Clostridium botulinum produces a potent neurotoxin, which causes the severe neuroparalytic illness in humans and animals referred to as botulism. The neurotoxin is serologically differentiated according to its neutralization with type-specific antitoxins into seven types designated by the letters A through G(1) . These neurotoxin (BoNT) ()types are proteins of about 150 kDa, which naturally exist as one of the components of progenitor toxic complexes: as the M complex (about 300 kDa) consisting of BoNT associated with a nontoxic-nonhemagglutinin (NTNH) protein of about 150 kDa, or as the L and LL complexes (about 500 and 900 kDa, respectively) in which the M complex associates with hemagglutinin protein(s) (designated as hemagglutinin 33 (HA-33, about 33 kDa) and hemagglutinin-II (HA-II, about 17 kDa); (2) and (3) ). The genes of these components of the toxin complexes are linked in clusters in C. botulinum types A, B, and C(2, 4) .

With the exception of C. botulinum types C and D, strains of C. botulinum usually produce only one neurotoxin type. There are, however, occasional reports of strains that produce two toxin types, of which one is at much higher titer than the other. Examples include types Af, Bf, and Ba(1) , where the major toxin type is identified by an uppercase letter and the minor type by a lowercase letter. Franciosa et al.(5) , using polymerase chain reaction (PCR) methodologies, reported the detection of genes encoding type B neurotoxin (BoNT/B) in a large number (43 of 79) of strains of C. botulinum type A, only one of which produced any demonstrable type B toxin. The one producing both toxins would correspond to type Ab according to the above convention. We use the designation A(B) in this article for the strains that produce type A toxin and possess the type B toxin gene, but do not express it. Cordoba et al.(6) recently confirmed the presence of silent or unexpressed BoNT/B genes in some strains of C. botulinum type A using PCR-restriction fragment length polymorphism and probing methodologies.

The surprisingly high frequency of a silent BoNT/B toxin gene in C. botulinum strains producing only type A toxin raises the important question of why the gene is not expressed. Although the silent type B gene was detected by PCR in three of the type A strains (type A(B)) using four different primer pairs or combinations designed for amplification of 5`-terminal, central, and 3`-terminal regions, as well as nearly the entire gene(5) , sequences within those segments could be altered in major or subtle ways. It is also not known whether genes encoding other components (e.g. NTNH and HA-33) normally associated with BoNT production are present.

This article reports the collaborative work of two laboratories directed toward explaining the silence of the type B gene of C. botulinum A(B) strains. The work at the University of Wisconsin focused on determining the presence, linkage, copy number, and orientation of the genes in the type A and B gene clusters. The Institute of Food Research, Reading Laboratory cloned and sequenced different parts of the A and B gene clusters and analyzed the sequence data. Collectively, the study reveals major alterations in the organization, structure, and expression of the genes encoding proteins of the toxin complexes.

EXPERIMENTAL PROCEDURES

Cultures and Cultivation

Oligonucleotides

DNA Preparation and Hybridization

PCR Amplification and Cloning

Sequence Determination

Hemagglutinin Assay

RESULTS

Confirmation of the Presence of Toxin A and B Genes

For Southern hybridization analyses, chromosomal DNAs were digested with BglII; the fragments were separated by electrophoresis and then hybridized with type A (BoNT-334a) and type B (BoNT-120b) toxin-specific gene probes. Hybridization confirmed the PCR findings with the four type A(B) strains and the Ab strain, giving positive signals with both probes (data not shown).

Presence and Copy Number of Neurotoxin Cluster Genes

Figure 1: Southern hybridization analyses of XbaI, BglII, and HindIII digests of chromosomal DNA from toxin types Ab (588) and A(B) (667) strains of C. botulinum to determine presence and copy number of neurotoxin cluster genes. A, Southern hybridization with BoNT/A gene probe (BoNT-334a); B, Southern hybridization with BoNT/B gene probe (BoNT-120b); C, Southern hybridization with NTNH gene probe (NN-2913); D, Southern hybridization with HA gene probe (Ha-1a). Lanes 1, 3, and 5, 588 (Ab); lanes 2, 4, and 6, 667 (A(B)); lanes 1 and 2, XbaI digests; lanes 3 and 4, BglII digests; lanes 5 and 6, HindIII digests. phage DNA/HindIII digest was used as a molecular size marker.

Linkage and Physical Relationship of BoNT and Other Neurotoxin Cluster Genes

Figure 2: Southern hybridization of PvuII, BamHI, and NdeI digests of chromosomal DNA from toxin types Ab (588) and A(B) (667) strains of C. botulinum for determining the linkage and physical relationship of BoNT and other toxin cluster genes. DNA digests were separated by PFGE, and the pulse time was ramped for 0.5-10 s over 18 h at 180 V; two duplicates were made. A, Southern hybridization with BoNT/A gene probe (BoNT-334a); B, Southern hybridization with BoNT/B gene probe (BoNT-120b); C, Southern hybridization with NTNH gene probe (NN-180); D, Southern hybridization with HA gene probe (Ha-1a). Lanes 1, 3, and 5, 588 (Ab); lanes 2, 4, and 6, 667 (A(B)); lanes 1 and 2, PvuII digests; lanes 3 and 4, BamHI digests; lanes 5 and 6, NdeI digests. A 5-kb DNA ladder was used as a molecular size marker.

To determine the relationship of the type A and B gene clusters, genomic DNAs were digested with rare cut enzymes (SmaI and XhoI), separated by PFGE, and hybridized with BoNT-334a and BoNT-120b probes (Fig. 3). In SmaI digests, a 106-kb fragment from strain 588 and a 60-kb fragment from strain 667 hybridized with both BoNT-334a and BoNT-120b probes. In XhoI digests, a similar size fragment (80 kb) in strain 588 and 667 was revealed with BoNT-334a and BoNT-120b probes. Since type A and B gene clusters themselves constitute a total of approximately 20 kb, these hybridization results suggest the distance between the gene clusters in strains 588 and 667 is less than 60 and 40 kb, respectively.

Figure 3: Southern hybridization of SmaI and XhoI digests of chromosomal DNA from toxin types Ab (588) and A(B) (667) strains of C. botulinum for estimating the distance between the type A and type B gene clusters. DNA digests were separated by PFGE, and the pulse time was ramped for 1-40 s over 22 h at 200 V. A, Southern hybridization with BoNT/A gene probe (BoNT-334a); B, Southern hybridization with BoNT/B gene probe (BoNT-120b). Lanes 1 and 3, 588 (Ab); lanes 2 and 4, 667 (A(B)); lanes 1 and 2, SmaI digests; lanes 3 and 4, XhoI digests. phage DNA ladder was used as a molecular size marker.

Characterization and Sequence Analysis of the Type A Gene Cluster

Figure 4: PCR amplification/cloning strategy and gene organization. A, type A gene cluster; B, type B gene cluster, of C. botulinum type A(B) strain 667.

A stretch of approximately 800 bases at the 5`-end of the BoNT/A gene was sequenced, which upon computer analysis showed 100% amino acid identity with published BoNT/A(13) , thereby confirming its serological assignment. The NTNH gene immediately upstream of BoNT/A gene consisted of an ORF encoding 1162 amino acid residues. Notable was a 33-amino acid deletion (from amino acid 115), which also occurs in the NTNH of non-proteolytic C. botulinum types E and F(14) . The partial sequence of the unidentified ORF linked to NTNH/A gene displayed high homology with the analogous ORF in C. botulinum type E (84.4% amino acid similarity/64.4% identity for a comparison of 82 amino acids) and nonproteolytic type F (96.6% amino acid similarity/89.2% identity).

Characterization and Sequence Analysis of the Type B Gene Cluster

The NTNH gene within the type B cluster consisted of an ORF encoding 1198 amino acids. Comparative analysis revealed only 92.8% and 88.3% overall amino acid sequence similarity and identity, respectively, with NTNH of proteolytic type B C. botulinum (data not shown). Upon closer inspection, the NTNH gene sequence was found to be unusual; regions of 5`-and 3`-ends of the gene exhibited high homology (amino acids 1-550, 99.6% and 99.5% sequence similarity and identity; amino acids 1022-1198, 97.7% and 97.2% sequence similarity and identity) with proteolytic type B NTNH, whereas a central stretch of 471 amino acid residues (from amino acids 551 to 1021) displayed much lower relatedness (83.2% and 71.9% sequence similarity and identity). Sequence analysis surprisingly revealed the central moiety of this NTNH to be highly homologous (99.8% and 99.2% amino acids sequence similarity and identity) with the equivalent region of the NTNH gene encoded in the type A cluster. Fig. 5shows a pairwise comparison of deduced amino acid sequences of NTNH genes within the A and B clusters of the strain 667, and illustrates the ``chimeric'' nature of the gene in the latter cluster.

Figure 5: Alignment of deduced amino acid sequences of NTNH genes of type A and type B cluster of C. botulinum type A(B) strain 667. Identical amino acids between the two proteins are shown in boldface type.

The PCR strategy employed (Fig. 4B) resulted in the determination of 3296 nucleotides of the silent BoNT/B gene, which correspond to the first 3304 nucleotides in the neurotoxin gene sequence from the C. botulinum (proteolytic) strain Danish, published by Whelan et al.(15) . The deviations from the published nucleotide sequence are listed in Table 3. Analysis of the determined sequence revealed a stop codon at amino acid position 128 due to the substitution of a T in place of a G at nucleotide 438. Deletion of nucleotides 1038-1043 results in amino acid deletions at positions 328 and 329. Interestingly, these amino acid residues are conserved in all published BoNT/A through G neurotoxins (as well as in the tetanus toxin). Two base deletions were also evident in the silent BoNT/B gene at nucleotide position 2389 (T deleted) and 2944 (A deleted). These deletions resulted in multiple stop signals due to two reading frameshifts. Over the course of the 3240 nucleotides corresponding to 3248 nucleotides in the published sequence coding for the amino acids (compensating for the deletions), we found 70 substituted and 8 deleted nucleotides (2.3% deviation). The nucleotide substitutions resulted in the stop codon and 46 amino acid changes, while 23 subsitutions would not have changed the amino acid.

DISCUSSION

The unexpected findings of Franciosa et al.(5) that 42 of 79 strains of Clostridium botulinum type A strains contained a ``silent'' or unexpressed type B neurotoxin gene raised some interesting questions regarding the origin, structure, and organization of the A and B toxin genes. In the present study, analyses of the genes of the toxin complexes by Southern hybridizations showed that the genome of type A(B) and Ab strains contains one copy of the type A toxin gene, one of the type B toxin gene, two copies of the NTNH gene, and one copy of the HA gene. It was also demonstrated that these genes are distributed into two toxin gene-based clusters, the BoNT/A toxin gene and one of the NTNH genes in one cluster and the remaining genes in a second cluster in the series HA-NTNH-BoNT/B toxin gene. Since the alignment of the genes of the second cluster is the same as the analogous genes in C. botulinum type A, B, and E reference strains(6) , this does not explain the silence of the type B toxin gene of A(B) strains.

A reasonable explanation for the silence of the type B toxin gene became evident when the nucleotide sequence of the type B toxin gene in strain 667 was determined and compared to that published for a proteolytic C. botulinum type B strain(15) . Although transcription studies are desirable to support the following interpretation, the most striking difference was the finding of a stop codon and two deletions in the gene of the A(B) strain. Although it is not known if the other A(B) strains also possess nucleotide changes that stop expression, this appears to be a likely explanation. The effects of the noted amino acid substitutions and deletions on biological activity if the toxin molecule had been expressed are not known.

The presence on a 60-kb (C. botulinum 667; A(B)) and an 80-kb (C. botulinum 588; Ab) restriction fragment of chromosomal DNAs, which hybridized with the probes for BoNT/A and/B toxin genes, lends the suggestion that less than 40 and 60 kb separate the A and B toxin gene clusters in the genome of those respective strains. However, the observation does not show the orientation of the gene clusters on the chromosome, information that could help in understanding why in strains that produce two toxin types, the toxicity of one of the types is significantly higher than the other. The toxicity ratio of the major to minor toxins is about 10 to 1 in a previously recognized Af strain (16) and about 10,000 to 1 for the Ab strain 588(5) . Such different toxicity levels could result if the molecules are expressed in different quantities, as might occur if the relative positions of the toxin gene clusters affect transcription and/or translation. The gene for the toxin of lower activity may be affected in transcription initiation or termination, or may produce a truncated protein with decreased toxicity. Alternatively, the quantities of toxin molecules synthesized may be similar, but their specific toxicities are different due to changes in protein structure or from postranslational modifications. Transcription and translation experiments would clearly be desirable to investigate the mechanism of low toxin production in the Ab and Af strains.

While the hemagglutinin genes HA-33 and HA-II are present in the type B neurotoxin gene cluster, they do not appear to be expressed, because no hemagglutinin activity was detectable in cultures of strain 667 or in cultures of any of the other four strains possessing both neurotoxin genes. This lack of hemagglutinin expression cannot be dependent on the defective nature of the type B neurotoxin gene, because there was no hemagglutinin expression detected in strain 588, which produces a low level of type B neurotoxin activity. While the HA genes are absent from the type A gene cluster, an unidentified ORF similar to an ORF previously reported only in nonproteolytic type E and F strains was found immediately upstream of the NTNH. Recent observations ()indicate its occurrence in infant C. botulinum type A strain Kyoto-F.

Although two NTNH genes (one type A and one chimeric) were identified in this study, it is not known if either or both were expressed since there is no known activity or immunologic assay for NTNH. The NTNH gene in the B cluster had a surprising chimeric composition; the nucleotide sequences of the two end regions indicate the gene's type B origin, while the sequence of the central region is more like that of the NTNH gene of a type A organism. The chimeric structure probably arose through recombination events by unknown mechanisms. The chimeric structure of the NTNH also presents an interesting possibility that the expression and structure of this protein may contribute to expression of the neurotoxin gene. Possibly the NTNH gene contains a strong promoter, whose transcription initiates and continues through to the toxin gene. Changes in the structure of this gene could lead to inefficient transcription of the polycistronic RNA encoding the NTNH and toxin proteins.

C. botulinum strains that contain the genes for two toxin types probably arise when a culture producing one toxin type acquires the genes encoding a toxin of a different antigenic type. This transfer possibly involves not only the neurotoxin gene but genes of the entire cluster. The mechanism of transfer is not known but would require a vector capable of transferring large regions of donor DNA. This transfer could occur by bacteriophages or conjugative transposable elements. Our findings confirm the possibility that the botulinal toxin genes are associated with transmissible vectors that can transfer toxicity to normally nontoxigenic clostridia such as Clostridium butyricum and Clostridium baratii, which have been isolated with increasing frequency during the last decade from humans suffering from botulism(1) .

FOOTNOTES

*

Research at the University of Wisconsin was sponsored by sponsors of the Food Research Institute and a Hatch grant. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked ``advertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank(TM)/EMBL Data Bank with accession number(s) X87848[GenBank]-X87850[GenBank].

§

To whom correspondence should be addressed: DBMD, NCID, MS C07, Centers for Disease Control and Prevention, 1600 Clifton Rd., N.E., Atlanta, GA 30333. Tel.: 404-639-3867; Fax: 404-639-3333; clh2{at}ciddbd2.em.cdc.gov.

()

The abbreviations used are: BoNT, botulinum neurotoxin; PCR, polymerase chain reaction; ORF, open reading frame; kb, kilobase(s); HA, hemagglutinin; NTNH, non-toxin nonhemagglutinin; PFGE, pulsed field gel electrophoresis.

()

A. K. East, personal communication.

ACKNOWLEDGEMENTS

REFERENCES

1. Hatheway, C. L. (1993) in Clostridium botulinum: Ecology and Control in Foods (Hauschild, A. H. W., and Dodds, K. L., eds) pp. 3-20, Marcel Dekker, Inc., New York
2. Tsuzuki, K., Kimura, K., Fujii, N., Yokosawa, N., Indoh, T., Murakami, T., and Oguma, K. (1990) Infect. Immun. 58, 3173-3177 [Abstract/Free Full Text]
3. Tsuzuki, K., Kimura, K., Fujii, N., Yokosawa, N., and Oguma, K. (1992) Biochem. Biophys. Res. Commun. 183, 1273-1279 [CrossRef][Medline] [Order article via Infotrieve]
4. East, A. K., Stacey, J. M., and Collins, M. D. (1994) System. Appl. Microbiol. 17, 306-312
5. Franciosa, G., Ferreira, J. L., and Hatheway, C. L. (1994) J. Clin. Microbiol. 32, 1911-1917 [Abstract/Free Full Text]
6. Cordoba, J. J., Collins, M. D., and East, A. K. (1995) System. Appl. Microbiol. 18, 13-22
7. Deleted in proof
8. Zhou, Y., Sugiyama, H., Nakano, H., and Johnson, E. A. (1995) Infect. Immun. 63, 2087-2091 [Abstract]
9. Lin, W. J., and Johnson, E. A. (1995) Appl. Environ. Microbiol. 61, 4441-4447 [Abstract]
10. East, A. K., Richardson, P. T., Allaway, D., Collins, M. D., Roberts, T. A., and Thompson, D. E. (1992) FEMS Microbiol. Lett. 96, 225-230 [CrossRef]
11. Somers, E., and DasGupta, B. R. (1991) J. Protein Chem. 10, 415-425 [CrossRef][Medline] [Order article via Infotrieve]
12. Fujii, N., Kimura, K., Yokosawa, N., Yashiki, T., Tsuzuki, K., and Oguma, K. (1993) J. Gen. Microbiol. 139, 79-86 [Abstract/Free Full Text]
13. Thompson, D. E., Brehm, J. K., Oultram, J. D., Swinfield, T. J., Shone, C. C., Atkinson, T., Melling, J., and Minton, N. P. (1990) Eur. J. Biochem. 189, 73-81 [Medline] [Order article via Infotrieve]
14. East, A. K., and Collins, M. D. (1994) Curr. Microbiol. 29, 69-77 [CrossRef][Medline] [Order article via Infotrieve]
15. Whelan, S. M., Elmore, M. J., Bodsworth, N. J., Brehm, J. K., Atkinson, T., and Minton, N. P. (1992) Appl. Environ. Microbiol. 58, 2345-2354 [Abstract/Free Full Text]
16. Giménez, D. F., and Ciccarelli, A. S. (1970) Zentralbl. Bakteriol. I Abt. Orig. A 215, 212-220

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. J. Jacobson, G. Lin, B. Raphael, J. Andreadis, and E. A. Johnson Analysis of Neurotoxin Cluster Genes in Clostridium botulinum Strains Producing Botulinum Neurotoxin Serotype A Subtypes Appl. Envir. Microbiol., May 1, 2008; 74(9): 2778 - 2786. [Abstract] [Full Text] [PDF]

				G. Franciosa, F. Floridi, A. Maugliani, and P. Aureli Differentiation of the Gene Clusters Encoding Botulinum Neurotoxin Type A Complexes in Clostridium botulinum Type A, Ab, and A(B) Strains Appl. Envir. Microbiol., December 1, 2004; 70(12): 7192 - 7199. [Abstract] [Full Text] [PDF]

				P. Fach, S. Perelle, F. Dilasser, J. Grout, C. Dargaignaratz, L. Botella, J.-M. Gourreau, F. Carlin, M. R. Popoff, and V. Broussolle Detection by PCR-Enzyme-Linked Immunosorbent Assay of Clostridium botulinum in Fish and Environmental Samples from a Coastal Area in Northern France Appl. Envir. Microbiol., December 1, 2002; 68(12): 5870 - 5876. [Abstract] [Full Text] [PDF]

				M. Dahlenborg, E. Borch, and P. Radstrom Development of a Combined Selection and Enrichment PCR Procedure for Clostridium botulinum Types B, E, and F and Its Use To Determine Prevalence in Fecal Samples from Slaughtered Pigs Appl. Envir. Microbiol., October 1, 2001; 67(10): 4781 - 4788. [Abstract] [Full Text]

				G. Schiavo, M. Matteoli, and C. Montecucco Neurotoxins Affecting Neuroexocytosis Physiol Rev, April 1, 2000; 80(2): 717 - 766. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hutson, R. A. Articles by Sugiyama, H. Search for Related Content PubMed PubMed Citation Articles by Hutson, R. A. Articles by Sugiyama, H. Social Bookmarking                      What's this?