Establishing genetic baselines for historic coregonine diversity in Lake Superior

Amanda Ackiss (USGS, aackiss@usgs.gov), Owen Gorman (USGS)

New research surveying morphological and genetic data across contemporary diversity in the cisco species complex has highlighted critical gaps in our understanding of the historic deepwater diversity in Lake Superior. Historic ciscoe diversity in Lake Superior, which includes 4 described species (Coregonus artedi, C. hoyi, C. kiyi, and C. zenithicus) and 3 described subspecies (C. nigripinnis cyanopterus, C. reighardi dymondi, and C. artedi albus), has been the subject of some debate among biologists (Eshenroder et al. 2016). In 1980, C. r. dymondi and C. n. cyanopterus were synonymized with C. zenithicus (Todd & Smith, 1980), however, Schmidt et al. (2011) found significant isotopic differences in museum specimens of C. r. dymondi suggesting it could be a valid subspecies. Contemporarily, the 4 described species are considered extant with all but C. zenithicus considered common, along with rare occurrences of a fifth C. nigripinnis-like form (Eshenroder et al. 2016). Recent genomic and isotopic data support the presence and differentiation of the 3 common species (C. artedi, C. hoyi, C. kiyi) but have not indicated evidence for significant differences in specimens identified as C. zenithicus (Ackiss et al. 2020, Bernal et al. 2022); both cited studies, however, were restricted to small sample sizes and/or limited geographic scopes. A FY20 GLRI-funded project entitled “Morphologic, geographic and genetic variation among Lake Superior ciscoes” led by co-PI Gorman, sought to address these issues by comprehensively surveying the contemporary morphological and genomic diversity of thousands of Lake Superior ciscoes across multiple geographic locations (Gorman et al, 2022a, b). Preliminary morphological data show evidence for a Lake Superior fish matching the description of C. reighardi by Koelz (1924, 1929) from Lake Michigan. These putative C. reighardi also exhibit a distinct genotype from the 3 common ciscoes and clusters with genotypes from C. reighardi scales collected in Lake Michigan prior to extinction. Despite the expanded dataset compiled by Gorman, the status of C. zenithicus and C. nigripinnis remains in question as contemporary specimens identified through morphology frequently genotype as C. hoyi, periodically as C. kiyi or C. reighardi, or somewhere intermediate. Unlike what was seen with C. reighardi, when historic Lake Michigan C. zenithicus genotypes were combined with contemporary Lake Superior data, those samples grouped out as a fifth discrete genetic cluster, rather than align with any Lake Superior C. zenithicus. There are numerous hypothetical explanations for the lack of contemporary evidence for discrete C. zenithicus and C. nigripinnis genotypes in Lake Superior. For example, it is possible that the historic forms were not true species or subspecies but rather zenithicus- or nigripinnis-like morphotypes of other species arising from phenotypic plasticity. Alternately, one or both may have at one point been genetically discrete and subsequently introgressed with another ciscoe species such as C. hoyi (proposed in Bernal et al. 2022). Stressors that significantly reduce the population sizes of closely related species, such as those that led to the declines of Great Lake coregonines, have been shown to induce a phenomenon known as reverse speciation which has been documented in complexes of European whitefish (Vonlanthen et al. 2012, Frei et al. 2022). With the impending resurrection of an extinct species (e.g., C. reighardi) and the lack of genetic support for what were recently assumed to be extant, rare species or subspecies, our understanding of the extant coregonine diversity in Lake Superior has fundamentally shifted, raising important questions that impact the conservation and restoration of ciscoes throughout the Great Lakes basin. The Lake Superior Biological Station (LSBS) in Ashland, WI contains an archive of thousands of historic scale samples collected from ciscoes from the late 1950s through the 1980s. We propose to survey this historic Lake Superior diversity using the same genomic methods used to recently survey contemporary diversity. These methods were successfully used on historic scale samples from 1920s-1990s in Lake Michigan for a previous GLRI grant entitled “Historical and Contemporary Genetic Diversity of Coregonus species from Lake Michigan,” and will enable direct comparisons between prior and proposed research. Historic genomic baselines will help shed light on the validity of Lake Superior C. zenithicus and C. nigripinnis, provide valuable comparisons between pre-collapse and post-collapse Lake Superior ciscoe diversity, and guide well-informed conservation and restoration actions to 1) preserve extant native preyfish diversity in Lake Superior and 2) leverage it as a resource for restoring lost coregonine biodiversity across the rest of the Great Lakes basin.