Chloroplast phylogeography of Dirca palustris L. indicates populations near the glacial boundary at the Last Glacial Maximum in eastern North America

Abstract

Aim Little is known about the distribution of temperate forest shrubs in eastern North America during the Last Glacial Maximum (LGM), and how far they subsequently migrated to occupy previously glaciated landscapes. We assessed whether populations of Dirca palustris L. persisted near the glacial margin during the LGM. Such populations might explain how the species colonized northern latitudes, despite scant evidence for contemporary long-distance seed dispersal.Location Eastern North America, Mexico and California.Methods We conducted ecological niche modelling (ENM) and sequenced five non-coding regions of the chloroplast genome from 390 plants among 104 populations representing the four species of Dirca. We used a Bayesian approach to identify relationships among individuals and to estimate timing of lineage splits. The spatial genetic structure of D. palustris was evaluated qualitatively and by statistical methods, including Monmonier's algorithm and analysis of molecular variance (AMOVA).Results Ecological niche modelling indicated suitable climate at the LGM in the Gulf Coastal Plain, but also north to the Ozark Plateaus Province (northern Arkansas), Interior Low Plateaus and Appalachian Plateaus Provinces (portions of Tennessee and Kentucky), and the Piedmont and Coastal Plain Provinces of the Carolinas and Virginia. Sequence variation was extensive and indicated deep lineage splits within D. palustris. Numerous haplotypes, many private, were identified at middle latitudes (c. 35-40°N), Monmonier's algorithm identified multiple genetic discontinuities within this region, and northern and southern regions differed by AMOVA.Main conclusions Ecological niche modelling and genetic analyses indicate that populations likely existed within several hundred km of the Laurentide ice sheet at the LGM. Deep Pleistocene diversification, along with weak geographical fidelity of haplotypes, probably resulted from repeated isolation, differentiation, and secondary contact during Quaternary glacial cycles. The absence of many northern haplotypes south of 35°N supports the hypothesis that temperate understorey shrubs accompanied temperate trees of eastern North America in populations near the glacial margin at the LGM.

Phylogeographic assessment of mtDNA paraphyly and the evolution of unisexuality inCalligrapha(Coleoptera: Chrysomelidae)

“…These range expansions toward coastal areas in the east ( p EL1 ) and particularly the latter toward the south ( p EL2 ) are coincident in time with the height of the Wisconsinan glaciation (12–35 ka BP) and consistent with the predictions of species range changes during the glaciations (Hewitt, , ). However, in our phylogeographic reconstruction, they also require that descendant populations remained close to their ancestral ranges, since there is no inferred evidence for recolonization from the hypothetical eastern and southern refugia into the ancestral range (as proposed for other organisms; e.g., Peterson & Graves, ). Moreover, the p EL lineage did not colonize, at least from a mtDNA perspective, the interior areas now occupied by descendants of the first group of C. philadelphica , indicating the possibility of some exclusive competition or demographic, leading‐edge effects in this group (Hewitt, ).…”

“…The current distribution of most Calligrapha species in North America, particularly of hybrid unisexual species, around the Great Lakes area (Gómez‐Zurita, ), as well as the estimated age of hybridization waves in these leaf beetles strongly suggest that climate change may have had a major role as the driver of range changes, secondary contacts and evolutionary processes in this group, as demonstrated for several other organisms, mainly in plants (e.g., Beatty & Provan, ; Peterson & Graves, ; Swenson & Howard, ). Indeed, this hypothesis is unquestionably orthodox, considering that current species and genetic pool ranges in North America are generally understood taking into account the oscillating climatic change during the Quaternary (Hewitt, , ).…”

“…In every interglacial, these ice age survivors colonized newly available habitat after the retreating ice and the reestablishment of favorable conditions in previously glaciated areas relatively quickly, at least in the case of plants (Clark et al., ). These migrations took advantage of the physiographic configuration of North America, offering limited barriers to south–north dispersal (Soltis et al., ), but also to the mentioned fact that temperate species may have remained close to the newly available habitat as suggested by the analysis of dispersal rates and recent phylogeographic evidence in plants (e.g., Clark et al., ; Peterson & Graves, ; Rowe et al., ; Snell & Cowling, ). The cyclical range expansions throughout Ice Ages changed the distribution of species following a variety of, sometimes relatively congruent patterns (e.g., Soltis et al., ).…”

Genomic evidence of a widespread southern distribution during the Last Glacial Maximum for two eastern North American hickory species

“…Although both species were likely fairly broadly distributed during the LGM, the fossil pollen record suggests that most LGM forest communities across southern ENA were conifer‐dominated with no modern analogues (Jackson & Williams, ; Jackson et al., ). Despite uncertainty about the LGM extent, or even the existence, of the temperate deciduous forest biome (Prentice et al., ), several other widespread temperate deciduous tree and plant species are also believed to have expanded from a large area covering much of southern ENA (Bennett, ; Magni et al., ; McLachlan et al., ; Peterson & Graves, ). In general, these results support the emerging understanding (Lumibao et al., ) that the genetic consequences of Pleistocene glaciation on widespread ENA temperate tree species were very different from those produced by expansion from distinct refugia in Europe (Hewitt, , ; Petit et al., ).…”

“…Third, despite numerous phylogeographical studies, well‐delineated glacial refugia generally shared by most species have not conclusively been identified. Proposed refugial locations include the Gulf Coast, the Atlantic Coast, Florida, Texas, the Ozark Plateau, the Lower Mississippi River Valley, the Appalachians and interior areas near ice sheets (Barnard‐Kubow, Debban, & Galloway, ; Griffin & Barrett, ; Jaramillo‐Correa et al., ; Magni, Ducousso, Caron, Petit, & Kremer, ; McCarthy & Mason‐Gamer, ; Morris, Graham, Soltis, & Soltis, ; Peterson & Graves, ; Soltis et al., ), which together sum to nearly the entire unglaciated region of ENA. While some species may have survived in one or more of these distinct refugia, other temperate taxa were likely not restricted to distinct LGM refugia, but were widespread over vast areas of the southeastern United States (Bennett, ; Lumibao et al., ; Magni et al., ; McLachlan et al., ; Peterson & Graves, ).…”