See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/317786078

Effects of biofloc technology on reproduction and ovarian recrudescence in

Nile tilapia

Article in Aquaculture Research · June 2017

DOI: 10.1111/are.13420

CITATIONS READS

14 769

6 authors, including:

Érika Alvarenga Cláudia Regina dos Santos

Federal University of Minas Gerais Federal University of Minas Gerais
47 PUBLICATIONS 858 CITATIONS 2 PUBLICATIONS 30 CITATIONS

SEE PROFILE SEE PROFILE

Eduardo Maldonado Turra

Federal University of Minas Gerais
43 PUBLICATIONS 760 CITATIONS

SEE PROFILE

All content following this page was uploaded by Érika Alvarenga on 01 November 2017.

The user has requested enhancement of the downloaded file.

Received: 19 January 2017 | Revised: 6 May 2017 | Accepted: 13 May 2017

DOI: 10.1111/are.13420

ORIGINAL ARTICLE

Effects of biofloc technology on reproduction and ovarian

recrudescence in Nile tilapia


Erika
lio Soares de Brito |
Ramos de Alvarenga | Suellen Cristina Moreira de Sales | Tu
Cl
audia Regina Santos | Rebeca Dias Serafim Corre
^a |
Gabriel Francisco de Oliveira Alves | Ludson Guimar~
aes Manduca |
Eduardo Maldonado Turra

rio de Aquacultura (LAQUA),

Laborato
Escola de Veterin

aria da Universidade Abstract
Federal de Minas Gerais, Belo Horizonte, The objectives of this study were to evaluate the reproductive cycle, morphological
MG, Brazil
changes of ovary and mobilization of energy reserves in Nile tilapia reared with biofloc
Correspondence technology (BFT). In general, the growth and reproductive performance were highly
Eduardo Maldonado Turra, Laborato
rio de
Aquacultura (LAQUA), Escola de Veterin
aria similar between BFT and Control system (clear water). Difference between the sys-
da Universidade Federal de Minas Gerais, tems was found in the hepatosomatic index (using mixed-effects models), which sug-
Belo Horizonte, MG, Brazil.
Email: [email protected] gested that BFT can alter the energy mobilization in the post-spawning period. The
absolute and relative fecundity, fertilization rate, number of larvae produced per
Funding information
Conselho Nacional de Desenvolvimento female, gonadosomatic index, proportion of oogenesis cells, number of post-ovulatory

gico (CNPq); Fundacß~ao
Cient
ıfico e Tecnolo and atretic follicles were similar between the two systems. We also did not detect a
de Amparo a Pesquisa do Estado de Minas
Gerais (FAPEMIG); Pro
-Reitoria de Pesquisa reduction in the reproductive cycle length in Nile tilapia reared in BFT. Because there
da UFMG (PRPQ-UFMG) was no evidence of the negative effects of BFT on Nile tilapia reproduction, we con-
cluded that BFT might be used for breeder stocking of this species.

KEYWORDS
biofloc technology, Oreochromis niloticus, ovary morphometry, post-spawn period

1 | INTRODUCTION ammonia) into microbial biomass (bioflocs) is maximized and the

resulting bioflocs can be utilized back by the cultured organisms as a
The major problems for seed production of Nile tilapias Oreochromis food source (Avnimelech, 2009). Due the nutrient recycle, the BFT
niloticus (Linnaeus, 1758) are the low egg production during spawning has recently gained attention as an economic and environmentally
and a lack of spawning synchrony (Bhujel, 2000). To overcome these sustainable system, especially for the main filter feeders species in
issues, a variety of management strategies have been tested and recom- aquaculture, including tilapias and shrimps. In addition to improving
mended (Campos-Mendoza, McAndrew, Coward & Bromage, 2004; El- the feed efficiency, recent studies have demonstrated the effect of
Sayed & Kawanna, 2008; Little, Coward, Bhujel, Pham & Bromage, 2000; BFT on fish and shrimp reproduction (Ekasari et al., 2016; Emeren-
Ridha & Cruz, 1999; Tsadik & Bart, 2007; Yoshida, Oliveira, Kunita, Riz-
valo & Gaxiola, 2014). For example, the relative
ciano, Cuzon, Are
zato & Ribeiro, 2015). Beyond management strategies, the choice of pro- fecundity of African catfish Clarias gariepinus (Burchell, 1822) brood-
duction system may contribute to improving the reproductive stock in BFT was 26% higher than that of the control on day 122 of
performance of this species (Ekasari et al., 2015). For example, Ekasari experiment (Ekasari et al., 2016); the shrimp Farfantepenaeus duo-
et al. (2015) showed that biofloc technology (BFT) improved the repro- rarum (Burkenroud, 1939) raised in BFT achieved better spawning
ductive performance of tilapia and increased the number of fry produced. performance in terms of number of spawns per ablated female, num-
The BFT is an aquaculture system based on limited water ber of eggs per spawn, relative fecundity and egg size as compared
exchange, where microbial conversion of nutrient waste (mainly clear-water system (Emerenciano et al., 2014).

Aquaculture Research. 2017;1–8. wileyonlinelibrary.com/journal/are © 2017 John Wiley & Sons Ltd | 1
2 | ALVARENGA ET AL.

Despite the results of these studies, a more comprehensive view molasses was added to the BFT tanks when the total ammonium
1
of the effects of BFT on the reproductive parameters of fish has not nitrogen concentration exceeded 0.5 mg L . The carbon content of
been conducted to date. How does BFT affect the reproductive this compound was considered to calculate the amount added to the
parameters? Could the increase in the egg mass be caused by faster system. The carbon input was determined according to Ebeling, Tim-
ovary recrudescence in BFT? Could BFT be a suitable system for mons and Bisogni (2006) employing a C:N ratio of 6:1 based on the
breeder stocking? Could the dynamics of energy mobilization in the total ammonia nitrogen (TAN). To control of solids in BFT, the water
post-spawning period be changed in tilapia reared in BFT? To renewal rate was 5% per day.
answer these questions, we developed a model to study tilapia In Control, the animals were maintained in a clear-water system
reproductive cycle and evaluated the effects of BFT on ovarian with water renewal rate of 50% per day to maintain water quality.
recrudescence as well as the reproductive and growth performance To avoid different temperatures between treatments, the water
of female Nile tilapia breeders. renewal was made with water that was previously heated (27–28°C)
in both systems.
1
Sodium chloride was added until a 2 g L concentration was
2 | MATERIALS AND METHODS
reached to reduce nitrite toxicity in all the tanks. A pH correction
was performed with calcium hydroxide (Ca(OH)2) when the pH was
2.1 | Location
below 6.5.
The experiments were conducted at the Aquaculture Laboratory of
the Veterinary School of Federal University of Minas Gerais (UFMG),
2.3 | The experimental model and procedures
Brazil. All procedures were previously approved by the Ethics Com-
mittee on Animal Use of UFMG under protocol number 208/2014. To evaluate the system effects, we needed to compare animals in
the same stage of reproductive cycle. However, the Nile tilapia has
an asynchronous gonad development (Melo et al., 2014). In a tank,
2.2 | Breeders stocking and experimental
for example, there are fish in different gonad development stages.
conditions
For that reason, the synchronization of reproductive cycle of Nile
Adult Nile tilapias of the Chitralada strain were tagged with a unique tilapia and the determination of the exact moment of spawning were
passive integrated transponder device for individual recognition. For necessary.
this experiment, the female tilapias were reared for one month (ap- To synchronize the reproduction of this species, we used a hor-
proximate duration of a reproductive cycle) in BFT and the Control monal protocol (as described by Fernandes et al., 2013). Females
system before spawning induction. To analyse the ovary recrudes- that presented an enlarged and reddish genital papilla and cambered
cence, these animals were reared in both systems 21 days more ventral region (i.e., “ready to spawn”) were chosen from each system
after spawning induction. The experimental conditions are detailed for spawning induction. Next, we weighed the animals and identified
below. them by microchip number. These females considered in similar
3
In both systems, the animals were stocked in 1-m fibre glass stage of reproductive cycle were injected intramuscularly with two
tanks (female and male kept in different tanks), which were main- doses of human chorionic gonadotropin (hCG) (Chorulonâ Intervet
tained in a greenhouse. The initial stoking density was 32 animals/ Schering-Plough Animal Health, Netherland) (first dose = 10% of
tank (6.5 kg/m3). For females, we used 96 animals in BFT and total dose, total dose = 3000 IU/kg; dose interval of 18 hr). The
another 96 tilapias in the Control system. Males were used only for stripping was performed 24 hr after the second dose (according to
semen collection and were reared in a tank. Aeration was provided procedure used by Fernandes et al., 2013).
using an air blower. The fish were fed two times a day with 4- to 6-
mm commercial pellets containing 32% protein at a feeding rate of
2.4 | Eggs and larvae production
1.5% body weight.
A daily assessment of dissolved oxygen, pH and salinity was con- We compared the eggs and larvae production of breeders from dif-
ducted using the multiparameter probe YSI Model 6920 V2â (Yellow ferent systems. For this purpose, one semen pool was used to fertil-
Springs Incorporated—YSI, OH, USA). Three times a week, the total ize the oocytes from Control and BFT females.
ammonia nitrogen (TAN) and nitric nitrogen (N-NO2 ) concentrations We distributed the sperm over the oocytes, water from the
were measured by spectrophotometry according to the methodology hatching system was added, and the eggs were transferred to an
generated by UNESCO (1983) and Bendschneider and Robinson incubator. The eggs were separated and maintained in incubators
(1952) respectively. The total settleable solids (TSS) were determined with clear water at a temperature of 28°C.
weekly with an Imhoff Cone (Avnimelech, 2009). Alkalinity and total After fertilization, the total egg volume and a 1.0 ml sample of
suspended solids were evaluated weekly according to APHA (1989) eggs from each female were obtained. The samples were fixed in
and Strickland and Parsons (1972) respectively. formaldehyde and used to estimate the egg production (i.e., the
In BFT treatment, the tanks were completely filled with mature number of eggs per ml multiplied by total egg volume). Subse-
biofloc previously development in our laboratory. Sugarcane quently, the relative fecundity (number of eggs per gram of body
ALVARENGA ET AL. | 3

weight) was obtained for each female. To obtain the fertilization

2.6 | Growth performance
rate, a sample of 50 eggs was collected from each female and anal-
ysed under a dissecting microscope according to the methods of To evaluate whether BFT affects the growth performance of female
Fernandes et al. (2013). To estimate the hatchability, we counted breeders, we measured the initial and final individual weight, survival
the number of hatched larvae in proportion to the number of eggs rate, initial and final stocking density and the daily weight gain. The
that were estimated. To validate this estimate, we counted two hun- final individual weight (g) was calculated by dividing the final biomass
dred eggs from each female, separated the eggs into a different (g) by the final number of animals per tank. The survival rate (SR)
incubator and compared the results of the hatchability rate between was determined as SR = (final number of animals/initial number of
the counted and estimated samples. animals) 9 100. The initial and final density (kg m ³) was obtained
by dividing the biomass (kg) by the volume in cubic metre (m³) at the
beginning and at the end of experiment respectively. The daily
2.5 | Ovary recrudescence
weight gain (g) was obtained by dividing the individual biomass
To evaluate the ovary recrudescence, after spawning, the hormon- increment (mean final weight – mean initial weight) by the number
ally induced females were allocated again to their original systems. of days the animals remained in the experiment.
These animals were identified by microchip number, were weighed
to obtain body weight (BW) and were euthanized 0, 7, 14 and
2.7 | Statistical analyses
21 days after spawning. To obtain the suitable number of females,
this experiment was performed in two blocks in time. The experi- The data of egg and larvae production did not follow a normal distri-
mental conditions in each block were very similar and described bution according to modified Shapiro–Wilk test and were analysed
above. Animals used in the first block were not used in the sec- using the Mann–Whitney test and Fisher’s exact test (p < .05).
ond one. In each block, twelve tilapias from each system posi- Spearman’s correlation (p < .05) was used to correlate the results of
tively responded to hormonal induction and spawned. Thus, in the hatchability rate between the counted and estimated samples.
each block, three females of each system were collected each The number of AO and POF per area of the total number of cells
time (0, 7, 14 and 21 days after spawning), totalizing 24 females analysed did not present a normal distribution according to modified
for each system. The viscera, ovaries and livers were collected Shapiro–Wilk test and were analysed with Kruskal–Wallis test
separately and weighed, and the visceral somatic index (VSI = vis- (p < .05). For evaluating the growth performance, an ANOVA was
cera weight/BW 9 100), gonadosomatic index (GSI = ovary applied (p < .05), except for the survival rate, which was evaluated
weight/BW 9 100) and hepatosomatic index (HSI = liver weight/ using Fisher’s exact test (p < .05). These statistical analyses were
BW 9 100) were calculated. conducted using the InfoStat program (Di Rienzo et al., 2015).
A central fragment of ovaries from all 48 euthanized females was As explained before, to obtain the suitable number of females
immersed in Bouin’s fixative for 24 hr, transferred to 70% alcohol for analysis of ovary recrudescence, this experiment was performed
and subjected to routine histological techniques (i.e., dehydration in two blocks in time. Therefore, the data of water quality parame-
through a graded series of ethanol, cleaning in xylol, paraffin embed- ters were analysed in randomized blocks and compared by analysis
ding, 5-lm sections and staining with haematoxylin–eosin). of variance (ANOVA) (p < .05) using InfoStat program. An analysis of
We analysed the number and the area of oogonia and oocytes in the relationship between the system, time and the influence of the
primary and secondary growth. We also counted the number of atre- system over specific indexes (GSI, HSI, IVS) and oogenesis cells was
tic oocytes (AO) and post-ovulatory follicles (POF) and determined conducted using restricted maximum likelihood in a linear mixed-
the number of AO or POF per area of the total number of cells anal- effects model that considered the experiment block as a random
ysed. The greater the number of sections that were analysed, the effect and system and time as fixed effects (Pinheiro & Bates, 2000).
more stable and reliable the count will be. However, the analysis of This analysis was conducted using the lmer function in the R/lme4
a high number of sections could make the count impracticable. package. Correlation and covariance structure were chosen
Therefore, to determine the number of sections necessary to achieve according to the best Bayesian information criterion (BIC).
a stable and reliable count, we analysed twelve sections from four
animals. Each section was at least 80 lm apart from the other sec-
3 | RESULTS
tions (in fact, the distance between two sections was larger than
80 lm in most of the cases, due the pursuit of high quality sections).
3.1 | Water quality parameters
The 80 lm distance ensured that the oogonias and primary oocytes
(the most instable cells in the count) did not appear in two consecu- All water quality parameters that were analysed were within the rec-
tive sections. After this analysis, we found that it was necessary to ommended range for normal development of tilapia (Table 1). Control-
have at least four sections for an accurate count, and this number ling water temperature in the experiment ensured the temperature
was used for histological analysis of all 48 females used for our was consistent between treatments. Although the dissolved oxygen
ovary recrudescence study. The number of ovary cells was deter- level was lower in BFT as compared to control, it remained above
1
mined as a percentage. 5.0 mg L in both treatments, which was the level suggested by Boyd
4 | ALVARENGA ET AL.

and Tucker (1998) to not limit the growth performance of fish. In addi- spawning) was observed for all biological indices, whereas the sys-
tion, other water parameters, including pH, alkalinity, TAN and nitric tem effect was detected for just HSI. For GSI and HSI, we noticed
nitrogen concentrations, salinity, TSS and dissolved solids, were signifi- that the cubic term (time) was significant, but the linear and quadra-
cantly different between the treatments. These differences occurred tic terms were not. The linear and quadratic terms (time) were signif-
due to the different features of each system. icant for VSI. The lower values (means: BFT = 2.1; Control = 1.3) of
GSI occurred one week after the spawning. GSI increased during the
2nd week (means: BFT = 3.58; Control = 3.36) and the values stabi-
3.2 | Eggs and larvae production
lized during the 3rd week (means: BFT = 3.23; Control = 2.74).
The percentage of females that were considered “ready to spawn” Immediately after spawning, the HSI values were 1.98 and 1.68% for
was similar between the different treatments (Table 2). Data about the BFT and Control groups respectively. The HSI increased one
fecundity (number of eggs/gram of body weight), fertilization rate week after spawning (means: BFT = 3.07; Control = 2.30), and then,
and hatchability from the females that were hormone-induced to HSI was reduced and stabilized in the 2nd (means: BFT = 2.33; Con-
spawn were not significantly different. A Spearman correlation trol = 1.86) and 3rd weeks (means: BFT = 2.85; Control = 2.2). VSI
between the hatchability obtained by count and by estimate was was more stable than the other indexes, and in general, the VSI val-
0.94, which indicated that the protocol used for the hatchability ues were lower immediately after spawning.
estimate was suitable. The analyses of ovary cell percentage (%) for different times and
systems using mixed-effects models are presented in Table 4. For
primary and secondary oocyte proportions, we noticed that the lin-
3.3 | Ovary recrudescence
ear term (time) was significant, and their values decreased and
The analyses of biological indices using the mixed-effects models are increased over time respectively. System and time effects were not
given in Figure 1 and Table 3. The time effect (i.e., days after detected for oogonia.
We did not find differences between the systems for the number
of POF and AO per area of ovary cells (lm2) (Table 5). However, we
T A B L E 1 Water quality parameters (mean) and their coefficients
noticed differences in these variables during the post-spawning
of variation (CV) in the Control and BFT systems
period.
Water parameters Control BFT CV (%) References
Temperature (T°C) 29.29 29.08 6.08 27–32a
T (8 am) 27.55 27,30 7,15
3.4 | Growth performance
T (4 pm) 31.05 30.94 4.90 The results regarding growth and survival are given in Table 6. The
Dissolved oxygen 6.44 6.14* 6.42 >4b growth performance of tilapias was highly similar between the two
(DO mg L 1) systems, although daily weight gain was better in the Control sys-
DO (8 am) 6.68 6.48 6.34 tem. However, these results were not considered conclusive because
DO (4 pm) 6.09 5.66* 6.53 the subtly higher survival rate in the BFT system could explain the
pH 7.19 6.42* 2.62 6–9d results.
pH (8 am) 6,99 6.42* 2.58
pH (4 pm) 7.65 6.69* 3.48
Alkalinity (mg de 82.86 41.92* 33.06 >20c 4 | DISCUSSION
CaCO L 1)
Salinity (g L 1) 1.72 0.36* 18.50 <30b Moving towards a more comprehensive view of the effects of
TSS (ml L 1
) 0.91 24.32* 13.90 30–40 d BFT on Nile tilapia female breeders, this study presented a data

Total dissolved 77.5 549.5* 28.80 <1,000 d set including reproductive performance, ovary morphometry, bio-
solids (mg L 1) logical indices and parameters for growth. Ekasari et al. (2015)
Total ammonia 0.44 0.26* 107.0 <1d suggested that BFT enhanced tilapia reproductive performance
1
nitrogen (mg L ) because the system likely offered a positive nutritional effect and
Unionized NH3 0.0035 0.0005* 107.0 <0.1a better water quality. The focus of this study was to evaluate how
(mg L 1) BFT affects physiological factors, including the duration of the
Nitric nitrogen 1.19 0.27* 95.0 <8a reproductive cycle, morphological changes and mobilization of
(mg L 1)
energy reserves.
a
El-Sayed (2006). Except for temperature, all water parameters were significantly
b
Nile tilapia reproduction stopped completely at salinity above 30 (mg different between the systems. However, the water quality was
L 1).
c appropriate in both the Control and BFT systems, and the parameter
Wedemeyer (1996).
d
Avnimelech (2009). values were within the adequate ranges for tilapia reproduction. A
*Values differ by ANOVA (p < .05). higher concentration of total ammonia and nitric nitrogen in the
ALVARENGA ET AL. | 5

T A B L E 2 Hatching data for females in the Control and BFT groups

Control BFT

% ready-to-spawn females* 58.5 62.5

Median Min. Max. Median Min. Max.

Number of eggs/gram of female** 4.78 1.52 9.70 4.24 1.59 6.8
Number of eggs/female** 1,137 375 2,750 1,065 341 2,356
Fertilization rate** 82.5 0 98 84.9 59 100
a,
Larvae/spawned female ** 0 0 909 5 0 519
Larvae/femaleb,** 232 101 909 207 106 519
Hatchability (%)a,** 0 0 51.8 0.55 0 22
b,
Hatchability (%) ** 37.4 8.7 51.8 18.1 11.7 22
a
All females were used in the statistical analysis.
b
Only females that produced more than 10 larvae were used in the statistical analysis.
*Values were not different according to Fisher’s exact test (p > .05).
**Values were not different according to the Mann–Whitney test (p > .05).

F I G U R E 1 Mixed-effects models for gonadosomatic (GSI), hepatosomatic (HSI) and viscerosomatic (VSI) indexes of females in the BFT and
Control systems during ovary recrudescence

T A B L E 3 Parameters of the mixed-effects models for biological indices

Response Intercept Standard error System coef System x time Time-coef. linear Time-coef. quadratic Time-coef. cubic
GSI 2.58*** 0.24 – – – – 1.43**
HSI 2.01*** 0.08 0.54** – – – 0.41*
VSI 10.48*** 0.38 – – 1.63* –1.85** –

*p-value ≤ .05.
**p-value ≤ .005.
***p-value ≤ .0001.

Control system could explain a subtle reduction (although not signifi- found effects of post-spawning time on HSI, GSI and VSI. A system
cant) in survival in this system compared to BFT. effect was found for HSI (using mixed-effects models). The analysis
All of the studied reproductive parameters were similar of biological indices allowed us to examine some hypotheses: (1)
between the Control and BFT systems. Ekasari et al. (2015) found BFT presented little influence on energy mobilization to reproduc-
differences between these systems for absolute fecundity (number tion because GSI was similar between the systems. Alternatively, (2)
of eggs/female) with higher values in the BFT system. However, BFT favoured energy mobilization to reproduction, but because the
they also found a higher average body weight in the BFT system. tilapia was reared in BFT during the only one reproductive cycle, this
These data could suggest that the effects of BFT on tilapia repro- period was not long enough to see a BFT effect. Corroborating the
duction were indirect and were related to better growth, which second hypothesis, the late effects of BFT on reproduction were
was a higher possibility than direct physiological mechanism of bio- also observed by Ekasari et al. (2016). These authors found similar
floc on reproduction. relative fecundity in African catfish (Clarias gariepinus) reared in BFT
The dynamics of energy mobilization during the post-spawning and Control systems on culture days 56 and 86, whereas this param-
period promoted changes in the biological indices. As expected, we eter was 26% higher in the BFT group on day 122. Therefore, more
6 | ALVARENGA ET AL.

T A B L E 4 Parameters for the mixed-effects models for the proportion of oogenesis cells
Response (%) Intercept Standard error System coef System x time Time-coef. linear Time-coef. quadratic Time-coef. cubic
Oogonia 0.09*** 0.010 – – – – –
Primary oocyte 0.67*** 0.032 – – 0.14** – –
Secondary oocyte 0.22*** 0.020 – – 0.13* – –

*p-value ≤ .05.
**p-value ≤ .005.
***p-value ≤ .0001.

T A B L E 5 Number of post-ovulatory follicles (POF) and atretic oogenesis cells respectively. The post-spawning time presented a lin-
oocytes (AO) per lm2 during gonadal recrudescence in Nile tilapia ear effect on these variables with an increase in the proportion of
reared in BFT and Control systems secondary oocytes and a decrease in primary oocytes. These
POF (number/lm2) AO (number/lm2) changes occur due to the differentiation of primary to secondary
Treatment N Median* Min. Max. Median* Min. Max. oocytes and preparation of ovaries for a new reproductive cycle
BFT t = 0 6 0.250 ab
0.000 2.950 0.67 a
0.11 1.14 ~o, 2009). Because we
(Dietrich & Krieger, 2009; Grier, Uribe & Patin

BFT t = 7 6 0.017abc 0.000 1.200 0.26ab 0.09 3.23 did not find differences in the proportion of oogenesis cells, the BFT
apparently did not increase the recruitment of these cells to a next
BFT t = 14 6 0.000 c
0.000 0.000 0.08 abc
0.02 0.38
reproductive cycle.
BFT t = 21 6 0.000ac 0.000 0.010 0.39abc 0.05 1.01
The post-ovulatory follicle consists of follicular layers that
Control t = 0 6 1.170b 0.000 2.880 0.64a 0.12 2.84
remained in the ovaries of fish after the release of mature oocytes
Control t = 7 6 0.000abc 0.000 0.630 0.57a 0.17 6.75
during spawning. The occurrence of POF is evidence of spawning
Control t = 14 6 0.000ac 0.000 0.040 0.07bc 0.06 0.12
and its value can reflect successful spawning (Grier et al., 2009). In
Control t = 21 6 0.000ac 0.000 0.049 0.19abc 0.06 0.52
this study, we did not find a difference in the POF number or reab-
*Values with different letters in the same column are significantly sorption rate between systems. A higher number of these structures
different according to the Kruskal–Wallis test (p < .05). were found on the spawning day, but they virtually disappeared
after seven days in both the BFT and Control systems.
T A B L E 6 Growth performance of tilapia in the Control and BFT The follicular atresia is the process of degeneration and removal
systems of ovarian follicles from the ovary. This process can be natural,
Control BFT CV (%) which permits the recovery and recycling of oocyte components, or

Initial number (fish tank 1

) 32 32 induced by stressors such as hormonal disruption or toxic substances

Initial individual weight (g)** 163.7 163.7 35.1

(Carrillo, Zanuy & Bayarri, 2009; Grier et al., 2009; Schreck, Contr-
eras-Sanchez & Fitzpatrick, 2001). Therefore, the increase in follicu-
Final individual weight (g)** 277.52 257.91 25.9
lar atresia could indicate poor conditions for reproduction. The
Survival rate (%)*** 95.3 98.4
3
number of atretic oocytes was very similar between systems and
Initial stocking density (kg.m )** 6.5 6.5
was apparently normal. In both systems, higher AO occurred in sec-
3
Final stocking density (kg.m )** 8.7 8.6
ondary oocytes in the first seven days, decreased during the second
1
Daily weight gain (g day )* 1.8 1.46 43
week and increased again in the third week. Thus, there was no evi-
*Values differed according to ANOVA (p < .05). dence of negative effects of BFT on Nile tilapia reproduction.
**Values were not different according to ANOVA (p > .05). In general, the growth performance of tilapia female breeders
***Values were not different according to Fisher’s exact test (p > .05).
was highly similar between the systems. Differences were observed
only for daily weight gain, which was higher in the Control system.
studies are necessary to confirm the late effects of BFT on Nile tila- This outcome could be an indirect effect of subtle mortality in the
pia reproduction and to evaluate the contribution of this system to Control system during the first week of the experiment, which
yields in seed production units. reduced the density and promoted better growth for the animals
GSI changes and the proportion of oogenesis cells were similar reared in this system. In addition, for daily weight gain, the coeffi-
between the systems. Therefore, we did not detect a reduction in cient of variation (CV) was unusually higher as compared to CVs
reproductive cycle length in Nile tilapia reared in BFT. As expected, obtained from monosex male Nile tilapia population. This probably
the proportion of oogonia did not change during the reproductive occurred because we considered all females that were in BFT and
cycle, and this cellular type presented a small proportion of oogene- Control systems to obtain the mean, including females that spawned
sis cells that was smaller than 10% of the total cells. Primary and and those did not spawn. The data on growth performance in BFT
secondary oocytes represented approximately 70% and 20% of are controversial. Some authors described lower growth performance
ALVARENGA ET AL. | 7

rez-Fuentes,
of Nile tilapia reared in BFT (Azim & Little, 2008; Pe Ebeling, J. M., Timmons, M. B., & Bisogni, J. J. (2006). Engineering analy-
Hern
andez-Vergara, Pe

rez-Rostro & Fogel, 2016). On the other sis of the stoichiometry of photoautotrophic, autotrophic and hetero-
trophic removal of ammonia–nitrogen in aquaculture systems.
hand, Nile tilapia reared in BFT presented a higher mean weight gain
Aquaculture, 257, 346–358.
and biomass (according to Luo et al., 2014) and significantly Ekasari, J., Suprayudi, M. A., Wiyoto, W., Hazanah, R. F., Lenggara, G. S.,
increased fish specific growth rate, net yield, weight gain, individual Sulistiani, R., . . . Zairin, M. (2016). Biofloc technology application in
fish weight and protein efficiency ratio compared to the Control African catfish fingerling production: The effects on the reproductive
performance of broodstock and the quality of eggs and larvae. Aqua-
group (Long, Yang, Li, Guan & Wu, 2015). These differences may be
culture, 464, 349–356.
related to the specificities of the culture phase, system design, stock- Ekasari, J., Zairin, M., Putri, D. U., Sari, N. P., Surawidjaja, E. H., & Bossier,
ing density and management strategies used in these studies. A P. (2015). Biofloc-based reproductive performance of Nile tilapia
meta-analysis study may be necessary to determine which factors Oreochromis niloticus L. broodstock. Aquaculture Research, 46, 509–
512.
contribute to the success or failure of BFT.
El-Sayed, A.-F. M. (2006). Tilapia Culture. Oxfordshire: CABI Publishing.
In conclusion, we found no negative effects of BFT on reproduc- El-Sayed, A.-F. M., & Kawanna, M. (2008). Effects of dietary protein and
tive or growth parameters. Due to the positive effects of this system energy levels on spawning performance of Nile tilapia (Oreochromis
in terms of controlling water quality, this system may be recom- niloticus) broodstock in a recycling system. Aquaculture, 280, 179–
184.
mended for female breeder stocking of Nile tilapia.
Emerenciano, M., Cuzon, G., Are
valo, M., & Gaxiola, G. (2014). Biofloc
technology in intensive broodstock farming of the pink shrimp Far-
ACKNOWLEDGMENTS fantepenaeus duorarum: Spawning performance, biochemical compo-
sition and fatty acid profile of eggs. Aquaculture Research, 45, 1713–
We thank Conselho Nacional de Desenvolvimento Cient
ıfico e Tec- 1726.

gico (CNPq), Fundacß~ao de Amparo a Pesquisa do Estado de
nolo Fernandes, A., Alvarenga, E.,
Oliveira, D., Aleixo, C., Prado, S., Luz, R., . . .

-Reitoria de Pesquisa da UFMG

Minas Gerais (FAPEMIG) and Pro Turra, E. M. (2013). Production of oocytes of Nile tilapia (Ore-
ochromis niloticus) for in vitro fertilization via hormonal treatments.
(PRPQ-UFMG) who provided financial support.
Reproduction in Domestic Animals, 48, 1049–1055.
Grier, J. H., Uribe, M. C., & Patin ~o, R. (2009). The ovary, folliculogenesis
REFERENCES and oogenesis in teleosts. In B. J. M. Jamieson (Ed.), Reproductive biol-
ogy and phylogeny of fishes (agnathans and bony fishes): Phylogeny,
APHA (1989). Standard Methods for the Examination of Water and reproductive system, viviparity, spermatozoa (pp. 25–84). Enfield, USA:
Wastewater. 17th ed. Washington: American Public Health Associa- CRC Press, Science Publishers.
tion, American Water Works Association, Water Pollution and Con- Little, D. C., Coward, K., Bhujel, R. C., Pham, T. A., & Bromage, N. R.
trol Federation. (2000). Effect of broodfish exchange strategy on the spawning per-
Avnimelech, Y. (2009). Biofloc Technology - A Practical Guide Book. Baton formance and sex steroid hormone levels of Oreochromis niloticus
Rouge: The World Aquaculture Society. broodfish in hapas. Aquaculture, 186, 77–88.
Azim, M. E., & Little, D. C. (2008). The biofloc technology (BFT) in indoor Long, L., Yang, J., Li, Y., Guan, C., & Wu, F. (2015). Effect of biofloc tech-
tanks: Water quality, biofloc composition, and growth and welfare of nology on growth, digestive enzyme activity, hematology, and
Nile tilapia (Oreochromis niloticus). Aquaculture, 283, 29–35. immune response of genetically improved farmed tilapia (Oreochromis
Bendschneider, K., & Robinson, R. J. (1952). A new spectrophotometric niloticus). Aquaculture, 448, 135–141.
method for the determination of nitrite in sea water. Journal of Mar- Luo, G., Gao, Q., Wang, C., Liu, W., Sun, D., Li, L., & Tan, H. (2014).
ine Research, 8, 1–18. Growth, digestive activity, welfare, and partial cost-effectiveness of
Bhujel, R. C. (2000). A review of strategies for the management of Nile genetically improved farmed tilapia (Oreochromis niloticus) cultured in
tilapia (Oreochromis niloticus) broodfish in seed production systems, a recirculating aquaculture system and an indoor biofloc system.
especially hapa-based systems. Aquaculture, 181, 37–59. Aquaculture, 42, 2–423.
Boyd, C. E., & Tucker, C. S. (1998). Pond Aquaculture: Water Quality Melo, R. M. C., Martins, Y. S., Teixeira, E. A., Luz, R. K., Rizzo, E., & Baz-
Management. New York, USA: Springer Science Business Media. zoli, N. (2014). Morphological and quantitative evaluation of the
700 pp. ovarian recrudescence in Nile Tilapia (Oreochromis niloticus) after
Campos-Mendoza, A., McAndrew, B. J., Coward, K., & Bromage, N. spawning in captivity. Journal of Morphology, 275, 348–356.
(2004). Reproductive response of Nile tilapia (Oreochromis niloticus)
rez-Fuentes, J. A., Hern
andez-Vergara, M. P., Pe
Pe
rez-Rostro, C. I., &
to photoperiodic manipulation: Effects on spawning periodicity, Fogel, I. (2016). C:N ratios affect nitrogen removal and production of
fecundity and egg size. Aquaculture, 231, 299–314. Nile tilapia Oreochromis niloticus raised in a biofloc system under high
Carrillo, M., Zanuy, S., & Bayarri, M. J. (2009). El control ambiental de la density cultivation. Aquaculture, 452, 247–251.
reproduccio
n de los peces con especial referencia al control del ciclo Pinheiro, J. C., & Bates, D. M. (2000). Mixed-Effects Models in S and S-
sexual, de la pubertad y de la precocidad. In M. A. C. Este
vez (Ed.), PLUS. New York: Springer.
La Reproduccio n de los Peces: Aspectos Basicos y Sus Aplicaciones En Ridha, M. T., & Cruz, E. M. (1999). Effect of different broodstock densi-
Acuicultura (pp. 173–246). Madrid: Publicaciones Cient
ıficas y Tec- ties on the reproductive performance of Nile tilapia, Oreochromis
nolo
gicas de la Fundacio
n Observatorio Espan ~ol de Acuicultura. niloticus (L.), in a recycling system. Aquaculture Research, 30, 203–
Di Rienzo, J. A., Casanoves, F., Balzarini, M. G., Gonzalez, L., Tablada, M., 210.
& Robledo, C. W. (2015). InfoStat versio n 2015. Argentina: Grupo Schreck, C. B., Contreras-Sanchez, W., & Fitzpatrick, M. S. (2001). Effects
InfoStat, FCA, Universidad Nacional de Co
rdoba, http://www.infosta of stress on fish reproduction, gamete quality, and progeny Aquaculture,
t.com.ar 197, 3–24.
Dietrich, D. R., & Krieger, H. O. (2009). Histological Analysis of Endocrine Strickland, J. D. H., & Parsons, T. R. (1972). A Practical Handbook of
Disruptive Effects in Small Laboratory Fish. Hoboken, New Jersey: John Seawater Analysis. 2nd edition. Ottawa: Fisheries Research Board of
Wiley & Sons, Inc. Canada, Bulletin 167.
8 | ALVARENGA ET AL.

Tsadik, G. G., & Bart, A. N. (2007). Effects of feeding, stocking density under different environments and age classes. Acta Scientiarum. Ani-
and water-flow rate on fecundity, spawning frequency and egg mal Sciences, 37, 221–226.
quality of Nile tilapia, Oreochromis niloticus (L.). Aquaculture, 272,
380–388.
UNESCO. (1983). Chemical Methods for use in Marine Environmental Moni-
toring. Manual and Guides(pp. 29–36), Paris: Intergovernmental
de Sales SCM, de
How to cite this article: de Alvarenga ER,
Oceanographic Commission. Brito TS, et al. Effects of biofloc technology on reproduction
Wedemeyer, G. (1996). Physiology of Fish in Intensive Culture Systems,
and ovarian recrudescence in Nile tilapia. Aquac Res.
Springer. New York: Chapman & Hall.
Yoshida, G. M., Oliveira, C. A. L. De, Kunita, N. M., Rizzato, G. S., & 2017;00:1–8. https://doi.org/10.1111/are.13420
Ribeiro, R. P. (2015). Reproduction performance of female Nile tilapia

View publication stats