Abstract
Study of parallel (or convergent) phenotypic evolution can provide important insights into processes driving sympatric, ecologically-mediated divergence and speciation, as ecotype pairs may provide a biological replicate of the underlying signals and mechanisms. Here, we provide evidence for a selective sweep creating an island of divergence associated with reproductive behavior in sockeye salmon (Oncorhynchus nerka), identifying a series of linked single nucleotide polymorphisms across a ~22,733 basepair region spanning the leucine-rich repeat-containing protein 9 gene exhibiting signatures of divergent selection associated with stream- and shore-spawning in both anadromous and resident forms across their pan-Pacific distribution. This divergence likely occurred ~3.8 Mya (95% HPD = 2.1–6.03 Mya), after sockeye separated from pink (O. gorbuscha) and chum (O. keta) salmon, but prior to the Pleistocene glaciations. Our results suggest recurrent evolution of reproductive ecotypes across the native range of O. nerka is at least partially associated with divergent selection of pre-existing genetic variation within or linked to this region. As sockeye salmon are unique among Pacific salmonids in their flexibility to spawn in lake-shore benthic environments, this region provides great promise for continued investigation of the genomic basis of O. nerka life history evolution, and, more broadly, for increasing our understanding of the heritable basis of adaptation of complex traits in novel environments.
Similar content being viewed by others
Introduction
Understanding how and why populations diverge into new species or ecotypes are principal goals of evolutionary biology1. Of particular interest are the processes driving sympatric, ecologically-mediated divergence and speciation2. Recent studies have begun exploring the genetic bases of these processes, detecting genome-wide changes associated with divergent local adaptation3, 4, and the specific genes that underlie reproductive isolation5, 6.
Parallel (or convergent) phenotypic evolution, the repeated independent emergence of a specific phenotype associated with a particular habitat, can provide important insights into ecological divergence, as each ecotype pair provides a replicate of the underlying signals and mechanisms4. When similar environmental pressures and associated selective pressures occur in different populations within a species range, parallel evolution may lead to similar phenotypic changes7, 8 that may have the same underlying genetic basis9, 10. While such parallel divergent adaptation may originate from novel mutations in the same gene(s), it is often attributed to changes in frequencies of existing alleles11,12,13. A classic example of selection occurring on pre-existing genetic variation comes from the parallel evolution of freshwater forms of three-spined stickleback (Gasterosteus aculeatus), where anciently derived alleles in a gene responsible for armor reduction (ectodysplasin-A) have been recurrently selected in multiple freshwater populations12.
Studies of the genomic bases of local adaptation have been facilitated by the advent of high-throughput genotyping methods that allow for the identification and genotyping of thousands of genetic polymorphisms throughout the genome enabling population genomic and association studies in non-model organisms14,15,16. Such data further allow for the investigation of the genetics of adaptation through divergence mapping, where large suites of markers are screened for signatures of divergent selection among ecologically distinct populations17,18,19,20. The key to this approach is that substitutions physically linked to a beneficial allele will ‘hitchhike’, creating a region of lower diversity of fixed alleles in a process known as a ‘selective sweep’21, 22. These linked loci will then show higher differentiation between populations than neutral unlinked loci23.
Salmonids are an exemplary taxonomic group to study the genetic bases for ecologically-driven, sympatric divergence because abundant environmental variation among river drainages, combined with precise natal homing, create great potential for differential local adaptation24. Sockeye salmon, in particular, exhibit tremendous life history and morphological variation, with the repeated parallel evolution of several morphologically and ecologically divergent ecotypes linked to migratory and spawning behaviour24,25,26. All sockeye salmon spawn and spend their early life in freshwater, with anadromous ecotypes then migrating out to sea, and resident ecotypes (kokanee) remaining in freshwater lakes throughout their lifecycle26, 27. While the kokanee phenotype is similar across catchments, kokanee populations are polyphyletic, having evolved multiple times from anadromous sockeye salmon through independent postglacial freshwater colonization events25, 26, 28, 29.
Both anadromous sockeye salmon and kokanee can be further subdivided into reproductive ecotypes, with each population exhibiting a specific spawning habitat preference. These include the classical ‘stream (or river)-spawning’ ecotypes, ‘shore (or beach)-spawning’ ecotypes that spawn on the shallow submerged shorelines of lakes or island beaches, and ‘black’ kokanee that also spawn on the lake benthos, but at depths down to 70 m below the lake surface30. This variability in spawning habitat preference is unique to sockeye among Pacific salmon, with all other species in the genus spawning in streams31. In some lakes, multiple reproductive ecotypes co-occur, while in others only one may be present. Divergence between shore- and stream-spawning ecotypes can occur rapidly, with reintroduced sockeye salmon observed to form distinct, reproductively isolated populations of shore- and stream-spawning ecotypes in less than 13 generations32, 33. While it is possible that such divergence occurs due to philopatry or learned spawning habitat preferences, we predict there may be underlying genetic mechanisms behind ecotype divergence, potentially involving ‘speciation genes’34.
We recently conducted a landscape genomics study employing restriction-site associated DNA sequencing (RADseq) of paired population samplings of migratory (resident versus anadromous) and reproductive (shore- versus stream-spawning kokanee) ecotypes sampled from seven lakes and two rivers spanning three catchments (Columbia, Fraser, and Skeena drainages) in British Columbia, Canada35. We identified 334 outlier loci associated with life history variation, one of which was shared in both magnitude and direction of differentiation across all sampled lakes containing sympatric shore- and stream-spawning kokanee (R68810)35. The 100 base pair (bp) RAD tag containing this SNP mapped to leucine-rich repeat-containing protein 9 (LRRC9) in both the rainbow trout and Atlantic salmon genomes (Oncorhynchus mykiss chromosome 29, Salmo salar chromosome 9). LRRC proteins are involved in gene expression and participate in many biologically important processes, such as enzyme inhibition, hormone–receptor interactions, cell adhesion and cellular trafficking36. Furthermore, they have been associated with immunotoxicity in different fish species37, 38. Interestingly, this SNP was also recorded as highly divergent in a recent study by Nichols et al.39 between shore-spawning sockeye salmon and stream-spawning kokanee in Redfish Lake, Idaho, USA within the Snake River catchment, a tributary of the Columbia River. No other SNP was identified as being as significantly divergent between any pair of shore- or stream-spawning ecotypes in either of these studies, or in a recently published study comparing beach- and river-spawning sockeye salmon from Alaska (which did not include this locus)40. Furthermore, no other SNP has been recorded across multiple comparisons of shore- and stream-spawning ecotypes across these studies39, 40.
To test whether variation at this locus underlies sockeye salmon reproductive life history variation across the pan-Pacific distribution of the species, here we genotyped 1519 anadromous sockeye salmon and resident kokanee from 47 shore- and stream-spawning populations from Russia, Alaska and Canada. In addition, we sequenced approximately 23,000 bp spanning the entire LRRC9 gene to further characterize the surrounding genomic region and validate the role and history of divergent selection underlying reproductive ecotype divergence within sockeye salmon.
Results
Range-wide ecotype genotyping
Genotypic data collected using a newly developed TaqMan™ assay (One_LRRC9_68810; Table S1) targeting this SNP showed directional divergence in both anadromous sockeye salmon and resident kokanee across the natural range of O. nerka in Russia, Alaska and Canada (Table 1 and Figs 1 and 2). In general, the ‘G’ allele was most prevalent in shore-spawning sockeye salmon and kokanee populations, while the ‘T’ allele dominated in stream-spawning populations (Table 1 and Fig. 2). Notable exceptions were two shore-spawning sockeye salmon sites along island beaches in Illiamna Lake, Alaska and one shore-spawning kokanee site in the West Arm of Kootenay Lake, BC, all of which exhibited a higher frequency of the ‘T’ allele than observed in other shore-spawning populations. In addition, kokanee sampled while spawning near the mouth of Drew Creek in Tchesinkut Lake were fixed for the ‘G’ allele, identical to all sampled shore-spawning individuals from this system.
Five within-lake reproductive ecotype pairs spanning multiple catchments had >99% correct assignment to reproductive ecotype under a straight Mendelian assignment rule (GG = shore-spawning, GT or TT = stream-spawning), including Okanagan Lake kokanee shore- and stream-spawners, Christina Lake kokanee shore- and stream-spawners, Anderson Lake black kokanee/Portage Creek sockeye salmon, Seton Lake black kokanee/Portage Creek sockeye salmon, and Redfish Lake shore-spawning sockeye salmon/Fishhook Creek stream-spawning kokanee. In these systems, GG was recorded only once across the 238 stream-spawning individuals genotyped, and GT was only recorded four times within the 336 shore-spawning individuals (TT was never observed among shore-spawners) (Fig. 2 and Table S2). For the other two clearly differentiated co-occurring kokanee ecotype-pairs in Wood and Kalamalka Lakes, assignment accuracy using this rule was also high at >90%.
Flanking region sequencing
Sanger sequencing of 744 bp flanking this SNP in Okanagan Lake kokanee (Columbia River drainage), Anderson Lake kokanee and Portage Creek sockeye salmon (Fraser River drainage), revealed three additional SNPs in full linkage, suggesting this region was ancestrally inherited for both alleles in these populations spanning different river catchments.
To more broadly characterize this region, we successfully sequenced 22,773 bp flanking the One_LRRC9_68810 SNP from eight individuals each carrying homozygous genotypes of Okanagan Lake shore- and stream-spawning kokanee, respectively, including the entire LRRC9 gene (Genbank accession KY681681-KY681682). These two sequences were 4.6% divergent from each other, with 181 fixed differences including 23 multiple bp indels up to 308 bp long between the ‘shore-spawning’ and ‘stream-spawning’ alleles, suggesting the genomic region around LRRC9 has undergone a significant selective sweep (Fig. 3). Notably, the level of divergence between alleles increased markedly downstream (towards the 3′ end of the LRRC9 gene) (Fig. 3). Variation within each ecotype was far lower, with ten variable sites within the shore-spawning population, and nine variable sites within the stream-spawning population. While five SNPs were identified within the coding regions of the LRRC9 gene, we found no non-synonymous mutations between the ‘shore-spawning’ and ‘stream-spawning’ alleles.
Divergence timing of the two alleles
Using an estimated rainbow trout/sockeye salmon divergence time of approximately 11.4 million years ago (Mya)41, BEAST analysis revealed ‘shore-spawning’ and ‘stream-spawning’ alleles diverged from each other approximately 3.8 Mya (95%HPD = 2.1–6.03 Mya) (Fig. 4).
Discussion
Our results show the recurrent evolution of reproductive ecotypes across the native range of O. nerka is at least partially associated with divergent selection of pre-existing genetic variation within or linked to the region surrounding the LRRC9 gene. Although no non-synonymous changes were detected between LRRC9 ‘shore-spawning’ and ‘stream-spawning’ alleles, this does not preclude changes in splice sites or regulatory regions of the gene. Differentiation at this locus between shore- and stream-spawning populations was extremely high even when there was minimal neutral population differentiation. For example, pairwise F ST for One_LRRC9_68810 was 0.92 among reproductive ecotypes of Okanagan Lake kokanee despite the population genomic F ST of 0.008 based on 6,234 neutral SNPs35.
The few failures of this SNP to discriminate between ecotypes are likely explained by a combination of uncertainties in sampling, incomplete ecotype differentiation, and/or recombination events separating the SNP from the related genomic changes. For instance, in Tchesinkut Lake, the vast majority of individuals spawn on the shores of an island where the “shore-spawners” were originally sampled in 201042. The “stream-spawners” were sampled at an outflow and Drew Creek, however, the latter spawning activity only occurred once Ministry personnel cleared the mouth allowing kokanee access42. These observations, in tandem with the complete lack of divergence at 6,234 neutral SNPs, strongly suggest there may not be distinct ecotypes in Tchesinkut lake35. The other exceptions where stream-spawning populations had elevated levels of the G allele (Sinmax Creek kokanee, Tintina and Hanna Creek sockeye salmon) are immediately adjacent to shore-spawning sites in gravel near the creek mouths43. Finally, we obtained samples from three sockeye salmon beach-spawning sites in Illiamna Lake in Alaska, two on island beaches (Fuel Dump Island; Woody Island) and one mainland beach (Knutson Bay). While genotype frequencies within the mainland beach-spawning site were consistent with range-wide patterns (Fig. 2 and Table S2), the island-beach spawning sites showed relatively uniform genotype frequencies (Table S2). Specific beaches on these islands lack the upwelling groundwater typical of most beach-spawning salmonids44, and consequently, may not contain the site-specific olfactory signals to guide returning adults that would be necessary for promoting ecotype differentiation45.
In at least five lakes spanning the Columbia and Fraser River drainages (Anderson, Seton, Okanagan, Christina, Redfish), assignment accuracy to reproductive ecotype was >99%. Given our results, it appears that this SNP could be associated with a Mendelian trait, potentially creating a switch to an alternative spawning behaviour. Understanding how such a genetic switch might work in terms of influencing behaviour awaits future breeding and physiological experiments. If this SNP is linked to a gene that causes a change in spawning habitat preference, our data are consistent with the hypothesis that the T (stream) allele is associated with a preference or ability for stream-spawning, while the G (shore) allele could be a loss of function, leading to a lack of spawning habitat preference, enabling spawning anywhere with suitable substrate.
This hypothesis fits well with several other lines of evidence from shore- and stream-spawning populations of O. nerka. For example, in Lake Washington, Washington, USA, a reintroduced population of anadromous sockeye salmon diverged into reproductively isolated shore- and stream-spawning populations within 13 generations32, 33. In this system, both populations were derived from the same hatchery stock, and the population solely exhibited stream-spawning behaviour initially; shore-spawning was first recorded over 17 years after their introduction. The hatchery stock was derived from both stream-spawning and shore-spawning individuals32. This rapid heritable divergence into two spawning ecotypes matches the predictions for a previously existing polymorphism in a gene or genes that influence spawning behaviour, with shore-spawning recessive to stream-spawning. Similarly, in a study of anadromous sockeye salmon in Little Togiak Lake, Alaska where individuals genetically identified as coming from one population spawned in the alternate habitat, straying was rare and asymmetrical, primarily with individuals from the stream-spawning population using shore habitats46. These findings are consistent with the predicted pattern of a recessive allele promoting shore-spawning behaviour.
Sockeye salmon are known to have survived the late Wisconsin glaciation in several refugia, including areas south of the ice sheets such as the Columbia River, arid northern areas that remained largely free of ice in Beringia (region connecting Kamchatka and much of western Alaska), and in small inland mountain refugia where glaciers impeded access to the sea29. Sockeye salmon and kokanee descended from all of these refugia exhibit variability at this locus and evidence for divergent selection between ecotypes; therefore divergence between the two alleles must significantly predate the last glacial maximum. We estimate that the ‘shore-spawning’ and ‘stream-spawning’ alleles most likely diverged from each other around 3.8 Mya in the Pliocene, after sockeye separated from pink (O. gorbuscha) and chum (O. keta) salmon, but prior to the Pleistocene glaciations. While there are many simplifying assumptions for estimating the time to the most recent common ancestor for this region, particularly as it has likely been under selection, our divergence time estimate does highlight the great age of these alleles, both of which have been maintained in populations across the range of the species.
This example of recurrent selection of pre-existing variants in the population as a source for ecologically-driven sympatric divergence closely resembles the parallel evolution of freshwater forms of stickleback12. In this system, anadromous populations carry the anciently derived ‘freshwater’ armor genes at low frequency that then repeatedly went to fixation when stickleback colonized similar freshwater environments. Here, while the ‘G’ (shore) allele was uniformly prevalent in shore-spawning populations of both kokanee and sockeye salmon, it was never absent from stream-spawning populations. This suggests that (most) stream-spawning populations carry the ‘G’ allele, which is then strongly selected for as shore-spawning populations form.
Identifying the gene(s) under divergent selection
The size of the genomic region associated with a selective sweep of a locus under selection is determined by the strength of selection, local rate of recombination, and time since the beneficial mutation arose21, 47, 48. Because of these factors, genomic scans for signatures of selection may highlight regions spanning several megabase pairs (Mbp); therefore determining the gene(s) under selection in such cases remains challenging49. In a recent review of selective sweeps in cattle breeds, the size of genomic regions showing signals of selection ranged from 8.2 to 948 kilobase pairs (kbp), with a median of 78.7 kbp50. In dog breeds, which have a recent history of very strong selective pressures and inbreeding, selective sweeps may be up to 10 Mbp51. The most phylogenetically similar species to sockeye salmon with dense SNP panels is the Atlantic salmon (Salmo salar). In a recent study of this species, SNPs significantly associated with age at maturity were located in a selective sweep region covering ~370 kbp52.
There are 17 other genes within 250 kbp on either side of the LRRC9 gene in the S. salar genome (Table S4), and it remains possible that the genetic variation under divergent selection is outside this area, or that there are O. nerka structural differences compared to Atlantic salmon where other genes could be closer. While we highlight an ancient selective sweep between alleles closely linked to, and potentially underlying spawning behaviour in this genomic region, we have not yet identified the specific gene(s) linked to this divergent selection. Interestingly, this same genomic region of chromosome 9 in Atlantic salmon has been identified as an island of divergence between S. salar genetic clusters that differed in the length of sea migration and age at maturity53, 54. Of particular note, Barson et al.53 found an island of divergence spanning 250 kb centered in this region that was strongly associated with age at maturity. They hypothesized that variation in SIX6, a transcriptional regulator gene and distal forebrain enhancer55 that also regulates eye development across multiple taxa56, age at maturity in humans57, neuro-endocrine and gonad development58, might be the cause of this divergence. This gene is 142 kb away from the LRRC9 in Atlantic salmon. Whatever the underlying genetic mechanism, it is noteworthy that this region is associated with local adaptation and population divergence in multiple salmonids. A denser SNP map for sockeye salmon, and/or direct sequencing for several 100 kbp across multiple populations and ecotypes will be required to ascertain the underlying divergently selected gene(s). As it appears that the level of divergence between the ‘shore-spawning’ and ‘stream-spawning’ alleles increases towards the 3′ end (and the SIX6 gene in S. salar), this may be the direction to initially explore for the specific target(s) of divergent selection. As sockeye salmon are unique among the Pacific salmonids in their flexibility to spawn in lake-shore benthic environments31, this region provides great promise for future investigations of the genomic basis of O. nerka life history evolution, and more broadly, for increasing our understanding of the heritable basis of adaptation of complex traits in novel environments. From an applied perspective, this highly informative SNP has immediate utility for informing fisheries management throughout British Columbia59 and likely across the entire range.
Methods
Sampling
We used previously extracted DNA from 1519 anadromous sockeye salmon and resident kokanee from 47 shore- and stream-spawning populations in Russia, Alaska and Canada (Table 1). All original sampling and experimental procedures were conducted in accordance with institutional, national and international guidelines and regulations as cited within the original published work (Table 1).
SNP genotyping
We designed a new TaqMan™ assay (One_68810_LRRC9; Table 1) using a previously sequenced 100 bp RAD tag 68810 containing the SNP of interest35. SNP genotyping of all samples was performed using this TaqMan™ assay in 6 μl reactions: 2.5 μl TaqMan™ Universal PCR Master Mix (Life Technologies, Carlsbad, CA), 0.25 μl TaqMan™ Genotyping Assay (20x), 1.25 μl H2O and 2 μl of 1/10 diluted extracted DNA. Genotyping reactions were performed in 384 well plates using an Applied Biosystems ViiA7™ Real-Time PCR system (Life Technologies, Carlsbad, CA).
Flanking region sequencing and comparison
We used BLAST-n to locate and align the 100 bp RAD tag 68810 with the Atlantic salmon (Salmo salar) ICSASG_V2 (ssa09: 24,748,525–24,748,624) and rainbow trout (Oncorhynchus myskiss)60 genomes (chrUn_29: 1,729,057–1,729,247). We then aligned ~60 kbp of the flanking genomic regions of these two species centered on the 68810 SNP using a global alignment with open ends, assuming 70% similarity as implemented in Geneious 9.0.561.
Using this alignment (O. mykiss chromosome 29, S. salar chromosome 9), we designed PCR primers (Table S1) in PRIMER362 to amplify a ~750 bp fragment immediately surrounding the SNP for known homozygotes for four individuals from each of: Anderson Lake black kokanee, Portage Creek stream-spawning sockeye salmon, Okanagan Lake stream-spawning kokanee, and Okanagan Lake shore-spawning kokanee. All PCRs were carried out on an ABI Veriti thermal cycler in 25 μl reactions containing: 20–50 ng of DNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 200 μM dNTPs, 0.5 μM of each primer, 20 μg bovine serum albumin (BSA) and 0.5 U of AmpliTaq Gold DNA polymerase (Applied Biosystems). Cycling conditions were as follows: 95° (5 minutes), 30 cycles of 94° (20 seconds), 57° (30 seconds), 72° (45 seconds), and a final extension of 72 °C (7 minutes). All PCR products were purified by ExoSAP-IT (USB Products, Santa Clara, CA, USA) and Sanger sequenced using an ABI 3130XL Genetic Analyzer (Applied Biosystems).
The resulting sequences of the immediate flanking regions, along with the S. salar and O. mykiss alignment, were used to design two sets of primers in PRIMER362 for long-range PCRs in each direction targeting two overlapping fragments each of ~11 kbp (total contiguous sequence length of ~21 kbp). Long range PCRs were conducted using the LongAmp® Taq PCR kit (NEB) for eight individuals of each homozygous genotype at the 68810 SNP from Okanagan Lake shore- and stream-spawning kokanee, respectively. Each long-range PCR was carried out in 25 μl reactions containing: ~100 ng of template DNA, 60 mM Tris-SO4, 20 mM (NH4)2SO4, 2 mM MgSO4, 3% Glycerol, 0.06% IGEPAL® CA-630, 0.05% Tween® 20, 300 µM dNTPs, 0.5 μM of each primer, and 5 U LongAmp® Taq polymerase. Cycling conditions were as follows: 94° (30 seconds), 30 cycles of 94° (20 seconds), 58° (30 seconds), 65° (12 minutes), and a final extension of 65 °C (10 minutes). These PCR products were purified using a Qiagen MinElute gel extraction kit and individuals for each PCR product were pooled. Sequencing libraries were constructed by shearing the PCR products to ~400 bp and using the Illumina TruSeq DNA kit. Libraries were subsequently sequenced using the Illumina MiSeq PE250 platform. Library construction and sequencing were performed at the McGill University and Génome Québec Innovation Centre, Montréal, Canada.
Obtained sequence reads were assembled using the Geneious 9.0.561 de novo assembler, with medium sensitivity, five iterations and a maximum indel size of 1000 bp. The libraries for each ecotype were subsequently combined using a pairwise local alignment and variants detected using a minimum minor allele frequency of 0.25.
As the long-range PCRs did not cover the entire LRRC9, we designed two further primer pairs covering ~1 kbp each (LRRC_FR1 & LRRC_FR2; Table S2) to span the missing portions of the gene to give a final contiguous sequence length of ~23 kbp. All PCRs were carried out using the same individuals in 25 μl reactions containing: 20–50 ng of DNA, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 200 μM dNTPs, 0.5 μM of each primer, 20 μg BSA and 0.5 U of AmpliTaq Gold DNA polymerase (Applied Biosystems). Cycling conditions were as follows: 95° (5 minutes), 30 cycles of 94° (20 seconds), 56° (30 seconds), 72° (60 seconds), and a final extension of 72 °C (7 minutes). All PCR products were purified by ExoSAP-IT (USB Products, Santa Clara, CA, USA) and Sanger sequenced using an ABI 3130XL Genetic Analyzer (Applied Biosystems).
The concatenated sequences of each sequence (‘shore’ and ‘stream’) were then aligned in Geneious 9.0.561 using a global alignment with free end gaps, assuming a 93% similarity. The percentage differentiation between these aligned sequences was then calculated, and the number of divergent fixed SNPs and indels counted. We used plotcon in EMBOSS63 with a window size of 500 bp to display the pattern of differentiation between the sequences. We also translated the resulting LRRC9 DNA sequence using the S. salar CDS (LOC106610979) as a guide to identify non-synonymous changes in the coding region. To do this, the five published isoforms of the S. salar gene (XM_014210737.1 – XM_014210742.1) were aligned with the two O. nerka sequences, and these aligned exons were then translated and aligned with each other to detect any non-synonymous changes between the shore- and stream-alleles – all performed in Geneious 9.0.561.
Divergence timing
The combined sequences for each ecotype were aligned with each other, and with O. mykiss using the Geneious 9.0.5 local alignment (Smith & Waterman) tool assuming 70% similarity. This alignment was used to estimate the time of divergence between the ‘shore-spawning allele’ and the ‘stream-spawning allele’ conducted in BEAST64. Analyses were implemented using an HKY substitution model, an estimated divergence time between O. mykiss and O. nerka of 11.4 Mya (95%CI = 9.8–13 Mya)41, a normal distribution prior, and a relaxed lognormal clock with a Yule Birth-Death tree prior. Three independent runs consisting of 100 million generations were conducted, with a 25% burn-in. Outputs were assessed in Tracer65 and tree files combined in LogCombiner. Resulting tree files were annotated in TreeAnnotator and visualized in FigTree66.
References
Coyne, J. A. & Orr, H. A. Speciation. (Sinauer Associates, 2004).
Nosil, P. Ecological Speciation. (Oxford University Press, 2012).
Marques, D. A. et al. Genomics of Rapid Incipient Speciation in Sympatric Threespine Stickleback. PLOS Genetics 12, e1005887, doi:10.1371/journal.pgen.1005887 (2016).
Roesti, M., Gavrilets, S., Hendry, A. P., Salzburger, W. & Berner, D. The genomic signature of parallel adaptation from shared genetic variation. Molecular Ecology 23, 3944–3956, doi:10.1111/mec.12720 (2014).
Orr, H. A. The genetic basis of reproductive isolation: Insights from Drosophila. Proceedings of the National Academy of Sciences of the United States of America 102, 6522–6526, doi:10.1073/pnas.0501893102 (2005).
Noor, M. A. F. & Feder, J. L. Speciation genetics: evolving approaches. Nature Reviews Genetics 7, 851–861, doi:10.1038/nrg1968 (2006).
Waples, R. S., Teel, D. J., Myers, J. M. & Marshall, A. R. Life-history divergence in Chinook salmon: Historic contingency and parallel evolution. Evolution 58, 386–403, doi:10.1111/j.0014-3820.2004.tb01654.x (2004).
Rosenblum, E. B. & Harmon, L. J. “Same Same but Different”: Replicated Ecological Speciation at White Sands. Evolution 65, 946–960, doi:10.1111/j.1558-5646.2010.01190.x (2011).
Nachman, M. W., Hoekstra, H. E. & D’Agostino, S. L. The genetic basis of adaptive melanism in pocket mice. Proceedings of the National Academy of Sciences of the United States of America 100, 5268–5273, doi:10.1073/pnas.0421157100 (2003).
Paaby, A. B., Blacket, M. J., Hoffmann, A. A. & Schmidt, P. S. Identification of a candidate adaptive polymorphism for Drosophila life history by parallel independent clines on two continents. Molecular Ecology 19, 760–774, doi:10.1111/j.1365-294X.2009.04508.x (2010).
Barrett, R. D. H. & Schluter, D. Adaptation from standing genetic variation. Trends in Ecology & Evolution 23, 38–44, doi:10.1016/j.tree.2007.09.008 (2008).
Colosimo, P. F. et al. Widespread parallel evolution in sticklebacks by repeated fixation of ectodysplasin alleles. Science 307, 1928–1933, doi:10.1126/science.1107239 (2005).
Pearse, D. E., Miller, M. R., Abadia-Cardoso, A. & Garza, J. C. Rapid parallel evolution of standing variation in a single, complex, genomic region is associated with life history in steelhead/rainbow trout. Proceedings of the Royal Society B-Biological Sciences 281, doi:10.1098/rspb.2014.0012 (2014).
Baird, N. A. et al. Rapid SNP discovery and genetic mapping using sequenced RAD markers. PLOS One 3, doi:10.1371/journal.pone.0003376 (2008).
Miller, M. R., Dunham, J. P., Amores, A., Cresko, W. A. & Johnson, E. A. Rapid and cost-effective polymorphism identification and genotyping using restriction site associated DNA (RAD) markers. Genome Research 17, 240–248, doi:10.1101/gr.5681207 (2007).
Davey, J. W. et al. Genome-wide genetic marker discovery and genotyping using next-generation sequencing. Nature Reviews Genetics 12, 499–510, doi:10.1038/nrg3012 (2011).
Jones, F. C. et al. The genomic basis of adaptive evolution in threespine sticklebacks. Nature 484, 55–61, doi:10.1038/nature10944 (2012).
Storz, J. F. Using genome scans of DNA polymorphism to infer adaptive population divergence. Molecular Ecology 14, 671–688, doi:10.1111/j.1365-294X.2004.02437.x (2005).
Hohenlohe, P. A. et al. Population genomics of parallel adaptation in Threespine Stickleback using sequenced RAD Tags. PLOS Genetics 6, e1000862, doi:10.1371/journal.pgen.1000862 (2010).
Oleksyk, T. K., Smith, M. W. & O’Brien, S. J. Genome-wide scans for footprints of natural selection. Philosophical Transactions of the Royal Society B-Biological Sciences 365, 185–205, doi:10.1098/rstb.2009.0219 (2010).
Maynard-Smith, J. & Haigh, J. Hitch-hiking effect of a favorable gene. Genetics Research 23, 23–35, doi:10.1017/s0016672300014634 (1974).
Kaplan, N. L., Hudson, R. R. & Langley, C. H. The hitchhiking effect revisted. Genetics 123, 887–899 (1989).
Foll, M. & Gaggiotti, O. A genome-scan method to identify selected loci appropriate for both dominant and codominant markers: a Bayesian perspective. Genetics 180, 977–993, doi:10.1534/genetics.108.092221 (2008).
Quinn, T. P. The behavior and ecology of Pacific salmon and trout. (University of Washington Press, 2005).
Wood, C. C., Bickham, J. W., Nelson, R. J., Foote, C. J. & Patton, J. C. Recurrent evolution of life history ecotypes in sockeye salmon: implications for conservation and future evolution. Evolutionary Applications 1, 207–221, doi:10.1111/j.1752-4571.2008.00028.x (2008).
Dodson, J. J., Aubin-Horth, N., Theriault, V. & Paez, D. J. The evolutionary ecology of alternative migratory tactics in salmonid fishes. Biological Reviews 88, 602–625, doi:10.1111/brv.12019 (2013).
Gustafson, R. G. et al. Status review of sockeye salmon from Washington and Oregon. U.S., 282 (Department of Commerce, NOAA, Seattle, Washington, 1997).
Schluter, D. Ecological speciation in postglacial fishes. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 351, 807–814, doi:10.1098/rstb.1996.0075 (1996).
Taylor, E. B., Foote, C. J. & Wood, C. C. Molecular genetic evidence for parallel life-history evolution within a Pacific salmon (sockeye salmon and kokanee, Oncorhynchus nerka). Evolution 50, 401–416, doi:10.2307/2410810 (1996).
Moreira, A. L. & Taylor, E. B. The origin and genetic divergence of “black” kokanee, a novel reproductive ecotype of Oncorhynchus nerka. Canadian Journal of Fisheries and Aquatic Sciences 72, 1584–1595, doi:10.1139/cjfas-2015-0145 (2015).
Groot, C. & Margolis, L. Pacific salmon life histories. 543 (UBC Press, 1991).
Hendry, A. P. Adaptive divergence and the evolution of reproductive isolation in the wild: an empirical demonstration using introduced sockeye salmon. Genetica 112, 515–534, doi:10.1023/a:1013367100865 (2001).
Hendry, A. P., Wenburg, J. K., Bentzen, P., Volk, E. C. & Quinn, T. P. Rapid evolution of reproductive isolation in the wild: Evidence from introduced salmon. Science 290, 516–518, doi:10.1126/science.290.5491.516 (2000).
Nosil, P. & Schluter, D. The genes underlying the process of speciation. Trends in Ecology & Evolution 26, 160–167, doi:10.1016/j.tree.2011.01.001 (2011).
Veale, A. J. & Russello, M. A. Genomic changes associated with reproductive and migratory ecotypes in sockeye salmon (Oncorhynchus nerka). bioRxiv, doi:10.1101/117648 (2017).
Linhoff, M. W., Harton, J. A., Cressman, D. E., Martin, B. K. & Ting, J. P. Y. Two distinct domains within CIITA mediate self-association: Involvement of the GTP-binding and leucine-rich repeat domains. Molecular and Cellular Biology 21, 3001–3011, doi:10.1128/mcb.21.9.3001-3011.2001 (2001).
Xu, Y. L. et al. Studies on quantitative trait loci related to superoxide dismutase in mirror carp (Cyprinus carpio L.). Aquaculture Research 44, 1860–1871, doi:10.1111/j.1365-2109.2012.03191.x (2013).
Zelikoff, J. T. Biomarkers of immunotoxicity in fish and other non-mammalian sentinel species: predictive value for mammals? Toxicology 129, 63–71, doi:10.1016/s0300-483x(98)00064-x (1998).
Nichols, K. M., Kozfkay, C. C. & Narum, S. R. Geomic signatures among Oncorhynchus nerka ecotypes to inform conservation and management of endangered Sockeye Salmon. Evolutionary Applications 9, 1285–1300, doi:10.1111/eva.12412 (2016).
Larson, W. A. et al. Genomic islands of divergence linked to ecotypic variation in sockeye salmon. Molecular Ecology 26, 554–570, doi:10.1111/mec.13933 (2017).
Shedko, S. V., Miroshnichenko, I. L. & Nemkova, G. A. Phylogeny of salmonids (Salmoniformes: Salmonidae) and its molecular dating: analysis of mtDNA data. Russian Journal of Genetics 49, 623–637, doi:10.1134/s1022795413060112 (2013).
Frazer, K. K. & Russello, M. A. Lack of parallel genetic patterns underlying the repeated ecological divergence of beach and stream-spawning kokanee salmon. Journal of Evolutionary Biology 26, 2606–2621, doi:10.1111/jeb.12250 (2013).
Hume, J. M. B. et al. Evaluation of restoration efforts on the 1996 Upper Adams river sockeye salmon run. Canadian Technical Report of Fisheries and Aquatic Sciences 2466, pp. 57 (2003).
Wood, C. C. In Evolution and the Aquatic Ecosystem: Defining Unique Units in Population Conservation Vol. 17 American Fisheries Society Symposium Series (ed Nielsen, J. L.) 195–216 (1995).
Stewart, I. J., Quinn, T. P. & Bentzen, P. Evidence for fine-scale natal homing among island beach spawning sockeye salmon. Oncorhynchus nerka. Environmental Biology of Fishes 67, 77–85, doi:10.1023/a:1024436632183 (2003).
Lin, J., Quinn, T. P., Hilborn, R. & Hauser, L. Fine-scale differentiation between sockeye salmon ecotypes and the effect of phenotype on straying. Heredity 101, 341–350, doi:10.1038/hdy.2008.59 (2008).
Nielsen, R. et al. Genomic scans for selective sweeps using SNP data. Genome Research 15, 1566–1575, doi:10.1101/gr.4252305 (2005).
Messer, P. W. & Neher, R. A. Estimating the strength of selective sweeps from deep population diversity data. Genetics 191, 593–605, doi:10.1534/genetics.112.138461 (2012).
Huber, C. D., Degiorgio, M., Hellmann, I. & Nielsen, R. Detecting recent selective sweeps while controlling for mutation rate and background selection. Molecular Ecology 25, 142–156, doi:10.1111/mec.13351 (2016).
Boitard, S., Boussaha, M., Capitan, A., Rocha, D. & Servin, B. Uncovering adaptation from sequence data: lessons from genome resequencing of four cattle breeds. Genetics 203, 433–50, doi:10.1534/genetics.115.181594 (2016).
Schlamp, F. et al. Evaluating the performance of selection scans to detect selective sweeps in domestic dogs. Molecular Ecology 25, 342–356, doi:10.1111/mec.13485 (2016).
Ayllon, F. et al. The vgll3 locus controls age at maturity in wild and domesticated Atlantic salmon (Salmo salar L.) males. Plos Genetics 11, e1005628, doi:10.1371/journal.pgen.1005628 (2015).
Barson, N. J. et al. Sex-dependent dominance at a single locus maintains variation in age at maturity in salmon. Nature 528, 405–408, doi:10.1038/nature16062 (2015).
Johnston, S. E. et al. Genome-wide SNP analysis reveals a genetic basis for sea-age variation in a wild population of Atlantic salmon (Salmo salar). Molecular Ecology 23, 3452–3468, doi:10.1111/mec.12832 (2014).
Lee, B. et al. Direct transcriptional regulation of Six6 is controlled by SoxB1 binding to a remote forebrain enhancer. Developmental Biology 366, 393–403, doi:10.1016/j.ydbio.2012.04.023 (2012).
Agathocleous, M. et al. A directional Wnt/beta-catenin-Sox2-proneural pathway regulates the transition from proliferation to differentiation in the Xenopus retina. Development 136, 3289–3299, doi:10.1242/dev.040451 (2009).
Perry, J. R. B. et al. Parent-of-origin-specific allelic associations among 106 genomic loci for age at menarche. Nature 514, 92–97, doi:10.1038/nature13545 (2014).
Xie, H. M. et al. Homeodomain proteins SIX3 and SIX6 regulate gonadotrope-specific genes during pituitary development. Mol. Endocrinol. 29, 842–855, doi:10.1210/me.2014-1279 (2015).
Lemay, M. A. & Russello, M. A. Genetic evidence for ecological divergence in kokanee salmon. Molecular Ecology 24, 798–811, doi:10.1111/mec.13066 (2015).
Berthelot, C. et al. The rainbow trout genome provides novel insights into evolution after whole-genome duplication in vertebrates. Nature Communications 5, doi:10.1038/ncomms4657 (2014).
Kearse, M. et al. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28, 1647–1649, doi:10.1093/bioinformatics/bts199 (2012).
Untergasser, A. et al. Primer3Plus, an enhanced web interface to Primer3. Nucleic Acids Research 35, W71–W74, doi:10.1093/nar/gkm306 (2007).
Rice, P., Longden, I. & Bleasby, A. EMBOSS: The European molecular biology open software suite. Trends in Genetics 16, 276–277, doi:10.1016/s0168-9525(00)02024-2 (2000).
Drummond, A. J., Suchard, M. A., Xie, D. & Rambaut, A. Bayesian Phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution 29, 1969–1973, doi:10.1093/molbev/mss075 (2012).
Tracer v1.6, Available from http://beast.bio.ed.ac.uk/Tracer (2014).
FigTree available from http://tree.bio.ed.ac.uk/software/figtree/ (2009).
R: A language and environment for statistical computing (R Foundation for Statistical Computing, Vienna, Austria, 2016).
Kahle, D. & Wickham, H. ggmap: spatial visualization with ggplot2. The R Journal 5, 144–161 (2013).
Beacham, T. D. et al. Estimation of stock composition of sockeye salmon in the North Pacific Ocean. (NPAFC Doc. 783). (Department of Fisheries and Oceans, Pacific Biological Station, Nanaimo, B. C., Canada V9T 6N7, 2004).
Frazer, K. K. Investigating the divergence of reproductive ecotypes in kokanee salmon PhD thesis, The University of British Columbia Okanagan (2012).
Veale, A. J. & Russello, M. A. Sockeye salmon repatriation leads to population re-establishment and rapid introgression with native kokanee. Evolutionary Applications 9, 1301–1311, doi:10.1111/eva.2016.9.issue-10 (2016).
Lemay, M. A. & Russello, M. A. Neutral loci reveal structure by geography, not ecotype, in Kootenay Lake kokanee. North American Journal of Fisheries Management 32, 282–291, doi:10.1080/02755947.2012.676383 (2012).
Morbey, Y. E., Jensen, E. L. & Russello, M. A. Time scale matters: genetic analysis does not support adaptation-by-time as the mechanism for adaptive seasonal declines in kokanee reproductive life span. Ecology and Evolution 4, 3714–3722, doi:10.1002/ece3.1214 (2014).
Acknowledgements
Rick Taylor, Amanda Moreira, Ruth Withler, Richard Bussanich, Paul Askey, Hillary Ward, and James Baxter all provided samples for this project. Matthew Waterhouse provided assistance in the lab. Stephen Lougheed, Matthew Lemay and Evelyn Jensen offered valuable feedback on the manuscript. Jim Seeb, Wesley Larson, Ryan Waples, Alexei Drummond and Nic Rawlence provided valuable feedback on analyses. Funding for this work was provided to MR by Genome British Columbia, Okanagan Aquatic Enterprises, BC Ministry of Forests, Lands and Natural Resource, and BC Hydro.
Author information
Authors and Affiliations
Contributions
A.V. carried out the molecular lab work and data analyses, participated in the design of the study and drafted the manuscript; M.R. conceived, designed and coordinated the study, contributed to data collection and analyses, and helped draft the manuscript.
Corresponding author
Ethics declarations
Competing Interests
The authors declare that they have no competing interests.
Additional information
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Veale, A.J., Russello, M.A. An ancient selective sweep linked to reproductive life history evolution in sockeye salmon. Sci Rep 7, 1747 (2017). https://doi.org/10.1038/s41598-017-01890-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-017-01890-2
This article is cited by
-
Genome-wide assessment of kokanee salmon stock diversity, population history and hatchery representation at the northern range margin
Conservation Genetics (2022)
-
Maturation in Atlantic salmon (Salmo salar, Salmonidae): a synthesis of ecological, genetic, and molecular processes
Reviews in Fish Biology and Fisheries (2021)
-
Whole genome re-sequencing reveals recent signatures of selection in three strains of farmed Nile tilapia (Oreochromis niloticus)
Scientific Reports (2020)
-
Genotyping-by-sequencing supports a genetic basis for wing reduction in an alpine New Zealand stonefly
Scientific Reports (2018)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.