Genomic footprints of Quaternary colonization and population expansion in the Patagonian‐Fuegian region rules out a separate southern refugium in Tierra del Fuego

Abstract

Aim During the glacial phases, the Andean slopes of southern South America were covered by ice sheets, causing the isolation of many species in glacial refugia. This work aimed to test the number and putative location of glacial refugia in southern South America and establish the sources, timing and direction of post‐glacial colonization events. Location Southern South America, Patagonia. Taxon Olive mouse (Abrothrix olivacea). Methods Using exome capture, we genotyped 7339 SNPs in 172 specimens of A. olivacea collected at 21 localities distributed across the steppe, grasslands, and forests in the Patagonian‐Fuegian region and seven specimens from two other species of Abrothrix. Based on the effective migration rate, principal component analysis, and admixture composition, we first study the population structure of A. olivacea. The source and direction of colonization events were inferred using the directionality index and standard population genetic analyses. To study the role of the glacial phases, we performed a bayesian demographic analysis to infer divergences times and ancestral population sizes. Results Population genomic analyses based on SNPs of more than 7000 exons identified three Patagonian genetic clusters: Valdivian forests, continental Magellan forests, and a combination of continental open biomes and Tierra del Fuego. The directionality index supports a single source in northwestern forests, in line with a Valdivian forest refugium, from which the post‐glacial colonization may have started. The effective diversity rate, the onset of population expansions and divergence times also support a Valdivian forest refugium and rule out a separate southern refugium in Tierra del Fuego. Main conclusions Our results support a coastal refugium located in or near the previously proposed Valdivian refugium and reject in situ differentiation in Tierra del Fuego. More in general, our demographical analysis reveals the role of the Last Glacial Maximum in affecting the population dynamics (i.e., contraction and expansion) of A. olivacea and in fostering the geographical diversification of populations.

The skull variation of the olive field mouseAbrothrix olivacea(Cricetidae: Abrotrichini) is localized and correlated to the ecogeographic features of its geographic distribution

“…The assessments of genetic variation of Abrothrix olivacea advanced by Giorello, D’Elía & Lessa (2021) and Quiroga-Carmona et al (2022) have mostly different geographic coverage between each other as well as with that of the present study; however, some of the uncovered cranial discontinuities are consistent with some of the identified genetic breaks. For instance, the genetic studies recognize that populations from northern Chile comprise a distinct genetic group, whose distinction match with cranial differentiation showed by populations that occur in CS and DF ( Figs.…”

“…This assessment also served to explore potential morphological differences among the populational clusters that can be recognized according to the recent studies about the structure of the genetic variation of this species. On this respect, Giorello, D’Elía & Lessa (2021) based on a panel of single nucleotide nuclear polymorphisms (nuclear SNPs), focused on Patagonian and Fueguian populations, indicates that the genomic variation of A . olivacea is structured in three main clusters, distributed in: Tongoy, Coquimbo region, Chile (site corresponding to the localities referred to DF), the Valdivian forest ( i.e ., northwest Patagonia; here referred as HF), and continuously throughout mainland Patagonia (except HF) and Tierra del Fuego, along Magellanic forests and Patagonian steppes (here referred as TF and PS, respectively; see Quiroga-Carmona et al, 2022 , Fig.…”

“…xanthorhina ). Thus, the grouping of individuals according to the environmental units present in their collection localities, closely considers the schemes derived from Quiroga-Carmona et al (2022 ; see also Giorello, D’Elía & Lessa, 2021 ). Cranial differences or resemblances among these groups were initially explored with a principal component analysis (PCA), performed with a variance-covariance matrix computed based on all craniodental measurements, using the R package FactoMineR ( Lê, Josse & Husson, 2008 ).…”

Climate influences the genetic structure and niche differentiation among populations of the olive field mouse Abrothrix olivacea (Cricetidae: Abrotrichini)

“…Thus, future research strategies focused on these species should contemplate the possibility that different evolutionary mechanisms are acting with uneven intensities on their populations, requiring the development of methodologies and approaches that cover distinct populations simultaneously and that, in turn, consider the heterogeneity of the landscape in which they exist. A good starting point under this premise could be achieved by using genetic data generated by previous studies but analyzing them with the approaches and methodologies used here, see 80 , 125 , ideally those containing comprehensive and exhaustive genetic samplings 34 , 118 . In this line, the characterization and analysis of the dimensions that constitute the ecological niche can become a mandatory procedure when attempting to study the dynamics of species with significant levels of population differentiation 58 , 126 .…”

“…Reconstructions of the historical demography of A. olivacea and their lineages based on mitochondrial variability were recently performed and indicate that Patagonian and Fueguian populations have undergone expansions that predates the LGM (around 70 Kay BP in populations from continental Patagonia [CS-Ch-Ar], and 55 Kay BP in those distributed in the Tierra del Fuego Island and surrounding continental regions [TdF-SCh]); see 37 . Meanwhile, demographic inferences based on nuclear genomic variation, but with a narrower geographic and specimen sampling, indicate that Patagonian populations expanded more recently (around 13 Kay BP), and likely in response to the retreat of the ice sheet at the end of the LGM 118 . This pattern is consistent with the expansion of climatically suitable areas as inferred with the model projections in the climatic scenarios corresponding to the LGM and the Mid-Holocene (SM, Figs.…”