Genetic variation and genetic structure within metapopulations of two closely related selfing and outcrossing Zingiber species (Zingiberaceae)

Abstract

Habitat fragmentation strongly affects the genetic diversity of plant populations, and this has always attracted much research interest. Although numerous studies have investigated the effects of habitat fragmentation on the genetic diversity of plant populations, fewer studies have compared species with contrasting breeding systems while accounting for phylogenetic distance. Here, we compare the levels of genetic diversity and differentiation within and among subpopulations in metapopulations (at fine-scale level) of two closely related Zingiber species, selfing Zingiber corallinum and outcrossing Zingiber nudicarpum. Comparisons of the genetic structure of species from unrelated taxa may be confounded by the effects of correlated ecological traits or/and phylogeny. Thus, we possibly reveal the differences in genetic diversity and spatial distribution of genetic variation within metapopulations that relate to mating systems. Compared to outcrossing Z. nudicarpum, the subpopulation genetic diversity in selfing Z. corallinum was significantly lower, but the metapopulation genetic diversity was not different. Most genetic variation resided among subpopulations in selfing Z. corallinum metapopulations, while a significant portion of variation resided either within or among subpopulations in outcrossing Z. nudicarpum, depending on whether the degree of subpopulation isolation surpasses the dispersal ability of pollen and seed. A stronger spatial genetic structure appeared within subpopulations of selfing Z. corallinum potentially due to restricted pollen flow and seed dispersal. In contrast, a weaker genetic structure was apparent in subpopulations of outcrossing Z. nudicarpum most likely caused by extensive pollen movement. Our study shows that high genetic variation can be maintained within metapopulations of selfing Zingiber species, due to increased genetic differentiation intensified primarily by the stochastic force of genetic drift among subpopulations. Therefore, maintenance of natural variability among subpopulations in fragmented areas is key to conserve the full range of genetic diversity of selfing Zingiber species. For outcrossing Zingiber species, maintenance of large populations is an important factor to enhance genetic diversity. Compared to outcrossing Z. nudicarpum, the subpopulation genetic diversity in selfing Z. corallinum was significantly lower, but the metapopulation genetic diversity did not differ. Most genetic variation resided among subpopulations in selfing Z. corallinum metapopulations, while a significant portion of variation resided either within or among subpopulations in outcrossing Z. nudicarpum, depending on whether the degree of subpopulation isolation surpasses the dispersal ability of pollen and seed. Our study shows that selfing Z. corallinum could maintain high genetic diversity through differentiation intensified primarily by the stochastic force of genetic drift among subpopulations at fine-scale level, but not local adaptation.

Keywords: Fragmentation; Zingiber corallinum; Zingiber nudicarpum; genetic differentiation; mating system.

Figure 1.

Location of the metapopulations and distribution of subpopulations in metapopulations of Zingiber corallinum (A—GDZJ; B—GDYX) and Z. nudicarpum (C—HNCJ; D—HNBT). Each number (1–4) represents one subpopulation while each dot represents a sampled individual, and blue lines indicate streams. The map was drawn by the authors with reference to Google Maps (https://maps.google.com/).

Figure 2.

Distribution of gene frequency in metapopulations of Zingiber corallinum (A—GDZJ; B—GDYX) and Z. nudicarpum (C—HNCJ; D—HNBT) and their mean values (E).

Figure 3.

Metapopulations’ genetic group structure analysis by InStruct for Zingiber corallinum (A—GDZJ, K = 2; B—GDYX, K = 4) and STRUCTURE for Z. nudicarpum (C—HNCJ, K = 2; D—HNBT, K = 3). Each individual vertical bar represents an individual and the black vertical bars separate the subpopulations, while different colours represent different gene pools.

Figure 4.

UPGMA dendrogram based on Dice coefficient for individuals in metapopulations of Zingiber corallinum (A—GDZJ; B—GDYX).

Figure 5.

UPGMA dendrogram based on Dice coefficient for individuals in metapopulations of Zingiber nudicarpum (A—HNCJ; B—HNBT).

Figure 6.

Correlogram showing the spatial autocorrelation coefficient (r) within subpopulations in metapopulations of Zingiber corallinum and Z. nudicarpum. U and L represent the 95 % two-tailed confidence interval, which was calculated based on 999 permutations; (A–D) subpopulations 1–4 in the GDYX metapopulation of Z. corallinum; (E and F) subpopulations 1–2 in the HNCJ metapopulation of Z. nudicarpum.

Figure 7.

UPGMA dendrogram (A) based on Dice coefficient and unrooted NJ tree (B) based on Nei’s genetic distance for 45 seedlings and all adults within the GDZJ-2 subpopulation in the GDZJ metapopulation of Zingiber corallinum (M1–M5, potential maternal plants; M1-1–M5-4, seedlings; 2-1–2-39, other adult plants).