Abbreviations
BPP
Bayesian posterior probabilities
BS
bootstrap
COI-5P
5′ region of cytochrome c oxidase subunit I
CUK
herbarium of Chosun University
GTR
General Time-Reversible
ML
maximum likelihood
MABIK
Marine Biodiversity Institute of Korea, Korea
PCR
polymerase chain reaction
RAxML
Randomized Axelerated Maximum Likelihood
INTRODUCTION
The genus Chondrophycus (J. Tokida & Y. Saito) Garbary & J. T. Harper was originally proposed as a subgenus within the genus Laurencia (Saito 1967). Laurencia was divided into two subgenera: (1) subgenus Laurencia, which encompassed the sections Pinnatifidae, Forsterianae, Laurencia, and Planae, and (2) subgenus Chondrophycus, which included the sections Cartilagineae, Palisadae, and Articulatae (Saito 1967, Saito and Womersley 1974, Zhang and Xia 1985, McDermid 1988b). The subgenus Chondrophycus has been characterized by the followed features: (1) two pericentral cells per axial segment, (2) first pericentral cell with side position of trichoblast, (3) absence of radially elongated cortical cells, (4) absence of lenticular thickenings, (5) tetrasporangia initially forming abaxially, (6) spermatangial development of the trichoblast-type, and (7) procarp-bearing segments with five pericentral cells (Saito 1969a, Nam et al. 1994, Nam 1999). Garbary and Harper (1998) raised Chondrophycus to the level of a genus, designating C. cartilagineus (Yamada) Garbary & J. T. Harper as the type species. This decision was based on a cladistic analysis of the vegetative and reproductive morpho-anatomical features of various species. Subsequently, the genus has been characterized by the combination of the following features: (1) Chondrophycus: absence of secondary pit connections between cortical cells; two sterile pericentral cells in tetrasporangial axial segments; tetrasporangia with a right-angle arrangement, (2) Kangjaewonia: similar to Chondrophycus but with tetrasporangia arranged in a parallel pattern, (3) Palisada: similar to Chondrophycus but with one sterile pericentral cell in tetrasporangial axial segments, and (4) Yuzurua: similar to Chondrophycus but with secondary pit connections between cortical cells (Nam 1999). Later, through a reevaluation of the phylogenetic relationships within the Laurencia complex, Palisada was proposed as a genus by Nam (2006, 2007). Palisada can be distinguished from Chondrophycus by the position of the first pericentral cell, which is located beneath the trichoblast rather than on its side, and by its tetrasporangial axes, which have one sterile pericentral cell instead of two (Nam 2006). Additionally, the elevation of Yuzurua to generic status was supported by molecular and morphological evidence, distinguishing it from Palisada by the presence of secondary pit connections between cortical cells (Martin-Lescanne et al. 2010). Currently, the genus, Chondrophycus is composed of two subgenera: Chondrophycus and Kangjaewonia.
Fourteen species of the genus Chondrophycus have been reported worldwide (Guiry and Guiry 2024). Of them, 12 species of Chondrophycus have been recorded in the Pacific Ocean (Guiry and Guiry 2024). In Korea, three species have been recorded based on morpho-anatomical analyses: C. cartilagineus, C. kangjaewonii (K. W. Nam & C. H. Sohn) D. J. Garbary & J. T. Harper, and C. undulatus (Yamada) Garbary & Harper (Lee and Kang 2001, Nam 2011). Chondrophycus cartilagineus, the type species of the genus, was described from Mogi, Fukuoka Prefecture, Japan (McDermid 1988a). It is a well-represented species with a distribution ranging from Heuksando to Ullendong in Korea (Lee and Kang 2001, Nam 2011). Chondrophycus kangjaewonii is an endemic species in Korea (Lee and Kang 2001) and distributed in the subtidal zone (2–3 meters), notably found in Yongdeuk, Busan, and Hongdo (Lee and Kang 2001, Nam 2011). Chondrophycus undulatus was described from Enoshima, Kanagawa Prefecture, Japan (Silva et al. 1996) and well reported in Korea (Nam 1999, 2011, Lee and Kang 2001). These Korean Chondrophycus species have been identified based on morphology alone, and no molecular data are available for any of these Korean species. Furthermore, there are limitations in accurately identifying each species of Chondrophycus due to the plasticity of morphological features in vegetative thalli and in those altered by their reproductive structures (Tseng 1943, Cribb 1958, Saito 1969b, Saito and Womersley 1974, Zhang and Xia 1980, Nam and Saito 1990, Nam and Sohn 1994, Nam 1999, 2011, Wynne et al. 2005, Sentíes et al. 2016, Popolizio et al. 2022). Molecular analyses have enabled the identification of Chondrophycus species, including C. anabeliae Sentíes, M. T. Fujii, Cassano & Dreckmann, C. dotyi (Y. Saito) K. W. Nam, and C. planiparvus Popolizio, C. W. Schneider & C. E. Lane (Sentíes et al. 2016, Popolizio et al. 2022). However, the molecular phylogenetic studies of the genus Chondrophycus still face challenges due to the difficulties of traditional morphology-based species recognition.
We collected the unidentified samples of Chondrophycus along the Korean coastal shores, identified the species, and recognized some of them as a novel species of Chondrophycus based on molecular and morphological analyses. In this study, we characterize them morphologically and examine their phylogenetic relationships by rbcL and COI-5P gene sequences.
MATERIALS AND METHODS
Sampling
A total of 25 samples were collected between 2008 and 2023 from the coasts of Korea and Japan. The collections were gathered from the lower intertidal zone at a depth of 1 meter. Specimens were processed for herbaria, preserved using silica and fixed in 4% formalin seawater for subsequent analyses. Voucher specimens were stored in the herbarium collections of Chosun University (CUK) and the Marine Biodiversity Institute of Korea (MABIK) in Korea.
DNA extraction and amplification
Genomic DNA from 25 samples was extracted using the NucleoSpin Plant II Kit (Macherey-Nagel, Düren, Germany). The plastidial rbcL and mitochondrial COI-5P genes were amplified using the following primer combinations: FrbcLstart-R753, F645-RrbcSstart, F57-R753, F577-R1150, F492-R1150, and F993-RrbcSstart for rbcL (Freshwater and Rueness 1994, Lin et al. 2001) and GazF1-GazR1 or GazF2-DumR1 for COI-5P (Saunders 2005, Lane et al. 2007). Polymerase chain reaction (PCR) amplification was carried out using a Veriti 96-well Thermal cycler (Applied Biosystem, Foster City, CA, USA). The PCR amplification for rbcL was modified from Gavio and Fredericq (2002), consisting of an initial denaturation at 94°C for 1 min, followed by 2 cycles of denaturation at 94°C for 1 min, annealing at 40°C for 1 min, and extension at 68°C for 1 min, with subsequent 40 cycles of denaturation at 94°C for 1 min, annealing at 42°C for 30 s, extension at 68°C for 1 min, and a final extension of 5 min at 72°C. The PCR amplification protocol for COI-5P followed either Cassano et al. (2019) using the primer combination GazF1-GazR1 or Peña et al. (2015) using GazF2-DumR1.
Sequence analyses
Phylogenetic trees were inferred from rbcL and COI-5P sequence data. All sequences used in the analyses are listed in Supplementary Table S1. Two datasets of rbcL and COI-5P were edited and aligned using Geneious Prime 2022.02. Chondria capillaris (Hudson) M. J. Wynne, Ch. collinsiana M. Howe, Ch. dasyphylla (Woodward) C. Agardh, Campylaephora kondoi (Yendo) Barros-Barreto & Maggs, and Vertebrata foetidissima (Cocks ex Bornet) Díaz-Tapia & Maggs were selected as outgroups. PartitionFinder2 (Guindon et al. 2010, Lanfear et al. 2012, 2017) was utilized to determine the best-fit partitioning schemes and models of molecular evolution for both datasets, rbcL and COI-5P. The selected model for both datasets was the General Time-Reversible model of nucleotide substitution with invariant sites and gamma-distributed rates for the variable sites (GTR + Γ + I) under the corrected Akaike information criterion. Bayesian inference was conducted using MrBayes 3.2.6 (Ronquist et al. 2012), employing Metropolis-coupled Markov chain Monte Carlo. To assess posterior probabilities, two runs were executed, each with four chains (one cold chain and three heated), for 2,000,000 generations, with tree sampling occurring every 1,000 generations under the GTR + Γ + I evolutionary model. A burn-in value of 25% was employed to eliminate suboptimal trees from the final consensus tree. Maximum likelihood analyses were carried out using raxmlGUI 2.0 (Edler et al. 2021) utilizing the GTR + G + I model, with statistical support derived from 1,000 bootstrap replicates. Interspecific pairwise distances were estimated using the p-distance model in MEGA11: Molecular Evolutionary Genetics Analysis version 11 (Tamura et al. 2021).
Morphological analyses
The freshly collected samples were photographed using a digital camera (Nikon D40; Nikon, Japan) and a stereomicroscope (SZX7; Olympus, Tokyo, Japan) to capture their external morphology. Internal morphological observations were conducted using fresh material, rehydrated herbarium specimens, or samples preserved in a solution of 4% formalin and seawater. Thallus examination involved the preparation of both cross and longitudinal sections using a stainless-steel razor blade or a freezing microtome, yielding slices with a thickness of 5–10 μm (Shandon Cryotome FSE; Thermo Shandon, Ltd., Loughborough, UK). The sliced samples were mounted on slides with distilled water. Subsequently, the samples were stained with a 1 : 1 mixture of aqueous aniline blue and acetic acid to enhance the clarity of structural details. The stained sections were then mounted in 50% corn syrup and examined under a microscope. Images were captured using an Olympus DP27 camera attached to an Olympus microscope (BX51TRF; Olympus). Measurements of the sections were conducted utilizing ImageJ software (Schneider et al. 2012).
RESULTS
Phylogenetic analyses
A total of 25 sequences with lengths of 1,359 bp for rbcL and 550 bp for COI-5P were generated from Korean and Japanese samples. Phylogenetic analyses using rbcL and COI-5P sequences showed that the Chondrophycus specimens from Korea and Japan were distributed into four separate clades. Three of these clades correspond to C. cartilagineus, C. kangjaewonii, and C. undulatus, while the fourth clade represents a novel species, C. coreanus sp. nov. In the rbcL phylogenetic tree, C. coreanus sp. nov. formed a distinct and fully supported clade within Chondrophycus genus (Fig. 1). The COI-5P phylogenetic tree (Supplementary Fig. S1) showed large congruency with rbcL phylogeny. The interspecific genetic distance between C. coreanus sp. nov. and other Chondrophycus species was found to be 4.2–9.5% for rbcL sequences and 5.5–11.6% for COI-5P sequences. Specifically, the sequence divergences between C. coreanus sp. nov. and C. cartilagineus were 5.3–6.2% for rbcL and 6.1–6.8% for COI-5P; between C. coreanus sp. nov. and C. kangjaewonii, they ranged from 7.2–7.4% for rbcL and 10.2–10.4% for COI-5P; and between C. coreanus sp. nov. and C. undulatus, they were 7.1–7.3% for rbcL and 8.1–8.7% for COI-5P. The intraspecific genetic distance for C. coreanus sp. nov. is identical in rbcL and 0.4–1.3% in COI-5P, whereas for C. cartilagineus, it ranges from 0–0.5% in rbcL and 0–0.9% in COI-5P.
Morphological observations
Chondrophycus coreanus P. Romero-Orozco, B. Y. Won & T. O. Cho sp. nov
Diagnosis
Thalli are cartilaginous, up to 8 cm in height, subcompressed when vegetative, terete when reproductive. Holdfast is discoid. The branching pattern is alternating or opposite. Branching order is two or three. On the surface view, cortical cells are polygonal to longitudinally elongated. In longitudinal section, the cortical cells are elongate-shaped near the apex, projecting beyond the surface, and without secondary pit connection. In cross-section, obconical-shaped cortical cells are less pigmented than subcortical cells. Elongated-shaped subcortical cells are connected by secondary pit connections. Medullary cells are rounded to elongated, colorless, thickened medullary cell walls without lenticular thickenings. Two pericentral cells are produced from axial cell. Cystocarps are conical, with protuberant ostiole, 569–1,320 μm in length, and 460–1,077 μm in diameter. The pericarp of the cystocarp consists of 4–6 cell layers. Male gametophytes and tetrasporophytes were not observed.
Holotype
MABIK00101051 (= CUK21260) deposited in CUK herbarium. GenBank accession Nos.: PP898383 for rbcL and PP898405 for COI-5P.
Type locality
Janggil-ri, Guryongpo-eup, Nam-gu, Pohang-si, Gyeongsangbuk-do, Korea (35°57′06″ N, 129° 32′49″ E), female gametophyte, Oct 12, 2023.
Paratypes
CUK21259 (= PH37), vegetative, Gampo-ro, Gampo-eup, Gyeongju-si, Gyeongsangbuk-do, Korea (35°48′23″ N, 129°30′40″ E), May 19, 2021; CUK4248, vegetative, Takero-kaigan, Kayakimachi, Nagasaki, Japan (32°38′20″ N, 129°47′54″ E), Apr 23, 2008.
Vegetative morphology
Thalli are red to brown, dark brown when they get dry, cartilaginous, and 4–8 cm in height (Fig. 2A). Holdfast is discoid. Branching pattern of branches and ultimate branchlets is alternate to opposite (Fig. 2B). The branching order is two or three (Fig. 2B). The main axes are slightly wider in the middle part with 2–4 mm in diameter, and narrower in the lower part with 1–2 mm in diameter (Fig. 2B). Ultimate branchlets are terete to clavate truncate at the apices, and 0.3–2 mm in length (Fig. 2B & C). In surface view of the thallus, the cortical cells are polygonal to longitudinally elongated, measuring 25–49 μm in length and 17–34 μm in width, without secondary pit connections, and projecting beyond the surface (Fig. 2D & E). Thalli are subcompressed and composed of cortical and medullary structures in cross-section (Fig. 2F). Two pericentral cells are produced from an axial cell (Fig. 2G). Cortical cells are obconical-shape lacking the formation of a palisade arrangement, less pigmented than the subcortical cells, 25–46 μm in length, and 12–30 μm in width (Fig. 2H & I). Subcortical cells are elongated, connected by secondary pit connections, 31–75 μm in length, and 20–58 μm in width (Fig. 2I). Medullary cells are round to elongated, colorless, with thick walls, with annular thickenings, 65–134 μm in length, and 63–120 μm in width (Fig. 2H).
Reproductive morphology
The female thalli are terete to partially compressed and 8 cm in height (Fig. 2J & K). The main axes are, 1–2 mm in diameter on the upper part, 1.5–3 mm in diameter on the middle part, 2 mm in diameter on the lower part, and having thickened medullary cell walls (Fig. 2K–M). Cystocarps are conical, closely developed together on branchlets, 569–1,320 μm long, 460–1,077 μm in diameter, with 4–6 cell-layered pericarp, with protuberant ostioles (Fig. 2K & N). Carposporangia are clavate (Fig. 2N & O). Male gametophytes and tetrasporophytes were not observed.
DISCUSSION
Based on molecular (rbcL and COI-5P) and morphological data, we described a new species, Chondrophycus coreanus sp. nov. offering useful insights into algal. C. coreanus sp. nov. was nested in a distinct clade within Chondrophycus and it was closely related to C. tronoi (E. Ganzon-Fortes) K. W. Nam in the rbcL phylogenetic tree and to C. anabeliae in the COI-5P tree (Fig. 1, Supplementary Fig. S1). Although C. coreanus formed an independent and well-supported lineage within the genus Chondrophycus, its relationships with C. dotyi, C. tronoi, and C. undulatus were not resolved by the rbcL phylogenetic tree. However, the divergence values between C. coreanus and C. dotyi (KX815263) showed a divergence of 7.2–7.4%, between C. coreanus and C. tronoi (AF489864, MN636853) 4.8–5%, between C. coreanus and C. undulatus (PQ043266) 7.1–7.3%, and between C. coreanus and C. cf. undulatus (FJ785307, FJ785308) 6.4–6.6%. Additionally, for COI-5P, the divergence values between C. coreanus and C. anabeliae ranges from 6.2 to 7.7%. These values align with divergence ranges reported in previous studies, which were 1.8 to 9.5% for rbcL and 5.4 to 9.3% for COI-5P further supporting the recognition of C. coreanus as a distinct species (Cassano et al. 2012, 2020, Sentíes et al. 2016, Popolizio et al. 2022). In the COI-5P analyses, our samples of C. coreanus from Korea and Japan cluster with a sample identified as Chondrophycus sp. (GenBank accession No. HQ423057) from Hawaii, USA. Additionally, the intraspecific between our collection from Korea and Japan is 1.3% on COI-5P, being higher than recorded before for Chondrophycus species (Cassano et al. 2020, Popolizio et al. 2022). C. coreanus sp. nov. is characterized by (1) terete to partially compressed thalli, (2) cortical cells that project near the branchlet apex, (3) thickened medullary cell walls, and (4) a cystocarp with a protuberant ostiole (Table 1).
The thallus forms have been used to distinguish some Chondrophycus species (Nam 1999, 2006). There are three types: terete, found in C. tronoi, C. articulatus (C. K. Tseng) K. W. Nam, C. carolinensis, C. columellaris (Børgesen) E. Coppejans & A. J. K. Millar, C. verticillatus (J. F. Zhang & B. M. Xia) K. W. Nam; subcompressed, seen in C. ceylanicus (J. Agardh) M. J. Wynne, Serio, Cormaci & G. Furnari, C. dotyi, C. kangjaewonii, C. planiparvus Popolizio, C. W. Schneider & C. E. Lane, C. succisus (A. B. Cribb) K. W. Nam, and C. undulatus, and a mix of terete vegetative and subcompressed reproductive thalli, characteristic of Chondrophycus anabeliae, C. brandenii (Y. Saito & H. B. S. Womersley) K. W. Nam, C. cartilagineus (Tseng 1943, Børgesen 1945, Cribb 1958, Saito 1969b, Zhang and Xia 1980, Ganzon-Fortes 1982, Nam and Saito 1990, Nam and Sohn 1994, Nam 1999, Wynne et al. 2005, Sentíes et al. 2016, Popolizio et al. 2022). C. coreanus exhibits a subcompressed thallus during the vegetative stage and a terete thallus during the reproductive stage. The mixed thallus form observed in C. coreanus is similar to that of C. anabeliae, C. brandenii, and C. cartilagineus; however, they exhibit an inverse combination with terete vegetative and subcompressed reproductive thalli (Saito and Womersley 1974, Nam and Saito 1990, Sentíes et al. 2016). The feature by having subcompressed vegetative thalli and terete reproductive thalli of C. coreanus is a novel finding within the genus.
The presence or absence of lenticular thickness in the walls of medullary cells has been used as a characteristic to distinguish the species within the Laurencia complex including Chondrophycus (Saito 1967). Within the genus Chondrophycus, lenticular thickness has been found in only two species: C. articulatus and C. verticillatus (Tseng 1943, Zhang and Xia 1980). In contrast, C. coreanus sp. nov. lack these lenticular thickenings. This feature, described as lenticular in shape when viewed in the cross-section, has been associated with a structural support role (Gordon-Mills and Womersley 1984). Additionally, C. coreanus is distinguished from other species within Chondrophycus by the presence of projections from the cortical cells near the branchlet apices. This feature has not been found in C. ceylanicus, C. kangjaewonii, C. succisus (A. B. Cribb) K. W. Nam, C. undulatus, C. tronoi, and C. verticillatus (Cribb 1958, Zhang and Xia 1980, Ganzon-Fortes 1982, Nam and Sohn 1994, Nam 1999, Wynne et al. 2005). C. coreanus also differs from C. dotyi, which is characterized by thin medullary cell walls (Saito 1969b), C. planiparvus, which has secondary pit connections between cortical cells (Popolizio et al. 2022), and C. articulatus, which features an articulate appearance and arcuate ultimate branchlets (Tseng 1943).
Chondrophycus cartilagineus, C. kangjaewonii, and C. undulatus have been reported based on morpho-anatomical analyses in Korea (Lee and Kang 1986, 2001). In this study, we identified them based on molecular and morphological analyses. C. cartilagineus was initially described by Yamada as Laurencia cartilaginea (Yamada 1931, McDermid 1988a), and its morphology, including the type material, was thoroughly examined by Nam and Saito (1990). Our specimens of C. cartilagineus have terete to subcompressed thalli, a discoid holdfast, projections from the cortical cells near the apex, and medullary cells without lenticular thickenings (Fig. 3A–E). These characteristics are corresponded with their observations by Nam and Saito (1990). Although C. cartilagineus is closest to the clade formed by “C. dotyi” (HQ422621) and “C. succisus”(GU223884) in the COI-5P analysis, the gene sequence divergence between C. cartilagineus and “C. dotyi” is 3.1–3.5%, and between C. cartilagineus and C. succisus is 3.5–4%. These values are sufficient to distinguish different species, as supported by previous studies in genera of the Laurencia complex (Machín-Sánchez et al. 2016, Cassano et al. 2019, Popolizio et al. 2022). C. kangjaewonii was recognized as a species within the genus Chondrophycus by Nam and Sohn (1994). Later, Nam (1999) proposed it as one of two subgenera within genus Chondrophycus, with this species being the only representative of its subgenus based on morphological evidence. Morphologically, it exhibits all the features of the other subgenus Chondrophycus, with the exception of the arrangement of parallel tetrasporangia. Our specimens of C. kangjaewonii have the compressed and soft thalli, a discoid holdfast, flat or somewhat round ultimate branchlets, medullary cells without lenticular thickenings, and parallel tetrasporangia arrangement (Fig. 3F–J). These characteristics are corresponded with the original descriptions of C. kangjaewonii (Nam and Sohn 1994). Our rbcL phylogenetic tree reveals that C. kangjaewonii does not show distinct separation at the subgenus level from other Chondrophycus species (see Fig. 1). Instead, it forms a well-supported clade (99/1) with other species within the genus. We suggest that, based on molecular data, the taxonomic status of subgenus Kangjaewonia should be revised. Chondrophycus undulatus was described by Yamada (1931) as Laurencia undulata and later transferred to the genus Chondrophycus by Garbary and Harper (1998). Our specimens of C. undulatus have the subcompressed thalli, mainly distichous, irregularly subopposite-pinnate branching pattern, terete ultimate branchlets at the apices, cortical cells without projection, and an undulate margin (Fig. 3K–O). These characteristics are corresponded with the original descriptions of C. undulatus (Nam 1999). Our phylogenetic analysis of rbcL shows that the C. undulatus sequence is 99% similar to a sequence from a specimen of C. dotyi collected near type locality in Hawaii, USA. Both species exhibit similar thallus forms, but C. dotyi differs by lacking a primary leading branch, having a radial branching pattern, and possessing cortical cell projections near the apex (Saito 1969b). In contrast, C. undulatus has a primary leading branch, a distichous branching pattern, and does not have cortical cell projections. This suggests that a review of the type material for both species might be necessary.
