Phylogeny of Neotropical Sicarius sand spiders suggests frequent transitions from deserts to dry forests despite antique, broad-scale niche conservatism
Graphical abstract
Introduction
Patterns of variation in species diversity, from communities to biomes, have long intrigued biogeographers. Different biomes have different biotic composition, and to understand how such diversity was attained it is crucial to assess the relative importance of biome conservatism vs. biome switching (Crisp et al., 2009). The biome conservatism hypothesis predicts that lineages tend to retain their ancestral ecological traits, and that colonization of novel biomes over evolutionary history is rare. This is mainly due to phylogenetic niche conservatism (PNC), which has been defined as an evolutionary process in which related species are more similar in the ecological niche they occupy than would be expected by chance (Losos, 2008). Empirical evidence for the biome conservatism hypothesis, however, is controversial and largely focused on testing biome conservatism vs. switching in plant lineages (but see Antonelli et al., 2018). Studies addressing the potential contribution of conservatism in Neotropical biomes are relatively scant (see Donoghue and Edwards, 2014 for a review). One of such few studies showed, for instance, an unexpected high degree of biome switching between wet forests and savannas in South American plant lineages (Simon et al., 2009). This indicates that biome conservatism might not be as common as suggested in the literature. On the other hand, niches are always conserved to some extent, and it is expected that related species occupy similar biomes (Wiens and Graham, 2005). Thus, to better understand the distribution of organisms we should not only investigate if the niches are conserved or not, but also identify at what level niche shifts happened.
Besides wet forests and savannas, the Neotropical region also houses a considerable extent of drought-associated biomes (henceforth dry biomes). These biomes harbour unique species and lineage diversity (Pennington et al., 2009, DRYFLOR, 2016), but are often overlooked in studies aiming at understanding the assembly of the Neotropical biota. Studying dry biomes is no easy task because they occur in several disjunct areas, or nuclei, scattered along the Neotropical region (Pennington et al., 2000, Werneck, 2011; see also Fig. 1). Despite their current geographical isolation, a significant part of the biota of these nuclei is shared among them, at least at the generic or family levels (Prado and Gibbs, 1993, Pennington et al., 2000, Werneck, 2011, Magalhaes et al., 2017). Some authors have argued that such fragmented distribution has persisted over long enough evolutionary timescales to have influenced the evolution and biogeography of dry biomes (Pennington et al., 2009). This led to the prediction that, due to historical dispersal limitation associated with island biogeographic rules, there is high geographical structure in phylogenetic relatedness in dry biomes (Pennington et al., 2009). This pattern holds for plant lineages (Lavin, 2006), with closely-related taxa tending to occupy the same nuclei. This suggests that dispersal limitation plays a prominent role in the assembly of the biota of Neotropical dry biomes. However, dry biomes are a heterogeneous assemblage, ranging from warmer, seasonally dry tropical forests (SDTFs) to subtropical deserts, where climatic conditions are more unpredictable. This finer-scale variation in climatic conditions could also have shaped the diversification of organisms associated with dry biomes, especially if these are evolving under strong phylogenetic niche conservatism. Nonetheless, this question remains unexplored.
The evolutionary timing is a third factor that should be taken in to account to understand processes of biome shift or conservatism. Older lineages likely have had more opportunities to either shift between biomes or to colonize distant areas with similar conditions. The timing of assembly of Neotropical dry biomes is still a matter of discussion. It has been hypothesized that these dry biomes share a significant portion of their biota because they have been connected in the past. Influential studies have suggested that these connections may have taken place in the drier, colder periods of the Pleistocene – the Pleistocene arc hypothesis (PAH) (Prado and Gibbs, 1993, Pennington et al., 2000). This hypothesis predict that lineages from different SDTF nuclei should have diverged recently, no later than the Pleistocene. However, many phylogenetic studies of taxa associated to SDTFs have found divergence times that are much deeper than the ones expected by the PAH, mostly in the Miocene (Pennington et al., 2004, Werneck et al., 2012, Beati et al., 2013, Magalhaes et al., 2014, Côrtes et al., 2015). While it seems likely that Neotropical dry biomes have been connected in the past, the timing and causes of such events remain unclear. Dated phylogenies would not only shed light on this question, but also put niche conservatism in a time scale and allow quantifying the number of niche shifts in a given period of evolutionary time.
A sound phylogenetic basis is fundamental to addressing these biogeographical questions. There are several dated, large-scale phylogenetic studies of plants associated with dry environments, particularly SDTFs (e.g. Pennington et al., 2004, Côrtes et al., 2015). However, there is only a handful of phylogenies focusing on animals from Neotropical dry biomes, (e.g. Werneck et al., 2012, Beati et al., 2013, Palma et al., 2014, Lanna et al., 2018, Fonseca et al., 2018). The majority of these studies included only a few focal species, are geographically restricted, and do not cover the wide range of climatic conditions within the dry biomes. The focal group of this study, the sand spiders (subfamily Sicariinae), are a particularly interesting model to study the evolution dry biomes. These spiders are widespread and restricted to several SDTF nuclei, deserts and xeric scrublands in America and Africa. These medium to large spiders have special setae that adhere to soil particles, so they are able to camouflage on the ground to ambush their prey. The subfamily includes two genera: the 21 species of Sicarius Walckenaer, 1867 occur in South and Central America, while the eight species of Hexophthalma Karsch, 1879, are restricted to southern Africa (Magalhaes et al., 2017, Lotz, 2018). Sicariids seem to be very poor dispersers: their diversity is highly structured geographically (Binford et al., 2008, Magalhaes et al., 2014, Magalhaes et al., 2017) and they do not disperse by ballooning (rafting in the wind using silk strands, often over long distances) (see Binford et al., 2008). The divergence between the Sicarius and Hexophthalma has been estimated to have occurred before the separation of South America and Africa (Binford et al., 2008). Thus, these genera have long evolutionary history spanning most of the major geological events that could have shaped current Neotropical and southern African biodiversity. Their ancient association with a wide array of dry environments suggests that niche conservatism might play an important role in their evolution. All of this suggests they would be excellent models to test the relative roles of niche conservatism and dispersal limitation in the diversification of a dry-associated group of organisms.
To increase our understanding of the biogeographic history of Neotropical dry biomes, we examined the timing of diversification of Sicariinae lineages and explored the relative importance of historical dispersal limitation and phylogenetic niche conservatism in explaining their current distribution. We focused particularly on Sicarius because this genus is more diverse, occupies a wider range of climatic conditions and biomes, and has a more disjunct distribution, and thus is an ideal model to explore these questions. We gathered a comprehensive, specialist-based dataset on the distribution of Sicariinae species, characterized the climatic space occupied by these spiders, and generated a robust, dated phylogeny for the subfamily based on multiple gene regions and morphological data. Specifically, we established three hypotheses to test whether patterns of phylogenetic niche conservatism and dispersal limitation shaped the evolutionary history of Sicarius: (H1) phylogenetic niche conservatism was a major force shaping the distribution of Sicarius, and lineages inhabiting SDTFs and deserts are reciprocally monophyletic; (H2) due to historical dispersal limitation within each of these two groups closely related species occur in geographically close areas; and (H3) divergence among Sicarius species pre-date the Pleistocene, suggesting that phylogenetic niche conservatism has persisted over long evolutionary timescales.
Section snippets
Sicariinae distribution, biome occupancy and characterization of climatic niche
Georeferenced records of Sicarius and Hexophthalma were obtained from Magalhaes et al., 2017, Lotz, 2012, representing virtually all the published records of the subfamily to date (available as online Supp. file S8). We took two complementary approaches to investigate the biomes and climatic niches occupied by Sicariinae spiders. To characterize biome occupancy, we sorted Sicariinae records using the biome classification of Olson et al. (2001), with two modifications following
Characterization of climatic space and biome occupancy
All variables indicative of higher temperatures and precipitation correlate positively with PC1, while those indicative of higher daily/annual variation in temperatures and precipitation correlate negatively (Supp. Fig. S1). Biologically, we interpret higher values of PC1 as representing areas with hotter, wetter, and more stable climates. A visual inspection of the frequency histogram of PC1 values associated with Sicariinae records indicates a clear bimodal distribution allowing us to define
Discussion
Estimated divergence between species of spider groups that occur in different continents are frequently much younger than the break-up of ancient landmasses, and their present distribution is best explained by recent long-distance dispersal (e.g., euophryine salticids, Zhang and Maddison, 2013; nephiline araneids, Kuntner et al., 2013; lycosids, Piacentini and Ramírez, 2019; but see Chousou-Polydouri et al., 2019 for evidence of continental vicariance in orsolobids). In contrast, our results
Author contributions
AJS and ILFM designed the study. ADB, AJS and ILFM performed fieldwork. ADB, AJS, THDAV and FRS provided reagents and equipment for molecular work. ILFM identified specimens, gathered sequences and specimen records, gathered climatic niche and biome data, and performed phylogenetic analyses. DMN designed and performed variation partitioning analyses. ILFM, DMN and AJS led the writing of the first version of the manuscript. All authors worked on the final version of the text.
Acknowledgements
A. Dippenaar and P. Marais (AcAT, South Africa), L. Esposito and D. Ubick (CAS, USA), L. Carvalho (CHNUFPI, Brazil), P. Motta (DZUB, Brazil), M. Ramírez (MACN, Argentina), A. Kury (MNRJ, Brazil), R. Pinto da Rocha (MZSP, Brazil) and R. Orellana (UNSAAC, Peru) kindly lent specimens from the collections under their care. We thank A.O. Porta, A. Taucare Ríos, B.T. Faleiro, C. Calitra, C. Veloso, E. Florez-Daza, F. Cala-Riquelme, F.M. Hughes, G.F.B. Pereira, H.A. Iuri, J. Ochoa, J. Cabra García,
Funding
This study was supported by Consejo Nacional de Investigaciones Científicas y Técnicas, Argentina (doctoral fellowship to ILFM and grants PIP 2012-0943 and iBOL Argentina 2012), Fondo para la Investigación Científica y Tecnológica, Argentina (PICT 2015-0283), Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brazil (301776/2004-0 to ADB; 308613/2016-3 to FRS; 407288/2013-9, 405795/2016-5, 306222/2015-9 to AJS), Fundação de Amparo à Pesquisa do Estado de Minas Gerais, Brazil (
References (72)
- et al.
Phylogenetic relationships of Loxosceles and Sicarius spiders are consistent with Western Gondwanan vicariance
Mol. Phylogenet. Evol.
(2008) - et al.
All-scale spatial analysis of ecological data by means of principal coordinates of neighbour matrices
Ecol. Model.
(2002) - et al.
A molecular phylogeny of nephilid spiders: evolutionary history of a model lineage
Mol. Phylogenet. Evol.
(2013) - et al.
The evolutionary history of Lygodactylus lizards in the South American open diagonal
Mol. Phylogenet. Evol.
(2018) - et al.
Hunting the wolf: a molecular phylogeny of the wolf spiders (Araneae: Lycosidae)
Mol. Phylogenet. Evol.
(2019) - et al.
Post-Eocene climate change across continental Australia and the diversification of Australasian spiny trapdoor spiders (Idiopidae: Arbanitinae)
Mol. Phylogenet. Evol.
(2017) - et al.
Age constraint on Burmese amber based on U-Pb dating of zircons
Cretac. Res.
(2012) Gondwanan break-up: legacies of a lost world?
Trends Ecol. Evol.
(2008)The diversification of eastern South American open vegetation biomes: Historical biogeography and perspectives
Quat. Sci. Rev.
(2011)- et al.
Fossil calibrations for the arthropod Tree of Life
Earth Sci. Rev.
(2016)