322

Abstract—Twenty homologous mi- Development and characterization of crosatellite markers, or simple se- quence repeats (SSRs), were devel- markers for blackfin tuna oped for blackfin tuna (Thunnus at- lanticus) through the use of a direct (Thunnus atlanticus) with the use of Seq-to-SSR approach. The number Illumina paired-end sequencing of alleles per locus ranged between 5 and 26, and the expected hetero- zygosity ranged between 0.640 and Luca Antoni1 0.969. Three loci displayed signifi- Patricia L. Luque1 cant departure from Hardy-Wein- Kaylie Naghshpour1 berg equilibrium expectations, likely reflecting occurrence of null alleles. Lionel Reynal2 Another locus showed consecutive Eric A. Saillant (contact author)1 alleles that differed by one base pair only. Consequently, this locus may be Email address for contact author: [email protected] prone to elevated rates of scoring er- rors. The remaining 16 loci will be 1 Gulf Coast Research Laboratory valuable for studies in conservation The University of Southern Mississippi genetics of blackfin tuna. 703 East Beach Drive Ocean Springs, Mississippi 39564 2 Institut Français de Recherche pour l’Exploitation de la Mer (IFREMER) Unité Biodiversité et Environnement de la Martinique 79 Route de Pointe-Fort 97231 Le Robert Martinique, France

The blackfin tuna (Thunnus atlanti- such a situation, the analysis of ge- cus) is a small tuna widely distribut- netic variation at molecular markers ed in tropical and subtropical waters is expected to provide valuable in- of the western Atlantic Ocean from formation on the structure of stocks the mid-Atlantic region of the East to develop sustainable management Coast of the United States south to of fisheries (Carvalho and Hauser, northern Brazil (Collette and Nauen, 1994). are the most 1983). Although this species is ex- widely used markers for genetic ploited by fisheries in several coun- studies of exploited fishes; however tries, management of this resource to date homologous markers for the in the United States and abroad is blackfin tuna are not available. almost inexistent and, in particu- The objective of this work was to lar, it is not known to date whether develop homologous microsatellite multiple stocks of this species oc- markers, or simple sequence repeats cur within its recorded distribution (SSRs), for studies in conservation range (Mathieu et al., 2013). genetics of this species. The appli- Tagging studies conducted in the cation of next-generation sequenc- island nations of St. Vincent and the ing technologies greatly enhances Grenadines and Bermuda have indi- microsatellite discovery through di- cated that blackfin tuna exhibit some rect screening of short sequences of degree of site fidelity for a significant genomic DNA, hereafter reads, for Manuscript submitted 7 January 2014. proportion of tagged fish that have repeat arrays without the need for Manuscript accepted 25 August 2014. been recaptured in close proximity cost-prohibitive steps, such as cloning Fish. Bull. 112:322–325 (2014). of their tagging location, sometimes of genomic libraries and screening of doi:10.7755/FB.112.4.8 over multiple years (Luckhurst et individual clones through Sanger se- al., 2001; Singh-Renton and Renton, quencing. The Seq-to-SSR method of The views and opinions expressed or implied in this article are those of the 2007). However, long distance move- Castoe et al. (2012) is based on direct author (or authors) and do not necessarily ments also were reported for some of screening of unassembled sequencing reflect the position of the National the individuals tagged in the study of reads, an approach that further in- Marine Fisheries Service, NOAA. Singh-Renton and Renton (2007). In creases the cost effectiveness of the Antoni et al: Microsatellite markers for Thunnus atlanticus 323

procedure. In our study, this method was applied to ther testing, and the 5′ end of 1 of the 2 primers was rapidly identify microsatellite loci in the blackfin tuna. labeled by using one of the fluorescent dyes 6-Fam, Hex, or Ned to allow detection and scoring on an auto- mated DNA sequencer. Evaluation of scoring reliability Materials and methods of the tested loci was based on DNA samples from 8 blackfin tuna specimens. The optimal aT for amplifica- Genomic DNA from fin tissue of one blackfin tuna spec- tion of each locus was then determined during PCR re- imen collected offshore of the Louisiana coastline was actions by using DNA from 4 blackfin tunas, and was extracted by using a standard phenol-chloroform pro- carried out at 6 annealing temperatures held constant tocol (Sambrook et al., 1989). An Illumina1 paired-end through the 35 amplification cycles and ranging from library, which allows sequencing of both ends of DNA 52°C to 62°C. PCR products were run on an ABI Prism fragments, was prepared and sequenced on an Illumina 377 96-lane DNA Sequencer (, Life HiSeq 2000 platform (Illumina, San Diego) according Technologies, Carlsbad, CA). The obtained electrophe- to methods described by Castoe et al. (2012). Reads rograms were analyzed with Applied Biosystems Ge- were quality controlled and trimmed for low-quality neScan software, vers. 3.1.2 (Life Technologies), and data (phred scores <30). In the program PAL_FINDER, alleles were scored in Applied Biosystems Genotyper vers. 0.02.03, a Perl script developed by Castoe et al. software, vers. 2.5 (Life Technologies). The polymorphic [2012] and available at http://sourceforge.net/projects/­ loci that could be scored reliably were characterized palfinder/), 5,874,294 reads were screened for microsat- statistically on the basis of the genotypes of 35 black- ellite arrays that contained a minimum of 12 repeats. fin tuna specimens provided by C. Pau and L. Reynal, Use of a minimum of 12 repeats, in our experience, en- both of IFREMER, La Martinique. Specimens had been sures that the selected microsatellite loci are likely to caught offshore of the island of Martinique in the sum- be polymorphic. The search in PAL_FINDER was con- mer of 2013. tinued until 45 dinucleotide loci and 5 tetranucleotide loci were discovered, requiring screening of 286,240 reads and the whole database (5,874,294 reads) for di- Results and tetranucleotides, respectively. Specific polymerase chain reaction (PCR) primers Amplification tests were conducted on 50 primer pairs, were designed with the open-source software Primer3 45 potentially amplifying dinucleotide microsatel- (Untergasser et al., 2012; Koressaar and Remm, 2007; lites, and 5 amplifying tetranucleotide microsatellites. source code available at http://sourceforge.net/projects/ Twenty-four loci were amplified consistently across the primer3/). The designed primers were tested for ampli- tested specimens and were all labeled with fluorescent fication success and scoring reliability and then evalu- dyes for further evaluation and optimization of anneal- ated at different annealing temperatures using blackfin ing temperature. Twenty loci (19 dinucleotides and 1 tuna samples provided by the Louisiana Department of tetranucleotide) gave scorable PCR products and are Wildlife and Fisheries. Samples had been collected off- described in Table 1. The number of alleles (A), expect- shore of the Louisiana coastline during the spring of ed heterozygosity (He), and inbreeding coefficient (FIS) 2013. Amplification success of the candidate loci during were calculated with the software FSTAT, vers. 2.9.3.2 PCR was tested by assaying 6 specimens. PCR reac- (Goudet, 1995). Per locus estimates ranged from 5 to tions were performed in a total volume of 5.6 µL that 26 for A, from 0.640 to 0.969 for He, and from −0.003 contained 7–13 ng of genomic DNA, 2.2 pmol of each to 0.268 for FIS. Conformance of genotypic proportions primer, 1.1 nmol of dNTPs (Promega Corp., Madison, to Hardy-Weinberg (H-W) equilibrium expectations was WI), 8.4 nmol of MgCl2 (Promega), 0.28 U of Taq poly- tested with exact tests in the software GENEPOP, vers. merase (Promega), and 1X of buffer (Promega). Ampli- 4.1 (Raymond and Rousset, 1995; Rousset, 2008). Geno- fication by PCR consisted of an initial denaturation at typic proportions did not depart significantly from H-W 95°C for 5 min, 35 cycles of 95°C for 30 s, annealing equilibrium expectations, except for the loci BT6, BT27, temperature (Ta) for 30 s, 72°C for 45 s, and a final BT47, and BT91. The departure at locus BT27 was not extension of 15 min at 72°C. Ta was 62°C for the first significant after Bonferroni correction (Rice, 1989). 7 cycles, 60°C for the following 7 cycles, and 56°C for Analyses in Micro-Checker, vers. 2.2.3 (Van Oosterhout the remaining 21 cycles. The obtained PCR products et al., 2004) revealed possible occurrence of null alleles were evaluated through electrophoresis on high-reso- at loci BT6, BT47, and BT91. There was no evidence of lution NuSieve GTG Agarose gels (Lonza Group, Basel, scoring artifacts at locus BC27. Switzerland). Loci that showed consistent amplification success and polymorphic PCR products were selected for fur- Discussion

1 Mention of trade names or commercial companies is for iden- On the basis of the significant departure of genotypic tification purposes only and does not imply endorsement by proportions from Hardy-Weinberg equilibrium expec- the National Marine Fisheries Service, NOAA. tations and the inference of null alleles at loci BT6, 324 Fishery Bulletin 112(4)

H-W 0.75 0.108 0.898 0.0008 0.0005 0.0006 P 0.0972 0.0197 0.0503 0.0698 0.5812 0.2354 0.6784 0.8843 0.1326 0.5591 0.2107 0.3411 0.7854 0.2059 : expected het - expected : e H

IS 0.16 0.05 F -0.16 0.038 0.052 0.263 0.061 0.011 0.268 0.145 0.039 0.055 0.001 0.042 0.137 0.013 -0.086 -0.017 -0.003 -0.051 e

H 0.89 0.67 0.64 0.89 0.905 0.923 0.646 0.776 0.815 0.943 0.952 0.937 0.929 0.969 0.949 0.903 0.915 0.852 0.924 0.897 : number of alleles; alleles; of number : A

96–136 95–113 134–176 233–273 304–316 126–156 137–163 225–305 229–387 144–158 155–211 172–224 142–242 198–258 163–243 162–208 365–403 210–232 151–197 190–234 (base pairs) Allele size range

5 8 7 A 17 15 14 13 10 26 25 20 20 30 22 18 19 12 12 19 14 : annealing temperature; temperature; annealing : a T

(°C)

58 54 62 62 54 62 62 62 62 62 62 62 62 62 56 62 58 62 58 58 a T Table 1 Table Thunnus atlanticus ). Thunnus

) label –3 ′ : probability of conformance to Hardy-Weinberg (H-W) equilibrium expectations. Values in bold represent significant in bold represent significant Values (H-W) equilibrium expectations. probability of conformance to Hardy-Weinberg : H-W Primer sequence (5 ′ CAGGAACATCAACAGGAATGAA R: (FAM) TGTATCACACCGCCAGGTAA R: CAGTGGTGAGTAACAGCATGTC R: (FAM) GGTTGAAGTGCTTCAGGTGTAA F: (HEX) CCATAAAACACTGCCTGTCAGA F: CCACATTACAATCCGTGATCC F: (HEX) TGATCAGCTATGAACGGTCCTT F: (FAM) GGATGGTGCACAGGTTCTC F: CAGGCTGAGGATGGTGCTAC F: (HEX) GTAAGCACCAAGCGTGTGAA F: TTAACGAGGCTGTACTTGCTCA F: (FAM) GCCTGCTACTCCGAGGTAA F: (HEX) AGGCCTGACTACAAATCTGTT F: CCTGCCACAGAATGTCTGAA R: (FAM) GCGTTTAGGTAACTGCTGGTGT R: GTGCTTGCTCCTGAAGCTCT R: (HEX) GTGAGGGTCACGCTTTATGC R: TGTTGTTCTGTAAACCCAGCTT R: CGCTTCCAGGGTTAGTTGTT R: F: CAGACCCTTAGTGGCTTGCT (NED) CAGACCCTTAGTGGCTTGCT F: ATGGCATGGTAGGTGATGGT R: F: GCGAGACTTGTGCCTGACTT (HEX) F: CTTGTGGCAGAAGCTCTTGA R: AGGAGCAGATGCTCTCAGGA R: F: ACAGCAGACTGTCAAACACAAA (HEX) ACAGCAGACTGTCAAACACAAA F: (FAM) AACCTTAACCACTGACCCAAA F: (HEX) TATCTGATAAGCAGGGAAGACC F: CACGACAGGCCTAATCAATC F: CATGACCCAGGAAACACCTT F: (FAM) AGGCTCAAGGGTCCACATC F: (FAM) AGCTTGGTGTAATTGCAAAGAC F: ATTGTGCAAAGGGAGGTGTT F: CTGGTTAGAGCTTTGTCATTGC R: AAATTTGCAGTGCGTCCTAA R: (NED) TTGAGGCCTACAGATGTCAGA R: (FAM) AAATGCCCTTGATGCTGTATG R: GGGCCAAGCTGAACAGTAAA R: (HEX) TCTTTCTCGTGACCGATGTG R: AGAAACGCTTGGATGCAACT R: AGCAGGCTGAGCTTAACTGG R: * 4 20 13

12 * 12 inbreeding coefficient; P inbreeding coefficient; * * TA(CA) * * * TACA * * TA(CA) CG(CA) IS : 15 2 17 13 12 14 18 17 14 25 12 13 3 26 13 13 19 12 F (CA) (CA) (GA) (CA) (CA) (CA) (CA) (CA) (CA)

Repeat motif (CA) (CA)

(CA) (GATA)

(CA) (CA) (CA) CACC(CA) (CA) (CA) (CA) erozygosity; erozygosity; departures from H-W expectations after Bonferroni correction. An asterisk (*) indicates that the sequence of the microsatellite array could not be determined in could not be determined in An asterisk (*) indicates that the sequence of microsatellite array departures from H-W expectations after Bonferroni correction. its entirety because of insufficient overlap the 2 paired-end reads (the reported number is highest repeats con tained within 1 paired- BT6 BT29 BT47 BT67 BT68 BT73 BT81 BT88 BT4

Summary data for 20 microsatellites developed for the blackfin tuna ( tuna blackfin the for developed microsatellites 20 for data Summary end readings). Locus BT95 BT22

BT11 BT91

BT18 BT27 BT31 BT37 BT83 BT20 BT71 Antoni et al: Microsatellite markers for Thunnus atlanticus 325

BT47, and BT91, we caution against the use of these Tomback, S. J. Oyler-McCance, J. A. Fike, S. L. Lance, J. W. 3 microsatellites for population genetic studies of the Streicher, E. N. Smith, and D. D. Pollock. blackfin tuna. Genotypic frequencies at locus BT27 did 2012. Rapid microsatellite identification from Illumina paired-end genomic sequencing in two birds and a not depart significantly from Hardy-Weinberg equilib- snake. PLoS ONE 7(2):e30953. doi:10.1371/journal. rium expectations, but analyses in Micro-Checker in- pone.0030953 dicated that null alleles may occur at this microsatel- Collette, B. B., and C. E. Nauen. lite. This marker, therefore, will need to be evaluated 1983. FAO species catalogue, vol. 2. Scombrids of the further with larger sample sizes to determine its suit- world. An annotated and illustrated catalogue of tunas, ability for studies of population genetics of the black- mackerels, bonitos and related species known to date. fin tuna. Finally, locus BT71 showed occurrence of rare FAO Fish. Synop. 125, 137 p. FAO, Rome. Goudet, J.. shifts of one base pair (i.e., consecutive alleles differed 1995. FSTAT (vers. 1.2): a computer program to calcu- by only one base pair, departing from the expected pat- late F-statistics. J. Hered. 86:485–486. tern of variation at microsatellites). Because of the lim- Luckhurst B. E., T. Trott, and S. Manuel. ited resolution of acrylamide gels, the ability to reliably 2001. Landings, seasonality, catch per unit effort and score alleles that differ by one base pair is challenging tag-recapture results of yellowfin tuna and blackfin and, consequently, this locus may be prone to elevated tuna at Bermuda. Am. Fish. Soc. Symp. 25:225–234. rates of scoring errors. The microsatellite markers de- Koressaar T., and M. Remm. 2007. Enhancements and modifications of primer design veloped during our study will be available for conduct- program Primer3. Bioinformatics 23:1289–1291. ing studies of the genetic stock structure of blackfin Mathieu, H., C. Pau, L. Reynal, and D. Theophille. tuna, investigations that are needed to assist in the 2013. What do we know about blackfin tuna Thunnus( design of management plans for the sustainability of atlanticus)? In Proceedings of the 65th Gulf and Ca- this species. ribbean Fisheries Institute; Santa Marta, Colombia, 5–9 November 2012, p. 245–249. Gulf and Caribbean Fisheries Institute, Fort Pierce, FL. Acknowledgments Raymond, M., and F. Rousset. 1995. GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. J. Hered. We thank K. Jones for help with the implementa- 86:248–249. tion of PAL_FINDER software and C. Pau for assis- Rice, W. R. tance with collection of blackfin tuna specimens. This 1989. Analyzing tables of statistical tests. Evolution work was supported by the Moored Fish Aggregating 43:223–225. Device in the Lesser Antilles (MAGDELESA) project Rousset, F. cofounded by IFREMER and the Fonds européen de 2008. GENEPOP’007: a complete re-implementation of développement régional (FEDER) within the frame- the GENEPOP software for Windows and Linux. Mol. Ecol. Resour. 8:103–106. work of the INTERREG Caraïbes Programme (con- Sambrook, J., E. F. Fritsch, and T. Maniatis. tract no. 13/1210870). K. Naghshpour was supported 1989. Molecular cloning: a laboratory manual, 2nd ed., by the Tidelands Trust Fund Program of the Missis- 1659 p. Cold Spring Harbor Laboratory Press, Cold sippi Department of Marine Resources (award no. S- Spring Harbor, NY. 13-RIP-USM/GCRL-01). Views expressed in this article Singh-Renton, S., and J. Renton. are those of the authors and do not necessarily reflect 2007. CFRAMP’s large pelagic fish tagging program. views of the sponsors. Gulf Caribb. Res. 19(2):99–102. Untergasser, A., I. Cutcutache, T. Koressaar, J. Ye, B. C. Fair- cloth, M. Remm, and S. G. Rozen. 2012. Primer3—new capabilities and interfaces. Nucle- Literature cited ic Acids Res. 40(15):e115. doi:10.1093/nar/gks596 Van Oosterhout, C., W. F. Hutchinson, D. P. M. Wills, and P. Carvalho, G. R., and L. Hauser. Shipley. 1994. Molecular genetics and the stock concept in 2004. MICRO-CHECKER: software for identifying and ­fisheries. Rev. Fish Biol. Fish. 4:326–350. correcting errors in microsatellite data. Castoe, T. A., A. W. Poole, A. P. J. de Koning, K. L. Jones, D. F. Mol. Ecol. Notes 4:535–538.