Introduction
To what range a species can distribute and how its intraspecific genetic variations are structured are fundamentally determined by geographical and ecological factors (Melville & Burchett, 2002). Various forms of geographical barriers, on the one hand, constrain the margins of a species’ distribution ranges and, on the other hand, cause abrupt discontinuities within the range (Hartl & Clark, 1997). To the coastal plants that disperse propagules by seawater, the geographical barriers are usually landmasses, open oceans (Thornhill, Mahon, Norenburg, & Halanych, 2008), and ocean currents. Moreover, limitations of dispersal ability may further shape their populations, following a pattern called isolation by distance (IBD), which assumes that the level of differentiation is positively correlated with geographic distance.
The geographic landscape and ecological landscape are not persistent throughout history, making the populations of organisms highly dynamic. Geological actions had changed the geographical landscape gradually in a macroevolutionary timescale, such as the uplift of the Qinghai-Tibetan Plateau (Ramstein, Fluteau, Besse, & Joussaume, 1997; Zhisheng, Kutzbach, Prell, & Porter, 2001) and the closure of the Central America isthmus (Lessios, 2008). In a smaller timescale, the historic climate changes, e.g. the Quaternary glaciers, had reshaped the Earth continuously. For coastal plants, the historic sea-level changes due to climate changes had caused geographic barriers to emerge and vanish repeatedly. The emergence of a barrier may subdivide populations while the vanish of a barrier may reunite diverging populations (Ge et al., 2015). Hence, the population structure within a species is an integrated output of the forces dividing populations and those mixing populations.
Other than isolating and connecting populations, historic geographical and climatic changes also shaped the demographic size of populations. Typically, the populations of a species contract to refugia or even become extinct when geographic or ecological conditions are too hostile to survive (Foufopoulos, Kilpatrick, & Ives, 2011; Hallam & Wignall, 1999; Jackson, Winston, & Coates, 1985; Ozawa, 2010; Paulay, 1990), and recover and expand when the conditions are ameliorated. During these processes of “shrinkage-expansion” (i.e., bottleneck), even when the species survive, a reduction in intraspecific diversity is common (Haanes, Røed, Flagstad, & Rosef, 2010; Mamuris, Stoumboudi, Stamatis, Barbieri, & Moutou, 2005; Moum & Árnason, 2001; Tsuchida, Kudô, & Ishiguro, 2014). In addition to reducing intraspecific diversity, how would such a bottleneck process influence the population structure within a species? During the “shrinkage-expansion” process, the refugial population usually sampled only a small subset of the total ancestral polymorphisms, and the small population is more likely to lose inherent polymorphisms and fix new mutations due to intensified genetic drift and relaxed purifying selection. Moreover, the contracted population is usually less likely to exchange genes with other populations due to enhanced isolation. Hence, we hypothesized the bottleneck process is highly potential in generating novel genetic structures within species.
Several previous studies have referred to this issue indirectly. In a coral reef fish, the strong nonequilibrium genetic structure was shown to be generated by genetic bottlenecks/founder effects associated with population reduction/extinctions and asymmetric migration or recolonization (Bay, Caley, & Crozier, 2008). Repeated bottleneck events during colonization of the parasite Geomydoecus aurie have been shown to impart genetic structure of a population due to allele surfing (Demastes, Hafner, Hafner, Light, & Spradling, 2019). The fine-scale genetic structure of Rhizophora racemosa in the Cameroon estuary complex was attributed to contemporary processes such as restricted propagule dispersal, bottleneck events, and recolonization by founders from ancient local refugia (Ngeve, Van der Stocken, Menemenlis, Koedam, & Triest, 2017). Despite the awareness that bottleneck may generate genetic structure, the hypothesis needs to be comprehensively and explicitly addressed.
The model with which to test this hypothesis should live in a highly dynamic habitat and be sensitive to habitat change. Mangrove plants, a group of typical coastal plants, are therefore an ideal model. Mangrove plants are distributed linearly along coasts and strictly inhabit tropical and subtropical intertidal zones. In the Quaternary cycles of glacial and interglacial conditions, climate changes drove the sea level to change in cycles (Miller et al., 2005; Zachos, Pagani, Sloan, Thomas, & Billups, 2001). The coastlines shifted following sea-level rise and drop, and mangrove forests were forced to advance and retreat repeatedly (Woodroffe & Grindrod, 1991). Hence, we tested the above hypothesis in mangrove plants, but the underlying reasonability may be general in other taxa.
Several studies had determined the genetic structure of different mangrove species, with periodic geographical barriers frequently identified in places such as the Malay Peninsula and the Wallacea region (W. Guo et al., 2018; Z. Guo et al., 2016; J. Li et al., 2016; Urashi, Teshima, Minobe, Koizumi, & Inomata, 2013; Y. Yang et al., 2016, 2017), in addition to a set of permanent barriers (Duke, 2017). As the historic sea-level changes eroded barriers periodically, populations of mangrove species were isolated or connected by intermittent gene flow, which was previously employed to illustrate the observed population structure (divergence and admixture) of many coastal species (Banerjee et al., 2020, 2021; Z. Guo, Guo, et al., 2018; Z. Guo et al., 2016; Westberg & Kadereit, 2009). Such intermittent gene flow was demonstrated to have promoted the speciation of mangrove species via a “mixing-isolation-mixing” mechanism (Z. He et al., 2019). However, it’s unclear how demographic size changes driven by sea-level changes have shaped population structure, although some studies have also reported that bottleneck events were involved in some mangrove species’ demographic histories and caused population diversity reductions (W. Guo et al., 2020, 2018; Zhou et al., 2010). It’s intriguing to test the hypothesis that bottleneck events could promote population subdivision. We performed such a study in the typical mangrove plant Aegiceras corniculatum , which is always a frontier species in mangrove forests.
Aegiceras corniculatum is distributed widely from India across Southeast Asia and South China to Australia and west Pacific islands (Duke, 2014), likely facilitated by its long-distance dispersal via buoyant propagules. Ge and Sun et al. had used inter-simple sequence repeat markers to reveal between-population differentiation in the populations of A. corniculatum in China (Ge & Sun, 1999). Deng et al. had used amplified fragment length polymorphism markers to reveal genetic divergences among populations in China, Malay Peninsula, and Sri Lanka (Deng et al., 2009). However, the genetic pattern of its whole distribution range remains unaddressed. We sampled 18 populations, covering the distribution range of A. corniculatum , and used both Sanger and Illumina Solexa sequencing to obtain single nucleotide polymorphism (SNP) markers, with which we could ascertain the population structure of A. corniculatum across its distribution range. Due to the dominant role of Sundaland and Wallacea barrier in coastal plants in the IWP, we expected to see the populations of A. corniculatumfrom the Indian Ocean, Southeast Asia (including South China), and Australasia is exclusively grouped. Moreover, we expected the historic sea-level changes had shaped demographic size changes of A. corniculatum populations, particularly, bottleneck event in history is expectable. Particular interest was devoted to whether bottleneck events had played a role in generating additional population structure not attributed to geographic barriers. The findings will deepen our understanding of population evolution in dynamic geographic and climate changes and guide efforts to conserve genetic diversity below the species level.