Genetic variability of broodstocks of restocking programs in Brazil Variabilidad genética de lotes de reproductores de programas de repoblación en Brasil

Objective. The aim of this study was evaluate the genetic diversity of the following broodstocks: piapara (Leporinus elongatus), dourado (Salminus brasiliensis), jundiá (Rhamdia quelen) and cachara (Pseudoplatystoma fasciatum) already useful for restocking programs in the Paranapanema, Iguaçu and Paraná Brazilian Rivers. Materials and methods. Samples from the caudal fin of 122 fish were analyzed. DNA was extracted by NaCl protocol. PCR products were separated by a horizontal agarose gel electrophoresis. The fragments were visualized by staining with ethidium bromide. Results. The amplification of 25 primers generated different fragments in studied species that allowed characterizing 440 fragments of 100-2900 bp. High percentage of polymorphic fragments (66.67 to 86.29), Shannon index (0.365 to 0.486) and genetic diversity of Nei (0.248 to 0.331) were detected. Conclusions. The level of genetic variability in the broodstocks was adequate for allowing their use in restocking programs in the studied Rivers. However, periodical monitoring studies of genetic variability in these stocks, the mating system, reproductive system and general management must be made to guarantee the preservation of wild populations.


INTRODUCTION
In the last seven years, fishing of the world aquatic organisms has pointed out toward decreases and stabilization.In 2005 and 2010, only 92.478.416 and 88.970.124tons were, respectively, captured, which means a reduction of 3.508.292tons.In 2011, in contrast, 93.494.340tons had 4.524.216tons higher than in 2010 (1).The fish farming, otherwise, has been more significant.For example, in 2005 and 2010 the growth was higher than the catching, when there was an increase from 57.815.836 to 78.028.539tons, respectively, representing a net output of 20.212.703tons.In 2011, furthermore, 83.675.661tons of fish were raised; an output of 5.647.122tons higher than in 2010 (1).
In Brazil, based on data from the Aquaculture and Fisheries Ministry (2), the fishing exploitation had an increase from 785.366 in 2010 to 803.270 tons in 2011, or 2.3%.Otherwise, the fish farming increased from 479.398 in 2010 to 628.704 tons in 2011, or 31.1%.These data suggest an overlap of fish farming on fishing exploitation in the next years as well as detected in other segments of the animal production (3).The piapara (Leporinus elongatus), dourado (Salminus brasiliensis), jundiá (Rhamdia quelen) and cachara (Pseudoplatystoma fasciatum) species has been highlighted because they are present in several Brazilian basins (4) with several factors encouraging their farming as the good growth and performance, the easy reproduction and the rusticity (5)(6)(7)(8).These factors have impacted the Brazilian market of fish.For example, in 2011 were produced 4309.3 and 1747.3 tons of Rhamdia and Leporinus (2).
Otros factores, sin embargo, como la polución del agua, deforestación, destrucción del hábitat, sobrepesca, introducción de especies exóticas y represamiento de los ríos (9,10) han reducido las poblaciones naturales de esos peces de agua dulce en el Brasil.Entre las decisiones tomadas para mitigar la reducción de las poblaciones naturales de peces, resaltamos los programas de repoblación (11).Aunque han sido aplicados en Brasil por décadas, las investigaciones sobre su eficiencia genética y ambiental no se han hecho endangered species have been to avoid the complete losses of gene clusters, preserve as much as possible the genetic diversity and reach the first step without compromise the survivorship of the populations in a long term.Many programs, however, were unsuccessful because the mismanagement of parental stocks (11) and the genetic non-monitoring.Thus, the genetic evaluation of the stocks for restocking has to be done.The molecular markers have been one of the most useful methods to approach the genetic diversity and the population structure of the various species and fish lines (14,15).
The aim of the current study was to evaluate the genetic diversity of the piapara (Leporinus elongatus), dourado (Salminus brasiliensis), j u n d i á ( R h a m d i a q u e l e n ) a n d c a c h a ra (Pseudoplatystoma fasciatum) broodstocks in restocking programs applied in the Paranapanema, Iguaçu and Paraná Brazilian Rivers.Experimental procedure.The DNA was extracted using the protocol described by Lopera-Barrero et al (16).The DNA extracted was quantified in the Shimadzu spectrophotometer using the absorbance of 260 nm.The samples were diluted to 10 ng/µL, and the DNA quality was evaluated in 1% agarose gel electrophorese using a buffer TBE 1X (500 mM Tris-HCl, 60 mM boric acid and 83 mM EDTA) at 70 V for one hour.The gel was visualized under UV radiation after treatment with ethidium bromite (0.5 mg/ml) for one hour.Thereafter, the image was photographed using the program Kodak EDAS (Kodak 1D Image Analysis 3.5, EUA).
The reactions were amplified in Eppendorf Mastercycler Gradient (EUA), and the cycles were based on the species.L. elongatus and S. brasiliensis had the DNA denaturized at 96ºC for 5 min and soon after was conducted 40 cycles of 1 min at 94ºC, 1 min and 30 s annealing the primer at 36ºC and extension at 72ºC for 2 min.Thereafter, the final extension was at 72ºC for 7 min.Otherwise, R. quelen had DNA denaturized at 92ºC for 4 min and soon after was conducted 40 cycles of 1 min at a 92ºC, 1 min of annealing at 40ºC and 2 min at 72ºC for extension.Soon after, was did the final extension at 72ºC for 5 min.Finally, P. fasciatum had amplification programmed for 40 cycles, with one initial step of denaturation at 95ºC for 5 min and the final step at 72ºC for 7 min.Every cycle had 1 min at 94ºC, 1 min at 36ºC and two min at 72ºC.Twenty five primers from the Kits OPA, OPW and OPX (Operon Technologies Ltd., Valencia, USA) with 10 bases for every species were tested, when those of best definition and reproducibility were chosen (Table 1).
The amplified products were separated in agarose gel at 1.5% using horizontal electrophoresis.Fifteen microliters of amplified product in conjunction with 2 μL of sample buffer (40% sucrose and 0.25% bromophenol blue) were inserted into the gel.This electrophoresis was carried out in buffer TBE 0.5X (45 mM Tris-Borate and 1 mM EDTA) at 70 V for 4 hours.The quantification and amplification gel were visualized under UV radiation after ethidium bromide dyeing (0.5 μg/mL) for 1 hour.The images were photographed using the program EDAS (Kodak 1D Image Analysis 3.5, New York, USA).

Computer software and statistical analysis.
The size of the fragments was estimated by comparing them to the standard ladder of 100 bp (Invitrogen, Carlsbad, USA).The presence or absence of fragments with similar molecular size was used to build a similarity matrix based on the Jaccard coefficient where 1 was the presence and 0 was the absence.The percentage of polymorphic fragments (95% criterion), the Shannon´s index and the genetic diversity of Nei were calculated by the PopGene 1.31 program (17).

RESULTS
The 25 primers (Table 1) had different amplification standards for every fish species.A total of 440 fragments from 100 bp (OPW14-Pseudoplatystoma fasciatum) to 2900 bp (OPW03 -Leporinus elongatus) were amplified.The total number of fragments and number of polymorphic fragments were different among the species.In L. elongatus, 107 in 124 fragments were polymorphic; S.
El porcentaje de fragmentos polimórficos, índice de Shannon y parámetros de diversidad genética de Nei indicaron una alta variabilidad genética que es esencial para un buen desempeño de programas de repoblación.Alelos perdidos de resistencia a enfermedades y reducciones en la capacidad de adaptación son consecuencias de la reducción de variabilidad genética (14).Los datos publicados de marcadores moleculares con las mismas especies estudiadas se muestran en la tabla 3.No se encontraron referencias a P. Lopera-Barrero -Genetic variability of broodstocks of restocking programs About the percentage of polymorphic fragments high values were found for all the species (66.67% to 86.29%).The Shannon index values were similarly high.The results from the genetic diversity of Nei showed high estimates, ranging from 0.248 (S. brasiliensis) to 0.331 (R. quelen) (Figure 1).

DISCUSSION
The total number of fragments found is consistent to the conclusions of Telles et al. (18) that determined the most important number of fragments obtained (at least 50) than the number of primers used for the calculation of genetic diversity.
The percentage of polymorphic fragments, Shannon index and genetic diversity of Nei parameters indicated high genetic variability that is essential for a good performance of restocking programs.Losses alleles of disease resistance and reduction in the adaptation capacity are consequences of dropping in the genetic variability (14).Literature data from molecular markers with the same studied species are displayed in the table 3. References to P. fasciatum were not found and only one  manuscript to R. quelen was reported.Not all the cited references available had evaluated the same parameters investigated in the current experiment.
About the percentage of polymorphic fragments (PPF) and Shannon index (SI) from other manuscripts (Table 3), just the percentage of polymorphic fragments of R. quelen found by Lidani et al (19) was higher than was found in the current investigation.Both parameters (PPF and SI) were slightly lower in all manuscripts about L. elongatus and much more pronounced for S. brasiliensis.
Most of the samples evaluated by these authors were collected in Paranapanema River.Working with S. brasiliensis, Lopes et al (20) stated that the genetic variability found in their stocks was the consequence of the small population in the sites where these fish were collected (Canoas complex, Paranapanema River).Thus, despite the river from where was collected the specimens of L. elongatus and S. brasiliensis had been the same the founder effect may have had influence on the genetic variability.Based on Kang et al (21), the small effective number of parental fish in the stock formation may eliminate low frequent alleles, and consequently reducing the genetic variability.This was the factor reported by Lopera Barrero et al (22) which induced high variability in Prochilodus lineatus because the stocks investigated were formed by fish populations from different Rivers.Similarly, Gomes et al (23) found low genetic variability in a group of seven S. brasiliensis with 33.33% of polymorphic fragments and Shannon index of 0.174, the lowest values found in these groups.
The genetic diversity of Nei is a parameter which determines the level of population differentiation.
Based on the values obtained in this study, it can be affirmed that there was high heterogeneity between stocks, suggesting that they do not have a common origin.For this parameter, the values expressed in table 3 were classified from low to very high.Based on low values of genetic diversity (0.086), Ramos et al (10) stated that groups collected different periods in the transposition ladder of the UHE in Canoas I and Canoas II were part of only one population with several sub-populations.Similar conclusion was found by Lopez et al (20) to observe a value of 0.018.They also stated that this and other groups were structured in just one population from Capivara reservoir The efficiency of restocking programs has also to be highlighted beyond proper genetic variability, i.e. the program intensity, number of parents using in the mating, mating system, general management, and the genetic differentiation between wild and broodstock populations.
Evaluando la intensidad de repoblación de Salvelinus fontinalis en la diversidad genética de poblaciones naturales, Marie et al (24) estableció que en lagos con mayor intensidad había mayores niveles de diversidad, probablemente debido a la introducción de alelos domésticos.No obstante, Lopera Barrero Lopera-Barrero -Genetic variability of broodstocks of restocking programs Evaluating the intensity of restocking of Salvelinus fontinalis on the genetic diversity of wild populations, Marie et al (24) stated that in the lakes with higher intensity had higher levels of diversity, likely because of the introduction of domestic alleles.However, Lopera Barrero et al (22) reported the importance of genetic monitoring of populations to avoid losses in genetic variability (outbreeding depression), while preventing the introduction of genetic material com high genetic differentiation in relation to the wild population.
On the other hand, showing the negative effect of impropriated number of parents, Machado-Schiaffino et al (25) reported the variability losses in juveniles of Atlantic salmon (Salmo salar L.) from restocking programs in comparison to wild populations likely because the bottleneck effect caused by the reduced number used in breeding matings.In contrast, Lopera-Barrero et al (26) stabilized the variability using 24 parental fish (12 males and 12 females) of Brycon orbignyanus showing the positive effect of adequate number of parents, despite the semi-natural reproductive system that also had influence in maintain the genetic variability.
Another factor with influence in the genetic variability of fish stocks has been the mating system.Gomes et al (14) found higher estimates of genetic variability in the progeny of S. brasiliensis than in the parental fish using the semi-natural reproductive system.When evaluating the paternity of progeny Brycon orbignyanus in semi-natural and extrusion reproductive systems, Lopera Barrero et al (27) found multiple paternities and differentiated reproductive contribution in both systems.The authors, however, suggest more experiments to confirm the reproductive efficiency of semi-natural system in maintaining the genetic variability in other species.
The management of the offspring can also influence the genetic variability of restocking programs.Almeida et al (28) found that instead of releasing each offspring lot separately into the water, mixing specimens produced in the various fry stocks before releasing them in the river would be more feasible.
The restocked population will have a genetic structure closer to wild populations.On the other hand, Rodriguez-Rodriguez et al (29) suggested that in the progeny aimed at restocking programs a first genetic analysis must be carried in the larval stage (3 days), which will provide a general overview of the new genetics generation.Then, at 60 or 90 days, depending on environmental conditions of the ecosystem should be performed a second analysis to objectively determine the final genetic variability with which individuals will be released into the river.This analysis should et al (22) reportó la importancia del monitoreo genético de las poblaciones para evitar pérdidas en la variabilidad genética (depresión exogámica), además de impedir la introducción de material genético con alta diferenciación genética en relación a la población natural.

Agradecimiento
Los autores agradecen a Duke Energy International por proporcionar parte del material biológico y a los investigadores, técnicos y estudiantes que apoyaron el presente experimento.
be performed to determine which individuals will be released and determine the true feasibility of restocking.
Finally, high genetic differentiation between wild and broodstocks can induce important gene losses associated with local adaptation.Similarly, González-Wangüemert et al (30) found high genetic differentiation between stocks of Diplodus sargus (for restocking the Castellamare gulf, Italy) and samples from the wild population.Based on these authors, this differentiation was likely verified by periodic inclusion of wild individuals from other places and the management adopted in captivity.
Based on the current results, the level of genetic variability in the L. elongatus, S. brasiliensis, R. quelen and P. fasciatum broodstocks are therefore adequate for allowing their use in restocking programs in the Paranapanema, Iguaçu and Paraná Rivers.However, periodical monitoring of the genetic variability of these stocks, the mating system, proper reproductive system and general management should be studied to guarantee the preservation of these wild populations.

Table 1 .
Nucleotide sequences and percentage of G+C bases of the primers amplified for the Leporinus elongatus, Salminus brasiliensis, Rhamdia quelen and Pseudoplatystoma fasciatum broodstocks.

Table 3 .
Variability and genetic diversity using molecular markers in L. elongatus, S. brasiliensis and R. quelen.