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


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

N . R . C A M P B E L L , * S . J . A M I S H , † V . L . P R I T C H A R D , ‡ K . M . M C K E L V E Y , § M . K . Y O U N G , §

M . K . S C H W A R T Z , § J . C . G A R Z A , ‡ G . L U I K A R T †, ¶ a n d S . R . N A R U M *

*Columbia River Inter-Tribal Fish Commission, 3059 F National Fish Hatchery Rd, Hagerman, ID 83332, USA, †Division of

Biological Sciences, Fish and Wildlife Genomics Group, University of Montana, Missoula, MT 59812, USA, ‡Southwest Fisheries

Science Center and the University of California, Santa Cruz, 110 Shaffer Rd, Santa Cruz, CA 95060, USA, §USDA Forest Service,

RMRS, Forestry Sciences Laboratory, 800 E. Beckwith Avenue, Missoula, MT 59801, USA, ¶Flathead Lake Biological Station,

University of Montana, Polson, MT 59860, USA

Abstract

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 GST of 0.74. A smaller subset of the markers (N = 86; average GST = 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).

Keywords: admixture, hybridization, KASPar, Oncorhynchus clarki, population genomics, rainbow trout, SNP, Taqman,

westslope cutthroat trout

Received 27 January 2012; revision received 9 April 2012; accepted 22 April 2012

Introduction

Cutthroat trout (Oncorhynchus clarki) are native to cold

water environments in western North America from

Alaska to New Mexico, USA. In the late Miocene and

early Pliocene, cutthroat trout spread throughout por-

tions of western North America via paleohydrological

connections influenced by orogeny and climate (Smith

et al. 2002). Subsequent isolation in major river basins led

to the evolution of up to 14 subspecies that are geographi-

cally, phenotypically and genetically distinct (Behnke

2002). The coastal subspecies (O. c. clarki) inhabits lakes

and streams west of the Cascade crest and includes both

resident and anadromous populations (Behnke 1992).

The remaining three major lineages (westlope, Yellow-

stone group and Lahontan group) are found in streams

and lakes of the Rocky Mountains on both sides of the

continental divide (Behnke 1992; Wilson & Turner 2009).

All cutthroat trout are sensitive to changes in their

environment, and human impact on water temperature

and quality has caused the decline of many populations

(Shepard et al. 2005). Introductions of non-native species

to their habitat have also supplanted native cutthroat

trout populations by way of competition for resources,

predation and interbreeding (Griffith 1988; Young 1995;

Ruzycki et al. 2003; Muhlfeld et al. 2009). These effects
Correspondence: Nathan R. Campbell, Fax: 208-837-6047;

E-mail: camn@critfc.org

� 2012 Blackwell Publishing Ltd

Molecular Ecology Resources (2012) doi: 10.1111/j.1755-0998.2012.03161.x



have led to the extinction of two cutthroat subspecies

(O. c. alvordensis and O. c. macdonaldi) and another three

subspecies, Lahontan (O. c. henshawi), Paiute (O. c. selen-

iris) and Greenback (O. c. stomias) cutthroat trout are

listed as threatened under the US Endangered Species

Act (ESA). Moreover, declining populations of wests-

lope (O. c. lewisi), Yellowstone (O. c. bouvieri), Bonneville

(O. c. utah), Colorado (O. c. pleuriticus) and Rio Grande

(O. c. virginalis) cutthroat trout subspecies are of

conservation concern although not currently protected

under the ESA.

The most widely distributed of the cutthroat trout

subspecies is the westslope cutthroat trout (Behnke 2002).

Historically, this subspecies ranged from the Rocky

Mountains of southern Canada in the north to central

Idaho in the south (both east and west of the continental

divide) with isolated populations in the eastern Cascade

Range of British Columbia, Washington and Oregon

(Behnke 1992). However, recent estimates have revealed

that genetically pure westslope cutthroat trout now

inhabit only 25% of their native range (COSEWIC 2006;

May 2009). Despite these dismal statistics, many popula-

tions of this subspecies still persist across a wide geo-

graphical area suggesting that a great deal of genetic

diversity still exists. Genetic tools to better understand

the genetic relationships between westslope cutthroat

trout populations and their habitats could be an impor-

tant tool in their conservation and are the main focus of

this study.

Genetic analyses within cutthroat trout subspecies

using microsatellite markers have revealed that popula-

tions are often highly genetically distinct from one

another, even within the same drainage (e.g. Taylor et al.

2003; Wofford et al. 2005; Drinan et al. 2011). These sub-

stantial genetic differences are attributable to physical

barriers to fish passage that reduce gene flow, and rela-

tively small breeding populations with high rates of

genetic drift (Waples et al. 2001; Fausch et al. 2009; Young

2011). Human impacts on fish passage and water quality

have likely led directly to the isolation of wild cutthroat

to headwater streams (Fausch et al. 2009) and, in many

streams, introduction of rainbow trout and subsequent

hybridization has further isolated pure cutthroat trout to

small tributary streams.

Rainbow trout and cutthroat trout are sister species,

and hybridization between them is common when they

share habitat. Hybridization with introduced rainbow

trout has been described as the primary threat to the sur-

vival of many cutthroat trout subspecies because of

diminished fitness and reproductive success of hybrids

and their offspring (Leary et al. 1985; Allendorf & Leary

1988; Muhlfeld et al. 2009). Mechanisms for reduced fit-

ness may include genetic incompatibilities such as chro-

mosomal rearrangements, possible differences in the

numbers of chromosomes (rainbow trout: N = 58–64; cut-

throat trout: N = 64–68) and the disruption of co-adapted

gene complexes, which contribute to outbreeding depres-

sion (Leary et al. 1985). Despite these potentially deleteri-

ous effects, rainbow trout DNA can spread rapidly

through a cutthroat trout population because of the rela-

tively high survival and reproductive success of first gen-

eration (F1) hybrids (Muhlfeld et al. 2009). Owing to the

difficulty in the phenotypic identification of hybrids,

molecular and genetic markers such as allozymes,

AFLPs and microsatellites have been used to study

hybridization (Allendorf & Leary 1988; Hitt et al. 2003;

Pritchard et al. 2007). More recently, however, diagnostic

single nucleotide polymorphisms (SNP) loci have

become available for detection of invasive rainbow trout

DNA (Kalinowski et al. 2010; Hohenlohe et al. 2011;

Pritchard et al. 2012).

In this study, we identified variation in cutthroat trout

DNA sequence data and developed assays for SNPs in

westslope cutthroat trout and for substitutions that dis-

tinguish westslope cutthroat trout from other cutthroat

trout subspecies and from rainbow trout. Variable sites

were identified by sequence alignment and assays were

developed for the most polymorphic loci, focusing on

SNPs that were variable within westslope cutthroat trout

or substitutions diagnostic between taxa. Another seven

SNP loci were developed based on restriction-site-associ-

ated DNA (RAD) sequence data generated using rainbow

and cutthroat trout samples (Hohenlohe et al. 2011). All

assays were evaluated by genotyping a set of cutthroat

trout samples from 11 populations and a single rainbow

trout population. Assays with poor plot quality (i.e. geno-

type resolution) and significant deviations from Hardy–

Weinberg proportions were removed from the data set.

The remaining assays were evaluated for their ability to

distinguish populations of westslope cutthroat trout and

identify cutthroat trout subspecies and hybrids with rain-

bow trout.

Methods

SNP discovery: ascertainment panel 1

Salmonid DNA sequences, mostly from mRNA tran-

scripts, were collected from online database entries in

TIGR and NCBI. Sequences containing 300+ bases of

3-prime un-translated region (UTR) were chosen

preferentially as these tend to contain more variation

than coding sequence. Primers were designed using

either PRIMER3 (http://frodo.wi.mit.edu/primer3/) or

the embedded primer design tool in NCBI’s database to

target gene regions 400–800 bp in length. All primer pairs

(N = 163) were initially tested for amplification of a

single product using DNA from two westslope cutthroat

� 2012 Blackwell Publishing Ltd

2 N . R . C A M P B E L L E T A L .



trout. PCR conditions were 1· multiplex master mix
(Qiagen, Valencia, CA, USA), 250 nM primers and 1–2 lL
of extract DNA in a 10 lL reaction. Thermal cycling con-
ditions were [95 �C for 15 min (95 �C for 30 s, 60 �C for
30 s, 72 �C for 45 s, repeated 40 cycles) and 4 �C hold].
Amplified products were size separated on a 1% agarose

gel and stained using SYBR-safe reagent (Life Technolo-

gies, Grand Island, NY, USA). Primer pairs that produced

clean PCR products (N = 134) were used to screen an

ascertainment panel of 63 individuals (Table S3).

Westslope cutthroat trout were the best represented

subspecies in the ascertainment panel, and samples were

chosen from 48 collections from 27 different drainages in

Idaho, Montana and British Columbia. Another two sam-

ples each of Yellowstone, Bonneville and coastal cut-

throat trout were included in the ascertainment panel as

well as one doubled haploid rainbow trout (Table S3).

Prior to PCR, all DNA extracts were quantified and

individuals from the same drainage ⁄ collection were
duplexed (two individual extracts combined into one

aliquot) to increase the number of screened genomes

while keeping sequencing costs low.

Ascertainment samples were then PCR amplified

using multiplexed primers such that each reaction pro-

duced four products and with the same PCR and cycling

conditions as in the initial amplification test. PCR prod-

ucts were cleaned by treatment with Exonuclease 1 and

shrimp alkaline phosphatase (New England Biolabs,

Ipswich, MA, USA) to remove unincorporated dNTPs

and primers. Sequencing reactions were performed in

one direction using the Big Dye v.1.1 sequencing kit (Life

Technologies) and with 2 lL of each product (as the
sequencing primers are different for each amplicon, there

is no need to separate them prior to sequencing). The

reaction products were purified by ethanol-EDTA-NaAc-

etate precipitation prior to sequencing by capillary elec-

trophoresis using a 3730 automated DNA Analyzer

instrument (Life Technologies). Chromatogram data

were analysed with Sequencing Analysis v.5.4 (Life Tech-

nologies), and Sequencher v.4.7 (Gene Codes Corpora-

tion, Ann Arbor, MI, USA) was used to align and edit the

data. Observed SNPs and small indels were recorded

and minor allele frequencies calculated for each.

Sequences containing sites chosen for assay design were

submitted to dbSNP (Tables S1 and S2).

SNP discovery: ascertainment panel 2

A second Sanger sequencing SNP discovery effort using

a distinct ascertainment panel (Table S3) focused on vari-

ation within westslope cutthroat trout and between cut-

throat trout subspecies. An expanded version of this

same panel was used previously for identification of

sites diagnostic between rainbow and cutthroat trout

(Pritchard et al. 2012). Initial PCR was performed using

242 primer pairs, designed from randomly chosen

nuclear tentative consensus sequences generated from

O. mykiss expressed sequence tags reported to the

Harvard ⁄ DFCI Gene Index project (http://compbio.dfci.
harvard.edu/tgi/tgipage.html). PCR, subsequent visuali-

zation, product purification and sequencing were as

described in Pritchard et al. 2012. Variable sites were

identified by visual examination of sequence data aligned

using Sequencher v4.10.1 (Gene Codes Corporation, Ann

Arbor, MI, USA). Sequences containing sites chosen for

assay design were submitted to dbSNP (Tables S1 and

S2). It is noted there is no overlap between the Pritchard

et al. assays and the assays reported here as that study

focuses strictly on diagnostic markers between O. mykiss

and O. clarki. Assays reported here are either variable

within westslope cutthroat trout or diagnostic between

cutthroat subspecies.

SNP discovery: ascertainment panel 3

Restriction-site-associated DNA sequence data generated

on an Illumina GAII instrument for identification of diag-

nostic SNPs between rainbow and cutthroat trout

(Hohenlohe et al. 2011) had also identified SNPs within

westslope cutthroat trout. RAD-tagged sequences from

11 westslope cutthroat trout samples included fish from

wild populations from both sides of the continental

divide in Montana (Table S3). Strict filtering based on

observed heterozygosity and deviations from Hardy–

Weinberg (HW) proportions was applied to remove

homologs (described in Hohenlohe et al. 2011). Fifty of

the loci with highest minor allele frequency that were

sequenced in at least ten of the samples were identified

for assay design. A search with these 48-bp sequences

against a draft of the O. mykiss genome (M. Miller per-

sonal communication – miller.michael.ryan@gmail.com)

was used to generate enough flanking sequence for

designing assays.

Assay design and genotyping

Suitable SNP and diagnostic sites were used to design

either TaqmanTM (Life Technologies) or KASPar

(KBioscience, Herts, England, UK) genotyping assays,

which use different chemistry, by submitting sequence

data to each company. Both types of assay were designed

for some of the SNP sites, but only one for most sites.

These assays, along with another eight previously pub-

lished diagnostic assays (Harwood & Phillips 2011), were

used to genotype 12 collections of approximately 30 fish

each representing each of the following species ⁄ subspe-
cies: rainbow trout (one collection), westslope (six

collections), Yellowstone (two collections), Bonneville

� 2012 Blackwell Publishing Ltd

S N P D E T E C T I O N A S S A Y S F O R C U T T H R O A T T R O U T 3



(one collection), coastal (one collection) and Lahontan

cutthroat trout (one collection). All genotyping was per-

formed using the EP1 system with 96.96 Dynamic Geno-

typing Arrays (Fluidigm). TaqmanTM assay genotyping

was modified from the manufacturers’ suggested proto-

cols to include a sample preamplification step and an

increase to 50 cycles of PCR. The sample preamplification

used 1· Qiagen multiplex master mix, 50 nM pooled
primers and 2 lL of extract DNA in a 7 lL reaction,
[thermal cycler conditions: 95 �C–15 min (95 �C–15 s,
60 �C–4 min, 14 cycles) and 4 �C–hold]. KASPar assay
genotyping used reagents and thermal cycling conditions

as recommended by the manufacturer and also included

a preamplification step.

Data analysis

The program GENEPOP (Raymond & Rousset 1995) was

used to evaluate deviations from HW Proportions and

test for statistically significant linkage disequilibrium

(LD), after corrections for multiple tests with B-Y false

discovery rate (Narum 2006). Assays that produced poor

quality genotyping plots or significant deviations from

HW proportions in multiple collections were removed

from further testing. The program GENETIX (Belkhir

et al. 2004) was used to calculate GST and evaluate genetic

differentiation by factorial correspondence analysis

(FCA). Individual assignment tests were performed

using the program GENECLASS2 (Piry et al. 2004) to

assess the ability of this set of assays to assign individuals

back to their collection of origin. Program settings used

a Bayesian method (Rannala & Mountain 1997) for

individual assignment and a score threshold of 0.05.

Results

Of the 335 primer pairs producing clean PCR fragments,

260 also generated adequate sequence data for SNP

detection. Another 98 190 RAD-tagged DNA fragments

(Hohenlohe et al. 2011) were examined for identification

of SNPs within westslope cutthroat trout. There were 246

sites chosen for assay design with either variation among

westslope cutthroat trout samples or that were putatively

diagnostic between westslope cutthroat trout and either

rainbow trout or other cutthroat trout subspecies. Of

these, a total of 220 were successfully converted into

genotyping assays (TaqmanTM = 126; KASPar = 94;

both = 22). Assay primer ⁄ probe sequences are listed in
Supplemental Tables S1 and S2.

The designed assays were then evaluated by genotyp-

ing samples from 11 cutthroat trout collections and a sin-

gle rainbow trout collection (Table 1). These samples

were also genotyped using 12 rainbow trout assays and

12 previously published diagnostic assays (Supplemental

Tables S1 and S2). Assays with poor quality genotyping

plots, low genotyping success and duplicated loci were

removed from further analysis (N = 35). Mitochondrial

SNPs were collapsed into haplotypes, and the 10 unique

haplotypes observed were coded into single genotypes.

Significant deviations from HWE within multiple popu-

lations were observed in eight SNP loci. Two of these

were sex-linked SNPs and were retained, because the

observed heterozygote deficits were expected, and the

other six were removed from further analysis.

Statistically significant LD was detected in 190 of

35 532 pairwise tests for one or more of the six westslope

cutthroat trout collections (P-value adjusted for multiple

tests <0.004). Seven of these locus pairs, involving 11 total

SNPs, were in significant LD in at least two of the collec-

tions (Supplemental Table S4). Assays were excluded

based on significant LD only for FCA and assignment

testing (exclusion of four assays resolved all seven signif-

icant pairs). Only westslope cutthroat trout were used for

this analysis because of the lack of variation within the

other subspecies collections. Several pairs of loci known

to be from the same sequence were not in significant LD

Table 1 Assignment test among collections of cutthroat trout

subspecies and rainbow trout using GENECLASS2

Collection Drainage N

% Correct

assignment

Ave.

Confidence

(%)

Wilson Cr.

(O. mykiss)

Salmon R. 27 100.0 100.0

Dog Cr.

(O. c. clarki)

Hood R. 22 100.0 100.0

Glenwood Hat.

(O. c. utah)

Hatchery stock 31 100.0 100.0

L. Lenore

(O. c. heshawi)

Hatchery stock 31 100.0 100.0

Barnes Cr.

(O. c. bouvieri)

South Fork

Snake R.

31 100.0 100.0

Gibson Cr.

(O. c. bouvieri)

Portneuf R. 24 100.0 99.9

Trout Cr.

(O. c. lewisi)

Kootenai R. 30 100.0 100.0

McGuire Cr.

(O. c. lewisi)

Kootenai R. 29 100.0 100.0

Schwartz Cr.

(O. c. lewisi)

Upper Clark R. 24 100.0 100.0

McCabe Cr.

(O. c. lewisi)

Blackfoot R. 30 100.0 100.0

Flat Cr.

(O. c. lewisi)

Lower Clark R. 30 100.0 100.0

Cache Cr.

(O. c. lewisi)

Salmon R. 30 100.0 100.0

N, number of individuals; Ave. confidence, Average of the likeli-

hood score for correct assignment calculated for each individual

within a population by GENECLASS2.

� 2012 Blackwell Publishing Ltd

4 N . R . C A M P B E L L E T A L .



because of the diagnostic nature (near fixation between

taxa) of one or both of the sites. Likewise, lack of hetero-

zygosity within the westslope cutthroat trout collections

may have masked LD for many other pairs of loci.

The remaining genotypes were used to evaluate the

utility of these assays to differentiate populations of

westslope cutthroat trout as well as distinguish cutthroat

trout subspecies from one another and from rainbow

trout. Analysis of the GST values for each locus reveals

that more than half of the assays are informative only for

differentiation between the tested species ⁄ subspecies
(Fig. 1 and Supplemental Tables S1 and S2). This leaves

87 markers with significant differentiation between the

six included westslope cutthroat trout collections (GST >

0.10), with the average GST of 0.41, suggesting a high

degree of differentiation between these six westslope cut-

throat trout collections. Average heterozygosity within

these collections for that subset of SNPs ranged from

0.096 to 0.333 indicating that heterozygosity within indi-

vidual westslope cutthroat trout populations is quite var-

iable. The population with the lowest observed

heterozygosity was McGuire Cr. in the upper Kootenai

River drainage. The Trout Cr. population from the lower

Kootenai River, however, showed much higher heterozy-

gosity (HOBS = 0.27) implying isolation of the McGuire

Cr. population from the lower Kootenai River gene pool.

The genotyped collections separated extremely well

using this set of markers as evidenced by the FCA plots

(Fig. 2) and assignment tests (Table 1). The large number

of markers that were informative for differentiation

between both the different cutthroat subspecies and

between cutthroat ⁄ rainbow trout allowed the major lin-
eages to be easily distinguished from one another

(Fig. 2a). Five fish from the coastal cutthroat collection

were identified as rainbow trout, and four fish were iden-

tified as rainbow ⁄ cutthroat trout hybrids by their inter-
mediate positions on the FCA graph. The misidentified

and hybrid fish were then removed from the data set for

FCA and assignment testing of westslope cutthroat trout

collections.

Separation of westslope cutthroat trout collections on

the FCA plot was less defined (Fig. 2b), as it relied on

variation in a much smaller subset of SNPs in more

genetically similar groups. The first two axes of the

westslope FCA, however, only explained 26.1% of the

total variation observed with these markers. Subsequent

assignment testing showed that every individual fish

could be correctly assigned back to their collection of ori-

gin using only genetic data from these markers (Table 1).

Moreover, the confidence of the assignments was high

and no misassignments were observed even between

populations from the same lineage, including the two

populations of Yellowstone cutthroat trout for which dif-

ferences in allele frequency of 20% or more occur in only

11 of the markers. Supplemental Tables S1 and S2 contain

minor allele frequencies and population differentiation

information that describes the utility of each marker.

Discussion

This study used multiple methods for SNP discovery that

were influenced by different ascertainment biases. For

example, one of the Sanger sequencing efforts used an

ascertainment panel directed at finding SNPs within the

westslope cutthroat trout subspecies while the other

included roughly equal representation of all extant cut-

throat trout subspecies. As each panel contained individ-

uals from several subspecies, it was possible to obtain

both putatively diagnostic markers and SNPs that were

variable within westslope cutthroat trout from both

efforts. However, there was significant ascertainment

bias for variation in a small number of westslope cut-

throat trout populations when the second panel was

used, because of the limited number and origin of wests-

lope cutthroat trout samples included in discovery

(Pritchard et al. 2012). A total of 36 SNP sites variable

within westslope cutthroat trout were chosen from this

sequencing panel. Likewise, limited representation of the

cutthroat trout subspecies in all three panels made the

identification of truly diagnostic markers from sequence

Fig. 1 Graph of GST for each of the 190

single nucleotide polymorphism loci.

Grey bars indicate GST when all collections

are included and black dots indicate GST
within only the westslope cutthroat trout

populations.

� 2012 Blackwell Publishing Ltd

S N P D E T E C T I O N A S S A Y S F O R C U T T H R O A T T R O U T 5



data unreliable. Of the 86 sites chosen from sequence

data as putatively diagnostic, only 17 had fixed geno-

types in all subsequently genotyped taxa. However,

although not strictly diagnostic, many of these markers

were nearly so, given the low frequency of alternate

alleles, and were still powerful for population identifica-

tion. The occurrence of alternate diagnostic alleles may

also be due to limited introgression within some of the

tested populations. Moreover, as only a single population

of rainbow trout was genotyped, the representation of

O. mykiss diversity was far from exhaustive.

Previous studies using microsatellite loci have shown

marked genetic distinctions among westslope cutthroat

trout populations (Young et al. 2004; Drinan et al. 2011).

These genetic differences, often even within the same

watershed, have led to recommendations for stream-level

management (Young et al. 2004). Although population

coverage is limited, our results also reveal strong genetic

differentiation among westslope cutthroat trout popula-

tions. We also observed low heterozygosity within some

populations, suggesting a risk of inbreeding depression,

but it is unclear the extent to which this is owing to ascer-

tainment bias. The average heterozygosity of westslope

cutthroat trout informative SNPs (HOBS = 0.209) is some-

what lower than that observed in similar SNP panels for

populations of anadromous, less genetically isolated

sockeye and coho salmon (HOBS = 0.267 and 0.260;

Campbell & Narum 2011). This is likely a result of limited

migration and gene flow in relatively isolated popula-

tions of westslope cutthroat trout.

The development of informative genetic markers for

the study of nonmodel organisms is the most costly

(a)

(b)

Fig. 2 Factorial correspondence analysis plots generated using the program GENETIX (190 markers). (a) Individuals from all

species ⁄ subspecies collections. F1 rainbow ⁄ cutthroat trout hybrids are indicated in the red ellipse. (b) Individuals in the six westslope
cutthroat trout populations.

� 2012 Blackwell Publishing Ltd

6 N . R . C A M P B E L L E T A L .



segment of most population genetics studies. Once a set

of markers has been developed and validated, the costs

associated with population genetic analysis decreases

significantly. Here, we describe a set of 216 SNP and

diagnostic marker genotyping assays and their relative

utility in differentiation of populations of westslope

cutthroat trout and diagnosis for several cutthroat trout

subspecies and rainbow trout. The SNP assays evaluated

in this study will be useful for a broad range of research

questions, including westslope cutthroat trout popula-

tion structure, detection of other cutthroat trout subspe-

cies and identification of introgression and hybrids

between related taxa.

Acknowledgements

We would like to thank Matt Campbell, Denise Hawkins,

Jessica Metcalf, Kirk Patten, Mary Peacock, Dennis Shiozawa,

Christian Smith, Chris Rossel, Robb Leary and Montana Fish

Wildlife and parks for supplying cutthroat trout samples for

genotyping. Thanks also to Sally Painter and Megan Moore

for help with DNA extraction and genotyping. Funding for

this study was provided by the Bonneville Power Administra-

tion, the Rocky Mountain Research Station, the US Fish and

Wildlife Service (Lahontan National Fish Hatchery and

Abernathy Fish Technology Center) and the National Oceanic

and Atmospheric Administration (Southwest Fisheries Science

Center).

References

Abadı́a-Cardoso A, Clemento AC, Garza JC (2011) Discovery and charac-

terization of single nucleotide polymorphisms in steelhead ⁄ rainbow
trout, Oncorhynchus mykiss. Molecular Ecology Resources, 11(Suppl. 1),

31–49.

Allendorf FW, Leary RF (1988) Conservation and distribution of genetic

variation in a polytypic species, the cutthroat trout. Conservation Biol-

ogy, 2, 170–184.

Behnke RJ (1992) Native Trout of Western North America. American Fisher-

ies Society, Bethesda, MD.

Behnke RJ (2002) Trout and Salmon of North America. The Free Press, New

York, NY.

Belkhir K, Borsa P, Chikhi L, Raufaste M, Bonhomme F (2004) GENETIX

4.05, logiciel sous WindowsTM pour la Génétique des Populations. Montpel-

lier, France, Laboratoire Génome, Populations, Interactions, CNRS

UMR 5171, Université de Montpellier II.

Brunelli JP, Thorgaard GH (2008) Single-Nucleotide polymorphisms

associated with allozyme differences between inland and coastal

rainbow trout. Transactions of the American Fisheries Society, 137, 1292–

1298.

Campbell NR, Narum SR (2011) Development of 54 novel single-

nucleotide polymorphism (SNP) assays for sockeye and coho

salmon and assessment of available SNPs to differentiate stocks

within the Columbia River. Molecular Ecology Resources, 11(Suppl. 1),

20–30.

Castaño-Sanchez C, Smith TPL, Wiedmann RT et al. (2009) Single nucleo-

tide polymorphism discovery in rainbow trout by deep sequencing of

a reduced representation library. BMC Genomics, 10, 559–567.

COSEWIC (2006) COSEWIC Assessment and Update Status Report on the

Westslope Cutthroat trout Oncorhynchus Clarkii lewisi (British Columbia

population and Alberta population) in Canada. Committee on the Status of

Endangered Wildlife in Canada, Ottawa.

Drinan DP, Kalinowski ST, Vu NV, Shepard BB, Muhlfeld CC, Campbell

MR (2011) Genetic variation in westslope cutthroat trout Oncorhynchus

clarkii lewisi: implications for conservation. Conservation Genetics, 12,

1513–1523.

Fausch KD, Rieman BE, Dunham JB, Young MK, Peterson DP (2009) Inva-

sion versus isolation: trade-offs in managing native salmonids with

barriers to upstream movement. Conservation Biology, 23, 859–870.

Finger AJ, Stephens MR, Clipperton NW, May B (2009) Six diagnostic sin-

gle nucleotide polymorphism markers for detecting introgression

between cutthroat and rainbow trouts. Molecular Ecology Resources, 9,

759–763.

Griffith JS (1988) Review of competition between cutthroat trout and other

salmonids. In: Status and Management of Cutthroat Trout (ed. Gresswell

RE), pp. 134–140. American Fisheries Society Symposium 4, Bethesda,

MD.

Harwood AS, Phillips RB (2011) A suite of twelve single nucleotide poly-

morphism markers for detecting introgression between cutthroat and

rainbow trout. Molecular Ecology Resources, 11, 382–385.

Hitt NP, Frissell CA, Muhlfeld CC, Allendorf FW (2003) Spread of hybrid-

ization between native westslope cutthroat trout, Oncorhynchus clarki

lewisi, and nonnative rainbow trout, Oncorhynchus mykiss. Canadian

Journal of Fisheries and Aquatic Science, 60, 1440–1451.

Hohenlohe PA, Amish SJ, Catchen JM, Allendorf FW, Luikart G (2011)

Next-generation RAD sequencing identifies thousands of SNPs for

assessing hybridization between rainbow and westslope cutthroat

trout. Molecular Ecology Resources, 11(Suppl. 1), 117–122.

Kalinowski ST, Novak BJ, Drinan DP, Jennings RDeM, Vu NV (2010)

Diagnostic single nucleotide polymorphisms for identifying westslope

cutthroat trout (Onchorhynchus clarki lewisi), Yellowstone cutthroat

trout (Oncorhynchus clarkii bouvieri), and rainbow trout (Oncorhynchus

mykiss). Molecular Ecology Resources, 11, 389–393.

Leary RF, Allendorf FW, Knudsen KL (1985) Developmental instability

and high meristic counts in interspecific hybrids of salmonid fishes.

Evolution, 39, 1318–1326.

May BE (2009) Westslope Cutthroat Trout Status Update Summary. Wild

Trout Enterprises LLC, Bozeman.

Muhlfeld CC, Kalinowski ST, McMahon TE et al. (2009) Hybridization

rapidly reduces fitness of a native trout in the wild. Biology Letters, 5,

328–331.

Narum SR (2006) Beyond Bonferroni: less conservative analyses for con-

servation genetics. Conservation Genetics, 7, 783–787.

Piry S, Alapetite A, Cornuet JM, Paetkau D, Baudouin L, Estoup A (2004)

GeneClass2: a software for genetic assignment and first-generation

migrant detection. Journal of Heredity, 95, 536–539.

Pritchard VL, Abadı́a-Cardoso A, Garza JC (2012) Discovery and charac-

terization of a large number of diagnostic markers to discriminate

Oncorhynchus mykiss and O. clarki. Molecular Ecology Resources, doi:

10.1111/j.1755-0998.2012.03149.x.

Pritchard VL, Jones K, Cowley DE (2007) Estimation of introgression in

cutthroat trout populations using microsatellites. Conservation Genetics,

8, 1311–1329.

Rannala B, Mountain JL (1997) Detecting immigration by using multilo-

cus genotypes. Proceedings of the National Acadamy of Science, 94, 9197–

9221.

Raymond M, Rousset F (1995) GENEPOP (version 1.2): population genet-

ics software for exact tests and ecumenicism. Journal of Heredity, 86,

248–249.

Ruzycki JR, Beauchamp DA, Yule DL (2003) Effects of introduced lake

trout on native cutthroat trout in Yellowstone Lake. Ecological Applica-

tions, 13, 23–37.

Shepard BB, May BE, Urie W (2005) Status and conservation of westslope

cutthroat trout within the western United States. North American Journal

of Fisheries Management, 25, 1426–1440.

Smith GR, Dowling TE, Gobalet KW, Lugaski T, Shiozawa DK, Evans RP

(2002) Biogeography and timing of evolutionary events among Great

� 2012 Blackwell Publishing Ltd

S N P D E T E C T I O N A S S A Y S F O R C U T T H R O A T T R O U T 7



Basin fishes. Great Basin Aquatic Systems History, Hershler R, Madsen

DB & Curry DR, editors. Smithsonian Contributions to the Earth Sciences

33, 175–234.

Taylor EB, Stamford MD, Baxter JS (2003) Population subdivision in

westslope cutthroat trout (Oncorhynchus clarki lewisi) at the northern

periphery of its range: evolutionary inferences and conservation impli-

cations. Molecular Ecology, 12, 2609–2622.

Waples RS, Gustafson RG, Weitkamp LA et al. (2001) Characterizing

diversity in salmon from the Pacific Northwest. Journal of Fish Biology,

59(Suppl. A), 1–41.

Wilson WD, Turner TF (2009) Phylogenetic analysis of the Pacific

cutthroat trout (Oncorhynchus clarki ssp.: Salmonidae) based on

partial mtDNA ND4 sequences: a closer look at the highly frag-

mented inland species. Molecular Phylogenetics and Evolution, 52,

406–415.

Wofford JEB, Gresswell RE, Banks MA (2005) Influence of barriers to

movement on within-watershed genetic variation of coastal cutthroat

trout. Ecological Applications, 15, 628–637.

Young MK(ed.) (1995) Conservation Assessment for Inland Cutthroat Trout.

US Forest Service General Technical Report. RM-256. Rocky Mountain

Research Station, Fort Collins, CO, 61pp.

Young MK (2011) Generation-scale movement patterns of cutthroat trout

(Oncorhynchus clarkii pleuriticus) in a stream network. Canadian Journal

of Fisheries and Aquatic Sciences, 68, 941–951.

Young SF, McLellan JG, Shaklee JB (2004) Genetic integrity and microgeo-

graphic population structure of westslope cutthroat trout, Oncorhyn-

chus clarki lewisi, in the Pend Oreille Basin in Washington.

Environmental Biology of Fishes, 69, 127–142.

Study design, sample collection, and oversight by KS

McKelvey, MK Young, MK Schwartz, JC Garza, G Lui-

kart and SR Narum. SNP identification using Sanger

sequencing was performed by NR Campbell and VL

Pritchard. SNPs were identified within RAD sequencing

data by SJ Amish and G Luikart. Assay design, genotyp-

ing, and assay evaluation was completed by NR Camp-

bell, SJ Amish, and VL Pritchard.

Data accessibility

SNP loci: dbSNP accessions ss494474925 – ss494475112.

The accessions are matched with reported assay names

in supplemental data tables.

SNP detection assays: Tables with primer ⁄ probe
sequences uploaded as online supplemental material.

Genotype Data: Genotypes at all tested loci for indi-

viduals in all 12 collections, GenePop files used for analy-

sis, and related raw data available in the Dryad database

(doi:10.5061/dryad.2m3h2m5 p).

Supporting information

Additional supporting information may be found in the online

version of this article.

Table S1 Describes each of the TaqmanTM assays, the panel used

for SNP discovery, their minor allele frequencies for each popu-

lation tested, and their informativeness for distinguishing popu-

lations (GST).

Table S2 Describes each of the KASPar assays, the panel used

for SNP discovery, their minor allele frequencies for each popu-

lation tested, and their informativeness for distinguishing popu-

lations (GST).

Table S3 Describes the 3 ascertainment panels used for SNP dis-

covery.

Table S4 Statistically significant linkage disequilibrium (LD)

between marker pairs within 2 or more populations.

Please note: Wiley-Blackwell are not responsible for the content

or functionality of any supporting information supplied by the

authors. Any queries (other than missing material) should be

directed to the corresponding author for the article.

� 2012 Blackwell Publishing Ltd

8 N . R . C A M P B E L L E T A L .