Genetic diversity among Great Lakes cisco species: exploring taxonomic and population boundaries

Contributing Authors

Wendylee Stott (DFO,

Executive Summary

Ciscoes (Coregonus artedi, hoyi, kiyi, zenithicus, and nigripinnis) once formed a diverse species flock in the Great Lakes. While several taxa appear to have been extirpated, those that remain are an important part of the Great Lakes food web. Deepwater forms (C. hoyi, kiyi, zenithicus, and nigripinnis) are most abundant in Lake Superior, bloater and cisco (C. artedi) are observed in lakes Huron and Michigan, and only cisco is observed in Lake Ontario. An important first step to maintaining or restoring within and among species diversity is describing extant diversity in compromised and healthy ecosystems. While past genetic analyses of bloater and cisco detected differences among the Great Lakes (Turgeon and Bernatchez 2003; Favé and Turgeon 2007), the number of collections per lake was low and not spatially comprehensive. In this project, we collected ~4,500 samples (C. artedi =2,533, C. hoyi=1,613, C. kiyi=255, C. zenithicus=93, C. nigripinnis=8) from all extant members of the cisco species flock in the Great Lakes and analyzed their genetic diversity using mitochondrial sequence variation, microsatellite DNA genotype data, and single nucleotide polymorphisms (SNPs) generated from restriction site associated DNA sequencing (RAD-seq). The larger sample sizes and greater spatial resolution allowed us to examine Great Lakes-wide population structure and analyze our ability to distinguish among taxa. Samples and sequence data collected as part of this objective are being used by Dr. Amanda Ackiss (USGS-GLSC) to develop and test a GTSeq panel that will distinguish among taxa and may have some use in population analyses and other activities such as hatchery monitoring programs. We saw two groups in all lakes and taxa (Figure 1) based on mitochondrial DNA diversity in sub-set of the samples (N=225) from all taxa and all lakes, supporting the idea that the Great Lakes represented a contact zone during post-glacial dispersal (Figure 2). With the expanded data set, we observed additional 25 mitochondrial DNA haplotypes, including several found only in C. kiyi (Figure 1). The new haplotypes found in C. kiyi were associated with the ‘clade B’ group identified in previous genetic analysis (Turgeon & Bernatchez 2003, Conserv. Genet. 4:67-81). There were significant differences among the frequencies of the clades among taxa, clade B was more common in C. artedi, C. kiyi, and C. zenithicus. In contrast, there were no differences in the frequencies of the two clades among lakes. While diversity was low among all haplotypes, taxa, and sample sites, we did observe more spatial variation within some lakes than previously observed (Figure1). In particular, there were several new haplotypes observed only in C. artedi from Georgian Bay (Lake Huron), which was not sampled in previous work. These haplotypes were most consistent with the ‘clade A' group. Genotype data from 14 microsatellite DNA loci for the entire data set showed that there were genetic differences among taxa and lakes (Figure 3). Genotype data could be used to distinguish between C. artedi and the deepwater ciscoes (C. hoyi, C. kiyi, C. zenithicus) with greater than 90% accuracy suggesting that genetic differences do exist among some taxa. The database of samples developed in this project was further developed with the additional of genotype data from lake whitefish (C. clupeaformis) to provide a reference for the identification of juvenile coregonines with greater than 90% accuracy (e.g., McKenna et al. 2020, J Gr Lakes Res. 46:1485-1490). The microsatellite DNA data was also able to detect population structure among samples of C. artedi among lakes and significant structure within Lake Huron. Several genetic populations were identified in the northern part of Lake Huron and in Georgian Bay (Figure 4). In contrast, less structure was observed within lakes among samples of C. hoyi (Figure 5), and very little structure was observed among samples of C. kiyi and C. zenithicus from Lake Superior, but sample sizes of the last two species may have been too low for a robust analysis. Genomic data from a sub-set of the C. artedi samples also detected population structure in Lake Huron and additional structure was detected in Lake Superior (Figure 6). Samples from the northern part of Lake Superior were different than those from the west.
Coalescent methods reconstruct contributions of natural colonization and stocking to origins of Michigan inland Cisco (Coregonus artedi) | Journal of Great Lakes Research
Spatial segregation of cisco (Coregonus artedi) and lake whitefish (C. clupeaformis) larvae in Chaumont Bay, Lake Ontario | Journal of Great Lakes Research
Genetic population structure of cisco, Coregonus artedi, in the Laurentian Great Lakes | Journal of Great Lakes Research
Genetic species identification of larval Coregonines from Chaumont Bay (New York), Lake Ontario | USGS

Funded In

Funding Agency


Restoration Framework Phase

Project Impact



Project Subjects