Is Amazonia a ‘museum’ for Neotropical trees? The evolution of the Brownea clade (Detarioideae, Leguminosae)

https://doi.org/10.1016/j.ympev.2018.04.029Get rights and content

Highlights

  • The Brownea clade diversified gradually, following the ‘museum’ model.

  • The Brownea clade originated in the Eocene, and diversified throughout the Neogene.

  • The clade diversified mainly in Amazonia, with subsequent migrations across the Neotropics.

Abstract

The flora of the Neotropics is unmatched in its diversity, however the mechanisms by which diversity has accumulated are debated and largely unclear. The Brownea clade (Leguminosae) is a characteristic component of the Neotropical flora, and the species within it are diverse in their floral morphology, attracting a wide variety of pollinators. This investigation aimed to estimate species divergence times and infer relationships within the group, in order to test whether the Brownea clade followed the ‘cradle’ or ‘museum’ model of diversification, i.e. whether species evolved rapidly over a short time period, or gradually over many millions of years. We also aimed to trace the spatio-temporal evolution of the clade by estimating ancestral biogeographical patterns in the group. We used BEAST to build a dated phylogeny of 73 Brownea clade species using three molecular markers (ITS, trnK and psbA-trnH), resulting in well-resolved phylogenetic relationships within the clade, as well as robust divergence time estimates from which we inferred diversification rates and ancestral biogeography. Our analyses revealed an Eocene origin for the group, after which the majority of diversification happened in Amazonia during the Miocene, most likely concurrent with climatic and geological changes caused by the rise of the Andes. We found no shifts in diversification rate over time, suggesting a gradual accumulation of lineages with low extinction rates. These results may help to understand why Amazonia is host to the highest diversity of tree species on Earth.

Introduction

The tropical Americas are home to a great number of seed plant species (ca. 100,000 species (Antonelli and Sanmartín, 2011, Hughes et al., 2013)), however the evolutionary and biogeographical forces which led to the formation of this superlative diversity are unclear for many groups. The time period over which this diversity has accumulated has been a topic of much debate, largely surrounding two theories. The first is the ‘museum’ model of diversification, whereby radiations dating from the Neogene or earlier have persisted in Amazonia due to climatic stability and niche conservatism, allowing many thousands of lineages to accumulate over the past 30 million years (Antonelli and Sanmartín, 2011). By contrast, the ‘cradle’ model of diversification suggests that geologically recent events, such as the orogeny of the Andes and the rise of the Panama isthmus, triggered large scale climatic, geographical and edaphic changes, resulting in rapid speciation due to reproductive isolation (Richardson et al., 2001, Hughes and Eastwood, 2006). It is worth noting that these theories are not mutually exclusive, and indeed both are evident in different lineages, such as Bromeliaceae (Givnish et al., 2011), Orchidaceae (Pérez-Escobar et al., 2017a, Pérez-Escobar et al., 2017b) and Annonaceae (Couvreur et al., 2011). In addition to this, the interactions between speciation, extinction and geographical range fluctuation are also considered to be important in the accumulation of diversity (Jablonski et al., 2017).

In the low-to-mid elevation forests of the Neotropics plant families show a bipartite pattern of diversity (Gentry, 1982). The first of these two major portions of Neotropical plant diversity includes the Andean-centred taxa, which are those lineages whose centre of diversity lies in North-western South America, within the foothills, valleys and montane zones of the Andes. The majority of the species in this portion are epiphytes and understory shrubs (Gentry and Dodson, 1987). The second portion includes the Amazonian-centred taxa, whose taxonomic diversity is highest in the Amazon basin. The majority of the species in this portion are canopy trees or lianas (Gentry, 1982, Prance, 1994). Several overlapping hypotheses are used to explain this biogeographical pattern (known as the ‘Gentry Pattern’ (Antonelli and Sanmartín, 2011)), the most prominent of which relates the rapid, punctuated rise of the Andes to the evolution of the Neotropical flora (Gentry, 1982). For the Amazonian-centred taxa it is well documented that the Andean orogeny drove much of the diversification of the modern flora (Hoorn et al., 2010), as mountain building processes would have driven large scale hydrological and climatic changes throughout what is now Amazonia (Gregory-Wodzicki, 2000).

Around 50 Ma, during the Eocene, the lowland forests of northern South America were much more extensive than in modern times, with an area known as ‘pan-Amazonia’ stretching over much of South America unimpeded (Hoorn et al., 2010). The Andean orogeny, beginning in earnest around the early Miocene (∼23 Ma), would have caused the formation of the Pebas system, an enormous matrix of wetland and terra firme forest, and subsequently the Acre system (another vast area of wetlands in what is now western Amazonia) which could have driven speciation in situ through allopatry (Hoorn et al., 2010, Antonelli and Sanmartín, 2011). During the Miocene, as the Andes rose, the drainage of the Amazon basin changed direction, resulting in a west-to-east flow; this would have gradually reduced the size of the Pebas and subsequent Acre wetlands until their disappearance. As well as this, fluvial regimes arising from the Andes would have deposited nutrient-rich sediments into the western Amazon basin, resulting in an ‘edaphic mosaic’ on a large scale (Hoorn et al., 2010). This high edaphic heterogeneity was likely also a strong driver for local adaptation and concurrent speciation (Pennington and Lavin, 2016). As such, hydrological regimes, edaphic heterogeneity and dispersal with subsequent diversification of taxa into novel niches (Erkens et al., 2007) may help to explain the hyperdiversity of tree species found in the western Amazon (Valencia et al., 2004).

Trees belonging to the Leguminosae (alternatively ‘Fabaceae’) are among the most common species found in Neotropical rainforests (Terborgh and Andresen, 1998). As such, they are an excellent system for studying the evolution of plant species in the Neotropics. The Brownea clade, which belongs to the legume subfamily Detarioideae (Legume Phylogeny Working Group (LPWG), 2017) is made up of 111 species belonging to nine genera (Mackinder, 2005), which are all sub-canopy or canopy-level trees found in lowland Neotropical rainforests (Redden and Herendeen, 2006). While the majority of species in the Brownea clade occur in Amazonia, there are several species that are endemic to only one region, such as the Chocó-Darien moist forests or Central America. The seeds of most species within the clade are dispersed by explosively dehiscing woody pods, and as such many species are locally common, forming stands (Klitgaard, 1991), with relatively low dispersal. A few species, such as Macrolobium acaciifolium, are, however, very widely dispersed and are ‘hyperdominant’ in Amazonia (ter Steege et al., 2013) due to their hydrophilic habitat preference and floating fruits. The Brownea clade is particularly diverse in floral morphology and attracts a wide variety of pollinators. Many species are adapted to attract vertebrate pollinators (mainly hummingbirds and bats) (Knudsen and Klitgaard, 1998, Fleming et al., 2009), although exactly which species is visited by which pollinator is as yet unclear for most species. These factors make the Brownea clade an interesting system for testing evolutionary hypotheses relating to the diversification of Neotropical trees. However, inter-and-infrageneric relationships are as yet poorly resolved in the Brownea clade. Previous work has mainly focussed on resolving tribes within the traditionally circumscribed subfamily Caesalpinioideae (which included the Brownea clade (Legume Phylogeny Working Group (LPWG), 2017)) using plastid trnL and matK sequences (Bruneau et al., 2001, Bruneau et al., 2008), or focussed on several genera within the Brownea clade (namely Paloue, Paloveopsis, Elizabetha and Heterostemon) using plastid trnL sequences, nuclear ITS sequences and morphological data (Redden et al., 2010). In order to study the patterns and rates of evolution within the Brownea clade a multilocus, dated phylogeny is required, from which phylogenetic patterns, divergence time estimates and rates of evolution may be inferred. This will provide some insight into the evolution of the hyperdiverse tree flora of the Neotropics, especially within Amazonia.

This investigation aims to answer major questions surrounding the evolution of the Brownea clade, and by extension how a major portion of the Neotropical tree flora evolved:

  • (1)

    Did the diversification of the Brownea clade follow the ‘museum’ or the ‘cradle’ model of lineage accumulation; that is, was diversification gradual, resulting in the accumulation of lineages over long time scales, or was diversification more recent and rapid?

  • (2)

    Did the Brownea clade diversify under a specific regime of evolutionary rates, and were there any discrete shifts in diversification rates? In other words, did the majority of the lineages within the Brownea clade accumulate through bursts of rapid speciation?

  • (3)

    Were ancestral species widely spread across many different regions, with subsequent diversification in Amazonia? Additionally, did diversification involve range shifts from Amazonia to the Andean region, or vice versa?

Section snippets

Taxon sampling, DNA sequencing and phylogenetic analyses

Our sampling represents all nine genera within the Brownea clade, comprising 73 of 111 species (excluding subspecies) and 109 accessions (Appendix A.1). A species list of the Brownea clade was compiled using the Plant List, Tropicos and generic monographs (Cowan, 1953, Klitgaard, 1991, Redden, Unpublished) in order to ensure taxonomically valid species were included in our analyses.

Analyses were run on 211 DNA sequences in total, belonging to three loci. DNA samples were extracted from 53

Phylogenetic relationships and divergence time estimates

Bayesian phylogenetic analyses reveal the Brownea clade to be well-supported and monophyletic, showing a posterior probability (PP) of 1 (Node A, shown in Fig. 2). The ancestral Brownea clade lineage diverged from the rest of the of the Detarioideae subfamily around 34 million years ago (Ma), during the Eocene, and began to diversify around 30 Ma during the Oligocene. Subsequently, the Ecuadendron/Brachycylix subclade diverged from Brownea/Browneopsis subclade around 28 Ma, during the

Diversification of the Brownea clade

Bayesian phylogenetic analyses of both nuclear and plastid sequence data resulted in a monophyletic Brownea clade (PP = 1; Node A in Fig. 2), which diverged from the rest of the Detarioideae subfamily around 34 Ma ago. Ancestral range estimations performed by de la Estrella et al. (2017) show that the Brownea clade most likely diverged from rainforest-dwelling Gondwanan ancestors, the closest relatives of which now live in the Paleotropics, with a centre of diversity in Africa. Generic

Conclusions

Overall, our analyses suggest that the Brownea clade evolved gradually over a long period, mostly within Amazonia, with subsequent occurrences of founder-event speciation into the Northern Andes and Central America. This implies that some tree species in Amazonian forests may have been subject to low extinction rates, allowing lineages to accumulate over the long time period for which Amazonia has existed, followed by dispersal and speciation into other regions of the Neotropics. This, coupled

Acknowledgements

The authors would like to thank the NERC SSCP DTP for funding the research, specifically laboratory work and herbarium trips to NY and US. Funding was also provided by a grant from the Natural Sciences and Engineering Research Council of Canada to Anne Bruneau. We would like to thank L. Fruscella at NY as well as A. Egan and M. Vatanparast at US for coordinating sampling from these herbaria. We would like to extend our thanks to L. Csiba, P. Malakasi and D. Devey for help in the laboratory, as

References (73)

  • R.H. Erkens et al.

    A rapid diversification of rainforest trees (Guatteria; Annonaceae) following dispersal from Central into South America

    Mol. Phylogenet. Evol.

    (2007)
  • A. Antonelli et al.

    Why are there so many plant species in the Neotropics?

    Taxon

    (2011)
  • G. Baele et al.

    Improving the accuracy of demographic and molecular clock model comparison while accommodating phylogenetic uncertainty

    Mol. Biol. Evol.

    (2012)
  • J.M. Beaulieu et al.

    Extinction can be estimated from moderately sized molecular phylogenies

    Evolution

    (2015)
  • A. Bruneau et al.

    Phylogenetic relationships in the Caesalpinioideae (Leguminosae) as inferred from chloroplast trnL intron sequences

    Syst. Bot.

    (2001)
  • A. Bruneau et al.

    Phylogenetic patterns and diversification in the caesalpinioid legumes

    Botany

    (2008)
  • Burnham, K.P., Anderson, D.R., 2003. Model Selection and Multimodel Inference: A Practical Information-Theoretic...
  • T.L. Couvreur et al.

    Early evolutionary history of the flowering plant family Annonaceae: steady diversification and boreotropical geodispersal

    J. Biogeogr.

    (2011)
  • R.S. Cowan

    Brachycylix, a new genus of tropical Leguminosae (Caesalpinioideae)

    Proc. Ser. C: Biol. Med. Sci.

    (1975)
  • R.S. Cowan

    A Taxonomic Revision of the Genus Macrolobium (Leguminosae-Caesalpinioideae)

    (1953)
  • D. Darriba et al.

    jModelTest 2: more models, new heuristics and parallel computing

    Nat. Methods

    (2012)
  • M. de la Estrella et al.

    Insights on the evolutionary origin of Detaroideae, a clade of ecologically dominant tropical African trees

    New Phytol.

    (2017)
  • C. Dick et al.

    Spatial scales of seed and pollen mediated gene flow in tropical forest trees

    Trop. Plant Biol.

    (2008)
  • J. Doyle et al.

    Genomic plant DNA preparation from fresh tissue-CTAB method

    Phytochem. Bull.

    (1987)
  • A.J. Drummond et al.

    Bayesian phylogenetics with BEAUti and the BEAST 1.7

    Mol. Biol. Evol.

    (2012)
  • A.J. Drummond et al.

    Relaxed phylogenetics and dating with confidence

    PLoS Biol.

    (2006)
  • T.H. Fleming et al.

    The evolution of bat pollination: a phylogenetic perspective

    Ann. Bot.

    (2009)
  • A.H. Gentry

    Neotropical floristic diversity: phytogeographical connections between Central and South America, Pleistocene climatic fluctuations, or an accident of the Andean orogeny?

    Ann. Mo. Bot. Gard.

    (1982)
  • A.H. Gentry et al.

    Diversity and biogeography of neotropical vascular epiphytes

    Ann. Mo. Bot. Gard.

    (1987)
  • T.J. Givnish et al.

    Phylogeny, adaptive radiation, and historical biogeography in Bromeliaceae: insights from an eight-locus plastid phylogeny

    Am. J. Bot.

    (2011)
  • K.M. Gregory-Wodzicki

    Uplift history of the Central and Northern Andes: a review

    Geol. Soc. Am. Bull.

    (2000)
  • S. Guindon et al.

    A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood

    Syst. Biol.

    (2003)
  • P.S. Herendeen et al.

    Reproductive and vegetative evidence for the occurrence of Crudia (Leguminosae, Caesalpinioideae) in the Eocene of southeastern North America

    Bot. Gaz.

    (1990)
  • P.S. Herendeen et al.

    Fossil legumes from the Middle Eocene (46.0 Ma) Mahenge Flora of Singida, Tanzania

    Am. J. Bot.

    (2000)
  • S.Y. Ho et al.

    Accounting for calibration uncertainty in phylogenetic estimation of evolutionary divergence times

    Syst. Biol.

    (2009)
  • C. Hoorn et al.

    Amazonia through time: andean uplift, climate change, landscape evolution, and biodiversity

    Science

    (2010)
  • C. Hughes et al.

    Neotropical plant evolution: assembling the big picture

    Bot. J. Linn. Soc.

    (2013)
  • C. Hughes et al.

    Island radiation on a continental scale: exceptional rates of plant diversification after uplift of the Andes

    Proc. Natl. Acad. Sci. USA

    (2006)
  • D. Jablonski et al.

    Shaping the latitudinal diversity gradient: new perspectives from a synthesis of paleobiology and biogeography

    Am. Nat.

    (2017)
  • S. Katoh

    MAFFT multiple sequence alignment software version 7: improvements in performance and usability

    Mol. Biol. Evol.

    (2013)
  • M. Kearse et al.

    Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data

    Bioinformatics

    (2012)
  • B.B. Klitgaard

    Ecuadorian Brownea and Browneopsis (Leguminosae-Caesalpinioideae): taxonomy, palynology, and morphology

    Nord. J. Bot.

    (1991)
  • J.T. Knudsen et al.

    Floral scent and pollination in Browneopsis disepala (Leguminosae: Caesalpinioideae) in western Ecuador

    Brittonia

    (1998)
  • L.P. Lagomarsino et al.

    The abiotic and biotic drivers of rapid diversification in Andean bellflowers (Campanulaceae)

    New Phytol.

    (2016)
  • M.J. Landis et al.

    Bayesian analysis of biogeography when the number of areas is large

    Syst. Biol.

    (2013)
  • Legume Phylogeny Working Group (LPWG): Azani, N., Babineau, M., Bailey, C.D., Banks, H., Barbosa, A.R., Pinto, R.B.,...
  • Cited by (27)

    • Glaciation-based isolation contributed to speciation in a Palearctic alpine biodiversity hotspot: Evidence from endemic species

      2018, Molecular Phylogenetics and Evolution
      Citation Excerpt :

      Studies in the past few decades concerning the speciation mechanisms of biodiversity hotspots have focused on several general hypotheses, such as the “cradle vs. museum” hypothesis (“cradle”: biodiversity hotspots are sites of high speciation rates leading to the birth of many young lineages vs. “museum”: biodiversity hotspots have low extinction rates relative to other parts of the world, leading to the retention of many old lineages) (Stebbins, 1974; Jablonski et al., 2006; Pirie et al., 2016), climatic oscillations (Lovette, 2005; Oswald et al., 2016) and alpine uplift (Hoorn et al., 2010; Lei et al., 2014; Xing and Ree, 2017). However, these general hypotheses cannot be used to explain recent speciation events in all cases (Sandel et al., 2011; Liu et al., 2016; Renner, 2016; Schley et al., 2018). Moreover, researchers normally refer to climatic oscillations such as glacial-interglacial cycles as drivers of biodiversity, but the specific roles of glaciation versus inter-glaciation periods are unclear (Lovette, 2005; Liu et al., 2016; Oswald et al., 2016).

    • From tree tops to the ground: Reversals to terrestrial habit in Galeandra orchids (Epidendroideae: Catasetinae)

      2018, Molecular Phylogenetics and Evolution
      Citation Excerpt :

      No niche differentiation was detected between terrestrial and epiphytic Galeandra, nor between long and short-spurred Galeandra (Fig. S6 and Fig. S7). Galeandra arose towards the Late Miocene, about 10 Ma, close to the climatic optimum when several plant clades in the Neotropical Region originated (Antonelli et al., 2009; Gustafsson et al., 2010; Hoorn et al., 2010; Pérez-Escobar et al., 2017c, Schey et al. 2018). At that time, the Guiana and the Brazilian Shields became large islands.

    View all citing articles on Scopus
    View full text