Skip to Content
MDPIMDPI

Advanced

PlantsPlants
PDF
  • Article
  • Open Access

20 February 2026

23 Pages

Resolving Species Limits and Evolutionary Distinctiveness of the Libyan Endemic Arbutus pavarii (Ericaceae) Using Multilocus DNA Barcoding and Phylogenetic Analyses

,
,
,
,
and
1
Botany Department, Faculty of Sciences, University of Benghazi, Benghazi 1308, Libya
2
Centre for Plant Diversity and Systematics, The Harborne Building, School of Biological Sciences, University of Reading, Whiteknights, Reading RG6 6UB, UK
3
Botany Department, Faculty of Science, Fayoum University, Fayoum 63514, Egypt
4
National Center for Vegetation Cover Development and Combating Desertification, Riyadh 13312, Saudi Arabia
Plants2026, 15(4), 653;https://doi.org/10.3390/plants15040653
This article belongs to the Special Issue Plant Molecular Phylogenetics and Evolutionary Genomics IV
Version Notes
Order Reprints

Abstract

The taxonomic status of Arbutus pavarii Pamp., a rare and geographically restricted species endemic to northeastern Libya, has long been debated, with some treatments considering it a synonym of A. unedo. To resolve this uncertainty, we applied an integrative molecular framework that combined multilocus DNA barcoding, phylogenetic inference, and multivariate statistical analyses. Five barcode loci—nrITS, matK, rbcL, trnH–psbA, and rps16—were analyzed using barcode-gap diagnostics, TaxonDNA identification tests, and single-locus and concatenated phylogenetic analyses. Barcode-gap analyses based on Kimura 2-parameter distances revealed clear and reproducible separation between intra- and interspecific variation for A. pavarii, particularly for nrITS and the concatenated multilocus dataset, whereas conserved plastid loci showed limited discriminatory power when used individually. Phylogenetic reconstructions consistently recovered A. pavarii as a strongly supported monophyletic lineage, distinct from A. unedo and other Mediterranean congeners, with congruent topologies across the nuclear, plastid, and combined datasets. Multivariate analyses, including principal component analysis and heatmap clustering, further corroborate the genetic cohesion and distinctiveness of A. pavarii samples. Collectively, these results provide robust molecular evidence supporting the recognition of Arbutus pavarii as a distinct evolutionary lineage, rather than an intraspecific variant of A. unedo. This study established a reproducible multilocus framework for species delimitation in Arbutus and highlighted the importance of integrating nuclear and plastid markers to resolve complex taxonomic relationships. The clarified taxonomic status of A. pavarii has important implications for biodiversity assessment and conservation planning in the Mediterranean region, particularly in the Cyrenaican floristic province.

1. Introduction

The genus Arbutus L. (Ericaceae, subfamily Arbutoideae) comprises evergreen shrubs and small trees distributed across the Mediterranean Basin, Macaronesia, and western North America [1,2,3,4]. First described by Linnaeus in 1753, the name “Arbutus” is derived from the Latin term for “strawberry-tree” and the Celtic word “arboise,” which means “rough fruit” [5]. Approximately ten species have traditionally been recognized within the genus [3,6]; however, Plant of the World Online (POWO), maintained by the Royal Botanic Gardens, Kew, currently recognizes 12 accepted species, with Arbutus pavarii Pamp. is the only species endemic to Libya [7,8].
The genus Arbutus L. is distinguished from other Ericaceae by its multiple-ovulate berries with a papillate surface, a key diagnostic trait within the family [6,9]. Four species are commonly acknowledged in the Mediterranean–Atlantic region: A. unedo L., A. andrachne L., A. canariensis Veill., and A. pavarii Pamp. [2]. Arbutus pavarii was originally described in the Cyrenaican (Al-Akhdar) Mountains of eastern Libya [10,11], where it remains geographically restricted and ecologically specialized.
Historically, A. pavarii has been distinguished from the widespread A. unedo by a suite of vegetative and reproductive characters documented in its protologue and subsequent regional floristic treatment. These diagnostic traits include oblong-lanceolate leaves with dense, persistent tomentum on the abaxial surface, bark exfoliating in thin, papery layers, and comparatively larger fruits reaching up to 3 cm in diameter, together with differences in floral dimensions and inflorescence structure [7,11]. Despite this distinction, the taxonomic delimitation of A. pavarii remains contentious because of its substantial morphological overlap with A. unedo.
This ambiguity is further exacerbated by the pronounced phenotypic plasticity within the genus Arbutus. Recent studies on A. unedo have documented considerable interannual and genotypic variation in morpho-pomological, physiological, and physio-biochemical traits under natural conditions, highlighting the instability of several characters traditionally regarded as taxonomically diagnostic [12,13]. Such plasticity, combined with complex reproductive biology and developmental dynamics, has contributed to the repeated treatment of A. pavarii as a synonym of A. unedo in some taxonomic accounts and global databases [14].
In contrast, ecological and demographic investigations consistently recognize A. pavarii as a geographically restricted lineage endemic to Al-Akhdar Mountains, characterized by distinct population dynamics and significant conservation concern. The species is classified as Vulnerable on the IUCN Red List [15]; lately recoded as Near Threatened under criteria A2c [16], based on detailed demographic studies, further underscoring its threatened status within the Cyrenaican landscape [17]. Its endemic status is acknowledged by major biodiversity platforms, such as GBIF [18] and POWO. Ecologically, A. pavarii plays an important role in local ecosystems. It is used as a fuel source, its flowers provide nectar and pollen for pollinators, and its leaves, fruits, and bark are utilized locally for tanning, fodder, and human consumption. The fruits are nutritionally valuable, being rich in carbohydrates, antioxidants, and vitamin C [19].
Given the persistent taxonomic uncertainty surrounding A. pavarii and the limitations of morphology-based approaches in the presence of phenotypic plasticity, molecular methods are essential for clarifying species boundaries. DNA barcoding provides a standardized framework for plant identification and species delimitation by comparing short, conserved genomic regions with curated reference databases [20,21,22,23,24,25,26]. The nuclear ribosomal internal transcribed spacer (nrITS) is a highly effective marker for species-level resolution in plants, owing to its relatively rapid rate of sequence evolution [27,28,29], whereas plastid markers such as rbcL, matK, trnH-psbA, and rps16 provide complementary phylogenetic signal and enable robust multi-locus inference [30,31,32,33].
Previous phylogenetic studies have suggested that Arbutus is not monophyletic, with Mediterranean species showing genetic ties closer to Arctostaphylos, Arctous, and Comarostaphylis than to their North American counterparts. This pattern has been attributed to the Madrean–Tethyan disjunction hypothesis, which is associated with the Paleogene-Neogene biogeographic history [34,35,36]. Although Mediterranean and Canary Island species have been studied morphologically and genetically, A. pavarii has remained absent from global barcoding and phylogenetic datasets, leaving its evolutionary status unresolved. Recent ecological niche modeling has further predicted contrasting future habitat trajectories under different climate scenarios, emphasizing the vulnerability of this endemic species [37].
To resolve the taxonomic status of A. pavarii, this study employs a multi-locus DNA barcoding framework for the first time. We aim to develop a reference barcode library for A. pavarii, determine whether it constitutes a genetically distinct evolutionary lineage or represents intraspecific variation within A. unedo, and elucidate its phylogenetic relationships within the Mediterranean–Atlantic Arbutus complex. By integrating five barcode regions (rbcL, trnH–psbA, matK, rps16, and nrITS), this study provides an evidence-based resolution to long-standing taxonomic ambiguity and establishes a crucial foundation for the conservation of this vulnerable Libyan endemic.

2. Results

2.1. Sequencing Characteristics and Marker Performance

In this study, five candidate plant barcode loci—nrITS, psbA–trnH, rbcL, matK, and rps16—were successfully amplified and sequenced for all Arbutus specimens (Table 1). Sequencing was 100% successful across all loci, although amplification success varied, with nrITS, psbA–trnH, and rbcL achieving full PCR success, whereas rps16 had the lowest success rate at 87.5% (Supplementary Figure S1). The aligned sequence lengths varied by locus, from 400 bp for psbA–trnH to 852 bp for matK and 842 bp for rps16.
Table 1. Specimen voucher and GenBank accession for the five barcode loci in Arbutus. For each specimen, the table lists the voucher identifier and GenBank accession numbers for nrITS, psbA–trnH, rbcL, matK, and rps16 genes.
The mean Kimura 2-parameter (K2P) distances consistently indicated greater interspecific than intraspecific divergence for all loci, with barcoding gap ratios (inter/intra) exceeding 1 in every instance (Table 2). The highest barcoding-gap ratio was found for rbcL (5.24), followed by psbA–trnH (3.94), matK (3.52), nrITS (3.16), and rps16 (2.63). These findings underscore rbcL, psbA–trnH, and matK as the most informative single markers for species identification, with nrITS providing reliable performance and rps16 offering supportive but limited resolution of species identification.
Table 2. Performance comparison of five candidate DNA barcoding regions (n = 16). PCR success (%), aligned length (bp), mean intra- and interspecific K2P distance, and barcoding gap ratio (inter/intra). The sample size (n) refers to the number of PCR attempts per marker.

2.1.1. Barcode Gap Analysis and Marker Performance Based on K2P Genetic Distances

The median K2P distances intra- and interspecific indicated strong or moderate barcode gap signals for most loci (Supplementary Table S1). The nrITS locus had the highest inter/intra ratio (~13.1), followed by rps16 (~6), and the combined multilocus dataset (~4.5). The markers psbA-trnH and rbcL showed favorable patterns with zero intra-median distances, but their minimum interspecific distances were nearly zero, leading to some overlap. In contrast, matK exhibited the broadest range of distances, including a significant intraspecific outlier (>22%), which may suggest alignment variability or biological structures that require careful analysis in future studies.
Barcode-Gap Patterns Across Individual Loci
Density plots comparing intraspecific and interspecific K2P genetic distances (Figure 1) revealed statistically significant separation between the two distributions across all barcode loci, although the magnitude and clarity of the separation varied substantially among markers.
Figure 1. Barcoding gap evaluation using K2P genetic distance. Density distributions for intraspecific (blue) and interspecific (red) distances are shown for all sampled species across each barcode locus and the combined dataset. The vertical dashed line in each panel indicates the Best Close Match (BCM) threshold, calculated as the 95th percentile of intraspecific variation.
The nuclear nrITS locus showed a pronounced barcode gap pattern, with intraspecific distances tightly clustered near zero and interspecific distances shifted toward markedly higher values. The overlap between the two distributions was limited, indicating strong discriminatory power and consistent species-level resolution.
Among plastid loci, psbA–trnH displayed low intraspecific distances and a broad interspecific distribution; however, overlap at lower genetic distances was evident, resulting in partial separation between intra- and interspecific variations. The matK locus showed a similar pattern of moderate barcode-gap structure, characterized by generally low intraspecific divergence but including extreme intraspecific values that expanded the overlap region with interspecific distances.
In contrast, rbcL and rps16 exhibited weak or absent barcode gap structures. For both loci, intra- and interspecific distances were concentrated at low K2P values, with extensive overlap across most of the distance range, indicating a limited capacity for species-level discrimination when used individually.
The concatenated multilocus dataset produced the most consistent barcode-gap pattern, with intraspecific distances remaining tightly constrained and interspecific distances clearly shifted toward higher divergence, resulting in a minimal overlap between distributions.
Overall, these results demonstrate that species boundaries in Arbutus are most clearly resolved using a multilocus barcode framework, particularly when combining nrITS, matK, and psbA–trnH, with the distinctive genetic profile of A. pavarii consistently evident across the most informative loci.
Species-level barcode-gap scatterplots (Figure 2) further validated the presence of a distinct barcoding gap for most taxa, with A. pavarii points not overlapping with those of A. unedo points. For each species, the maximum intraspecific K2P distance (x-axis) was plotted against the minimum interspecific K2P distance (y-axis). The dashed red line (y = x) represents the threshold for a barcode gap, with points above the line indicating greater interspecific divergence than intraspecific divergence. This study highlights the effectiveness of DNA barcoding in differentiating Arbutus species within the Mediterranean region. Arbutus pavarii stands out with a distinct barcode gap, indicating its good potential for accurate species identification. In contrast, A. canariensis is noted for its considerable phylogenetic divergence, suggesting that it represents a unique evolutionary lineage within the genus. While the rps16 marker proved to be less effective because of overlapping sequences among species, the inclusion of markers such as nrITS, psbA–trnH, and matK, along with a concatenated dataset, revealed a much clearer differentiation among species. Statistical analysis using Wilcoxon one-sided tests further corroborated the significant interspecific divergence across all utilized markers, affirming the robustness of these findings (p < 0.05). This suggests that a combination of markers, especially those noted for strong species-level separability, is critical for resolving complex relationships within the genus Arbutus and enhancing the utility of DNA barcoding in this context.
Figure 2. Barcode gap validation for species delimitation in Arbutus. Scatterplots were used to compare the maximum intraspecific distance to the minimum interspecific distance for all sampled species across each barcode locus and the combined dataset.

2.2. Species Identification Success Under TaxonDNA Best Match and Best Close Match Criteria

TaxonDNA analyses were conducted using the Best Match (BM) and Best Close Match (BCM; 95th-percentile intraspecific threshold) identification criteria (Table 3). Identification success varied markedly among the loci under the BM criterion. The nrITS marker achieved perfect identification success, correctly assigning all samples (100%; 16/16). The concatenated multilocus dataset also performed strongly, yielding an accuracy of 87.5%. In contrast, rbcL showed moderate performance, with 56.3% correct identifications and a relatively high proportion of incorrect assignments (25%). Both matK and rps16 exhibited lower discriminatory ability, each achieving 37.5% correct identifications and a moderate proportion of incorrect or ambiguous results. The plastid spacer psbA–trnH showed the poorest performance, with many identifications classified as ambiguous (81.3%).
Table 3. Taxon DNA identification outcomes under Best Match (BM) and Best Close Match (BCM; threshold = 95th percentile of intraspecific K2P distance) for six barcode options (nrITS, psbA–trnH, rbcL, matK, rps16, and combined; n = 16 queries per marker).
The application of the Best Close Match (BCM) criterion using the 95th-percentile intraspecific distance threshold resulted in a substantial increase in ambiguous identifications across all loci. This threshold proved overly permissive for the present dataset, allowing multiple heterospecific sequences to fall within the “close match” range, thereby inflating ambiguity. This effect was particularly pronounced for matK and psbA–trnH, each yielding 87.5% ambiguous identifications. Under the BCM criterion, the combined multilocus dataset failed to produce any correct identification and was dominated by ambiguous outcomes (87.5%). Collectively, these results indicate that while BM-based TaxonDNA analyses provide meaningful insights into marker performance, the BCM approach with a 95th-percentile threshold is insufficiently stringent for reliable species discrimination in this genus.

2.3. Phylogenetic Analysis

2.3.1. Nuclear Dataset (nrITS), nr (DNA) Marker

The nrITS dataset provided the strongest phylogenetic signal, yielding a well-resolved species-level topology with high posterior probability support. Most internal nodes were strongly supported, with posterior probability values from 0.95 to 1.00. Mediterranean taxa (Arbutus andrachne, A. canariensis, A. unedo, and A. pavarii) formed a distinct and well-supported clade that was clearly separated from Western North American taxa (e.g., A. arizonica, A. menziesii, and A. occidentalis), reflecting well-defined geographic structuring within the genus.
Arbutus pavarii was recovered as a strongly supported monophyletic lineage (posterior probability = 0.96) and formed a sister relationship with A. andrachne. In contrast, A. unedo constituted a separate, well-supported lineage, clearly distinct from A. pavarii, indicating substantial genetic differentiation among Mediterranean congeners (Figure 3).
Figure 3. Bayesian phylogenetic tree based on nuclear ribosomal ITS sequences, rooted with Enkianthus campanulatus (Ericaceae). Posterior probability values are shown at nodes, and branch lengths are proportional to the estimated number of substitutions per site. The tree strongly supports the separation of A. pavarii (Libyan endemic) and A. unedo into distinct clades, confirming their status as separate species.

2.3.2. Multilocus Plastid Markers Dataset, cp (DNA) Markers

Among the four plastid markers evaluated (matK, rbcL, trnH–psbA, rps16), phylogenetic tree resolution varied among loci but remained broadly congruent in overall topology, with each marker contributing distinct strengths to phylogenetic reconstruction. The rps16 intron provided moderate support for deeper backbone relationships while consistently clustering conspecific accessions, indicating its utility for resolving species boundaries despite limited resolution at higher taxonomic levels. The rbcL marker yielded a stable and well-resolved species-level topology with strong clade support (posterior probability values 0.95–1.00), reflecting its conserved yet informative nature across Arbutus. The matK region offered good resolution at terminal nodes but showed comparatively weaker support along the backbone, consistent with a higher substitution rate and reduced performance in resolving deeper relationships. In contrast, the psbA–trnH intergenic spacer, despite shallow phylogenetic splits and limited sequence divergence, proved to be a reliable tool for clustering conspecific accessions with high support, highlighting its effectiveness for species-level discrimination rather than deep phylogenetic inference (Supplementary Figures S2–S5).
The concatenated plastid dataset yielded a topology largely congruent with the nrITS phylogeny at deeper nodes, supporting the major geographic and taxonomic structuring within Arbutus. However, it revealed a differing resolution of relation among Mediterranean taxa, particularly concerning the placement of A. unedo and A. pavarii.
Across all plastid analyses, Arbutus pavarii consistently emerged as a distinct lineage and well-supported lineage, clearly separate from A. unedo or A. andrachne, while Mediterranean and Canary Island taxa formed a cohesive and well-supported cluster (Figure 4).
Figure 4. Bayesian phylogenetic tree of Arbutus species based on the concatenated analysis of four chloroplast markers (matK, psbA, rbcL, rps16). The tree is rooted with Pyrola rotundifolia and Enkianthus campanulatus (Ericaceae) as outgroups. Posterior probability values are shown at nodes, and branch lengths are proportional to the estimated number of substitutions per site. The concatenated chloroplast dataset improves species-level resolution and provides increased nodal support relative to single-locus analyses, thereby reducing stochastic error.

2.3.3. Multilocus Concatenated Dataset

The five-locus concatenated phylogeny, combining the nuclear nrITS region with plastid loci, produced the most robust and highly resolved topology of the study, with strong posterior probability support (≥0.95) recovered for all interspecific nodes. This multilocus analysis consistently resolved Arbutus pavarii as a distinct and strongly supported monophyletic lineage (posterior probability = 1.00), clearly separated from other Mediterranean congeners. The topology further revealed a pronounced biogeographic structure, with Mediterranean species forming a well-supported clade distinct from Western North American taxa. Species represented by multiple accessions were consistently recovered as monophyletic, each supported by high posterior probability values, indicating strong phylogenetic coherence across the genus. Overall, the concatenated dataset provides a stable and well-supported phylogenetic framework for Arbutus (Figure 5).
Figure 5. Bayesian phylogenetic tree based on combined nuclear (nrITS) and plastid (psbA–trnH, rbcL, matK, rps16) sequence data, rooted with Ericaceae outgroups (Pyrola rotundifolia, Vaccinium ovatum, Enkianthus campanulatus). Posterior probability values are shown at nodes, and branch lengths are proportional to the estimated number of substitutions per site. This total-evidence phylogeny integrates signals from both nuclear and chloroplast genomes, providing the most comprehensive and robust reconstruction of relationships within the genus to date. Strong concordance between nuclear and plastid markers reinforces species delimitations and supports the distinctiveness of A. pavarii from A. unedo, among other clades.

2.3.4. Genetic Structure and Multivariate Analyses (Heatmap and PCA)

A phylogenetic correlation matrix derived from the concatenated dataset confirmed clear genetic separation among the four Mediterranean taxa, showing noticeably higher genetic similarity within species than between them (Figure 6). As anticipated, the outgroup genera (Enkianthus, Pyrola, and Vaccinium) formed a distinct cluster, exhibiting maximal genetic divergence from all members of the genus Arbutus, collectively affirming the strong phylogenetic structure recovered in this study.
Figure 6. Heatmap of pairwise genetic distances validates species boundaries and biogeographic structure. The matrix shows Kimura 2-parameter (K2P) distances based on the concatenated five-locus dataset (nrITS + psbA–trnH + rbcL + matK + rps16), with hierarchical clustering of rows and columns. Cool colors (blue) indicate low genetic divergence, while warm colors (yellow/red) indicate high divergence. The tight clustering and low intraspecific divergence (dark blue squares along the diagonal) for species like Arbutus pavarii confirm strong genetic cohesion. The clear separation between the Mediterranean clade (containing A. pavarii, A. unedo, A. andrachne, A. canariensis) and the western North American clade (e.g., A. menziesii, A. arizonica) is evident as a distinct warm-colored block, corroborating the deep phylogenetic split.

2.3.5. Principal Component Analysis (PCA)

Strongly confirmed robust clustering consistent with species boundaries, with the first two principal components accounting for most of the total variance. PC1 (68.9%) clearly separated the Mediterranean and Western North American lineages, while PC2 (18.3%) captured the fundamental variation within these major clades (Figure 7). Notably, A. pavarii formed a tight, well-defined group distinct from other Mediterranean species, such as A. unedo and A. andrachne, which also formed discrete clusters. In contrast, North American taxa exhibited greater dispersion along the axes, suggesting higher levels of intraspecific heterogeneity. The complete isolation of outgroups on the positive PC1 axis further underscored the resolution of the analysis. Combined with the genetic distance heatmaps, these PCA results validate the power of the multilocus dataset to delineate species and reveal major biogeographic patterns within Arbutus.
Figure 7. Principal Component Analysis (PCA) of genetic variation within Arbutus based on concatenated barcode loci. The plot shows clear genetic structuring, with individuals color-coded by biogeographic group: Mediterranean species (blue), western North American species (orange), and Ericaceae outgroups (gray). The first two principal components explain 87.2% of the total variance (PC1 = 68.9%, PC2 = 18.3%). The tight clustering of Arbutus pavarii samples, distinct from A. unedo and other congeners, provides multivariate support for its genetic distinctiveness. Convex hulls emphasize the strong east–west (Mediterranean vs. North American) phylogeographic partition within the genus.

3. Discussion

3.1. Taxonomic Resolution of Arbutus pavarii Using Multilocus DNA Barcoding

Taxonomic resolution of Arbutus pavarii using multilocus DNA barcoding by integrating five DNA barcode markers (nrITS, matK, rbcL, trnH–psbA, and rps16) with species-level divergence metrics, TaxonDNA identification tests, and multilocus phylogenetic analyses establishes a robust and reproducible framework for resolving the taxonomic status of the Libyan endemic. The combined application of barcode-gap analyses, distance-based identification criteria, and phylogenetic inference provides complementary lines of molecular evidence supporting species delimitation, consistent with recent advances in plant DNA barcoding methodology and analytical optimization [21,22,31].
Across loci, A. pavarii exhibited a pronounced and reproducible barcode gap, characterized by a clear disparity between interspecific and intraspecific K2P distances as defined by [38]. Key markers—particularly nrITS, matK, and trnH–psbA—consistently distinguished A. pavarii from its closest Mediterranean congeners, demonstrating their effectiveness for species-level identification in angiosperms and corroborating findings from recent barcoding studies [31,39]. Notably, no overlap or misidentification was detected between A. pavarii and A. unedo, a result that directly contradicts earlier taxonomic treatments that synonymized the two species [7] and highlights the limitations of morphology-based classifications in this group.

3.2. Phylogenetic Evidence for Species Delimitation in Arbutus pavarii

The phylogenetic analyses provide strong molecular evidence for species delimitation within the genus Arbutus. The nrITS phylogeny, widely recognized for its high resolving power in plant systematics [28,40,41], consistently recovered Arbutus pavarii as a well-supported monophyletic lineage, resolved as sister to A. andrachne. This topology is congruent with previously reported relationships among Mediterranean Arbutus species [34,36].
Plastid and concatenated multilocus phylogenies yielded broadly congruent topologies, reinforcing the robustness of this relationship and minimizing the likelihood that the observed pattern reflects locus-specific artifacts. Across datasets, A. pavarii was consistently resolved outside the Mediterranean–Macaronesian assemblage comprising A. unedo, A. andrachne, and A. canariensis, indicating an independent evolutionary trajectory. This pattern is consistent with recent biogeographic reconstructions within the genus [2] and supports the interpretation that historical isolation within the Cyrenaican floristic province has played a key role in shaping the evolutionary history of A. pavarii.
Insights into the genomic structure of A. unedo, particularly the presence of extensive chloroplast genome rearrangements, further underscore the evolutionary complexity of Arbutus. In parallel, multilocus comparisons between wild and cultivated Arbutus populations [42], together with evidence for deep-time diversification across Ericales [43], highlight pronounced phylogeographic structuring within the genus. Collectively, these independent lines of evidence position A. pavarii as an evolutionary distinct rather than a derivative of A. unedo, underscoring the importance of integrated nuclear, plastid, and concatenated datasets when resolving species boundaries in Mediterranean woody taxa.
Comparison of nuclear and plastid phylogenies reveals a high degree of overall congruence in the major geographic and taxonomic structure of Arbutus, particularly in the consistent separation of Mediterranean and Western North American lineages. Both datasets strongly support the monophyly and evolutionary distinctiveness of A. pavarii, which is consistently recovered as an independent lineage across all analyses. Minor topological differences are evident in the relative placement of Mediterranean congeners, especially A. undeo and A. andrachne, between nuclear (ITS) and plastid datasets. In particular slight variation was observed in the broader placement of A. pavarii relative to A. andrachne; however, the central results—strong monophyly and robust support for A. pavarii as a lineage distinct from A. undeo—are fully congruent across all datasets.
Such cytonuclear incongruence is well documented in recently diverged plant lineages and reflects fundamental properties of gene tree-species tree discordance under the multispecies coalescent framework [44,45]. In plants, additional contributing factors may include historical introgression, chloroplast capture, and the contrasting inheritance modes and effective population sizes of nuclear versus plastid genomes [46,47]. Importantly, none of these processes undermines the primary conclusion of this study; on the contrary, the concordant recovery of A. pavarii as a distinct lineage across both nuclear and plastid genomes provides robust multilocus evidence supporting its recognition as an evolutionarily independent species.

3.3. Statistical and Multivariate Support for Genetic Distinctiveness

Statistical and multivariate analyses further corroborated the phylogenetic evidence for the distinctiveness of A. pavarii. Principal Component Analysis (PCA) based on multilocus genetic distance matrices revealed a compact and well-defined cluster corresponding to A. pavarii, clearly separated from Mediterranean and western North American Arbutus species. This pattern is consistent with expectations for a well-delimited species exhibiting low internal genetic variance relative to interspecific divergence.
Heatmap-based clustering analyses produced concordant results, highlighting strong genetic cohesion among A. pavarii samples and pronounced divergence from other species. These findings agree with earlier population-level studies in A. unedo that employed ISSR and RAPD markers [48], which documented substantial genetic structuring and geographic differentiation. In the context of the present study, the observed clustering patterns support the hypothesis that long-term geographic isolation has contributed to the maintenance of a distinct genetic identity in A. pavarii. Together, the multivariate and statistical analyses provide independent, quantitative support for species-level separation and reinforce conclusions drawn from phylogenetic and barcode-gap evidence.

3.4. Implications for Taxonomy and Conservation

The recognition of A. pavarii as a distinct species has important implications for both taxonomy and conservation. Its recovery as a genetically cohesive and evolutionarily independent lineage reflects a prolonged history of adaptation within a geographically and ecologically distinct region [6,9]. Given ongoing pressures from habitat loss and climate change, particularly in parts of North Africa where conservation planning remains limited, acknowledging the independent taxonomic status of A. pavarii is a necessary step toward informed conservation prioritization [37]. More broadly, these results underscore the value of integrating phylogenetic, barcoding, and multivariate statistical approaches to resolve complex species boundaries in Mediterranean plant lineages.

3.5. Limitations and Future Perspectives

Although strong and consistent evidence supports the classification of Arbutus pavarii as a distinct evolutionary lineage, several limitations should be acknowledged. Sequence completeness varied among loci and taxa, particularly for sequences obtained from public databases, resulting in uneven representation across datasets. Missing data can influence phylogenetic resolution, although its impact is reduced when a strong and congruent signal exists across independent loci [49,50]. Future studies using complete plastid genomes or genome-wide nuclear data will further improve resolution [51].
In addition, the relatively small sample size for some populations may underestimate intraspecific genetic diversity. Expanded sampling across the species’ range is needed to better characterize population structure and historical demography [52,53]. Finally, integrative approaches combining molecular, morphological, anatomical, and ecological data will further strengthen taxonomic conclusions [54,55].

4. Materials and Methods

4.1. Plant Material

Sampling of the genus Arbutus was designed to support molecular phylogenetic inference through accurate species identification, traceable provenance, and adequate taxonomic and geographic representation. Herbarium vouchers from the University of Reading Herbarium (RNG) were selected because they provide reliably identified material curated under standardized conditions, which is essential for comparative molecular studies. A total of seventeen vouchers representing Arbutus pavarii, A. unedo, A. andrachne, and A. canariensis were selected to achieve balanced interspecific coverage while avoiding overrepresentation of any single taxon (Table 4, Scheme 1). This sampling strategy provides sufficient representation of the Mediterranean–Atlantic Arbutus complex for evaluating species boundaries and phylogenetic relationships. To expand the range and increase phylogenetic signal, additional Arbutus sequences available from GenBank were incorporated in a complementary manner. Only accessions with clear taxonomic identification, voucher information, and adequate sequence length were retained to minimize potential errors associated with misidentification or incomplete data.
Table 4. Herbarium vouchers are used for DNA extraction and barcoding.
Scheme 1. Sampling localities of Arbutus species included in the analysis.
Outgroup sampling included three Ericaceae taxa—Enkianthus campanulatus, Pyrola rotundifolia, and Vaccinium ovatum— selected to provide an appropriate external framework for rooting the tree while maintaining close familial affinity (Table 5). The overall sampling design follows widely accepted practices in molecular systematics and ensures that downstream analyses are based on a coherent, well-documented, and phylogenetically informative dataset.
Table 5. GenBank accession numbers for each barcode locus.

4.2. DNA Extraction, Amplification, and Marker Selection

Genomic DNA was extracted from approximately 20–30 mg of dried leaf material obtained from herbaria or silica-dried specimens. Leaf tissue was finely pulverized in liquid nitrogen to ensure effective disruption of cell walls, following protocols optimized for archival and degraded plant material. DNA extraction was performed using a modified cetyltrimethylammonium bromide (CTAB) protocol [56,57,58], with additional adjustments to enhance DNA yield and purity from degraded tissues and samples rich in secondary metabolites [24,59].
The preheated CTAB extraction buffer contained CTAB, Tris-HCl, EDTA, NaCl, and polyvinylpyrrolidone (PVP), with β-mercaptoethanol added immediately prior to use to reduce oxidative damage and bind phenolic compounds. Samples were incubated at 65 °C with gentle inversion to facilitate cell lysis and disruption of nucleoprotein complexes. Proteins and other contaminants were separated by chloroform: isoamyl alcohol (24:1) extraction followed by centrifugation. DNA was precipitated from the recovered aqueous phase using cold isopropanol, pelleted by centrifugation, washed with 70% ethanol to remove residual salts and impurities, air-dried, and resuspended in sterile TE buffer.
DNA quality and integrity were evaluated by electrophoresis on 1% agarose gels stained with ethidium bromide. DNA concentrations were estimated by visual comparison of band intensity against HyperLadder I (Bioline, London, UK). Samples exhibiting clear high-molecular-weight bands with minimal covering were considered suitable for downstream molecular analyses, whereas samples showing lower yield or quality were subjected to additional purification steps where necessary. This approach integrates previously validated CTAB modifications tailored for challenging plant materials and demonstrates robust performance across diverse taxa prepared from silica-dried and herbarium sources [24,57].
Five barcode loci were targeted for amplification: the nuclear ribosomal marker (nrITS) and four plastid markers (rbcL, trnH–psbA, matK, and rps16). These loci were selected based on established plant DNA barcoding frameworks and their demonstrated universality, amplification success, and discriminatory power across angiosperms [60,61]. Polymerase chain reactions (PCRs) were performed in 25 μL reaction volumes using a commercial premixed polymerase system, supplemented with bovine serum albumin (BSA) to improve amplification from herbarium-derived DNA. Reaction conditions followed protocols successfully applied in previous molecular systematic studies of woody plants and archival material [62,63]. Successful amplification and expected fragment sizes were verified by agarose gel electrophoresis prior to sequencing.
Primer sequences, annealing temperatures, and locus-specific PCR cycling conditions are provided in (Supplementary Table S2). Successful amplification and expected fragment sizes were confirmed by agarose gel electrophoresis prior to sequencing.
PCR products were enzymatically purified using ExoSAP-IT (USB Corporation, Cleveland, OH, USA) according to the manufacturer’s recommendations to remove residual primers and unincorporated dNTPs prior to sequencing. Purified amplicons were then outsourced to Eurofins Genomics Germany GmbH (Ebersberg, Germany) for bidirectional Sanger sequencing (forward and reverse directions) using an ABI 3730XL Genetic Analyzer (Life Technologies Corporation, Carlsbad, CA, USA).
Raw chromatogram reads from both directions were assembled into contigs and edited to produce high-confidence consensus sequences using GENEIOUS Prime® v.R9 (Biomatters Ltd., ST, USA, https://www.geneious.com/). During assembly and editing, ambiguous base calls were checked and corrected where supported by chromatogram evidence, and low-quality terminal regions were trimmed to minimize sequencing artifacts prior to downstream analyses.

4.3. Data Analysis

4.3.1. DNA Barcode Analyses

Sequence alignments were performed independently for each locus using ClustalW as implemented in BioEdit v7.0 (https://bioedit.software.informer.com/) [64,65]. Resulting alignments were examined by visual inspection and manually refined to correct obvious misalignments, particularly in regions affected by insertions, deletions, and length polymorphisms, which are common in plant barcode loci. Curated single-locus alignments were subsequently concatenated to generate a multilocus matrix for evaluating species discrimination under an integrative DNA barcoding framework.
All DNA barcoding analyses and quality-control procedures were conducted in the R statistical environment (v4.x; https://www.r-project.org/) using RStudio (01 January 2023; Posit Software, PBC, https://posit.co/products/open-source/rstudio/, accessed on 6 December 2025) [66,67]. Summary statistics, PCR and sequencing success rates, and pairwise genetic distances were calculated using the packages ape and pegas. Genetic divergence was estimated under the Kimura two-parameter (K2P) model, which remains widely used in plant barcoding studies for comparative purposes [60].
Barcode performance and species-level discriminatory power were quantitatively assessed using the Barcoding Gap Ratio, defined as the ratio between mean interspecific and mean intraspecific K2P distances following [38]. This metric provides an objective evaluation of whether genetic divergence between species exceeds within-species variation, thereby indicating the presence of a barcode gap. Because this study represents the first multilocus DNA barcoding assessment of Arbutus pavarii, particular emphasis was placed on combining distance-based criteria with phylogenetic inference to ensure robust species delimitation.
Species identification success was further assessed using the TaxonDNA criteria of Best Match (BM), Best Close Match (BCM), and All Species Barcodes [68]. All barcoding analyses were implemented in the R environment using the packages ape, dplyr, ggplot2, and phangorn. Statistical differentiation between intra- and interspecific genetic distance distributions was tested using one-sided Wilcoxon rank-sum tests based on median values, providing a conservative, non-parametric evaluation of barcode discrimination.
Diagnostic visualization was employed to complement numerical metrics and to explicitly assess DNA barcoding effectiveness. This included (i) density plots comparing intra- and interspecific K2P distance distributions to evaluate the presence and extent of a barcoding gap, and (ii) species-level barcode-gap scatterplots plotting maximum intraspecific against minimum interspecific distances relative to the y = x boundary, facilitating the identification of taxa exhibiting unambiguous molecular discrimination versus those showing distance overlap. Collectively, these analyses establish a rigorous, reproducible, and diagnostically informative DNA barcoding framework suitable for systematic and evolutionary inference of Arbutus.

4.3.2. Phylogenetic Analyses

Three molecular datasets were assembled, incorporating newly generated and GenBank-accessible sequences: (i) nrDNA: ITS region (25 sequences; nine taxa), (ii) cpDNA (matK, rbcL, rps16, trnH–psbA), and (iii) Concatenated: nrITS four plastid loci (25 sequences; seven taxa).
The best-fitting substitution models were determined using MrModeltest v2.3 (https://github.com/nylander/MrModeltest2, accessed on 6 December 2025) [69] and PAUP* v4b10 (http://phylosolutions.com/paup-test/, accessed on 6 December 2025) [70]: HKY + I + G for nrDNA, GTR + G for cpDNA, and GTR + I + G for the concatenated matrix. Bayesian inference (BI) was performed in MrBayes v3.2 (http://nbisweden.github.io/MrBayes/, accessed on 6 December 2025) [71] using four Metropolis-coupled MCMC chains run for 10 million generations, sampling every 1000 generations. Convergence was assessed in Tracer v1.5 (https://beast.community/tracer, accessed on 6 December 2025) [72]; the first 10% of samples were discarded as burn-in. Posterior probabilities were calculated from post-burn-in trees to generate a 50% majority-rule consensus; the resulting trees were visualized and annotated using FigTree v1.4.0 (http://tree.bio.ed.ac.uk/software/figtree/, accessed on 6 December 2025) [73].
To further explore the genetic structure, a phylogenetic correlation matrix was computed using ape package [74]. Among-taxon similarity patterns were summarized using heatmaps generated with the pheatmap and ggplot2 packages [75,76] and pairwise K2P distance matrices. Multivariate patterns were visualized using Principal Component Analysis (PCA) via factoextra and ggplot2, generating species-level scatterplots that highlight clustering patterns and geographical structure [77,78,79].

5. Conclusions

This research undertakes a thorough molecular reevaluation of the taxonomic classification of the Libyan endemic species, Arbutus pavarii, resolving persistent ambiguities about its systematic categorization. Utilizing multilocus DNA barcoding, barcode-gap diagnostics, phylogenetic inference, and multivariate statistical analyses, we confirm that A. pavarii is a genetically cohesive and evolutionarily distinct lineage, clearly separate from A. unedo and other Mediterranean relatives. The strong agreement among nuclear, plastid, and concatenated datasets reinforces the reliability of this conclusion and underscores the importance of using complementary molecular markers in species delimitation. Beyond defining species boundaries, this study establishes a reproducible molecular framework that aligns with current standards in plant systematics and is broadly applicable to taxonomically intricate Mediterranean lineages. Recognizing A. pavarii as a distinct species has significant implications for biodiversity assessment and conservation planning, especially in the Cyrenaican region, where endemic species are increasingly threatened by environmental pressures. By providing conclusive molecular evidence for the evolutionary uniqueness of A. pavarii, this research enhances the understanding of diversity within the Arbutus genus and supports informed conservation and management strategies for Mediterranean plant biodiversity.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/plants15040653/s1. Supplementary Tables S1 and S2; Supplementary Table S1. Median K2P genetic distances within and between species, along with range overlap and summary verdict per marker. Supplementary Table S2. Primer pairs and PCR conditions for the five barcoding regions. Primer names, sequences (5′–3′), original references, annealing temperatures, expected product length, and thermal cycling profiles for nrITS, rbcL, matK, psbA–trnH, and rps16. Supplementary Figures S1–S5; Supplementary Figure S1. PCR amplification success rates for five DNA barcode loci in Arbutus. Supplementary Figure S2. Maximum-likelihood phylogeny of Arbutus based on the plastid rps16 intron. Supplementary Figure S3. Maximum-likelihood phylogeny of Arbutus based on plastid rbcL. Supplementary Figure S4. Maximum-likelihood phylogeny of Arbutus based on plastid matK. Supplementary Figure S5. Maximum-likelihood phylogeny of Arbutus based on the plastid spacer psbA–trnH.

Author Contributions

Conceptualization: Data Curation: A.M.H.G., A.C., S.L.J., A.E.-B.; Formal Analysis: A.M.H.G., F.Y.E., A.E.-B.; Funding: A.M.H.G., A.C., S.L.J., A.A.A.; Acquisition: A.M.H.G., A.C.; Investigation: A.M.H.G., A.C., F.Y.E., A.A.A., S.L.J., A.E.-B.; Methodology: A.M.H.G., A.C., F.Y.E., S.L.J., A.E.-B.; Project Administration: A.M.H.G., A.C., S.L.J.; Resources: A.M.H.G., A.C., F.Y.E., S.L.J., A.E.-B.; Software: A.M.H.G., A.C., F.Y.E., A.A.A., S.L.J., A.E.-B.; Supervision: A.C., S.L.J., A.E.-B.; Validation: A.C., S.L.J., A.A.A., A.E.-B.; Visualization: A.M.H.G., F.Y.E., A.E.-B.; Writing—original Draft: A.M.H.G., F.Y.E., A.E.-B.; Writing—Review & Editing: A.A.A., A.E.-B. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable for this section.

Data Availability Statement

Data is contained within the article.

Conflicts of Interest

The authors declare that they have no competing interests.

References

  1. Angiosperm Phylogeny Group. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG III: APG III. Bot. J. Linn. Soc. 2009, 161, 105–121. [Google Scholar] [CrossRef] [Scilit]
  2. De Santis, S.; Michelangeli, F.; Spada, F.; Magri, D. Longitudinal population dynamics of Mediterranean-Atlantic Arbutus during the last 30 ka. Rev. Palaeobot. Palynol. 2024, 325, 105099. [Google Scholar] [CrossRef] [Scilit]
  3. Stevens, P.F.; Luteyn, J.; Oliver, E.G.H.; Bell, T.L.; Brown, E.A.; Crowden, R.K.; George, A.S.; Jordan, G.J.; Ladd, P.; Lemson, K.M.; et al. The Families and Genera of Vascular Plants. volume, VI. In Flowering Plants: Dicotyledons. Celastrales, Oxalidales, Rosales, Cornales, Ericales; Kubitzki, K., Ed.; Springer: Berlin/Heidelberg, Germany, 2004. [Google Scholar]
  4. Sutton, J. ‘Arbutus’ from the website Trees and Shrubs Online. Trees and Shrubs Online. 2021. Available online: https://www.treesandshrubsonline.org/articles/arbutus/ (accessed on 13 August 2025).
  5. Gledhill, D. The Names of Plants; Cambridge University Press: Cambridge, UK, 2008. [Google Scholar]
  6. Sørensen, P.D. Arbutus Linnaeus. Flora Neotropica Monogr. 1995, 66, 194–221. Available online: https://www.nybg.org/bsci/res/lut2/arbutus.html (accessed on 13 August 2025).
  7. Jafri, S.M.H.; El-Gadi, A. El-Gadi, Flora of Libya; Al Faateh University Press: Tripoli, Libya, 1978. [Google Scholar]
  8. POWO. Arbutus in Kew Science Plants of the World Online. 2025. Available online: https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:30008667-2 (accessed on 6 December 2025).
  9. Diggs, G.M.; Breckon, G.J. Generic Circumscription in the Arbuteae (Ericaceae). Ph.D. Thesis, The University of Wisconsin—Madison, Madison, WI, USA, 1981. [Google Scholar]
  10. Pampanini, R. L’Arbutus della Cirenaica. Arch. Bot. 1936, 12, 169–174. [Google Scholar]
  11. Pampanini, R. Un manipolo di piante della Cirenaica, Pellas, 1912.
  12. D’Agostino, A.; Di Marco, G.; Canini, A.; Gismondi, A. Megagametophyte maturation dynamics and flavonol changes during Arbutus unedo flower development. Front. Plant Sci. 2025, 16, 1694629. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  13. Zitouni, H.; Charafi, J.; Ourradi, H.; Zerhoune, M.; Hanine, H. Interannual and Genotypic Variation of Morpho-Pomologicaland Physico-Biochemical Traits of Strawberry Tree (Arbutus unedo L.) Genotypes Grown Under Natural Moroccan Conditions. Vegetos. 2026. Available online: https://link.springer.com/10.1007/s42535-025-01620-4 (accessed on 25 January 2026).
  14. WFO. Arbutus pavarii Pamp. 2025. Available online: https://www.worldfloraonline.org/taxon/wfo-0000543169 (accessed on 6 December 2025).
  15. IUCN. Arbutus pavarii: World Conservation Monitoring Centre: The IUCN Red List of Threatened Species 1998:E.T30323A9535904. 1998. Available online: https://www.iucnredlist.org/species/30323/9535904 (accessed on 25 January 2026).
  16. IUCN. Arbutus pavarii: Harvey-Brown, Y.: The IUCN Red List of Threatened Species 2022: E.T30323A199350094. 2021. Available online: https://www.iucnredlist.org/species/30323/199350094 (accessed on 25 January 2026).
  17. Kabiel, H.F.; Hegazy, A.K.; Lovett-Doust, L.; Al-Rowaily, S.L.; Borki, A.E.N.E. Demography of the threatened endemic shrub, Arbutus pavarii, in the Al-Akhdar mountainous landscape of Libya. J. For. Res. 2016, 27, 1295–1303. [Google Scholar] [CrossRef] [Scilit]
  18. GBIF.org 2025. Available online: https://www.gbif.org (accessed on 13 January 2025).
  19. Hegazy, A.K.; Al-Rowaily, S.L.; Faisal, M.; Alatar, A.A.; El-Bana, M.I.; Assaeed, A.M. Nutritive value and antioxidant activityof some edible wild fruits in the Middle East. J. Med. Plant Res. 2013, 7, 938–946. [Google Scholar] [CrossRef]
  20. Antil, S.; Abraham, J.S.; Sripoorna, S.; Maurya, S.; Dagar, J.; Makhija, S.; Bhagat, P.; Gupta, R.; Sood, U.; Lal, R.; et al. DNAbarcoding, an effective tool for species identification: A review. Mol. Biol. Rep. 2023, 50, 761–775. [Google Scholar] [CrossRef] [Scilit]
  21. Barman, R.; Banik, D. DNA barcoding clubbed with morphomatrix of commonly distributed species of Myristicaceae from North East India. 3 Biotech 2025, 15, 424. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  22. Botha, I.; De Canha, M.N.; Oberlander, K.; Botes, J.; Lall, N.; Berger, D.K. DNA barcoding and anti-tyrosinase activities of threespecies-representative populations of the genus Greyia Hook & Harv. S. Afr. J. Bot. 2026, 189, 55–67. [Google Scholar] [CrossRef] [Scilit]
  23. Chac, L.D.; Thinh, B.B. Species Identification through DNA Barcoding and Its Applications: A Review. Biol. Bull. 2023, 50, 1143–1156. [Google Scholar] [CrossRef] [Scilit]
  24. El-Banhawy, A.; Nour, I.H.; Acedo, C.; ElKordy, A.; Faried, A.; Al-Juhani, W.; Gawhari, A.M.H.; Olwey, A.O.; Ellmouni, F.Y. Taxonomic Revisiting and Phylogenetic Placement of Two Endangered Plant Species: Silene leucophylla Boiss. and Silene schimperiana Boiss. (Caryophyllaceae). Plants 2021, 10, 740. [Google Scholar] [CrossRef] [Scilit]
  25. Faried, A.; El-Banhawy, A.; Elqahtani, M. Taxonomic, DNA Barcoding and Phylogenetic Reassessment of the Egyptian Ephedra, L. (Ephedraceae). Catrina Int. J. Environ. Sci. 2018, 17, 1–13. [Google Scholar] [CrossRef] [Scilit]
  26. Miralles, A.; Puillandre, N.; Vences, M. DNA Barcoding in Species Delimitation: From Genetic Distances to Integrative Taxonomy. In DNA Barcoding; DeSalle, R., Ed.; Springer: New York, NY, USA, 2024; pp. 77–104. [Google Scholar] [CrossRef] [Scilit]
  27. Ahmed, H.I.S.; Badr, A.; El-Shazly, H.H.; Watson, L.; Fouad, A.S.; Ellmouni, F.Y. Molecular Phylogeny of Trifolium, L. Section Trifolium with Reference to Chromosome Number and Subsections Delimitation. Plants 2021, 10, 1985. [Google Scholar] [CrossRef] [Scilit]
  28. Hollingsworth, P.M.; Li, D.-Z.; van der Bank, M.; Twyford, A.D. Telling plant species apart with DNA: From barcodes to genomes. Philos. Trans. R. Soc. B Biol. Sci. 2016, 371, 20150338. [Google Scholar] [CrossRef] [Scilit]
  29. Paloi, S.; Luangsa-Ard, J.J.; Mhuantong, W.; Stadler, M.; Kobmoo, N. Intragenomic variation in nuclear ribosomal markers andits implication in species delimitation, identification and barcoding in fungi. Fungal Biol. Rev. 2022, 42, 1–33. [Google Scholar] [CrossRef] [Scilit]
  30. Fuoad, A.S.; Badr, A.; Faried, A.; Ellmouni, F.Y. Systematic implications of morphological traits variation and rbcL sequencepolymorphism on inter-species relationships of the subtribe Plucheinae (tribe Inuleae—Asteraceae). Egypt. J. Bot. 2024, 64, 358–368. [Google Scholar] [CrossRef] [Scilit]
  31. Letsiou, S.; Madesis, P.; Vasdekis, E.; Montemurro, C.; Grigoriou, M.E.; Skavdis, G.; Moussis, V.; Koutelidakis, A.E.; Tzakos, A.G. DNA barcoding as a plant identification method. Appl. Sci. 2024, 14, 1415. [Google Scholar] [CrossRef] [Scilit]
  32. Liu, Y.; Zhang, M.; Chen, X.; Chen, X.; Hu, Y.; Gao, J.; Pan, W.; Xin, Y.; Wu, J.; Du, Y.; et al. Developing an efficient DNA barcoding system to differentiate between Lilium species. BMC Plant Biol. 2021, 21, 465. [Google Scholar] [CrossRef] [Scilit]
  33. Mahadani, P.; Dasgupta, M.; Vijayan, J.; Kar, C.S.; Ray, S. DNA Barcoding in Plants: Past, Present, and Future. In Plant Genomicsfor Sustainable Agriculture; Singh, R.L., Mondal, S., Parihar, A., Singh, P.K., Eds.; Springer Nature: Singapore, 2022; pp. 331–350. [Google Scholar] [CrossRef] [Scilit]
  34. Gillespie, E.L.; Kron, K. Molecular phylogenetic relationships and a revised classification of the subfamily Ericoideae (Ericaceae). Mol. Phylogenet. Evol. 2010, 56, 343–354. [Google Scholar] [CrossRef] [Scilit]
  35. Gillespie, E.L.; Kron, K.A. Molecular phylogenetic relationships and morphological evolution within the tribe Phyllodoceae (Ericoideae, Ericaceae). Syst. Bot. 2013, 38, 752–763. [Google Scholar] [CrossRef] [Scilit]
  36. Hileman, L.C.; Vasey, M.C.; Parker, V.T. Phylogeny and biogeography of the Arbutoideae (Ericaceae): Implications for theMadrean–Tethyan hypothesis. Syst. Bot. 2001, 26, 131–143. [Google Scholar]
  37. Farahat, E.A.; Tashani, A.F.; Mahmoud, A.R. The sensitivity and response of the threatened endemic shrub Arbutus pavarii tocurrent and future climate change. BMC Ecol. Evol. 2025, 25, 36. [Google Scholar] [CrossRef] [Scilit]
  38. Čandek, K.; Kuntner, M. DNA barcoding gap: Reliable species identification over morphological and geographical scales. Mol.Ecol. Resour. 2015, 15, 268–277. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  39. Jin, D.-P.; Sim, S.; Park, J.-W.; Choi, J.-E.; Yoon, J.; Lim, C.E.; Kim, M.-H. Identification of the plant family Caryophyllaceae inKorea using DNA barcoding. Plants 2023, 12, 2060. [Google Scholar] [CrossRef] [Scilit]
  40. El-Banhawy, A.; Acedo, C.; Qari, S.; Elkordy, A. Molecular identification and phylogenetic placement of Rosa arabica Crép. (Rosaceae), a critically endangered plant species. Life 2020, 10, 335. [Google Scholar] [CrossRef] [Scilit]
  41. El-Banhawy, A.; Qari, S.H.; Alrehaili, M. DNA barcoding and tribal placement of Forsskaolea tenacissima L. (Urticaceae) in western Saudi Arabia: Insights from rbcL and ITS DNA markers. Egypt J. Bot. 2024, 64, 290–300. [Google Scholar] [CrossRef] [Scilit]
  42. McLamb, F.; Vazquez, A.; Olander, N.; Vasquez, M.F.; Feng, Z.; Malhotra, N.; Bozinovic, L.; Najera Ruiz, K.; O’Connell, K.; Stagg, J.; et al. Comparative three-barcode phylogenetics and soil microbiomes of planted and wild Arbutus strawberry trees. Plant Direct 2025, 9, e70078. [Google Scholar] [CrossRef] [Scilit]
  43. Rose, J.P.; Kleist, T.J.; Löfstrand, S.D.; Drew, B.T.; Schönenberger, J.; Sytsma, K.J. Phylogeny, historical biogeography, and diversification of angiosperm order Ericales suggest ancient Neotropical and East Asian connections. Mol. Phylogenet. Evol. 2018, 122, 59–79. [Google Scholar] [CrossRef] [Scilit]
  44. Degnan, J.H.; Rosenberg, N.A. Gene tree discordance, phylogenetic inference and the multispecies coalescent. Trends Ecol. Evol. 2009, 24, 332–340. [Google Scholar] [CrossRef] [Scilit]
  45. Maddison, W.P. Gene trees in species trees. Syst. Biol. 1997, 46, 523–536. [Google Scholar] [CrossRef]
  46. Feliner, G.N.; Rosselló, J.A. Better the devil you know? Guidelines for insightful utilization of nrDNA ITS in species-level evolutionary studies in plants. Mol. Phylogenet. Evol. 2007, 44, 911–919. [Google Scholar] [CrossRef] [Scilit]
  47. Birky, C.W. The inheritance of genes in mitochondria and chloroplasts: Laws, mechanisms, and models. Annu. Rev. Genet. 2001, 35, 125–148. [Google Scholar] [CrossRef] [Scilit]
  48. Lopes, L.; Sá, O.; Pereira, J.A.; Baptista, P. Genetic diversity of Portuguese Arbutus unedo L. populations using leaf traits andmolecular markers: An approach for conservation purposes. Sci. Hortic. 2012, 142, 57–67. [Google Scholar] [CrossRef] [Scilit]
  49. Wiens, J.J.; Morrill, M.C. Missing data in phylogenetic analysis: Reconciling results from simulations and empirical data. Syst. Biol. 2011, 60, 719–731. [Google Scholar] [CrossRef] [Scilit]
  50. Wiens, J.J. Missing data, incomplete taxa, and phylogenetic accuracy. Syst. Biol. 2003, 52, 528–538. [Google Scholar] [CrossRef] [PubMed]
  51. Lemmon, E.M.; Lemmon, A.R. High-throughput genomic data in systematics and phylogenetics. Annu. Rev. Ecol. Evol. Syst. 2013, 44, 99–121. [Google Scholar] [CrossRef] [Scilit]
  52. Funk, W.C.; McKay, J.K.; Hohenlohe, P.A.; Allendorf, F.W. Harnessing genomics for delineating conservation units. Trends Ecol.Evol. 2012, 27, 489–496. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  53. Fujita, M.K.; Leaché, A.D.; Burbrink, F.T.; McGuire, J.A.; Moritz, C. Coalescent-based species delimitation in an integrativetaxonomy. Trends Ecol. Evol. 2012, 27, 480–488. [Google Scholar] [CrossRef] [Scilit]
  54. Padial, J.M.; Miralles, A.; De La Riva, I.; Vences, M. The integrative future of taxonomy. Front. Zool. 2010, 7, 16. [Google Scholar] [CrossRef] [Scilit]
  55. Dayrat, B. Towards integrative taxonomy. Biol. J. Linn. Soc. 2005, 85, 407–415. [Google Scholar] [CrossRef] [Scilit]
  56. Doyle, J.J.; Doyle, J.L. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 1987, 19, 11–15. [Google Scholar]
  57. Jamaludin, A.A.; Bakar, A.A.; Shahimi, S.; El-Banhawy, A. An Effective CTAB Method for Isolation of CpDNA from Silica-DriedFrond Tissues of Several Tree Fern Species from Peninsular Malaysia. In Proceedings of the International Conference on Education, Mathematics and Science 2020, Tg. Malim, Malaysia, 12 December 2020; Volume 23, p. 1797. [Google Scholar] [CrossRef]
  58. Doyle, J.J.; Doyle, J.L. Isolation of plant DNA from fresh tissue. Focus 1990, 12, 1–3. [Google Scholar]
  59. Porebski, S.; Bailey, L.G.; Baum, B.R. Modification of a CTAB DNA extraction protocol for plants containing high polysaccharideand polyphenol components. Plant Mol. Biol. Rep. 1997, 15, 8–15. [Google Scholar] [CrossRef] [Scilit]
  60. CBOL Plant Working Group; Hollingsworth, P.M.; Forrest, L.L.; Spouge, J.L.; Hajibabaei, M.; Ratnasingham, S.; Van Der Bank, M.; Chase, M.W.; Cowan, R.S.; Erickson, D.L.; et al. A DNA barcode for land plants. Proc. Natl. Acad. Sci. USA 2009, 106, 12794–12797. [Google Scholar] [CrossRef] [Scilit]
  61. 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] [Scilit]
  62. Binnoubah, A.; Hamdy, R.; Ragab, O.G.; El-Taher, A.M.; Abou El-Yazied, A.; Safhi, F.A.; Elzilal, H.A.; Althobaiti, A.T.; Alshamrani, S.M.; Abd El Moneim, D.; et al. Anatomical and molecular identification of ornamental plant Ficus, L. species. Phyton 2023, 92, 1329–1347. [Google Scholar] [CrossRef] [Scilit]
  63. El-Banhawy, A.; Al-Juhani, W. DNA Barcoding and Phylogeny of Phlomis aurea (Lamiaceae) Endemic to Sinai peninsula, Egypt. Pak. J. Bot. 2019, 51. Available online: http://pakbs.org/pjbot/paper_details.php?id=7751 (accessed on 26 December 2025). [CrossRef] [Scilit]
  64. Hall, T.A. BioEdit: A user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. NucleicAcids Symp. Ser. 1999, 41, 95–98. [Google Scholar]
  65. Thompson, J.D.; Higgins, D.G.; Gibson, T.J. CLUSTAL W: Improving the sensitivity of progressive multiple sequence alignmentthrough sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994, 22, 4673–4680. [Google Scholar] [CrossRef] [Scilit]
  66. R Core Team, R. 2016. Available online: https://www.r-project.org/ (accessed on 15 February 2026).
  67. RStudio Team. RStudio: Integrated Development for R; Rstudio: Boston, MA, USA, 2015; Available online: https://posit.co/download/rstudio-desktop/ (accessed on 15 February 2026).
  68. Meier, R.; Shiyang, K.; Vaidya, G.; Ng, P.K.L. DNA barcoding and taxonomy in Diptera: A tale of high intraspecific variabilityand low identification success. Syst. Biol. 2006, 55, 715–728. [Google Scholar] [CrossRef] [Scilit]
  69. Nylander, J.A.A. MrModeltest; Uppsala University: Uppsala, Sweden, 2004; Available online: https://mybiosoftware.com/mrmodeltest-2-3-program-selecting-dna-substitution-models-paup.html (accessed on 15 February 2026).
  70. Swofford, D.L. PAUP. Sinauer; 2002. (Phylogenetic Analysis Using Parsimony (and Other Methods). Available online: https://paup.phylosolutions.com// (accessed on 15 February 2026).
  71. Ronquist, F.; Teslenko, M.; van der Mark, P.; Ayres, D.L.; Darling, A.; Höhna, S.; Larget, B.; Liu, L.; Suchard, M.A.; Huelsenbeck, J.P. MrBayes:Efficient Bayesian phylogenetic inference and model choice across a large model space. Syst. Biol. 2012, 61, 539–542. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  72. Rambaut, A.; Suchard, M.A.; Xie, D.; Drummond, A.J. Tracer. 2014. Available online: http://beast.bio.ed.ac.uk/Tracer (accessed on 15 February 2026).
  73. Rambaut, A. FigTree. 2012. Available online: http://tree.bio.ed.ac.uk/software/figtree (accessed on 15 February 2026).
  74. Paradis, E.; Claude, J.; Strimmer, K. APE: Analyses of Phylogenetics and Evolution in R language. Bioinformatics 2004, 20, 289–290. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  75. Kolde, R. pheatmap: Pretty Heatmaps, version 1.0.13; CRAN: Vienna, Austria, 2025. [Google Scholar] [CrossRef] [Scilit]
  76. Wickham, H. ggplot2: Elegant Graphics for Data Analysis, 2nd ed.; Springer: Cham, Switzerland, 2016. [Google Scholar] [CrossRef] [Scilit]
  77. Kassambara, A. ggpubr:‘ggplot2’Based Publication Ready Plots; R Package Version 0.4; R Core Team: Vienna, Austria, 2020. [Google Scholar]
  78. Kassambara, A.; Mundt, F. factoextra: Extract and Visualize the Results of Multivariate Data Analyses. 2020. Available online: https://github.com/kassambara/factoextra (accessed on 15 February 2026).
  79. Viscosi, V.; Cardini, A. Correction: Leaf Morphology, Taxonomy and Geometric Morphometrics: A Simplified Protocol for Beginners. PLoS ONE 2012, 7. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Soil Stoichiometry-Regulated Microbial Carbon Use Efficiency Between Rhizosphere and Bulk Soils in the Temperate Forests of Northeastern China
Previous
Antimutagenic and Anticarcinogenic Evidence of the Genus Ficus L.
Next