|Home | About | Journals | Submit | Contact Us | Français|
A real-time PCR assay based on the 16S rRNA gene sequence was designed for differentiation of blackleg-causing Clostridium chauvoei and Clostridium septicum, a phylogenetically closely related bacterium responsible for malignant edema. In order to exclude false-negative results, an internal amplification control was included in the assay. A set of three probes, one specific for C. chauvoei, one specific for C. septicum, and one specific for both species, permitted unequivocal detection of C. chauvoei in tests of 32 Clostridium sp. strains and 10 non-Clostridium strains. The assay proved to be sensitive, detecting one genome of C. chauvoei or C. septicum per PCR and 1.79 × 103 C. chauvoei cells/g artificially contaminated muscle tissue. In tests of 11 clinical specimens, the real-time PCR assay yielded the same results as an established conventional PCR method.
Clostridium chauvoei is a strictly anaerobic, Gram-positive, spore-forming rod and a common microbe found in soil as well as the guts of ruminants. It is the causal agent for blackleg, a peracute, nontraumatic, endogenous infection in cattle, presenting as gas gangrene. After blunt trauma, which reduces oxygen levels in muscle tissue, spores are able to germinate. In sheep, C. chauvoei is expressed mainly as an exogenous wound infection resulting from shearing, castration, or docking (3, 12, 18). Clostridium septicum belongs to the pathogens involved in the gas edema complex and is very similar to C. chauvoei.
One case of human fulminant gas gangrene caused by C. chauvoei has been reported. The actual prevalence is suspected to be higher if genetic characterization is included in the diagnostic procedure. Routine clinical microbiology laboratory tests may falsely identify C. septicum instead of C. chauvoei (12). Differentiation of the two pathogens is also needed to obtain governmental financial support for blackleg in certain countries or districts (8).
Conventional microbiological methods are hampered by the slow growth of C. chauvoei, which is often overgrown by other anaerobes or swarming C. septicum colonies, as both might be present in a sample (8, 15). Immunological detection methods are reported to be prone to cross-reactions between the two species (6). In addition, production of antisera for their detection is complex because it involves the use of laboratory animals.
DNA-based PCR identification proved to be a good alternative, as it is more reliable, fast, and easy to perform. Conventional PCR systems target the 16S rRNA gene, the 16S-23S rRNA gene spacer region, or the flagellin gene (1, 6, 8, 13, 14, 15, 16, 18). On one hand, conventional PCR bears the risk of yielding false-positive results due to contamination during post-PCR handling of the amplicons for detection. On the other hand, published assays do not include internal amplification controls (IACs) to permit detection of false-negative results due to PCR-inhibitory components in the sample matrix.
Real-time PCR employs fluorescently labeled probes for amplicon detection. Thus, no post-PCR processing of the samples is required. This reduces the risk of laboratory contamination and also saves time (4). Multichannel detection permits parallel analysis of more than one target region by use of differently labeled probes. Thus, an IAC may easily be integrated into the procedure.
The aim of the present study was to develop a real-time PCR assay for discrimination of C. chauvoei and C. septicum. An IAC was included in the assay. Inclusivity and exclusivity, as well as the detection limit, were tested, and the optimized assay was applied to clinical specimens.
The strains used in this study are listed in Table Table1.1. Clostridial strains were grown anaerobically on semifluid reinforced clostridial medium (RCM; Oxoid, Basingstoke, United Kingdom) at 37°C for 48 h. The purity of the strains was examined on Columbia agar with sheep blood (Oxoid, Basingstoke, United Kingdom). The strains were stored at −80°C in a mixture containing 500 μl of 60% glycerol and 1,500 μl of culture grown in RCM. Strains for exclusivity testing were grown overnight in tryptone soy broth with 6% yeast (TSB-Y; Oxoid, Hampshire, United Kingdom) at 37°C.
Clinical specimens were obtained from the Austrian Agency for Health and Food Safety, Vienna, Austria, and were previously tested for the presence of C. chauvoei and C. septicum by the fluorescent-antibody technique (FAT), using an anti-Clostridium chauvoei polyclonal antiserum conjugated to fluorescein isothiocyanate (FITC) and an anti-Clostridium septicum polyclonal antiserum conjugated to FITC (VMRD, Inc., Pullman, WA), respectively.
For inclusivity and exclusivity testing, as well as preparation of the genomic DNA standard for real-time PCR quantification, DNAs from pure cultures were isolated from 1 ml of enrichment broth (RCM or TSB-Y) by use of a NucleoSpin tissue kit and the support protocol for bacteria (Macherey-Nagel, Düren, Germany) according to the manufacturer's instructions. The DNA concentration was measured fluorometrically with a HoeferDyNA Quant 200 apparatus (Pharmacia Biotech, San Francisco, CA). For inclusivity/exclusivity testing, the DNA was diluted in double-distilled water (ddH2O) to 0.1 ng/μl.
DNAs from clinical specimens stored at −20°C were extracted directly from 25 mg of muscle or 10 mg of spleen. For each sample, two to four grain-sized pieces were dissected from areas which showed typical signs of necrosis and hemorrhage. DNA was extracted using a QIAamp DNA Mini kit (Qiagen, Hilden, Germany) and the support protocol for Gram-positive bacteria. Tissue was digested with ALT buffer (Qiagen) and proteinase K (Qiagen) for at least 3 h. After complete lysis, the bacteria were pelleted by centrifugation for 10 min at 5,000 × g, and DNA isolation was performed according to the support protocol for Gram-positive bacteria. Extracted DNA was subjected to real-time PCR analysis in duplicate.
To assess the detection limit of the method for clinical material, 25 mg of minced meat was spiked with 50 μl of a serial dilution of a pure culture containing 107 to 100 cells of C. chauvoei/ml. The cell number was determined using a Live/Dead BacLight bacterial viability kit (Invitrogen, San Diego, CA). Minced meat and bacterial suspension were blended with a pipette tip, and vigorous vortexing was performed. DNA was isolated as described above, using a QIAamp DNA Mini kit. These experiments were performed in duplicate. As a negative control, 25 mg of uninoculated minced meat was subjected to the same procedure. Real-time PCR analysis was performed in duplicate on the extracted DNA.
Primer and probe sequences for real-time PCR were analyzed for melting temperature by use of RaW-probe, version 0.15β, freeware (MRC-Holland; http://www.mrc-holland.com/WebForms/WebFormMain.aspx?Tag) and for the formation of secondary structures by use of the mfold web server (20) (Fig. (Fig.1).1). Sequences of the 16S rRNA genes of C. chauvoei and its closest phylogenetic relatives (7) were aligned with the MultAlign web server (2). Based on the alignment, a primer pair was designed for the detection of both C. chauvoei and C. septicum, producing a 143-bp fragment. A set of three differently labeled probes was designed in order to achieve real-time PCR detection and quantification of the amplicon and included a C. chauvoei-specific probe, a C. septicum-specific probe, and a probe specific for both C. chauvoei and C. septicum. To enhance the impact of the mismatch discriminating C. septicum from C. chauvoei, locked nucleic acid (LNA) modifications were incorporated into the C. septicum-specific probe. Primers and conventional TaqMan probes were purchased at MWG Biotech (Ebersberg, Germany), and the probe incorporating the LNA modifications was ordered from Sigma-Aldrich (St. Louis, MO).
A 122-bp region of the 12S rRNA gene IO3 of Boa constrictor imperator served as a noncompetitive IAC target and was artificially constructed (MWG Biotech). Primers producing a 122-bp amplicon and a TaqMan probe were designed using Primer Express software v2.0 (Applied Biosystems Inc., Foster City, CA).
The amplification reactions were carried out in a total volume of 25 μl containing 20 mM Tris-HCl, 50 mM KCl, 3.5 mM MgCl2, 300 nM Clostridium-specific primers, 150 nM IAC-specific primers, a 200 nM concentration of each probe, a 200 μM concentration (each) of dATP, dTTP, dGTP, and dCTP, 1.5 U of Platinum Taq DNA polymerase (Invitrogen, Lofer, Austria), and 5 μl of template DNA. Amplification was performed in an Mx3000P real-time PCR thermocycler (Stratagene, La Jolla, CA) following initial denaturation at 94°C for 2 min, with 45 cycles of 94°C for 20 s and 62°C for 1 min.
For determination of the detection limit of the real-time PCR assay, an average of 1.18 × 106 genomes/ng DNA was calculated. This calculation was based on the genome sizes of Clostridium species listed on the website of the J. Craig Venter Institute, San Diego, CA (http://cmr.jcvi.org/). In addition, fluorometrically determined quantities of genomic DNA were adjusted for the average GC content of clostridia (29.8%), which was calculated from data posted at the same website.
Due to the close phylogenetic relationship of C. chauvoei and C. septicum and the similar clinical signs caused by the two bacteria, a real-time PCR assay that permits detection of both microorganisms was designed. The 16S rRNA gene was selected for this purpose. The selected primers yielded a 143-bp amplicon with an internal region of genetic heterogeneity in both bacteria (Fig. (Fig.1).1). Thus, it was possible to design different probes, each specific for C. chauvoei or C. septicum. To enhance the diagnostic reliability of the assay, a third probe was added, targeting a homogenous region in both species. The final setup included an IAC to identify false-negative results.
Primer, probe, and MgCl2 concentrations, as well as the combined annealing/extension temperature, were optimized to yield specific signals for each individual pathogen while still maintaining the efficiency of the PCR. The assay performed best when a combined annealing/extension temperature of 62°C was used. At lower temperatures, nonspecific signals were achieved for DNA extracted from clinical material. The PCR efficiency calculated from the slope of the DNA standard curve was 99% for C. chauvoei and 98% for C. septicum (Fig. (Fig.22 and and3).3). The optimal copy number of the IAC target was 50 copies/PCR, as this quantity of IAC permitted amplification of the IAC target without interfering with the main reaction.
Inclusivity testing for the C. chauvoei-specific probe was performed on 20 C. chauvoei strains and yielded positive results for all strains. Threshold cycle (CT) values at a fixed threshold of 1,000 fluorescence units ranged from 17.4 to 24.55 (mean, 19.7; standard deviation, 1.8) for the C. chauvoei-specific probe and from 16.34 to 23.89 (mean, 18.9; standard deviation, 1.9) for the C. chauvoei/C. septicum-specific probe.
Thirteen C. septicum strains were subjected to real-time PCR for inclusivity testing of the C. septicum-specific probe. All strains yielded positive results, and CT values at a fixed threshold of 1,000 fluorescence units ranged from 18.23 to 23.68 (mean, 20.9; standard deviation, 1.8) for the C. septicum-specific probe and from 17.20 to 25.12 (mean, 20.4; standard deviation, 2.4) for the C. chauvoei/C. septicum-specific probe.
No C. chauvoei strain tested positive with the C. septicum probe and vice versa. Further exclusivity testing was performed on 25 strains, including 14 Clostridium sp. strains and 9 non-Clostridium species (Table (Table1).1). Within the tested clostridia species, 6 strains, including C. quinii, C. celatum, C. carnis, C. novyi, C. haemolyticum, and C. histolyticum, yielded a signal for the C. septicum-specific probe. C. celatum and C. carnis could be detected with the C. chauvoei/C. septicum-specific probe. Thus, a positive signal with the C. chauvoei- and C. chauvoei/C. septicum-specific probes indicated the presence of C. chauvoei and a positive signal with the C. septicum- and C. chauvoei/C. septicum-specific probes indicated the presence of C. septicum, which could not be distinguished from C. celatum or C. carnis. None of the non-Clostridium species tested yielded a signal with any of the probes.
To investigate the detection limit of the real-time PCR assay, 20 replicates containing an average of one C. chauvoei or C. septicum genome per PCR were tested. For both tested species, all replicates yielded positive signals at this target concentration.
The detection limit of the entire analysis procedure, including DNA isolation, was assessed in artificially contaminated muscle tissue. The lowest contamination level yielding positive results for all tested replicates was 1.79 × 103 cells/g muscle tissue.
Real-time PCR analysis of 11 clinical specimens previously analyzed by FAT yielded similar results for the two methods (Table (Table2).2). With the exception of one sample, all samples testing positive with FAT for either C. chauvoei or C. septicum yielded the same species in the real-time PCR analysis. One sample was C. chauvoei positive by FAT but yielded real-time PCR signals for both tested species. On the other hand, three samples which tested positive for both Clostridium species with FAT yielded a positive real-time PCR signal for only one species.
A real-time PCR system was designed for two reasons. Differentiation of C. chauvoei from other gas gangrene-causing clostridia is an important issue, and only conventional PCR systems have been described for this purpose.
The 16S rRNA gene was used as a target to design the real-time PCR system, permitting differentiation of C. chauvoei and C. septicum. Considering the close phylogenetic relationship between the 16S rRNA gene sequences of these two pathogens (99.3% identity) (7), an LNA modification was incorporated into the C. septicum-specific probe in order to prevent cross-reaction of that probe with C. chauvoei. As described previously (9, 19), LNA oligonucleotides bind complementary nucleic acids with greater affinity and specificity. A triplet of LNA residues centered on the discriminatory site of the C. septicum probe performed best. In addition to the C. chauvoei- and C. septicum-specific probes, a third probe, specific for both species, was incorporated into the assay to enhance its diagnostic reliability. This probe setup permitted unequivocal detection of C. chauvoei. With regard to the detection of C. septicum, 6 strains, including C. quinii, C. celatum, C. carnis, C. novyi, C. haemolyticum, and C. histolyticum, yielded a signal with the C. septicum-specific probe. With the exception of C. carnis--which is not considered to be a pathogen of ruminants (18)—and C. celatum, these strains could be distinguished from C. septicum since they did not yield a signal with the C. chauvoei/C. septicum-specific probe. However, both species are considered not to be involved in malignant edema. As an alternative, primer pairs and a probe targeting the hemolysin gene of C. septicum could be included in the assay (17).
Although inclusivity was 100% for both the C. chauvoei- and C. septicum-specific probes, variations in the CT values obtained with these probes were observed, suggesting differences in the amplification efficiencies of the tested strains. Variations in amplification efficiency due to mismatches in primer or probe binding regions have been reported (11). Sequence data for C. chauvoei and C. septicum do not suggest the presence of mismatches in the primer and probe binding regions. However, few data are present in the GenBank database, and sequencing of additional strains could provide further insights.
With clinical and diagnostic applications in mind, it was considered mandatory to include an IAC in the assay design. An IAC permits recognition of false-negative results due to inhibitory effects of the sample matrix (5). Conventional PCR assays for the detection of C. chauvoei published thus far do not include an IAC (1, 6, 8, 13, 14, 15, 16, 18).
Due to the multiplicity of the 16S rRNA gene target, a detection limit of 1 genome equivalent/PCR could easily be achieved. Since no data about the genome sizes of C. chauvoei and C. septicum are available so far, this parameter had to be estimated on the basis of the average genome size of clostridial species. The copy number of the 16S rRNA gene target within the genomes of C. chauvoei and C. septicum was estimated by an experimental approach in which a purified PCR amplicon was used as a standard. About 10 operons appeared to be present in each of these species (data not shown). The rRNA database, which does not contain information about C. chauvoei and C. septicum, yielded an average of 9.43 operons for the clostridial species listed there (10). With respect to contamination levels of 107 to 105 CFU/g found in diseased muscle tissue (6, 14), the detection limit of the proposed assay achieved in artificially contaminated muscle tissue (1.79 × 103 CFU/g) suggested reliable detection of the pathogen in clinical specimens.
However, the testing of clinical specimens revealed some discrepancies, mainly in terms of positive FAT results for both C. chauvoei and C. septicum within single samples which yielded only C. septicum by real-time PCR analysis. Parallel testing of the DNA isolated from the sample material with three different conventional PCR systems targeting the 16S-23S rRNA gene spacer region and the flagellin gene (13, 14, 16) yielded the same results as those obtained with the newly designed real-time PCR assay (data not shown). Thus, these discrepancies cannot be assigned to misidentification of the Clostridium species at the DNA level. Since only a small quantity of tissue was subjected to DNA isolation (25 mg), these results suggest that the DNAs of other species were not present in this part of the tissue. Further advancements of the procedure should be focused on creating more representative samples by increasing the quantity of sampled tissue and by pooling tissues taken from different parts of the muscle. The possibility of cross-reactions of related clostridia such as C. chauvoei and C. septicum, which may occur in immunological tests (18), should be borne in mind.
In summary, the proposed real-time PCR assay combines all essential features of a diagnostic tool to distinguish C. chauvoei from C. septicum in clinical specimens. The results gained thus far have been promising. Further testing of a larger number of clinical specimens will be performed in the future.
We thank Christian Seyboldt at the Friedrich-Loeffler Institute, Jena, Germany, Peter Kuhnert at the University of Bern, Switzerland, and Helge Boehnel at the Georg-August University of Goettingen, Germany, for providing Clostridium strains, as well as Alice Wallner, Monika Matt, and Tanja Urbanke at the Austrian Agency for Health and Food Safety for providing clinical specimens. We are also grateful to Martin Lange at the Friedrich-Loeffler Institute, Jena, Germany, for providing the results of the conventional PCR systems.
This work was funded by the Austrian Christian Doppler Research Society.
Published ahead of print on 3 February 2010.