| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Bundeswehr Institute of Microbiology, Munich, Germany;2 Qiagen Diagnostics GmbH, Hamburg, Germany;3 Military Medical Training School, Vienna, Austria;4 Central Veterinary Research Laboratory, United Arab Emirates;
aaddress correspondence to this author at: Bundeswehr Institute of Microbiology, Neuherbergstrasse 11, 80937 Munich, Germany; fax 49-89-3168-3292, e-mail herbert.tomaso{at}web.de
Abstract
Background: Burkholderia mallei is a potential biological agent that causes glanders or farcy in solipeds, a disease notifiable to the Office International des Epizooties (OIE). The number of reported outbreaks has increased steadily during the last decade, but diagnosis is hampered by the low bacterial load in infected tissues and excretions.
Methods: We developed a B. mallei-specific 5'-nuclease real-time PCR assay that targets the fliP gene of B. mallei and includes an internal amplification control. Specificity was assessed with 19 B. mallei strains, 27 Burkholderia pseudomallei strains, other Burkholderia strains of 29 species, and clinically relevant non-Burkholderia organisms.
Results: Amplification products were observed in all B. mallei strains but in no other bacteria. The linear range of the B. mallei real-time PCR covered concentrations from 240 pg to 70 fg of bacterial DNA/reaction. The detection limit was 60 fg of B. mallei DNA. The clinical applicability of the assay was demonstrated by use of organ samples from diseased horses of a recent outbreak that was reported to the OIE by the United Arab Emirates in 2004.
Conclusions: Compared with conventional PCR, our rapid 5'-nuclease real-time PCR assay for the specific identification of B. mallei has a lower risk of carryover contamination and eliminates the need for post-PCR manipulations. This real-time PCR assay also shortens the turnaround time for results and has the potential for automation.
Burkholderia mallei is a gram-negative bacterium causing glanders and farcy in solipeds (1)(2). Glanders primarily causes pneumonia, purulent nasal discharge, and poor general condition, whereas farcy is a chronic cutaneous disease with nodules developing into ulcers. Equines are the only known reservoir, but sporadic human infections may occur (3)(4). Timely identification of the agent is crucial for adequate therapy, as B. mallei is resistant to many antimicrobial agents (5). The clinical picture of B. mallei infections resembles melioidosis, and detection is difficult because of the low number of bacteria in tissues and excretions (3)(6)(7). During the last decade, the number of reports on infections in horses in Asia and South America has increased (7)(8)(9)(10)(11)(12). Beyond that, B. mallei has been classified by the CDC as a potential category B biological agent, making the development of a specific and sensitive PCR assay an urgent need (13).
For our assay, we used the MX3000PTM real-time PCR system (Stratagene) for amplification and detection in 96-well plates (ThermoFast 96 ABGeneTM; Rapidozym). The oligonucleotides were designed based on differences of the flagellin P gene (fliP) sequences from B. mallei ATCC 23344T (accession nos. NC_006350 and NC_006351) and Burkholderia pseudomallei K96243 (accession nos. NC_006348 and NC_006349). Fluorogenic probes were synthesized with 6-carboxyfluorescein (6FAM) or Yakima Yellow (YAK) at the 5' end and black hole quencher 1 (BHQ1) or dabcyl (DB) at the 3' end. Oligonucleotides targeting fliP included the following: Bma-flip-f (5'-CCCATTGGCCCTATCGAAG-3'), Bma-flip-r (5'-GCCCGACGAGCACCTGATT-3'), and Bma-probe (5'-6FAM-CAGGTCAACGAGCTTCACGCGGATC-BHQ1). As an internal inhibition control, a bacteriophage 
-PCR system was used as described previously (14): -F (5'-ATGCCACGTAAGCGAAACA-3'), -R (5'-GCATAAACGAAGCAGTCGAGT-3'), and Lam-YAK (5'-YAK-ACCTTACCGAAATCGGTACGGATACCGC-DB-3'). Oligonucleotides were designed in cooperation with and obtained from TIB MOLBIOL. The 25-µL reaction mixture consisted of 12.5 µL of 2x TaqManTM Universal MasterMix (Applied Biosystems), 0.1 µL of each primer (10 pmol/µL), 0.1 µL of the TaqMan probes (10 pmol/µL), and 4 µL of DNA. Thermal cycling conditions were as follows: 50 °C for 2 min, 95 °C for 10 min, and 45 (or 50) cycles at 95 °C for 25 s and 63 °C for 1 min. A negative result was assigned when no amplification occurred or when the threshold cycle (CT) value was >40 cycles. In pilot experiments, primer concentrations and annealing temperature were optimized. In various experiments, positive CT values were confirmed by agarose gel electrophoresis. Bacteriophage -DNA was titrated to yield a CT value of 
35.
We assessed assay specificity with 19 B. mallei, 27 B. pseudomallei, other Burkholderia strains of 29 species, and clinically relevant non-Burkholderia organisms (see Tables 1 and 2 in the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol52/issue2/). Crude DNA was prepared as described previously (14). For quantitative assays, standard phenol–chloroform extraction was used. The DNA concentration was determined spectrophotometrically (GeneQuantTM RNA/DNA Calculator).
In 2004, an outbreak of glanders in horses was reported to the Office International des Epizooties (OIE) by the United Arab Emirates. Clinical samples of skin, liver, lung, spleen, and conchae from horses involved in the outbreak were fixed in formalin (100 mL/L), cut into pieces of 0.5 x 0.5 x 0.5 cm, washed twice in deionized water (10 mL), incubated overnight in sterile saline at 4 °C, and minced using liquid nitrogen and a mortar and pestle. Total DNA was prepared from 50 mg of tissue, using the QIAamp Tissue KitTM (Qiagen), and eluted with 80 µL of PCR-grade H2O.
We determined the linear dynamic range according to Clinical and Laboratory Standards Institute (CLSI; formerly NCCLS) standards (15), using B. mallei strain Bogor DNA in PCR-grade water in triplicate (Fig. 1
). We assessed the effect of horse DNA and the internal amplification control on the fliP assay by use of serial logarithmic dilutions of B. mallei DNA with and without the respective DNAs, performed in triplicate. The SD of the CT observed with 8 replicates was calculated. Statistical analysis was performed with SPSS 11 for Mac OSx (SPSS Inc.) and MicrosoftTM Excel X for MacTM (Microsoft GmbH).
|
Amplification products were observed in all B. mallei strains but in no other bacteria (Table 1
; also see Fig. 1 in the online Data Supplement). Agarose gel analysis showed the presence of amplification products of the expected length and the absence of other amplicons. Only at lower annealing temperatures did we observe nonspecific amplification curves, with CT values close to 40 for other Burkholderia strains. The linear range of the B. mallei real-time PCR covered concentrations from 240 pg to 70 fg of bacterial DNA/reaction (Fig. 1
). The lowest amount of DNA that was always detectable in 3 runs, with 24 measurements altogether, was 60 fg. The intraassay variability of the fliP PCR assay for 35 pg of DNA/reaction was assessed to be 0.68% (based on CT values) and for 875 fg was 1.34%, respectively. The SD for CT values obtained with the flip PCR assay was calculated to be <0.7 for DNA concentrations within the linear range. Horse DNA and the internal amplification control had no influence on the efficiency of the assay (see Fig. 2 in the online Data Supplement). B. mallei was detected in organs of 2 horses (Fig. 3 in the online Data Supplement). These findings were confirmed by culture, pathology, and a previously developed real-time PCR assay (14). To confirm the specificity and stability of the fliP recombination in B. mallei, the fliP genomic regions, including their recombination sites, were amplified and subsequently sequenced from strains Bogor, Zagreb, NCTC 3708, NCTC 120, and Dubai 7 with the primers Bma-IS407-flip-f (5'-TCAGGTTTGTATGTCGCTCGG-3') and Bma-flip-r (5'-CTAGGTGAAGCTCTGCGCGAG-3), generating fragments of 989 bp. PCRs were performed in 50 µL of ready-to-go master mix (Eppendorf GmbH) in a GeneAmp2400TM thermal cycler (Perkin-Elmer, Applied Biosystems) with 15 pmol of each primer. Cycling was as follows: 30 cycles of 30 s at 94 °C, 30 s at 66 °C, and 1 min at 72 °C, followed by final elongation for 7 min at 72 °C. The nucleotide sequences were determined by dideoxynucleotide sequencing of both strands with primers Bma-IS407-flip-f and Bma-flip-r, carried out in an ABI PRISMTM 3100 Genetic Analyzer (Applied Biosystems). Nucleotide sequences were further analyzed by the Chromas 1.45 program. Multiple sequence alignments of the fliP genomic region sequences were performed with ClustalW 1.8 (available at http://clustalw.genome.jp). The fliP genomic region of B. mallei strains SAVP1, GB8 horse 4, 10229, and NCTC 10247 were determined by the Sanger Institute and The Institute of Genomic Research and are available for Blast analysis via http://www.tigr.org/msc/mallei/mallei.shtml. The fliP genomic regions of B. mallei strains Bogor, Zagreb, Dubai 7, NCTC 120, and NCTC 3708 were deposited in the EMBL Nucleotide Sequence Database under accession nos. AM087433 to AM087437, respectively.
|
Because glanders has regained importance as a reemerging disease and a potential biological agent, the purpose of the present study was to develop a real-time PCR assay for the identification of B. mallei. The alignment of the available fliP sequences of B. pseudomallei K96243 and B. mallei ATCC 23344 revealed that B. pseudomallei encodes a functional fliP gene (762 bp), whereas in B. mallei, a recombination event at nucleotide position 235 of fliP was encountered. At this recombination site, a large DNA segment of 60 kbp was inserted within the fliP coding region with involvement of the insertion sequence IS407A (data not shown). The stability of this recombination in B. mallei was confirmed by sequencing and sequence comparison of the fliP region from 5 selected B. mallei strains, including the oldest strain of our collection (NCTC 120) and the strain of a recent outbreak (Dubai 7). In these analyses, the corresponding sequences of B. mallei strains that have been made available recently for sequence analysis (SAVP1, GB8 horse 4, 10229, and NCTC 10247) were included. In these analyses, all 9 sequences were identical (see accession nos. AM087433 to AM087437), which indicates that this specific recombination is highly conserved in B. mallei independent of the geographic origin and source of isolation. This recombination was absent from each B. pseudomallei tested. Because fliP is part of the flagellar complex, the disruption of its open reading frame might explain the amotility of B. mallei, in contrast to B. pseudomallei, which is motile. On the basis of this genetic difference, we designed oligonucleotides for the specific detection of B. mallei. The detection limit of 60 fg was markedly lower than that of conventional PCR (16) and in the same range as the detection limit of indirect real-time PCR assays (17).
We detected B. mallei-specific DNA in 2 horses from Dubai confirmed as having the infection by culture and pathology. Recently, a conventional multiplex PCR and a real-time PCR for the detection and differentiation of B. mallei from B. pseudomallei and Burkholderia thailandensis have been published (16)(17). In our hands, the real-time PCR described by Thibault et al. (17) misidentified Burkholderia sordicola and 2 B. thailandensis strains as B. pseudomallei. One Burkholderia caribensis, 1 Burkholderia phenazinium, and 1 B. thailandensis strain were detected in the supposedly B. mallei/B. pseudomallei-specific PCR targeting orf13. These findings can be explained by the gene targets, which were derived from the operon of the type III secretion apparatus that is widely distributed among gram-negative bacteria. The clinical usability of the conventional PCR described by Lee et al. (16) will have to be confirmed, but it has a high detection limit of 100 pg.
B. mallei is considered a potential biological agent because it can be dispersed as a highly infective aerosol causing incapacitating or fatal disease. The only known reservoirs are horses, making the effect of a deliberate release predictable and a continuing epidemic in humans unlikely (3). Real-time PCR assays could offer a dramatic decrease in turnaround time for results and have the potential for automation. We previously described 5'-nuclease real-time PCR assays for the simultaneous screening for B. mallei and B. pseudomallei (14). The combination of both assays allows species identification of B. pseudomallei by exclusion. To our knowledge this is the first report of a real-time PCR assay for the specific identification and detection of B. mallei in clinical samples.
Acknowledgments
We thank C. Lodri, G. Echle, and C. Kleinemeier for excellent technical assistance.
Footnotes
1 these authors contributed equally to this work; 
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
