Wolbachia, Hertig, 1936

Iorgu, Ionuț Ştefan, Iorgu, Elena Iulia, Stalling, Thomas, Puskás, Gellért, Chobanov, Dragan, Szövényi, Gergely, Moscaliuc, Liviu Aurel, Motoc, Rozalia, Tăuşan, Ioan & Fusu, Lucian, 2023, Ant crickets and their secrets: Myrmecophilus acerŋorum is not always parthenogenetic (Insecta: Orthoptera: Myrmecophilidae), Zoological Journal of the Linnean Society 197 (1), pp. 211-228 : 214

TESTING FOR WOLBACHIA INFECTION

We also tested our M. acerƲorum samples for Wolbachia infections in an attempt to explain the cause of parthenogenetic populations. For this, we extracted genomic DNA from the whole individual and we screened them for Wolbachia using the W-Spec diagnostic primers: W-spec f (5′– CATACCTATTCGAAGGGATAG–3′) / W-spec r (5′– AGCTTCGAG TGAAACCAATTC–3′) ( Jeong et al., 2009) following the protocol implemented by Jeong et al. (2012).After the initial screening we genotyped the strains on three gene regions: ftsZ, aesp and 16S. In order to amplify the Wolbachia 16S rRNA gene we performed a nested PCR using an initial universal 16S rDNA primer set: 27f (5′–AGAGTTTGATCMTGGCTCAG–3′) / 1513r (5′–ACGGYTACCTTGTTACGACTT–3′) ( Weisburg et al., 1991) and a second primer set: 76f (5′–TTGTAGCCTGCTATGGTATAAYT–3′) / 1012r (5′–GAATAGGTATRATTTYCATGT–3′) ( O’Neill et al., 1992). The reaction protocol and PCR conditions were adapted from Jeong et al. (2012). The partial Wolbachia aesp gene for outer surface protein was obtained using the 81F (5′–TGGTCCAATAAGTGATGAAGAAAC–3′) / 691R (5′–AAAAATTAAACGCTACTCCA–3′) primer set ( Zhou et al., 1998) following the PCR conditions from Vaishampayan et al. (2007). Finally, the ftsZ gene was amplified using the primers ftsZ_F1 (5′–ATYATGGARCATATAAARGATAG– 3′) / ftsZ_R1 (5′–TCRAGYAATGGATTRGATAT–3′) and the PCR protocol from Baldo et al. (2006).

PHYLOGENETIC ANALYSES

Sequences were aligned using MEGA 7 ( Kumar et al., 2016). We also compared our COI sequences with those found in GenBank and BOLD databases (Supporting Information, Table S1 ). Pairwise genetic distances were estimated using the Kimura 2-parameter (K2P) nucleotide substitution model implemented in MEGA 7. Haplotype identities and diversity (h), as well as nucleotide diversity (Π) were determined using DnaSP v.5.1 ( Librado & Rozas, 2009). Haplotype networks were constructed using PopART v.1.7 ( Leigh & Bryant, 2015) with a medianjoining algorithm ( Bandelt et al., 1999), in order to illustrate phylogenetic and geographic patterns. The best-fit model for nucleotide substitution, for each of the two mtDNA markers (16S and COI), was estimated using jModelTest v.2.1.5 ( Guindon & Gascuel, 2003; Darriba et al., 2012) with the Akaike information criterion (AIC). Also, the best partitioning scheme for the concatenated set was determined using PartitionFinder2 ( Lanfear et al., 2016), both softwares were implemented on the CIPRES Science Gateway ( Miller et al., 2010).

Phylogenetic trees were generated using the maximum likelihood (ML) and Bayesian inference (BI) methods. We used RAxML v.8 ( Stamatakis, 2006) implemented online on the CIPRES Science Gateway for ML analyses, using 1000 bootstrap resampling and other parameters as default. Mr Bayes v.3.2 ( Ronquist & Huelsenbeck, 2003) was used for BI analyses. Two separate runs, with four Monte Carlo Markov chains were used with 10 × 106 generations with trees sampled every 1000 generations, the first 25% of samples being discarded as burn-in. BI posterior branch probabilities were calculated by the majority-rule consensus of the sampled trees and obtained tree diagrams were visualized using FigTree v.1.4.2 ( Rambaut, 2014).

SEQUENCE- BASED SPECIES DELINEATION TESTS

In order to identify independent evolutionary units or operational taxonomic units (OTUs) in our datasets, we applied four DNA sequence-based species delineation approaches on our COI sequences as well as on concatenated sequences of COI and 16S. The COI gene was used for delimiting species in many taxonomic groups, providing evidence for independently evolving lineages and recognizing genetic patterns within groups, to support morphological evidence or other traditional taxonomic studies ( Hamilton et al., 2014; Costa-Silva et al., 2015).

The first test is the statistical parsimony network analysis implemented in TCS v. 1.21 ( Clement et al., 2000). The program computes the maximum number of mutations that constitutes a parsimonious connection between two haplotypes with a 95% probability and then reconstructs a network following the algorithms of Templeton et al. (1992). Each network is considered an OTU. The second test is the automatic barcode gap discovery (ABGD), implemented on the ABGD web platform ( Puillandre et al., 2012). The principle of the test is that it uses the so-called ‘barcodegap’ in the distribution of the pairwise differences from the COI sequences to assign organisms into hypothetical species. The third test is the general mixed yule-coalescent model (GMYC) applied on another web-based server (https://species.h-its. org/gmyc/). The GMYC method is a likelihood method for delimiting species by fitting within- and between-species branching models to reconstructed gene trees ( Fujisawa & Barraclough, 2013). The fourth test is the Poisson Tree Process (PTP model). It is a model for delimiting species on a rooted phylogenetic tree, implemented also on an online platform (https://species.h-its.org/ptp/; Zhang, 2013; Zhang et al., 2013).