Nuclear DNA identification of migrating bull trout captured at the Puget Sound Energy diversion dam on the White River, Washington State
Abstract
Bull trout (Salvelinus confluentus) is a char listed as threatened under the United States Endangered Species Act throughout its range in the coterminous United States. Substantial morphological similarities between bull trout and Dolly Varden (S. malma) make field identification difficult. This has resulted in an incomplete understanding of their distribution and abundance in Washington State where these two species occur sympatrically. We used three diagnostic nuclear loci to determine the species of char collected at a trap on the White River in southern Puget Sound (Washington State, USA). Each of the 104 samples revealed the expected bull trout genotype at all three loci. This work presents three principle results: (i) the presence of a migratory bull trout population in southern Puget Sound; (ii) no evidence of migratory Dolly Varden over 3 years; and (iii) no evidence of hybridization was detected. These results also demonstrate how molecular markers can provide information essential to the conservation and management of these species.
Introduction
Morphologically similar, bull trout (Salvelinus confluentus) and Dolly Varden (S. malma) are easily confused in areas of sympatry. Until recently, bull trout were believed to be a subspecies of the Dolly Varden taxon (Cavender 1978), however, multiple lines of morphometric and molecular evidence now demonstrate the specific status of bull trout (Cavender 1978; Grewe et al. 1990; Hass & McPhail 1991; Crane et al. 1994; Phillips et al. 1994, 1999). Further, fixed differences within mitochondrial DNA (mtDNA) sequences and several nuclear loci clearly and reliably distinguish bull trout from Dolly Varden (Cavender 1978; Grewe et al. 1990; Hass & McPhail 1991; Crane et al. 1994; Phillips et al. 1994, 1999; Taylor et al. 2001). Additionally, bull trout and Dolly Varden hybridize producing viable offspring and these markers can and have been used to detect hybrids.
This state of morphological similarity and ability to hybridize is similar to that seen in coastal cutthroat trout (Oncorhynchus clarki clarki) and coastal rainbow trout/steelhead (Oncorhynchus mykiss irideus) of western North America. These two species are sympatric and hybridize widely (Johnson et al. 1999) yet the fate of hybrids and the impact of hybridization on populations is unclear (Campton & Utter 1985; Hawkins & Quinn 1996; Hawkins 1997). Equally, the bull trout/Dolly Varden system provides another example and opportunity to study the ecology and hybridization of two very similar species and provide essential biological information to resource managers.
The bull trout is a char native to coastal and inland waters of western North America. Although bull trout populations are chiefly found inland throughout its range, exceptions to this distribution are known to occur in the streams of Puget Sound, Hood Canal, coastal Washington State, and in the Fraser River of British Columbia (Cavender 1978; Hass & McPhail 1991; McPhail & Taylor 1995; Leary & Allendorf 1997; Taylor et al. 2001). In these coastal areas, the distribution and life history of bull trout are unresolved because of the co-occurrence of Dolly Varden.
The United States Fish and Wildlife Service (USFWS) has identified five Distinct Population Segments (DPS) of bull trout in the coterminous United States and determined that each of these is threatened as allowed by the Endangered Species Act (ESA) (United States Fish and Wildlife Service 1999). The Coastal–Puget Sound bull trout DPS is particularly interesting for three reasons. First, it is the only DPS of the coterminous United States where bull trout and Dolly Varden are known to occur in sympatry (although the extent of overlap is unclear, see below). Second, it is one of two DPS that may include anadromous stocks (Cavender 1978; McPhail & Taylor 1995). Lastly, the precise distribution of bull trout, as well as Dolly Varden, within this DPS is uncertain (Cavender 1978; Hass & McPhail 1991; McPhail & Taylor 1995).
This study focuses on char from the Puyallup River system, in the southern portion of the Coastal–Puget Sound DPS. Presently, consensus on a management strategy for char in this region is lacking because basic information such as species composition and distribution is not available. To illustrate this, the USFWS recognizes two bull trout subpopulations in the Puyallup River system, while the Washington Department of Fish and Wildlife (WDFW) recognizes three and manages bull trout and Dolly Varden under a single plan as ‘native char’ (Washington Department of Fish and Wildlife 1998). Furthermore, the USFWS describes the status of the upper Puyallup River subpopulation as ‘unknown’ and the lower Puyallup subpopulation as ‘depressed’, whereas the WDFW describes the status of the three stocks it recognizes as ‘unknown.
In an effort to gain a better understanding of which char species are present in the Puyallup River system, the Northwest Fisheries Science Center, Conservation Biology Division was asked by the Puyallup Tribal Fisheries to determine the species of 104 char collected in the Puyallup River system from 1999 to 2001. We employed a multilocus approach using nuclear DNA markers that exhibit species-specific variation demonstrated in other studies of bull trout and Dolly Varden (McPhail & Taylor 1995; Baxter et al. 1997; Taylor et al. 1999). All 104 samples were unambiguously shown to be bull trout, and we explore the likelihood of hybridization, the presence of resident Dolly Varden, and the possibility of true anadromy in this migratory Puyallup River population.
Materials and methods
Sample collection
Samples for this study were collected by personnel of the Puyallup Tribal Fisheries in 1999, 2000, and 2001 (Figure 1). Ninety-seven of the 104 fin-clip tissue samples were collected from upstream migrating char at the Puget Sound Energy diversion dam on the White River near Buckley, WA, USA. During 2000, one sample was collected from the lower Puyallup River tidewater, and six samples were collected from the Electron Project dam on the Puyallup River. To minimize stress and handling, no attempt was made to identify the fish or to collect morphological measures other than fork length in the field. Each fish was finclipped and the tissue was placed into individual vials containing 95% ethanol. Although locally derived controls were not available, four known bull trout and Dolly Varden samples from elsewhere in Puget Sound were included as controls to verify the expected species-specific genotypes.
Map of study sampling locations. Site 1, Puget Sound Energy diversion dam; Site 2, Electron Project dam; Site 3, lower Puyallup River.
Sample preparation
Total genomic DNA was extracted from test samples (Conservation Biology Division Tissue Archive accession numbers 32868, 32869, and 32870) using a NuceloSpin High-Throughput kit (Clonetech). Eight samples from population number 32870 were extracted using a NucleoSpin Nucleic Acid Purification kit. No differences were evident in quality or yield associated with the different formats of what is essentially the same DNA extraction system (data not shown). Genomic DNA extractions were visualized by electrophoresis 2 µL of each sample through a 0.8% agarose gel, staining with SybrGold, and documentation with Polaroid 660 film. Each sample was quantified using an FLx 800 plate reader (Bio Tek Instruments) and diluted to 10 ng/µL. Control DNA samples of bull trout and Dolly Varden were kindly provided by Dr Eric Taylor of the University of British Columbia.
Molecular methods
We used the polymerase chain reaction (PCR) to amplify three loci known to discriminate bull trout from Dolly Varden (Taylor et al. 2001): growth hormone (GH), metallothionien B (MTB), and the internal transcribed spacer 1 of the ribosomal gene (ITS-1). GH and MTB are insertion/deletion (indel) variants whereas the ITS-1 is a restriction fragment length polymorphism (RFLP). Typical PCR conditions were as follows: 1×Taq buffer [500 mm KCl, 100 mm Tris–HCl (pH 9.0) and 1.0% Triton X-100], 1.5 mm MgCl2, 200 µm each dNTP, 0.1 µm each primer, 0.25 unit Promega Taq DNA polymerase and 10 ng genomic DNA template per 20 µL reaction. The thermal profile consisted of one cycle at 95 °C for 2 min to denature the DNA, followed by 40 cycles of; 94 °C for 1 min, 55 °C for 1 min and 72 °C for 1 min and a final extension cycle of 72 °C for 2 min See Taylor et al. (2001) for primer sequences and annealing temperatures for each locus.
After PCR amplification, GH and MTB products were visualized by electrophoresis of 5 µL of each PCR product through a 3.0% equivalent agarose/Synergel (Diversified Biotech) (see manufacturer's directions) using 1× TAE buffer, staining with SybrGold, and photodocumentation with Polaroid 660 film. Following amplification of ITS-1, 2 µL of PCR product was digested with SmaI for 24 h in a 15-µL total reaction volume. The restriction digest was visualized by electrophoresing the entire reaction volume as described above. All fish of the sample were genotyped at each locus from the gel photographs.
Results/Discussion
The results presented here convincingly indicate that all samples collected from 1999 to 2001 were bull trout. Among the 104 samples, 616 alleles represented by three nuclear loci were screened and only bull trout alleles were detected (seven individuals failed amplification for the GH and ITS-1 loci, though they were scored as bull trout for MTB). Moreover, it is unlikely that any of the samples were hybrids. At GH, each sample revealed a PCR product 540 bp in length, as expected for bull trout, and 60 bp smaller than expected for Dolly Varden. Likewise, every MTB sample revealed the expected bull trout PCR product 700 bp in length. Finally, 101 of 104 samples produced ITS-1 PCR products, and restriction digestion of these products revealed the expected bull trout fragments of 490 and 210 bp while none showed the expected Dolly Varden fragment of 700 bp.
At present the exact distribution of bull trout and Dolly Varden within Puget Sound and coastal Washington State is not known. Moreover, little is known about their life history in this region. Elsewhere, Dolly Varden are generally considered anadromous, and bull trout tend toward residency. However, both life history strategies are found in each species. It is not clear where anadromous and resident populations exist in Puget Sound and coastal Washington State, or whether both life history forms are expressed within a given population. This lack of basic life history information makes management of the species difficult and points to specific research opportunities.
In some salmonids such as coastal steelhead and coastal cutthroat trout, migratory life history types (adfluvial and anadromous) are associated with increased growth relative to resident individuals. However, all bull trout life histories can attain comparable size (R. Ladley, unpublished data). Because of this, although fork-length measurements of the individuals sampled in this study were available, no strong inference can be made about anadromy. It is clear that 103 of the fish captured were migratory because they passed the weirs at predictable times and directions during the year. However, the potential for anadromy exists because the Puyallup River system flows into Puget Sound and bull trout of this system may migrate to tidewater, if not into Puget Sound proper. In fact, the one sample not collected at a weir was collected in tidewater. More work, such as stable isotope analysis (Kline et al. 1998), is needed to draw conclusions regarding anadromy in bull trout of the Puyallup River.
We found no evidence of Dolly Varden in the samples analysed for this study. It may be that Dolly Varden do not occur in central and southern Puget Sound and therefore in the Puyallup River system as no Dolly Varden have been identified in taxonomic (Cavender 1978; Hass & McPhail 1991) or molecular studies (Leary & Allendorf 1997; Taylor et al. 2001). Personnel of both the WDFW and USFWS have expended considerable effort to detect either species of char in central and southern Puget Sound, yet there is no verified record of Dolly Varden south of the Skagit River (F. Goetz, US Corps of Engineers, 4/17/02, personal communication). This uncertainty regarding the distribution of Dolly Varden is in contrast to the clear documentation of bull trout in the Puyallup River (from which the bull trout type specimen was described; Cavender 1978)
We found no evidence that migratory Dolly Varden are present in the Puyallup River. We do not think this is a case of missed-opportunity as the trap is operated year-round by the United States Army Corp of Engineers and is considered to be an absolute barrier to migration. Nevertheless, it is possible that resident Dolly Varden may exist in the upstream reaches of the Puyallup River. Because residents do not make substantial migrations it is unlikely that resident individuals would be taken at the trap. We conclude that if Dolly Varden are present they are probably resident, whereas bull trout are migratory (and possibly anadromous). Our conclusion is consistent with the observation that the typical migratory behaviour observed in these species is reversed in populations of western Washington and southwestern British Columbia relative to populations in the remainder of their range (McPhail & Taylor 1995).
A multilocus approach to species identification as presented here can result in a greater probability of positively identifying the species of the sample and detecting hybrids because more of the genome is sampled. Considering no heterospecific alleles were detected in this study, it is unlikely that hybridization occurs in this migratory population. Other studies have shown that bull trout and Dolly Varden hybridize in headwater reaches, and hybridization may negatively effect bull trout populations (McPhail & Taylor 1995; Baxter et al. 1997). Hybridization between bull trout and Dolly Varden has been documented where they occur in sympatry (McPhail & Taylor 1995; Baxter et al. 1997; Taylor et al. 2001). The directional nature of hybridization (male Dolly Varden mating with female bull trout; McPhail & Taylor 1995) may put bull trout populations at risk because of wasted reproductive effort and lower hybrid fitnesses (McPhail & Taylor 1995; Baxter et al. 1997; Hagen & Taylor 2001).
To conclude, we determine that the char samples examined in this study are bull trout, and it is unlikely that any hybrids are present in the migratory population. The data presented here also show that no migratory Dolly Varden were present in the lower Puyallup River over a 3-year period. Furthermore, ecological and molecular surveys should be considered to determine if resident Dolly Varden are present in the headwaters and to evaluate hybridization further. Taking into account the fact that the USFWS considers the lower Puyallup River subpopulation ‘depressed’, the positive identification of bull trout in this system is important to the managers who are given the task of this species’ conservation.
Acknowledgements
Funding for this project was provided by the United States Army Corps of Engineers. The authors would like to thank Dr Eric Taylor for providing the control samples and both Dr Taylor and Dr Paul Spruell for consultation and advice on diagnostic molecular markers.
References
This work is part of a continuing collaboration between the National Marine Fisheries Service, Puyallup Tribal Fisheries, and the United State Corps of Engineers that is focused on studying the distribution and biology of bull trout and Dolly Varden in the Puget Sound region. Jon Baker is a forensic geneticist interested in using population genetics to identify endangered salmonid species and populations. Paul Moran is a population geneticist whose current research involves understanding reproductive success and genetic interactions between wild and hatchery salmonid populations. Russ Ladley is a fisheries biologist for the Puyallup Tribal Fisheries and works on the restoration and management of salmonid populations.
