A historical stepping-stone path for an island-colonizing cactus across a submerged "bridge" archipelago

Abstract

Here we use population genomic data (ddRAD-Seq) and ecological niche modeling to test biogeographic hypotheses for the divergence of the island-endemic cactus species Cereus insularis Hemsl. (Cereeae; Cactaceae) from its sister species C. fernambucensis Lem. The Cereus insularis grows in the Fernando de Noronha Islands (FNI), a Neotropical archipelago located 350 km off the Brazilian Atlantic Forest (BAF) coast. Phylogeographic reconstructions support a northward expansion by the common ancestor of C. insularis and C. fernambucensis along the mainland BAF coast, with C. insularis diverging from the widespread mainland taxon C. fernambucensis after colonizing FNI in the late Pleistocene. The morphologically distinct C. insularis is monophyletic and nested within C. fernambucensis, as expected from a progenitor-derivative speciation model. We tested alternative biogeographic and demographic hypotheses for the colonization of the FNI using Approximate Bayesian Computation. We found the greatest support for a stepping-stone path that emerged during periods of decreased sea level (the "bridge" hypothesis), in congruence with historical ecological niche modeling that shows highly suitable habitats on stepping-stone islands during glacial periods. The outlier analyses reveal signatures of selection in C. insularis, suggesting a putative role of adaptation driving rapid anagenic differentiation of this species in FNI.

Fig. 1. Map of eastern Brazil highlighting the study area and the locations of Cereus sampled during this study (details in Table S1).

FNI, Fernando de Noronha Islands; FNC, Fernando de Noronha Chain; BCS, Brazilian Continental Shelf; AB, Abrolhos Bank; VTC, Vitória-Trindade Chain. The dotted line highlights the seamounts of FNC. Photos: Franco, FF.

Fig. 2. Demographic models compared to test alternative scenarios for colonization of FNI by C. insularis.

Arrows represent the probability of gene flow in models with a non-zero migration rate (m ≠ 0). Below each model, a schematic demonstrates a putative geographic rationale. Model 2 (outlined by a dotted rectangle) is the best fit for the data. pp = posterior probability for each model; BF: Bayes Factor.

Fig. 3. Phylogeography and population structure.

A Maximum Likelihood tree based on the concatenated supermatrix (522,328 bp, 10,584 SNPs; 25.6% missing data) generated using RaxML. B, C STRUCTURE barplot (k = 6) and PCA scatterplots showing population structure. D Population sampling locations are colored by the majority population cluster assignment. E Summary of Bayesian spatiotemporal diffusion (spherical continuous phylogeography analysis; see Fig. S4) for the six discrete genetic clusters, showing a northward range expansion along restinga forest, eventually reaching Fernando de Noronha Island (FNI). F Bayesian estimation of node ages under the multispecies coalescent model and a relaxed molecular clock in BPP. Asterisks next to taxon names in (B) indicate non-admixed individuals selected for divergence time estimation.

Fig. 4. Highlights of the Ecological niche modeling (ENM) of Cereus fernambucensis and C. insularis (first and second columns, respectively).

All periods modeled can be found in Figure S7. Warmer colors on the SDM maps indicate more suitable areas for species occurrence, highlighting the Fernando de Noronha Chain (FNC), including the Fernando de Noronha Islands (FNI) and Rocas Atoll (RA).

Fig. 5. Global sea-level change during the last 120 ka (adapted from Rijsdijk et al. 2014).

The gray bar represents the 95% Highest Posterior Density (HPD) interval (46 Ka–125 Ka), and the black dot is the mean for the stem age of C. insularis (79 Ka). The dashed line represents the hypothesized sea level threshold (−80 m) associated with the presence or absence of the transient land bridge in the Fernando de Noronha Chain (FNC), as determined by our paleomodeling (Fig. 4) and assuming the same impact along the late Pleistocene. t1, putative period of long-term colonization followed by divergence facing recurrent gene flow; t2, putative period of maximum extension of the transient bridge in the LGM followed by gradual disconnection (Fig. 4); t3, isolation and further differentiation. TMRCA, Time to the Most Recent Common Ancestor.