Climatic niche pre-adaptation facilitated island colonization followed by budding speciation in the Madeiran ivy (Hedera maderensis, Araliaceae)

Abstract

The path followed by species in the colonization of remote oceanic islands ultimately depends on their phylogenetic constraints and ecological responses. In this study, we aim to evaluate the relative role of geographical and ecological forces in the origin and evolution of the Madeiran ivy (Hedera maderensis), a single-species endemic belonging to the western polyploid clade of Hedera. To determine the phylogenetic placement of H. maderensis within the western polyploid clade, we analyzed 40 populations (92 individuals) using genotyping-by-sequencing and including Hedera helix as outgroup. Climatic niche differences among the study species were evaluated using a database with 867 records representing the entire species ranges. To test species responses to climate, 13 vegetative and reproductive functional traits were examined for 70 populations (335 individuals). Phylogenomic results revealed a nested pattern with H. maderensis embedded within the south-western Iberian H. iberica. Gradual niche differentiation from the coldest and most continental populations of H. iberica to the warm and stable coastal population sister to H. maderensis parallels the geographical pattern observed in the phylogeny. Similarity in functional traits is observed for H. maderensis and H. iberica. The two species show leaves with higher specific leaf area (SLA), lower leaf dry matter content (LDMC) and thickness and fruits with lower pulp fraction than the other western polyploid species H. hibernica. Acquisition of a Macaronesian climatic niche and the associated functional syndrome in mainland European ivies (leaves with high SLA, and low LDMC and thickness, and fruits with less pulp content) was a key step in the colonization of Madeira by the H. iberica/H. maderensis lineage, which points to climatic pre-adaptation as key in the success of island colonization (dispersal and establishment). Once in Madeira, budding speciation was driven by geographical isolation, while ecological processes are regarded as secondary forces with a putative impact in the lack of further in situ diversification.

Keywords: Hedera; Madeira; budding speciation; climatic pre-adaptation; functional traits; genotyping-by sequencing; geographic isolation; island colonization.

FIGURE 1

Occurrence-based map of the distribution ranges of the three species of the western polyploidy clade of Hedera (H. hibernica, H. iberica, H. maderensis) and H. helix. Circles indicate samples included for the climatic niche study. Stars indicate samples also included in the phylogenetic study.

FIGURE 2

Phylogenetic tree of the western polyploid clade of Hedera based on phylogenomic analysis of GBS data (c80m15p6r3), and using H. helix as outrgoup. The maximum-likelihood tree obtained in RAxML is shown, with bootstrap support values ≥75% indicated above branches. Posterior probability values obtained in the Bayesian analysis in ExaBayes are shown below branches. The inset shows a map of the Iberian Peninsula and Madeira indicating sampling localities for the study of functional traits and the geographical lineages of H. iberica.

FIGURE 3

Climatic niche ordination analysis. (A) Climatic niche space of the three species of the western polyploidy clade of Hedera (H. hibernica, H. iberica, H. maderensis) and H. helix as inferred from their entire geographical ranges. The colors of the density envelopes correspond to species, with shading intensity indicating the density based on point occurrences. The solid black line delimits the multivariate space corresponding to the 100% of the available environment for all the species, and the dotted line contains the 95% of the available environment. (B) Climatic space occupied by the geographical lineages of H. iberica according to the GBS phylogeny (Figure 2). Contributions of the original WorldClim variables are provided: bio03, isothermality; bio07, temperature annual range; bio08, mean temperature of wettest quarter; bio09, mean temperature of driest quarter; bio10, mean temperature of warmest quarter; bio11, mean temperature of coldest quarter; bio15, precipitation seasonality; bio16, precipitation of wettest quarter; bio18, precipitation of warmest quarter. HEL, Hedera helix; HIB, Hedera hibernica; IBE, Hedera iberica; MAD, Hedera maderensis.

FIGURE 4

Fruit (A-D), vegetative leaf (E-H), and reproductive leaf (I-L) functional trait differences among the species of the western polyploid clade of Hedera: H. hibernica (HIB), H. iberica (IBE), and H. maderensis (MAD). The crossbar within the boxplot shows the median, and the length of the box indicates the interquartile range. Shape of the violin plot reflects the kernel density plot of data. Level of significance is shown. ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, ^nsp > 0.05. Marginal significance is indicated with the corresponding p-value.