Next Article in Journal
The Common Ice Plant (Mesembryanthemum crystallinum L.)–Phytoremediation Potential for Cadmium and Chromate-Contaminated Soils
Previous Article in Journal
Comparison of the Ability to Control Water Loss in the Detached Leaves of Wedelia trilobata, Wedelia chinensis, and Their Hybrid
Article

Molecular Study of Selected Taxonomically Critical Taxa of the Genus Iris L. from the Broader Alpine-Dinaric Area

1
Institute of Agriculture and Tourism, Karla Huguesa 8, HR-52440 Poreč, Croatia
2
Department of Life Sciences, Imperial College London, South Kensington Campus, London SW7 2AZ, UK
3
Biotechnical Faculty, University of Ljubljana, Jamnikarjeva 101, SI-1000 Ljubljana, Slovenia
4
Department of Biology, Faculty of Science, University of Zagreb, Rooseveltov trg 6, HR-10000 Zagreb, Croatia
5
Public Health Institute of the Istrian Region, Nazorova 23, HR-52100 Pula, Croatia
6
Public Institution Natura Histrica, Riva 8, HR-52100 Pula, Croatia
7
Medical and Chemical School, Ante Šupuk Street 29, HR-22000 Šibenik, Croatia
8
Nature History Museum Split, Poljana kneza Trpimira 3, HR-21000 Split, Croatia
*
Author to whom correspondence should be addressed.
Joint last authors.
Plants 2020, 9(9), 1229; https://doi.org/10.3390/plants9091229
Received: 1 August 2020 / Revised: 10 September 2020 / Accepted: 16 September 2020 / Published: 18 September 2020
(This article belongs to the Section Plant Systematics, Taxonomy, Nomenclature and Classification)

Abstract

Some wild, morphologically diverse taxa of the genus Iris in the broad Alpine-Dinaric area have never been explored molecularly, and/or have ambiguous systematic status. The main aims of our research were to perform a molecular study of critical Iris taxa from that area (especially a narrow endemic accepted species I. adriatica, for which we also analysed genome size) and to explore the contribution of eight microsatellites and highly variable chloroplast DNA (ndhJ, rpoC1) markers to the understanding of the Iris taxa taxonomy and phylogeny. Both the microsatellite-based UPGMA and plastid markers-based maximum likelihood analysis discriminated three main clusters in the set of 32 analysed samples, which correspond well to the lower taxonomic categories of the genus, and support separate status of ambiguous regional taxa (e.g., I. sibirica subsp. erirrhiza, I. x croatica and I. x rotschildii). The first molecular data on I. adriatica revealed its genome size (2C = 12.639 ± 0.202 pg) and indicated the existence of ecotypes. For future molecular characterisation of the genus we recommend the utilisation of microsatellite markers supplemented with a combination of plastid markers.
Keywords: Iridaceae; Europe; chloroplast DNA; microsatellites; phylogeny; taxonomy Iridaceae; Europe; chloroplast DNA; microsatellites; phylogeny; taxonomy

1. Introduction

Iris L. (family Iridaceae) is a diverse genus with over 300 taxa distributed worldwide, mostly in the northern hemisphere [1,2]. In addition to conservational importance, many wild and cultivated taxa provide great horticultural value [3]. Phylogenetic and evolutionary studies of relationships of wild Iris taxa have long been challenging for several reasons. Namely, wide distribution, morpho-ecological diversity, multiple hybridisations, and convergent evolution processes, make definitive statements of the origin and evolution of taxa in the genus Iris very difficult [4,5]. To resolve a myriad of uncertainties and issues related to taxonomic and phylogenetic relationships within the genus Iris, extensive work was performed on morpho-anatomical features, palynology, phytochemical constituents’ analysis, cytogenetic traits, and molecular analysis [6,7,8,9]. Despite different approaches to lower (and individual) taxonomic categories, most authors agree on the classification of the genus Iris into six subgenera, which are divided into sections and series [1,10,11].
Most of the European native taxa of the genus Iris belong to the subgenus Iris L., section Iris L. (so-called “Pogoniris”), represented by numerous rhizomatous Iris taxa characterised by bearded outer tepals. Less prevalent on the European territory are taxa from the subgenus Limniris (Tausch) Spach, section Limniris (Tausch) Spach (so-called “Apogoniris”), which are rhizomatous irises whose outer tepals are without a beard [1,3]. The broad Alpine-Dinaric, as well as the surrounding Mediterranean and Pannonian area of Europe (where irises for our study were sampled) is characterised by peculiar eco-climate conditions which have caused a great morphological variability of some Iris populations and groups. Their variety has resulted in ambiguous systematic status of some regional, especially endemic, Iris taxa [5,8]. Some of them are recognised in the national and regional floras [12,13] and still have an unclear phylogenetic and classification status. Some of them neither are accepted in the World Checklist of Selected Plant Families [2], nor are molecularly researched in detail. Therefore we intended to molecularly study some, insufficiently researched and/or globally neglected taxa; namely: I. x croatica Horvat et M. D. Horvat (endemic in Croatia and Slovenia), I. illyrica Tomm. ex Vis.(endemic in Croatia, Slovenia, and Italy), I. sibirica L. subsp. erirrhiza (Posp.) Wraber (endemic in Bosnia and Herzegovina, Croatia, and Slovenia) and I. x rotschildii Degen (endemic in Croatia). However, in this study we paid special attention to the validly described [14] and accepted [2], molecularly unexplored endemic species I. adriatica Trinajstić ex Mitić (Figure 1 and Figure 2).
I. adriatica (Figure 1a–c) is a narrow endemic plant from the I. pumila complex, characterised by an extremely dwarf stem (one of the smallest species within the genus Iris) and relatively large yellow, violet, or purple solitary flowers (Figure 1a–c) [14].
It is confined to a few Croatian localities in the wider area of Dalmatia and classified as a NT (near threatened) species [13]. Given that the localities newly recorded by authors are spatially distant from the previously catalogued specimens (Figure 2), questions of subspeciation or higher-level genetic divergence can arise. All the more so as the recent metabolic profiling [15] revealed a notable diversity between the ecotypes and their pharmacological and chemotaxonomic potential.
Since the 1990s, when molecular biology techniques have become widely accessible, taxonomical biology has been driven towards using molecular methods to establish and re-establish evolutionary relationships between species [16,17]. Tang et al. [18] developed 400 ortholog-specific EST-SSR (Expressed Sequence Tag—Simple Sequence Repeats) markers, which can be reliably used to distinguish between the species in the Iris genus, providing a cheap and efficient way to resolve taxonomical discrepancies. Simple Sequence Repeats or microsatellites are present in most species; they are usually locus-specific, multiallelic, polymorphic, and co-dominant and are as such ideal candidates for discriminating between Iris species [19].
AChloroplast gene sequences are often used for plant phylogenetic studies and DNA barcoding because of the relatively low evolutionary mutation rates, their uniparental inheritance, high level of genetic diversity, and absence of recombination. Many candidate plastid regions have been suggested as the plant barcode and have as such been extensively tested [20,21,22]. However, to this end, a single marker has not yet been found which could reliably distinguish between a majority of plant species. Different combinatorial approaches have been used in different instances, to set on a final consortium [23]. Plastid DNA regions rpoC1 and ndhJ used previously to evaluate plant phylogeny with low taxonomic variation [22] seemed appropriate for our study.
One of the basic genomic parameters that characterise the species and represent one of the important plant traits is the total amount of DNA in the unreplicated haploid or gametic cell nuclei, referred to as the C value or genome size [24]. Genome size data have numerous applications: They can be used in comparative studies on genome evolution, or as a tool to estimate the cost of whole-genome sequencing programs [25]. Currently, the largest updated plant genome size database—Plant DNA C-values database contains data for 12,273 species and among them 65 C-values for 44 species of genus Iris [26]. For most species involved in our study C-values are measured in several studies [27,28,29]. Different methods were used for the measurement of plant DNA content, but flow cytometry has become the method of choice due to its reliability, simplicity, and relatively low cost [30,31].
A noticeable lack of efforts to molecularly resolve remaining issues in Iris phylogeny and taxonomy on the Alpine-Dinaric area (including the adjacent areas of Mediterranean and the Pannonian Plain) in the context of conservation was extremely important when designing the study. Hence, to provide molecular insights into phylogenetic relationships of selected wild Iris taxa of the wider Alpine-Dinaric area, with a special emphasis on regional endemics and molecular evidence for their conservation, the aims of our research were: (i) To characterise representative and critical Iris taxa from the wider Alpine-Dinaric area by nuclear (SSR) markers; (ii) to clarify the genetic divergence within and between several wild (local endemic) and cultivated Iris populations through chloroplast DNA (cpDNA) markers; (iii) to present the first molecular description of a nearly threatened narrow endemic dwarf species I. adriatica; and (iv) contribute to the efforts of establishing optimal molecular markers for detecting taxonomic and phylogenetic relationships within critical taxa of the genus Iris.

2. Results

2.1. SSR Analysis

In total, 32 Iris samples across the Alpine-Dinaric region were analysed (Supplementary Table S1). Parameters of genetic diversity evaluation are presented in Table 1. SSR marker analysis was able to identify a total of 71 alleles (Supplementary Table S2). The observed number of alleles per locus ranged from 6 (at locus IM123) to 12 (at loci IM196 and IM327) with an average of 8.8 alleles and 4.3 effective alleles per locus. At locus IM348, out of eight alleles, allele 125 showed a frequency of 0.71; thus locus polymorphism information content (PIC) was 0.466, while at locus IM164, allele 324 showed a frequency of 0.68 resulting in locus polymorphism of PIC = 0.480. In general, the number of effective alleles was relatively low, indicating that rare and frequent alleles are present in the examined group of samples. The highest numbers of effective alleles (5.5 and 6.2) were observed at loci IM196 and IM327, respectively, where the frequencies of alleles were equally distributed. PIC values ranged from 0.466 (at locus IM348) to 0.845 (at locus IM391), indicating sufficient polymorphism information content of all loci. Loci IM164 and IM348 were moderately informative (0.25 < PIC < 0.5), while the rest were highly informative (PIC > 0.5). The expected heterozygosity varied between 0.490 (IM348) and 0.877 (IM391), with an average of 0.728. The highest observed heterozygosity (0.871) was found at locus IM123, and the lowest (0.129) was characteristic of locus IM164. The observed heterozygosity was lower than expected on all loci except IM123. The probability of identity (PI) values were in a range from 0.072 to 0.357, and the total PI calculated for all loci was 2.01 × 10−7, indicating a low probability of identical genotypes.
The UPGMA clustering analysis (Figure 3) discriminated 28 genotypes and revealed three distinct groups of samples. The first cluster contained samples of mostly tall bearded Alpine-Dinaric taxa: I. x croatica, I. x germanica L., I. illyrica, I. pallida Lam., I. pumila L., I. reichenbachii Heuff., and I. x rotschildii, grouped in two subclusters. The second cluster (with several smaller subclusters) consists of all samples of narrow endemic dwarf species I. adriatica, its closely related species I. attica Boiss. & Heldr. as well as I. barbata cultivar, a horticulturally-widespread variety in the region. All samples of I. sibirica L. sensu lato grouped in the third cluster, together with I. pseudacorus L. and I. graminea L. within a separate subcluster.

2.2. Chloroplast Barcodes Analysis

The maximum likelihood (ML) analysis was used in reconstructing phylogenetic relationships of a heterogeneous group of Iris species based on two plastid markers (rpoC1, ndhJ). ML analysis discriminated three major clusters of which seven groups of taxa and 10 different genotypes (Figure 4). In the ML dendrogram, three main groups of irises were discriminated, with I. reichenbachii separated from the rest. The first group consisted of five undiscriminated mostly Alpine-Dinaric species. Dwarf bearded irises I. adriatica, I. pumila, and I. attica were not separated and grouped with I. barbata cult. in the second cluster. Both subspecies of I. sibirica grouped in a third cluster, together with the out grouped I. graminea and I. pseudacorus. The samples accessed from NCBI gene repository I. missouriensis Nutt., I. sanguinea Hornem., and I. gatesii Foster were grouped appropriately, according to their classification within the genus Iris.

2.3. Genome Size

4′,6-diamidino-2-phenylindole (DAPI) fluorochrome is known to bind to DNA, specifically, to AT base pairs and therefore lower values for absolute genome size analysis are found [32]. The determination of total DNA content of plants from DAPI stained cells was: For I. adriatica 12.639 ± 0.202 pg (2C) and for I. x germanica 24.249 ± 0.163 pg (2C) respectively as compared to the Pisum sativum cv. Kleine Rheinländerin (9.07 pg/nucleus) internal standard.

3. Discussion

In our study, we applied 8 SSR markers developed by Tang et al. [18] which proved to be highly polymorphic and amplified alleles across the 39 Iris ecotypes and cultivars. We were not able to utilise the IM61 marker recommended but the remaining markers provided sufficient resolution to distinguish between our samples. We observed the greatest allelic diversity on IM196 and IM327 in concurrence with the aforementioned study; however, the observed number of alleles per locus in our study was significantly lower (average 8.8) suggesting greater phylogenetic similarity across all of our samples. Although it is comparable with the average number of alleles per locus observed within the group of 13 yellow-flag, Siberian, and tall-bearded Iris cultivars analysed by [18]. In our case, a small population size could be the reason for low allele frequency. Genetic similarity ranged from 0.23 to 0.8 and 0.26 to 1.00 among Alpine-Dinaric taxa from the subgenus Iris (section Iris) grouped in the UPGMA clusters I and II, respectively. The highest genetic similarity was intraspecific (Dice = 1; I19 and I21; I13 and I19; I30 and I31), whilst the lowest were interspecific (Dice = 0.23; I22 and I41 in cluster I; Dice = 0.26; I16 and I10; I16 and I11 in cluster II). Genetic similarity between endemic dwarf ecotypes of I. adriatica grouped within a separate subcluster, and correlated with the locations of origin, ranged from 0.55 to 1.00, implying significant and disperse genetic diversity among ecotypes. Taxa from the subgenus Limniris (section Limniris) displayed genetic similarity in a range from 0.07 to 0.72, the highest between samples of I. sibirica subsp. erirrhiza. Only a few SSR markers were needed to identify (distinguish) ecotypes and species.
The unique microsatellite profiles were established as described in the method section below, nevertheless, we acknowledge that the SSR analysis can differ from lab to lab as the method inherently produces high numbers of edge cases where a judgment call has to be made. An example of an edge case is the apparent presence of 3 alleles in what we presumed (and confirmed for I. adriatica) to be 2n = 2x species. As described, this was resolved by establishing a common SSR profile for those particular samples, since our subsequent analysis methods rely on the binary presence or absence of a particular allele and a presence of 3 alleles would likely confound the result and be factually incorrect. To resolve such an edge case a full sequencing run could reveal genomic mutations, such as translocation, or perhaps other properties of the genome at that position which would allow the probes to bind in this particular way. Further, as I. x germanica is a suspect tetraploid [5,18,33], the additional genetic information could skew the subsequent phylogenetic analysis as additional peaks appeared in positions only in one individual and could thus not be compared to any other values in the study, carrying an extremely low PIC. For our analysis such peaks were considered to be outliers; however, we are not suggesting they are not valid data in different subsamples.
This means that for any analysis the attribution of a particular profile needs to be internally consistent and cannot be used at face value form any further studies which want to include the same dataset. In our case, we employed the algorithm described in the methods to come to a conclusion which was cross-examined within the research group to preserve the established logic of sorting different cases. The final analysis of genetic relationship relies on the presence and absence of specific alleles so for our purposes the aim was to obtain the same profiles for the same species when attributing an SSR profile, without knowing which species the profile belongs to. Since a matching algorithm can only be established ad-hoc after accessing the reads, there is a potential to introduce some bias into edge-case decision making. Nevertheless, we are confident in our results several reasons; sample duplicates were included as an internal control and independently produced the same profiles using the same “blind” determination method, the chloroplast marker analysis largely produced the same clustering, profile differences between presumed same species are minimal, our described SSR relationship mirrors the relationships which were confirmed or predicted using taxonomic, botanical or other methods.
Different combination of chloroplast genome sequences were proposed for species discrimination, such as rpoC1, rpoB, and matK; rpoC1, matK, and psbA-trnH; [34] and rbcL and trnH-psbA [35]. In a recent review [23], authors Saddhe and Kumar discussed the utility of plastid markers to differentiate between different species within plant divisions, where they establish ndhJ as a good candidate marker for barcoding angiosperms. Additionally, rpoC1 is often used as a supplementary marker to increase the barcoding depth of samples [36]. Plant Working Group (PWG) of the Consortium for the Barcoding of Life (CBOL) recommended the combination of rbcL and matK as the plant barcode [20], while rpoB and here applied rpoC1 showed markedly lower discriminatory power. Chloroplast marker matK is recommended as one of the best DNA barcoding candidates for species discrimination [20,37]. However, this chloroplast region proved to be difficult to amplify and sequence in certain taxa, and additional universal primers and optimisation of PCR reactions were necessary [38,39]. In our study, the preliminary amplification of matK sequences was unsuccessful and the testing of additional plastid markers is foreseen. However, the combination of ndhJ and rpoC1 revealed to be adequate for discrimination up to the series taxonomic level, indicating the possibility of applying additional candidates for the species discrimination. As discussed, a plastid marker with sufficient resolution would be operationally favourable for widespread utility in discriminating between different species. Up to date a few phylogenetic studies based on chloroplast markers were carried out on Iris [6,40,41,42]. Neither ndhJ nor rpoC1 was not tested in any Iris genus study.
Groupings of the Iris taxa from the broader Alpine-Dinaric area, observed in our research by both sets of molecular markers (Figure 3 and Figure 4), mostly correspond to proposed phylogenetic relationships based on palynological features [8]; a clear distinction between the subgenera Limniris and Iris and within the majority of the lower taxonomic Iris categories of sections and series emerges. The anticipated exception is the position of analysed NCBI sequence of Middle Eastern species I. gatesii (Figure 4), which separated within the subgenus Iris in an individual cluster, as it belongs to the different series Oncocyclus (Siemssen) Baker [1]. However, the unexpected exceptions are positions of the species I. pumila based on SSR markers (Figure 3), and of I. reichenbachii based on ML analysis (Figure 4). Molecular analysis of both sets of markers (Figure 3 and Figure 4) in principle resulted in the creation of three main clusters: Two of three clusters covering rhizomatous taxa from the subgenus Iris, section Iris, with a beard (“Pogoniris”, [3]), while the taxa from the subgenus Limniris, section Limniris, rhizomatous irises with falls without a beard (“Apogoniris”, [3]) were grouped in the third cluster (Figure 3 and Figure 4). For the ML analysis control NCBI sequences: Of I. sanguinea (subgenus Limniris; sect. Limniris, series Sibiricae (Diels) Lawrence) and I. missouriensis (subgenus Limniris; sect. Limniris, series Longipetalae (Diels) Lawrence), grouped with other members of the same subgenus (Figure 4); and of I. gatesii (subgenus Iris; section Oncocyclus) made a separate branch between samples of “Apogoniris” and the rest of the “Pogoniris” (Figure 4). Such results are in agreement with previous studies and monographs of the genus Iris [1,3,11,41,43].
Within the subgenus Iris, section Iris, on the series level, one cluster (based on both sets of molecular markers; Figure 3 and Figure 4) comprises the group of mostly tall bearded irises and covers the series Elatae Lawr. [10]. The second cluster covers the group of dwarf bearded irises and matches the series Pumilae Lawr. [10], except for I. pumila grouping in the first cluster based on SSR markers analysis (Figure 3). However, plastid markers (Figure 4) did not discriminate analysed taxa within neither series Elatae (the only exception is I. reichenbachii) nor Pumilae. In our study chloroplast markers ndhJ and rpoC1 provide a weaker resolution into the species, concurrent with other authors [22]; however, we acknowledge that the analysis of sequence data is quicker and much less prone to human error and enables clustering comparison across different studies if the sequences are made publicly available. Further, our study looked at only two plastid regions, as compared to eight microsatellite loci. Therefore, we would recommend the utilisation of SSR markers for subsequent analysis supplemented by a plastid marker combination for the genus Iris, until a single plastid marker combination is established as a convention.
According to SSR markers analysis (Figure 3), within the cluster I, two subgroups were formed: In the first are two samples of tall bearded I. x croatica, I. x germanica, and, unexpectedly, dwarf bearded I. pumila, whereas one sample of I. x croatica is grouped with other analysed tall bearded irises within the second subgroup. Although its taxonomic position is critical and still unresolved, the taxon I. x croatica is considered as a native endemic taxon in northern Croatia and Slovenia [12,13,44]. Likely due to morphological similarities, it is often mixed with and named as a synonym for I. x germanica [1,2,5,13], which is, in our opinion, distributed worldwide only as a cultivated hybrid species [1,9]. The fact that the WCSP [2] wrongly “declares” I. croatica Horvat & M.D. Horvat as an illegitimate name, due to an incorrect replacement with I. croatica Prodan, provokes further taxonomic confusion [45], explained in detail in [5]. The close relationship between I. x croatica and I. x germanica was noticed by examining both plant and pollen morphology [8] (B. Mitić, personal observations) and is confirmed with our results—their joint sub clustering (Figure 3). However, they are both tetraploids of yet unresolved origin with reported chromosome numbers of 2n = 44 for I. x germanica, and 2n = 48 for I. x croatica [5,46]. Two earlier speculations about (auto) tetraploid origin of I. x croatica both agreed that the progenitor species for that hybrid is I. pallida, although this is yet to be cytogenetically confirmed [5,8]. Grouping a sample of I. x croatica together with I. pallida and I. illyrica within the second subgroup in our results (Figure 3) confirms the proximity of tetraploid I. x croatica and presumed progenitor species I. pallida.
Considering the clear discrimination within lower taxonomic subgroups such as series, obtained by the applied marker systems (Figure 3), the status of other closely related taxa from the so-called I. pallida complex could be discussed. Taxonomic relationships within the complex have not been fully explored and it is not yet clear whether the taxa of this complex have the status of species or subspecies. Namely, the majority of taxa from this complex (including representatives from our research—I. pallida and I. illyrica) were defined at the species level and extracted, apart from the series Elatae into the new series Pallidae (A. Kern.) Trinajstić [47]. Although earlier taxonomic researches of I. pallida complex [48,49] have supported such taxonomic treatment of its taxa, a later palynological study [8] indicated their return into the status of subspecies level (as classified by WCSP [2]), and of the series Pallidae back into the series Elatae. Results of our study are in accordance with the last hypothesis as both marker systems (Figure 3 and Figure 4) grouped members of those series closely together.
The taxon I. x rotschildii from the series Elatae also garners considerable attention in the context of this study. So far, this narrow endemic iris is known from a single locality on Mt. Velebit (Croatia). It is described as a natural hybrid between species I. illyrica and I. variegata L. [1,50] with observed morphological, palynological, and cytogenetic variabilities [46]. Some of the mentioned features confirm the hybridogenous origin of this taxon. Despite this, no further molecular studies have been done on the taxon to confirm its claimed status. This is the likely reason it was recently considered as a synonym of I. x germanica by WCSP [2]. Unfortunately, due to hard-to-reach mountainous terrain (with mines still present in the area) and the small number of specimens in the only known population on Mt. Velebit (B. Mitić, personal observations), only one sample of this taxon was included in our analysis. Bearing this in mind, the SSR profile of I. x rotschildii that shares at least one allele on all analysed loci with I. illyrica as well as their position in the same UPGMA subcluster additionally support their parent-sibling relationship (Figure 3). Moreover, although I. x germanica and I. x rotschildii are presumed synonyms [2], their discrimination by SSR could disprove that assumption and would favour the placement of I. x rotschildii within a separate taxonomic position. However, further extensive detailed molecular study of I. x rotschildii and its presumed parents is needed to confirm both its separate taxonomic status and its difference with I. x germanica.
Furthermore, unexpected discrepancies occur in the placement of I. reichenbachii, which was positioned in the same UPGMA subcluster as I. illyrica, I. pallida, and I. x rotschildii (Figure 3) and also as an outgroup in ML dendrogram (Figure 4). Namely, I. reichenbachii is native in mountainous regions of the Balkan Peninsula and SW Romania, known as parental species of some natural hybrids [5], and according to [10] was firstly placed in the series Pumilae. However, according to both chromosome numbers 2n = 24, 48 [43] and pollen analyses [8] it seemed to fit better in the series Elatae. Nevertheless, outgrouping of I. reichenbachii in our ML analysis (Figure 4) might indicate its specific position between two series that still needs to be explored, as it has the same number of chromosomes [5] and pollen grains [8] as tall bearded irises and is morphologically quite dwarfish [43]. Further, its genome size (1C value) is intermediate between some members of both series Elatae and Pumilae [33].
Cluster II in our study (Figure 3 and Figure 4) covers mostly dwarf bearded irises. However, except for I. pumila based on SSR markers, grouping within the first cluster (Figure 3), together with tall bearded I. x croatica and I. x germanica. Given current evidence, we speculate that the grouping may have happened due to the normalisation of the chromosomal content applied, and treatment of SSR data as codominant, with maximally two alleles counted. An additional element could be genetic variability of I. pumila, evident from genome size of this tetraploid species (2n = 32), differing in several previous studies (e.g., 1C = 13.20 pg [27]; 1C = 6.81 pg [33]; 1C = 10.64 pg [51]). Furthermore, this taxon is supposed to have the same hypothetical ancestor as tall bearded irises (with x = 4 [3,43]), and is often known as the parental species (together with some tall bearded irises as second parents) of many native and artificial hybrids [43].
On the contrary, all other investigated samples of dwarf bearded irises of the series Pumilae [10] grouped in a separate cluster II based on plastid markers (Figure 4). Such results are in accordance with pollen morphology of dwarf bearded irises [8,52] and confirm their separate taxonomic position, and belonging to the same I. pumila complex [14]. Since I. attica is the only member of the complex with the status of a subspecies, and with others having equal rank of species in the WCSP [2], our results (Figure 3 and Figure 4) suggest that they should be treated at the same taxonomic rank. Therefore, further research is needed to corroborate (or disprove) our statement about taxonomic relationships within the whole I. pumila complex.
Meanwhile, special attention in our study was dedicated to one member of the complex—a relatively-recently described diploid (2n = 16) species I. adriatica [14], native and endemic to Croatia. Namely, to prepare the basis for its conservation, because of its nearly threatened species status [13], we were particularly focused on its molecular features. Evidence about taxonomic and phylogenetic values of palynological and phytochemical features of I. adriatica are well documented [8,15]. However, thus far, this species has not been researched on a molecular level. In the present results (Figure 3) we documented diversity of different populations of the species I. adriatica, showing the existence of geographical ecotypes. In particular, the UPGMA grouping (Figure 3) of established ecotypes corresponds well with the geographical origins of the samples (Figure 2, Supplementary Table S1): The island populations (sample numbers I26 island of Brač; I10, I11, and I12 island of Cres) have separated from the land coastal populations (Figure 3, Supplementary Table S1, other samples). Therefore, we assume that island populations might be a specific ecotype of the typical species. Within inland populations, we were particularly interested in the population of the hinterland population “Brnjica-Pokrovnik”, which has been singled out as an ecotype based on phytochemical analysis [15]. In our analysis (Figure 3, sample no. I18) it has a separate branch in the dendrogram, although it is “surrounded” by other inland populations. Therefore, it is obvious that potential inland ecotype(s) require additional investigations. One more reason in favour of the separation of ecotypes is the fact that “Brnjica-Pokrovnik” population is growing on an open calcareous meadows (mainly belonging to the Festuco-Koelerietum splendentis Horvatić 1963 association), whilst the rest of the researched populations grow on limited rocky pastures and hills (mainly belonging to the Stipo-Salvietum officinalis Horvatić 1985 association), very often endangered by the succession, i.e., overgrowth with macchia.
Furthermore, in this study we present the first genome size estimation of I. adriatica measured by flow cytometry and expressed according to [24] as 2C value = 12.639 ± 0.202 pg. Observed value of genome size for I. adriatica we could hardly compare with values of all other members of the complex I. pumila, since the data are known only for tetraploid species I. pumila [27,33,51]. However, as previously mentioned, data for this tetraploid species indicates its variability. Our results of genome size value for diploid species I. adriatica are the first data about genome size for this strictly endemic, near threatened species and should contribute to its future conservation. The 1C value of I. adriatica is similar to that of tetraploid I. pumila obtained by [33]. Such results should confirm belonging of both species to the same complex. Additionally, similar deviations in 1C values as in the species I. pumila, were observed for the species I. x germanica: Our results of 2C = 24.249 pg for this “control” species could be compared to the result (1C = 12.45) of [27], while the value of 1C = 5.87 for the same species was observed by [33]. Therefore, the genome sizes of critical taxa of the genus Iris require further, more complex research.
In our results within the third cluster (Figure 3 and Figure 4) all samples of so-called “Apogoniris” taxa [3] grouped together, further all are representatives of the subgenus Limniris, section Limniris. Such results are in accordance with some previous research of molecular phylogeny of these taxa [40,53,54]. Additionally, our analysis based on both sets of markers (Figure 3 and Figure 4) resulted with different subclusters within the subgenus Limniris. Namely, mentioned subgroups correspond well to the series as a lower taxonomic level (according to [1,10]): Laevigatae (Diels) Lawrence (I. pseudacorus), Sibiricae (Diels) Lawrence (both subspecies of I. sibirica), and Spuriae (Diels) Lawrence (I. graminea). The analysed NCBI sequences of “Apogoniris” taxa (I. missouriensis and I. sanguinea) additionally support that distinction (Figure 4), they grouped with other members of the subgenus Limniris, section Limniris. Moreover, I. sanguinea, which belongs to the series Sibiricae [1], grouped close to other members of this series.
Furthermore, all samples of I. sibirica sensu lato (series Sibiricae) grouped apart from other members of the subgenus Limniris (Figure 3 and Figure 4), and created further subclusters (Figure 3). This was especially interesting because of the still unclear position of the Alpine-Dinaric mountain populations described as subspecies of the typical I. sibirica species [55]. Although plastid markers (Figure 4) did not discriminate I. sibirica subspecies, the results of SSR analysis (Figure 3) confirmed their differentiation. This is also in accordance with the presumption that I. sibirica subsp. erirrhiza might be a mountain ecotype [46], which differs from the typical lowland subspecies I. sibirica subsp. sibirica [55]. This is particularly interesting for further conservation of wild, especially endemic irises from that area. Namely, I. sibirica subsp. erirrhiza was found only in several localities in Bosnia and Herzegovina, Croatia, and Slovenia where it might be an endemic taxon [46,55]. The subclustering of I. sibirica subsp. erirrhiza samples in our research and an extra subcluster of typical I. sibirica subsp. sibirica (Figure 3) additionally confirms this distinction of subspecies as ecotypes. Unfortunately, in our study we did not have a sample of the population of I. sibirica subsp. erirrhiza from Mt. Bjelolasica (Croatia), the supposed link between the subgenera Limniris and Iris in the territory of Southern Europe [8]. Further research focused on broader ecotype samples of I. sibirica sensu lato is needed to give a better insight into the phylogenetic structure within this complex taxon.
Regarding other representatives of the subgenus Limniris in our study, we can comment on the specific position of the species I. graminea, which separated in the distinct cluster in both trees (Figure 3 and Figure 4). Therefore, our results might support the hypothesis that the species I. graminea is probably the most primitive member of the subgenus Limniris on the Southern European territory [8]. Besides this, our analysis of microsatellites (Figure 3) might also confirm the opinion based on palynological observations, that the subgenus Iris is more advanced than the subgenus Limniris [8,56].
In closure, we can confirm that our results of the molecular study of Alpine-Dinaric taxa of the genus Iris correspond well with their positions within the subgenera Iris and Limniris, and are in accordance with some other recent molecular researches of taxa of the genus Iris [41,57]. Additionally, our results present the first molecular data on narrow endemic and near threatened species I. adriatica and also support the separate taxonomic status of investigated ambiguous regional taxa (e.g., I. sibirica subsp. erirrhiza, I. x croatica and I. x rotschildii).

4. Materials and Methods

4.1. Plant Material and DNA Extraction

Plants of the genus Iris distributed across the broader Alpine-Dinaric region were collected either in their natural habitats during the vegetation seasons 2016–2018, retrieved from botanical collections of the National Botanical gardens in Zagreb (Croatia) and Ljubljana (Slovenia) (Supplementary Table S2). Most vouchers are live specimens deposited within the Iris collections of the mentioned Botanical Gardens in Zagreb and Ljubljana, and one in the private garden of the corresponding author. Herbarium voucher specimens are deposited in the herbarium of the Istrian Botanical Society, Vodnjan, Croatia (not yet registered in the Index Herbariorum). Total genomic DNA was isolated from 25–100 mg dried or fresh leaves, depending on the sample, using the commercial kit PureLink® Plant Total DNA Purification Kit (InvitrogenTM; Waltham, Massachusetts, USA), in accordance with the manufacturer’s instructions. One sample (I24; I. sibirica subsp. sibirica; Supplementary Table S2) was excluded from SSR analysis due to poor imaging signals.

4.2. Microsatellite and Chloroplast Barcodes Amplification

Eight SSR markers [18] were used for genotyping (Supplementary Table S2), following the optimised procedures described in [18]. Forward SSR primers were end-labelled with one of three fluorophores, 6FAM, HEX, or TAMRA (Supplementary Table S3). Briefly, the initial denaturation step was performed at 95 °C for 3 min, followed by 1 cycle of 94 °C for 30 s, 55–64 °C (depending on optimal annealing temperature (Ta)) for 30 s and 72 °C for 45 s. The annealing temperature was decreased 1 °C per cycle in subsequent 7 cycles until reaching the optimal Ta (Supplementary Table S3) at which 35 cycles were carried out, with a final extension at 72 °C for 20 min. The PCR products were checked on 2% agarose gels to confirm amplification. The length of the PCR products was determined through capillary gel-electrophoresis (Macrogen Europe B.V., Amsterdam, the Netherlands). SSR alleles were resolved on the ABI3730XL DNA Analyser (Applied BiosystemsTM; Waltham, Massachusetts, USA), using GeneMarker® Software V2.7.0 (SoftGenetics, State College, Pennsylvania, USA) and 400HD ROXTM dye-labelled internal size standard marker. SSR peak estimates were determined using inbuilt software on pre-set settings. Each peak was individually evaluated. False positives were eliminated by looking at peak values appearing at the same position in reads where no SSR probe was present for a particular analyte, judged to be innate background. Due to slight shifts occurring at each read, peaks from different runs, which were consistently different in length were judged to be the same SSR profile [58]. All samples were described using a maximal value of two alleles at each SSR locus examined normalised to a 2n = 2 × chromosomal content (Supplementary Table S2). Where more than two alleles (peaks) were apparent their pattern was cross-examined with other available samples to determine their unique descriptive allelic values.
A combinatorial approach of ndhJ and rpoC1 plastid markers (Supplementary Table S3) was used for barcoding according to the procedure of [59]. The procedure consisted of an initial denaturation step at 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 55 °C for 30 s, 72 °C for 60 s and a final extension step at 72 °C for 10 min. Before sequencing PCR products were additionally purified using exonuclease I and shrimp alkaline phosphatase to remove unincorporated nucleotides and primers. The barcodes were Sanger Sequenced using ABI 3130XL capillary machine (Biotechnology Faculty, University of Ljubljana, Ljubljana, Slovenia) and submitted to GenBank (Supplementary Table S2). Further three additional sequences (I. gatesii, GenBank: KM014691.1; I. missouriensis, NCBI Reference Sequence: NC_042827.1; I. sanguinea, NCBI Reference Sequence: NC_029227.1) were mined from the NCBI repository. Sequences were aligned using Codon Code Aligner V9.0.1 (CodonCode Corporation, Centerville, MA, USA).

4.3. Data Analysis

Genetic parameters were calculated for 32 Iris samples over eight microsatellite loci (Table 1). The number of amplified microsatellite alleles (n), number of effective alleles (ne), observed heterozygosity (Ho), and expected heterozygosity (He) were calculated using POPGENE, version 1.32 [60]. Polymorphic Information Content (PIC) was calculated with the program Cervus, Version 3.0.7 [61] and probability of identity (PI) was determined using IDENTITY v.1.0 program [62]. Genetic distances between all pairwise combinations of the samples were calculated using Dice’s coefficient of similarity. The dendrogram was constructed from the resultant matrices via the UPGMA distance-matrix method using the PAST software [63]. Statistical support for the tree topology was assessed by 1000 bootstrap replicates. The two chloroplast loci (rpoC1 and ndhJ) sequence data were aligned using the “Create Alignment” algorithm implemented in CLC Genomics Workbench 20.0.2. Alignments were joined together and a Maximum Likelihood Neighbour-Joining tree was constructed using the “Maximum Likelihood Phylogeny” algorithm of CLC using the Jukes–Cantor nucleotide substitution model.

4.4. Genome Size Analysis

The DNA content of I. adriatica and I. x germanica plants were analysed by flow cytometry analysis according to the method reported in [32]. A portion of the fresh young leaves tissue of approximately 1 cm2 was used in sample preparation. For an internal standard, the Pisum sativum cv. Kleine Rheinländerin (9.07 pg/nucleus) was used for reference. Both the sample and the standard were chopped finely using a razor and released into 0.1 M citric acid containing 0.5% Tween 20. The homogeneous mixture was filtered through a 30-μm nylon filter removing larger particles. 4′,6-diamidino-2-phenylindole (DAPI) was used as the genome staining dye. A 3–4-fold volume of staining buffer containing 4 μg ml−1 of DAPI in 0.4 M Na2HPO4 × 12H2O was added to the specimens.
Samples were analysed with a Partec CyFlow® Space flow cytometer using linear scale. FloMax® software (Partec, Münster, Germany) was used for the calculation of relative nuclear DNA content.

5. Conclusions

In the present molecular study of selected representative and critical Iris taxa from the wider Alpine-Dinaric area, we enhanced the current knowledge and understanding of the genus Iris taxonomy and phylogeny of the area; important for their further protection and conservation in the study area. Our research showed taxonomic positions of investigated taxa within the genus Iris, which is mostly in accordance with previous comprehension of the genus Iris. We were especially focused on getting the first molecular data on the nearly threatened narrow endemic dwarf species I. adriatica, hitherto molecularly unexplored. The results of molecular analysis showed that the 2C value for this species is 12.639 ± 0.202 pg, pointing to its relationship with other dwarf irises from the I. pumila complex, and indicating the existence of ecotypes. Additionally, we stressed some, presently unresolved, key taxonomic questions about certain critical groups and/or taxa of the genus Iris from that area, and the most pertinent are: Taxonomic and phylogenetic relationships of some complex Iris groups from this area (e.g., I. x germanica, I. pallida, I. pumila and I. sibirica groups) and the taxonomic status of regionally recognised, but globally neglected endemic taxa: I. sibirica subsp. erirrhiza, and natural hybrids I. x croatica and I. x rotschildii. For mentioned groups and taxa our study establishes baseline taxonomic and phylogenetic relationships across the Alpine-Dinaric region, but more precise confirmation of their phylogenetic and taxonomic status require further, more complex molecular analysis on a broader set of Iris samples. Regarding the contribution to the efforts of establishing optimal molecular markers for detecting taxonomic and phylogenetic relationships within critical taxa of the genus Iris, we would recommend the utilisation of SSR markers for subsequent analysis supplemented with a combination of plastid markers until a plastid marker combination for the genus is established and fully validated as convention. Chloroplast markers ndhJ and rpoC1 provide a weaker resolution into the species; however, analysis of sequence data is quicker and much less prone to human error. Further, our SSR study looked at 8 microsatellite loci as compared to two plastid regions. Chloroplast markers can give further context to SSR analysis and provide independent control despite their lower resolution as they can confirm broader clusters. For future studies of the genus Iris we would additionally recommend the inclusion of other appropriate barcoding regions to serve the same purpose and hopefully increase the sequencing resolution.
Molecular evidences obtained in this study, besides contribution to the knowledge on taxonomy and phylogeny of the genus Iris in the Alpine-Dinaric, Mediterranean and Pannonian area, should also help in further understanding about the importance of wild, especially endemic Iris taxa and encourage their more intensive conservation efforts.

Supplementary Materials

The following are available online at https://www.mdpi.com/2223-7747/9/9/1229/s1, Table S1: Alpine-Dinaric taxa of the Genus Iris used in the present molecular study; Table S2: Genotypes of the analysed Iris samples at eight microsatellite loci (allele sizes in bp); Table S3: SSR and chloroplast markers used in the present molecular study of the Alpine-Dinaric taxa of the genus Iris.

Author Contributions

Conceptualization, D.P., B.S., B.M., and T.W.; methodology, D.P., J.J., B.B., T.W.; sampling, B.M., D.H.; S.B., N.L., M.M., and D.V.; formal analysis, T.W., D.P., J.J., B.B.; writing—original draft preparation, D.P., T.W., B.M., B.S., and D.H.; writing—review and editing, J.J., B.B., B.S., B.M., T.W., D.P., D.H., S.B., N.L., M.M., and D.V.; visualisation, T.W., D.P., J.J., and D.H.; All authors have read and agreed to the published version of the manuscript.

Funding

The parts of this research were funded by the Slovenian Research Agency (SRA–ARRS), grant number P4-0077, and the University of Zagreb, grant number 106-F20-00025. University of Westminster Distant Horizons Award fund provided mobility support.

Acknowledgments

The authors would like to thank Miroslav Mitić, Radnić family, and Nediljko Ževrnja for fieldwork assistance. We would also like to thank Jože Bavcon and Janja Makše, University Botanic Gardens Ljubljana for their help in collecting selected Iris samples, and Botanical Garden, Faculty of Science, University of Zagreb for providing us the living plant material of some taxa of the genus Iris from its collections. Thanks are also extended to anonymous reviewers, whose valuable comments contributed to the final appearance of the manuscript. We also thank the Institute of Agriculture and Tourism, University of Zagreb, and University of Ljubljana for in kind support. Field research was performed with the permission of the Croatian Ministry of Environment and Nature Protection (Decision no. UP/1-612-07/15-48/23).

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

References

  1. Mathew, B. The Iris, Featured Edition; BT Batsford Ltd.: London, UK, 1989. [Google Scholar]
  2. WCSP-World Checklist of Selected Plant Families. Facilitated by the Royal Botanic Gardens, Kew. Available online: http://wcsp.science.kew.org (accessed on 27 May 2020).
  3. Köhlein, F. Iris; Verlag Eugen Ulmer: Stuttgart, Germany, 1981. [Google Scholar]
  4. Boltenkov, E.V. Taxonomic notes on Iris ser. Lacteae (Iridaceae) with typifications of fifteen names and one new combination. Phytotaxa 2018, 383, 283–292. [Google Scholar] [CrossRef]
  5. Niketić, M.; Tomović, G.; Šiljak-Yakovlev, S. A new spontaneous hybrid between the cultivated and wild Iris species from Serbia. Bull. Nat. Hist. Mus. 2018, 11, 189–210. [Google Scholar] [CrossRef]
  6. Tillie, N.; Chase, M.W.; Hall, T. Molecular studies in the genus Iris L.: A preliminary study. Ann. Bot. 2000, 58, 105–112. [Google Scholar] [CrossRef]
  7. Wang, H.; Cui, Y.; Zhao, C. Flavonoids of the genus Iris (Iridaceae). Mini Rev. Med. Chem. 2010, 10, 643–661. [Google Scholar] [CrossRef]
  8. Mitić, B.; Halbritter, H.; Šoštarić, R.; Nikolić, T. Pollen morphology of the genus Iris L. (Iridaceae) from Croatia and surrounding area: Taxonomic and phylogenetic implications. Plant Syst. Evol. 2013, 299, 271–288. [Google Scholar] [CrossRef]
  9. Wilson, C.A. Resolving sectional relationships in subgenus Iris, the Eurasian bearded species. Syst. Bot. 2017, 42, 392–401. [Google Scholar] [CrossRef]
  10. Lawrence, G.H.M. A reclassification of the genus Iris. Gentes Herb. 1953, 8, 346–371. [Google Scholar]
  11. Mathew, B. The Iris, 1st ed.; BT Batsford Ltd.: London, UK, 1981. [Google Scholar]
  12. Wraber, T. Iris. In Mala flora Slovenije: Ključ za Določanje Praprotnic in Semenk; Martinčić, A., Wraber, T., Jogan, N., Ravnik, V., Podobnik, A., Turk, B., Vreš, B., Eds.; Tehniška Založba Slovenije: Ljubljana, Slovenia, 1999; pp. 657–659. [Google Scholar]
  13. Nikolić, T. Flora Croatica Database; Department of Biology, Faculty of Science, University of Zagreb: Zagreb, Croatia, 2019; Available online: http://hirc.botanic.hr/fcd (accessed on 29 March 2020).
  14. Mitić, B. Iris adriatica (Iridaceae), a new species from Dalmatia (Croatia). Phyton 2002, 42, 305–314. [Google Scholar]
  15. Alperth, F.; Mitić, B.; Mayer, S.; Maleš, Ž.; Kunert, O.; Hruševar, D.; Bucar, F. Metabolic profiling of rhizomes of native populations of the strictly endemic Croatian species Iris adriatica. Plant Biosyst. 2018, 153, 317–324. [Google Scholar] [CrossRef]
  16. Chase, M.W.; Soltis, D.E.; Olmstead, R.G.; Morgan, D.; Les, D.H.; Mishler, B.D.; Duvall, M.R.; Price, R.A.; Hills, H.G.; Qiu, Y.-L.; et al. Phylogenetics of seed plants: An analysis of nucleotide sequences from the plastid gene rbcL. Ann. Missouri Bot. Gard. 1993, 80, 528. [Google Scholar] [CrossRef]
  17. Soltis, P.S.; Folk, R.A.; Soltis, D.E. Darwin review: Angiosperm phylogeny and evolutionary radiations. Proc. R. Soc. B Biol. Sci. 2019, 286, 20190099. [Google Scholar] [CrossRef]
  18. Tang, S.; Okah, R.A.; Cordonnier-Pratt, M.M.; Pratt, L.H.; Johnson, V.E.; Taylor, C.A.; Arnold, M.L.; Knapp, S.J. EST and EST-SSR marker resources for Iris. BMC Plants Biol. 2009, 9, 72. [Google Scholar] [CrossRef] [PubMed]
  19. Kalia, R.K.; Rai, M.K.; Kalia, S.; Singh, R.; Dhawan, A.K. Microsatellite markers: An overview of the recent progress in plants. Euphytica 2011, 177, 309–334. [Google Scholar] [CrossRef]
  20. CBOL (Consortium for the Barcode of Life) Plant Working Group. A DNA barcode for land plants. Proc. Natl. Acad. Sci. USA 2009, 106, 12794–12797. [Google Scholar] [CrossRef]
  21. Hollingsworth, P.M.; Graham, S.W.; Little, D.P. Choosing and using a plant DNA barcode. PLoS ONE 2011, 6, e19254. [Google Scholar] [CrossRef]
  22. Dong, W.; Liu, J.; Yu, J.; Wang, L.; Zhou, S. Highly variable chloroplast markers for evaluating plant phylogeny at low taxonomic levels and for DNA barcoding. PLoS ONE 2012, 7, e35071. [Google Scholar] [CrossRef]
  23. Saddhe, A.A.; Kumar, K. DNA barcoding of plants: Selection of core markers for taxonomic groups. Plant Sci. Today 2017, 5, 9–13. [Google Scholar] [CrossRef]
  24. Greilhuber, J.; Doležel, J.; Lysak, M.A.; Bennett, M.D. The origin, evolution and proposed stabilization of the terms ‘Genome size’ and ‘C-value’ to describe nuclear DNA contents. Ann. Bot. 2005, 95, 255–260. [Google Scholar] [CrossRef]
  25. Garcia, S.; Leitch, I.J.; Anadon-Rosell, A.; Canela, M.A.; Galvez, F.; Garnatje, T.; Gras, A.; Hidalgo, O.; Johnston, E.; Mas de Xaxars, G.; et al. Recent updates and developments to plant genome size databases. Nucleic Acids Res. 2014, 42, D1159–D1166. [Google Scholar] [CrossRef]
  26. Leitch, I.J.; Johnston, E.; Pellicer, J.; Hidalgo, O.; Bennett, M.D. Angiosperm DNA C-Values Database (release 9.0, April 2019). 2019. Available online: https://cvalues.science.kew.org/ (accessed on 15 May 2020).
  27. Zonneveld, B.J.M.; Leitch, I.J.; Bennett, M.D. First nuclear DNA amounts in more than 300 angiosperms. Ann. Bot. 2005, 96, 229–244. [Google Scholar] [CrossRef]
  28. Šiljak-Yakovlev, S.; Muratović, E.; Bogunić, F.; Bašić, N.; Šolić, M.E.; Boškailo, A.; Bourge, M.; Valentin, N.; Stevanović, V.; Garnatje, T.; et al. Genome size of Balkan flora: A database (GeSDaBaF) and C-values for 51 taxa of which 46 are novel. Plant Syst. Evol. 2020, 306, 40. [Google Scholar] [CrossRef]
  29. Šmarda, P.; Knápek, O.; Březinová, A.; Horová, L.; Grulich, V.; Danihelka, J.; Veselý, P.; Šmerda, J.; Rotreklová, O.; Bureš, P. Genome sizes and genomic guanine+cytosine (GC) contents of the Czech vascular flora with new estimates for 1700 species. Preslia 2019, 91, 117–142. [Google Scholar] [CrossRef]
  30. Doležel, J.; Greilhuber, J.; Suda, J. Flow Cytometry with Plants: An overview. In Flow Cytometry with Plant Cells; Doležel, J., Greilhuber, J., Suda, J., Eds.; WILEY-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2007. [Google Scholar] [CrossRef]
  31. Bourge, M.; Brown, S.C.; Siljak-Yakovlev, S. Flow cytometry as tool in plant sciences, with emphasis on genome size and ploidy level assessment. Genet. Appl. 2018, 2, 1–12. [Google Scholar] [CrossRef]
  32. Bohanec, B. Ploidy determination using flow cytometry. In Doubled Haploid Production in Crop Plants; Maluszynski, M., Kasha, K.J., Forster, B.P., Szarejko, I., Eds.; Springer: Dordrecht, The Netherlands, 2003; pp. 397–403. [Google Scholar] [CrossRef]
  33. Šiljak-Yakovlev, S.; Pustahija, F.; Šolić, E.M.; Bogunić, F.; Muratović, E.; Bašić, N.; Catrice, O.; Brown, S.C. Towards a genome size and chromosome number database of Balkan flora: C-values in 343 taxa with novel values for 242. Adv. Sci. Lett. 2010, 3, 190–213. [Google Scholar] [CrossRef]
  34. Chase, M.W.; Cowan, R.S.; Hollingsworth, P.M.; Petersen, G.; Seberg, O.; Jorgsensen, T.; Cameron, K.M.; Carine, M. A proposal for a standardised protocol to barcode all land plants. Taxon 2007, 56, 295–299. [Google Scholar] [CrossRef]
  35. Kress, W.J.; Erickson, D.L. A two-locus global DNA barcode for land plants: The coding rbcL gene complements the non-coding trnH-psbA spacer region. PLoS ONE 2007, 2, e508. [Google Scholar] [CrossRef]
  36. Blazier, J.C.; Jansen, R.K.; Mower, J.P.; Govindu, M.; Zhang, J.; Weng, M.; Ruhlman, T.A. Variable presence of the inverted repeat and plastome stability in Erodium. Ann. Bot. 2016, 117, 1209–1220. [Google Scholar] [CrossRef]
  37. Sun, X.-Q.; Zhu, Y.-J.; Guo, J.-L.; Peng, B.; Bai, M.-M.; Hang, Y.-Y. DNA Barcoding the Dioscorea in China, a Vital Group in the Evolution of Monocotyledon: Use of matK Gene for Species Discrimination. PLoS ONE 2012, 7, e32057. [Google Scholar] [CrossRef]
  38. Dunning, L.T.; Savolainen, V. Broad-scale amplification of matK for DNA barcoding plants, a technical note. Bot. J. Linn. Soc. 2010, 164, 1–9. [Google Scholar] [CrossRef]
  39. Yu, J.; Xue, J.H.; Zhou, S.L. New universal matK primers for DNA barcoding angiosperms. J. Syst. Evol. 2011, 49, 176–181. [Google Scholar] [CrossRef]
  40. Wilson, C.A. Phylogeny of Iris based on chloroplast matK gene and trnK intron sequence data. Mol. Phyl. Evol. 2004, 33, 402–412. [Google Scholar] [CrossRef] [PubMed]
  41. Wilson, C.A. Subgeneric classification in Iris re-examined using chloroplast sequence data. Taxon 2011, 60, 27–35. [Google Scholar] [CrossRef]
  42. Ikinci, N.; Hall, T.; Lledo, M.D.; Clarkson, J.J.; Tillie, N.; Seismus, A.; Saito, T.; Harley, M.; Chase, M. Molecular phylogenetics of the juno irises, Iris subgenus Scorpiris (Iridaceae), based on six plastid markers. Bot. J. Linn. Soc. 2011, 167, 281–300. [Google Scholar] [CrossRef]
  43. Colasante, M. Iridaceae Present in Italy; Sapienza Università Editrice: Roma, Italy, 2014. [Google Scholar]
  44. Horvat, I.; Horvat, M. Iris croatica—Nova vrsta perunike u Hrvatskoj. Acta Bot. Croat. 1961, 20, 7–20. (In Croatian) [Google Scholar]
  45. Prodan, J. Die Flora der Dobrudscha und ein kurzer überblick über die Flora der Meeresküste Rumäniens. Bul. Minist. Agric. Domen. 1931, 3, 3–95. [Google Scholar]
  46. Mitić, B.; Cigić, P. Endemic taxa of the Genus Iris, L. (Iridaceae) in Croatia. Hrvat. Misao 2007, 7, 50–81. [Google Scholar]
  47. Trinajstić, I. Beitrag zur Taxonomie des Iris pallida Komplexes. Biosistematika 1976, 2, 67–78. [Google Scholar]
  48. Mitić, B.; Nikolić, T.; Liber, Z. Morphological and anatomical relationships in Alpine-Dinaric populations of the genus Iris L., Pallidae series (A. Kern.) Trinajstić (Iridaceae). Acta. Soc. Bot. Polon. 2000, 69, 285–291. [Google Scholar] [CrossRef]
  49. Mitić, B.; Nikolić, T.; Liber, Z. Morphological and karyological relationships within Alpine-Dinaric populations of the genus Iris L., Pallidae series (A. Kern.) Trinajstić (Iridaceae). Acta. Soc. Bot. Polon 2001, 70, 221–227. [Google Scholar] [CrossRef]
  50. Degen, A. Flora Velebitica; Ungar. Akad. Wiss.: Budapest, Hungary, 1936; Volume 1, p. 646. [Google Scholar]
  51. Veselý, P.; Bureš, P.; Šmarda, P.; Pavlíček, T. Genome size and DNA base composition of geophytes: The mirror of phenology and ecology. Ann. Bot. 2012, 109, 65–75. [Google Scholar] [CrossRef]
  52. Colasante, M.; Difford, M.; Vosa, C.G. Scanning electron microscopy of some critical bearded Irises pollen: Preliminary observations. Webbia 1989, 43, 339–350. [Google Scholar] [CrossRef]
  53. Makarevitch, I.; Golovnina, K.; Scherbik, S.; Blinov, A. Phylogenetic relationships of the Siberian Iris species inferred from noncoding chloroplast DNA sequences. Int. J. Plans Sci. 2003, 164, 229–237. [Google Scholar] [CrossRef]
  54. Wilson, C.A. Phylogenetic relationships among the recognised series in Iris section Limniris. Syst. Bot. 2009, 34, 277–284. [Google Scholar] [CrossRef]
  55. Wraber, T. Iris sibirica L., Sp. Pl. 39 (1753) subsp. erirrhiza (Posp.) T. Wraber, comb. nova. Hladnikia (Ljubl) 2001, 11, 27–43. [Google Scholar] [CrossRef]
  56. Pinar, N.M.; Oybak Dönmez, E. Pollen morphology of Turkish Iris L. (Iridaceae) with reference to evolutionary trends at the infrageneric level. Isr. J. Plants Sci. 2000, 48, 129–141. [Google Scholar] [CrossRef]
  57. Jiang, Y.L.; Huang, Z.; Liao, J.Q.; Song, H.X.; Luo, X.M.; Gao, S.P.; Lei, T.; Jiang, M.Y.; Jia, Y.; Chen, Q.B.; et al. Phylogenetic analysis of IRIS L. from China on chloroplast TRNL-F sequences. Biologia 2018, 73, 459–466. [Google Scholar] [CrossRef]
  58. Trapnell, D.W.; Hamrick, J.L.; Parker, K.C.; Braungart, K.W.; Glenn, T.C. Evaluating the utility of microsatellites for investigations of autopolyploid taxa. J. Hered. 2011, 102, 473–478. [Google Scholar] [CrossRef]
  59. Anvarkhah, S.; Khajeh-Hosseini, M.; Rashed-Mohassel, M.H.; Panah, A.D.E.; Hashemi, H. Identification of three species of genus Allium using DNA barcoding. Int. J. Agric. Crop Sci. 2013, 5, 1195–1203. [Google Scholar]
  60. Yeh, F.C.; Boyle, T.; Ye, Z.; Xiyan, J.M. Popgene Version 1.31: Microsoft Windows-Based Freeware for Population Genetic Analysis; University of Alberta and Center for International Forestry Research: Edmonton, AB, Canada, 1999. [Google Scholar] [CrossRef]
  61. Kalinowski, S.T.; Taper, M.L.; Marshall, T.C. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Mol. Ecol. 2007, 16, 1099–1106. [Google Scholar] [CrossRef]
  62. Wagner, H.W.; Sefc, K.M. Identity 1.0; Center for Applied Genetics, University of Agricultural Sciences Vienna: Vienna, Austria, 1999. [Google Scholar]
  63. Hammer, Ø.; Harper, D.A.T.; Ryan, P.D. PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontol. Electron. 2001, 4, 9. [Google Scholar]
Figure 1. Narrow endemic wild Alpine-Dinaric endemic species Iris adriatica: (ac) Individuals of different colours (Photo: Miroslav Mitić).
Figure 1. Narrow endemic wild Alpine-Dinaric endemic species Iris adriatica: (ac) Individuals of different colours (Photo: Miroslav Mitić).
Plants 09 01229 g001
Figure 2. (a) Distribution map of the narrow endemic Alpine-Dinaric species Iris adriatica included in our study (all localities are in Croatia, and are incorporated in the national Flora Croatica Database (https://hirc.botanic.hr/fcd)—FCD; marks: Plants 09 01229 i001—earlier data from the FCD; Plants 09 01229 i002—localities of collected specimens in our study); (b) habitat on the locality Brnjica-Pokrovnik; (c) habitat on the island of Cres.
Figure 2. (a) Distribution map of the narrow endemic Alpine-Dinaric species Iris adriatica included in our study (all localities are in Croatia, and are incorporated in the national Flora Croatica Database (https://hirc.botanic.hr/fcd)—FCD; marks: Plants 09 01229 i001—earlier data from the FCD; Plants 09 01229 i002—localities of collected specimens in our study); (b) habitat on the locality Brnjica-Pokrovnik; (c) habitat on the island of Cres.
Plants 09 01229 g002
Figure 3. UPGMA dendrogram obtained with Dice’s similarity coefficient based on eight SSR markers for 31 out of 32 collected Iris samples (as explained in the Material and Methods section); Bootstrap percentages (>50) are shown in the nodes of the dendrogram; labels I–III denote major clusters.
Figure 3. UPGMA dendrogram obtained with Dice’s similarity coefficient based on eight SSR markers for 31 out of 32 collected Iris samples (as explained in the Material and Methods section); Bootstrap percentages (>50) are shown in the nodes of the dendrogram; labels I–III denote major clusters.
Plants 09 01229 g003
Figure 4. Maximum Likelihood (ML) tree of 32 Iris samples and sequences from NCBI (http://www.ncbi.nlm.nih.gov) (I. missouriensis, I. sanguinea, I. gatesii) based on two plastid markers (rpoC1, ndhJ). Bootstrap percentages are shown in the nodes of the dendrogram; labels I–III denote major clusters.
Figure 4. Maximum Likelihood (ML) tree of 32 Iris samples and sequences from NCBI (http://www.ncbi.nlm.nih.gov) (I. missouriensis, I. sanguinea, I. gatesii) based on two plastid markers (rpoC1, ndhJ). Bootstrap percentages are shown in the nodes of the dendrogram; labels I–III denote major clusters.
Plants 09 01229 g004
Table 1. Values of observed (Ho) and expected (He) heterozygosity, number of alleles (n), effective number of alleles (ne), polymorphic information content (PIC), and probability of identity (PI) of 8 microsatellite loci for all studied samples of the Alpine-Dinaric taxa of the genus Iris.
Table 1. Values of observed (Ho) and expected (He) heterozygosity, number of alleles (n), effective number of alleles (ne), polymorphic information content (PIC), and probability of identity (PI) of 8 microsatellite loci for all studied samples of the Alpine-Dinaric taxa of the genus Iris.
LocusnneHoHePICPI
IM9394.10.4520.7690.7270.131
IM12364.00.8710.7630.7120.172
IM16472.00.1290.5180.4800.292
IM196125.50.5000.8330.8050.079
IM20083.40.3870.7210.6720.178
IM327126.20.5930.8550.8210.085
IM34881.90.1940.4900.4660.357
IM39197.20.7140.8770.8450.072
Average8.84.30.4800.7280.691-
Total-----2.01 × 10−7
Back to TopTop