype 1 isolates were lethal and all belonged to the same group of V. splendidus biotype 1 isolates. CONCLUSIONS: In a commercial turbot hatchery, the microbial flora of the larval gut was dominated by V. splendidus biotype 1. Four of the 11 V. splendidus biotype 1 isolates caused mortalities in larval turbot and all belonged to one group of the biotype 1 strains identified. SIGNIFICANCE AND IMPACT OF THE STUDY: Identification of four isolates of V. splendidus that are pathogenic for turbot larvae from three separate batches of larval turbot will allow these to be compared with avirulent isolates to define how V. splendidus causes mortalities in larval turbot.     

Identification of Vibrio spp. (other than V. vulnificus) recovered on CPC agar from marine natural samples.

Two hundred and eighty four presumptive but not confirmed Vibrio vulnificus isolates grown on cellobiose-polymixin B-colistin agar (CPC) at 40 degrees C, recovered from sea water samples from Valencia, Spain, during a microbiological survey for V. vulnificus, were phenotypically identified. Most of the isolates (91%) corresponded to Vibrio species. V. harveyi (24%) and V. splendidus(19%) were the most abundant species identified, followed by V. navarrensis (13%), V. alginolyticus (8%) and V. parahaemolyticus (5%). The ability to grow on CPC agar and ferment cellobiose of several V. vulnificus strains from different origins and serovars, including reference strains, was tested. Most serovar E isolates and 25% of non-serovar E isolates could not grow on CPC agar.     

Pathogenicity of vibrio splendidus strains associated with turbot larvae, scophthalmus maximus

Turbot larvae were challenged with eight strains of Vibrio splendidus isolated from diseased larvae, plus a ninth strain pathogenic to scallop larvae (A515; Nicolas et al . 1996 ). Six strains caused heavy mortality but the scallop pathogen and the other two strains did not. All the strains shared a large number of phenotypic traits, and an attempt was made to relate virulence to genotype and phenotype. Five of the six pathogenic strains were very similar, as shown by RAPD fingerprinting and phenotypic characteristics. The relatedness of the other strains was intermediate between the main pathogenic group and V. splendidus ATCC 33125, but the DNA–DNA homology between the pathogenic group and the reference strain was still high (78% of reassociation rate). The non-pathogenic isolates may be a useful tool for determining the possible virulence factors, as all the isolates differed by few characteristics.     

Bacterial influences on Atlantic halibut Hippoglossus hippoglossus yolk-sac larval survival and start-feed response

A bacteria-free halibut larval rearing system was used to test 20 bacterial isolates, from British halibut hatcheries, for their toxicity towards halibut yolk-sac larvae under microbially controlled conditions. The isolates tested spanned a range of genera and species (Pseudoalteromonas, Halomonas marina, Vibrio salmonicida-like, Photobacterium phosphoreum and V. splendidus species). A pathogen of turbot, Scophthalmus maximus, V. anguillarum 91079, and 2 isolates from adult halibut were also included. Isolates were inoculated, at a concentration of 5 x 10(2) cfu ml(-1), into flasks containing 25 recently hatched axenic halibut larvae, using a minimum of 3 flasks for each treatment. Control survivals to 38 d post-hatch for the 3 experiments averaged 84, 51.5 and 49%, respectively. With the exception of V. anguillarum 91079, which was highly pathogenic towards halibut yolk-sac larvae, there was no statistically significant difference in survival between the controls and the different treatments. This suggests that most of the bacteria routinely isolated from halibut hatcheries are not harmful to yolk-sac larvae, even though most flasks contained in excess of 5 x 10(6) cfu m(-1) of the inoculated organism when the experiments were terminated. Three organisms previously shown to inhibit growth of bacteria in vitro were tested for their ability to protect halibut yolk-sac larvae against invasion by V. anguillarum. In 4 separate challenge experiments none of the test isolates, a Pseudoalteromonas strain and 2 Carnobacterium-like organisms, showed any protective effect. To investigate how particular bacteria influence their start-feed response, larvae were fed axenic and gnotobiotic Artemia colonized with a range of different Vibrio spp., and examined after 8 d. There were no statistically significant between-treatment differences in the proportion of Artemia-containing larvae, indicating that bacterial contamination of the live food does not appear to influence initiation of the feeding response.     

香港养殖海鲷弧菌致病菌药物敏感性及耐药质粒研究

从发病海鲷(Sparus sarba)中共分离到51株弧菌(\%Vibrio)\%,经API20E细菌快速鉴定系统及Alsina和Blanch关键生理生化特性分析鉴定为7个种,它们分别是：溶藻胶弧菌(\%V.alginolyticus)(24株),创伤弧菌(V.vulnificus)(12株)和副溶血弧菌(V.parahaemolyticus)(7株),火神弧菌(V.logei)(4株),远洋弧菌Ⅱ菌(V.pelagius Ⅱ)(2株),河弧菌(V.fluvialis)(1株)和地中海弧菌(V.mediterranei)(1株)\%。其中3种优势菌溶藻胶弧菌创伤弧菌和副溶血弧菌证实对海鲷有致病性。另外采用平板稀释法检测了51株菌对16种抗菌素的敏感性。发现所有菌株对ceftriaxone,链霉素,萘啶酮酸和利福霉素敏感,几乎所有菌株对ceftazidime, netilimicin,氯霉素和sulfamethoxazole敏感.大部分菌株对氨苄青霉素 (60.8%),cefuroxime(667%),丁胺卡那霉素(55%),卡那霉素(588%)和三甲氧苄氨嘧啶(765%)等具有较强的耐药性。通过对菌株中所含有的耐药质粒进行分析,发现15株菌株含有1～4个质粒,分子量范围为9～123kb之间,对12株既含有较大分子量质粒又具有耐药性的菌株进行了质粒转化试验,结果其中9株菌的质粒具有转化能力,转化率为１０－１１～１０－９,表明所分离的菌株的抗药性是由于细菌染色体相关突变造成的。     

Molecular determination of oxytetracycline-resistant bacteria and their resistance genes from mariculture environments of China

AIMS: To assess the diversity of antibiotic-resistant bacteria and their resistance genes in typical maricultural environments. METHODS AND RESULTS: Multidrug-resistant bacteria and resistance genes from a mariculture farm of China were analysed via cultivation and polymerase chain reaction (PCR) methods. Oxytetracycline (OTC)-resistant bacteria were abundant in both abalone and turbot rearing waters, accounting for 3.7% and 9.9% of the culturable microbes. Multidrug resistance was common, with simultaneous resistance to OTC, chloramphenicol and ampicillin the most common resistance phenotype. 16S rDNA sequence analyses indicate that the typical resistant isolates belonged to marine Vibrio, Pseudoalteromonas or Alteromonas species, with resistance most common in Vibrio splendidus isolates. For OTC resistance, tet(A), tet(B) and tet(M) genes were detected in some multidrug-resistant isolates, with tet(D) being the most common molecular determinant. For chloramphenicol resistance, cat II was common, and floR was also detected, especially in marine Pseudoalteromonas strains. CONCLUSIONS: There is the risk of multidrug-resistant bacteria contamination in mariculture environments and marine Vibrio and Pseudoalteromonas species serve as reservoirs of specific antibiotic resistance determinants. SIGNIFICANCE AND IMPACT OF THE STUDY: This paper and similar findings from Korea and Japan indicate the potential for widespread distribution of antibiotic resistance genes in mariculture environments from the East Asian region of the world.     

Selection and identification of autochthonous potential probiotic bacteria from turbot larvae (Scophthalmus maximus) rearing units

The purpose of this study was to select, identify and characterise bacteria as a disease control measure in the rearing of marine fish larvae (turbot, Scophthalmus maximus). Thirty-four out of 400 marine bacterial strains exhibited in vitro anti-bacterial activity against three fish larval pathogens. Two strains originated from culture collections and thirty two strains were isolated directly from turbot larvae rearing units using a pre-selection procedure to facilitate detection of antagonists. Approximately 8,500 colonies from colony-count plates were replica-plated on agar seeded with Vibrio anguillarum, and 196 of them caused zones of clearing in the V. anguillarum agar layer. Of these, 32 strains exhibited reproducible antibacterial properties in vitro when tested against the fish pathogens V. anguillarum 90-11-287, V. splendidus DMC-1 and a Pseudoalteromonas HQ. Seventeen antagonists were identified as Vibrio spp. and four of twelve tested were lethal to yolk-sac larvae. The 15 remaining strains were identified as Roseobacter spp. based on phenotypic criteria and 16S rDNA gene sequence analysis of two strains representing the two major RAPD groups. Most of the remaining 164 strains selected in the initial replica plating were identified as Vibrionaceae or Pseudoalteromonas. Roseobacter spp. were not lethal to egg yolk sac turbot larvae and in two of three trials, the mortality of larvae decreased (p > 0.001) in treatments where 10(7) cfu/ml Roseobacter sp. strain 27-4 was added, indicating a probiotic potential.     

Characterization of Vibrio anguillarum and closely related species isolated from farmed fish in Norway.

A total of 264 bacterial strains tentatively or definitely classified as Vibrio anguillarum were examined. The strains were isolated from diseased or healthy Norwegian fish after routine autopsy. With the exception of five isolates from wild saithe (Pollachius virens), the strains originated from nine different species of farmed fish. The bacteria were subjected to morphological, physiological, and biochemical studies, numerical taxonomical analyses, serotyping by slide agglutination and enzyme-linked immunosorbent assay, DNA-plasmid profiling, and in vitro antimicrobial drug susceptibility testing. The results of the microbiological studies were correlated to anamnestic information. The bacterial strains were identified as V. anguillarum serovar O1 (n = 132), serovar O2 (n = 89), serovar O4 (n = 2), serovar O8 (n = 1), and not typeable (n = 1) as well as Vibrio splendidus biovar I (n = 36) and biovar II (n = 1), Vibrio tubiashii (n = 1), and Vibrio fischerii (n = 1). V. anguillarum serovar O1 or O2 was isolated in 176 out of 179 cases of clinical vibriosis in Atlantic salmon (Salmo salar). V. anguillarum serovar O1 was the only serovar isolated from salmonid fish species other than Atlantic salmon, while V. anguillarum serovar O2 was isolated from all marine fish suffering from vibriosis. A 48-Mda plasmid was isolated from all V. anguillarum serovar O1 isolates examined. Serovar O2 isolates did not harbor any plasmids. Resistance against commonly used antibiotic compounds was not demonstrated among V. anguillarum isolates. Neither V. splendidus biovar I nor other V. anguillarum-related species appeared to be of clinical importance among salmonid fish.(ABSTRACT TRUNCATED AT 250 WORDS)     

Vibrios isolated from the cultured manila clam (Ruditapes philippinarum): numerical taxonomy and antibacterial activities

AIMS: A numerical taxonomic study of halophilic Vibrio isolated from healthy and brown ring disease (BRD) affected manila clams (Ruditapes philippinarum), harvested from the Atlantic coast of south-western Spain, was performed. METHODS AND RESULTS: Characterization of 123 presumptive Vibrio spp. was carried out using 94 phenotypic tests. Simple matching and Jaccard similarity coefficients were used for numerical analysis. Cluster analysis by the unweighted pair group method with arithmetic averages yielded 15 phena defined at 0.81 similarity. Large phena corresponded to Vibrio tubiashii, V. splendidus biotype I and V. harveyi (phena 1, 5 and 9, respectively). The species V.splendidus biotype II, V. natriegens, V. mediterranei and V. alginolyticus were also represented. The inhibitory effect of diffusible extracellular products of the isolates against 27 strains of V.tapetis, the aetiological agent of BRD, was also investigated. Only five V. tubiashii isolates inhibited the growth of V. tapetis strains. The antimicrobial effect was inhibited by heating and depended on the culture medium. CONCLUSIONS: The main Vibrio species associated with manila clams were V. tubiashii, V.spendidus and V. harveyi. The antagonistic relationship established between V. tapetis and the Vibrio spp. clam microbiota may explain the failure of isolation in plating medium of V.tapetis from BRD-affected clams on the south Atlantic coast of Spain. SIGNIFICANCE AND IMPACT OF THE STUDY: Some of the strains isolated from manila clams correspond to agarolytic strains that constitute phenon 7 and they do not fit into any of the currently described Vibrio species.     

Characterization by numerical taxonomy and ribotyping of Vibrio splendidus biovar I and Vibrio scophthalmi strains associated with turbot cultures

Twelve Vibrio strains were examined phenotypically in 91 biochemical characters and genotypically by ribotyping. Ten were isolated from sea water and two from diseased turbot (Scophthalmus maximus). All isolates originated from one experimental system located in Ría de Vigo (Galicia, north-west Spain). Different type strains were used for comparative purposes. The taxonomic position was analysed with the NTSYST-pc and similarities among strains were calculated by the Simple Matching coefficient (SSM). rRNA gene restriction patterns were performed with the HindIII enzyme. The SSM coefficient separated the 12 Vibrio strains into two groups which included strains that showed a SSM coefficient quite similar to V. splendidus biovar 1 (ATCC 33125) and V. scophthalmi (CECT 4638). None of 91 phenotypical characters were specific in distinguishing both species. The ribotyping confirmed the taxonomic classification of strains. The pathogenicity of each strain was evaluated; 10 environmental strains were avirulent and two, isolated from diseased turbot, were virulent. Different biotypes and ribotypes were found among the avirulent isolates. This work showed ribotyping to be a valuable tool for identification and confirmed the necessity of extending the ribotype database within closely related Vibrio species in order to clarify the taxonomic position.     

Taxonomy of four marine bacterial strains that produce tetrodotoxin

Four strains of tetrodotoxin-producing bacteria isolated from a red alga and from pufferfish were characterized. Two of these strains are members of the genus Listonella MacDonell and Colwell. The phenotypic characteristics, guanine-plus-cytosine contents, and base sequences of the 16S rRNAs of these organisms indicated that they are members of Listonella pelagia (Vibrio pelagius) biovar II. The other two strains are members of the genus Alteromonas Baumann et al. and the genus Shewanella MacDonell and Colwell. These two strains are mutually distinct and distinct from the previously described Alteromonas and Shewanella species and therefore are placed in new species. The names Shewanella alga and Alteromonas tetraodonis are proposed for these organisms; the type strains are strains OK-1 and GFC, respectively.     

Low incidence of Vibrio vulnificus among Vibrio isolates from sea water and shellfish of the western Mediterranean coast

A specific search for Vibrio vulnificus in natural marine samples from the Spanish Mediterranean Sea was carried out by nested PCR and cultural approaches using thiosulphate-citrate-bile salts-sucrose agar (TCBS) and cellobiose-polymixin B-colistin agar (CPC), incubated at 40 degrees C, as selective media. Presumptive colonies were identified by PCR using specific primers against 23S rRNA sequences. This species was isolated from sea water and edible bivalves, mainly after preenrichment in alkaline peptone water (APW) at 40 degrees C followed by CPC agar. None of the V. vulnificus isolates identified corresponded to serovar E. Dominant Vibrio species on directly inoculated TCBS plates incubated at 25 degrees C were V. splendidus below 20 degrees C and V. harveyi and V. mediterranei above that temperature. Low percentages of several pathogenic vibrios were recorded but V. vulnificus was never recovered at this incubation temperature. The incidence of this species in the samples studied was lower than that described for other geographical areas, probably due to the high salinity values of the Mediterranean Sea.     

Genome sequence of Vibrio splendidus: an abundant planctonic marine species with a large genotypic diversity

Vibrio splendidus is a dominant Vibrio species in seawater presenting a remarkable genetic diversity; several strains have been linked to invertebrate's mortality. We report the complete genome sequence of V. splendidus LGP32, an oyster pathogen, and its comparison with partial genome sequences from related strains. As is typical for the genus, V. splendidus LGP32 contains two chromosomes (3.29 and 1.67 Mb) and most essential cellular processes are encoded by chromosome 1. Comparison with two other V. splendidus partial genome sequences (strains 12B01 and Med222) confirms the previously suggested high genotypic diversity within this species and led to the identification of numerous strain-specific regions that could frequently not be assigned to a specific mechanisms of recombination. Surprisingly, the chromosomal integron, the most variable genetic element in all other Vibrio species analysed to date, is absent from 12B01 and inactivated by a mobile element in Med222, while in LGP32 it only contains a limited number of cassettes. Finally, we found that the LGP32 integron contains a new dfrA cassette, related to those found in resistance integrons of Gram-negative clinical isolates. Those results suggest that marine Vibrio can be a source of antibiotic resistance genes.     

Numerical taxonomy of Vibrionaceae isolated from cultured amberjack (Seriola dumerili) and surrounding water

A numerical taxonomic study was performed on 148 isolates of Gram-negative, heterotrophic, facultative anaerobic bacteria isolated from amberjack (Seriola dumerili) and its surrounding culture water. The study included 30 type and reference strains belonging to genera Vibrio, Listonella, and Photobacterium. The strains were characterized by 109 morphological, biochemical, physiological, and nutritional tests. Cluster analysis of similarity matrices obtained with S(SM) and S(J) coefficients was carried out. UPGMA (unweighted pair group mathematical average) analysis defined 11 phena at S(SM) values > or = 86%. Nine phena were identified as Vibrio alginolyticus, V. fischeri, V. harveyi, V. carchariae, V. mediterranei, V. splendidus, V. furnissii, V. parahaemolyticus, and Photobacterium damselae subsp. damselae. The two latter comprised strains isolated from diseased fish.     

Reduction of adhesion properties of Ruditapes philippinarum hemocytes exposed to Vibrio tapetis

Vibrio tapetis is the causative agent of brown ring disease (BRD), which affects a species of clam, Ruditapes philippinarum. After incubation with V. tapetis, hemocytes lose filopods and become rounded, indicating cytotoxic activity of the bacterium. To rapidly quantify this cytotoxicity, a flow-cytometry test was developed based on the capacity of V. tapetis to inhibit adhesion of clam hemocytes to plastic. Several bacteria:hemocyte ratios, the cytotoxicity of other Vibrio spp. pathogenic to bivalves, and that of various V. tapetis isolates were tested. Inhibition of adherence is detectable with as few as 5 bacteria per hemocyte. The greater cytotoxic activity of V. tapetis compared to that of V. splendidus and V. pectenicida suggests a specific pathogenicity of V. tapetis to R. philippinarum hemocytes. Although all V. tapetis isolates inhibited adhesion, significant variations in cytotoxicity among isolates was demonstrated.     

Vibrio tasmaniensis sp. nov., isolated from Atlantic salmon (Salmo salar L.)

We describe the polyphasic characterization of four Vibrio isolates which formed a tight AFLP group in a former study. The group was closely related to V. cyclitrophicus, V. lentus and V. splendidus (98.2-98.9% similarity) on the basis of the 16S rDNA sequence analysis, but by DNA-DNA hybridisation experiments it had at maximum 61% DNA similarity towards V. splendidus. Thus, we propose that the isolates represent a new Vibrio species i.e. V. tasmaniensis (LMG 20012T; EMBL under the accession numbers AJ316192; mol% G+C of DNA of the type strain is 44.7). Useful phenotypical features for discrimination of V. tasmaniensis from other Vibrio species include gelatinase and beta-galactosidase activity, fatty acid composition (particularly 14:0), utilisation and fermentation of different compounds (e.g. sucrose, melibiose and D-galactose) as sole carbon source.     

Virulence and molecular typing of Vibrio harveyi strains isolated from cultured dentex,gilthead sea bream and European sea bass

Vibrio harveyi was isolated from internal organs or ulcers of diseased and apparently healthy gilthead sea bream (Sparus aurata) and European sea bass (Dicentrarchus labrax) cultured in several fish farms located on the Spanish Mediterranean coast. The prevalence of the bacterium was significantly higher in European sea bass than in gilthead sea bream, and was closely related to the season in both fish species, occurring almost exclusively on warm months (June to November). After phenotypic characterization, a selection of forty five isolates from gilthead sea bream, sea bass, and several isolates previously obtained from common dentex (Dentex dentex) of the same area, were molecularly typed by automated ribotyping and random amplified polymorphic DNA (RAPD) analysis. Cluster analysis of data established 8 RAPD types and 13 ribotypes among wild isolates, and the combination of both techniques allowed to define fourteen different groups and a clear discrimination of all outbreaks and samplings. Several strains isolated from diseased gilthead sea bream and sea bass and also from asymptomatic sea bream, were tested for virulence in both fish species by intracoelomic injection. All the isolates (11) were pathogenic for sea bass, with nine out of the eleven LD50 values ranging from 1.5 x 10(5) to 1.6 x 10(6) cfu/fish. Gilthead sea bream was unaffected by the seven tested strains, even by those more virulent for sea bass, and only one strain caused a 10% mortality at 4.2 x 10(7) cfu/fish. This is the first report on virulence of V. harveyi for sea bass.     

Carriage of potentially fish-pathogenic bacteria in Sparus aurata cultured in Mediterranean fish farms

A bacteriological survey of gilthead sea bream Sparus aurata from different fish farms and culture systems on the Spanish Mediterranean coast was conducted. Three different studies were performed. Study A included hatchery-reared larvae; Study B, periodic examination of randomly sampled growing fish; and Study C, growing fish sampled only during mortality/morbidity events. In Studies B and C, sea cages, earth ponds and indoor tanks were surveyed, and in both cases diseased (showing clinical signs) and non-diseased fish were included. In Study A, a shift from Vibrio spp. (30 d after hatching) to oxidative species (60 d after hatching) was detected, and no mortality events were registered. The percentage of fish yielding bacterial growth were similar in Studies B and C, reaching 57.4 and 61.3%, respectively. A statistically significant relationship between the bacterial carriage and the type of facility was only found in Study B, showing that fish from sea cages had a higher bacterial occurrence than fish from other facilities. A statistically significant relationship between bacterial carriage and signs of disease was found, although the pattern differed in each study. Thus, in Study B only 36.2% of fish yielding abundant bacterial growth were diseased, versus 68.0% in Study C. In total, 25.0% of the fish examined were diseased. Bacterial species composition was similar in asymptomatic and diseased fish, except for a group of V. ichthyoenteri-like isolates that occurred almost exclusively in asymptomatic fish. Dominant bacterial species were V. harveyi and V. splendidus, followed by V. ichthyoenteri-like isolates, Photobacterium damselae ssp. damselae and V. fisheri. Non-fermenters were less frequent but, among them, unidentified halophilic Cytophaga-Flavobacterium isolates and Pseudoalteromonas haloplanktis were the most abundant. An association of individual species with disease was not clear, which suggests the involvement of mixed infections.