Research Article |
Corresponding author: Elżbieta Cieślak ( e.cieslak@botany.pl ) Academic editor: Pamela S. Soltis
© 2024 Elżbieta Cieślak, Michał Ronikier, Magdalena Szczepaniak.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Cieślak E, Ronikier M, Szczepaniak M (2024) Glacial history of Saxifraga wahlenbergii (Saxifragaceae) in the context of refugial areas in the Western Carpathians. PhytoKeys 246: 295-314. https://doi.org/10.3897/phytokeys.246.118796
|
Despite the wealth of data available for mountain phylogeography, local-scale studies focused on narrow endemic species remain rare. Yet, knowledge of the genetic structure of such species biogeographically linked to a restricted area is of particular importance to understand the history of the local flora and its diversity patterns. Here, we aim to contribute to the phylogeographical overview of the Western Carpathians with a genetic study of Saxifraga wahlenbergii, one of the most characteristic endemic species of this region. We sampled populations from all discrete parts of the species’ distribution range to apply sequencing of selected non-coding cpDNA and nuclear ribosomal DNA (ITS) regions, as well as Amplified Fragment Length Polymorphism (AFLP) fingerprinting. First, while ITS sequences showed weak diversification, the genetic structure based on cpDNA sequences revealed two well-differentiated groups of haplotypes. One of them is restricted to the main center of the distribution range in the Tatra Mountains (Mts), while the second group included a series of closely related haplotypes, which in most cases were unique for particular isolated groups of populations in peripheral mountain ranges and in the south-eastern part of the Tatra Mts. AFLP fingerprinting also revealed a pattern of divergence among populations, while only partly corroborating the division observed in cpDNA. Taking into account all the data, the pattern of genetic structure, supported by the high levels of unique genetic markers in populations, may reflect the historical isolation of populations in several local refugia during the last glacial period. Not only the center of the range in the Tatra Mts, but also other, neighboring massifs (Malá Fatra, Nízke Tatry, Chočské vrchy, Muránska planina), where populations are characterized by separate plastid DNA haplotypes, could have acted as separate refugia.
AFLP, haplotype, high mountain plant, narrow endemic, phylogeography, refugial areas, Tatra Mts
Mountainous systems of temperate Europe are biodiversity hotspots, due, among others, to their high floristic richness, including endemic species that contribute to the natural uniqueness of a given biogeographical unit. The evolution of mountain species is strongly related to historical environmental and climatic factors, including the Pleistocene climatic fluctuations that caused alternating glacial and interglacial periods (
Endemic species are a special element of the biodiversity of mountain areas. In this group, species with highly limited local ranges are of particular importance, as they most often reflect evolution in specific microclimatic isolation in geographically small and isolated areas (
It has been confirmed that the Western Carpathians are an important, independent center of endemism with both Pan-Carpathian species and a group of endemic species specific to this region (
In this study, we address Saxifraga wahlenbergii Ball, one of the most characteristic endemic species of the Western Carpathians, with wide ecological preferences, including a large elevational range and limited bedrock restriction. It occurs in several massifs with its range center in the Tatra Mts (
Interestingly, in the framework of the phylogenetic analysis, regional genetic variation was found in the populations of S. wahlenbergii including two distinct cpDNA clades, which was a direct motivation to undertake a more detailed analysis of the species’ phylogeographical structure. While unequivocally dating the hybrid origin of S. wahlenbergii could not be assessed, it could theoretically precede the Pleistocene and several arguments pointed to a possibly ancient rather than recent age of this species (
In the region of the Western Carpathians, with its high topographic and habitat heterogeneity, S. wahlenbergii, a plant with a rather wide altitudinal range, could survive in the massifs where it occurs today, over the Pleistocene climatic oscillations, following the altitudinal shifts or persistence in long-term, non-glaciated microrefugia. However, as cold (glacial) periods have led to a significant increase in habitats suitable for alpine plants in the lower parts of the Carpathians (
The main objective of this study is to determine the range-wide genetic structure of S. wahlenbergii, to provide insight into its glacial history. For this purpose, an extended sequence analysis of selected nuclear and plastid DNA regions complementing data from
Based on the above data, an attempt was made to resolve whether the contemporary distribution results from long-term survival in several isolated areas (local refugia) and thus is a relic of ancient events or whether the species recently spread from a single refugium (likely located in the Tatra Mts – central part of the range).
Saxifraga wahlenbergii Ball (sect. Saxifraga; Saxifragaceae) is an endemic perennial species of the Western Carpathians (in Poland and Slovakia). Its range includes the massifs of Tatra Mts, Malá Fatra Mts, Chočské vrchy Mts, Nízke Tatry Mts, and Muránska planina. However, it is a common species only in the Tatra Mts (the highest and environmentally most complex massif of the Western Carpathians) – it is abundant at higher altitudes above the tree line (up to 2540 m a.s.l.) and also descends to lower elevations, e.g., along streams (880 m a.s.l.) (
Plant material of Saxifraga wahlenbergii was sampled in natural populations, spanning the entire natural distribution area of this species in the Western Carpathians. Populations were assigned to regional geographical units, which were further assigned into predefined groups: the Western Tatra Mts, the Eastern Tatra Mts and those outside of the Tatra Mts, including localities from: the Malá Fatra Mts, Chočské vrchy Mts, Nízke Tatry Mts and Muránska planina (
Localities of populations of Saxifraga wahlenbergii used in the study and parameters of their genetic variability based on AFLP, nrDNA (ITS) and cpDNA sequences. NA/NS – population sampling for AFLP analysis and DNA sequencing; P/% – number and percentage of polymorphic markers; He – mean (±SD) Nei’s gene diversity; I – mean (±SD) Shannon’s Index; DW – frequency down-weighted marker values; R – ribotypes (variants of ITS of nrDNA) and H – haplotypes (variants of cpDNA) in population (the number of individuals representing a particular ribotype or haplotype is given in parentheses). Country code: PL – Poland, SK – Slovakia. Collectors code: AD – Anna Delimat, AR – Anna Ronikier, MR – Michał Ronikier, RL – Roman Letz, PM – Patrik Mráz, PT – Peter Turis.
Code | Locality | NA/NS | AFLP | ITS | cpDNA | |||
---|---|---|---|---|---|---|---|---|
P/% | He | I | DW | R(No.) | H(No.) | |||
Western Tatra Mts (Tatry Zachodnie, Západné Tatry) | ||||||||
S1 | PL, Dolina Chochołowska valley, 1370 m a.s.l., 49°14'N, 19°48'E (AD) | 5/2 | 59/27.31 | 0.10 (±0.18) | 0.15 (±0.26) | 21.77 | R1(2) | H1(2) |
S2 | PL, between the Gaborowa Przełęcz pass and Bystra Przełęcz pass, ~1930 m a.s.l., 49°12'N, 19°49'E (RL, PM) | 4/2 | 61/28.24 | 0.12 (±0.19) | 0.17 (±0.28) | 20.06 | R2(2) | H1(1) H3(1) |
S3 | PL, Przełęcz pod Kopą Kondracką pass, 1500 m a.s.l., 49°14'N, 19°55'E (AD) | 5/2 | 61/28.24 | 0.11 (±0.18) | 0.16 (±0.26) | 25.07 | R1(2) | H1(1) H4(1) |
S4 | PL, Piekiełko (Piekło) valley, 1640 m a.s.l., 49°14'N, 19°56'E (AD) | 9/2 | 91/42.10 | 0.15 (±0.20) | 0.23 (±0.28) | 58.57 | R2(1) R3(1) | H1(2) |
Eastern Tatra Mts (High Tatra Mts, Tatry Wysokie, Vysoké Tatry) | ||||||||
S5 | PL, N slopes of the pass Zawrat, 2100 m a.s.l., 49°13'N, 20°01'E (MR) | 4/2 | 79/36.57 | 0.14 (±0.19) | 0.20 (±0.28) | 25.46 | R2(1) R3(1) | H1(1) H2(1) |
S6 | PL, Mięguszowiecki Szczyt Czarny Mt., 2220 m a.l.s., 49°11'N, 20°03'E (AD) | 7/2 | 83/38.43 | 0.14 (±0.20) | 0.20 (±0.28) | 36.37 | R1(2) | H1(1) H7(1) |
S7 | SK, Hrubý vrch Mt, ~ 2350 m a.s.l., 49°10'N, 20°01'E (MR) | 8/2 | 85/39.35 | 0.14 (±0.19) | 0.21 (±0.28) | 40.09 | R3(2) | H7(2) |
Malá Fatra Mts | ||||||||
S8 | SK, Veľký Rozsutec Mt., 1550 m a.s.l., 49°14'N, 19°06'E (MR, AR) | 5/2 | 76/35.19 | 0.13 (±0.19) | 0.19 (±0.27) | 26.90 | R1(1) R5(1) | H10 (2) |
Chočské vrchy Mts | ||||||||
S9 | SK, Veľký Choč Mt., 1600 m a.s.l., 49°09'N, 19°20'E (MR) | 4/2 | 60/27.78 | 0.10 (±0.18) | 0.15 (±0.26) | 23.36 | R1(1) R4(1) | H5(1) H6(1) |
Nízke Tatry Mts | ||||||||
S10 | SK, Siná Mt., 1422 m a.s.l., 49°00'N, 19°33'E (RL, PT) | 4/2 | 61/28.24 | 0.11 (±0.19) | 0.27 (±0.27) | 29.54 | R1(2) | H8(1) H9(1) |
Spišsko-gemerský kras | ||||||||
S11 | SK, Muránska planina, Vel’ka Stožka, 1242 m a.s.l., 48°46'N, 19°58'E (PT) | –/2 | – | – | – | – | R1(1) R2(1) | H11 (2) |
Location of studied populations of Saxifraga wahlenbergii and their genetic variability based on DNA sequence data A distribution of 11 populations of S. wahlenbergii and haplotypes and ribotypes in the populations B haplotype network based on the combined chloroplast regions: rps16-trnK and rpl32-trnL C ribotype network based on ITS region. Networks obtained from TCS based on a 95% connection limit. The relative sizes of circles in networks are proportional to haplotype and ribotype frequencies. For population acronyms see Table
The total genomic DNA was isolated from 5–10 mg of dried leaf tissue of collected samples using the DNeasy Plant Mini Kit system (Qiagen, Hilden, Germany) according to the manufacturer’s protocol (final elution step was carried out using 2×50 μL of elution buffer). DNA quality and concentration were estimated against λ-DNA on 1% agarose gel stained with ethidium bromide. The purified DNA isolates were the basis of DNA sequencing and AFLP analyses. Samples from the Muránska planina population (S11) were collected later than the core sample set and they could only be used in the sequencing analysis.
The non-coding chloroplast DNA regions (cpDNA) – rps16-trnK and rpl32-trnL (
AFLP analysis was performed according to
Analyses of cpDNA and ITS of nrDNA regions were performed separately. Forward and reverse DNA sequences data were automatically assembled and aligned based on ClustalW algorithm (
Gene diversity (h) and nucleotide diversity (π) were calculated based on cpDNA and ITS sequence variation for the total sample of Saxifraga wahlenbergii and for predefined region groups within the species range using the DNAsp 5.0 program (
The genetic diversity of Saxifraga wahlenbergii at species and within-species level (populations) was assessed on the basis of binary AFLP data matrix by calculating the genetic parameters, including the number (P) and percentage of polymorphic markers (%) , Nei’s gene diversity (He), Shannon’s information index (I) and gene flow (Nm) using POPGENE v. 1.32 software (
The sequences of cpDNA and ITS of nrDNA regions were obtained from 22 individuals from eleven populations of Saxifraga wahlenbergii (Table
The alignments of rps16-trnK and rpl32–trnL regions of cpDNA were 796 bp and 655 bp in length, respectively (concatenated cpDNA alignment – 1451 bp in length). 15 variable sites were found – 4 singleton variable sites and 11 parsimony informative sites, which represented transitions (C-T, 3 A-G) and transversions (3 G-T, 2 A-T and 2 A-C). In the rpl32–trnL region a thirty-one-nucleotide insertion/deletion was also identified. Eleven haplotypes (H1–H11; Fig.
Each population harbored one or two cpDNA haplotypes, mostly specific for individual populations and/or mountain ranges. H1–H4 and H7 haplotypes occurred only in the Tatra Mts, with the most frequent H1 haplotype present in almost all Tatra populations (six out of seven populations) (Fig.
Genetic diversity of nrDNA and cpDNA sequences of Saxifraga wahlenbergii calculated for a priori delimitation of regional groups. R – ribotypes (variants of nrDNA ITS) and H – haplotypes (variants of cpDNA); h – mean (±SD) gene diversity; π – mean (±SD) nucleotide diversity; D – Tajima’s D statistic value; *non-significant at the 5% level (P > 0.05).
Regional groups | ITS | cpDNA | ||||||
---|---|---|---|---|---|---|---|---|
R | h | π | D | H | h | π | D | |
Western Tatra Mts | R1, R2, R3 | 0.00 | 0.00 | 0.00 | H1, H3, H4 | 0.46 (±0.20) | 0.0003 (±0.0007) | -1.31* |
Eastern Tatra Mts | R1, R2, R3 | 0.00 | 0.00 | 0.00 | H1, H2, H7 | 0.73 (±0.16) | 0.0036 (±0.0004) | 1.80* |
Tatra Mts (as a whole) | R1, R2, R3 | 0.00 | 0.00 | 0.00 | H1, H2, H3, H4, H7 | 0.66 (±0.12) | 0.0024 (±0.0008) | -0.15* |
Outside of the Tatra Mts | R1, R2, R4, R5 | 0.00 | 0.00 | 0.00 | H5, H6, H8, H9, H10, H11 | 0.86 (±0.11) | 0.0013 (±0.0003) | -0.33* |
Overall, S. wahlenbergii displays a moderate gene diversity (h = 0.81 ±0.06) and nucleotide diversity (π = 0.0036 ±0.0003) of cpDNA. At the local level of predefined regional groups, the highest gene diversity (h = 0.86 ±0.11) was found in the group of populations outside of the Tatra Mts (i.e., in Malá Fatra Mts, Chočské vrchy Mts, Nízke Tatry Mts and Muránska planina) with low nucleotide diversity (π = 0.0013 ±0.0003; Table
The obtained ITS alignment was 730 bp long, with very low sequence diversity. Only four indels (three poly-A and one poly-G stretches) were found and on this basis five ribotypes were established (R1–R5) (Fig.
The AFLP analysis yielded 213 DNA markers, of which 181 (84.98%) were polymorphic for 55 individuals from ten populations of Saxifraga wahlenbergii (Table
At the species level, Nei’s gene diversity (He) was 0.16 (±0.17) and ranged from 0.10 (S1) to 0.15 (S4), with an average value of 0.12 (±0.02). The frequency of down-weighted markers (DW) was similar across most populations and ranged from 20 to 30, with much higher values in populations S4 (59), S7 (40) and S6 (36) (Table
The further analysis of PCoA performed on the entire dataset revealed that S. wahlenbergii populations are not clearly genetically divergent and form partially overlapping groups. In general, the population’s scatter is characterized by the west-east gradient across the distribution range of S. wahlenbergii. In 1–3 axes arrangement, the populations from the disjunct parts of the range (populations: S1, S8, S9 and S10) are opposite to the highest locations of Hrubý vrch Mt. and Mięguszowiecki Szczyt Czarny Mt. (the Eastern Tatra Mts). In the central part of plot, individuals from population of the Western Tatra Mts (S3, S4) and Malá Fatra Mts were located. The first three factors of the PCoA accounted for 35.39% of the total variation in the dataset (Fig.
Phylogeographic structure within of Saxifraga wahlenbergii based on AFLP dataset (55 individuals from 10 populations) A Principal Component Analysis diagram, ordination at 1 vs 2 vs 3 axes B Neighbor-Net diagram with the bootstrap values derived from an analysis of 2,000 replicates above 50% has been given. Both diagrams were prepared based on the Nei-Li coefficient. For population acronyms see Table
The Neighbor-Net diagram demonstrated two groups, each consisting of clusters representing single, spatially isolated populations, but with different bootstrap support. The first group included those with higher bootstrap values, such as Nízke Tatry Mts (98%), Dolina Chochołowska valley (96%), Gaborowa Przełęcz pass (85%), Veľký Choč Mt. (62%) and Malá Fatra Mts (61%). The second group consisted of populations with very low support (Fig.
Analysis of genetic variation (AMOVA) showed that a major part of S. wahlenbergii variation is attributed to the within-population level – 72.09%, in relation to among-population variation – 27.91% (FST = 0.28, P < 0.001). The same pattern can be found when analyzing geographical groups (Table
AMOVA analysis based on AFLP data for the populations of Saxifraga wahlenbergii calculated for all populations and a priori delimitation of regional groups. Significance tests based on 1023 permutations, ***P < 0.001, **P < 0.01.
Source of variation | d.f. | Sums of Squares | Variance components | % Total variance | F statistics |
---|---|---|---|---|---|
Among populations | 9 | 487.480 | 6.753 | 27.91*** | F ST = 0.28 |
Within populations | 45 | 784.738 | 17.439 | 72.09 | |
Total | 54 | 1272.218 | 24.192 | ||
Among regional groups – Western Tatra Mts vs. Eastern (High) Tatra Mts vs. outside of the Tatra Mts | 2 | 145.588 | 1.206 | 4.92** | F CT = 0.05 |
Among populations | 7 | 341.892 | 5.872 | 23.95*** | F SC = 0.25 |
Within population | 45 | 784.738 | 17.439 | 71.13 | F ST = 0.29 |
Total | 54 | 1272.218 | 24.517 |
AFLP data indicates a low level of gene flow between populations of S. wahlenbergii (Nm = 0.66). In the Structure analysis of AFLP data, the stable and optimal number of population groups was selected based on the kink in the envelope of lnP(D) values. As can be seen from Fig.
A The histograms representing the assignment of 55 individuals of Saxifraga wahlenbergii to different clusters by Bayesian spatial clustering (STRUCTURE software). Each vertical bar corresponds to an individual, highlighted in gray for clarity, contrasting with the cluster assignments, respectively, at K = 2 and K = 3 B in P(Data) values in function of K are shown. For population acronyms see Table
Phylogenetic analyses of the Saxifraga section indicated the hybrid origin of Saxifraga wahlenbergii and allowed us to estimate the possible oldest age of its hybrid origin to the late Neogene (4.7 Ma). These analyses also provided insights into its internal diversity (
In mountain conditions, the process of isolation by distance contributed to historical interruption of gene flow between populations, leading to geographically driven groups of populations (e.g.,
In the Tatra Mts, subalpine populations of S. wahlenbergii are more closely related to those geographically closest from the same mountain range than to their subalpine counterparts from other mountain ranges. This suggests that the source area of their recolonization could have been populations from low elevations, such as the extant population from the Dolina Chochołowska valley, a site which remained outside the glaciation area (
The Tatra Mts, unlike most of the Western Carpathians, were strongly, albeit unevenly, glaciated during the Pleistocene glaciations (
On the other hand, low values of FST (the lowest in relation to the compared pairs of populations) observed in the AFLP data from the Tatra Mts can be the result of the maintenance of gene flow between populations during recolonization of this area after the last glaciation and may counteract incipient differentiation processes, thereby avoiding bottlenecks, genetic drift, and the loss of genetic diversity. Characteristically, the highest values of the FST were noted between populations from the areas with the highest altitudes, namely the Tatra Mts and the Nízke Tatry Mts. These results suggest that mountain ridges acted as a stronger barrier for gene flow more effectively than the elevation differences between subalpine and lower-lying areas within the same ranges. In addition, the genetic structure of S. wahlenbergii, a relic mountain plant species, certainly reflects processes acting in different time periods. Populations that survived when environmental conditions became unfavorable could retain genetic variability. Becoming a source of remigration in new conditions, they could also host new local mutation fixations. Therefore, it can be assumed that both Quaternary climatic oscillations and ecological divergence have played a role in shaping the distribution and divergence patterns observed in S. wahlenbergii. Similarly, in the species complex Alyssum montanum–A. repens, a clear elevational shift was identified, indicating that differential ecological adaptation occurred in the respective mountain areas (
We would like to thank the Editor and Reviewer for their useful comments and suggestions; Tony Dixon for improving our English; Anna Delimat, Anna Ronikier, Roman Letz, Patrik Mráz and Peter Turis for their valuable help in sampling. Collecting permits were granted by Tatrzański Park Narodowy, Poland (no. Bot-203) and by Ministerstvo Životného Prostredia Slovenskej Republiky (Rozhodnutie MZP SR no. 6188/2017-6.3 from 13.12.2017).
The authors have declared that no competing interests exist.
No ethical statement was reported.
This study was funded from the statutory funds of the W. Szafer Institute of Botany, Polish Academy of Sciences.
Elżbieta Cieślak: Research concept and design, Data analysis and interpretation, Writing the article, Critical revision of the article, Final approval of the article. Michał Ronikier: Research concept and design, Collection and/or assembly of data, Critical revision of the article, Final approval of the article. Magdalena Szczepaniak: Data analysis and interpretation, Critical revision of the article, Final approval of the article.
Elżbieta Cieślak https://orcid.org/0000-0002-6267-9333
Michał Ronikier https://orcid.org/0000-0001-7652-6787
Magdalena Szczepaniak https://orcid.org/0000-0002-7483-3932
All of the data that support the findings of this study are available in the main text or Supplementary Information.
AFLP dataset of Saxifraga wahlenbergii
Data type: xls
The datasets of nrDNA and cpDNA of Saxifraga wahlenbergii
Data type: docx
AMOVA analysis based on AFLP data for populations of Saxifraga wahlenbergii calculated with a priori delimitation of regional groups
Data type: docx
Explanation note: Significance tests based on 1023 permutations; *P < 0.001. Regional grouping of populations see in Materials and methods.
Pairwise genetic divergence (FST) across 10 populations of Saxifraga wahlenbergii based on AFLP data
Data type: docx
Explanation note: Significance tests based on 1023 permutations; P < 0.001. For population acronyms see Table
Spatial arrangement of varying divergences (FST) among populations within distribution range of Saxifraga wahlenbergii
Data type: pdf
Explanation note: Above the line, the average FST values are given; the line colors correspond to scale in the right corner of the map. For population acronyms see Table