Discovery and characterization of novel genetic markers for use in the management of Lahontan cutthroat trout (Oncorhynchus clarkii henshawi)

The Lahontan cutthroat trout (Oncorhynchus clarkii henshawi) is threatened by habitat destruction, over‐harvest and hybridization with nonnative trout. Currently, three Geographic Management Units (GMUs) are recognized within the taxon. Here, we describe a suite of 68 single‐nucleotide polymorphism (SNP) genetic markers for use in the study and management of Lahontan cutthroat trout and a closely related subspecies, the Paiute cutthroat trout (O. c. seleneris). These include markers variable within the two subspecies (n = 35), diagnostic for the two subspecies (n = 23) and diagnostic for Yellowstone cutthroat trout (O. c. bouvieri) and other closely related subspecies (n = 10). Sixty‐three markers were discovered by Sanger sequencing of 171 EST loci in an ascertainment panel including Lahontan cutthroat trout from four populations representing all GMUs. Five markers were identified in a secondary sequencing effort with a single population of Lahontan cutthroat trout. TaqMan assays were validated on six Lahontan cutthroat trout populations and a diverse panel of other trout. Over 90% of the markers variable in Lahontan cutthroat trout were polymorphic in at least two populations, and 66% were variable within all three GMUs. All Lahontan diagnostic markers were also fixed for the Lahontan allele in Paiute cutthroat trout. Most of the Yellowstone diagnostic markers can also be used for this purpose in other cutthroat trout subspecies. This is the first set of SNP markers to be developed for Lahontan cutthroat trout, and will be an important tool for conservation and management.     

Characterization of tetranucleotide microsatellites for Rio Grande cutthroat trout and rainbow trout,and their cross‐amplification in other cutthroat trout subspecies

We describe the isolation and characterization of 12 tetranucleotide microsatellites for Rio Grande cutthroat trout (Oncorhynchus clarkii virginalis) and rainbow trout (Oncorhynchus mykiss), and subsequently investigate their performance in Colorado River cutthroat trout (Oncorhynchus clarkii pleuriticus), greenback cutthroat trout (Oncorhynchus clarkii stomias) and Yellowstone cutthroat trout (Oncorhynchus clarki bouvieri). All 12 loci are polymorphic in all subspecies of O. clarkii examined.     

Development and evaluation of 200 novel SNP assays for population genetic studies of westslope cutthroat trout and genetic identification of related taxa

DNA sequence data were collected and screened for single nucleotide polymorphisms (SNPs) in westslope cutthroat trout (Oncorhynchus clarki lewisi) and also for substitutions that could be used to genetically discriminate rainbow trout (O. mykiss) and cutthroat trout, as well as several cutthroat trout subspecies. In total, 260 expressed sequence tag‐derived loci were sequenced and allelic discrimination genotyping assays developed from 217 of the variable sites. Another 50 putative SNPs in westslope cutthroat trout were identified by restriction‐site‐associated DNA sequencing, and seven of these were developed into assays. Twelve O. mykiss SNP assays that were variable within westslope cutthroat trout and 12 previously published SNP assays were also included in downstream testing. A total of 241 assays were tested on six westslope cutthroat trout populations (N = 32 per population), as well as collections of four other cutthroat trout subspecies and a population of rainbow trout. All assays were evaluated for reliability and deviation from Hardy–Weinberg and linkage equilibria. Poorly performing and duplicate assays were removed from the data set, and the remaining 200 assays were used in tests of population differentiation. The remaining markers easily distinguished the various subspecies tested, as evidenced by mean G_ST of 0.74. A smaller subset of the markers (N = 86; average G_ST = 0.40) was useful for distinguishing the six populations of westslope cutthroat trout. This study increases by an order of magnitude the number of genetic markers available for the study of westslope cutthroat trout and closely related taxa and includes many markers in genes (developed from ESTs).     

Novel molecular markers differentiate Oncorhynchus mykiss (rainbow trout and steelhead) and the O. clarki (cutthroat trout) subspecies

A suite of 26 PCR‐based markers was developed that differentiates rainbow (Oncorhynchus mykiss) and coastal cutthroat trout (O. clarki clarki). The markers also differentiated rainbow from other cutthroat trout subspecies (O. clarki), and several of the markers differentiated between cutthroat trout subspecies. This system has numerous positive attributes, including: nonlethal sampling, high species‐specificity and products that are easily identified and scored using agarose gel electrophoresis. The methodology described for developing the markers can be applied to virtually any system in which numerous markers are desired for identifying or differentiating species or subspecies.     

Absence of developmental incompatibility in hybrids between rainbow trout and two subspecies of cutthroat trout

We examined the developmental rate of hybrids between rainbow trout (Salmo gairdneri) and two subspecies of cutthroat trout: westslope cutthroat trout (Salmo clarki lewisi) and Yellowstone cutthroat trout (Salmo clarki bouvieri). These taxa show considerable genetic divergence at 42 structural loci encoding enzymes; the mean Nei's d between the rainbow trout and the two species of cutthroat trout is 0.22. We used four measures of developmental rate: time of hatching and yolk resorption, rate of increase in activity of four enzymes, and time of initial detection of seven isozyme loci. The two cutthroat trout subspecies reached hatching and yolk resorption earlier than rainbow trout. Cutthroat trout had higher relative enzyme activities than rainbow trout from deposition of eye pigment to hatching. There was no difference in the rate of increase in enzyme activity or time of initial expression of these loci between these species. Hybrids showed developmental rates intermediate or similar to that of the parental species using all measures. Our results indicate an absence of regulatory and developmental incompatibility between these taxa.This research was supported by NSF Grants ISP-8011449 and BSR-8300039. M.M.F. was supported by a postgraduate scholarship from the Natural Sciences and Engineering Research Council of Canada.     

Y chromosome phylogeny for cutthroat trout (Oncorhynchus clarkii) subspecies is generally concordant with those of other markers

Sequence divergence was evaluated in the non-recombining, male-specific OmyY1 region of the Y chromosome among the subspecies of cutthroat trout (Oncorhynchus clarkii) in the western United States. This evaluation identified subspecies-discriminating OmyY1-haplotypes within a ～1200 bp region of the OmyY1 locus and localized the region to the end of the Y chromosome by FISH analysis. OmyY1 sequences were aligned and used to reconstruct a phylogeny of the cutthroat trout subspecies and related species via maximum-parsimony and Bayesian analyses. In the Y-haplotype phylogeny, clade distributions generally corresponded to the geographic distributions of the recognized subspecies. This phylogeny generally corresponded to a mitochondrial tree obtained for these subspecies in a previous study. Both support a clade of trout vs. Pacific salmon, of rainbow trout, and of a Yellowstone cutthroat group within the cutthroat trout. In our OmyY1 tree, however, the cutthroat “clade”, although present topologically, was not statistically significant. Some key differences were found between trees obtained from the paternally-inherited OmyY1 vs. maternally-inherited mitochondrial haplotypes in cutthroat trout compared to rainbow trout. Other findings are: The trout OmyY1 region evolves between 3 and 13 times slower than the trout mitochondrial regions that have been studied. The Lahontan cutthroat trout had a fixed OmyY1 sequence throughout ten separate populations, suggesting this subspecies underwent a severe population bottleneck prior to its current dispersal throughout the Great Basin during the pluvial phase of the last ice age. The Yellowstone group is the most derived among the cutthroat trout and consists of the Yellowstone, Bonneville, Colorado, Rio Grande and greenback subspecies. Identification of subspecies and sex with this Y-chromosome marker may prove useful in conservation efforts.     

Polymorphic microsatellite markers for the goosander (Mergus merganser)

To investigate the population genetic structure of the goosander (Mergus merganser) in Europe and to identify populations with a significant conservation value, we isolated nine microsatellite loci, and screened them in the subspecies Mergus merganser merganser and Mergus merganser americanus. All markers were polymorphic with two to 15 alleles per locus. Average observed and expected heterozygosity values were 0.422 and 0.624, respectively, for the European and 0.497 and 0.667, respectively, for the North American subspecies. Only one marker departed significantly from Hardy–Weinberg expectations in both subspecies. This marker was highly variable but homozygous in all females, suggesting a sex‐linked inheritance.     

Diagnostic single nucleotide polymorphisms for identifying westslope cutthroat trout (Oncorhynchus clarki lewisi), Yellowstone cutthroat trout (Oncorhynchus clarkii bouvieri) and rainbow trout (Oncorhynchus mykiss)

We describe 12 diagnostic single nucleotide polymorphism (SNP) assays for use in species identification among rainbow and cutthroat trout: five of these loci have alleles unique to rainbow trout (Oncorhynchus mykiss), three unique to westslope cutthroat trout (O. clarkii lewisi) and four unique to Yellowstone cutthroat trout (O. clarkii bouvieri). These diagnostic assays were identified using a total of 489 individuals from 26 populations and five fish hatchery strains.     

Identification and characterization of nine polymorphic microsatellite loci in the North American pika,Ochotona princeps

Nine polymorphic microsatellite loci were developed for the North American pika (Ochotona princeps) from di‐ and tetranucleotide repeat‐enriched genomic libraries. Polymorphism was assessed for 165 individuals from eight geographical locations in the western United States. All loci were polymorphic. The number of alleles per locus ranged from three to 14, with observed heterozygosity between 0.189 and 0.822. All loci were in Hardy–Weinberg equilibrium (< 0.05). Regional differences were evident with unique alleles at multiple loci in six of eight populations.     

Isolation and characterization of microsatellite loci in Tripterygion delaisi

We isolated 49 microsatellite loci from a genomic library of Tripterygion delaisi×anthosoma enriched for CA and GA repeats. Ten loci were screened in 30 individuals with high numbers of alleles per locus (averaging 15.5 ± 2.86) and observed heterozygosity (averaging 0.765 ± 0.052). No deviations from Hardy–Weinberg expectations were detected. These highly polymorphic markers will be useful in determining the spatial patterns of genetic diversity between and within subspecies of Tripterygion delaisi.     

Microsatellite analysis of genetic population structure in an endangered salmonid: the coastal cutthroat trout (Oncorhynchus clarki clarki)

The genetic population structure of coastal cutthroat trout ( Oncorhynchus clarki clarki ) in Washington state was investigated by analysis of variation in allele frequencies at six highly polymorphic microsatellite loci for 13 anadromous populations, along with one outgroup population from the Yellowstone subspecies ( O. clarki bouvieri) (mean heterozygosity = 67%; average number of alleles per locus = 24). Tests for genetic differentiation revealed highly significant differences in genotypic frequencies for pairwise comparisons between all populations within geographical regions and overall population subdivision was substantial ( F _ST = 0.121, R _ST = 0.093), with 44.6% and 55.4% of the among-population diversity being attributable to differences between streams ( F _SR = 0.054) and between regions ( F _RT = 0.067), respectively. Analysis of genetic distances and geographical distances did not support a simple model of isolation by distance for these populations. With one exception, neighbour-joining dendrograms from the Cavalli-Sforza and Edwards' chord distances and maximum likelihood algorithms clustered populations by physiogeographic region, although overall bootstrap support was relatively low (53%). Our results suggest that coastal cutthroat trout populations are ultimately structured genetically at the level of individual streams. It appears that the dynamic balance between gene flow and genetic drift in the subspecies favours a high degree of genetic differentiation and population subdivision with the simultaneous maintenance of high heterozygosity levels within local populations. Results are discussed in terms of coastal cutthroat trout ecology along with implications for the designation of evolutionarily significant units pursuant to the US Endangered Species Act of 1973 and analogous conservation units.     

MITOCHONDRIAL GENOTYPES HAVE NO DETECTABLE EFFECTS ON MERISTIC TRAITS IN CUTTHROAT TROUT HYBRID SWARMS

Efforts to detect effects of cytoplasmic genes are confounded by the problem of partitioning nuclear and cytoplasmic effects. In this study we test for effects of mtDNA haplotype on development in hybrid populations of cutthroat trout (Oncorhynchus clarki) with randomly associated nuclear and mtDNA genotypes. We have previously described several intraspecific hybrid swarms formed by interbreeding of westslope cutthroat trout (O. c. lewisi) and Yellowstone cutthroat trout (O. c. bouvieri). Genetic distance between these subspecies is high (Nei's D = 0.30; mtDNA P = 0.02), and diagnostic alleles at multiple nuclear loci and two distinct mtDNA haplotypes are present in the hybrids. Historical associations between alleles at nuclear loci and between cytotypes and nuclear alleles have largely decayed. We test for differences in meristic characters between fish with alternate mtDNA haplotypes. Counts and fluctuating bilateral asymmetry for these characters have been previously shown to be sensitive indicators of genetic differences that affect development. No differences were found between mtDNA types in meristic counts or fluctuating asymmetry. Therefore, the alternate mtDNA haplotypes had no detectable effect on development as measured by meristic counts in these hybrid populations. However, diagnostic alleles at one nuclear allozyme locus (CK-CI) were associated with several fin ray counts.     

Isolation and characterization of polymorphic microsatellite loci for the study of paternity and population structure in the little penguin Eudyptula minor

We isolated and characterized eight novel microsatellite loci in the little penguin Eudyptula minor, using nonradioactive polymerase chain reaction‐based techniques to screen GA and GAAA repeats from enriched genomic DNA libraries. All eight loci were polymorphic and seven were variable in our main study population (mean H_E = 0.613, mean N_A = 7.14). Cross‐amplification using a microsatellite primer developed in Spheniscus demersus (African penguin) yielded one additional polymorphic locus. This locus combined with six of the little penguin loci is suitable for paternity assignment in little penguins (exclusion probability for seven unlinked loci = 0.993).     

Isolation and cross‐familial amplification of 41 microsatellites for the brook charr (Salvelinus fontinalis)

The brook charr (Salvelinus fontinalis; Osteichthyes: Salmonidae) is a phenotypically diverse fish species inhabiting much of North America. But relatively few genetic diagnostic resources are available for this fish species. We isolated 41 microsatellites from S. fontinalis polymorphic in one or more species of salmonid fish. Thirty‐seven were polymorphic in brook charr, 15 in the congener Arctic charr (Salvelinus alpinus) and 14 in the lake charr (Salvelinus namaycush). Polymorphism was also relatively high in Oncorhynchus, where 21 loci were polymorphic in rainbow trout (Oncorhynchus mykiss) and 16 in cutthroat trout (Oncorhynchus clarkii) but only seven and four microsatellite loci were polymorphic in the more distantly related lake whitefish (Coregonus clupeaformis) and Atlantic salmon (Salmo salar), respectively. One duplicated locus (Sfo228Lav) was polymorphic at both duplicates in S. fontinalis.     

Characterization of 13 microsatellites for Lahontan cutthroat trout (Oncorhynchus clarki henshawi) and cross-amplification in six other salmonids

Thirteen newly developed tri- and tetranucleotide repeat microsatellite markers were developed for Lahontan cutthroat trout (Oncorhynchus clarki henshawi), a threatened subspecies endemic to the Lahontan hydrographic basin in the western USA. These loci are highly polymorphic with five to 30 alleles per locus and observed heterozygosities ranging from 0.4 to 0.7. Cross-species amplification of these markers was most successful in the closely related rainbow trout, Oncorhynchus mykiss, with only three loci amplifying in brown trout, Salmo trutta. Nonoverlapping allelic distributions for many of these loci among the six salmonid species screened suggest these markers may be useful for hybrid determination.     

Assessing thermal adaptation using family‐based association and FST outlier tests in a threatened trout species

Discovering genetic markers associated with phenotypic or ecological characteristics can improve our understanding of adaptation and guide conservation of key evolutionary traits. The Lahontan cutthroat trout (Oncorhynchus clarkii henshawi) of the northern Great Basin Desert, USA, demonstrated exceptional tolerance to high temperatures in the desert lakes where it resided historically. This trait is central to a conservation hatchery effort to protect the genetic legacy of the nearly extinct lake ecotype. We genotyped full‐sibling families from this conservation broodstock and samples from the only two remaining, thermally distinct, native lake populations at 4,644 new single nucleotide polymorphisms (SNPs). Family‐based genome‐wide association testing of the broodstock identified nine and 26 SNPs associated with thermal tolerance (p < 0.05 and p < 0.1), measured in a previous thermal challenge experiment. Genes near the associated SNPs had complex functions related to immunity, growth, metabolism and ion homeostasis. Principal component analysis using the thermotolerance‐related SNPs showed unexpected divergence between the conservation broodstock and the native lake populations at these loci. F_ST outlier tests on the native lake populations identified 18 loci shared between two or more of the tests, with two SNPs identified by all three tests (p < 0.01); none overlapped with loci identified by association testing in the broodstock. A recent history of isolation and the complex genetic and demographic backgrounds of Lahontan cutthroat trout probably limited our ability to find shared thermal tolerance loci. Our study extends the still relatively rare application of genomic tools testing for markers associated with important phenotypic or environmental characteristics in species of conservation concern.     

ASSOCIATIONS BETWEEN MITOCHONDRIAL AND NUCLEAR GENOTYPES IN CUTTHROAT TROUT HYBRID SWARMS

We examined mtDNA and nuclear allozyme genotypes in hybrid populations formed from interbreeding of westslope cutthroat trout (Oncorhynchus clarki lewisi) and Yellowstone cutthroat trout (O. c. bouvieri). These subspecies show substantial genetic divergence (Nei's D = 0.30; mtDNA P = 0.02). Diagnostic alleles at multiple nuclear loci and two distinct mtDNA haplotypes segregate in the hybrids. Nuclear and mtDNA genotypes are largely randomly associated, although there is slight disequilibrium in both nuclear and cytonuclear measures in some samples. Consistent positive gametic disequilibria for three pairs of nuclear loci confirm one previously reported linkage, and indicate two more. Allele frequencies provide no evidence for selection on individual chromosome segments. However, westslope mtDNA haplotype frequencies exceed westslope nuclear allele frequencies in all samples. This may be explained by differences in the frequency of occurrence of reciprocal F₁ matings, by viability, fertility, or sex ratio differences in the progeny of reciprocal matings, or by weak selection on mtDNA haplotypes.     

Characterization of microsatellite loci in polyploid Lavatera assurgentiflora ssp. assurgentiflora and ssp. glabra (Malvaceae)

Lavatera assurgentiflora (Malvaceae) is one of four species of the genus Lavatera native to California and Baja California. Two geographically defined subspecies are recognized: L. a. assurgentiflora on the northern islands and L. a. glabra on the southern islands. We isolated nine polymorphic microsatellite loci that amplify in both subspecies of L. assurgentiflora. Substantial levels of polymorphism were observed at many of the loci. Four loci exhibited more than 10 alleles, polymorphism information content ranged from 0.4 to 0.8, and up to six alleles were found in some individuals, supporting reports that these taxa are hexaploid. All loci also amplified in Lavatera lindsayi from Guadalupe Island, and we anticipate that they will cross‐amplify in other California Lavatera species as well.     

Summer and fall microhabitat utilization of juvenile bull trout and cutthroat trout in a wilderness stream, Idaho

Microhabitat use and availability were evaluated and compared between different size classes of juvenile resident bull trout (Salvelinus confluentus) and cutthroat trout (Oncorhynchus clarki) in a small wilderness stream within the South Fork Clearwater River basin, Idaho. The objective was to determine if utilization of measured habitat characteristics changed from summer to late fall. Sampling of fish was conducted with night snorkeling. During the summer, smaller juvenile bull trout (<66 mm) total length (TL) were associated with shallow stream margins over coarse substrates. In the fall, they moved to significantly deeper, lower velocity water, and closer to cover (p<0.05), but maintained their association with coarse substrates. During the summer, larger juvenile bull trout and larger juvenile cutthroat trout (66–130 mm TL) occupied significantly deeper water than smaller juvenile bull trout (p<0.05). Generally, larger juvenile bull trout were found closer to the bottom and in lower velocity water than larger juvenile cutthroat trout (p<0.05). In the fall, larger juvenile bull trout and larger juvenile cutthroat trout were associated with significantly deeper, lower velocity water located closer to cover than in summer (p<0.05). However, cutthroat trout occupied slightly deeper water over finer substrates than bull trout. Deep water with low velocities evidently provide important rearing areas for large bull trout and large cutthroat trout in the fall. Land management practices that maintain such environments will benefit these species.     

Bi‐parentally inherited species‐specific markers identify hybridization between rainbow trout and cutthroat trout subspecies

Eight polymerase chain reaction primer sets amplifying bi‐parentally inherited species‐specific markers were developed that differentiate between rainbow trout (Oncorhynchus mykiss) and various cutthroat trout (O. clarki) subspecies. The primers were tested within known F₁ and first generation hybrid backcrosses and were shown to amplify codominantly within hybrids. Heterozygous individuals also amplified a slower migrating band that was a heteroduplex, caused by the annealing of polymerase chain reaction products from both species. These primer sets have numerous advantages for native cutthroat trout conservation including statistical genetic analyses of known crosses and simple hybrid identification.