Limited phylogeographic and genetic connectivity in Acacia species of low stature in an arid landscape

Abstract

Widespread plant species are expected to maintain genetic diversity and gene flow via pollen and seed dispersal. Stature is a key life history trait that affects seed and potentially pollen dispersal, with limited stature associated with limited dispersal and greater genetic differentiation. We sampled Hill's tabletop wattle (Acacia hilliana) and curry wattle (Acacia spondylophylla), two co-distributed, widespread, Acacia shrubs of low stature, across the arid Pilbara region of north-western Australia. Using chloroplast sequence and nuclear microsatellite data we evaluated patterns of population genetic and phylogeographic diversity and structure, demographic signals, ratios of pollen to seed dispersal, evidence for historical refugia, and association between elevation and diversity. Results showed strong phylogeographic (chloroplast, G _ST = 0.831 and 0.898 for A. hilliana and A. spondylophylla, respectively) and contemporary (nuclear, F _ST = 0.260 and 0.349 for A. hilliana and A. spondylophylla, respectively) genetic structure in both species. This indicates limited genetic connectivity via seed and pollen dispersal associated with Acacia species of small stature compared to taller tree and shrub acacias across the Pilbara bioregion. This effect of stature on genetic structure is superimposed on moderate levels of genetic diversity that were expected based on widespread ranges (haplotype diversity h = 25 and 12; nuclear diversity He = 0.60 and 0.47 for A. hilliana and A. spondylophylla, respectively). Contemporary genetic structure was congruent at the greater landscape scale, especially in terms of strong genetic differentiation among geographically disjunct populations in less elevated areas. Measures of diversity and connectivity were associated with traits of greater geographic population proximity, population density, population size, and greater individual longevity, and some evidence for range expansion in A. hilliana. Results illustrate that low stature is associated with limited dispersal and greater patterns of genetic differentiation for congenerics in a common landscape and highlight the complex influence of taxon-specific life history and ecological traits to seed and pollen dispersal.

Keywords: Acacia; Pilbara; arid zone; genetic connectivity; genetic diversity; stature.

FIGURE 1

Geographic distribution of sampled populations, chloroplast DNA haplotypes, and evolutionary relationships amongst chloroplast DNA haplotypes of (a) Acacia hilliana and (b) Acacia spondylophylla in the Pilbara bioregion. Populations are represented by pie charts. Segment size within each pie‐chart corresponds to the proportion of individuals with that haplotype. Population codes correspond to those in Table 1. Greyscale indicates elevation. Top inset shows species known geographic distribution as blue circles. Pilbara bioregion indicated. Bottom inset Median‐Joining networks show the evolutionary relationships between cpDNA haplotypes in sampled populations. Haplotypes are represented by colored circles. Color of circle corresponds to haplotypes in the geographic map. Circle size is proportional to the number of individuals with that haplotype and branch lengths are proportional to the number of mutations with lengths greater than one indicated by cross bars. Small black circles indicate unsampled hypothetical haplotypes. Panels to the right show the prostrate nature of representative individuals

FIGURE 2

Assignment of individuals and populations to genetic clusters identified via Bayesian analysis of multilocus nuclear microsatellite genotype data for Acacia hilliana in the Pilbara bioregion. (a) Mean values of the ad hoc test statistic Ln probability of the data (L(K)). (b) Values of the ad hoc test statistic Delta K (ΔK). (c) Bar plot indicating the assignment of individuals and populations to one of nine genetic clusters. Each individual is represented as a vertical line partitioned into nine segments whose length is proportional to the individual coefficients of membership in each of nine clusters. Population codes on the x‐axis correspond to those in Table 1. (d) Geographic distribution of genetic clusters. Colors correspond to those in (c). Sections of pie charts show the proportion of assignment of the given population to each genetic subcluster. Labels correspond to population codes in Table 1

FIGURE 3

Assignment of individuals and populations to genetic clusters identified via Bayesian analysis of multilocus nuclear microsatellite genotype data for Acacia spondylophylla in the Pilbara bioregion. (a) Mean values of the ad hoc test statistic ln probability of the data (L(K)), values of the ad hoc test statistic DeltaK (ΔK), and bar plot indicating the assignment of individuals and populations to one of two optimally determined genetic clusters. Each individual is represented as a vertical line partitioned into two segments whose length is proportional to the individual coefficients of membership in each of two clusters. (b) Hierarchical clustering analysis of populations from Cluster One (Orange) in (a). Mean values of the ad hoc test statistic ln probability of the data (L(K)), values of the ad hoc test statistic DeltaK (ΔK), and bar plot indicating the assignment of individuals and populations to one of two optimally determined genetic clusters. Each individual is represented as a vertical line partitioned into two segments whose length is proportional to the individual coefficients of membership in each of two clusters. (c) Hierarchical clustering analysis of populations from Cluster Two (Blue) in (a). Mean values of the ad hoc test statistic ln probability of the data (L(K)), values of the ad hoc test statistic DeltaK (ΔK), and bar plot indicating the assignment of individuals and populations to one of nine optimally determined genetic clusters. Each individual is represented as a vertical line partitioned into nine segments whose length is proportional to the individual coefficients of membership in each of nine clusters. (d) Geographic distribution of eleven genetic clusters from the hierarchical analysis. Colors correspond to those in (b) and (c). Sections of pie charts show the proportion of assignment of the given population to each genetic subcluster. All population codes correspond to those in Table 1

FIGURE 4

Principal co‐ordinates analysis of the genetic distance between sampled individuals of (a) Acacia hilliana and (b) Acacia spondylophylla in the Pilbara bioregion. Population codes correspond to those in Table 1

FIGURE 5

Heatmaps of pairwise population genetic divergence (F _ST) for nuclear microsatellite data for (a) Acacia hilliana and (b) Acacia spondylophylla in the Pilbara bioregion. Population codes correspond to those in Table 1