Production and Characterization of Antibodies against Each of the Three Subunits of the Bacillus cereus Nonhemolytic Enterotoxin Complex

Abstract

The nonhemolytic enterotoxin (Nhe) is one of the two three-component enterotoxins which are responsible for diarrheal food poisoning syndrome caused by Bacillus cereus. To facilitate the detection of this toxin, consisting of the subunits NheA, NheB, and NheC, a complete set of high-affinity antibodies against each of the three components was established and characterized. A rabbit antiserum specific for the C-terminal part (15 amino acids) of NheC was produced using a respective synthetic peptide coupled to a protein carrier for immunization. Using purified B. cereus exoprotein preparations as immunogens, one monoclonal antibody against NheA and several antibodies against NheB were obtained. No cross-reactivity with other proteins produced by different strains of B. cereus was observed. Antibodies against the NheB component were able to neutralize the cytotoxic activity (up to 98%) of Nhe. Based on indirect enzyme immunoassays, the antibodies developed in this study were successfully used in the characterization of the enterotoxic activity of several B. cereus strains. For the first time, it could be shown that strains carrying the nhe genes usually express the complete set of the three components, including NheC. However, the amount of toxin produced varies considerably between the different strains.

Bacillus cereus is known to cause two different types of food poisoning (for reviews, see references 9, 15, and 26), which are characterized by either emesis or diarrhea. At present, two different protein complexes, each consisting of three exoproteins, as well as a single protein (cytotoxin K) (19), are discussed as causative agents (9, 13). The enterotoxin described by Beecher and Wong (3, 5), consisting of the components B, L₁, and L₂, showed hemolytic activity and was therefore named hemolysin BL (HBL). HBL has been characterized intensively in view of the biological activity (5) as well as genetically (14, 25). The nonhemolytic enterotoxin (Nhe) described by Lund and Granum (17) contains the protein components NheA (41.0 kDa), NheB (39.8 kDa), and NheC (36.5 kDa). The genes encoding the components of Nhe have been cloned and characterized, and it has been shown that they are transcribed as one operon (10, 16).

Specific monoclonal antibodies (MAbs) for immunochemical studies on the protein level are available only for HBL (6). Due to this limitation, the detection of the B. cereus enterotoxins is still not satisfactory, and a range of in vivo and in vitro tests is used to estimate the toxicity of B. cereus isolates, e.g., the mouse lethality test, the rabbit ileal loop test, the vascular permeability reaction, and cell culture assays (1, 3, 5, 27). These assays, however, do not allow differentiation between the specific activities of the individual toxins. On the other hand, several studies published during recent years showed that nearly all strains of B. cereus harbor nhe, whereas hbl genes were detected in only about 50% of the tested isolates (7, 11, 12, 21, 22, 28). Although there is some evidence that both complexes are coexpressed frequently (1, 7, 11, 24), quantitative data on the amount of toxin produced by B. cereus isolates carrying both hbl and nhe genes are not available. To provide tools for such detailed studies and to improve the detection of Nhe, we describe here the production of specific antibodies against NheA, NheB, and NheC.

MATERIALS AND METHODS

B. cereus strains, culture medium, and culture conditions.

Enterotoxic strains of B. cereus used in this study were B-4ac (DSM 4384; DSM, Germany) and NVH 0075/95 and NVH 391/98 (nhe deficient) (17, 19). A B. subtilis strain expressing recombinant NheB and Escherichia coli strains producing either NheA or NheC (16) were also used. All other B. cereus strains (prefix MHI) were isolated from infant food or dried milk products (2). For cytotoxicity testing and indirect enzyme immunoassay (EIA) analyses, cells were grown in casein hydrolysate-yeast extract broth (3) supplemented with 1% glucose (CGY medium) for 6 h at 32°C with shaking. To inhibit proteolytic cleavage of the toxins by metalloproteases, EDTA (1 mM) was added at the time of harvesting. Cell-free supernatants obtained by centrifugation (10,000 × g at 4°C for 20 min) and filtration through 0.2-μm Millipore filters were used for purification of proteins and as coating antigens in the EIA. For the production of recombinant NheB and NheC, B. subtilis and E. coli strains were grown in brain heart infusion supplemented with 50 μg kanamycin per ml for 24 h.

Production of MAbs.

Purified NheA was prepared according to the method of Lund and Granum (17, 18) and used as an immunogen. Additionally, an exoprotein preparation of strain B-4ac was produced as described by Dietrich et al. (6), and the fraction B2 obtained by gel filtration on Sephadex G-75sf was used for the immunization of mice. Two groups of 12-week-old female mice (three BALB/c strain mice and three mice of a hybrid strain of BALB/c × [NZW × NZB] per group) were immunized by intraperitoneal injection with 30 μg of the respective protein preparation, which had been dissolved in 0.01 M Tris-HCl buffer (pH 8.6) and emulsified in Freund's complete adjuvant (1:3). At day 84, the animals received a booster injection of the same amount of immunogen in incomplete Freund's adjuvant. Finally, at day 142, 3 days before cell fusion, the animals got a final booster injection of 45 μg of antigen dissolved in Tris-HCl buffer. Cell fusion experiments, establishment of hybridomas, and antibody purification were done according to previously published protocols (6).

Production of NheC antisera.

Due to the low immunogenicity of the recombinant protein, several attempts to produce monoclonal antibodies against NheC failed. Therefore, two rabbits were immunized with a synthetic peptide derived from the C terminal part of the NheC sequence (VKDYTEKLHEGVAK) and coupled to keyhole limpet hemocyanin as the protein carrier. The rabbits were immunized subcutaneously with 300 μg of the peptide protein conjugate emulsified in complete Freund's adjuvant and received booster injections at week 18 with 250 μg of the peptide protein conjugate in incomplete Freund's adjuvant. Blood samples were taken in 2- to 3-week intervals, and the serum was obtained by centrifugation of the clotted blood.

Antibody screening by indirect EIA.

To screen for antibody secreting hybridomas, an indirect EIA system (6) was established by using recombinant NheA and NheB preparations as coating antigens. A similar approach was applied to determine the relative antibody titers of the rabbit antisera against the NheC peptide. For this purpose, plates were coated with a peptide-ovalbumin conjugate at a concentration of 0.25 μg/ml.

Cytotoxicity and neutralization assay.

The cytotoxic activity of B. cereus culture supernatants was determined using Vero cells as previously described (6, 20). To check the neutralization capacity of the antibodies, serial dilutions of B. cereus culture supernatants (0.1 ml) were placed into microtiter plates together with 10 μg of the purified MAbs (1 mg/ml phosphate-buffered saline [PBS]). Cell suspensions (0.1 ml; 10³ cells/well) were added immediately afterwards. For the control, an identical preparation containing 10 μg of an unrelated MAb (MAb 1A6 against 3-acetyldeoxynivalenol, a Fusarium mycotoxin) (unpublished results) was incubated in parallel on the same plate. After 24 h, the mitochondrial activity of viable cells was determined by adding a tetrazolium salt (WST-1; Roche Diagnostics, Mannheim, Germany). The resulting dose-response curve was used to calculate the 50% inhibitory value (expressed as the reciprocal dilution that resulted in 50% loss of mitochondrial activity) by linear interpolation. Neutralization capacities of the antibodies tested were calculated by comparing the cytotoxicity titers seen in the samples with and without specific antibody.

SDS-PAGE.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out by using the PhastSystem (Amersham Biosciences, Freiburg, Germany) and precast minigels (PhastGel gradient 10 to 15). Separated proteins were stained with Coomassie brilliant blue.

Immunoblotting.

To further characterize the specificity of the monoclonal antibodies, culture supernatants containing recombinant Nhe components and exoprotein preparations from B. cereus were separated by SDS-PAGE and electrophoretically transferred to a polyvinylidene difluoride membrane, Immobilon-PSQ (Millipore, Bedford, Mass.). Immunochemical staining was performed by blocking the membrane with Tris-buffered saline (50 mM of Tris-HCl, 150 mM of NaCl [pH 7.5]) containing sodium caseinate (10 g/liter) for 30 min and then incubating it with protein A-purified monoclonal antibodies (2 μg/ml) or rabbit antisera precipitated with ammonium sulfate. After a washing step, bound antibodies were detected with alkaline phosphatase-labeled secondary antibodies (1:1,000) by using 4-nitroblue tetrazolium chloride-5-bromo-4-chloro-3-indolylphosphate (NBT-BCIP) as the chromogenic substrate according to the instructions of the manufacturer (Roche Diagnostics).

Immunoaffinity chromatography (IAC).

Monoclonal antibody 1E11 (10 mg) was attached to 1 g of CNBr-activated Sepharose 4b (Amersham Biosciences, Freiburg, Germany) according to the manufacturer's instructions. The purification procedure was comprised of the following steps: (i) storage buffer (PBS containing 0.1% sodium azide) was replaced with PBS, (ii) sample (B. cereus supernatant diluted five times in PBS) was applied, (iii) column was washed with PBS, (iv) bound NheB was eluted with glycine-HCl buffer (pH 2.5), and (v) the column was washed with PBS and stored in storage buffer. During all steps, the flow rate was set to 1 ml/min.

Analyses of B. cereus culture supernatants.

The HBL and Nhe titers of cell-free culture supernatants of B. cereus strains were determined using an indirect EIA as described recently (6), with slight modifications. Briefly, plates were coated with serial dilutions (in carbonate-bicarbonate buffer, 0.05 mol/liter, pH 9.6) of crude cell-free supernatants of the strains grown in CGY medium. After the blocking step with 3% sodium caseinate-PBS for 45 min, 100 μl of purified monoclonal antibody was added to separate wells on the plates at a concentration of 2 μg ml⁻¹ for 1 h. For the determination of NheC, the rabbit antiserum was used in a dilution of 1:1,000. After a washing step, secondary antibodies (rabbit anti-mouse immunoglobulin [Ig] [DakoCytomation, Hamburg, Germany] or goat anti-rabbit IgG [Sigma-Aldrich, Taufkirchen, Germany]) labeled with horseradish peroxidase (1:3,000 in 1% sodium caseinate-PBS) was added and incubated for 1 h at room temperature. Then the plate was washed again, and 100 μl/well of substrate-chromogen solution (1 mM 3,3′,5,5′-tetramethylbenzidine-3 mM H₂O₂ per liter of potassium citrate buffer, pH 3.9) was added. After 20 min, the color development was stopped with 1 M H₂SO₄ (100 μl/well) and the absorbance was measured at 450 nm. Antigen titers were defined as the reciprocal of the highest dilution of crude supernatants that gave an absorbance value of 1.0 unit under these conditions.

PCR.

DNA from 100-μl aliquots of overnight cultures was extracted using the DNeasy tissue kit (QIAGEN, Germany) according to the manufacturer's instructions. The strains were tested for the presence of all genes of the nhe operon (nheA, nheB, and nheC) and for hblC of the hbl operon. Primers and PCR conditions are summarized in Table 1. Primers for the detection of nhe genes were deduced from the sequences for GenBank accession number Y19005 (10), and for the detection of hblC, from the sequences from GenBank accession number U63928 (25) by use of the program Primer3 (23). For the simultaneous detection of nheB and nheC, a single primer pair was designed that spans the nheB-nheC intergenic region. Primers were produced by custom synthesis by MWG Biotech (Ebersberg, Germany). PCR (30 cycles) was performed in a total volume of 50 μl containing 25 pM concentrations of each primer, 200 μM concentrations of each deoxynucleoside triphosphate (peqlab, Erlangen, Germany), 1.5 mM MgCl₂ (for hblC, 2 mM), 1.5 U Taq polymerase (for hblC, 1.75 U; ABgene, Hamburg, Germany), 5 μl 10× polymerase buffer, and 1 μl DNA preparation. The PCR products were visualized by ethidium bromide staining after electrophoresis on agarose gels (2%).

TABLE 1.

PCR primers and conditions used in this study

Primer pair	Nucleotide sequence	Target	Product size (bp)	PCR conditions	Reference
45c1	GAGGGGCAAACAGAAGTGAA	nheA	186	94°C, 60 s; 52°C, 60 s; 72°C, 60 s	20
45c2	TGCGAACTTTTGATGATTCG
nheBC1	ACATTGCGAAAGATAGCTGGA	nheB/nheC	300	94°C, 60 s; 48°C, 60 s; 72°C, 60 s	This study
nheBC2	TGTTCTGCTGCAAAAGGATG
L2aF	CGAAAATTAGGTGCGCAATC	hblC	411	94°C, 60 s; 51°C, 60 s; 72°C, 60 s	20
L2aR	TAATATGCCTTGCGCAGTTG

RESULTS

Antibody production.

Culture supernatants of B. cereus strain B-4ac were purified by gel chromatography as described by Dietrich et al. (6), and one of the resulting exoprotein preparations, namely, fraction Sephadex G-75 B2, contained mainly proteins in the 35- to 42-kDa range. Using this fraction for the immunization of mice, a broad range of MAbs was obtained and screened against the recombinant proteins NheA and NheB. A total of 25 hybridoma cell lines secreting antibodies reactive with recombinant NheB were identified, and these antibodies were further characterized by immunoblotting. However, none of the antibodies obtained from this immunization showed reactivity with NheA, and therefore, mice were immunized with a purified preparation of this component. A total of eight hybridoma cell lines, secreting specific antibodies against NheA, were obtained from two different fusions of mouse splenocytes with myeloma cells. However, most of these antibodies were of the IgM class, and only one high-affinity IgG antibody (MAb 1A8) was identified. For the production of antibodies against NheC, a peptide-protein conjugate was used. This conjugate was highly immunogenic when the respective antisera were tested in an indirect EIA using a peptide-ovalbumin conjugate as the coating antigen. Detectable antibody titers were >1:50,000 under the conditions described above.

Immunoblot analyses.

Immunoblot analyses were done to further characterize the specificity of the produced antibodies. By using the supernatants of strains producing the recombinant proteins NheA, NheB, or NheC, the results of the screening EIAs were confirmed. Monoclonal antibody 1A8 reacted with NheA, several antibodies with NheB, and the polyclonal antisera with NheC (Fig. 1, lanes 4 through 6). However, one of the rabbit antisera also showed a considerable reactivity with the recombinant NheB (results not shown) and was therefore not used for further experiments. In parallel, the antibodies were tested against crude culture supernatants of three B. cereus reference strains representing the different diarrheal toxin profiles commonly found in this food poisoning organism. In detail, strain B-4ac produces both Nhe and HBL (7, 20), strain NVH 0075/95 produces only Nhe (17, 18), and strain 391/98 lacks both the hbl and nhe genes (19). Analyzing culture supernatants of these strains by immunoblotting the specificity of the monoclonal antibody 1A8 against NheA and the antiserum against NheC was demonstrated. Single bands were obtained for both strains producing Nhe (Fig. 1A and B, lanes 1 and 2), whereas strain NVH 391/98, which has been found to be nhe deficient by PCR (19), gave negative results (Fig. 1A and B, lanes 3). However, testing the antibodies raised against NheB revealed that most of the anti-NheB antibodies (20 out of 25) showed an additional reactivity with an unidentified exoprotein expressed by the Nhe-deficient control strain. This reactivity is exemplified in Fig. 1D by the antibody 3F1. The remaining five antibodies (1E11, 1A5, 2C3, 2B11, and 3G5) lacking this cross-reactivity showed a uniform reactivity pattern with crude culture supernatants of strains NVH 0075/95 and B-4ac, which was characterized by a strong reactivity zone between approximately 33 kDa and 38 kDa and an additional band at approximately 97 kDa (Fig. 1C, lanes 1 and 2). Characteristics of the monoclonal antibodies are summarized in Table 2. To further verify the specificity of these antibodies, culture supernatants of strain NVH 0075/95 were purified by an immunoaffinity approach using the antibody 1E11 exhibiting the highest apparent affinity for NheB in the indirect EIA. The fractions obtained after the procedure described above were separated by SDS-PAGE (Fig. 2), and the N-terminal amino acid sequence of the protein band obtained for the eluate was determined by Edman degradation (TOPLAB, Martinsried, Germany). The sequence found was VAKAYNDYEEYSL, a nicked form of NheB (17, 18).

FIG. 1.

Immunoblot reactivity of the monoclonal antibodies 1A8, anti-NheA (A), 1E11, anti-NheB (C), and 3F1, anti-NheB (D), and anti-NheC rabbit antiserum (B) with exoprotein preparations from B. cereus strain B-4ac (lanes 1), B. cereus strain NVH 0075/95 (lanes 2), B. cereus strain NVH 391/98 (lanes 3), and recombinant NheA (lanes 4), NheB (lanes 5), and NheC (lanes 6).

TABLE 2.

Characteristics of the monoclonal antibodies against Nhe of B. cereus

Monoclonal antibody	Subtype	Specificity	Neutralization of cytotoxic activity (%)a
1A8	IgG1	NheA	15
1E11	IgG1	NheB	98
1A5	IgG1	NheB	87
2C3	IgG1	NheB	98
2B11	IgG2b	NheB	87
3G5	IgG2a	NheB	92

^aTested with culture supernatants of a strain producing only Nhe and not HBL (strain NVH 0075/95).

FIG. 2.

SDS-PAGE documentation of the NheB purification by IAC using monoclonal antibody 1E11: lane 1, crude culture supernatant of the B. cereus strain NVH 0075/95; lane 2, flowthrough of the IAC column; lane 3, rinse fraction of the IAC column; lane 4, eluted NheB.

Neutralizing properties of the antibodies.

Besides the antiserum against NheC, the purified monoclonal antibodies against NheA and NheB were tested for neutralizing properties by using the Vero cell culture assay (Table 2). For this purpose, 10 μg of each monoclonal antibody, as well as the polyclonal antibodies against NheC, was added to a serially diluted supernatant of strain NVH 0075/95 (producing only Nhe) and assayed in the cytotoxicity test. While the addition of the antibody 1A8 against NheA or of the NheC antiserum had only a minor effect on the apparent cytotoxic activity of the supernatants tested, all of the antibodies against NheB significantly reduced the cytotoxic effect of Nhe (Table 2). For supernatants of strain B-4ac containing both HBL and Nhe, the residual cytotoxicity amounted to approximately 40% of the untreated control (Table 3). By using previously described (6) monoclonal antibodies against the HBL components, it could be verified that this residual cytotoxicity could be ascribed to HBL (details not shown).

TABLE 3.

Cytotoxic activity and reactivity in the indirect EIA and PCR of different strains of B.cereus

Strain	Reactivitya						Cytotoxic activityc
	Indirect EIAb			PCR
	1A8	1E11	NheC serum	nheA	nheB-nheC	hblC
Reference strains
DSM 4384	3	3	1	+	+	+	666 (60)
NVH 0075/95	4	4	1	+	+	−	1,250 (>95)
NVH 391/98	−	−	−	−	−	−	335 (<5)
Food-related strains
MHI 1	3	4	1	+	+	−	1,250 (93)
MHI 13	3	3	1	+	+	−	862 (89)
MHI 24	1	1	1	+	+	−	270 (>95)
MHI 48	3	3	1	+	+	−	1,000 (>95)
MHI 52	4	4	1	+	+	−	1,666 (94)
MHI 61	1	1	1	+	+	−	135 (>95)
MHI 64	2	2	1	+	+	−	357 (>95)
MHI 69	2	2	1	+	+	−	526 (>95)
MHI 97	4	4	1	+	+	−	769 (>95)
MHI 109	3	3	1	+	+	−	555 (>95)
MHI 124	3	2	1	+	+	−	322 (>95)
MHI 126	2	2	1	+	+	−	277 (>95)
MHI 144	2	2	1	+	+	−	400 (>95)
MHI 183	2	2	1	+	+	−	152 (>95)
MHI 195	2	1	1	+	+	−	217 (>95)
MHI 1476	2	4	1	+	+	−	1,316 (>95)
MHI 1493	3	4	1	+	+	−	2,500 (87)
MHI 1496	3	4	1	+	+	−	1,818 (>95)
MHI 1563	4	4	1	+	+	−	1,389 (>95)
MHI 1662	2	3	1	+	+	−	690 (82)

^aPositive (+) and negative (−) results are indicated for the indirect EIA and PCR.

^bIndirect EIAs were based on antibody 1A8 against NheA and antibody 1E11 against NheB and the polyclonal antiserum against NheC. Values represent the relative reactivity of the culture supernatants tested: 1, antigen titer of >10 to 500; 2, antigen titer of 500 to 1,500; 3, antigen titer of 1,500 to 2,500; 4, antigen titer of >2,500.

^cValues represent the reciprocal values of the titers obtained from the cytotoxicity test. Percentages of neutralization of the cytotoxicity after the addition of antibody 1E11 are given in parentheses.

Characteristics of B. cereus strains.

A total of 50 different strains of B. cereus, including the reference strains, were screened by PCR for hblC, nheA, and nheB-nheC. It turned out that 49 of these strains harbored the nhe genes, 28 were positive for hblC, and strain NVH 391/98 was negative for both. To test the applicability of the antibodies for the characterization of cytotoxic activity of B. cereus strains and to avoid the influence of HBL on the cytotoxicity results, only the 20 food-borne isolates which were positive for nhe and negative for hblC were used for further analyses. Cell-free culture supernatants of these strains were tested in parallel for cytotoxicity and reactivity in the indirect EIA (Table 3). It could be demonstrated that all of these nhe-positive strains produced the complete enterotoxin complex; all three Nhe components could be detected in the respective assays. Antigen titers, defined as the reciprocal dilution giving an absorbance value of 1.0 in the EIA, ranged from 180 to 6,000 for NheA and NheB. In contrast, measurable NheC titers were comparatively low, ranging from 5 to 180. The corresponding cytotoxicity values ranged from 135 to 2,500. Neutralization assays confirmed that most of the cytotoxicity measurable in the culture supernatants was due to the activity of Nhe (Table 3).

DISCUSSION

During recent years, it has become clear that, besides HBL, another enterotoxin complex, namely, Nhe, represents a major pathogenicity factor involved in the etiology of B. cereus diarrheal syndrome (17). Though multiple studies showed that nearly all B. cereus isolates possess the nhe genes (7, 11, 12, 22), only limited studies on the expression efficiency are available (8, 11). This is mainly caused by the lack of specific detection tools, such as well-characterized antibodies.

The primary objective of the work was therefore the development of specific antibodies against each of the three Nhe components, thus facilitating detection and quantification of this enterotoxin complex. Based on experiences made during the development of MAbs against HBL (6), initially impure preparations of exoproteins were used for immunization purposes. This approach worked well for the NheB but failed for the other two components insofar as no specific antibodies for these components were detectable in the respective antisera. Since the recombinant proteins were not available at this time, alternative immunogens were produced as described in Materials and Methods. As none of the used immunogens contained the complete tripartite Nhe, all of the applied preparations were well tolerated by the animals and adverse reactions were not observed.

Owing to the formation of weakly immunogenic self-aggregates of the purified NheA, only one MAb of the IgG subtype (1A8) was obtained against this component. This MAb, however, exhibited an extraordinary high affinity for the toxin, enabling positive EIA results with culture supernatants of the reference strain diluted up to 1:10,000. Immunoblot analyses proved the desired specificity, with no cross-reactivity observed either within the Nhe complex or with other exoproteins produced by B. cereus, particularly not with the L₂ component of HBL sharing sequence similarities with NheA (9, 10). For the detection of NheA, a commercial test kit, Bacillus diarrheal enterotoxin visual immunoassay (Tecra), is available and frequently used for the analysis of enterotoxin production of Bacillus species (7, 11, 24). However, the specificity of the employed polyclonal antibodies is unclear. Using immunoblot analyses, Beecher and Wong (4) showed that, besides NheA, a broad range of other exoproteins with apparent molecular masses ranging from 41 to 114 kDa is recognized by the antibody-enzyme conjugate supplied with the EIA kit. Therefore, published results obtained with this test kit have to be interpreted carefully.

After a first screening, a broad panel of MAbs reactive with recombinant NheB was obtained. However, detailed immunoblot analyses (Fig. 1D) revealed that most of these antibodies additionally reacted with an unidentified exoprotein expressed by the B. cereus strain NVH 391/98, which lacks the nhe gene (19). In this context, it is interesting to note that the previously described MAb 1C2, which detects the L₁ component of HBL and exhibits cross-reactivity with NheB (6), shows exactly the same reactivity pattern. Therefore, we assumed that a common epitope exists in L₁, NheB, and the unknown protein. On the other hand, none of the MAbs against NheB showed reactivity toward the recombinant NheC, as was expected due to the substantial sequence homologies of these two proteins (10). To enable the specific detection of NheB, consequently, further work concentrated on the remaining five MAbs which were characterized by a broad reactivity zone in the range from 36 to 39 kDa and an additional band at the top of the blot (Fig. 1C). An NheB aggregate with an apparent molecular mass of 105 kDa was also seen in a recent study (8) in which the extracellular proteome of B. cereus was first separated by two-dimensional electrophoresis and then analyzed by matrix-assisted laser desorption ionization-time of flight mass spectrometry. It was speculated that this spot could be due to the formation of a stable complex between NheB and another protein (8). In our study, this top protein was also found in the recombinant NheB preparations, suggesting that only NheB contributes to the formation of this complex. So far, in all our experiments with Nhe-positive B. cereus isolates, this complex was commonly observed in the crude culture supernatants. However, despite the distinct reactivity of the MAb 1E11 with this complex, neither by SDS-PAGE (Fig. 2) nor by an immunoblot analysis (data not shown) was it possible to detect the corresponding band after IAC. A possible explanation for this unexpected result could be that the complex is unstable under the acidic conditions used for the elution process.

After purification of culture supernatants from B. cereus strain NVH 0075/95 by IAC, a single band with an apparent M_r of 36,000 could be seen in the SDS-PAGE gel (Fig. 2) after the gel was stained with Coomassie blue. Use of the more sensitive immunoblot technique revealed that the preparation also contained some minor bands (results not shown). Subsequent analysis of the N-terminal amino acid sequence showed that the dominant protein represents a truncated form of the NheB enterotoxin, proving the specificity of the MAb 1E11. A similar degradation product has been obtained by Lund and Granum (17, 18) after purifying NheB by chromatographic techniques. Obviously, the first 12 amino acids of the NheB protein are commonly degraded by simultaneously expressed proteases to a variable degree. This finding could also explain the observed broad reactivity zone of the MAbs (Fig. 1), which was probably caused by a mixture of differently truncated NheB derivates. Overall, the IAC purification yielded an average amount of approximately 300 μg per run corresponding to an NheB concentration of at least 6 μg/ml of the crude supernatant, indicating that NheB belongs to the highly expressed exoproteins, and this finding is consistent with data reported by Gohar et al. (8). Taking into account this level of productivity and the antigen titers measured by indirect EIA, it can be concluded that the detection limit of this assay is below 1 ng NheB/ml, a further indication for the high affinity of the produced MAbs.

All five MAbs against NheB possess distinct neutralizing properties. For instance, after the addition of MAb 1E11, negligible cytotoxic activity was observed for culture supernatants of strains producing only Nhe and not HBL and cytotoxicity titers decreased from >1:1,000 to <1:20. NheB represents the binding unit of the enterotoxin complex, as shown recently by Lindbäck et al. (16). Consequently, the question arises if the neutralizing effect of the MAbs can be ascribed to an interaction with the receptor binding epitope of the toxin or inhibition of binding of the other components to NheB, thus preventing the formation of an active toxin complex. Preliminary results (unpublished data) suggest that the latter case applies, but these findings need to be clarified in a more detailed study.

Recent findings (16) also proved the relevance of the NheC component for the formation of an active toxin complex despite the fact that several approaches failed to detect this component in culture supernatants of B. cereus (8, 10, 18). To produce polyclonal antibodies against this component, a peptide sequence corresponding to the C-terminal end of NheC was chosen for the immunization since the C terminus of a peptide coupled to a carrier protein usually represents an immunodominant epitope. As this peptide shows a certain sequence homology with NheB, i.e., XKDYXEK, at the N terminus, it was not surprising that only one of the two rabbits immunized produced antibodies specific for NheC (Fig. 1). By using the specific antiserum, NheC could be detected for the first time in culture supernatants of B. cereus both under native (EIA) and denaturing (immunoblot) conditions. The EIA results also suggest that in untreated B. cereus supernatants, the C terminus of NheC represents a freely accessible epitope and is not assembled within the tertiary structure of the protein. However, during these experiments, it also became obvious that, compared with the EIA systems for the detection of NheA and NheB, relatively low NheC antigen titers could be found in the cell-free supernatants. For instance, whereas NheC titers in a culture supernatant of strain NVH 0075/95 were 1:80, the corresponding NheA and NheB titers, however, were approximately 1:2,500. Whether this finding is due to a considerably lower affinity of the polyclonal antibodies against NheC or to a low expression level of this component needs to be clarified. In a recent publication based on recombinant proteins, Lindbäck et al. (16) reported that the maximum cytotoxic activity of the Nhe enterotoxin complex was obtained when the NheA:NheB:NheC molar ratio was 10:10:1 and it was postulated that, compared to NheA and NheB, B. cereus strains express only small amounts of NheC, possibly due to a predicted stem-loop structure between the nheB and nheC genes (10).

To demonstrate the applicability of the developed immunochemical methods and to get a first indication of the Nhe expression of B. cereus, 50 isolates were enriched under optimized conditions and the resulting supernatants were analyzed by EIA (presence/absence tests). Besides this, the nhe and hblC genes were detected by PCR assays. By applying EIA and PCR methods, consistent results were obtained. All isolates carrying the hblC gene expressed the L₂ component of HBL (details not shown), and all isolates carrying the nhe genes expressed the complete set of the three components. Thus, the PCR primers used in our study correctly identified all nhe-positive isolates despite commonly observed nhe gene polymorphisms (11). These results indicate that the primer sequences used are located in rather conserved sections of the genes. The results obtained also suggest that the monoclonal antibodies against NheA (1A8) and NheB (1E11) react with a stable epitope, which is not affected by amino acid substitutions.

To avoid biased cytotoxicity results due to the simultaneous production of HBL, cell culture analyses concentrated on the 20 strains producing only Nhe (PCR negative for hblC). Comparison of cytotoxicity and Nhe antigen titers of the tested culture supernatants revealed a consistency between these two parameters. Simultaneously performed neutralization assays corroborated these findings. The addition of MAb 1E11 possessing neutralizing properties decreased the observable cytotoxic activity of most of the tested culture supernatants by more than 95% (Table 3). Thus, it could be demonstrated that the cytotoxicity results are not biased by other virulence factors, such as proteases, hemolysins, and phospholipases (8), possibly produced by B. cereus under the enrichment conditions used for the preparation of the cell-free culture supernatants. With regard to the Nhe productivity, marked differences between the particular isolates could be observed; for example, NheB antigen titers ranged from 1:180 to 1:6,000. Similar results have been described for the productivity of HBL (6, 11, 21, 28).

In conclusion, this study describes the production and characterization of the first complete set of high-affinity antibodies against all three components of the Nhe enterotoxin produced by B. cereus. The antibodies enable the specific and sensitive detection of these compounds in culture supernatants and may be used together with the previously produced MAbs against the HBL enterotoxin for the quantitative evaluation of the toxin expression of B. cereus strains. Furthermore, the neutralizing properties in particular of the MAbs against NheB enable detailed studies on the mode of action of this toxin and provide a basis for further differentiated studies on B. cereus strains expressing simultaneously the HBL and the Nhe enterotoxin complex.