On the origins of Balkan endemics: the complex evolutionary history of the Cyanus napulifer group (Asteraceae)

Katarína Olšavská; Marek Slovák; Karol Marhold; Eliška Štubňová; Jaromír Kučera

doi:10.1093/aob/mcw142

. 2016 Jul 21;118(6):1071–1088. doi: 10.1093/aob/mcw142

On the origins of Balkan endemics: the complex evolutionary history of the Cyanus napulifer group (Asteraceae)

Katarína Olšavská ^1,^*, Marek Slovák ¹, Karol Marhold ^1,², Eliška Štubňová ¹, Jaromír Kučera ¹

PMCID: PMC5091721 PMID: 27443297

Abstract

Background and Aims The Balkan Peninsula is one of the most important centres of plant diversity in Europe. Here we aim to ﬁll the gap in the current knowledge of the evolutionary processes and factors modelling this astonishing biological richness by applying multiple approaches to the Cyanus napulifer group.

Methods To reconstruct the mode of diversiﬁcation within the C. napulifer group and to uncover its relationships with potential relatives with x = 10 from Europe and Northern Africa, we examined variation in genetic markers (amplified fragment length polymorphisms [AFLPs]; 460 individuals), relative DNA content (4′,6-diamidino-2-phenylindole [DAPI] flow cytometry, 330 individuals) and morphology (multivariate morphometrics, 40 morphological characters, 710 individuals). To elucidate its evolutionary history, we analysed chloroplast DNA (cpDNA) sequences of the genus Cyanus deposited in the GenBank database.

Key Results The AFLPs revealed a suite of closely related entities with variable levels of differentiation. The C. napulifer group formed a genetically well-defined unit. Samples outside the group formed strongly diversified and mostly species-specific genetic lineages with no further geographical patterns, often characterized also by a different DNA content. AFLP analysis of the C. napulifer group revealed extensive radiation and split it into nine allopatric (sub)lineages with varying degrees of congruence among genetic, DNA-content and morphological patterns. Genetic admixture was usually detected in contact zones between genetic lineages. Plastid data indicated extensive maintenance of ancestral variation across Cyanus perennials.

Conclusion The C. napulifer group is an example of a rapidly and recently diversified plant group whose genetic lineages have evolved in spatio-temporal isolation on the topographically complex Balkan Peninsula. Adaptive radiation, accompanied in some cases by long-term isolation and hybridization, has contributed to the formation of this species complex and its mosaic pattern.

Keywords: Cyanus napulifer group, Centaurea, Compositae, Balkan Peninsula, AFLP, cpDNA, flow cytometry, multivariate morphometrics, endemics, allopatric speciation, hybridization, incomplete lineage sorting

INTRODUCTION

The Balkan Peninsula is one of the most important European centres of plant diversity. Compared with any other region of Europe, it harbours the richest flora, including a high percentage of endemics (e.g. Eastwood, 2004; Kryštufek and Reed, 2004; Stevanović et al., 2007; Hewitt, 2011). Its astonishing diversity was likely generated by a combination of factors and processes. One of the most important is undoubtedly the refugial character of the Balkan Peninsula. The paleo-environmental conditions in this region allowed the survival of relic taxa, including some of tertiary origin (Tzedakis, 2004; Peev and Delcheva, 2007). Furthermore, the territory of today’s Balkan Peninsula served as a crossroad for Asian lineages and taxa during east–west colonization of Europe since the Early Oligocene (32 million years ago; Hewitt, 2011; Manafzadeh et al., 2014). The heterogeneous climate and complex physical geography of the Balkan Peninsula provided a wide range of conditions for different species and vegetation types to evolve in. The patchy, topographically varied landscape of the Balkans, with its mosaic of mountain ranges, islands and peninsulas that acted as migration barriers for most organisms, significantly contributed to the formation of Balkan biodiversity (Stevanović, 1996; Reed et al., 2004; Meshinev, 2007; Peev and Delcheva, 2007; Velikov and Stoyanova, 2007; Radford and Odé, 2009). The Balkan flora is nevertheless still largely unexplored and calls for further investigation (but see Perný et al., 2005; Frajman and Oxelman, 2007; Stefanović et al., 2008; Frajman and Schneeweiss, 2009; Schönswetter and Schneeweiss, 2009; Bardy et al., 2010; Kolarčik et al., 2010; Kučera et al., 2010; Surina et al., 2011; Kuzmanović et al., 2013; Lakušić et al., 2013; Ronikier and Zalewska-Gałosz 2014).

The Balkan Peninsula accommodates almost 27% (18 out of 67 species) of perennial taxa of Cyanus sect. Protocyanus (Dobrocz.) Olšavská (Olšavská et al., 2013) and is considered its evolutionary centre (Hellwig, 2004). A large proportion of Cyanus perennials from the Balkans belong to the taxonomically extremely complex group of C. napulifer (Olšavská et al., 2013), which is the focus of the present study. The C. napulifer group, as defined by Olšavská et al. (2013), comprises five Balkan endemics – C. napulifer (Rochel) Soják (s.str.), C. orbelicus (Velen.) Soják, C. velenovskyi (Adamović) Wagenitz & Greuter, C. nissanus (Petrović) Soják and C. tuberosus (Vis.) Soják; one species occurring in the Balkan Peninsula and Turkey [C. thirkei (Sch. Bip.) Holub]; and one recently described Turkish species [C. eflanensis Kaya & Bancheva (Greuter, 2006–2009; Kaya and Bancheva, 2009)]. Although their mutual phylogenetic relationships remain unresolved, previous genetic studies indicate their close evolutionary relationships. However, C. thirkei seems to occupy a separate position, and no information is available on C. eflanensis (Boršić et al., 2011; Löser, 2012). Morphologically, the C. napulifer group is clearly characterized by a suite of common features, mainly by rhizomatous and/or tuberous roots and narrow stem leaves. All the taxa are diploids and share the base chromosome number of x = 10 (no data are available for C. eflanensis), save for a few triploids (2n∼3x∼30) found in C. nissanus and C. thirkei (Olšavská et al., 2013). Two other Balkan species, C. diospolitanus Bancheva & Stoyanov and C. pseudoaxillaris (Stef. & T. Georgiev) Holub, previously included in Cyanus sect. Napuliferi (Stef. & T.Georgiev) Bancheva & Raimondo (Bancheva and Raimondo, 2003; Bancheva and Stoyanov, 2009), were proved to be genetically, karyologically and morphologically unrelated to the C. napulifer group (Boršić et al., 2011; Olšavská et al., 2013) and so were not included in the present study.

The C. napulifer group represents a promising model for the study of homoploid speciation promoted mainly by geographical isolation and adaptation to new ecological niches. However, previous efforts to infer the evolutionary relationships of Cyanus perennials and the mechanisms of their diversification largely failed because of the insufficient or conflicting phylogenetic signals inferred from genetic sequence markers (Boršić et al., 2011; Löser, 2012). To uncover the evolutionary relationships among members of this group and the factors responsible for their diversification, we applied a combination of approaches that has proved to be effective in similar studies (e.g. Schönswetter et al., 2009; Bardy et al., 2010; Kučera et al., 2010; Olšavská et al., 2011). Specifically, we employed fast-evolving multilocus amplified fragment length polymorphism (AFLP) markers, chloroplast DNA (cpDNA) data, flow-cytometric data and multivariate morphometrics.

The aims of our current study were: (1) to reveal which populations/taxa constitute the C. napulifer group and to uncover their relationships with other Cyanus taxa with x = 10 occurring in Europe and Northern Africa; (2) to infer the overall genetic, karyological and morphological patterns of Balkan members of the C. napulifer group and to examine their congruence; and (3) to identify the evolutionary processes and factors that triggered speciation and divergence within the C. napulifer group and to infer biogeographical implications.

MATERIALS AND METHODS

Sampling design

The sampling strategy was focused on members of the Cyanus napulifer group occurring predominantly in the Balkan Peninsula. A total of 38 population samples of C. napulifer (s.str.), C. orbelicus, C. velenovskyi, C. nissanus, C. thirkei and C. tuberosus were collected from natural habitats in 2009–13 (May/August) throughout the Balkan Peninsula [Bulgaria (BU), Croatia (HR), Greece (GR), Former Yugoslav Republic of Macedonia (MK), Serbia (RS)] (Fig. 1, Table 1).

Fig. 1. — Distribution map of *Cyanus* samples used in molecular, DNA-content and morphometric analyses. Samples of *C. napulifer* group are assigned to genetic lineages (coloured solid lines) and sublineages (coloured dashed lines) according to the results of AFLP data analyses. Symbols with thick lines are used for populations of presumably hybrid origin.

Table 1.

Locality details, including geographical coordinates, altitude, the date of collection and the collector(s) of plants of Cyanus species included in the molecular (AFLP), DNA content flow cytometric and morphometric analyses

Population (voucher) code^*	Locality and collection details^†	No. of plants analysed
Population (voucher) code^*	Locality and collection details^†	AFLP	Flow cytometry N/P^‡ (ploidy; ST^§)	Morphometric analyses
Cyanus napulifer group
Cyanus napulifer + orbelicus BU (central Bulgarian) genetic lineage
TRI 214	Bulgaria; Stara Planina Mts, Korduna peak; 42°45′19″ N, 24°04′36″ E; 1605 m; (A) 2.viii.2011; (B) 8.vii.2013, MP & KO	8	4/3 (2n ∼ 20; SL)	20
TRI 215	Bulgaria; Stara Planina Mts, 2 km SE from Vezhen cottage; 42°45′47·5″ N, 24°24′10·8″ E; 1869 m; 2.viii.2011, KO	5	3/3 (2n ∼ 20; SL)	15
TRI 216	Bulgaria; Stara Planina Mts, Troyanski prokhod, Beclemento; 42°46′37″ N, 24°36′52″ E; 1601 m; 3.viii.2011, KO	8	3/3 (2n ∼ 20; SL)	12
TRI 217	Bulgaria; Stara Planina Mts, Troyanski prokhod, Mt Viloto; 42°46′47″ N, 24°39′56″ E; 1603 m; 3.viii.2011, KO	8	4/3 (2n ∼ 20; SL)	11
TRI 300	Bulgaria; Stara Planina Mts, Buzludzha; 42°44′07·0″ N, 25°23′51·8″ E; 1400 m; 9.vii.2013, MP & KO	8	6/0 (2n ∼ 20; SL)	20
TRI 222	Bulgaria; Rila Mts, Belmeken dam; 42°09′51″ N, 23°46′57″ E; 1963 m; 9.viii.2011, KO	8	2/4 (2n ∼ 20; SL)	15
TRI 153	Bulgaria; Rila Mts; (A) 2·5 km NE from Granchar Hut; 42°08′08″ N, 23°36′54″ E; 2294 m; 19.vi.2010, KO; (B) 1·2 km N from Granchar Hut; 42°07′51″ N, 23°35′45″ E; 2277 m; 20.vi.2010, KO	8	2/6 (2n ∼ 20; SL)	31
Cyanus orbelicus BU/RS (Bulgarian–Serbian) genetic sublineage
TRI 296	Serbia; Kopaonik Mts Karaman; 43°17′34·0″ N, 20°49′47·7″ E; 1910; 6.vii.2013, MP & KO	8	5/0 (2n ∼ 20; SL)	20
TRI 260	Serbia; Mt Besna Kobila; (A) below the peak; 42°32′57·9″ N, 22°13′56·2″ E; 1662 m; (B) N of the peak; 42°33′29·9″ N, 22°14′09·4″ E; 1559 m; 15.vi.2012, CL & KO	8	0/4 (2n ∼ 20; SL)	41
TRI 298	Bulgaria; Ossogovska planina; (A) Kulin Kamak; 42°11′41·4″ N, 22°35′54·3″ E; (B) Gramadite; 42°11′17·9″ N, 22°36′31·6″ E; 1636 m; 1710 m; 7.vii.2013, MP & KO	8	8/0 (2n ∼ 20; SL)	32
TRI 302	Bulgaria; Rhodopi Mts, Golyam Perelik Mt; 41°36′00·8″ N, 24°34′48·5″ E; 2140 m; 11.vii.2013, MP & KO	8	5/0 (2n ∼ 20; SL)	20
Cyanus velenovskyi GR (Greek) genetic sublineage
TRI 144	Greece; Pella, E slope of Mt Kaimaktsalan; 40°55′03″ N, 21°48′41″ E; 2160 m; 13.vi.2010, MO & KO	8	0/4 (2n ∼ 20; SL)	20
TRI 241	Greece; Florina, Vernon Mts, W slope of Mt Vitsi; 40°38′53·0″ N, 21°22′47·1″ E; 1921 m; (A) 22.v.2012, MO & KO; (B) 13.vii.2013, MP & KO	8	5/4 (2n ∼ 20; SL)	12
TRI 305	Greece; Florina, Varnous Mts, Mt Vigla; 40°48′35·3″ N, 21°15′07·0″ E; 1990 m; 13.vii.2013, MP & KO	8	5/0 (2n ∼ 20; SL)	20
Cyanus velenovskyi RS (Serbian) genetic sublineage
TRI 258	Serbia; Stara Planina Mts, Mt Midžor; (A) near the ski resort Babin zub; 43°22′45·6″ N, 22°37′39″ E; 1592 m; 15.vi.2012, KO; (B) S slope; 43°22′44″ N, 22°40′48″ E; 1662 m; 15.vi.2012, CL	8	3/4 (2n ∼ 20; SL)	29
Cyanus nissanus RS genetic sublineage
TRI 256	Serbia; Svrljiska Planina Mts, Mt Popova glava, summit area; 43°21′23·4″ N, 22°05′30·2″ E; 1008 m; (A) 14.vi.2012, CL & KO; (B) 19.v.2013, MO, JO & KO	8	1/2 (2n ∼ 20; SL)0/2 (2n ∼ 30; SL)	20
Cyanus orbelicus Vich (Vichren) genetic sublineage
TRI 151	Bulgaria; Pirin Mts, Mt Vihren, above Vihren cottage; 41°45′27″ N, 23°24′17″ E; 2440 m; (A) 17.vi.2010; (B) 10.viii.2010; (C) Bakushevata Mura; 41°45′54″ N, 23°24′44″ E; 2217 m; 10.viii.2011, KO	7	3/7 (2n ∼ 20; SL)	24
Cyanus tuberosus GR/MK (Greek–Macedonian) genetic sublineage
TRI 240	Greece; Larisa, Deskati, E of Loutro; 39°56′55·4″ N, 21°55′40·8″ E; 732 m; 21.v.2012, MO & KO	8	2/0 (2n ∼ 20; SL)	15
TRI 170	Greece; Grevena, Vourinos Mts, S from Exarchos, 40°09′02″ N, 21°37′30″ E, 684 m; 28.v.2011, KO	5	0/4 (2n ∼ 20; SL)	13
TRI 149	Greece; Kozani, Pieria Mts, SE of Velvendo; 40°13′52″ N, 22°06′39″ E; 1278 m; 16.vi.2010, KO	8	0/4 (2n ∼ 20; SL)	18
TRI 162	Greece; Kilkis, Paiko Mts; (A) S of Livadhia; 40°59′40″ N, 22°17′46″ E; 1160 m; (B) NW of Livadhia; 41°01′ N, 22°16′30″ E; 1180 m; 26.v.2011, KO	7	11/0 (2n ∼ 20; SL)	35
TRI 238	Greece; Thesaloniki, Mt Hortiatis; 40°35′57·8″ N, 23°06′01·2″ E; 826 m; 20.v.2012, MO & KO	8	2/0 (2n ∼ 20; SL)	14
TRI 237	Greece; Serres; (A) Lailas; 41°14′22·7″ N, 23°34′22·0″ E; 1393 m; (B) between Kato Vrondous and Eleonas; 41°13′57·2″ N, 23°39′45·3″ E; 915 m; 19.v.2012, MO & KO	8	5/0 (2n ∼ 20; SL)	27
TRI 235	Greece; Drama, Voras Mts; (A) Falakro; 41°17′36·6″ N, 24°02′29·7″ E; 1467 m; (B) Volokas; 41°18′10·9″ N, 23°59′34·9″ E; 1012 m; 19.v.2012, MO & KO	8	1/0 (2n ∼ 20; SL)	35
TRI 273	Macedonia/Greece; Kožuf Mts; (A) Mt Keci Kaj; 41°11′48·0″ N, 22°14′33·8″ E; 1715 m; 22.vi.2012, KO; (B) above ski centre; 41°11′01·1″ N, 22°12′59·7″ E; 1912 m; 22.vi.2012, CL; (C) Mt Mala Rupa; 41°09′52·5″ N, 22°14′37·6″ E; 1863 m; 22.vi.2012, CL	8	3/0 (2n ∼ 20; SL)	52
Cyanus tuberosus MK (central Macedonian) genetic lineage
TRI 270	Macedonia; Dren Mts, pass above Štavica; 41°15′03·5″ N, 21°35′41·1″ E; 1132 m; (A) 21.vi.2012, CL & KO; (B) 23.v.2013, MO, JO & KO	8	4/2 (2n ∼ 20; SL)	20
TRI 271	Macedonia; Dren Mts; (A) Veprchani; 41°16′11·9″ N, 21°45′24·4″ E; 832 m; 22.vi.2012, KO; (B) 20.v.2013, MO, JO & KO; (B) above Dunje; 41°27′27·2″ N, 21°46′07·4″ E; 1479 m; 22.vi.2012, CL	8	5/4 (2n ∼ 20; SL)	10
TRI 262	Macedonia; Ramno Mts, above Brodec; (A) 42°08′35·8″ N, 21°26′57·5″ E; 1100 m; 15.vi.2012, CL & KO; (C) 42°09′50·9″ N, 21°26′53·0″ E; 1386 m; 25.v.2013, MO, JO & KO	8	8/4 (2n ∼ 20; SL)	19
TRI 264	Macedonia; Vodno Mts, Skopje, near Millenium cross; (A) 42°58′11·7″ N, 21°23′54·6″ E; 922 m; 16.vi.2012, CL; (B) 41°57′58·34″ N, 21°23′43·34″ E; 1022 m; 24.v.2013, MO, JO & KO	8	5/1 (2n ∼ 20; SL)	20
TRI 266	Macedonia; Babuna Mts, Prisad pass; 41°26′54·0″ N, 21°37′50·2″ E 41°26′24″ N, 21°36′55″ E; 809–1100 m; (A) 19.vi.2012, CL & KO; (B) 24.v.2013, MO, JO & KO	8	5/3 (2n ∼ 20; SL)	20
TRI 265	Macedonia; Babuna Mts, above Nezhilovo; 41°40′11·8″ N, 21°29′29″ E; 2064 m; 17.vi.2012, CL	8	1/4 (2n ∼ 20; SL)	16
Cyanus tuberosus HR (Croatian) genetic lineage
TRI 247	Croatia; Lika, Mt Poštak, along a road to Rastičevo; 44°13′28·4″ N, 16°05′48·8″ E; 752 m; 5.vi.2012, MO, JO & KO	8	2/4 (2n ∼ 20; SL)	17
TRI 252	Croatia; Split-Dalmatia County, above Sinj; (A) near Zelovo; 43°45′04·4″ N, 16°33′12·2″ E; 800 m; 7.vi.2012, MO, JO & KO; (B) along a road to Sutina; 43°42′04·0″ N, 16°36′52·5″ E; 402 m; 7.vi.2012, MO, JO & KO	8	6/8 (2n ∼ 20; SL)	11
TRI 253	Croatia; Mosor Mts, near Rašca; 43°33′48·8″ N, 16°42′33·8″ E 43°33′22·3″ N, 16°42′22·5″ E; 312–361 m; 8.vi.2012; MO, JO & KO	8	3/4 (2n ∼ 20; SL)	8
Cyanus pindicola (Griseb.) Soják
TRI 145	Greece; Pella, NE slope of Mt Piperitsa; 40°51′38″ N, 21°44′32″ E; 1812 m; 14.vi.2010, MO & KO	6	0/3 (2n ∼ 20*; SL)	–
TRI 148	Greece; Pieria, Oros Olympos Mts, below Zolotas cottage; 40°04′56″ N, 22°22′50″ E; 1709 m; 15.vi.2010, MO & KO	6	0/1 (2n ∼ 40*; SL)	–
TRI 166	Vermio Mts, Seli 40° 32′ 03″ N, 22°01′30″ E; 1662 m; 27.v.2011, KO	5	3/0 (2n ∼ 40; SL)	–
TRI 168	Vourinos Mts 40° 11′ 56·2″ N, 21°39′27·7″ E; 1330 m; 28.v.2011, KO	5	3/0 (2n ∼ 40; SL)	–
TRI 175	Greece; Larisa, Kato Olympos Mts, 2 km NE from Kallipefki; 39°58′25″ N, 22°28′46″ E; 1141 m; 29.v.2011, CL & KO	6	0/5 (2n ∼ 40*; SL)	–
TRI 267	Greece; Galičica Mts; (A) Galičica Mt; 40° 57′ 43·2″ N, 20° 49′ 41·3″ E; 1843 m; 20.vi.2012, KO; (B) Magaro Mt; 40°56′22·1″ N, 20°49′18·7″ E; 2 163 m; 20.vi.2012, CL	6	1/0 (2n ∼ 20; SL)	–
Cyanus epirotus (Halácsy) Holub
TRI 182	Greece; Ioannina, Tzoumerka Mts, Kryopigi; 39°27′09″ N, 21°07′39″ E; 1443 m; 2.vi.2011, CL & KO	3	0/3 (2n ∼ 20; SL)	–
TRI 183	Greece; Ioannina, Tzoumerka Mts, close to Mparos pass; 39°36′36·6″ N, 21°08′32·9″ E; 1676 m; 2.vi.2011, CL & KO	6	0/3 (2n ∼ 20; SL)	–
TRI 185	Greece; Ioannina, Mitsikeli Mts, above Ioannina; 39°43′01·6″ N, 20°53′0·7″ E; 1669 m; 3.vi.2011, CL & KO	6	0/5 (2n ∼ 20; SL)	–
TRI 187	Greece; Ioannina, Timfi Mts, above the road to Vradeto; 39°55′03″ N, 20°48′21″ E; 1738 m; 3.vi.2011, CL & KO	6	0/3 (2n ∼ 20; SL)	–
Cyanus thirkei (Sch.Bip.) Holub
TRI 228	Bulgaria; Stara Planina Mts, Varbishki prohod, above Sadovo, 42°53′17·6″ N, 26°39′05·8″ E; 686 m; (A) 15.v.2012, MO & KO; (B) 20.v.2013, MO, JO & KO	8	3/0 (2n ∼ 20; BP) 0/7 (2n ∼ 30; BP)	–
TRI 229	Bulgaria; Balgarova near Burgas; 42°37′58·3″ N, 27°16′48·2″ E; 99 m; (A) 16.v.2012, KO; (B) 21.v.2013, MO, JO & KO	8	3/4 (2n ∼ 20; BP)	–
TRI 230	Bulgaria; Otmanli near Burgas; 42°25′49·1″ N, 27°32′59·1″ E; 52 m; 16.v.2012, MO & KO	8	0/4 (2n ∼ 20; BP)	–
TRI 294	Bulgaria; Karnobatska Planina Mts, between Valchin and Lazarevo; 42°44′26·6″ N, 26°53′54·7″ E; 271 m; 21.v.2013, MO, JO & KO	8	6/0 (2n ∼ 20; BP)	–
TRI 295	Bulgaria; Karnobatska Planina Mts, N of Aytos; 42°45′04·6″ N, 27°15′16·7″ E; 347 m; 21.v.2013, MO, JO & KO	8	6/0 (2n ∼ 20; BP)	–
Cyanus pichleri
TRI 179	Greece; Peloponnese, Mt Panachiko; 38°15′52″ N, 21°51′06″ E; 1146 m; 31.v.2011, CL & KO	6	0/2 (2n ∼ 40; SL)	–
TRI 186	Greece; Ioannina, Timfi Mts, Vradeto; 39°53′53·2″ N, 20°46′23·8″ E; 1339 m; 3.vi.2011, CL & KO	5	0/3 (2n ∼ 40; SL)	–
TRI 188	Greece; Kastoria, between Nea Kotyli and Kato Nestorio; 40°22′16·6″ N, 21°02′41·4″ E; 1268 m; 4.vi.2011, CL & KO	5	0/2 (2n ∼ 40; SL)	–
TRI 231	Bulgaria; Strandzha Mts, Golyamo Bukovo, above road E79; 42°14′28·3″ N, 27°06′36·6″ E; 340 m; 17.v.2012, MO & KO	6	0/1 (2n ∼ 40; SL)	–
TRI 239	Greece; Mt Chortiatis, above village of Chortiatis; 40°35′57·8″ N, 23°06′01·2″ E; 826 m; 20.v.2012, MO & KO	6	0/4 (2n ∼ 40; SL)	–
Cyanus fuscomarginatus (K.Koch) Greuter
TRI 278	Ukraine; Crimea, Aj-Petrinskaja Jajla, Mt Aj-Petri; 44°27′29″ N, 34°03′37″ E; 1178 m; 7.vii.2012, KO	8	0/4 (2n ∼ 40; SL)	–
TRI 282	Ukraine; Crimea, Čatyr-Dag Jalja, near Marmornaja cave; 44°47′25″ N, 34°16′48″ E; 973 m; 9.vii0·2012, KO	8	0/3 (2n ∼ 40; SL)	–
Cyanus lingulatus (Lag.) Holub
JAV	Spain; Sierra de Javalambre; 40°13′31 N, 00°20′56 W; 1853 m; 3.vi.2010; SS, KM & JZ	8	0/5 (2n ∼ 20; BP)	–
PUR	Spain; Sierra del Moncayo, Purujosa; 41°42′47 N, 1°44′54 W; 1280 m; 8.v.2011; SS, KM & JZ	7	0/5 (2n ∼ 20*; BP)	–
TRI 289	Morocco; Moyen Atlas, between Sources de l′Oum-er-Rbia and Âin-Leuch; 33°12·421′ N, 05°20·637′ W; 1734 m; 10·5·2013; AG, JK, MS & KO	8	6/0 (2n ∼ 20; BP)	–
TRI 291	Morocco; Rif Mts, Tleta Ketama; 34°52·406′ N, 04°33·610′ W; 1573 m; 11.v.2013; AG, JK, MS & KO	8	1/0 (2n ∼ 20; BP)	–
Cyanus graminifolius (Lam.) Olšavská
TRI 65	France; Hautes Alpes, Montagne de Céüse; 44°31′02″ N, 05°52′08″ E; 1572 m; 15.vii.2007, KO, MP & DD	8	0/5 (2n ∼ 40*; BP)	–
TRI 101	France; Alpes de Hte Provence, Montagne du Cheiron, Pic de Fourneuby; 43°49′29·4″ N, 06°53′32·5″ E; 1422 m; 6.vii.2008; KO, MP & VK	6	0/7 (2n ∼ 40; BP)	–
TRI 196	Italy; Pollino Mts, Colle del Dragone; 39°54′46·4″ N, 16°07′30·8″ E; 1531 m; 14.vi.2011, KO (SAV: TRI 196)	6	0/3 (2n ∼ 40; BP)	–
TRI 198	Italy; Calabria, La Sila Mts, 4 km SW from Moccone; 39°20′05·7″ N, 16°23′56·5″ E; 1613 m; 14.vi.2011, KO (SAV: TRI 198)	6	0/5 (2n ∼ 40; BP)	–

Open in a new tab

Populations of C. napulifer group are assigned to genetic lineages and sublineages according to the results of AFLP data analyses.

*Population codes of populations with presumably hybrid origin are in bold.

^†Collectors: AG, A. Guttová; IH, I. Hodálová; VK, V. Kolarčik; CL, C. Löser; KM, K. Marhold; JK, J. Kučera; JO, J. Olšavská; MO, M. Olšavský; MS, M. Slovák; KO, K. Olšavská; MP, M. Perný; SS, S. Španiel; JZ, J. Zozomová-Lihová.

^‡Number of newly analysed plants/number of plants analysed in a previous study (Olšavská et al., 2013)

^§Standard used in flow cytometry analysed: BP, Bellis perennis; SL, Solanum lycopersicum.

To elucidate the evolutionary relationships of the C. napulifer group, material from 30 populations of other morphologically similar European and North African species of the section Protocyanus, with the base chromosome number of x = 10 , was collected for AFLP and DNA content analyses (Table 1, Fig. 1): C. pindicola (Griseb.) Soják, C. epirotus (Halácsy) Holub, C. lingulatus (Lag.) Holub, C. graminifolius (Lam.) Olšavská, C. fuscomarginatus (K.Koch) Greuter, and C. pichleri (Olšavská et al., 2013).

Plant material

Intact fresh leaves were collected from three to eight plants per locality and dried in silica gel for AFLP analyses. Roots/rhizomes from up to 14 plants were transferred to an experimental garden in Banská Bystrica, Slovakia (48 °45′08·9″ N, 19 °09′29·0″ E, 390m a.s.l.) for flow cytometry estimation. Depending on the abundance of flowering individuals, 10–52 plants of the C. napulifer group (not including C. thirkei) were collected at original localities for morphometric analyses. The type of bedrock and biotope were recorded during the ﬁeld research. Each population of the species under investigation had at least several hundreds of individuals, and the taxa under study are neither endangered nor protected by law. Voucher specimens were deposited in the herbarium of the Institute of Botany of the Slovak Academy of Sciences (SAV).

DNA extraction and AFLP generation

Total genomic DNA was extracted using the DNeasy Plant Mini Kit (Qiagen, Düsseldorf, Germany).

The AFLP procedure (Vos et al., 1995) followed a modified protocol described by Mereďa et al. (2008). To obtain high-quality AFLP profiles, we tested 38 selective primer combinations using four individuals from different, geographically distant taxa. Final analyses were performed using four primer combinations labelled with different fluorescent dyes and producing clear, unambiguous and polymorphic profiles: EcoRI-ACA-(6-FAM)/MseI-CTG, EcoRI-AAG-(VIC)/MseI-CAG, EcoRI-AAC-(NED)/MseI-CTG and EcoRI-ACA-(PET)/MseI-CAC (Applied Biosystems). Amplification products were pooled and fragment analysis was performed on an ABI 3100 Avant capillary sequencer of the BITCET Consortium, Department of Molecular Biology, Comenius University, Bratislava. The internal size standard GeneScan –500 LIZ (Applied Biosystems) was employed for size calibration of the AFLP profiles obtained. AFLP trace files were read and analysed using DAx software (Van Mierlo Software Consultancy, Netherlands). Only unambiguously scorable markers in the size range of 50–500 base pairs (bp) were scored manually as present (1) or absent (0).

The reproducibility of AFLP profiles was determined using 35 replicates (7·6% of the final dataset). This dataset comprised randomly chosen re-extracted samples of all taxa under study, which were analysed and scored independently. The error rate (Bonin et al., 2004) was calculated as the ratio of mismatches (1 versus 0) to matches (1 versus 1) between replicated samples.

AFLP data analyses

To reveal the genetic pattern of the taxa under study, we assembled four AFLP datasets. The first dataset, AFLP dataset 1, comprised all analysed taxa (460 individuals); the second one, AFLP-dataset 2 (264 individuals), comprised only accessions of the C. napulifer group excluding C. thirkei, which early on turned out to be genetically clearly distinct from the rest of the group (see the Results section). AFLP dataset 3 (233 individuals) comprised accessions of the C. napulifer group excluding four genetically admixed populations (TRI 151, TRI 153, TRI 265 and TRI 273; see the Results section). For the purposes of multispecies coalescent analyses, we created AFLP dataset 4 (120 individuals), composed of two randomly selected individuals for each population analysed except four genetically admixed ones (TRI 151, TRI 153, TRI 265 and TRI 273; see also below in this section).

The overall genetic pattern of relationships among populations and individuals was investigated using the following approaches based on AFLP datasets 1–3: (1) neighbour-joining tree (NJ; Saitou and Nei, 1987) based on Sørensen’s ( = Dice) similarities transformed into a distance matrix with bootstrap support for clades and subclades computed using 5000 replicates; (2) split network (neighbour-net, NN) analysis based on Sørensen’s similarities transformed into a distance matrix (Huson and Bryant, 2006); and (3) principal coordinates analysis (PCoA; Krzanowski, 1990) based on Jaccard’s similarity matrix. The NJ tree analysis and PCoA were performed using Famd 1·108 beta software (Schlüter and Harris, 2006), while NN was conducted using SplitsTree 4 (Huson and Bryant, 2006).

To search for genetically homogeneous groups within the C. napulifer group encompassing only diploid taxa (AFLP dataset 2), Bayesian model-based clustering as implemented in Structure 2·2·3 (Falush et al., 2007) was employed. The settings for the analysis were the same as those used in Slovák et al. (2012a, b). The R script Structure-sum-2009 (part of AFLPdat; Ehrich, 2006) was used to calculate the statistics necessary for determining the correct K value (Evanno et al., 2005). Graphical outputs of analyses were generated using Clumpp v1·1.1 (Jakobsson and Rosenberg, 2007) and Distruct (Rosenberg, 2004).

To infer hierarchical relationships between genetic lineages of the C. napulifer group and related species with x = 10, we computed a species trees in a multispecies coalescent framework using SNAPP v2.2.0 (Bryant et al., 2012). A priori ‘species’ delimitation was done with respect to the results of NN, NJ, PCoA and structure analyses, with the exceptions of C. lingulatus and C. graminifolius. These were regarded as a single species to ensure the monophyly of the group. Due to the large computational demands of the algorithm, we analysed a subsample comprising two randomly selected individuals for each population (AFLP dataset 4). We ran two analyses with different θ priors in order to allow for different current and ancestral population sizes, as applied in Kolář et al. (2016): (1) mean θ prior of 0·043 (γ distribution, α = 1·5, β = 35) and (2) mean θ prior = 0·1 (γ distribution, α = 12, β = 110 prior for large population sizes); the remaining parameters were left at their default values. The analyses were checked for convergence using Tracer v1.6, making sure that Bayesian runs reached an effective sample size exceeding 200 after burn-in. The posterior distributions of species trees were visualized using DensiTree v2.2.0 (Bouckaert and Heled, 2014).

Genetic diversity within and among the taxa/populations/genetic (sub)lineages under study (AFLP datasets 1, 2 and 4) was estimated using the following statistical parameters: (1) total number of AFLP multilocus genotypes; (2) average number of AFLP fragments (± s.d.); (3) average proportion of pairwise differences between individuals (Nei’s gene diversity, D_Nei; Nei and Li, 1979) using the R script AFLPdat (Ehrich, 2006); and (4) percentage of polymorphic markers (P%) using Pop-Gene 1·32 (Yeh et al., 1997). Genetic divergence among analysed taxa/populations/genetic (sub)lineages was explored by calculating the number of (5) private fixed fragments (diagnostic); (6) private fragments (exclusive); (7) rare fragments (present at a frequency of < 10% of investigated individuals); and (9) unique fragments, using FAMD 1·108 beta software (Schlüter and Harris, 2006) and by calculating of the frequency down-weighted marker value (Schönswetter et al., 2005), as implemented in AFLPdat (DW1; Ehrich, 2006).

Within the C. napulifer group (AFLP dataset 3), we also explored within- and between-group variance using analyses of molecular variance (AMOVA). AMOVAs were performed using Arlequin 3·11 (Excoffier et al., 2005) based on Euclidean pairwise distances and a significance test with 10 000 permutations.

cpDNA data analyses

To address the evolutionary relationships among perennial representatives of the genus Cyanus, we explored cpDNA sequences deposited in the GeneBank database (NCBI, http://www.ncbi.nlm.nih.gov/nucleotide), particularly those generated by C. J. Löser, G. Akaydin and F. H. Hellwig (University of Jena, Germany). We used 114 sequences of trnC-ycf6 and 99 sequences of 3′rps16-5′trnK^(UUU) cpDNA regions (Shaw et al., 2005, 2007) for 15 Cyanus taxa. This dataset included not only an overwhelming majority of species analysed in our study with x = 10, but also species with x = 11 (Supplementary Data Table S1).

All cpDNA sequences were manually edited and aligned using BIOEDIT (v7.0.4.1; Hall, 1999). Ambiguously aligned regions were removed with Gblocks (Castresana, 2000; Talavera and Castresana, 2007). The method of statistical parsimony was employed to generate a haplotype network (TCS v.1.18; Clement et al., 2000) depicting the general pattern of relationships among the haplotypes. TCS was run with a 95% connection limit and with indels coded as missing data.

Flow cytometric analyses

The dataset resulting from relative nuclear DNA content analyses comprised both newly generated (136 individuals) and previously published data (194 individuals; Olšavská et al., 2013); between one and 14 plants per population were analysed by flow cytometry using the fluorochrome 4′,6-diamidino-2-phenylindole (DAPI, specific for A–T base pairs) (Table 1). DAPI flow cytometry was chosen because a previous study, which was focused on the C. triumfetti and C. montanus groups (Olšavská et al., 2012), proved a strong correlation between absolute DNA content (estimated using the intercalating fluorochrome propidium iodide) and relative DNA content (estimated using DAPI) (r = 0·99, P<0·001). Moreover, DAPI flow cytometry is more accurate and is particularly useful for detecting small differences in genome size (Marhold et al., 2010; Suda et al., 2010).

To ensure the accuracy of relative DNA content estimations, each plant was analysed separately, and fresh leaf material was used in all cases. As a consequence, cultivated plants subjected to flow cytometry analysis only partly overlapped with those used for generating AFLP data. Details of the method of flow cytometry analysis are given in Olšavská et al. (2011, 2012). Simultaneous analyses of samples differing by >5% in genome size were performed to confirm the reliability of the estimated values. The relationship between chromosome numbers and DNA content values was verified using previously published chromosome counts (Table 1; Olšavská et al., 2013). Relative DNA content was calculated as the ratio of the position of the G1 peak of the standard [Solanum lycopersicum ‘Stupické polní rané’, 2C DNA = 1·96pg (Doležel et al., 1992) or Bellis perennis, 2C DNA = 3·38pg (Schönswetter et al., 2007)] to that of the sample. When B. perennis was used as the internal standard, the resulting relative DNA content was recalculated relative to the S. lycopersicum standard based on the ratio between B. perennis and S. lycopersicum (0·578) calculated from eight estimations performed at different times. To compare samples differing in ploidy level, relative DNA content per monoploid genome was used. The Tukey–Kramer test (Tukey’s test for unequal sample size; Tukey, 1977) was used to test for differences in relative DNA content between populations of the C. napulifer group (triploids from population 256, investigated only in the previous study, were not included) and in monoploid relative DNA content between taxa or genetic (sub)lineages. All analyses were carried out using STATISTICA 9·1 (StatSoft Inc., 2013).

For populations of C. nissanus and C. thirkei (48 plants in total), ploidy levels were estimated by flow cytometry for each individual subjected to AFLP because in a previous study (Olšavská et al., 2013), two ploidy levels (2x and 3x) were identified in some populations of these taxa. Sample preparation followed the procedure described above, but silica-dried material was used, and one to three plants were analysed simultaneously; such flow cytometric estimations were not used to calculate relative DNA content.

Morphometric analyses

Morphological variability was studied using multivariate morphometric analyses based on 710 individuals from 34 populations of the C. napulifer group (Table 1). A total of 24 quantitative and nine qualitative characters were measured or scored on herbarium material, and three ratios were subsequently computed (Supplementary Data Table S2). Depending on the number of states recorded (Table S2), multistate qualitative characters were binary-coded for further analyses as two (ROOT, SI, LBWS, LBAS, LLI, LUI, BUL, BBL), three (FECO) or five (SECR) dummy binary variables. Each multistate qualitative character with s states was decomposed into s–1 dummy variables (to avoid complete linear dependency of the sth variable).

First, all quantitative characters were tested for normality of distribution (Shapiro–Wilk test). Subsequently, Spearman’s (non-parametric) correlation coefficient was computed for all characters because several characters departed from a normal distribution. No pair of characters showed a strong correlation (arbitrary value >0·95) that would potentially distort further computations.

To gain insight into the overall phenetic relationships and to estimate the level of congruence among morphological, genome size and genetic (AFLP) variation, non-metric multidimensional scaling (NMDS, based on Euclidean distance and standardization by variable range; Kruskal, 1964) and canonical discriminant analysis (CDA; Krzanowski, 1990) were performed. Population samples used in the NMDS analyses were characterized by mean values of all 40 characters. The CDA analyses were based on individual plants and 26 quantitative and ratio characters. Qualitative characters were excluded because they were uniform in some groups of individuals. The analyses were based both on the whole set of data (NMDS 1) and on the partial datasets representing genetic (sub)lineages as revealed by Bayesian clustering (NMDS 2–6, CDA 1–5). The analyses were carried out using STATISTICA 9·1 (StatSoft Inc., 2013), Canoco 5·0 (ter Braak and Šmilauer 2012) and SAS 9·3 (SAS Institute Inc., 2011).

Ecological and phenological trait analyses

To reveal the life strategies of populations of the C. napulifer group, the following ecological and phenological data were collected for each population: altitude, type of bedrock, type of biotope and flowering time (Table S2). These data, together with selected morphological characters describing the root system (Table S2), were subjected to NMDS analyses (NMDS 7) in the same way as morphological characters alone. Nei’s gene diversity, frequency down-weighted marker values and mean DNA content of particular populations were entered as supplementary characters.

RESULTS

AFLP analyses

The overall reproducibility of AFLP dataset 1 (entire dataset) was high because the error rate was minute, being only 0·19%. The analysis yielded 442 unambiguously scorable fragments from 460 profiles and four selective primer combinations. Fragment length ranged from 51 to 488bp, and 436 (98·6%) of fragments were polymorphic. The overwhelming majority of profiles (453, 99·1%) were unique, and identical ones were fairly rare (Supplementary Data Table S3). The average (± s.d.) number of fragments per population varied from 60±4·8 to 91±4·4 in diploids and from 75±7·4 to 110±1·6 in tetraploids (Table S3).

All three genetic analyses, namely NJ (not shown), NN (Fig. 2) and PCoA (Supplementary Data Fig. S1), all based on AFLP dataset 1, revealed essentially similar patterns, with the ‘core’ genetic lineage encompassing solely diploids from the Balkan Peninsula, assignable to the C. napulifer group. This core lineage indicates that the C. napulifer group represents a genetically well-defined unit. Samples outside the core genetic lineage formed well-delimited and species-specific genetic lineages with high bootstrap support (Fig. 2, Fig. S1): C. epirotus (2x), C thirkei (2x) and C. pichleri (4x) from the Balkan Peninsula, and C. fuscomarginatus (4x) from Crimea. Exceptions were populations of C. graminifolius (4x) from south-eastern France and Calabria, which clustered together with populations of C. lingulatus (2x) from Spain and Morocco.

Bayesian clustering based on AFLP dataset 2 (the C. napulifer group not including C. thirkei; Fig. 3B) yielded stable results (with a similarity coefficient between pairs of runs equal to 1) only for K = 5. The results support the recognition of five genetic lineages, the composition of which largely contradicts recent taxonomic concepts: (1) C. napulifer and some accessions of C. orbelicus from Bulgaria (C. napulifer + C. orbelicus BU; coloured pink in Fig. 3B); (2) C. orbelicus from Serbia and from Bulgaria, and C. velenovskyi from Greece (coloured blue in Fig. 3B); (3) C. tuberosus from Greece and the border region of Greece and the Former Yugoslav Republic of Macedonia, C. velenovskyi, C. nissanus from Serbia, and C. orbelicus from Mt Vichren (Bulgaria) (coloured yellow in Fig. 3B); (4) C. tuberosus from the central Former Yugoslav Republic of Macedonia (C. tuberosus MK; coloured orange in Fig. 3B); and (5) C. tuberosus from Croatia (C. tuberosus HR; coloured red in Fig. 3B). The last lineage was genetically most distinct from all the others. Obvious traces of genetic admixture were found in four populations (TRI 151, TRI 153, TRI 265 and TRI 273). Further NJ (not shown), NN (Fig. 3A) and PCoA analyses (Supplementary Data Fig. S2) based on AFLP dataset 2 showed deeper structuring within the second and third genetic lineages that correlated with geographical origin. Within the second genetic lineage (coloured blue in Fig. 3B), samples of C. velenovskyi from Greece formed their own well-supported cluster (C. velenovskyi GR sublineage) separated from samples of C. orbelicus from Serbia and Bulgaria (C. orbelicus BU/RS sublineage). The most heterogeneous third lineage (coloured yellow in Fig. 3B) was split into several mostly well-supported sublineages: (1) C. tuberosus from Greece (C. tuberosus GR/MK sublineage), being genetically closest to the C. tuberosus MK lineage; (2) C. velenovskyi from Serbia (C. velenovskyi RS sublineage, TRI 258); (3) C. nissanus from Serbia (C. nissanus RS sublineage, TRI 256); and (4) C. orbelicus from Mt Vichren (C. orbelicus Vich; TRI 151). Whereas C. velenovskyi RS and C. nissanus RS were resolved in close proximity of C. tuberosus GR/MK and C. tuberosus HR in all analyses; the position of C. orbelicus Vich was uncertain and varied across analyses. In summary, C. orbelicus Vich showed genetic affinities not only to C. orbelicus BU/RS and C. napulifer + orbelicus BU (Fig. 3A), but also to populations of C. tuberosus GR/MK (Fig. S2).

The results of Bayesian analysis based on the dataset including the C. napulifer group without the four genetically heterogeneous populations (AFLP dataset 3) revealed the same five groups as resulted from AFLP dataset 2, but with increased bootstrap support and resolution (Fig. S3).

Analyses of species trees in the framework of the multispecies coalescent implemented in SNAPP with different θ priors produced the same topologies, so we only report the results of the second θ prior settings, which led to higher likelihood values and posterior probabilities (PPs). The analyses jointly revealed the C. napulifer group as a genetically well-defined unit (PP 1·00) in sister position (PP 1·00) to a moderately supported group formed by the Balkan species C. pindicola and C. epirotus (PP 0·87). Within the C. napulifer group, the sequence of divergences suggested the first divergence of C. tuberosus HR lineages with high support (PP 0·99). The remaining (sub)lineages of the group had little support, except apparent traces (PP 1·00) of a common evolutionary history of the C. velenovkyi GR and C. orbelicus BU/RS sublineages.

Genetic diversity and divergence statistics are summarized in Table 2 and Table S3. In general, no private fixed fragments were found for the C. napulifer group. Within the group, the greatest number of private fixed fragments was harboured by the genetic lineages C. tuberosus HR (5) and C. nissanus RS (1). The highest values of the rarity index (DW) were found in the following populations: TRI 256 representing C. nissanus RS, and TRI 264 and TRI 271 of C. tuberosus MK. Nei’s gene diversity among populations reached its highest values in populations TRI 264 and TRI 271 of C. tuberosus GR/MK and in two peculiar populations, TRI 265 and TRI 273 (Fig. 5).

Table 2.

Distribution of AFLP fragments across taxa of the genus Cyanus (A) and genetic (sub)lineages of the C. napulifer group (B)

Taxon/genetic (sub)lineage	N_IND	N_PHEN	N_FRAGM	P ( %)	N_DG	N_PR	N_RARE	N_UNI	DW	Nei
(A) Taxa of the genus Cyanus (AFLP dataset 1)
C. napulifer group	264	258	67 ± 5·61	74·88	0	41	287	9	–	0·11
C. pindicola	34	34	69 ± 7·4	42·63	0	4	86	0	–	0·11
C. epirotus	21	21	74 ± 3·8	25·35	0	7	54	0	–	0·07
C. thirkei	40	40	76 ± 4·58	33·41	1	6	70	1	–	0·09
C. pichleri	28	28	84 ± 3·85	35·02	0	3	63	0	–	0·11
C. fuscomarginatus	16	15	99 ± 3·28	23·96	1	6	39	0	–	0·1
C. lingulatus	45	45	81 ± 7·12	36·41	1	9	58	2	–	0·08
C. graminifolius	12	12	104 ± 7·1	20·97	2	1	25	0	–	0·09
(B) Genetic lineages of the C. napulifer group (AFLP dataset 3)
C. napulifer + orbelicus BU	45	44	71 ± 4·51	39·35	1	19	44	5	264	0·09
C. orbelicus BU/RS	32	30	66 ± 4·29	32·25	0	6	25	2	83	0·1
C. velenovskyi GR	24	23	65 ± 4·01	27·51	0	3	29	0	91	0·09
C. velenovskyi RS	8	8	68 ± 5·5	10·95	0	2	11	1	20	0·05
C. nissanus RS	8	8	64 ± 4·63	17·16	1	4	22	1	38	0·08
C. orbelicus Vich	–	–	–	–	–	–	–	–	–	–
C. tuberosus GR/MK	52	52	64 ± 4·9	48·82	0	113	62	5	150	0·11
C. tuberosus MK	40	40	66 ± 6·3	53·55	1	33	71	16	243	0·11
C. tuberosus HR	24	24	72 ± 4·36	27·51	5	12	41	4	243	0·07

Open in a new tab

N_IND, number of individuals analysed; N_PHEN, number of AFLP multilocus phenotypes; N_FRAGM, average number of AFLP fragments per individual ± standard deviation; P (%), percentage of polymorphic markers; N_DG, number of private fixed (diagnostic) fragments; N_PR, number of private (exclusive) fragments; N_RARE, number of rare fragments (present at a frequency of <10 % of the investigated individuals); N_UNI, number of unique fragments; DW, frequency down-weighted marker values; Nei, Nei’s average proportion of pairwise differences between individuals (Nei’s gene diversity).

Fig. 5. — (A) Nei’s gene diversity and (B) frequency down-weighted marker values of *Cyanus napulifer* populations. Dot size is directly proportional to the depicted values (Table S2). The positions and colours of symbols for populations are the same as in Fig. 1.

Although all AMOVAs attributed a substantial part of the variation to within-population variation, a still considerable proportion of the variation was down to variation among populations and (sub)lineages. This implies significant genetic differentiation also at the inter-population and genetic (sub)lineage levels (Table 3).

Table 3.

Analysis of molecular variance (AMOVA) based on AFLP data for 30 populations of the Cyanus napulifer group

Source of variation	d.f.	Percentage of variation
Two-level AMOVA
Among populations	29	52·96^*
Within-population variation	203	47·04
Three-level AMOVA, five genetic lineages
Variation among genetic lineages	4	30·22^*
Among populations within genetic lineages	25	25·49^*
Within-population variation	203	44·29^*
Three-level AMOVA, eight genetic (sub)lineages
Variation among genetic lineages	7	34·59^*
Among populations within genetic lineages	22	20·49^*
Within-population variation	203	44·92^*

Open in a new tab

*P<0·0001; P values were estimated based on a permutation test (10000 randomizations).

cpDNA analyses

The overall length of the trnC-ycf6 and 3′rps16-5′trnK^(UUU) regions was 1081 and 913bp, respectively. The statistical parsimony network comprised 50 trnC-ycf6 haplotypes, while 51 haplotypes were found in the 3′rps16-5′trnK^(UUU) region. Both networks were formed by two loose, closely related groups of haplotypes connected by a few mutation steps. Accordingly, in both parsimony networks, accessions of Cyanus with x = 10, including those of the C. napulifer group, appeared spread across the networks in ancestral as well as in derived positions, but clearly intermingled with accessions of Cyanus with x = 11 (Supplementary Data Fig. S4).

Flow cytometric analyses

Flow cytometric analyses of genome size resulted in high-resolution histograms (Fig. 6); the mean coefficient of variation value of G1 peaks for analysed samples of Cyanus was 1·92 ± 0·44%, and the mean coefficient of variation values of G1 peaks of the internal standards were 1·81 ± 0·41% for S. lycopersicum and 1·99 ± 0·47% for B. perennis.

The individuals of the C. napulifer group included in the study were shown to be exclusively diploid with 2n∼2x∼20, as were all investigated samples of C. nissanus and C. thirkei. Relative DNA content varied within the C. napulifer group by 23·06% at the diploid level, ranging from 1·07 in the C. orbelicus BU/RS sublineage to 1·32 in the C. tuberosus HR lineage. The results revealed minute intra-population variation in DNA content, which varied from 0·37% to 5·70% (Supplementary Data Table S4). Within the C. napulifer group, the largest and significantly different genomes were found in Croatian populations of the C. tuberosus HR lineage. Populations of (sub)lineages C. tuberosus MK, C. tuberosus GR/MK and C. orbelicus Vich had intermediate relative DNA content and differed significantly from populations of C. nissanus RS, C. orbelicus BU/RS, C. velenovskyi GR and C. velenovskyi RS and some populations of C. napulifer + orbelicus BU. All genetic lineages and sublineages represented by several populations were essentially homogeneous with respect to their DNA content. Exceptions were represented by two genetically admixed populations, TRI 153 and TRI 265 (Fig. 7A, Table S4).

Fig. 7. — Box-and-whisker plots of (A) relative DNA content of the *Cyanus napulifer* group (in total 211 individuals) and (B) relative DNA content per monoploid genome for European perennial taxa of the genus *Cyanus* with the base chromosome number of 10 (in total 302 individuals). Samples of the *C. napulifer* group are assigned to genetic lineages and sublineages according to the results of AFLP data analyses. The numbers of plants investigated for DNA content by DAPI flow cytometry are given in brackets.

As regards relative DNA content per monoploid genome, C. pindicola and C. epirotus were closest of all the other taxa under study to the C. napulifer group. On the other hand, C. thirkei, with its significantly lower monoploid relative DNA content, occupies a very distinct position with respect to the C. napulifer group (Fig. 7B, Supplementary Data Table S5). In general, the results of our genome size analyses indicate that relative DNA content tends to be taxon-specific.

Morphometric analyses

The ordination diagram of NMDS 1 (Fig. 8A), based on 34 populations of the C. napulifer group and 40 characters, shows patterns that partly corroborate and partly contradict those revealed by AFLP analyses. Out of the five AFLP lineages revealed by Bayesian clustering (Fig. 3B), only the C. tuberosus HR genetic lineage (coloured red in Fig. 3B) was morphologically homogeneous and separated from all other lineages.

The partial NMDS 2 analysis (Fig. 8B) of seven populations belonging to the C. napulifer + orbelicus BU genetic lineage (marked in pink in Fig. 3B) showed an unambiguously distinct position of population TRI 153. The morphological separation of TRI 153 individuals was supported also by the CDA 1 analysis (124 individuals, two groups, population TRI 153 versus the rest of the lineage; Fig. 8C); these plants differed from the remaining ones mainly in having fewer leaves (LN), smaller uppermost leaves (LUL), a greater number of florets (FEN, FIN), larger involucral bracts (BN, BL) and a broader margin of involucral bracts (AMAW) (Supplementary Data Table S6).

The results of another partial NMDS 3 analysis (Fig. 8D) of seven populations of the second lineage revealed by Bayesian clustering of AFLP data (marked in blue in Fig. 3B) essentially supported the separation of genetic sublineages C. orbelicus BU/RS and C. velenovskyi GR, predicted by NN and PCoA analyses of AFLP data (Fig. 3A, Fig. S2). Only weak separation of these sublineages was revealed by the CDA 2 analysis of 163 individual plants, based purely on quantitative characters (Fig. 8E). Plants of the C. velenovskyi GR sublineage differed in having a shorter foliated part of the stem (SLL/SL), shorter distances between stem leaves (SLL/LN), which were sessile (BUL, BBL), and larger (LUL, LUW, LBL) and more often lobed basal stem leaves (LBLN, LBLA) compared with plants of the C. orbelicus BU/RS sublineage (Table S6).

The results of morphological analyses of 11 populations (NMDS 4, CDA 3; Fig. 8F, G) and 230 individual plants of the third lineage resulting from Bayesian clustering (coloured yellow in Fig. 3B) supported the clear separation of four genetic sublineages: C. velenovskyi GR, C. nissanus RS, C. orbelicus Vich and C. tuberosus GR/MK. The number of leaves (LN), the size and shape of basal stem leaves (LBW, LBLN, LBMW, LBLA) and the length of fimbria of appendages of involucral bracts (AFAL) were identified by the CDA analysis as characters useful for distinguishing between the sublineages morphologically (Table S6).

Analysis NMDS 5 of populations and analysis CDA 4 of 105 individual plants (two groups, population TRI 265 versus the rest of the lineage; Fig. 8H, I) revealed a very distinct position of the genetically admixed population TRI 265 from the fourth Bayesian clustering lineage C. tuberosus MK (coloured orange in Fig. 3B). This population differed markedly in a number of morphological characters, particularly in having larger uppermost stem leaves (LUL, LUW), longer inner florets (FIL), larger involucres (IL) and involucral bracts (BL, BW), broader margins (AMAW, AMMW) and longer fimbriae (AFML) (Table S6).

Finally, we examined the morphological differentiation of the C. tuberosus GR/MK lineage and the C. tuberosus MK sublineage, as their populations overlapped on the NMDS 1 diagram. Analysis NMDS 6 (Fig. 8J) of 14 populations found them to be morphologically close, except for population TRI 265, which again occupied a distinct position. It was therefore impossible to compare the intermediate genetic position of population TRI 273 with morphological data. Not even analysis CDA 5 (265 individuals, without populations TRI 265 and 273; Fig. 8K) revealed any characters that would enable their separation (Table S6).

Congruences and incongruences among genetic, DNA-content and morphological variation

When compared, the results of our AFLP, flow-cytometric and morphometric analyses show wildly differing patterns, from complete congruence to almost full discord. Molecular and DNA content data are congruent in that they indicate a close relationship of C. napulifer to C. epirotus and C. pindicola (Fig. 4 versus Fig. 7). Within the C. napulifer group, the most obvious cases of congruence between the three markers were the C. tuberosus HR and partly also the C. nissanus RS genetic (sub)lineage (Figs 3 and 5 versus Fig. 7, Table S4 versus Fig. 8A). Furthermore, several populations bore traces of genetic admixture, which were clearly apparent, especially in four populations: TRI 151, TRI 153, TRI 265 and 273 (Fig. 3B). This is underlined by their distinct DNA content, morphology, or both (Fig. 3 versus Fig. 7, Table S4 versus Fig. 8B, C, F, G, H, I).

Fig. 4. — Analyses of species trees in the framework of the multispecies coalescent implemented in SNAPP of AFLP data: (A) complete tree set; (B) consensus tree. Posterior probabilities are given at each node.

On the other hand, there were two mutually inverse cases of incongruence between genetic, DNA-content and morphological variation. We found (1) no differences in morphology and relative DNA content between the genetically well-separated lineages C. tuberosus MK and C. tuberosus GR/MK (Fig. 3B versus Fig. 7, Supplementary Data Table S4 versus Fig. 8J, K), which stands in contrast to (2) the poor genetic diversification of C. orbelicus BU/RS and C. velenoversuskyi GR accompanied by their distinct morphology and DNA content (Fig. 3 versus Fig. 7, Supplementary Data Table S4 versus Fig. 8D, E).

Ecological, phenological and morphological traits analyses

The ordination diagram of NMDS 7 (Fig. 9), based on 34 populations of the C. napulifer group and six phenological and ecological characters, shows the separation of the genetic (sub)lineages C. napulifer + orbelicus BU, C. orbelicus BU/RS, C. velenovskyi GR and C. velenovskyi RS from the remaining genetic (sub)lineage along the first axis. The populations on the left side of the diagram include plants with long creeping rhizomes that prefer granite substrates and occur predominantly in alpine meadows in mountainous and subalpine regions (1400–2400m). The populations on the right occur at lower altitudes (300–1500m), in stony and steppe biotopes on calcareous or schismatic bedrock. Interestingly, genome size of plants of the C. napulifer group was negatively correlated with the first axis: plants with rhizomes, which predominated at higher altitudes, had smaller genomes in comparison with geophytes, which occurred at lower altitudes. A slight trend towards higher genetic diversity and divergence in geophytes was also visible.

DISCUSSION

AFLPs help disentangle the evolution and biogeography of Cyanus perennials: new implications for Mediterranean representatives

Perennial members of the genus Cyanus have been notorious for their complex variation, and all attempts to treat them in a clear, straightforward way and to elucidate their evolutionary relationships have essentially failed (Hellwig, 2004; Boršić et al., 2011; Olšavská et al., 2013). Investigations based on various nuclear and plastid DNA sequence markers have resulted in poorly resolved gene trees and networks with particular alleles crossing taxon borders (Boršić et al., 2011; Löser, 2012; see cpDNA analyses in the Results section). Recent investigations into the tribe Carduae (Barres et al., 2013) and the genus Centaurea (Hilpold et al., 2014) estimated by molecular dating that the crown diversification between Cyanus annuals and perennials as well as between the two chromosome lineages (x = 10 and x = 11) of Cyanus perennials dates back to the late Tertiary (Late Miocene/Pliocene; ∼3·8–8·8 million years ago). The extensive sharing of plastid haplotypes among accessions of both chromosome lineages detected in our study thus suggests that there was not enough time between each speciation event for informative mutations to emerge and for lineage sorting to take place (Jakob and Blattner, 2006; Maddison and Knowles, 2006; Moreno-Letelier et al., 2013). The presence of insufficient or contradictory phylogenetic signals may in this case also be attributed to the maintenance of ancestral variation formed by rapid radiation during the first phases of group diversification or, alternatively, to ancient introgression between distinct populations or genetic lineages (e.g. Frajman and Oxelman, 2007; Boršić et al., 2011; Moody and Rieseberg, 2012; Olšavská et al., 2013; Slovák et al., 2014).

Our AFLP analyses uncovered a suite of closely related genetic lineages among European and North African perennial members of the genus Cyanus with x = 10. Furthermore, they revealed two unequal levels of genetic and DNA content differentiation: (1) the core grouping corresponding to the C. napulifer group and possessing a more diffuse genetic and genome size structure; and (2) the remaining samples, forming strongly diversified genetic lineages with no further geographical patterns (in spite of their broad distribution; e.g. C. lingulatus and C. pichleri). We suggest that the observed patterns reflect different historical time frames of evolution. Whereas the diffuse AFLP pattern occurring in the core C napulifer group can be attributed to rather recent extensive radiation (see also below), the distinct genetic lineages formed by the remaining samples might be explained by long-term geographical or reproductive isolation.

If we extrapolate from the presented data, it seems that there is no straightforward biogeographical explanation for the origin of Mediterranean Cyanus perennials with x = 10. We assume that multiple colonization events took place from Western Asia to the Mediterranean via both southern and northern migration routes across land bridges that recurrently emerged and disappeared during Tertiary and Quaternary climatic oscillations (Petit et al., 2003; Quézel, 2004; Sanmartín et al., 2010; Hewitt 2011). For example, the genetic divergence of C. lingulatus and C. graminifolius from the remaining taxa (Fig. 1, Fig. S1) might indicate colonization of the West Mediterranean through North Africa (cf. Barres et al., 2013; Hilpold et al., 2014).

Notes on polyploid evolution in Mediterranean Cyanus perennials with x = 10

AFLPs proved to be a powerful tool also in tracing the origin of polyploids in the genus Cyanus. The lack of genetic divergence between diploids and tetraploids in C. pindicola (Greece) and between diploid samples of C. lingulatus (Spain and Marocco) versus tetraploids of C. graminifolius (south-eastern France and Calabria) supports the autopolyploid origin of these tetraploids. The relatively distinct position of C. graminifolius samples from Calabria in our genetic analysis (Fig. 2, Fig. S1) might be explained by long-term isolation from their close relatives. Alternatively, Calabrian tetraploids could be of hybrid origin, most likely derived from a cross between a tetraploid formed by polyploidization of C. lingulatus and another tetraploid related to C. pichleri from Greece, which possibly came into contact especially during the Pleistocene era (Petit et al., 2003; Hewitt, 2011, Barres et al., 2013; Hilpold et al., 2014).

Allopatric speciation within the C. napulifer group accompanied by long-term isolation and hybridization

Our data prove that the C. napulifer group is a genetically well-defined unit with respect to other members of the genus Cyanus with x = 10. Although C. thirkei had until recently been considered a member of the C. napulifer group (Bancheva and Raimondo 2003; Olšavská et al., 2013), our data prove that it is genetically much closer to C. pichleri from Greece than to the C. napulifer group.

We assume that (sub)lineages of the C. napulifer group evolved after rapid and relatively recent diversification. Incongruences in genetic, DNA-content and morphological variation between allopatric groups of populations (C. tuberosus MK and C. tuberosus GR/MK, C. orbelicus BU/RS and C. velenovskyi GR, see the Results section) indicate that the speciation processes in this group are still ongoing and that not enough time has elapsed for these lineages to form completely clear-cut biological entities (Alarcón et al., 2012; Slovák et al., 2012a).

It seems that the new entities within the C. napulifer group originated in allopatry after fragmentation of the once continuous distribution area of their common ancestor (cf. Kropf et al., 2008, 2012; Bardy et al., 2010; Hilpold et al., 2011; Alarcón et al., 2012; Dobeš et al., 2013; Kuzmanović et al., 2013; Manafzadeh et al., 2014). Spatial isolation is considered one of the crucial mechanisms of speciation in homoploid plant groups (Watanabe, 1986; Gross and Rieseberg, 2005; Martín-Bravo et al., 2010), especially as members of the C. napulifer group are unlikely to be reproductively isolated (cf. Olšavská and Löser, 2013). Noteworthy is the patchy nature of the Balkan landscape (Stevanović, 1996; Reed et al., 2004; Meshinev, 2007; Velikov and Stoyanova, 2007), where numerous borders between geomorphological units probably served as effective barriers to species migration or gene flow.

Both of the genetically most distinct (sub)lineages C. tuberosus HR and C. velenovskyi RS occur at the margin of the C. napulifer group’s distribution area. We should therefore also consider a peripatric speciation scenario (Barraclough and Vogler, 2000; Ikeda et al., 2012). These (sub)lineages might be descendants of peripheral populations that evolved and diversified in isolation following range contractions of their more widespread parental ancestors (Castellanos-Morales et al., 2016 and references therein). Whether the (sub)lineages diversified in allopatry or peripatry, the accumulation of private fixed fragments and the high values of the DW index and Nei’s gene diversity (Fig. 5) indicate their long-term in situ survival in habitats that most probably served as glacial refugia (Stehlik et al., 2002; Schönswetter et al., 2005, 2009; Kropf et al., 2006; López-Sepúlveda et al., 2013; Kučera et al., 2013).

Apparent genetic admixture in four populations from the C. napulifer group (TRI 151, TRI 153, TRI 265 and 273) most probably evidences more recent hybridization events followed by genetic homogenization towards one of the parental taxa. Nevertheless, traces of hybridization are still apparent at the morphological and DNA content levels. Secondary contacts between taxa or their lineages were most probably triggered by their range expansions and contractions during the Pleistocene (inter)glacial periods (e.g. Surina et al., 2011; Alarcón et al., 2012).

Distinct life strategies of the C. napulifer group point to adaptive radiation

Adaptations of organisms to ecologically divergent niches possibly triggered their extensive diversification and mostly allopatric speciation (Thompson, 1999; Gross and Rieseberg, 2005; Thompson et al., 2005; López-Sepúlveda et al., 2013). It seems that the climatic and topographical heterogeneity of the Balkan Peninsula (Stevanović, 1996; Meshinev, 2007; Velikov and Stoyanova, 2007; Radford and Odé, 2009) played a crucial role in the diversification of the C. napulifer group. The adaptation process in the C. napulifer group has resulted in an array of different morphological and ecological traits, some of which even lead to different life strategies. Within the group, there are two types of plants with fundamentally different life strategies: (1) geophytes possessing a tuberous root system and (2) plants with long creeping rhizomes. Species from the first group use nutrition from tubers to very quickly develop a flowering stem in May/June. During the rest of the vegetation season, they form new stocks of nutrition, and during the hot and dry summer period they wither completely (e.g. C. tuberosus HR, MK, GR/MK and C. nissanus RS). Members of this group mainly inhabit open stony and steppe communities in the seasonally arid Mediterranean climate. The second group includes plants with long creeping rhizomes allowing persistence and clonal dispersal in montane and alpine grasslands dominated by highly competitive clonal plants. As expected, clonality is connected with lower genetic diversity within populations (Figs 5 and 9). The finding of a higher DNA content in populations from lower altitudes contrasts with results on genome size variation in the C. triumfetti group (Olšavská et al., 2012). We therefore postulated that the increase in DNA content in the C. napulifer group could be linked to geophytism. In general, species with large genomes are largely represented by geophytes (Veselý et al., 2012 and reference therein).

Conclusions

Our multi-approach study sheds new light on the evolution and biogeography of Cyanus perennials with x = 10 occurring in the Mediterranean and Northern Africa. Their overall genetic patterns within the group suggest that multiple colonization events took place from Western Asia to the Mediterranean via both southern and northern migration routes.

Most importantly, the results of genetic analyses support the status of the C. napulifer group as a genetically well-defined unit and an example of a rapidly and recently diversified plant group with several strictly allopatric genetic lineages and sublineages. Incongruences among genetic, DNA-content and morphological variation point to still ongoing speciation processes and secondary contacts between genetic lineages in the past. The climatic and topographical heterogeneity of the Balkan Peninsula contributed markedly to the adaptive radiation of this intricate group of mountain plants.

The presented data show only little congruence with the current taxonomy of the C. napulifer group. The overwhelming majority of genetic lineages detected within the group, however, represent bio-ecologically well-defined entities and thus deserve formal taxonomic classification at the species or subspecies rank (K. Olšavská et al., unpubl. res.). Regardless of which rank is assigned to these entities, they must be regarded as valuable elements of Balkan biodiversity.

SUPPLEMENTARY DATA

Supplementary data are available online at www.aob.oxfordjournals.org and consist of the following. Table S1: material investigated for cpDNA (including GenBank accession numbers). Table S2: list of morphological, ecological and phenological characters and their abbreviations used in this study. Table S3: distribution of AFLP fragments across the investigated populations. Table S4: relative DNA content of the C. napulifer group and results of the Tukey–Kramer test. Table S5: relative DNA content and relative DNA content per monoploid genome for European perennial taxa of the genus Cyanus with the base chromosome number of x = 10 and results of the Tukey–Kramer test. Table S6: total canonical structure expressing correlations of morphological characters with canonical axes. Figure S1: principal coordinates analyses based on AFLP profiles of 460 Cyanus individuals. Figure S2: principal coordinates analyses based on AFLP profiles of the C. napulifer group; populations of presumably hybrid origin were omitted. Figure S3: neighbour-net diagram based on AFLP profiles of the C. napulifer group; populations with presumably hybrid origin were omitted. Figure S4: maximum parsimony network of cpDNA haplotypes of Cyanus plants.

Supplementary Data

supp_118_6_1071__index.html^{(686B, html)}

ACKNOWLEDGEMENTS

We thank Eliška Záveská for help with species tree analyses in SNAPP. This work was financially supported by the Slovak Research and Development Agency (APVV-0320-10) and the Scientific Grant Agency of the Slovak Republic (VEGA 2/0096/15). It is also a result of the implementation of project ITMS 6240120014 (The Centre of Excellence for Biodiversity and Land-Use Conservation), supported by the Research and Development Operational Programme funded by the European Regional Development Fund (ERDF). STRUCTURE calculations were performed at the Computing Centre of the Slovak Academy of Sciences using the infrastructure acquired within projects ITMS 26230120002 and ITMS 26210120002 (Slovak infrastructure for high-performance computing), supported by the Research and Development Operational Programme funded by the ERDF.

LITERATURE CITED

Alarcón M, Vargas P, Sáez L, Molero J, Aldasoro JJ. 2012. Genetic diversity of mountain plants: two migration episodes of Mediterranean Erodium (Geraniaceae). Molecular Phylogenetics and Evolution 63: 866–876. [DOI] [PubMed] [Google Scholar]
Bancheva S, Raimondo FM. 2003. Biosystematic studies of seven Balkan species from genus Cyanus (Compositae). Bocconea 16: 507–527. [Google Scholar]
Bancheva S, Stoyanov S. 2009. A new species of Cyanus (Asteraceae, Centaureinae) from Southeastern Bulgaria. Novon 19: 421–425. [Google Scholar]
Barraclough TG, Vogler AP, 2000. Detecting the geographical pattern of speciation from species-level phylogenies. American Naturalist 155: 419–434. [DOI] [PubMed] [Google Scholar]
Bardy KE, Albach DC, Schneeweiss GM, Fischer MA, Schönswetter P. 2010. Disentangling phylogeography, polyploid evolution and taxonomy of a woodland herb (Veronica chamaedrys group, Plantaginaceae s.l.) in southeastern Europe. Molecular Phylogenetics and Evolution 57: 771–786. [DOI] [PMC free article] [PubMed] [Google Scholar]
Barres L, Sanmartín I, Anderson CL. et al., 2013. Reconstructing the evolution and biogeographic history of tribe Cardueae (Compositae). American Journal of Botany 100: 867–882. [DOI] [PubMed] [Google Scholar]
Bonin A, Bellemain E, Eidesen PB, Pompanon F, Brochmann C, Taberlet P. 2004. How to track and assess genotyping errors in population genetic studies. Molecular Ecology 13: 3261–3273. [DOI] [PubMed] [Google Scholar]
Boršić I, Susanna A, Bancheva S, Garcia-Jacas N. 2011. Centaurea sect. Cyanus: nuclear phylogeny, biogeography and life-form evolution. International Journal of Plant Sciences 172: 238–249. [Google Scholar]
Bouckaert R, Heled J. 2014. DensiTree 2: seeing trees through the forest. bioRxiv: 012401. doi:http://dx.doi.org/10.1101/012401. [Google Scholar]
ter Braak CJF, Šmilauer P. 2012. Canoco reference manual and user’s guide: software for ordination, version 5.0. Ithaca: Microcomputer Power. [Google Scholar]
Bryant D, Bouckaert R, Felsenstein J, Rosenberg NA, RoyChoudhury A. 2012. Inferring species trees directly from biallelic genetic markers: bypassing gene trees in a full coalescent analysis. Molecular Biology and Evolution 29: 1917–1932. [DOI] [PMC free article] [PubMed] [Google Scholar]
Castellanos-Morales C, Gámez N, Castillo-Gámez RA, Eguiarte LE. 2016. Peripatric speciation of an endemic species driven by Pleistocene climate change: the case of the Mexican prairie dog (Cynomys mexicanus). Molecular Phylogenetics and Evolution 94: 171–181. [DOI] [PubMed] [Google Scholar]
Castresana J. 2000. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution 17: 540–552. [DOI] [PubMed] [Google Scholar]
Clement M, Posada D, Crandall K. 2000. TCS: a computer program to estimate gene genealogies. Molecular Ecology 9: 1657–1660. [DOI] [PubMed] [Google Scholar]
Dobeš C, Rossa J, Paule J, Hülber K. 2013. Morphology, DNA-molecular variation, karyology, ecogeography, and phytosociology suggest allopatric differentiation and species rank for Potentilla rigoana (Rosaceae). Taxon 62: 733–745. [Google Scholar]
Doležel J, Sgorbati S, Lucretti S. 1992. Comparison of three DNA ﬂuorochromes for ﬂow cytometric estimation of nuclear DNA content in plants. Journal of Plant Physiology 85: 625–631. [Google Scholar]
Eastwood WJ. 2004. East Mediterranean vegetation and climate change In: Griffiths HI, Kryštufek B, Reed JM, eds. Balkan biodiversity. Pattern and process in the European hotspot. Dordrecht: Kluwer Academic Publishers, 25–48. [Google Scholar]
Ehrich D. 2006. AFLPDAT: a collection of R functions for convenient handling of AFLP data. Molecular Ecology Notes 6: 603–604. [Google Scholar]
Evanno G, Regnaut S, Goudet J. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14: 2611–2620. [DOI] [PubMed] [Google Scholar]
Excoffier L, Laval G, Schneider S. 2005. Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1: 47–50. [PMC free article] [PubMed] [Google Scholar]
Falush D, Stephens M, Pritchard JK. 2007. Inference of population structure using multilocus genotype data: dominant markers and null allele. Molecular Ecology Notes 7: 574–578. [DOI] [PMC free article] [PubMed] [Google Scholar]
Frajman B, Oxelman B. 2007. Reticulate phylogenetics and phytogeographical structure of Heliosperma (Sileneae, Caryophyllaceae) inferred from chloroplast and nuclear DNA sequences. Molecular Phylogenetics and Evolution 43: 140–155. [DOI] [PubMed] [Google Scholar]
Frajman B, Schneeweiss GM. 2009. A campanulaceous fate: the Albanian stenoendemic Asyneuma comosiforme in fact belongs to isophyllous Campanula. Systematic Botany 34: 595–601. [Google Scholar]
Greuter W. 2006–2009. Compositae (pro parte majore) In: Greuter W, von Raab-Straube E. Compositae. Euro+Med Plantbase – the information resource for Euro-Mediterranean plant diversity. http://ww2.bgbm.org/EuroPlusMed. [Google Scholar]
Gross BL, Rieseberg LH. 2005. The ecological genetics of homoploid hybrid speciation. Journal of Heredity 96: 241–252. [DOI] [PMC free article] [PubMed] [Google Scholar]
Hall TA. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95–98. [Google Scholar]
Hellwig FH. 2004. Centaureinae (Asteraceae) in the Mediterranean: history of ecogeographical radiation. Plant Systematics and Evolution 246: 137–162. [Google Scholar]
Hewitt GM. 2011. Mediterranean peninsulas: the evolution of hotspots In: Frank E, Zachos FE, Habel JC, eds. Biodiversity hotspots. Distribution and protection of conservation. Berlin: Springer, 123–147. [Google Scholar]
Hilpold A, Schönswetter P, Susanna A, Garcia-Jacas N, Vilatersana R. 2011. Evolution of the central Mediterranean Centaurea cineraria group (Asteraceae): evidence for relatively recent, allopatric diversification following transoceanic seed dispersal. Taxon 60: 528–538. [Google Scholar]
Hilpold A, Vilatersana R, Susanna A. et al. 2014. Phylogeny of the Centaurea group (Centaurea, Compositae) – geography is a better predictor than morphology. Molecular Phylogenetics and Evolution 77: 195–215. [DOI] [PubMed] [Google Scholar]
Huson DH, Bryant D. 2006. Application of phylogenetic networks in evolutionary studies. Molecular Biology and Evolution 23: 254–267. [DOI] [PubMed] [Google Scholar]
Ikeda H, Carlsen T, Fujii N, Crochmann C, Setoguchi H, 2012. Pleistocene climatic oscillations and the speciation history of an alpine endemic and a widespread arctic-alpine plant. New Phytologist 194: 583–594. [DOI] [PubMed] [Google Scholar]
Jakob SS, Blattner FR. 2006. A chloroplast genealogy of Hordeum (Poaceae): long-term persisting haplotypes, incomplete lineage sorting, regional extinction, and the consequences for phylogenetic inference. Molecular Biology and Evolution 23: 1602–1612. [DOI] [PubMed] [Google Scholar]
Jakobsson M, Rosenberg NA. 2007. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 23: 1801–1806. [DOI] [PubMed] [Google Scholar]
Kaya Z, Bancheva S. 2009. A new species of Cyanus (Centaurea p.p.) sect. Napuliferi (Asteraceae) from Turkey. Novon 19: 175–177. [Google Scholar]
Kolarčik V, Zozomová-Lihová J, Mártonfi P. 2010. Systematics and evolutionary history of the Asterotricha group of the genus Onosma (Boraginaceae) in central and southern Europe inferred from AFLP and nrDNA ITS data. Plant Systematics and Evolution 290: 21–45. [Google Scholar]
Kolář F, Fuxová G, Záveská E, et al. 2016. Northern glacial refugia and altitudinal niche divergence shape genome-wide differentiation in the emerging plant model Arabidopsis arenosa. Molecular Ecology, in press doi:10.1111/mec.13721. [DOI] [PubMed] [Google Scholar]
Kropf M, Comes HP, Kadereit JW. 2006. Long-distance dispersal vs vicariance: the origin and genetic diversity of alpine plants in the Spanish Sierra Nevada. New Phytologist 172: 169–184. [DOI] [PubMed] [Google Scholar]
Kropf M, Comes HP, Kadereit JW. 2008. Causes of the genetic architecture of south-west European high mountain disjuncts. Plant Ecology & Diversity 1: 217–228. [Google Scholar]
Kropf M, Comes HP, Kadereit JW. 2012. Past, present and future of mountain species of the French Massif Central – the case of Soldanella alpina L. subsp. alpine (Primulaceae) and a review of other plant and animal studies. Journal of Biogeography 39: 799–812. [Google Scholar]
Kruskal JB. 1964. Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis. Psychometrika 29: 1–27. [Google Scholar]
Kryštufek B, Reed JM. 2004. Patterns and process in Balkan biodiversity: an overview In: Griffiths HI, Kryštufek B, Reed JM, eds. Balkan biodiversity. Pattern and process in the European hotspot. Dordrecht: Kluwer Academic Publishers, 1–8. [Google Scholar]
Krzanowski WJ. 1990. Principles of multivariate analysis. Oxford: Clarendon Press. [Google Scholar]
Kučera J, Marhold K, Lihová J. 2010. Cardamine maritima group (Brassicaceae) in the amphi-Adriatic area: a hotspot of species diversity revealed by DNA sequences and morphological variation. Taxon 59: 148–164. [Google Scholar]
Kučera J, Turis P, Zozomová-Lihová J, Slovák M. 2013. Cyclamen fatrense (Primulaceae), only a myth or a true West Carpathian endemic? Genetic and morphological evidence. Preslia 85: 133–158. [Google Scholar]
Kuzmanović N, Comanescu P, Frajman B. et al. 2013. Genetic, cytological and morphological differentiation within the Balkan-Carpathian Sesleria rigida sensu Fl. Eur. (Poaceae): a taxonomically intricate tetraploid-octoploid complex. Taxon 62: 458–472. [Google Scholar]
Lakušić D, Liber Z, Nikolić T. et al. 2013. Molecular phylogeny of the Campanula pyramidalis species complex (Campanulaceae) inferred from chloroplast and nuclear non-coding sequences and its taxonomic implications. Taxon 62: 505–524. [Google Scholar]
López-Sepúlveda P, Tremetsberger K, Ángeles Ortiz M, Baeza CM, Peñailillo P, Stuessy TF. 2013. Radiation of the Hypochaeris apargioides complex (Asteraceae: Cichorieae) of southern South America. Taxon 62: 550–564. [Google Scholar]
Löser CJ. 2012. Speciation and genomic conflicts in Cyanus Mill. PhD Thesis, Friedrich Schiller University Jena, Germany.
Maddison WP, Knowles LL. 2006. Inferring phylogeny despite incomplete lineage sorting. Systematic Biology 55: 21–30. [DOI] [PubMed] [Google Scholar]
Manafzadeh S, Salvo G, Conti E. 2014. A tale of migrations from east to west: the Irano-Turanian floristic region as a source of Mediterranean xerophytes. Journal of Biogeography 41: 366–379. [Google Scholar]
Marhold K, Kudoh H, Pak J-H, Watanabe K, Španiel S, Lihová J. 2010. Cytotype diversity and genome size variation in eastern Asian polyploid Cardamine (Brassicaceae) species. Annals of Botany 105: 249–264. [DOI] [PMC free article] [PubMed] [Google Scholar]
Martín-Bravo S, Valcárcel V, Vargas P, Luceño M. 2010. Geographical speciation related to Pleistocene range shifts in the western Mediterranean mountains (Reseda sect. Glaucoreseda, Resedaceae). Taxon 59: 466–482. [Google Scholar]
Mereďa P, Hodálová I, Mártonfi P, Kučera P, Lihová J. 2008. Intraspecific variation in Viola suavis in Europe: parallel evolution of white-flowered morphotypes. Annals of Botany 102: 443–462. [DOI] [PMC free article] [PubMed] [Google Scholar]
Meshinev T. 2007. Vegetation and phytogeography: a brief characteristic In: Fet V, Popov A, eds. Biogeography and ecology of Bulgaria. Dordrecht: Springer, 581–588. [Google Scholar]
Moody ML, Rieseberg LH. 2012. Sorting through the chaff, nDNA gene trees for phylogenetic inference and hybrid identification of annual sunflowers (Helianthus sect. Helianthus). Molecular Phylogenetics and Evolution 64: 145–155. [DOI] [PubMed] [Google Scholar]
Moreno-Letelier A, Ortíz-Medrano A, Piñero D. 2013. Niche divergence versus neutral processes: combined environmental and genetic analyses identify contrasting patterns of differentiation in recently diverged pine species. PLoS One 8: e78228. [DOI] [PMC free article] [PubMed] [Google Scholar]
Nei M, Li WH. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy of Sciences of the USA 76: 5269–5273. [DOI] [PMC free article] [PubMed] [Google Scholar]
Olšavská K, Löser CJ. 2013. Mating system and hybridization of the Cyanus triumfetti and C. montanus groups (Asteraceae). Folia Geobotanica 48: 537–554. [Google Scholar]
Olšavská K, Perný M, Kučera J, Hodálová I. 2011. Biosystematic study of the Cyanus triumfetti group in Central Europe. Preslia 83: 59–98. [Google Scholar]
Olšavská K, Perný M, Španiel S, Šingliarová B. 2012. Nuclear DNA content variation among perennial taxa of the genus Cyanus (Asteraceae) in Central Europe and adjacent areas. Plant Systematics and Evolution 298: 1463–1482. [Google Scholar]
Olšavská K, Perný M, Löser CJ, Stimper R, Hodálová I. 2013. Cytogeography of European perennial species of Cyanus (Asteraceae). Botanical Journal of the Linnean Society 173: 230–257. [Google Scholar]
Peev D, Delcheva M. 2007. Florogeographical elements and evolutionary trends in the Bulgarian flora In: Fet V, Popov A, eds. Biogeography and ecology of Bulgaria. Dordrecht: Springer, 571–579. [Google Scholar]
Perný M, Tribsch A, Anchev ME. 2005. Infraspecific differentiation in diploid Cardamine acris (Brassicaceae) from Balkan Peninsula: morphological and molecular evidence. Folia Geobotanica 39: 405–429. [Google Scholar]
Petit RJ, Aguinagalde I, de Beaulieu J-L. et al. 2003. Glacial refugia: hotspots but not melting pots of genetic diversity. Science 300: 1563–1565. [DOI] [PubMed] [Google Scholar]
Quézel P. 2004. Large-scale post-glacial distribution of vegetation structures in the Mediterranean region In: Mazzoleni S, di Pasquale G, Mulligan M, di Martino P, Rego F, eds. Recent dynamics of the Mediterranean vegetation and landscape. Chichester: John Wiley & Sons, 3–12. [Google Scholar]
Radford EA, Odé B, eds. 2009. Conserving important plant areas: investing in the Green Gold of South East Europe. Salisbury: Plantlife International. [Google Scholar]
Reed JM, Kryštufek B, Eastwood WJ. 2004. The physical geography of the Balkans and nomenclature of place names In: Griffiths HI, Kryštufek B, Reed JM, eds. Balkan biodiversity. Pattern and process in the European hotspot. Dordrecht: Kluwer Academic Publishers, 9–22. [Google Scholar]
Ronikier M, Zalewska-Gałosz J. 2014. Independent evolutionary history between the Balkan ranges and more northerly mountains in Campanula alpina s.l. (Campanulaceae): genetic divergence and morphological segregation of taxa. Taxon 63: 116–131. [Google Scholar]
Rosenberg NA. 2004. DISTRUCT: a program for the graphical display of population structure. Molecular Ecology Notes 4: 137–138. [Google Scholar]
Saitou N, Nei M. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4: 406–425. [DOI] [PubMed] [Google Scholar]
Sanmartín I, Anderson CL, Alarcon M, Ronquist F, Aldasoro JJ. 2010. Bayesian island biogeography in a continental setting: the Rand Flora case. Biology Letters 6: 703–707. [DOI] [PMC free article] [PubMed] [Google Scholar]
SAS Institute Inc. 2011. SAS®9.3 Product documentation. Cary: SAS Institute. [Google Scholar]
Schönswetter P, Schneeweiss GM. 2009. Androsace komovensis sp. nov., a long mistaken local endemic from the southern Balkan Peninsula with biogeographic links to the Eastern Alps. Taxon 58: 544–549. [Google Scholar]
Schönswetter P, Stehlik I, Holderegger R, Tribsch A. 2005. Molecular evidence for glacial refugia of mountain plants in the European Alps. Molecular Ecology 14: 3547–3555. [DOI] [PubMed] [Google Scholar]
Schönswetter P, Suda J, Popp M, Weiss-Schneeweiss H, Brochmann C. 2007. Circumpolar phylogeography of Juncus biglumis (Juncaceae) inferred from AFLP ﬁngerprints, cpDNA sequences, nuclear DNA content and chromosome numbers. Molecular Phylogenetics and Evolution 42: 92–103. [DOI] [PubMed] [Google Scholar]
Schönswetter P, Solstad H, Garcìa PE, Elven R. 2009. A combined molecular and morphological approach to the taxonomically intricate European mountain plant Papaver alpinum s.l. (Papaveraceae) – taxa or informal phylogeographical groups? Taxon 17: 1326–1343. [Google Scholar]
Schlüter PM, Harris SA. 2006. Analysis of multilocus fingerprinting data sets containing missing data. Molecular Ecology Notes 6: 569–572. [Google Scholar]
Shaw J, Lickey E, Beck JT. et al. 2005. The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal of Botany 92: 142–166. [DOI] [PubMed] [Google Scholar]
Shaw J, Lickey E, Schilling EE, Small RL. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. American Journal of Botany 94: 275–288. [DOI] [PubMed] [Google Scholar]
Slovák M, Kučera J, Marhold K, Zozomová-Lihová J. 2012a. The morphological and genetic variation in the polymorphic species Picris hieracioides (Compositae, Lactuceae) in Europe strongly contrasts with traditional taxonomical concepts. Systematic Botany 37: 258–278. [Google Scholar]
Slovák M, Kučera J, Turis P, Zozomová-Lihová J. 2012b. Multiple glacial refugia and postglacial colonisation routes inferred for a woodland geophyte, Cyclamen purpurascens: patterns concordant with the Pleistocene history of broadleaved and coniferous tree species. Biological Journal of the Linnean Society 105: 741–760. [Google Scholar]
Slovák M, Kučera J, Záveská E, Vďačný P. 2014. Dealing with discordant genetic signal caused by hybridisation, incomplete lineage sorting and paucity of primary nucleotide homologies: a case study of closely related members of the genus Picris subsection Hieracioides (Compositae). PLoS One 9: e104929. [DOI] [PMC free article] [PubMed] [Google Scholar]
StatSoft Inc. 2013. Electronic statistics textbook. Tulsa: Statsoft; http://www.statsoft.com/Textbook. [Google Scholar]
Stehlik I, Blattner FR, Holderegger R, Bachmann K. 2002. Nunatak survival of the high Alpine plant Eritrichium nanum (L.) Gaudin in the central Alps during the ice ages. Molecular Ecology 11: 2027–2036. [DOI] [PubMed] [Google Scholar]
Stefanović S, Lakušić D, Kuzmina M, Mededović S, Tan K, Stevanović V. 2008. Molecular phylogeny of Edraianthus (grassy bells; Campanulaceae) based on non-coding plastid DNA sequences. Taxon 57: 452–475. [Google Scholar]
Stevanović V. 1996. Analysis of the Central European and Mediterranean orophytic element on the mountains of the W. and Central Balkan Peninsula, with special reference to endemics. Bocconea 5: 77–97. [Google Scholar]
Stevanović V, Tan K, Petrova A. 2007. Mapping the endemic flora of the Balkans – a progress report. Bocconea 21: 131–137. [Google Scholar]
Suda J, Trávníček P, Mandák B, Berchová-Bímová K. 2010. Genome size as a marker for identifying the invasive alien taxa in Fallopia section Reynoutria. Preslia 82: 97–106. [Google Scholar]
Surina B, Schönswetter P, Schneeweiss GM. 2011. Quaternary range dynamics of ecologically divergent species (Edraianthus serpyllifolius and E. tenuifolius, Campanulaceae) within the Balkan refugium. Journal of Biogeography 38: 1381–1393. [Google Scholar]
Talavera G, Castresana J. 2007. Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Systematic Biology 56: 564–577. [DOI] [PubMed] [Google Scholar]
Thompson JD. 1999. Population differentiation in Mediterranean plants: insights into colonisation history and implications for species diversification. Heredity 82: 229–236. [DOI] [PubMed] [Google Scholar]
Thompson JD, Lavergne S, Affre L, Gaudeu M, Debussche M. 2005. Ecological differentiation of Mediterranean endemic plants. Taxon 54: 967–976. [Google Scholar]
Tukey JW. 1977. Exploratory data analysis. Reading, MA: Addison-Wesley. [Google Scholar]
Tzedakis PC. 2004. The Balkans as prime glacial refugial territory of European temperate trees In: Griffiths HI, Kryštufek B, Reed JM, eds. Balkan biodiversity. Pattern and process in the European hotspot. Dordrecht: Kluwer Academic Publishers, 49–68. [Google Scholar]
Velikov V, Stoyanova M. 2007. Landscapes and climate of Bulgaria In: Fet V, Popov A, eds. Biogeography and ecology of Bulgaria. Dordrecht: Springer, 589–605. [Google Scholar]
Veselý P, Bureš P, Šmarda P, Pavlíček T. 2012. Genome size and DNA base composition of geophytes: the mirror of phenology and ecology? Annals of Botany 109: 65–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
Vos P, Hogers R, Bleeker M. et al. 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23: 4407–4414. [DOI] [PMC free article] [PubMed] [Google Scholar]
Watanabe K. 1986. Cytogeography of the genus Eupatorium (Compositae). A review. Plant Species Biology 1: 99–116. [Google Scholar]
Yeh FC, Yang RC, Timothy BJ, Ye Z, Judy M. 1997. Pop Gene, the user-friendly shareware for population genetic analysis. Alberta: Alberta Molecular Biology and Biotechnology Centre, University of Alberta. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data

supp_118_6_1071__index.html^{(686B, html)}

supp_mcw142_suppl_data.zip^{(2MB, zip)}

[mcw142-B1] Alarcón M, Vargas P, Sáez L, Molero J, Aldasoro JJ. 2012. Genetic diversity of mountain plants: two migration episodes of Mediterranean Erodium (Geraniaceae). Molecular Phylogenetics and Evolution 63: 866–876. [DOI] [PubMed] [Google Scholar]

[mcw142-B2] Bancheva S, Raimondo FM. 2003. Biosystematic studies of seven Balkan species from genus Cyanus (Compositae). Bocconea 16: 507–527. [Google Scholar]

[mcw142-B3] Bancheva S, Stoyanov S. 2009. A new species of Cyanus (Asteraceae, Centaureinae) from Southeastern Bulgaria. Novon 19: 421–425. [Google Scholar]

[mcw142-B4] Barraclough TG, Vogler AP, 2000. Detecting the geographical pattern of speciation from species-level phylogenies. American Naturalist 155: 419–434. [DOI] [PubMed] [Google Scholar]

[mcw142-B5] Bardy KE, Albach DC, Schneeweiss GM, Fischer MA, Schönswetter P. 2010. Disentangling phylogeography, polyploid evolution and taxonomy of a woodland herb (Veronica chamaedrys group, Plantaginaceae s.l.) in southeastern Europe. Molecular Phylogenetics and Evolution 57: 771–786. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B6] Barres L, Sanmartín I, Anderson CL. et al., 2013. Reconstructing the evolution and biogeographic history of tribe Cardueae (Compositae). American Journal of Botany 100: 867–882. [DOI] [PubMed] [Google Scholar]

[mcw142-B7] Bonin A, Bellemain E, Eidesen PB, Pompanon F, Brochmann C, Taberlet P. 2004. How to track and assess genotyping errors in population genetic studies. Molecular Ecology 13: 3261–3273. [DOI] [PubMed] [Google Scholar]

[mcw142-B8] Boršić I, Susanna A, Bancheva S, Garcia-Jacas N. 2011. Centaurea sect. Cyanus: nuclear phylogeny, biogeography and life-form evolution. International Journal of Plant Sciences 172: 238–249. [Google Scholar]

[mcw142-B9] Bouckaert R, Heled J. 2014. DensiTree 2: seeing trees through the forest. bioRxiv: 012401. doi:http://dx.doi.org/10.1101/012401. [Google Scholar]

[mcw142-B10] ter Braak CJF, Šmilauer P. 2012. Canoco reference manual and user’s guide: software for ordination, version 5.0. Ithaca: Microcomputer Power. [Google Scholar]

[mcw142-B11] Bryant D, Bouckaert R, Felsenstein J, Rosenberg NA, RoyChoudhury A. 2012. Inferring species trees directly from biallelic genetic markers: bypassing gene trees in a full coalescent analysis. Molecular Biology and Evolution 29: 1917–1932. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B12] Castellanos-Morales C, Gámez N, Castillo-Gámez RA, Eguiarte LE. 2016. Peripatric speciation of an endemic species driven by Pleistocene climate change: the case of the Mexican prairie dog (Cynomys mexicanus). Molecular Phylogenetics and Evolution 94: 171–181. [DOI] [PubMed] [Google Scholar]

[mcw142-B13] Castresana J. 2000. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution 17: 540–552. [DOI] [PubMed] [Google Scholar]

[mcw142-B14] Clement M, Posada D, Crandall K. 2000. TCS: a computer program to estimate gene genealogies. Molecular Ecology 9: 1657–1660. [DOI] [PubMed] [Google Scholar]

[mcw142-B15] Dobeš C, Rossa J, Paule J, Hülber K. 2013. Morphology, DNA-molecular variation, karyology, ecogeography, and phytosociology suggest allopatric differentiation and species rank for Potentilla rigoana (Rosaceae). Taxon 62: 733–745. [Google Scholar]

[mcw142-B16] Doležel J, Sgorbati S, Lucretti S. 1992. Comparison of three DNA ﬂuorochromes for ﬂow cytometric estimation of nuclear DNA content in plants. Journal of Plant Physiology 85: 625–631. [Google Scholar]

[mcw142-B17] Eastwood WJ. 2004. East Mediterranean vegetation and climate change In: Griffiths HI, Kryštufek B, Reed JM, eds. Balkan biodiversity. Pattern and process in the European hotspot. Dordrecht: Kluwer Academic Publishers, 25–48. [Google Scholar]

[mcw142-B18] Ehrich D. 2006. AFLPDAT: a collection of R functions for convenient handling of AFLP data. Molecular Ecology Notes 6: 603–604. [Google Scholar]

[mcw142-B19] Evanno G, Regnaut S, Goudet J. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14: 2611–2620. [DOI] [PubMed] [Google Scholar]

[mcw142-B20] Excoffier L, Laval G, Schneider S. 2005. Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1: 47–50. [PMC free article] [PubMed] [Google Scholar]

[mcw142-B21] Falush D, Stephens M, Pritchard JK. 2007. Inference of population structure using multilocus genotype data: dominant markers and null allele. Molecular Ecology Notes 7: 574–578. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B22] Frajman B, Oxelman B. 2007. Reticulate phylogenetics and phytogeographical structure of Heliosperma (Sileneae, Caryophyllaceae) inferred from chloroplast and nuclear DNA sequences. Molecular Phylogenetics and Evolution 43: 140–155. [DOI] [PubMed] [Google Scholar]

[mcw142-B23] Frajman B, Schneeweiss GM. 2009. A campanulaceous fate: the Albanian stenoendemic Asyneuma comosiforme in fact belongs to isophyllous Campanula. Systematic Botany 34: 595–601. [Google Scholar]

[mcw142-B24] Greuter W. 2006–2009. Compositae (pro parte majore) In: Greuter W, von Raab-Straube E. Compositae. Euro+Med Plantbase – the information resource for Euro-Mediterranean plant diversity. http://ww2.bgbm.org/EuroPlusMed. [Google Scholar]

[mcw142-B25] Gross BL, Rieseberg LH. 2005. The ecological genetics of homoploid hybrid speciation. Journal of Heredity 96: 241–252. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B26] Hall TA. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95–98. [Google Scholar]

[mcw142-B27] Hellwig FH. 2004. Centaureinae (Asteraceae) in the Mediterranean: history of ecogeographical radiation. Plant Systematics and Evolution 246: 137–162. [Google Scholar]

[mcw142-B28] Hewitt GM. 2011. Mediterranean peninsulas: the evolution of hotspots In: Frank E, Zachos FE, Habel JC, eds. Biodiversity hotspots. Distribution and protection of conservation. Berlin: Springer, 123–147. [Google Scholar]

[mcw142-B29] Hilpold A, Schönswetter P, Susanna A, Garcia-Jacas N, Vilatersana R. 2011. Evolution of the central Mediterranean Centaurea cineraria group (Asteraceae): evidence for relatively recent, allopatric diversification following transoceanic seed dispersal. Taxon 60: 528–538. [Google Scholar]

[mcw142-B30] Hilpold A, Vilatersana R, Susanna A. et al. 2014. Phylogeny of the Centaurea group (Centaurea, Compositae) – geography is a better predictor than morphology. Molecular Phylogenetics and Evolution 77: 195–215. [DOI] [PubMed] [Google Scholar]

[mcw142-B31] Huson DH, Bryant D. 2006. Application of phylogenetic networks in evolutionary studies. Molecular Biology and Evolution 23: 254–267. [DOI] [PubMed] [Google Scholar]

[mcw142-B32] Ikeda H, Carlsen T, Fujii N, Crochmann C, Setoguchi H, 2012. Pleistocene climatic oscillations and the speciation history of an alpine endemic and a widespread arctic-alpine plant. New Phytologist 194: 583–594. [DOI] [PubMed] [Google Scholar]

[mcw142-B33] Jakob SS, Blattner FR. 2006. A chloroplast genealogy of Hordeum (Poaceae): long-term persisting haplotypes, incomplete lineage sorting, regional extinction, and the consequences for phylogenetic inference. Molecular Biology and Evolution 23: 1602–1612. [DOI] [PubMed] [Google Scholar]

[mcw142-B34] Jakobsson M, Rosenberg NA. 2007. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 23: 1801–1806. [DOI] [PubMed] [Google Scholar]

[mcw142-B35] Kaya Z, Bancheva S. 2009. A new species of Cyanus (Centaurea p.p.) sect. Napuliferi (Asteraceae) from Turkey. Novon 19: 175–177. [Google Scholar]

[mcw142-B36] Kolarčik V, Zozomová-Lihová J, Mártonfi P. 2010. Systematics and evolutionary history of the Asterotricha group of the genus Onosma (Boraginaceae) in central and southern Europe inferred from AFLP and nrDNA ITS data. Plant Systematics and Evolution 290: 21–45. [Google Scholar]

[mcw142-B37] Kolář F, Fuxová G, Záveská E, et al. 2016. Northern glacial refugia and altitudinal niche divergence shape genome-wide differentiation in the emerging plant model Arabidopsis arenosa. Molecular Ecology, in press doi:10.1111/mec.13721. [DOI] [PubMed] [Google Scholar]

[mcw142-B38] Kropf M, Comes HP, Kadereit JW. 2006. Long-distance dispersal vs vicariance: the origin and genetic diversity of alpine plants in the Spanish Sierra Nevada. New Phytologist 172: 169–184. [DOI] [PubMed] [Google Scholar]

[mcw142-B39] Kropf M, Comes HP, Kadereit JW. 2008. Causes of the genetic architecture of south-west European high mountain disjuncts. Plant Ecology & Diversity 1: 217–228. [Google Scholar]

[mcw142-B40] Kropf M, Comes HP, Kadereit JW. 2012. Past, present and future of mountain species of the French Massif Central – the case of Soldanella alpina L. subsp. alpine (Primulaceae) and a review of other plant and animal studies. Journal of Biogeography 39: 799–812. [Google Scholar]

[mcw142-B41] Kruskal JB. 1964. Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis. Psychometrika 29: 1–27. [Google Scholar]

[mcw142-B42] Kryštufek B, Reed JM. 2004. Patterns and process in Balkan biodiversity: an overview In: Griffiths HI, Kryštufek B, Reed JM, eds. Balkan biodiversity. Pattern and process in the European hotspot. Dordrecht: Kluwer Academic Publishers, 1–8. [Google Scholar]

[mcw142-B43] Krzanowski WJ. 1990. Principles of multivariate analysis. Oxford: Clarendon Press. [Google Scholar]

[mcw142-B44] Kučera J, Marhold K, Lihová J. 2010. Cardamine maritima group (Brassicaceae) in the amphi-Adriatic area: a hotspot of species diversity revealed by DNA sequences and morphological variation. Taxon 59: 148–164. [Google Scholar]

[mcw142-B45] Kučera J, Turis P, Zozomová-Lihová J, Slovák M. 2013. Cyclamen fatrense (Primulaceae), only a myth or a true West Carpathian endemic? Genetic and morphological evidence. Preslia 85: 133–158. [Google Scholar]

[mcw142-B46] Kuzmanović N, Comanescu P, Frajman B. et al. 2013. Genetic, cytological and morphological differentiation within the Balkan-Carpathian Sesleria rigida sensu Fl. Eur. (Poaceae): a taxonomically intricate tetraploid-octoploid complex. Taxon 62: 458–472. [Google Scholar]

[mcw142-B47] Lakušić D, Liber Z, Nikolić T. et al. 2013. Molecular phylogeny of the Campanula pyramidalis species complex (Campanulaceae) inferred from chloroplast and nuclear non-coding sequences and its taxonomic implications. Taxon 62: 505–524. [Google Scholar]

[mcw142-B48] López-Sepúlveda P, Tremetsberger K, Ángeles Ortiz M, Baeza CM, Peñailillo P, Stuessy TF. 2013. Radiation of the Hypochaeris apargioides complex (Asteraceae: Cichorieae) of southern South America. Taxon 62: 550–564. [Google Scholar]

[mcw142-B49] Löser CJ. 2012. Speciation and genomic conflicts in Cyanus Mill. PhD Thesis, Friedrich Schiller University Jena, Germany.

[mcw142-B50] Maddison WP, Knowles LL. 2006. Inferring phylogeny despite incomplete lineage sorting. Systematic Biology 55: 21–30. [DOI] [PubMed] [Google Scholar]

[mcw142-B51] Manafzadeh S, Salvo G, Conti E. 2014. A tale of migrations from east to west: the Irano-Turanian floristic region as a source of Mediterranean xerophytes. Journal of Biogeography 41: 366–379. [Google Scholar]

[mcw142-B52] Marhold K, Kudoh H, Pak J-H, Watanabe K, Španiel S, Lihová J. 2010. Cytotype diversity and genome size variation in eastern Asian polyploid Cardamine (Brassicaceae) species. Annals of Botany 105: 249–264. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B53] Martín-Bravo S, Valcárcel V, Vargas P, Luceño M. 2010. Geographical speciation related to Pleistocene range shifts in the western Mediterranean mountains (Reseda sect. Glaucoreseda, Resedaceae). Taxon 59: 466–482. [Google Scholar]

[mcw142-B54] Mereďa P, Hodálová I, Mártonfi P, Kučera P, Lihová J. 2008. Intraspecific variation in Viola suavis in Europe: parallel evolution of white-flowered morphotypes. Annals of Botany 102: 443–462. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B55] Meshinev T. 2007. Vegetation and phytogeography: a brief characteristic In: Fet V, Popov A, eds. Biogeography and ecology of Bulgaria. Dordrecht: Springer, 581–588. [Google Scholar]

[mcw142-B56] Moody ML, Rieseberg LH. 2012. Sorting through the chaff, nDNA gene trees for phylogenetic inference and hybrid identification of annual sunflowers (Helianthus sect. Helianthus). Molecular Phylogenetics and Evolution 64: 145–155. [DOI] [PubMed] [Google Scholar]

[mcw142-B57] Moreno-Letelier A, Ortíz-Medrano A, Piñero D. 2013. Niche divergence versus neutral processes: combined environmental and genetic analyses identify contrasting patterns of differentiation in recently diverged pine species. PLoS One 8: e78228. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B58] Nei M, Li WH. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy of Sciences of the USA 76: 5269–5273. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B59] Olšavská K, Löser CJ. 2013. Mating system and hybridization of the Cyanus triumfetti and C. montanus groups (Asteraceae). Folia Geobotanica 48: 537–554. [Google Scholar]

[mcw142-B60] Olšavská K, Perný M, Kučera J, Hodálová I. 2011. Biosystematic study of the Cyanus triumfetti group in Central Europe. Preslia 83: 59–98. [Google Scholar]

[mcw142-B61] Olšavská K, Perný M, Španiel S, Šingliarová B. 2012. Nuclear DNA content variation among perennial taxa of the genus Cyanus (Asteraceae) in Central Europe and adjacent areas. Plant Systematics and Evolution 298: 1463–1482. [Google Scholar]

[mcw142-B62] Olšavská K, Perný M, Löser CJ, Stimper R, Hodálová I. 2013. Cytogeography of European perennial species of Cyanus (Asteraceae). Botanical Journal of the Linnean Society 173: 230–257. [Google Scholar]

[mcw142-B63] Peev D, Delcheva M. 2007. Florogeographical elements and evolutionary trends in the Bulgarian flora In: Fet V, Popov A, eds. Biogeography and ecology of Bulgaria. Dordrecht: Springer, 571–579. [Google Scholar]

[mcw142-B64] Perný M, Tribsch A, Anchev ME. 2005. Infraspecific differentiation in diploid Cardamine acris (Brassicaceae) from Balkan Peninsula: morphological and molecular evidence. Folia Geobotanica 39: 405–429. [Google Scholar]

[mcw142-B65] Petit RJ, Aguinagalde I, de Beaulieu J-L. et al. 2003. Glacial refugia: hotspots but not melting pots of genetic diversity. Science 300: 1563–1565. [DOI] [PubMed] [Google Scholar]

[mcw142-B66] Quézel P. 2004. Large-scale post-glacial distribution of vegetation structures in the Mediterranean region In: Mazzoleni S, di Pasquale G, Mulligan M, di Martino P, Rego F, eds. Recent dynamics of the Mediterranean vegetation and landscape. Chichester: John Wiley & Sons, 3–12. [Google Scholar]

[mcw142-B67] Radford EA, Odé B, eds. 2009. Conserving important plant areas: investing in the Green Gold of South East Europe. Salisbury: Plantlife International. [Google Scholar]

[mcw142-B68] Reed JM, Kryštufek B, Eastwood WJ. 2004. The physical geography of the Balkans and nomenclature of place names In: Griffiths HI, Kryštufek B, Reed JM, eds. Balkan biodiversity. Pattern and process in the European hotspot. Dordrecht: Kluwer Academic Publishers, 9–22. [Google Scholar]

[mcw142-B69] Ronikier M, Zalewska-Gałosz J. 2014. Independent evolutionary history between the Balkan ranges and more northerly mountains in Campanula alpina s.l. (Campanulaceae): genetic divergence and morphological segregation of taxa. Taxon 63: 116–131. [Google Scholar]

[mcw142-B70] Rosenberg NA. 2004. DISTRUCT: a program for the graphical display of population structure. Molecular Ecology Notes 4: 137–138. [Google Scholar]

[mcw142-B71] Saitou N, Nei M. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4: 406–425. [DOI] [PubMed] [Google Scholar]

[mcw142-B72] Sanmartín I, Anderson CL, Alarcon M, Ronquist F, Aldasoro JJ. 2010. Bayesian island biogeography in a continental setting: the Rand Flora case. Biology Letters 6: 703–707. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B73] SAS Institute Inc. 2011. SAS®9.3 Product documentation. Cary: SAS Institute. [Google Scholar]

[mcw142-B74] Schönswetter P, Schneeweiss GM. 2009. Androsace komovensis sp. nov., a long mistaken local endemic from the southern Balkan Peninsula with biogeographic links to the Eastern Alps. Taxon 58: 544–549. [Google Scholar]

[mcw142-B75] Schönswetter P, Stehlik I, Holderegger R, Tribsch A. 2005. Molecular evidence for glacial refugia of mountain plants in the European Alps. Molecular Ecology 14: 3547–3555. [DOI] [PubMed] [Google Scholar]

[mcw142-B76] Schönswetter P, Suda J, Popp M, Weiss-Schneeweiss H, Brochmann C. 2007. Circumpolar phylogeography of Juncus biglumis (Juncaceae) inferred from AFLP ﬁngerprints, cpDNA sequences, nuclear DNA content and chromosome numbers. Molecular Phylogenetics and Evolution 42: 92–103. [DOI] [PubMed] [Google Scholar]

[mcw142-B77] Schönswetter P, Solstad H, Garcìa PE, Elven R. 2009. A combined molecular and morphological approach to the taxonomically intricate European mountain plant Papaver alpinum s.l. (Papaveraceae) – taxa or informal phylogeographical groups? Taxon 17: 1326–1343. [Google Scholar]

[mcw142-B78] Schlüter PM, Harris SA. 2006. Analysis of multilocus fingerprinting data sets containing missing data. Molecular Ecology Notes 6: 569–572. [Google Scholar]

[mcw142-B79] Shaw J, Lickey E, Beck JT. et al. 2005. The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal of Botany 92: 142–166. [DOI] [PubMed] [Google Scholar]

[mcw142-B80] Shaw J, Lickey E, Schilling EE, Small RL. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. American Journal of Botany 94: 275–288. [DOI] [PubMed] [Google Scholar]

[mcw142-B81] Slovák M, Kučera J, Marhold K, Zozomová-Lihová J. 2012a. The morphological and genetic variation in the polymorphic species Picris hieracioides (Compositae, Lactuceae) in Europe strongly contrasts with traditional taxonomical concepts. Systematic Botany 37: 258–278. [Google Scholar]

[mcw142-B82] Slovák M, Kučera J, Turis P, Zozomová-Lihová J. 2012b. Multiple glacial refugia and postglacial colonisation routes inferred for a woodland geophyte, Cyclamen purpurascens: patterns concordant with the Pleistocene history of broadleaved and coniferous tree species. Biological Journal of the Linnean Society 105: 741–760. [Google Scholar]

[mcw142-B83] Slovák M, Kučera J, Záveská E, Vďačný P. 2014. Dealing with discordant genetic signal caused by hybridisation, incomplete lineage sorting and paucity of primary nucleotide homologies: a case study of closely related members of the genus Picris subsection Hieracioides (Compositae). PLoS One 9: e104929. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B84] StatSoft Inc. 2013. Electronic statistics textbook. Tulsa: Statsoft; http://www.statsoft.com/Textbook. [Google Scholar]

[mcw142-B85] Stehlik I, Blattner FR, Holderegger R, Bachmann K. 2002. Nunatak survival of the high Alpine plant Eritrichium nanum (L.) Gaudin in the central Alps during the ice ages. Molecular Ecology 11: 2027–2036. [DOI] [PubMed] [Google Scholar]

[mcw142-B86] Stefanović S, Lakušić D, Kuzmina M, Mededović S, Tan K, Stevanović V. 2008. Molecular phylogeny of Edraianthus (grassy bells; Campanulaceae) based on non-coding plastid DNA sequences. Taxon 57: 452–475. [Google Scholar]

[mcw142-B87] Stevanović V. 1996. Analysis of the Central European and Mediterranean orophytic element on the mountains of the W. and Central Balkan Peninsula, with special reference to endemics. Bocconea 5: 77–97. [Google Scholar]

[mcw142-B88] Stevanović V, Tan K, Petrova A. 2007. Mapping the endemic flora of the Balkans – a progress report. Bocconea 21: 131–137. [Google Scholar]

[mcw142-B89] Suda J, Trávníček P, Mandák B, Berchová-Bímová K. 2010. Genome size as a marker for identifying the invasive alien taxa in Fallopia section Reynoutria. Preslia 82: 97–106. [Google Scholar]

[mcw142-B90] Surina B, Schönswetter P, Schneeweiss GM. 2011. Quaternary range dynamics of ecologically divergent species (Edraianthus serpyllifolius and E. tenuifolius, Campanulaceae) within the Balkan refugium. Journal of Biogeography 38: 1381–1393. [Google Scholar]

[mcw142-B91] Talavera G, Castresana J. 2007. Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Systematic Biology 56: 564–577. [DOI] [PubMed] [Google Scholar]

[mcw142-B92] Thompson JD. 1999. Population differentiation in Mediterranean plants: insights into colonisation history and implications for species diversification. Heredity 82: 229–236. [DOI] [PubMed] [Google Scholar]

[mcw142-B93] Thompson JD, Lavergne S, Affre L, Gaudeu M, Debussche M. 2005. Ecological differentiation of Mediterranean endemic plants. Taxon 54: 967–976. [Google Scholar]

[mcw142-B94] Tukey JW. 1977. Exploratory data analysis. Reading, MA: Addison-Wesley. [Google Scholar]

[mcw142-B95] Tzedakis PC. 2004. The Balkans as prime glacial refugial territory of European temperate trees In: Griffiths HI, Kryštufek B, Reed JM, eds. Balkan biodiversity. Pattern and process in the European hotspot. Dordrecht: Kluwer Academic Publishers, 49–68. [Google Scholar]

[mcw142-B96] Velikov V, Stoyanova M. 2007. Landscapes and climate of Bulgaria In: Fet V, Popov A, eds. Biogeography and ecology of Bulgaria. Dordrecht: Springer, 589–605. [Google Scholar]

[mcw142-B97] Veselý P, Bureš P, Šmarda P, Pavlíček T. 2012. Genome size and DNA base composition of geophytes: the mirror of phenology and ecology? Annals of Botany 109: 65–75. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B98] Vos P, Hogers R, Bleeker M. et al. 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23: 4407–4414. [DOI] [PMC free article] [PubMed] [Google Scholar]

[mcw142-B99] Watanabe K. 1986. Cytogeography of the genus Eupatorium (Compositae). A review. Plant Species Biology 1: 99–116. [Google Scholar]

[mcw142-B100] Yeh FC, Yang RC, Timothy BJ, Ye Z, Judy M. 1997. Pop Gene, the user-friendly shareware for population genetic analysis. Alberta: Alberta Molecular Biology and Biotechnology Centre, University of Alberta. [Google Scholar]

PERMALINK

On the origins of Balkan endemics: the complex evolutionary history of the Cyanus napulifer group (Asteraceae)

Katarína Olšavská

Marek Slovák

Karol Marhold

Eliška Štubňová

Jaromír Kučera

Abstract

INTRODUCTION

MATERIALS AND METHODS

Sampling design

Fig. 1.

Table 1.

Plant material

DNA extraction and AFLP generation

AFLP data analyses

cpDNA data analyses

Flow cytometric analyses

Morphometric analyses

Ecological and phenological trait analyses

RESULTS

AFLP analyses

Fig. 2.

Fig. 3.

Table 2.

Fig. 5.

Table 3.

Flow cytometric analyses

Fig. 6.

Fig. 7.

Morphometric analyses

Fig. 8.

Congruences and incongruences among genetic, DNA-content and morphological variation

Fig. 4.

Ecological, phenological and morphological traits analyses

Fig. 9.

DISCUSSION

AFLPs help disentangle the evolution and biogeography of Cyanus perennials: new implications for Mediterranean representatives

Notes on polyploid evolution in Mediterranean Cyanus perennials with x = 10

Allopatric speciation within the C. napulifer group accompanied by long-term isolation and hybridization

Distinct life strategies of the C. napulifer group point to adaptive radiation

Conclusions

SUPPLEMENTARY DATA

ACKNOWLEDGEMENTS

LITERATURE CITED

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases