SGM Journals
DIAGNOSTICS, TYPING AND IDENTIFICATION

Using real-time PCR to specifically detect Burkholderia mallei

Melanie P. Ulrich, David A. Norwood, Deanna R. Christensen, Ricky L. Ulrich

  • Diagnostic Systems Division1 and Bacteriology Division2, United States Army Medical Research Institute of Infectious Diseases, Fort Detrick, Frederick, MD 21702-5011, USA
  • Correspondence
    Ricky L. Ulrich
    Ricky.Ulrich{at}AMEDD.ARMY.MIL

    • Journal of Medical Microbiology 2006; 55(5):551–559 · https://doi.org/10.1099/jmm.0.46350-0

    View at publisher PubMed

    Abstract

    Burkholderia mallei is the causative agent of human and animal glanders and is a category B biothreat agent. Rapid diagnosis of B. mallei and immediate prophylactic treatment are essential for patient survival. The majority of current bacteriological and immunological techniques for identifying B. mallei from clinical samples are time-consuming, and cross-reactivity with closely related organisms (i.e. Burkholderia pseudomallei) is a problem. In this investigation, two B. mallei-specific real-time PCR assays targeting the B. mallei bimAma gene (Burkholderia intracellular motility A; BMAA0749), which encodes a protein involved in actin polymerization, were developed. The PCR primer and probe sets were tested for specificity against a collection of B. mallei and B. pseudomallei isolates obtained from numerous clinical and environmental (B. pseudomallei only) sources. The assays were also tested for cross-reactivity using template DNA from 14 closely related Burkholderia species. The relative limit of detection for the assays was found to be 1 pg or 424 genome equivalents. The authors also analysed the applicability of assays to detect B. mallei within infected BALB/c mouse tissues. Beginning 1 h post aerosol exposure, B. mallei was successfully identified within the lungs, and starting at 24 h post exposure, in the spleen and liver. Surprisingly, B. mallei was not detected in the blood of acutely infected animals. This investigation provides two real-time PCR assays for the rapid and specific identification of B. mallei.

    INTRODUCTION

    Burkholderia mallei, the aetiologic agent of glanders, is a Gram-negative, non-motile bacillus that is a highly adapted obligate parasite of equines and is unable to persist outside of its host (DeShazer & Waag, 2004). Until the early twentieth century and the development of motorized transportation, glanders was common throughout the world (DeShazer & Waag, 2004). With the development of the mallein test (a serological assay for identifying B. mallei-infected horses), there have not been any naturally occurring cases of human glanders reported in the USA since the 1930s. However, sporadic cases of glanders still occur in Asia, Africa, the Middle East, and South America. Human glanders is now uncommon, and individuals such as veterinarians, slaughterhouse workers and laboratory scientists, whose occupation exposes them to infection, are most susceptible to acquiring glanders. In horses, clinical symptoms of glanders include nasal discharge, possible ulceration of the nasal septum, enlarged lymph nodes, and characteristic ulcers, pustules and nodules in the lungs and on the skin (DeShazer & Waag, 2004).

    Acute human glanders is normally fatal without immediate antibiotic treatment. Infected individuals may present clinically with swollen lymph nodes, ulcerating nodules in the alimentary and respiratory tracts, and weight loss, as well as numerous subcutaneous abscesses (DeShazer & Waag, 2004; Srinivasan et al., 2001). The route of exposure for human glanders is normally via contact with infectious material (i.e. nasal discharge from infected horses) through skin abrasions or cuts, which usually results in a chronic form of disease. In contrast, and likely the route of exposure in the deliberate release of B. mallei, contraction from aerosols causes acute glanders, and if not treated immediately, death occurs in a matter of days (Howe & Miller, 1947; Hunter, 1936).

    With the potential of B. mallei weaponization, rapid methods for the identification of this infectious Burkholderia species are essential to ensure that adequate prophylactic therapies can be initiated following exposure. The conventional laboratory methods for detecting B. mallei rely on biochemical testing, colony morphology, and motility assays, all of which can take up to 7 days to obtain preliminary results. In addition, it has recently been demonstrated that the API 20NE and RapID NF commercial kits are 0–60 % successful in positively identifying B. mallei (Glass & Popovic, 2005).

    With the physiological and genetic relatedness between B. mallei, Burkholderia pseudomallei and Burkholderia thailandensis (Brett et al., 1997; Godoy et al., 2003; Holden et al., 2004; Nierman et al., 2004), PCR-based and biochemical techniques for identifying and differentiating these Burkholderia species are difficult. For B. pseudomallei, the 16S and 23S rRNA, 16S–23S intergenic region, fliC, heat-shock protein 70 (Hsp70), and a type three secretion (TTS) system have been targeted for the development of molecular-based assays (Antonov et al., 2005; Bauernfeind et al., 1998; Gee et al., 2003; Hagen et al., 2002; Lee et al., 2005; Sprague et al., 2002; Tanpiboonsak et al., 2004; Thibault et al., 2004; Tomaso et al., 2004, 2005; Tyler et al., 1995; U'Ren et al., 2005). One study required the use of two separate assays and a negative PCR result to discriminate B. mallei from B. pseudomallei. Several other investigations could differentiate B. pseudomallei from B. mallei, but these reports were based on a single base pair difference between the two species. The lack of a target region with substantial nucleotide variation between B. pseudomallei and B. mallei further demonstrates the genetic similarity between these Burkholderia species.

    In this investigation, we have designed a real-time PCR assay for the definitive identification of B. mallei and separation from B. pseudomallei. The assay was tested for specificity against a panel of diverse bacterial species, and for the ability to detect B. mallei within the lungs, liver and spleen of aerosol-challenged BALB/c mice.

    METHODS

    Bacterial culture conditions.

    All B. mallei isolates were cultured using Luria–Bertani (LB) broth containing 4 % (v/v) glycerol (Sigma) (LBG) at 37 °C with aeration (250 r.p.m.). The remaining Burkholderia isolates were cultured as described above using LB broth. A representative panel of eight B. mallei, eight B. pseudomallei and eight B. thailandensis strains, obtained from various geographical, clinical and environmental locations throughout the world, was investigated (Table 3⇓). Additional closely related Burkholderia species tested included Burkholderia cepacia, Burkholderia stabilis, Burkholderia multivorans and Burkholderia vietnamiensis. Other bacterial isolates tested for cross-reactivity are described below.

    B. mallei animal challenges and tissue collection.

    Female BALB/c mice, 6–8 weeks old, were obtained from Charles River Laboratories (National Cancer Institute, Frederick, MD) and challenged by the aerosol route with approximately 10 LD50 of B. mallei ATCC 23344. Briefly, whole-body exposures were performed on 20 mice by nebulizing a 10 ml overnight culture of B. mallei ATCC 23344 which delivers an inhaled dose of approximately 1000 c.f.u., or 10 LD50. At 1, 24 and 48 h following exposure, three mice for each time-point were euthanized (CO2 chamber) and lungs (1, 24 and 48 h), spleens and livers (24 and 48 h) were harvested. Individual organs were placed in scintillation tubes containing 1 ml sterile PBS and organs were disrupted using a Tissue-Tearor (BioSpec Products) by administering 20 s pulses (3×) on setting thirty or until the tissue was appropriately homogenized. Organ bacterial loads were determined by serially diluting each organ extract (100 μl tissue into 900 μl PBS), followed by cultivation on LBG plates at 37 °C for 48 h. For blood collection, three animals (same mice used for organ extractions) at each time-point were exsanguinated by intracardial puncture. All research was conducted in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals, and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals, National Research Council, 1996. The facility where this research was conducted is fully accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care International.

    Bacterial and tissue DNA extraction.

    Burkholderia species were cultured as described above, and genomic DNA from cultures as well as collected tissues was prepared using the Qiagen QIAamp DNA Mini kit with 180 μl of an overnight culture or homogenized tissue, with the following modifications to the manufacturer's protocol. The sample (180 μl) was mixed with 20 μl proteinase K (17·8 mg ml−1) and 200 μl buffer AL, and incubated at 55 °C for 1 h. Ethanol (96–100 %, 210 μl) was added and mixed, and the sample was loaded onto the column and the column washed using buffers AW1 and AW2. Columns were centrifuged at 16 000 g for 3 min to dry residual ethanol from the column. Buffer AE (100 μl), preheated to 70 °C, was added to each column, incubated at 70 °C for 5 min, and DNA was eluted.

    The DNA used in this study for cross-reactivity testing was from reference material obtained from recognized culture collections, commercial vendors, clinics, or unique entries from previous United States Army Medical Research Institute of Infectious Diseases (USAMRIID) collections. Pure cultures of each agent were grown on appropriate media at optimal temperatures and harvested for DNA isolation. The DNA was extracted using either Bactozol Kits (Molecular Research Center) according to the manufacturer's recommendations or Qiagen kits as previously described (Coyne et al., 2004). Extracted DNA was quantified and checked for purity with a spectrophotometer by measuring absorbance at 260, 280 and 320 nm. The ability of the DNA to be amplified was established with a universal primer set to the 16S rRNA gene (Bricker & Halling, 1994). The presence of high-molecular-mass DNA was confirmed by running 400 ng ml−1 of each genomic DNA on 0·7 % agarose gels. Identity was confirmed for each DNA by sequence analysis of the entire 16S rRNA gene and comparison with published data.

    Assay design and optimization.

    The real-time PCR assay primer and probe sequences are listed in Table 1⇓. The B. mallei bimAma (Burkholderia intracellular motility A) homologue encoded by gene locus BMAA0749 was used for assay development. It has recently been demonstrated that the N-terminal nucleotide sequence of bimAma contains a unique B. mallei region not present in the B. pseudomallei and B. thailandensis bimA genes (Stevens et al., 2005a). The specific primer and TaqMan minor groove binding (MGB) probe sequences were designed using Visual OMP version 4.1.0.0 for Windows (DNA Software). All primer and probe sequences were analysed using blastn to ensure specificity for their respective targets. All primers were purchased from Invitrogen. The probes were synthesized by Applied Biosystems, and contained 6-carboxyfluorescein (FAM) at the 5′ end and a non-fluorescent quencher with the MGB protein at the 3′ end. Primer candidates were used to amplify 10 pg target template in the presence of SYBR green dye. Melt curves and agarose gels were analysed to eliminate inefficient and/or dimer-producing primer pairs. MgCl2 and primer concentrations were optimized sequentially as follows: MgCl2 was optimized in 1 mM increments from 3 to 7 mM, and primers were optimized symmetrically in 0·1 μM increments from 0·1 to 1·0 μM. The combinations that exhibited the earliest crossing threshold (CT) value and generated the highest end-point fluorescence (EPF) value were chosen as the optimal conditions for each assay (Table 1⇓).

    Table 1.

    Primer and probe sequences for B. mallei-specific real-time PCR assay

    Amplicon sizes, sequences and optimized conditions are shown.

    The optimized assays were carried out in 20 μl volumes. Each assay contained 1× PCR buffer (50 mM Tris, pH 8·3, 250 μg BSA ml−1) (Idaho Technology, Inc.), 0·2 mM dNTP mix (Idaho Technology, Inc.) and 1·0 U Platinum Taq DNA polymerase (Invitrogen). Optimal concentrations of primers, probe and MgCl2 were added, and the master mix was distributed to reaction tubes. Five microlitres of control/template/sample DNA was added just before analysis on the instrument. Thermal cycling conditions were standardized for both assays, and consisted of 1 cycle at 95 °C for 2 min, followed by 45 cycles of 95 °C for 5 s and 60 °C for 20 s. Fluorescence readings were taken at the end of each 60 °C step. For the LightCycler 2.0 instrument (Roche), each reaction capillary tube was read in channel 1 (F1) with data analysed by the LightCycler Data Analysis (LCDA) software version 4.0.

    Assay linearity and estimated detection limit.

    The linearity of the assay was assessed by standard curve analysis in order to determine the amplification efficacy, efficiency, and quantitative utility. Additionally, serial dilutions were used to estimate the limit of detection. Serial dilutions in Tris/EDTA buffer (Eppendorf) were prepared at the following per-reaction concentrations: 1 ng, 500 pg, 100 pg, 50 pg, 10 pg, 5 pg, 1 pg, 500 fg, 100 fg. Triplicate assays were performed at each dilution, the standard curve was generated by the LightCycler 4.0 software, and slopes were used to calculate amplification efficiency and standard error. An estimate of the detection limit was determined by the lowest concentration of template per reaction that produced three positive results.

    Inclusivity and exclusivity experiments.

    A general panel of 60 organisms was analysed using the optimized conditions to establish inclusivity and exclusivity (Table 2⇓). In addition, a panel of 30 genetic near neighbours was also analysed (Table 3⇓). Panels were constructed to include: other threat organisms; nearest genetic neighbours to threat organisms; organisms sharing a clinical niche with B. mallei, especially respiratory pathogens, opportunists, and normal respiratory flora; organisms observed repeatedly in clinical and environmental samples. In all cases, 100 pg genomic DNA per reaction was used to determine if the assays cross-reacted with nucleic acid from other organisms. Inclusivity and exclusivity testing was performed on the LightCycler 2.0 instrument and utilized the automated calling capability of LightCycler software version 4.0.

    Table 2.

    DNAs that encompass the general cross-reactivity panel analysed in this investigation

    All assays were performed with 100 pg template DNA per reaction from each organism to ensure specificity.na, Not applicable.

    Table 3.

    Near-neighbour analysis for the two B. mallei-specific real-time PCR assays

    CT values are provided for those strains that produced positive results for each assay. un, Unknown information; Neg, negative assay result.

    RESULTS

    Identification of a B. mallei unique genomic DNA sequence

    Stevens et al. (2005b) have recently identified a putative protein (BimA) synthesized by B. pseudomallei that is involved in host-cell actin polymerization. In silico analysis of the B. mallei and B. thailandensis genomes indicates that these Burkholderia species also encode a BimA orthologue. Interestingly, considering that the B. mallei and B. thailandensis (a closely related avirulent Burkholderia species) BimA proteins are capable of complementing a B. pseudomallei bimA mutant, significant nucleotide sequence deviations have been identified within the N-terminal coding regions of this allele (Stevens et al., 2005a). Using this unique DNA sequence within the 5′ region of the B. mallei bimAma, two real-time PCR primer and probe sets (Table 1⇑) were designed and tested for the capability to detect B. mallei within infected lungs, spleens and livers of challenged BALB/c mice. Cross-reactivity to eight B. pseudomallei isolates and eight B. thailandensis strains obtained from various geographical, clinical and environmental sources over a 70 year period was tested.

    Assay optimization

    Primers were designed from a portion of the 5′ region of the B. mallei bimAma sequence and initially tested by rapid PCR amplification in the presence of SYBR Green. All primer combinations resulting in PCR products smaller than 160 bp were tested for amplification efficiency. Initial testing was performed with five forward and five reverse primers in a matrix that produced 13 combinations. All combinations amplified B. mallei DNA from strain ATCC 23344. Two combinations, F49 with R144 and F52 with R146, generated the largest amount of amplified DNA. In addition, no template controls with these two primer sets produced minimal to no primer dimers. After primers were tested and selected using standard PCR, five compatible probes were tested for performance with both of these primer sets. For both primer sets, probe P104R consistently produced the earliest CT and the highest EPF values. Assays were further optimized for primer and MgCl2 concentrations to produce maximum sensitivity, earliest CT and highest EPF. The final primer and probe pairs and optimized assay conditions are listed in Table 1⇑.

    Sensitivity/efficiency

    Standard curve analysis was utilized to assess the amplification efficiency and approximate detection limit of the assays. Standard curves were designed to contain triplicate samples at a minimum of six detectable concentrations of DNA. The assays had amplification efficiencies of 1·97 (F49/R144) and 1·92 (F52/R146) with errors of 0·018 and 0·057, respectively. An amplification efficiency of 2·00 is considered ideal, and corresponds to a doubling of copy number for every PCR cycle. Both assays detected all three replicates at 1 pg DNA template per reaction, which corresponds to 424 genome copies, based on a genome size of 2·3 Mbp (this corresponds to B. mallei ATCC 23344 chromosome 2).

    Inclusivity/exclusivity

    Using standard PCR (Epicentre FailSafe kit with buffer J, Epicentre Technologies), with primers targeting the unique 5′ DNA sequence of the B. mallei bimAma (AT5, 5′-TTCGATCGATTCCTGCTATC-3′, and AT4, 5′-GCGTTAAACGCCGTACTTTC-3′), we have previously tested a collection of 29 B. mallei and 34 B. pseudomallei isolates to determine if this unique B. mallei nucleotide sequence is conserved (Ulrich et al., 2006). Surprisingly, given the extensive genetic similarity between B. mallei and B. pseudomallei (Godoy et al., 2003), only B. mallei genomic DNA produced a detectable amplicon with PCR primers (listed above) specifically designed for bmaAma. Likewise, both real-time PCR assays (Table 1⇑) were analysed for their ability to specifically detect strains of B. mallei in the absence of non-specific detection of both genetic near neighbours as well as a broader collection of micro-organisms. The near-neighbour analysis included a panel of 30 DNAs, including eight strains of B. thailandensis, eight strains of B. mallei, eight strains of B. pseudomallei and six strains of other Burkholderia species, including B. cepacia, B. multivorans, B. stabilis and B. vietnamiensis. As observed with primer pairs AT4 and AT5 (standard PCR primers targeting bimAma), both assays detected only B. mallei, and consistent CT values were observed across all strains and assays tested (Table 3⇑). Both assays were negative for other genetic near neighbours, as well as for a panel of 60 micro-organisms from a general microbial DNA panel (Tables 2 and 3⇑⇑).

    In vivo analysis

    To determine if our real-time PCR assays were capable of detecting B. mallei within infected tissues and blood, we challenged (10 LD50 or 1000 c.f.u.) 20 female BALB/c mice via the aerosol route and followed organ colonization for 48 h post exposure (p.e.). At 1 h p.e., our real-time PCR primer and probe sets definitively identified B. mallei within the lungs of mice, indicated by mean CT values of 30·06 for primer pair 49F/144R and 30·15 for set 52F/146R (Table 4⇓). At 24 h p.e., dissemination of B. mallei to the liver was detected in two out of three animals with primer pairs 49F/144R (mean CT of 34·56) and 52F/146R (mean CT of 35·19) (Table 4⇓). Within the spleen, primers 52F/146R detected B. mallei in two out of three animals (mean cycle value of 35·57), whereas primers 49F/144R only detected B. mallei in one out of the three animals (CT of 33·42) at 24 h p.e. (Table 4⇓). At 48 h post challenge, all organ extracts (lungs, livers and spleens) tested positive for B. mallei (Table 4⇓). The mean CT values for the triplicate lungs were 23·50 for primer pair 49F/144R and 23·59 for primer pair 52F/146R. For extracted livers, the mean CT values were 28·50 and 28·82 for 49F/144R and 52F/146R, respectively, and for spleen tissue, 31·56 with pair 49F/144R and 31·60 with set 52F/146R. It should be noted that BALB/c mice challenged by the aerosol route with B. mallei develop acute glanders and normally succumb to infection 48–72 h post challenge; therefore, additional time-points were not analysed. All control organ extracts, from non-infected animals housed within the same biosafety level 3 containment laboratory, were negative for the presence of B. mallei by analysis with the real-time PCR primers and probes described in this study (Table 4⇓). Surprisingly, and likely a result of the low bacterial burden in the blood, our real-time PCR assays were not capable of detecting B. mallei in the blood of infected mice (Table 4⇓). The presence of PCR inhibitors in all sample eluates was tested by using an inhibition assay previously described (Hartman et al., 2005). The extracted DNA material from whole blood required a dilution of 1 : 8 to overcome the effects of PCR inhibitors present in the sample (Table 4⇓).

    Table 4.

    Analysis of B. mallei-specific real-time PCRassays on infected mouse tissues

    Neg, negative assay result.

    DISCUSSION

    Rapid identification of B. mallei in the mammalian host is essential for the management of disease progression and prevention of acute glanders. Standard laboratory methods for the diagnosis of B. mallei from clinical samples focus on biochemical and phenotypical testing that can take in excess of 7 days (DeShazer & Waag, 2004), which is inadequate for potential acute cases of glanders (patients normally die 2 days p.e.). With the potential weaponization and deliberate release of B. mallei into the public domain, rapid molecular methods for the identification of this highly infectious pathogen are crucial.

    As previously mentioned, the current laboratory methods for identifying B. mallei in clinical samples rely on biochemical and phenotypical testing. However, it has been shown that commercially available biochemical testing kits (API 20NE and RapID NF) cross-react with other non-virulent bacterial species, leading to false-positive reactions (Glass & Popovic, 2005; Inglis et al., 1998). An ELISA has been developed using irradiation-killed B. mallei whole cells; however, cross-reactivity with irradiation-killed B. pseudomallei was observed, suggesting that this immunological assay is not B. mallei specific (unpublished results).

    The development of PCR-based methodologies for the identification and differentiation of B. mallei from B. pseudomallei and B. thailandensis has been troublesome due to the level of genetic similarity between these Burkholderia species (Brett et al., 1998; Godoy et al., 2003; Holden et al., 2004; Nierman et al., 2004). Various chromosomal markers have been used to detect B. pseudomallei, and include the 16S and 23S rRNA, 16S–23S intergenic region, the filament-forming flagellin fliC gene, heat-shock protein Hsp70, and a putative type three secretion operon (Antonov et al., 2005; Bauernfeind et al., 1998; Gee et al., 2003; Hagen et al., 2002; Lee et al., 2005; Sprague et al., 2002; Thibault et al., 2004; Tomaso et al., 2004, 2005; Tyler et al., 1995; U'Ren et al., 2005). Several of the assays that have been described are targeted toward single-nucleotide differences in the 23S rDNA region, the 16S region, and a putative antibiotic resistance gene. However, targeting a single base pair could be problematic for the identification of B. mallei, as it has been shown that the B. mallei genome is unstable upon passage. B. mallei ATCC 23344 has been passaged in culture, in a mouse and in a horse, and two isolates have been obtained from the accidental infection of a scientist (Srinivasan et al., 2001). The genomes of these isolates were sequenced and compared to the published B. mallei sequence. Single-nucleotide polymorphisms were noted after passage, mainly located within coding regions of the genome. The two human isolates obtained from the accidental infection each possessed a distinct set of sequence variations as well as two sequence variations in common (unpublished data). These data suggest that the B. mallei population in a host is not identical and that accumulated genome-sequence variation may occur.

    The B. mallei-specific real-time PCR assay described in this study was designed using a unique 5′ nucleotide region encoded within bimAma (BMAA0749) (Stevens et al., 2005a). In B. pseudomallei, BimA has been shown to be involved in host-cell actin polymerization (Stevens et al., 2005b). Surprisingly, the unique DNA sequence encoding the N-terminal region of BimAma, the portion that is exposed at the bacterial cell surface, is conserved among numerous B. mallei isolates (29 strains obtained from various clinical sources over a 70 year period; Ulrich et al., 2006), suggesting that this sequence divergence occurred early in the evolution of B. mallei from B. pseudomallei. These findings also suggest that this chromosomally encoded gene is genetically stable and will provide an ideal target for the rapid identification of B. mallei. Despite these deviations in nucleotide sequence, heterologous expression of the B. mallei bimAma in a B. pseudomallei bimAps mutant fully complements the defective actin motility phenotype observed in this strain, suggesting that bimAma encodes a functional protein (Stevens et al., 2005a).

    The specificity of our B. mallei real-time PCR assays was tested against an extensive cross-reactivity panel comprising bacterial (Gram-positive and -negative species), viral, fungal and human DNAs (Table 2⇑). In addition to the cross-reactivity panel, a set of 30 closely related Burkholderia species was tested, including B. mallei, B. pseudomallei, B. thailandensis, B. stabilis, B. multivorans, B. cepacia and B. vietnamiensis (Table 3⇑). With our real-time PCR primer and probe sets, B. mallei was definitively identified and differentiated from B. pseudomallei and other closely related species (Table 3⇑).

    To further test the applicability of our B. mallei-specific real-time PCR assay, serial animal sacrifices were performed on BALB/c mice infected by the aerosol route with B. mallei ATCC 23344. As anticipated, B. mallei was positively detected in organ extracts processed at 1 h (lungs), 24 h (lungs, spleen and liver) and 48 h p.e. (lungs, spleen and liver) (Table 4⇑). The data produced in this study indicate the proliferation of bacterial cells in the organs of infected mice based on the decreasing CT values for each of the time-points tested. These findings suggest that the real-time PCR primer and probe sets described in this work will provide a rapid method for the identification of B. mallei in clinical samples. Unlike previously described techniques for detecting B. mallei in human clinical samples (i.e. culturing abscess biopsies, GLC of cellular fatty acids, and 16S rRNA gene-sequence analysis), real-time PCR provides an alternative method for identifying B. mallei and eliminates the need for bacterial pre-enrichment (i.e. total DNA can be isolated from organ abscesses and real-time PCR performed). Unfortunately, the sensitivity (see below) of our B. mallei-specific real-time PCR assays was not sufficient to detect B. mallei in the blood of infected animals. These findings are likely the result of either the low bacterial load found within the blood of acutely infected BALB/c mice or the presence of PCR inhibitors (Fritz et al., 2000). Even when incorporating spiked BALB/c control blood with B. mallei at concentrations of 1000, 500 and 100 c.f.u. ml−1, we failed to identify this pathogen using the primer and probe sets described in this work (data not shown). However, we were able to detect B. mallei in spiked human blood (human use protocol FY04-14) at a concentration of 500 c.f.u. ml−1 with primer set 49F/144R (three out of three replicates) (data not shown). These findings suggest that, despite using a commonly employed commercial DNA extraction kit (Qiagen QIAamp DNA Mini kit), mouse blood contains PCR inhibitors not present in human blood. Further studies to address these extraction limitations will be needed. Similar reductions in the ability to detect B. mallei and B. pseudomallei in spiked blood have been reported by Tomaso et al. (2005), who targeted the B. pseudomallei and B. mallei 16S rDNA, fliC and rpsU genes.

    For serological detection of B. mallei, the Russian scientists Gelman and Kalning developed the mallein assay in the late 1800s (DeShazer & Waag, 2004). This immunological test utilized antigens prepared from 4–8 month culture filtrates of B. mallei and was utilized by the USA and Canada to identify horses infected with B. mallei (Steele, 1979). Despite the rapid (48 h) identification of B. mallei using the mallein test in horses, serological conversion does not occur in humans for 3–4 weeks p.e., which suggests that this assay is ineffective for detecting human glanders. It should be noted that this serological test also cross-reacts with B. pseudomallei (DeShazer & Waag, 2004). To date, the only B. mallei-specific detection assay reported in the literature incorporates the bacterial phage phiE125, which has been shown to be specific for B. mallei capable of biosynthesizing LPS (Woods et al., 2002).

    The findings of this investigation provide two real-time PCR assays for the specific identification of B. mallei and its differentiation from B. pseudomallei and closely related Burkholderia species. The described assays did not cross-react with any of the bacterial species included in our culture collection, and were capable of detecting B. mallei in the lungs, spleen and liver of infected mice beginning at 1 h p.e. This B. mallei-specific real-time PCR assay will undoubtedly complement other methodologies used in the clinical laboratory for the rapid identification of this obligate mammalian pathogen.

    Acknowledgments

    We would like to thank David DeShazer for providing the nucleotide sequence used in designing this assay. We would like to thank Katheryn Kenyon and David Heath for reviewing the manuscript. The research described herein was sponsored by National Institute of Allergy and Infectious Diseases (NIAID) Interagency Agreement Y1-AI-5004-01 and JSRO/DTRA plan no. G0015_04_RD_B of DTO CB.56. Opinions, interpretations, conclusions and recommendations are those of the authors and are not necessarily endorsed by the US Army.

    References

    1. Antonov, V. A. & Iliukhin, V. I. (2005). Molecular-genetic approaches to diagnosis and intraspecific typing of causative agents of glanders and melioidosis. Mol Gen Mikrobiol Virusol 2, 3–9 (in Russian).
    2. Bauernfeind, A., Roller, C., Meyer, D., Jungwirth, R. & Schneider, I. (1998). Molecular procedure for rapid detection of Burkholderia mallei and Burkholderia pseudomallei. J Clin Microbiol 36, 2737–2741.
    3. Brett, P. J., DeShazer, D. & Woods, D. E. (1997). Characterization of Burkholderia pseudomallei and Burkholderia pseudomallei-like strains. Epidemiol Infect 118, 137–148.
    4. Brett, P. J., DeShazer, D. & Woods, D. E. (1998). Burkholderia thailandensis sp. nov., a Burkholderia pseudomallei-like species. Int J Syst Bacteriol 48, 317–320.
    5. Bricker, B. J. & Halling, S. M. (1994). Differentiation of Brucella abortus bv. 1, 2, and 4, Brucella melitensis, Brucella ovis, and Brucella suis bv. 1 by PCR. J Clin Microbiol 32, 2660–2666.
    6. Coyne, S. R., Craw, P. D., Norwood, D. A. & Ulrich, M. P. (2004). Comparative analysis of the Schleicher and Schuell IsoCode Stix DNA isolation device and the Qiagen QIAamp DNA Mini Kit. J Clin Microbiol 42, 4859–4862.
    7. DeShazer, D. & Waag, D. (2004). Glanders: new insights into an old disease. In Biological Weapons Defense: Infectious Diseases and Counterbioterrorism, pp. 209–237. Edited by L. Lindler, F. Lebeda & G. W. Korch. Totowa, NJ: Humana Press.
    8. Fritz, D. L., Vogel, P., Brown, D. R., Deshazer, D. & Waag, D. M. (2000). Mouse model of sublethal and lethal intraperitoneal glanders (Burkholderia mallei). Vet Pathol 37, 626–636.
    9. Gee, J. E., Sacchi, C. T., Glass, M. B., De, B. K., Weyant, R. S., Levett, P. N., Whitney, A. M., Hoffmaster, A. R. & Popovic, T. (2003). Use of 16S rRNA gene sequencing for rapid identification and differentiation of Burkholderia pseudomallei and B. mallei. J Clin Microbiol 41, 4647–4654.
    10. Glass, M. B. & Popovic, T. (2005). Preliminary evaluation of the API 20NE and RapID NF plus systems for rapid identification of Burkholderia pseudomallei and B. mallei. J Clin Microbiol 43, 479–483.
    11. Godoy, D., Randle, G., Simpson, A. J., Aanensen, D. M., Pitt, T. L., Kinoshita, R. & Spratt, B. G. (2003). Multilocus sequence typing and evolutionary relationships among the causative agents of melioidosis and glanders, Burkholderia pseudomallei and Burkholderia mallei. J Clin Microbiol 41, 2068–2079.
    12. Hagen, R. M., Gauthier, Y. P., Sprague, L. D., Vidal, D. R., Zysk, G., Finke, E. J. & Neubauer, H. (2002). Strategies for PCR based detection of Burkholderia pseudomallei DNA in paraffin wax embedded tissues. Mol Pathol 55, 398–400.
    13. Hartman, L. J., Coyne, S. R. & Norwood, D. A. (2005). Development of a novel internal positive control for Taqman based assays. Mol Cell Probes 19, 51–59.
    14. Holden, M. T., Titball, R. W., Peacock, S. J. & 45 other authors (2004). Genomic plasticity of the causative agent of melioidosis, Burkholderia pseudomallei. Proc Natl Acad Sci U S A 101, 14240–14245.
    15. Howe, C. & Miller, R. (1947). Human glanders: report of six cases. Ann Intern Med 26, 93–115.
    16. Hunter, D. (1936). Saints and martyrs. Lancet 231, 1131–1134.
    17. Inglis, T. J., Chiang, D., Lee, G. S. & Chor-Kiang, L. (1998). Potential misidentification of Burkholderia pseudomallei by API 20NE. Pathology 30, 62–64.
    18. Lee, M. A., Wang, D. & Yap, E. H. (2005). Detection and differentiation of Burkholderia pseudomallei, Burkholderia mallei and Burkholderia thailandensis by multiplex PCR. FEMS Immunol Med Microbiol 43, 413–417.
    19. Nierman, W. C., DeShazer, D., Kim, H. S. & 30 other authors (2004). Structural flexibility in the Burkholderia mallei genome. Proc Natl Acad Sci U S A 101, 14246–14251.
    20. Sprague, L. D., Zysk, G., Hagen, R. M., Meyer, H., Ellis, J., Anuntagool, N., Gauthier, Y. & Neubauer, H. (2002). A possible pitfall in the identification of Burkholderia mallei using molecular identification systems based on the sequence of the flagellin fliC gene. FEMS Immunol Med Microbiol 34, 231–236.
    21. Srinivasan, A., Kraus, C. N., DeShazer, D., Becker, P. M., Dick, J. D., Spacek, L., Bartlett, J. G., Byrne, W. R. & Thomas, D. L. (2001). Glanders in a military research microbiologist. N Engl J Med 345, 256–258.
    22. Steele, J. H. (1979). Glanders. In CRC Handbook Series in Zoonoses, pp. 339–362. Boca Raton, FL: CRL Press.
    23. Stevens, J., Ulrich, R., Taylor, L., Wood, M., DeShazer, D., Stevens, M. & Galyov, E. (2005a). Actin-binding proteins from Burkholderia mallei and B. thailandensis can functionally compensate for the actin-based motility defect of a B. pseudomallei bimA mutant. J Bacteriol, (in press).
    24. Stevens, M. P., Stevens, J. M., Jeng, R. L., Taylor, L. A., Wood, M. W., Hawes, P., Monaghan, P., Welch, M. D. & Galyov, E. E. (2005b). Identification of a bacterial factor required for actin-based motility of Burkholderia pseudomallei. Mol Microbiol 56, 40–53.
    25. Tanpiboonsak, S., Paemanee, A., Bunyarataphan, S. & Tungpradabkul, S. (2004). PCR-RFLP based differentiation of Burkholderia mallei and Burkholderia pseudomallei. Mol Cell Probes 18, 97–101.
    26. Thibault, F. M., Valade, E. & Vidal, D. R. (2004). Identification and discrimination of Burkholderia pseudomallei, B. mallei, and B. thailandensis by real-time PCR targeting type III secretion system genes. J Clin Microbiol 42, 5871–5874.
    27. Tomaso, H., Scholz, H. C., Al Dahouk, S., Pitt, T. L., Treu, T. M. & Neubauer, H. (2004). Development of 5′ nuclease real-time PCR assays for the rapid identification of the Burkholderia mallei//Burkholderia pseudomallei complex. Diagn Mol Pathol 13, 247–253.
    28. Tomaso, H., Pitt, T. L., Landt, O., Al Dahouk, S., Scholz, H. C., Reisinger, E. C., Sprague, L. D., Rathmann, I. & Neubauer, H. (2005). Rapid presumptive identification of Burkholderia pseudomallei with real-time PCR assays using fluorescent hybridization probes. Mol Cell Probes 19, 9–20.
    29. Tyler, S. D., Strathdee, C. A., Rozee, K. R. & Johnson, W. M. (1995). Oligonucleotide primers designed to differentiate pathogenic pseudomonads on the basis of the sequencing of genes coding for 16S–23S rRNA internal transcribed spacers. Clin Diagn Lab Immunol 2, 448–453.
    30. Ulrich, R. L., Ulrich, M. P., Schell, M. A., Kim, S. & DeShazer, D. (2006). Development of a PCR assay for the specific identification of Burkholderia mallei and differentiation from Burkholderia pseudomallei and other closely related Burkholderiaceae. Diagn Microbiol Infect Dis (in press).
    31. U'Ren, J. M., Van Ert, M. N., Schupp, J. M., Easterday, W. R., Simonson, T. S., Okinaka, R. T., Pearson, T. & Keim, P. (2005). Use of a real-time PCR TaqMan assay for rapid identification and differentiation of Burkholderia pseudomallei and Burkholderia mallei. J Clin Microbiol 43, 5771–5774.
    32. Woods, D. E., Jeddeloh, J. A., Fritz, D. L. & DeShazer, D. (2002). Burkholderia thailandensis E125 harbors a temperate bacteriophage specific for Burkholderia mallei. J Bacteriol 184, 4003–4017.