DNA elution was conducted with TE buffer (10 mM Tris, 1 mM EDTA, pH 8). Mitochondrial DNA fragments of approximately 920 bp were Romidepsin amplified by PCR. These fragments are part of the cytochrome oxidase I gene (approximately 780 bp), leucine transfer RNA (70 bp), and part of the cytochrome oxidase II (approximately 60 bp).

The amplifications were carried out with a final volume of 25 μL, containing 250–500 ng of DNA template, 0.2–0.4 μM (5–10 pmol) of each primer, using the Ready-to-go kit (Amersham Pharmacia Biotech). The thermal cycler was programmed as proposed by Ross and Shoemaker (1997): 1 min at 94 °C (initial denaturation) and 35 cycles at 94 °C for 1 min, annealing temperature of 48 °C for 1 min, and extension temperature of 68 °C for 2 min, followed by a final extension step at 72 °C for 5 min. The primers used were: C1-J-2195 (COI-RLR) (5′-TTGATTTTTTGGTCATCCAGAAGT-3′) and DDS-COII-4 (5′-TAAGATGGTTAATGAAGAGTAG-3′) (Ahrens et al., 2005 and Ross and Shoemaker, 1997). When the combination of primers did not amplify the desired fragment, the second primer was used instead of DDS-COII-4,

named JerryGarcia-CI (5′-GGGAATTAGAATTTTGAAGAG-3′) (Shoemaker et al., 2006), which produces fragments of approximately 780 bp that includes only the gene cytochrome oxidadese I (COI). Two pairs of primers were used to examine the presence of Wolbachia in ants. The first pair was the control: EF1α-532F (5′-AGGCAAATGTCTTATTGAAG-3′) and EF1α-610R (5′-GCGGGTGCGAAGGTAACAAC-3′) ( Shoemaker et al., 2000) that amplify a fragment of 400 bp of the nuclear gene EF1α (elongation factor). The second pair amplifies Sorafenib the variable fragment of a gene that decodes a surface protein of the bacteria of approximately 600 bp, named wsp81F (5′-TGGTCCATTAAGTGATGAAGAAAC-3′) and wsp691R (5′-AAAAATTAAACGCTACTCCA-3′) ( Braig et al.,

1998 and Zhou et al., 1998). The presence of the control primer (EF1α) fragment and the absence of the Wolbachia-specific fragment (wsp) most likely reflects an absence of the bacteria rather than low quality (low yield of PCR product), a high concentration of genomic DNA or an error associated with the PCR setup ( Shoemaker et al., 2000). However, in the Thymidylate synthase absence of the EF1α fragment and of the wsp gene fragment, it is not possible to conclude the absence of the endobacteria. In this case, the genomic DNA was diluted and the PCR protocol repeated. The amplifications were carried out with final volume of 25 μL, with 250–500 ng of DNA template, 0.2–0.4 μM (5–10 pmol) of each primer, using the Ready-to-go kit (Amersham Pharmacia Biotech). The thermal cycler was programmed according to Braig et al. (1998) and Zhou et al. (1998). The confirmation of the amplification was visualized in 2% agarose gel. The presence of noise in the electropherogram of the sample of the sequenced wsp gene required cloning of the sample to separate the strains. PCR products were cloned using the CloneJET PCR Cloning Kit (Fermentas Life Sciences).