Towards an evolutionary history of European-Alpine Trechus ground beetles: Species groups and wing reduction
Graphical abstract
Introduction
Trait loss or reduction, also known as vestigialisation, is a common yet understudied evolutionary phenomenon and an expected consequence of relaxed selection (Fong et al., 1995, Lahti et al., 2009). Loss of a trait may be caused by direct selection, for example, in the case of stout wings in penguins co-opted for swimming (Fong et al., 1995). However, such clear cases are rarely reported, and most studies invoke indirect selection driven by energy economy as possible explanation, for example through the reduction of flight muscles, or antagonistic pleiotropy, for example trade-offs between pesticide resistance and fitness-correlated life-history traits. Non-selectionist alternative scenarios include the accumulation of neutral mutations, founder effects accelerating drift, and gene flow, which may result in or at least contribute to vestigialisation (Fong et al., 1995, Lahti et al., 2009).
A frequently cited example for vestigialisation are pterygote insects, in which flight capability has been repeatedly lost during evolution (Wagner and Liebherr, 1992, Roff, 1994). Although, at first glance, this may be counterintuitive given that the evolution of wings and flight was essential for the radiation of the class, there is evidence of thousands events of secondary flight loss, in almost all insect orders (Roff, 1994). One possible explanation for these multiple, independent losses of flight are trade-offs between the ability to fly and fitness components suggesting that flightlessness allows for the re-allocation of energy otherwise needed for flying and maintenance of flight capability to reproduction (reviewed by Fong et al., 1995, Grzywacz et al., 2018). Flightlessness has also frequently been associated with habitat stability, reducing the importance of dispersal for long-term survival, with habitat isolation, as dispersers may face a higher risk of mortality, and with cold and windy habitats, in which the energetic costs of flight are increased (reviewed by Wagner and Liebherr, 1992, McCulloch et al., 2009, Dussex et al., 2016, Venn, 2016). Unsurprisingly, flightlessness seems to be especially abundant among insect species inhabiting mountains (Darlington, 1943, McCulloch et al., 2009). Numerous studies investigating flightlessness in insects in the context of vestigialisation lead to interesting questions stimulating further research: How often does flightlessness evolve independently within a clade? How much variation in the vestigial trait is retained? Is the process reversible and, if so, what are frequency and mechanisms of reversal? What are the most important conditions under which wing reduction evolves?
When searching for conditions conducive to and following from flightlessness, considering the relationships of flightlessness and other characteristics of species can be useful. Because in many studies, including this one, wing status is assessed but not flight capability, from here, the term wing reduction is used as a proxy for flightlessness, albeit we are aware that not all fully-winged insects can fly (e.g., McCulloch et al., 2009). One approach to searching for general patterns is using biogeography, and a negative relationship of reduced wings and distribution-area size has been established, concerning (near-) complete wing loss (Gutierrez and Menendez, 1997) as well as less drastic wing reductions (McCulloch et al., 2017). Such effects were observed both on the mainland and on islands (reviewed by Venn, 2016), indicating that range size could both be a consequence of and a reason for wing reduction. Another approach to understanding wing-status evolution is considering the ecology of species (Venn, 2016), and, among others, four ecological proxies have been considered, namely body size, stratum inhabited, elevation, and hygrophily. Body size has been a very general proxy for ecology (e.g., Funk et al., 2006) and was found to be constrained by full-wing status but not by reduced-wing status, that is, the latter has been inferred to allow development of relatively larger body size (Liebherr, 1988, McCulloch and Waters, 2018). The stratum inhabited may be relevant as soil dwelling has been identified to be correlated with wing reduction (Wagner and Liebherr, 1992). For the characterization of stratum preferences, body-surface pigmentation can be used as a proxy given that lighter pigmentation is frequently found in below-ground species whereas darker pigmentation is common in above-ground species. (Culver and Pipan, 2013). As already mentioned, the elevational zone has an effect since isolation, wind, and cold increase with elevation and are usually correlated with wing reduction (Venn, 2016). Finally, hygrophily was found to correlate negatively with wing reduction (Darlington, 1943, Shepard, 2019), possibly because more humid habitats tend to be more unstable (Darlington Jr., 1943).
In light of the aforementioned aspects, we here propose Trechini ground beetles as a promising system for studying the evolution of wing reduction. The tribe is a monophyletic, globally distributed, and ecologically diverse group of small beetles featuring an extraordinary diversity of species (Maddison et al., 2019). It shows variation in wing size among and within species (Jeannel, 1927, Desender et al., 1980a, Desender et al., 1980b, Liebherr and Krushelnycky, 2007). Besides winged species, the majority of these beetles are incapable of flight and adapted to a hidden life on or in the ground, preferring moist and cool microclimates. This epi- or endogean mode of life likely facilitated the evolution of hundreds of species with restricted distribution and often highly derived body shapes, especially in the subterranean environment (Contreras-Díaz et al., 2007). The tribe was monographed by Jeannel, 1926, Jeannel, 1927, Jeannel, 1928 at the beginning of the last century. Due to the lack of a robust phylogenetic framework, Jeannel’s arrangement into genera, subgenera, and species groups, based on genital morphology, has been in use until today (Casale and Laneyrie, 1982, Moravec et al., 1810), although dozens of new genera and hundreds of new species have been described since then. Interestingly, Jeannel followed two contrasting taxonomic strategies within Trechini: On the one hand, most of the morphologically 'primitive' and uniform, epigean species of the northern hemisphere were lumped together in the huge genus Trechus, without any division into subgenera. Informal species groups were used instead, partly defined by geography, partly defined on the basis of morphological characters. On the other hand, the morphologically more 'derived' and diverse, endogean taxa were split into many distinct genera.
The extraordinarily diverse genus Trechus Clairville, 1806 with approximately one thousand species in the northern hemisphere (Casale and Laneyrie, 1982, Moravec et al., 1810) is the nominate genus of the tribe Trechini and mainly consists of epigean and Holarctic species. Most species are aggregated in the subgenus Trechus s. str. The taxonomy of the genus is complicated because of the uniform appearance of most species on the one hand and high intraspecific variability (caused by low dispersal capacities and small population sizes, Faille et al., 2013) on the other hand. Thus, taxonomy is based primarily on characteristics of the male reproductive organs (aedeagi) (Jeannel, 1926). Neither the genus nor some of the species groups can be defined by any set of characters, and the genus in the present sense may be understood as all standard Trechini of the northern hemisphere lacking specific characteristics.
Particularly interesting with regard to the evolution of wing reduction are Trechus from the comparatively small European Alps (c. 120,000 km2, Körner, 2007), which are especially diverse with 85 species, most of them with restricted distribution (Donabauer, 2017). Their evolutionary history is largely unresolved. For example, it is currently unknown how often wing reduction emerged in the European-Alpine species of Trechus. The major hurdle that prevented answering such questions (comparable to that in other groups, Grzywacz et al., 2018) is that a phylogeny with reproducible species groups and consistent assessment of character traits was never elaborated due to two reasons: First, it is difficult to overview so many species, and the sampling in collections is incomplete due to rarity and extremely restricted distributions. Second, potential convergence in external features of distinct species inhabiting similar ecological niches complicates morphology-based phylogenetic reconstruction.
In recent years, Faille et al., 2010, Faille et al., 2011, Faille et al., 2012, Faille et al., 2013, Faille et al., 2014 sequenced mitochondrial and nuclear genes of many Trechini from the western Palearctic region and reported very interesting results about the phylogenetic relationship of several derived subterranean clades. A fundamental reorganisation of European Trechini was called for but has not been implemented to date. In addition, the genus Trechus was demonstrated to be both polyphyletic and paraphyletic. Additional phylogenies have been published about Trechus from the Canary Islands (Contreras-Díaz et al., 2007) and about a Trechus group restricted to the alpine elevation zone (Lohse et al., 2011), both corroborating the need of further phylogenetic work.
Encouraged by Faille’s work, a collection of 88 species of Trechus, including 72 out of 85 Alpine species and six widespread and partly winged European species, has been established by M. Donabauer and is here for the first time used for reconstructing the phylogeny of the Alpine species of Trechus using two commonly used genes, the mitochondrial cytochrome c oxydase subunit I (COI) and the nuclear 28S ribosomal DNA (28S). Based on this phylogeny, the aims of this study were (i) rearranging the Alpine species of Trechus into monophyletic species groups and (ii) elucidating wing status and the evolutionary history of wing reduction in our sample with a strong focus on the Alpine species of Trechus. For the latter, we assessed wing status by calculating hindwing-to-forewing length ratios and mapped wing ratio, biogeography (area of distribution), and four proxies of ecology (body size, pigmentation, elevational zone, hygrophily) onto our phylogeny to then test for correlations of wing ratios with traits of each of these characters using independent contrasts. In addition, we reconstructed ancestral states for wing status and the other characters, interpreted the inferred ancestral situation for a subset of completely sampled subclades, and discussed alternative, but not necessarily mutually exclusive, scenarios of wing-status evolution. We here present the currently most complete phylogeny and a species-group classification for the Alpine species of Trechus based on molecular markers. Demonstrating variation in wing status among and within species and clades, we also set the stage for studying the evolution of wing reduction in this group and outline avenues for future research.
Section snippets
Insect collection and morphological determination
Between 2011 and 2017, in total 838 individuals of Trechini were collected from 117 locations by M. Donabauer (Table S1; for the locations in the European Alps, see Fig. 1) and immediately stored in absolute ethanol for later genetic analyses; additional samples were provided by Alexander Szallies (Table S1). From each sampling location, additional specimens from the same site were used for morphological determination following the keys as listed in Table 1, mostly using characters located on
PCR, sequencing, and linking to available GenBank entries
DNA of a total of 415 insects was extracted in a non-destructive way. After manual curation, 277 and 353 sequences were identified for COI and 28S, respectively, that passed quality screening. Alignment length was 766 bp for COI and 761 bp for 28S. The subset with high-quality sequences in both genes contained 273 individuals from 91 species (GenBank entries MH186597-MH187154; Table S1; Figs. S1 and S2). In this subset, 179 and 150 unique haplotypes were present for COI and 28S, respectively.
Discussion
In the genus Trechus, the major systematic classifications are almost a hundred years old and primarily based on genital characters (Jeannel, 1926, Jeannel, 1927, Jeannel, 1928). More recent work using molecular data showed that the genus is not monophyletic (Faille et al., 2011, Faille et al., 2012, Faille et al., 2013), but a comprehensive phylogeny of Trechus from the European Alps has been lacking so far. Here, we reconstructed the phylogeny of 72 of the 85 species of the European Alps
Author contributions
MD, WA, BCS-S, and FMS designed the study. WA, FMS, and BCS-S coordinated the study. MD did the field work and the species identifications. WA contributed to the molecular experiments, FMS to the morphometric analyses for the wing-ratio assessments. MHM, WA, BCS-S, and FMS analysed the data. MHM, MD, WA, BCS-S, and FMS produced the figures. MHM, MD, WA, BCS-S, and FMS interpreted the data. MHM, WA, BCS-S, and FMS wrote the draft manuscript with contributions from MD, which was revised by all
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We thank Philipp Andesner for assistance in the laboratory and for sequence editing; Elisabeth Zangerl for wing measurements and GenBank submissions; Philipp Bertemes, Johanna Meusburger, and Agnes Strasser for assistance with a pilot study; Alexander Szallies for providing samples; David R. Maddison for sharing information on 28S sequences; Alfried P. Vogler and an anonymous reviewer for constructive criticism that very much helped in improving the paper. The computational results presented
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References (96)
- et al.
Evolution and diversification of the forest and hypogean ground-beetle genus Trechus in the Canary Islands
Mol. Phylogenet. Evol.
(2007) - et al.
Subterranean ecosystems
- et al.
A molecular phylogeny shows the single origin of the Pyrenean subterranean Trechini ground beetles (Coleoptera: Carabidae)
Mol. Phylogenet. Evol.
(2010) The use of 'altitude' in ecological research
Trends Ecol. Evol.
(2007)- et al.
Relaxed selection in the wild
Trends Ecol. Evol.
(2009) - et al.
Phylogeny of the beetle supertribe Trechitae (Coleoptera: Carabidae): Unexpected clades, isolated lineages, and morphological convergence
Mol. Phylogenet. Evol.
(2019) Phylogenetic relationships of the carabid subfamily Harpalinae (Coleoptera) based on molecular sequence data
Mol. Phylogenet. Evol.
(2002)- et al.
Flightlessness in insects
Trends Ecol. Evol.
(1992) - et al.
The ground beetle tribe Trechini in Israel and adjacent regions (Coleoptera, Carabidae)
Spixiana
(2012) Wing-length determination in 2 wing-dimorphic Calathus species (Coleoptera, Carabidae)
Hereditas
(1990)