The genus Amara Bonelli, 1810 is a very speciose and taxonomically difficult genus of the Carabidae. The identification of many of the species is accomplished with considerable difficulty, in particular for females and immature stages. In this study the effectiveness of DNA barcoding, the most popular method for molecular species identification, was examined to discriminate various species of this genus from Central Europe. DNA barcodes from 690 individuals and 47 species were analysed, including sequences from previous studies and more than 350 newly generated DNA barcodes. Our analysis revealed unique BINs for 38 species (81%). Interspecific K2P distances below 2.2% were found for three species pairs and one species trio, including haplotype sharing between Amara alpina/Amara torrida and Amara communis/Amara convexior/Amara makolskii. This study represents another step in generating an extensive reference library of DNA barcodes for carabids, highly valuable bioindicators for characterizing disturbances in various habitats.

Central Europe, cytochrome c oxidase subunit I, German Barcode of Life, mitochondrial DNA, molecular specimen identification, Zabrus

With the rise of modern sequencing technologies in the early 1990s, DNA sequences have been increasingly used as supplementary markers for species description, identification, and classification (Raupach et al. 2016). In this context, DNA barcoding has become the most popular approach for the assignment of specimens throughout all life stages to described and classified species following the Linnean guidelines (Hebert et al. 2003a, 2003b). In the case of animals, an app. 660 base pair (bp) fragment of the mitochondrial cytochrome c oxidase subunit I (COI) gene has been chosen as standardized barcode marker (Hebert et al. 2003a, 2003b). The concept of DNA barcoding is based on a simple assumption: every species will most likely have unique DNA barcodes with low intraspecific variation and interspecific variation exeeds the variability within species, generating a so-called DNA barcoding gap that highly depends on the studied taxonomic groups (Hebert et al. 2003a, 2003b, Čandek and Kuntner 2015, Koroiva and Kvist 2017). In spite of the fact that various effects can limit the usefulness of DNA barcodes and mitochondrial DNA in general, e.g., the presence of pseudogenes or numts (e.g., Bensasson et al. 2001, Leite 2012, Jordal and Kambestad 2014, Haran et al. 2015), heteroplasmy (e.g., Magnacca and Brown 2010, Robinson et al. 2015), effects of Wolbachia infections within terrestrial arthropods (e.g., Hurst and Jiggins 2005, Werren et al. 2008, Smith et al. 2012), or general critics on the concept (e.g., Will and Rubinoff 2004, Collins and Cruickshank 2013), numerous studies have demonstrated that DNA barcoding yields excellent results across a broad range of various animal taxa (e.g., Costa et al. 2007, Aliabadian et al. 2013, Knebelsberger et al. 2014, Lobo et al. 2015, Raupach et al. 2015, Barco et al. 2016). Today, barcode data can be easily managed and analysed using the public Barcode of Life data base (BOLD; www.boldsystems.org; Ratnasingham and Hebert 2007). This core data retrieval interface offers various analytical tools, including the Barcode Index Number (BIN) system (Ratnasingham and Hebert 2013).

In term of arthropods, most DNA barcoding studies focus on insects (Raupach and Radulovici 2015), e.g., the Ephemeroptera, Plecoptera and/or Trichoptera (Zhou et al. 2009, Zhou et al. 2011, Ruiter et al. 2013, Moriniére et al. 2017), Heteroptera (Jung et al. 2011, Park et al. 2011, Raupach et al. 2014), Hymenoptera (Smith and Fisher 2009, Smith et al. 2013, Schmidt et al. 2015), Lepidoptera (e.g., Hajibabaei et al. 2006, Hausmann et al. 2011, Hausmann et al. 2013, Kekkonen et al. 2015), and others (e.g., Glover et al. 2010, Moriniére et al. 2014, Hawlitschek et al. 2017). In comparison to the high number of described species, however, the number of studies analysing the Coleoptera (e.g., Greenstone et al. 2011, Woodcock et al. 2013, Pentinsaari et al. 2014, Hendrich et al. 2015, Oba et al. 2015, Rougerie et al. 2015, Han et al. 2016), and in particular the Carabidae or ground beetles (Greenstone et al. 2005, Raupach et al. 2010, 2011, 2016b), is still low.

Ground beetles represent highly valuable and frequently used bioindicators for the characterization of disturbances in various habitats such as forests, meadows, fens, or river banks (e.g., Lövei and Sunderland 1996, Rainio and Niemelä 2003, Koivula 2011, Kotze et al. 2011). Within the Carabidae, Amara Bonelli, 1810 is a large genus in the tribe Zabrini Bonelli, 1810. Many species are Holarctic, but a few are Neotropical or occur in Eastern Asia. About 150 European species are known (Luff 2007), with 52 recorded for Germany (Trautner et al. 2014). Beetles of this genus are typically characterized by their rather oval and parallel-sided form, with females that are often somewhat duller than the males and may even differ in body shape (Luff 2007) (Fig. 1). While ground beetles are mostly carnivorous, numerous Amara species feed on plant seeds as both larvae and adults (e.g., Hůrka 1996, Jørgensen and Toft 1997, Holland 2002, Klimeš and Saska 2010), although some species consume seeds only as a supplement to their predominantly predatory diet (e.g., Goldschmidt and Toft 1997, Holland 2002, Koprdova et al. 2008). They typically require dry habitats, uncultivated areas and open vegetation on light soils, such as sand, gravel, or chalk (e.g., Kromp 1989, Thomas et al. 2001). As a consequence of their more or less homogenous habitus and very subtle morphological differences between species (e.g., the shape of the pronotum or coloration of antennomeres), Amara is known as the most challenging genus of ground beetles in terms of species identification in Central Europe. Nevertheless, Fritz Hieke (1930–2015) devoted his scientific career to this genus and thoroughly cleared up the difficult taxonomic assessment of this genus at all levels (e.g., Hieke 1984, 1988, 2005). In this context he published a list of valid names and their synonyms, with over 560 specific and subspecific, and 47 subgeneric names (Hieke 1995, 2011).

Figure 1. 

An image collection of some representative species of the analysed ground beetles. AAmara (Amara) similata (Gyllenhal, 1810) BAmara (Amarocelia) erratica (Duftschmid, 1812) CAmara (Bradytus) fulva (Müller, 1776) DAmara (Curtonotus) convexiuscula (Marsham, 1802) EAmara (Leirides) spectabilis Schau, 1858 FAmara (Paracelia) quenseli (Schönherr, 1806) GAmara (Xenocelia) cursitans Zimmermann, 1931 HAmara (Zezea) kulti Fassati, 1947, and I Zabrus tenebrioides Goeze, 1777. Scale bars 1 mm. All images were obtained from www.eurocarabidae.de.

Here we present the next step in building-up a comprehensive DNA barcode library of Central European species of ground beetles as part of the German Barcode of Life project (GBOL), focusing on the genus Amara. The analysed barcode library included 46 Amara species as well as one species of Zabrus Clairville, 1806 which represents the second genus of the tribe Zabrini known from Central Europe. Four species (Amara littorea Thomson, 1857, Amara makolskii Roubal, 1923, Amara sabulosa Audinet-Serville, 1821, and Amara spectabilis Schaum, 1858) were not covered by previous studies (Raupach et al. 2010, Pentinsaari et al. 2014, Hendrich et al. 2015). In summary, 358 new barcodes were generated and a total number of 690 DNA barcodes examined.

In total, 690 DNA barcode sequences of 47 carabid beetle species were examined. A full list of the species is presented in the supporting information (Suppl. material 1). In total, 46 species of the genus Amara were studied, with 41 (79%) of the 52 species documented for Germany. Five analysed species, Amara alpina (Paykull, 1790) (n = 3), Amara hyperborea Dejean, 1831 (n = 1), Amara interstitialis Dejean, 1828 (n = 1), Amara spectabilis Schaum, 1858 (n = 3), and Amara torrida Panzer, 1796 (n = 4), are not known from Germany. All these specimens were collected from other countries (see above). Fragment lengths ranged from 307 (n =14) to a full length of 657 bp. Base frequencies analysis revealed low GC-contents (average: 32%) for the barcode fragment, as it is known from insects and other arthropods. The individual mean nucleotide contents were A = 0.29, C = 0.15, G = 0.17, and T = 0.39. Intraspecific K2P distances ranged from zero to 2.18% (Amara bifrons (Gyllenhal, 1810)). Interspecific K2P distances had values between zero and a maximum of 10.06%.

The BIN analyses were performed on January 11^th 2018. Unique BINs were revealed for 38 species (81%). Three species pairs shared a BIN: Amara alpina Paykull, 1790 and Amara torrida (Panzer, 1796) were both included in ACF5385, Amara familiaris (Duftschmid, 1812) and Amara lucida (Duftschmid, 1812) in AAC4901, and Amara ovata (Fabricius, 1792) and Amara similata (Gyllenhal, 1820) in AAJ5377. Furthermore, one BIN (ACF1000) contained three species: Amara communis (Panzer, 1797), Amara convexior Stephens, 1828, and Amara makolskii Roubal, 1923 (the so-called Amara communis complex).Interspecific distances of zero were found for Amara alpina and Amara torrida as well as for Amara communis, Amara convexior and Amara makolskii.

The NJ analyses based on K2P distances revealed non-overlapping clusters with bootstrap support values >90% for 33 species (70% of all studied species) with more than one studied specimen (Fig. 2). A comprehensive topology is presented in the supporting information (Suppl. material 2).

Figure 2. 

Neighbor joining topology of the analysed ground beetle species based on Kimura 2-parameter distances. Triangles show the relative number of individual’s sampled (height) and sequence divergence (width). Red triangles indicate species pairs with interspecific distances <2.2%. Numbers next to nodes represent non-parametric bootstrap values >90% (1,000 replicates). Asterisks indicate species not recorded in Germany. All images were obtained from www.eurocarabidae.de.

Figure 2. 

Continue.

Our statistical maximum parsimony analysis revealed closely related haplotypes for Amara ovata (Fabricus, 1792) and Amara similata (Gyllenhal, 1810) (Fig. 3a). The dominant haplotypes of both species (Amara ovata: h1, Amara similata: h2) were separated by six mutational steps. An even lower number of mutational steps were found between Amara familiaris (Duftschmid, 1812) and Amara lucida (Duftschmid, 1812) (Fig. 3b): the only examined specimen of Amara lucida (h5) was separated from the dominant haplotype of Amara familiaris (h1) by two mutational steps. Furthermore, multiple haplotypes shared by more than one species were found in the Amara communis complex (n = 49; Fig. 4) and for Amara alpina (n = 3) with Amara torrida (n = 4) (Fig. 5). For the Amara communis complex, eight different haplotypes with two dominant ones (h1, h2) were identified. Whereas haplotype h1 was shared by 18 specimens with all three species (Amara communis: n = 6, Amara convexior: n = 2, Amara makolskii: n = 10), haplotype h2 was found exclusively in specimens of Amara convexior (n = 17). Haplotype h3, located between h1 and h2 in the network, was shared by specimens of Amara communis (n = 8) and Amara convexior (n = 1). In addition, five haplotypes represented by one specimen only (singletons) were located at the periphery of the network (Amara communis: h4, h5, Amara convexior: h8, Amara makolskii: h6, h7). In the case of Amara alpina and Amara torrida, the statistical maximum parsimony analysis revealed four haplotypes, with one haplotype (h2) shared by specimens of both species (Amara alpina: 2 specimens, Amara torrida: 1 specimen). This haplotype was separated by four additional steps from haplotype h1 that was restricted to specimens of Amara torrida. Furthermore, two singletons (h3: two additional mutational steps; h4: one additional mutational step) were connected with haplotype h1, generating a compact network that contained only a few mutational steps.

Figure 3. 

Maximum statistical parsimony networks of two species pairs: A Amara ovata (Fabricius, 1792) and Amara similata (Gyllenhal, 1810), and B Amara familiaris (Duftschmid, 1812) and Amara lucida (Duftschmid, 1812). Used parameters included default settings for connection steps whereas gaps were treated as fifth state. Each line represents a single mutational change whereas small black lines indicate missing haplotypes. The numbers of analysed specimens (n) are listed, the diameter of the circles is proportional to the number of haplotypes sampled (see given open half circles with numbers). Scale bars 1 mm. Beetle images were obtained from www.eurocarabidae.de.

Figure 4. 

Maximum statistical parsimony network of the Amara communis complex. Used parameters included default settings for connection steps whereas gaps were treated as fifth state. Each line represents a single mutational change whereas small black lines indicate missing haplotypes. The numbers of analysed specimens (n) are listed, the diameter of the circles is proportional to the number of haplotypes sampled (see given open half circles with numbers). Scale bars 1 mm. Beetle images were obtained from www.eurocarabidae.de.

Figure 5. 

Maximum statistical parsimony network of Amara alpina (Paykull, 1790) and Amara torrida Panzer, 1796. Used parameters included default settings for connection steps whereas gaps were treated as fifth state. Each line represents a single mutational change whereas small black lines indicate missing haplotypes. The numbers of analysed specimens (n) are listed, the diameter of the circles is proportional to the number of haplotypes sampled (see given open half circles with numbers). Scale bars 1 mm. Beetle images were obtained from www.eurocarabidae.de.

Within the past few years, DNA-based approaches have become more and more popular for the assessment of biodiversity and identification of specimens, in particular where the traditional morphology-based identification has proved problematic (Taberlet et al. 2012). As a consequence of this development and the rise of new concepts (Hebert et al. 2003a, 2003b), the analysis of single specimens, bulk samples (metabarcoding) or environmental DNA (eDNA) will be performed routinely as part of modern species diversity assessment studies in the near future (e.g., Scheffers et al. 2012, Cristescu 2014, Kress et al. 2015). However, such studies highly rely on comprehensive on-line sequence libraries that act as references (e.g., Brandon-Mong et al. 2015, Creer et al. 2016, Staats et al. 2016). Therefore, our DNA barcode library represents an important step for the molecular characterization of ground beetles in Central Europe and adjacent regions. The current results demonstrate that DNA barcodes distinguish Central European species of the taxonomically challenging genus Amara remarkably well. Our analysis revealed unique BINs for 38 (81%) of the 47 analysed species. The results coincide with high rates of successful species identification of previous barcoding studies on ground beetles (Raupach et al. 2010, 2011, Pentinsaari et al. 2014, Hendrich et al. 2015, Raupach et al. 2016b). In contrast to other carabid genera, e.g., Bembidion Latreille, 1802 (Raupach et al. 2016b) or Calathus Bonelli, 1810 (Hendrich et al. 2015), no evidence was found for high intraspecific distances (above 2.2%) within the analysed Amara species. In contrast to this, low intraspecific distances (below 2.2%) and shared haplotypes for various species pairs were revealed. Such low distances are typically indicative of a recent ancestry and/or ongoing gene flow for various species pairs (e.g., Tautz et al. 2003, Frezal and Leblois 2008, Raupach et al. 2010). We will discuss these cases in more detail.

Used alone or in combination with DNA metabarcoding on environmental samples (Taberlet et al. 2012), DNA barcoding is becoming a standard for basic and applied research in ecology, evolution and conservation across taxa, communities and ecosystems (Zinger and Philippe 2016). In this context, our study clearly encourages the use of DNA barcodes for the identification of ground beetles species of the taxonomically difficult genus Amara. However, DNA barcodes of additional eleven Amara species documented for Germany are currently missing. The analysis of these missing species may include other, so far undetected problematic cases. For example, Amara chaudoiri Schaum, 1858 and Amara concinna Zimmermann, 1832 are morphologically very similar species. Nevertheless, our data set and results represent another important step in building-up a comprehensive barcode library for the Carabidae in Germany and Central Europe which can be used in modern molecular biodiversity assessment studies. Despite the fact that DNA barcoding failed to deliver a valid species identification for some species in this study, it narrows the options to a pair (or in one case trio) of closely related species. Especially for the almost impossible identification of immature stages and/or females within various species of Amara, this is a very encouraging result.

We would like to thank Christina Blume, Claudia Etzbauer (both ZFMK, Bonn) and Jana Deppermann (DZMB, Wilhelmshaven) for their laboratory assistance. Furthermore we are very grateful to Karl-Hinrich Kielhorn (Berlin) and Frank Köhler (Bonn) for providing various specimens, and to Ortwin Bleich for giving permission to use his excellent photos of ground beetles taken from www.eurocarabidae.de. We also thank David Kavanaugh, Hongbin Liang, and Mikko Pentisaari for their helpful comments. This publication was partially financed by German Federal Ministry for Education and Research (FKZ01LI1101A, FKZ01LI1101B, FKZ03F0664A), the Land Niedersachsen and the German Science Foundation (INST427/1-1), as well as by grants from the Bavarian State Government (BFB) and the German Federal Ministry of Education and Research (GBOL2: 01LI1101B). We are grateful to the team of Paul Hebert in Guelph (Ontario, Canada) for their great support and help and in particularly to Sujeevan Ratnasingham for developing the BOLD database infrastructure and the BIN management tools. Sequencing work was partly supported by funding from the Government of Canada to Genome Canada through the Ontario Genomics Institute, whereas the Ontario Ministry of Research and Innovation and NSERC supported development of the BOLD informatics platform.