Wolbachia detection in insects through LAMP: loop mediated isothermal amplification

Background The bacterium Wolbachia is a promising agent for the biological control of vector-borne diseases as some strains have the ability to block the transmission of key human disease-causing pathogens. Fast, accurate and inexpensive methods of differentiating between infected and uninfected insects will be of critical importance as field-based trials of Wolbachia-based bio-control become increasingly common. Findings We have developed a specific and sensitive method of detecting Wolbachia based on the isothermal DNA amplification. This technique can be performed in an ordinary heat block without the need for gel-based visualisation, and is effective for a wide variety of insect hosts. Conclusion Here we present the development of a rapid, highly sensitive and inexpensive method to detect Wolbachia in a variety of insect hosts, including key mosquito disease vectors.


Findings
Vector-borne diseases (e.g. malaria, dengue, filariases and Chagas disease) occur in more than 100 countries, primarily within the tropics, with the annual, global death rate in the millions. Despite the existence of a variety of vector control measures disease incidence is generally increasing, and consequently there is an urgent need to develop new and effective control strategies [1]. Ideally these strategies should be able to be used in conjunction with existing control methods, and in this context the bacterium Wolbachia pipientis has proven to be a promising alternative, given its ability to restrict pathogen development in many different mosquitopathogen combinations [2][3][4][5], and recently being applied in the field in Australia, Vietnam, Indonesia and soon in other dengue endemic countries (www.eliminatedengue.org).
Wolbachia is a vertically-transmitted bacterial endosymbiont of arthropods that is able to manipulate its host's reproductive system and consequently spread rapidly through wild populations [6]. Wolbachia was originally identified in the ovaries of the mosquito Culex pipiens [7], and recent studies have estimated that 40% of terrestrial arthropod species are infected with the bacterium [8]. Critically, infections do not occur in key vector species including the dengue mosquito Aedes aegypti or in the Anopheles genus [9]. Wolbachia is typically detected within the host by PCR against genes such as wsp (Wolbachia surface protein) [10], ftsZ (cell division protein) [11], and 16S ribosomal protein [12], or through fluorescence-based assays [13]. These techniques can be expensive to perform for large numbers of samples, and require the use of laboratory equipment.
The LAMP (loop mediated isothermal amplification) technique utilizes Bst DNA polymerase to create standdisplacement amplification, and operates at a constant temperature [14]. Examples of its use are very broad, ranging from the detection of pathogenic microorganisms to sex determination in embryos [15]. The LAMP product can be detected by the visualisation of turbidity, fluorescence or a metal ion indicator producing a colorimetric change [16] without the need to run a gel. These characteristics make LAMP a suitable method to detect Wolbachia under circumstances where minimum infrastructure is available. Here we report the use of LAMP to detect Wolbachia in a diverse range of insect species, which will prove beneficial to monitoring the progress of fieldbased trials utilising the bacterium.

Methods
We examined many different species including laboratoryreared mosquito lines of Aedes fluviatilis, naturally infected by the Wolbachia strain wFlu, A. aegypti transinfected with the wMelPop and wMel strains [17][18][19], and uninfected Anopheles aquasalis mosquitoes. These lines were reared at FIOCRUZ Minas as previously described [20]. We also examined field-captured mosquitoes including A. aegypti, A. albopictus, and Culex sp. Furthermore, insects belonging to a diverse range of orders, which were known to be Wolbachia-infected (unpublished results) were also included.

Results and discussion
Loop mediated isothermal amplification has been used to detect different organisms, such as Leishmania [22], Toxoplasma gondii [23] and also bacteria belonging to the same class as Wolbachia (Alphaproteobacteria); Brucella spp [24], Anaplasma ovis [25] and Ehrlichia ruminantium [26]. We designed a LAMP-based assay to detect Wolbachia in different hosts using the bacterial 16S rRNA sequence [12]. Initially we designed the four basic LAMP primers (F3, B3, FIP and BIP) but due to low specificity we then included the extra Loop primers [14].
To optimize the LAMP reaction for Wolbachia, samples of A. fluviatilis and wMel-infected A. aegypti were used to conduct a time course assay where we determined that  amplification first occurred after 60 minutes of incubation (at 63°C using a thermocycler -Additional file 1). This was subsequently standardized at 90 minutes in order to account for samples with low bacterial density. Assay sensitivity was assessed by serially diluting plasmid DNA containing the target sequence. Dilutions containing 10 7 , 10 5 , 10 3 , 10 1 and 10°copies of the target gene were incubated for 90 min, using a thermocycler ( Figure 1A) or heat block ( Figure 1B). Sensitivity was high using either method, with amplification observed from as little as 10°c opies of the target sequence, although it should be noted that amplification from plasmids is more effective than from genomic DNA.
To check the specificity of our assay we analyzed mosquito specimens (Diptera: Culicidae) both infected and uninfected with Wolbachia. No amplification was observed in uninfected specimens, suggesting that the assay was highly specific and did not react to other bacteria infecting the mosquito gut [27]. We were also correctly able to identify Wolbachia infection in insect specimens from other orders ( Figure 2B). It is important to mention though, as lateral gene transfer events are quite common in Wolbachia [28] further confirmation of the presence of this bacterium (MLST strain typing or in situ hybridization for instance) will likely be appropriate in specific studies, where the high throughput screening for positive individuals is not the main objective.
The assay included the metal ion indicator Hydroxy Naphthol Blue [16] (which reacts with the pyrophosphate ion by-product of the reaction) to check whether we could detect Wolbachia without running agarose gels. We were able to clearly visualise a difference in colour after incubating the samples of A. fluviatilis (infected with wFlu) and A. aegypti (wild type and infected with wMel), besides the negative A. aquasalis, for 150 min at 63°C in the thermocycler (Figure 3). This step could be utilized when access to laboratory equipment is limited and also has an advantage of reducing the risk of contamination (no need to open the tubes).
In terms of the cost per reaction, LAMP is approximately half as expensive as conventional PCR. If the cost of equipment is also included, LAMP becomes comparatively even cheaper given that it can be performed using only a heat block. As a further cost-saving measure, or if required in the field, homogenization of specimens could be performed using micro-pestles rather than a beadbeater.  (7). We also examined insects from a range of different orders that were known to be infected with Wolbachia: (B) Hymenoptera (8), Hemiptera (9), Lepidoptera (10), Orthoptera (11), Siphonaptera (12), Coleoptera (13), Isoptera (14). M = 1 Kb Plus, Invitrogen. NC = negative control. All Wolbachia-infected specimens amplified, while no amplification was observed for uninfected specimens indicating that the assay was both highly reliable and highly specific. We have demonstrated the reliability and sensitivity of using LAMP to detect Wolbachia, with detection possible for low-density infections, and the results easily visualized by the naked eye. Consequently, this technique could be readily applied in the field, without the need to use expensive laboratory equipment. This is the first report of the use of LAMP to detect Wolbachia infection in different hosts. The technique could potentially be applied to differentiating between multiple strains that could potentially occur within the same host species.

Additional file
Additional file 1: Standardization of the LAMP assay incubation time.
To determine the optimal incubation time for the assay, LAMP reactions were performed using; (1) Ae. fluviatilis naturally infected with Wolbachia (wFlu), and (2) Ae. aegypti artificially infected with wMel. Samples were incubated at 63°C in a thermocycler and the tubes were removed at different times: 30 min (a), 60 min (b), 90 min (c), 120 min (d), 150 min (e). The products were visualised on a 1.5% agarose gel stained with EtBr. Visible products were present after 60 minutes of incubation. NC = negative control.