Primary immunization using low antigen dosages and immunological tolerance in rainbow trout.

Enteric redmouth disease (ERM), caused by the Gram negative enterobacterium Yersinia ruckeri, affects farming of salmonids, but vaccination against ERM confers a certain degree of protection dependent on the administration route. Recent studies on oral vaccination of rainbow trout suggest that immunological tolerance may be induced by primary immunization using a low antigen dosage. We have examined if low dosages of Y. ruckeri antigens, applied in feed or bath exposure over a prolonged period of time, leave rainbow trout more susceptible to infection. Groups of rainbow trout were immunized, either by immersion or feeding using different vaccine dosages, and subsequently challenged by live Y. ruckeri. Survival was recorded and immune reactions in surviving fish were evaluated (ELISA and qPCR). Trout, bath-vaccinated in a highly diluted vaccine or fed the same amount of bacterin in feed over 10 days, were not protected against Y. ruckeri challenge infection and in some cases these sub-optimally immunized fish experienced lower survival compared to non-primed controls. Genes encoding FoxP3 and immune-suppressive cytokines were down-regulated in fish vaccinated with a high antigen dosage when compared to groups exposed to low antigen dosages, suggesting a higher regulatory T cell activity in the latter fish groups. The study suggests that repeated exposure to low antigen concentrations induces some degree of immune tolerance in rainbow trout and we recommend application of high antigen dosages for primary immunization of trout.

Secondary immune response of rainbow trout following repeated immersion vaccination.

Teleosts are able to raise a protective immune response, comprising both innate and adaptive elements, against various pathogens. This is the basis for a widespread use of vaccines, administered as injection or immersion, in the aquaculture industry. It has been described that repeated injection vaccination of fish raises a secondary immune response, consisting of rapid, accelerated and increased antibody reaction. This study reports how rainbow trout responds to repeated immersion vaccination against yersiniosis (ERM) caused by the bacterial pathogen Yersinia ruckeri. It was found that rainbow trout does not raise a classical secondary response following repeated immersion vaccination. Serum antibody titres were merely slightly increased even after three immunizations, using 30-s immersion into a bacterin consisting of formalin-inactivated Y. ruckeri (serotype O1, biotypes 1 and 2), performed over a 3-month period. The densities of IgM-positive lymphocytes in spleen of fish immunized three times were increased compared to control fish, but no general trend for an increase with the number of immunizations was noted. The lack of a classical secondary response following repeated immersion vaccination may partly be explained by limited uptake of antigen by immersion compared to injection.

In vivo adherence of Flavobacterium psychrophilum to mucosal external surfaces of rainbow trout (Oncorhynchus mykiss) fry.

The adherence of Flavobacterium psychrophilum to surfaces of epithelial tissues has been inconclusively suggested as a mechanism, which enables the bacterium to invade the host. Hence, the present study aimed to examine the adherence of the cells of two colony phenotypes, smooth and rough, of F. psychrophilum to mucosal tissues of rainbow trout fry and to test the skin mucus as a nutrient for the growth of F. psychrophilum. Fish were immersed in water containing 106  CFU  mL-1 F. psychrophilum for each colony phenotype. Mucosal tissue samples from fins, gills, skin and eyes, and swab samples from spleen and kidney were taken and inoculated onto TYES agar plates. Colony phenotypes of F. psychrophilum were identified and number of colonies counted. The results showed that cells of both phenotypes initially (0 h) adhered to all mucosal surfaces, but only the rough cells were still present on tissues 1 h post-immersion. Both phenotypes showed a tissue tropism with the fin tissue being the most adhered. Furthermore, skin mucus promoted the growth of both colony phenotypes. We suggest that the growth of F. psychrophilum cells in skin mucus apparently facilitates the bacterial adherence to mucosal surfaces, and the subsequent invasion into the host.

Antimicrobial peptide CAP18 and its effect on Yersinia ruckeri infections in rainbow trout Oncorhynchus mykiss (Walbaum): comparing administration by injection and oral routes.

The antimicrobial peptide CAP18 has been demonstrated to have a strong in vitro bactericidal effect on Yersinia ruckeri, but its activity in vivo has not been described. In this work, we investigated whether CAP18 protects rainbow trout Oncorhynchus mykiss (Walbaum) against enteric red mouth disease caused by this pathogen either following i.p. injection or by oral administration (in feed). It was found that injection of CAP18 into juvenile rainbow trout before exposure to Y. ruckeri was associated with lowered mortality compared to non-medicated fish although it was less effective than the conventional antibiotic oxolinic acid. Oral administration of CAP18 to trout did not prevent infection. The proteolytic effect of secretions on the peptide CAP18 in the fish gastrointestinal tract is suggested to account for the inferior effect of oral administration.

Infection routes of Aeromonas salmonicida in rainbow trout monitored in vivo by real-time bioluminescence imaging.

Recent development of imaging tools has facilitated studies of pathogen infections in vivo in real time. This trend can be exemplified by advances in bioluminescence imaging (BLI), an approach that helps to visualize dissemination of pathogens within the same animal over several time points. Here, we employ bacterial BLI for examining routes of entry and spread of Aeromonas salmonicida susbp. salmonicida in rainbow trout. A virulent Danish A. salmonicida strain was tagged with pAKgfplux1, a dual-labelled plasmid vector containing the mutated gfpmut3a gene from Aequorea victoria and the luxCDABE genes from the bacterium Photorhabdus luminescens. The resulting A. salmonicida transformant exhibited growth properties and virulence identical to the wild-type A. salmonicida, which made it suitable for an experimental infection, mimicking natural conditions. Fish were infected with pAKgfplux1 tagged A. salmonicida via immersion bath. Colonization and subsequent tissue dissemination was followed over a 24-h period using the IVIS spectrum imaging workstation. Results suggest the pathogen's colonization sites are the dorsal and pectoral fin and the gills, followed by a progression through the internal organs and an ensuing exit via the anal opening. This study provides a tool for visualizing colonization of A. salmonicida and other bacterial pathogens in fish.

Detection and quantification of Aeromonas salmonicida in fish tissue by real-time PCR.

Furunculosis, a septicaemic infection caused by the bacterium Aeromonas salmonicida subsp. salmonicida, currently causes problems in Danish seawater rainbow trout production. Detection has mainly been achieved by bacterial culture, but more rapid and sensitive methods are needed. A previously developed real-time PCR assay targeting the plasmid encoded aopP gene of A. salmonicida was, in parallel with culturing, used for the examination of five organs of 40 fish from Danish freshwater and seawater farms. Real-time PCR showed overall a higher frequency of positives than culturing (65% of positive fish by real-time PCR compared to 30% by a culture approach). Also, no real-time PCR-negative samples were found positive by culturing. A. salmonicida was detected by real-time PCR, though not by culturing, in freshwater fish showing no signs of furunculosis, indicating possible presence of carrier fish. In seawater fish examined after an outbreak and antibiotics treatment, real-time PCR showed the presence of the bacterium in all examined organs (1-482 genomic units mg-1 ). With a limit of detection of 40 target copies (1-2 genomic units) per reaction, a high reproducibility and an excellent efficiency, the present real-time PCR assay provides a sensitive tool for the detection of A. salmonicida.

Positive correlation between Aeromonas salmonicida vaccine antigen concentration and protection in vaccinated rainbow trout Oncorhynchus mykiss evaluated by a tail fin infection model.

Rainbow trout, Oncorhynchus mykiss (Walbaum), are able to raise a protective immune response against Aeromonas salmonicida subsp. salmonicida (AS) following injection vaccination with commercial vaccines containing formalin-killed bacteria, but the protection is often suboptimal under Danish mariculture conditions. We elucidated whether protection can be improved by increasing the concentration of antigen (formalin-killed bacteria) in the vaccine. Rainbow trout juveniles were vaccinated by intraperitoneal (i.p.) injection with a bacterin of Aeromonas salmonicida subsp. salmonicida strain 090710-1/23 in combination with Vibrio anguillarum serotypes O1 and O2a supplemented with an oil adjuvant. Three concentrations of AS antigens were applied. Fish were subsequently challenged with the homologous bacterial strain administered by perforation of the tail fin epidermis and 60-s contact with live A. salmonicida bacteria. The infection method proved to be efficient and could differentiate efficacies of different vaccines. It was shown that protection and antibody production in exposed fish were positively correlated to the AS antigen concentration in the vaccine.

Epidemiological cut-off values for Flavobacterium psychrophilum MIC data generated by a standard test protocol.

Epidemiological cut-off values were developed for application to antibiotic susceptibility data for Flavobacterium psychrophilum generated by standard CLSI test protocols. The MIC values for ten antibiotic agents against Flavobacterium psychrophilum were determined in two laboratories. For five antibiotics, the data sets were of sufficient quality and quantity to allow the setting of valid epidemiological cut-off values. For these agents, the cut-off values, calculated by the application of the statistically based normalized resistance interpretation method, were ≤16 mg L(-1) for erythromycin, ≤2 mg L(-1) for florfenicol, ≤0.025 mg L(-1) for oxolinic acid (OXO), ≤0.125 mg L(-1) for oxytetracycline and ≤20 (1/19) mg L(-1) for trimethoprim/sulphamethoxazole. For ampicillin and amoxicillin, the majority of putative wild-type observations were 'off scale', and therefore, statistically valid cut-off values could not be calculated. For ormetoprim/sulphadimethoxine, the data were excessively diverse and a valid cut-off could not be determined. For flumequine, the putative wild-type data were extremely skewed, and for enrofloxacin, there was inadequate separation in the MIC values for putative wild-type and non-wild-type strains. It is argued that the adoption of OXO as a class representative for the quinolone group would be a valid method of determining susceptibilities to these agents.

Effect of hydrogen peroxide and/or Flavobacterium psychrophilum on the gills of rainbow trout, Oncorhynchus mykiss (Walbaum).

The immune response and morphological changes in the gills of rainbow trout fry after immersion in hydrogen peroxide (H2O2), Flavobacterium psychrophilum or combined exposure were examined. The gills were sampled 4, 48, 125 and 192 h after exposure, and the regulation of expression of the following genes was investigated using qPCR: IgT, IgM, CD8, CD4, MHC I, MHC II, IL-4/13A, TcR-ß, IL-10, IL-1ß, IL-17, SAA and FoxP3. Bacteria were not observed in haematoxylin-and-eosin-stained gill tissue, but the presence of F. psychrophilum 16S rRNA was detected using qPCR. The 16S rRNA levels were correlated with gene expression. Although pretreatment with H2O2 before immersion in F. psychrophilum did not significantly alter the amount of bacteria found in the gill, the immune response was influenced: exposure to F. psychrophilum resulted in a negative correlation with expression of IL-17c1, MHC I and MHC II, while pretreatment with H2O2 resulted in a positive correlation with IL-4/13A and IgM. Exposure to either H2O2 or F. psychrophilum influenced the regulation of gene expression and damaged tissue. Exposure to both combined altered the immune response to infection and postponed healing of gill tissue.

Evaluation of the immune response in rainbow trout fry, Oncorhynchus mykiss (Walbaum), after waterborne exposure to Flavobacterium psychrophilum and/or hydrogen peroxide.

The immune response in rainbow trout fry against Flavobacterium psychrophilum was elucidated using an immersion-based challenge with or without prior exposure to hydrogen peroxide (H2O2). Samples were taken from the head kidney 4, 48, 125 and 192 h after immersion, and the regulation of several genes was examined. Bacterial load was assessed based on the presence of 16S rRNA and correlated with gene expression, and the levels of specific antibodies in the blood were measured 50 days post-infection. Separately, both H2O2 and F. psychrophilum influenced gene expression, and pre-treatment with H2O2 influenced the response to infection with F. psychrophilum. Pre-treatment with H2O2 also affected correlation between gene regulation and pathogen load for several genes. A delay in antibody production in H2O2-treated fish in the early phase of infection was indicated, but H2O2 exposure did not affect antibody levels 50 days post-infection. An increasing amount of F. psychrophilum 16S rRNA was found in the head kidneys of infected fish pre-treated with H2O2 relative to the F. psychrophilum group. The results show that a single pre-treatment with H2O2 impairs the response against F. psychrophilum and may intensify infection.

Diversity of Flavobacterium psychrophilum and the potential use of its phages for protection against bacterial cold water disease in salmonids.

Flavobacterium psychrophilum causes rainbow trout fry syndrome (RTFS) and cold water disease (CWD) in salmonid aquaculture. We report characterization of F. psychrophilum strains and their bacteriophages isolated in Chilean salmonid aquaculture. Results suggest that under laboratory conditions phages can decrease mortality of salmonids from infection by their F. psychrophilum host strain. Twelve F. psychrophilum isolates were characterized, with DNA restriction patterns showing low diversity between strains despite their being obtained from different salmonid production sites and from different tissues. We isolated 15 bacteriophages able to infect some of the F. psychrophilum isolates and characterized six of them in detail. DNA genome sizes were close to 50 Kbp and corresponded to the Siphoviridae and Podoviridae families. One isolate, 6H, probably contains lipids as an essential virion component, based on its chloroform sensitivity and low buoyant density in CsCl. Each phage isolate rarely infected F. psychrophilum strains other than the strain used for its enrichment and isolation. Some bacteriophages could decrease mortality from intraperitoneal injection of its host strain when added together with the bacteria in a ratio of 10 plaque-forming units per colony-forming unit. While we recognize the artificial laboratory conditions used for these protection assays, this work is the first to demonstrate that phages might be able protect salmonids from RTFS or CWD.

Comparative protection of two different commercial vaccines against Yersinia ruckeri serotype O1 and biotype 2 in rainbow trout (Oncorhynchus mykiss).

Differentially extended specific protection by two commercial vaccines against Yersinia ruckeri serotype O1 biotype 2 was studied following 30s immersion exposure. Rainbow trout were challenged intra-peritoneally (i.p.) with Y. ruckeri serotype O1, biotype 2 (≈10(6) to 10(7)CFU/fish) at 4, 6 and 8 months after vaccination with vaccines containing either biotype 1 (AquaVac(®) ERM) or both biotypes 1 and 2 (AquaVac(®) RELERA™). The specific pattern of vaccine-mediated protection was evaluated by relative percentage survival (RPS) analysis at 4 and 6 months post-vaccination and by obtaining gross pathological observations at 4 and 8 months respectively. We determined specific significant and superior protection in terms of increased survivability in AquaVac(®) RELERA™ vaccinated fish and observed correspondingly fewer pathological changes. The challenge trials indicated a longer protection for at least 6 months without any booster vaccination. A specific and adaptive response induced by AquaVac(®) RELERA™ vaccine against Y. ruckeri biotype 2 was clearly indicated. In addition, some degree of cross protection rendered by AquaVac(®) ERM containing biotype 1 during infection with Y. ruckeri biotype 2 was also noted.

Chemical surface disinfection of eggs of Baltic cod, Gadus morhua L.

The effect of two disinfectants on eggs and larvae of Baltic cod, Gadus morhua, was investigated. The eggs were disinfected for 10 min using various concentrations of either glutaraldehyde (100, 200, 400, 600 and 800 mg L(-1)) or iodophor (10, 50, 100 and 150 mg L(-1)), 1-4-days post-fertilization. Bactericidal effect of disinfection, survival to hatching, hatching success and larval abnormalities were assessed. Larval survival was recorded at 5-, 10- and 15-days post-hatch (dph). Although Baltic cod eggs have an unusually thin chorion, they could tolerate surface disinfection. A reduction in bacterial growth was observed with increased concentrations of disinfectant (3.0 x 10(7)-1.6 x 10(1) CFU mL(-1)). Abnormalities in newly hatched larvae were not related to disinfection. Survival of the yolk sac larvae was significantly better for eggs treated with 400 mg L(-1) glutaraldehyde for 10 min at 10 and 15 dph. Effective disinfection was also recorded using 100 mg L(-1) Actomar K30. Egg batch effect rather than initial bacterial concentration, disinfectant type or incubation method determined the survival of the eggs to hatching and survival of larvae. Because of the carcinogenic effect of glutaraldehyde, iodophor is recommended for routine disinfection of cod eggs.

Water recirculation and good management: potential methods to avoid disease outbreaks with Flavobacterium psychrophilum.

Flavobacterium psychrophilum infections cause high mortality among rainbow trout, Oncorhynchus mykiss, fry in Danish fish farms and hatcheries. Hatcheries based entirely on bore-hole water recirculation systems have been suggested as a possibility for eliminating F. psychrophilum or at least keeping the amount of this bacterium low. The occurrence of the bacterium in a bore-hole water recirculation system was compared with a combined bore-hole water and stream water flow-through system in a hatchery where outbreaks of rainbow trout fry syndrome caused by F. psychrophilum often occurred. Broodfish, unfertilized and fertilized eggs, eyed eggs and fry, as well as water samples from the tanks/troughs with broodfish/fry, were examined. Suspect yellow bacterial colonies were either confirmed or rejected as F. psychrophilum by growth characteristics and by PCR. As both virulent and less virulent F. psychrophilum isolates are known, isolates were characterized. The isolates were ribotyped and grouped according to ribotyping patterns. Representatives of the groups were serotyped. Fry isolates were very homogeneous whereas isolates from broodfish were heterogeneous, whether the isolates originated from external surfaces of the fish (mucus from skin and gills, haemorrhages and ulcers) or internal organs. Flavobacterium psychrophilum was isolated from broodfish in both water systems; 56% of investigated broodfish from the borehole/flowthrough system and 36% from the recirculation facility harboured the bacterium. In the recirculation system, the bacterium was isolated from fish (ulcers, milt, liver, abdominal cavity) kept in the system for 11 months. Flavobacterium psychrophilum was found in milt and ovarian fluid as well as on the surface of fertilized eggs, but not inside the eggs. Fry also harboured F. psychrophilum, but in the water recirculation system the bacterium was first isolated from the fry after they had been graded. Flavobacterium psychrophilum was found regularly in other parts of the hatchery (outside the recirculation facility), including at the time of grading, suggesting that the occurrence of F. psychrophilum in the fry recirculation facility was due to contamination from the borehole/flow-through hatchery. It is suggested that the combination of bore-hole water recirculation systems and good management procedures (including egg disinfection) is a possible method for hatcheries to avoid disease outbreaks due to F. psychrophilum.

Profiling acylated homoserine lactones in Yersinia ruckeri and influence of exogenous acyl homoserine lactones and known quorum-sensing inhibitors on protease production.

AIMS: To profile the quorum-sensing (QS) signals in Yersinia ruckeri and to examine the possible regulatory link between QS signals and a typical QS-regulated virulence phenotype, a protease. METHODS AND RESULTS: Liquid chromatography-high resolution mass spectrometry (HPLC-HRMS) showed that Y. ruckeri produced at least eight different acylated homoserine lactones (AHLs) with N-(3-oxooctanoyl)-L-homoserine lactone (3-oxo-C8-HSL) being the dominant molecule. Also, some uncommon AHL, N-(3-oxoheptanoyl)-L-homoserine lactone (3-oxo-C7-HSL) and N-(3-oxononanoyl)-L-homoserine lactone (3-oxo-C9-HSL), were produced. 3-oxo-C8-HSL was detected in organs from fish infected with Y. ruckeri. Protease production was significantly lower at temperatures above 23 degrees C than below although growth was faster at the higher temperatures. Neither addition of sterile filtered high-density Y. ruckeri culture supernatant nor the addition of pure exogenous AHLs induced protease production. Furthermore, three QS inhibitors (QSIs), sulfur-containing AHL analogues, did not inhibit protease production in Y. ruckeri. CONCLUSIONS: Exogenous AHL or sulfur-containing AHL analogues did not influence the protease production indicating that protease production may not be QS regulated in Y. ruckeri. SIGNIFICANCE AND IMPACT OF THE STUDY: The array of different AHLs produced indicates that the QS system of Y. ruckeri is complex and could involve several regulatory systems. In this case, neither AHLs nor QSI would be likely to directly affect a QS-regulated phenotype.

Iron acquisition mechanisms of Flavobacterium psychrophilum.

Forty strains of Flavobacterium psychrophilum were tested for the production of siderophores using the universal Chrome Azurol S (CAS) assay. The majority of the strains (85%) were CAS positive (CAS+) and some (15%) were CAS negative (CAS-). The cryptic plasmid pCP1 was carried by all positive strains and was lacking from negative strains. While a weak catechol reaction was detectable in CAS+ culture supernatants, the CAS reaction was, to some extent, heat sensitive, questioning whether the positive reaction was caused only by siderophores. The ability to grow in vitro under iron-restricted conditions did not correlate with the CAS reactivity, as growth of both CAS+ and CAS- strains was similarly impaired under iron restriction induced by 2,2 dipyridyl. Suppressed growth under these conditions was restored by addition of FeCl3, haemoglobin and transferrin for both CAS+ and CAS- strains.

Standardization of a broth microdilution susceptibility testing method to determine minimum inhibitory concentrations of aquatic bacteria.

A multiple laboratory study was conducted in accordance with the standards established by the Clinical and Laboratory Standards Institute (CLSI), formerly the National Committee for Clinical Laboratory Standards (NCCLS), for the development of quality control (QC) ranges using dilution antimicrobial susceptibility testing methods for bacterial isolates from aquatic animal species. QC ranges were established for Escherichia coli ATCC 25922 and Aeromonas salmonicida subsp. salmonicida ATCC 33658 when testing at 22, 28 and 35 degrees C (E. coli only) for 10 different antimicrobial agents (ampicillin, enrofloxacin, erythromycin, florfenicol, flumequine, gentamicin, ormetoprim/sulfadimethoxine, oxolinic acid, oxytetracycline and trimethoprim/sulfamethoxazole). Minimum inhibitory concentration (MIC) QC ranges were determined using dry- and frozen-form 96-well plates and cation-adjusted Mueller-Hinton broth. These QC ranges were accepted by the CLSI/NCCLS Subcommittee on Veterinary Antimicrobial Susceptibility Testing in January 2004. This broth microdilution testing method represents the first standardized method for determining MICs of bacterial isolates whose preferred growth temperatures are below 35 degrees C. Methods and QC ranges defined in this study will enable aquatic animal disease researchers to reliably compare quantitative susceptibility testing data between laboratories, and will be used to ensure both precision and inter-laboratory harmonization.

Flavobacterium psychrophilum in rainbow trout, Oncorhynchus mykiss (Walbaum), hatcheries: studies on broodstock, eggs, fry and environment.

The occurrence of Flavobacterium psychrophilum at four rainbow trout hatcheries was investigated to provide more knowledge about the reservoirs and transmission of this bacterium. Broodstock were sampled at stripping (including both unfertilized and fertilized eggs), and the offspring were then sampled at the eyed egg and fry stages. Water and surface samples (e.g. hatchery trays) were also sampled. Flavobacterium psychrophilum was found in ovarian fluid and milt, indicating that broodstock may serve as a reservoir and are latent carriers of the pathogen. Flavobacterium psychrophilum was not found on or inside eggs, but further egg studies will be necessary to elucidate the possibility of vertical transmission of the pathogen. Flavobacterium psychrophilum was isolated from water samples, but only from water that had been in close contact with farmed rainbow trout or eggs. Flavobacterium psychrophilum isolates were characterized and compared with well-characterized strains, using degradation of elastin, serotype and ribotype profiles. Different ribotypes of F. psychrophilum were found between hatcheries, but a common ribotype A was found at all four hatcheries. Different ribotypes were found in broodstock without clinical disease, whereas only a few ribotypes (mostly ribotype A) were found in diseased fry. The same ribotype A was found in broodstock, in water samples from hatchery trays and in fry, which suggests the possibility of transmission of F. psychrophilum between broodstock and offspring.

Concomitant exposure of rainbow trout fry to Gyrodactylus derjavini and Flavobacterium psychrophilum: effects on infection and mortality of host.

Although rainbow trout fry mortality syndrome caused by the bacterium Flavobacterium psychrophilum is widespread in fish farms it is difficult to reproduce infection of rainbow trout in the laboratory using immersion exposure with bacterial suspensions. It has therefore been speculated that ectoparasites could act as enhancers of bacterial infections under natural conditions. In the present study rainbow trout fry were exposed to infections with F. psychrophilum (immersion for 30 min or 10 h) alone, exposed to the ectoparasitic monogenean Gyrodactylus derjavini alone or exposed to both pathogens in combination. Infection levels and host mortality were subsequently monitored to elucidate if the ectoparasitic monogeneans could enhance infection of fish with the bacterium. Immersion of fish in bacterial suspensions alone did not result in infection. Only one fish became infected with the bacterium and this fish belonged to the combination exposure group. The parasite populations increased differently in the various groups and it was found that host mortality was correlated to gyrodactylid infection levels (r=0.94) but not to bacterial exposure. The results emphasise the pathogenicity of the parasite G. derjavini, the relative resistance of intact fish to direct exposure to F. psychrophilum but provide only a weak indication of a possible enhancement of bacterial invasion due to ectoparasitic infections. It cannot be excluded that higher parasite burdens and/or prolonged immersion (more than 10 h) in bacterial suspensions may result in bacterial invasion.

Incidence, distribution, and spread of tetracycline resistance determinants and integron-associated antibiotic resistance genes among motile aeromonads from a fish farming environment.

A collection of 313 motile aeromonads isolated at Danish rainbow trout farms was analyzed to identify some of the genes involved in high levels of antimicrobial resistance found in a previous field trial (A. S. Schmidt, M. S. Bruun, I. Dalsgaard, K. Pedersen, and J. L. Larsen, Appl. Environ. Microbiol. 66:4908-4915, 2000), the predominant resistance phenotype (37%) being a combined oxytetracycline (OTC) and sulphadiazine/trimethoprim resistance. Combined sulphonamide/trimethoprim resistance (135 isolates) appeared closely related to the presence of a class 1 integron (141 strains). Among the isolates containing integrons, four different combinations of integrated resistance gene cassettes occurred, in all cases including a dihydrofolate reductase gene and a downstream aminoglycoside resistance insert (87 isolates) and occasionally an additional chloramphenicol resistance gene cassette (31 isolates). In addition, 23 isolates had "empty" integrons without inserted gene cassettes. As far as OTC resistance was concerned, only 66 (30%) out of 216 resistant aeromonads could be assigned to resistance determinant class A (19 isolates), D (n = 6), or E (n = 39); three isolates contained two tetracycline resistance determinants (AD, AE, and DE). Forty OTC-resistant isolates containing large plasmids were selected as donors in a conjugation assay, 27 of which also contained a class 1 integron. Out of 17 successful R-plasmid transfers to Escherichia coli recipients, the respective integrons were cotransferred along with the tetracycline resistance determinants in 15 matings. Transconjugants were predominantly tetA positive (10 of 17) and contained class 1 integrons with two or more inserted antibiotic resistance genes. While there appeared to be a positive correlation between conjugative R-plasmids and tetA among the OTC-resistant aeromonads, tetE and the unclassified OTC resistance genes as well as class 1 integrons were equally distributed among isolates with and without plasmids. These findings indicate the implication of other mechanisms of gene transfer besides plasmid transfer in the dissemination of antibiotic resistance among environmental motile aeromonads.