DNA barcoding approaches for fishing authentication of exploited grouper species including the endangered and legally protected goliath grouper Epinephelus itajara

1 Laboratório de Genômica Evolutiva e Ambiental, Departamento de Zoologia, Universidade Federal de Pernambuco, Av. Prof. Nelson Chaves, s/n, Cidade Universitária, Recife, PE, Brazil, CEP 50760-420. E-mail: rodrigotorres@ufpe.br 2 Pontifícia Universidade Católica de Minas Gerais, Belo Horizonte, MG, Brazil. 3 Programa Pós Graduação em Ecologia e Conservação – Universidade Federal do Paraná, Setor de Ciências Biológicas, Caixa Postal 19031, CEP 81531-980, Curitiba, PR, Brazil. 4 Universidade Federal do Espírito Santo, CEUNES, DCAB, Campus de São Mateus, São Mateus, ES, Brazil. 5 Departamento de Oceanografia, Universidade Federal de Pernambuco, Recife, PE, Brazil.


INTRODUCTION
Molecular DNA markers are now sufficiently sophisticated to recognize genetic variants of a species.The low cost of the techniques has resulted in a recent increase in the number of studies using this approach.For example, one of the simplest, lowest-cost methods uses polymerase chain reaction-restriction fragment length polymorphisms (PCR-RFLPs) (Schlötterer 2004).
PCR-RFLP markers are commonly used and highly repeatable (Meyer et al. 1995) for the identification of a variety of species in several phyla (Wolf et al. 1999, McDonald et al. 2005, Torres 2006, Santaclara et al. 2007, Spergser and Rosengarten 2007, Fernandez-Tajes and Méndez 2007, Sowmya et al. 2007, Kang et al. 2008).Although DNA sequencing and analysis is accurate and authentic, it is costly, time-consuming and not suitable for routine species identification studies.PCR-RFLP has been proven to be a practical, simple and rapid technique (Meyer et al. 1995, Partis et al. 2000) and a high level of expertise in molecular genetics is not necessary for interpreting results obtained in agarose gels.The technique allows mislabeled or fraudulent fish products to be detected easily (Hsieh et al. 2010).Additionally, this technique has recently been used as a forensic tool for solving crimes ("forensic genetics", Butler 2005, Jobling andGill 2004).For example, it has been used to determine the origin of cattle meat (Verkaar et al. 2002) and illegal wildlife trading (Dubey et al. 2010).
Because of the abundance of mutations in the genomes of many species (Morin et al. 2004), single nucleotide polymorphisms (SNPs) are very effective markers for resolving similar issues.Using SNPs has become one of the most recent innovations for ecological and conservation management, especially that of the fishery industry (Morin et al. 2004, Hauser andSeeb 2008).Technological advances in fisheries have resulted in an industrial level of fishing with a much greater and widespread impact.As a consequence, regulation is required to control the exploitation of fish and prevent depletion of stocks (Pauly et al. 2002).Following the worldwide trend for marine fisheries (Pauly et al. 2005, Pauly 2009), most grouper fisheries are sharply declining, collapsing or already depleted (Morris et al. 2000, Frédou and Ferreira 2005, Mitcheson et al. 2012).
As a management tool, accurate identification of these species based on molecular authentication (e.g.DNA barcoding) is very important because species may be identified from tissue samples in the absence of morphological characters (Frézal and Leblois 2008).In addition, several molecular techniques have been used to distinguish between legal and illegal products, to relocate animals for their natural populations, and to mark and track DNA profiles (Alacs et al. 2012).Often, fish fillets are mislabeled as different species for marketing purposes or to disguise illegal capture and retail (Jacquet and Pauly 2008, Miller and Mariani 2010, Carvalho et al. 2011a, b).DNA authentication through DNA barcoding has applications in ecology, medicine, epidemiology, evolutionary biology, biogeography and conservation biology (Frézal andLeblois 2008, Alacs et al. 2012), such as the use of cytochrome C oxidase for species delimitation and fishing authentication (Hebert et al. 2003, 2004a, Sass et al. 2007, Hajibabaei et al. 2007, Linacre and Tobe 2011, Carvalho et al. 2011b).
The grouper family (Epinephelidae) comprises 62 genera and 449 species (Heemstra andRandall 1993, Nelson 2006), many of which are exploited throughout the world (Morris et al. 2000, Mitcheson, et al. 2008, Mitcheson et al. 2012) and comprise some of the most important fishery resources in the tropical west Atlantic (Heemstra andRandall 1993, Claro et al. 2001).Due to over-fishing, the goliath grouper (Epinephelus itajara) is critically endangered and consequently its capture is prohibited in several countries, including Brazil (IBAMA 2007, IUCN 2008).Also, because of its slow development in brackish waters, it requires many years to recover (Sadovy andEklund 1999, Frias-Torres 2007).Traditional knowledge about the ecology and habits of the grouper among local, small-scale fishermen, has been very important for the conservation of the goliath grouper in Brazil (Hostim-Silva et al. 2005, Gerhardinger et al. 2009).Other species of grouper, including the dusky (formerly Mycteroperca marginata sensu Craig and Hastings 2007), red (Epinephelus morio) and black (M.bonaci), are also exploited, yet their conservation receives less attention worldwide (Gimenez-Hurtado et al. 2005, Machado et al. 2003, Teixeira et al. 2004, Gerhardinger et al. 2006, Freitas et al. 2011).Therefore, they may suffer even greater extinction risks.
Taxonomic identification of fish is a difficult task because the way in which the fish are processed leads to a lack of diagnostic morphological characters.This favours the illegal trade of prohibited species because the fish can be sold under the name of similar, legally caught fish (Wong and Hanner 2008).In the case of E. itajara in Brazil, knowledge about the fish ban is widespread, so illegal traders also use filleting to trade it mislabeled as other species of grouper.In these cases, environmental law enforcement is often hampered by absence of proof.Thus, DNA-based procedures are valuable tools for law enforcement since they can attest the species in which is marketed.We adopted PCR-RFLP because it is less time-consuming and more cost-effective than DNA sequencing.Such methodology also requires equipment readily available in most molecular laboratories, and it has proven its utility in species identification (Palo and Merila 2003, Rohilla and Tiwari 2008, Dubey et al. 2010).Here, we examined the variability of the barcode region (COI mtDNA gene) in groupers to develop DNA authentication tools (PCR-RFLPs and SNPs), specifically the genetic signatures of the 13 commercially important species, especially the goliath (Epinephelus itajara), red (E.morio), dusky (Mycteroperca marginata) and black (M.bonaci) groupers.
accurate identification among species levels (Ward et al. 2008), COI variations were not considered.In addition, negligible nucleotide variation is expected in the barcode region between several geographically different marine fish populations (F ST <0.01/p>0.24;Ward et al. 2008).
We collected fin, liver and muscle samples that were stored in 96% ethanol (Merck) at -20°C.DNA extraction through the DNEasy Blood and Tissue kit (QIAGEN) followed the instructions of the manufacturer.Templates were dissolved in Tris+EDTA (pH 8.0) and diluted at 1:20 for PCR.
RFLP reactions occurred in 90 min at the temperatures recommended by the manufacturer of each enzyme.Resultant products were resolved by electrophoresis in 1.5% agarose gel immersed in TBE buffer, stained with GelGreen™ (Biotium), and then photographed.The molecular weight of the fragments was estimated using a 100-bp molecular marker ladder (New England Biolabs).Few different reagent brands were tested and no different RFLP profiles were detected.
Apart from BamHI and EcoRI (data not shown), other enzymes yielded diagnosable RFLP profiles (Figs 3-5).A single undigested PCR fragment (undigested control lane) was observed with BamHI.Experiments with EcoRI also revealed a single band in all species that was slightly smaller than the undigested control lane.The remaining enzymes allowed different unique or shared PCR-RFLP profiles to be identified.For example, AluI resulted in ten species-linked profiles (Fig. 3, lanes 1-5, 9, 11 and 12) and two shared profiles (Fig. 3, lanes 6-7 and 8-10).Despite lower resolution, enzymes MboI and BtsCI allowed eight PCR-RFLP profiles to be identified (Fig. 4a, b).HaeIII, HhaI and Taq α I gave seven, six and five profiles, respectively (Fig. 5a, b, c).

PCR-RFLPs
Our PCR-RFLP analyses allowed us to correctly identify 12 species of grouper, including the endangered goliath grouper, using individual or combined genetic profiles obtained with a few restriction enzymes (Table 2,Figs 3,4a,b,.Thus, our results provide useful DNA authentication tools for identifying fishing of goliath (Epinephelus itajara), dusky (Mycteroperca marginata), and black (Mycteroperca bonaci) groupers without morphological characters in hand, as in fish markets.Considering the extreme risk of extinction of the analysed species (Mitcheson et al. 2008, Mitcheson et al. 2012), our finding are very welcome for grouper fishery management.While laws protect endangered fish such as goliath grouper, these laws have been futile due to the lack of an effective means of identifying illegal catches.Thus, our results provide the resolution of this important problem, especially for the goliath, dusky (M.marginata) and black (M.bonaci) groupers.
With a properly amplified barcode segment for all species (Fig. 2), a small adjustment improved resolution for those three species.That is, the replacement of the reverse primer (Fig. 2) improved resolution by avoiding the formation of nonspecific PCR products (possibly from COI).Thus, F1 and R1/R2 primers are specific for COI amplifications, and homology was shown by identical molecular weight for these three species with F1 and R2 primers (Ward et al. 2005).
In considering the use of the eight enzymes in procedures for DNA authentication of fishing (PCR-RFLPs), the enzymes BamHI and EcoRI yielded very similar RFLP profiles with a single band for each of the 12 species (data not shown).It is possible that a priori these enzymes were not used because the terminal COI region in all 12 species may have been cut into tiny fragments of low molecular weight.Nonetheless, these enzymes may be effectively used to prove that a putative fish sample was not a grouper when more than a single band appears.
Clear authentication of illegal fishing of these fish can be resolved through the use of five enzymes.The goliath grouper had RFLP profiles shared by two of the six enzymes (MboI and Taq α I; Figs 3 and 5, respectively), yet had unique RFLP profiles for AluI, BtsCI and HhaI (Figs 3,4b,5b).Thus, these data show that a combination of these profiles will provide clear forensic evidence for the goliath grouper.
Similarly, the black grouper (Mycteroperca bonaci) had a RFLP profile shared with the goliath grouper in the enzyme Taq α I (Fig. 5c).With enzymes AluI and MboI, identification is much better, with a single band of ~180 bp (AluI) and two bands of ~250 and 350 pb (MboI, Figs 3 and 4a).Also, the dusky grouper (M.marginata) is identified by the combined RFLP profiles of AluI (Fig. 3), HhaI (Fig. 5b), and Taq α I (Fig. 5).By using the genetic signatures of each species (Table 2), the necessary steps of using a single or combined genetic profile are shown for DNA authentication of the black (Mycteroperca bonaci), dusky (M.marginata) and goliath (E.itajara), and other groupers.

SNPs
A total of 94 SNPs were observed for correct identification both among and within genera, and among species (Fig. 6).Previously, a minimum of 60 SNPs was suggested as necessary for good forensic/ authentication diagnostics (Sobrino et al. 2005).The evidence provided herein (94 SNPs) indicate the robustness of the genetic features observed regarding a DNA authentication method.Traditional (as shown herein) and alternative barcoding regions have been used to examine the authentication and traceability of cattle meat (Fontanesi et al. 2010), wildlife (Sato et al. 2010, Ferreira et al. 2011) and fish (Baker and Palumbi 1994, Comi et al. 2005, Ogden 2008, Rasmussen and Morrissey 2008, Holmes et al. 2009, Ardura et al. 2010, Supernault et al. 2010, Carvalho et al. 2011a).Thus, correct species identification by SNP positions (Table 4) illustrates that the current barcode regions are very useful for fishery management.A clear example is the red grouper (Epinephelus morio), which has 29 autapomorphic SNPs (Table 4, Fig. 6) and which now has an important tool to prevent its exploitation (Morris et al. 2000, Sadovy 2001, Olavo et al. 2005, Freitas et al. 2011).
In addition, the barcode distinctiveness observed between the two Mycteroperca and Epinephelus spp.(Table 3) indicates the efficiency of the presented protocol.According to Ward et al. (2009) andCarvalho et al. (2011b), congeneric fish species are genetically (COI) different at a minimum rate of 8.4%.Thus, the minimum COI difference of 10.3% (Mycteroperca marginata X M. bonaci, Table 3) seems to be high enough to state that the presented SNP-based protocol is an accurate DNA authentication procedure.In practi- cal terms, sites 93 and 258 (Fig. 6, arrows) support this statement because they allow the simultaneous identification of those four overexploited and legally protected species as E. itajara by presenting a different nucleotide in each of the species.
In summary, we developed and tested the DNA barcoding as an effective tool for monitoring fisheries and illegal trade in groupers, including the endangered goliath grouper.

Fig. 6 .
Fig. 6. -Aligned sequences from Mycteroperca bonaci, M. marginata, Epinephelus morio and E. itajara.Boxes indicate the SNP positions among the analyzed species; arrows indicate two SNP positions that uniquely identify all species.

Table 1 .
-The list of the enzymes used in this study, their restriction sites (*), and their temperatures of use.

Table 2 .
-PCR-RFLP profiles (in base pairs of DNA fragments) of the barcode region of the groupers in this study (species numbers as in Fig.1).