A narrow zone extending from the western to the southern coast of the Korean Peninsula, encompassing approximately 3,000 islands, is characterized by the dominance of warm-temperate plant species (Lee and Yim, 2002; Jang et al., 2023). This distinct vegetation composition defines the area as the Southern Coast Floristic Province (Lee and Yim, 2002). Although the Jejudo Floristic Province, which has a similar climate, exhibits somewhat different flora, the majority of warm-temperate plant species on the Korean Peninsula are widely distributed in both regions. While both the Southern Coast and Jejudo Floristic Provinces are primarily warm-temperate, these regions display a subtle inclination towards subtropical conditions, as indicated by the presence of many plants typically associated with subtropical forests. For example, forests in these regions are predominantly composed of evergreen broad-leaved trees from the genera Machilus Nees, Neolitsea (Bentham & J. D. Hooker) Merrill, and Quercus L. [subgenus Cyclobalanopsis (Oersted) C.K. Schneid.] (Lee and Choi, 2010).

Within this context, Hepatica Mill., which comprises 11 taxa and is distributed across temperate regions of the Northern Hemisphere (Park and Park, 2021), includes H. insularis Nakai, an endemic species of the warm-temperate regions on the Korean Peninsula. This species, first described by Nakai in 1937, is particularly notable for its smaller size, which distinguishes it from H. asiatica Nakai, a species found in temperate regions on the Korean Peninsula and in cool-temperate regions of Manchuria (Park and Park, 2021). Aside from size, the two species exhibit few other distinct morphological differences, making their taxonomic separation challenging. Despite this, the status of H. insularis as a distinct species from H. asiatica has been the subject of ongoing debate among researchers. Kim and Lee (1994) and Woo et al. (2002) reported that, although the two taxa exhibit morphological similarities, genetic differentiation was observed through allozyme and isozyme analyses. However, Pfosser et al. (2011) found no clear genetic patterns that could distinguish the two taxa using nuclear AFLP and plastid marker analyses.

We speculate that the morphological and ecological differences between these two taxa may be influenced by various phylogeographic scenarios, as suggested by previous studies on plants in similar forest biomes (Lee et al., 2013, 2014; Han et al., 2020, 2023). To explore the evolutionary history of H. insularis, we developed 15 polymorphic microsatellite markers. We also tested the applicability of these markers to H. asiatica.

To generate high-throughput sequencing data, we collected a fresh leaf sample of H. insularis from Jokeunnokkome Oreum, Jejudo Island, Korea, and extracted its genomic DNA using the DNeasy Plant Mini Kit (QIAGEN, Seoul, Korea) according to the manufacturer’s protocol. A voucher specimen (Voucher no. LeeHi190323V1) was deposited in the herbarium of Biological Education, Chonnam National University (BEC). The quality of the extracted DNA was evaluated using a Micro-Spectrophotometer Nano-300 (Allsjeng, Hangzhou, China), which showed a DNA concentration of 36.8 ng/μL, with A260/280 and A260/230 ratios of 1.94 and 2.22, respectively. A shotgun library was subsequently prepared using the Illumina MiSeq platform (LAS, Seoul, Korea), resulting in a total of 7,934,669 reads with paired-end sequencing (2 × 301 bp).

To identify microsatellites within these reads, we utilized the SSR pipeline v. 0.951 (Miller et al., 2013), configuring the settings to detect di-, tri-, and tetra-nucleotide motifs with flanking regions larger than 100 bp and containing at least 10, 6, or 4 repeats, respectively. Reads containing AT and TA repeats, those with GC content below 30% or above 80%, and reads with additional repeats were excluded. The sequences were categorized based on their repeat motifs and mapped to the reference genome using Geneious R 10.1.3 (Kearse et al., 2012). Contigs with sequence identities below 80% or quality scores under 50% were discarded. We selected sequences that exhibited unique patterns, had fewer than 20 reads, contained two distinct alleles, showed minimal variation at the primer attachment site, and had no additional single nucleotide polymorphisms in the flanking regions. These selected reads were deduplicated through de novo assembly and subsequently used for microsatellite marker development. We designed 90 primer pairs (18–22 bp in length) using the Primer 3 software within Geneious, targeting a melting temperature of 53–60°C and GC content of 35–65%. Forward primers were tagged with one of three M13 sequences (5′-CACGACGTTGTAAACGAC-3′, 5′-TGTGGAATTGTGAG CGG-3′, and 5′-CTATAGGGCACGCGTGGT-3′) and labeled with 6-FAM, VIC, or NED fluorescent dyes, respectively. For multiplex PCR, each primer set comprised nine primers ranging in length from 100 to 250 bp.

To assess the effectiveness of the newly developed microsatellite loci, we sampled 44 individuals of H. insularis from the Jjo and JS populations (Table 1). Cross-species amplification was also tested by sampling 28 individuals from the DG population of the related species H. asiatica (Table 1). PCR amplifications were performed in a final volume of 5 μL, containing 15–20 ng of DNA, 2.5 μL Multiplex PCR Master Mix (QIAGEN), 0.01 μM forward primer, 0.2 μM reverse primer, and 0.1 μM M13 primer (fluorescently labeled). The PCR protocol consisted of an initial denaturation at 95°C for 15 min, followed by 35 cycles of 95°C for 30 s, annealing at 56°C for 1.5 min, and extension at 72°C for 1 min, with a final extension at 72°C for 10 min. PCR products were analyzed using an ABI 3730XL sequencer with the GeneScan 500 LIZ Size Standard (Thermo Fisher Scientific, Waltham, MA, USA), and allele sizes and peaks for each sample were determined using Peak Scanner Software version 2.0 (Thermo Fisher Scientific). Genetic parameters, including the number of alleles, expected heterozygosity (H_E), and observed heterozygosity (H_O), were calculated with GenAlEx 6.5 (Peakall and Smouse, 2006). Deviations from Hardy-Weinberg equilibrium (HWE) and linkage disequilibrium (LD) were evaluated using GENEPOP version 4.6.9 (Rousset, 2008).

Using 90 newly designed primer pairs, we developed 15 polymorphic microsatellite loci, each exhibiting clear and strong peaks for every allele in 44 individuals of H. insularis from two populations (Table 2). The number of alleles per locus ranged from 2 to 11, with an average of 5.07. The 15 marker sequences developed for H. insularis were deposited in GenBank (MW092027-MW092045). The expected heterozygosity (H_E) and observed heterozygosity (H_O) values ranged from 0.180 to 0.802 and from 0.000 to 0.933, respectively (Table 3). Cross-amplification in the closely related species H. asiatica indicated that 8 loci were successfully amplified and were polymorphic, with 1 to 5 alleles per locus (Table 3). Although only one locus (HeI051) showed significant deviations from HWE after Bonferroni correction (p < 0.0033) across three populations, this deviation is likely due to population-specific factors rather than being marker-specific. Moreover, no significant LD was detected at any locus after Bonferroni correction, further supporting the reliability of the markers across the studied populations.

In conclusion, we developed a set of 15 polymorphic microsatellite markers from H. insularis and determined that 8 of these loci are applicable to the related species H. asiatica. The microsatellite markers described here will serve as a powerful genetic tool for elucidating the evolutionary patterns of H. insularis endemic to the Korean Peninsula. Furthermore, they will significantly contribute to advancing our understanding of both historical and contemporary gene flow between H. insularis and H. asiatica.