Population differentiation in relation to conservation: nuclear microsatellite variation in the Canary Island endemic Lotus sessilifolius (Fabaceae)

Molecular phylogenetics of Lotus (Leguminosae) with emphasis in the tempo and patterns of colonization in the Macaronesian region

With a wide distribution range including Europe and Asia, Lotus (Leguminosae) represents the largest genus within Loteae. It is particularly diverse in the Mediterreanean region and in the five archipelagos of Macaronesia (Atlantic Ocean). However, little is known about the relationships among the 14 sections currently recognized within Lotus and about the timing and patterns of its colonization in the Macaronesian region. In this investigation, we use four DNA regions (nuclear ribosomal ITS plus three plastid regions) in the most comprehensive sampling of Lotus species to date (some endemic species within the Canary Islands were poorly represented in previous phylogenetic analyses) to infer relationships within this genus and to establish patterns of colonization in Macaronesia. Divergence time estimates and habitat reconstruction analyses indicate that Lotus likely diverged about 7.86 Ma from its sister group, but all colonization events to Macaronesia occurred more recently (ranging from the last 0.23 to 2.70 Ma). The diversification of Lotus in Macaronesia involved between four and six independent colonization events from four sections currently distributed in Africa and Europe. A major aspect shaping the current distribution of taxa involved intra-island colonization of mainly new habitats and inter-island colonization of mostly similar habitats, with Gran Canaria and Tenerife as the major sources of diversification and of further colonization events. Section Pedrosia is the most diverse in terms of colonization events, number of species, and habitat heterogeneity, including a back-colonization event to the continent. Subsections within Pedrosia radiated into diverse habitat types recently (late Pleistocene, ca 0.23-0.29 Ma) and additional molecular markers and sampling would be necessary to understand the most recent dispersal events of this group within the Canary Islands and Cape Verde.

A generalized framework for AMOVA with multiple hierarchies and ploidies

The analysis of molecular variance (AMOVA) is a widely used statistical method in population genetics and molecular ecology. The classic framework of AMOVA only supports haploid and diploid data, in which the number of hierarchies ranges from two to four. In practice, natural populations can be classified into more hierarchies, and polyploidy is frequently observed in extant species. The ploidy level may even vary within the same species, and/or within the same individual. We generalized the framework of AMOVA such that it can be used for any number of hierarchies and any level of ploidy. Based on this framework, we present four methods to account for data that are multilocus genotypic and allelic phenotypic (with unknown allele dosage). We use simulated datasets and an empirical dataset to evaluate the performance of our framework. We make freely available our methods in a new software package, polygene, which is freely available at https://github.com/huangkang1987/polygene.

The restoration of the endangered Sambucus palmensis after 30 years of conservation actions in the Garajonay National Park: genetic assessment and niche modeling

The translocation of individuals or the reinforcement of populations are measures in the genetic rescue of endangered species. Although it can be controversial to decide which and how many individuals must be reintroduced, populations can benefit from reinforcements. Sambucus palmensis is a critically endangered endemic to the Canary Islands. During the past 30 years, the Garajonay National Park (La Gomera) has carried out an intensive program of translocations using cuttings, due to the low germination rates of seeds. To assess the effect of the restorations on the population genetics of S. palmensis in La Gomera, we collected 402 samples from all the restored sites and all known natural individuals, which were genotyped with seven microsatellite markers. In addition, we conducted a species distribution modeling approach to assess how restorations fit the ecological niche of the species. Results show that there is a high proportion of clone specimens due to the propagation method, and the natural clonal reproduction of the species. Nonetheless, the observed heterozygosity has increased with the restorations and there still are private alleles and unique genotypes in the natural populations that have not been considered in the restorations. The population of Liria constitutes a very important genetic reservoir for the species. To optimize future reintroductions, we have proposed a list of specimens that are suitable for the extraction of seeds or cuttings in a greenhouse, as well as new suitable areas obtained by the species distribution models.