Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/de-gruyter/extending-dna-barcoding-coverage-for-lake-whitefish-coregonus-aWW1WDtn9n
- Publisher
- de Gruyter
- Copyright
- Copyright © 2015 by the
- ISSN
- 2299-1077
- eISSN
- 2299-1077
- DOI
- 10.1515/dna-2015-0007
- Publisher site
- See Article on Publisher Site
Abstract
DNA barcoding is a useful tool for both species identification and discovery, but the latter requires denser sampling than typically used in barcode studies. Lake Whitefish (Coregonus clupeaformis) is a valuable species, fished traditionally, commercially, and recreationally in Lake Huron. Based on the natural geographic and bathymetric separation of the three major basins in Lake Huron, the potential separation of Lake Whitefish within these basins, and the variation among life history (early and late spawning), we predicted that Lake Huron might harbour cryptic lineages of Lake Whitefish at the basin level. To test this prediction, DNA barcodes of the mitochondrial 5' cytochrome c oxidase subunit I (COI) gene sequences were recovered from spawning phase Lake Whitefish (n = 5 per site), which were collected from sites (n = 28) around Lake Huron during Fall 2012. These sequences, combined with other publically available DNA barcodes from the Barcode of Life Data System (BOLD), revealed twelve unique haplotypes across North America, with seven unique to Lake Huron. The dominant haplotype was found throughout Lake Huron and east to the St. Lawrence River. No deep divergences were revealed. This comprehensive lake-wide sampling effort offers a new perspective on C. clupeaformis, and can provide insight for environmental assessments and fisheries management. Keywords: haplotype, COI, Coregoninae 1 Introduction Fish identification can be challenging because of ontogenetic variation and a limited understanding of taxon boundaries [1]. DNA barcoding is a tool for species identification [2] that applies to all life stages. It relies on the existence of species-specific differences found in short standardized mitochondrial 5' cytochrome c oxidase subunit I (COI) gene sequences, and has successfully been used for species discrimination in fishes (e.g., [3-5]). As a tool, barcoding lies at the interface of population genetics and phylogenetics [6]. Although DNA barcodes are primarily used for species identification, they also support species discovery in the well studied North American fauna by revealing deeply divergent genetic lineages that may represent cryptic species [7,8]. In order to appropriately apply DNA barcoding as a tool for species identification and discovery, there is a need to improve our knowledge of genetic variation both within and between species. This variation is important because species-level identification relies on the existence of a "barcode gap", which is a separation between the maximum intraspecific variation versus the minimum interspecific distance [9,10]. The Fish Barcode of Life (FISH-BOL) initiative has focused on generating DNA barcodes for all fish species globally [3,11]. DNA barcoding has proven useful in identifying North American freshwater fishes [4,11,12], with a 95% coverage of barcodes for Canadian freshwater fish species [4]. Although a sample size of five individuals per species was recommended at the commencement of FISH-BOL [13], there have been more recent suggestions that greater spatial coverage and increased sample size per *Corresponding author:Lauren M. Overdyk, Department of Integrative Biology, University of Guelph, 50 Stone Road East, Guelph, Ontario, Canada N1G 2W1, E-mail: loverdyk@uoguelph.ca Stephen S. Crawford, Robert H. Hanner: Department of Integrative Biology, University of Guelph, 50 Stone Road East, Guelph, Ontario, Canada N1G 2W1 Heather E. Braid: Institute for Applied Ecology New Zealand, Auckland University of Technology, Private Bag 92006, Auckland, New Zealand 1010 Stephen S. Crawford: Chippewas of Nawash, Unceded First Nation, 135 Lakeshore Blvd, Nevaashiinigmiing, Ontario, Canada N1G 2W1 Robert H. Hanner: Biodiversity Institute of Ontario, University of Guelph, 50 Stone Road East, Guelph, Ontario, Canada N1G 2W1 © 2015 Lauren M. Overdyk et al. licensee De Gruyter Open. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License. species are necessary to accurately represent intraspecific genetic variation [8]. Lake Whitefish, Coregonus clupeaformis, is a culturally and socio-economically important species in the Laurentian Great Lakes. Lake Huron is comprised of three major basins (Main Basin, North Channel and Georgian Bay), and is the second largest of the Great Lakes and fourth largest lake in the world [14]. Within the lake, the three major basins are partially separated bathymetrically by ridges and channels [15,16]. Lake Whitefish are found throughout the lake, with a heterogeneous distribution of spawning grounds through each of the basins. There have been some reports at some spawning grounds of different phases of whitefish spawners through the spawning season (e.g. early and late spawning stages), typically extending from October to December with phenotypes that have been distinguished by First Nations and harvesters [17,18]. For example, interviews with whitefish commercial fishermen suggested that when fishing for lake whitefish in Lake Huron there was a chance of seeing up to three different morphologies of Lake Whitefish; some forms were long and not so deep, while others were more rounded [17]. It has also been suggested that smalled sized whitefish were present in the south end of Georgian Bay, while the more northern fish were `thicker' [17]. DNA barcodes have shown in many cases that subtle morphological variation is correlated with barcode divergence; corroboration from independent data partitions highlights the existence of cryptic diversity [19-21]. The variation in morphology and spawning chronology, combined with the inherent challenges associated with whitefish taxonomy generally, suggest the possibility of cryptic Lake Whitefish lineages between and/or within basins. In previous studies, DNA barcoding has been effective in discriminating Lake Whitefish from other species (i.e., [12]). However, the results should be considered preliminary due to limited sample sizes. Increased sampling would be needed to detect potential cryptic lineages as previously demonstrated in an investigation of sculpins (Cottus spp.) in a North American riverine system [8]. To date, there have been few Lake Whitefish barcodes available on the Barcode of Life Data system (BOLD; [22]) with little representation of possible haplotype variation. The goal of this study was to investigate basin-level haplotype variation and potential cryptic diversity of Lake Whitefish in Lake Huron, by expanding the barcode library across known spawning sites and times. This study is novel in that it represents the first comprehensive barcode sampling effort for a species inhabiting a large and bathymetrically complex lake ecosystem. 2 Materials and Methods 2.1 Specimen Collection A lake-wide sampling effort for adult Lake Whitefish was undertaken in the fall of 2012 in Lake Huron (Figure 1; Table 1). Spawning-phase Lake Whitefish were sampled from twenty-eight locations across the three basins of Lake Huron: the Main Basin (n = 16), the North Channel (n = 4), and Georgian Bay (n = 8). Early and late samples were taken from Sites 16-18 (Cape Hurd, Stokes Bay, and Howdenvale) along the Western shore of the Saugeen (Bruce) Peninsula based on known variation in spawning activity of Lake Whitefish in these areas (Table 1). Five different individuals, including at least two males and two females, were sampled from each site for DNA barcoding. Two muscle fillets were taken from each individual and maintained at -20oC, thawed briefly for subsampling of approximately 1 gram of tissue for DNA extraction. We had previously obtained representative Lake Whitefish from Stokes Bay and Sharbot Lake, which were processed in the same manner. 2.2 DNA Barcoding DNA was extracted from the selected individuals using Xytogen Animal Extraction Kits (Xytogen, Perth, Australia) following manufacturing guidelines. DNA was diluted with ddH2O to 1:10 before use in PCR (~50 ng/µl). The 652 bp "barcode" region of COI [2] was amplified using universal fish primers VF1i_t1 (TGT AAA ACG ACG GCC AGT TCT CAA CCA ACC AIA AIG AIA TIG G) and VR1i_t1 (CAG GAA ACA GCT ATG ACT AGA CTT CTG GGT GIC CIA AIA AIC A) [23]. PCR amplification was carried out in 12.5 µL reaction volumes with 6.25 µL 10% trehalose, 2 µL ddH2O, 1.25 µL 10X buffer, 0.625 µl MgCl2 (50 mM), 0.1 µL VF1i_t1 (10 µm), 0.1 µL VR1i_t1 (10 µm), 0.0625 µL 10 mM dNTPs, 0.06 µL Platinum Taq polymerase (5 U/µL), and 2 µL of DNA. The reaction profile was: an initial hot start at 94°C; followed by five cycles with a denaturation at 94°C for 30 sec, annealing temperature of 50°C for 40 sec, and extension at 72°C for 1 min; followed by thirty five cycles with a denaturation at 94°C for 30 sec, annealing temperature of 54°C for 40 sec, and extension at 72°C for 1 min; a final extension at 72°C for 10 min and hold at 4°C indefinitely. Amplification success of PCR products was ascertained visually using 2% Agarose E-gels (Invitrogen), where a clear band on the gel indicated successful amplification. DNA Barcoding Lake Huron Lake Whitefish Figure 1: DNA barcode haplotype variation found in Lake Whitefish (Coregonus clupeaformis) across North American ecoregions. Each colour represents a different haplotype. Haplotypes were included from Yukon River, Swan Lake, St. Laurence River, Sharbot Lake, and Lake Huron. The size of each pie chart is proportional to the sample size from each location; however, the most dominant haplotype (A) has been excluded from Lake Huron for the purpose of visualization of the less prevalent haplotypes, but all haplotypes for Lake Huron can be seen in Figure 2. Sequencing reactions for PCR products used BigDye v3.1 and the same primers used for amplification, and sequenced bidirectionally. Sequences were edited using Sequencher version 4.9 (Gene Codes Corporation, Ann Arbor, MI, USA) and aligned using MEGA 5.2 [24]. Sequences were uploaded to the Barcode of Life Database (BOLD; [22]) under the container project "Lake Whitefish (Coregonus clupeaformis)" (project code: LOLW), in the public project "Lake Wide Lake Huron Lake Whitefish" (project code: LWLHW; accession numbers KP978068KP978225). These were combined with other sequences for Lake Whitefish that we generated from Sharbot Lake (60 sequences, project code: SLHL; accessione numbers KP978226 - KP978312) and Stokes Bay (28 sequences, project Stokes Bay, Ontario, Lake Whitefish (project code: SBOLW; accession numbers KP978018-KP978067). Other sequences were obtained from public projects available on BOLD: eight sequences from locations across Canada were taken from the project Barcoding of Canadian Freshwater Fishes (project code: BCF). Together, these sequences were combined into a single "data set" (DS-CORG) on BOLD and given a unique digital object identification number (http://dx.doi.org/10.5883/DS-CORG). Novel sequences from this study were also submitted to GenBank (accession numbers KP978068- KP978225). 2.3 Haplotype Analysis A haplotype network was generated for the haplotypes of Lake Whitefish using a median-joining network [25] constructed in Network 4.6.1.2 using default settings (www.fluxus-engineering.com). Distribution maps of haplotypes were created using ArcGIS 10.2 (Environmental Systems Research Institute [ESRI], Redlands, CA). The map of DNA barcode haplotypes from across North America, excluding the dominant haplotype A, includes a shapefile of ecoregions based on Abell et al. [26] from http://www.feow.org. The Lake Huron coastline data were mapped using GIS layers (glgis_gl_shore_noaa_70k and lake_huron_bathymetry) published by the Great Lakes Information Network (http://gis.glin.net/). The size of each pie is proportional to the number of individuals, and each colour symbolizes a different haplotype (Figure 1). Sample sizes are proportional in Figure 1, except for Lake Huron due to the large number of haplotype A, which would have obscured the figure. Table 1: Collection site data for the lake whitefish (Coregonus clupeaformis) for the 2012 sampling effort in Lake Huron. Map number corresponds to the site number indicated on Figure 1; Basin name abbreviations are NC = North Channel, GB = Georgian Bay, and MB = Main Basin; the sex of fish sampled is indicated by M (male) and F (female); early designated the start of the fall spawning season, late designates the end of the fall spawning season. Site name Thessalon South Cockburn Blind River Vidal Bay Burnt Island Providence Bay Bedford Island Wiki-Smith Bay Bad River Henvey Inlet Christian Island Maurice Point Meaford Cape Croker South Sydney Bay Cape Hurd Map number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Basin NC MB NC NC MB MB NC GB GB GB GB GB GB GB GB MB Latitude 46.23 45.87 46.16 45.94 45.8 45.64 46.01 45.86 45.91 45.84 44.88 44.72 44.62 44.85 44.91 45.18 Longitude -83.43 -83.46 -82.92 -83.09 -82.46 -82.31 -82.07 -81.63 -80.99 -80.73 -80.16 -80.07 -80.53 -80.94 -81.07 -81.75 Number of fish sampled 2M, 3F 3M, 2F 3M, 2F 3M, 2F 2M, 3F 2M, 3F 2M, 3F 2M, 3F 2M, 3F 3M, 2F 2M, 2F 2M, 3F 2M, 3F 3M, 2F 2M, 2F 2M, 3F early 3M, 2F late Stokes Bay 17 MB 45.02 -81.48 3M, 2F early 3M, 2F late Howdenvale 18 MB 44.85 -81.35 1M, 4F early 3M, 2F late Scougall Bank Kettle Point Sarnia Outer Saginaw Bay North Island East Tawas Alpena North Point Hammond Bay Search Bay 19 20 21 22 23 24 25 26 27 28 MB MB MB MB MB MB MB MB MB MB 44.39 43.20 43.03 44.15 43.87 44.27 44.86 45.03 45.50 45.98 -81.53 -82.04 -82.31 -83.08 -83.44 -83.46 -83.25 -83.24 -84.03 -84.50 2M, 3F 3M, 2F 2M, 3F 2M,2F 3M, 1F 3M, 2F 3M, 2F 3M, 2F 2M, 3F 2M, 3F DNA Barcoding Lake Huron Lake Whitefish 3 Results A total of 208 Lake Whitefish sequences were analyzed, associated with a single Barcode Index Number (BIN) for all known specimens, yielding a total of 12 different haplotypes. The BIN system forms clusters of barcode sequences that can document taxonomic diversity [27]. From the lake-wide Lake Huron sampling, full-length DNA barcodes were successfully obtained for 148 out of the 154 (96.1%) specimens representing nine haplotypes, seven of which were unique to Lake Huron (Figure 1, 2). In eastern North America, one dominant haplotype (A, dark blue) representing 91.8% (191 individuals) of the sequences analyzed appears to be the most predominant, with occurrences throughout Lake Huron and east to the St. Nicolas River near the St. Lawrence River (Figure 2). Four additional haplotypes were found in locations across Canada. Two individuals from the St. Nicholas River represented the unique haplotype J (light blue), which had a single base pair different from the dominant haplotype A. Three haplotypes were found that differed from the dominant haplotype A by two base pairs: three individuals from Swan Lake, British Columbia represented haplotype G (teal); two unique haplotypes were identified from single individuals from the Yukon River in the Yukon Territory for haplotype H (grey) and I (brown). In Lake Huron, the seven unique haplotypes occurred at low frequencies, with each differing from the dominant haplotype A by only a single base pair. Haplotype, B (dark green), was represented by three individuals from outer Saginaw Bay (Main Basin), North Island (Main Basin), and Bedford Island (North Channel). Haplotype C (light green) represented two individuals, one respectively from Stokes Bay and Loscombe Bank (Main Basin). There were five additional singleton unique haplotypes found in Lake Huron: Haplotype E (red), on the eastern shores of the Main Basin; Haplotypes K (pink) and L (purple) found in Stokes Bay on the eastern side of Main Basin; Haplotype F (yellow) on the western shores of the Main Basin; and Haplotype D (orange) in the North Channel. Of the three sites that were sampled in early and late spawning periods, both Stokes Bay (Site 17) and Howdenvale (Site 18) respectively exhibited one unique haplotype in the late spawning phase along with the dominant haplotype, while Cape Hurd (Site 16) did not show any unique haplotypes. The dominant haplotype was found across all spawning sites at all sampling periods (Figure 2B). 4 Discussion After extensive lake-wide sampling of individuals of spawning phase Lake Whitefish from Lake Huron, very little variation was found in the DNA barcodes, providing no evidence for basin level cryptic lineages, consistent with a single species (e.g. no disconnected haplotype networks, [28]). Haplotype A, which was dominant in the entire collection, was also dominant throughout all three Figure 2: DNA barcode haplotype variation found in Lake Whitefish (Coregonus clupeaformis); site numbers correspond to Table 1 and L1 and L2 are two sites sampled in BOLD project Stokes Bay, Ontario, Lake Whitefish [SBOLW]. (A) Haplotype network analysis of DNA barcode sequences from across North America; the size of the nodes corresponds to the number of individuals that share each haplotype; the colour of each unique haplotype corresponds to the pie charts in B; (B) Geographic distribution of haplotypes from Lake Huron at each site sampled during 2012, the size of each pie chart is proportional to the sample size from each location. basins of Lake Huron; including both the early and late spawning stages. In contrast, haplotype B (dark green), was found only in Lake Huron in a limited number of locations. Haplotype B was found in both Saginaw Bay (Main Basin) on the south western coast of Lake Huron and north of Douglas Point (Main Basin) on the east side of the bay. Additionally, haplotype B was found in the north side of Manitoulin Island (North Channel). These patterns of haplotype distributions do not suggest cryptic lineages [29,30]. Our comprehensive lake-wide sampling of Lake Whitefish barcodes does not provide evidence for the existence of cryptic species among three main basins of Lake Huron. Of the three sites that were sampled in both early and late spawning stages, two sites were found to have rare haplotypes in the late stage, but those individuals were always found to co-occur with individuals that had the dominant haplotype. The occurrence of unique haplotypes among late stage spawners may suggest a temporal, rather than a spatial trend in cryptic lineages; however, due to the co-occurrence of the dominant haplotype, this seems unlikely. The goal of this study was to investigate basin-level haplotype variation and examine the potential for cryptic diversity within Lake Whitefish in Lake Huron. While new haplotype variants were discovered at some sites (rare and in low frequencies), these haplotypes were only 1-2 nucleotides different from the dominant haplotype and always co-occurred with the dominant haplotype at the same collection site/time. From the substantial expansion of the Lake Whitefish barcode library, no cryptic diversity (e.g., deeply divergent lineage) was noted in Lake Huron and no basin-specific haplotype variation was revealed. This study represents one of the most comprehensive barcode surveys of an individual species ever undertaken and stands in contrast to Young et al. [8] riverine system, which discovered substantial cryptic diversity with increased sampling effort. This study fails to support the hypothesis that cryptic lineages of Lake Whitefish exist in Lake Huron and although it does reveal the presence of rare barcode haplotypes that seem to be unique to specific sampling sites, it does not address population structure. Such investigations require more sensitive methods and, ideally, sampling more individuals from each putative population, including individuals that were morphologically identified as different. It is important to note that population discrimination of Lake Whitefish in Lake Huron has been identified as a key ecological uncertainty for fisheries management and environmental assessments [18,31,32] and should be the focus of future work. However, further sampling of Lake Whitefish among the Laurentian Great Lakes and throughout the species range may yet provide evidence for cryptic whitefish diversity elsewhere and should be a topic for further exploration. Extensive sampling is a necessary step in the development of a robust species-specific PCR identification assay for any given species of interest [33]. Hence, this study paves the way for developing a more cost-effective method of identifying whitefish specimens, which might be a concern because of the species' status as a valued ecosystem component. Acknowledgements: This research was supported by the Saugeen Ojibway Nation (Chippewas of Nawash Unceded First Nation, Saugeen First Nation) in collaboration with Bruce Power Limited. We would like to also thank Nikole Freeman, Kelly Mulligan, Grace Burke, Ashley Wincikaby, Rebecca Eberts (University of Regina), Chris Somers (University of Regina), Wendy Lee Stott (USGS for samples), and Chris Ho (BOLD for bioinformatics support). Conflict of interest: Ms. Overdyk reports other financial support from Saugeen Ojibway Nation during the conduct of the study.
Journal
DNA Barcodes – de Gruyter
Published: Jan 1, 2015