Anthropogenic and natural drivers of gene flow in a temperate wild fruit tree: a basis for conservation and breeding programs in apples

Abstract

Gene flow is an essential component of population adaptation and species evolution. Understanding of the natural and anthropogenic factors affecting gene flow is also critical for the development of appropriate management, breeding, and conservation programs. Here, we explored the natural and anthropogenic factors impacting crop-to-wild and within wild gene flow in apples in Europe using an unprecedented dense sampling of 1889 wild apple (Malus sylvestris) from European forests and 339 apple cultivars (Malus domestica). We made use of genetic, environmental, and ecological data (microsatellite markers, apple production across landscapes and records of apple flower visitors, respectively). We provide the first evidence that both human activities, through apple production, and human disturbance, through modifications of apple flower visitor diversity, have had a significant impact on crop-to-wild interspecific introgression rates. Our analysis also revealed the impact of previous natural climate change on historical gene flow in the nonintrogressed wild apple M. sylvestris, by identifying five distinct genetic groups in Europe and a north-south gradient of genetic diversity. These findings identify human activities and climate as key drivers of gene flow in a wild temperate fruit tree and provide a practical basis for conservation, agroforestry, and breeding programs for apples in Europe.

Keywords: SPIPOLL; admixture; crabapple; dispersal; glacial refugia; global changes; pollinators.

Figure 1

Geographical quantification of introgression rates and numbers of hybrids among European wild apple sites in Europe and, more specifically, in France. (A) STRUCTURE barplot for K = 6, for the European wild apple gene pool (yellow) and the cultivated apple gene pool (black) at the European scale (N = 2228, 66 sites); below, associated country-averaged introgression rates (P_DOM→SYL) and number of hybrids. GB: Great Britain, BE: Belgium, FR (west): Western France, FR (east): Eastern France, ESP: Spain, IT: Italy, AUT: Austria, HUN: Hungary, BIH: Bosnia Herzegovina, RO: Romania, POL: Poland, BGR: Bulgaria, UKN: Ukraine, DEU: Germany, DA: Denmark, NO: Norway. (B) Site-averaged introgression rates (P_DOM→SYL) and number of hybrids in France (N = 1092, 20 sites).

Figure 2

Geographical genetic diversity, genetic structure, and admixture of Malus sylvestris in Europe (A) Population structure for the nonintrogressed Malus sylvestris dataset (N = 1376, 62 sites), inferred with TESS, for K_max = 8, showing five distinct genetic clusters. Each vertical line represents an individual. Individuals were grouped by country. Colors represent the inferred ancestry from K ancestral populations. For clarity, the 62 sites are grouped by country: GB: Great Britain, BE: Belgium, FR (West): Western France, FR (East): Eastern France, ESP: Spain, IT: Italy, AUT: Austria, HUN: Hungary, BIH: Bosnia Herzegovina, RO: Romania, POL: Poland, BGR: Bulgaria, UKN: Ukraine, DEU: Germany, DA: Denmark, NO: Norway; (B) Map representing the mean membership proportions inferred by TESS for the five detected clusters, for samples of Malus sylvestris collected from the same site (62 sites across Europe). (C) and (D) Maps of interpolated allelic richness A_R and observed heterozygosity H_O, respectively, across Europe.

Figure 3

Number of observations for the six orders of insect visitors to apple flowers in France (N = 306), in areas with more intensive (white) and less intensive (black) cultivation and management. ALL: number of observations for all orders considered together. ***: P < 0.0001; *: P < 0.01.