Historical contingency or effective niche differentiation as drivers for the emergence of endemism centres? A commentary on ‘The evolution of ecological specialization underlies plant endemism in the Atlantic Forest’

Abstract

This article comments on:

Eduardo K. Nery, Mayara K. Caddah, Matheus F. Santos and Anselmo Nogueira. The evolution of ecological specialization underlies plant endemism in the Atlantic Forest, Annals of Botany, Volume 131, Issue 6, 9 May 2023, Pages 921–940, https://doi.org/10.1093/aob/mcad029

Centres of endemism, namely geographical regions that contain a high number of narrow-range species, are of paramount importance in establishing plant conservation priorities, as they hold nearly half of all extant species of vascular plants (Myers et al., 2000). Not surprisingly, these centres have been the subject of many descriptive studies aimed at characterizing their outstanding biodiversity, yet the mechanisms favouring the accumulation of narrow-range species remain largely unknown, with fewer studies exploring the links between geographical patterns of endemism and climate, geomorphology or other factors that may relate to the accumulation of endemic species (Harrison and Noss, 2017).

In this issue of Annals of Botany, Nery et al. (2023) present a process-based study aimed at elucidating the eco-evolutionary mechanisms explaining the emergence of the Atlantic Forest (AF) centre of endemism in the Neotropics, using 66 species of Miconia supersect. Discolores (Melastomataceae) as a case study (85 % of the species in the group). The authors argue that the accumulation of narrow-range species is commonly assumed to result from shrinking and confinement of ancestral widespread ranges within climatic refugia and subsequent fragmentation (range reduction model, Fig. 1). Here, they challenge this view to instead propose that environmental heterogeneity can promote effective dispersion of ancestors out of their niche space and further specialization across sympatric areas with contrasting ecological conditions (niche expansion model, Fig. 1). Based on a biogeographical deconstruction approach that differentiates between three biogeographical elements, namely AF-endemics, AF-occasional and biogeographically indifferent species, the niche expansion hypothesis is supported by environmental niche modelling and phylogenetic comparative analyses. Two lines of evidence supporting the niche expansion model for the emergence of Miconia AF-endemics are that (1) these species are surrounded by a greater environmental heterogeneity in all the ecological gradients analysed (and particularly at high elevation), and (2) they are more specialized (narrower in niche breadth) across the solar radiation and soil pH axes relative to species in the other two biogeographical elements. These findings support the view that endemics grow at sites not only for historical reasons but also because of a fit between their phenotypes and prevailing local conditions, hence pointing to mountainous topography and soil properties as major drivers for the emergence of endemism (Buira et al., 2021). Montane landscapes favour high levels of endemism by enabling species to survive periods of climatic oscillations (Loarie et al., 2009), and they also provide greater climatic heterogeneity over short spatial distances than surrounding lowlands, circumstances that may foster sympatric speciation across contrasting microclimatic niches. Besides, many endemics grow in unusual substrates such as serpentine, limestone, gypsum and heavy metal soils, because these habitats exert selective pressures that promote adaptive speciation and may allow newly emergent species (so-called edapho-endemics) to avoid widespread competitors (Kruckeberg and Rabinowitz, 1985), and this could be the case with the Miconia AF-endemics of the study. Do they sit towards the most stressful ends of the soil gradient, or do they simply have narrower soil niches than non-endemics? Should the former pattern prove correct, Miconia AF-endemics would add to a growing list of edapho-endemics that grow under harsh soil conditions.

Fig. 1.

The range reduction and niche expansion models as alternative mechanisms for the accumulation of narrow-range species. The figure shows a mountain range with contrasting soil types (horizontal belts) where the sea-oriented slope acts as a climatic refugium (lower temperature oscillation relative to the inland-oriented slope). The range reduction model assumes that the range of a widely distributed species (1a) is shrunken due to local extinction of populations outside the refugium (2a), and then emerging reproductive barriers (e.g. restricted mobility of pollinators due to lower temperature, 2a) promote speciation via fragmentation of the ancestral range. The model assumes no ecological specialization, and thus the newly emerged species can effectively disperse along environmental gradients (altitude and contrasting soil types, 3a). Under the niche expansion model, environmental heterogeneity implies ecological opportunity for ancestral species (1b) to effectively disperse (2b) and specialize across different but spatially close areas with contrasting ecological conditions (3b), a process that is favoured by presumed outbreeding depression. Because ecological specialization implies a reduction in niche breadth, effective dispersal out of the niche space of the newly emerged species is impaired (3b). The flower icon is adapted from the Noun Project (https://thenounproject.com/icon/lily-flower-1069667/), protected under Creative Commons licence CC-BY 3.0, and credited to ANTON.

If the ecological fitness of Miconia AF-endemics is as finely tuned as niche modelling suggests, they should also show evidence of distinct phenotypes that allow their persistence under local conditions. Lavergne et al. (2004) reported lower investment in pollen transfer and limited seed production in narrow-range species relative to widespread congenerics, suggesting that local persistence is a key feature of the population ecology of endemics. The scenario of ecological specialization advanced by Nery et al. (2023) suggests that these and other traits related to local persistence could have evolved across Miconia AF-endemics to avoid outbreeding and dispersion out of their niche optima, a functional syndrome that would support the ecological specialization hypothesis. In the future, the specialization of Miconia AF-endemics to their inferred niche optima could be empirically evaluated through estimates of plant fitness in common garden and reciprocal transplant experiments.

Species can have narrow ranges because they have evolved in a relatively recent time (lack of historical opportunity to expand their ranges across suitable habitats) or because they were once more widespread than they currently are (relict species in a late state of their biogeographical ontogeny). The former are commonly referred to as neo-endemics, because they are newly formed and thus share a relatively recent common ancestor with their sister species, and the latter are referred to as palaeo-endemics, which are old and evolutionarily distinct species that have persisted through long time periods (Stebbins and Major, 1965). Most of the species that are exclusively found within endemism centres are probably neo-endemics by virtue of their tempo of evolution. However, the predominant mode of evolution within the centres has received less attention, and the work of Nery et al. (2023) contributes to filling in this gap.

The authors used a phylogenetic comparative approach to conclude that Miconia AF-endemics have evolved under a selection-driven scenario towards narrower niches and smaller distribution ranges than non-endemics. These traits also showed more constrained evolutionary trajectories (i.e. lower instantaneous rates σ²) in the former group. The authors interpreted this evolutionary regime (Ornstein–Uhlenbeck model with multiple optima and instantaneous rates σ²) as an adaptive response of AF-endemics to the pressure imposed by environmental heterogeneity, as outbreeding and dispersion out of their specialized niches would result in reduced fitness (Fig. 1). Although these results are in line with the ecological specialization scenario favoured by the authors, they should be taken with caution because Ornstein–Uhlenbeck models are often as equally supported as simple ‘white noise’, a null model in which trait values vary around some global mean independently of phylogeny (Cooper et al., 2016). Such a ‘control’ scenario in which phylogeny is completely irrelevant for trait evolution was not tested in the study.

The phylogenetic perspective in Nery et al. (2023) provides a nice step forward in our understanding of the tempo and mode of evolution within endemism centres, and their findings may lead to promising new avenues of research. For example, chromosome rearrangement and hybridization are integral components of speciation that could underlie the postulated genetic capacity for ecological differentiation and adaptation across Miconia AF-endemics (Rieseberg et al., 2003). Even though Lewis (1966) long argued that polyploidization may have had a significant role in the evolution of woody plants, whether such rapid chromosome reorganization events have expedited the accumulation of endemic species remains unexplored.

Perhaps reflecting how far we are from a comprehensive understanding of the eco-evolutionary mechanisms that underly the emergence of endemism centres, the research conducted by Nery et al. (2023) ultimately raises more questions than it resolves. This is undoubtedly a symptom of science on the right track, and it is an engaging and well-conceived study that will inspire evolutionary biologists interested in understanding why biodiversity is so unevenly distributed on Earth.