Journal of Applied Microbiology 2000, 89, 317322

Lactic acid bacteria associated with the digestive tract of

Atlantic salmon (Salmo salar L.)
E. Ring1, H.R. Bendiksen1, M.S. Wesmajervi2, R.E. Olsen3*, P.A. Jansen4{ and H. Mikkelsen5
1

Department of Arctic Veterinary Medicine, The Norwegian School of Veterinary Science, Troms, 2Department of Medical
Microbiology, University of Troms, Troms, 3Department of Fisheries, Finnmark College, Alta, 4Finnmark Research Center,
Hammerfest, 5Norwegian Institute of Fisheries and Aquaculture Ltd, Troms, Norway
222/3/2000: received 10 March 2000, revised 31 March 2000 and accepted 4 April 2000
E. RING, H.R. BENDIKSEN, M.S. WESMAJERVI, R.E. OLSEN, P.A. JANSEN AND H. MIKKELSEN.
2000. The present study reports the effect of excessive handling stress and starvation on the
lactic acid bacteria associated with the digestive tract of Atlantic salmon (Salmo salar L.). A
relatively low population level (approximately 2  103 bacteria per gram wet tissue) of viable
adherent heterotrophic bacteria was associated with the digestive tract (foregut, midgut and
hindgut). Of the 752 bacterial isolates isolated from diet, water and the digestive tract, 201
isolates belonged to the carnobacteria. Of these isolates, one from the diet, one from the
rearing water and 80 from the gastrointestinal tract, were further identied on the basis of
16S rDNA sequence analysis. All these isolates were identied as being Carnobacterium
piscicola-like. Daily repeated stress and starvation of the sh over 11 d had no inuence on
the total culturable bacterial numbers or population level of C. piscicola associated with the
digestive tract. C. piscicola-like isolates colonizing the various intestinal regions (foregut,
midgut and hindgut) were also screened for their ability to produce growth inhibitory
compounds active against the sh pathogen Aeromonas salmonicida. Of the 199 C. piscicola
isolates tested, 139 inhibited growth of the pathogen.

INTRODUCTION

During the last decade interest has been expressed in the

activities of lactic acid-producing bacteria associated with
epithelial surfaces in the digestive tracts of sh, and it is generally accepted that lactic acid bacteria are part of the normal
intestinal ora of sh (for reviews see Joborn 1998; Ring and
Gatesoupe 1998). Evidence that lactic acid bacteria can
associate with intestinal surfaces would be of particular interest in the use of probiotic isolates to produce a benecial
effect on the host animal, and there are reports that lactic acid
bacteria isolated from the sh intestine are antagonistic
towards sh pathogens (for review see Ring and Gatesoupe
1998). In sh, little information is available on the effect of
handling stress on the allochthonous intestinal bacterial ora
(Lesel and Sechet 1979), and no information is available on
the effect of stress on the population level of lactic acid bacCorrespondence to: E. Ring, Department of Arctic Veterinary Medicine, The
Norwegian School of Veterinary Science, No-9292, Troms, Norway (e-mail:
einar.ringo@veths.no).
*Present address: Institute of Marine Research, Matre Aquaculture Research
Station; Matredal, Norway, {Department of Zoology, Norwegian University
of Science and Technology, Trondheim, Norway
= 2000 The Society for Applied Microbiology

teria colonizing the digestive tract. The aim of the present

study was therefore to examine how population level and
composition of lactic acid bacteria associated with the gut are
affected by daily repeated stress and starvation.
As lactic acid bacteria isolated from sh are known to
produce substances which are bactericidal in vitro
(Schrder et al. 1980; Strm 1988; Pilet et al. 1995; Joborn
1998; Ring 1999), this study screened the ability of intestinal carnobacteria to inhibit growth of the sh pathogen
Aeromonas salmonicida.

MATERIALS AND METHODS

Fish and experimental conditions

Atlantic salmon (Salmo salar L.) from a fth generation

bred stock (Gjedrem et al. 1991) were used in the experiments. The salmon had been hatched and reared at
Ishavssmolt Ltd, Finnmark, Norway, and were delivered as
yearling smolts to Finnmark Research Centre, Station of
Mariculture Research in June 1995, where the experiments
were carried out. The sh, mean average weight of about
400 g, were held in sea water (S 33  /oo) at approximately

318 E . R I N G E T A L .

4  C in indoor rearing tanks, and were fed by hand a commercial diet (Skretting Ltd, Stavanger, Norway) in excess.
The sh were divided into two tanks, each containing 50
salmon. The self-cleaning PVC tanks (diameter 4 m, h
0
4 m) were continuously supplied with running water
(1 30 l min1). Both groups were allowed a 2-week adaptation period. Two weeks before the start of the experiment
one of the groups was starved to allow the sh gut to
empty. At the experiment start, six sh from both groups
were sampled for bacteria. Thereafter, stress was induced
in both tanks by reducing the water level to 510 cm
depth, and subsequently chasing the sh in the tank with a
pole for 5 min. This procedure was used as water level
reduction alone is known to elicit stress responses in
Atlantic salmon (Einarsdottir and Nilsen 1996). No salmon
were injured during the stress procedure and the seawater
supply was not stopped during the experiments. The stress
procedure was repeated daily for 11 d. During the experiment no mortalities occurred.
Bacteriological examination

Samples of gut, sea water and diet were collected for bacteriological analysis. Indigenous gut microora were
sampled at the experimental start from six individual sh
from the fed and non-stressed rearing group and six individual sh from the starved and non-stressed group. Post
stress, the gut microora from both rearing groups were
sampled after 5 h (n 6), 24 h (n 6) and after 11 d of
daily handling stress from both fed and starved sh (n 6).
The aseptic removal of the digestive tract from the sh is
described in detail elsewhere (Ring 1993). The gastrointestinal tract was divided into three regions. Region A, the
foregut, was dened as the region from the last pyloric caecum and 3 cm down, region B, the midgut (3 cm before its
increase and 3 cm further down) and region C, the hindgut
(the remaining part of the intestine). The different regions
were emptied and thoroughly rinsed three times in sterile
0
9% saline to remove nonadherent bacteria before homogenization of the different regions (approximately 0
30
4 g
of each) in sterile plastic bags and a Stomacher (Seward
Laboratory, London, UK). Homogenates of the different
regions of the gut were diluted in 0
9% saline and 0
1 ml
volumes of appropriate dilutions were spread on the surface
of tryptic soy agar plates containing glucose (5 g l1) with
(TSAgs) or without (TSAg) addition of 2% NaCl. The
plates were incubated at 12  C for 4 weeks.
Phenotypic identification. After 4 weeks' incubation, 752
isolates from diet, water and the digestive tract were randomly picked and re-streaked on fresh plates similar to
those from which they were originally isolated to ensure

purity before phenotypic characterization. Pure cultures of

all isolates were subjected to standard tests; Gram-staining,
morphology, motility, production of catalase and oxidase,
Hugh and Leifson's fermentation test and gas from glucose. Gram-positive bacteria belonging to the lactic acid
bacteria group were further tested for growth on acetate
agar (pH 5
4) and cresol red thallium acetate sucrose
(CTAS) agar (pH 9
1) (Hamnes et al. 1992). Isolates
obtained from the gastrointestinal tract which did not
belong to the lactic acid bacteria were classied according
to Muroga et al. (1987), Ring (1993) and Ring and Olsen
(1999).
Preparation of DNA, oligonucleotide primers, general PCR
conditions and DNA sequence analysis of carnobacteria.

One isolate (F201) from the diet, one isolate (V202) from
the rearing water, 80 out of 199 isolates (randomly chosen)
from the digestive tract and four type strains of
Carnobacterium (C. divergens CCUG 30094, C. piscicola
CCUG 34645, C. gallinarum CCUG 30095 and C. mobile
CCUG 30096) obtained from the University of
Gothenburg Culture Collection were analysed by 16S
rDNA sequence analysis. A detailed description of the preparation of DNA, oligonucleotide primers, general PCR
conditions and DNA sequencing is presented by Ring
et al. (1998).
Fatty acid analysis of carnobacteria isolates. Membrane
fatty acid methyl esters of carnobacteria cells were prepared
and analysed as described elsewhere (Jstensen and
Landfald 1996).
Growth inhibition of Aeromonas salmonicida

Inhibition of the sh pathogen A. salmonicida isolated from

an Atlantic salmon infected with furunculosis were performed by a microtitre plate assay described by Gram et al.
(1999). Pre-cultures of the test bacteria were inoculated
into tryptic soy broth with added glucose and NaCl
(TSBgs). The asks were incubated at room temperature
with agitation (150 rev min1). Samples were withdrawn
after 24 h and sterile-ltered (0
22 mm Millex1-GV,
Millipore). The possible inhibitory activity of the sterile-ltered supernatant was tested by adding 50 ml supernatant to
150 ml fresh medium (TSBgs) in microtiter wells (Nunc
microwell 96 F) and inoculation with 10 ml A. salmonicida
yielding approximately 106 cfu ml1. Controls consisted of
inoculating the sh pathogen in 200 ml fresh medium. Each
combination was tested in triplicate, and the growth was
estimated as optical density at 600 nm (O.D.600) using a
Labsystems Multiscan1 Bichromatic microtitre plate
reader (Labsystems, Helsinki, Finland).

= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 89, 317322

319

LAB IN ATLANTIC SALMON

Table 1 Log total viable counts (log TVC) per gram wet mass, total number of isolates and frequency of lactic acid bacteria isolated from

three different regions of the digestive tract of fed Atlantic salmon (Salmo salar L.) compared with those and starved for two weeks prior
to the start of experimentation. Values of log TVC are means of six sh.
Fed sh
gut regions
A

Day 0
Log TVC
S.E.M.
Total no. of isolates
Lactic acid bacteria
Carnobacterium sp.
Streptococcus sp.
5h
Log TVC
S.E.M.
Total no. of isolates
Lactic acid bacteria
Carnobacterium sp.
Lactobacillus sp.
24 h
Log TVC
S.E.M.
Total no. of isolates
Lactic acid bacteria
Carnobacterium sp.
Day 11
Log TVC
S.E.M.
Total no. of isolates
Lactic acid bacteria
Carnobacterium sp.
Lactobacillus sp.

Starved sh
gut regions

2
62
2
07
20

3
00
2
62
12

2
45
2
09
13

3
31
2
70
23

3
96
3
26
18

3
76
3
48
17

14
ND

8
ND

8
ND

3
3

1
1

4
ND

2
69
2
04
57

2
81
2
19
84

2
39
1
41
33

2
70
2
21
50

3
51
2
91
34

3
96
3
34
36

9
ND

12
2

10
ND

11
ND

8
ND

10
ND

2
78
2
19
21

2
61
1
88
32

2
85
2
34
38

2
96
2
42
24

3
50
3
00
32

3
53
3
00
31

11

16

15

10

3
03
2
37
16

3
09
2
50
14

2
68
2
12
24

3
10
2
49
21

3
28
2
80
21

3
42
3
02
19

4
ND

4
ND

7
ND

11
ND

6
ND

5
1

Regions: A foregut; B midgut; C hindgut ND not detected.

S.E.M. standard error of the mean.

RESULTS

The total viable counts of diet and rearing water were 2
7
 102 g1 and 2
3  103 ml1, respectively. The random
bacterial isolates chosen included 24 from the feed, 38 from
water (results not shown), and 690 from the gastrointestinal
tract (foregut, midgut and hindgut) of fed and starved sh
(Table 1). Of the 24 bacterial isolates isolated from the
feed, one (F201) was identied as a Carnobacterium sp. Of a
total of 38 isolates from the rearing water one (V202) was
classied as a Carnobacterium sp. In addition, the rearing
water contained one Leuconostoc sp. and one Streptococcus
sp. isolate.
When fed and starved Atlantic salmon were exposed to
handling stress no clear reduction was observed in total

viable counts (TVC) in the different regions (foregut, midgut and hindgut) of the gastrointestinal tract (Table 1). A
total of 690 culturable isolates were isolated from the digestive tract of both experimental groups. Of these 206
(29
9%) belonged to lactic acid bacteria (Table 1), of which
199 isolates (28
8%) were similar to Carnobacterium spp.,
four (0
6%) to Streptococcus spp. and three (0
4%) to
Lactobacillus spp. The lactobacilli isolates were identied as
Lactobacillus group 1 according to the scheme of Gancel
et al. (1997). The remaining isolates (484) which were not
lactic acid bacteria were classied into 12 taxonomic
groups; Brevibacterium, Microbacterium, Micrococcus,
Staphylococcus, Acinetobacter, Aeromonas, Alcaligenes,
Cytophaga/Flexibacter,
Moraxella,
Photobacterium,
Pseudomonas and Xanthomonas (results not shown). These

= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 89, 317322

320 E . R I N G E T A L .

12 taxonomic groups were detected in the digestive tract

immediately before handling stress and 5 h after stress.
However, after 11 d of regular stress we were only able to
detect Acinetobacter, Pseudomonas and Xanthomonas associated with the intestine of both groups (results not
shown).
To obtain more taxonomic information about the carnobacteria, one isolate from the diet, one isolate from the
rearing water and 80 isolates (randomly chosen) from the
gastrointestinal tract were subjected to partial sequencing
of the 16S rDNA. The sequencing data were compared
with sequences for Carnobacterium type strains obtained
from the Culture Collection, University of Gothenburg
(Sweden). The above 82 isolates matched 100% over a 578
nucleotide region compared with the type strain C. piscicola
CCUG 34645.
Fatty acid analysis of carnobacteria isolates

The most abundant cellular fatty acid of the carnobacteria

isolated from the gastrointestinal tract, water and diet was
found to be 18 : 1 (n-9), and a comparison of the amount of
18 : 1 (n-9) in sh isolates with the four type strains is
shown in Table 2. The C18 : 1 isomer represented between
47
5 and 52
8% of the cellular fatty acid of carnobacteria
isolates isolated from water, diet and the gastrointestinal
tract of Atlantic salmon. The content of 18 : 1 (n-9) was
highly variable in the cellular lipid of the four type strains,
as 51
3% was produced by C. piscicola while less than 29%
was observed in C. divergens. In addition to 18 : 1 (n-9), all
carnobacteria isolates investigated in the present study were

strongly dominated by the straight-chained fatty acids 14 :

0, 16 : 0 and 16 : 1 in cellular lipid (Table 2).

Inhibition of growth of Aeromonas salmonicida

The growth inhibition of A. salmonicida in the microtitre

plate assay (Table 3) revealed that of 199 C. piscicola isolates from the digestive tract of Atlantic salmon, 139 inhibited pathogen growth. The number of isolates inhibiting
the pathogen was highest in sh sampled prior to handling
stress (37/38). Later in the experimental period, after 11 d
of daily handling stress only 16/37 isolates inhibited
growth of A. salmonicida. Of the 139 isolates able to inhibit
growth of A. salmonicida, 50 were isolated from the foregut,
42 from the midgut and 47 from the hindgut.

DISCUSSION
Starvation

The gastrointestinal tract of sh harbours a complex collection of microbes, and in experiments evaluating the intestinal microora of sh it has been considered important to
avoid stress, such as starvation. In an early study, Margolis
(1953) reported that bacteria do not persist in the intestine
of fasting sh. These results are in contrast to later ndings
of Trust (1975), Lesel (1979), Conway et al. (1986) and the
current study, showing that fasting does not change total
numbers of viable autochthonous aerobic bacteria.

Table 2 Cellular fatty acid composition of Carnobacterium isolates from water, diet and the gastrointestinal tract (GIT) of Atlantic salmon

compared with those of (A) Carnobacterium divergens CCUG 30094 (B) C. piscicola CCUG 34645 (C) C. gallinarum CCUG 30095 and (D)
C. mobile CCUG 30096
Fatty acids

Water

Diet

GIT

14 : 0
14 : 1
16 : 0
16 : 1 (n-7)
16 : 1 (n-9)
18 : 0
18 : 1 (n-7)
18 : 1 (n-9)
cyclo 19 : 1
20 : 1

10
2
1
9
15
9
1
5
13
6
2
0
Tr.
50
9
ND
Tr.

11
5
2
0
16
0
2
4
10
8
1
5
Tr.
48
2
ND
2
0

7
313
9
1
83
7
14
618
8
1
02
7
7
714
5
1
13
3
Tr.
43
552
8
ND
1
03
3

12
6
Tr.
15
0
Tr.
12
4
5
3
Tr.
27
5
20
5
Tr.

15
1
Tr.
17
8
Tr.
12
6
3
0
Tr.
50
2
ND
Tr.

16
7
Tr.
21
8
Tr.
12
6
2
0
Tr.
42
8
ND
Tr.

4
4
Tr.
16
4
Tr.
22
9
2
2
Tr.
50
1
ND
Tr.

Tr. trace, < 0
5%.

ND not detected.
No. of isolates tested: water, 1; Diet, 1; GIT, 80; A, 1; B, 1; C, 1, D, 1.
= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 89, 317322

LAB IN ATLANTIC SALMON

Table 3 Growth inhibition of Aeromonas salmonicida ssp.

salmonicida strain LFI 4038 by Carnobacterium piscicola-like

isolates isolated from different regions of the gastrointestinal tract
of fed Atlantic salmon. Data shows the number of inhibitory
strains of the total isolates tested
Proportion of samples inhibiting
growth of A. salmonicida
Foregut

Midgut

Hindgut

Ultimately before stress

5 h after stress
24 h after stress

17/17
16/20
11/21

8/9
16/20
13/20

12/12
14/20
16/23

Stress

The digestive tract is a system that reacts to stress-like stimuli, but there is little information available on how handling stress affects the balance of the gastrointestinal
microbiota (Lesel and Sechet 1979). In the present study
no clear effect of stress was seen on population levels of
viable autochthonous aerobic bacteria or composition of
carnobacteria. Our results with Atlantic salmon contradict
ndings in endothermic animals that population level of
lactic acid bacteria in the digestive tract decrease during
environmental stress (Tannock and Savage 1974; Tannock
1983).
16S rDNA sequence analysis of carnobacteria

The presence of carnobacteria in the digestive tract of sh

has been reported in several investigations (for review see
Ring and Gatesoupe 1998), while the presence of intestinal C. piscicola has only been reported in three investigations (Strm 1988; Ring et al. 1997, 1998). In the present
study, all 80 intestinal carnobacteria isolates identied on
the basis of 16S rDNA sequence analyses strongly suggest
they can be classied as C. piscicola-like. Secondly, their
high content of 18 : 1 (n-9) in cellular lipids (Table 3) correspond well to known content of C18 : 1 in cellular lipid of C.
piscicola (Hamnes et al. 1992; Ring et al. 1998).
The primers used in the present study to identify the
carnobacteria isolates covered a region of 578 bp of the
total rDNA. However, to make a more complete identication of all isolates a larger region should to be sequenced,
in combination with other molecular methods such as
AFLP2 microbial ngerprinting.
Growth inhibition of fish pathogens

The bacteriology of the sh intestinal tract has been the

focus of many reviews throughout the years (Horsley 1977;

321

Cahill 1990; Ring and Gatesoupe 1998; Ring and

Birkbeck 1999), and there is evidence that dense microbial
populations occur in the digestive tract with numbers of
bacteria higher than those of the surrounding water.
Because the sh gut microbiota can play a role against
invading pathogens (Joborn 1998; Ring and Gatesoupe
1998; Ring 1999) one can wonder whether this implies
production of antibacterial compounds (e.g. bacteriocins).
During the last two decades, several studies have shown
that the gastrointestinal tract of various sh species contains lactic acid bacteria which produce antibacterial compounds able to inhibit the growth of Gram-negative sh
pathogens in vitro (Schrder et al. 1980; Strm 1988; Pilet
et al. 1995; Joborn 1998; Ring 1999). In the present work,
we demonstrated that 139 out of 199 C. piscicola-like isolates isolated from the gastrointestinal tract of Atlantic salmon inhibited growth of A. salmonicida. However, it is not
clear why the frequency of C. piscicola-like isolates able to
inhibit the pathogen decreased during the experiment. It
could be that, on subculture, the inhibitory qualities were
lost and this hypothesis calls for further studies.
Based on the results in the current study it is suggested
that the intestinal microbiota may be involved in the protection of sh against certain pathogenic bacteria.

ACKNOWLEDGEMENTS

Thanks to Ms R. Larsen for technical assistance, and to

Ms Kirsten Zachariassen for correction of the English.
Financial support from Norwegian Research Council (grant
no. 116154/122 and no. 122851/122) is gratefully acknowledged.

REFERENCES
Cahill, M.M. (1990) Bacterial ora of shes: a review. Microbial
Ecology 19, 2141.
Conway, P.L., Maki, J., Mitchel, R. and Kjelleberg, S. (1986)
Starvation of marine ounder, squid and laboratory mice and
its effect on the intestinal microbiota. FEMS Microbiology
Ecology 38, 187195.
Einarsdottir, I.E. and Nilsen, K.J. (1996) Stress responses of
Atlantic salmon (Salmo salar L.) elicited by water level reduction in rearing tanks. Fish Physiology and Biochemistry 15, 395
400.
Gancel, F., Dzierszinski, F. and Tailliez, R. (1997) Identication
and characterization of Lactobacillus species isolated from llets
of vacuum-packed smoked and salted herring (Clupea harengus).
Journal of Applied Microbiology 82, 722728.
Gjedrem, T., Gjen, H.M. and Gjerde, B. (1991) Genetic origin
of Norwegian farmed Atlantic salmon. Aquaculture 98, 4150.
Gram, L., Melchiorsen, J., Spanggard, B., Huber, I. and Nielsen,
T.F. (1999) Inhibition of Vibrio anguillarum by Pseudomonas

= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 89, 317322

322 E . R I N G E T A L .

uorescens AH2, a possible probiotic treatment of sh. Applied

and Environmental Microbiology 65, 969973.
Hamnes, W.P., Weiss, N. and Holzapfel, W. (1992) The Genera
Lactobacillus and Carnobacterium in the Prokaryotes. A Handbook
on the Biology of Bacteria: Ecophysiology, Isolation, Identication,
Applications, Vol. II, ed. Balows, A., Truper. H.G., Dworkin,
M., Harder, W. and Schleifer, K.H. pp. 15361594. New York,
NY: Springer-Verlag.
Horsley, R.W. (1977) A review of the bacterial ora of teleosts
and elasmobranchs, including methods for its analysis. Journal
of Fish Biology 10, 529553.
Joborn, A. (1998) The Role of the Gastrointestinal Microbiota in
the Prevention of Bacterial Infections in Fish. PhD Thesis,
Goteborg: Goteborg University.
Jstensen, J.-P. and Landfald, B. (1996) Inuence of growth conditions on fatty acid composition of a polyunsaturated fattyacid-producing Vibrio species. Archives of Microbiology 165,
306310.
Lesel, R. (1979) Donnees preliminaires sur la microore du tube
digestif de la truite arc-en-ciel Salmo gairdneri Richardson.
Revue Biologie et Ecologie Mediterraneenne 4, 167174.
Lesel, R. and Sechet, J. (1979) Inuence d'un stress de manipulation sur le transit de la microore bacterienne digestive chez la
truite arc-en-ciel Salmo gairdneri Richardson. Acta Oecologica
Oecologica Applicata 3, 2333.
Margolis, L. (1953) The effect of fasting on the bacterial ora of
the intestine of sh. Journal of Fishery Resources Board Canada
10, 6263.
Muroga, K., Higashi, M. and Keitoku, H. (1987) The isolation of
intestinal microora of farmed red seabream (Pagrus major) and
black seabream (Acanthopagrus schlegeli) at larval and juvenile
stages. Aquaculture 65, 7988.
Pilet, M.-F., Dousset, X., Barre, R., Novel, G., Desmazeaud, M.
and Piard, J.-C. (1995) Evidence for two bacteriocins produced
by Carnobacterium piscicola and Carnobacterium divergens isolated
from sh and active against Listeria monocytogenes. Journal of
Food Protection 58, 256262.
Ring, E. (1993) The effect of chromic oxide (Cr2O3) on aerobic
bacterial populations associated with the intestinal epithelial
mucosa of Arctic charr, Salvelinus alpinus (L.). Canadian
Journal of Microbiology 39, 11691173.
Ring, E. (1999) Lactic acid bacteria in sh: antibacterial effect
against sh pathogens. In Effects of Antinutrients on the

.,
Nutritional Value of Legume Diets, Vol. 8, ed. Krogdahl, A
Mathiesen, S.D. and Pryme, I. COST 98. pp. 7075.
Luxembourg: EEC Publication.
Ring, E., Bendiksen, H.R., Gausen, S.J., Sundsfjord, A. and
Olsen, R.E. (1998) The effect of dietary fatty acids on lactic
acid bacteria associated with the epithelial mucosa and from faecalia of Arctic charr, Salvelinus alpinus (L.). Journal of Applied
Microbiology 85, 855864.
Ring, E. and Birkbeck, T.H. (1999) Intestinal microora of sh
larvae and fry. Aquaculture Research 30, 7393.
Ring, E. and Gatesoupe, F.-J. (1998) Lactic acid bacteria in sh:
a review. Aquaculture 160, 177203.
Ring, E. and Olsen, R.E. (1999) The effect of diet on aerobic
bacterial ora associated with intestine of Arctic charr
(Salvelinus alpinus L.). Journal of Applied Microbiology 86, 22
28.
Ring, E., Olsen, R.E., verli, . and Lvik, F. (1997) Effect of
hierarchy formation on aerobic microbiota associated with
epithelial mucosa of subordinate and dominant individuals of
Arctic charr, Salvelinus alpinus (L.). Aquaculture Research 28,
901904.
Schrder, K., Clausen, E., Sandberg, A.M. and Raa, J. (1980)
Psychrotropic Lactobacillus plantarum from sh and its ability to
produce antibiotic substances. In Advances in Fish Science and
Technology ed. Connell, J.J. pp. 480483. Surry: Fishing News
Books.
Strm, E. (1988) Melkesyrebakterier i Fisketarm. Isolasjon, karakterisering og egenskaper. MSc Thesis. The Norwegian College
of Fishery Science, University of Troms, Troms, Norway.
[In Norwegian.]
Tannock, G.W. (1983) Effect of dietary and environmental stress
on the gastrointestinal microbiota. In Human Intestinal
Microora in Health and Disease ed. Hentgens, D.J. pp. 517
539. New York, NY: Academic Press.
Tannock, G.W. and Savage, D.C. (1974) Association of microorganisms with epithelium in the alimentary tract of Aspicularis
tetraptera. Infection and Immunity 9, 475476.
Trust, T. (1975) Facultative anaerobic bacteria in the digestive
tract of chum salmon (Oncorhynchus keta) maintained in freshwater under dened culture conditions. Applied Microbiology 29,
663668.

= 2000 The Society for Applied Microbiology, Journal of Applied Microbiology, 89, 317322