Taxonomic revision of Paris kwantungensis (Melanthiaceae) based on morphological and molecular evidence
Article information
Abstract
Paris kwantungensis was proposed initially as a monophyletic entity by Miau in 1982, but its taxonomic status remained contentious. We recently collected and studied this taxon at its original locality in Nujiang, Yunnan province, China. A molecular phylogenetic analysis and morphological observations of P. kwantungensis and related species support its recognition as a distinct species. Phylogenetic evidence shows that P. kwantungensis is a sister to P. polyphylla var. chinensis. Morphologically, P. kwantungensis is distinguished by its petals being nearly twice as long as its sepals (as opposed to petals shorter than sepals), reflexed sepals, and a globose capsule with numerous surface granules. This study provides a comprehensive species description of P. kwantungensis and compares it with related species in the genus Paris. The combination of morphological traits and molecular analyses, particularly of the internal transcribed spacer region sequences, clearly establishes P. kwantungensis as a unique and independent species within the genus.
INTRODUCTION
Paris L., the largest genus in the family Melanthiaceae, is widely distributed across Eurasia, with its center of diversity in the Himalayan-Hengduan Mountain region (Li, 1984, 1988). Notably, China harbors more than 30 species of Paris, nearly half of which are endangered (Liu et al., 2017; Fan et al., 2020). Previous studies have shown that Paris species exhibit relatively remarkable morphological diversity, characterized by simple, net-veined leaves, actinomorphic flowers, and variable fruit types, including rounded or angular berries and berry-like capsules.
The morphological diversity of Paris has led to considerable taxonomic complexity within the genus. For example, the P. polyphylla Smith complex exhibits substantial morphological variation and has been subdivided into multiple varieties (Wu and Raven, 2013), including P. polyphylla var. polyphylla, P. polyphylla var. chinensis (Franch.) Hara, P. polyphylla var. yunnanensis (Franch.) Hand.-Mazz., P. polyphylla var. nana H. Li, P. polyphylla var. alba H. Li et R. J. Mitchell, and P. polyphylla var. stenophylla Franch. (Li, 1988). However, this classification has long been debated. Some species, such as P. kwantungensis Miau, were initially described as distinct based on unique stamen morphology (Miau, 1982). Nevertheless, Takhtajan treated P. kwantungensis as a synonym of Daiswa polyphylla (Smith) Rafinesque (≡ P. polyphylla var. polyphylla), relying on R. H. Miau’s morphological descriptions in his revision of Daiswa (Takhtajan, 1983). This circumscription was later supported by Ji and by The Plant List (Ji et al., 2007, 2013), which listed P. kwantungensis as a synonym of P. polyphylla var. polyphylla. However, this treatment was not universally accepted. After reexamining the holotype, Chen and Liang (1995) argued that P. kwantungensis should be regarded as a variety of P. polyphylla var. polyphylla rather than a synonym, and accordingly revised its accepted name to P. polyphylla var. kwantungensis (R. H. Miau) S. C. Chen & S. Y. Liang (Chen and Liang, 1995). Interestingly, Ji later proposed that P. kwantungensis is conspecific with P. polyphylla var. chinensis (Franch.) H. Hara (≡ P. chinensis Franch.) (Ji, 2020).
In summary, the classification of P. kwantungensis remains highly controversial because it relies almost exclusively on morphological evidence and lacks molecular support. Therefore, it is urgent to conduct molecular studies using DNA markers to clarify its taxonomic status. To address this issue, we integrated morphological analysis with molecular phylogenetics (internal transcribed spacer [ITS]) to reevaluate the status of P. kwantungensis. This study provides a more robust classification framework and contributes to the targeted conservation of this taxon.
MATERIALS AND METHODS
Sample collection and molecular analysis
For 14 Paris species, 39 fresh samples were collected from various regions of China (Table 1). For each species, at least two samples, each comprising approximately 1.0 g of fresh leaves, were collected and dried with silica gel. Professor Baozhong Duan of Dali University authenticated all samples. Voucher specimens were deposited in the herbarium at Dali University.
Specimens and associated GenBank accession numbers for the taxa used in the genetic relationship analysis.
Total genomic DNA was extracted from silica gel-dried material using a modified CTAB protocol (Doyle and Doyle, 1987). The nuclear ribosomal ITS was amplified in a 25 μL reaction containing 2 μL (20–100 ng) of DNA template, 8.5 μL of ddH2O, 12.5 μL of 2× Taq Plus PCR MasterMix (Beijing Aidlab Biotech Co., Beijing, China), and 1 μL each of forward and reverse primers (2.5 μmol/L). Cycling conditions followed Liu et al. (2019): an initial denaturation at 94°C for 5 min, followed by 40 cycles of 94°C for 30 s, 56°C for 30 s, and a final extension at 72°C for 45 s. PCR products were purified with polyethylene glycol (20%) and analyzed on an ABI 3730XL genetic analyzer (Applied Biosystems, Foster City, CA, USA).
Sequences were edited and aligned by eye using CodonCode Aligner (ver. 9.0, https://www.codoncode.com/index.htm). The maximum parsimony (MP) analysis was performed in MEGA X (ver. 10.2, https://www.megasoftware.net), and the maximum-likelihood (ML) bootstrap analysis was conducted with RAxML (Stamatakis, 2014). Bootstrap percentages were calculated from 1,000 replicates.
Morphology
P. Specimens of P. kwantungensis were collected from plantation areas on hill slopes in Yunnan, Guizhou, and Sichuan provinces. Morphological characteristics of P. kwantungensis and related species, such as P. polyphylla var. polyphylla and P. polyphylla var. chinensis, were recorded in both plantation areas and laboratory settings. Comparative morphological analyses were conducted using descriptions from the literature (Miau, 1982; Takhtajan, 1983; Chen and Liang, 1995; Ji et al., 2007; Ji, 2020).
RESULTS
Sequence characteristics and phylogenetic analysis
The ITS sequences used for the phylogenetic analysis included 47 newly determined sequences from Paris. Smilax lanceifolia Roxburgh and Smilax hypoglauca Bentham served as outgroups. All new sequences were deposited in GenBank (Table 1). The results showed that sequence lengths ranged from 633 to 750 bp, with 131 variable sites detected, of which 98 were parsimony-informative. The average GC content of the ITS sequences was 55.1%.
For phylogenetic inference, MP analysis was performed, which yielded eight equally parsimonious trees. These MP trees had a tree length of 479 steps, a consistency index of 0.833, and a retention index of 0.914. There were differences between the MP and ML trees (Figs. 1, 2), but both recovered the same three clades, as labeled in the ML tree (Fig. 1: clades A–C). After collapsing poorly supported nodes, the topologies of the ML and MP trees were identical. The MP and ML analyses showed that P. kwantungensis (9 accessions) was most closely related to P. polyphylla var. chinensis (4 accessions). The two species formed mutually distinct groups with strong support (Fig. 1, maximum likelihood bootstrap percentage 99; Fig. 2, maximum parsimony bootstrap percentage 95). Within this clade, P. kwantungensis clustered with other species in a large clade that included P. dunniana, P. cronquistii var. xichouensis, P. vietnamensis, and P. cronquistii var. cronquistii. The other two clades (A and B) comprised the remaining species of the genus. Interestingly, prior studies suggested that P. kwantungensis was a synonym of P. polyphylla var. polyphylla (Takhtajan, 1983; Ji, 2007). However, the molecular data presented here place P. polyphylla var. polyphylla and P. kwantungensis in distinct clades (clades B and C, respectively).
Maximum likelihood (ML) tree of Paris based on the nuclear DNA (ITS) sequences under the Kimura 2-parameter distance model. Two species of Smilax as outgroups.
Maximum parsimony (MP) tree of Paris based on the nuclear DNA (ITS) sequences. Two species of Smilax as an outgroup.
Overall, we formally propose that P. kwantungensis be treated as a distinct species rather than as a synonym or variety of P. polyphylla var. polyphylla or P. polyphylla var. chinensis, based on molecular findings and morphological evidence.
Morphological diagnostic characters
Paris kwantungensis R. H. Miau in Acta Sci. Nat. Univ. Sunyatseni. 1982: 74, 1982 (Figs. 3, 4).—TYPE: CHINA. Guangdong (广东): Xinyi (信宜), 21 Mar 1932, Chi Wang (黄志) 32199 (holotype: SYS!).
Paris kwantungensis. 1, leaf; 2, sepal; 3, the upper part of the pedicel (including petals, capsules); 4, flowering plant; 5, the open fruit, showing parietal placentas after all seeds are removed; 6, fructus; 7, rhizome and roots. Heying Zhao draws them.
A. A population. B. Leaf (left is the adaxial surface; right is the abaxial surface). C. Sepal (left is the adaxial surface; right is the abaxial surface). D. Rhizome and roots. E. Top view of the plant. F. A dehiscing fruit, seeds and removed testa. G. The open fruit, showing parietal placentas after all seeds are removed.
Herbs, perennial, glabrous, oblique or erect. Stem single, 15.5–45.5 cm tall, 0.4–1.2 cm in diameter, primarily curved, base yellowish white. Rhizome horizontal, dark brown externally, white internally, 4.3–8.2 × 1.2–3.8 cm, variably segmented, with well-developed roots; shoot tips white, ca. 5.9 cm. Leaves 9–13, in a terminal whorl; petiole 0.9–2.4 × 0.2–0.4 cm; blades oblanceolate or lanceolate, 5.5–21.0 × 1.9–6.5 cm, green or yellowish green, apex attenuate, base cuneate or narrowing toward petiole, margin entire or minutely undulate, lateral veins 4–7 pairs, first pair being subbasal, straight to margin, others alternate or opposite in basal third, arching and fading near the apex. Flowers solitary; pedicels 6.8–17.0 × 0.3–0.4 cm. Sepals 3–8, green adaxially, pale green abaxially, ovate or elliptic, 3.8–6.7 × 1.5–2.5 cm. Petals 4–7, equalling sepals in number, green, filamentous, 3.0–8.9 × ca. 0.2 cm, ca. twice as long as sepals. Stamens twice the number of petals, green, persistent. Ovary subglobose, 0.9–2.4 cm long, 1-loculed, with 4–8 placentas; ovules anatropous, numerous, yellowish white, ridged longitudinally; styles and stigmas purple-black, 0.3–0.6 cm long, equal to the sepals in number, separate, curved; style base deep purple at anthesis, yellowish white in fruit, inflating to cover ovary apex. Fruits capsules, globose, 1.4–3.3 cm in diam., with 4–8 winged ridges, surface granulate, yellowish green to dark purple (immature) or purplish green (mature). Seeds numerous, ovoid, with red sarcotesta, white endotesta ca. 0.5 cm.
Distribution: Known thus far only from SW Guangdong (Xinyi), and Yunnan provinces, China.
Phenology: Flowering from April to June and fruiting from July to September.
Notes: Paris kwantungensis is classified in subg. Daiswa (Rafinesque 1836: 18; Li 1984: 356) based on its thick rhizome, green ovary with a single locule, and thick sarcotesta with 4–6 green longitudinal ridges. This species is particularly similar to P. polyphylla var. polyphylla and P. polyphylla var. chinensis. A comparative analysis of these morphologically similar taxa is presented in Table 2 and Fig. 5.
Morphological characters of Paris kwantungensis, P. polyphylla var. chinensis, and P. polyphylla var. polyphylla.
Paris kwantungensis and morphologically similar species. A. P. kwantungensis. B. P. polyphylla var. chinensis. C. P. polyphylla var. polyphylla.
Table 2 highlights three major distinctions. First, P. kwantungensis has a consistently tuberculate pericarp, whereas the other two taxa have a smooth pericarp. Second, the petals of P. kwantungensis are significantly longer than the sepals. By contrast, in P. polyphylla var. chinensis, the petals are only one-third to two-thirds the length of the sepals and are typically reflexed, whereas in P. polyphylla var. polyphylla, the petals are nearly equal in length to the sepals, yellowish-green, slightly broader in the middle, and tapering at both ends. Third, P. kwantungensis generally bears more than ten leaves, whereas the other two taxa usually have fewer than 10.
DISCUSSION
Our analysis aligns with the initial morphological phylogenetic study by Miau (1982). Although many researchers have considered P. kwantungensis a synonym of P. polyphylla var. polyphylla and P. polyphylla var. chinensis, we maintain that P. kwantungensis is monophyletic and should be treated as a separate species. Both molecular and morphological differences support this conclusion. On the one hand, molecular evidence strongly supports the monophyly of P. kwantungensis, confirming its sister relationship to P. polyphylla var. chinensis, with each species placed in distinct clades. On the other hand, these molecular findings are corroborated by distinct morphological differences. While P. kwantungensis is similar to P. polyphylla var. chinensis, it is distinguished by longer petals, nearly twice as long as the sepals (rather than shorter), and reflexed sepals, which allow for easy differentiation from other related species. It is important to note that the name P. polyphylla var. kwantungensis has previously been used to refer to a variety of P. polyphylla. However, in our phylogenetic analysis, P. kwantungensis was positioned distantly from other P. polyphylla varieties. Therefore, we propose that the species name be changed to P. kwantungensis. This study provides compelling evidence to resolve the long-standing debate over the potential synonymization of these species.
Although the ITS sequences used here provide a reliable basis for phylogenetic inference, future research should incorporate additional molecular markers, such as the chloroplast genome. A multi-marker approach would yield a more comprehensive phylogenetic framework, improving resolution and providing stronger support for recognizing P. kwantungensis as an independent species.
Notes
ACKNOWLEDGMENTS
We thank Yuyu Liu and Zefei Du for their valuable help in collecting plant samples. We also acknowledge financial support from the National Natural Science Foundation of China (No. 31860080), the Yunnan Provincial Basic Research Program (No. 202501AU070183), the Yunnan Provincial Department of Education Science Research Fund (No. 2024J0830), and the Open Project of the National Experimental Teaching Demonstration Center for Military Pharmacy (No. JSYXKF-2506).
CONFLICTS OF INTEREST
The authors declare that there are no conflicts of interest.