OUP user menu

Pathogenic potential of fifty Bacillus weihenstephanensis strains

Lotte P Stenfors, Ralf Mayr, Siegfried Scherer, Per Einar Granum
DOI: http://dx.doi.org/10.1111/j.1574-6968.2002.tb11368.x 47-51 First published online: 1 September 2002


The aim of this study was to evaluate the food poisoning potential of strains of the new species in the Bacillus cereus group, B. weihenstephanensis. Fifty strains were tested for cytotoxicity in a Vero cell assay, and 23 of the strains were also tested for production of enterotoxin components with commercial antibody kits, and for presence of enterotoxin gene components by polymerase chain reaction (PCR). The majority of the strains (72%) were not cytotoxic, although all of the strains that were tested with PCR and commercial kits had part of at least one of the B. cereus enterotoxins Hbl, Nhe or CytK.

Key words
  • Pathogenic
  • Enterotoxin
  • Food poisoning
  • Bacillus weihenstephanensis
  • Bacillus cereus


Bacterial strains belonging to the Bacillus cereus group can be isolated from several types of food, as they are common soil inhabitants in various climatic zones[1] and can easily contaminate raw foods as well as food processing equipment. Pasteurised milk and milk products are a common source for isolation of psychrotolerant strains of Bacillus species [25]. B. cereus can produce off-flavours in milk already at low counts, as well as causing the defect called bitty cream.

In 1998, Bacillus weihenstephanensis was suggested as a new species on the basis of sequence differences in ribosomal RNA genes and cold-shock protein genes[6]. Although the psychrotolerant strains clearly constituted a separate group in the 1998 study, it was also clear from the high degree of similarity in the DNA sequences, that the B. cereus group species were closely related. Since the B. weihenstephanensis strains differ in one physiological characteristic, the optimum temperature of growth, there is a need to establish which other properties are different from those of B. cereus. The pathogenicity of B. cereus is well established, as it can cause local and systemic opportunistic infections as well as being a major contributor to the food poisoning statistics of some countries (Norway, Japan, the Netherlands). B. cereus produces one emetic toxin and at least three enterotoxins associated with the diarrhoeal type of food poisoning[7]. The haemolytic enterotoxin Hbl has been well characterised and has several biological effects as well as causing food-borne gastroenteritis[8]. Hbl consists of three proteins, B, L1 and L2, which is similar to the composition of the non-haemolytic enterotoxin Nhe[9]. A third enterotoxin involved in B. cereus gastroenteritis was described recently after an outbreak in France, in which three persons died. This toxin was named CytK, and it consists of a single necrotic and haemolytic protein of 34 kDa[7]. To this point, we have not seen any documentation of food-poisoning outbreaks caused by B. weihenstephanensis, as this species and the methods for its identification are still new. It seems important to clarify if the new species in the B. cereus group has a similar pathogenicity profile as its relatives.

2Materials and methods

2.1 Bacterial strains, media and growth conditions

Eight strains (Weihenstephan Bacillus Collection, WSBC 10201, -02, -04, -06, -08, -09, -10, -11) have been characterised in the article proposing the new species, including the type strain for B. weihenstephanensis[6]. These and two others (WSBC 10203, 10212) have been isolated from German pasteurised milk after enrichment at 7°C[10]. All other strains have been isolated from German pasteurised milk or lab-heated raw milk (80°C, 10 min), respectively, after enrichment at 10°C (Ralf Mayr, unpublished results). PEMB Agar (Oxoid) was used for isolation and purification of presumptive B. cereus/weihenstephanensis and identification was performed by polymerase chain reaction (PCR) according to Francis et al.[11] (Mayr, unpublished results). For strains that had not been characterised earlier by Lechner et al.[6], growth temperatures were determined within 9 days as agitated liquid cultures in Plate Count Broth (8 ml, inoculated with 20 μl of a 30°C overnight culture). The cultures were checked for visible growth.

Cell extracts for cytotoxicity tests were produced as follows: 5 ml of BHIG (brain heart infusion, Difco, with 1% glucose added) was inoculated from blood agar plates and incubated overnight at 32°C with agitation (approximately 100 rpm). 0.5 ml of the overnight cultures was transferred to 50 ml BHIG and incubated at 32°C with agitation for 6 h. Extra cellular components were harvested by centrifugation of the cultures at 12 000×g, 4°C for 20 min. Aliquots of the supernatants containing the extra cellular components were immediately frozen at −20°C.

2.2 Vero cell assay for cytotoxicity

The cytotoxicity of all the strains was tested using a method described by Sandvig and Olsnes[12]. This assay measures the inhibition of protein synthesis in the Vero cells, caused by the toxin(s). Inhibition is measured by incorporation of [14C]leucine into proteins, so that cytotoxic strains obtain very low radioactive counts. For each of the 50 strains, two parallels of 3, 10, 30 and 100 μl crude toxin were applied on the Vero cell monolayers. The B. cereus reference strains 1230-88 or CCUG 6514 were used as positive control (100 μl). Seven strains that initially showed a cytotoxicity of 20–30% were tested once more, after 10-fold concentration of the toxin extract by precipitation with ammonium sulfate. The precipitate was resuspended in a 20-mM Tris–HCl buffer, pH 7.6, and dialysed overnight against this buffer in Spectra/Por® dialysis membrane tubing (Spectrum Laboratories, Inc., USA) at 4°C. The degree of inhibition of protein synthesis was calculated from the mean of duplicates: 100%−(100×mean count of test/mean count of negative control−mean count of positive control).

2.3 PCR amplification of enterotoxin gene components

Twenty-three of the B. weihenstephanensis strains were tested by PCR for presence of gene components encoding the enterotoxins Hbl, Nhe and CytK. The standard programme was as follows: 95°C for 1 min, 30 cycles of 95°C for 1 min, annealing temperature for 1 min and 72°C for 1 min, followed by an extension step at 72°C for 7 min. The primers and annealing temperatures used are listed in Table 3. Template in the reactions was genomic DNA, extracted from BHIG cultures by an isoamylalcohol/chloroform method[13].

View this table:
Table 3

PCR primers used in this study

Primer nameSequence (5′–3′)Amplified geneAccession numberPrimer pos. (5′–3′)Annealing temp. (°C)Fragment length (bases)
L1FCGCTCAAGAACAAAAAGTAGGhblD (L1)U639282664–268455802

All the PCR reactions were performed in an MJ Research Minicycler™ PTC-150 equipped with heated lid. A mastermix was prepared for the 50-μl reactions, using DyNAzyme II DNA polymerase and dNTP Mix from Finnzymes.

2.4 Bacillus cereus enterotoxin kits (TECRA® and Oxoid)

The culture supernatants from the selected 23 strains were also tested for enterotoxin production with two commercial kits, manufactured by Oxoid Ltd. and TECRA® International Pty. Ltd., respectively. The Oxoid kit uses antibody-covered latex particles to detect the L2 component (encoded by hblC) of the hemolytic B. cereus enterotoxin Hbl. The TECRA® test is performed in antibody-covered wells and detects the 41-kDa protein component of the non-hemolytic enterotoxin Nhe, encoded by nheA.

3 Results and discussion

The results of the growth tests (Table 1) meet the phenotype of B. weihenstephanensis as described by Lechner et al.[6] and thus confirm the identification of the isolates performed by PCR (Mayr, unpublished results). Hence no intermediate types were included in this study, which can be observed sometimes [1,14] and are interpreted as a ‘snapshot’ of ongoing thermal adaptation within the B. cereus group[1].

View this table:
Table 1

Growth of B. weihenstephanensis at different temperatures

WSBC strain no.Growth intensity (°C)a
  • aGrowth intensity: −= none, += weak, ++= medium, +++= intensive.

Of the 50 strains tested in this study, six were highly cytotoxic, causing 80% or more inhibition of the protein synthesis in the Vero cell assay (Table 2). In the food poisoning strains we encounter at the Norwegian reference laboratory for B. cereus, this is the level of cytotoxicity we usually see. Additionally four strains had a relatively high level of cytotoxicity, over 50% inhibition when 100 μl was applied. Thirty-six strains, constituting the major part of the tested strains (72%), were not cytotoxic in our assay. This is an unusually large proportion compared with what we normally find when testing B. cereus strains, but is in accordance with results from a study by Prüss et al.[15] where the majority of 15 B. weihenstephanensis strains were only weakly cytotoxic. One of the exceptions in that study was strain WSBC 10209, which was also found to be highly cytotoxic in our assay.

View this table:
Table 2

Cytotoxicity of B. weihenstephanensis strains, and results from PCRa and toxin kits on 23 strains

WSBC strain no.hblD (L1)nheB (39 kDa)cytKOxoid kit (L2)TECRA® kit (41 kDa)Cytotoxicityb
  • ND = not determined.

  • aIn the PCR assays, ‘+’ means amplification of the correctly sized band, ‘–’ means that no PCR product could be detected in gel electrophoresis.

  • bInhibition percentage calculated from 100 μl test volume results. >90%=+++, 80–90%=++(+), 50–80%=++, 30–50%=+, <20%=–.

  • cWSBC 10204 is the type strain of B. weihenstephanensis.

In the PCRs, the correct fragment could be amplified from all the 23 tested strains with the nheB primers. It seems, from our earlier experiments, that the vast majority of B. cereus strains possess the nhe genes[14] (unpublished results from Norwegian reference laboratory for B. cereus). Other species belonging to the B. cereus group may share this characteristic; in a study done on 74 strains of Bacillus thuringiensis, the nheBC genes were found in all the strains by PCR[16].

With the hblD primers, approximately half the tested strains (14 of 23) yielded the positive PCR fragment. This result is also in accordance with our earlier experiences with B. cereus strains[14] (unpublished results from Norwegian reference laboratory for B. cereus). In a study of the prevalence of Hbl in all the species of the B. cereus group, the authors found that nine of 15 B. weihenstephanensis strains, and 10 of 23 B. cereus strains, carried hblA, encoding the toxin component B.

With the cytK primers, only one strain was positive in the PCR. So far, very few of the B. cereus strains we have tested with these cytK primers have given a PCR product, though with more degenerate primers we have found some strains with a cytK-like gene [7,14]. The cytK positive strain was also positive for Nhe and Hbl in both PCR and the TECRA® and Oxoid kits, and showed a high level of cytotoxicity in the Vero cell assay (Table 2).

The Oxoid kit detects the L2 component of the haemolytic B. cereus enterotoxin. In this study, only nine of the 23 tested strains were positive in this assay. Using the TECRA® kit, detecting the 41-kDa protein from the non-haemolytic enterotoxin, we found 20 positive strains out of the 23 tested. When comparing the results from the commercial toxin detection kits with the PCR results, we find a relatively good accordance between the methods detecting Nhe. Exceptions are the three strains that are negative in the TECRA® test, while all strains are PCR positive for nheB. In the case of the haemolytic enterotoxin, seven strains that give the correct PCR product with the hblD primers are negative in the Oxoid test. Two strains lacking the hblD gene express the hblC gene (L2 protein) to a level detectable in the Oxoid kit.

4 Conclusion

The tests done in this study detect two out of three of the components of each of the enterotoxins Hbl and Nhe, as well as the recently described, lethal B. cereus enterotoxin, CytK. While PCR results cannot be claimed to confirm the presence of a complete and functional gene, they certainly give a good indication of whether the gene is at all present, especially using a careful selection of primers. When PCR results are evaluated together with other detection methods, in this study antibody detection of other enterotoxin proteins and the cytotoxicity assay, they indicate that many B. weihenstephanensis strains have the genetic makeup for producing essential pathogenicity factors, and that some do so under laboratory conditions.


We thank The Research Council of Norway (Grant 124097/110 to L.P.S.) for supporting this work.


  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].
View Abstract