Importance of biofilm for water quality and nourishment in intensive shrimp culture

Abstract

Experiments were conducted to test the usefulness of biofilms—a microbial consortium associated with extracellular polymeric substances attached to submersed surfaces—in reducing the levels of ammonium and phosphate of rearing system water, and as a food source for the shrimp Farfantepenaeus paulensis. A mature biofilm, which is able to keep ammonium and phosphate at low levels, occurred 10–15 days after tank cleanup, and was characterized by chlorophyll-a concentration around 5 μg/cm2. It was mainly composed of pennate diatoms (Amphora, Campylopyxis, Navicula, Sinedra, Hantschia and Cylindrotheca; ca. 9×104 cells/mg of biofilm) and filamentous cyanobacteria (Oscillatoria and Spirulina; ca. 2×105 cells/mg), though bacteria (max. 1.48×107/mg), flagellates (max. 1.08×103/mg) and ciliates (max. 3.51×102/mg) were also present. Pennate diatoms and filamentous cyanobacteria were responsible for the largest uptake of ammonium from the water, but nitrifying bacteria also played an important role. The presence of a biofilm lead to reduced exportation of phosphorus (33% less phosphate) and to a higher output of nitrate+nitrite, instead of ammonium. Biofilm was also an important complementary food source for the shrimp, increasing their growth.

Full text

Ž.
Aquaculture 203 2002 263–278 www.elsevier.comrlocateraqua-online
Importance of biofilm for water quality and
nourishment in intensive shrimp culture
Fabiano Lopes Thompson1, Paulo Cesar Abreu),
Wilson Wasielesky
Departamento de Oceanografia, Fund. UniÕersidade Federal do Rio Grande — FURG, Cx. P. 474,
96201-900 Rio Grande, RS Brazil
Received 8 June 2000; received in revised form 19 December 2000; accepted 31 March 2001
Abstract
Experiments were conducted to test the usefulness of biofilms—a microbial consortium
associated with extracellular polymeric substances attached to submersed surfaces—in reducing
the levels of ammonium and phosphate of rearing system water, and as a food source for the
shrimp Farfantepenaeus paulensis. A mature biofilm, which is able to keep ammonium and
phosphate at low levels, occurred 10–15 days after tank cleanup, and was characterized by
chlorophyll-aconcentration around 5 mgrcm
2
. It was mainly composed of pennate diatoms
Ž
4
Amphora,Campylopyxis,NaÕicula,Sinedra,Hantschia and Cylindrotheca; ca. 9=10 cellsrmg
.Ž
5
.
of biofilm and filamentous cyanobacteria Oscillatoria and Spirulina; ca. 2 =10 cellsrmg ,
Ž
7
.Ž
3
.Ž
though bacteria max. 1.48=10 rmg , flagellates max. 1.08=10 rmg and ciliates max.
2
.
3.51=10 rmg were also present. Pennate diatoms and filamentous cyanobacteria were responsi-
ble for the largest uptake of ammonium from the water, but nitrifying bacteria also played an
Ž
important role. The presence of a biofilm lead to reduced exportation of phosphorus 33% less
.
phosphate and to a higher output of nitrateqnitrite, instead of ammonium. Biofilm was also an
important complementary food source for the shrimp, increasing their growth. q2002 Elsevier
Science B.V. All rights reserved.
Keywords: Shrimp; Biofilm; Microorganisms; Water quality; Food
)Corresponding author. Tel.: q55-532-336509; fax: q55-532-336602.
Ž. Ž.
E-mail addresses: Fabiano.Thompson@rug.ac.be F.L. Thompson , docpca@super.furg.br P.C. Abreu .
1Present address: Laboratory for Microbiology, Gent University, K.L. Ledeganckstraat 35, 9000 Gent,
Belgium.
0044-8486r02r$ - see front matter q2002 Elsevier Science B.V. All rights reserved.
Ž.
PII: S0044-8486 01 00642-1

()
F.L. Thompson et al.rAquaculture 203 2002 263–278264
1. Introduction
Accumulation of dissolved nitrogen, especially ammonium 2, as a result of food
addition and excretion of organisms reared at high density, is one of the main problems
in intensive shrimp culture systems, affecting their food ingestion, growth and survival
Ž
rates Tomasso, 1994; Wasielesky et al., 1994; Ostrensky and Wasielesky, 1995; Cavalli
.
et al., 1996 . Moreover, shrimp exposure to high ammonium concentrations seems to
Ž.
reduce their resistance to diseases Brock and Main, 1994 . Additionally, effluents from
such rearing systems reach most water bodies without previous treatment, thus carrying
high loads of nutrients, which cause environmental degradation and give rise to
Ž.
pathogenic microorganisms Hopkins et al., 1995 . To keep dissolved nutrient at low
levels, large amounts of water must be exchanged daily, increasing the costs of shrimp
production.
Ž
An alternative way to maintain high water quality is biological treatment Wheaton,
.
1977 , based on the use of filters with a high surfacervolume ratio, pre-colonized by
microorganisms that absorb excess nutrients from the water. A similar process occurs in
nature, where biofilms—a microbial consortium associated with a matrix of extracellular
polymeric substances bound to any submersed surfaces—are responsible for many
Ž
biogeochemical cycles in aquatic ecosystems, especially nitrogen cycling Decho 1990;
.
Meyer-Reil, 1994 .
Our main objective was to test the efficiency of biofilms in maintaining a high water
Ž.
quality through uptake of dissolved inorganic nutrients ammonium and phosphate . In
addition, we investigated whether biofilm could be used as a complementary food source
during the culture of the shrimp Farfantepenaeus paulensis.
2. Material and methods
2.1. Experimental design
Five experiments were conducted at the Marine Aquaculture StationrFURG in
Ž.
southern Brazil 328S, 528W , using healthy juveniles of the shrimp F. paulensis,
produced at the Marine Shrimp Laboratory of the Universidade Federal de Santa
Catarina, Brazil. In all experiments, shrimp were fed with a specific ration for penaeid
Žw.Ž
shrimp Sibra . The experiments were conducted either in circular nursery tanks 10
2323
.
m surface and 3–5 m volume; ca. 3 m of biofilmrm of water , or in glass fiber
Ž232 3
.
containers 0.3 m and 0.06 m ; ca. 13 m of biofilmrm of water . Before the
beginning of each experiment, the tanks and containers were emptied, and all organic
material accumulated at the bottom was removed. After that, they were filled again with
sand-filtered coastal water.
2Žq.
Throughout the text, the term ammonium will be used as the sum of the ionized ammonium—NH and
4
Ž.
anionized ammonia—NH forms. Ammonium is the dominant form in natural water, while ammonia,
3
though in smaller proportion, is the toxic form.

()
F.L. Thompson et al.rAquaculture 203 2002 263–278 265
2.1.1. Experiment 1
The main objective of this experiment was to test the efficiency of the biofilm in
removing ammonium from shrimp culture water. The assay was carried out during 10
Ž.Ž
days in two circular tanks: 1F without biofilm; with food addition and 1BF with
.Ž
4-month-old biofilm and food addition . Each tank had 200 shrimp 0.14"0.05 g wet
.23
weight making 20 shrimprmor9grm . Food was added daily at a rate equivalent to
200% of the total shrimp biomass. During the first 3 days of the experiment, water was
Ž.
renewed daily 33% of total volume .
2.1.2. Experiment 2
In this experiment, we tested the nutrient uptake efficiency of biofilms under
conditions of high shrimp density and high food input. The experiment was conducted in
Ž.
six glass fiber containers with biofilm distributed into two treatments in triplicate as
Ž.
follows: 2LD—low density—with 70 shrimp 0.43"0.02 g wet weight , corresponding
to 500 g shrimp wet weightrm3and a daily input of 83 g rationrm3; 2HD—high
Ž3.
density—with 140 shrimp 1000 g shrimprm and addition of food at a daily rate of
166 grm3. The experiment lasted for 14 days. The organic matter accumulated at the
bottom of the containers was removed by siphoning, and water was added to complete to
the previous level, representing a daily water exchange rate of 2.5% of the total volume.
2.1.3. Experiment 3
This experiment was a repetition of experiment 2, but with biofilm removal. The
amount of biomass present in the containers and food addition were the same as in the
Ž.
previous experiment. However, bigger shrimp 1.44"0.07 g wet weight were used and
the 3LD and 3HD treatments received 21 and 42 individuals, respectively.
2.1.4. Experiment 4
Ž.
This experiment was designed to determine: 1 the minimum time for the establish-
ment of mature biofilm, i.e., when the biofilm shows maximum ammonium uptake; and
Ž.
2 the potential of biofilm as a food source. The experiment was carried out in fiber
Ž. Ž
containers distributed in three treatments in triplicate : treatment 4B with biofilm,
.Ž .
without food addition ; treatment 4F without biofilm, with food addition ; and treatment
Ž. Ž
4BF with biofilm and food addition . Each treatment had 20 shrimp 0.24"0.02 g wet
.23
weight corresponding to 60 shrimprm or80grm . The total amount of food added
daily corresponded to 50% of the shrimp biomass. There was no water exchange during
the 28 days of experiment.
2.1.5. Experiment 5
The main objectives of this experiment were to determine the export of N and P from
tanks with and without biofilm and to analyze the shrimp growth rate under these
Ž
conditions. Two circular tanks 5F—without biofilm and with food addition and
.Ž
5BF—with biofilm and food addition were stocked with 10,000 juveniles 0.30"0.03
3.
g wet weight; 1000 grm and received daily rations equivalent to 50% of the total
Ž.
shrimp biomass. Water was renewed daily 50% of the total water volume during all
Ž.
experiment 49 days .

()
F.L. Thompson et al.rAquaculture 203 2002 263–278266
2.2. Physical and chemical analysis
Ž.
In all experiments, water temperature mercury thermometer "0.18C and salinity
Ž. Ž
refractometer "1 were monitored daily. The concentration of ammonium N—NH q
3
q.
NH ; UNESCO, 1983 in water was measured daily, immediately after sampling
4Žyy
.
experiment 1 and twice a week in experiments 2–5. Nitriteqnitrate N—NO qNO
23
concentrations were determined every day in experiment 1 and once a week in
Žy.
experiment 4, while nitrite N—NO was measured separately twice a week in
2Žy.
experiments 2, 3 and 5, as well as nitrate N—NO in experiment 5. Phosphate
3
Ž3y.
P—PO was measured at days 7, 14, 17, 21 and 49 in experiment 5. The analysis of
4Ž.
these nutrients was conducted according to Strickland and Parsons 1972 . Biofilm
Ž.
protein content wet weight was measured at the beginning and at the end of all
Ž
experiments according to the Association of Official Analytical Chemists AOAC,
.
1995 .
The loads of nitrogen and phosphorus exported from the tanks during experiment 5
were determined considering the concentration of ammonium, nitrite, nitrate and phos-
phate and the water volume exchanged. Total export was calculated by plotting the
nutrient load at different times and integrating the area under the curve by the Gauss
method.
2.3. Biological analysis
During the first experiment, the abundance of microorganisms was determined in the
biofilm scraped off the tank wall. The biofilm wet weight was determined using a
Ž.
Sartorius balance 0.0001 g , and microorganism abundance was estimated per gram of
Ž.
biofilm wet weight . In the other four experiments, the concentration of microorganisms
and chlorophyll-awere measured in the biofilm, which had developed on flexible PVC
Ž.
tubes 0.5–15 cm length; 0.5 cm diameter that had been introduced in the tanks
immediately after biofilm cleanup. Total microorganism abundance and pigment concen-
tration were estimated considering the area of the tube covered by the biofilm. Ancillary
tests showed no significant differences between microalgae and microorganism abun-
Ž.
dance on tank walls versus tubes t-test; P)0.05 .
Biofilm chlorophyll-awas determined from the PVC tubes. The tubes were removed
Žv.
from the tanks and placed in vials with 20 ml acetone 90%rand stored for 24 h at
v
y128C in the dark, for pigment extraction. Pigment concentration was determined using
Ž. Ž
a calibrated fluorometer Turner TD 700 with acidification of samples Strickland and
.
Parsons, 1972 .
For microalgae and microorganism enumeration, the biofilm scraped off the tank wall
Žv.
in experiment 1 and that present on PVC tubes were fixed in formalin 4%r. Before
v
enumeration, the biofilm was detached from the PVC tubes and disintegrated using an
Ž.
ultrasonic homogenizer 4710 Series, Cole Parmer , applying three to five times a
20-kHz frequency during 10–15 s, with the same interval between homogenization.
Auxiliary tests showed that this procedure does not disrupt microorganisms. The
material was then diluted with water from the containers, filtered through GFrF glass
Ž.
fiber filters and polycarbonate membrane filters Nuclepore—0.2 mm pore size .

()
F.L. Thompson et al.rAquaculture 203 2002 263–278 267
Biofilm scraped off the tank wall during experiment 1 was homogenized and diluted in a
similar way.
Bacteria and flagellates present in the biofilm were quantified in 1 ml aliquots filtered
Ž.
through polycarbonate membrane filters Nuclepore—0.2 mm pore , previously dark-
ened with irgalan black. After filtration, the cells were dyed with acridine orange
Žw.Ž.
0.1% raccording to the method described by Hobbie et al. 1977 , with some
vŽ.
adaptations Thompson et al., 1999 . Microorganisms were enumerated in 30 fields,
chosen at random, using a Zeiss epifluorescence microscope equipped with blue light
Ž.
filter BP 450–490; FT 510; LT 520 . The differentiation between auto- and hetero-
trophic cells was made by observing the bright red fluorescence, characteristic of
chlorophyll-a. For the determination of diatom, cyanobacteria and ciliate abundances,
2–10 ml of diluted biofilm was poured into sedimentation chambers and left there for at
least 24 h. After this time, organisms were counted in 30 fields, chosen at random, using
Ž.
an inverted light microscope Utermohl, 1958 . Identification of diatoms, cyanobacteria
¨Ž.
and ciliates at the genus level was performed according to Round et al. 1990 and
Ž.
Whitford and Schumacher 1973 .
In all experiments, shrimp survival was determined at the end of the study period.
The final weight of the shrimp was also determined. Shrimp gut content was examined
Ž.
as described elsewhere Thompson et al., 1999 .
2.4. Statistical analysis
Ž.
Significant differences P-0.05 among treatments of experiments 1, 2, 3 and 5
Ž.
were determined using the t-test for small samples Sokal and Rohlf, 1969 . ANOVA
Ž.
and the a posteriori Tukey test P-0.05; Sokal and Rohlf, 1969 were used to
differentiate treatments in experiment 4.
3. Results
The values of temperature, salinity, biofilm chlorophyll-a, shrimp survival and
increase in weight are registered in Table 1.
3.1. Biofilm efficiency
There was a significant increase in the ammonium concentration in the 1F treatment
Ž.
without biofilm of experiment 1, especially after the third day, when water exchange
Ž. Ž .
stopped Fig. 1A . Maximum ammonium concentration 83.99 mM was registered 9
days after the beginning of the experiment. On the other hand, levels of ammonium were
Ž.
low and stable in the tank with biofilm 1BF , varying from 5.94 to 16.09 mM. In this
tank, there was a constant increase of nitriteqnitrate, reaching maximum concentration
Ž.Ž.
21.93 mM on day 9 Fig. 1B . In the 1F treatment, nitriteqnitrate started to increase
after 6 days.

()
F.L. Thompson et al.rAquaculture 203 2002 263–278268
Table 1
Ž. Ž . Ž. Ž. Ž.
Water temperature C , salinity, biofilm chlorophyll a mgrl variation, survival rate % , final weight g and increase of weight % of F. paulsensis at the end of
Ž.
the experiments, mean "s.e.
Ž. Ž.
Different letters superscript indicate significant differences PF0.05 .
Ž. Ž.
Experiment Treatment Water temperature Salinity Biofilm Survival % Final weight g Increase in
Ž. Ž.
8C chlorophyll weight %
2
Ž.
mgrcm
Ž. Ž. Ž .
1 1F 25.0 "0.5 20.0 "0.3 – 62 0.20 "0.07 42.8
Ž. Ž. Ž .
1BF 25.0 "0.5 20.0 "0.3 – 58 0.20 "0.09 42.8
Ž. Ž. Ž. Ž .
2 2LD 24.2 "0.4 25.0 "0.3 5.0–11.8 98 "9 0.57 "0.01 32
Ž. Ž. Ž. Ž .
2HD 24.2 "0.4 25.0 "0.3 7.4–8.6 78 "11 0.56 "0.01 30
Ž. Ž. Ž .
3 3LD 20.8 "0.2 29.0 "0.5 0–1.0 100 1.66 "0.02 15
Ž. Ž. Ž. Ž .
3HD 20.8 "0.2 29.0 "0.5 0–0.7 89 "8 1.63 "0.01 13
a
Ž. Ž. Ž. Ž .
4 4B 24.0 "0.4 25.4 "0.2 2.5–30.7 98 "2 0.24 "0.02 0
b
Ž. Ž. Ž .
4F 24.0 "0.4 25.4 "0.2 0–11.4 100 0.61 "0.06 54
b
Ž. Ž. Ž .
4BF 24.0 "0.4 25.4 "0.2 1.5–21.7 98 0.64 "0.11 66a
a
Ž. Ž. Ž .
5 5F 23.8 "0.3 25.9 "0.2 0–18.2 51 0.40 "0.02 33b
b
Ž. Ž. Ž .
5BF 23.8 "0.3 25.9 "0.2 4.9–36.2 53 0.61 "0.05 103

()
F.L. Thompson et al.rAquaculture 203 2002 263–278 269
Ž. Ž. Ž . Ž .
Fig. 1. Mean values of A ammonium and B nitriteqnitrate concentration mM in the 1F without biofilm
Ž. Ž .
and 1BF with biofilm treatments—experiment 1 bar sstandard error .
Ammonium concentration was also lessened in containers with biofilm and low
Ž. Ž.
shrimp density 2LD of experiment 2. In the high density treatment 2HD , concentra-
tion of this nutrient was also low at the beginning, however, there was an increase in
Ž.
ammonium levels reaching 624.20 mM after 14 days Fig. 2A—with biofilm . Both
Ž
treatments showed an increase in nitrite, with maximum occurrence at day 7 Fig.
.Ž.
2B—with biofilm . The containers where biofilms were removed experiment 3 showed
much higher ammonium concentration, with maximum of 742.57 and 1784.11 mMin
Ž.
the 3LD and 3HD treatments, respectively Fig. 2A—without biofilm . On the other
hand, the amount of nitrite in these treatments was low, compared to the containers with
Ž.
biofilm Fig. 2B .
Ž.
In experiment 4, the highest ammonium concentration 544.48 mM was measured in
Ž.
the 4F treatment without biofilm at day 10. After this, ammonium concentration

()
F.L. Thompson et al.rAquaculture 203 2002 263–278270
Ž. Ž. Ž . Ž .
Fig. 2. A Ammonium and B nitrite mean concentration mM in the 2LD low shrimp density and 2HD
Ž. Ž.Ž.
high shrimp density treatments —experiment 2 with biofilm —and in the 3LD low shrimp density and
Ž. Ž.Ž .
3HD high shrimp density treatments—experiment 3 without biofilm barsstandard deviation .
Ž.
reached similar levels as the 4B and 4BF treatments Fig. 3A . The decline in
ammonium in the 4F treatment corresponded to an increase in the nitriteqnitrate, which
Ž.
reached its maximum 1006.3 mM at day 14. At the end of the experiment, there was
Ž.
no significant difference P)0.05 of nitriteqnitrate concentrations among treatments
Ž.
Fig. 3B .
Ž.
During experiment 5, the highest ammonium concentration 173.66 mM was mea-
sured after 10 days in the 5F tank. A second peak of ammonium occurred at day 28,
Ž.
decreasing again afterwards Fig. 4A . In this treatment, the decline of ammonium
Ž.
matched the increase of nitrite Fig. 4B . However, the nitrate concentration remained

()
F.L. Thompson et al.rAquaculture 203 2002 263–278 271
Ž. Ž.Ž. Ž. Ž. Ž 2.
Fig. 3. A Ammonium mM , B nitriteqnitrate mM and C chlorophyll-amgrcm concentration in the
Ž.Ž.Ž
4B with biofilm, without food addition , 4F without biofilm, with food addition and 4BF with biofilm and
.Ž.
food addition treatments—experiment 4 barsstandard error .
Ž. Ž.
low during most of the experiment Fig. 4C . In the tank with biofilm 5BF , ammonium
and nitrite concentrations were lower than in the 5F treatment. Nitrate, on the other
Ž.
hand, showed higher concentration Fig. 4A–C . Phosphate showed similar trends in

()
F.L. Thompson et al.rAquaculture 203 2002 263–278272
Ž. Ž. Ž. Ž. Ž . Ž
Fig. 4. A Ammonium, B nitrite, C nitrate and D phosphate concentration mM in the 5F without
.Ž . Ž .
biofilm and 5BF with biofilm treatments—experiment 5 barsstandard error .
both treatments, although values in the 5F treatment were higher than in the 5BF tank
Ž.
Fig. 4D .
3.2. Establishment of a mature biofilm
Ž
In the process of biofilm formation, smaller heterotrophic bacteria cocci and rod
.
shaped were present in high numbers in the 1F treatment after the second day of the
Ž7.
experiment 0–1.48=10 rmg , while filamentous heterotrophic bacteria appeared later
Ž4.Ž . Ž 3.Ž
0–1.48=10 rmg Fig. 5A . Flagellates 0–1.08 =10 rmg and ciliates 0–3.51=
2.Ž.
10 rmg showed peaks at the second day and at the end of the experiment Fig. 5B .
Decrease of ammonium in all experiments was related to the augmentation of
chlorophyll-ain the biofilm. In experiment 4, for example, maximum chlorophyll-a
Ž2.
measured in the 4F containers 11.39 mgrcm occurred after 14 days of experiment
Ž. Ž .
Fig. 3C . Even the tanks with biofilm 4B and 4BF treatments showed an increase in
Ž.
chlorophyll-aconcentration Fig. 3C . The biofilm was dominated by pennate diatoms
ŽAmphora,Campylopyxis,NaÕicula,Sinedra,Hantschia,Cylindrotheca and Gy-
.Ž.
rosigma and filamentous cyanobacteria Oscillatoria and Spirulina . Diatoms abun-
dance varied from 0 to 2.49=104rmg of biofilm, while filamentous cyanobacteria
4Ž.
varied between 0 and 0.35=10 rmg Fig. 5C .
3.3. Biofilm as food
Gut content analysis showed that F. paulensis actively feeds non-selectively on the
biofilm. There was no significant difference of shrimp survival among the treatments

()
F.L. Thompson et al.rAquaculture 203 2002 263–278 273
Ž.Ž. Ž.
Fig. 5. Mean abundance cellsrmg of biofilm of A small and filamentous heterotrophic bacteria, B
Ž. Ž .
flagellates and ciliates and C diatoms and cyanobacteria in the 1F treatment without biofilm —experiment 1
Ž.
barsstandard error .
Ž.
P)0.05 . However, shrimps showed a higher final weight in the tanks with biofilm,
Ž.
leading to higher shrimp biomass at the end of experiments Table 1 . The biofilm of all
Ž.
experiments had a low protein content mean 6% of wet weight .
3.4. Biofilm and nutrient export
The amount of nitrogen and phosphorus exported from the 5F tank during experiment
Žq.Ž
y.
5 was: 127 g of ammonium N—NH qNH ; 59 g of nitrite N—NO ; 6 g of nitrate
34 2
Žy.Ž
3y.
N—NO and 6.34 g of phosphate P—PO . The 5BF treatment, on the other hand,
34

()
F.L. Thompson et al.rAquaculture 203 2002 263–278274
Ž.Ž
Fig. 6. Exported load of ammonium, nitrite and nitrate from treatments 5F without biofilm and 5BF with
.Ž .
biofilm —experiment 5 gr49 days and % of total N .
exported 76 g of ammonium; 48 g of nitrite; 62 g of nitrate and 4.23 g of phosphate
Ž.
Fig. 6 .
4. Discussion
Biofilm formation begins with the accumulation of organic molecules on any
submersed surface. This is a physical–chemical process, and occurs a few seconds to
minutes after the immersion of any surface in the liquid. Few hours after the establish-
Ž.
ment of the macromolecular film, bacterial colonization begins Whal, 1989 . In our
study, we observed high numbers of attached bacteria 24 h after the start of the
experiments, though bacterial abundance was, in many cases, controlled by the grazing
exerted by flagellates and ciliates. It was not possible to precisely determine how this
predator–prey interaction could influence the ammonium and nitriteqnitrate production
in the tanks, however, it is likely that ciliates work as mineralizer when they excrete
Ž
ammonium and return to the water the nitrogen incorporated by the bacteria Caron,
.
1994 . We also observed a decrease of flagellate and ciliate abundances during some
experiments, which possibly resulted from the predation pressure exerted by F. paulen-
sis juveniles, or rotifers and nematodes present in the biofilm. Similarly, we did not
evaluate the effect of this trophic interaction on the production of the different nitrogen
forms.

()
F.L. Thompson et al.rAquaculture 203 2002 263–278 275
Some studies primarily attributed the decrease of ammonium in culture water to the
Ž.
action of nitrifying bacteria Langis et al., 1988; Ramesh et al., 1999 . In fact, the
decrease in ammonium and parallel augmentation in nitrite and nitrate concentrations in
most of the experiments indicates that nitrifying bacteria present in the biofilm play an
important role. However, the levels of nitriteqnitrate generated in the tanks were
frequently smaller than that of ammonium. Moreover, ammonium in the tanks was not
Ž
as high as that normally measured when nitrifying bacteria are dominant Kaiser and
.
Wheaton, 1983 .
Actually, the decline in ammonium concentrations in all experiments was mainly
related to an increase in chlorophyll-ain biofilms. It seems that the ammonium was
mainly absorbed by the microalgae that use this element to produce new biomass.
Diatoms and cyanobacteria appeared 4–5 days after the beginning of the experiments,
reaching maximum values 2 weeks later, as demonstrated by the chlorophyll-avalues.
However, the largest ammonium uptake occurred 10–15 days after the beginning of the
experiments, when chlorophyll-aconcentration was around 5 mgrcm2, i.e., below the
maximum value. At this time, microalgae community was dominated by pennate
Ž.
diatoms Amphora,Campylopyxis,NaÕicula,Sinedra,Hantschia and Cylindrotheca
Ž.
and filamentous cyanobacteria Oscillatoria and Spirulina .
The variability of chlorophyll-ain the biofilm can be controlled by the predation
pressure exerted by shrimp and nematodes. However, it seems that the light availability
in the tanks are the main controlling factor of microalgae abundance. As an example,
highest ammonium concentration measured in containers with biofilm and high shrimp
Ž.
density treatment 2HD in experiment 2 was probably caused by the mortality of
biofilm diatoms at the bottom of the containers, due to the accumulation of faeces and
uneaten food that precluded light from reaching the lowest part of the tanks.
The fact that a biofilm effectively absorbs or transforms ammonium present in the
water column has important implications for the health of F. paulensis juveniles, since
Ž.Ž.
this shrimp tolerates high amounts of nitrate )15000 mM and nitrite )1000 mM
Ž. Ž
Cavalli et al., 1996 , but high ammonium concentrations are lethal Ostrensky and
.Ž
Wasielesky, 1995 , or may seriously inhibit their food intake and growth Miranda Fo,
.
1997; Wasielesky et al., 1994 .
Biofilms have been considered as reservoirs of pathogenic bacteria, like Vibrio
Ž.
harÕeyi, which can affect shrimp cultures Karunasagar et al., 1996 . Pathogenic bacteria
present in biofilms are difficult to eliminate, even using great amounts of antibiotics
Ž.
Costerton et al., 1999 . Therefore, the asepsis of culture tanks by cleaning off the
Ž.
biofilm is normally recommended Karunasagar et al., 1996; Austin and Austin, 1999 .
However, our results indicate that nitrogen uptake by a biofilm may help to reduce the
occurrence of pathogenic bacteria, since these microorganisms normally occur in
Ž
situations where nitrogenous compounds reach extremely high values Austin and
.
Austin, 1999; Brock and Main, 1994 . Moreover, many microalgae present in biofilms
Ž
are able to produce antibiotics that prevent pathogenic bacterial growth Austin and Day,
.
1990; Alabi et al., 1999 . Protozoa that inhabit biofilms could also control the abundance
Ž.
of pathogenic bacteria through grazing Thompson et al., 1999 . Thus, it is possible that,
contrary to the expected effect, biofilm removal could increase the risk of developing
pathogenic bacteria.

()
F.L. Thompson et al.rAquaculture 203 2002 263–278276
Results of experiment 5 show that tanks with biofilm export less phosphorus. Total
nitrogen load exported by tanks with and without biofilm were, on the other hand,
similar. However, the amount of ammonium exported from tanks with biofilm was
smaller than that without a mature biofilm. Eutrophication can be accelerated if the main
form of exported nitrogen is ammonium. This happens because primary producers use
less energy to incorporate this N source into amino acids and proteins, while nitrate has
Ž
to be transformed inside the cells into ammonium, with some energy costs Syrett,
.
1981 . Therefore, autotrophic cells grow more rapidly in the presence of ammonium
than nitrate. Thus, the presence of biofilm would reduce eutrophication in the water
bodies that receive aquaculture effluents by effective phosphate uptake, or slow this
process by the release of more nitrate, instead of ammonium.
Ž
Biofilm has already been considered an important food source for Daphnia Langis et
.Ž .Ž
al., 1988 , Nile tilapia Shrestha and Knud-Hansen, 1994 and carp Ramesh et al.,
.
1999 . Despite the low protein contents measured in the biofilm of this study, microor-
ganisms present in the tanks may supply essential elements to cultured shrimp, such as
Ž
polyunsaturated fatty acids, sterols, amino acids, vitamins, and carotenoids Thompson
.
et al., 1999 . In this way, the biofilm probably contributed to the increase of weight and
total biomass of F. paulensis juveniles measured in our experiments.
Finally, the presence of a biofilm reduces the necessity of water exchange, which
certainly decreases the costs of shrimp production. It was also demonstrated in other
studies that reduction, or even suppression of water exchange, did not cause any damage
Ž.
to the cultured organisms Hopkins et al., 1995; McIntosh et al., 2000 .
In conclusion, our results point out the importance of biofilm for the improvement of
water quality and as a complementary food source for reared F. paulensis juveniles.
However, further studies are needed in order to evaluate the biofilm efficiency in
large-scale culture systems, and how the different trophic interactions among microor-
ganisms may affect the nitrogen cycle in these ecosystems.
Acknowledgements
We would like to thank Sandro Fabres for his technical support during the develop-
ment of the experiments. We also acknowledge the contribution of J. Vandenberghe, J.
Muelbert, U. Seeliger, W. Graneli and three anonymous referees for their suggestions on
´
a previous version of this manuscript. F. Thompson and P.C. Abreu had financial
support from the Conselho Nacional de Desenvolvimento Cientıfico e Tecnologico-
´´
CNPq, Brazil. This study had the financial support from the Fundac¸ao de Amparo a
ˆ
Pesquisa do Estado do Rio Grande do Sul-FAPERGS.
References
Alabi, A.O., Cob, Z.C., Jones, D.A., Latchford, J.W., 1999. Influence of algal exudates and bacteria on growth
Ž.
and survival of white shrimp larvae fed entirely on microencapsulated diets. Aquacult. Int. 7 3 , 137–158.
Ž.
Association of Official Analytical Chemists AOAC , 1995. Official Methods of Analysis of AOAC
International. Microchemical Determination of Nitrogen, vol. 1, American Chemical Society, USA, p. 7.

()
F.L. Thompson et al.rAquaculture 203 2002 263–278 277
Austin, B., Austin, D., 1999. Bacterial Fish Pathogens: Disease of Farmed and Wild Fish. 3rd edn. Springer,
Chichester, 457 pp.
Austin, B., Day, J.G., 1990. Inhibition of prawn pathogenic Vibrio spp. by a commercial spray-dried
preparation of Tetraselmis suecica. Aquaculture 90, 389–392.
Brock, J.A., Main, K.L., 1994. A Guide to the Common Problems and Diseases of Cultured Penaeus
Õannamei. The World Aquaculture Society, Baton Rouge, 242 pp.
Caron, D.A., 1994. Inorganic nutrients, bacteria and the microbial loop. Microb. Ecol. 28, 295–298.
Cavalli, R.O., Wasielesky Jr., W., Franco, C.S., Miranda Filho, C., 1996. Evaluation of the short-term toxicity
Ž.
of ammonia, nitrite and nitrate to Penaeus paulensis CRUSTACEA, DECAPODA Broodstock. Arq.
Ž.
Biol. Tecnol. 39 3 , 567–575.
Costerton, J.W., Stewart, P.S., Greenberg, E.P., 1999. Bacterial biofilms: a common cause of persistent
infections. Science 284, 1318–1322.
Ž.
Decho, A.W., 1990. Microbial exopolymer secretions in ocean environments: their role s in food webs and
marine processes. Oceanogr. Mar. Biol. Annu. Rev. 28, 73–153.
Hobbie, J.E., Daley, R.J., Jasper, S., 1977. Use of nuclepore filters for counting bacteria by fluorescence
microscopy. Appl. Environ. Microbiol. 3, 1225–1228.
Hopkins, J.S., Sandifer, P.A., Browdy, C.L., 1995. A review of water management regimes which abate the
Ž.
environmental impacts of shrimp farming. In: Browdy, C.L., Hopkins, J.S. Eds. , Swimming Through
Troubled Water. Proceedings of the Special Session Shrimp Farming, The World Aquaculture Society,
Baton Rouge, pp. 157–166.
Kaiser, G.E., Wheaton, F.W., 1983. Nitrification filters for aquatic culture systems: state of the art. J. World
Maric. Soc. 14, 302–324.
Karunasagar, I., Otta, S.K., Karunasagar, I., 1996. Biofilm formation by Vibrio harÕeyi on surfaces.
Aquaculture 140, 241–245.
Langis, R., Proulx, D., Noue, J., Couture, P., 1988. Effects of bacterial biofilm on intensive Daphnia culture.
¨
Aquacult. Eng. 7, 21–38.
McIntosh, D., Samocha, T.M., Jones, E.R., Lawrence, A.L., McKee, D.A., Horowitz, S., Horowitz, A., 2000.
The effect of a commercial bacterial supplement on the high-density culturing of Litopenaeus Õannamei
Ž.
with a low-protein diet in an outdoor tank system and no water exchange. Aquacult. Eng. 21 3 , 215–227.
Ž.
Meyer-Reil, M., 1994. Microbial life in sedimentary biofilms—the challenge to microbial ecologists. Review
Mar. Ecol.: Prog. Ser. 112, 303–311.
Miranda Fo, K.C., 1997. Efeito da amonia na sobrevivencia e crescimento de juvenis de camarao-rosa
ˆˆ ˜
Ž.Ž
Penaeus paulensis Perez-Farfante, 1967 CRUSTACEA: DECAPODA Effect of ammonium on the
´Ž
survival and growth of the pink-shrimp Penaeus paulensis Perez-Farfante, 1967 CRUSTACEA: DE-
´
..
CAPODA juveniles . MSc Thesis, Fundac
¸ao Universidade do Rio Grande, Rio Grande, RS, Brazil, 122
˜
pp.
Ostrensky, A., Wasielesky Jr., W., 1995. Acute toxicity of ammonia to various life stages of Sao Paulo shrimp,
˜
Penaeus paulensis Perez-Farfante, 1967. Aquaculture 132, 339–347.
´
Ramesh, M.R., Shankar, K.M., Mohan, C.V., Varghese, T.J., 1999. Comparison of three plant substrates for
enhancing carp growth through bacterial biofilm. Aquacult. Eng. 19, 119–131.
Round, F.E., Crawford, R.M., Mann, D.G., 1990. The Diatoms: Biology and Morphology of the Genera.
Cambridge Univ. Press, Cambridge, 747 pp.
Shrestha, M.K., Knud-Hansen, C.F., 1994. Increasing attached microorganism biomass as a management
Ž.
strategy for Nile Tilapia Oreochromis niloticus production. Aquacult. Eng. 13, 101–108.
Sokal, R.R., Rohlf, F.J., 1969. Biometry. The Principles and Practices of Statistics in Biological Research.
Freeman, San Francisco, 776 pp.
Strickland, J.D.H., Parsons, T.R., 1972. A Practical Handbook of Seawater Analysis. Fish. Research Board of
Canada, Ottawa, 310 pp.
Ž.
Syrett, P.J., 1981. Nitrogen metabolism of microalgae. In: Platt, T. Ed. , Physiological Bases of Phytoplank-
ton Ecology. Can. Bull. Fish. Aquat. Sci. Department of Fisheries and Oceans, Ottawa, Canada, pp.
182–210.
Thompson, F.L., Abreu, P.C., Cavalli, R.O., 1999. The use of microorganisms as food source for Penaeus
paulensis larvae. Aquaculture 174, 139–153.
Ž.
Tomasso, J.R., 1994. Toxicity of nitrogenous wastes to aquaculture animals. Rev. Fish. Sci. 2 4 , 291–314.

()
F.L. Thompson et al.rAquaculture 203 2002 263–278278
UNESCO, 1983. Chemical methods for use in marine environmental monitoring. Intergovernmental Oceano-
graphic Commission. Manual and guides, 12, Paris.
Utermohl, H., 1958. Zur vervollkommnurg der quantitativen phytoplankton methodik. Int. Ver. Theor. Angew.
¨
Limnol. 9, 1–38.
Wasielesky Jr., W., Marchiori, M.A., Santos, M.H., 1994. Efeito da amonia no crescimento de pos-larvas do
ˆ´
Ž.w
camarao rosa, Penaeus paulensis, Perez Farfante, 1967 DECAPODA: PENAEIDAE . Effect of ammo-
˜´ Ž
nium on the growth of the pink-shrimp Penaeus paulensis, Perez Farfante, 1967 DECAPODA: PENAEI-
´
.x
DAE post larvae Nauplius 2, 99–105.
Whal, M., 1989. Marine epibiosis I. Fouling and antifouling: some basic aspects. Mar. Ecol.: Prog. Ser. 58,
175–189.
Wheaton, F.W., 1977. Aquacultural Engineering. Wiley-Interscience, New York, 708 pp.
Whitford, L.A., Schumacher, G.J., 1973. A Manual of Fresh-Water Algae. Sparks Press, Raleigh, 322 pp.