the ambient air in the building and remained at
26F1 8C throughout the trial. A 12 h light:12 h
dark photoperiod was maintained with fluorescent
lights controlled by timers.
Each experimental diet was fed to three groups
of small fish and one group of large fish for 16
weeks. All groups were fed to apparent satiation in
the morning and evening, 7 days each week.
Growth and feed efficiency were monitored monthly
by collectively weighing each group of fish. Weight
gain was expressed as the increase in total
cumulative biomass per tank. After week 16, a
chronic mycobacterial infection became severe and
was confirmed by the TVMDL to be caused by M.
marinum based on biochemical tests. At that
juncture, fish were fed to apparent satiation once
per day for additional 5 weeks. Mortality was
monitored twice daily during this period. All
procedures were approved by the Animal Care and
Use Committee of Texas A&M University.
2.3. Sample collection and analysis
At the end of the 16-week feeding trial, three
apparently healthy fish (no obvious skin lesions and
visceral granulomas) from each tank (12 fish per
treatment) were anesthetized with tricaine methane
sulfonate (MS-222), and approximately 2 ml of
blood was collected from the caudal vasculature
using a 3-ml syringe and 23-gauge needle. These
representative fish were euthanized and head kidney
samples were pooled for macrophage isolation and
assay of respiratory burst of head kidney leukocytes.
The assay of extracellular and intracellular superoxide
anion followed the procedure of Secombes
(1990), as modified by Sealey and Gatlin (2002a).
The amount of extracellular superoxide anion was
calculated by the formula of Pick and Mizel (1981).
Whole blood neutrophil oxidative radical production
was determined as described by Siwicki et al.
(1994). Absorbance was converted to nitro blue
tetrazolium (NBT) units based on a standard curve
of NBT diformazan/ml blood. Plasma was separated
as previously described (Li and Gatlin, 2003) and
lysozyme activity was determined by a turbidimetric
assay (Jørgensen et al., 1993). A lysozyme activity
unit was defined as the amount of enzyme
producing a decrease in absorbance of 0.001 min1
at pH 6.1. Serum peroxidase was analyzed as
described by Rodrı´guez et al. (2003).
2.4. Statistics
Data from the feeding trial, immune response
assays and the bacterial challenge were subjected to
analysis of variance according to a randomized
complete block design using a significant level of
P V0.05. If a significantly main effect was observed,
treatment means were separated using the LSD
comparison test. All statistical analysis was conducted
with StatistixR Analytical Software (Tallahassee,
FL).
3. Results
3.1. Growth performance
Generally, fish fed the diets supplemented with
brewers yeast and GroBioticR-A had better growth
performance during the 16-week feeding trial (Table
2). After 4 weeks of feeding, fish fed 2% brewers
yeast had a significantly (P b0.05) higher weight gain
than fish fed the basal diet and the diet supplemented
with 2% GroBioticR-A. After 12 weeks, fish fed 1%
and 2% brewers yeast and 2% GroBioticR-A had
significantly higher weight gain than fish fed the basal
diet, and feed efficiency showed a similar trend. At
the end of the 16-week period, fish fed 2% brewers
yeast and GroBioticR-A had the highest biomass,
which tended to be significant (P =0.11). Feed
efficiency of fish fed 2% brewers yeast was significantly
greater than the other treatments.
3.2. Immune response assays
No significant effects of the various diets were
observed on neutrophil oxidative radical production,
serum lysozyme and intracellular superoxide anion
production by head kidney macrophages of sub-adult
hybrid striped bass after the 16-week period (Table 3).
However, fish fed 1% and 2% brewers yeast had a
significantly (P b0.01) higher serum peroxidase level
than fish fed the basal diet and the diet supplemented
with 2% GroBioticR-A diet. The extracellular superoxide
anion production of head kidney macrophages