© 2005 International Council for the Exploration of the Sea
Bacteria in the gut of juvenile cod Gadus morhua fed live feed enriched with four different commercial diets
a Institute of Marine Research PO Box 1870, Nordnes, N-5817 Bergen, Norway
b Bodø Regional University N-8049 Bodø, Norway
c Nordland Research Institute N-8049 Bodø, Norway
*Correspondence to Ø. Bergh: tel: +47 5523 6370; fax: +47 5523 8586. e-mail: oivind.bergh{at}imr.no.
Atlantic cod, Gadus morhua L., larvae were fed rotifers, Brachionus plicatilis and Artemia franciscana enriched on four different commercial media, using the manufacturers' protocols. Pooled samples of 20 cod larvae were homogenized, diluted, and plated out on Petri dishes. The number of colony-forming units per larva was estimated, and the dominant strains subsequently sampled for sequencing of 16S rDNA. Bacteria showing high sequence similarity to a pathogen characteristic of cod and other fish species, Listonella anguillarum, were present in all four groups. Other taxa present among the dominating bacterial colonies were Pseudoalteromonas sp., and Vibrio sp. However, these bacteria could be assigned to genera only. The different enrichments probably affected the number of colony-forming bacteria per millilitre in the enrichment cultures as well as in the larval gastrointestinal (GI) tract. Also, the composition of the microbiota associated with the larval GI tract was probably affected by the enrichment media.
Keywords: Atlantic cod larvae, bacteria, enrichment, live feed
Received 22 September 2004; accepted 28 October 2005.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Aquaculture of Atlantic cod, Gadus morhua L., is currently at the verge of a commercial breakthrough in Norway, having received large private investments during the past 5 years (Svåsand et al., 2004). However, as with many fish species, the cultivation of the early life stages is hampered by high mortality, induced by infection with various opportunistic bacteria. The combination of a supply of bacterial substrates from the live food and high density of a host with a poorly developed immune system provides good conditions for r-selected (i.e. fast-growing) opportunistic pathogenic bacteria (Vadstein et al., 2004).
In intensive aquaculture, cod larvae are offered rotifers Brachionus plicatilis and brine shrimp, mostly Artemia franciscana as live feed (Svåsand et al., 2004). These live feed organisms are cultured and fed different enrichment media. Both rotifers and Artemia are filter-feeding bacteriovores capable of concentrating large amounts of bacteria, and the live feed constitutes a major influx of bacteria to the gastrointestinal tract of the fish (Nicolas et al., 1989; Skjermo and Vadstein, 1993; Makridis et al., 2000a, b, 2001). It seems likely that variation in the enrichment media and conditions affect the composition of the microbiota associated with the live feed organisms, and hence, the composition of the influx of bacteria to the fish larvae. However, apart from studies aimed at using live feed as vectors for probiotic bacteria (for review, see Verschuere et al., 2000), little information is available on the impact of commonly used enrichment media on the intestinal microbiota of fish larvae. The aim of the present study was to identify and compare bacteria isolated from the gut of Atlantic cod larvae fed rotifers enriched on four different media.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Enrichment diets
Four commercial products were used for rotifer enrichment. AlgaMac-2000 (Aquafauna Bio-Marine, Hawthorne, CA, USA) and AquaGrow® Advantage (Advanced Bionutrition, Columbia, MD, USA) both consist of spray-dried algae, Schistochytrium sp. and Crypthecodinium, respectively. Marol E (SINTEF, Trondheim, Norway) is an emulsion, based on oils from the alga Chrypthecodinium cohnii, with added fish oil and vitamins. For Protein Selco®, the producer (INVE, Baasrode, Belgium) provides no information on composition. Chemical composition, according to the producer's datasheets, is shown in Table 1.
|
Live feed cultivation and enrichment
Rotifers ("Tinfoss" strain) (lorica length: 133.7 ± (s.d.) 13.4 µm) were grown semi-continuously using Selco 3000 (INVE, Belgium) as growth medium. Prior to enrichment, rotifers were harvested and rinsed in running water. Temperature and salinity during rinsing and enrichment were 25°C and 22–25, respectively. All four enrichments were mixed in freshwater using a kitchen blender for 1 min before addition to the enrichment tanks. Rotifer density in enrichments was 5 x 105 l–1. Detailed procedures for enrichment are given in Table 2 and are in accordance with producer recommendations.
|
Following completed enrichment, one portion of enriched rotifers for each treatment was cleansed and fed to the larvae. Remaining rotifers were cold-stored (aeration and gradual cooling to 12°C, no rinsing before cold storage), then rinsed and fed to larvae throughout the day in accordance with the feeding schedule.
Brine shrimp (Artemia salina) were hatched from EG cysts (INVE, Belgium) and enriched with A1 (INVE, Belgium), according to the producer's recommendations. Cysts were decapsulated to allow hatching and enrichment in the same tank without the need of an internal harvest at 24 h. Salinity was 22–25 and temperature was 28–30°C. The tank was strongly aerated and light intensity kept at about 2000 lux, measured immediately above the water surface. After 48 h, the production was harvested and fed to the larvae or cold-stored for later feeding.
Fish origin, egg incubation, and hatching
Fertilized eggs of Atlantic cod were transported to the research facilities at Nordland Research Institute, Bodø, Norway from the nearby commercial hatchery Nord-Marine AS. Eggs were disinfected upon arrival (400 ppm glutaraldehyde) in order to avoid bacterial attacks and increased mortality during the egg stage. An egg sample was split into five subsamples of 1 ml each, counted, and the number of eggs estimated to about 420 ml–1. Remaining eggs were placed in a 270-l cylindrical egg incubator and incubated in darkness. Water temperature was 7.1°C, salinity 35, and water exchange 5 l min–1. Central aeration was provided to prevent eggs from clogging. Dead eggs were removed daily from the incubator by turning off aeration and inlet flow, thus allowing dead eggs to descend to the conical tank bottom before draining them out through the bottom outlet. When hatching started, egg volumes equivalent to a larval density of 120 larvae l–1 (28.4 ml) were transferred to each of the experimental tanks. One egg sample was taken for examination of hatching success and for defining day 0 (start of trial). Day 0 was defined when 50% of eggs in the sample hatched. The estimate of initial larval density was adjusted for hatching success (93%) to 110 larvae l–1.
Experimental design and procedures
The four different rotifer enrichments constituted the experimental treatments. Each treatment was assigned in triplicate to 100-l black, flat-bottomed cylindrical tanks with a central outlet and central aeration at 30 cm depth. Seawater supply was taken from 250 m depth and filtered through a 1-µm filter (Harmsco® Filtration Products, North Palm Beach, FL, USA). Water temperature throughout the trial was 7.0°C, s.d. = 0.18, n = 40 (Testo 925 digital thermometer), and salinity was 35. Mean gas saturation (O2) was 99%, s.d. = 1.01, n = 242 (Oxyguard, Handy Delta). Light intensity during the first 8 days was 179 lux (s.d. = 24.1, n = 12; OM 210 digital luxmeter, Robin Electronics Ltd.). Owing to heat production, bulbs were changed on day 9, giving an intensity of 172 lux (s.d. = 24.8, n = 12) for the rest of the trial. Water exchange during the trial, defined as tank volumes per day, was as follows – day 0: 1.8, days 1–6: 5.8, days 7–16: 7.2, and days 17–41: 13. Tanks were cleaned daily, and dead larvae were removed.
Algae (Isochrysis galbana) were added to the tanks at a concentration of about 2.45 x 108 cells l–1, three times a day (08:00, 15:00, and 20:00) from days 1 to 25 post-hatch. Rotifers were fed at a density of 10 animals ml–1, up to four times a day at regular times (08:00, 12:00, 16:00, and 20:00). The feeding scheme for rotifers (times per day) was: days 2–3: 1, days 4–9: 2, days 10–18: 3, and days 18–21: 4 times. Transition from rotifers to Artemia was done during days 22–24. During co-feeding, one rotifer feeding was replaced with Artemia each day, starting with the first-feeding, etc. From day 25 on, only Artemia were fed to the larvae. Artemia were fed at a density of 3 animals ml–1.
At day 0, 50 larvae were taken from the incubator for standard length measurements. Also, samples of five larvae for standard length were taken from each tank at days 8, 14, 21, 28, 35, and 41 post-hatch.
Estimates and statistical analysis
At termination (day 41), remaining live larvae from each tank were counted, and survival was estimated as: survival (%) = (survivors/estimated number at day 0) x 100. Survival was not adjusted for sampling throughout the trial and, therefore, constitutes a minimum estimate.
For all sampling dates, length data were tested for homogeneity of variance and normality using the Levene test and Kolmogorov–Smirnov test, respectively. Owing to homogenous variance, triplicate samples from each treatment were pooled before running separate one-way ANOVAs for each sampling date. When ANOVAs showed significant F-values, Bonferroni post hoc tests were run to decide which treatments were significantly different (p < 0.05). All statistical analyses were carried out using the program SPSS 11.0 for Windows.
Bacterial sampling
Sampling from the intestine of cod larvae used a procedure modified from Muroga et al. (1987) at days 9, 15, 21, and 29. Groups of 20 larvae were immersed in 0.1% benzalkonium chloride (Sigma, Germany) solution for 1 min in order to sterilize the larval surface, then washed in autoclaved seawater with a salinity of 30 for 20 s and homogenized in a glass homogenizer. The homogenate was diluted in sterile seawater and was plated out on two different cultivation media: (i) TSA – Tryptone Soya Broth (Oxoid) and (ii) MA – Marine Agar (Difco). Samples were incubated for 96 h before counting. Incubation temperature was 18°C. Colony-forming units per larvae were calculated, and dominant strains, based on visual judgement of colony morphology, were sampled for identification by sequencing of 16S rDNA. Samplings from enrichment suspensions were incubated at 25°C but were otherwise treated equally.
PCR amplification and 16S rDNA sequence analysis
Bacterial strains were grown on MA plates overnight at 18°C, and a few colonies were harvested and suspended in 500 µl nuclease-free water and used as a template in the PCR reaction.
PCR was performed in 50 µl reaction mixtures using the universal primers 27f (AGAGTTTGATC(A:C)TGGCTCAG) and 1492r (TACGG(C:T)TACCTTGTTACGACTT) covering nearly full length 16S rDNA (Weisbergh et al., 1990). The following PCR reaction mix was used for amplification of this region: 1 µl bacteria template, 50 pmol of each primer, 0.2 mM of each dNTP, 1.25 mM MgCl2, 1x PCR buffer (Promega), and 2.5 U Taq polymerase (Promega). Amplification was initiated with a denaturation step at 95°C for 5 min, followed by 30 cycles each consisting of DNA denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and elongation at 72°C for 1 min. The PCR run was completed with a 10-min step at 72°C. The reaction products were examined on an agarose gel for size and purity before sequencing. To remove the unincorporated primers and dNTP before sequencing, a PCR clean-up kit (Millipore) was used. Purified PCR products were subjected to DNA sequencing using the ABI Prism Big Dye Terminator Cycle Sequencing kit 3.1 (Applied Biosystems). The nucleotide sequences were analysed using the Vector NTITM application programmes (InforMax®). The 16S rDNA sequence was searched for nucleotide–nucleotide matches in the BLAST database at the NCBI homepage (http://www.ncbi.nlm.nih.gov/BLAST/) to establish strain identity.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Mean standard length of larvae at day 0 was 4.2 mm (s.d. = 0.27, n = 50). Standard length did not differ significantly among treatments, with one exception. Variance analysis showed a significant difference among treatments at 28 days post-hatch (ANOVA: F = 7.19, p < 0.001, d.f. = 3), with the Marol E treatment significantly lower than all other groups.
Mortality was very high during the trial (Table 3). During the first 2 days, high mortality was observed in most of the tanks. For three of the tanks, two from the AlgaMac and one from the Protein Selco treatment, early mortality was so high that these tanks were terminated at 35 days post-hatch, and all remaining tanks were terminated at 41 days post-hatch.
|
Large differences were found between the treatments in colony-forming units (CFU) in enrichment media (Table 4). Low bacterial numbers were observed in the AquaGrow group, intermediate bacterial numbers in the Marol E group, and high numbers of CFU in the two remaining groups, as well as in the homogenates of cod larvae (Table 5). In the TSA medium, lower numbers of CFU were generally observed from the AquaGrow group compared with other groups throughout the experiment. No clear pattern could be deduced from the CFU counts in the MA medium.
|
|
" indicating that the CFU number represents the minimum quantity.Bacterial isolates could be assigned to different taxa from BLAST searches performed with the 16S rDNA sequences (Altschul et al., 1990). Many isolates showed high similarity to several previously described marine species. However, in the AlgaMac group almost half the isolates could not be assigned to a taxon, and this was also the case with some isolates from the other three groups. About 15–26% of the isolates from the four experimental groups were highly similar to a well known (Wiik et al., 1989; Espelid et al., 1991; Samuelsen and Bergh, 2004) pathogen of cod and other fish species, Listonella anguillarum (previously Vibrio anguillarum) (Table 6). The sequences showing highest similarity corresponded to a previous isolate of Listonella anguillarum serotype O2a (Wiik et al., 1989) that have been used in challenge models with cod juveniles (Samuelsen and Bergh, 2004).
|
Other taxa present among the dominating colonies were Pseudoalteromonas sp. and Vibrio sp., which could be assigned to genera only. Among these genera, differences were found between the four experimental groups (Table 6). Pseudoalteromonas sp. constituted a large fraction of isolates from the Protein Selco group, whereas only one such strain was present among isolates from the AquaGrow group. Vibrio sp. were present in all groups. A few single isolates were assigned to the genera Pseudomonas sp., Shewanella sp., Microbacterium sp., and Micrococcus sp.
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
With one exception, no significant differences in growth among enrichment media were found, but major differences were found regarding bacteria associated with the larvae, although mortality in all groups was high. The different enrichment media affected the concentration of bacteria in the enrichment cultures as well as in the larval gastrointestinal tract. Also, the composition of the microbiota associated with the larval gastrointestinal tract was different among the four experimental groups, i.e. among groups given live feed enriched on different media. Hence, the results may indicate that the different enrichment media influenced the microbiota in terms of both quantitative and qualitative.
Enrichment with AquaGrow generally gave the lowest amount of culturable bacteria in the enrichment media and in homogenates of larvae at the TSA medium. Towards the end of the experiment, no obvious quantitative differences in the bacterial load could be seen. Interestingly, the AquaGrow group survived best, although it must be stated that survival of Atlantic cod larvae was poor with all four enrichment media.
Different enrichment media also affected the composition of bacteria in the larval gut, as demonstrated by the searches in the BLAST database. It is important to note that this method does not provide a confirmed identification of the different bacteria, but should only be regarded as a tentative identification. In many cases, the most similar sequences corresponded to unidentified Vibrio sp. or Pseudoalteromonas sp. The potential for accumulation of opportunistic pathogens in enrichment cultures constitutes a major shortcoming of the current enrichment methods. Bacteria from live feed is the major influx to the intestinal microbiota of the larvae from the onset of feeding (Nicolas et al., 1989; Bergh et al., 1994; Makridis et al., 2001; Jensen et al., 2004), so the composition of the microbiota associated with live feed is probably a factor of importance for larval survival.
Bacteria similar to Listonella anguillarum appeared abundant among isolates from all four experimental groups. There is some evidence of an oral pathway of infection of fish larvae with this bacterium. Grisez et al. (1996) concluded that juvenile turbot (Scophthalmus maximus) challenged with Artemia sp., released bacteria in the intestine, then transported them through the intestinal epithelium by endocytosis, and released them in the lamina propria, after which the bacteria were transported by blood to the different organs. Although L. anguillarum was also present on larvae from the group fed rotifers enriched with AquaGrow, the lower amounts of bacteria, in terms of CFU per larva in early samples and in the enrichment suspensions, suggest the challenge pressure in this group was lower. However, challenge experiments and pathological studies of larvae, including re-isolation of the pathogenic bacteria from moribund larvae, would be required to ascertain the cause of death.
It is generally believed that certain bacterial species, when abundant in fish larval incubators are opportunistically able to induce mortality (Vadstein et al., 2004). Although existing knowledge about the modes of action of beneficial bacteria are limited, reviews suggest that addition of such bacteria as probiotics may enhance larval development and survival (Gatesoupe, 1999; Hansen and Olafsen, 1999; Ringø and Birkbeck, 1999). The pioneering experiment by Strøm and Ringø (1993) demonstrated that the addition of certain lactic acid bacteria to rearing water could enhance cod larval culture. Cod larvae ingest bacteria even before active feeding commences (Olafsen and Hansen, 1992). Although the role of this uptake is poorly known, intestinal colonization and uptake of bacteria probably influence larval survival. Challenge experiments with yolk-sac larvae of halibut, Hippoglossus hippoglossus (Bergh et al., 1992) and turbot, Scophthalmus maximus (Bergh et al., 1997; Hjelm et al., 2004) demonstrate that the presence of certain bacterial species in sufficient amounts may induce mortality, whereas other bacteria may even increase survival (Hjelm et al., 2004).
In conclusion, mortality was high in all groups, but large differences were found in the type and number of microbiota associated with larvae. Although mortality in general was high, the results indicate that different enrichment media may cause major differences in the microbiota composition.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Altschul S.F., Gish W., Miller W., Myers E.W., Lipman D.J. (1990) Basic local alignment search tool. Journal of Molecular Biology 215:403–410.[CrossRef][Web of Science][Medline]
Bergh Ø., Hansen G.H., Taxt R.E. (1992) Experimental infection of eggs and yolk sac larvae of halibut, Hippoglossus hippoglossus L. Journal of Fish Diseases 15:379–391.[CrossRef][Web of Science]
Bergh Ø., Hjeltnes B., Skiftesvik A.B. (1997) Experimental infection of turbot, Scophthalmus maximus and halibut Hippoglossus hippoglossus yolk sac larvae with Aeromonas salmonicida subsp. salmonicida. Diseases of Aquatic Organisms 29:13–20.[Web of Science]
Bergh Ø., Naas K.E., Harboe T. (1994) Shift in the intestinal microflora of Atlantic halibut (Hippoglossus hippoglossus) larvae during first feeding. Canadian Journal of Fisheries and Aquatic Sciences 51:1899–1903.
Espelid S., Rødseth O.M., Jørgensen T.Ø. (1991) Vaccination experiments and studies of the humoral immune-responses in cod, Gadus morhua L, to 4 strains of monoclonal-defined Vibrio anguillarum. Journal of Fish Diseases 14:185–197.[CrossRef][Web of Science]
Gatesoupe F.J. (1999) The use of probiotics in aquaculture. Aquaculture 180:147–165.[CrossRef][Web of Science]
Grisez L., Chair M., Sorgeloos P., Ollevier F. (1996) Mode of infection and spread of Vibrio anguillarum in turbot Scophthalmus maximus larvae after oral challenge through live feed. Diseases of Aquatic Organisms 26:181–187.[Web of Science]
Hansen G.H. and Olafsen J.A. (1999) Bacterial interactions in early life stages of marine cold water fish. Microbial Ecology 38:1–26.[CrossRef][Web of Science][Medline]
Hjelm M., Bergh Ø., Riaza A., Nielsen J., Melchiorsen J., Jensen S., Duncan H., Ahrens P., Birkbeck H., Gram L. (2004) Selection and identification of autochthonous probiotic bacteria from turbot larvae (Scophthalmus maximus) rearing units. Systematic and Applied Microbiology 27:360–371.[CrossRef][Web of Science][Medline]
Jensen S., Øvreås L., Bergh Ø., Torsvik V. (2004) Phylogenetic analysis of bacterial communities associated with larvae of the Atlantic halibut propose succession from a uniform normal flora. Systematic and Applied Microbiology 27:728–736.[CrossRef][Web of Science][Medline]
Makridis P., Bergh Ø., Skjermo J., Vadstein O. (2001) Addition of bacteria bioencapsulated in Artemia metanauplii to a rearing system for halibut larvae. Aquaculture International 9:225–235.[CrossRef]
Makridis P., Fjellheim A.J., Skjermo J., Vadstein O. (2000) Colonization of the gut in first feeding turbot by bacterial strains added to the water or bioencapsulated rotifers. Aquaculture International 8:367–380.[CrossRef][Web of Science]
Makridis P., Fjellheim A.J., Skjermo J., Vadstein O. (2000) Control of the bacterial flora of Brachionus plicatilis and Artemia franciscana by incubation in bacterial suspensions. Aquaculture 185:207–218.[CrossRef][Web of Science]
Muroga K., Higashi M., Keitoku H. (1987) The isolation of intestinal microflora of farmed red sea bream (Pagrus major) and black sea bream (Acanthopagrus schlegeli) at larval and juvenile stages. Aquaculture 65:79–88.[CrossRef][Web of Science]
Nicolas J.L., Robic E., Ansquer D. (1989) Bacterial flora associated with a trophic chain consisting of microalgae, rotifers and turbot larvae: influence of bacteria on larval survival. Aquaculture 83:237–248.[CrossRef][Web of Science]
Olafsen J.A. and Hansen G.H. (1992) Intact antigen uptake in intestinal epithelial cells of marine fish larvae. Journal of Fish Biology 40:141–156.[CrossRef][Web of Science]
Ringø E. and Birkbeck T.H. (1999) Intestinal microflora of fish larvae and fry. Aquaculture Research 30:73–93.[CrossRef][Web of Science]
Samuelsen O.B. and Bergh Ø. (2004) Efficacy of orally administered florfenicol and oxolinic acid for the treatment of vibriosis in cod (Gadus morhua). Aquaculture 235:27–35.[CrossRef][Web of Science]
Skjermo J. and Vadstein O. (1993) Characterization of the bacterial flora of mass cultivated Brachionus plicatilis. Hydrobiologia 255/256:185–191.
Strøm E. and Ringø E. (1993) Changes in the bacterial composition of early developing cod, Gadus morhua (L.) larvae following inoculation of Lactobacillus plantarum into the water. In Walther B.T. and Fyhn H.J. (Eds.). Physiological and Biochemical Aspects of Fish Development(Zoological Institute, University of Bergen, Norway) pp. 226–228.
Svåsand T., Otterå H., Taranger G.L. (2004) Atlantic cod. In Moksness E., Kjørsvik E., Olsen Y. (Eds.). Culture of Cold-water Marine Fish(Blackwell Scientific Publications Ltd., Oxford) pp. 433–444 528 pp.
Vadstein O., Mo T.A., Bergh Ø. (2004) Microbial interactions, prophylaxis and diseases. In Moksness E., Kjørsvik E., Olsen Y. (Eds.). Culture of Cold-water Marine Fish(Blackwell Scientific Publications Ltd., Oxford) pp. 28–72 528 pp.
Verschuere L., Rombaut G., Sorgeloos P., Verstraete W. (2000) Probiotic bacteria as biological control agents in aquaculture. Microbiology and Molecular Reviews 64:655–671.[CrossRef]
Weisbergh W.G., Barns S.M., Pellitier D.A., Lane D.J. (1990) 16S ribosomal DNA amplification for phylogenetic study. Journal of Bacteriology 173:697–703.[Web of Science]
Wiik R., Hoff K.A., Andersen K., Daae F.L. (1989) Relationships between plasmids and phenotypes of presumptive strains of Vibrio anguillarum isolated from different fish species. Applied and Environmental Microbiology 55:826–831.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||