OUP user menu

Rapid and sensitive identification of pathogenic and apathogenic Bacillus anthracis by real-time PCR

Heinz Ellerbrok, Herbert Nattermann, Muhsin Özel, Lothar Beutin, Bernd Appel, Georg Pauli
DOI: http://dx.doi.org/10.1111/j.1574-6968.2002.tb11324.x 51-59 First published online: 1 August 2002


Bacillus anthracis spores have been shown to be an efficient biological weapon and their recent use in bioterrorist attacks has demonstrated the need for rapid and specific diagnostics. A TaqMan real-time PCR for identification of B. anthracis was developed, based on the two plasmids, pX01 and pX02, both of which are necessary for pathogenicity, as well as on the chromosomally encoded rpoB gene. Bacteria picked from colonies or pelleted from liquid cultures were directly inoculated into the PCR mix, thus avoiding time-consuming DNA preparation and minimizing handling risks. B. anthracis spores were cultivated for a few hours in enrichment broth before PCR analysis, or used directly for real-time PCR, thus allowing to confirm or exclude potential attacks approximately 2–3 h after the material has arrived in the laboratory.

  • Anthrax
  • Bacillus cereus group
  • real-time PCR
  • spores
  • bioterrorist attack
  • Bacillus anthracis

1 Introduction

Bacillus anthracis spores are considered as an effective biological weapon due to their high stability. Their highly pathogenic nature and efficiency as a weapon were demonstrated in 1979 by an accidental release from a facility for biological weapons in Sverdlovsk in the former Soviet Union [1,2] and by the recently reported cases of terrorist attacks in Florida, New Jersey, New York and Washington, DC[3]. Three forms of anthrax disease in humans are known: cutaneous, gastrointestinal and pulmonary anthrax. While cutaneous anthrax is often self-limiting, an anthrax infection is generally fatal when endospores enter the body by inhalation or ingestion if not treated with antibiotics before or at the time the first symptoms develop.

Diagnosis and confirmation of B. anthracis infections from clinical specimens are easily achieved by conventional microbiological techniques[4]. In contrast, anthrax diagnosis from environmental samples is difficult and time consuming since B. anthracis needs to be distinguished from other, closely related non-pathogenic Bacillus species[5]. The need for additional enrichment steps may further slow down the diagnostic procedure. This is particularly true in the case of a potential bioterrorist attack, when various kinds of suspicious substances may have to be analyzed. A suspected bioterrorist anthrax attack could mean prophylactic post-exposition treatment with antibiotics for a large number of people, a strategy bearing its own risks[6]. In addition, expenses for decontamination of affected areas and buildings have to be taken into account. Therefore a rapid and reliable diagnostics is needed to introduce or continue counter-measures or, in the case of a confirmed negative result, to lift preventive counter-measures or to completely avoid unnecessary decontamination or medication.

Specific detection of B. anthracis by PCR is complicated by the close homology between B. anthracis, B. cereus and B. thuringiensis, which some authors consider as genetically just one species [79]. Recently, the rpoB gene has been used for the discrimination of bacterial species [10,11], and a lightcycler-based rpoB-specific PCR for the identification of B. anthracis was described[12]. However, this approach does not distinguish between pathogenic and apathogenic B. anthracis strains. Pathogenic B. anthracis carries two plasmids, the 174-kb plasmid pX01 with the toxin genes pag, lef and cya and the 95-kb plasmid pX02 with the genes capA, capB and capC involved in capsule formation [1315]. The capsule confers resistance to phagocytosis[16]. The anthrax toxin is composed of three proteins, lethal factor, edema factor and protective antigen (PA), and the molecular mechanisms resulting in the destruction of target cells are understood in some detail [1720]. For a fully pathogenic strain both plasmids have to be present [21,22]. As a consequence, a PCR-based B. anthracis diagnostics for the evaluation of a bioterrorist threat should not only be based on the detection of a specific chromosomal marker but has also to prove the presence or absence of the two plasmids in order to distinguish between pathogenic and apathogenic anthrax strains.

Here we describe a quantitative real-time PCR (TaqMan) assay specific for pag from pX01, capC from pX02 and the chromosomal rpoB. In addition to a sequence-specific primer pair, this recently introduced method uses a probe labeled with fluorescent reporter and quencher dyes. The probe, designed to hybridize internally to the flanking primers, is fragmented during cycling due to a 5′–3′ endonuclease activity of the Taq DNA polymerase as it extends the PCR primers, leading to increased fluorescence of the reporter that can be detected online[23]. This method is highly sensitive and can detect even a few or single cells in a background of other cells [24,25].

2 Materials and methods

2.1 Bacterial strains

B. anthracis strains were either from a panel of B. anthracis field isolates from the ‘Bundesamt für Gesundheitlichen Verbraucherschutz und Veterinärmedizin’ (BgVV) collected before 1962 or a gift from W. Beyer (University of Hohenheim, Germany). Other bacilli strains were either from the ‘German Collection of Microorganisms and Cell Cultures’ (Braunschweig, Germany) or from the ‘American Type Culture Collection’. B. anthracis and all other Bacillus species were grown on sheep blood agar Petri dishes at 37°C. Centrifugation of bacteria and spore suspensions was performed in an aerosol-tight centrifuge (Kendro minifuge, Kendro, Hanau, Germany).

2.2 DNA preparation from B. anthracis

Bacterial colonies were taken from blood agar plates with disposable plastic spatulas and mixed with 200 μl of ATL buffer (tissue preparation kit, Qiagen, Hilden, Germany). Suspensions were incubated for 30 min at 80°C, centrifuged for 5 min at maximum speed in an Eppendorf centrifuge to sediment spores, and supernatant was pre-diluted and used for PCR. Alternatively, the suspensions were autoclaved (20 min at 121°C) and subsequently total DNA was purified using the Qiagen tissue kit.

2.3 B. anthracis spore preparation and titration

B. anthracis was grown on blood agar. A single colony was transferred into 10 ml of liquid broth and incubated at 37°C for 24 h. This broth was plated on 50 ml manganese sulfate agar (0.01% MnSO4) in a Roux bottle and incubated at 37°C for 3 days. Subsequently, the bottle was kept for 4 weeks at room temperature and protected from light. Spores were harvested with 2×10 ml of sterile water, filtrated through several layers of gauze and centrifuged (4°C, 30 min, 5000×g). The pellet was resuspended in 65% isopropanol and kept at room temperature for 1 h. The spore suspension was diluted with an equal volume of water and recentrifuged. Sediment was washed several times with sterile water, centrifuged, resuspended in 10 ml sterile water and stored at 4°C until use. Purity of the spore preparation was analyzed by scanning electron microscopy and by Rakette staining except that spores were fixed by a 2-h incubation with 10% of formaldehyde instead of heat fixation. Triplicates from serial dilutions of spores were plated on blood agar plates and incubated at 37°C for 48 h. Colonies were counted and the number of spores per ml were calculated.

2.4 Primer selection

Using the Primer Express software (ABI Weiterstadt, Germany), three primer/TaqMan probe combinations were defined on sequences from the NCBI public data base specific for (i) the rpoB gene (AF205326) as a chromosomal marker, (ii) the pag gene (M22589) for pX01 and (iii) the cap gene (M24150) for pX02. Primer and probe design was optimized in cooperation with TIB Molbiol (Berlin, Germany). Primers for rpoB (rpoB-F and rpoB-R) were taken from Qi et al.[12] and a matching TaqMan probe, rpoB-TM, was defined. Primers and probes (Table 1) were synthesized by TIB Molbiol.

View this table:
Table 1

Primers and probes for B. anthracis real-time PCR

Target and primers and probesPrimer sequencePrimer positionaFragment size (bp)
rpoB gene175
PA-TM (probe)CTCgAACTggAgTgAAgTgTTACCgCAAAT3312–3341

2.5 PCR

Standard PCR reactions contained 2.5 U of Taq polymerase (InVitek, Berlin, Germany), 300 nM of each primer, all four dNTPs at 200 nM each and 2 mM MgCl2. Purified DNA and water were added to a total volume of 50 μl. PCR reaction was performed in a 9600 thermal cycler (Applied Biosystems, Weiterstadt, Germany) with 2 min of denaturation at 94°C and 33 cycles with 20 s at 94°C, 20 s at 60°C and 30 s at 72°C. Amplicons were analyzed by conventional DNA electrophoresis in a 2% agarose gel.

2.6 B. anthracis-specific real-time PCR

Real-time PCR reaction mix consisted of 25 μl of Universal Master Mix (Perkin Elmer) containing dNUTPs, MgCl2, reaction buffer and Ampli Taq Gold, 300 nM of each primer and 100 nM of fluorescence-labeled TaqMan probe. DNA and water were added to a final volume of 50 μl. For colony PCRs the reaction mix was filled up to 50 μl with water, and bacteria were inoculated directly into the mix without volume correction. All PCRs were performed in duplicate. For diagnostical approaches B. anthracis wild-type and B. megaterium were included as positive and negative controls, respectively. Also 104 and 102 copies of cloned target DNA were included for the corresponding PCR reaction. Quantified plasmids were prepared in cooperation with GenExpress (Berlin, Germany). Real-time PCR was performed in an SDS 7700 or an SDS 7000 as follows: 2 min at 52°C, 10 min at 95°C, 40 cycles at 15 s 95°C and 1 min at 60°C. Data were analyzed with the Sequence Detector software (Applied Biosystems).

3 Results

3.1 Detection of B. anthracis by real-time PCR

Pathogenicity of B. anthracis depends on the presence of two plasmids, pX01 coding for the toxins and pX02 coding for the capsule. In order to distinguish between pathogenic and apathogenic strains, we selected sequences from both plasmids (pag gene from pX01, capC gene from plasmid pX02) and a B. anthracis-specific region of the rpoB gene as a chromosomal marker to establish real-time PCR systems. Care was taken to select sequence and lengths of primers and probes for the different PCR assays in a way to obtain similar annealing temperatures, thus allowing to perform analysis of B. anthracis in parallel with all three PCR systems in the same cycler. Since handling of B. anthracis strains is restricted to laboratories with the BSL3 safety level, we decided to use short subgenomic cloned target sequences that can be handled in a normal laboratory setting (BSL1) in order to facilitate the set up of parameters for the different PCR reactions. The defined target sequences were amplified from a B. anthracis SDS lysate, using the three PCR primer pairs selected for the pag, cap and rpoB targets, and were analyzed by gel electrophoresis. Each PCR reaction generated just a single band of the expected size (not shown). The PCR fragments were cloned and inserts were verified by sequencing. Purified plasmid preparations were quantified with a photospectrometer and plasmid copy numbers were calculated. Aliquots of these plasmid DNAs were used in real-time PCR reactions to establish PCR conditions for the different primer/probe combinations. Serial dilutions for each plasmid were applied to define the sensitivity of the method. For each target sequence as little as 10 copies per reaction could be detected, and quantification of initial copy numbers was linear over a range from 107 to 10 copies of plasmid standard DNA (Fig. 1).

Figure 1

Real-time TaqMan PCR of pag, capC and rpoB standards. Serial dilutions of standard plasmids ranging from 10 to 107 were run in pag-specific (A), capC-specific (B) and rpoB-specific (C) real-time PCR. The curves represent fluorescence changes over cycles. Initial copy numbers for all three PCR assays are indicated in the central box. The standard curves for the three PCR assays are shown in panel D.

For conventional PCR we used SDS lysates from B. anthracis colonies incubated for 30 min at 80°C in SDS containing ATL buffer (Qiagen tissue kit). Lysates were centrifuged in order to eliminate or reduce the number of spores. It was observed by plating that in some of the preparations residual viable spores were present (data not shown). In addition, these lysates had to be pre-diluted at least 100-fold to avoid interference of SDS with the PCR reaction, thus reducing the sensitivity of the assay without completely solving the safety problems. We then suspended the bacteria in ATL buffer and autoclaved them at 121°C for 20 min. This treatment readily inactivated spores in the lysate, and with the Qiagen tissue kit spore-free B. anthracis DNA could be prepared. Since this procedure was laborious and time consuming, valuable time would have been lost in an emergency. To circumvent these problems, 50 μl aliquots of the complete PCR mix for pag, cap and rpoB, respectively, were pipetted into the wells of a 96-well PCR plate. B. anthracis colonies were picked with sterile pointed toothpicks and the adhering bacteria were resuspended in the PCR mix by swirling the toothpicks between two fingers (approx. 2–3 s). Several B. megaterium, B. cereus, B. mycoides and B. subtilis strains were included as negative controls, and quantified plasmids were used as standards. The plates were heat sealed with a translucent plastic foil, thus avoiding potential contamination with anthrax spores. The samples were subjected to real-time PCR and the collected data were analyzed with the Sequence Detector software. The sealed 96-well plates could be autoclaved immediately after the PCR reaction since it was not necessary to open them for post-PCR treatment of the products, thus eliminating risks linked to potentially surviving anthracis spores. For B. anthracis pag, cap and rpoB PCRs threshold cycles (Ct) of 20 or less were generated corresponding to at least 106 copies of the target sequence. While B. mycoides, B. subtilis and one of the two B. cereus strains were negative in all three reactions, the second B. cereus strain and B. megaterium generated weak signals in the rpoB PCR with Ct values above 30 (Fig. 2).

Figure 2

B. anthracis-specific real-time PCR on bacterial colonies. Colonies from different bacteria were picked with sterile toothpicks and transferred into PCR mixes. Bacteria were analyzed for pag, cap-C and rpoB. For non-anthracis strains only the rpoB PCR is shown.

To analyze the reproducibility of the method, we picked 19 replicates of B. anthracis from a single plate for each PCR target, i.e. pag, cap and rpoB, respectively. Each toothpick was used only once. All PCR reactions were positive for the corresponding target sequence with Ct values for pag ranging from 16.1 to 18.4, for cap ranging from 17.2 to 20.5 and for rpoB ranging from 16.4 to 18.6 with one exception with a Ct value of 24.2 (data not shown). Following PCR, 40 reaction mixes were spotted onto blood agar plates and incubated for 48 h. Only one reaction mix gave rise to bacterial colonies, indicating residual viable B. anthracis spores in this particular PCR reaction.

3.2 Specificity of B. anthracis real-time PCR

We then analyzed 12 different B. anthracis isolates and Bacillus species, previously identified as field isolates, and two B. anthracis strains known to have only the toxin plasmid (‘spike strain’) or none of the two plasmids (‘host strain’), respectively. All 12 field isolates gave strong positive signals in the rpoB PCR and in the cap PCR with Ct values of approx. 20. Ten isolates were positive in the pag PCR and two remained negative, indicating the absence of plasmid pX01, suggesting that these two B. anthracis isolates were attenuated. Also, real-time PCR confirmed the absence of capC for the ‘spike strain’ and the absence of capC and pag for the ‘host strain’.

To test the specificity of the assay, also 36 non-anthracis bacilli isolates including members of the B. cereus group were included as controls. All control strains were negative for pag and capC, while four out of five B. cereus and one out of three B. megaterium isolates were positive for rpoB, however, in contrast to the B. anthracis isolates with Ct values above 30. The results are summarized in Table 2.

View this table:
Table 2

Bacillus strains tested in pag, cap and rpoB real-time PCR

Bacillus speciesStrain IDPAaCAParpoBa
B. anthracisDU III-7++++++
B. anthracisBehring−−++++
B. anthracis4463++++++
B. anthracisB22/39++++++
B. anthracis13/38++++++
B. anthracis53/59++++++
B. anthracis19/37++++++
B. anthracis19/57++++++
B. anthracisB19/39++++++
B. anthracis527++++++
B. anthracis5261−−++++
B. anthracisB11/38++++++
B. anthracisspike++−−++
B. anthracishost−−−−++
B. azotoformansDSM 1046−−−−−−
B. brevisDSM 5619−−−−−−
B. brevisDSM 30−−−−−−
B. cereusDSM 31−−−−(+) −
B. cereusDSM 2301−−−−(+) (+)
B. cereusATCC13061−−−−(+) (+)
B. cereusATCC 11774−−−−(+) (+)
B. cereusI-H−−−−−−
B. circulansDSM 11−−−−−−
B. circulansDSM 1315−−−−−−
B. firmusDSM 1530−−−−−−
B. lentusDSM 9−−−−−−
B. lentusDSM 5221−−−−−−
B. licheniformisDSM 12369−−−−−−
B. licheniformisATCC 12759−−−−−−
B. licheniformisC-S−−−−−−
B. megateriumDSM 90−−−−−−
B. megateriumDSM 32−−−−−−
B. megateriumAD01/22/2−−−−(+) −
B. mycoidesLB−−−−−−
B. pumilusDSM 27−−−−−−
B. pumilusDSM 13835−−−−−−
B. spaericusDSM 369−−−−−−
B. spaericusDSM 28−−−−−−
B. stearothermophilusATCC 7953−−−−−−
B. stearothermophilusATCC 12980−−−−−−
B. subtilisDSM 10−−−−−−
B. subtilisDSM 13−−−−−−
B. subtilisDSM 347−−−−−−
B. subtilisATCC 9372−−−−−−
B. subtilisIr−−−−−−
B. subtilisBr−−−−−−
B. subtilisEr−−−−−−
B. thuringiensisDSM 350−−−−−−
B. thuringiensisDSM 2046−−−−−−
B. trimusDSM 12−−−−−−
  • aTwo reactions per target were performed: + positive (Ct<21); (+) positive (Ct>30); − negative.

3.3 Detection of B. anthracis in mixed cultures

Due to the presence of other aerobic spore-forming bacteria, cultivation of bacterial specimens from unknown origin in liquid media may result in mixed cultures where B. anthracis might represent only a minority among the bacteria introduced into the PCR, resulting in potential interference with the anthracis-specific PCRs. Therefore, 1-ml culture aliquots containing approximately 106B. subtilis or B. megaterium, respectively, were mixed with 10-fold serial dilutions of B. anthracis, pelleted and transferred directly into the pre-made PCR mixes for pag, cap and rpoB and analyzed in real-time PCR. For all PCRs a 1/1000 dilution of B. anthracis in a background of B. subtilis or B. megaterium, respectively, was reliably detected. For some of the PCR reactions even a 1/100 000 dilution, representing 10 B. anthracis cells in a background of 106 heterologous bacteria, was detected (Table 3). Therefore this method allows to analyze mixtures of different bacteria and to detect B. anthracis even when it is just a minor component of this mixture with a level of sensitivity that can not be obtained with conventional microbiological methods.

View this table:
Table 3

PA, CAP and rpoB PCRs for B. anthracis in a background of different bacilli

B. anthracis aB. subtilis1 bB. megaterium b
  • aNumber of B. anthracis bacteria added.

  • b106 bacteria per reaction as background.

  • cTwo reactions per target were performed: + positive (Ct<21); (+) positive (Ct>30); − negative.

3.4 Detection of B. anthracis DNA from spores

In case of a potential bioterrorist attack with weaponized B. anthracis, the primary material to analyze most likely would not be vegetative bacteria but spore preparations. In order to analyze such samples by microbiological tests and by B. anthracis-specific PCR systems, the spore material would have to be cultured in liquid media or on agar plates long enough to obtain detectable levels of vegetative bacteria before analyses can be performed. To reduce this loss of time, we determined the minimum incubation time necessary in enrichment broth for a spore preparation to become accessible to PCR diagnosis.

At different time points 105 spores from an apathogenic B. anthracis lacking both plasmids were inoculated into 1 ml of medium and incubated at 37°C. At the respective end of the incubation periods all samples were centrifuged in parallel, yielding samples with 0, 1, 2, 4 and 6 h of incubation. Pellets were resuspended in 50 μl of sterile water and 5 μl were used for rpoB real-time PCR. Already at time point 0 h the PCR was clearly positive, indicating that spores introduced into the PCR reaction were accessible to analysis without further treatment. The signal increased with incubation time (Fig. 3).

Figure 3

Detection of B. anthracis sequences from cultures inoculated with spores. 105B. anthracis spores per ml were incubated in growth medium. 1-ml aliquots were pelleted after different incubation times. Aliquots of the pellet were analyzed in rpoB-specific real-time PCR. Different copy numbers of quantified standard DNA were run as a reference and used to calculate the copy number of target DNA for the kinetics.

We then introduced 8×104, 8×102 and eight spores, respectively, directly into the PCR reaction mix and rpoB-specific real-time PCR was performed. More than 40% of the 80 000 spores introduced were detected. Of 800 spores 20% were detected and only the experiment using eight spores gave a negative PCR result. Therefore, direct analysis of spores in the B. anthracis-specific real-time PCR assay allows to considerably shorten the delay in reacting to a potential threat.

4 Discussion

The TaqMan real-time PCR for the identification of pathogenic B. anthracis is based on the detection of sequences of two pathogenicity plasmids pX01 and pX02 and on the chromosomally encoded rpoB gene. In our opinion this approach is essential for the risk assessment when analyzing material from a potential anthrax attack. Apathogenic B. anthracis strains lacking either one or both plasmids have been isolated from the environment[21] and strains with just the pX01 have been used for vaccination [26,27]. A vaccine strain has been used in an anthrax attack in Japan, but without casualities. Furthermore, the pX02 plasmid has been transferred into other bacillus species, and genes from the pX01 plasmid have been successfully expressed in other bacteria [17,19,28]. Such recombinant bacteria would generate positive results in at least one of the plasmid-specific PCR assays. Therefore only the identification of both plasmids, pX01 and pX02, together with the chromosomal rpoB gene demonstrates the presence of pathogenic B. anthracis. Apathogenic B. anthracis strains could lead to misinterpretation if not properly identified, and as a result expensive counter-measures might be introduced.

For detection of the B. anthracis rpoB gene we have adopted primers from a lightcycler assay recently described as specific for B. anthracis[12] and combined them with a TaqMan-suitable fluorescence-labeled probe. This assay identified B. anthracis chromosomal DNA with a high sensitivity reliably detecting 10 copies or less of target DNA. To our surprise, four out of five B. cereus strains tested in colony PCR gave positive signals for rpoB (Table 2), however, with Ct values over 10 cycles later than those seen for B. anthracis, suggesting the presence of copy numbers at least 1000 times lower, although comparable numbers of bacteria had been introduced into the reaction. Specificity of the rpoB PCR primers depends on single nucleotide mismatches at the 3′-terminus of both the sense and the reverse primer for B. cereus target sequences[12]. Sequencing of the rpoB gene from the B. cereus isolates confirmed the presence of the mismatches which distinguish B. cereus from B. anthracis and weak signals were also observed with cloned B. cereus rpoB sequences excluding low-level contamination of B. cereus colonies with B. anthracis (not shown). One possible explanation may be that the Taq DNA polymerase used in these experiments is capable of ‘mispriming’ with a 3′ unmatched primer nucleotide, but at a very low level not detectable by conventional PCR methods, thus simulating the presence of a few copies of B. anthracis. This could for example be the result of a residual 3′ nuclease activity eliminating the 3′ mismatch. Nevertheless, rpoB-reactive B. cereus is unlikely to be mistaken for pathogenic B. anthracis since the Ct value was remarkably higher than for B. anthracis, and in addition B. cereus was always negative for the cap and pag PCR assays.

When dealing with a potential anthrax attack, one of the most important tasks is rapid verification or falsification. We have shown that bacteria picked from colonies on agar plates or pelleted from liquid cultures can be inoculated directly into the PCR mix, thus saving time-consuming DNA preparation and handling of potentially highly infectious agents. Furthermore, it could be shown that pure preparations of B. anthracis spores can be analyzed directly in real-time PCR without prior cultivation steps, thus allowing to identify an anthrax attack within 2–3 h after arrival of the sample in the laboratory. Since additives in the spore preparation or in bacterial cultures may interfere with the PCR reaction, we have developed the following protocol. The PCR reaction is controlled by adding either wild-type or apathogenic B. anthracis cells to a parallel sample. This control reaction has to give a positive signal comparable to the positive control, even when the sample to be investigated remains negative. Cultivation of samples in enrichment broth and on blood agar plates should be started in parallel and, in the case of a negative PCR result, analysis should also be performed on the cultured material. Since additives might interfere with bacterial growth or the germination of spores, or other bacteria present in the sample could inhibit the growth of B. anthracis in enrichment medium, a parallel culture is spiked with B. anthracis spores. It must be mentioned that primary material as well as enrichment cultures should be analyzed both directly and in a 1:1000 dilution in water, since high concentrations of bacteria or spores may result in inhibition of the PCR reaction (data not shown). Cultivation time can be minimized when the bacteria are pelleted by centrifugation of the culture after a few hours of incubation and the bacterial pellet is used directly for PCR. The different approaches taken together may overcome both the problem of samples containing substances interfering with the PCR, and the growth inhibition in liquid medium by using single colonies on agar plates. In consequence, bacterial cultures and single colonies should always be processed for classical microbiological and biochemical analysis, but within this protocol PCR analysis should be performed as early as possible to save valuable time.

The primer system described here can also be used in a conventional PCR where the PCR products are analyzed by DNA agarose electrophoresis (not shown). However, samples analyzed in the conventional PCR assay would have to be pre-treated to efficiently eliminate all spores. Otherwise, analysis of the PCR products would also have to be performed in a BLS3 lab.


We are greatful to S. Pociuli, H. Emmel and T. Franz for expert technical assistance, to A. Nitsche and O. Landt for helpful discussions and to C. Drosten, K. Fleischer, D. Schimmel, A. Rassbach and W. Beyer for bacterial strains and DNA samples. We thank U. Erikli for expert help in preparation of the manuscript.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
View Abstract