Barcoding Nemo: DNA-Based Identifications for the Ornamental Fish Trade

Citation: Steinke D, Zemlak TS, Hebert PDN ( Barcoding Nemo: DNA-Based Identifications for the Ornamental Fish Trade Dirk Steinke. 0 Tyler S. Zemlak. 0 Paul D. N. Hebert 0 Robert DeSalle, American Museum of Natural History, United States of America 0 Canadian Centre for DNA Barcoding, Biodiversity Institute of Ontario, University of Guelph , Guelph, Ontario , Canada Background: Trade in ornamental fishes represents, by far, the largest route for the importation of exotic vertebrates. There is growing pressure to regulate this trade with the goal of ensuring that species are sustainably harvested and that their point of origin is accurately reported. One important element of such regulation involves easy access to specimen identifications, a task that is currently difficult for all but specialists because of the large number of species involved. The present study represents an important first step in making identifications more accessible by assembling a DNA barcode reference sequence library for nearly half of the ornamental fish species imported into North America. Methodology/Principal Findings: Analysis of the cytochrome c oxidase subunit I (COI) gene from 391 species from 8 coral reef locations revealed that 98% of these species exhibit distinct barcode clusters, allowing their unambiguous identification. Most species showed little intra-specific variation (adjusted mean = 0.21%), but nine species included two or three lineages showing much more divergence (2.19-6.52%) and likely represent overlooked species complexes. By contrast, three genera contained a species pair or triad that lacked barcode divergence, cases that may reflect hybridization, young taxa or taxonomic over-splitting. Conclusions/Significance: Although incomplete, this barcode library already provides a new species identification tool for the ornamental fish industry, opening a realm of applications linked to collection practices, regulatory control and conservation. - . These authors contributed equally to this work. Over the last 50 years, the international trade in ornamental fishes has grown rapidly. Beginning as a small export fishery in parts of the Indo-Pacific region during the early 20th century, the industry now involves most tropical and subtropical regions, generating some US$200300 million annually for fishes alone [1]. Target species derive from freshwater and marine environments and include invertebrates (corals, crustaceans, anemones) and vertebrates (fishes) from both natural and captive breeding sources. Most marine fishes derive from wild populations collected from coral reef habitats along the coastal margins of the Atlantic, Pacific and Indian Oceans. Some 800 marine fish species, about 5% of all marine taxa, are involved in this trade with 70% of sales directed to North America [1]. DNA barcoding, the analysis of sequence diversity in a standardized gene region, has gained considerable validation as a tool for species identification and discovery. Several studies have demonstrated its effectiveness for identifying both marine and freshwater fishes [24], provoking an effort to build a barcode library for all fish species [5]. Currently, records are available for 41771 fishes, representing 6566 fish species on the Barcode of Life Data System, BOLD [6]. DNA barcoding also provides an independent means of testing the validity of existing taxonomic systems, revealing cases of inappropriate synonymy or overlooked taxa. For example, Ward et al. [7] and Zemlak et al. [8] found several likely cases of overlooked diversity in marine fishes. These results suggest that the species boundaries need to be examined for the heavily exploited populations targeted by the aquarium trade, to properly inform conservation strategies and planning. The current study has constructed a DNA barcode database for marine fishes that are commonly imported by the pet trade to Canada. This investigation not only provides a further test of the capacity of DNA barcoding to deliver accurate species identifications, but also employs DNA barcodes to highlight potentially cryptic species and discusses some likely impacts of a DNA-based identification system on the ornamental fish trade. Materials and Methods Taxonomic Coverage Whenever possible, at least 5 adults were analyzed per species with a total of 1638 individuals, representing 391 species. All specimens are deposited as vouchers in the Biodiversity Institute of Ontario, Guelph, Canada. Collection details are available from the Barcode of Life website (www.barcodinglife.org) in the project file Aquarium Imports and are listed in Table S1 by taxonomic rank following Nelson [9]. All samples were wild caught, dead on arrival specimens provided by a Canadian importer of marine ornamental fishes. Specimens were frozen immediately and subsequently imaged on a flatbed scanner following a standard protocol [10]. DNA Analysis A sample of muscle tissue from each specimen was extracted using an automated Glass Fiber protocol [11]. The 650 bp barcode region of COI was subsequently amplified under the following thermal conditions: 2 min at 95uC; 35 cycles of 0.5 min at 94uC, 0.5 min at 52uC, and 1 min at 72uC; 10 min at 72uC; held at 4uC. The 12.5 ml PCR reaction mixes included 6.25 ml of 10% trehalose, 2.00 ml of ultrapure water, 1.25 ml 10X PCR buffer [200 mM Tris-HCl (pH 8.4), 500 mM KCl], 0.625 ml MgCl2 (50 mM), 0.125 ml of each primer cocktail (0.01 mM, using primer cocktails C_FishF1t1 and C_FishR1t1 from [12], 0.062 ml of each dNTP (10 mM), 0.060 ml of PlatinumH Taq Polymerase (Invitrogen), and 2.0 ml of DNA template. PCR amplicons were visualized on a 1.2% agarose gel E-GelH (Invitrogen) and bidirectionally sequenced using sequencing primers M13F or M13R [12] and the BigDyeH Terminator v.3.1 Cycle Sequencing Kit (Applied Biosystems, Inc.) on an ABI 3730 capillary sequencer following manufacturers instructions. Sequence data are available on both the Barcode of Life Data System (BOLD, http://www.boldsystems.org, see [6]) and GenBank (Accession numbers in Table S1). Specimen and collection data, sequences, specimen images, and trace files are listed in the same project folder as collection data (Aquarium Imports) on BOLD. A Kimura 2-parameter (K2P) distance metric was employed for sequence comparisons [13]; genetic distances and initial Neighbor-joining (NJ) clustering used the BOLD Management & Analysis System. Confidence in estimated relationships of NJ tree topologies was evaluated by a bootstrap analysis with 1,000 replicates with MEGA version 3.1 [14]. A threshold of 2.0% intraspecific sequence divergence was employed to screen for overlooked species following the recommendation that a sequence divergence value set at 10X the average within species variation (0.21 in this study- see later) is likely to be effective in this regard [15]. COI amplicons were recovered from all 1638 individuals and there was no evidence of indels or stop codons which (...truncated)