Methods for the rapid, specific identification of pathogenic Burkholderia species are important for timely recognition of glanders or melioidosis in patients with general clinical symptoms that could fit various diseases. Both B. pseudomallei and B. mallei are intrinsically resistant to many widely-used antibiotics, and susceptibilities differ between the two species
[6, 14]. Identification of and differentiation between the two pathogens could thus help healthcare workers to choose the right antibiotics to treat infected patients.
In regions where these species are endemic, infection with Burkholderia mallei or B. pseudomallei can often be pinpointed by consideration of prevalences and infection risk factors. The presence of the pathogens is less evident in non-endemic areas. Numerous cases of imported melioidosis have been reported
[3, 6], and are likely to continue to occur as travel and trade increase. Fast diagnostic methods are therefore essential for use in non-endemic areas, not only to support appropriate patient treatment but also to ensure the safety of laboratory workers culturing unknown organisms for diagnostic purposes. Another scenario in which rapid identification of pathogenic Burkholderia including species distinction is a challenging but essential task is that of deliberate release of these biothreat agents, where appropriate assay methods may be needed e.g. for forensic tracking.
The assay presented in this paper permits sensitive, reliable detection of pathogenic Burkholderia, thanks to the use of qPCR (real-time polymerase chain reaction). It reduces false-negative measurements, thanks to the inclusion of an internal control and the high sensitivity that was achieved.
The idea of adding spores from B. thuringiensis as internal controls was based on their properties as highly refractory biological structures. The detection of the B. thuringiensis signature sequence cry1 guarantees successful DNA extraction and amplification from any microorganism in the sample
[11, 12]. The target BuMC was used as a sensitive, specific indicator for the presence of pathogenic Burkholderia strains. The selection of this target based on its presence in multiple copies in all 21 publicly available genomes and its usefulness for sensitive detection was evidenced by the in vitro validation showing lower Cq values (Table
2). However, the benefit of more sensitive detection did not hold for all strains, since BuMC could not be amplified from three of the B. pseudomallei strains tested (Table
2). The absence of this target from the clinical samples is congruent with its absence from strain BD10-00211, which had been isolated from the patient from which the clinical samples were taken. It is possible that the rather diverse species B. pseudomallei contains a phylogenetic cluster that has the absence of the targeted ISBma2 transposase gene as a common feature. Further research is required to substantiate this assumption, however.
We did not find any strains that were not detected by the species-specific signature sequences we developed. However, probe design is always limited by the available sequences and strains. Hence, the possibility that some strain exists or could arise which does not possess one of the signature sequences used in this assay cannot be excluded. This is true of any assay, however, and the multiplex qPCR assay described here has the advantage of possessing two signatures for each strain, thus reducing the risk that some strains will escape detection.
The measured linearity and efficiency show that the qPCR assay is very suitable for quantitative measurement. The calculated LODs were very low, particularly when based on multicopy target sequence BuMC. The detection limits of 0.2 fg per reaction for B. mallei and 3.7 fg per reaction for B. pseudomallei (Table
3) correspond to approximately 0.03 and 0.5 genomic equivalents respectively. The LODs were lower than or similar to those reported for other assays
[6, 8, 14–19]. However, it is difficult to make a direct comparison between reported LODs due to the differences between the methods used to measure and calculate them, and to measure the DNA concentration of the standards. We used the DNA intercalating dye picogreen for accurate determination of the concentration of double-stranded DNA in our stocks and probit analysis as a basis for robust calculation of the concentration at which the probability of detecting the target is 95%
The high reliability and sensitivity of the qPCR assay described here make it very useful for screening of samples containing few organisms and potential inhibitors, as is the case in many environmental and clinical samples. It can furthermore be used to supplement other assays, including molecular assays based on other signature sequences, for definitive identification. Burkholderiaceae are highly recombining organisms
[6, 9, 10, 18] and emerging novel strains will continue to challenge the coverage and sensitivity of detection assays. This qPCR assay offers the potential of continuing to meet this challenge effectively in the foreseeable future.