Introduction

Spider mites (Acari: Tetranychidae) are the most economically important mite family (Van Leeuwen et al. 2010). They can feed on various host plants and cause economic yield losses (Van Leeuwen et al. 2009). So far, more than 1300 tetranychid species have been recorded worldwide (Migeon and Dorkeld 2022). Misidentification of spider mites, as with pests in general, may cause failure in pest control.

Identification of tetranyhid mites is traditionally based on morphological characters (Seeman and Beard 2011), and there are several reasons that lead to species misidentification. High levels of experience and knowledge are required for species discrimination, particularly because of the minute size of the characters observed. Furthermore, among some mite genera, only the males, which are less frequent in nature (Helle and Sabelis 1985), allow species discrimination. In addition, traditional morphological keys do not allow species discrimination of immature stages (Pritchard and Baker 1955).

DNA-based identification offers a solution to overcome these difficulties (Hebert et al. 2003; Ros and Breeuwer 2007). Several molecular markers have been used in species identification of mites so far (Navajas and Fenton 2000; Cruickshank 2002). The family Tetranychidae is among the first Acari groups used in molecular identification (Navajas et al. 1992). However, 30% of the sequences belonging to Tetranychidae in the public GenBank have been reported to be unreliable (de Mendonça et al. 2011). The internal transcribed spacer (ITS) nuclear gene region (ITS1-5.8S-ITS2) and mitochondrial Cytochrome oxidase I (COI) have been successfully used in the molecular identification of spider mites (Ben-David et al. 2007; Ros and Breeuwer 2007; Matsuda et al. 2014). Still, additional sequences are required to fully elucidate the intra- and inter-specific variations that allow more precise species identification.

Although about 30 spider mite species have been recorded in Turkey so far (Çobanoğlu et al. 2015; Altunç and Akyazı 2020; Migeon and Dorkeld 2022), Tetranychus urticae Koch is the only species having COI sequences (İnak et al. 2019), and there are no ITS sequences obtained from spider mites in Turkey. In this study, we molecularly characterize ten tetranychid species using two genes collected from various hosts in Turkey. In addition, a new record belonging to the genus Eotetranychus has been reported using integrative taxonomy.

Materials and methods

Collection and identification of tetranychid mites

Tetranychid mites were collected from various regions of Turkey during 2019–21. Only plants that showed spider mite damage were sampled. Mites were transferred to 70 and 96% alcohol for morphological and molecular identification, respectively. Locations, hosts, collection dates and accession numbers are presented in Table 1.

Table 1 Collection dates and locations, host plants and accession numbers of spider mite species obtained in this study
Full size table

Hoyer’s medium was used for the permanent slides and male individuals have been positioned laterally to examine the aedeagus. The permanent slides are deposited at Ankara University, Department of Plant Protection, and at INRAE (Montpellier, France) in case of future verification.

Morphological identifications were performed mainly using the following publications: Pritchard and Baker (1955), Jeppson et al. (1975), Hatzinikolis and Emmanouel (1991), Zhang (2003), Auger et al. (2003) and Auger and Migeon (2014). In addition, original descriptions of Eotetranychus species were used when necessary.

Genomic DNA extraction

The mites kept in 96% alcohol were placed on filter paper allowing for the evaporation of alcohol. Genomic DNA was extracted from individual adult females using a Qiagen DNaeasy Blood and Tissue Kit following the manufacturer’s instructions. In the final step, 50 μl of elution buffer was used for each sample (İnak et al., 2020). DNA extracts were stored at −20 °C until used in PCR reactions.

Amplification of target gene/gene regions

Primers used for amplifying the target genes (ITS and COI) are given in Table 2. The PCR reaction was performed in a total volume of 30 μl, containing 5 μl of mite DNA, 0.5 μl of both forward and reverse primers, 18 μl of ultrapure nuclease-free water and 6 μl of FIREPol Master Mix (Solis Biodyne, Estonia).

Table 2 Primers and melting temperatures (TM) used in amplification of the target gene/gene regions
Full size table

DNA amplification was performed with a thermal cycler (BioRad T100) under the following conditions: 4 min at 94 °C, 40 cycles of 60 s at 92 °C, 60 s at 48–51 °C and 90 s at 72 °C, followed by a final extension of 5 min at 72 °C. Electrophoresis was carried out on a 1.5% agarose gel in 0.5 × TAE buffer for 35 min at 100 V to verify the success of the PCR.

The post-PCR products were purified using the HighPrep PCR clean-up system (MagBio Genomics, Gaithersburg, MD, USA) and sequencing was performed at the company BM (Ankara, Turkey).

Phylogenetic analysis

The obtained sequencing chromatographs were checked for errors based on visual inspection using BioEdit v.7.0.5 (Hall 1999). Then, all sequences herein obtained (MW535070-MW535104 for COI, MW534197-MW534222 for ITS) and retrieved from the public GenBank were aligned using MAFFT v.7 with ‘Auto’ strategy (Katoh et al. 2019) and refined manually. Maximum likelihood (ML) phylogenetic trees were built with MEGA X using the GTR + G + I for COI and T92 + G + I model for ITS (identified to be the best-fit model in MEGA X) (Tamura 1992) with 1000 bootstraps.

Intra- and interspecific genetic distances were calculated using all available sequences in the public GenBank (see Table S1 for the accession numbers of the used sequences) with the K2Parameter model in Mega X via selecting the target sequences as groups and computing the distances between/within them (Kumar et al. 2018).

Results and discussion

Although most of the species obtained in the present study have been reported previously in Turkey, no sequence data belonging to these species (except COI sequences of T. urticae) have been available so far. Considering the importance of the determination of genetic variation within certain species from different geographical areas, there was a gap in the literature for the sequences of Turkish spider mites.

A total of 70 sequences (COI + ITS, 35 sequences for each) belonging to ten tetranychid species have been obtained: Bryobia kissophila van Eyndhoven, Bryobia rubrioculus (Scheuten), Eutetranychus orientalis (Klein), Eurytetranychus buxi (Garman), Amphitetranychus viennensis (Zacher), Eotetranychus populi (Koch), Eotetranychus quercicola Auger & Migeon, Panonychus ulmi (Koch), T. urticae and Tetranychus turkestani Ugarov & Nikolski. Morphological identification was further supported by sequence comparison (COI max distance = 1.74%) and thus the integrative approach allowed to identify a new record of E. quercicola for the mite fauna of Turkey—this species was only known from South France, collected on Quercus pubescens (see Auger and Migeon 2014 for the detailed description).

The phylogenetic tree based on COI sequences is presented in Fig. 1. The tree shows a good clustering pattern at the genus level, although it was not supported by high bootstrap values. The genus Eotetranychus shows a polyphyletic evolutionary history in line with Matsuda et al. (2014; 2018). The genera Eutetranychus and Eurytetranychus, which belong to the tribe Eurytetranychini, clustered together, in line with Matsuda et al. (2014), supporting their closely related history. However, there is only a limited number of COI sequences belonging to spider mites in the public GenBank; therefore, more sequences from various species are needed to analyse the suitability of COI as a molecular marker for genus-level discrimination. At the species-level, the COI gene seems to allow species discrimination with some exceptions (i.e., closely related species such as T. urticae and T. turkestani, only 1.9% genetic difference between Turkish specimens), in line with Hinomoto et al. (2007) and Ros and Breeuwer (2007). Similarly, Matsuda et al. (2013) separated 13 tetranychid species from Japan using COI sequences successfully, on the other hand, only 10 out of these 13 species have been identified using the ITS region.

Fig. 1
figure 1

Phylogenetic tree of obtained tetranychid species based on COI sequences. The red dots indicate the sequences obtained in the present study. Only the bootstrap values higher than 60% are shown

Full size image

The intra- and inter-specific genetic variation for three tetranychid species that have more than 25 sequences available in GenBank were assessed (Table 3). The gap between intra- and inter-specific genetic distance (also known as barcoding gap) indicates the usefulness of COI sequences in species identification of analysed spider mites. Panonychus ulmi and B. kissophila had the lowest (1.09%) and highest (4.48%) intra-specific genetic distance, respectively. Besides commonly distributed species herein analysed, assessment of barcoding gaps should be performed for cryptic and closely related spider mite species where species delimitation is of significant importance in further studies. Rather than using a single or few methods in species delimitation, a wide range of analyses should be applied and all available data should be integrated to establish a standard barcoding gap (Carstens et al., 2013).

Table 3 Mean (± SE) intra- and interspecific genetic distances for four tetranychid species based on COI sequences
Full size table

Although Navajas et al. (1999) and Choi et al. (2018) reported higher genetic distances within A. viennensis when all sequences available in GenBank were analysed (up to 4.1 and 6.9%, respectively), the mean genetic distance was 2.63%. Still, COI has been determined to be a suitable marker for identification of A. viennensis in the present study, in line with Arabuli et al. (2019) that discriminated all three species belonging to the genus Amphitetranychus using COI sequences.

ITS2 evolves very quickly, which makes it a suitable marker in phylogenetic analysis (Yao et al. 2010). Although alignment of ITS2 sequences might sometimes be difficult due to its high sequence divergence, its secondary structure has been conserved considerably. It therefore allows identification at species-level (Schultz and Wolf 2009; Koetschan et al. 2010). In addition, Navajas et al. (1998) reported that ITS evolves 2.5 × faster than COI in spider mites, indicating the presence of a high level of differentiation within and between species.

The ITS2-based phylogenetic tree is presented in Fig. 2. Contrary to the COI-based tree, the genera Amphitetranychus and Tetranychus were seperated with high bootstrap values (91%) in the tree constructed using ITS2 sequences. Both trees showed the monophyly of the genera Tetranychus, Amphitetranychus and Bryobia, although the genus Tetranychus has been suggested to be polyphyletic (Matsuda et al. 2014, 2018). However, the number of sequences analysed in the present study is not sufficient to show a high-resolution phylogenetic relationship. The dataset and markers analysed (even different partial fragments of the same gene/gene regions) could also lead to such incongruence in the results. In addition, concatenated trees using multiple genes or phylogenomics, if possible, may allow overcoming this issue (Gadagkar et al. 2005; Delsuc et al. 2005). Still, misidentified and contaminated DNA sequences should also be taken into account before drawing a conclusion (de Mendonça et al. 2011).

Fig. 2
figure 2

Phylogenetic tree of obtained tetranychid species based on ITS sequences. The red dots indicate the sequences obtained in the present study. Only the bootstrap values higher than 60% are shown

Full size image

Tetranychus urticae and T. turkestani were separated according to nucleotide substitution at certain positions in ITS as previously described (Navajas and Boursot 2003; de Mendonça et al. 2011). These species did not cluster together in the ITS-based phylogenetic tree (contrary to the COI-based tree), which might allow the discrimination of even such closely related species using phylogenetic trees, in line with Navajas and Boursot (2003) and Ben-David et al. (2007).

The discrepancy between the phylogenetic trees based on nuclear and mitochondrial markers (although showing similar higher-level taxonomic composition) might be associated with insufficient data, homoplasy, incomplete lineage sorting and different hybridization events.

Conclusion

Accurate and early identification of spider mites is the initial step in pest management. Although DNA-based species identification is gaining more and more importance, the lack of reliable reference sequences and knowledge on genetic variation within and between genera/species often limits the use of this technique. In this study, ten tetranychid species collected in Turkey were molecularly characterized using the COI and ITS sequences. Whereas almost all the sequences obtained represent the first sequences from Turkey, some of them were submitted to the public GenBank for the first time. In addition, DNA-based species identification allowed us to provide a new spider mite record for Turkey. In the future, together with increasing reference sequences, it might be possible to perform faunistic studies using molecular data for mite groups as indiviual samples or pools of mites/samples using next generation sequencing. In this context, the establishment of a reference mite database based on integrative taxonomy that provides more reliable results would substantially contribute to all aspects of acarological studies including ecology, evolution, biological control and pest management. Finally, the number of species and individuals used in genetic distance assessments varied from one specimen to another, which might cause misconceptions; therefore, more sequences are required to obtain more accurate results.