Probiotic Modulation of the Gut Microbiota of Fish - Aquaculture Nutrition: Gut Health, Probiotics and Prebiotics - page 191

groups after 90 days feeding on the probiotic diets. Contrary to this, an increase of coliforms
(by ca. log 1.0 CFU g -1 ) was observed in the control fish. As the dietary concentration of
B. subtilis in the study was insignificant with regards to the general reduction of intestinal
bacterial counts, the authors suggested that a continuous administration of 10 6 -10 7 CFU g -1
is sufficient to yield a healthy balance between B. subtilis and the indigenous gut microbiota.
Although Bacillus spp. are well equipped to make similar impacts upon the gastric micro-
bial ecosystems of other fish species, the present data are somewhat ambiguous. Chang and
Lui (2002) demonstrated that after dietary provision of B. toyoi to European eel ( Anguilla
anguilla ) for 14 days it was not possible to recover viable B. toyoi isolates from the digestive
tract using a culture based approach. Despite this, however, a clear effect on the indigenous
microbiota was observed as the total viable heterotrophic population was reduced by 90%
compared to initial levels even though the bacterial community composition was not affected.
However, as the authors starved the fish for 24 h prior to sampling one cannot draw a full con-
clusion. Similarly, after feeding dietary B.subtilis C-3102 (Calsporin ® , Calpis Co. Ltd, Tokyo,
Japan) to koi carp Cyprinus carpio for five weeks, He et al . (2011) reported, using DGGE,
that the probiont was not present at detectable levels in the intestine of probiotic fed fish at
any sampling time point. In spite of this the probiotic treatment could impact the intestinal
microbial community, particularly at the early stages of the trial. Yang et al . (2012) reported
from DGGE analysis that the B. clausii was not detectable in the fore-, mid- or hindgut of
grouper Epinephelus coioides juveniles after dietary provision at 10 8 cells g -1 for 60 days.
The microbial fingerprints revealed complex, and generally similar, bacterial communities in
the respective GI regions of the two treatment groups. Differences in the communities in the
different gut regions were observed, which were characterized by regional specialization of
some operational taxonomic units (OTUs) and clustering of hindgut replicates, irrespective of
treatment, together in the dendrogram. No differences in the species richness or diversity were
observed between the treatments. However, four OTUs (closest known relatives identified as
Enterococcus sp. (92% similarity), Photobacterium sp. (100%), Bacillus pumilus (90%) and
an uncultured bacterium (97%)) were observed to be present only in the probiotic group and
three OTUs (closest known relatives identified as Staphylococcus sp. HB4 (90%), uncultured
Gammaproteobacterium clone (93%) and Vibrio ponticus (99%)) were present only in the
control group (though not in all replicates of all gut regions).
The same research group fed grouper juveniles B. pumilus (originally isolated from the
GI tract of juvenile grouper: Sun et al . 2009) at 10 8 cells g -1 for 60 days (Sun et al . 2011a).
Complex DGGE patterns were observed but no significant difference in the total number of
OTUs or bacterial diversity was detected between the probiotic group and the control group.
Despite this, UPGMA analysis showed clustering into three groups, one for each gut region
investigated (fore-, mid- and hindgut), and each region contained two sub-clusters represent-
ing the two dietary treatments. Most observed OTUs, including an OTU at the same migration
point as the probiotic, were common to all groups. Four OTUs (closest known relatives identi-
ied as uncultured Bacillus sp. clone (100% similarity), uncultured bacterium clone LL14117
(89%), Nitratireductor sp. YCSC5 (100%) and an uncultured bacterium clone G2-66 (95%))
were present only in the probiotic group, and two OTUs ( Staphylococcus saprophyticus (99%)
and uncultured bacterium clone SW-12 (98%)) were only present in the control group (though
not in all replicates). Furthermore, the relative abundance (determined by the intensity of the
band as a proportion of the total band intensities) of one OTU, identified as uncultured bac-
terium clone D3T-094 (95%), was significantly higher in the probiotic samples. Although the