Phylogeographic Analysis and Genetic Structure of an Endemic Sino-Japanese Disjunctive Genus Diabelia (Caprifoliaceae)

Abstract

The Sino-Japanese Floristic Region (SJFR) is a key area for plant phylogeographical research, due to its very high species diversity and disjunct distributions of a large number of species and genera. At present, the root cause and temporal origin of the discontinuous distribution of many plants in the Sino-Japanese flora are still unclear. Diabelia (Caprifoliaceae; Linnaeoideae) is a genus endemic to Asia, mostly in Japan, but two recent discoveries in China raised questions over the role of the East China Sea (ECS) in these species' disjunctions. Chloroplast DNA sequence data were generated from 402 population samples for two regions (rpl32-trnL, and trnH-psbA) and 11 nuclear microsatellite loci were screened for 549 individuals. Haplotype, population-level structure, combined analyses of ecological niche modeling, and reconstruction of ancestral state in phylogenies were also performed. During the Last Glacial Maximum (LGM) period after the Tertiary, Diabelia was potentially widely distributed in southeastern China, the continental shelf of the East China Sea and Japan (excluding Hokkaido). After LGM, all populations in China have disappeared except those in Zhejiang which may represent a Glacial refuge. Populations of Diabelia in Japan have not experienced significant bottleneck effects, and populations have maintained a relatively stable state. The observed discontinuous distribution of Diabelia species between China and Japan are interpreted as the result of relatively ancient divergence. The phylogenetic tree of chloroplast fragments shows the characteristics of multi-origin evolution (except for D. sanguinea). STRUCTURE analysis of nuclear Simple Sequence Repeat (nSSR) showed that the plants of the Diabelia were divided into five gene pools: D. serrata, D. spathulata, D. sanguinea, D. ionostachya (D. spathulata var. spathulata-Korea), and populations of D. ionostachya var. ionostachya in Yamagata prefecture, northern Japan. Molecular evidence provides new insights of Diabelia into biogeography, a potential glacial refuge, and population-level genetic structure within species. In the process of species differentiation, ECS acts as a corridor for two-way migration of animals and plants between China and Japan during glacial maxima, providing the possibility of secondary contact for discontinuously distributed species between China and Japan, or as a filter (creating isolation) during glacial minima. The influence of the ECS in speciation and biogeography of Diabelia in the Tertiary remains unresolved in this study. Understanding origins, evolutionary histories, and speciation will provide a framework for the conservation and cultivation of Diabelia.

Keywords: Caprifoliaceae; Diabelia; Sino-Japanese disjunct distribution; ecological niche modeling; phylogeography.

Figure 1

Species range, geographical distribution of haplotypes, and phylogenetic relationship among four Diabelia species. Geographical distribution of 37 chloroplast haplotypes in sampled locations. Shading indicates the native range of each species (Hara, ; Landrein and Farjon, 2019). Four haplotypes (H13, H25, H28, and H29) are multiple species shared haplotypes, of which H25 is shared haplotype of D. ionostachya, D. sanguinea, and D. spathulata. H13 is the shared haplotype for D. ionostachya and D. serrata; H28 is the shared haplotype for D. ionostachya and D. sanguinea; H29 is the shared haplotype for D. ionostachya and D. spathulata, respectively.

Figure 2

Maximum Likelihood tree inferred from RAxML for 37 haplotypes in Diabelia based on concatenated sequences of chloroplast DNA (trnH-psbA, rpl32-trnL) with Dipelta yunnanensis as the outgroup. Numbers associated with branches are ML bootstrap support values. Population migration direction inferred (a, b, and c) from haplotype phylogenetic tree of Diabelia.

Figure 3

Results of STRUCTURE analysis for all individuals of Diabelia studied based on nSSR data. Assignment of individuals to genetic clusters at K = 5.

Figure 4

Comparison of potential distributions as probability of occurrence for Diabelia serrata, D. spathulata, D. sanguinea, and D. ionostachya (Landrein and Farjon, 2019), at the CCSM climatic scenarios of the Last Glacial Maximum (LGM, ca. 21,000 years BP). The maximum training sensitivity plus specificity logistic threshold has been used to discriminate between suitable (cutline 0.1–1 area) and unsuitable habitat. The darker color indicates a higher probability of occurrence.