Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx 1

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev 5 6

4

Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow q

7

Sofie Meeus a,b,⇑, Steven Janssens a,c, Kenny Helsen a, Hans Jacquemyn a

3

8 9 10 12 11 13 1 2 5 9 16 17 18 19 20 21 22 23 24 25 26 27 28

a

KU Leuven, Department of Biology, Plant Population and Conservation Biology, Kasteelpark Arenberg 31, 3001 Heverlee, Belgium Biological and Environmental Sciences, University of Stirling, Stirling FK9 4LA, UK c Botanic Garden Meise, Nieuwelaan 38, BE-1860 Meise, Belgium b

a r t i c l e

i n f o

Article history: Received 9 May 2015 Revised 28 November 2015 Accepted 30 November 2015 Available online xxxx Keywords: Hybridization Lineage sorting Introgression Distyly Mating barriers Interspecific gene flow

a b s t r a c t The distylous genus Pulmonaria contains approximately 18 species that are widely distributed across Eurasia. Previous studies have shown that species delimitation in the genus is problematic, but have not yet explored the evolutionary history of the genus. Premating reproductive barriers between European species appear to be weak, as several species have strongly overlapping distribution areas, flower at the same time and share the same pollinators, suggesting that hybridization may have contributed to the evolutionary history of Pulmonaria. To test this hypothesis, phylogenetic analyses of nuclear ITS and plastid data (rps16, trnH-psbA, rpl16) from 48 allopatric and four sympatric populations were performed to (1) provide a molecular phylogeny for nine of the most common Pulmonaria species in Europe, (2) detect current and ancient hybridization events, and (3) assess the contribution of hybridization versus incomplete lineage sorting to the inferred phylogenetic patterns. Our results showed that gene trees displayed widespread, strongly supported incongruence associated with the conflicting position of hybrid samples rather than incomplete lineage sorting. Evidence was found of different degrees of hybridization, ranging from current interspecific gene flow at secondary contact zones to introgression at the population level and at least one event of hybrid speciation. Overall, these results suggest that hybridization and introgression were – and could still be – important processes affecting speciation in the genus Pulmonaria. Ó 2015 Elsevier Inc. All rights reserved.

30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

49 50

1. Introduction

51

Hybridization has long been acknowledged to be important in the process of speciation and is known to result in varied outcomes with respect to species formation (reviewed in Mallet, 2007; Abbott et al., 2013). Interspecific gene flow at secondary contact zones may slow down or even cause lineage convergence due to introgression and occurs more readily among related species as a result of the gradual increase of reproductive isolation with genetic divergence (Coyne and Orr, 1989, 1997; Moyle et al., 2004; Scopece et al., 2007). On the other hand, hybridization may promote rapid reproductive isolation of sympatric species due to selection against the formation of unfit hybrid offspring (‘reinforcement’), thereby accelerating the process of speciation (Servedio and Noor, 2003). Furthermore, new favorable gene combinations resulting from

52 53 54 55 56 57 58 59 60 61 62 63

q

This paper was edited by the Associate Editor Elizabeth Zimmer. ⇑ Corresponding author at: KU Leuven, Department of Biology, Plant Population and Conservation Biology, Kasteelpark Arenberg 31, 3001 Heverlee, Belgium. E-mail address: [email protected] (S. Meeus).

hybridization may give rise to new species, which can become instantaneously reproductively isolated from the parental species (Vereecken et al., 2010). Hybridization is considered a regular event in species biology with 25% of the plant species known to hybridize with at least one other species (Mallet et al., 2007). The genus Pulmonaria (Boraginaceae) has been of interest for centuries to botanists worldwide because of its ornamental and medicinal properties (Hewitt, 1994; Bennett, 2003). In addition, Pulmonaria is known for its complex taxonomy and problematic delimitation of species (Sauer, 1975; Bolliger, 1982) as well as its characteristic distylous breeding system throughout the whole genus (Darwin, 1877; Olesen, 1979; Richards and Mitchell, 1990; Champluvier and Jacquemart, 1999; Meeus et al., 2012a,b). Based on a previous molecular phylogeny, Pulmonaria is placed within the generic complex Nonea/Elizaldia/Pulmonaria/Paraskevia (Boraginaceae, Boragineae) (Selvi et al., 2006). Pulmonaria separated from a common Tertiary ancestor with Paraskevia cesatiana, a tetraploid (2n = 28) Greek endemic characterized by a non-rhizomatous root system, the absence of heterostyly and the prefloral development of foliage leaves, traits that differentiate this species from the

http://dx.doi.org/10.1016/j.ympev.2015.11.022 1055-7903/Ó 2015 Elsevier Inc. All rights reserved.

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 2 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145

S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

typical Pulmonaria species (Selvi et al., 2006; for the most recent placement of Pulmonaria within the Boraginaceae, see Cohen, 2013). Morphologically, Pulmonaria species are fairly similar in having pink-blue1 actinomorphic tube-shaped flowers and hairy leaves (Fig. 1). As a result, they were often considered as varieties of the same species rather than as separate species (Darwin, 1877). According to the most recent monographic treatment of the genus by Bolliger (1982), the genus Pulmonaria comprises 18 species (8–11 subspecies), of which eight occur in Western Europe. Pulmonaria obscura, P. officinalis, P. longifolia and P. angustifolia are the most common species in Western Europe and have largely overlapping distribution areas, which together cover almost the entire range of Pulmonaria (Fig. 2). Although the majority of species occurs most often in shaded woodland habitat (e.g. P. officinalis, P. obscura), some species (e.g. P. longifolia) prefer dry and sunny habitats and are, therefore, often found in grasslands (Hill et al., 2004). Although the genus Pulmonaria is morphologically uniform, chromosome numbers vary widely among and even within Pulmonaria species and subspecies, having a basic haploid chromosome number of n = x = 7. Therefore, chromosome counts have been intensively used for species identification and have provided evidence of polyploidization events and the occurrence of aneuploid species (i.e. having incomplete sets of or deviations from the basic chromosome number) within the genus (Sauer, 1975; Bolliger, 1982). Chromosome numbers (2n) range from 14 (present in at least 40% of the Pulmonaria species) to 38 with a total of 10 different cytotypes (Sauer, 1975). Pulmonaria contains diploids (2n = 2x = 14), tetraploids (2n = 4x = 28) and a series of aneuploids (2n = 16, 18, 20, 22, 24, 26, 30, 38) (Sauer, 1975; Bolliger, 1982). Little is known about the strength of the different isolating barriers in Pulmonaria, however, the apparent lack of premating barriers (i.e. strong overlap in geographic distribution, flowering synchrony, similar pollinator preference) and potential strong postmating barriers resulting from the high variation in chromosome numbers within the genus has made that the extent to which Pulmonaria species hybridize and how this process contributed to speciation within the genus has been subject to debate during the last century (Gams, 1927; Merxmüller and Sauer, 1972). Moreover, historical records report opposite results regarding compatibility among the species and the incidence of hybridization. For example, Darwin (1877) reported a failure of seed set after pollinating Pulmonaria longifolia populations from the Isle of Wight (UK) with pollen from P. officinalis. In horticulture, however, Pulmonaria is known to hybridize freely among species, resulting in popular varieties such as Pulmonaria ‘Mawson’s Blue’ and Pulmonaria ‘Sissinghurst White’ (Bennett, 2003). These popular ornamental hybrids are, in turn, used to generate new varieties implying that hybrids are not only viable, but also able to reproduce sexually (Bennett, 2003). Moreover, few interspecific crosses performed by Bolliger (1982) suggests that first-generation hybrids show regular bivalent pairing and high pollen viability (60–89%) at meiosis both for intra- and interploidy crosses. Nevertheless, taxonomists agree on the species status of these common Pulmonaria species based on their morphological traits and chromosome numbers (Sauer, 1975; Bolliger, 1982). Heretofore, only few attempts have been made to reconstruct a molecular phylogeny of Pulmonaria (Kirchner, 2004) and to test specific hypotheses about the role of hybridization in affecting the complex evolutionary history of the genus. Reconstruction of phylogenetic relationships using multiple genetic markers, however, is problematic as hybridization is a major cause of topological

1

For interpretation of color in Fig. 1, the reader is referred to the web version of this article.

incongruence between gene trees (McBreen and Lockhart, 2006). Horizontal gene transfer between closely or more distantly related species often results in a strong conflict between gene trees with the traditional bifurcating (hierarchical) representation of species diversification and is therefore best presented by a reticulation network (Hennig, 1966; Legendre and Makarenkov, 2002; Huson et al., 2005; Hegarty and Hiscock, 2005; McBreen and Lockhart, 2006). However, studying the incongruence between gene trees of hybridizing taxa offers an opportunity to detect hybrid speciation (Sang and Zhong, 2000). Moreover, by comparing differently inherited markers – e.g. nuclear (biparentally inherited) and plastid genomes (maternally inherited in case of Pulmonaria) – it is possible to identify the paternal and maternal species involved in current and historical hybridization events (Linder and Rieseberg, 2004). Several tests have been proposed to assess the statistical significance of phylogenetic incongruences (Farris et al., 1995; Shimodaira and Hasegawa, 1999; Shimodaira, 2002) and significant topological differences are increasingly attributed to interspecific hybridization (McBreen and Lockhart, 2006 and references therein). However, confounding population genetic processes such as lineage sorting (i.e. the stochastic sorting of alleles following divergence from a polymorphic ancestor) might mislead inference of the real contribution of hybridization to the observed pattern of gene tree incongruence (Linder and Rieseberg, 2004; Kubatko, 2009; de Villiers et al., 2013) and is especially common among closely related species where lineage sorting has not yet been completed, leading to non-monophyletic species assemblages (e.g. Primula; Schmidt-Lebuhn et al., 2012). Incomplete lineage sorting is especially important when the effective population size of a given lineage is large with respect to the time elapsed since divergence, so that genetic drift is unlikely to result in fixed alleles subsequent to divergence (Rosenberg and Nordborg, 2002). Since uniparentally inherited plastids have an effective population size which is smaller than that of biparentally inherited nuclear markers, the rate of lineage sorting is expected to be faster in plastids (Palumbi et al., 2001; Hedrick, 2007). Deviations from this expectation could, therefore, be caused by hybridization events (Chan and Levin, 2005). A third approach to detect hybridization between a group of species which has been increasingly used in several studies concerns the analysis of sequences from multiple gene families (e.g. ITS) to look for different intra-individual copies that hold information about the hybrid’s parentage to get a more direct assessment of hybrid evolution of which patterns are not subjected to incomplete lineage sorting (e.g. Koch et al., 2003; Feliner et al., 2004; Hodacˇ et al., 2014). In the present study, we investigated whether a statistically significant phylogenetic conflict between nuclear and plastid genomes was apparent in the genus Pulmonaria, providing support that hybridization has played a significant role in intrageneric speciation processes. For the specific detection of potential hybrid (‘‘conflicting”) populations and the assessment of ongoing gene flow, morphologically intermediate individuals from sympatric populations were included in the analysis and ITS polymorphisms were investigated. Furthermore, the Genealogical Sorting Index (GSI) was calculated to infer the contribution of incomplete lineage sorting versus hybridization to the observed phylogenetic pattern.

146

2. Materials and methods

203

2.1. Sampling and species identification

204

Leaf samples were collected across seven European countries (Belgium, France, England, Germany, Czech Republic, Italy, Estonia) during spring 2013 from 48 allopatric populations of nine species

205

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202

206 207

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

3

Fig. 1. The nine study species of the genus Pulmonaria: (A) P. collina, (B) P. mollis with detailed inlay of glandular hairs on the calyx, (C) P. officinalis, (D) P. longifolia, (E) P. montana (spots can be present or absent in this species and are presented by dashed lines), (F) P. affinis, (G) P. saccharata, (H) P. angustifolia, (I) P. obscura. Photos A, B, H, I courtesy of Kenny Helsen, photo E courtesy of Kasper Van Acker, the others taken by Sofie Meeus at the sampling site GY (D) and private garden (C, G), and photo F taken by Steven Janssens at the Botanic Garden Meise (Belgium). Inlayed leaf diagrams present the average summer leaf proportions and the presence of spots in the nine species sampled in this study.

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 4

S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

Fig. 2. Distribution ranges of the nine study species. Map reconstructed using following sources: Sauer (1975), Bolliger (1982), Stewart et al. (1994) and Brewis et al. (1996).

208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236

of Pulmonaria: P. saccharata (N = 3 populations), P. angustifolia (N = 6), P. mollis (N = 4), P. collina (N = 2), P. longifolia (N = 7), P. montana (N = 7), P. affinis (N = 2), P. officinalis (N = 8) and P. obscura (N = 9) for molecular phylogenetic analyses (Table S1). These taxonomic species were identified using the identification key of Bolliger (1982), which is mainly based on geographic provenance, the shape of summer leaves (length-to-width ratios, the position of the widest part) and leaf variegation (spots versus no spots). In the Czech Republic, additional leaf samples were collected from four sympatric populations (PD, KV, VnV, BRZ) where two different Pulmonaria species co-occurred within a radius of 200 m. In these populations, P. obscura co-occurred with either P. officinalis (PD), P. angustifolia (BRZ) or P. mollis (KV, VnV). Specimen names consist of abbreviated population names (two to three letters) that refer to the location of the focal population (e.g. BH). In case of sympatric populations, a subsequent lowercase letter is added to indicate with which species the morphology of the specimen corresponds (e.g. BRZa refers to a specimen morphologically resembling P. angustifolia), the extension ‘‘hybrid” is used for morphological intermediates, and a capital ‘‘B” indicates the specimen was found in the nearby P. obscura population. For species identification, five to ten individuals in each of the populations (unless populations consisted of less than five individuals) were measured for the maximum length (L) and maximum width (Wmax) of their summer leaves to characterize the shape (Table S1). As the timing of summer leaf emergence differed between species and between northern and southern populations, species identification based on these leaves was restricted to the length–width ratio and position of the widest part of the leaf

(measured from the top) (P(Wmax)) instead of actual lengths and widths. The three main morphological characteristics related to the summer leaves that are used to distinguish among Pulmonaria species (i.e. the length–width ratio, the position of the widest part, presence of spots) were plotted in Fig. S1 and the same characteristics were redrawn for each species in Fig. 1 (inlays) based on their summer leaf measurements. The two species with heart-shaped leaves, P. officinalis (with spots) and P. obscura (without spots) were characterized by length-to-width ratios of less than 2 (L/ Wmax = 1.72 and 1.88, respectively) with the widest point of their leaves centered in the lowest part of the leaf (‘heart-shaped’; P (Wmax)/L = 0.73 and 0.72, respectively) (Fig. 1C and I; Table S1; Fig. S1). At the other extreme are the species with lanceolate leaves such as P. longifolia (with spots) and P. angustifolia (without spots) having leaves which are more than four times as long than wide (L/ Wmax = 4.29 and 4.70, respectively) and with the widest point of their leaves measured more or less halfway the leaf (P(Wmax)/ L = 0.58 and 0.61, respectively)) (Fig. 1D and H; Table S1; Fig. S1). Species with a leaf shape that was intermediate to these extreme cases were P. affinis, P. mollis, P. collina, P. montana and P. saccharata, P. mollis (Table S1; Fig. S1). Although overlapping in leaf shape P. affinis and P. saccharata, and P. mollis and P. montana could be easily distinguished based on their non-overlapping distribution ranges and the presence of glandular hairs (characteristic to P. mollis), respectively (Figs. 1B and 2). In three (KV, PD, BRZ) out of four sympatric populations, morphological intermediates were identified in the field and referred to as ‘hybrid’ (e.g. BRZ(hybrid), KV(hybrid); Table S1). In population PD, a sympatric population of the species P. officinalis

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

269

(PDof) and P. obscura (PD(B)), a continuous variation in leaf variegation was observed ranging from no spots at all (P. obscura) to leaves densely covered with clearly delineated white spots (P. officinalis).

270

2.2. Molecular protocols

271

284

Total genomic DNA was extracted from silica dried leaf material using a modified version of the CTAB protocol (Saghai-Maroof et al., 1984; Doyle and Doyle, 1987). Primers and temperature programs used for the amplification of chloroplast rps16, trnH-psbA and ITS follow Oxelman et al. (1997), Sang et al. (1997), and White et al. (1990), respectively. For rpl16, genus-specific Pulmonaria primers were newly designed based on known Pulmonaria sequences from GenBank to increase amplification (rpl16-PulmF: 50 -TCC GGA TCT AAA GTC TCG GTC A-30 ; rpl16-PulmR: 50 -CAC CTC ATC CGG CTC CTC GC-30 ; Ann.T: 54 °C). Purified amplification products were sent to Macrogen Inc. (Seoul, South Korea) for sequencing. Sequences were submitted to Genbank: rps16 (KT737605–KT737662), trnH-psbA (KT737495–KT737552), rpl16 (KT737553–KT737604) and ITS (KT737663–KT737721).

285

2.3. Phylogenetic analyses

286

322

Contiguous sequences were assembled using Geneious v7.0.6 (Biomatters, New Zealand). Automatic alignments were carried out with MAFFT (Katoh et al., 2002) under the E-INS-i algorithm, a 100PAM/k = 2 scoring matrix, a gap open penalty of 1.3 and an offset value of 0.123. Subsequent manual fine-tuning of the aligned dataset was done in Geneious v 7.0.6. Gaps were treated as missing data. The best-fit nucleotide substitution model for each plastid and nuclear dataset was determined using jModelTest 2.1.4 (Posada, 2008) under the Akaike information criterion (AIC). A mixedmodel approach was used in which the combined dataset was partitioned according to the different substitution models for each gene marker (Ronquist and Huelsenbeck, 2003). Bayesian inference (BI) analyses were conducted with MrBayes v3.1 (Huelsenbeck and Ronquist, 2001) on four individual data partitions (rps16, trnHpsbA, rpl16 and ITS) and a combined data matrix (combined chloroplast). For each Bayesian analysis, two simultaneous runs with four chains (one cold, three heated) were initiated from a random starting tree and run for 10 million generations. Trees were sampled every 1000 generations. Inspection of chain convergence and ESS parameters was done with TRACER v.1.4 (Rambaut and Drummond, 2007). The 25% burn-in was discarded after log likelihood stationarity was reached. Clades with posterior probability values (PP) P 0.95 were considered to be strongly supported (Suzuki et al., 2002). Maximum Likelihood analyses were carried out on the CIPRES web portal using RAxML 7.2.8 (Stamatakis et al., 2008) under the GTRGAMMA model. Non-parametric ML bootstrapping analysis was calculated with 1000 bootstrap replicates. Branches with bootstrap values (BS) P 75% were considered to be strongly supported. On the basis of these aligned sequences two haplotype networks were created based on statistical parsimony using TCS 1.21 with a 95% cutoff as the maximum number of mutational connections between pairs of sequences justified by the parsimony criterion (Clement et al., 2000). Nuclear DNA haplotypes were calculated from the ITS marker, and cpDNA haplotypes were calculated from the three combined chloroplast markers.

323

2.4. Test of incongruence

324

In order to search for incongruence between the chloroplast and nuclear datasets a series of incongruence length difference tests

266 267 268

272 273 274 275 276 277 278 279 280 281 282 283

287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321

325

5

(ILD; Farris et al., 1995) were carried out with PAUP⁄ v.4.0b10 (Swofford, 2002) using the following parameters: simple taxon addition, TBR branch swapping and heuristic searches of 1000 repartitions of the data. Pairwise tests were not only conducted between chloroplast and nuclear datasets, but also between the different chloroplast partitions. In addition, topological incongruence between chloroplast and nuclear datasets was examined with the Approximately Unbiased (AU; Shimodaira, 2002) and Shimodaira–Hasegawa (SH; Shimodaira and Hasegawa, 1999) tests. The combined chloroplast dataset and the nrITS dataset were treated as two independent partitions for the AU and SH test. Conflicting topologies (or nodes) were obtained from the ML analyses (most resolved topology). Site-wise log-likelihoods for constrained and unconstrained phylogenies were calculated with PAUP⁄ v.4.0b10 (Swofford, 2002) and used as input for multi-scale bootstrap resampling using CONSEL v0.1j (Shimodaira and Hasegawa, 2001) to evaluate the probability value of each alternative topology. Multiscale bootstrap resampling was conducted with ten sets of 10,000 replicates. Significance level was set at 0.05 for all tests. Taxa were identified as ‘incongruent’ when they emerged in different clades between the two trees.

326

2.5. Genealogical Sorting Index

347

To study the relative contribution of ancestral polymorphism or incomplete lineage sorting to the observed phylogenetic patterns, we calculated the Genealogical Sorting Index (GSI), a statistic for quantifying the degree of exclusive ancestry of Pulmonaria species by assessing the significance of clustering of the various populations within a certain species (Cummings et al., 2008). GSI is calculated using topological information only and ranges between 0 and 1, with 1 indicating monophyly of the species and 0 indicating dispersal over the entire tree. GSI calculation and random permutations tests (10,000) to assess statistical significance were conducted using GSI version 0.92 made available on the website http://molecularevolution.org/. P-values below a significance level of a = 0.05 indicate significant deviation of GSI = 0 and result in the rejection of the null hypothesis that the species is of mixed ancestry.

348

3. Results

363

3.1. Phylogeny

364

Alignment and sequence characteristics are summarized in Table 1 and ITS polymorphism are reported in Table S1. Using JModelTest, the GTR model was selected as best fit model for ITS and trnH-psbA, whereas the GTR + I model was selected as best substitution model for rps16 and rpl16. In general, the nrITS matrix yielded a more resolved tree than the combined plastid matrix (Fig. 3). At high intrageneric level, each gene tree consists of two major clades (Clade I and Clade II), which differ in species composition. Clade I (BS: 78, PP: 0.72) in the nuclear phylogeny (subsequently referred to as nrClade I, Fig. 3) contains mainly specimens of P. officinalis, P. obscura, P. affinis, P. mollis and P. collina, whereas Clade II (BS: 89, PP: 0.85) in the nuclear phylogeny (subsequently referred to as nrClade II, Fig. 3) consists in general of specimens of P. angustifolia, P. longifolia, P. saccharata and P. montana. In contrast, Clade I (BS: 100, PP: 1.0) in the chloroplast phylogeny (subsequently referred to as cpClade I, Fig. 3) groups all specimens of P. officinalis and P. obscura (including one specimen ‘WET’ of P. angustifolia), whereas Clade II (BS: 94, PP: 0.95) consists of all the remaining Pulmonaria species (subsequently referred to as cpClade II, Fig. 3). Within nrClade I, specimens belonging to P. mollis and

365

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346

349 350 351 352 353 354 355 356 357 358 359 360 361 362

366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 6

S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

Table 1 Alignment and sequence characteristics of the different partitions (excluding outgroup specimens).

N° taxa Sequence length range Aligned sequence range Parsimony Informative characters Variable characters Constant characters AIC model

ITS

rpl16

rps16

trnH-psbA

59 694–696 696 30 (4.3%) 35 (5.0%) 661 GTR

52 831–858 858 14 (1.6%) 45 (5.2%) 813 GTR + I

58 777–790 798 8 (1.0%) 37 (4.6%) 761 GTR + I

58 370–373 373 14 (3.75%) 17 (4.60%) 356 GTR

414

P. collina (except for one P. mollis specimen) were grouped in one lineage (BS: 82, PP: 0.91), which is sister to the lineage containing P. obscura, P. officinalis and P. affinis (BS: 95, PP: 0.89). In addition, specimens of P. officinalis and P. affinis form one clade in nrClade I (BS: 74, PP: 0.92). Within cpClade I, there is a less clear delimitation of lineages as three specimens of P. officinalis (and one specimen of P. angustifolia) fall among the specimens of P. obscura. CpClade II is the least resolved lineage, yet this might be partially caused by the fact that four basal taxa (RY, BW, FP and GR) could not be sequenced for rpl16 despite the use of genus-specific designed primers. Both TCS haplotype networks (Fig. 4), based on mutational steps, showed the same grouping pattern with two major clusters of haplotypes detected in the nrDNA (nrClade I, II) and cpDNA (cpClade I, II) phylogenetic trees. These two groups were separated from each other by 7 and 17 mutational steps, respectively. Outgroups Borago officinalis and Symphytum asperum fell well beyond the maximum connection limit of 11 connection steps for ITS and 20 connection steps for cpDNA markers. The TCS haplotype network of both ITS and cpDNA sequences confirmed the results obtained in the phylogenetic trees and retrieved the same number of haplogroups with 12 haplotypes found for the ITS marker and 13 for the combined chloroplast markers (Figs. 3 and 4). Although the two major clades were well separated for each haplotype network, some admixture across clades was detected (‘‘conflicting” populations; see Section 3.2). Within the two major clades, haplotypes were frequently shared between species (represented as different colors in Fig. 4), especially in cpClade II in which chloroplast sequences showed low differentiation (Figs. 3 and 4).

415

3.2. Gene tree incongruence

416

In addition to the presence of soft incongruences between plastid and nuclear phylogenies (weakly supported conflicting relationships between the data matrices, and polytomous relationships for one dataset versus strongly or weakly supported in the other dataset), several cases of strong topological conflict are also present between both data sets when taking BS and PP support values into account (Fig. 3). None of the supported clades in both gene trees contained the same subset of taxa and additionally, discrepancies that are visually observed between both topologies (Figs. 3 and 4) were significantly substantiated by the different congruence tests. The ILD test demonstrated a significant incongruence between nrITS and all plastid markers, the latter ones either analyzed separately or combined (Table 2). AU and SH tests corroborate the topological incongruence that was observed in the ILD tests. Either nuclear or chloroplast topology as constraint showed no significant fit for the other data set (Table 3). Topological conflict was especially apparent in specimens collected from the four sympatric populations (BRZ, KV, VnV, PD) included in this study. The morphological intermediate specimens (‘hybrid’) from the sympatric Czech populations BRZ (P. angustifolia/P. obscura) and KV (P. mollis/P. obscura) showed high affinity with the putative paternal species for the biparentally inherited marker (nrITS) and with the putative maternal species for the

386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413

417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438

maternally inherited plastid markers (Fig. 3). Specimens KVm and VnVm which were morphologically identified as members of P. mollis did not share the same chloroplast haplotypes of P. mollis specimens from Germany and France, which form a clade that is sister to these Czech populations (Fig. 3). In addition, VnVm clustered with all of the P. obscura specimens in the nrITS gene tree. KVo and KV(B), collected from a sympatric population consisting of P. obscura and P. mollis clustered with P. angustifolia samples in the cpDNA topology. PDof which was morphologically identified as P. officinalis showed high affinity in nrITS sequence with the rest of the P. officinalis samples whereas it clustered with P. obscura in the cpDNA gene tree. Similar conflicts between gene topologies were found for allopatric populations such as WET, CdL, ZG, KB, COR and VAR which show high affinity with another species in one of both gene trees (Fig. 3). A higher level of incongruence was observed for P. mollis and P. collina which were grouped together both in the nrITS (BS: 82, PP: 0.91) and combined plastid topology (BS: 79) (Fig. 3). However, the entire clade had conflicting positions within both gene trees as it was embedded in nrClade I along with P. officinalis and P. obscura, and in cpClade II with the rest of the species. A similar pattern is observed for P. affinis, but is less clear as this species is not delineated as a monophyletic clade, but is part of a large polytomy in cpClade II (Fig. 3). ITS polymorphisms occurred at nucleotide positions 70, 105, 114, 115, 123, 129, 143, 146, 162, 174, 215, 216, 225, 260, 491, 492, 580, 630 and are summarized in Table S2. Ambiguity codes ‘‘Y” (C/T) and ‘‘W” (A/T) at positions 123 and 146, respectively, reveal polymorphisms between the two main nrClades, as nrClade I is represented by 123 = ‘‘T” and 146 = ‘‘T” and nrClade II by 123 = ‘‘C” 146 = ‘‘A”. Polymorphisms at these sites were observed in all populations of P. mollis (ALB, CHW, CHB, VOS, KVm, VnVm) and P. collina (EN, WH) (Table S2), and in all but three P. angustifolia populations (BRZa, KAR, CdL). Nucleotides at positions 105, 114, 115, 174, 215, 260, 491 differentiated P. montana from the other Pulmonaria species included in this study (Table S2). Cytosine ‘‘C” at position 129 and thymine ‘‘T” at position 580 were unique for P. longifolia, guanine ‘‘G” at position 70 and ‘‘C” at position 492 were restricted to P. mollis, and ‘‘T” at position 630 only occurred in P. officinalis and one P. affinis (ML) population. ITS polymorphisms were observed in three of the abovementioned allopatric populations showing conflicts across gene trees (VAR, COR, CdL), and in two additional P. montana (RB, VES) and one P. saccharata population (ITA) showing no conflict between nrITS and cpDNA trees (Fig. 3, Table S2). Interestingly, no ITS polymorphisms were observed in sympatric populations (except for the fixed polymorphisms in P. mollis) and, none of the morphological intermediates (KV(hybrid), BRZ(hybrid), PDof) from the sympatric Czech populations showed ITS polymorphisms either (Table S2).

439

3.3. Genealogical Sorting Index

488

GSI values and levels of significance for all species are summarized in Table 4. The GSI values ranged from 0.237 to 1.000 for the nrITS analysis, from 0.127 to 1.000 for the combined plastid and

489

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487

490 491

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

7

Fig. 3. Maximum Likelihood (ML) trees for nine species of Pulmonaria based on a nuclear marker (ITS; left) and three chloroplast markers (trnH-psbA, rpl16, rps16; right) with two outgroups (Symphytum asperum, Borago officinalis). Support values above branches (./.) represent Bootstrap values (as inferred by ML) and Posterior Probabilities (as inferred by Bayesian analysis). Lines indicate conflicts of population haplotypes (‘‘conflicting populations”) between nrITS and combined chloroplast trees. An asterisk indicates unsupported nodes for either method used. Zig-zag lines represent the original length of the branches. Sample names consist of the population code, species name and the country of origin (G = Germany, F = France, C = Czech Republic, GB = Great Britain, B = Belgium, E = Estonia, I = Italy). Superscripts (m) and (I) indicate whether the British P. longifolia populations were found on the British mainland or on the Isle of Wight.

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 8

S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

Fig. 4. Statistical parsimony networks (TCS networks) of haplotypes detected in this study based on a nuclear marker (ITS; left) and three chloroplast markers (trnH-psbA, rpl16, rps16; right). Numbers denote haplotypes; colors denote the nine species of Pulmonaria sampled in this study (for color legend see Fig. 2, black: Pulmonaria obscura, hatched: morphological intermediates, ‘‘hybrid”); size of circles is proportional to the number of individuals (populations) sharing the haplotype; lines represent mutational steps separating the different haplotypes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Table 2 Incongruence length difference results (P-values) obtained via an ILD test (a) complete taxon sampling (b) conflicting taxa excluded. Partitions (a) rpl16 rps16 trnH-psbA ITS Combined chloroplast (b) rpl16 rps16 trnH-psbA ITS Combined chloroplast

492 493 494 495 496 497

rpl16 _

_

rps16 0.55 _

0.68 _

trnH-psbA 0.41 0.38 _

0.72 0.85 _

Table 3 Results of the approximately unbiased (AU) and Shimodaira–Hasegawa (SH) tests. Constraint topology

ITS 0.01 0.01 0.01 _ 0.01 0.01 0.01 0.01 _ 0.01

from 0.183 to 0.655 for both genomes analyzed together (Table 4). For both the nrITS dataset and the ensemble of both genomes all species except for P. affinis achieved significant p-values, indicating that the null-hypothesis of mixed ancestry can be rejected for these species. GSI values were generally lower in the combined plastid analysis than in the nrITS analysis except for Pulmonaria

nrITS Chloroplast

nrITS

Chloroplast

AU

SH

AU

SH

0.63 <0.01

0.71 <0.01

<0.01 0.79

<0.01 0.88

saccharata, indicating that most of the species showed a higher degree of exclusivity based on the nrITS topology than on the combined plastid topologies. For the combined plastid dataset, the null-hypothesis of mixed ancestry was rejected for the majority of species except for P. affinis (0.128), P. obscura (0.172), P. collina (0.153) and P. angustifolia (0.127) (Table 4).

498

4. Discussion

504

4.1. Hybridization in Pulmonaria

505

Statistically significant incongruence and strong topological conflicts were present between gene trees of differently inherited

506

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

499 500 501 502 503

507

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx Table 4 Genealogical Sorting Index and significance levels (symbols) for the rejection of the null-hypothesis of mixed ancestry. Species

nrITS

Combined plastid

Both genomes

P. P. P. P. P. P. P. P. P.

0.769*** 0.761*** 0.237 (n.s.) 0.310* 0.722*** 0.405*** 1.000** 0.556*** 1.000***

0.308** 0.172 (n.s.) 0.128 (n.s.) 1.000*** 0.192* 0.299** 0.153 (n.s.) 0.127 (n.s.) 0.281**

0.539*** 0.466*** 0.183 (n.s.) 0.655*** 0.457*** 0.352*** 0.576** 0.341*** 0.641***

officinalis obscura affinis saccharata montana mollis collina angustifolia longifolia

(n.s.), non-significant * P < 0.05. ** P < 0.01. *** P < 0.001.

508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555

genetic markers (cpDNA and ITS; Table 2; Figs. 3 and 4), and allowed to detect current interspecific gene flow in sympatric populations and ancient hybridization events which might have led to the evolution of new evolutionary lineages (Sang and Zhong, 2000). Together with the presence of ITS polymorphisms and the inclusion of morphologically intermediate individuals from sympatric populations, we were able to identify patterns of hybridization prior to GSI calculations. Evidence was found of different degrees and timescales of hybridization ranging from current interspecific gene flow in sympatric populations [KV(hybrid), BRZ(hybrid), PDof, VnVm] to introgression at the population level [VAR, WET, COR, CdL, PAN, HS, SS, ZG, KB] and at least one event of hybrid speciation or adaptive introgression [P. mollis/collina]. For sympatric populations, topological conflict was especially apparent in the morphological intermediate specimens (‘hybrid’) from the sympatric Czech populations (BRZ, KV) as they grouped with their paternal parent for the biparentally inherited marker (nrITS) and with the maternal parent for the maternally inherited plastid markers (Figs. 3 and 4). Conflict was also found in specimens from sympatric populations (PDof, VnVm) not identified as morphological intermediates but as representatives of pure species indicating that backcrossing into the parental species might occur after hybridization. The potential of backcrossing of hybrids in parental species is further corroborated by the lack of ITS polymorphisms in the individuals of sympatric populations (see Section 4.2). Yet, ITS polymorphisms were found in isolated populations showing conflict between gene trees (VAR, COR, CdL). This indicates that the evidence of hybridization provided by gene tree incongruence might as well be detected as polymorphisms in ITS which are not subjected to confounding population genetic processes such as lineage sorting. The chloroplast haplotype of the P. saccharata population COR, for example, which occur in close vicinity of BF (P. montana) shows high affinity with the other P. saccharata samples included in this study whereas in the nrITS gene tree it occurs clustered with P. montana (Figs. 3 and 4). In addition, COR contains ITS polymorphisms at nucleotide positions 105, 114, 115, 174, 215 which indicate the co-occurrence of alleles derived from P. montana on the one hand and one of the other species (P. saccharata) on the other hand (Table S2). Sometimes, ITS polymorphisms may also provide clues of hybridization events in case one of the gene trees is unsufficiently resolved. For example, in the case of RB (P. montana) ITS polymorphisms are found at nucleotide positions 129 and 580 that differentiate P. longifolia from the rest of the species in this study (Table S2). Since chloroplast haplotypes of P. montana and P. longifolia are identical, no evidence of this historical hybridization between these two species was provided by a conflicting position of this population between

9

gene trees. Two conflicting populations (ZG, KB) identified as P. officinalis, however, did not exhibit polymorphisms at nucleotide position 630, which may again be indicative of backcrossing, in this case of P. officinalis  P. obscura hybrids with P. officinalis as the paternal parent (Table S2). The most remarkable case of hybridization within the genus is represented by the P. mollis/collina lineage in which an ancient hybridization event gave rise to an entirely new monophyletic lineage. Comparison of maternally inherited chloroplast DNA sequences with biparentally inherited nuclear ITS sequences (Fig. 3), and the occurrence of ITS polymorphism (Table S2) indicates that P. mollis/collina as a clade originated from a cross between a species from Clade I (e.g. P. obscura) as the paternal ancestor and a species from Clade II (e.g. P. longifolia, P. montana) as the maternal ancestor. In addition, P. mollis/collina individuals have summer leaves that lack spots and are intermediate in shape between the two major immaculate representatives of the two clades: P. obscura (Clade I), which has wide heart-shaped leaves and P. angustifolia (Clade II), which has narrow elliptical leaves (Fig. 1; Table S1). Furthermore, the single (intermediate) chromosome number 2n = 18 might be indicative of a homoploid hybridization event between a species with 2n = 14 (e.g. P. obscura – Clade I) and 2n = 22 (e.g. P. montana – Clade II). This ancient hybrid lineage occurs in a similar habitat as the other species, but exhibits denser glandularity as compared with any other Pulmonaria species (Fig. 1). Another case of ancient hybridization between both clades was evidenced in P. angustifolia by the presence of ITS polymorphisms (Table S2) and recently confirmed by more in-depth study on intra-individual ITS polymorphisms in Pulmonaria angustifolia and P. obscura corroborating our results regarding the hybrid origin of the majority of P. angustifolia populations analyzed in this study (Kook et al., 2015). Hybridization might thus have played a significant role in shaping evolutionary relationships within the genus Pulmonaria and several clues on a minor role of incomplete lineage sorting on the observed non-monophyly among the morphologically distinct species and gene tree incongruence are provided. Firstly, results from the permutation test for significance of the GSI in the ensemble dataset (Table 4, ‘‘both genomes”) indicate that the null hypothesis of mixed ancestry can be rejected for all Pulmonaria species except for P. affinis. Since the power to reject the null hypothesis of mixed genealogical ancestry markedly increases as does sample size in calculating GSI, this latter result might be attributable to our sampling size (N = 2) rather than incomplete lineage sorting (Cummings et al., 2008). Secondly, given their smaller effective populations size as compared to that of biparentally inherited nuclear markers, lineage sorting of uniparentally inherited plastid markers is expected to occur more rapidly in absence of hybridization (de Villiers et al., 2013 and references therein) and any deviation of this pattern might be explained by hybridization (Chan and Levin, 2005). GSI values showed deviations from this pattern by revealing more monophyly for the nrITS marker than the combined chloroplast markers and generally lower GSI values for the chloroplast marker (Table 4). When all conflicting populations (Fig. 3) were removed from the analysis, levels of monophyly increased markedly, and P. obscura and P. angustifolia became significantly monophyletic (results not shown) indicating that non-monophyly and tree incongruence in Pulmonaria are mainly caused by introgression. Thirdly, even after omitting all of the conflicting populations significant topological incongruence remained (Table 2) indicating that hybridization at higher intrageneric levels (e.g. P. mollis/P. collina) might still be present. Also, after omitting these hybrid specimens, lower GSI values for the uniparentally inherited plastid markers as compared to the biparentally inherited nuclear marker remained which presents a deviation from the pattern expected in the case of lineage sorting and

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 10

S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

623

could, therefore, corroborate a remaining effect of hybridization at higher intrageneric levels.

624

4.2. Barriers to hybridization

625

From our results, reproductive barriers in Pulmonaria species seem to some extent to be permeable to interspecific gene flow as we found molecular evidence of multiple hybridization events. Little is known about the strength of the different isolating barriers in Pulmonaria, yet premating barriers appear to be largely absent within the genus, as many species have overlapping distribution areas (Fig. 2), flower at the same time and share most of their pollinators (most often bees and bumblebees). However, recent research in the distylous genus Primula has shown that even small differences in flower morphology and herkogamy can lead to inefficient pollen transfer between species with a distylous breeding system (Keller et al., 2012). Since pollen derived from the same morph type (pin-pin, thrum-thrum pollinations) does not contribute to seed set due to the heteromorphic incompatibility system, differences in herkogamy between cooccurring species resulting in illegitimate pollen transfer could represent a potential barrier that also limits gene flow between Pulmonaria species (Keller et al., 2012). On the other hand, Ma et al. (2014) found that intra-morph crosses between two Primula species yielded more viable seeds (30%) than intra-morph crosses within species (<10%), significantly increasing the chances of interspecific gene flow between distylous species. Moreover, there is some evidence that the self-incompatibility in Pulmonaria may be leaky (Brys et al., 2008a,b; Meeus et al., 2012b), which may also affect interspecific gene flow. Data on post-mating barriers are currently scarce for Pulmonaria. Nevertheless, high variability in chromosome number within the genus suggests some putative difficulties with the formation of fertile hybrids (Sauer, 1975). On the other hand, the fact that in horticulture popular hybrids have been repeatedly used to generate new varieties (Bennett, 2003) and that fertile hybrids can be obtained after experimental interspecific crosses (Bolliger, 1982) suggest that these interspecific hybrids might at least be partially fertile, which would have important consequences for the establishment of natural hybrids and their potential to backcross with the parental species (Linder and Rieseberg, 2004; Abbott et al., 2013). Morphological intermediates in the sampled sympatric Pulmonaria populations were represented by only a few individuals, which might be indicative of a low fitness of hybrids. However, if hybrids would indeed be partially fertile, this low representation could also be explained by backcrossing leading to introgression. No recent evidence or direct assessment of sexual reproduction and/or backcrossing has been reported so far in natural Pulmonaria hybrids. However, the reconstruction of the evolutionary history of independently inherited genes and the presence of ITS polymorphisms in this study suggests that Pulmonaria hybrids are to some extent fertile. ITS polymorphisms were only found in the P. mollis/collina lineage and in a few other conflicting populations, but not in recent hybrids from sympatric populations (Table S1). When two species hybridize, a ITS copy of each parental genome will be present in the hybrid offspring. However in time, either through backcrossing with the parental species or through the homogenizing effect of concerted evolution, this evidence of past interspecific hybridization might become erased (Dover, 1982; Avise, 1994; Koch et al., 2003). However, in some species, additivity of parental ITS variants might still be found even in ancient hybrids (‘‘non-concerted evolution”, e.g. Rosa [Wisseman, 1999]; Pyrus [Zheng et al., 2008]), and is expected to be retained longer in asexually reproducing hybrids (Hodacˇ et al., 2014). The fixed ITS polymorphisms that were found in all sampled P. mollis

622

626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685

and P. collina populations might therefore indicate that these ancient hybrids mainly reproduce asexually which is not an uncommon mode of reproduction in the genus (Meeus et al., 2013). However, we did identify a hybrid (VnVm) between P. mollis and P. obscura in one of the sympatric populations, indicating that P. mollis is fertile despite its hybrid origin (Fig. 3; Table S1). Moreover, the loss of ITS polymorphisms in hybrids of sympatric populations (Table S2) provides additional evidence for the fertility of Pulmonaria hybrids, either through concerted evolution or through backcrossing, for both of which at least one cycle of reproduction is needed. This evidence was found in a sympatric population of two species with the same chromosome number (BRZ), but also in sympatric populations of species differing in chromosome number (KV, VnV, PD). The pathway in which a hybridization event gave rise to a new monophyletic lineage (P. mollis/collina) in Pulmonaria is still largely unknown, although the capacity to reproduce sexually might have facilitated the establishment and dispersion of this hybrid lineage. On the other hand, its fertility might have interfered with divergence from the parental species, slowing down the process of speciation (Mallet, 2007; Soltis and Soltis, 2009).

686

5. Conclusions

707

In this study, we found evidence of different degrees of hybridization ranging from current interspecific gene flow in sympatric populations to introgression at the population level and at least one event of hybrid speciation or adaptive introgression. Our results thus show that hybridization and introgression are evolutionary processes that have played a significant role in the evolutionary history of the distylous genus Pulmonaria. The absence of geographic clustering in the phylogeny of Pulmonaria, however, indicates that hybridization in the genus is not as ubiquitous to lead to convergence of lineages since the phylogenies clearly show genetically distinct species. Our results indicate that interspecific gene flow is ongoing in secondary contact zones regardless of the postulated reproductive barriers related to the mating system and karyotype differences. Moreover, the data indicate that these hybrids are at least partially fertile and that at least one hybrid lineage was able to become established as a new fertile species. The means by which this species became established remains unknown, however our data indicate that introgression through backcrossing of hybrids with parental species might have been involved in the speciation process.

708

6. Uncited references

728

Drummond and Rambaut (2007), Hilger et al. (2004) and Rieseberg (1997).

729

Acknowledgments

731

The authors thank Krista Takkis, Kersti Püssa, Frank Müller, Peter Otto, Lenz Meierott, the State Museum of Natural History Stuttgart, Martin Sanford, Alex Prendergast, Jonathan Cox, Clive Chatters, the Hampshire & Isle of Wight Wildlife Trust, Jana ˇ epka and Jean-Marc Tison for help with locating Ku˚rová, Radomir R the populations. We would also like to thank Janne Swaegers, Tobias Ceulemans, Kasper Van Acker and Geert Meeus for assistance in the field and Mohamed Abdelaziz for providing comments on the manuscript. This work was supported by the Flemish Fund for Scientific Research (F.W.O.) (project G.0500.10) and the European Research Council (ERC starting grant 260601–MYCASOR) (H.J.).

732

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706

709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727

730

733 734 735 736 737 738 739 740 741 742 743

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx 744

Appendix A. Supplementary material

745 747

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ympev.2015.11. 022.

748

References

749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823

Abbott, R., Albach, D., Ansell, S., et al., 2013. Hybridization and speciation. J. Evol. Biol. 26, 229–246. Avise, M.L., 1994. Molecular Markers, Natural History and Evolution. Chapman & Hall, New York. Bennett, M., 2003. Pulmonarias and the Borage Family. BT Batsford, London, UK. Bolliger, M., 1982. Pulmonaria in Westeuropa. Gantner Verlag K.G., Vaduz, Liechtenstein. Brewis, A., Bowman, P., Rose, F., 1996. The Flora of Hampshire. Brill Academic Pub, UK. Brys, R., Jacquemyn, H., Hermy, M., Beeckman, T., 2008a. Pollen deposition rates and the functioning of distyly in the perennial Pulmonaria officinalis (Boraginaceae). Plant Syst. Evol. 273, 1–12. Brys, R., Jacquemyn, H., Beeckman, T., 2008b. Morph ratio variation, population size and female reproductive success in the distylous Pulmonaria officinalis (Boraginaceae). J. Evol. Biol. 21, 1281–1289. Champluvier, D., Jacquemart, A.-L., 1999. Quelques observations sur deux populations gaumaises (Lorraine belge) de Pulmonaria montana (Boraginaceae). Belg. J. Bot. 132, 13–25. Chan, K.M.A., Levin, S.A., 2005. Leaky prezygotic isolation and porous genomes: rapid introgression of maternally inherited DNA. Evolution 59, 720–729. Clement, M., Posada, D., Crandall, K.A., 2000. TCS: a computer program to estimate gene genealogies. Mol. Ecol. 9, 1657–1659. Cohen, J.I., 2013. A phylogenetic analysis of morphological and molecular characters of Boraginaceae: evolutionary relationships, taxonomy, and patterns of character evolution. Cladistics 30, 139–169. Coyne, J.A., Orr, H.A., 1989. Patterns of speciation in Drosophila. Evolution 43, 849– 857. Coyne, J.A., Orr, H.A., 1997. ‘‘Patterns of speciation in Drosophila” revisited. Evolution 51, 295–303. Cummings, M.P., Neel, M.C., Shaw, K.L., 2008. A genealogical approach to quantifying lineage divergence. Evolution 62, 2411–2422. Darwin, C., 1877. The Different Forms of Flowers on Plants of the Same Species. John Murray, London, UK. de Villiers, M.J., Pirie, M.D., Hughes, M., Möller, M., Edwards, T.J., Bellstedt, D.U., 2013. An approach to identify putative hybrids in the ‘coalescent stochasticity zone’, as exemplified in the African plant genus Streptocarpus (Gesneriaceae). New Phytol. 198, 284–300. Dover, G., 1982. Molecular drive: a cohesive mode of species evolution. Nature 299, 111–117. Doyle, J.J., Doyle, J.L., 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19, 11–15. Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol. 7, 214. Farris, J.D., Källersjö, M., Kluge, A.G., Bult, C., 1995. Constructing a significance test for incongruence. Syst. Biol. 44, 570–572. Feliner, G.N., Larena, B.G., Aguilar, J.F., 2004. Fine-scale geographical structure, intra-individual polymorphism and recombination in nuclear ribosomal internal transcribed spacers in Armeria (Plumbaginaceae). Ann. Bot. 93, 189– 200. Gams, H., 1927. Pulmonaria L. In: Hegi, G. (Ed.), Illustrierte Flora von Mitteleuropa, Band V, Teil 3. J.F. Lehmanns Verlag, München, Germany, p. 2220. Hedrick, P.W., 2007. Sex: differences in mutation, recombination, selection, gene flow and genetic drift. Evolution 61, 2750–2771. Hegarty, M.J., Hiscock, S.J., 2005. Hybrid speciation in plants: new insights from molecular studies. New Phytol. 165, 411–423. Hennig, W., 1966. Phylogenetic Systematics. University of Illinois Press, Urbana. Hewitt, J., 1994. Pulmonarias. The Hardy Plant Society, Worcestershire, UK. Hilger, H.H., Selvi, F., Papini, A., Bigazzi, M., 2004. Molecular systematics of Boraginaceae tribe Boragineae based on ITS1 and trnL sequences, with special reference to Anchusa s.l. Ann. Bot. 94, 201–212. Hill, M.O., Preston, C.D., Roy, D.B., 2004. PLANTATT. Attributes of British and Irish Plants: Status, Size, Life History, Geography and Habitats. Centre for Ecology and Hydrology, Monks Wood, Cambridgeshire, UK. Hodacˇ, L., Scheben, A.P., Hojsgaard, D., Paun, O., Hörandl, E., 2014. ITS polymorphisms shed light on hybrid evolution in apomictic plants: a case study on the Ranunculus auricomus complex. PLoS ONE 9 (7). Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17, 754–755. Huson, D., Kloepper, T., Lockhart, P., Steel, M., 2005. Reconstruction of reticulate networks from gene trees. In: Proceedings of the Ninth International Conference on Research in Computational Molecular Biology (RECOMB 2005), SpringerVerlag, Cambridge, MA, Heidelberg, LNBI 3500, pp. 233–249. Katoh, K., Misawa, K., Kuma, K., Miyata, T., 2002. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30, 3059–3066.

746

11

Keller, B., de Vos, J.M., Conti, E., 2012. Decrease of sexual organ reciprocity between heterostylous primrose species, with possible functional and evolutionary implications. Ann. Bot. 110, 1233–1244. Kirchner, D.E., 2004. Molekulare Phylogenie und Biogeographie der Gattung Pulmonaria L. (Boraginaceae). Verlag Mainz, Aachen, Germany. Koch, M.A., Dobes, C., Mitchell-Olds, T., 2003. Multiple hybrid formation in natural populations: concerted evolution of the internal transcribed spacer of nuclear ribosomal DNA (ITS) in North American Arabis divaricarpa (Brassicaceae). Mol. Biol. Evol. 20, 338–350. Kook, E., Vedler, E., Püssa, K., Kalamees, R., Reier, Ü., Pihu, S., 2015. Intraindividual ITS polymorphism and hybridization in Pulmonaria obscura Dumort. and Pulmonaria angustifolia L. (Boraginaceae). Plant Syst. Evol. 301, 893–910. Kubatko, L.S., 2009. Identifying hybridization events in the presence of coalescence via model selection. Syst. Biol. 58, 478–488. Legendre, P., Makarenkov, V., 2002. Reconstruction of biogeographic and evolutionary networks using reticulograms. Syst. Biol. 51, 199–216. Linder, C.R., Rieseberg, L.H., 2004. Reconstructing patterns of reticulate evolution in plants. Am. J. Bot. 91, 1700–1708. Ma, Y., Xie, W., Tian, X., Sun, W., Wu, Z., Milne, R., 2014. Unidirectional hybridization and reproductive barriers between two heterostylous primrose species in north-west Yunnan, China. Ann. Bot. 113, 763–775. Mallet, J., 2007. Hybrid speciation. Nature 446, 279–283. McBreen, K., Lockhart, P.J., 2006. Reconstructing reticulate evolutionary histories of plants. Trends Plant Sci. 11, 398–404. Meeus, S., Honnay, O., Jacquemyn, H., 2012a. Strong differences in genetic structure across disjunct, edge and core populations of the distylous forest herb Pulmonaria officinalis (Boraginaceae). Am. J. Bot. 99, 1809–1818. Meeus, S., Honnay, O., Brys, R., Jacquemyn, H., 2012b. Biased morph ratios and skewed mating success contribute to loss of genetic diversity in the distylous Pulmonaria officinalis. Ann. Bot. 109, 227–235. Meeus, S., Brys, R., Honnay, O., Jacquemyn, H., 2013. Biological Flora of the British Isles: Pulmonaria officinalis. J. Ecol. 101, 1353–1368. Merxmüller, H., Sauer, W., 1972. Pulmonaria L. In: Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M., Valentine, D.H., Walters, S.M., Webb, D.A. (Eds.), Flora Europaea, vol. 3. Cambridge University Press, Cambridge, UK, p. 100. Moyle, L.C., Olson, M.S., Tiffin, P., 2004. Patterns of reproductive isolation in three angiosperm genera. Evolution 58, 1195–1208. Olesen, J.M., 1979. Floral morphology and pollen flow in the heterostylous species Pulmonaria obscura Dumort (Boraginaceae). New Phytol. 82, 757–767. Oxelman, B., Liden, M., Berglund, D., 1997. Chloroplast rps16 intron phylogeny of the tribe Sileneae (Caryophyllaceae). Plant Syst. Evol. 206, 393–410. Palumbi, S.R., Cipriano, F., Hare, M.P., 2001. Predicting nuclear gene coalescence from mitochondrial data: the three-times rule. Evolution 55, 859–868. Posada, D., 2008. JModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25, 1253–1256. Rambaut, A., Drummond, A.J., 2007. Tracer [Computer Program]. Richards, A.J., Mitchell, J., 1990. The control of incompatibility in distylous Pulmonaria affinis Jordan (Boraginaceae). Bot. J. Linn. Soc. 104, 369–380. Rieseberg, L.H., 1997. Hybrid origins of plant species. Annu. Rev. Ecol. Syst. 28, 359– 389. Ronquist, F., Huelsenbeck, J.P., 2003. MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572–1574. Rosenberg, N.A., Nordborg, M., 2002. Genealogical trees, coalescent theory and the analysis of genetic polymorphisms. Nature 3, 380–390. Saghai-Maroof, M.A., Soliman, K.M., Jorgensen, R.A., Allard, R.W., 1984. Ribosomal DNA spacer-length polymorphisms in barley: Mendelian inheritance, chromosomal location, and population dynamics. Proc. Natl. Acad. Sci. 81, 8014–8019. Sang, T., Zhong, Y., 2000. Testing hybridization hypotheses based on incongruent gene trees. Syst. Biol. 49, 422–434. Sang, T., Crawford, D.J., Stuessy, T.F., 1997. Chloroplast DNA phylogeny, reticulate evolution, and biogeography of Paeonia (Paeoniaceae). Am. J. Bot. 84, 1120– 1136. Sauer, W., 1975. Karyosystematische Untersuchungen an der Gattung Pulmonaria (Boraginaceae): Chromosomen-Zahlen, Karyotyp-Analysen und allgemeine Hinweise auf die Entwicklungsgeschichte. Bibl. Bot. Heft 131, 1–85. Schmidt-Lebuhn, A.N., Keller, B., De Vos, J.M., Conti, E., 2012. Phylogenetic analysis of Primula section Primula reveals rampant non-monophyly among morphologically distinct species. Mol. Phylogenet. Evol. 65, 23–34. Scopece, G., Muscacchio, A., Widmer, A., Cozzolino, S., 2007. Patterns of reproductive isolation in Mediterranean deceptive orchids. Evolution 61, 2623–2642. Selvi, F., Bigazzi, M., Hilger, H.H., Papini, A., 2006. Molecular phylogeny, morphology and taxonomic re-circumscription of the generic complex Nonea/Elizaldia/ Pulmonaria/Paraskevia (Boraginaceae-Boragineae). Taxon 55, 907–918. Servedio, M.R., Noor, M.A.F., 2003. The role of reinforcement in speciation: theory and data. Annu. Rev. Ecol. Evol. Syst. 34, 339–364. Shimodaira, H., 2002. An approximately unbiased test of phylogenetic tree selection. Syst. Biol. 51, 492–508. Shimodaira, H., Hasegawa, M., 1999. Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Mol. Biol. Evol. 16, 1114–1116. Shimodaira, H., Hasegawa, M., 2001. CONSEL: for assessing the confidence of phylogenetic tree selection. Bioinformatics 17, 1246–1247. Soltis, P.S., Soltis, D.E., 2009. The role of hybridization in plant speciation. Annu. Rev. Plant Biol. 60, 561–588.

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909

YMPEV 5371

No. of Pages 12, Model 5G

30 December 2015 12 910 911 912 913 914 915 916 917 918 919 920

S. Meeus et al. / Molecular Phylogenetics and Evolution xxx (2015) xxx–xxx

Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for the RAxML web servers. Syst. Biol. 57, 758–771. Stewart, A., Pearman, D.A., Preston, C.D., 1994. Scarce Plants in Britain. JNCC, UK, pp. 242–243. Suzuki, K., Kita, Y., Kato, M., 2002. Comparative developmental anatomy of seedlings in nine species of Podostemaceae (subfamily Podostemoideae). Ann. Bot. 89, 755–765. Swofford, D.L., 2002. PAUP⁄. Phylogenetic Analysis using Parsimony (⁄and other methods). Sinauer Associates, Sunderland, MA. Vereecken, N.J., Cozzolino, S., Schiestl, F.P., 2010. Hybrid floral scent novelty drives pollinator shift in sexually deceptive orchids. BMC Evol. Biol. 10, 103.

White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J. (Eds.), PCR Protocols: A Guide to Methods and Applications. Academic Press, New York, USA, pp. 315–322. Wisseman, V., 1999. Genetic constitution of Rosa Sect. Caninae (R. canina, R. junzillii) and Sect. Gallicanae (R. gallica). J. Appl. Bot. 73, 191–196. Zheng, X., Cai, D., Yao, L., Teng, Y., 2008. Non-concerted ITS evolution, early origin and phylogenetic utility of ITS pseudogenes in Pyrus. Mol. Phylogenet. Evol. 48, 892–903.

Please cite this article in press as: Meeus, S., et al. Evolutionary trends in the distylous genus Pulmonaria (Boraginaceae): Evidence of ancient hybridization and current interspecific gene flow. Mol. Phylogenet. Evol. (2015), http://dx.doi.org/10.1016/j.ympev.2015.11.022

921 922 923 924 925 926 927 928 929 930