More than fishing in the dark: PCR of a dispersed sequence produces simple but ultrasensitive Wolbachia detection

Background Detecting intracellular bacterial symbionts can be challenging when they persist at very low densities. Wolbachia, a widespread bacterial endosymbiont of invertebrates, is particularly challenging. Although it persists at high titers in many species, in others its densities are far below the detection limit of classic end-point Polymerase Chain Reaction (PCR). These low-titer infections can be reliably detected by combining PCR with DNA hybridization, but less elaborate strategies based on end-point PCR alone have proven less sensitive or less general. Results We introduce a multicopy PCR target that allows fast and reliable detection of A-supergroup Wolbachia - even at low infection titers - with standard end-point PCR. The target is a multicopy motif (designated ARM: A-supergroup repeat motif) discovered in the genome of wMel (the Wolbachia in Drosophila melanogaster). ARM is found in at least seven other Wolbachia A-supergroup strains infecting various Drosophila, the wasp Muscidifurax and the tsetse fly Glossina. We demonstrate that end-point PCR targeting ARM can reliably detect both high- and low-titer Wolbachia infections in Drosophila, Glossina and interspecific hybrids. Conclusions Simple end-point PCR of ARM facilitates detection of low-titer Wolbachia A-supergroup infections. Detecting these infections previously required more elaborate procedures. Our ARM target seems to be a general feature of Wolbachia A-supergroup genomes, unlike other multicopy markers such as insertion sequences (IS).


Background
Detecting endosymbionts such as the widespread alphaproteobacterium Wolbachia in its host cell environment requires reliable and ideally simple but still sensitive molecular marker systems. When such bacteria are present at high titers, classic end-point PCR is sufficient to unambiguously determine infection status of an unknown specimen. Particularly for Wolbachia, a quite comprehensive set of diagnostic PCR markers has been developed and applied successfully. The most commonly used among these makers is the multi locus sequence typing (MLST) system [1][2][3] and the four hypervariable regions (HVRs) of the Wolbachia outer surface protein gene wsp [4,5]. Both MLST, comprising a set of five singlecopy Wolbachia genes, and the wsp locus were demonstrated to be highly useful for Wolbachia infection determination and consequent diversity assessment. However, those marker systems are limited if the endosymbiont persists at very low titers within the host, either only during a certain ontogenetic stage [6] or throughout all life stages. In both cases proper detection of the endosymbiont is hindered and this points towards the need of an alternative strategy for efficient, robust and fast Wolbachia detection. One approach to address this issue is to use multicopy Wolbachia gene markers for PCR analyses. Particularly insertion sequences (IS; [7,8]) represent a good strategy to increase the detection threshold [9,10]. However, this approach relies on the conservation of such elements and their copy-numbers in diverse strains, which might not be the case over longer evolutionary distances due to the mobile nature of these elements. Another approach to cope with the detection problem introduced by low-titer infections is 'nested PCR'. This method might help to increase the detection threshold but is also highly prone to contamination [6]. A third strategy combines standard PCR with consequent hybridization [6,11,12], which increases overall detection limit by four orders of magnitude [6]. On the other hand, this is an elaborate and time-consuming technique. Hence, we set out to find a more sensitive marker for detection of low-titer Wolbachia infections using standard PCR and identified ARM as such a simple but 'ultra-sensitive' marker for A-supergroup Wolbachia.

Results and discussion
Identification of a multicopy marker associated with tandem repeats in A-supergroup Wolbachia genomes (ARM) To find a marker that serves a highly sensitive detection method of low-titer Wolbachia strains we identified multicopy regions in the A-supergroup wMel genome (Wolbachia of Drosophila melanogaster; GenBank NC_002978). An intergenic region of 440 bp associated with the recently described hypervariable tandem repeat region ( Figure 1; [13]) was the most promising candidate, hereafter called ARM (A-supergroup repeat motif) as it was found in 24 almost identical copies dispersed throughout the wMel genome (Additional file 1). However, for a marker to be useful as a general tool it also needs to be conserved and present in multiple copies in other strains and we therefore used the wMel repeat sequence to search an additional 13 draft and complete Wolbachia genomes from four different Wolbachia supergroups for the same sequence. We were able to identify the presence of the repeat in seven A-supergroup Wolbachia genomes (wHa, wRi, wWil, wAna, wUni, wSuzi and wGmm; see Table 1), albeit in variable copy numbers. In the Drosophila associated Wolbachia strains, the copy numbers were around 20 per genome (Table 1), whereas the other two A-supergroup genomes (wUni and wGmm) contained about half the amount of copies. Low number of hits in wUni is most likely explained by the incomplete status of the genome resulting in an underestimation of the actual copy number. In the B-(wNo, wVitB, wPip), C-(wOo, wOv), and D-supergroup (wBm) genomes, ARM was not found. Even though some of the genomes in supergroups B, C, and D are incomplete, the total absence of the repeat in all genomes from these supergroups suggests that this motif might be Wolbachia A-supergroup specific. Additionally, VNTR-tandem repeats associated with ARM in A-supergroup infections are also absent from genomes of B-to D-supergroups, further indicating that this feature might indeed be A-supergroup specific.

ARM facilitates detection of low-titer Wolbachia from A-supergoup
ARM-targeting primer were tested via end-point PCR screen on DNA from high-and low-titer Wolbachia infections in Drosophila and Glossina (tsetse fly) species (Additional file 2). As shown in Figure 2, the classic Wolbachia singlecopy gene marker wsp (Wolbachia outer surface protein gene) is only applicable for samples with high-titer infections, since Wolbachia was only detected in high-titer D. paulistorum Orinocan semispecies (OR, Figure 2A) as well as in D. willistoni (Dw + , Figure 2B), D. melanogaster (Dm + , Figure 2B), D. simulans (Ds + , Figure 2B) and Glossina morsitans morsitans (Gmm, Figure 2B). The wsp primer failed to detect Wolbachia in low-titer strains like D. paulistorum Amazonian (AM) and Centroamerican (CA) semispecies plus Glossina swynnertoni (Figure 2A,B), indicating that a singlecopy gene like wsp is not suited for tracking low-titer infections. As multicopy gene markers like insertion sequences (IS) can be used to increase the detection limit, we ran PCR using primer for Insertion Sequence 5 (IS5; [8][9][10] on the same sample set. We observed increased sensitivity compared to wsp-PCR since Wolbachia was detected in low-titer CA2 ( Figure 2A) and in the A/O hybrid samples. However, IS5 primer failed at amplifying the target sequence in all three Glossina samples (Gmm, Gsw and Gs/Gm hybrid; Figure 2B) despite the overall high Wolbachia titer in Gmm [12].
We have recently shown that Wolbachia titers increase in D. paulistorum [11] and Glossina [12] hybrid backgrounds, which should significantly facilitate detection and strain characterization. Such titer increase was sufficient to detect Wolbachia with the IS5 primer set in A/O hybrids, but the low-titer Wolbachia infection in the AM mother still remained undetected ( Figure 2B). Failure of IS5-amplification in the Gs/Gm hybrid plus parents is explained by lacking homology between primer sequences and target, as no matches with the IS5 primer sequence were found in the wGmm genome [14]. This finding implies that IS5 is not suitable as a general Wolbachia A-supergroup marker. Figure 2A and B show that the ARM-marker system can be applied to address aforementioned problems arising with wsp and IS5 primer: sensitivity during PCR is increased significantly and all tested A-supergroup infections are unambiguously detected. Wolbachia was traced in all low-titer New world Drosophila species (AM1, AM2; CA1, CA2) plus the A/O hybrid. In contrast to IS5, the ARM primer set amplified Wolbachia from all three Glossina samples (Gmm, Gsw and Gs/Gm hybrid). As anticipated, all samples from high-titer Wolbachia infections (OR, Dw + , Dm + , Ds + ) showed bright bands with ARM, whereas Wolbachia-uninfected specimens (Dw -, Ds -) did not (Figure 2A,B). This argues for a high specificity of the ARM primer and against mis-amplification of a random host target rather than the specific symbiont target site.

Conclusions
We suggest that the new multicopy Wolbachia Asupergroup marker can be used as an 'ultra-sensitive' tool to trace low-titer infections by means of classic end-point PCR. First, ARM has the advantage of higher sensitivity compared to classic singlecopy Wolbachia markers like wsp and thus improves detection limit significantly. Particularly, ARM-PCR can be easily applied to screen larger numbers of untyped DNA specimens, even of low quality arising from long-term storage and/ or storage in inappropriate media, from laboratory stocks or samples directly from nature. This is of pivotal interest since classical detection tools might yield false negatives when examining species harboring Wolbachia at very low densities, and thereby lead to underestimating natural prevalence of A-supergroup infections. Given that 80% of the Dipteran infections are supergroup A [15], our new method will significantly facilitate and improve the sensitivity of such surveys. In addition our approach is an advantage over the classic IS5-marker, which fails in Wolbachia from the tsetse fly Glossina. Taken together, we show that a Wolbachia sequence motif found in multiple copies associated with the VNTR loci facilitates reliable Wolbachia screening of samples from low-titer infections and might thus serve as a great tool for the Wolbachia research community. Furthermore a similar approach might be applied to detect other symbionts such as Sodalis glossinidius (secondary symbiont of Glossina) and the primary symbiont Candidatus Sodalis pierantonius str. SOPE of the weevil Sitophilus orizae. Both symbiont genomes exhibit more than 20% of repetitive DNA rendering them appropriate candidates for repeatbased PCR analysis [16,17]. However, we anticipate that such a method reaches its limit when dealing with symbiont genomes, which have become highly streamlined in the course of tight host-symbiont coevolution.

Drosophila and Glossina strains plus hybrid samples
Drosophila specimens included members of New world and Old world clades (Additional file 2). Representatives of the new world clade were Drosophila paulistorum semispecies AM, CA and OR, together with Wolbachia- infected (Dw + ) and -uninfected (Dw -) D. willistoni (see Additional file 2 for details). The Old world clade was represented by Wolbachia-infected D. melanogaster (Dm + ) and Wolbachia-infected (Ds + ) and uninfected (Ds -) D. simulans (Additional file 2). Additionally, the tsetse fly species Glossina swynnertoni and G. morsitans morsitans (genus Glossina, superfamily Hippoboscoidea) and hybrids from D. paulistorum (A/O) and Glossina (Gs/Gm) were included (Additional file 2). Detailed descriptions of establishing hybrid samples can be found in [11,12].  Table 1; www.ncbi.nlm. nih.gov) and other analyses were performed using Geneious 5.6.6 software (Biomatters, New Zealand).
Diagnostic wsp-, IS5-, ARM-and 12S rRNA-PCR Primer pairs for diagnostic wsp-PCR were taken from [19] and the corresponding PCR set-up is described in [11]. Primers and PCR profile for IS5 can be found in [9]. We designed the following primer set targeting ARM: ARM-F 5'-TTCGCCAATCTGCAGATTAAA-3' and ARM-R 5'-GTTTTAAACGCTTGACAA-3'. Both primers are positioned in the flanking regions of the VNTR-105 locus in wMel [9,13], and produce an amplicon of 315 bp constant size. Composition of the locus is shown in Figure 1. Diagnostic ARM-PCR was performed in 20 μl reactions containing 1x reaction buffer, 3.0 mM MgCl 2 , 0.4 μM of forward and reverse primer, 35 μM dNTPs, 0.4 U of Taq Polymerase (Promega) and 2 μl of DNA template. PCR was performed using a profile of 2 min initial denaturation at 94°C followed by 30 cycles consisting of 45 sec denaturation at 94°C, 45 sec annealing at 55°C, and 1 min extension at 72°C. Final extension was performed for 10 min at 72°C. In order to assess DNA quality, we amplified part of the mitochondrial 12S rRNA gene with primer set 12SCFR 5′-GAGAGTGACGGGCGATATGT-3′ and 12SCRR 5′-AAACCAGGATTAGATACCCTAT-TAT-3′ [20]. PCR conditions are outlined in [21]. PCR amplicons were examined using gel-electrophoresis on a 1% agarose gel pre-stained with 0.05 mg ethidium bromide.

Ethics statement
This study did not involve any subjects and materials that require approval by an ethics committee (human, vertebrate, regulated invertebrates). No genetically modified organisms were part of this study.

Additional files
Additional file 1: Positions of ARM in the wMel and wRi genomes. Circular schemes of the wRi (Wolbachia symbiont of Drosophila simulans; NC_012416; [22]) and wMel genomes (Wolbachia, endosymbiont of D. melanogaster; NC_002978; [8]), showing that ARM (indicated by black bars) is equally dispersed throughout the genomes.
Additional file 2: Detailed information on Drosophila and Glossina specimens used in this study. First column refers to the abbreviated code used for each specimen in text, figures and figure legends. Last column lists reference and/or collector's name [31,11,[32][33][34]12].

Competing interests
The authors declare that they have no competing interests.
Authors' contributions DIS and WJM conceived the study. DIS, LK, AEL and WJM designed and performed the experiments. WJM provided material. DIS, LK, AEL and WJM analyzed the data. DIS, LK and WJM wrote the manuscript. All authors read and approved the final version of the manuscript.