Genetic structure of the Canarian palm tree (Phoenix canariensis) at the island scale: does the 'island within islands' concept apply to species with high colonisation ability?

Oceanic islands are dynamic settings that often promote within-island patterns of strong population differentiation. Species with high colonisation abilities, however, are less likely to be affected by genetic barriers, but island size may impact on species genetic structure regardless of dispersal ability. The aim of the present study was to identify the patterns and factors responsible for the structure of genetic diversity at the island scale in Phoenix canariensis, a palm species with high dispersal potential. To this end, we conducted extensive population sampling on the three Canary Islands where the species is more abundant and assessed patterns of genetic variation at eight microsatellite loci, considering different within-island scales. Our analyses revealed significant genetic structure on each of the three islands analysed, but the patterns and level of structure differed greatly among islands. Thus, genetic differentiation fitted an isolation-by-distance pattern on islands with high population densities (La Gomera and Gran Canaria), but such a pattern was not found on Tenerife due to strong isolation between colonised areas. In addition, we found a positive correlation between population geographic isolation and fine-scale genetic structure. This study highlights that island size is not necessarily a factor causing strong population differentiation on large islands, whereas high colonisation ability does not always promote genetic connectivity among neighbouring populations. The spatial distribution of populations (i.e. landscape occupancy) can thus be a more important driver of plant genetic structure than other island, or species' life-history attributes.

Patterns of pollen dispersal in a small population of the Canarian endemic palm (Phoenix canariensis).

The genetic diversity of small populations is greatly influenced by local dispersal patterns and genetic connectivity among populations, with pollen dispersal being the major component of gene flow in many plants species. Patterns of pollen dispersal, mating system parameters and spatial genetic structure were investigated in a small isolated population of the emblematic palm Phoenix canariensis in Gran Canaria island (Canary Islands). All adult palms present in the study population (n=182), as well as 616 seeds collected from 22 female palms, were mapped and genotyped at 8 microsatellite loci. Mating system analysis revealed an average of 5.8 effective pollen donors (Nep) per female. There was strong variation in correlated paternity rates across maternal progenies (ranging from null to 0.9) that could not be explained by the location and density of local males around focal females. Paternity analysis revealed a mean effective pollen dispersal distance of ∼71 m, with ∼70% of effective pollen originating from a distance of <75 m, and 90% from <200 m. A spatially explicit mating model indicated a leptokurtic pollen dispersal kernel, significant pollen immigration (12%) from external palm groves and a directional pollen dispersal pattern that seems consistent with local altitudinal air movement. No evidence of inbreeding or genetic diversity erosion was found, but spatial genetic structure was detected in the small palm population. Overall, the results suggest substantial pollen dispersal over the studied population, genetic connectivity among different palm groves and some resilience to neutral genetic erosion and subsequently to fragmentation.

Allozyme variation and structure of the Canarian endemic palm tree Phoenix canariensis (Arecaceae): implications for conservation.

Electrophoretic analysis of 18 allozyme loci was used to estimate the levels and structuring of genetic variation within and among natural populations of the protected endemic palm species from the Canary Islands (Phoenix canariensis) to evaluate its genetic relationship with the widespread congener P. dactylifera, and to assess comparatively the genetic variation in the populations where the two species coexist with morphologically intermediate plants (mixed populations). Our survey revealed that the within-population component explains roughly 75% of the genetic variation levels detected in P. canariensis (A=1.59; P=41.8; He=0.158), which rank higher than those reported for other species of the Arecaceae. A Principal Component analysis (PCA) based on allele frequencies consistently separates populations of P. canariensis and P. dactylifera, and reveals a close genetic relationship between P. canariensis and the mixed populations. Reduced levels of genetic variation in P. canariensis with respect to P. dactylifera, the fact that the genetic makeup of the Canarian endemic (with no unique alleles) is a subset of that found in P. dactylifera, and the high genetic identity between both species strongly suggest that P. canariensis is recently derived from a common ancestor closely related to P. dactylifera.

High prevalence of GB virus C strains genetically related to strains with Asian origin in Nicaraguan hemophiliacs.

The presence of hepatitis GB virus C (GBV-C), also known as hepatitis G virus (HGV), and hepatitis C virus (HCV) were investigated in sera from 45 hemophiliacs from nine locations in Nicaragua using a nested polymerase chain reaction (PCR). Primers used to detect GBV-C and HCV derived from the helicase region and 5'UTR, respectively. Seventeen (38%) patients were positive for GBV-C RNA in serum by PCR. Twelve (27%) patients were positive for HCV RNA by PCR. Six (13%) of these were coinfected with GBV-C. Anti-HCV was detected in all the 12 HCV RNA positive hemophiliacs and in another 14 (31%) individuals, in whom GBV-C RNA was found in 2. Ten patients (22%) lacked markers for both GBV-C and HCV. The mean age of the patients positive for GBV-C but negative for HCV by PCR was significantly lower than for those negative for GBV-C but positive for HCV by PCR (P < 0.05; Student's t-test), indicating that the risk for this group of hemophiliacs to acquire GBV-C infection is higher as compared to the risk of acquiring HCV infection. Eleven GBV-C strains were sequenced in the 5'UTR. Sequence comparison to previously published GBV-C strains revealed that all 11 strains were more similar to Asian strains than to strains of European and African origin. Sequences in the NS5-B region were available for 8 HCV strains, all of which were found to belong to genotype 1a. The similarity of the Nicaraguan GBV-C strains to strains from Asia indicates that the GBV-C strains in the region presumably have an Amerindian origin. It is also considered that the HTLV II strains in the New World aboriginal populations are ancient and brought there by the ancestral Amerindian populations from Asia. Further, the genotype F of hepatitis B virus, known to represent the strains in populations with Amerindian background, predominates in Central American populations with Hispanic background. It remains to be clarified why Amerindian strains of GBV-C as well as of HBV predominate also in populations with mixed ethnic background in Central America.