Populations of species with large spatial distributions are shaped by complex forces that differ throughout their ranges. To maintain the genetic diversity of species, genepool-based subsets of widespread species must be considered in conservation assessments. In this study, the population genetics of the lichenized fungus Lobaria pulmonaria and its algal partner, Symbiochloris reticulata , were investigated to determine population structure, genetic diversity, and degree of congruency in eastern and western North America. Data loggers measuring temperature and humidity were deployed at selected populations in eastern North America to test for climatic adaptation. To better understand the role Pleistocene glaciations played in shaping population patterns, a North American, range-wide species distribution model was constructed and hindcast to 22,000 years before present and at 500-year time slices from then to the present. The presence of two gene pools with minimal admixture was supported, one in the Pacific Northwest and one in eastern North America. Western populations were significantly more genetically diverse than eastern populations. There was no evidence for climatic adaptation among eastern populations, though there was evidence for range-wide adaptation to evapotranspiration rates. Hindcast distribution models suggest that observed genetic diversity may be due to a drastic Pleistocene range restriction in eastern North America, whereas a substantial coastal refugial area is inferred in the west. Taken together the results show different, complex population histories of L. pulmonaria in eastern and western North America, and suggest that conservation planning for each gene pool should be considered separately.

Methods

Sample Collection and Data Logger Deployment

A total of 19 sites and 369 individuals of L. pulmonaria were sampled across North America (Table 1). Twelve sites were in eastern North America, including one site in Newfoundland and 11 sites in the central and southern Appalachian Mountains. Seven sites were in the Pacific Northwest, including along the coast and in the Cascade Mountains. At each site, small pieces of lichen material were taken from up to three individual thalli per tree, from a maximum of ten trees, for an optimal total of 30 individuals per site. However, some sites had too few individuals to reach the optimal sample size, in which case, small pieces were taken from all detected thalli. The total number of individuals sampled per location is provided in Table 1. HOBO U23 Pro v2 Temperature/Relative Humidity Data Loggers (ONSET, Bourne, Massachusetts, USA) were deployed at five sites. The devices logged temperature and humidity every 30 minutes for one year (November 15, 2015 – November 15, 2016). Three loggers were deployed at each of the four sites in North Carolina: vicinity of the Chattooga River, Highlands Biological Station, Rough Butt Bald, and Roan Mountain. Five loggers were deployed in Newfoundland on the Avalon Peninsula near the settlement of Markland.

Molecular Methods

Samples were cleaned of debris, lyophilized, and DNA was extracted from whole thallus fragments using the DNeasy 96 Plant Kit (Qiagen, Hombrechtikon, Switzerland) according to the manufacturer protocol. Remaining lichen material was archived in cold storage. Eight fungal microsatellite loci (LPu03, LPu09, LPu15, LPu23, LPu24, LPu25, LPu28, and MS4) and nine green algal microsatellite loci (6825(1), 7007(2), 6816, 6819, 6820, 6863, 6828(2), 6861(3), 7000(2), LPu16, LPu19, LPu20, LPu26, and LPu27) were amplified according to previously established methods (Widmer et al., 2010; Dal Grande et al., 2012) using a PTC-100 thermal cycler (MJ Research, Poway, California, USA). A 3730 DNA Analyzer (Applied Biosystems, Waltham, Massachusetts, USA) was used for fragment analysis with LIZ-500 (Applied Biosystems, Waltham, Massachusetts, USA) as an internal size standard. GENEMAPPER 3.7 (Life Technologies) was used for genotyping.