Study species and population sampling

Utricularia bifida is a small annual with traps (0.5–1 mm long) on stolons and leaves. Inflorescences are erect and 2–40 cm long with 1–16 flowers. Corolla (6–10 mm long) is yellow, with a subulate spur as long as lower corolla lip. Capsules are broadly ellipsoid (2.5–3 mm long) and dorsiventrally compressed. Seeds are obliquely obovoid and 0.4–0.6 mm long (Li and Cheek, 2011). Utricularia caerulea is also a small annual with traps (0.2–1 mm long) on stolons and leaves. Inflorescences are erect and 5–44 cm long with 1–15 flowers. As one of the diagnostic characters, corollas (4–10 mm long) are violet, pink, or white. The spur is narrowly conic, usually longer than lower corolla lip. Capsules, similar to those of U. bifida, are also small and globose (1.5–2 mm long) and dorsiventrally compressed. Seeds are also tiny (0.2–0.3 mm long) (Li and Cheek, 2011). Although Utricularia species are usually self-compatible (Taylor, 1989; Araki and Kadono, 2003; Hobbhahn et al., 2006; Clivati et al., 2013), we do not know much about breeding systems of the two study species. Utricularia bifida and U. caerulea often co-occur in wet places and wetlands in southern Korea, where U. yakusimensis is rarely found, particularly in mountain wetlands along the southern “Nakdongjeongmaek” (J.-O. Hyun, pers. comm.).

In South Korea, U. bifida and U. caerulea have been listed as “LC” (“Least Concern”) in the red list of threatened species (MOE, 2014), i.e., they are not considered endangered. However, U. bifida has been listed as “VU” (“Vulnerable”) in the list of rare plants of Korea (KNA, 2012).

In proportion to total samples in each population, we randomly collected 341 individuals (U. bifida) and 208 individuals (U. caerulea) from 10 populations of each species from southern Korea (Fig. 1) in October 2007 and October 2008. The name of locations and sample sizes are included in the legend of Fig. 1.

Fig. 1.

The location and sample sizes of the 10 populations of Utricularia bifida (UB) and U. caerulea (UC) in southern examined in this study: 1, Sangju City (UB, n = 28; UC, n = 36); 2, Mujechi # 4-neup in Mt. Jeongjok (UB, n = 26; UC, n = 32); 3, Gijang Town (UB, n = 13; UC, n = 27); 4, Hogye Township (UB, n = 25; UC, n = 39); 5, Geoguem Island (UB, n = 22; UC, n = 36); 6, Hwasan Township (UB, n = 17; UC, n = 37); 7, Hwangsan Township (UB, n = 20; UC, n = 39); 8, Munnae Township (UB, n = 8; UC, n = 15); 9, Namchang Township (UB, n = 26; UC, n = 32); 10, Gunnae Township, Jin Island (UB, n = 18; UC, n = 32).

Enzyme electrophoresis

Using a phosphate-polyvinylpyrrolidone buffer (Mitton et al., 1979), we extracted enzymes by finely cutting leaf samples and then crushing them with a mortar and pestle. Enzyme extracts were absorbed onto 4 × 6 mm wicks cut from Whatman 3MM chromatography paper, which were then stored at −70°C until needed. Allozyme variation was assessed with horizontal starch-gel electrophoretic techniques. Starch gels (13%) were stained for 13 enzyme systems: alcohol dehydrogenase (ADH, E.C.1.1.1.1), cathodal peroxidase (CPX, E.C.1.11.1.7), diaphorase (DIA, E.C.1.6.99.–), fluorescent esterase (FE, E.C.3.1.1.–), isocitrate dehydrogenase (IDH, E.C.1.1.1.42), leucine aminopeptidase (LAP, E.C.3.4.11.–), malate dehydrogenase (MDH, E.C.1.1.1.37), malic enzyme (ME, E.C.1.1.1.40), phosphoglucoisomerase (PGI, E.C.5.3.1.9), phosphoglucomutase (PGM, E.C.2.7.5.1), 6-phosphogluconate dehydrogenase (6PGD, E.C.1.1.1.44), shikimate dehydrogenase (SKDH, E.C.1.1.1.25), and triosephosphate isomerase (TPI, E.C.5.3.1.1). A modification (Haufler, 1985) of Soltis et al. (1983) system 6 was used to resolve ADH, CPX, DIA, FE, LAP, ME, PGI, PGM, and TPI. The morpholine citrate buffer system (pH 6.1) of Clayton and Tretiak (1972) was used to resolve IDH, MDH, 6PGD, and SKDH. Stain recipes were taken from Soltis et al. (1983), except DIA (Cheliak and Pitel, 1984). Putative loci were designated sequentially, with the most anodally migrating isozyme designated as ‘1’ and the next ‘2.’ Alleles were also designated sequentially, with the most anodally migrating allele designated as ‘a’ the next ‘b,’ etc.

Lack of allozyme diversity in Utricularia bifida and U. caerulea: inference of population history

Consistent with our first prediction, no allozyme variation was found in both species. Although this finding can be partially attributed to their shared phylogenetic, ecological, and life-history traits (they are congeners that usually co-occur, and both are self-compatible with similar seed-dispersal mechanisms), we believe that their lack of allozyme variation support the first scenario; extant Korean populations of U. bifida and U. caerulea would have been derived from a single refugial population (a different one for each species). The genetic monomorphism (that is, all loci fixed for the same allele at all populations) detected for both species suggests that contemporary populations would have been established (or recolonized) by a single introduction event from an ancestral population (perhaps located in southern Japan or in southern China) with low (or no) genetic variability (Chung et al., 2013 and references therein). The lack of genetic polymorphism at both U. bifida and U. caerulea is, in addition, unlikely attributable to a common trait of the whole genus; in other words, that the genus Utricularia originally had low genetic diversity. Araki (2000) examined allozyme variation in the common, aquatic bladderwort U. australis across Japan (except Hokkaido and Ryukyu Islands) and found four polymorphic loci (Adh, Pgi-2, Pgm-1, and Pgm-2) at three enzyme systems (ADH, PGI, and PGM) (Fig. 1 in Araki [2000]), suggesting that at least some species within the genus are not genetically depauperate.

How the two terrestrial bladderworts recolonized the Korean Peninsula from southern Japan, southern China, or, perhaps, Jeju Island (Fig. 1) remains a matter of debate. Nevertheless, seeds of Utricularia species are tiny (0.2–0.6 mm long) and have large testa cells with a foveolate surface, features that are known to be favorable to the aggregation of air bubbles and thereby facilitating their floatability (i.e., increasing buoyancy in air or water) (Eriksson and Kainulainen, 2011; Menezes et al., 2014). Given these traits, it is likely that seeds were infrequently transported from southern Japan or southern China by strong gusts of wind (e.g., typhoons, which are very common in East Asia) from early summer to autumn, the flowering and fruiting time of U. bifida and U. caerulea. It is also possible that, for the case of the refugial population being located in southern China, seed transport to Korea was waterborne through the Kuroshio, the largest current in the Western Pacific (Chung et al., 2013). Although Kuroshio’s flow into the Korean Peninsula was significantly reduced at the LGM, it increased progressively during the deglaciation until its current levels at the onset of the Holocene (Kao et al., 2006). Seed dispersal through this current would have been possible during most of the deglaciation period.

Common features shared by plant taxa lacking allozyme variation on the Korean Peninsula

To date, studies on 16 plant taxa native to the Korean Peninsula that lack allozyme variation are available. In addition to the two Utricularia species studied herein, up to six orchid species (Habenaria dentata, Liparis pterosepala, Oreorchis coreana, Pecteilis radiata, Peristylus densus, and Tipularia japonica), plus the sundew Drosera peltata var. nipponica (Table 1) have a series of common features: (1) they are warm-temperate elements in East Asia (i.e., southern Korean Peninsula is one of their northern limits), (2) they are rare within the peninsula (suggesting that the occurrence of these species in Korea might be the result of a rare event of long-distance seed dispersal), and (3) they have tiny, dust-like seeds (indicating a potential for long-distance dispersal) (reviewed in Chung et al., 2017b).

Table 1.

Plant taxa native in South Korea lacking allozyme diversity.

The other five orchid species listed in Table 1 (Cephalanthera subaphylla, Cypripedium japonicum, Epipactis papillosa, Goodyera repens, and Liparis kumokiri) are, in contrast, boreal and/or temperate elements of Northeastern Asia (Chung et al., 2017b), and are also rare on the Korean Peninsula (except for L. kumokiri). We assume that these boreal and/or temperate orchid species, although likely managed to persist in situ on the Korean Peninsula at the LGM, would have done just in some favorable enclaves (microrefugia; Rull, 2009) (see Chung et al., 2017a, 2017b).

The two herbaceous perennials Lycoris radiata var. radiata and Sedum sarmentosum also lack allozyme variation (Table 1). The common features observed in these two plant taxa are triploidy (sterility) and, perhaps, a single introduction to the Korean Peninsula by vegetative means (e.g., bulbs for L. radiata var. radiata; shoots for S. sarmentosum).

In sum, the common features observed in the sexually-reproducing 14 species (among them 11 are orchids) are largely responsible of their lack of allozyme variation, primarily through founder effects (e.g., post-glacial migration) and bottlenecks (e.g. survival in microrefugia). A future study adopting highly variable DNA and cpDNA markers (and, for warm-temperate or subtropical plant species, including samples from their LGM putative refugia) will be very useful to unravel the trajectories of population establishment and survival on the southern Korean Peninsula.