Fish & Shellfish Immunology 35 (2013) 766e775

Contents lists available at SciVerse ScienceDirect

Fish & Shellfish Immunology

journal homepage: www.elsevier.com/locate/fsi

Effects of dietary arabinoxylan-oligosaccharides (AXOS) and

endogenous probiotics on the growth performance, non-specific
immunity and gut microbiota of juvenile Siberian sturgeon
(Acipenser baerii)
Zahra Geraylou a, b, *, Caroline Souffreau a, Eugene Rurangwa a, c, Luc De Meester a,
Christophe M. Courtin d, Jan A. Delcour d, Johan Buyse b, Frans Ollevier a
a
Laboratory of Aquatic Ecology, Evolution and Conservation, KU Leuven, Leuven, Belgium
b
Laboratory of Livestock Physiology, Immunology and Genetics, KU Leuven, Leuven, Belgium
c
Institute for Marine Resources and Ecosystem Studies (IMARES), Wageningen UR, Yerseke, The Netherlands
d
Laboratory of Food Chemistry and Biochemistry & Leuven Food Science and Nutrition Research Centre (LFoRCe), KU Leuven, Leuven, Belgium

a r t i c l e i n f o a b s t r a c t

Article history: We investigated the effects of administration of putative endogenous probiotics Lactococcus lactis spp.
Received 12 May 2013 lactis or Bacillus circulans, alone and in combination with arabinoxylan-oligosaccharides (AXOS), a new
Received in revised form class of candidate prebiotics, in juvenile Siberian sturgeon (Acipenser baerii). Eight experimental diets
3 June 2013
were tested: basal diet (Diet 1), basal diet supplemented with 2% AXOS (Diet 2), or L. lactis ST G81 (Diet
Accepted 10 June 2013
3), L. lactis ST G45 (Diet 4), B. circulans ST M53 (Diet 5), L. lactis ST G81 þ 2% AXOS (Diet 6), L. lactis ST
Available online 25 June 2013
G45 þ 2% AXOS (Diet 7), B. circulans ST M53 þ 2% AXOS (Diet 8). After four weeks, growth performance
and feed conversion ratio significantly improved in fish fed diet 7. Innate immune responses of fish were
Keywords:
Siberian sturgeon
boosted with both AXOS and probiotic diets, however synergistic effects of AXOS and probiotic diets
AXOS were only observed for phagocytic and alternative complement activity. Phagocytic and respiratory burst
Probiotics activity of fish macrophage increased in fish fed diet 2 and 7, while humoral immune responses only
Synbiotic increased in fish fed diet 7. Pyrosequencing analysis (16S rDNA) of the hindgut microbiota demonstrated
Immune responses that AXOS improved the colonization or/and growth capacity of L. lactis, as a higher relative abundance of
L. lactis was observed in fish receiving diet 7. However, no observable colonization of B. circulans was
found in the hindgut of fish fed diet 5 or 8, containing this bacterium. The dietary L. lactis ST G45 þ 2%
AXOS caused significant alterations in the intestinal microbiota by significantly decreasing in bacterial
diversity, demonstrated by the fall in richness and Shannon diversity, and improved growth performance
and boosted immune responses of Siberian sturgeon.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction infectious diseases is therefore vital. Traditional disease control

In recent years, sturgeon culture has developed into a very
successful industry, due to the precious eggs (caviar) and meat.
Intensive culture exposes the fish to different sources of stress such
as high stocking densities and manipulations. Development of
effective methods to boost the fish immune system and to avoid
* Corresponding author. Laboratory of Aquatic Ecology, Evolution and Conser-
vation, Zoological institute, KU Leuven, Charles Deberiotstraat 32, 3000 Leuven,
Belgium. Tel.: þ32 16 32 39 48; fax: þ32 16 32 45 75.
E-mail address: Zahra.Geraylou@biw.kuleuven.be (Z. Geraylou).
strategies include antibiotics and chemical disinfectants. However,
these are no longer recommended practices due to the emergence
of bacterial resistance [1] and concerns about environmental
impact and wildlife protection [2]. In the past decennium, natural
prophylactic supplements including probiotics and prebiotics have
received a great deal of attention to replace chemotherapeutics in
aquaculture. Probiotics are defined as live microbial feed supple-
ments which beneficially affect the host animal by improving its
intestinal microbial balance [3]. Several studies have demonstrated
that probiotics can improve growth performance, feed utilization,
digestibility of dietary ingredients, disease resistance and immune
responses of aquatic animals [4e7]. The most common probiotics

1050-4648/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.fsi.2013.06.014
 Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775
used in aquaculture are Lactobacillus spp., Bacillus spp., Vibrio spp.,
Saccharomyces spp. and Enterococcus spp. [8].
Evidence of the beneficial effects of probiotics gave birth to the
concept of prebiotics [9], which are defined as indigestible (by the
host) feed components that provide beneficial effects to the host
through their selective metabolism by favourable bacteria in the
gastrointestinal tract (GI) [9,10]. It appears more practical in aquatic
animals to manipulate the GI tract microbiota through the use of
prebiotics that alter the GI tract conditions to favour certain bacte-
rial species which may enhance fish growth efficiency and reduce
disease susceptibility of the host organism [11]. However, only
during the last decade that an improved understanding arose of the
importance of commensal microbiota in the fish intestine [11]. Since
then, positive effects of some prebiotics have been presented on
growth performance, innate immunity, haematological and serum
biochemical parameters, microbial fermentation and autochtho-
nous intestinal microbiota of some sturgeon species [12e17].
Arabinoxylan-oligosaccharides (AXOS) represent a new class of
candidate prebiotics. They are fragmentation products of arabi-
noxylans (AX), which occur in the cell wall of many cereal grains,
consisting of a main chain of beta-1,4-linked D-xylopyranosyl units
to which O-2 and/or O-3-L-arabinofuranosyl units are linked [18e
20]. Beneficial effects of AXOS on growth performance, gut micro-
biota composition and fermentation have already been reported for
pigs [21], broilers [22,23], rat [24] and humans [25e27]. Our pre-
vious study indicated that incorporation of AXOS in Siberian stur-
geon diet resulted in modulation of gut microbiota and their
metabolites in hindgut and subsequently improved innate immune
responses of Siberian sturgeon [28].
Synbiotics are products that contain both probiotics and pre-
biotics. Gibson and Roberfroid stated that the use of synbiotics
provides the benefits of both pre- and probiotics mainly due to
their synergistic effects [9]. The combined application of probiotics
and prebiotics, is based on the principle of providing a probiont
with a competitive advantage (a fermentable energy source) over
competing endogenous populations, by effectively improving the
survival and implantation of the live microbial dietary supplement
in the gastrointestinal tract of the host [9].
Although the use of pro- and prebiotics are now widely accepted
in aquaculture, only a few studies have focused on the effects of
synbiotics in farmed aquatic species. Nevertheless positive impacts
of synbiotics on growth, feed utilization, body composition, diges-
tive enzyme activity and immune responses have been reported
[29e34]. To the authors’ knowledge there is no information avail-
able regarding the combined effect of AXOS and probiotics. The
present study was undertaken to evaluate the potential effect of the
dietary administration of AXOS and selected endogenous pro-
biotics, each separately and combined, on growth parameters, gut
microbiota and immune responses of juvenile Siberian sturgeon
(Acipenser baerii). The candidate probiotics, two strains of Lacto-
coccus lactis ssp. lactis and one strain of Bacillus circulans, were
isolated from the hindgut of Siberian sturgeon [35].
2. Material and methods
2.1. Experimental setup
The experiment was conducted at the laboratory of Aquatic
Ecology, Evolution and Conservation (KU Leuven, Belgium), and
consisted of eight experimental diets with three replicates each
following a completely block design. It was run in 24 aquaria of 60 L
for 28 days using 36 fish per treatment.
The experimental treatments were as follows: Diet 1: basal diet
(control); Diet 2: basal diet þ 2% AXOS; Diet 3: basal diet þ L. lactis
spp. ST G81; Diet 4: basal diet þ L. lactis spp. ST G45; Diet 5: basal
767
diet þ B. circulans ST M53; Diet 6: basal diet þ L. lactis spp. ST
G81 þ 2% AXOS; Diet 7: basal diet þ L. lactis spp. ST G45 þ 2% AXOS;
Diet 8: basal diet þ B. circulans ST M53 þ 2% AXOS. The composition
of the basal feed and the experimental diets is given in Table S1.
2.1.1. Rearing
Juveniles of A. baerii (48.4  1.4 g) were obtained from Joosen
and Luyckx Aquabio (Turnhout, Belgium). The animals were accli-
matized for two weeks to the experimental conditions and basal
diet. Fish health status was verified by physical examination (excess
of mucous secretion, normal colouration, erosion of scales or fins,
skin, bulging of eyes and presence of cysts, spots or patches over
the body and gills) and behavioural signs (swimming and feeding
reflexes). Tanks of 60-L, supplied with water flow-through
(400 ml min1) and aeration were used. Water quality parame-
ters including dissolved oxygen (DO) and pH were measured daily,
whereas nitrate, nitrite and ammonia concentrations were moni-
tored biweekly. Nitrate, nitrite and ammonia concentrations were
determined using a portable Hach DR/2400 spectrophotometer.
Water temperature, DO, pH, ammonia and nitrate ranged from 20.8
to 21.4  C, 6.6e6.9 mg l1, 6.84e7.06 mg l1, 0.34e0.39 mg l1 and
9.12e10.53 mg l1, respectively. The lightedark cycle was fixed at
12 h light and 12 h dark.
2.1.2. Microorganisms
The potential probiotic bacteria Bacillus circulans ST M53
(Genbank accession number JQ765857), Lactococcus lactis subsp.
lactis ST G45 (Genbank accession number JQ765861) and L. lactis
subsp. lactis ST G81 (Genbank accession number JQ765865) were
previously isolated from the gut contents of healthy Siberian stur-
geon and identified as potential probiotics based on diverse in vitro
test [35]. More information concerning the identification, isolation
and selection methods can be found in Ref. Geraylou et al. [35].
Brain heart infusion (BHI) and Man Rogosa Sharpe broth (MRS)
were used to grow B. circulans ST M53 and L. lactic ssp. lactis strains
(24 h at 22  C), respectively. Cell density was calculated from OD600
values and correlated with colony forming unit (cfu) counts using
serial dilution and spread plating on BHI (B. circulans ST M53) or
MRS agar (L. lactis ssp. lactis strains). The bacteria were subse-
quently harvested by centrifugation at 1500 g for 15 min in sterile
phosphate-buffered saline (PBS). The quantified bacteria were
maintained at 4  C in a suspended form and were used for feed
preparation as required.
2.1.3. Diet preparation
AXOS-32-0.30 [average Degree of Polymerization (avDP) ¼ 32;
average Degree of Substitution (avDS) ¼ 0.30] was prepared as
described earlier [19,20] and kindly provided by Laboratory of Food
Chemistry and Biochemistry (KU Leuven, Belgium). The composi-
tion of the AXOS preparations is summarized in Table S1. Com-
mercial powder of grow-out feed for sturgeon (Joosen & Luyckx
Aquabio, Turnhout, Belgium) was taken as the basal diet for the
supplementation of the probiont, separately or in combination with
AXOS. The results of our previous study showed that incorporation
of 2% dry matter of AXOS-32-0.30 resulted in better performance in
juvenile Siberian sturgeon [28]. Thereafter in the current study,
AXOS-32-0.30 were supplemented at level of 3.2% (wt/wt) to the
basal diet, corresponding to 2% AXOS after correction for purity.
Based on literature [36,37], candidate probiotics were considered to
add to diet at final dose of 1  109 cfu g1. Briefly, commercial
sturgeon feed powder was mixed either with tap water and AXOS-
32-0.30 (diet 2) or bacteria suspension (diet 3, 4, 5) or AXOS-32-
0.30 and bacterial suspension (diet 6, 7, 8). Two percent binder
(carboxymethyl cellulose) was added to all diet. Cellulose was
replaced AXOS in control diet and diet 3, 4 and 5 (Table S2). The
768 Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775
mixture was minced and pelletized with a lab scale kitchen meat
grinder (Kenwood) to obtain 3 mm sized pellets. Feeds were dried
overnight at 50  C in a Heraeus oven with ventilation.
Viability of the candidate probiotics in feed was checked peri-
odically using the spread plate method. For four weeks, the fish
received a daily feeding rate of 3.0% body weight, using automatic
feeders. The total fish biomass in each aquarium was determined
every week and the daily ration adjusted accordingly.
2.2. Growth parameters
After four weeks, growth performance was assessed in terms of
weight gain (WG), Specific Growth Rate (SGR) and Feed Conversion
Ratio (FCR). The calculations were performed using the following
formulae: SGR ¼ 100 (ln Wt  ln W0)/t; FCR ¼ FO/(Wt  W0),
where Wt is the weight of the fish at day t, W0 is the initial weight
of the fish, t is the duration of feeding (in days), FO is the quantity of
feed offered. The survival (%) ¼ 100  survival number/initial
number and was calculated after four weeks.
2.3. Non-specific immunity parameters
2.3.1. Sample collection
At the end of trial, three apparently healthy fish (no obvious skin
lesions) from each replicate tank (9 fish per treatment) were
anesthetized with 120 mg l1 benzocaine. Blood was collected from
the caudal vein using 1.0 ml non-heparinized syringes. For serum
isolation, blood samples were left for 4 h at 4  C, centrifuged at
6000  g for 10 min and the supernatant retained as serum. Serum
of the three fish of each replicate was pooled, yielding three rep-
licates per treatment. Serum samples were stored at 80  C for
further analysis. The head kidneys were removed and pooled for
the three representative fish per replicate. Then macrophages iso-
lated as described by Secombes [38]. Cell viability was evaluated
using the trypan blue exclusion test and the cell density was
adjusted to 2  107 cells ml1.
2.3.2. Immunological assays
Phagocytic activity (PA) was assessed by incubating phagocytic
cells from the head kidney with congo red-stained yeast cells
overnight at 15  C. The PA was determined as the percentage of
phagocytic cells quantified from 200 cells observed under a mi-
croscope. Respiratory burst activity (RBA) of macrophages was
studied by monitoring their ability to reduce nitroblue tetrazolium
according to the method of Chung & Secombes [39]. In brief, the
wells of 96-microtitre plates were coated by 200 ml of cell sus-
pension of head kidney cells (2  107 cells ml1), which were
allowed to adhere for a 4e5 h. The non-adherent cells were washed
carefully and 100 ml of Nitroblue Tetrazolium (NBT) (2.0 mg ml1 in
Hank’s Buffered Salt Solution (HBSS) without phenol red, Sigma)
was added into each 96-microplate well. Subsequently, 100 ml of
zymosan solution (Sigma, 2.5 mg ml1 PBS) was added into each
well. The same volume of HBSS was added into control wells. The
cells were incubated for 90 min at room temperature. NBT solution
was then removed and the remaining cells were fixed for 5 min
using 200 ml 100% methanol. Cells were washed 3 times in 70%
methanol and allowed to dry. The reduced form of the NBT inside
the cells was solubilized by adding subsequently 120 ml of 2.0 N
KOH and 140 ml of Dimethyl Sulfoxide (DMSO). Finally, the optical
density was measured at 630 nm by a microplate reader (Bio-Rad
Benchmark plus microplate reader).
The total serum peroxidase activity (PO) was measured ac-
cording to the method of Quade and Roth [40]. We used 3, 30 , 5, 50 -
tetramethylbenzidine hydrochloride (TMB, Sigma) and hydrogen
peroxidase (H2O2, Sigma) as substrate for the peroxidase enzyme.
Fifteen ml of serum was diluted by 35 ml of Ca2þ and Mg2þ free HBSS
in flat bottomed 96-well plates. Lastly, 100 ml of solution of
peroxidase substrate was added. The colour-change reaction was
stopped after 3 min by adding 50 ml of 2.0 M sulphuric acid and the
optical density was read at 450 nm in a microplate reader (Bio-Rad
Benchmark plus microplate reader). Standard samples without
serum were analysed as control. The stimulation index was ob-
tained by dividing each sample value by its mean control value.
The alternative haemolytic complement activity was assayed
according to Yano [41] and based on haemolysis of sheep red blood
cells (SRBC). The washed SRBC were adjusted to 2  108 cells ml1
in ethylene glycol tetraacetic-magnesium-gelatin veronal buffer
(0.01M EGTA-MG-GVB). The diluted serum was adjusted in vol-
umes ranging from 0.1 to 0.25 ml using EGTA-MG-GVB and incu-
bated with 0.1 ml of SRBC suspension for 90 min at room
temperature (22  C) with occasional shaking. After centrifugation
at 1600  g at 4  C, absorbance of the supernatant was measured at
414 nm using a microplate reader (Bio-Rad Benchmark plus
microplate reader). The degree of haemolysis (Y) was estimated and
the lysis curve for each specimen was obtained by plotting Y/(1  Y)
against the volume of serum added (ml) on a logelog scaled graph.
The volume yielding 50% haemolysis was calculated and in turn
used for quantifying the complement activity of the sample (ACH50
value ¼ units ml1).
Total immunoglobulin (Ig) was measured by the method
described by Siwicki and Anderson [42]. Shortly, the plasma was
diluted 100 times with 0.85% NaCl and the Bradford method was
employed for determining the protein content. On the other hand,
an aliquot (0.1 ml) of each plasma sample was mixed with an equal
volume of 12.0% solution of polyethylene glycol (MW 10,000), and
incubated for 120 min. This precipitated the Ig molecules, which
were removed upon centrifuging at 5000 g at 4  C. The supernatant
was diluted 50 times with 0.85% NaCl and the protein content was
determined as mentioned above. The difference between the pro-
tein values of each untreated and polyethylene glycol treated
sample corresponds to the total Ig content and is expressed as
mg ml1.
2.4. Bacterial community analysis
2.4.1. Hindgut microflora sampling
A preliminary investigation of different sections of Siberian
sturgeon’s gastrointestinal tract using plating on BHIA revealed that
density and diversity of cultivable bacteria in the hindgut (spiral
vesicle) was significantly higher in comparison to the midgut and
caecum. Therefore the sampling was only performed from the
hindgut of the fish. The microflora characterization was performed
on the hindgut of the fish which were sampled for the immune
assays. After disinfection of the fish in a 0.1% benzalkonium chloride
solution, the hindgut was dissected using sterilized surgical scissors
and the content was frozen at 80  C for further DNA analysis.
2.4.2. DNA extraction and pyrosequencing
Total DNA was extracted from hindgut content using the
NucleoSpin 1 Tissue kit (Macherey-Nagel) according to the manu-
facturer’s protocol. For 454 pyrosequencing, an amplicon library
was prepared using eubacterial universal primers. Given the mean
sequencing length of 400e450 bp for the 454 GS FLX Titanium
platform. The 16S rDNA gene was amplified from extracted DNA
using the composite forward primer 50 -CGTATCGCCTCCCTCGCGC-
CATCAG MID ACTCCTACGGGAGGCAGCAGT-30 where the underlined
sequence is the 454 Life SciencesÒ primer A-key, MID is a 10 bp
barcode specific for each replicate treatment, and in italics is the
broad range bacterial primer E338F. The reverse primer was 50 -
CTATGCGCCTTGCCAGCCCGCTCAGGGGTATCTAATCCTG-0 3, where the
 Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775 769

underlined sequence is the 454 Life SciencesÒ primer B-key and in
italics is the broad range bacterial primer E797R. As sequencing was
done in only the forward direction, no barcode was necessary
within the reverse primer. Reaction conditions were as follows:
2.5 ml 10  PCR buffer II (Eurogentec), 1 ml MgCl2 (50 mM, Euro-
gentec), 2.5 ml deoxyribonucleoside triphosphates (2 mM), 1.0 ml
forward primer and 1.0 ml reverse primer (20 pmol/ml each) and
0.2 ml Silverstar Taq DNA polymerase (2.5 U, Eurogentec) and 50 ng
template DNA in a total reaction volume of 25 ml. The PCR condi-
tions were as follows: 5 min denaturing at 95  C followed by 30
cycles of 1 min at 95  C (denaturing), 1 min at 55  C (annealing) and
1 min at 72  C (elongation), with a final extension at 72  C for 5 min.
PCR products of three samples of the same treatment were pooled
and purified using Nucleofast (Macherey-Nagel, Germany). The 24
amplicon libraries were checked by agarose gel electrophoresis,
their concentration was estimated using Quant-IT PicoGreen
(Invitrogen), and they were then pooled to an equimolar concen-
tration. Pyrosequencing was performed with a Roche GS-FLX Ti-
tanium sequencer by Biogenomics at the Genomics Core Facility of
KU Leuven (Belgium).
2.4.3. Sequence processing
The sequences generated from pyrosequencing were mainly
analysed with the software PANGEA for identification of opera-
tional taxonomic unit (OTU), taxonomic assignment and commu-
nity comparison [43]. Raw 454 FLX data were parsed using the
trim2.pl script from the PANGEA to only keep the reads that passed
quality control. They include a minimum sequence length of
100 bp, minimum phred quality score of 20 and presence of a
correct barcode at the beginning of the sequence. The trimmed file
was splitted into different files based of their barcode using a perl
script named barcode.pl. The sequences were phylogenetically
classified using a standalone BLAST against a modified bacterial
RDP-II database prepared using Tax Collector (http://www.
microgator.org), which attaches complete taxonomic information
from domain to species to each sequence in the database and can be
obtained from http://www.microgator.org/. Sequences classified by
Megablast were clustered into OTUs based on the relatedness of
classification. The sequences not classified by Megablast were
captured using a perl script called unclassified_selector.pl and then
submitted to CD-HIT (Cluster Database at High Identity with Toler-
ance) to be clustered into OTUs based on the relatedness of the
sequences [43]. To assess the bacterial diversity in the DNA samples
in a comparable manner, the number of reads was normalized to
the same number of reads in each sample. This was done by
identifying the sample with the smallest number of reads and
randomly selecting this same number of sequences for each of the
samples.
2.5. Statistical analyses
Diet response in terms of growth performance and immune
responses were analysed by two-way analysis of variance (ANOVA),
using AXOS diet and probiotics as factors. When the differences
were significant at p < 0.05 level, Tukey’s Multiple Comparison test
was utilized to compare the mean values among the treatments
due to main effects. Statistical analyses were conducted using
STATISTICA (version 11). To analyse the overall effect of treatment
on gut microbiota, non-zero relative abundance OTUs were
assessed at phylum, class, family, genus and species level. It was
first determined whether the relative abundances were signifi-
cantly different from zero using a t-test, after which the effect of
treatment on relative abundance was assessed by two-way ANOVA
in STATISTICA (version 11) for non-zero relative abundances.
Relative abundances were log2 transformed and a threshold of
p < 0.05 was used to determine statistical significance. Taxonomic
richness was calculated by summing the number of OTUs per
replicate, including singlet OTUs (OTUs with only a single obser-
vation in a single replicate). Shannon diversity was calculated
P
following H0 ¼  pi ln (pi) where pi is the proportion of taxon i
using the script Shannon.pl of the PANGEA pipeline. A 99% similarity
level (species level) value was used for assigning an OTU.
3. Results
3.1. Growth
Growth parameters are summarized in Table 1. A two-way
ANOVA showed that final weight, SGR and FCR were positively
affected by AXOS treatments (p < 0.05). However, there was no
effect of the probiotics independently, and no interaction was found
between the candidate probiotics and AXOS group treatments
(p > 0.05). The WG and SGR significantly increased in fish fed diet 7
(basal feed þ L. lactis spp. lactis ST G45 þ AXOS) in comparison to
control fish. Moreover FCR decreased in fish receiving diet 7. The
survival was high in all groups (100%).
3.2. Non-specific immunity parameters
PA was significantly affected by dietary AXOS and candidate
probiotics separately (p < 0.05, Table 2). Moreover, a synergistic
effect between biotics was observed. (p ¼ 0.05). PA was significantly
higher in fish receiving dietary AXOS, L. lactis spp. lactis ST G45 and
a combination of L. lactis spp. lactis ST G45 and AXOS, compared to
fish fed the control diet (p < 0.05, Table 2). RBA was significantly
increased by dietary AXOS (p < 0.05), however, no significant ef-
fects of probiotics dietary (p ¼ 0.08) or synergistic effects of AXOS

Table 1
Growth performance, feed conversion ratio and survival of the Siberian sturgeon fed different experimental diets.

Diet Initial weight (g) Final weight (g) Weight gain (g) SGR (% day1) FCR Survival (%)
a ab a
Diet 1 (Basal feed) 50.7  0.4 87.8  5.5 37.0  5.1 1.95  0.15a 1.31  0.25a 100
Diet 2 (Basal feed þ AXOS) 47.8  2.8a 95.0  4.9ab 47.1  7.0ab 2.45  0.27ab 1.19  0.35ab 100
Diet 3 (Basal feed þ L. lactis spp. lactis ST G81) 48.0  1.2a 87.75  5.2ab 39.6  4.9ab 2.14  0.15ab 1.18  0.01ab 100
Diet 4 (Basal feed þ L. lactis spp. lactis ST G45) 48.0  2.5a 89.1  1.1ab 41.1  3.4ab 2.21  0.17ab 1.06  0.04ab 100
Diet 5 (Basal feed þ B. circulans ST M53) 49.5  0.6a 91.3  2.7ab 41.7  2.6ab 2.18  0.08ab 1.24  0.31ab 100
Diet 6 (Basal feed þ L. lactis spp. lactis ST G81 þ AXOS) 46.0  4.3a 85.0  4.8a 39.0  5.3ab 2.19  0.11ab 1.17  0.15ab 100
Diet7 (Basal feed þ L. lactis spp. lactis ST G45 þ AXOS) 49.0  1.3a 105.9  5.1b 56.8  4.7b 2.74  0.11b 0.92  0.01b 100
Diet 8 (Basal feed þ B. circulans ST M53 þ AXOS) 48.1  1.3a 91.2  14.3ab 43.0  13.4ab 2.24  0.38ab 1.14  0.22ab 100
Two-way ANOVA
p-Value
Probiotics 0.35 0.11 0.12 0.19 0.20
AXOS 0.15 0.09 0.03 0.02 0.04
Interaction probiotics * AXOS 0.46 0.13 0.15 0.25 0.29

Data are means of triplicate. Means in the same column sharing a same superscript letter are not significantly different (p < 0.05) (by Two-way ANOVA). SGR: specific growth
rate, FCR: feed conversion Ratio.
770 Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775

Table 2
Immune responses of juvenile Siberian sturgeon fed different experimental diets for 4 weeks.

Diet Phagocytic Respiratory burst Alternative complement Serum peroxidase Total

activity (%) activity activity (ACH 50 U mll) (stimulation index) immunoglobulin
(O.D. at 630 nm) (mg ml-1)

Diet 1(Basal feed) 35.0  1.55a 0.143  0.02a 0.39  0.05a 1.24  0.09a 29.3  3.2a
Diet 2 (Basal feed þ AXOS) 48.6  3.01b 0.258  0.02b 0.56  0.04ab 1.47  0.11ab 34.0  4.3a
Diet 3 (Basal feed þ L. lactis spp. lactis ST G81) 36.0  2.35ab 0.174  0.01ab 0.42  0.08a 1.33  0.13a 33.3  9.4a
Diet 4 (Basal feed þ L. lactis spp. lactis ST G45) 43.0  4.24b 0.221  0.03ab 0.41  0.11a 1.47  0.09ab 32.6  5.4a
Diet 5 (Basal feed þ B. circulans ST M53) 41.3  0.89ab 0.146  0.01a 0.41  0.07a 1.29  0.12a 28.0  5.4a
Diet 6 (Basal feed þ L. lactis spp. lactis ST G81 þ AXOS) 41.0  3.0ab 0.210  0.04ab 0.45  0.03ab 1.40  0.04ab 31.3  6.9a
Diet 7 (Basal feed þ L. lactis spp. lactis ST G45 þ AXOS) 50.0  2.94b 0.275  0.01b 0.61  0.03b 1.98  0.28b 30.6  5.6a
Diet 8 (Basal feed þ B. circulans ST M53 þ AXOS) 42.6  1.79ab 0.210  0.02ab 0.42  0.02a 1.31  0.27a 31.6  6.8a
Two-way ANOVA
p-Value
Candidate probiotics 0.01 0.08 0.06 0.01 0.53
AXOS 0.01 0.00 0.00 0.05 0.12
Interaction probiotics * AXOS 0.05 0.48 0.03 0.22 0.32

O.D. ¼ optical density.

Values are presented as mean  standard error (n ¼ 9).
Values with a different superscript in the same column are significantly different (p < 0.05).

and candidate probiotics were observed (p > 0.05, Table 2). RBA
was significantly higher in fish fed L. lactis spp. lactis ST G45 þ AXOS
(diet 7) or AXOS (diet 2) than in the control and in fish receiving
only B. circulans ST M53. Among the studied humoral parameters,
alternative complement activity was significantly boosted by a
synergistic effect of AXOS and the candidate probiotics as well as by
AXOS or candidate probiotics dietary separately. Alternative com-
plement activity was significantly higher in the serum of fish
receiving diet 7 (basal feed þ L. lactis spp. lactis ST G45 þ AXOS)
than other treatments, with the exception of fish fed AXOS (diet 2)
and diet 6 (basal feed þ L. lactis spp. lactis ST G81 þ AXOS) (Table 2).
Serum peroxidase content was positively affected by both AXOS
and probiotic diets. It was significantly higher in fish fed diet 7
(basal feed þ L. lactis spp. lactis ST G45 þ AXOS) than fish fed control
diet (diet 1), diet 3 (Basal feed þ L. lactis spp. lactis ST G81), diet 5
(Basal feed þ B. circulans ST M53) and diet 8 (Basal
feed þ B. circulans ST M53 þ AXOS). The total immunoglobulin
content was not affected by dietary treatment (Table 2).
3.3. Hindgut microbiota analyses
(basal feed þ L. lactis ssp. lactis ST G45 þ AXOS), in comparison to
fish fed the control diet (Table 4). The abundance of Fusobacteria
was affected by synergistic effects between AXOS and probiotics, so
that their abundance decreased in the hindgut of fish fed diet 7
(basal feed þ L. lactis ssp. lactis ST G45 þ AXOS) in comparison to
control fish. No significant changes in relative abundance of Pro-
teobacteria were observed (Table 4).
Among the Firmicutes, the classes Bacilli and Clostridia had the
highest relative abundances in all samples. The dietary biotic sup-
plementation significantly affected the relative abundance of bacilli
(p < 0.05, Table 4). Relative abundances of Bacilli were significantly
higher in the hindgut of fish fed a diet supplemented with L. lactis
spp. lactis ST G45 þ AXOS (diet 7) than in other treatments.
Moreover, Bacilli significantly increased in fish fed a L. lactis spp.
lactis ST G45 (diet 4) in comparison to the fish receiving the control
diet, Diet 2 (AXOS), diet 5 (basal feed þ B. circulans ST M53) or diet 8
(basal feed þ B. circulans ST M53 þ AXOS). Within the Bacilli, the
most striking difference between treatments was due to Lacto-
coccus, which had a significantly higher relative abundance in the
hindgut of fish fed diet 4 (basal feed þ L. lactis spp. lactis ST G45)

A total of 44,697 high quality pyrosequencing reads were ob-

tained after quality control processes, covering 4087 OTUs at 99% Table 3
similarity level. The number of sequences per sample varied be- Diversity indices of the hindgut microbiota of Siberian sturgeon fed different
experimental diets for 4 weeks.
tween 561 and 3174. The number of OTUs per sample averaged
between 106 and 410 at species level (99% similarity level). Experimental diet Richness Shannon
Two-way ANOVA results showed that both bacterial richness index

and Shannon index were affected by candidate probiotics dietary or Diet 1 (Basal feed) 140.3  3.84a 3.22  0.16a
AXOS, while no synergistic effects of candidate probiotics and AXOS Diet 2 (Basal feed þ AXOS) 117.0  9.4ab 2.86  0.25ab
Diet 3 (Basal feed þ L. lactis ssp. 99.0  8.05ab 2.05  0.34ab
on bacterial diversity were observed (Table 3). Bacterial richness
lactis ST G81)
was significantly reduced in fish fed diet 7 (basal diet þ L. lactis spp. Diet 4 (Basal feed þ L. lactis ssp. lactis ST G45) 82.3  14.01ab 2.34  0.31ab
lactis ST G45 þ AXOS) and diet 8 (basal diet þ B. circulans ST Diet 5 (Basal feed þ B. circulans ST M53) 105.3  21.2ab 2.68  0.51ab
M53 þ AXOS) in comparison to fish fed the control diet (diet 1). Diet 6 (Basal feed þ L. lactis ssp. lactis ST 126.0  6.27ab 3.06  0.21a
Shannon diversity was significantly lower in fish fed diet 7 (basal G81 þ AXOS)
Diet 7 (Basal feed þ L. lactis ssp. lactis ST 60.0  9.95b 1.28  0.26b
diet þ L. lactis spp. lactis ST G45 þ AXOS) than in fish fed control diet G45 þ AXOS)
and diet 6 (basal diet þ L. lactis spp. lactis ST G81 þ AXOS). Changes Diet 8 (Basal feed þ B. circulans 71.0  10.7b 2.00  0.21ab
in the relative abundance of the hindgut microflora were analysed ST M53) þ AXOS
at phylum, class, genus, and species level. At the phylum level, Two-way ANOVA
p-Value
Firmicutes, Proteobacteria and Fusobacteria were the most abun-
Candidate probiotics 0.00 0.01
dant. Two-way ANOVA showed that the relative abundance of AXOS 0.04 0.01
Firmicutes was significantly affected by the probiotic diets Interaction probiotics * AXOS 0.24 0.14
(Table 4), while the effect of AXOS on the abundance of Firmicutes Values are presented as mean  standard error (n ¼ 9).
was almost significant (p ¼ 0.08). The relative abundance of Fir- Values with a different superscript in the same column are significantly different
micutes significantly increased in fish fed diet 2 (AXOS) and diet 7 (p < 0.05).
Table 4
Relative abundances (% of sequences per treatment)  standard deviation of the most abundant bacteria at different taxonomy levels, in the hindgut of juvenile Siberian sturgeon fed with different experimental diets.

Diet 1 Diet 2 Diet 3 Diet 4 Diet 5 Diet 6 Diet 7 Diet 8 Two way-ANOVA

Probiotic AXOS Probiotics*AXOS

a b ab ab ab ab b ab
Phyllum Firmicutes 16.8  6.31 53.0  15.2 46.9  18.9 51.8  27.6 53.2  34.0 40.2  12.2 96.8  3.0 78.5  21.7 0.05 0.08 0.18
Proteobacteria 5.64  4.00 28.6  9.60 35.0  12.6 5.33  2.11 10.0  4.53 33.5  13.7 1.22  0.60 6.31  4.90 0.06 0.10 0.44

Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775

Fusobacteria 70.4  18.5a 15.2  9.21ab 10.0  0.24ab 41.7  14.2ab 34.9  12.4ab 22.8  10.6ab 1.35  1.0b 18.6  12.5ab 0.12 0.84 0.02
Class Bacilli 1.81  0.08a 2.62  1.57a 11.8  4.51ab 24.7  2.70b 2.76  1.71a 13.5  9.44ab 82.4  5.56c 1.09  0.30a 0.00 0.00 0.00
Clostridia 3.12  0.40 15.4  3.21 10.0  0.07 14.3  7.54 24.0  12.6 3.66  2.44 1.80  0.66 10.5  4.46 0.14 0.24 0.36
Fusobacteriia 69.7  18.5a 16.5  9.05ab 10.1  0.25ab 40.6  14.1ab 35.8  18.2ab 23.6  10.5ab 1.74  0.99b 20.7  13.9ab 0.12 0.84 0.02
Family Bacillaceae 0.65  0.21 0.64  0.42 1.02  0.34 0.30  0.12 1.41  0.81 1.26  0.94 0.90  0.44 0.45  0.37 0.74 0.95 0.67
Clostridiaceae 3.90  2.71a 46.6  24.2ab 39.4  18.3ab 21.8  16.8ab 36.2  16.9ab 28.4  24.2ab 11.4  6.20ab 71.3  24.5b 0.04 0.86 0.22
Lactobacillaceae 0.18  0.00a 4.92  1.22b 1.26  0.64a 1.26  0.59ac 0.62  0.3ac 4.68  2.11b 7.08  1.07b 0.00  0.00c 0.00 0.00 0.02
Streptococcaceae 0.09  0.02a 5.16  0.58a 6.36  1.60a 24.2  2.41c 0.31  0.20a 9.00  3.03a 81.4  6.00b 0.18  0.10a 0.00 0.00 0.00
Eubacteriaceae 5.13  1.47 4.56  1.64 2.10  0.50 5.64  1.67 13.3  8.27 0.44  0.34 1.08  0.73 6.60  5.90 0.16 0.23 0.85
Fusobacteriaceae 69.8  18.8a 16.4  9.17ab 10.45  0.58ab 40.6  13.3ab 35.2  19.5ab 23.2  10.0ab 1.74  1.00b 14.0  7.91ab 0.12 0.84 0.02
Rhodobacteraceae 0.97  0.41a 0.36  0.78a 0.65  0.30a 0.30  0.16a 2.58  1.26ab 15.9  6.40b 0.30  0.22a 3.42  1.77a 0.03 0.05 0.04
Enterobacteriaceae 29.9  10.2a 25.5  11.23a 2.22  0.78a 2.40  1.19a 4.80  3.68a 4.80  2.34a 0.30  0.22b 1.32  0.82a 0.01 0.56 0.99
Genus Lactococcus 0.18  0.02a 3.22  0.41a 6.48  3.65a 24.1  2.51b 0.10  0.07a 8.11  3.11a 80.2  6.20b 0.19  0.20a 0.00 0.00 0.05
Lactobacillus 0.18  0.00a 3.69  0.07ab 1.14  0.06a 1.08  0.56a 0.48  0.23a 4.32  0.25ab 6.07  1.02b 0.00  0.00a 0.00 0.00 0.02
Candidatus Arthromitus 1.08  0.63a 27.9  10.9ab 40.7  17.2ab 12.9  6.26ab 18.7  4.73ab 28.6  19.6ab 10.8  5.47ab 54.0  14.5b 0.05 0.97 0.08
Clostridium 1.35  0.67 0.54  0.29 0.24  0.08 0.30  0.09 0.72  0.19 0.12  0.09 1.22  0.70 0.90  0.20 0.21 0.23 0.82
Eubacterium 0.42  0.10 4.38  1.33 4.71  1.12 5.04  1.67 13.2  10.7 2.04  0.47 1.01  0.75 6.65  3.77 0.15 0.26 0.85
Bacillus 0.18  0.07 0.12  0.08 0.42  0.06 0.06  0.01 0.72  0.23 0.42  0.21 0.42  0.08 0.36  0.13 0.51 0.83 0.35
Rhodobacter 0.70  0.30ab 0.18  0.09a 0.54  0.41a 0.18  0.13 1.26  0.70 12.6  7.89b 0.18  0.10a 2.70  2.0ab 0.03 0.09 0.02
Cetobacterium 69.9  0.27a 16.3  9.06ab 9.7  4.4ab 41.7  14.3ab 34.3  18.2ab 22.2  8.10ab 1.27  0.97b 13.6  6.8ab 0.13 0.84 0.02
Species Lactococcus lactis 0.18  0.00a 2.33  0.75a 6.90  1.06c 14.0  6.62b 0.00  0.00a 5.70  3.15c 74.2  5.84b 0.00  0.00a 0.00 0.00 0.03
Lactobacillus aviarius 0.18  0.00a 4.26  1.36b 0.90  0.60a 0.18  0.00a 0.24  0.15a 3.84  1.15ab 6.24  1.25b 0.00  0.00a 0.01 0.00 0.01
Cetobacterium somerae 62.3  21.9a 14.2  8.11ab 9.09  0.29ab 36.0  12.5ab 31.5  15.9ab 20.36  12.1ab 1.56  0.95b 12.9  9.20ab 0.13 0.86 0.02
Rhodobacter sp. 0.63  0.35ab 0.18  0.10a 0.48  0.12ab 0.12  0.04a 1.20  0.66ab 9.24  2.16b 0.18  0.10a 2.34  0.98ab 0.04 0.10 0.06

Values with a different superscript in the same row are significantly different (p < 0.05).
Diet 1: Basal feed, Diet 2: Basal feed þ AXOS, Diet 3: Basal feed þ L. lactis spp. lactis ST G81, Diet 4: Basal feed þ L. lactis spp. lactis ST G45, Diet 5: Basal feed þ B. circulans ST M53, Diet 6: Basal feed þ L. lactis spp. lactis ST G81 þ
AXOS, Diet 7: Basal feed þ L. lactis spp. lactis ST G45 þ AXOS, Diet 8: Basal feed þ B. circulans ST M53.

771
772 Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775
and diet 7 (basal feed þ L. lactis spp. lactis ST G45 þ AXOS) (Table 4).
Lactobacillus, another dominant genus of the class of Bacilli, was
significantly affected by biotic diets. It was significantly higher in
fish fed diet 7 (basal feed þ L. lactis spp. lactis ST G45 þ AXOS) than
in fish receiving other diets, with the exception of diet 2 (AXOS) and
6 (basal feed þ L. lactis spp. lactis ST G81 þ AXOS). Within the class
Clostridia, relative abundances of Candidatus Arthromitus, the most
abundant genus of Clostridia in the hindgut of sturgeon fish, were
stimulated by probiotics. Abundances of Candidatus Arthromitus
were significantly higher in the hindgut of fish fed B. circulans ST
M53 þ AXOS than of fish fed the control diet (Table 4). Eubacterium
and Clostridium were the second and third most abundant genera
from the class of Clostridia, but no significant effects of biotics was
observed on their abundances. Within the phylum Fusobacteria,
the class of Fusobacteria and its most abundant genus, Cetobacte-
rium, had significantly higher abundances in fish fed the control
diet than in fish fed diet 7 (basal feed þ L. lactis spp. lactis ST
G45 þ AXOS).
At the species level, the abundance of L. lactis ssp. lactis was
affected by biotic supplementation. L. lactis spp. lactis significantly
increased in fish receiving the diets supplemented with L. lactis spp.
lactis (diets 3, 4, 6 and 7) (p < 0.05, Table 4). L. lactis composed on
average 6.9% and 5.7% or 14% and 74.2% of reads in the hindgut
samples of fish fed diets supplemented with L. lactis spp. lactis ST
G81 or L. lactis spp. lactis ST G45, respectively. No sequences of
L. lactis spp. lactis were detected in fish fed B. circulans ST M53 or
the control diet. In contrast, biotic supplementation of B. circulans
ST M53 (diet 5 and 8) did not affect the relative abundance of Ba-
cillus spp. in the hindgut of sturgeon and its relative abundance was
lower than 1%.
Interaction effects between AXOS and candidate probiotics
significantly affected the relative number of Lactobacillus aviarius,
Cetobacterium somerae and Rhodobacter sp. L. aviarius significantly
increased in fish fed diet 2 (AXOS) or diet 7 (basal feed þ L. lactis
spp. lactis ST G45 þ AXOS), in comparison to fish fed probiotic di-
etaries (diet 1, 3, 4, 5) and diet 8 (basal feed þ B. circulans ST
M53 þ AXOS). L. aviarius was not detected in fish fed diet 8. Relative
abundances of C. somerae were significantly higher in the hindgut
of fish fed the control diet than in fish fed diet 7 (basal
feed þ L. lactis spp. lactis ST G45 þ AXOS). The abundance of Rho-
dobacter sp. significantly increased in fish fed a diet supplemented
with L. lactis spp. lactis ST G81 þ AXOS in comparison to the other
treatments, with the exception of fish fed control diet (diet 1), diet 5
(basal feed þ B. circulans ST M53) or diet 8 (basal feed þ B. circulans
ST M53 þ AXOS) (Table 4).
4. Discussion
The current study demonstrates the benefits of supplementing
sturgeon feed with AXOS-32-0.30 þ L. lactis spp. lactis ST G45 on
the growth performance and feed utilization of juvenile Siberian
sturgeon. Information about the combined use of L. lactis and AXOS
were not yet available, but it is already known that the separate
supplementation of L. lactis or AXOS is able to improve growth
parameters in several organisms.
Stimulation of growth by L. lactis has been previously reported
in several aquatic animals and in teleost fish, including the rotifer
Brachionus plicatilis [44], Nile tilapia (Oreochromis niloticus) [45]
and Catla catla [46]. Bandyopadhyay et al. [46] showed that sup-
plementation of 2  105 cfu of L. lactis per g feed induced a
better growth and a significantly lower feed conversion ratio in C.
catla. However, in the present study, no influence of single sup-
plementation of L. lactis on growth parameters of Siberian sturgeon
was observed.
Positive growth effects of B. circulans have been reported for
Rohu (Labeo rohita) fingerlings [47], but this was not confirmed for
the strain used in the underlying study.
Growth-promoting effects of dietary arabinoxylans (AX) and
their hydrolysis products, such as xylooligosaccharides (XOS) and
AXOS, are less studied in fish in comparison to other animals. Xu
et al. [48] indicated that XOS supplementation promoted growth
performance of Crucian carp (Carassius auratus gibelio). In our
previous study, sturgeon juveniles fed 2% AXOS-32-0.30 had a
higher specific growth rate and weight gain, but not significantly
due to a large internal variation [28]. Likewise, in the present study,
separate administration of AXOS did not significantly affect the
growth of Siberian sturgeon. Moreover, no synergistic effect be-
tween AXOS and L. lactis observed. The significant improvement in
fish growth performance when simultaneously adding L. lactis spp.
lactis ST G45 and AXOS can be explained as a cumulative outcome
of the improvements observed when L. lactis spp. lactis ST G45 and
AXOS were applied independently.
Improvement of growth performance in fish fed a synbiotic diet
might be attributed to an elevated health status, an increased di-
gestibility of the prebiotic, or the improvement of survival and
colonization of the probiotic [29e34] compared to the separate
pre- or probiotic diets. It has been demonstrated that L. lactis can
provide disease protection by activating cellular and humoral im-
mune defenses in several aquatic animals including turbot (Psetta
maxima) [49], tilapia (O. niloticus) [45] and Olive flounder (Para-
lichthys olivaceus) [50]. Bacillus spp. have been successfully used as
probiotics in aquatic animals and their immunodulatory effects
have been proven in several studies, including the improvement of
the phagocytic ratio and phagocytic index of C. catla [46], and the
increase of resistance against salinity, temperature, ammonia and
pH stress in larvae of Acipenser percicus [51].
In the present study, administration of L. lactis ssp. lactis ST G45 in
feed of Siberian sturgeon significantly increased phagocytic activity,
respiratory burst activity, alternative complements activity and
serum peroxidase content, while no effects of L. lactis ssp lactis ST
G81 and B. circulans ST M53 on immune responses were observed.
Research on the effects of arabinoxylan (AX) and their hydrolysis
products on fish immune responses are scarce [52]. In a previous
study on juvenile Siberian sturgeon, AXOS augmented phagocytic
activity, alternative complement activity and total serum peroxidase
content [28]. Similarly, the immunodulatory effect of AXOS was
again proven in the current study. This study demonstrated that the
combined use of AXOS and L. lactis ssp. lactis ST G45, positively
affected cellular immunity as well as humoral immunity of juvenile
Siberian sturgeon. Such prebiotic induced immuno-modulation can
be due to selective increase/decrease in the abundances of specific
intestinal bacteria, interactions with carbohydrate receptors on in-
testinal epithelial cells and immune cells, or partial absorption
resulting in systemic contact with the immune system [53]. The
exact process by which AXOS and L. lactis act on the immune system
of Siberian sturgeon is however unclear and needs further study.
Because both pre- and probiotics potentially act through the
modulation of the intestinal microbial community, the hindgut
microflora was assessed by pyrosequencing. In addition, this allows
confirming colonization by the candidate probiotics in the hindgut
of the juvenile Siberian sturgeon. The inclusion of the combination
of L. lactis ssp. lactis ST G45 and AXOS in sturgeon feed resulted in a
significant reduction in the bacterial diversity in terms of specific
richness and Shannon index in comparison to the control diet.
Moreover, the combined use of B. circulans ST M53 and AXOS in
sturgeon feed resulted in a significant decrease in bacterial rich-
ness. A similar reduction in microfloral richness has already been
shown after dietary administration of carbohydrates in Arctic char
(Salvelinus alpinus) [54], a combination administration of dietary
 Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775
probiotics and microalgae in gilthead seabream (Sparus aurata)
[34], and inclusion of inulin in feed of Atlantic salmon (Salmo salar)
[55]. In contrast, Najdegerami et al. [56] indicated that a well-
balanced diet with poly-b-hydroxybutyrate (PHB) increased the
bacterial species richness in juvenile Siberian sturgeon and that
there was a relationship between a high growth rate, low mortality
and bacterial diversity in the GI tract of fish fed 2% PHB. The
reduction in bacterial diversity obtained in this study could be
related to the antimicrobial activity of L. lactis [49,57,58] or to a
competitive exclusion by L. lactis on the hindgut microflora [59]. L.
lactis ssp. lactis ST G45 and ST G81 indeed effectively inhibit the
growth of Aeromonas hydrophila, A. salmonicida, Yersinia ruckeri,
Flavobacterium columnare and Vibrio anguillarum in vitro [35],
showing the high antimicrobial capacities of these both strains.
Based on pyrosequencing analysis of 16S rDNA, L. lactis abun-
dance levels were significantly higher in juveniles fed L. lactis spp.
lactis ST G45, L. lactis spp. lactis ST G45 þ AXOS, L. lactis spp. lactis
ST G81, or L. lactis spp. lactis ST G81 þ AXOS. L. lactis was not
detected in the control group, neither in the group of fish fed
B. circulans ST M53. Strong synergism between L. lactis strains and
AXOS was also observed, since higher abundances of L. lactis were
present in the hindgut when the synbiotic was administrated,
which is in line with our previous observation that AXOS-32-0.30
stimulates the growth of L. lactis in Siberian sturgeon’s hindgut
[35]. Our previous in vitro results revealed no significant difference
in adhesion to and survival in sturgeon’s gut mucus between
L. lactis spp. lactis ST G45 and L. lactis spp. lactis ST G81. However,
results of the pyrosequencing analysis showed that L. lactis
composed 6.9% and 5.7% of sequences in fish fed L. lactis spp. lactis
ST G81 and L. lactis spp. lactis ST G81 þ AXOS, respectively, while
those value rose 14% and 74.2% for the treatments with L. lactis
spp. lactis ST G45 and L. lactis spp. lactis ST G45 þ AXOS, respec-
tively. It may be that the adherence capacity of L. lactis is affected
by environmental factors such as gastrointestinal acids and
competition for surface with other bacteria. Villamil et al. [49]
reported a difference in the adhesion capacity of L. lactis to gut
mucus of turbot (Psetta maxima) under in vivo versus in vitro
conditions. They found that even though L. lactis is able to adhere
in vitro to turbot intestinal mucus without specificity problems,
in vivo adherence and persistence in the intestinal mucus is more
variable. Structural differences in the cell wall compositions of
different Lactic Acid Bacteria (LAB) strains can be responsible for
efficacy differences [60].
In contrast to the successful colonization of hindgut by L. lactis,
the relative abundance of B. circulans was lower than 1% in fish fed
B. circulans ST M53 or B. circulans ST M53 þ AXOS. Dietary
B. circulans ST M53 did not affect the abundance of Bacillus spp. in
the hindgut of sturgeon. Although we previously showed that
B. circulans ST M53 is able to colonize and survive GI conditions
under in vitro conditions [35], this was not confirmed by the present
in vivo study. Ouwehand & Salminen [61] suggested that since many
factors interfere with the mucosal adhesion of probiotics, it is
difficult to extrapolate in vitro adhesion results reliably to the in vivo
situation in humans. It has to be pointed out that viability assess-
ments of B. circulans ST M53 and the two L. lactis candidate pro-
bionts, added to the feed and then stored at 4  C for up to four weeks,
showed an overall maintenance of their initial inoculation level.
In addition to the detection of candidate probiotics in hindgut
samples, we also investigated how the biotics affect hindgut
micoflora composition at different taxonomic levels (Table 4). The
relative abundance of Firmicutes bacteria was increased by pro-
biotic dietary. The increase of Firmicutes abundance in fish fed
L. lactis spp. lactis ST G45 þ AXOS can therefore be explained as a
cumulative effect of the interaction of AXOS and L. lactis spp. lactis
ST G45, as two way ANOVA did not show any significant synergism
773
between the prebiotic and probiotic bacteria (p ¼ 0.18). Lactic acid
bacteria, mostly L. lactis, composed the main Firmicutes bacteria in
the hindgut of juvenile Siberian sturgeon, which can be related to
colonization by the candidate probiotic L. lactis or/and growth
stimulation of L. lactis by the AXOS diet. The abundance of L. aviarius
was affected by the biotic dietaries. In vitro stimulation of growth of
Lactobacillus strains by different preparations of AXOS has been
proven in humans [24,62] and hindguts of Siberian sturgeon
[28,63]. Candidatus Arthromitus was another genus of the Firmi-
cutes and it was significantly affected by probiotic dietaries. Can-
didatus Arthromitus increased up to 54% in fish fed a B. circulans ST
M53 þ AXOS, however, at the species level (99% similarity) no
identifiable Candidatus Arthromitus sequences were retrieved.
Candidatus Arthromitus are host-specific intestinal symbionts that
comprise a distinct clade within the Clostridiaceae and are known
to (potentially) induce a multifaceted immune response. Although
an enteritic syndrome affecting farmed rainbow trout (rainbow
trout gastroenteritis) has been related to the accumulation of
Candidatus Arthromitus in the digestive tract of fish [64], some
papers report the non-pathogenic character of this bacterial genus.
To our knowledge, this is the first report of the presence of Candi-
datus Arthromitus in Sturgeon.
Fusobacteria had the highest relative abundance in fish fed the
control diet and significantly decreased in fish fed diet 7 (basal
feed þ L. lactis spp. lactis ST G45 þ AXOS). Combined with the
opposite pattern in abundance of L. lactis in fish fed these diets, we
hypothesize that these results could be explained by a much
higher competitive potential of L. lactis in the hindgut of fish fed
diet 7 for substrate and adhesion sites. The phylum Fusobacteria
was at the genus level almost exclusively represented by Ceto-
bacterium, with a high similarity to C. somerae at species level. C.
somerae has been previously reported in the intestinal tract of
freshwater fish [65] and was also detected in Siberian sturgeon
hindgut at relatively high abundances using Denaturing Gradient
Gel Electrophoresis [28]. The abundance of Rhodobacter sp. was
affected by probiotics. Rhodobacter spp. has been known as a
probiotic in aquaculture [66].
In conclusion, the ability of L. lactis ssp. lactis ST G45 to colonize
and modify the intestinal microbiota as a potential probiotic strain,
was confirmed. The selected probiotic strains isolated from Siberian
sturgeon are safe and capable of surviving and colonizing the fish
intestinal mucus, as well as antagonizing the resident microbiota.
The results of the present study strongly suggest that the dietary
combination of L. lactis ssp. lactis ST G45 and AXOS is cost effective
to promote growth performance and to boost some immune re-
sponses of Siberian sturgeon, compared to their separate supple-
mentation. They could therefore be considered as a useful
alternative to chemotherapeutic treatments to promote fish health.
Acknowledgements
We gratefully acknowledge the technical assistance of Bart
Hellemans, KU Leuven. Jeroen KJ Van Houdt from Biogenomics
(http://bio.kuleuven.be/eeb/lbeg/consulting.html), a KU Leuven
Research & Development-subdivision, is thanked for the meta-
genomics study design and NGS-based development of Genomic
resources. This research was co-funded by a KU Leuven IDO Project,
the IWT (Agency for Innovation through Science and Technology e
Flanders) SBO IMPAXOS Project, and is part of the Methusalem
programme ‘Food for the Future’ at the KU Leuven.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.fsi.2013.06.014.
774 Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775

References arabinoxylans, arabinoxylan-oligosaccharides, and xylooligosaccharides. Crit

Rev Food Sci Nutr 2011;51:178e94.
[28] Geraylou Z, Souffreau C, Rurangwa E, D’Hondt S, Callewaert L, Courtin CM,
[1] Nomoto K. Prevention of infections by probiotics. J Biosci Bioeng 2005;100:
et al. Effects of arabinoxylan-oligosaccharides (AXOS) on juvenile Siberian
583e92.
sturgeon (Acipenser baerii) performance, immune responses and gastrointes-
[2] Bachère E. Shrimp immunity and disease control. Aquaculture 2000;191:3e
tinal microbial community. Fish Shellfish Immunol 2012;33:718e24.
11.
[29] Rodriguez-Estrada U, Satoh S, Haga Y, Fushimi H, Sweetman J. Effects of single
[3] Fuller R. Probiotics in man and animals. J Appl Bacteriol 1989;66:365e78.
and combined supplementation of Enterococcus faecalis, mannanoligosaccharide
[4] Gatesoupe FJ. Updating the importance of lactic acid bacteria in fish farming:
and polyhydrobutyric acid on growth performance and immune response of
natural occurrence and probiotic treatments. J Mol Microbiol Biotechnol
rainbow trout Oncorhynchus mykiss. Aquac Sci 2009;57:609e17.
2008;14:107e14.
[30] Ai Q, Xu H, Mai K, Xu W, Wang J, Zhang W. Effects of dietary supplementation
[5] Kesarcodi-Watson A, Kaspar H, Lategan MJ, Gibson L. Probiotics in aquacul-
of Bacillus subtilis and fructooligosaccharide on growth performance, survival,
ture: the need, principles and mechanisms of action and screening processes.
non-specific immune response and disease resistance of juvenile large yellow
Aquaculture 2008;274:1e14.
croaker, Larimichthys crocea. Aquaculture 2011;317:155e61.
[6] Merrifield DL, Dimitroglou A, Foey A, Davies SJ, Baker RTM, Bøgwald J, et al.
[31] Ye JD, Wang K, Li FD, Sun YZ. Single or combined effects of fructo- and
The current status and future focus of probiotic and prebiotic applications for
mannanoligosaccharide supplements and Bacillus clausii on the growth, feed
salmonids. Aquaculture 2010;302:1e18.
utilization, body composition, digestive enzyme activity, innate immune
[7] Nayak SK. Probiotics and immunity: a fish perspective. Fish Shellfish Immunol
response and lipid metabolism of the Japanese flounder Paralichthys olivaceus.
2010;29:2e14.
Aquac Nutr 2011;17:902e11.
[8] Kumar R, Mukherjee SC, Ranjan R, Nayak SK. Enhanced innate immune pa-
[32] Geng X, Dong X-H, Tan B-P, Yang Q-H, Chi S-Y, Liu H-Y, et al. Effects of dietary
rameters in Labeo Rohita (Ham.) following oral administration of Bacillus
chitosan and Bacillus subtilis on the growth performance, non-specific im-
subtilis. Fish Shellfish Immunol 2008;24:168e72.
munity and disease resistance of cobia, Rachycentron canadum. Fish Shellfish
[9] Gibson R, Roberfroid MR. Dietary modulation of the human colonic microbiota
Immunol 2011;31:400e6.
introducing the concept of prebiotics. J Nutr 1995;125:1401e12.
[33] Mehrabi Z, Firouzbakhsh F, Jafarpour A. Effects of dietary supplementation of
[10] Gibson AR, Horn KM, Pong M. Activation of climbing fibers. Cerebellum
synbiotic on growth performance, serum biochemical parameters and carcass
2004;3:212e21.
composition in rainbow trout (Oncorhynchus mykiss) fingerlings. J Anim
[11] Burr G, Gatlin D, Ricke S. Microbial ecology of the gastrointestinal tract of fish
Physiol Anim Nutr 2012;96:474e81.
and the potential application of prebiotics and probiotics in finfish aquacul-
[34] Cerezuela R, Guardiola FA, Meseguer J, Esteban MA. Increases in immune
ture. J World Aquac Soc 2005;36:425e36.
parameters by inulin and Bacillus subtilis dietary administration to gilthead
[12] Jeney G, Jeney ZS. Application of immunostimulants for modulation of the
seabream (Sparus aurata L.) did not correlate with disease resistance to Pho-
non-specific defense mechanisms in sturgeon hybrid: Acipenser
tobacterium damselae. Fish Shellfish Immunol 2012;32:1032e40.
ruthenus  A. baerii. J Appl Ichthyol 2002;18:416e9.
[35] Geraylou Z, Vanhove MPM, Souffreau C, Rurangwa E, Buyse J, Ollevier F.
[13] Mahious AS, Van Loo J, Ollevier F. Impact of the prebiotics, inulin and oligo-
In vitro selection and characterization of putative probiotics isolated from the
fructose on microbial fermentation in the spiral valve of Siberian sturgeon
gut of Acipenser baerii (Brandt, 1869). Aquac Res 2012. http://dx.doi.org/
(Acipenser baerii). Florence, Italy: Aqua 2006, World Aquacult Society; 2006.
10.1111/j.1365-2109.2012.03232.x.
p. 564e5.
[36] Suralikar V, Sahu NP. Effect of feeding probiotic (Lactobacillus cremoris) on
[14] Mohajer Esterabadi M, Vahabzadeh H, Zamani AA, Soudagar M, Ghorbani
growth and survival of Macrobrachium rosenbergii postlarvae. J Appl Anim Res
Nasrabadi R. Effect of dietary immunogen prebiotics on growth and survival
2001;20:117e24.
indices of giant sturgeon (Huso huso Linne, 1758) juveniles. J Fish 2010;4:
[37] Mujeeb Rahiman KM, Jesmi Y, Thomas AP, Hatha AAM. Probiotic effect of
610e72.
Bacillus NL110 and Vibrio NE17 on the survival, growth performance and
[15] Ta’ati R, Soltani M, Bahmani M, Zamini AA. Effects of the prebiotics
immune response of Macrobrachium rosenbergii (de Man). Aquac Res 2010;41:
immunoster and immunowall on growth performance of juvenile beluga
120e34.
(Huso huso). J Appl Ichthyol 2011;27:796e8.
[38] Secombes CJ. Isolation of salmonid macrophages and analysis of their killing
[16] Hoseinifar SH, Mirvaghefi A, Merrifield DL, Amiri B, Yelghi S, Bastami K. The
activity. In: Stolen JS, Fletcher TC, Anderson DP, Robertson BS, Van
study of some haematological and serum biochemical parameters of juvenile
Muiswinkel WB, editors. Techniques in fish immunology. New Jersey: SOS
beluga (Huso huso) fed oligofructose. Fish Physiol Biochem 2011;37:91e6.
Publications; 1990. p. 137e52.
[17] Hoseinifar SH, Mirvaghefi A, Mojazi Amiri B, Rostami HK, Merrifield DL. The
[39] Chung S, Secombes CJ. Analysis of events occurring within teleost macro-
effects of oligofructose on growth performance, survival and autochthonous
phages during the respiratory burst. Comp Biochem Physiol 1998;89:539e44.
intestinal microbiota of beluga (Huso huso) juveniles. Aquac Nutr 2011;17:
[40] Quade MJ, Roth JA. A rapid, direct assay to measure degranulation of bovine
498e504.
neutrophil primary granules. Vet Immunol Immunopathol 1997;58:239e48.
[18] Grootaert C, Delcour JA, Courtin CM, Broekaert WF, Verstraete W, Van de
[41] Yano T. Assays of hemolytic complement activity. In: Stolen JS, Fletcher TC,
Wiele T. Microbial metabolism and prebiotic potency of arabinoxylan oligo-
Anderson DP, Kaattari SL, Rowley AF, editors. Techniques in fish immunology.
saccharides in the human intestine. Trends Food Sci Technol 2007;18:64e71.
New Jersey: SOS Publications; 1992. p. 131e41.
[19] Swennen K, Courtin CM, Van der Bruggen B, Vandecasteele C, Delcour JA.
[42] Siwicki AK, Anderson DP. Nonspecific defence mechanisms assay in fish II;
Ultrafiltration and ethanol precipitation for isolation of arabinox-
potential killing activity of neutrophils and manocytes, lysozyme activity in
ylooligosaccharides with different structures. Carbohydr Polym 2005;62:
serum and organs and total immunoglobulin (Ig) level in serum. In:
283e92.
Siwicki AK, Anderson DP, Waluga J, editors. Disease diagnosis and prevention
[20] Swennen MK, Courtin CM, Lindemans GCJE, Delcour JA. Large-scale produc-
methods 1993. p. 105e11. Olsztyn, Poland.
tion and characterization of wheat bran arabinoxylooligosaccharides. J Sci
[43] Giongo A, Crabb DB, Davis-Richardson AG, Chauliac D, Mobberley JM,
Food Agric 2006;86:1722e31.
Gano KA, et al. PANGEA: pipeline for analysis of next generation amplicons.
[21] Niewold TA, Schroyen M, Geens MM, Verhelst RSB, Courtin CM. Dietary in-
ISME J 2010;4:852e61.
clusion of arabinoxylan oligosaccharides (AXOS) down regulates mucosal
[44] Shiri Harzevili AR, Van Duffel H, Dhert PH, Swings J, Sorgeloos P. Use of a
responses to a bacterial challenge in a piglet model. J Funct Foods 2012;4:
potential probiotic Lactococcus lactis AR21strain for the enhancement of
626e35.
growth in the rotifer Brachionus plicatilis (Müller). Aquac Res 1998;29:411e7.
[22] Courtin CM, Broekaert WF, Swennen K, Lescroart O, Onagbesan O, Buyse J,
[45] Zhou X, Wang Y, Yao J, Li W. Inhibition ability of probiotic, Lactococcus lactis,
et al. Dietary inclusion of wheat bran arabinoxylooligosaccharides induces
against A. hydrophila and study of its immunostimulatory effect in tilapia
beneficial nutritional effects in chickens. Cereal Chem 2008a;85:607e13.
(Oreochromis niloticus). Int J Eng Sci Technol 2010;2:73e80.
[23] Courtin CM, Swennen K, Broekaert WF, Swennen Q, Buyse J, Decuypere E,
[46] Bandyopadhyay P, Das Mohapatra PK. Effect of a probiotic bacterium Bacillus
et al. Effects of dietary inclusion of xylooligosaccharides, arabinox-
circulans PB7 in the formulated diets: on growth, nutritional quality and
ylooligosaccharides and soluble arabinoxylan on the microbial composition of
immunity of Catla catla (Ham.). Fish Physiol Biochem 2009;35:467e78.
caecal contents of chickens. J Sci Food Agric 2008b;88:2517e22.
[47] Ghosh K, Sen SK, Ray AK. Supplementation of an isolated fish gut bacterium,
[24] Van Craeyveld V, Swennen K, Dornez E, Van de Wiele T, Marzorati M,
Bacillus circulans, in formulated diets for Rohu, Labeo rohita, fingerlings. Isr J
Verstraete W, et al. Structurally different wheat-derived arabinox-
Aquac Bamidgeh 2003;55:13e21.
ylooligosaccharides have different prebiotic and fermentation properties in
[48] Xu B, Wang Y, Li J, Lin Q. Effect of prebiotic xylooligosaccharides on growth
rats. J Nutr 2008;138:2348e55.
performances and digestive enzyme activities of allogynogenetic crucian carp
[25] Grootaert C, Van den Abbeele P, Marzorati M, Broekaert WF, Courtin CM,
(Carassius auratus gibelio). Fish Physiol Biochem 2009;35:351e7.
Delcour JA, et al. Comparison of prebiotic effects of arabinoxylan oligosac-
[49] Villamil L, Tafalla C, Figueras A, Novoa B. Evaluation of immunomodulatory
charides and inulin in a simulator of the human intestinal microbial
effects of lactic acid bacteria in turbot (Scophthalmus maximus). Clin Diagn Lab
ecosystem. FEMS Microbiol Ecol 2009;69:231e42.
Immunol 2002;9:1318e23.
[26] Sanchez JI, Marzorati M, Grootaert C, Baran M, Van Craeyveld V, Courtin CM,
[50] Heo WS, Kim YR, Kim EY, Bai SC, Kong IS. Effects of dietary probiotic, Lactococcus
et al. Arabinoxylan-oligosaccharides (AXOS) affect the protein/carbohydrate
lactis subsp. lactis I2, supplementation on the growth and immune response of
fermentation balance and microbial population dynamics of the simulator of
olive flounder (Paralichthys olivaceus). Aquaculture 2013;376:20e4.
human intestinal microbial ecosystem. Microb Biotechnol 2009;2:101e13.
[51] Faramarzi M, Jafaryan H, Roozbehfar R, Jafari M, Rashidi Y, Biria M. Influences
[27] Broekaert WF, Courtin CM, Verbeke K, Van de Wiele T, Verstraete W,
of probiotic bacilli via bioencapsulated Daphnia magna on resistance of Per-
Delcour JA. Prebiotic and other health-related effects of cereal-derived
sian sturgeon larvae against challenge tests. Glob Vet 2012;8:421e5.
 Z. Geraylou et al. / Fish & Shellfish Immunology 35 (2013) 766e775
[52] Li Y, Wang YJ, Wang L, Jiang KY. Influence of several non-nutrient additives on
nonspecific immunity and growth of juvenile turbot, Scophthalmus maximus L.
Aquac Nutr 2008;14:387e95.
[53] Watzl B, Girrbach S, Roller M. Inulin, oligofructose and immunomodulation. Br
J Nutr 2005;93:49e55.
[54] Ringø E, Olsen RE. The effect of diet on aerobic bacterial flora associated
with intestine of Arctic charr (Salvelinus alpinus L.). J Appl Microbiol
1999;86:22e8.
[55] Bakke-McKellep AM, Penn MH, Salas PM, Refstie S, Sperstad S, Landsverk T,
et al. Effects of dietary soyabean meal, inulin and oxytetracycline on intestinal
microbiota and epithelial cell stress, apoptosis and proliferation in the teleost
Atlantic salmon (Salmo salar L.). Br J Nutr 2007;97:699e713.
[56] Najdegerami EH, Ngoc Tran T, Defoirdt T, Marzorati M, Sorgeloos P, Boon N,
et al. Effects of poly-b-hydroxybutyrate (PHB) on Siberian sturgeon (Acipenser
baerii) fingerlings performance and its gastrointestinal tract microbial com-
munity. FEMS Microbiol Ecol 2011;79:25e33.
[57] Campos CA, Rodríguez Ó, Calo-Mata P, Prado M, Barros-Velázquez J. Pre-
liminary characterization of bacteriocins from Lactococcus lactis, Enterococcus
faecium and Enterococcus mundtii strains isolated from turbot (Psetta maxima).
Food Res Int 2006;39:356e64.
[58] Balcázar JL, Vendrell D, Blas ID, Ruiz-Zarzuela I, Muzquiz JL, Girones O.
Characterization of probiotic properties of lactic acid bacteria isolated from
intestinal microbiota of fish. Aquaculture 2008;278:188e91.
775
[59] Balcázar JL, Vendrell D, Blas ID, Ruiz-Zarzuela I, Girones O, Muzquiz JL. In vitro
competitive adhesion and production of antagonistic compounds by lactic
acid bacteria against fish pathogens. Vet Microbiol 2007;122:3e4.
[60] Gill HS. Stimulation of the immune system by lactic cultures. Int Dairy J
1998;8:535e44.
[61] Ouwehand AC, Salminen S. In vitro adhesion assays for probiotics and their
in vivo relevance: a review. Microb Ecol Health Dis 2004;15:175e84.
[62] Moura P, Barata R, Carvalheiro F, Girio F, Loureiro-Dias MC, Esteves MP.
In vitro fermentation of xylo-oligosaccharides from corn cobs autohydrolysis
by Bifidobacterium and Lactobacillus strains. LWT Food Sci Technol 2007;40:
963e72.
[63] Rurangwa E, Laranja JL, Van Houdt R, Delaedt Y, Geraylou Z, Van de Wiele T,
et al. Selected nondigestible carbohydrates and prebiotics support the growth
of probiotic fish bacteria mono-cultures in vitro. J Appl Microbiol 2009;136:
932e40.
[64] Klaasen HL, Van der Heijden PJ, Stok W, Poelma FG, Koopman JP, Van den
Brink ME, et al. Apathogenic, intestinal, segmented, filamentous bacteria
stimulate the mucosal immune system of mice. Infect Immun 1993;61:303e6.
[65] Tsuchiya C, Sakata T, Sugita H. Novel ecological niche of Cetobacterium som-
erae, an anaerobic bacterium in the intestinal tracts of freshwater fish. Lett
Appl Microbiol 2008;46:43e8.
[66] Qi Z, Zhang XH, Boon N, Bossier P. Probiotics in aquaculture of China e current
state, problems and prospect. Aquaculture 2009;290:15e21.