Ancient landscapes of the Namib Desert harbor high levels of genetic variability and deeply divergent lineages for Collembola.

Gemma E Collins,Don A Cowan,Ian D Hogg,Janine R Baxter,Gillian Maggs-Kölling

doi:10.1002/ece3.5103

Abstract

AimTo assess spatial patterns of genetic and species‐level diversity for Namib Desert Collembola using mitochondrial DNA cytochrome c oxidase subunit I (COI) gene sequences.LocationNamib Desert gravel plains.TaxonCollembola (springtails).MethodsA total of 77 soil samples were collected along NE‐SW (60 km) and E‐W (160 km) transects from within a 4,000 km2 area of the Namib Desert gravel plains. We extracted 434 springtails from the 37 samples which contained Collembola and sequenced them at the COI gene locus. In the absence of specific taxonomic keys and previous genetic data for these taxa, we used Generalized Mixed Yule Coalescent (GMYC) analyses to provide putative species‐level designations.ResultsWe obtained 341 successful COI sequences, 175 of which were unique haplotypes. GMYC analyses identified 30 putative species, with up to 28% sequence divergence (uncorrected p‐distance). The distribution of genetic variants was disjunct, with 97% of haplotypes and 70% of “GMYC species” found only at single sites.Main conclusionsDispersal events, although rare, may be facilitated by environmental events such as prevailing onshore winds or occasional flow of rainwater to the coast. We conclude that the high genetic diversity we observed is the result of ancient springtail lineages, patchy distribution of suitable habitats, and limited dispersal (gene flow) among habitable locations.

Highlights

The capacity to support eukaryotic life is limited within both hot and cold deserts owing to extreme temperatures, high ultraviolet radiation, high winds, and low nutrient and water availability (Ayal, 2007; Polis, 1991)
While relationships between environmental gradients and mi‐ crobial (e.g., Johnson, Ramond, Gunnigle, Seely, & Cowan, 2017; Scola et al, 2018; Stomeo et al, 2013) and plant (Hachfeld, 2000) communities within the Namib Desert are partly understood, only limited data are available for the invertebrate fauna and most of this research has focused on the macroarthropod taxa
In the absence of detailed taxonomic evaluation, Generalized Mixed Yule Coalescent (GMYC) analyses (Monaghan et al, 2009; Pons et al, 2006) were used alongside Barcode Index Number (BIN) designations to esti‐ mate species‐level diversity. While both methods have been shown appropriate for grouping sequences to species level (Hausmann, Haszprunar, & Hebert, 2011; Leliaert, Verbruggen, Wysor, & Clerck, 2009), we found extremely high genetic variability using the BIN algorithm (n = 43) relative to the GMYC analyses (n = 30)

Summary

| INTRODUCTION

The capacity to support eukaryotic life is limited within both hot and cold deserts owing to extreme temperatures, high ultraviolet radiation, high winds, and low nutrient and water availability (Ayal, 2007; Polis, 1991). The gravel plains can be categorized into three xeric zones (fog, arid, and rainfall; Figure 1) occurring along a gradient from the coast to the Escarpment (~1,000 m asl) (Scola et al, 2018). While relationships between environmental gradients and mi‐ crobial (e.g., Johnson, Ramond, Gunnigle, Seely, & Cowan, 2017; Scola et al, 2018; Stomeo et al, 2013) and plant (Hachfeld, 2000) communities within the Namib Desert are partly understood, only limited data are available for the invertebrate fauna and most of this research has focused on the macroarthropod taxa. We determine genetic variability, using COI, for Collembola across large‐scale Namib Desert soil transects, and assess the spatial distri‐ bution of COI haplotypes and putative species‐level groupings as an indication of population connectivity and dispersal

| MATERIALS AND METHODS

| DISCUSSION

Findings

| SUMMARY