Spatial genetic structure of a keystone long-lived semiarid shrub: historical effects prevail but do not cancel the impact of recent severe habitat loss on genetic diversity

摘要

Land-use conversion and habitat loss and degradation are among the factors affecting populations and species genetic integrity. Understanding how these factors govern the genetic structure of threatened plant populations is essential to design efficient conservation strategies. Here we analyze how environmental correlates, geographic location and anthropogenic disturbance (at local, landscape and regional scales) relate to population differentiation and genetic diversity of a keystone semiarid long-lived shrub,Ziziphus lotus. The European range of this species is restricted mainly to the southeast of the Iberian Peninsula (> 90%) but its area has shrunk considerably as a consequence of intensive agriculture and urban expansion since the 1950s. We characterize its genetic diversity and structure across the Iberian Peninsula with microsatellite markers, considering 21 populations. All populations showed high and relatively similar values of genetic diversity (mean/pop:H-o = 0.54;H-e = 0.55;A(R) = 3.59). Three main genetic clusters were defined by clustering analyses corresponding largely to west, central (core) and east of its distribution. Molecular variance analysis revealed that 87.3% of genetic variation was grouped within populations, 5% among populations and 7% between regions. Geographic location (37.88%) has stronger correlation with population genetic differences than landscape characteristics (5.82%). Natural habitat cover at the landscape level was the only environmental variable associated withF(ST)population differences. The detected population genetic patterns are probably related primarily to historical gene flow and secondarily to intensive land-use change of this area in the last 60 years. Our results suggest that, although the long-lived and resprouting strategy of the species is buffering the erosion of genetic diversity at population and regional level, there are already genetic consequences of landscape degradation, which could be aggravated in the future.