Ca. Branchiomonas cysticola, Ca. Piscichlamydia salmonis and Salmon Gill Pox Virus transmit horizontally in Atlantic salmon held in fresh water.

Elucidation of the role of infectious agents putatively involved in gill disease is commonly hampered by the lack of culture systems for these organisms. In this study, a farmed population of Atlantic salmon pre-smolts, displaying proliferative gill disease with associated Candidatus Branchiomonas cysticola, Ca. Piscichlamydia salmonis and Atlantic salmon gill pox virus (SGPV) infections, was identified. A subpopulation of the diseased fish was used as a source of waterborne infection towards a population of naïve Atlantic salmon pre-smolts. Ca. B. cysticola infection became established in exposed naïve fish at high prevalence within the first month of exposure and the bacterial load increased over the study period. Ca. P. salmonis and SGPV infections were identified only at low prevalence in exposed fish during the trial. Although clinically healthy, at termination of the trial the exposed, naïve fish displayed histologically visible pathological changes typified by epithelial hyperplasia and subepithelial inflammation with associated bacterial inclusions, confirmed by fluorescent in situ hybridization to contain Ca. B. cysticola. The results strongly suggest that Ca. B. cysticola infections transmit directly from fish to fish and that the bacterium is directly associated with the pathological changes observed in the exposed, previously naïve fish.

Aeromonas salmonicida infection levels in pre- and post-stocked cleaner fish assessed by culture and an amended qPCR assay.

Due to increasing resistance to chemical therapeutants, the use of 'cleaner fish' (primarily wrasse, Labridae, species) has become popular in European salmon farming for biocontrol of the salmon louse, Lepeophtheirus salmonis (Krøyer). While being efficient de-licers, cleaner fish mortality levels in salmon cages are commonly high, and systemic bacterial infections constitute a major problem. Atypical furunculosis, caused by Aeromonas salmonicida A-layer types V and VI, is among the most common diagnoses reached in clinical investigations. A previously described real-time PCR (qPCR), targeting the A. salmonicida A-layer gene (vapA), was modified and validated for specific and sensitive detection of all presently recognized A-layer types of this bacterium. Before stocking and during episodes of increased mortality in salmon cages, cleaner fish (primarily wild-caught wrasse) were sampled and screened for A. salmonicida by qPCR and culture. Culture indicated that systemic bacterial infections are mainly contracted after salmon farm stocking, and qPCR revealed A. salmonicida prevalences of approximately 4% and 68% in pre- and post-stocked cleaner fish, respectively. This underpins A. salmonicida's relevance as a contributing factor to cleaner fish mortality and emphasizes the need for implementation of preventive measures (e.g. vaccination) if current levels of cleaner fish use are to be continued or expanded.

vapA (A-layer) typing differentiates Aeromonas salmonicida subspecies and identifies a number of previously undescribed subtypes.

Sequence variation in a region of the virulence array protein gene (vapA; A-layer) was assessed in 333 ('typical' and 'atypical') isolates of the fish pathogenic bacterium Aeromonas salmonicida. Resulting similarity dendrograms revealed extensive heterogeneity, with nearly all isolates belonging to either of 14 distinct clusters or A-layer types. All acknowledged A. salmonicida subspecies (except ssp. pectinolytica, from which no vapA sequence could be obtained) were clearly separated, and notably, all isolates phenotypically identified as ssp. salmonicida formed a distinct and exclusive A-layer type. Additionally, an array of un-subspeciated atypical strains formed several equally prominent clusters, demonstrating that the concept of typical/atypical A. salmonicida is inappropriate for describing the high degree of diversity evidently occurring outside ssp. salmonicida. Most representatives assessed in this study were clinical isolates of spatiotemporally diverse origins, and were derived from a variety of hosts. We observed that from several fish species or families, isolates predominantly belonged to certain A-layer types, possibly indicating a need for host-/A-layer type-specific A. salmonicida vaccines. All in all, A-layer typing shows promise as an inexpensive and rapid means of unambiguously distinguishing clinically relevant A. salmonicida subspecies, as well as presently un-subspeciated atypical strains.

Multilocus variable number tandem repeat analysis of Edwardsiella piscicida isolates pathogenic to fish.

This study describes a novel multilocus variable number tandem repeat analysis (MLVA) based on six variable number of tandem repeat (VNTR) loci for genotyping of 37 Edwardsiella piscicida (previously Edwardsiella tarda) isolates from multiple sources. The number of alleles identified for each of the six VNTR loci ranged from 3 to 5 with VNTR loci 1 (DI = 0.632) and 3 (DI = 0.644), displaying the highest degrees of polymorphism. MLVA typing of the 37 E. piscicida isolates resulted in the identification of five major clusters consistent with their geographical origins, and were designated as MLVA types I, II, III, IV and V. Types III and V were resolved further into subtypes largely consistent with outbreak source. An MLVA profile comprising a string of integers representing the number of tandem repeats for each allele provided a unique identification for each MLVA type and/or strain. The MLVA protocol described in the current study is robust, relatively simple, has a higher power of resolution than multilocus sequence analysis (MLSA) and is capable of discriminating closely related isolates.

Edwardsiella piscicida sp. nov., a novel species pathogenic to fish.

AIMS: This study describes a novel species within the genus Edwardsiella based on phenotypic and genetic characterization of fish pathogenic Edwardsiella isolates previously identified as E. tarda. METHODS AND RESULTS: Phenotypic characterization, DNA-DNA hybridization and phylogenetic analysis of representative Edwardsiella isolates from fish previously identified as E. tarda were conducted and compared with E. tarda type strain (ATCC 15947(T)). Phenotypically, strains from fish grow with pin-point colonies producing slight ß-haemolysis under the colony. In contrast to the E. tarda type strain, fish strains did not [corrected] degrade ß-methyl-D-glucoside [corrected] (with the exception of NCIMB 2034), citric acid and L-proline. [corrected]. With the exception of strain ETK01, all fish strains were highly pathogenic to zebra fish, while ATCC 15947(T) and NCIMB 2034 were nonpathogenic. DNA-DNA hybridization (DDH) levels between representative fish isolates and the E. tarda type strain ranged from 15 to 43·6%, while NCIMB 2034 hybridised with the type strain at the level of 63·2%. DDH values between the various fish isolates ranged from 68·2 to 93·9% defining a new and separate DNA hybridization group differing from the E. tarda type strain consistent with the findings of phylogenetic analysis, in which the fish isolates comprised a separate clade. CONCLUSIONS: Phenotypical and genetic characterizations demonstrated that Edwardsiella isolates from fish described in this study do not belong to the species E. tarda or any of the previously established taxa within the genus Edwardsiella. The fish related strains studied here (excluding NCIMB 2034) represent, therefore, a novel species within the genus Edwardsiella for which we propose the name Edwardsiella piscicida sp. nov, with strain ET883(T) (NCIMB 14824(T) = CCUG 62929) as the type strain. SIGNIFICANCE AND IMPACT OF THE STUDY: The current finding will improve the diagnosis, understanding of the epidemiology and in establishment of effective control measures against this serious fish pathogen.

Multi-locus Sequence Analysis (MLSA) of Edwardsiella tarda isolates from fish.

Edwardsiella tarda is an enteric fish pathogen that has caused significant economic losses in a range of fish species residing in diverse ecological conditions. Several molecular methods relying on DNA fingerprinting (RAPD, RFLP and ERIC-PCR) and the gyrB gene marker have been used to characterize E. tarda isolates. However, all had drawbacks in resolving power and reproducibility. The present study was aimed at developing a novel Multi-locus Sequence Analysis (MLSA) scheme for genetic characterization of E. tarda isolates originating from multiple sources. MLSA has been described as an effective molecular tool with superior discriminatory power and reproducibility for exploring intra-species genetic diversity of several bacterial species. Nucleotide sequence fragments of eight protein coding housekeeping genes (gyrB, mdh, adk, dnaK, phoR, metG, pyrG and aroE2) were obtained from 23 fish pathogenic E. tarda isolates of different geographical origins, one human isolate and 3 reference strains. The phylogenetic relationships between isolates in individual gene analyses were not consistent, although some common patterns were apparent. Phylogenetic analysis based on concatenated sequences of seven gene loci, however, buffered the conflicting phylogenetic signals and resolved isolates according to their geographical origin and/or fish host. The MLSA revealed two major genetically diverging clusters in E. tarda isolates examined, one cluster representing isolates from fish and the other representing (in the main) human isolates, with E. ictaluri cluster situated in between. The results suggest, therefore, that the fish pathogenic E. tarda isolates may have been previously misclassified and probably represent one or more as yet unrecognized taxa within the genus Edwardsiella. The MLSA described here was robust enough in discriminating E. tarda isolates not only with respect to their geographical origins but also within different hosts from the same geographical location, high-lighting its potential application in tracing the source of infection and understand the epidemiological relationships among isolates of environmental, fish, other domestic animals or human origins.

Experimental mycobacteriosis in Atlantic cod, Gadus morhua L.

Piscine mycobacteriosis causes losses in a number of fish species both in the wild and in aquaculture worldwide. Mycobacterium salmoniphilum infections have on several occasions been reported in farmed Atlantic salmon, Salmo salar L. The present study tested and confirmed the susceptibility of Atlantic cod, Gadus morhua L., an important yet relatively novel aquaculture species, to infection with M. salmoniphilum. Atlantic cod injected intraperitoneally with a suspension of this bacterium were maintained together with cohabitant (COH) fish in a flow-through marine water system at 10-11 °C. The fish were supervised daily and samples taken at 2, 7, 14, 23, 34 and 53 weeks post-infection and examined pathologically, bacteriologically and using molecular biology. Injected mycobacteria were re-isolated in high concentrations from both injected and COH fish groups. Death attributable to mycobacterial infection was observed in both injected (47%) and COH (28%) fish groups. Extensive development of granuloma in visceral organs, mainly the mesenteries, spleen, kidney and liver (lesser extent) and at later stages of the infection in heart tissues and gills, was observed in both injected and COH fish. Granulomas underwent a temporal progression of distinct morphological stages, culminating in well-circumscribed lesions surrounded by normal or healing tissue. Acid-fast bacilli were detected in both granulomas and non-granulomatous tissues. This study confirms that Atlantic cod is highly susceptible to M. salmoniphilum infection and that this bacterial species may be a threat to cod both in the wild and in the aquaculture.

Mycobacterium salmoniphilum infection in farmed Atlantic salmon, Salmo salar L.

Multiple greyish-white visceral nodules containing abundant rapidly growing and acid-fast bacteria, subsequently identified as Mycobacterium salmoniphilum, were detected in moribund and newly dead market-sized fish during a period of increased mortality in an Atlantic salmon, Salmo salar, farm in western Norway. Isolates cultured from diseased fish were phenotypically consistent with Mycobacterium sp. previously isolated from Atlantic salmon [MT 1890 (= NCIMB13533), MT1892, MT1900 and MT1901] in the Shetland Isles, Scotland. Partial sequences of 16S rDNA, ribosomal RNA internal transcribed spacer (ITS1), 65-kDa heat-shock protein (Hsp65) and ß subunit of RNA polymerase (rpoB) revealed 97-99% similarity with M. salmoniphilum type strain ATCC 13758(T) . The source of infection was not confirmed. Koch's postulates were fulfilled following experimental challenge of Atlantic salmon with field isolate NVI6598 (FJ616988). Mortality was recorded in experimentally infected fish; however, the infection remained subclinical in the majority of affected fish over the 131-day challenge period.

Tenacibaculum sp. associated with winter ulcers in sea-reared Atlantic salmon Salmo salar.

Coldwater-associated ulcers, i.e. winter ulcers, in seawater-reared Atlantic salmon Salmo salar L. have been reported in Norway since the late 1980s, and Moritella viscosa has been established as an important factor in the pathogenesis of this condition. As routine histopathological examination of winter ulcer cases in our laboratory revealed frequent presence in ulcers of long, slender rods clearly different from M. viscosa, a closer study focusing on these bacteria was conducted. Field cases of winter ulcers during 2 sampling periods, 1996 and 2004-2005, were investigated and long, slender rods were observed by histopathological examination in 70 and 62.5% of the ulcers examined, respectively, whereas cultivation on marine agar resulted in the isolation of yellow-pigmented colonies with long rods from 3 and 13% of the ulcers only. The isolates could be separated into 2 groups, both identified as belonging to the genus Tenacibaculum based on phenotypic characterization and 16S rRNA sequencing. Bath challenge for 7 h confirmed the ability of Group 1 bacterium to produce skin and cornea ulcers. In fish already suffering from M. viscosa-induced ulcers, co-infection with the Group 1 bacterium was established within 1 h. Ulcers from field cases of winter ulcers and from the transmission experiments tested positive by immunohistochemistry with polyclonal antiserum against the Group 1 bacterium but not the Group 2 bacterium. Our results strongly indicate the importance of the Group 1 bacterium in the pathogenesis of winter ulcers in Norway. The bacterium is difficult to isolate and is therefore likely to be underdiagnosed based on cultivation only.

Immunohistochemical and Taqman real-time PCR detection of mycobacterial infections in fish.

Real-time PCR and immunohistochemistry (IHC) assays were developed to detect fish mycobacterial infections at the genus level, based on the RNA polymerase ß subunit (rpoB) gene and polyclonal anti-Mycobacterium rabbit serum, respectively. The PCR assay positively identified a number of pathogenic mycobacteria including Mycobacterium abscessus, M. avium ssp. avium, M. bohemicum, M. chelonae ssp. chelonae, M. farcinogenes, M. flavescens, M. fortuitum ssp. fortuitum, M. gastri, M. gordonae, M. immunogenicum, M. malmoense, M. marinum, M. montefiorense, M. phlei, M. phocaicum, M. pseudoshottsii, M. salmoniphilum, M. senegalense, M. shottsii, M. smegmatis, M. szulgi and M. wolinskyi. A detection limit equivalent to 10(2) cfu g(-1) was registered for M. salmoniphilum-infected fish tissue. The IHC precisely localized both free and intracellular mycobacteria in tissues and detected mycobacterial infections down to 10(2) cfu g(-1) tissue. Both assays were found to be more sensitive than Ziehl-Neelsen (ZN) staining, where the detection limit was below 8 × 10(3) cfu g(-1) tissue. Although specificity testing of the real-time PCR against a panel of non-Mycobacterium spp. revealed a degree of cross-reaction against pure DNA extracted from Nocardia seriolae and Rhodococcus erythropolis, no cross-reactions were identified (by either real-time PCR or IHC) on testing of formalin-fixed paraffin-embedded (FFPE) tissues confirmed to be infected with these bacteria. The broad applicability of both assays was confirmed by analysis of FFPE tissues from a range of fish species infected with diverse Mycobacterium spp. The results indicate that both assays, alone or in combination, constitute sensitive tools for initial, rapid diagnosis of mycobacteriosis in fish. This should in turn allow rapid application of more specific studies, i.e. culture based, to identify the specific Mycobacterium sp. involved.

Microbial and pathological findings in farmed Atlantic salmon Salmo salar with proliferative gill inflammation.

Proliferative gill inflammation (PGI) is an important cause of loss in seawater-farmed Atlantic salmon in Norway. Several microbes have been associated with PGI, including the commonly but not exclusively observed inclusions (epitheliocysts) within the gill lamellae related to infection with 'Candidatus Piscichlamydia salmonis'. Atlantic salmon transferred in the spring of 2004 to 12 seawater farms situated in mid- and southwest Norway were sampled throughout that year. Outbreaks of PGI, as evaluated by clinical examination, histology, and mortality data, were diagnosed in 6 of 7 farms in southwest Norway but not in the 5 farms studied in mid-Norway. Generally, mortality started 3 to 5 mo after seawater transfer and outbreaks lasted at least 1 to 3 mo. 'Ca. P. salmonis' was detected by real-time PCR only in fish from PGI-affected farms and our results indicate an association between 'Ca. P. salmonis' load and PGI severity. Likewise, although widely distributed in all 12 farms studied, epitheliocyst prevalence and number per fish as observed by histology appears associated with PGI prevalence and severity. However, the occurrence of epitheliocysts showed no association with molecular detection of 'Ca. P. salmonis', suggesting that at least 1 other organism is responsible for many of the observed inclusions. A microsporidian, Desmozoon lepeophtherii, was identified at high prevalence regardless of fish and farm PGI status, but at higher loads in fish with PGI. Our results support a multifactorial etiology for PGI in which 'Ca. P. salmonis', an unidentified epitheliocyst agent, and the microsporidian are contributing causes. No evidence for the involvement of Atlantic salmon paramyxovirus in PGI development was identified in the present study. High water temperatures and ectoparasites probably exacerbated mortality.

Epitheliocystis in Atlantic salmon, Salmo salar L., farmed in fresh water in Ireland is associated with 'Candidatus Clavochlamydia salmonicola' infection.

Intracellular inclusions containing chlamydia-like organisms are frequently observed in the gill epithelial cells of Atlantic salmon, Salmo salar L., cultured in fresh water in Ireland. In this study, the causative agent was identified in four separate freshwater sites, using 16s rRNA sequencing, as 'Candidatus Clavochlamydia salmonicola'. Histopathology and real-time (RT) PCR were used to further assess infections. The prevalence of infection ranged from 75-100% between sites and infection intensity was highly variable. No significant lesions were associated with these infections. As a diagnostic tool, RT-PCR proved marginally more sensitive than histopathology. The fate of 'Candidatus Clavochlamydia salmonicola' in Atlantic salmon post-seawater transfer was investigated in a 12-week marine longitudinal study. Both RT-PCR and histopathological examination indicate that the organism disappears from the gills 4-6 weeks post-transfer.

Previously unrecognised division within Moritella viscosa isolated from fish farmed in the North Atlantic.

Previously undocumented phenotypical and genetic variation was identified amongst isolates of Moritella viscosa collected from various geographical locations and from different fish species. The studied isolates could be split into 2 major phenotypically and genetically different clusters, one of which was consistent with the species type strain (NCIMB 13548). Isolates consistent with the type strain originated exclusively from Atlantic salmon farmed in Norway, Scotland and the Faroe Isles, although a single isolate from farmed Norwegian cod clustered closely with this group. The 'variant' cluster comprised isolates originating from Norwegian farmed rainbow trout, Icelandic farmed rainbow trout and salmon, Canadian farmed (Atlantic) salmon, Icelandic lumpsucker and only exceptionally from Norwegian salmon. With the exception of the single aforementioned cod isolate, all isolates from Norwegian farmed cod belonged to the variant cluster. Phenotypically, the clusters could be absolutely separated only by elevated haemolytic activity in the variant strain, although approximately half of these isolates also produced acid from mannose, in contrast to the typical (type) strain. While 16S rRNA gene sequencing was unable to separate the 2 clusters, Western blot analyses, plasmid profile analysis, pulsed field gel electrophoresis and gyrB gene sequence analysis produced clusters consistent with the phenotypic data. Macroscopically and histologically the disease in rainbow trout caused by the variant strain was consistent with that previously described in Atlantic salmon. The results of the present study may indicate a degree of host specificity of the typical strain for Atlantic salmon.

Real time PCR detection of Piscirickettsia salmonis from formalin-fixed paraffin-embedded tissues.

Piscirickettsia salmonis is the causative agent of piscirickettsiosis, a transmissible disease of salmonid fish. Diagnosis of piscirickettsiosis has traditionally been based upon identification of typical pathological changes by histological investigation, with confirmation by immunohistochemistry on formalin-fixed, paraffin-embedded tissues. However, implementation of more rapid confirmatory techniques, preferably with higher levels of sensitivity and possibilities for quantification, is desirable. A real-time polymerase chain reaction (PCR) assay was designed for specific detection of P. salmonis and tested on samples extracted from formalin-fixed paraffin-embedded material. Construction of a PCR-target mimic allowed determination of detection limits, linearity of the real-time PCR and quantitative detection of P. salmonis. The present study demonstrates the capability of the described real time PCR assay for detection of P. salmonis from paraffin-embedded material with a high degree of sensitivity and specificity. Implementation of this assay constitutes an important development for a rapid and secure diagnosis of piscirickettsiosis.