Cent. Eur. J. Biol. • 3(4) • 2008 • 442–450 DOI: 10.2478/s11535-008-0033-6 Central European Journal of Biology Genetic structure of the Anthyllis vulneraria L. s. l. species complex in Estonia based on AFLPs Research Article Egle Köster1*, Elena Bitocchi2, Roberto Papa2, Silvia Pihu1 1 Institute of Ecology and Earth Sciences, University of Tartu, 51005 Tartu, Estonia 2 Faculty of Agriculture, Agricultural University of Marche, 60131 Ancona, Italy Received 3 December 2007; Accepted 29 May 2008 Abstract: Anthyllis vulneraria L. (Fabaceae) s. lato includes many cryptic taxa, ranging from 25 to 60 subspecies according to different authors. The delimitation of intraspeciic taxa of A. vulneraria s. lato has always been complicated and inconsistent. Different data sets (multivariate analyses of morphological variation, allozymes, chloroplast SSRs and ITS) have not resolved the existing problem with distinguishing some subspecies. We used the ampliied fragment length polymorphism (AFLP) analysis to describe the differentiation in this species complex and to characterize variation on a geographic scale. Some correlation was found between genetic variability and geographic distribution (western-eastern directional variation), but AFLP data analysis did not reveal clear intraspeciic structure of the seven analysed taxa. The analysed specimens did not comprise groups correlated with the subspecies. Keywords: Anthyllis vulneraria • AFLP • Geographical pattern • Intraspeciic molecular variance © Versita Warsaw and Springer-Verlag Berlin Heidelberg. 1. Introduction The genus Anthyllis belongs to the tribe Loteae and family Fabaceae and is closely related to the genus Hymenocarpus [1-3]. The exact number of Anthyllis species is controversial and depends on the interpretation of their morphological-geographical boundaries with respect to active speciation and hybridisation [4]. The genus is considered to range from 25 [5] to 60 species [6]. Although some species in the genus are well deined and generally accepted, there are many cryptic forms that have been the subject of different interpretations. There are two schools of thought with respect to the taxonomy of the genus Anthyllis. The irst, most used in the area of the former Soviet Union, distinguishes numerous sibling species of Anthyllis vulneraria s. l. [6,7]. The second system, prevalent in most of Europe, recognises 18 European species [5,8-10] and some species of the irst school are classiied as subspecies and varieties of Anthyllis vulneraria s. l. Cullen [5] divided A. vulneraria s. l. into three major groups: subsp. vulneraria, subsp. maritima and subsp. polyphylla, which also includes all the seven taxa investigated in this paper. Anthyllis vulneraria L. s. l. occurs from the Volga River to England and from Northern Europe to the Mediterranean [11]. It has also been introduced into North America and New Zealand [12]. A few morphological characteristics in different keys and loras distinguish these taxa. Bicoloured rufous calyx teeth demarcate subsp. vulneraria, var. coccinea, A. x baltica and A. x colorata from the other four taxa, which have concolorous, green calyces. Another characteristic that is reasonably easy to detect is the disposition of hair on the stem and petiole. The subspecies polyphylla and A. x colorata have patent hairs on the stems and petioles, whereas the other taxa have appressed hairs [5,8,9,13-15]. The subspecies maritima can be distinguished from the other taxa by concolorous calyces, sericeous calyx pubescence and some inlorescences with a few lowers, which are sometimes not fully developed [13,14,16]. Long peduncles also characterize inlorescences of this species. The subspecies arenaria has well-developed * E-mail: egle.koster@ut.ee 442 Unauthenticated Download Date | 4/21/16 5:20 AM E. Köster et al. inlorescences that are sessile [14,15]. Branches of this species form an acute angle with the stem [16]. A. vulneraria var. coccinea is most readily distinguished from the other taxa by its red–coloured corollas [5,13-15]. Anthyllis x baltica has some undeveloped inlorescences on the axilla, as well as subsp. maritima [14,16]. Anthyllis vulneraria subsp. vulneraria has unbranched stems and mainly apical inlorescences [7,14,15]. There are also many infrequently used characteristics in other studies [5,13,17-19; Akulova, unpublished data]. These seven taxa can, according to the keys and loras, be well distinguished, yet individual plants of genus Anthyllis in natural stands are dificult to identify. The morphological variation and taxonomic continuum of the eight sibling species of A. vulneraria s. l. were analysed previously [20]. Most of the characteristics analysed were statistically signiicant in species discrimination. The results also showed that the Anthyllis species are morphologically rather indistinct in natural populations. Analysis supported the distinction of four groups of taxa within the species Anthyllis vulneraria s. l., which could be named as vulneraria, coccinea, maritima and macrocephala [20]. Kalinowski et al. [21,22] analysed the isoenzymatic variability of A. vulneraria, s. l. populations in Poland to determine whether there exists any differentiation between populations of inland and coastal areas. Different methods of multivariate statistical analysis all conirmed the differences between populations depending on geographic distance according to isoenzyme data. Molecular phylogeny of the genus was studied based on the sequences of the internal transcribed spacers ITS1 and ITS2 of the nuclear ribosomal DNA of ten Anthyllis species, including eleven subspecies of A. vulneraria and three subspecies of A. montana [23]. Additionally, the polymorphic chloroplast SSRs were used to quantify the genetic variation of Anthyllis. ITS sequences discriminate between some subspecies of A. vulneraria, but this genetic differentiation is inconsistent with taxonomic delimitation, based on morphological characters. cpSSRs showed some minor differences within A. vulneraria. These results suggest that the classiication of the subspecies of A. vulneraria should be revised according to their phylogenetic relationships [23]. Kropf et al. [24] investigated the ITS regions of nuclear ribosomal DNA by sequencing multiple accessions of Anthyllis montana L. and some closely related taxa. The ITS phylogeny implied a western Mediterranean origin followed by an eastward migration [24]. In addition, they analysed AFLP from 71 individuals of A. montana and revealed a major genetic (west/east) subdivision, probably caused by the massive glaciations of the Alps during the last glacial period [24]. Honnay et al. [25] studied habitat fragmentation effects on the population genetic structure of Anthyllis vulneraria in the Viroin Valley in southern Belgium. Their data show that the consequences of habitat fragmentation for genetic differentiation and diversity of A. vulneraria are moderately minor. These results can be conirmed by the fact that the historical seed exchange levels are quite high among fragments through the agency of grazing and roaming livestock [25]. The results of different molecular analysis of A. vulneraria s. lato are quite contentious [21-23]. Here we present a new molecular study of A. vulneraria s. l. In order to address questions of intraspeciic genetic variability of taxa and the role of geographical isolation, ampliied fragment length polymorphism (AFLP) markers were analysed. The study’s objectives were: (i) to investigate the level of AFLP variation between and within seven intraspeciic taxa of A. vulneraria s. l. in Estonia and (ii) to quantify the genetic differences of the taxa between four regions of contrasting edaphic and climatic conditions. 2. Experimental Procedures 2.1. Taxa, study sites and plant materials We investigated seven taxa belonging to A. vulneraria s. lato: (A. vulneraria subsp. arenaria Rupr., A. vulneraria var. coccinea L., A. vulneraria subsp. polyphylla (DC.) Nyman, A. vulneraria subsp. maritima (Schweigg) A. et G. and A. vulneraria subsp. vulneraria L., A. x colorata Juz. (A. vulneraria subsp. vulneraria x A. vulneraria subsp. polyphylla), A. x baltica Juz. (A. vulneraria subsp. vulneraria x A. vulneraria subsp. maritima) [5,9]. Anthyllis x baltica is considered to be endemic to the Baltic region and A. x colorata endemic to Estonia [6]. One more hybrid taxon A. x polyphylloides (A. vulneraria subsp. polyphylla x A. vulneraria subsp. arenaria) is found from Estonia [6], but it was not found during our collecting period. Estonian habitats are similar to those in Central and Southern Sweden, where several varieties of Anthyllis have been described and where populations of Anthyllis occur typically as hybrid complexes [26]. To analyse intraspeciic genetic variation of these taxa, 58 specimens from 4 regions in Estonia were collected and included in AFLP analyses (Table 1). Regions were deined to analyse the genetic differentiation on a geographic scale (Table 1,2). According to the loristic division of Europe, Estonia is divided into two main provinces: Western Estonia 443 Unauthenticated Download Date | 4/21/16 5:20 AM Genetic structure of the Anthyllis vulneraria L. s. l. species complex in Estonia based on AFLPs N° Sample N° taxon abbreviation A. vulneraria region 30 35 mari ssp. maritima East 1 1 cocc var. coccinea Islands 31 36 mari ssp. maritima East 2 3 vuln ssp. vulneraria Islands 32 37 aren ssp. arenaria East 3 4 vuln ssp. vulneraria Islands 33 40 aren ssp. arenaria West 4 5 cocc var. coccinea Islands 34 41 colo x colorata West 5 6 cocc var. coccinea Islands 35 42 colo x colorata West 6 8 vuln ssp. vulneraria Islands 36 43 balt x baltica West 7 9 mari ssp. maritima Islands 37 44 balt x baltica West 8 10 mari ssp. maritima Islands 38 45 mari ssp. maritima West 9 11 balt x baltica Islands 39 46 mari ssp. maritima West 10 12 balt x baltica Islands 40 47 balt x baltica West 11 13 balt x baltica Islands 41 48 balt x baltica West 12 14 balt x baltica Islands 42 49 cocc var. coccinea West 13 15 cocc var. coccinea Islands 43 50 cocc var. coccinea West 14 16 cocc var. coccinea Islands 44 51 vuln ssp. vulneraria West 15 17 vuln ssp. vulneraria Islands 45 52 vuln ssp. vulneraria West 16 18 vuln ssp. vulneraria Islands 46 53 colo x colorata North 17 20 vuln ssp. vulneraria Islands 47 54 colo x colorata North 18 21 vuln ssp. vulneraria Islands 48 55 colo x colorata North 19 23 cocc var. coccinea Islands 49 57 poly ssp. polyphylla North 20 24 cocc var. coccinea Islands 50 58 poly ssp. polyphylla North 21 25 aren ssp. arenaria Islands 51 59 colo x colorata North 22 26 aren ssp. arenaria Islands 52 60 colo x colorata North 23 27 poly ssp. polyphylla East 53 61 poly ssp. polyphylla North 24 28 poly ssp. polyphylla East 54 62 poly ssp. polyphylla North 25 29 aren ssp. arenaria East 55 63 poly ssp. polyphylla North 26 30 aren ssp. arenaria East 56 64 poly ssp. polyphylla North 27 32 colo x colorata East 57 65 colo x colorata North 28 33 vuln ssp. vulneraria East 58 66 colo x colorata North 29 34 vuln ssp. vulneraria East Table 1. Materials of Anthyllis vulneraria s. l. used in AFLP analysis. Habitat Bedrock Climate Isolation Floristic province North calcareous continental mainland Eastern European West calcareous marine mainland Middle European East sandstone continental mainland Eastern European Islands calcareous marine isolated Middle European Table 2. Characters describing four regions used in analysis [27,28]. belongs to the Middle European province and Eastern Estonia is a part of Eastern European province. The division between these two provinces bisects Estonia around 25°30´ N [27]. The other division of Estonia is characterized by bedrock. The borderline between limestone and sandstone is situated between 58°00´E (in Western Estonia) and 58°45´E (in Eastern Estonia). Estonian island area is characterised by marine climatic conditions and limestone bedrock and is considered separately from Western Estonia because of the isolation from the mainland [28]. Continental climatic conditions and sandstone bedrock describing Eastern Estonia and Northern Estonia differ from the latter by limestone bedrock [27,28] (Table 2). Leaf material was collected in summer 2004. The leaves were dried in silica gel and crushed in a MM300 Mixer Mill (QIAGEN GmbH, Hilden, Germany) with stainless-steel beads. DNA was extracted using QIAGEN DNeasy 96 Plant Kit, to maintain the quality of DNA as high as possible. The genomic DNA concentration for each DNA sample was determined on agarose gel by comparison 444 Unauthenticated Download Date | 4/21/16 5:20 AM E. Köster et al. with standardized DNA 25, 50, 100 and 150 ng/µl. The amount of DNA used in further analysis was adjusted accordingly. In addition, DNA of two species from Fabaceae family Lotus japonicus and Robinia pseudacacia was extracted and analysed with AFLP technique, and used as outgroups. can be calculated using the equation: IJ = J XY / J X JY , where X and Y are populations [33]. For randomly mating diploid populations (X and Y), Nei’s genetic distance (D) can be described as: D = - ln M XY / M X MY , 2.2. AFLP analysis The DNA ingerprinting technique called AFLP, based on the selective PCR ampliication of restriction fragments from a total digest of genomic DNA [29], can be a powerful technique and has been used previously by Kropf et al. [24] on A. montana and by Honnay et al. [25] on A. vulneraria. 200 ng of genomic DNA was digested with EcoRI and MseI, and double-stranded EcoRI and MseI adapters were ligated to the ends of the fragments [29]. In the following two-step ampliication, primers with one selective base (E+0, M+A) were used. In the last ampliication step primer combinations: A14/P06, A05/ P09 and C15/P13 were used. The primer sequences used were: E+0: 5´ GTAGACTGCGTACCAATTC 3´; M+1: 5´ GACGATGAGTCCTGAGTAAA 3´; E+2: E+0+TC (A14), E+0+CA (A05), E+0+TG (C15); M+3: M+1+CC (P06), M+1+GA (P09), M+1+TA (P13). Selective ampliication products were separated on 6% polyacrylamide gels (40% Acrylamide/Bisacrylamide, buffer TBE 1X and urea). Gels were run for 2 h on Genomix SC System (Beckman, Paulo Alto, CA, USA). Run parameters were: temperature 50°C, voltage 3000 V and power 100 W. 2.3. Data analyses Fluorescent fragments were scored on gels by visual observation. Each marker was coded, as 1 or 0 whether present or absent in an individual, to form a binary data matrix. Total genetic diversity was partitioned among regions deined as populations and the seven taxa of A. vulneraria s. l. by carrying out an analysis of molecular variance (AMOVA) based on pairwise genetic distances [30] using Arlequin [31]. Allele frequencies and expected heterozygosity across the total dataset were used to calculate Nei’s [32] gene diversity (HT), which Lowe et al. [33] deined as: i =K HT = 1- å pi2 , i =1 where p is the mean frequency of the ith of K alleles across all populations surveyed. Regional genetic variation was estimated on the basis of Nei’s [34] genetic identity and genetic distance with POPGENE [35]. The identity of genes (IJ) for a locus where MXY, MX and MY are the arithmetic means of JXY, JX and JY, respectively. JXY is the probability that an allele drawn from population X is the same as that from population Y. JX is the probability that two alleles drawn randomly from population X are the same. JY is the probability that two alleles created randomly from population Y are the same. A dendrogram was created based on Nei’s genetic distances using unweighted pair group method with arithmetic mean (UPGMA). Genetic variation of analyzed material was estimated using TREECON [36] by analysis of neighbour joining and simple matching. Bootstrap values were also calculated and added to the dendrogram if reaching over 25%. Neighbour-joining analysis was based on the genetic distances of Nei and Li [37]. 3. Results The three AFLP primer combinations resulted in a dataset containing information about 131 AFLP loci for 58 samples and two outgroup specimens. Genetic characteristics obtained with AMOVA analysis for the seven subspecies and four regions are presented in Tables 3 and 4. According to FST only populations of the West and North did not differ signiicantly. All other p-levels of FST for regions were below 0.05 (Table 2). FST values of taxa were signiicant for distinguishing A. vulneraria subsp. vulneraria from A. vulneraria subsp. maritima, A. x baltica from A. x colorata. The FST value of the latter taxon was signiicantly different from the values of A. vulneraria subsp. maritima and A. vulneraria subsp. polyphylla. A. vulneraria var. coccinea could also be differentiated from A. vulneraria subsp. arenaria and A. vulneraria subsp. baltica (Table 4). Analysis of genetic variation based on Nei’s (34) genetic distances supported the outcome of AMOVA analyses of specimens from the four regions. Similar to the preceding results the highest similarity exists between the North and the West (Figure 1). UPGMA dendrogram (Figure 1) accentuates differences between Island and Eastern regions of Estonia. Still values of genetic identity are very high and genetic distances between regions are inconsiderable (Table 5). 445 Unauthenticated Download Date | 4/21/16 5:20 AM Genetic structure of the Anthyllis vulneraria L. s. l. species complex in Estonia based on AFLPs Population Sample size No p-l % p-l h FST FST p-values Islands East West Islands 22 70 53.44 0.18 0.049 0 East 10 60 45.80 0.15 0.060 + 0 West 13 59 45.04 0.15 0.062 + + 0 13 64 48.85 0.18 0.051 + + – North Table 3. North 0 Characteristics of four regional subsets of Anthyllis vulneraria s. l. based on AFLP data. (No p-l – number of polymorphic loci, % p-l – percentage of polymorphic loci, h – mean value of Nei’s gene diversity for loci, FST – Fixation index, (signiicance levels of FST p<=0,05 are marked with + and p>0,05 are marked with –). Taxon Sample size No p-l % p-l h FST FST p-values Aren 6 52 39.69 0.15 0.043 Balt 8 51 38.93 0.14 0.047 – 0 Cocc 9 54 41.22 0.16 0.031 + + 0 Colo 10 59 45.04 0.17 0.027 – – – 0 Mari 6 46 35.11 0.13 0.056 – – – + 0 Poly 8 61 46.56 0.18 0.022 – – – + – 0 Vuln 11 69 52.67 0.18 0.012 – + – + + – Aren Table 4. Balt Cocc Colo Mari Poly Vuln 0 0 Characteristics of seven taxa of Anthyllis vulneraria s. l. based on AFLP data. (No p-l – number of polymorphic loci, % p-l – percentage of polymorphic loci, h – mean value of Nei’s gene diversity for loci, FST – Fixation index, (signiicance levels of FST p<=0,05 are marked with + and p>0,05 are marked with –), (Abbreviations of taxa from Table 1). 4. Discussion Figure 1. UPGMA dendrogram of Anthyllis specimens from four regions of Estonia based on Nei’s (1972) genetic distances (distances between nodes given on branches). Pop ID Islands East West North Islands X 0.96 0.96 0.96 East 0.05 X 0.96 0.94 West 0.04 0.04 X 0.97 North 0.04 0.06 0.03 X Table 5. Nei’s (1972) original measures of genetic identity (upper part of table) and genetic distance (lower part of table) of populations from four regions in Estonia. A phenogram of all AFLP phenotypes, based on a simple matching technique using the UPGMA clustering method (Figure 2) comprises many unresolved clusters. All of the larger clusters are weakly supported by bootstrap analysis and any regional or taxonomic groups could not be distinguished. The same tendency can be seen on the neighbourjoining tree calculated from the genetic distances of Nei and Li [37]. All taxa, populations and regions were disorderly located and some clusters from the same region were composed of only two or three specimens (Figure 3). P-values of ixation indexes are mostly signiicant except between specimens from the Islands and the Northern Estonia (see Table 3). That follows the proposition that there exists west/east directional variability in the lora and vegetation of Estonia [27,28,38]. In intraspeciic analysis FST measures were low, reaching maximally up to 0.056 for A. vulneraria subsp. maritima. The genetic divergence was, despite these results, signiicant (p-value<0.05) for several pairs of taxa (Table 4). Nevertheless, low values of FST and low mean values of Nei’s gene diversity refer to nearly panmictic populations (see Table 4). The value of Nei’s gene diversity can be at maximum 1.0 [33]. In our case the mean value for each analysed region and taxon was less than 0.2 that refers to the small probability that randomly chosen copies of the same gene are from different alleles (see Tables 3 and 4). The variability of analysed populations, according to the UPGMA dendrogram (Figure 1), shows the westerneastern directional trend, which is probably caused by different ecological conditions (climate and bedrock differences) occurring in the Western and the Eastern Estonia [27,28]. This interregional differentiation could be interpreted as ecotypes, but further investigation is needed. Evidence of western-eastern directional variation in A. vulneraria s. l. populations found in Estonia differs 446 Unauthenticated Download Date | 4/21/16 5:20 AM E. Köster et al. Figure 2. UPGMA dendrogram [based on simple matching method] of 58 AFLP phenotypes, plus two outgroup accessions. Lotus - Lotus japonicus, Robinia – Robinia pseudacacia. Individuals are denoted by taxon abbreviation (cf. Table 1) and followed by the regional subset (cf. Figure 1) and sample number. Bootstrap values over 25% given on branches. from the results obtained by Kropf and others [24] in A. montana in the Alps. It also differs from the work of Honnay and co-authors [25], who found no effects of the historical landscape coniguration on the genetic diversity of the populations of A. vulneraria. The UPGMA dendrogram based on Nei’s genetic distances (Figure 2) clearly demonstrates that the analysed specimens did not comprise groups correlated with the subspecies. AFLP results also do not support distinguishing of the four subgroups within A. vulneraria s. l. in Estonia, based on morphology [20]. Our data strongly support the results of Nanni et al. [23], who concluded from analysis of ITS sequences that different subspecies of A. vulneraria from a different origin and geographical distribution did not show signiicant differences in sequences. They explained these results 447 Unauthenticated Download Date | 4/21/16 5:20 AM Genetic structure of the Anthyllis vulneraria L. s. l. species complex in Estonia based on AFLPs with a complex mating system of A. vulneraria s. l. [23]. In conclusion, AFLP analysis can be used to describe the differentiation in complicated species, but our results did not reveal clear intraspeciic structure of A. vulneraria s. l. species complex. The analysed specimens did not comprise groups correlated with the subspecies. Still, some correlation was found between genetic variability and geographic distribution (westerneastern directional variation). Acknowledgements Our greatest gratitude belongs to the working group of plant genetics at Agricultural University of Marche, whose advice about laboratory work were most helpful. We also thank Tatjana Oja and Meelis Pärtel for their improvements to the manuscript and Marcus Denton for language corrections. The Estonian Science Foundation grant nr 5815 and Archimedes Foundation supported this work. Figure 3. Neighbour-joining tree, based on Nei & Li (1979) genetic distances of 58 multilocus AFLP phenotypes of Anthyllis vulneraria s. lato from four regional subsets, plus two outgroup accessions. Lotus - Lotus japonicus, Robinia – Robinia pseudacacia. Individuals are denoted by taxon abbreviation (cf. Table 1) and followed by the regional subset (cf. Figure 1) and sample number. References [1] Bisly F.A., Bickingham J., Harboune J.B., Phytochemical dictionary of the Leguminosae, Chapman & Hall, New York, 1994 [2] Allan G., Porter J., Tribal delimitation and phylogenetic relationships of Loteae and Coronilleae (Faboidaea: Fabaceae) with special reference to Lotus: evidence from nuclear ribosomal ITS sequences, Am. J. Bot, 2000, 87, 1871-1881 [3] Klitgaard B.B., Bruneau A., Advances in legume systematics (Vol. 10), Royal Botanic Gardens, Kew, 2003 [4] Yakovlev G.P., Sytin A.K., Roskov Y.R., Legumes of [5] [6] [7] [8] [9] Northern Eurasia – a check-list, Royal Botanic gardens, Kew, 1996 Cullen J., Anthyllis L., In: Tutin T.G. (Ed.), Flora Europaea (Vol. 2), Cambridge, UK, 1968 Fedorova V.A., Flora of the European part of USSR, Nauka, Leningrad, 1987, (in Russian) Komarov V., Flora URSS (Vol. 11), Leningrad, Moskva, 1945, (in Russian) Garcke A., Illustrierte Flora, Verlag Paul Parey, Berlin-Hamburg, 1972, (in German) Hegi G., Illustrierte Flora von Mitteleuropa (Vol. 4), Verlag Paul Parey, Berlin-Hamburg, 1975, (in German) 448 Unauthenticated Download Date | 4/21/16 5:20 AM E. Köster et al. [10] Hämet-Ahti L., Suominen J., Ulvinen T., Uotila P., Retkeilykasvio, Yliopistopaino, Helsinki, 1998, (in Finnish) [11] Hultén E., Fries M., Atlas of North European vascular plants. North of the tropic cancer (Vol. 2), Koeltz Scientiic Books, Königstein Federal Republic of Germany, 1986 [12] Hultén E., Fries M., Atlas of North European vascular plants. North of the tropic cancer (Vol. 3), Koeltz Scientiic Books, Königstein Federal Republic of Germany, 1986 [13] Krall H., Koldrohud (Genus Anthyllis), Eesti Loodus, 1983, 6, 388-393, (in Estonian) [14] Kuusk V., Tabaka L., Jankeviciene R., Flora of the Baltic countries (Vol. 2), Eesti Loodusfoto, Estonia, 1996 [15] Leht M., Eesti taimede määraja, Eesti Loodusfoto, Estonia, 1999, (in Estonian) [16] Roze I., Kidney vetch Anthyllis L. in the lora of Latvia, Proceedings of the Latvian Academy of Sciences, 2004, 58, 61-69 [17] Lukaszewska K., Kalinowski A., Kaczmarek Z., Szweykowski J., Analysis of the variability of natural Anthyllis vulneraria s. l. populations from Baltic coast. Variability of morphological traits on natural stands, Bulletin de la Societe des Amis des Sciences et des Lettres, 1983, 23, 45-55 [18] Lukaszewska K., Kalinowski A., Kaczmarek Z., Szweykowski J., Analysis of the variability of natural Anthyllis vulneraria s. l. populations from Baltic coast. Preliminary characteristics of morphological variability in sea coast and inland populations in successive stages of plant development, Bulletin de la Societe des Amis des Sciences et des Lettres, 1983, 23, 91-102 [19] Tihomirov V., Sokoloff D.D., Division of genus Anthyllis L. (Papilionaceae, Loteae) into subgenera and sections, Biul. MOIP Otd. biologii, 1996, 101, 61-73, (in Russian) [20] Puidet E., Liira J., Paal J., Pärtel M., Pihu S., Morphological variation of eight taxa of Anthyllis vulneraria L., s. l. (Fabaceae), Ann. Bot. Fenn., 2005, 42, 293-304 [21] Kalinowski A., Szweykowski J., Kaczmarek Z., Analysis of the variability of natural Anthyllis vulneraria s. l. populations from Baltic coast. Patterns of three enzyme systems following electrophoretic separation, Bulletin de la Societe des Amis des Sciences et des Lettres, 1983, 23, 67-78 [22] Kalinowski A., Szweykowski J., Kaczmarek Z., Analysis of the variability of natural Anthyllis vulneraria s. l. populations from Baltic coast. Patterns of esterases following separation by isoelectrofocus- [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] ing and general analysis of variation of enzymatic system, Bulletin de la Societe des Amis des Sciences et des Lettres, 1983, 23, 79-89 Nanni L., Ferradini N., Taffetani N., Papa R., Molecular Phylogeny of Anthyllis spp, Plant Biol., 2005, 6, 454-464 Kropf M.J., Kadereit W., Comes H.P., Late Quaternary distributional stasis in the submediterranean mountain plant Anthyllis montana L. (Fabaceae) inferred from ITS sequences and ampliied fragment length polymorphism markers, Mol. Ecol., 2002, 11, 447-458 Honnay O., Coart E., Butaye J., Adriaens D., Van Glabeke S., Roldán-Ruiz I., Low impact of present and historical landscape coniguration on the genetic of fragmented Anthyllis vulneraria populations, Biol. Conserv., 2006, 127, 411-419 Jalas J., Zur Kausalanalyze der Verbereitung einiger nordischen Os- und Sandplanzen, Ann. Soc. Zool. Bot. Fennici, 1950, 3, 123-127, (in German) Eilart J., Pontiline ja pontosarmaatiline element Eesti loras, H. Heidemanni nim. Trükikoda, Estonia, 1963, (in Estonian) Raukas A., Eesti loodus (Estonian Nature), Valgus, Tallinn, 1995, (in Estonian) Vos P., Hogers R., Bleeker M., Reijans M., van de Lee T., Hornes M., et al., AFLP: a new technique for DNA ingerprinting, Nucl. Ac. Res., 1995, 23, 21, 4407-4414 Weir B.S., Cockerham C.C., Estimating F-statistics for the analysis of population structure, Evolution, 1984, 38, 1358-1370 Excofier L., Laval G., Schneider S., Arlequin, ver. 3.0: An integrated software package for population genetics data analyses, Evolutionary Bioinformatics Online, 2005, 1, 47-50 Nei M., Analysis of gene diversity in subdivided populations, Proc. Natl. Acad. Sci. USA, 1973, 70, 3321-3323 Lowe A., Harris S., Ashton P., Ecological Genetics: Design, Analysis, and Application, Blackwell Publishing, 2004 Nei M., Genetic distance between populations, American Nature, 1972, 106, 283-292 POPGENE 1.31 – Microsoft Window-based Freeware for Population Genetic Analysis – A joint Project Development by Francic C. Yeh, Rong-cai Yang and Tim Boyle, 1999 Van de Peer Y., De Wachter R., TREECON for Windows, ver. 1,3b: a software package for the construction and drawing of evolutionary trees for the Microsoft Windows environment, CABIOS, 1994, 10, 569-570 449 Unauthenticated Download Date | 4/21/16 5:20 AM Genetic structure of the Anthyllis vulneraria L. s. l. species complex in Estonia based on AFLPs [37] Nei M., Li W.H., Mathematical model for studying genetic variation in terms of restriction endonucleases, Proc. Natl. Acad. Sci. USA, 1979, 76, 5269-5273 [38] Kull T., Kukk T., Leht M., Krall H., Kukk Ü., Kull K., et al., Distributional trends of rare vascular plant species in Estonia, Biodivers. Conserv., 2002, 11, 171-196 450 Unauthenticated Download Date | 4/21/16 5:20 AM