Low genetic diversity and potential inbreeding in an isolated population of alder buckthorn (Frangula alnus) following a founder effect

Low genetic diversity and potential inbreeding in an isolated population of alder buckthorn... www.nature.com/scientificreports OPEN Low genetic diversity and potential inbreeding in an isolated population of alder buckthorn (Frangula alnus) Received: 18 November 2016 following a founder effect Accepted: 25 April 2017 Published: xx xx xxxx 1 1 1,3 2 Caroline M. V. Finlay , Caroline R. Bradley , S. Jane Preston & Jim Provan Alder buckthorn (Frangula alnus) is one of Ireland’s rarest tree species, and in Northern Ireland the species is now restricted to a single population in Peatlands Park, Co. Armagh numbering ca. 140 mature trees. Genotyping of 95% of the trees at nine nuclear microsatellite loci revealed that levels of genetic diversity within this population were generally lower than those reported from larger populations in Spain. Analysis of six chloroplast microsatellite loci revealed no variation. The level of F was significantly higher than that in the Spanish populations, as well as in other populations across IS Europe, potentially indicating inbreeding. Spatial autocorrelation analysis indicated some evidence of fine-scale genetic structuring, most likely due to limited seed dispersal, but the overall level of differentiation between subpopulations was low, indicating high levels of gene flow, probably due to cross-pollination by bees. Our results are consistent with a gradual population expansion from a limited number of individuals. We suggest that more immediate conservation efforts might be best focused on ensuring suitable habitat for the continued recovery of this isolated population. Populations of endangered or threatened species tend to be small and/or isolated and are thus particularly vul- nerable to stochastic processes. These problems are further exacerbated at the genetic level, where the increased 1, 2 effects of genetic drift and potential for inbreeding can lead to low levels of genetic va . Thir s c iaa tn b ion e fur - ther compounded if such populations have been founded by a limited number of indi , v sin idu ce al g senetically 4, 5 depauperate populations tend to have reduced evolutionary potential, which can increase the risk o . f extinction Where populations are fragmented, as is often the case in threatened taxa, reduced levels of gene flow between fragments can also aggravate the problems associated with limited genetic diversity, as there is less scope for 6, 7 immigration of alleles to counter the effects of dr . Co ift nsequently, knowledge of the levels and patterns of genetic diversity in populations of threatened species are vital to the formation of well-informe-d, ee ff ctive con 8, 9 servation plans. Frangula alnus (alder buckthorn) is one of Ireland’s rarest tree species. Although widespread in temperate Europe, the species has a very limited and fragmented distribution in Ireland, where it has been in serious decline 10, 11 over the last few decades as a result of drainage of its preferred bogland habitat for alterna (Fig tive l . 1 a ). n d use In Northern Ireland, recent surveys suggested t . ah lnu at s i Fs restricted to the southern shores of Lough Neagh. Although there are records of t . a hle nu Fs previously occurring on the northern side of the Lough, as well as a single tree in Drumawhey Bog, County Dow , t n hese are now extinct, the former natural woodland having been replaced by a broadleaf plantat . The p ion resent-day surviving population has been part of Annagarriff Nature Reserve in Peatlands Park, County Tyrone since 1978, and the species is protected under the Wildlife (NI) Order (1985) and is a Priority Species for Conservation Action. The history of this population, which curre - ntly num bers ca. 140 individuals (see Methods and Fig 1), i .  s not well-documented. The earliest records mention “Twenty bushes on the NE margin of Annaghgarriff [sic] … before 1934” , and census numbers appear to have remained low for many years, with a record of “about 30 young plants” around 1987–89 . Since then, there has been a gradual increase in numbers to those found today, possibly due to removal of rhododendron from the area, but 1 2 School of Biological Sciences, Queen’s University Belfast, Belfast, BT9 7BL, UK. Institute of Biological, Environmental and Rural Sciences, Aberystwyth University, Aberystwyth, SY23 3DA, UK. Present address: ATECNI Environmental Consultancy, 31 Castlewellan Road, Banbridge, BT32 4JQ, UK. Correspondence and requests for materials should be addressed to J.P. (email: J.Provan@aber.ac.uk) Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 1 www.nature.com/scientificreports/ Figure 1. Location of the Peatlands Park population of Frangul in C a alno u. A s rmagh, Northern Ireland. Zoomed area shows the distribution o alf nu F. s in Peatlands Park, with the five fragments sampled (labeled A–E), mapped using ArcMap 10. The coloured area indicates a designated Special Area for Conservation (SAC). Land classes were taken from the CEH Land Cover Map (NERC/Centre for Ecology & Hydrology). DB – Drumawhey Bog (extinct population referred to in text). it is not known whether the original trees were remnants of a once larger population, or whether numbers have always been low due to an initial founder effect. In recent years, the use of polymorphic microsatel lite markers has allowed the testing of whether populations have gone through a bottleneck, based on theoretical expectations 15–18 under mutation-drift equilibrium at a single point in tim . C eonsequently, the aim of the present study was to determine the levels of and patterns of genetic diversity in the remaining population to discover (1) whether there is any evidence for a genetic bottleneck, (2) if the establishment from a relatively limited number of individuals has been accompanied by a degree of inbreeding, and (3) whether there is any significant genetic substructuring 19, 20 within the population. In addition,. a as lnu F s is considered an invasive pest species in many countri , o esur findings could also shed light on the genetic demography of this recently expanded population with respect to similar invasive populations. Results Current distribution of F. alnus in Northern Ireland. Surveys of sites where . a Flnus had been recorded previously found that the species is now restricted to a single location; Peatlands Park, Co. Armagh. The sole remaining population exists as five discrete clusters of plants numbering between 3–98 individuals, each sepa- rated by between 100–1,30 m 0 (Fig. 1; FigureS   1, Supplementary Material). In total, ther.e 1 a 4r 0 m e ca ature trees. e o Th nly other population recorded, at Drumawhey Bog on the northern edge of Strangford Lough, now appears to have been extirpated. According to the CEH Land Class Map 2007 and site visits, Subpopulation A resides in a suburban area and is the only subpopulation outside the SAC of Peatlands Par 2), S k (Fig ubp . opulation B resides in grass-dominated bog, Subpopulation C occurs on the edge of deciduous woodland, Subpopulation D is split between grass-dominated bog, heather-dominated bog and scrub, and Subpopulation E occurs on the boundary between a coniferous and deciduous woodland. The majority of the trees in the population occur in bog (40.1%), divided between grass dominated bog (17.4%) and heather dominated bog (22.7%). The next most common habitat type for this population is scrub (33.3%), then deciduous woodland (16.7%), suburban areas (8.3%) and the least com- mon habitat type these trees occur in is coniferous woodland (1.5%). Levels of genetic diversity. No evidence of linkage disequilibrium was detected between the nine loci studied. Genotypes were obtained for 132 of the 139 trees (95%). Allele frequencies by locus and subpopulation are given in Table  S1, Supplementary Material. Within-subpopulation levels of genetic diversity averaged across loci are given in Tab1 le  and ranged from 0.222 (Subpopulation E) to 0.404 (Subpopulation B) for observed heterozygosity (H mean = 0.314) and from 0.331 (Subpopulation C) to 0.423 (Subpopulation D) for expected heterozygosity (H mean = 0.387). Mean inbreeding coefficients ( F ) across loci (Tab1 le ) ra   nged from − 0.022 E IS Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 2 www.nature.com/scientificreports/ Figure 2. Correlogram of autocorrelation coefficient (θ; y-axis) plotted against distance (x-axis). 95% confidence intervals are indicated by dashed red lines. Subpopulation N H H F O E IS A 11 0.305 0.39 0.226** B 6 0.404 0.422 0.050 NS C 21 0.340 0.331 −0.022 NS D 91 0.299 0.423 0.293** E 3 0.222 0.372 0.625** Mean 132 0.314 0.387 0.234 Total 132 0.308 0.411 0.251** Table 1. Diversity statistics. N – number of samples; – o H bserved heterozygosity; – exp H ected O E heterozygosity F; – inbreeding coefficient. Significance of F : *P < 0.05; **P < 0.01; ***P < 0.001; NS – Not IS IS Significant. (Subpopulation C) to 0.625 (Subpopulation E; m = e a 0.234). n Three of the subpopulations had F values signif- IS icantly greater than zero. Diversity values and inbreeding coefficients calculated for Subpopulation E should be treated with some caution, as this fragment only contained three trees. Treating the five subpopulations as a single population gave values of 0.308, 0.411 and 0.251 H for , H and F respectively. Values of summary statistics by O E IS locus and subpopulation are given in T S2 ab , S le  upplementary Material. No evidence of a genetic bottleneck was detected under any of the three mutation models, with two of the nine nuclear loci studied showing a heterozy- gote excess under all three mutation models (T 2). able  The Peatlands Park population exhibited significantly lower levels of genetic diversity than two of the three Spanish populations based on the loci analyzed in the present s = tud 0.411 y (H vs. H = 0.608 for Ajibe E E [Mann-Whitney test, = z −1.99, P = 0.047]; H = 0.603 for Medio [Mann-Whitney tes t, = − z 1.99, P = 0.047]; (2) E (2) H = 0.470 for Puerto Oscuro [Mann-Whitney tes =t, −z 0.98, P = 0.327]; Fig. 3a). The mean value of F for E (2) IS the Peatlands Park population (0.251) was significantly higher than those from the three Spanish populations (F = 0.015 for Ajibe [Mann-Whitney test, = z 2.12, P = 0.034]; F = −0.027 for Medio [Mann-Whitney test, IS (2) IS z = 2.03, P = 0.042]; F = −0.070 for Puerto Oscuro [Mann-Whitney tes =t, 2.30, z P = 0.021]; Fig. 3b). (2) IS (2) No diversity was observed for the chloroplast microsatellites, with all six loci being monomorphic. Levels of genetic differentiation and spatial genetic structuring. Differentiation among subpopu- lations calculated from the analysis of molecular variance (AMO 3) V wa A; s TΦ able  = 0.0322 (p = 0.009). e Th ST BAPS analysis indicated that the first four subpopulations belonged to the same genetic cluster, with the fifth (Subpopulation E) being genetically distinct (Figur S2, Su e  pplementary Material). The spatial autocorrelation analysis did, however, suggest that there may be some n fi e-scale spatial genetic structuring within Subpopulation D, with the autocorrelation coefficient being significantly higher than zero at the smallest scale ( 2). <50 m; Fig.  Discussion e g Th enetic analysis of the sole remaining population of Frangula in N alnusorthern Ireland carried out in the pres- ent study suggests that establishment from a limited number of individuals has led to limited level -s of genetic var iation, accompanied by potential inbreeding during the recent expansion in census population size. A comparison with results from a study o . a n lnu F s in Spain suggests that the Peatlands Park population has significantly lower levels of genetic diversity than two of the three Spanish populations. The complete lack of genetic variation in the six chloroplast microsatellite loci studied is also consistent with a founder ee ff ct. Although comparable data for these markers are not available for the Spanish populations, chloroplast microsatellites represent the most variable regions of the chloroplast geno , m an ed have also been shown to be monomorphic in a previously well-documented population bottleneck in Torrey p . Thin e o ebserved differences in levels of diversity between Northern Ireland and Spain could also reflect longer-term historical factors, since the Spanish populations most likely represent ref- ugial populations during the last glaciation, and that the low levels observed in Northern Ireland are consistent Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 3 www.nature.com/scientificreports/ Locus B101 A110 B7 A104 B106 B4 A7 A3 B9 Empirical data   Sample size (haploid genomes) 262 244 264 264 262 230 264 256 258   Heterozygosity observed () H 0.482 0.713 0.067 0.269 0.407 0.348 0.593 0.244 0.578   Number of alleles observed ( ) k 3 5 5 8 7 8 7 3 7 IAM   Average heterozygosity ()H 0.276 0.453 0.447 0.610 0.566 0.616 0.600 0.282 0.609   Standard deviation (SD) 0.188 0.182 0.180 0.141 0.157 0.136 0.145 0.189 0.150   Standard deviate ([H − H ]/SD) 1.097 1.430 −2.113 −2.424 −1.010 −1.971 −0.046 −0.204 −0.210 O E   Probability H ( > H ) 0.206 0.037 0.021 0.026 0.167 0.053 0.396 0.471 0.330 TPM   Average heterozygosity ()H 0.396 0.598 0.602 0.745 0.710 0.748 0.740 0.389 0.741   Standard deviation (SD) 0.166 0.115 0.118 0.070 0.077 0.065 0.071 0.161 0.074   Standard deviate ([H − H ]/SD) 0.517 0.991 −4.516 −6.83 −3.966 −6.116 −2.064 −0.901 −2.218 O E   Probability H ( > H ) 0.394 0.121 0.000 0.000 0.006 0.001 0.042 0.218 0.033 SMM   Average heterozygosity ()H 0.438 0.649 0.649 0.786 0.751 0.785 0.785 0.435 0.783   Standard deviation (SD) 0.137 0.089 0.093 0.047 0.058 0.048 0.048 0.144 0.052   Standard deviate ([H − H ]/SD) 0.324 0.712 −6.268 −10.89 −5.902 −9.128 −4.206 −1.33 −3.968 O E   Probability H ( > H ) 0.471 0.264 0.000 0.000 0.000 0.000 0.004 0.128 0.009 Table 2. Results of the Bottleneck analysis. Values in bold indicate heterozygote excess. Values in italics indicate heterozygote deficiency. Variance Source of variation d.f components % variation Fixation index Between subpopulations 4 0.05651 3.22 Φ = 0.0322** ST Within subpopulations 259 1.70113 96.78 Table 3. Analysis of molecular variance (AMOVA). Significance o : f ** ΦP < 0.01. ST Figure 3. Boxplots showing values of ( H a) , and (b) F in the Peatlands Park population analyzed in the E IS present study and three Spanish populations analyzed in Riguier . (2009). o et al 24, 25 with the founder effects associated with postglacial recolonization (“southern richness and norther . n purity”) Interestingly, though, a further study on populations from Italy, France, Belgium and Sweden using 186 single nucle- otide polymorphisms (SNPs) showed no decrease in genetic diversity with l.atitude Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 4 www.nature.com/scientificreports/ Despite the low levels of genetic variation, the Wilcoxon test for heterozygote excess did not indicate the occurrence of a bottleneck. Seven of the nine loci indicated a heterozygote deficiency, which could be indicative of population stability or constant gr. o A wt lth hough there are no records predating 1934, evidence suggests that the population remained relatively stable from this time until the end of the 1980s, after which numbers 10, 13 increased to the current censc u as . 14 of 0 individuals . It may be unwise to take this at face value, though, since the power of the test is affected by several parameters, including number of loci used, number of generations since the bottleneck, the length of the bottleneck itself, and the magnitude of reduction in effective population size. Immigration, another potentially confounding factor, can most likely be ruled out, since the closest known F. alnus populations to that in Peatlands Park are fo . 125 und kca m away in Co. Westmeath, Ireland. Likewise, the mutation model is unlikely to have an effect, since the results were broadly consistent across the three models ana- lyzed, and changing the parameters of the two-phase model from 90% single-stepwise mutations to 70% also had no effect. The generation time of F. alnus has been estimated to be between 5 years in Central Europe to around 20 years in the South , so the maximum number of generations of growth is likely to be much less than 10, although the period of apparent population stability preceding this could be upwards of 10 generations. These time scales may limit the power of the heterozygote excess test to detect any possible bottleneck, but this may be balanced to some extent by the fact that, based on historical records, the effective populat) io an er ft size the (N bottleneck is unlikely to be any higher tha. 20 n ca . Removal of rhododendron from areas of the park since the designation of the Annagarriff Nature Reserve may have facilitated the increase in numbe.r a sl o nu f s F (Keith Stanfield, personal communication), but this has apparently been accompanied by a degree of inbreeding, based o va n lues. S F uch inbreeding is most likely a IS result of the relatively limited genetic base of the population, particularly compared to the other populations of F. alnus. The mean value of F for the Peatlands Park population is significantly higher than those from the three IS Spanish populations studied previou. I slt i y s also far higher than the range of values report a e lnu d in s pF opu . - lations across Europe based on SNPs ( =F −0.107–0.088, mean = −0.015) . Reproductive dominance by a rel- IS 28, 29 atively low number of highly fecund individuals within a population could give rise to biparental in , breeding but since F . alnus possesses a self-incompatibility mechanism, as indicated by crossing s , t tudies his would be dependent on the diversity o a f l Sleles in the population. Previous studies have indicated a breakdown in 31, 32 self-incompatibility following population bottlen , e bc u ks t such a possibility would have to be tested via con- trolled pollination experiments. The absence of high levels of spatial structuring of genetic variation in the Peatlands Park pop. ulation of F alnus indicates a general lack of barriers to dispersal, which is unsurprising given the small spatial scale relative to the potential dispersal distances of pollinators. The overall level of population differen = ti 0.032) ation ( wa Φ s ST lower than the average values for outcrossing species with seeds dispersed by gravity (0.152) or ingestion (0.200) quoted by Hamrick & God, a t s well as the average value for biparentally inherited markers in angiosperms (0.184) quoted by Petit et al . . The spatial autocorrelation analysis suggested some fine-scale genetic structure 27, 35 in Subpopulation D, probably due to limited seed dispersal. Se. ea dls o nus f c Fan be dispersed by birds , but in the closely related (congeneric in some classificat R io ha nm s)nus cathartica, it has been shown that 90% of fruits fall beneath the mother t . Thir s s ee cenario could at least in part be responsible for the hig obs h er F ved in IS Subpopulation D, by way of a Wahlund effect. e a Th rea surrounding Peatlands Park has a number of apple orchards with large numbers of wild bees in the vicinity, and the high incidence of fruits in the Peatland Fs . a Pla nu rk s population suggests substantial levels of cross-pollination. Bees have large foraging ranges, even across sub-optimal habitats, and are likely to be important 38, 39 in maintaining the connectivity observed between the fragments in the presen.t I n st c uo dn ytrast to the levels of fruiting found in Peatlands Park, where almost all trees had multiple fruits, a previous study on reproduction in populations of F a. lnus from Cádiz, Spain found that only 2.8% of open-pollinated flowers set f . A rui no tther study on southern range edge populations also indicated that the majority of ovule losses were due to cross-pollen limitation and extensive geitonogamy, and that seed output in the populations was limited to a fe . w large trees e hig Th h percentage of fruiting trees observed in Peatlands Park suggests that this population is more similar to those found in Central European populations, which have a shorter generation time and higher levels of fruit production compared to Southern Iberian popula. I tio f t n hse observed high F values reflect some degree of IS inbreeding, there has been no apparent impact on fitness, at least in terms of fruit production. Although the BAPS analysis assigned Subpopulation E to a separate genetic cluster from the other four sub- populations, this should be taken with some degree of caution for several reasons. Firstly, the subpopulation num- bers only three individuals, and thus allelic frequencies will be skewed. Secondly, it has been shown previously that the BAPS algorithm tends to over-estimate the true number of genetic clusters presen . Fin t in t all hy e d , ata only a single private allele is present in Subpopulation E, with the majority of genetic differentiation being due to the aforementioned differences in allele frequencies at several of the loci studied. Whilst it is true that knowledge of the evolutionary dynamics of natural populations with respect to demog- raphy and gene flow allows the management of threatened plant populations to go beyond simple “conservation gardening” , in the case of . F alnus in Northern Ireland it would appear that edaphic and ecological factors are of greater importance. The land classes that are correlated with occurrence of alder buckthorn include bog, conifer and pastures, and this is consistent with the literature that claims this species likes open lowland areas with moist, 43, 44 fertile soils . Nevertheless, there are large areas of apparently suitable habitat in Peatlands Park which have not been colonized during the recent expansion of the species in the nature reserve. This would suggest that although the low levels of genetic diversity and potential inbreeding revealed in the population are of some concern with respect to evolutionary potential, more immediate conservation efforts might be better focused o- n ensuring suit able habitat for the continued recovery of this isolated population. Additionally, consideration should be given to possible supplementation using material from the populations in the Republic of Ireland, which are larger and may harbor additional genetic diversity. Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 5 www.nature.com/scientificreports/ Methods Study species. Frangula alnus Miller (syn. Rhamnus frangul L.) i a s a small tree or shrub found across tem- perate Europe from northern Scandinavia and Russia to the Mediterranean, where it is compara . t In ively scarce Britain and Ireland, the species tends to be found in damp, boggy habitats, but can also colonize drier ground out- 19, 20, 35 side of its native range, particularly in North America where the species is now considered an invasi . ve alien Reproduction in F. alnus is exclusively sexua. lFlowers are hermaphroditic and well-adapted to insect polli- nation as they have nectar, colour and odour. They are generalised entomophiles, with orders Hymenoptera, Diptera and Coleoptera providing the main pollinators, and occasional pollination by L . C ep oido ntr p o tl era led pollination studies revealed almost no selfing or geitonogamy, indicating the existence of self-incompatibility mechanisms, although occasional selfing in the absence of insect pollinators has been. Fi rep eld or s tt eu ddies 30, 40 suggest that limited fruit initiation is primarily due to low levels of cross-po. S llin eed as a tio rn e dispersed 27, 35, 37 either by gravity or via ingestion by birds . Surveys and sampling. Surveys were carried out across Northern Ireland in the summer of 2007 at sites where F. alnus had been found previously based on records from the National Biodiversity Network (NBN) Gateway (http://data.nbn.org.uk) and the Centre for Environmental Data and Recording (CEDaR: http://www.habitas.org. uk/cedar/). Frangula alnus is restricted to the southern shores of Lough Neagh, where it exists as a fragmented pop- ulation from a single location in Annagarriff Nature Reserve in Peatlands Park, County T 1). A yro l n l m e (Fig atur .  e trees were numerically tagged with metal tags, GPS coordinates recorded, and a leaf sample from each obtained for genetic analysis. The GPS coordinates were loaded into ESRI ArcMap 10 and plotted on top of a CEH Land Cover Map 2007 to discover the main land class that the trees occupy. DNA was extracted from leaf material using the CTAB method . Genotypes were successfully obtained for 132 of the 139 trees sampled. Nuclear microsatellite analysis. We attempted to genotype all mature plants for sixteen previously described microsatellite loci f . ao ln r uF s . Of these, four (FaA103, FaA125, FaA8 and FaB8) could not be con- sistently amplified, and three (FaA12, FaB102 and FaA116) were monomorphic, leaving nine polymorphic loci: FaB101, FaA110, FaB7, FaA104, FaB106, FaB4, FaA7, FaA3 and FaB9. All reactions were carried out on a MWG Primus thermal cycler. PCR was carried out in a total volμ um l co e o nf 10 tainin g 100 ng g enomic DNA, 10p mol of dye-labelled forward primer (HEX), 1 pmo l of tailed forward primer p, 10 mo l reverse primer, 1x PCR reaction buffer, 200 μM each dNTP, 2.5mM MgC l and 0.25U G oTaq Flexi DNA polymerase (Promega). PCR conditions were as described previousl . G y enotyping was carried out on an AB3730xl capillary genotyping system. Allele sizes were scored using the GeneMapper software package (V5.0; Applied Biosystems) and LIZ-500 size stand- ards, and were checked by comparison with previously sized control samples. Chloroplast microsatellite analysis. Chloroplast microsatellite markers were developed by identifying mononucleotide regions of ten or more repeats in p F.a a rlt ni ua sl chloroplast genome sequences either from GenBank or in regions amplified and sequenced de n u ovo sing universal chloroplast prim . Ther e tr s n T-trnF, atpH-atpI, atpI-rpoC2, rps18-clpp, psbC-trnS, trnS-trnfM and rpl16-rps3 (UCP6) regions were amplified and 47 48 sequenced using the primers described in Grivet . e a t n ald Provan et al . . Species-specific primers were designed using the Primer3 program to amplify six chloroplast microsatelli S3 t, S es (T upp alem ble entary Material). Primers were tailed as described previously, and PCR was carried o μ ul t r in e 1 ac 0tions as described previously using the following conditions: initial denaturatio °n C a fo tr 9 3 4 min followed by 30 cycles of denaturation ° a Ct f 9 or 4 30 s, annealing at 56 °C f or 30s, ext ension at 72 °C f or 30s a nd a final extension at 72 °C f or 5min. G enotyping was carried out as described previously. Data analysis. Subpopulations of the Peatlands Park population were classed as distinct groups of trees with no other trees of this species growing within 100 m bet ween groups, resulting in five distinct subpopulations (Fig 1). .  GenePop V3.4 was used to test for linkage disequilibrium between nuclear loci. To estimate genetic diversity within the population, levels of observe) a d (H nd expected (H ) heterozygosity, and fixation indices ( ) w F ere O E IS 50 51 calculated using t Ar he lequin (V3.5.1.2) and Fstat (V2.9.3.2) sowa ft re packages respectively. Significance o F f IS was determined by 10,000 randomization steps. To test for the occurrence of a genetic bottleneck, the Wilcoxon test for heterozygote excess was performed under the infinite alleles model (IAM), the stepwise mutation model (SMM) and a two-phase model (TPM) incorporating 90% single-stepwise mutations using the program Bo ttleneck (V1.2) . The Wilcoxon test was used as it is recommended for a relatively low number of loci. To compare genetic diversity between the population analyzed in the present study, and those from three Spanish populations previously analyzed using the same nuclear micros, m ate el ali n va teslues for H and F E IS were calculated over the twelve loci successfully amplified in the present study, including the three loci which were monomorphic. Mann-Whitney tests were carried out to assess the significance of differen an ces in d F .H E IS The level of genetic differentiation between the five fragments was estimated usin , w g Φ hich gives an ana- ST 53 54 logue of F calculated within the analysis of molecular variance (AMOVA) fra u m sin ew g oAr rklequin. To ST identify possible spatial patterns of gene flow, the software package BAPS (V5) was used to identify clusters of genetically similar subpopulations using a Bayesian approach. Ten replicates were run for all possible values of the maximum number of cluster) u s (K p to K = 5, the number of subpopulations, with a burn-in period of 10 000 iterations followed by 50 000 iterations. Multiple independent runs always gave the same outcome. To further identify possible spatial patterns of gene flow, spatial autocorrelation analysis was carried out for Subpopulation 56 57 D, the largest of the five subpopulations, using SPAG (V1.4) eDi . Mean coancestry coefficients (θ ) between xy pairs of individuals were calculated a m di t 25 st ance class intervals, and plotted as a correlogram, with 95% con- fidence intervals calculated from 1,000 permutations of individuals within each distance class, and for estimates of θ using 1,000 permutations. xy Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 6 www.nature.com/scientificreports/ References 1. Hedrick, P. W. & Kalinowski, S. T. Inbreeding depression and conservation biology. Ann. Rev. Ec 31 ol. S , 139–162, do yst. i:10.1146/ annurev.ecolsys.31.1.139 (2000). 2. Cole, T. C. Genetic variation in rare and common plants. Ann. Rev. Ecol. Evo34 l. S , 213–237, do yst. i:10.1146/annurev. ecolsys.34.030102.151717 (2003). 3. Nei, M., Maruyama, T. & Chakraborty, R. The bottleneck effect and genetic variability in population 29 s. , 1–10, Evolut ion doi:10.2307/2407137 (1975). 4. Hansson, B. & Westerberg, I. On the correlation between heterozygosity and fitness in natural populatio 11 ns. , 2467–2474, Mol. Ecol. doi:10.1046/j.1365-294X.2002.01644.x (2002). 5. Frankham, R. Genetics and extinction. Biol. Cons 126 erv, 131–140, do . i:10.1016/j.biocon.2005.05.002 (2005). 6. Wright, S. Isolation by distance. Genet 28 ic , 114–138 (1943). s 7. Young, A., Boyle, T. & Brown, T. The population genetic consequences of habitat fragmentation for plants. Trends Ec 11o , l. Evol. 413–418, doi:10.1016/0169-5347(96)10045-8 (1996). 8. Haig, S. M. Molecular contributions to conservation. 79 Ec, 413–412 (1998). ology 9. Hedrick, P. W. Recent developments in conservation genetics. Forest Ecol. Ma197 nagem. , 3–19, do i:10.1016/j.foreco.2004.05.002 (2004). 10. Harron, J., Rushton, B. S. & Newbould, P. J. Flora of Lough Neagh. Ir. N 22 at, 1–270 (1986). . J. 11. Rich, T. C. G., Beesley, S. & Goodwillie, R. Changes in the vascular plant flora or Ireland between pre-1960 and 1987-1988, the BSBI Monitoring Scheme. Ir. Nat. J 26 . , 333–350 (2001). 12. Rippey, I. Alder buckthorn Frangula alnus Miller and other scarce plants in Co. Down (H38). 23, 223–224 (1990). Ir. Nat. J. 13. Rippey, I. Some new county records and other botanical notes. Ir 23 . N , 216–218 (1990). at. J. 14. Powell, W., Machray, G. C. & Provan, J. Polymorphism revealed by simple sequence repeats. Trends P 1l, 215–222, ant Sci. doi:10.1016/S1360-1385(96)86898-0 (1996). 15. Cornuet, J. M. & Luikart, G. Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genet i144 cs , 2001–2014 (1996). 16. Luikart, G. & Cornuet, J. M. Empirical evaluation of a test for identifying recently bottlenecked populations from allele frequency data. Conserv. Biol. 12, 228–237, doi:10.1111/j.1523-1739.1998.96388.x (1998). 17. Garza, J. C. & Williamson, E. G. Detection of reduction in population size using data from microsatellite lo 10, 305–318, ci. Mol. Ec ol. doi:10.1046/j.1365-294x.2001.01190.x (2001). 18. Peery, M. Z. Reliability of genetic bottleneck tests for detecting recent population decli 21 ne,s 3 . 4 M 0o 3l– . E 34 c1 o8 l., doi:10.1111/ j.1365-294X.2012.05635.x (2012). 19. McClain, W R. hamnus frangula. In: Invasive Plants: Weeds of the Global Garden (eds Randall J. M. & Marinelli J.), Brooklyn Botanic Garden Publications, New York, USA (1996). 20. Cunard, C. & Lee, T. D. Is patience a virtue? Succession, light and the death of invasive glossy buckthorn (Fra ). ng Bi uo la l. a lnus Invasions 11, 577–586, doi:10.1007/s10530-008-9272-8 (2009). 21. Rigueiro, C., Arroyo, J. M., Rodriguez, R., Hampe, A. & Jordano, P. Isolation and characterization of 16 polymorphic microsatellite loci for Frangula alnu (R s hamnaceae). Mol. Ecol. Resource9 s , 986–989, doi:10.1111/men.2009.9.issue-3 (2009). 22. Provan, J., Powell, W. & Hollingsworth, P. M. Chloroplast microsatellites: new tools for studies in plant ecology and systematics. Trends Ecol. Evol. 16, 142–147, doi:10.1016/S0169-5347(00)02097-8 (2001). 23. Provan, J., Soranzo, N., Wilson, N. J., Goldstein, D. B. & Powell, W. A low mutation rate for chloroplast microsat e 153 lli, tes. Genetics 943–947 (1999). 24. Provan, J. & Bennett, K. D. Phylogeographic insights into cryptic glacial refugia. Trends Ec 23o , 564–571, do l. Evol. i:10.1016/j. tree.2008.06.010 (2008). 25. Hewitt, G. M. Post-glacial recolonization of EuropeaB ni b ol. io Jt . L a.i nnean Soc. 68, 87–112, doi:10.1111/bij.1999.68.issue-1-2 (1999). 26. de Kort, H., Vandepitte, K., Mergeay, J. & Honnay, O. Isolation, characterization and genotyping of single nucleotide polymorphisms in the non-model tree species Frangula aln (R uh s amnaceae). Conserv. Genet. Resource6 s , 267–269, doi:10.1007/s12686-013-0083-6 (2014). 27. Hampe, A. & Bairlein, F. Modified dispersal-related traits in disjunct populations of bird-dispersed Fra (R ngu hla amn alnace us ae): a result of its Quaternary distribution shifts? Ecogr 23 ap , 603–613, do hy i:10.1034/j.1600-0587.2000.230511.x (2000). 28. Aldrich, P. R. & Hamrick, J. L. Reproductive dominance of pasture trees in a fragmented tropical forest m os 281 aic. , Science 103–105, doi:10.1126/science.281.5373.103 (1998). 29. Cascante, A., Quesada, M. & Lobo, J. J. Effects of tropical forest fragmentation on the reproductive success and genetic structure of the tree Samanea saman . Conserv. Biol. 16, 137–147, doi:10.1046/j.1523-1739.2002.00317.x (2002). 30. Medan, D. Reproductive biology of Frangula aln (R ushamnaceae) in southern SpaP in. lant Syst. Evol. 193, 173–186, doi:10.1007/ BF00983549 (1994). 31. Reinartz, J. A. & Lee, D. H. Bottleneck induced dissolution of self-incompatibility and breeding systems consequences in Aster furcatus (Asteraceae). Am. J. Bo81 t. , 446–455, doi:10.2307/2445494 (1994). 32. Busch, J. The evolution of self-compatibility in geographically peripheral populations of Leavenworthia a (B laba ram ssic icaaceae). Am. J. Bot. 92, 1503–1512, doi:10.3732/ajb.92.9.1503 (2005). 33. Hamrick, J. L. & Godt, M. J. W. Effects of life history traits on genetic diversity in plant species. Phil. Trans. R. Soc. L 351 on , don B 1291–1298, doi:10.1098/rstb.1996.0112 (1996). 34. Petit, R. J. et a . C l omparative organization of chloroplast, mitochondrial and nuclear diversity in plant populat 14 io, ns. Mol. Ecol. 689–701, doi:10.1111/j.1365-294X.2004.02410.x (2005). 35. Hampe, A. The role of fruit diet within a temperate breeding bird community in southern Spain. B 48 ird , 116–123, Study doi:10.1080/00063650109461209 (2001). 36. Richardson, J. E., Fay, M. F., Cronk, Q. C. B., Bowman, D. & Chase, M. W. A phylogenetic analysis of Rhamnaceae using rbcL and trnL-F plastid DNA sequences. Am. J. B 87 ot, 1309–1324, do . i:10.2307/2656724 (2000). 37. Archibold, O. W., Brooks, D. & Delanoy, L. An investigation of the invasive shrub European buckthorn Rhamnu L., n s catha ea rr tica Saskatoon, Saskatchewan. Can. Field N 111 at. , 617–621 (1997). 38. Steffan-Dewenter, I. & Kuhn, A. Honeybee foraging in differentially structured landP sr c oa c. R pes. . Soc. London B 270, 569–575, doi:10.1098/rspb.2002.2292 (2003). 39. Kreyer, D., Oed, A., Walther-Hellwig, K. & Frankl, R. Are forests potential landscape barriers for foraging bumblebees? Landscape scale experiments with Bombus terrest a rig sg. and Bombus pascuorum (Hymenoptera, Apidae). Biol. Conserv 116 . , 111–118, doi:10.1016/S0006-3207(03)00182-4 (2004). 40. Hampe, A. Fecundity limits in Frangula aln (R ushamnaceae) relict populations at the species’ southern range margin. 143 Oe, cologia 377–386, doi:10.1007/s00442-004-1811-0 (2005). 41. Latch, E. K., Dharmarajan, G., Glaubitz, J. C. & Rhodes, O. E. Relative performance of Bayesian clustering software for inferring population substructure and individual assignment at low levels of population differentiation. C7 ons , 295–302, erv. Gen et. doi:10.1007/s10592-005-9098-1 (2006). Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 7 www.nature.com/scientificreports/ 42. Hobbs, R. Managing plant populations in fragmented landscapes: restoration or gardening? Austr a 55 lia , 371–374, n J. Bot. doi:10.1071/BT06088 (2007). 43. Vedel, H. & Lange, J. Trees and Bushes in Wood and Hedgerows. Methuen and Co. Ltd., London, UK (1960). 44. Godwin, H. Frangula alnus Miller (Rhamnus frangul L.). a J. Ecol. 31, 77–92, doi:10.2307/2256793 (1943). 45. Morton, De. t al. Final Report for LCM2007 - the new UK Land Cover Map. Countryside Survey Technical Report No. 11/07 NERC/ Centre for Ecology & Hydrology: 112 (2011). 46. Doyle, J. J. & Doyle, J. L. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phy19 toc, 11–15 (1987). hem. Bull. 47. Grivet, D., Heinze, B., Vendramin, G. G. & Petit, R. J. Genome walking with consensus primers: application to the large single copy region of chloroplast DNA. Mol. Ecol. No 1t, 345–349, do es i:10.1046/j.1471-8278.2001.00107.x (2001). 48. Provan, J., Murphy, S. & Maggs, C. A. Universal plastid primers for Chlorophyta and Rhodophyta. Eur. J 39 . P , 43–50, do hycol. i:10. 1080/09670260310001636668 (2004). 49. Raymond, M. & Rousset, F. Genepop (version 1.2): population genetic software for exact tests and ecumenicism. 86 J. H , ered. 248–249, doi:10.1093/oxfordjournals.jhered.a111573 (1995). 50. Excoe ffi r, L. & Lischer, H. E. L. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol. Ecol. Resources10 , 564–567, doi:10.1111/men.2010.10.issue-3 (2010). 51. Goudet, J. FSTAT, A program to estimate and test gene diversities and fixation indices (Version 2.9.3.) http://www2.unil.ch/popgen/ softwares/fstat.htm Accessed October 2016 (2001). 52. Piry, A., Luikart, G. & Cornuet, J. M. Bottleneck: a computer program for detecting recent reductions in the ee ff ctive population size using allele frequency data. J. Her 90 e, 502–503, do d. i:10.1093/jhered/90.4.502 (1999). 53. Weir, B. S. & Cockerham, C. C. Estimating F-statistics for the analysis of population structur 38 e. , 1358–1370, Evolution doi:10.2307/2408641 (1984). 54. Excoffier, L., Smouse, P. E. & Quattro, J. M. Analysis of molecular variance inferred from metric distances among DNA haplotypes - application to human mitochondrial DNA restriction data. 131 Ge, 479–491 (1992). netics 55. Corander, J., Waldmann, P. & Sillanpää, M. J. Bayesian analysis of genetic differentiation between populatio 163 ns. , 367–374 Genetics (2003). 56. Hardy, O. J. & Vekemans, X. SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels. Mol. Ecol. No2 te , 618–620, do s i:10.1046/j.1471-8286.2002.00305.x (2002). 57. Loiselle, B. A., Sork, V. L., Nason, J. & Graham, C. Spatial genetic structure of a tropical understorey shrub, Psych otria officinalis (Rubiaceae). Am. J. Bot. 82, 1420–1425, doi:10.2307/2445869 (1995). Acknowledgements We are grateful to Keith Stanfield for providing helpful information on the history of the Peatlands Park population of . F alnus and to Gemma Beatty for assistance in the lab. Rob Paxton, Niall McKeown and two anonymous Referees provided valuable comments on the manuscript. This project was funded by the Northern Ireland Environment Agency (NIEA) through the Natural Heritage Research Partnership (NHRP) with Quercus, Queen’s University Belfast (QUB). Thanks to Peter McEvoy for conducting field surveys and Georgina Thurgate, Kathryn Turner and Tommy McDermott for assisting leaf sampling. IBERS receives strategic funding from the Biotechnology and Biological Sciences Research Council (BBSRC). Author Contributions S.J.P. and J.P. conceived the study. S.J.P. collected the samples. C.M.V.F and C.R.B. carried out the labwork. C.M.V.F. and J.P. carried out the data analyses and drae ft d the manuscript. All authors reviewed and commented on the manuscript. Additional Information Supplementary information accompanies this paper at doi:10.1038/s41598-017-03166-1 Competing Interests: The authors declare that they have no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2017 Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 8 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Scientific Reports Springer Journals

Low genetic diversity and potential inbreeding in an isolated population of alder buckthorn (Frangula alnus) following a founder effect

Free
8 pages

Loading next page...
 
/lp/springer_journal/low-genetic-diversity-and-potential-inbreeding-in-an-isolated-zqIWWeac60
Publisher
Springer Journals
Copyright
Copyright © 2017 by The Author(s)
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2045-2322
D.O.I.
10.1038/s41598-017-03166-1
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/scientificreports OPEN Low genetic diversity and potential inbreeding in an isolated population of alder buckthorn (Frangula alnus) Received: 18 November 2016 following a founder effect Accepted: 25 April 2017 Published: xx xx xxxx 1 1 1,3 2 Caroline M. V. Finlay , Caroline R. Bradley , S. Jane Preston & Jim Provan Alder buckthorn (Frangula alnus) is one of Ireland’s rarest tree species, and in Northern Ireland the species is now restricted to a single population in Peatlands Park, Co. Armagh numbering ca. 140 mature trees. Genotyping of 95% of the trees at nine nuclear microsatellite loci revealed that levels of genetic diversity within this population were generally lower than those reported from larger populations in Spain. Analysis of six chloroplast microsatellite loci revealed no variation. The level of F was significantly higher than that in the Spanish populations, as well as in other populations across IS Europe, potentially indicating inbreeding. Spatial autocorrelation analysis indicated some evidence of fine-scale genetic structuring, most likely due to limited seed dispersal, but the overall level of differentiation between subpopulations was low, indicating high levels of gene flow, probably due to cross-pollination by bees. Our results are consistent with a gradual population expansion from a limited number of individuals. We suggest that more immediate conservation efforts might be best focused on ensuring suitable habitat for the continued recovery of this isolated population. Populations of endangered or threatened species tend to be small and/or isolated and are thus particularly vul- nerable to stochastic processes. These problems are further exacerbated at the genetic level, where the increased 1, 2 effects of genetic drift and potential for inbreeding can lead to low levels of genetic va . Thir s c iaa tn b ion e fur - ther compounded if such populations have been founded by a limited number of indi , v sin idu ce al g senetically 4, 5 depauperate populations tend to have reduced evolutionary potential, which can increase the risk o . f extinction Where populations are fragmented, as is often the case in threatened taxa, reduced levels of gene flow between fragments can also aggravate the problems associated with limited genetic diversity, as there is less scope for 6, 7 immigration of alleles to counter the effects of dr . Co ift nsequently, knowledge of the levels and patterns of genetic diversity in populations of threatened species are vital to the formation of well-informe-d, ee ff ctive con 8, 9 servation plans. Frangula alnus (alder buckthorn) is one of Ireland’s rarest tree species. Although widespread in temperate Europe, the species has a very limited and fragmented distribution in Ireland, where it has been in serious decline 10, 11 over the last few decades as a result of drainage of its preferred bogland habitat for alterna (Fig tive l . 1 a ). n d use In Northern Ireland, recent surveys suggested t . ah lnu at s i Fs restricted to the southern shores of Lough Neagh. Although there are records of t . a hle nu Fs previously occurring on the northern side of the Lough, as well as a single tree in Drumawhey Bog, County Dow , t n hese are now extinct, the former natural woodland having been replaced by a broadleaf plantat . The p ion resent-day surviving population has been part of Annagarriff Nature Reserve in Peatlands Park, County Tyrone since 1978, and the species is protected under the Wildlife (NI) Order (1985) and is a Priority Species for Conservation Action. The history of this population, which curre - ntly num bers ca. 140 individuals (see Methods and Fig 1), i .  s not well-documented. The earliest records mention “Twenty bushes on the NE margin of Annaghgarriff [sic] … before 1934” , and census numbers appear to have remained low for many years, with a record of “about 30 young plants” around 1987–89 . Since then, there has been a gradual increase in numbers to those found today, possibly due to removal of rhododendron from the area, but 1 2 School of Biological Sciences, Queen’s University Belfast, Belfast, BT9 7BL, UK. Institute of Biological, Environmental and Rural Sciences, Aberystwyth University, Aberystwyth, SY23 3DA, UK. Present address: ATECNI Environmental Consultancy, 31 Castlewellan Road, Banbridge, BT32 4JQ, UK. Correspondence and requests for materials should be addressed to J.P. (email: J.Provan@aber.ac.uk) Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 1 www.nature.com/scientificreports/ Figure 1. Location of the Peatlands Park population of Frangul in C a alno u. A s rmagh, Northern Ireland. Zoomed area shows the distribution o alf nu F. s in Peatlands Park, with the five fragments sampled (labeled A–E), mapped using ArcMap 10. The coloured area indicates a designated Special Area for Conservation (SAC). Land classes were taken from the CEH Land Cover Map (NERC/Centre for Ecology & Hydrology). DB – Drumawhey Bog (extinct population referred to in text). it is not known whether the original trees were remnants of a once larger population, or whether numbers have always been low due to an initial founder effect. In recent years, the use of polymorphic microsatel lite markers has allowed the testing of whether populations have gone through a bottleneck, based on theoretical expectations 15–18 under mutation-drift equilibrium at a single point in tim . C eonsequently, the aim of the present study was to determine the levels of and patterns of genetic diversity in the remaining population to discover (1) whether there is any evidence for a genetic bottleneck, (2) if the establishment from a relatively limited number of individuals has been accompanied by a degree of inbreeding, and (3) whether there is any significant genetic substructuring 19, 20 within the population. In addition,. a as lnu F s is considered an invasive pest species in many countri , o esur findings could also shed light on the genetic demography of this recently expanded population with respect to similar invasive populations. Results Current distribution of F. alnus in Northern Ireland. Surveys of sites where . a Flnus had been recorded previously found that the species is now restricted to a single location; Peatlands Park, Co. Armagh. The sole remaining population exists as five discrete clusters of plants numbering between 3–98 individuals, each sepa- rated by between 100–1,30 m 0 (Fig. 1; FigureS   1, Supplementary Material). In total, ther.e 1 a 4r 0 m e ca ature trees. e o Th nly other population recorded, at Drumawhey Bog on the northern edge of Strangford Lough, now appears to have been extirpated. According to the CEH Land Class Map 2007 and site visits, Subpopulation A resides in a suburban area and is the only subpopulation outside the SAC of Peatlands Par 2), S k (Fig ubp . opulation B resides in grass-dominated bog, Subpopulation C occurs on the edge of deciduous woodland, Subpopulation D is split between grass-dominated bog, heather-dominated bog and scrub, and Subpopulation E occurs on the boundary between a coniferous and deciduous woodland. The majority of the trees in the population occur in bog (40.1%), divided between grass dominated bog (17.4%) and heather dominated bog (22.7%). The next most common habitat type for this population is scrub (33.3%), then deciduous woodland (16.7%), suburban areas (8.3%) and the least com- mon habitat type these trees occur in is coniferous woodland (1.5%). Levels of genetic diversity. No evidence of linkage disequilibrium was detected between the nine loci studied. Genotypes were obtained for 132 of the 139 trees (95%). Allele frequencies by locus and subpopulation are given in Table  S1, Supplementary Material. Within-subpopulation levels of genetic diversity averaged across loci are given in Tab1 le  and ranged from 0.222 (Subpopulation E) to 0.404 (Subpopulation B) for observed heterozygosity (H mean = 0.314) and from 0.331 (Subpopulation C) to 0.423 (Subpopulation D) for expected heterozygosity (H mean = 0.387). Mean inbreeding coefficients ( F ) across loci (Tab1 le ) ra   nged from − 0.022 E IS Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 2 www.nature.com/scientificreports/ Figure 2. Correlogram of autocorrelation coefficient (θ; y-axis) plotted against distance (x-axis). 95% confidence intervals are indicated by dashed red lines. Subpopulation N H H F O E IS A 11 0.305 0.39 0.226** B 6 0.404 0.422 0.050 NS C 21 0.340 0.331 −0.022 NS D 91 0.299 0.423 0.293** E 3 0.222 0.372 0.625** Mean 132 0.314 0.387 0.234 Total 132 0.308 0.411 0.251** Table 1. Diversity statistics. N – number of samples; – o H bserved heterozygosity; – exp H ected O E heterozygosity F; – inbreeding coefficient. Significance of F : *P < 0.05; **P < 0.01; ***P < 0.001; NS – Not IS IS Significant. (Subpopulation C) to 0.625 (Subpopulation E; m = e a 0.234). n Three of the subpopulations had F values signif- IS icantly greater than zero. Diversity values and inbreeding coefficients calculated for Subpopulation E should be treated with some caution, as this fragment only contained three trees. Treating the five subpopulations as a single population gave values of 0.308, 0.411 and 0.251 H for , H and F respectively. Values of summary statistics by O E IS locus and subpopulation are given in T S2 ab , S le  upplementary Material. No evidence of a genetic bottleneck was detected under any of the three mutation models, with two of the nine nuclear loci studied showing a heterozy- gote excess under all three mutation models (T 2). able  The Peatlands Park population exhibited significantly lower levels of genetic diversity than two of the three Spanish populations based on the loci analyzed in the present s = tud 0.411 y (H vs. H = 0.608 for Ajibe E E [Mann-Whitney test, = z −1.99, P = 0.047]; H = 0.603 for Medio [Mann-Whitney tes t, = − z 1.99, P = 0.047]; (2) E (2) H = 0.470 for Puerto Oscuro [Mann-Whitney tes =t, −z 0.98, P = 0.327]; Fig. 3a). The mean value of F for E (2) IS the Peatlands Park population (0.251) was significantly higher than those from the three Spanish populations (F = 0.015 for Ajibe [Mann-Whitney test, = z 2.12, P = 0.034]; F = −0.027 for Medio [Mann-Whitney test, IS (2) IS z = 2.03, P = 0.042]; F = −0.070 for Puerto Oscuro [Mann-Whitney tes =t, 2.30, z P = 0.021]; Fig. 3b). (2) IS (2) No diversity was observed for the chloroplast microsatellites, with all six loci being monomorphic. Levels of genetic differentiation and spatial genetic structuring. Differentiation among subpopu- lations calculated from the analysis of molecular variance (AMO 3) V wa A; s TΦ able  = 0.0322 (p = 0.009). e Th ST BAPS analysis indicated that the first four subpopulations belonged to the same genetic cluster, with the fifth (Subpopulation E) being genetically distinct (Figur S2, Su e  pplementary Material). The spatial autocorrelation analysis did, however, suggest that there may be some n fi e-scale spatial genetic structuring within Subpopulation D, with the autocorrelation coefficient being significantly higher than zero at the smallest scale ( 2). <50 m; Fig.  Discussion e g Th enetic analysis of the sole remaining population of Frangula in N alnusorthern Ireland carried out in the pres- ent study suggests that establishment from a limited number of individuals has led to limited level -s of genetic var iation, accompanied by potential inbreeding during the recent expansion in census population size. A comparison with results from a study o . a n lnu F s in Spain suggests that the Peatlands Park population has significantly lower levels of genetic diversity than two of the three Spanish populations. The complete lack of genetic variation in the six chloroplast microsatellite loci studied is also consistent with a founder ee ff ct. Although comparable data for these markers are not available for the Spanish populations, chloroplast microsatellites represent the most variable regions of the chloroplast geno , m an ed have also been shown to be monomorphic in a previously well-documented population bottleneck in Torrey p . Thin e o ebserved differences in levels of diversity between Northern Ireland and Spain could also reflect longer-term historical factors, since the Spanish populations most likely represent ref- ugial populations during the last glaciation, and that the low levels observed in Northern Ireland are consistent Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 3 www.nature.com/scientificreports/ Locus B101 A110 B7 A104 B106 B4 A7 A3 B9 Empirical data   Sample size (haploid genomes) 262 244 264 264 262 230 264 256 258   Heterozygosity observed () H 0.482 0.713 0.067 0.269 0.407 0.348 0.593 0.244 0.578   Number of alleles observed ( ) k 3 5 5 8 7 8 7 3 7 IAM   Average heterozygosity ()H 0.276 0.453 0.447 0.610 0.566 0.616 0.600 0.282 0.609   Standard deviation (SD) 0.188 0.182 0.180 0.141 0.157 0.136 0.145 0.189 0.150   Standard deviate ([H − H ]/SD) 1.097 1.430 −2.113 −2.424 −1.010 −1.971 −0.046 −0.204 −0.210 O E   Probability H ( > H ) 0.206 0.037 0.021 0.026 0.167 0.053 0.396 0.471 0.330 TPM   Average heterozygosity ()H 0.396 0.598 0.602 0.745 0.710 0.748 0.740 0.389 0.741   Standard deviation (SD) 0.166 0.115 0.118 0.070 0.077 0.065 0.071 0.161 0.074   Standard deviate ([H − H ]/SD) 0.517 0.991 −4.516 −6.83 −3.966 −6.116 −2.064 −0.901 −2.218 O E   Probability H ( > H ) 0.394 0.121 0.000 0.000 0.006 0.001 0.042 0.218 0.033 SMM   Average heterozygosity ()H 0.438 0.649 0.649 0.786 0.751 0.785 0.785 0.435 0.783   Standard deviation (SD) 0.137 0.089 0.093 0.047 0.058 0.048 0.048 0.144 0.052   Standard deviate ([H − H ]/SD) 0.324 0.712 −6.268 −10.89 −5.902 −9.128 −4.206 −1.33 −3.968 O E   Probability H ( > H ) 0.471 0.264 0.000 0.000 0.000 0.000 0.004 0.128 0.009 Table 2. Results of the Bottleneck analysis. Values in bold indicate heterozygote excess. Values in italics indicate heterozygote deficiency. Variance Source of variation d.f components % variation Fixation index Between subpopulations 4 0.05651 3.22 Φ = 0.0322** ST Within subpopulations 259 1.70113 96.78 Table 3. Analysis of molecular variance (AMOVA). Significance o : f ** ΦP < 0.01. ST Figure 3. Boxplots showing values of ( H a) , and (b) F in the Peatlands Park population analyzed in the E IS present study and three Spanish populations analyzed in Riguier . (2009). o et al 24, 25 with the founder effects associated with postglacial recolonization (“southern richness and norther . n purity”) Interestingly, though, a further study on populations from Italy, France, Belgium and Sweden using 186 single nucle- otide polymorphisms (SNPs) showed no decrease in genetic diversity with l.atitude Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 4 www.nature.com/scientificreports/ Despite the low levels of genetic variation, the Wilcoxon test for heterozygote excess did not indicate the occurrence of a bottleneck. Seven of the nine loci indicated a heterozygote deficiency, which could be indicative of population stability or constant gr. o A wt lth hough there are no records predating 1934, evidence suggests that the population remained relatively stable from this time until the end of the 1980s, after which numbers 10, 13 increased to the current censc u as . 14 of 0 individuals . It may be unwise to take this at face value, though, since the power of the test is affected by several parameters, including number of loci used, number of generations since the bottleneck, the length of the bottleneck itself, and the magnitude of reduction in effective population size. Immigration, another potentially confounding factor, can most likely be ruled out, since the closest known F. alnus populations to that in Peatlands Park are fo . 125 und kca m away in Co. Westmeath, Ireland. Likewise, the mutation model is unlikely to have an effect, since the results were broadly consistent across the three models ana- lyzed, and changing the parameters of the two-phase model from 90% single-stepwise mutations to 70% also had no effect. The generation time of F. alnus has been estimated to be between 5 years in Central Europe to around 20 years in the South , so the maximum number of generations of growth is likely to be much less than 10, although the period of apparent population stability preceding this could be upwards of 10 generations. These time scales may limit the power of the heterozygote excess test to detect any possible bottleneck, but this may be balanced to some extent by the fact that, based on historical records, the effective populat) io an er ft size the (N bottleneck is unlikely to be any higher tha. 20 n ca . Removal of rhododendron from areas of the park since the designation of the Annagarriff Nature Reserve may have facilitated the increase in numbe.r a sl o nu f s F (Keith Stanfield, personal communication), but this has apparently been accompanied by a degree of inbreeding, based o va n lues. S F uch inbreeding is most likely a IS result of the relatively limited genetic base of the population, particularly compared to the other populations of F. alnus. The mean value of F for the Peatlands Park population is significantly higher than those from the three IS Spanish populations studied previou. I slt i y s also far higher than the range of values report a e lnu d in s pF opu . - lations across Europe based on SNPs ( =F −0.107–0.088, mean = −0.015) . Reproductive dominance by a rel- IS 28, 29 atively low number of highly fecund individuals within a population could give rise to biparental in , breeding but since F . alnus possesses a self-incompatibility mechanism, as indicated by crossing s , t tudies his would be dependent on the diversity o a f l Sleles in the population. Previous studies have indicated a breakdown in 31, 32 self-incompatibility following population bottlen , e bc u ks t such a possibility would have to be tested via con- trolled pollination experiments. The absence of high levels of spatial structuring of genetic variation in the Peatlands Park pop. ulation of F alnus indicates a general lack of barriers to dispersal, which is unsurprising given the small spatial scale relative to the potential dispersal distances of pollinators. The overall level of population differen = ti 0.032) ation ( wa Φ s ST lower than the average values for outcrossing species with seeds dispersed by gravity (0.152) or ingestion (0.200) quoted by Hamrick & God, a t s well as the average value for biparentally inherited markers in angiosperms (0.184) quoted by Petit et al . . The spatial autocorrelation analysis suggested some fine-scale genetic structure 27, 35 in Subpopulation D, probably due to limited seed dispersal. Se. ea dls o nus f c Fan be dispersed by birds , but in the closely related (congeneric in some classificat R io ha nm s)nus cathartica, it has been shown that 90% of fruits fall beneath the mother t . Thir s s ee cenario could at least in part be responsible for the hig obs h er F ved in IS Subpopulation D, by way of a Wahlund effect. e a Th rea surrounding Peatlands Park has a number of apple orchards with large numbers of wild bees in the vicinity, and the high incidence of fruits in the Peatland Fs . a Pla nu rk s population suggests substantial levels of cross-pollination. Bees have large foraging ranges, even across sub-optimal habitats, and are likely to be important 38, 39 in maintaining the connectivity observed between the fragments in the presen.t I n st c uo dn ytrast to the levels of fruiting found in Peatlands Park, where almost all trees had multiple fruits, a previous study on reproduction in populations of F a. lnus from Cádiz, Spain found that only 2.8% of open-pollinated flowers set f . A rui no tther study on southern range edge populations also indicated that the majority of ovule losses were due to cross-pollen limitation and extensive geitonogamy, and that seed output in the populations was limited to a fe . w large trees e hig Th h percentage of fruiting trees observed in Peatlands Park suggests that this population is more similar to those found in Central European populations, which have a shorter generation time and higher levels of fruit production compared to Southern Iberian popula. I tio f t n hse observed high F values reflect some degree of IS inbreeding, there has been no apparent impact on fitness, at least in terms of fruit production. Although the BAPS analysis assigned Subpopulation E to a separate genetic cluster from the other four sub- populations, this should be taken with some degree of caution for several reasons. Firstly, the subpopulation num- bers only three individuals, and thus allelic frequencies will be skewed. Secondly, it has been shown previously that the BAPS algorithm tends to over-estimate the true number of genetic clusters presen . Fin t in t all hy e d , ata only a single private allele is present in Subpopulation E, with the majority of genetic differentiation being due to the aforementioned differences in allele frequencies at several of the loci studied. Whilst it is true that knowledge of the evolutionary dynamics of natural populations with respect to demog- raphy and gene flow allows the management of threatened plant populations to go beyond simple “conservation gardening” , in the case of . F alnus in Northern Ireland it would appear that edaphic and ecological factors are of greater importance. The land classes that are correlated with occurrence of alder buckthorn include bog, conifer and pastures, and this is consistent with the literature that claims this species likes open lowland areas with moist, 43, 44 fertile soils . Nevertheless, there are large areas of apparently suitable habitat in Peatlands Park which have not been colonized during the recent expansion of the species in the nature reserve. This would suggest that although the low levels of genetic diversity and potential inbreeding revealed in the population are of some concern with respect to evolutionary potential, more immediate conservation efforts might be better focused o- n ensuring suit able habitat for the continued recovery of this isolated population. Additionally, consideration should be given to possible supplementation using material from the populations in the Republic of Ireland, which are larger and may harbor additional genetic diversity. Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 5 www.nature.com/scientificreports/ Methods Study species. Frangula alnus Miller (syn. Rhamnus frangul L.) i a s a small tree or shrub found across tem- perate Europe from northern Scandinavia and Russia to the Mediterranean, where it is compara . t In ively scarce Britain and Ireland, the species tends to be found in damp, boggy habitats, but can also colonize drier ground out- 19, 20, 35 side of its native range, particularly in North America where the species is now considered an invasi . ve alien Reproduction in F. alnus is exclusively sexua. lFlowers are hermaphroditic and well-adapted to insect polli- nation as they have nectar, colour and odour. They are generalised entomophiles, with orders Hymenoptera, Diptera and Coleoptera providing the main pollinators, and occasional pollination by L . C ep oido ntr p o tl era led pollination studies revealed almost no selfing or geitonogamy, indicating the existence of self-incompatibility mechanisms, although occasional selfing in the absence of insect pollinators has been. Fi rep eld or s tt eu ddies 30, 40 suggest that limited fruit initiation is primarily due to low levels of cross-po. S llin eed as a tio rn e dispersed 27, 35, 37 either by gravity or via ingestion by birds . Surveys and sampling. Surveys were carried out across Northern Ireland in the summer of 2007 at sites where F. alnus had been found previously based on records from the National Biodiversity Network (NBN) Gateway (http://data.nbn.org.uk) and the Centre for Environmental Data and Recording (CEDaR: http://www.habitas.org. uk/cedar/). Frangula alnus is restricted to the southern shores of Lough Neagh, where it exists as a fragmented pop- ulation from a single location in Annagarriff Nature Reserve in Peatlands Park, County T 1). A yro l n l m e (Fig atur .  e trees were numerically tagged with metal tags, GPS coordinates recorded, and a leaf sample from each obtained for genetic analysis. The GPS coordinates were loaded into ESRI ArcMap 10 and plotted on top of a CEH Land Cover Map 2007 to discover the main land class that the trees occupy. DNA was extracted from leaf material using the CTAB method . Genotypes were successfully obtained for 132 of the 139 trees sampled. Nuclear microsatellite analysis. We attempted to genotype all mature plants for sixteen previously described microsatellite loci f . ao ln r uF s . Of these, four (FaA103, FaA125, FaA8 and FaB8) could not be con- sistently amplified, and three (FaA12, FaB102 and FaA116) were monomorphic, leaving nine polymorphic loci: FaB101, FaA110, FaB7, FaA104, FaB106, FaB4, FaA7, FaA3 and FaB9. All reactions were carried out on a MWG Primus thermal cycler. PCR was carried out in a total volμ um l co e o nf 10 tainin g 100 ng g enomic DNA, 10p mol of dye-labelled forward primer (HEX), 1 pmo l of tailed forward primer p, 10 mo l reverse primer, 1x PCR reaction buffer, 200 μM each dNTP, 2.5mM MgC l and 0.25U G oTaq Flexi DNA polymerase (Promega). PCR conditions were as described previousl . G y enotyping was carried out on an AB3730xl capillary genotyping system. Allele sizes were scored using the GeneMapper software package (V5.0; Applied Biosystems) and LIZ-500 size stand- ards, and were checked by comparison with previously sized control samples. Chloroplast microsatellite analysis. Chloroplast microsatellite markers were developed by identifying mononucleotide regions of ten or more repeats in p F.a a rlt ni ua sl chloroplast genome sequences either from GenBank or in regions amplified and sequenced de n u ovo sing universal chloroplast prim . Ther e tr s n T-trnF, atpH-atpI, atpI-rpoC2, rps18-clpp, psbC-trnS, trnS-trnfM and rpl16-rps3 (UCP6) regions were amplified and 47 48 sequenced using the primers described in Grivet . e a t n ald Provan et al . . Species-specific primers were designed using the Primer3 program to amplify six chloroplast microsatelli S3 t, S es (T upp alem ble entary Material). Primers were tailed as described previously, and PCR was carried o μ ul t r in e 1 ac 0tions as described previously using the following conditions: initial denaturatio °n C a fo tr 9 3 4 min followed by 30 cycles of denaturation ° a Ct f 9 or 4 30 s, annealing at 56 °C f or 30s, ext ension at 72 °C f or 30s a nd a final extension at 72 °C f or 5min. G enotyping was carried out as described previously. Data analysis. Subpopulations of the Peatlands Park population were classed as distinct groups of trees with no other trees of this species growing within 100 m bet ween groups, resulting in five distinct subpopulations (Fig 1). .  GenePop V3.4 was used to test for linkage disequilibrium between nuclear loci. To estimate genetic diversity within the population, levels of observe) a d (H nd expected (H ) heterozygosity, and fixation indices ( ) w F ere O E IS 50 51 calculated using t Ar he lequin (V3.5.1.2) and Fstat (V2.9.3.2) sowa ft re packages respectively. Significance o F f IS was determined by 10,000 randomization steps. To test for the occurrence of a genetic bottleneck, the Wilcoxon test for heterozygote excess was performed under the infinite alleles model (IAM), the stepwise mutation model (SMM) and a two-phase model (TPM) incorporating 90% single-stepwise mutations using the program Bo ttleneck (V1.2) . The Wilcoxon test was used as it is recommended for a relatively low number of loci. To compare genetic diversity between the population analyzed in the present study, and those from three Spanish populations previously analyzed using the same nuclear micros, m ate el ali n va teslues for H and F E IS were calculated over the twelve loci successfully amplified in the present study, including the three loci which were monomorphic. Mann-Whitney tests were carried out to assess the significance of differen an ces in d F .H E IS The level of genetic differentiation between the five fragments was estimated usin , w g Φ hich gives an ana- ST 53 54 logue of F calculated within the analysis of molecular variance (AMOVA) fra u m sin ew g oAr rklequin. To ST identify possible spatial patterns of gene flow, the software package BAPS (V5) was used to identify clusters of genetically similar subpopulations using a Bayesian approach. Ten replicates were run for all possible values of the maximum number of cluster) u s (K p to K = 5, the number of subpopulations, with a burn-in period of 10 000 iterations followed by 50 000 iterations. Multiple independent runs always gave the same outcome. To further identify possible spatial patterns of gene flow, spatial autocorrelation analysis was carried out for Subpopulation 56 57 D, the largest of the five subpopulations, using SPAG (V1.4) eDi . Mean coancestry coefficients (θ ) between xy pairs of individuals were calculated a m di t 25 st ance class intervals, and plotted as a correlogram, with 95% con- fidence intervals calculated from 1,000 permutations of individuals within each distance class, and for estimates of θ using 1,000 permutations. xy Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 6 www.nature.com/scientificreports/ References 1. Hedrick, P. W. & Kalinowski, S. T. Inbreeding depression and conservation biology. Ann. Rev. Ec 31 ol. S , 139–162, do yst. i:10.1146/ annurev.ecolsys.31.1.139 (2000). 2. Cole, T. C. Genetic variation in rare and common plants. Ann. Rev. Ecol. Evo34 l. S , 213–237, do yst. i:10.1146/annurev. ecolsys.34.030102.151717 (2003). 3. Nei, M., Maruyama, T. & Chakraborty, R. The bottleneck effect and genetic variability in population 29 s. , 1–10, Evolut ion doi:10.2307/2407137 (1975). 4. Hansson, B. & Westerberg, I. On the correlation between heterozygosity and fitness in natural populatio 11 ns. , 2467–2474, Mol. Ecol. doi:10.1046/j.1365-294X.2002.01644.x (2002). 5. Frankham, R. Genetics and extinction. Biol. Cons 126 erv, 131–140, do . i:10.1016/j.biocon.2005.05.002 (2005). 6. Wright, S. Isolation by distance. Genet 28 ic , 114–138 (1943). s 7. Young, A., Boyle, T. & Brown, T. The population genetic consequences of habitat fragmentation for plants. Trends Ec 11o , l. Evol. 413–418, doi:10.1016/0169-5347(96)10045-8 (1996). 8. Haig, S. M. Molecular contributions to conservation. 79 Ec, 413–412 (1998). ology 9. Hedrick, P. W. Recent developments in conservation genetics. Forest Ecol. Ma197 nagem. , 3–19, do i:10.1016/j.foreco.2004.05.002 (2004). 10. Harron, J., Rushton, B. S. & Newbould, P. J. Flora of Lough Neagh. Ir. N 22 at, 1–270 (1986). . J. 11. Rich, T. C. G., Beesley, S. & Goodwillie, R. Changes in the vascular plant flora or Ireland between pre-1960 and 1987-1988, the BSBI Monitoring Scheme. Ir. Nat. J 26 . , 333–350 (2001). 12. Rippey, I. Alder buckthorn Frangula alnus Miller and other scarce plants in Co. Down (H38). 23, 223–224 (1990). Ir. Nat. J. 13. Rippey, I. Some new county records and other botanical notes. Ir 23 . N , 216–218 (1990). at. J. 14. Powell, W., Machray, G. C. & Provan, J. Polymorphism revealed by simple sequence repeats. Trends P 1l, 215–222, ant Sci. doi:10.1016/S1360-1385(96)86898-0 (1996). 15. Cornuet, J. M. & Luikart, G. Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genet i144 cs , 2001–2014 (1996). 16. Luikart, G. & Cornuet, J. M. Empirical evaluation of a test for identifying recently bottlenecked populations from allele frequency data. Conserv. Biol. 12, 228–237, doi:10.1111/j.1523-1739.1998.96388.x (1998). 17. Garza, J. C. & Williamson, E. G. Detection of reduction in population size using data from microsatellite lo 10, 305–318, ci. Mol. Ec ol. doi:10.1046/j.1365-294x.2001.01190.x (2001). 18. Peery, M. Z. Reliability of genetic bottleneck tests for detecting recent population decli 21 ne,s 3 . 4 M 0o 3l– . E 34 c1 o8 l., doi:10.1111/ j.1365-294X.2012.05635.x (2012). 19. McClain, W R. hamnus frangula. In: Invasive Plants: Weeds of the Global Garden (eds Randall J. M. & Marinelli J.), Brooklyn Botanic Garden Publications, New York, USA (1996). 20. Cunard, C. & Lee, T. D. Is patience a virtue? Succession, light and the death of invasive glossy buckthorn (Fra ). ng Bi uo la l. a lnus Invasions 11, 577–586, doi:10.1007/s10530-008-9272-8 (2009). 21. Rigueiro, C., Arroyo, J. M., Rodriguez, R., Hampe, A. & Jordano, P. Isolation and characterization of 16 polymorphic microsatellite loci for Frangula alnu (R s hamnaceae). Mol. Ecol. Resource9 s , 986–989, doi:10.1111/men.2009.9.issue-3 (2009). 22. Provan, J., Powell, W. & Hollingsworth, P. M. Chloroplast microsatellites: new tools for studies in plant ecology and systematics. Trends Ecol. Evol. 16, 142–147, doi:10.1016/S0169-5347(00)02097-8 (2001). 23. Provan, J., Soranzo, N., Wilson, N. J., Goldstein, D. B. & Powell, W. A low mutation rate for chloroplast microsat e 153 lli, tes. Genetics 943–947 (1999). 24. Provan, J. & Bennett, K. D. Phylogeographic insights into cryptic glacial refugia. Trends Ec 23o , 564–571, do l. Evol. i:10.1016/j. tree.2008.06.010 (2008). 25. Hewitt, G. M. Post-glacial recolonization of EuropeaB ni b ol. io Jt . L a.i nnean Soc. 68, 87–112, doi:10.1111/bij.1999.68.issue-1-2 (1999). 26. de Kort, H., Vandepitte, K., Mergeay, J. & Honnay, O. Isolation, characterization and genotyping of single nucleotide polymorphisms in the non-model tree species Frangula aln (R uh s amnaceae). Conserv. Genet. Resource6 s , 267–269, doi:10.1007/s12686-013-0083-6 (2014). 27. Hampe, A. & Bairlein, F. Modified dispersal-related traits in disjunct populations of bird-dispersed Fra (R ngu hla amn alnace us ae): a result of its Quaternary distribution shifts? Ecogr 23 ap , 603–613, do hy i:10.1034/j.1600-0587.2000.230511.x (2000). 28. Aldrich, P. R. & Hamrick, J. L. Reproductive dominance of pasture trees in a fragmented tropical forest m os 281 aic. , Science 103–105, doi:10.1126/science.281.5373.103 (1998). 29. Cascante, A., Quesada, M. & Lobo, J. J. Effects of tropical forest fragmentation on the reproductive success and genetic structure of the tree Samanea saman . Conserv. Biol. 16, 137–147, doi:10.1046/j.1523-1739.2002.00317.x (2002). 30. Medan, D. Reproductive biology of Frangula aln (R ushamnaceae) in southern SpaP in. lant Syst. Evol. 193, 173–186, doi:10.1007/ BF00983549 (1994). 31. Reinartz, J. A. & Lee, D. H. Bottleneck induced dissolution of self-incompatibility and breeding systems consequences in Aster furcatus (Asteraceae). Am. J. Bo81 t. , 446–455, doi:10.2307/2445494 (1994). 32. Busch, J. The evolution of self-compatibility in geographically peripheral populations of Leavenworthia a (B laba ram ssic icaaceae). Am. J. Bot. 92, 1503–1512, doi:10.3732/ajb.92.9.1503 (2005). 33. Hamrick, J. L. & Godt, M. J. W. Effects of life history traits on genetic diversity in plant species. Phil. Trans. R. Soc. L 351 on , don B 1291–1298, doi:10.1098/rstb.1996.0112 (1996). 34. Petit, R. J. et a . C l omparative organization of chloroplast, mitochondrial and nuclear diversity in plant populat 14 io, ns. Mol. Ecol. 689–701, doi:10.1111/j.1365-294X.2004.02410.x (2005). 35. Hampe, A. The role of fruit diet within a temperate breeding bird community in southern Spain. B 48 ird , 116–123, Study doi:10.1080/00063650109461209 (2001). 36. Richardson, J. E., Fay, M. F., Cronk, Q. C. B., Bowman, D. & Chase, M. W. A phylogenetic analysis of Rhamnaceae using rbcL and trnL-F plastid DNA sequences. Am. J. B 87 ot, 1309–1324, do . i:10.2307/2656724 (2000). 37. Archibold, O. W., Brooks, D. & Delanoy, L. An investigation of the invasive shrub European buckthorn Rhamnu L., n s catha ea rr tica Saskatoon, Saskatchewan. Can. Field N 111 at. , 617–621 (1997). 38. Steffan-Dewenter, I. & Kuhn, A. Honeybee foraging in differentially structured landP sr c oa c. R pes. . Soc. London B 270, 569–575, doi:10.1098/rspb.2002.2292 (2003). 39. Kreyer, D., Oed, A., Walther-Hellwig, K. & Frankl, R. Are forests potential landscape barriers for foraging bumblebees? Landscape scale experiments with Bombus terrest a rig sg. and Bombus pascuorum (Hymenoptera, Apidae). Biol. Conserv 116 . , 111–118, doi:10.1016/S0006-3207(03)00182-4 (2004). 40. Hampe, A. Fecundity limits in Frangula aln (R ushamnaceae) relict populations at the species’ southern range margin. 143 Oe, cologia 377–386, doi:10.1007/s00442-004-1811-0 (2005). 41. Latch, E. K., Dharmarajan, G., Glaubitz, J. C. & Rhodes, O. E. Relative performance of Bayesian clustering software for inferring population substructure and individual assignment at low levels of population differentiation. C7 ons , 295–302, erv. Gen et. doi:10.1007/s10592-005-9098-1 (2006). Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 7 www.nature.com/scientificreports/ 42. Hobbs, R. Managing plant populations in fragmented landscapes: restoration or gardening? Austr a 55 lia , 371–374, n J. Bot. doi:10.1071/BT06088 (2007). 43. Vedel, H. & Lange, J. Trees and Bushes in Wood and Hedgerows. Methuen and Co. Ltd., London, UK (1960). 44. Godwin, H. Frangula alnus Miller (Rhamnus frangul L.). a J. Ecol. 31, 77–92, doi:10.2307/2256793 (1943). 45. Morton, De. t al. Final Report for LCM2007 - the new UK Land Cover Map. Countryside Survey Technical Report No. 11/07 NERC/ Centre for Ecology & Hydrology: 112 (2011). 46. Doyle, J. J. & Doyle, J. L. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phy19 toc, 11–15 (1987). hem. Bull. 47. Grivet, D., Heinze, B., Vendramin, G. G. & Petit, R. J. Genome walking with consensus primers: application to the large single copy region of chloroplast DNA. Mol. Ecol. No 1t, 345–349, do es i:10.1046/j.1471-8278.2001.00107.x (2001). 48. Provan, J., Murphy, S. & Maggs, C. A. Universal plastid primers for Chlorophyta and Rhodophyta. Eur. J 39 . P , 43–50, do hycol. i:10. 1080/09670260310001636668 (2004). 49. Raymond, M. & Rousset, F. Genepop (version 1.2): population genetic software for exact tests and ecumenicism. 86 J. H , ered. 248–249, doi:10.1093/oxfordjournals.jhered.a111573 (1995). 50. Excoe ffi r, L. & Lischer, H. E. L. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol. Ecol. Resources10 , 564–567, doi:10.1111/men.2010.10.issue-3 (2010). 51. Goudet, J. FSTAT, A program to estimate and test gene diversities and fixation indices (Version 2.9.3.) http://www2.unil.ch/popgen/ softwares/fstat.htm Accessed October 2016 (2001). 52. Piry, A., Luikart, G. & Cornuet, J. M. Bottleneck: a computer program for detecting recent reductions in the ee ff ctive population size using allele frequency data. J. Her 90 e, 502–503, do d. i:10.1093/jhered/90.4.502 (1999). 53. Weir, B. S. & Cockerham, C. C. Estimating F-statistics for the analysis of population structur 38 e. , 1358–1370, Evolution doi:10.2307/2408641 (1984). 54. Excoffier, L., Smouse, P. E. & Quattro, J. M. Analysis of molecular variance inferred from metric distances among DNA haplotypes - application to human mitochondrial DNA restriction data. 131 Ge, 479–491 (1992). netics 55. Corander, J., Waldmann, P. & Sillanpää, M. J. Bayesian analysis of genetic differentiation between populatio 163 ns. , 367–374 Genetics (2003). 56. Hardy, O. J. & Vekemans, X. SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels. Mol. Ecol. No2 te , 618–620, do s i:10.1046/j.1471-8286.2002.00305.x (2002). 57. Loiselle, B. A., Sork, V. L., Nason, J. & Graham, C. Spatial genetic structure of a tropical understorey shrub, Psych otria officinalis (Rubiaceae). Am. J. Bot. 82, 1420–1425, doi:10.2307/2445869 (1995). Acknowledgements We are grateful to Keith Stanfield for providing helpful information on the history of the Peatlands Park population of . F alnus and to Gemma Beatty for assistance in the lab. Rob Paxton, Niall McKeown and two anonymous Referees provided valuable comments on the manuscript. This project was funded by the Northern Ireland Environment Agency (NIEA) through the Natural Heritage Research Partnership (NHRP) with Quercus, Queen’s University Belfast (QUB). Thanks to Peter McEvoy for conducting field surveys and Georgina Thurgate, Kathryn Turner and Tommy McDermott for assisting leaf sampling. IBERS receives strategic funding from the Biotechnology and Biological Sciences Research Council (BBSRC). Author Contributions S.J.P. and J.P. conceived the study. S.J.P. collected the samples. C.M.V.F and C.R.B. carried out the labwork. C.M.V.F. and J.P. carried out the data analyses and drae ft d the manuscript. All authors reviewed and commented on the manuscript. Additional Information Supplementary information accompanies this paper at doi:10.1038/s41598-017-03166-1 Competing Interests: The authors declare that they have no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2017 Scientific Repo R ts | 7: 3010 | DOI:10.1038/s41598-017-03166-1 8

Journal

Scientific ReportsSpringer Journals

Published: Jun 7, 2017

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off