INTRODUCTION
The genus Pentactina Nakai (Tribe Spiraeeae; Amygdaloideae; Rosaceae) was recognized as a monotypic genus endemic to the central part of the Korean Peninsula (Nakai, 1917). The type species of the genus, Pentactina rupicola Nakai, has an extremely restricted geographic distribution and limited populations in the Kumgangsan Mountains, with an estimated occupancy range of 24–57.5 km2 (Lee and Hong, 2011; Chang et al., 2016). Due to its rarity and restricted occurrence, P. rupicola has been assessed as a critically endangered (CR) species on the IUCN Red List (Kim et al., 2016). As a rare monotypic genus, the taxonomic treatment of the genus Pentactina has been controversial since Nakai’s original publication (Nakai, 1917; Lee and Hong, 2011), with some taxonomic reviews recognizing this genus as a synonym of the genus Spiraea L. (Hutchinson, 1964; Kalkman, 2004). Despite limited accessions of P. rupicola, a series of palynological (Lee et al., 1993; Song et al., 2017), phylogenetic (Lee and Hong, 2011), and morphological studies (Song et al., 2020a, 2020b) supported the taxonomic treatment of Pentactina as a distinct genus. Meanwhile, a recent publication proposed a new status for Pentactina schlothauerae (Ignatov & Vorosch.) Jakubov, a species endemic to the Badzhal Range of the Russian Far East, challenging the recognition of Pentactina as a monotypic genus (Yakubov, 2014).
Pentactina schlothauerae was originally published as Spiraea schlothauerae Ignatov & Vorosch. Voroshilov and Ignatov (1987) published this new species as a taxon in the genus Spiraea, and described that this species has an affinity to the subgenus Protospiraea Nakai with paniculate inflorescences but also has obvious differences in the shape of leaves, length of stamens, and width of petals. However, Yakubov (2014) found that this species is much similar to P. rupicola in terms of morphological characteristics, and published a new combination of P. schlothauerae, transferring it from the genus Spiraea to Pentactina. He suggested the shape of leaves and the distribution as the key characteristics for identifying the two Pentactina species. Following this taxonomic treatment, Chung et al. (2017) excluded the genus Pentactina from the list of Korean vascular endemic genera. A biochemical study found differences in the qualitative composition of phenolic compounds, supporting the taxonomic distinction between P. schlothauerae and Spiraea species (Kostikova, 2018).
Despite these developments, the taxonomic position of P. schlothauerae has not been evaluated using molecular phylogenetic analyses. In this study, we determined genetic sequences of the nuclear ribosomal DNA (nrDNA) regions of P. schlothauerae accessions, and assembled a complete chloroplast (cp) genome of one of these accessions. For the first time, we assessed the taxonomic position of P. schlothauerae and the status of the genus Pentactina within the tribe Spiraeeae of Rosaceae through the molecular phylogenetic and comparative genomic analyses, by utilizing both nuclear and chloroplast DNA sequences.
MATERIALS AND METHODS
Sample collection and DNA isolation
A total of three accessions of P. schlothauerae were collected from the field, with each accession collected from a distinct population separated by a distance of ≥500 m. The leaf materials of collected accessions were silica gel-dried and deposited with specimen vouchers in the Herbarium of the Botanical Garden-Institute of the Far Eastern Branch of the Russian Academy of Sciences (FEB RAS) (VBGI) (Table 1). Total genomic DNA (gDNA) of each accession was isolated from silica gel-dried leaf tissues using DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. The nucleotide sequences of P. rupicola, some closely related taxa in the tribe Spiraeeae, and outgroup taxa were retrieved from GenBank for subsequent phylogenetic analyses (Appendix 1).
DNA amplification and sequencing
To infer the biparentally inherited nrDNA-based phylogenetic position and relationships of the Pentactina species, internal transcribed spacer (ITS) regions of nrDNA were amplified using polymerase chain reaction (PCR). The reaction included 0.2 μM of modified ITS5 (5′-GGA AGG AGA AGT CGT AAC AAG G-3′) (White et al., 1990; Sang et al., 1995) as the forward primer, 0.2 μM of ITS4 (5′-TCC TCC GCT TAT TGA TAT GC-3′) (White et al., 1990) as the reverse primer, and Inclone Taq DNA polymerase kit (Inclone Biotech Co., Ltd., Seoul, Korea) with 2 mM MgCl2, 0.2 mM of each dNTP, and 1.25 units of Taq polymerase. The condition for PCR amplification was (1) 95°C initial denaturation for 2 min, (2) 35 cycles of 95°C denaturation for 20 s, 54°C annealing for 40 s, and 72°C extension for 1 min, and (3) 72°C final extension for 5 min. The amplified PCR products were purified using Inclone Gel & PCR purification kit (Inclone Biotech Co., Ltd.) following the manufacturer’s instructions. The purified PCR products were Sanger sequenced by GenoTech Corp. (Daejeon, Korea).
To infer the uniparentally inherited cpDNA-based phylogenetic position and relationships of the Pentactina species, genomic sequence data for an accession of P. schlothauerae (accession ID: 248) were generated using next-generation sequencing to assemble a complete cp genome. The sequencing library was prepared from gDNA by random fragmentation, followed by 5′ and 3′ adapter ligation, using TruSeq Nano DNA kit (Illumina, Inc., San Diego, CA, USA). The fragmented library was subjected to size selection and sequenced using the Illumina HiSeq X Ten platform (Illumina, Inc.) by Macrogen, Inc. (Seoul, Korea).
Chloroplast genome assembly and structural annotation
Genomic sequence reads were trimmed to remove sequencing adapters, PCR dimers, and low-quality sequences using BBDuk v38.26 (Bushnell, 2014). Trimmed sequence reads were de novo assembled into a complete circular cp genomic sequence using NOVOPlasty v4.2 (Dierckxsens et al., 2017). Genomic structure was annotated using GeSeq (Tillich et al., 2017), incorporating referenced annotation using BLAT alignment (Kent, 2002) based on the reference cp genome of P. rupicola (NCBI RefSeq accession: NC_016921) (Kim and Kim, 2016), and ab initio annotation using Chloë v0.1.0 (Small and Castleden, 2020) and tRNAscan-SE v2.0.7 (Chan et al., 2021). A graphical map of the genomic structure was generated using OGDRAW v1.3.1 (Greiner et al., 2019).
Phylogenetic analyses
The genetic sequences of nrDNA-ITS regions were aligned using Geneious R 10.2.3 (Biomatters, Ltd., Auckland, New Zealand), whereas the coding DNA sequences (CDSs) of cp genomic sequences were aligned using MAFFT v7.508 (Katoh and Standley, 2013) for subsequent phylogenetic analyses. Each sequence alignment was gap-coded to minimize phylogenetic bias from large gaps. The phylogenetic position and relationships of Pentactina species were inferred by maximum likelihood (ML) inferences using IQ-TREE v1.6.12 (Nguyen et al., 2015), with the TIM+G4 substitution model for the nrDNA-ITS alignment and the TVM+I+G4 model for the cp genome-wide CDS alignment respectively, which were selected using ModelFinder (Kalyaanamoorthy et al., 2017). To evaluate the robustness of the inferred phylogenetic trees, branch support in each tree was assessed by a bootstrap approximation with 5,000 bootstrap replicates using UFBoot2 (Hoang et al., 2018). The phylogenetic position and relationships of Pentactina species were additionally inferred by Bayesian inference using MrBayes v3.2.7 (Ronquist et al., 2012), with the GTR+G4 substitution model for the nrDNA-ITS alignment and the GTR+I+G4 model for the cp genome-wide CDS alignment respectively selected from ModelFinder. For Bayesian inference, four Markov chain Monte Carlo chains were run for 10,000,000 generations, with sampling of parameters and trees every 1,000 generations.
Comparative genomic analyses
Genomic structures, nucleotide diversity (π), and genetic distances (D) of cp genomes of Pentactina and Spiraea species were compared to assess the taxonomic status of P. schlothauerae. The cp genomic synteny was analyzed using Mauve v2.4 (Darling et al., 2004), and the gene content of cp genomes was compared based on their structural annotations. Sliding window analyses were performed with the 800-bp window length and 200-bp step size to compare pairwise nucleotide diversity of the cp genomes of four species pairs: (1) P. schlothauerae and P. rupicola, (2) P. schlothauerae and Spiraea japonica L.f., (3) P. schlothauerae and Spiraea thunbergii Siebold ex Blume, and (4) S. japonica and S. thunbergii. Whole-genome nucleotide diversity and genetic distances were assessed through pairwise comparisons between species within the genera Pentactina and Spiraea.
Morphological characteristics
The morphological characteristics of P. schlothauerae were investigated using herbarium specimens stored in the Herbarium of the Federal Scientific Center of the East Asia Terrestrial Biodiversity FEB RAS (VLA; 11 vouchers) and VBGI (seven vouchers) (Appendix 2), and 25 living samples cultivated in the Botanical Garden-Institute FEB RAS during the flowering and fruiting periods. Detailed morphologies were investigated using a Stemi 2000-C Stereo Microscope (Carl Zeiss, Oberkochen, BW, Germany). The quantitative traits on the microscope were measured using AxioVision v4.8 (Carl Zeiss). The numerical and dimensional characteristics are presented as ranges or the maximum of recorded values.
RESULTS
Chloroplast genome assembly and structural annotation
The whole-genome sequence reads of P. schlothauerae were assembled into a complete circular cp genome, with the 157,043-bp sequence length and the 4,116× depth of coverage (Appendix 3). The cp genome of P. schlothauerae was found to have a typical quadripartite structure, in which a large single-copy region (85,157 bp) and a small single-copy region (18,894 bp) were separated by two inverted repeats (26,496 bp each). For gene content, 113 unique genes were identified in the cp genome, consisting of 79 protein-coding, 30 tRNA, and four rRNA genes (Appendix 4).
Phylogenetic inference on Pentactina schlothauerae
Both phylogenetic inferences using nrDNA ITS and cp genomes indicated a sister relationship between P. schlothauerae and P. rupicola, which was strongly supported in the cp-genome phylogenetic tree (100% bootstrap support [BS] and 1.0 posterior probability [PP]) and moderately supported in the nrDNA-ITS tree (77% BS and 0.99 PP), with identical tree topologies from ML and Bayesian inferences from each dataset (Figs. 1, 2). These two sister taxa were resolved as a monophyletic group representing the genus Pentactina, which have diverged into a distinct phylogenetic lineage with long branch lengths, and each of the other genera in the tribe Spiraeeae was monophyletic in both trees. The genus Sibiraea Maxim. was inferred to be a close relative of Pentactina in the nrDNA-ITS tree with a low branch support value (≤50% BS) (Fig. 1), whereas a clade including the genera Spiraea, Sibiraea, Kelseya (S.Watson) Rydb. and Petrophytum (Nutt.) Rydb. was inferred to be a close relative of Pentactina in the cp-genome tree with high branch support values (100% BS and 1.0 PP) (Fig. 2).
Comparative genomic analyses between the genera Pentactina and Spiraea
Synteny analysis indicated that a single synteny block was shared in the cp genomes of the genera Pentactina and Spiraea (Appendix 5A). Gene content and gene order were highly conserved among the cp genomes of two genera (Appendices 4, 5B). The only difference was that infA was intact in the cp genomes of P. schlothauerae, Spiraea salicifolia Elliott and Spiraea tomentosa L., whereas it was pseudogenized in the other cp genomes, being infA content homoplasious.
In sliding window analyses, nucleotide diversity between the two Pentactina species, and between the two Spiraea species was lower (median π = 0.0038 in both) than nucleotide diversity between P. schlothauerae and one of the two Spiraea species (median π = 0.0100 in both) (Fig. 3A, B). In pairwise whole-genome comparisons on nucleotide diversity and genetic distances, both π and D were 0.005 between P. schlothauerae and P. rupicola, and π was 0.001–0.006 and D was 0.001–0.007 between two Spiraea species (Fig. 3C). However, both π and D were much higher in pairs between the genera Pentactina and Spiraea (π = 0.011–0.012; D = 0.016–0.019).
DISCUSSION
The phylogenetic analyses in this study inferred that P. schlothauerae w as a sister species o f P. rupicola, and phylogenetically distinct from Spiraea species (Figs. 1, 2), corroborating the Yakubov’s taxonomic revision (2014), which moved P. schlothauerae from the genus Spiraea to Pentactina based on morphological characters. Given that both nrDNA-ITS regions and cp genomes efficiently delimited generic boundaries with long branch lengths and high BS values within the tribe Spiraeeae, an independent and isolated lineage of Pentactina species in the phylogenetic analyses implied a robust taxonomic status of the genus Pentactina. This distinct phylogenetic position of the genus Pentactina corresponds to the unique palynological, morphological, and biochemical characteristics of Pentactina, which are different from those of other genera in the tribe Spiraeeae (Lee et al., 1993; Kostikova, 2018; Song et al., 2020a, 2020b). However, in the nrDNA-ITS phylogenetic tree, the monophyly of the genus Pentactina was moderately supported (77% BS and 0.99 PP), and its phylogenetic position within the tribe Spiraeeae was weakly supported (Fig. 1). In a previous phylogenetic study using nrDNA-ITS, the genus Petrophytum was suggested as a close relative of Pentactina, whereas Sibiraea was a close relative in this study, although both were weakly supported (Lee and Hong, 2011). Although the phylogenetic analysis of chloroplast genomes resolved robust genus-level phylogenetic relationships in the tribe Spiraeeae (Fig. 2), cp DNA may have some systematic problems such as a low rate of evolution, rare recombination, and uniparental inheritance (Harris and Ingram, 1991). Given that molecular data from multiple nrDNA regions may improve the resolution of phylogenetic relationships (Cummings et al., 1995), further studies with improved datasets including multi-locus nuclear markers can elucidate the robust phylogenetic position of the genus Pentactina and its relationship with other genera in the tribe Spiraeeae.
Comparative cp genomic analyses also agreed with the taxonomic status of P. schlothauerae, indicating lower genetic divergence with P. rupicola than with Spiraea species, although no significant structural variation was found between the cp genomes of the genera Pentactina and Spiraea (Fig. 3, Appendices 4, 5). The only difference in genomic structure was the pseudogenization of infA, which appeared to be homoplasious within the tribe Spiraeeae (Appendix 4). It has been previously demonstrated that the loss of infA occurs independently in diverse lineages of flowering plants (Millen et al., 2001). In both genome-wide sliding window analyses and whole-genome comparisons, nucleotide diversity and genetic distances were lower in congeneric pairs than between the genera Pentactina and Spiraea (Fig. 3). Additionally, each chloroplast genome of P. schlothauerae and P. rupicola was almost equally divergent from those of Spiraea species (Fig. 3C). These results did not support the taxonomic position of P. schlothauerae within the genus Spiraea. Given the similar levels of nucleotide diversity and genetic distances between P. schlothauerae and P. rupicola to those between Spiraea species, P. schlothauerae can be recognized as an independent species that is clearly distinguished from P. rupicola by their genetic backgrounds. Differences in morphology (Nakai, 1917; Voroshilov and Ignatov, 1987; Yakubov, 2014), habitat (Nakai, 1917; Voroshilov and Ignatov, 1987; Yakubov, 2014), and karyotype (Sax, 1931; Darlington and Wylie, 1956; Probatova et al., 2017) also support the distinct taxonomic identity of P. schlothauerae from P. rupicola (Fig. 4, Table 2). Given that polyploidy is not frequently observed in the tribe Spiraeeae, except in its largest genus Spiraea (Sax, 1931; Dickinson et al., 2007), tetraploid P. schlothauerae may have recently diverged into a distinct apoendemic lineage from the ancestral diploid lineage, occupying a new habitat in the Badzhal Range (Stebbins and Major, 1965).
TAXONOMIC TREATMENT
We compiled the first detailed and illustrated description of P. schlothauerae, the morphology of which has been published only in the protologue so far (Voroshilov and Ignatov, 1987) (Figs. 5, 6). Empowered with detailed morphological characteristics of P. schlothauerae, the two Pentactina species could be diagnosed, although the morphological characteristics of P. rupicola were referenced from its protologue (Nakai, 1917) due to its limited accessibility and availability.
Pentactina schlothauerae (Ignatov & Vorosch.) Jakubov in Komarovskie Chteniya (Vladivostok) 62: 232, 2014; Spiraea schlothauerae Ignatov & Vorosch. in Byull. Moskovsk. Obshch. Isp. Prir., Otd. Biol. 92: 132, 1987.—TYPE: Russia. Khabarovsk Krai, the Badzhal Range, Valley of the Vstrechnoy stream, 1,500 m above sea level, in the canopy of Pinus pumila, 28 Jul 1984, S. D. Schlothauer (holotype: MHA).
Plant sprawling geoxyl shrub, up to 30 (dimensions of cultivated plants, 85) cm tall (Fig. 5B). Shoots numerous, thin, with red-brown bark, erect with a straight or inclined apex, or inclined; inclined shoots up to 50 (dimensions of cultivated plants, 110) cm long. Buds lanceolate, 2 times longer than the petiole; outer bud scales glabrous; inner scales pubescent with white hairs. Young annual shoots more or less pubescent with erect hairs. Leaves alternate, perpendicularly deflected from the stem, slightly arched, on short petioles 0.1–0.3 cm long (Fig. 5C); blades lanceolate, up to 4 (dimensions of cultivated plants, 7) cm long, 1 (dimensions of cultivated plants, 2.1) cm wide, bases rounded; apices acute; margins in the upper third with a few acute teeth; along the edge and on the lower side with scattered pubescence from long appressed silky hairs. Inflorescences apical, up to 5 (dimensions of cultivated plants, 18) cm long, 3 (dimensions of cultivated plants, 8) cm wide, paniculate, narrow pyramidal-ovate; the branches of the inflorescence simple, deflected almost 90°, with a pubescent axis (Fig. 5A). Pedicels short, 0.15–0.2 mm long, with a short linear bract under the calyx, 0.5–2 mm long (Fig. 6C, D). Calyx naked on the outside; the bottom of the hypanthium covered with long and thin hairs; nectar rings entire, with a smooth or slightly wavy, thickened edge, 0.10–0.13 mm in height (Fig. 6B); calyx lobes triangular, slightly longer than the tube, deflected and pressed against the tube during flowering, the length 1/4–1/3 longer than the width, ca. 1.2 mm long, 0.9 mm wide, purple colored along the edge and at the apex, on the inside at the apex with hairs (Fig. 6C). Petals white, 3.8–5 mm long, 0.5–0.7 mm wide, linear, narrowed towards the base, rounded at the apex, and occasionally notched. Stamens 20, 2.5–3 times shorter than the petals; filaments thin, white, 1.5–2 mm long; anthers light purple, 0.3 mm long, 0.4 mm wide, violet after opening, then brown; pollen white, dried pollen grains elliptical. Carpel 5; ovaries ovoid, glabrous, not fused with the hypanthium; styles longer than the ovaries, white, slightly widened at the apex, extending from the ventral side of the ovary (Fig. 6B, C). Fruits dry cyclic apocarpous five-carpel follicles, 1.8–1.9 mm in diameter, with dorso-ventral dehiscence of the carpels; follicles light brown, lanceolate, 1.8 mm long, 0.65 mm wide; styles long, ca. 0.9 mm long, usually break off, their basal parts remain in the form of a spout, bent outward almost horizontally (Fig. 6A). Seeds lanceolate, flattened, light brown, shiny, 1.3 mm long, 0.3 mm wide, with a cellular sculpture.
Distribution and habitat: Pentactina schlothauerae is narrowly endemic to the Badzhal Range in the Russian Far East, and inhabits rocky screes or slopes near rivers or in coniferous forests. The distribution and habitat of P. schlothauerae are clearly distinguished from those of P. rupicola, which is endemic to the Kumgangsan Mountains of North Korea and inhabits the clefts of rocks or rocky outcrops (Fig. 4, Table 2).
Diagnosis: The morphological characteristics of P. schlothauerae were compared with those of P. rupicola provided in the protologue (Nakai, 1917) to identify the distinctive features of both species. The two Pentactina species show large similarities in their general morphological characteristics. A noticeable morphological difference between them was observed in the shape of the leaf blades (see Table 2). Additionally, several minor differences were observed. The bract in P. schlothauerae is attached at the end of the pedicel under the calyx, but not in the middle of the pedicel as in P. rupicola. In P. schlothauerae, bract lengths vary over a larger range and include the range of variation for P. rupicola, although the characterization of P. rupicola may have been observed in limited materials. The ovaries of P. rupicola are reddish as noted in the protologue, whereas those of P. schlothauerae are light green. A further detailed study of P. rupicola will complement its morphological characteristics and allow the comparison to reveal more distinctive features between the two Pentactina species.
Notes: A study of P. schlothauerae on living materials indicated that the color of floral structures can be influenced by lighting conditions at the growing sites. In much illuminated areas, the nectar rings of P. schlothauerae are light pink, and the tips of the sepals are dark pink. In shady places, the pink color of these flower elements may be weakly expressed or absent (see Fig. 6B, C).
CONCLUSIONS
In this study, we identified the taxonomic status of P. schlothauerae as a distinct species within the genus Pentactina, and confirmed the status of genus Pentactina as an independent genus-level lineage from Spiraea or other related genera in the tribe Spiraeeae, based on the phylogenetic and comparative genomic analyses, and previous studies. Although the findings do not support Pentactina as a monotypic genus endemic to Korea, this genus remains important in biodiversity as a distinct phylogenetic lineage of which the two species are rare and narrowly endemic to restricted areas. Due to the limited availability of P. rupicola, which was recognized as the only species in the genus, the systematics, physiology, development and conservation of Pentactina remain poorly understood. As a result of the close relationship between the two Pentactina species, which are both rare and narrowly endemic, further understanding of their life-history traits, levels of genetic diversity, population differentiation, demographic history, and threats will allow us to develop conservation strategies for this genus.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation 
Print 
