Probing Diversity in Freshwater Fishes from Mexico and Guatemala With
Total Page:16
File Type:pdf, Size:1020Kb
Journal of Fish Biology (2009) 74, 377–402 doi:10.1111/j.1095-8649.2008.02077.x, available online at http://www.blackwell-synergy.com Probing diversity in freshwater fishes from Mexico and Guatemala with DNA barcodes M. VALDEZ-MORENO*†, N. V. IVANOVA‡, M. ELIAS´ -GUTIERREZ *, S. CONTRERAS-BALDERAS§ AND P. D. N. HEBERT‡ *El Colegio de la Frontera Sur, Av. Centenario km 5.5, Chetumal 77014, Quintana Roo, Mexico, ‡Biodiversity Institute of Ontario, University of Guelph, Guelph, Ontario, N1G 2W1 Canada and §Bioconservacio´n A. C., A.P. 504, San Nicolas´ de los Garza 66450, Nuevo Leo´n, Mexico (Received 19 October 2007, Accepted 21 August 2008) The freshwater fish fauna of Mexico and Guatemala is exceptionally diverse with >600 species, many endemic. In this study, patterns of sequence divergence were analysed in representatives of this fauna using cytochrome c oxidase subunit 1 (COI) DNA barcodes for 61 species in 36 genera. The average divergence among conspecific individuals was 0Á45%, while congeneric taxa showed 5Á1% divergence. Three species of Poblana, each occupying a different crater lake in the arid regions of Central Mexico, have had a controversial taxonomic history but are usually regarded as endemics to a single lake. They possess identical COI barcodes, suggesting a very recent history of isolation. Representatives of the Cichlidae, a complex and poorly understood family, were well discriminated by barcodes. Many species of Characidae seem to be young, with low divergence values (<2%), but nevertheless, clear barcode clusters were apparent in the Bramocharax–Astyanax complex. The symbranchid, Opisthernon aenigmaticum, has been re- garded as a single species ranging from Guatemala to Mexico, but it includes two deeply divergent barcode lineages, one a possible new endemic species. Aside from these special cases, the results confirm that DNA barcodes will be highly effective in discriminating freshwater fishes from Central America and that a comprehensive analysis will provide new important insights for understanding diversity of this fauna. # 2009 The Authors Journal compilation # 2009 The Fisheries Society of the British Isles Key words: biodiversity; BOLD; COI; mtDNA; taxonomy. INTRODUCTION The fauna of Mexico includes >500 freshwater fish species, almost matching the combined total for the U.S.A. and Canada (Miller, 2005). Another 151 freshwater fishes are known from Guatemala (Froese & Pauly, 2007). High endemism has been recognized in both countries (Miller, 2005; Valdez-Moreno et al., 2005; Froese & Pauly, 2007). Reflecting this diversity, species and generic boundaries in several families are controversial (Miller, 2005). Moreover, †Author to whom correspondence should be addressed. Tel.: þ52 983 8350440 ext. 4307; fax: þ52 983 8350440 ext. 4321; email: [email protected] 377 # 2009 The Authors Journal compilation # 2009 The Fisheries Society of the British Isles 378 M. VALDEZ-MORENO ET AL. species discovery in this region remains active with at least 13 new species described from Mexico within the past 5 years, including one new genus of cichlids (Rocio) and a new family of catfishes, the Lacantuniidae (Barbour, 2002; Lozano-Vilano, 2002; Minckley et al., 2002; Kallman et al.,2004;Lyons& Mercado-Silva, 2004; Rodiles-Hernandez´ et al., 2005; Strecker, 2005; Schmitter- Soto, 2007). In most cases, understanding of this fish diversity is solely reliant on morphological studies; few taxa have been subjected to molecular analysis (Strecker, 2004; Barriga-Sosa et al., 2005). Mitochondrial sequence diversity has been used to distinguish closely allied species for >20 years (Avise, 1994). More recently, ‘DNA barcoding’, the survey of sequence diversity in a 648 bp segment of the mtDNA gene cytochrome c oxi- dase subunit 1 (COI) has been proposed as a standard tool for species-level iden- tifications of all animals (Hebert et al., 2003). Aside from the benefits of creating a DNA-based identification system, DNA barcoding is an effective tool for gain- ing an initial sense of the patterning of genetic divergences. Because of this, sev- eral authors have suggested that DNA barcoding will aid rapid progress in traditional taxonomic work by speeding the discovery of new species and in the recognition of synonymies (Gregory, 2005). Although barcoding remains controversial in some circles (Lipscomb et al., 2003; Moritz & Cicero, 2004; Fitzhugh, 2006), it has now been shown to be highly effective in distinguishing species of collembolans (Hogg & Hebert, 2004), ants (Smith et al., 2005), butter- flies (Hebert et al.,2004a), birds (Hebert et al.,2004b) and mammals (Clare et al., 2007; Borisenko et al., 2008). Although fishes represent nearly 50% of all vertebrates, only one large-scale DNA barcoding study has been carried out on this group, and it focused on Australian marine species. This study re- vealed that DNA barcodes were able to identify 100% of the 207 species exam- ined (Ward et al., 2005) and that barcodes were subsequently used to identify marine fish larvae from Australian waters (Pegg et al., 2006; Victor, 2007). The present study builds on this work, by beginning assembly of a DNA bar- code library for the freshwater fishes of Mexico and Guatemala with a focus on species whose taxonomic status has been controversial. MATERIALS AND METHODS Fishes were collected in 24 diverse freshwater environments from Mexico and Guatemala, ranging from southern tropical forests to semi-deserts in the north (Fig. 1). Details on collect- ing localities, co-ordinates and dates are available within the project file ‘Freshwater Fishes of Mexico’ on the Barcode of Life Data System (BOLD) (Ratnasingham & Hebert, 2007). A small piece (1–3 mm3) of muscle was removed from the left side of each fish and placed in 100% ethanol using tools that were flame sterilized before sampling each specimen. The remainder of each fish is preserved as a reference voucher in the Fish Collection of El Colegio de la Frontera Sur, Chetumal Unit (ECOCHP), Escuela Na- cional de Ciencias Biologicas´ (ENCB, IPN) and the Universidad Autonoma´ de Nuevo Leon´ (Monterrey, Mexico), and accession numbers are included in BOLD. Whenever possible, at least five adults of each species were sampled. In the case of difficult taxa, more specimens were obtained. All identifications were based on specialized literature and consultation with taxonomic specialists in cases when the identification was partic- ularly difficult. Names follow FishBase (Froese & Pauly, 2007). DNA barcoding was carried out at the Canadian Centre for DNA Barcoding using standard protocols (Hajibabaei et al., 2005). DNA was extracted from 1 mm3 tissue plugs # 2009 The Authors Journal compilation # 2009 The Fisheries Society of the British Isles, Journal of Fish Biology 2009, 74, 377–402 FRESHWATER FISH BARCODES 379 FIG. 1. Approximate locations of all collection sites in Mexico and Guatemala for freshwater fish barcoded. Additionally, the following marks denote positive localities for: , Astyanax mexicanus; , Astyanax aeneus; , Bramocharax caballeroi, only found in Catemaco lake; , Bramocharax baileyi; , Poblana species. that were sub-sampled into a vertebrate lysis buffer with proteinase K and digested over- night at 56° C. Genomic DNA was subsequently extracted using a membrane-based approach on a Biomek NXÒ (www.pall.com) liquid handling station using AcroPrep 96 (www.beckman.com), 1 ml filter plates with 1Á0 mm PALL glass fibre media (Ivanova et al., 2006). A 652–658 bp segment of COI was amplified using different combinations of fish primers: FishF1, FishR1, FishF2, FishR2 (Ward et al., 2005) or the M13-tailed primer cock- tails C_Fish F1t1 – C_FishR1t1 and C_VF1LFt1 – C_VR1LRt1 (Ivanova et al., 2007). The 12Á5 ml polymerase chain reaction (PCR) mixes included 6Á25 ml 10% trehalose, 2 ml ultrapure water, 1Á25 ml10Â PCR buffer, 0Á625 MgCl2 (50 mM), 0Á125 ml of each primer (0Á01 mM), 0Á0625 ml of each dNTP (0Á05 mM), 0Á625 ml Taq polymerase (www.net.com) and 2Á0 ml DNA template. Amplification protocols followed those described in earlier publications (Hajibabaei et al., 2005). PCR products were visualized on pre-cast agarose gels (E-GelsÒ; www.invitrogen.com), and the most intense products were selected for sequencing. Products were labelled with the BigDyeÒ Terminator v.3.1 Cycle Sequencing Kit (www.appliedbiosystems.com) using standard methods (Hajibabaei et al., 2005) and sequenced bidirectionally using an ABI 3730 capillary sequencer. Sequences were edited and aligned using SEQSCAPE v.2.1.1 (Applied Biosystems Inc.). Sequence divergences were calculated using the Kimura two-parameter (K2P) distance model (Kimura, 1980). Sequence data, electropherograms, trace files, primer details, photo- graphs and collection localities for specimens are available within the project file on BOLD (http://www.barcodinglife.org; see further information). Sequences have also been deposited in GenBank (http://www.ncbi.nlm.nih.gov/Genbank). All accession num- bers in both databases are included in Appendix I. Neighbour-joining (NJ) trees of K2P distances were created to provide a graphical representation of the patterning of divergence between species (Saitou & Nei, 1987). # 2009 The Authors Journal compilation # 2009 The Fisheries Society of the British Isles, Journal of Fish Biology 2009, 74, 377–402 380 M. VALDEZ-MORENO ET AL. A simplified ID tree of all the species was generated with MEGA 3 software (Kumar et al., 2004). RESULTS In total, 427 fishes were sequenced including 61 species in 36 genera and 15 families, representing 10% of the known fauna. Accession numbers to BOLD and GenBank sequences for each specimen are provided in Appendix I. Read lengths were all over 600 bp long, and no insertions, deletions or stop codons were observed. The full K2P–NJ tree is provided in Appendix II, but Fig. 2 provides an overview of the patterning of sequence divergences. The average K2P distance among conspecific individuals for all species was 0Á45% compared with 5Á1% for species within the genera (Table I). Overall, 93% of nominal species recognized by prior taxonomic work were dis- criminated by barcodes. The most conspicuous case of failed resolution involved three species of Poblana (Atherinopsidae).