Keynote Speakers and Titles of their Lectures

Topics of Symposium Sessions

Scientific Programme

Abstracts

Poster

Keynote Speakers and Titles of their Lectures

G. Gottschalk
Introductory lecture: The anaerobic world

E. Stackebrandt
Phylogenetic evidence for a taxonomic dissection of the genus Clostridium

S. Kozaki
Immunobiology of Clostridia: Dissection of antigenic structure of the neurotoxins produced by Clostridium botulinum and the related organisms

M. Popoff
Diagnosis of Clostridioses

S. Finegold
Treatment and prophylaxis of clostridial intoxications and infections in humans

Topics of Symposium Sessions

• Identification of Clostridia
• Immunobiology of Clostridia
• Diagnosis of Clostridioses
• Treatment and Prevention of Clostridioses

Scientific Programme

Abstracts

The anaerobic world

G. Gottschalk
Institut für Mikrobiologie und Genetik, Georg-August-Universität, Göttingen, Germany

Molecular oxygen in appreciable amounts is found only in those areas on earth that are in direct contact with air or are inhabited by organisms carrying out oxygenic photosynthesis. Large habitats on earth, rich in organic material, are devoid of molecular oxygen and a diverse and vast population of anaerobic organisms is around there. Among them are eukaryotes such as protozoa, organisms grazing on bacteria, and a great variety of bacteria and archaea. The latter form, the so called anaerobic food chain, which starts out with the degradation of macromolecules such as cellulose, starch, pectins, proteins and nucleic acids; bacteria and archaea specialized on a variety of electron acceptors carry out the terminal steps of the anaerobic food chain. There is a hierachy among them. Nitrate commes first, then ferric ions, sulfate and carbon dioxide. The importance of these reductive processes and the bacteria involved will be outlined. Then focus will be on the role of the clostridia in anaerobic habitats. Among the clostridia we find specialists such as the uric acid degrading Clostridium acidiurici or Clostridium kluyveri which grows on the combination of eathanol, acetate and CO₂, and most noteworthy a great variety of species are able to grow on macromolecules and therefore, secrete exoenzymes. This is the reason why clostridia exhibit an important ecological role in nature and why there is a small percentage of species which is pathogenic.

Session I: Identification of Clostridia

Phylogenetic evidence for a taxonomic dissection of the genus Clostridium

E. Stackebrandt, H. Hippe, I. Kramer and Jolantha Swiderski
DSMZ-German Collection of Microorganisms and Cell Cultures GmbH, Mascheroder Weg 1b, 38124 Braunschweig, Germany

For nearly 20 years classification of prokaryotes is undergoing major revisions. Based upon a rather stable phylogenetic backbone, many of the traditional taxa have been revised and new taxa described by a polyphasic approach. Today, traditional properties of assumed taxonomic weight have lost their dominating role in describing taxa at the levels of genera and families. One of the best examples of taxa, originally described because of the emphasis placed on some of such traditional properties are those which have been established on the basis of Gram staining behavior, shape, relationship to oxygen and spore formation - or combinations of one, two or all of the latter three properties.
Based upon 16S rDNA sequence analysis the phylogenetic heterogeneity has guided the taxonomic dissection of Bacillus and is presently the basis for a major restructuring of Clostridium. In the pioneering work of Collins et al. (199x) 21 phylogenetically separate clusters have been recognized within the Gram-positive bacteria with a DNA G+C content of less than 50 mol %. Many of these clusters contain species of the genus Clostridium, some of which have already been reclassified as members of novel genera, such as Synthrophospora, Caloramator, Sporohalobacter, Oxalophagus, Oxobacter, Moorella, Filifactor and Thermo-anaerobacter. The majority of Clostridium species are located in cluster 1, including the type species and many of the human and animal pathogenic representatives. Taxonomic descriptions of the recognized phylo-genetic Clostridium clusters other than cluster 1 will have to be reclassified in the future not only because of their moderate to distant relationships with the type species but because of the intermixing with organisms with non-Clostridium-type characters. Taxonomic progress, however, is hampered by the lack of phenetic properties that would allow the circumscription of these clusters.
We have recently completed the 16S rDNA database of all validly described and available Clostridium species, thus providing the molecular basis for a future reclassification strategy. Knowledge about the natural relatedness among the type strains will facilitate the search for phenetic properties needed to taxonomically delineate the molecularly defined clusters. Except for C. cylindrosporum that may be considered a representative of a new cluster, the other 21 species are members of the previously recognized cluster, with the majority of strains belonging into cluster 1.

Collins et al. (1994). J. Syst. Bacteriol. 44, 812-826.

Identification of Clostridium botulinum type E strains with the API 20 A, Rapid ID 32 A and RapID ANA II System

M. Lindström^*, H. Jankola, S. Hielm, E. Hyytiä and H. Korkeala
Faculty of Veterinary Medicine, Department of Food and Environmental Hygiene, Helsinki University, Finland

Three commercially available test systems for the identification of anaerobic bacteria, API 20 A, Rapid ID 32 A (bioMérieux SA, Marcy-l’Etoile, France) and RapID ANA II System (Innovative Diagnostic Systems, Inc., Norcross, GA, USA), were evaluated for their ability to identify Clostridium botulinum type E strains. Sixty strains were confirmed as type E by botulinum neurotoxin(BoNT)-specific PCR detection (Hielm et al., 1996), and 23 strains were presumed as C. botulinum by phenotypic characterization (colony morphology, a positive lipase reaction on egg yolk agar). The latter strains gave negative results with the PCR and mouse lethality tests (referred to as BoNT-negative strains).
Of the 60 type E strains, the API 20 A identified 53 (88%) strains to the species level and one strain to the genus level, and gave an unacceptable identification profile for six strains. The Rapid ID 32 A identified 59 (98%) strains as C. botulinum (group II) and one strain as Clostridium spp. The RapID ANA II System recognized all strains as C. botulinum (II).
With the API 20 A, 18 (78%) of the 23 BoNT-negative strains were classified as C. botulinum and four strains as Clostridium spp., with one strain given an unacceptable identification profile. The Rapid ID 32 A recognized all the BoNT-negative strains as C. botulinum. With the RapID ANA II System 21 (91%) strains were identified as group II C. botulinum and two (9%) strains were not identified at all.
The Rapid ID 32A and the RapID ANA II System recognized a majority of the BoNT/E-positive strains (59 and 60, respectively) as C. botulinum (II). However, they incorrectly classified most of the BoNT-negative strains into the same group which obviously decreases the reliability of the two test systems. On the other hand, although the API 20 A could not identify all the BoNT/E-positive strains as C. botulinum it did not give as many false positive results as the two other tests either. It seems that with the aid of the test systems based on phenotypic characterization, no distinction between C. botulinum strains and those strains not producing botulinum neurotoxin can be done.

Reference

Hielm S., E. Hyytiä, J. Ridell, and H. Korkeala, (1996). Detection of Clostridium botulinum in fish and environmental samples using polymerase chain reaction. Int.J.Food Microbiol. 31: 357-365.

Molecular typing methods for Clostridium botulinum

S. Hielm*, J. Björkroth, E. Hyytiä, M. Lindström and H. Korkeala
Department of Food and Environmental Hygiene, Faculty of Veterinary Medicine, Helsinki University, Helsinki, Finland

We have developed DNA-based protocols for the identification and typing of Clostridium botulinum species and strains. There is an obvious need for such methods, as the mouse lethality test-based serotyping scheme is neither accurate enough or ethically justifiable. Foods in a certain geographical region are commonly contaminated with a single C. botulinum serotype, which renders serotyping a useless approach when investigating botulism outbreaks. For the same reasons, it also lacks applicability in biodiversity studies.
Pulsed-field gel electrophoresis (PFGE) typing is generally accepted as the current ‘gold standard’ in the typing of bacteria. It has outstanding reproducibility and a very high discriminatory index, thus it has been applied to virtually all relevant bacterial species since its introduction a decade ago. However, as PFGE demands intact, whole DNA, the typing of clostridia has been effectively hindered by the DNases produced by many clostridial species, notably the non-proteolytic C. botulinum group II. We have recently described a protocol (Hielm et al. 1998) where this problem has been overcome to a large extent by formalin fixation of cells prior to DNA isolation. Using the restriction enzymes SmaI or XhoI, it is possible to distinguish most strains of C. botulinum from each other, making it an ideal tool for epidemiological studies. Although some species-specific fragments in the macrorestriction profiles would seem to exist, PFGE is clearly not a taxonomic nor phylogenetic method.
Ribotyping (Grimont and Grimont, 1986), on the other hand, promises to be better suited for taxonomy studies. Its discriminatory index for C. botulinum species is inferior to that of PFGE, but since its modus operandi is based on highly conservative genome structures, our results gained by ribotyping correlate very well with modern clostridial taxonomy. We have performed both manual and automated (Riboprinter, Qualicon Inc.) ribotyping, and both approaches are equally good for C. botulinum species identification. Although expensive to use, the Riboprinter gives highly reproducible fingerprints in only 8 h, making it an ideal identification tool when quick answers are needed.

References

Grimont, F. and P.A.D. Grimont, (1986). Ribosomal ribonucleic acid gene restriction as potential taxonomic tools. Ann.Inst.Pasteur/Microbiol. 137 B, 165-175.
Hielm, S., J. Björkroth, E. Hyytiä and H. Korkeala, (1998). Genomic analysis of Clostridium botulinum group II by pulsed-field gel electrophoresis. Appl.Env.Microbiol. 64:703-708.

Molecular methods for the analysis of Clostridium perfringens relevant to food hygiene.

Barbara Schalch*, Brigitte Sperner, H. Eisgruber and A. Stolle
Institute for Hygiene and Technology of Food of Animal Origin, Ludwig-Maximilians-Universität München, Tierärztliche Fakultät, Veterinärstraße 13, 80539 München, Germany

Clostridium (C.) perfringens continues to be a common cause of food-borne disease. C. perfringens produces an enterotoxin (CPE) which is released upon lysis of the vegetive cell during sporulation in the intestinal tract. Catering premises with insuffient cooling and reheating devices often seem to be the cause for outbreaks of C. perfringens food poisoning.
Typing of C. perfringens isof great importance for investigating sources of food poisoning cases and for studying the epidemiology of this microorganism. This report describes the examination of 172 C. perfringens isolates by molecular methods. Isolates were taken from 10 food poisoning outbreaks and cases (n = 34, food and fecal isolates) and from meat (n = 121). Isolates were characterized by plasmid profiling, ribotyping, and/or macrorestriction analysis by pulse-field gel electrophoresis (PFGE).
Results show that all three methods are suitable for classifying strains below the species level. Ribopatterns and PFGE patterns can be interpreted more easily than plasmid profiling results and were most useful for the epidemiologic investigation of foodborne disease outbreaks and cases caused by C. perfringens. In eight out of then outbreaks and cases the ribotype patterns of all food and stool isolates were identical. Therefore, it can be assumed that those isolates caused the disease. Among the food isolates studied, a broad variety of patterns was detected by ribotyping and PFGE.

Short protocol for pulsed-field gel electrophoresis of a variety of Clostridia species.

Brigitte Sperner*, Barbara Schalch, H. Eisgruber and A. Stolle
Institute for Hygiene and Technology of Food of Animal Origin, Ludwig-Maximilians-Universität München, Tierärztliche Fakultä, Veterinärstraße 13, 80539 München, Germany

While pulsed-field gel electrophoresis (PFGE) has become an important tool for genotyping of bacteria, one of its drawbacks is that standard methods are rather time-consuming. In order to overcome this problem shortened procedures for DNA preparation have been developed for some bacterial species. The aim of this study was to examine if a short procedure used by Hielm et al. (1998) for PFGE of Clostridium botulinum could be applied to other Clostridia species. For this the protocol was modified and used to prepare the DNA of 34 strains of 25 different Clostridia species. In contrast to a standard procedure (Maslow et al., 1993), which takes at least five days from DNA extraction to completion of the electrophoresis, this protocol yielded results within two days. In order to directly compare the results of the short protocol with those of the standard long procedure, parallel DNA preparations were performed using both methods and the two DNA samples thus obtained per strain were then run on the same gel.
Briefly, the procedure was as follows: After embedding the bacterial cells in agarose, the agarose blocks were incubated for 1 h in lysis solution containing lysozyme, mutanolysin, lysostaphin and RNase. This was followed by a 1 h Proteinase K treatment. Then, slices were cut from the agarose blocks and washed for 15 min in TE buffer; these washes were repeated four times with fresh TE. After a 2 h restriction with Smal, electrophoresis was carried out overnight.

Random amplified polymorphic DNA (RAPD) for fingerprinting of Clostridium botulinum strains: preliminary results

G. Franciosa*, S. Hodzic, L. Fenicia, P. Aureli
Food Microbiology Laboratory, Food Department, Istituto Superiore di Sanità, Rome, Italy

The genetic typing of food-borne bacterial pathogens has become a powerful tool for epidemiological as well as other research purposes. Nonetheless, so far a little interest has been spent in genotyping of Clostridium botulinum, perhaps because the characterization by specific antisera of the neurotoxin type produced is generally sufficient to correlate between clinical and food isolates from episodes of botulism. As a matter of fact, among the numerous existing different techniques of DNA fingerprinting only the pulsed- field gel electrophoresis (PFGE) has been applied to C. botulinum strains, with problems due to the DNAse activity of some strains. However, a better understanding of the ecology and biology of the pathogen is still needed; also, the more and more frequent implication of commercial foods in food-borne botulism requires a rapid identification of the spores source and the elucidation of the causes of the product contamination. Finally a reliable, fast and easy typing technique might help finding the still unrecognized sources of spores which account for the infectious forms of botulism (infant, infant-like, and wound botulism).
In the present study, we report for the first time the early results of the application of a recently developed PCR-based typing technique (random amplified polymorphic DNA, RAPD) for the genotyping of proteolytic strains of C. botulinum (group I), which are the most frequently isolated in Italy. A preliminary screening of 15 different short primers selected from the literature and used either for the typing of other Clostridia spp. or different gram-positive food-borne pathogens was performed with eight strains of C. botulinum producing all different neurotoxins. Two primers, OPG-5 and ARP, were chosen for the experiments since they discriminated well between the strains with a good number of bands. Fifty unrelated proteolytic strains were then tested (24 type A; 24 type B and 2 type F) in order to define the capability of the primers to separate inside the group. Finally, some epidemiologically related A and B strains from cases of foodborne botulism occurred in our Country were also assayed. Our results show that the technique is easy to perform, rapid, and efficient in discriminating strains and can be successfully used for analyzing a large number of isolates.

Utilisation of PCR (Polymerase Chain Reaction) for typing of Clostridium perfringens and for determination of its enterotoxin

B. Kadra *(1), J.P. Guillou (2), M. Popoff (3)
(1) Sanofi Santé Nutrition Animale, Z. I. La Ballastière BP 126, 33501- Libourne, France
(2) CNEVA - LCRV, 22, rue Pierre Curie, BP. 67, 94703 Maison - Alfort, France
(3) Institut Pasteur, 28, rue du Docteur Roux, 75724 Paris, France

A simple procedure was developed to identify toxitypes of C. perfringens in microbial cultures.
90 strains of C. perfringens were identified by classical bacteriological methods, typing of the strains was made by a seroneutralization test on mice (Table 1).

Production of enterotoxin was revealed by the following tests:

• Reverse passive latex agglutination test
• Cytopathogenicity test on Vero cells

All the strains were analysed by PCR using:

• gene of toxin alpha
• gene of toxin epsilon
• gene of toxin beta
• gene of enterotoxin

We proceeded by simple amplification (amplifying one gene), duplex and triplex (aiplifying 2 and 3 genes simultaneously). Amplified DNA was subjected to gel electrophoresis in 2% agarose in TBE after ethidium bromide staining. Products of PCR are revealed by enzymatic restriction and by hybridization with DIG-labelled oligonucleotide probes.
In the conditions of the experimentation, the PCR method has proved performant. The specificity and sensitivity is excellent and superior to the standard techniques.
Confirmation of the results by oligonucleotide hybridization is an essential part of the PCR technique.

Comparison of Polymerase Chain Reaction (PCR) with seroneutralisation bio assay and ELISA for typing sheep clinical isolates of Clostridium perfringens.

S. El Jai, A.H. El Idrissi*
Département de Microbiologie et Maladies Contagieuses, Institut Agronomique et Vétérinaire Hassan II, B.P. 6202, Rabat-Instituts, Morocco

Typing of Clostridium perfringens into toxigenic types (A through E) is based on the production of the majors toxins (alpha, beta, epsilon and iota). This typing is routinely performed by the conventional seroneutralisation test in mice or guinea pigs. The ELISA method also has been extensively used to assay the toxigenicity of Clostridium perfringens. In order to compare these serological assays with the recent gene detection assays as an alternative method for typing Clostridium perfringens, 56 strains isolated from sheep suspected of having enterotoxemia were typed by a multiplex PCR and the results were compared with seroneutralisation in mice and ELISA. The PCR results agreed with both serological tests in 46 strains (82%). However, the 10 remaining strains were revealed either non typable (no toxin produced, 7 strains) or mistyped as type A (only alpha toxin is produced) by both serological assays while they were all genotyped as type D (presence of epsilon toxin gene). The non or low expression of the epsilon toxin despite the presence of its gene in these strains was further studied by SDS-PAGE and northern blot analysis. The finding of unexpressed toxin d nes in clinical isolates suggests that PCR could be a powerful test to detect non typable isolates such as potentially toxigenic strains and low toxin producing strains.which usually result in typing problems for routine laboratories

Isolation of Sarcina ventriculi, Clostridium fallax and Clostridium sordellii from lambs with abomasal disorders.

Synnøve Vatn
Department of Sheep and Goat Research, Norwegian College of Veterinary Medicine, Sandnes, Norway

An abomasal syndrome with a clinic of acute abdominal distension, colic or sudden death occurs in Norwegian lambs aged 2-5 weeks. Some farms report up to 30% prevalence. Abomasal post mortem findings vary, and ranges from simple tympany, sometimes combined with hemorrhage and/or ulcers to hemorrhage and ulcers without gaseous accumulation. In most lambs with acute abomasal tympany a Sarcina-like organism was detected on direct smears or in tissue sections. Isolation was only successful in one case and the isolate was identified as Sarcina ventriculi. In less than half of the cases Clostridium fallax was isolated as well. In lambs with more pronounced mucosal damage, S ventriculi was found in more than half of the lambs. Clostridium sordellii was also isolated from some of these lambs, sometimes together with the Sarcina-like bacteria. None of these microbes were isolated or detected in control lambs. S ventriculi has earlier been associated with abomasal disorders in neonatal calves (1). Sarcina-like bacteria have been seen histopathologically in goatkids with fatal abomasal bloat (2). C fallax has not earlier been reported from animals with abomasal disorders. C sordellii is reported from Great Britain as causing sudden death in sheep of all ages, the youngest with abomasal lesions (3).
S ventriculi is probably the main causal micro-organism of abomasal disorders in Norwegian lambs, sometimes in synergism with C fallax or C sordellii. Sarcina ventriculi is an anaerobic gram positive coccus, that forms bundle shaped structures. It is extremely pH tolerant and is reported to grow in media ranging from pH 1 to 9,8. According to Willems and Collins (4) the phylogenetic placement of S ventriculi should be within Group 1 Clostridium. The cultivation of S ventriculi has proven to be difficult. So far the evaluation of direct smears by phase contrast microscopy has been the most sensitive diagnostic method.

References

Mills K.W., A. Boerger-Fields, M.M. McAllister, (1994). Sarcina ventriculi: Isolation and identification in abomasitis and bloat of neonatal calves. (Abstract)
DeBey B.M, P.C. Blanchard, P.T. Durfee, (1996). Abomasal bloat associated with Sarcina-like bacteria in goat kids. J.Am.Vet Med Ass;209:1468-1469.
Lewis C.J, R. Naylor, (1998). Sudden death in sheep associated with Clostridium sordellii. Vet Rec;142:417-421.
Willems A, M.D. Collins, (1994). Phylogenetic placement of Sarcina ventriculi and Sarcina maxima within group I Clostridium, a possible problem for future revision of the genus Clostridium. Request for an opinion. Int.J.Syst Bacteriol;44:591-593.

Session II: Immunobiology of Clostridia

Immunobiology of Clostridia: Dissection of antigenic structure of the neurotoxins produced by Clostridium botulinum and the related organisms

Shunji Kozaki*,¹⁾ Seijiro Kawaguchi,²⁾ and Motohide Takahashi ^3)
1)Department of Veterinary Science, College of Agriculture, Osaka Prefecture University, ²⁾Chiba Serum Institute, and ³⁾Department of Bacterial and Blood products, National Institute of Infectious Diseases, Japan

Members of genus Clostridium are distributed widely in nature and found in soil as well as in fresh water and marine sediments in the world. Limited number of species are encountered as agents of infection in properly collected human and animal clinical specimens. Although most strains can be identified by the determination of their morphological and cultural characteristics, some species are classified into immunological types of the toxins produced. In the past, the use of antibody against toxin has been limited to typing, and, in some case, therapy of patient. However, the toxin has been highly purified, and polyclonal antibody is available as a reagent that reacts specifically with the toxin. Moreover, techniques have been developed for production of monoclonal antibodies. During the last decade, we have been characterizing the antigenic structures of botulinum neurotoxins by use of polyclonal and monoclonal antibodies. It is well known that the neurotoxin is nicked by a certain proteases like trypsin, resulting in transformation to a di-chain toxin molecule, made up of a light and a heavy chain. In a series of studies, we found that the neurotoxin is composed of at least three antigenically and functionally different domains, the light chain, the amino-terminal and the carboxyl-terminal halves of the heavy chain. By use of the monoclonal antibodies, we scrutinized the antigenic properties of the neurotoxin produced by the isolates associated with infant botulism. C. butyricum neurotoxin is antigenically similar but not identical to type E neurotoxin and such immunological difference is not attributable to a particular portion of the toxin molecule. Type A neurotoxin of isolates implicated in infant botulism in Japan also shows immunological dissimilarity to the authentic neurotoxin. However, our results indicate that the distinguishable characteristics are ascribable to the heavy chain but not the light chain. Recently, we encountered a low toxic neurotoxin associated with type B infant botulism. In this neurotoxin, the carboxyl-terminal half of the heavy chain differs antigenically from that of the authentic neurotoxin, causing a decreased capacity of binding to the receptor. From the foregoing knowledge, analysis of the antigenic structure of the neurotoxins will not only provide valuable information on their immunological diversity, but also contribute to revealing on the structure-function relationship of the neurotoxin.

Characterisation of a microtitre based endopeptidase assay for botulinum neurotoxin A and E utilising synthetic SNAP-25.

D. Sesardic*, K. McLellan, H.C. Martin and T.A.N. Ekong.
Division of Bacteriology, National Institute for Biological Standards and Control, South Mimms, Potters Bar, Hertfordshire EN6 3QG, UK.

Botulinum neurotoxins A (BoNT/A) and E (BoNT/E) are metalloproteases with a unique specificity for SNAP-25, a synaptosomal membrane associated protein which is an essential component of the neuroexocytotic machinery. Whereas BoNT/A cleaves SNAP-25 at bond Q₁₉₇-R₁₉₈, BoNT/E cleaves the same protein at position R₁₈₀-I₁₈₁. Microtitre based assay monitoring zinc-dependent endopeptidase activity for both toxins has been developed using a synthetic peptide of SNAP-25 (SNAP25_137-206) and targeted antibodies selective for either BoNT/A or BoNT/E cleavage site on SNAP-25. Targeted antibodies were prepared by immunisation of rabbits with KLH coupled to N-terminal peptide of SNAP-25_190-197 or SNAP-25_173-180 for detection of BoNT/A and BoNT/E proteolysis, respectively. Each antibody was specific for the C-terminus of the corresponding toxin cleaved substrate and did not cross-react with the native SNAP-25 peptide on immunoassay. That each neurotoxin cleaves synthetic SNAP-25 peptide at the position reported for intact synaptosomal protein was confirmed by molecular weight analysis of the cleavage products, separated by HPLC, which corresponded to those calculated theoretically for both toxins. Specificity of the endopeptidase reaction was further confirmed by the inhibitory profile of known inhibitors of zinc-dependent endopeptidases; zinc ions (1mM) and EDTA (10mM). BoNT/A endopeptidase activity was completely inhibited by addition of BoNT/E, as this toxin cleaves SNAP-25 downstream from the BoNT/A toxin cleavage site. Neither reaction was inhibited by other bacterial toxins studied (BoNT/B, BoNT/F, tetanus or diphtheria). BoNT/A activity was inhibited >95% by polyclonal antiserum to type A toxoid and a monoclonal antibody to H₂L fragment of BoNT/A but other botulinum toxin serotype specific antibodies were non-inhibitory. Likewise BoNT/E activity was inhibited (>95%) only by a polyclonal antibody to BoNT/E toxoid.
Using optimum conditions for BoNT/A and BoNT/E activity, both assays were more sensitive than the mouse bioassay (limit of detection 0.3-0.5 mouse LD₅₀/ml). Although use of such assays will require extensive validation the endopeptidase assay for potency testing of therapeutic formulations containing BoNT/A looks promising. We have demonstrated an excellent agreement between the mouse and endopeptidase assays for 23 different batches of therapeutic formulations (r=0.95, slope=1.03).

HPLC for monitoring botulinum neurotoxin types A and E endopeptidase activity

T.A.N. Ekong, P.H. Corran², C. Gee³, and D.Sesardic*^1
1Division of Bacteriology, ²Endocrinology and ³Laboratory for Molecular Structure. National Institute for Biological Standards and Control, Potters Bar, Hertfordshire, EN6, 3QG, UK.

The role of botulinum neurotoxin types A and E (BoNT/A and BoNT/E) as extremely potent food poisons, and the emerging use of BoNT/A as a therapeutic tool in the management of a growing number of neuromuscular conditions, has created the need for sensitive, specific and biologically relevant assays for these toxins. Recent studies have shown that these toxins are zinc endopeptidases which act by cleaving the neuronal protein SNAP-25 at distinct sites.
We have used a synthetic peptide of SNAP-25 spanning the cleavage site of both toxins (residues 137 to 206) as an in vitro substrate and reverse-phase HPLC (RP-HPLC) to evaluate the proteolytic activity of these toxins. Toxin activity was measured by monitoring the peak areas of products following RP-HPLC of cleavage mixtures. Reaction specificity was determined by electrospray-mass spectrometry (ES-MS).
The peptide substrate was cleaved by both toxins at the expected peptide bonds (Q₁₉₈-R₁₉₈ for BoNT/A and R₁₈₀-I₁₈₁ for BoNT/E). Proteolytic activity was completely inhibited by a divalent metal ion chelators (10 mM EDTA), 1 mM ZnCl₂ and by 10 µg/ml of the respective specific antitoxin. Neither toxin was antagonized by captopril (10 mM), phosphoramidon (1 mM) and O-phenanthroline (1 mM), other clostridial toxin serotypes (BoNT/ B and BoNT/F and TeNT) or their antitoxins. Although both toxins were activated by up to 3 to 6 - fold by bovine serum albumin, a general activator of metalloproteases, they were distinguishable by their sensitivities to other cleavage conditions, including pH, ionic strength and DTT concentration. Under optimum conditions of cleavage, a minimum of 0.5 ng of pure BoNT/A and 50 ng of pure BoNT/E, equivalent to 100 and 200 MLD50/ml respectively, could be detected.
These results demonstrate that the synthetic peptide substrate was effectively cleaved by BoNT/A and BoNT/E and that RP-HPLC is suitable for assessing their endopeptidase activity.

Detection of etx gene (epsilon toxin inducer) in plasmids of high molecular weight in Clostridium perfringens type D.

Adriana Bentancor*, M. Rodriguez Fermepin, L.D. Bentancor and R.A. de Torres
Catedra de Microbiologia Facultad de Ciencias Veterinarias y Catedra de Microbiologia, Facultad de Farmacia y Bioquimica. Universidad de Buenos Aires, Argentina

There is evidence of an important epizootiological prevalence of infections in animals by Clostridium perfringens type D in Argentine.
It is of special interest to study a set of strains in order to select strains high and persistent epsilon toxin producers.
A group of 19 strains, from different origin were previously classified as Clostridium perfringens. All strains were retyped using thirty conventional taxonomic criteria and the presence of the etx gene was explored by polimerase chain reaction (PCR) using the primers:

• TAC TCA TAC TGT GGG AAC TTC GAT ACA AGC
• CTC ATC TCC CAT AAC TGC ACT ATA ATT TCC

All strains were positive for the presence of the etx gene by this technique. Nine strains were selected for a more efficient production of epsilon toxin, measure as DL50 by IP challenge in mice (WHO standard methods).
Plasmids of low molecular weight were isolated from each one of these nine strains and in all cases the etx gene was not detected by PCR.
However, the etx gene was detected by PCR in plasmids of high molecular weight isolated by lisis in situ.
The presence of the etx gene in plasmids of high molecular weight was confirmed by Southern Blot, using a chemioluminiscent detection develpmed by a UTP-digoxigenine labelled probe. The localisation of the gene was in more than one plasmid of at least one strain.
We believe that this association of the etx gene with plasmids of high molecular weight is correlated with the levels of production of the epsilon toxin and with the stability of epsilon toxin expression during several passages.

Detection of Clostridium novyi type B alpha toxin using cell culture systems

E. Borrmann, F. Schulze
Federal Institute for Health Protection of Consumers and Veterinary Medicine, Division 4, Jena, Germany

Clostridium novyi type B is the causative agent of the infectious necrotic hepatitis (black disease) especially in sheep and also in other animals. The production of the alpha toxin by these bacterium results in severe losses. An early prophylactic immunisation with vaccines containing C. novyi alpha toxoid as protective antigen is recommended in endangered areas.
According to the European Pharmacopoeia the potency of clostridial vaccines are measured in vivo using the mouse neutralisation assay. The neutralisation test is time-consuming, cumbersome and inconvenient and causes severe distress and suffering of animals. The aim of our study was to investigate if a cell culture system can replace the toxin neutralisation test in mice for the potency testing of vaccines.
The alpha toxin produced by type B is a classical lethal exotoxin. This toxin with a molecular mass of approximately 250 kDa belongs to the family of large clostridial toxins. The toxin is cytotoxic and causes irreversible changes in cell shape based on the action on the actin cytoskeleton. This was the basis for the development of our cell culture assay. Nine permanent cell lines were examined for their reaction to the alpha toxin. The cytotoxic effect of the toxin was determined after three days by microscopic examination and MTT assay. The toxin exhibited the strongest effect on the ESH-L cells. We were able to show that the cytotoxic effect was neutralised by the international standard for gas gangrene antitoxin (C.novyi) but never by heterologous antisera. Our results showed that the ESH-L cell line was a suitable indicator for the detection of the cytotoxic effect of alpha toxin.

Session III: Diagnosis of Clostridioses

Diagnosis of Clostridioses

M.R. Popoff
Toxines Microbiennes, Institut Pasteur, Paris

Clostridium are widespread in the environment, and some species cause disease in human and animals. These bacteria can enter organisms either by oral route or disruption of tegument integrity, and thereby induce gastrointestinal diseases and gangrene respectively. Clostridium multiply in the site of infection and produce toxins and hydrolytic enzymes which are responsible for the symptoms and lesions. Three elements are useful for the diagnosis of clostridial infections: clinical observations, toxin detection, and evidence of toxigenic Clostridium.
The symptomatology is specific in certain diseases such as those due to neurotoxigenic Clostridium. botulism and tetanus. Evocative lesions in gastrointestinal diseases and gangrene are: necrotic tissues, presence of gas, particular small ...
Toxin assays include biological methods (mouse lethality, cytotoxicity, ...), immunological methos (precipitation, agglutination, ELISA ...) and biochemical methods based on the toxin enzymatic activity. Toxin detection is valuable to identify the botulism and certain gastrointenstinal diseases (C. difficile colitis, C. perfringens food poisoning, enterotoxemia).
Identification of toxigenic Clostridium is usually performed by conventional methos. DNA based techniques allow a rapid toxin typing of these microorganisms. A toxigenic Clostridium evidenced in muscle with gangrene lesions, is generally considered as the caustive agent. In the gastrointestinal disease, the presence of toxigenic Clostridium in the intesinal content, which is the primary site of infection, supports the diagnosis of clostridiose if the Clostridium count is high (>10⁶/g) and/or accompanied by a production of toxin

PCR ribotyping of Clostridium difficile. Experiences of a national reference laboratory and investigations into a possible clone of C. difficile in hospitals throughout England and Wales.

J.S. Brazier*, S.L.J. Stubbs and B.I. Duerden
PHLS Anaerobe Reference Unit, Public Health Laboratory, University Hospital of Wales, Cardiff, UK.

The Anaerobe Reference Unit of the Public Health Laboratory Service of England and Wales provides a typing service to hospitals wishing to investigate putative outbreaks of Clostridium difficile diarrhoea. In the three year period of March 1995 to April 1998, a total of 1,244 isolates from 33 hospitals throughout England and Wales have been typed by the modified PCR ribotyping method of O’Neill et al.. This method is based on the variability that exists in the multiple copies of the 16S-23S rRNA genes of C.difficile; it has, to date, discriminated over 120 different PCR ribotypes among strains analysed from a wide variety of origins. Of the 1,244 isolates originating from symptomatic hospitalised patients or their ward environment, a total of 56 different PCR ribotypes were identified. However, just 16 PCR ribotypes accounted for over 90% of strains. One type, PCR ribotype 1 was present in 97% (32/33) of hospitals that submitted isolates. This type accounted for 68% of the total number of hospital patient isolates examined and is the type causing the majority of C.difficile infections in 24 (72%) of the 33 hospitals. The next most common type, type 106 accounted for only 7.2% of hospital patient isolates. Investigations were undertaken to attempt to discriminate between 40 type 1 isolates originating from 22 different hospitals by the ARDRA (amplified ribosomal DNA restriction analysis) method with 11 different restriction enzymes. This analysis revealed no differences in the amplified 16S-23S intergenic spacer region, indicating a very high level of homogeneity within this ribotype. Further molecular investigations are in progress that may either provide evidence that PCR ribotype 1 is a single strain of C.difficile which has become predominant in hospitals across England and Wales, or enable sub-typing within this strain which will enhance our understanding of its epidemiology. The results of these investigations will be presented at the conference.

Development and evaluation of various enzyme linked immunosorbent assays for the detection of Clostridium perfringens beta antitoxins

B. Krt
Institute for Microbiology and Parasitology, Veterinary faculty, University of Ljubljana, Ljubljana, Slovenia

The aim of our work was to develop an enzyme linked for the detection of antibodies against the Clostridium perfringens beta toxin. For this purpose five different ways of performing of ELISA were investigated. Positive and negative sera of different animals and partially purified beta toxin were used. In all ELISA tests microplates were first coated with monoclonal antibodies against the Clostridium perfringens beta toxin. Actually the first three ways of performing of ELISA proved to be an inhibition or a block ELISA. In the first of these modifications the examined serum was added on a microplate after the toxin. In the second two tests they were added simultaneously after they had been incubated together (60 minutes at room temperature or over night at 4 ° C). An antitoxin conjugate was used for the detection. It was also used in competitive ELISA, where it was added together with the examined serum on the microplate, to which the toxin had been already bound. The fifth way of performing of ELISA differed from others by the use of conjugated antispecies immunoglobulin for the detection.
The biggest differences in absorbances between positive and negative sera were found in the block ELISA, where the mixture of the toxin and the examined serum had been previously incubated over night at 4 ° C. The smallest differences in absorbances were found when antispecies conjugates were used.

Detection of the ß2 toxin gene of Clostridium perfringens in diarrhoeic piglets in The Netherlands

H.L.B.M. Klaasen^*,1, M.J.C.H. Molkenboer¹, J.F. van den Bosch¹, J. Bakker², J. Frey³ and M.R. Popoff^4
1Bacteriological R & D, Intervet International BV, Boxmeer, The Netherlands;
²Animal Health Service, Boxtel, The Netherlands;
³Institute of Veterinary Bacteriology, University of Bern, Switzerland; ⁴Unité des Toxines Microbiennes, Institut Pasteur, Paris, France

In the early nineties, a growing number of piglets (aged 0-2 wks) with diarrhoea or acute death were submitted for necropsy at the Animal Health Service in Boxtel, The Netherlands. In a significant proportion of animals, necrotic enteritis was found in association with C. perfringens apparently of type A, and not type C. The present study was initially done to determine the occurrence of C. perfringens type A in piglets suspected of necrotic enteritis. For this purpose, 44 piglets from 22 farms, not older than 2 wks and presenting with diarrhoea or enteritis, were selected, and complete post-mortem examination was carried out. The toxin type of each C. perfringens isolate was determined with 4 different PCRs using primer sets specific for the a , ß, e and enterotoxin gene, respectively (Buogo et al., J Vet Med B 1995, 42, 51-58). Since a novel toxin - ß2 toxin - and its gene were recently described by Gibert et al. (Gene 1997, 203, 65-73), and associated with necrotic enteritis in piglets, also a PCR was done to detect the ß2 toxin gene. The ß toxin is henceforth designated ß1 toxin, and the ß1 and ß2 toxin genes cpb1 and cpb2.
The numbers of farms with either diarrhoea or acute death were 14 and 7 out of 22 farms, respectively. For necrotic enteritis and other pathology, these numbers were 15/21 and 5/21, respectively. C. perfringens was isolated from piglets from 21/22 farms. According to the PCR results, all isolates were positive for the ß toxin gene, which confirmed the identification as C. per-fringens. Furthermore, none of the isolates possessed the e toxin gene or enterotoxin gene. From 14 farms with diarrhoea, none were positive for cpb1 only, 5 were positive for cpb1 + cpb2, and – most importantly – 8 for cpb2 only. Alternatively, from 7 farms with acute death, 3 farms had cpb1 only, 2 had cpb1 + cpb2, and one farm had cpb2 only. Regarding fibrinous/necrotic enteritis, the numbers of positive farms were: only cpb1, 4 farms; cpb1 + cpb2, 5 farms; and only cpb2, 4 farms. In conclusion, the prevalence of these two genes on farms with severe enteritis was similar, but in diarrhoeic piglets cpb2 was detected at a significantly higher frequency than cpb1. These results suggest a certain role of cpb2 and the ß2 toxin in the pathogenesis of diarrhoeal disease in young piglets.

Experimental necrotic enteritis of the fowl: requirements of disease induction and diagnostic methods.

M. Kaldhusdal*¹, M. Hofshagen¹, A. Løvland¹ and Keith Redhead^2
1National Veterinary Institute, Oslo, Norway
²Hoechst Roussel Vet, Milton Keynes, UK

Necrotic enteritis of the fowl (NE) is caused by Clostridium perfringens types A (CPA) or C. The intestinal lesions of NE have been produced by intraduodenal infusion of washed CPA organisms and/or CPA crude toxins. However, infection per se is insufficient for the field type disease to occur. Predisposing factors are numerous, but many are ill-defined and experimental results have been contradictory. Consequently, NE is considered difficult to reproduce. This lack of reliable experimental models hampers efforts to evaluate the predisposing and preventive effect of putative risk factors and preventive measures. The objective of this work was to evaluate the significance of some factors in the reproduction of NE.
Trials were conducted with broiler chickens raised on litter. Disease production was attempted by giving 14 to 19 days old birds feed mixed with CPA-inoculated broth. Diagnostic methods included post mortem examination of dead birds, examination of the intestinal mucosa of randomly selected, euthanatized birds, and quantitative examination for Clostridium perfringens (CP) of caecal contents or freshly voided, urate-capped faeces.
Two CPA strains were tested. Inoculation of the most potent a -toxin producing strain (as measured in vitro) was associated with increased CP counts and increased frequencies of NE-associated intestinal lesions.
In order to control the proliferation of wild-type CP before inoculation, the effect of supplementing the pre-inoculation feed with narasin and bacitracin was tested. These compounds kept CP numbers significantly lower than in birds offered non-supplemented feed, but there was no distinct difference in frequencies of intestinal NE lesions between the groups after withdrawal of antibiotics and inoculation with CPA.
The administration of inoculated feed on just one day was compared with giving inoculated feed on four days. No marked difference between the groups was detected with regard to CP numbers or intestinal NE lesions.
The associations between administration of inoculated feed, and CP numbers and NE lesions were tested in 2 experiments. In a pen trial CP numbers were 3 to 4 logs higher in inoculated birds on day 7 post inoculation, and a concurrent NE-specific mortality was detected in this group. In a cage experiment the CP numbers were moderately higher in the inoculated birds, but no clear difference in frequencies of NE lesions was found.

The development of novel assays for the detection of botulinum toxins

M. Wictome *, K. Newton, K. Jameson, K. Foster and C. C. Shone
Centre for Applied Microbiology and Research, Porton Down, Salisbury. SP4 OJG, UK

Food-borne botulism, characterised by flaccid paralysis, is a result of ingestion of pre-formed neurotoxin and is often fatal. Currently the only accepted method for the detection of toxin in samples is the mouse bioassay. Although highly sensitive this test has a number of drawbacks: it is expensive to perform; lacks specificity and involves the use of animals. With increasing resistance to such animal tests there is a need to replace the bioassay with a reliable in vitro test.
Over the past three years it has been demonstrated that all the botulinum neurotoxins act intracellularly as highly specific zinc endoproteases, cleaving proteins involved in the control of secretion of neurotransmitter. In the work described, this endopeptidase activity has been utilised in assay formats for the detection in foods of neurotoxin from the serotypes involved in food-borne outbreaks in man. The assays have a greater sensitivity, speed and specificity than the mouse bioassay, and it is envisaged that they will prove realistic alternatives to such animal based tests.

Development of an in vitro assay for the detection of Botulinum Neurotoxins - preliminary results.

F. Gessler, S. Behrens, P. Loch and H. Böhnel
Institute for Crop and Animal Production in the Tropics, Georg-August-Universität Göttingen, Germany

Botulism is a severe, often lethal intoxication, which affects man and animal. It is caused by the neurotoxins of Clostridium botulinum (BoNT), an anaerobic spore-former. At present the toxin types A-G have been identified. Types C and D are of vetererinary importance, types A, B and E are of special interest in human medicine. Due to the very high biologic activity of the BoNT, the toxin still needs to be detected in the mouse assay. Based on the investigations presented, a serological test shall be developed, which concentrates the sample and detects the toxin in one step.
The C. botulinum strains, received from culture collections have been intensively investigated. Considering these results, especially the toxicity of the isolates, one strain out of each group (toxin type A-E) was chosen for the development of the serological toxin detection test. Starting with type D the fermentation parameters in the "Goettinger bioreactor" had been optimized and toxin was produced in batch cultures. The toxin was purified using filtration with hollow fiber membranes and three chromatographic steps, including HIC and IEX methods. The purity of the toxin preparation was checked electrophoretically in SDS-PAGE and chromatographically in SEC.
The purified substance was treated with 0,4 % formalin. The toxoid was applied to rabbits and goats for antibody production. From both animal species high titer antisera could be obtained. Titres have been measured using an ELISA.
Besides the polyclonal sera monoclonal antibodies have been developed. The primary clones are about to be tested.
Polyclonal antibodies from goats were purified using ion exchange and affinity chromatography. The purified antibodies were bound to different carrier materials in the test tubes. In preliminary tests approx. 1 ng toxoid could be detected in 750 µL sample volume. Since the sample volume, which can be applied in the assay is nearly unlimited, at least as far as the test is concerned, one mouse lethal dose can be detected, if 10 mL of the sample are used in the test. However, the assay needs to be optimized modifying the detection and colour reaction parameters. A sensitivity of about 10-100 pg BoNT should be achieved. Specifity testing will follow.

Research efforts to identify the causative agent of an outbreak of bovine paraplegic mortality in the eastern plains of Colombia.

E. Benavides*, D. Ortiz, D. Duque, C. Estupinan, R. Altuzarra, and J. Benavides.
Veterinary Epidemiology Program, Corporacion Colombiana de Investigacion Agropecuaria, CORPOICA, Santafe de Botota, Colombia

The eastern plains of Colombia (Llanos Orientales, Orinoquia) are extensive savannas, around 17 millions of hectares (ha) extended from the Andean chains to Venezuela, characterized by their acidic soils (Oxisoils, high aluminium saturation). The "Altillanura plana" is a well drained subregion of 3.5 millions of ha, 2500 mm annual rainfall and the most limiting climatic and edafic conditions of the region, where extractive extensive ranching cattle production systems predominate. In the Altillanura, since 1994 a dramatic increase in the cases of mortality of adult cattel (mainly cows on milk) has been observed, disease characterized by neuroparalytic symptoms. Botulism has been suggested as the cause of the mortality due to the association of phosphorus deficiency in the region and the evidence of bone chewing by the cows. This paper summarizes the research efforts performed in order to confirm such diagnosis; due to the lack of previous expertise on the subject, a basic Pasterian approach has been used, improving protocols and methods with time. A first step was to demonstrate toxic activity (production of typical symptoms -wasp waist- in mice after injection) of liver or ruminal and intentinal contents, taken from affected animals; thermolobility (inactivation of the toxin by heating at 100°C), and induction of toxicity by trypsin treatment were also evaluated. Afterwards, specific antitoxins were acquired from CDC and seroneutralization protocols are being applied on samples maintained under deep freeze temperatures. By now, 43 cases (samples collected from diverse animals in a herd of a specific date) have been examined (total 90 samples) of which 16 cases, have shown toxicity with thermolability features and typical symptoms. Seroneutralization tests in mice have been completed on four cases, demonstrating the activity of C and D botulism toxins on a case, but in the other three cases, a toxic thermolabile agent, not neutralized by known botulism antitoxins. On the other hand, anaerobic culture was conducted on 34 soil samples from 8 affected herds, testing supernatiants for toxicity in mice and using further isolation and biochemical tests on sediments. The biological test showed the presence of toxicity on superantants from 32 soils, of which 21 were thermolabile, besides this, at least 203 isolates of sporulated, anaerobic, catalase negative rods were obtained. These have been tentatively diagnosed as Clostridium bifermentans (86), Clostridium novyi (88), Clostridium histolyticum (3) and Clostridium botulinum (17). Toxic soils are under evaluation of type of toxin incriminated by mouse senoneutralization test.

Occurrence and distribution of Clostridium botulinum type C and D spores in buffalo breeding areas of the lowlands of the State of Maranhao, Brazil

T.M.D. Silva, I.S. Dutra*, R. Nonato and J. Doebereiner
Department of Production and Animal Health, Unesp-Campus de Aracatuba, SP 16015-050, Brazil

Botulism is enzootic in buffaloes raised in the Baixada Maranhense region, the lowlands of the State of Maranhão, Brazil, near the Atlantic coast. This paper deals with the occurrence of spores of Clostridium botulinum type C and D and their distribution in soil, mud and buffalo faeces. Sampling was done by chance in temporarily flooded area where buffaloes were mantained, in 4 municipalities of the Baixada Maranhense. Thirty five samples were soil, 65 mud and 40 were faeces of buffaloes. The presence of spores was detected by culturing the samples in CMM and by subsequent inoculation of filtered culture supernatants in mice. Typification of positive samples was done by micro-complement fixation. In 104 (74.28%) of the 140 samples examined C. botulinum was encountered. There were no significant differences (P> 0.05) regarding the frequency of samples positive for C. botulinum spores between soil (77.1%), mud (60.0%) and faeces (95.0%). A total of 28 positive samples were used for typification: Four (14.29%) were type C, 23 (82.14%) type D and one (3.5%) was of the CD complex. The results revealed a high contamination of the environment by C. botulinum in buffalo breeding areas of the lowlands of Maranhao, Brazil, indicating a permanent potential for the occurrence of botulism in animals.

Study of the presence of the spores of Clostridium botulinum in honey in Brazil

R.P. Schocken-Iturrino*, M.C. Carneiro, E.C. Kato, J.O.B. Sorbara, O.D. Rossi Jr.Fac. Ciências Agrárias e Veterinárias – Jaboticabal - UNESP, Brazil

Infantile botulism was firstly registered in the USA, having been notified more than 600 cases since the identification of the disease in 1976. The symptoms of botulism are: constipation, muscle weakness, decrease of the reflexes and of the movements, multiple disfunctions of the cranial nerve and paralysis. Considering the potency of the toxin, it can even take to death. The botulism occurs by the colonization of the digestive tract by the bacteria Clostridium botulinum and production in vivo of the toxin, which is absorbed at intestinal level and it acts in the autonomous nervous system. This fact only happens in babies and children under one year old due to the intake of spores and germination of these in the digestive tract with less acid pH, differently from individuals older one year in which the pH is lower. The honey as a quickly available source of sugar, has been recommended thoroughly for babies' diet. This deserves special attention, because several studies in other countries have been indicating the honey as a risk factor for infantile botulism. This work had as objective to evaluate the presence of spores of Clostridium botulinum in honey marketed in Brazil. Eighty five (85) samples of honey from different origins were analyzed for the presence of spores of Clostridium botulinum. All the samples were slightly heated up for homogenization, before the removal of an aliquot of 10 g, which was diluted in 25 ml of sterilized saline solution, sown in Roberts media, submitted to a thermal shock (80-82 ^oC during 10 minutes and chilled at 20 ^oC in water with ice) and incubated at 37 ^oC for up to 10 days. In the 5 and 10 day of incubation, the samples were tested for the presence of botulin toxin by using the biological method in mice. The samples under suspicion were submitted to the isolation of Clostridium botulinum sowing them in plates of reinforced Clostridium agar (R.F.C.A.-Oxoid) and incubated at 37 ^oC in anaerobic conditions with the Gas-Pak system. The characteristic colonies were submited to smears stained by the Gram’s method and grown in tubes with Roberts medium and incubated in the same conditions mentioned for execution of biochemical test for identification and typing of the toxin. From the 85 samples analyzed, 6 were positive for botulism confirmed by biochemical test and typed as being type A, B and D toxin. Considering the wide use of the honey in the infantile diet and the possibility of this product be related to the sudden death in children with less than one year, it is necessary more studies on this agent present in honey, to alert and to avoid its consumption by these children.

Financial Support: CNPq.

Session IV: Treatment and Prevention of Clostridioses

Treatment and prophylaxis of clostridial intoxications and infections in humans

S.M. Finegold
Infectious Diseases Section, Veterans Administration Medical Center West Los Angeles and University of California School of Medicine, Los Angeles, USA

The two principal clostridial intoxications of man are tetanus and botulism. Therapy of tetanus involves antitoxin, antimicrobials, surgical management, and intensive care management. Prophylaxis of tetanus is carried out by active/passive immunization and wound care. There are three principal types of botulism–food-poisoning, wound botulism and infant botulism. Therapy and prophylaxis vary according to the type and will be discussed. Infections due to clostridia are variable in nature and severity. They include all the major categories of infection involving any bacteria, but clostridial myonecrosis (gas gangrene) and serious soft tissue infections are typically of the greatest concern. Gastrointestinal infection due to Clostridium species may also be serious; included in this category are enteritis necroticans (necrotizing jejunitis), neutropenic enterocolitis, and pseudomembranous colitis due to C. difficile. Therapy of non-gastrointestinal clostridial infections involves debridement and drainage and antimicrobials. Increasing resistance of certain clostridia to a number of antimicrobial drugs is a concern and must be considered in choosing an antimicrobial agent; in particular, resistance is seen in 20-35% of non-C. perfringens clostridia with clindamycin and cefoxitin. A small percent of clostridia exhibits resistance to penicillin and rare strains are resistant to metronidazole. Ironically, the principal mode of therapy for antimicrobial agent-induced pseudomembranous colitis is the use of antimicrobial agents–chiefly oral metronidazole, vancomycin, or bacitracin. In seriously ill patients, particularly those with toxic megacolon, surgery may be required. Relapse following medical therapy occurs in 20-30% of patients and may be difficult to manage; different strategies to try in this situation will be discussed. Prophylaxis of gas gangrene and other clostridial infections involves appropriate surgical management and the use of antimicrobial agents. Prevention of pseudomembranous colitis revolves around the rational use (or restriction) of antimicrobial agents and infection control measures. Ultimately, we hope to be able to provide immunization and to develop effective and safe means of biological interference.

Quantitation of commercial equine tetanus antitoxin by competitive enzyme-linked immunosorbent assay

D. R. Kolbe*
USDA, APHIS, VS, Center for Veterinary Biologics-Laboratory, Ames, IA, USA

In the United States, the potency of commercially prepared equine tetanus antitoxin is determined by the method outlined in the Code of Federal Regulations, Title 9, Part 113.451. In the current test, commercial antitoxin is tested by a toxin neutralization test (TNT) in guinea pigs. The in vivo test measures antitoxin content through effectiveness of protection of guinea pigs injected with diluted mixtures of antitoxin based on the volume of antitoxin in the final container and the number of antitoxin units claimed on the container label. The competitive enzyme-linked immunosorbent assay (cELISA) measures antitoxin content based on a competitive reaction between known or unknown serum and murine monoclonal antibody specific for tetanus toxin. The monoclonal antibody delays death in mouse passive protection studies, reacts with the C fragment of tetanus toxin, and is directly conjugated with horseradish peroxidase. No cross-reaction was observed with toxins of Clostridium chauvoei, C. novyi, C. perfringens, or C. sordellii. The cELISA is designed to measure the antitoxin content of test samples containing 100 to 1500 American units of antitoxin. Tetanus antitoxin titers obtained using the guinea pig TNT compared favorably with results obtained using the cELISA. The cELISA serves as a feasible alternative to the guinea pig TNT because it can be completed in less time, is reproducible, and eliminates the pain caused to the test animals.

Testing of neutralising antibodies to botulinum A toxin: implications for potency testing of therapeutic antitoxins.

H. C. Martin* and D. Sesardic.
Division of Bacteriology, National Institute for Biological Standards and Control, Blanche Lane, Potters Bar, Hertfordshire, EN6 3QG, UK.

Botulism is a rare but life threatening disease caused by toxins produced by Clostridium botulinum. Human, including infant, botulism is generally associated with A, B & E toxins where effective treatment is offered by i.v. or i.m. injection of botulism antitoxins. Therapeutic antitoxins are traditionally hyperimmune equine preparations containing >500 IU/ml of A, B & E antitoxins, although USAMRID have prepared equine derived botulinum F(ab’)₂ antitoxin heptavalent (A, B, C, D, E, F, G). Human botulism immunoglobulin (BIG) has been prepared from human subjects immunised with pentavalent (A, B, C, D, E) botulinum toxoid and successfully used in an infant botulism clinical trial in California.
We have developed an in vitro assay selective for type A toxin. The assay is dependent on the BoNT/A L (light) chain metalloprotease activity on synthetic SNAP-25 substrate. We have previously shown that the synthetic substrate is as effective as the native synaptosomal membrane associated protein which is an essential component of the neuroexocytotic machinery and the intracellular target for BoNT/A (Ekong et al., 1997). Further, we have demonstrated that BoNT/A in vitro activity is specifically inhibited by type A toxin antiserum but not by antiserum to other neurotoxins, suggesting that the assay could also be used for detection of toxin neutralising antibodies. Using 60-80 mouse LD₅₀/ml of purified type A toxin (>1.5 x 10⁸ mouse LD₅₀/mg protein) and equine International Standard (I.S.) for type A antitoxin (500 IU/vial) or BIG (25 IU/ml of type A antitoxin) it was possible to detect 0.001 IU/ml of type A antitoxin activity for both preparations, as the lowest concentration of antitoxin to inhibit BoNT/A proteolytic activity. This is at least 10-fold more sensitive than the mouse bioassay. Sensitivity was not affected if equivalent activity of a less pure preparation of BoNT/A was used. There was no inhibition if I.S. for tetanus antitoxin, equine (TE, 1400 IU/vial) or I.S. for human anti-tetanus IgG (TE-3) was used at >100-times the concentration. Three different preparations of equine antitoxins and BIG with in vivo determined potency for A antitoxin of 25, 440, 850 and 2500 IU/ml were tested against I.S. in vitro. Comparable results (30, 350, 300 and 2000 IU/ml) were obtained suggesting that an in vitro assay could be used for assessing the potency of equine or human IgG derived anti-botulism therapeutic antitoxins. Further validation studies will need to be performed.

Detection of neutralizing antibodies against a -toxin of different C.septicum strains in cell culture

F. Roth*, K. Jansen and S. Petzke
Institute for Applied Biotechnology in the Tropics, Göttingen, Germany

C.septicum as well as C.chauvoei, C.novyi or C.sordellii is a causal organism of the so called gas gangrene. C.septicum is the pathogen, which causes the classical malignant oedema. The organism occurs in the soil and intestine of humans and animals. Because of its strong cytotoxical a -toxin the infections often are lethal. To prevent losses in animals vaccination with a -toxoid vaccines is carried out.
Quality control of the vaccines by a neutralization test in mice is no longer acceptable. A lot of different in vitro methods are developed and after their standardization and validation alternatives to the in vivo methods. For the detection of neutralizing antibodies against the lethal cytotoxic a -toxin of C.septicum a cytotoxin inhibition test was standardized. The standardization included the test of seven different permanent cell lines for their sensitivity against the a -toxin and the optimization of their growth conditions, the production of reference a -toxin and laboratory reference anti-a -toxin from serum of rabbits, and the alignment of the laboratory reference anti-a -toxin and the international reference anti-a -toxin. In the studies a -toxin from three different strains of C.septicum were compared. One strain is an international C.septicum reference strain from the National Collection of Type Culture (No.: NC 547), the two others are field strains from outbreaks in Germany. Sera from five heterologous polyvalent and three monovalent vaccines from seven rabbits groups were available. The monovalent vaccines are divided into two groups. Group one is a heterologous monovalent C.chauvoei vaccine, group two are homologous monovalent C.septicum vaccines. Each vaccination was carried out according to the procedure of the German Pharmacopeia.
In three out of the five sera of the groups vaccinated with the heterologous polyvalent vaccine, cytotoxin neutralizing antibodies were detected. High antibody titres were observed in sera of rabbits vaccinated with a vaccine of the strain NC 547, lower titres in the sera of rabbits vaccinated with a vaccine of a field strain. No cytotoxin neutralizing antibodies could be found in the sera of rabbits vaccinated with the monovalent C.chauvoei vaccine. Here no cross-reaction between C.septicum and C.chauvoei could be observed. The toxins of all strains showed the same ranking of the vaccines. Vaccines which cause high antibody titres in the animals were detected by all toxins as such, as well as vaccines which have middle or low antibody inducing capacitiy. The results are indepent of the C.septicum strain used for the production of a -toxin.

Prevalidation of two different ELISA systems for the potency testing of C. perfringens b - and e -toxoid containing veterinary vaccines

E. Ebert*, V. Öppling, E. Werner and K. Cussler
Paul-Ehrlich-Institut, Federal Institute for Sera and Vaccines, Langen, Germany

The requirements for the quality control of C. perfringens vaccines for veterinary use are described in the monograph 363 of the European Pharmacopoeia (Ph. Eur.) and in §113.111 of the US Code of Federal Regulations (CFR). For the current used potency estimation the vaccines have to be administrated in rabbits. After collecting and pooling of the rabbit sera the neutralising antibodies against C. perfringens b - and e -toxin are determined in a mouse neutralisation test (MNT). In this quantitative assay the protective capacity of the test serum is compared to an international standard preparation with defined antitoxin units.
Two ELISA methods were developed for the replacement of the MNT.
Both methods use monoclonal antibodies to determine the quantity of specific antibodies against b -toxoid (Capture-ELISA) and against e -toxoid (Competitive-ELISA) in vitro. A high specificity and a good reproducibility are evident for both test systems. In parallel to the routine batch potency test in mice, the b - and e -antitoxin levels in 523 rabbit serum samples were estimated with the ELISA procedures. Linear regression analysis showed significant correlation coefficients for both ELISA systems at a significance level of p < 0.01.
An interlaboratory prevalidation study was carried out in four laboratories to evaluate the transferability of the ELISA procedures. Respective three code labelled serum samples were multiple tested by each participant to evaluate the inter- and intralaboratory reproducibility of the in vitro methods. The ranking of the test sera, based on the in vivo potency and on the in vitro potency estimated with the Capture-ELISA, in all four laboratories are in agreement. The ranking obtained with the Competitive-ELISA in three laboratories are in agreement, however, one laboratory showed a different ranking.
In the alteration of the monograph for "Clostridium Perfringens Vaccine For Veterinary Use" (PHARMEUROPA Vol.9 No.4, 1997) is established that for the batch potency test a suitable validated alternative method could be used if the in vitro method show a satisfactory correlation with the in vivo potency test.
It is concluded that both prevalidated ELISAs seems to be suitable alternative methods for assessing the potency of C. perfringens b - and e -toxoid in batches of mono- and multicomponent vaccines for veterinary use.
However, further studies like a formal validation is needed for the final replacement of the required MNT in the Ph. Eur. monograph 363.

The control of necrotic enteritis in sucking piglets by use of a Clostridium perfringens toxoid vaccine

S. Springer* und H.-J. Selbitz
Impfstoffwerk Dessau-Tornau GmbH, Research and Development Department, P.O. Box 214, 06855 Roßlau, Germany

Necrotic enteritis (NE) in sucking piglets constitutes a serious problem in piglet rearing units due to its high morbidity and mortality.
This disease is caused mainly by Clostridium (C.) perfringens type C. The beta-toxin plays a decisive role in the pathogenesis of this disease.
On the basis of this experience, a C. perfringens toxoid vaccine for use in sows has been developed. The sows are vaccinated twice at an interval of three weeks, immediately before farrowing. Antibodies to the beta-toxin protect the piglets against. perfringens-C-infections passively by intake of the colostrum.
According to the monography of Clostridium perfringens vaccines for animals, the German Pharmacopoeia prescribes a method of efficacy testing. This method is based on the immunisation of rabbits, the extraction of pooled sera and the following detection of antitoxic antibodies in mice by means of an appropriate test toxin. A vaccine is considered effective when a minimum of 10 I.U. beta-antitoxin per millilitre rabbit serum is induced. Own investigations established from 17.14 I.U. to 98.23 I.U. beta-antitoxin per millilitre rabbit serum induced by a C. perfringens toxoid vaccine sample.
The vaccine has been applied under field conditions in different rearing units at the same time; mostly as emergency vaccination after outbreak of the disease.
The effect of vaccination has been evaluated by record of the total numbers of born and lossed piglets. The use of the vaccine in combination with other measures decreased the losses by about 30 %.
Own results will finally be discussed.

Liver lesions and prduction performance data as indicators of necrotic enteritis in broiler flocks.

A. Løvland* and M. Kaldhusdal
National Veterinary Institute, Oslo, Norway

In most Norwegian poultry processing plants, carcass condemnations and broiler performance data is recorded in a way that with some effort, it is possible to use this data as a part of disease surveilance. Until now the use of the data has been limited, and used mainly as an economic performance report to the farmers. Avoparcin was used as an antibacterial feed additive in Norway from 1987 up to July 1995. In 1995 the use of Avoparcin as feed additive was prohibited in Norway (in the EU from 1997), and the poultry industry decided not to use growth promoters. Avoparcin has in earlier studies shown potent inhibition of Clostridium perfringens (CP) growth.
During the second half of 1995 post mortem reports from the processing plants indicated steep increases in condemnation due to liver lesions (increasing from 0.02% to 0.30% of the carcasses). By pathological and bacteriological examinations, these types of lesions were diagnosed as cholangiohepatitis and focal hepatitis, with isolation of CP.
We have studied the association between production performance and frequencies of carcasses with liver lesions at slaughter. Preliminary results indicate losses at a lvel that cannot be explained by condemnation figures alone. Necrotic enteritis (NE) causes by CP appeared as a clinical problem at the same time as the frequencies of CP-associated liver lesions at slaughter started in increase. Further, when the heaptitis problem started to decrease, the clinical NE problem gradually disapeared. However, the hepatitis problem persisted after the clinical problem was gone. This co-variation of two conditions associated with CP suggests the presence of subclinical necoritic enteritis depressing production performance in flocks with increased levels of liver lesions.
The aim of presenting our preliminary results is to make other European researchers working on Clostridioses aware of the possiblity of monitoring the incidence of necrotic enteritis and subclinical necrotic enteritis in broilers by systematically using data from processing plants. Necrotic enteritis may emerge as a problem in other countries with intensive poultry production if the use of antibacterial components becomes more restricted.

Occurrence and determination of the potential toxical risks of Botulism Neurotoxine C1 and their relation to ecological parameters of flat saltwater pools in the national park Neusiedler See - Seewinkel, Burgenland, Austria

Zechmeister T*, Farnleitner A †, Kirschner AKT ‡
* Biozentrum Wien, Inst. f. Biochemie Medizin. Fakultät,
Dr. Bohrg. 7, A-1030 Vienna, Austria
†Bundesanstalt f. Wassergüte, Schiffmühlenstr. 120, A-1220 Vienna, Austria
‡ Inst. f. Medizin. Biologie, Währingerstr. 10, A-1090 Vienna, Austria

Botulism neurotoxine, serotype C (BoNt/C) causes animal botulism. This toxine is produced under anaerobic conditions by Clostridium botulinum strains, infected with phages containing the toxigenic gene tox+. Type C1 is worldwide common in avian species, causing massive outbreaks in wild waterfowl. A variety of different lysogenic types of tox+ phages and different Clostridium botulinum strains lead to different type C1-toxines of different toxicity. Especially in the flat salt water pools and surrounding marshlands of the national park Neusiedler See-Seewinkel, Austria, avian botulism has been diagnosed for several times since 1982, often leading to the collapse of several endangered bird populations.
In our planned work we want to elucidate the connections between the potential occurrence of Botulism Neurotoxine C1 production and the ecological parameters in selected areas of the national park. The ecological parameters include chemophysical data (temperature, pH, O2, phosphorus and nitrogen fractions, etc.) and microbiological data (bacterial numbers, biomass and production, etc.) in both the sediments and the water column of the flat saltwater pools. In addition, microcosm experiments will be conducted simulating different ecological conditions (enrichment with sugars, aminoacids, antibiotics, phages, evertebrates etc.) in order to find out the key factors being responsible for successful toxine production. After extraction and enrichment toxine is detected via mouse bioassay. Extracted and purified DNA of the environmental samples is analysed for different types of tox+ phages by PCR technique, DGGE and sequencing.
The results of this study shall be the basis for future monitoring and management programs, which should dam in massive outbreaks of the desease.

Water holes – incubation areas for botulism in Brazil?

I.S. Dutra* and H.S.H. Seifert**
* Department of Production and Animal Health, Unesp-Campus de Aracatuba, SP 16015-050, Brazil
**Institute for Crop and Animal Production in the Tropics, Georg-August-University, Göttingen, Germany

Since botulism was first observed in the State of Piaui at the end of the Sixties, it has become a serious problem in the extensive beef cattle production system in Brazil. It is well known that the classical pattern of pathogenesis, osteophagia - due to the deficiency of phosphorus in the pasture - is one of the causes for the emergence of the disease. However, it does not explain the continuously increasing incidence of botulism. The Laboratory of Microbiology of the Dep. of Production and Animal Health of the Unesp-Campus in Aracatuba, SP. was able to demonstrate active botulism type C- and D-toxin in water and mud collected from water-holes from pastures where cattle died with symptoms of botulism. Thus it may be supposed that the environment of the water-hole which contains large amounts of cattle dung is an incubator area in which the spores introduced with infected faeces germinate and produce toxin. Continuation and increase of grazing cattle on these sites will lead to the accumulation of infection of the water-holes and the chance of toxin development. Further research is required to prove this hypothesis.
ung is an incubator area in which the spores introduced with infected faeces germinate and produce toxin. Continuation and increase of grazing cattle on these sites will lead to the accumulation of infection of the water-holes and the chance of toxin development. Further research is required to prove this hypothesis.

Poster

An exceptional serious case of botulism in the Wesermarsch, Northwest Germany in Summer 1996

J. Altmann
Landkreis Wesermarsch, Amt für Veterinärwesen und Lebensmittelüberwachung, Postfach 1435, 26919 Brake, Germany

On a family-run well-managed dairy farm with 60 cows in Wesermarsch in June 1996 botulism occurred in the dairy herd, causing the loss of 46 animals.
The pastures were not fenced but surrounded by ditches which were part of the water drainage system in Wesermarsch and from where the cows could drink at libitum.
In the hot summer 1996 the farmer reported to the local veterinary authorities a "suspected case of intoxication" among his cows. Three cows had been slaughtered and post mortem examination was done. Another twenty cows showed also symptoms of the "suspected intoxication". Visiting veterinary authorities could not make a specific diagnosis. Clinically the following symptoms were present: Central nervous depression, motoric paralyis, restlessness, progressive and unphysiological lying down and general progressive weakness, uncoordinated movements, stumbling, partial paralysed tail, no signs of resistance when pulling out the tongue, and stronger salivation than usal.
Differential diagnosis resulted in excluding lead poisoning and Aujeszky disease. There were no indications of the existence of any poisonous plants in the pastures. The decaying carcass of a hare was found in one of the surrounding ditches. Based on clinical symptoms and epidemiology, botulism was suspected, slaughtering was prohibited and the already slaughtered cows were declared unfit for human consumption.
Within three days another 43 cows died. Six slightly sick cows were treated with C. botulinum antiserum (Onderstepoort): Three cows died, three survived., six non treated animals recovered as well.
For the first time in Wesermarsch an animal diease caused so many casulties in such a short of time. Unrest was seen not only with farmers but also among the inhabitants of the nearby villages. Large-scale and extensive research in this specific case followed, not only to exclude a contagious disease but also to look for financial support from the Animal Health Fund of Lower-Saxony.
Finally it became clear that the fly blown carcass of the hare in the drinking water was the reason for this outbreak. As there is no botulism-vaccin available in Germany, dairy farmers were strongly advised to bar the ditches and to use tanks for drinking-water.

Clostridium perfringens Enterotoxaemia in Hunting Falcons in Dubai/UAE

H. Böhnel
Institute for Crop and Animal Production in the Tropics, Georg-August-University, Göttingen, Germany

In the United Arab Emirates numerous falcons are kept in captivity. Recently there were several casualities, suspected as Enterotoxaemia (1).
For isolation and identification 16 faecal specimens were received, and possibly for vaccine production, originating from healthy and diseased animals.
By our standard laboratory techniques (2) 8 strains were isolated and identified by gaschromatography and tested for haemolytic and cytotoxic metabolites (3). These isolates were compared with the strains of our Clostridia strain collection. Surprisingly no local strains from Dubai were found. However, the incriminated strains were grouped together with some of German origin.
The explication for these inattended results: the animals had been imported from Germany about 6 months prior to disease onset.

References

1. Wernery, U., J. Kinne, A. Sharma, H. Boehnel, J. Samour, (1998). Clostridium enterotoxaemia: An emerging disease in Falconiformes in the United Arab Emirates (UAE) (in press)
2. Heitefuß, S., (1991). Untersuchungen zur Identifizierung von aeroben, anaeroben und fakultativ anaeroben Bakterien mit gaschromatographischen Methoden. Gött.Beitr.Land.Forstwirt.Trop.Subtrop. Vol 57
3. Schaper, R. (1991). Methodische Untersuchungen zur Produktions- und Wirksamkeits-kontrolle von Rauschbrandvakzinen. Gött.Beitr.Land.Forstwirt.Trop.Subtrop.Vol 61

 

Clostridia isolates originating from the freshwater clam Corbicula fluminea in the rivers Rhein and Neckar (Germany)

H. Böhnel
Institute for Crop and Animal Production in the Tropics, Georg-August-University, Göttingen, Germany

During the hot summer 1977 there was a high mortality of freshwater clams Corbicula fluminea in the rivers Rhein and Neckar. These clams are of tropical origin and are invading South-German rivers.
The dead clams disintegrate, the bodies are floating on the water surface due to gas bubbles (Vobis, pers. comm, 1977). Suspicion arose that the causative pathogen might be Clostridia.
Alive and dead clams where tested for the presence of Clostridia according to our standard procedures (Heitefuß, 1991). Identification by gaschromatography and toxicity tests in mice and guinea pigs were performed.
Two strains could be isolated

No 2528 is a proper field strain; No. 2529 is related to C. clostridioforme.
Infectivity and pathogenicity tests in clams need to be done.

Reference

Heitefuß, S. (1991): Untersuchungen zur Identifizierung von aeroben, anaeroben und fakultativ anaeroben Bakterien mit gaschromatographischen Methoden. Gött.Beitr.Land.Forstwirt.Trop.Subtrop. 57

In vitro evaluation methods for Clostridium botulinum type C and D vaccines.

C.E. Ellis* , M. Hamman, R. de Bruin and H. Harris
Onderstepoort Veterinary Institute, South Africa

Animal botulism is caused by Clostridium botulinum type C and D. In South Africa active vaccination is required to combat this disease. The present vaccine is produced by Onderstepoort Biological Products and it consists of a combination of the C and D neurotoxins, in toxoided form. The quality standards for the vaccine are a neutralising antibody level of 2 IU/ml for the C component and 5 IU/ml for the D component, as measured by the potency assay. The potency assay involves the vaccination of a group of guinea pigs with the vaccine to be tested, followed by measurement of the antibody levels in the antisera with the mouse neutralisation test. As this vaccine is frequently produced, the quality control tests in animals add to the cost of the final vaccine. Together with the increasing international demand to replace and or reduce the use of animals for vaccine testing, alternative in vitro assays were developed.
ELISA’s were developed to quantify the neurotoxins after production, as well as determining the concentration of the toxoids after detoxification. As a replacement of the mouse neutralisation test, ELISA’s were also developed to determine the antibody levels in guinea pig antisera, as part of the potency assay. Results will be presented to show the degree of correlation between the in vivo and in vitro assays.
Estimation of the C neurotoxin and toxoid concentrations with ELISA correlated with the results of the potency assay, thus making it possible to formulate the concentration of the C antigen in the final vaccine. Difficulties with the D antigen were encountered, as the concentration of the D toxoid could not be correlated with the outcome of the potency assay. Good correlations between the ELISA and mouse neutralisation test were found for measurement of the antibody levels in guinea pigs immunised with the separate C or D antigens. However, the ELISA ‘s for measurement of the C and D components in guinea pig antisera, immunised with the combination vaccine, can not be used due to cross reactions. The cross reactions are due to common epitopes on the heavy chain of the C and D neurotoxins. Polyclonal antibodies to the light chain of the C and D neurotoxins will be prepared, in order to develop an ELISA that can distinguish between antibodies to the C and D antigens.

New recovery of neurotoxigenic Clostridium butyricum type E from a case of infant botulism

G. Franciosa*, F. Anniballi, L. Fenicia, P. Aureli
Food Microbiology Laboratory, Food Department, Istituto Superiore di Sanità, Rome, Italy

At the end of July, 1998, a case of infant botulism came to our attention. The baby, a 5 months-old female living in Mestre (province of Venice), presented the typical symptoms presumptive of botulism except for tympanic abdomen that is rather unusual. After excluding many other hypothesis, physicians tentatively diagnosed infant botulism and sent to our laboratory the first clinical samples (serum and feces) for confirmation of the diagnosis, after about 20 days from the onset of the illness. The serum was negative for botulinum neurotoxin; however, the feces contained botulinum toxin type E. Recovery of the neurotoxigenic organism from the specimen was achieved by isolating single colonies from egg-yolk agar (EYA) plates streaked with the enrichment cultures from the feces and submitting them to PCR assay with primers specific for the type E neurotoxin gene amplification. The colonies yielding a positive PCR product were shown to produce the botulinum neurotoxin type E by standard neutralization mouse assay and were lipase and lecithinase negative. The organism was then characterized by sugar fermentation patterns and GLC analysis of metabolic products by the Virginia Polytechnic Institute procedure and classified as Clostridium butyricum.
The baby had been breast fed; a sample of honey used to sweeten her pacifier was also analyzed but it was negative for neurotoxigenic clostridial spores. Hence, the source of the spores remains unknown. Detection of neurotoxin and spores in the feces samples subsequently collected from the patient is still under progress.
Two similar organisms were isolated in Italy 14 years ago during investigation of two cases of infant botulism occurred in Rome. In 1992 and 1993 two cases of infant-like botulism in adults (respectively occurred in Ferrara and Rovigo, two cities geographically located next to Venice) were also caused by C. butyricum type E. This is the fifth strain of neurotoxigenic type E C. butyricum that is isolated in Italy, where it appears to solely cause infectious forms of botulism characterized by similar symptomatology, e.g. presence of large amounts of gas in the intestine probably due to the metabolic fermentation of lactose by the strain.
Other strains of C. butyricum type E have recently been isolated in China from cases of foodborne botulism and from the environment.

Toxinotyping of Clostridium perfringens by sandwich ELISA.

Ginter A.*, Coppe Philippe°
*Bio-X Bvd. Edm. Machtens 75/16 B-1080 Bruxelles
°CER 1, rue du Carmel B-6900 Marloie

The classical toxinotyping test, based on toxine-antitoxine neutralization in mice, has been advantageously replaced by an in-vitro detection of toxines alpha, beta, epsilon and a constitutional protein of C.perfringens. For this purpose we developed a sandwich ELISA kit using a combination of polyclonal and monoclonal antibodies. This combined sandwich Elisa was used on animal intestinal content samples as well as on isolated C.perfringens strains. The results were compared to those obtained with DNA probes and are discused in this paper.

Detection of antibodies against a -toxin of Clostridium septicum in sera of specifically immunized cows and their offspring by a cell culture test

K. Jansen* and F. Roth
Institute for Crop and Animal Production in the Tropics, Georg-August-Universität, Göttingen, Germany

Clostridium (C.) septicum is the classical causative agent of the malignant oedema. The bacterium is pathogenic to animals as well as to humans. It is an inhabitant of the soil and of the intestines if incorporated with fodder. Infections with C. septicum in wounded tissues often are fatal due to the activity of the lethal a -toxin produced by C. septicum. The a -toxin, an exoprotein of 45-48 Kd, invokes poreforming in host cells, thus determining their death.
In 1996, a locality specific outbreak of malignant oedema was diagnosed at a dairy farm in West Germany. Cows died after a characteristically rapid decline shortly following their giving birth to calves. A field strain of C. septicum was isolated from a sample and identified as the causative pathogen.
The strain, laboratory No. 2258, was cultivated in order to produce a locality specific a -toxoid-vaccine for the cattle at the farm. The vaccine was applied following a scheme which involved a first vaccination, a booster vaccination after four weeks and a freshening up in the following year. Monitoring was carried out firstly by watching the stock for further losses. Secondly, blood samples of the cows were taken: a) before vaccination (0-sera), b) at the second vaccination (antisera) and c) from the calves of vaccinated cows, at two to six weeks of age. The calves had received colostral milk from their mothers.
The obtained sera were investigated in a cytotoxin neutralisation test with cell cultures. A dilution series of the serum samples was incubated with active a -toxin from the vaccine production strain 2258 and sensitive cells. Serumneutralizing units per ml (AU/ml) that indicate the dilution of a serum needed to protect 50% of the cells were given.
22 sets of sera with 0- and antiserum of cow and antiserum of calf were investigated, with reproducible results. It was shown that the reaction towards vaccination in general was highly variable. 63 % (16 cows) reacted by increasing antibody titres by 100 % and more. 8 of these cows had detectable antibody titres in the 0-sera and reached 1000-5500 AU/ml in the antisera, the other 8 cows had few or no antibodies in the 0-sera and only increased antibody titres up to about 500 AU/ml. 6 cows showed no significant difference of antibody titre before and after vaccination.
Sera from the offspring only in one case contained a significantly higher titre of antibodies than the mother´s 0-serum, regardless of the mother´s reaction towards vaccination.
No further losses due to C. septicum were reported.

Detection of Clostridium perfringens beta antitoxins in animal sera with blocking ELISA and comparison with biological assay

B. Krt
Institute for Microbiology and Parasitology, Veterinary faculty, University of Ljubljana, Ljubljana, Slovenia

The purpose of the present work was to evaluate blocking ELISA for the detection of antibodies against the Clostridium perfringens beta toxin in comparison with the toxin neutralization test in mice. A number of swine, sheep, bovine, rabbit and horse serum were searched for. International standard beta antitoxin was used as a standard in both tests.
By comparing the results of ELISA with the results of the biological assay we have found out that the methods matched in the qualitative detection of beta antitoxins, but in the quantitative differences were often found. However, ELISA could be used for approximate determination of the potency of beta antitoxin thus reducing the use of laboratory animals. Using this method the number of sera for examination can be much more higher whereas the price and duration of examination are lower than in biological assay.

Evaluation of Rapid ID 32 A and API 20 A for the Identification of Clostridium botulinum Types A – G

P. Loch
Institute for Crop and Animal Production in the Tropics,
Georg-August-University, Göttingen, Germany

The Rapid ID 32 A and the API 20 A (BioMérieux SA, Marcy-l`Etoile, France), two commercially available micromethods for the identification of anaerobic bacteria were tested for their ability to identify 29 reference strains of Clostridium botulinum types A-G and 6 field strains. Furthermore all strains were biochemically investigated according to the methods of the Anaerobe Laboratory Manual of Holdeman and Moore (Virginia Polytechnic Institute/Blacksburg USA) and were tested with regard to their production of botulinumneurotoxin (BoNT) in the mouse.
After the recommended incubation period of 4h the Rapid ID 32 A was able to identify 14 (48.3 %) reference and 3 (50 %) field strains correctly as C. botulinum. It is remarkable that this micromethod failed to correctly identify any of the 9 reference strains producing neurotoxin of types C and D. All these strains were considered as being non-growing bacteria whereas the investigation of all 11 strains of C. botulinum type A or B led to a correct identification. With strains showing metabolic activity in the Rapid ID 32 A an identification down to genus level could be achieved in 88.9 % (16 strains). Two strains (11.1 %) were incorrectly identified but when using additional tests an identification down to genus level was possible. The main problems arose in identifying strains that produce BoNT/F, BoNT/E or BoNT/G.
The Rapid ID 32 A was able to identify 5 field strains (83.3 %) correctly to genus level. Here again one strain producing BoNT/D was identified as being non-growing.
The API 20 A was able to identify all of the investigated strains correctly to genus level but only 3 (10 %) reference strains and 1 (16 %) field strain were correctly identified as C. botulinum. Here again strains of type C and D were almost inactive and so falsely classified as C. spp.. Strains of the biochemically more active types A, B or F were mainly identified as C. sporogenes and very seldom correctly as C. botulinum. These results were obtained after 48 h and could not be improved by varying the time of incubation.
Reviewing these results it can be concluded that both tested micromethods might be helpful for the diagnosis of strains of C. botulinum pathogenic to humans but are of almost no use for the strains of veterinary importance.

Improved method for the detection of botulinum toxin in soils rich in organic matter

K. Lube
Institute for Crop and Animal Production in the Tropics, Georg-August-University, Göttingen, Germany

The propagation of C. botulinum and its toxigenesis is dependent on many conditions such as, supply of nutrients, anaerobic conditions, pH, moisture content and temperature. Based on their examinations of samples from landfill sites, N.E. Ortiz and G.R. Smith concluded that the presence of the spores of C. botulinum together with the rotting organic matter and the generated heat, leads to proliferation and toxigenesis of the bacterium. We assume that in soils rich in organic matter there are favourable conditions for the survival of clostridial spores and their toxigenesis.
Our method for detection of the toxin in samples of soil is based on those used for soil, environmental samples, landfills and food.
The toxin detection contains two parts: at first we tried to isolate the toxin from a buffer-extract of the sample, then the cultivation of C. botulinum in an enrichment broth Reinforced Clostridium Medium (RCM), for 3 to 5 days followed, and the examination of its ability to develop the toxin.
The toxin (A-E) was detected in the mouse bioassay including its typification by neutralization with polyclonal antisera.
Important for the detection of the toxins respectively for their stability during the steps of neutralisation are the following steps:

• incubation and the thorough homogenization of the sample in gelatin-phosphate buffer (pH 6,2) at a ratio 1:1
• storage of the sample in buffer overnight at 4°C
• cold centrifugation at 9000 U/min for 10 minutes
• realization of the mouse assay with trypsinization and without heating the injected RCM-broth.

At first we investigated 31 samples of soil from rotting organic material with the described method. In 22 of 31 samples botulinum toxin was detected after incubation in the enrichment broth.
The question arises whether the method of detection of the toxin from the enrichment tubes is too sensitive for materials which contain spores of C. botulinum a priori.
In order to develop a preliminary insight into this problem, we examined samples (8) of soil and 20 garden moulds for their ability to develop the toxin after incubation in the enrichment broth. The toxin was found in 3 garden moulds but in no soil samples.
These first results suggest that it is not possible to isolate the botulinum toxin after the enrichment from any soil that contains botulinum spores.
In order to clarify which nutritive, physicochemical and microbiological conditions may support the toxigenesis of botulinum spores in soil, further investigations are necessary.

A Cytotoxin Inhibition Test as an Alternative to the Mouse Neutralisation Test for the Quality Control of C.septicum a -Toxoid Vaccines

Frauke Roth
Institute for Applied Biotechnology in the Tropics, Georg-August-University , Göttingen

As alternatives for the potency testing of Clostridial toxoid vaccines a lot of research was done to replace in vivo methods by in vitro methods as for example ELISA or the ToBi test (potency test for C.tetani vaccines). Recently no further investigations were done to standardize a cell culture test for the quality control of C.septicum a -toxoid-vaccines. The advantage of a cell culture test is the possibility on the one hand to detect specific antibodies which can neutralize the toxin and on the other handside to measure the biological activity of the toxin.
The preliminary results shown here deal with the optimization and standardization of a cytotoxin inhibition test (CIT) in cell culture. For the optimization of the test seven mammalia cell lines were tested for their sensitivity against the a -toxin. For the standardization of the in vitro method, international reference serum and a standardized a -toxin of the reference strain NC 547 of the National Collection of Type Culture were used. Rabbit test sera from the quality control of six different commercial vaccines and rabbit test sera from vaccination with an a -toxin preparation of the reference strain were available.
The highest sensitivity of all tested cell lines was obtained with the BHK21-BSR/PK5/88 cell line. The a -toxin of the reference strain was neutralized by the international reference serum as well as by the sera of the different commercial vaccines. A ranking of the vaccines was possible and the results were reproducible in various tests.

The Occurrence of Clostridium perfringens in the Intestine of Commercial Broiler

R.P. Schocken-Iturrino*, L.F.S.A. de M. Gama, F. Cruz, A. Berchieri Jr.
Fac. Ciências Agrárias e Veterinárias – Jaboticabal - UNESP, Brazil

It was studied the presence of Clostridium perfringens from the intestine content and mucous of the intestinal wall of commercial broiler from farms of the region of Ribeirão Preto, S.P., Brazil. Were analysed 560, 46-day-old broilers with dirty cloaca, sacrificed and the intestines collected and submitted to macroscopic analysis. For the microbiological analysis samples were submitted to a 10 fold dilution and plated onto SPS selective agar medium, incubated in anaerobic jars at 37^oC for 24 hr. Discrete black colonies designated "perfringens-like" were selected for confirmation of species, transferred to cooked meat medium tubes and incubated in anaerobically. Clostridium perfringens was identified on the basis of anaerobioses, morphology on Gram-stained smears presence of double zone of hemolysis on blood agar and submited to identification using the anaerobic API 20 A system. The results showed that from the 560 samples, 374 (66,78%) presented growth of anaerobic sporulated rods (Clostridium) and 186 (33,21%) presented growth of other species of bacteria. From the 374 isolates of Clostridium, were confirm through biochemical and toxicity tests 94 (25,13%) as been Clostridium perfringens and 280 (74,87%) as other types of Clostridium. Between these strains 19 were pathogenic; 3 C. chauvoei; 9 C. sordelli; 3 C. bifermentans; 3.C.septicum; and 1 C tetani., and the other 261 were no pathogenic.

Financial Support: CNPq.

Detection of botulin toxin and isolation of Clostridium botulinum in poultry litter samples used in feed-lot cattle

R.P. Schocken-Iturrino*, F. Ávila, J.O.B. Sorbara, M.C. Carneiro, A. Medeiros
Fac. Ciências Agrárias e Veterinárias – Jaboticabal - UNESP, Brazil

Botulism is normally a foodborne intoxication caused by the toxin of Clostridium botulinum always due to the ingestion of the toxin together with water or contamined food, where the toxin was preformed by the bacterium. The high interest in the utilization of poultry litter for feeding cattle, justifies the study of the risks that this type of feeding stuff will caused to the animals from the microbiological point of view. The samples of poultry litter, used in this study came from feed-lots of the region of Ribeirão Preto - SP, with a suspicion of having provoked death by botulism in bovine. The extraction of botulinical toxin was done in gelatin-phosphate buffered and the toxicity tested in mice (Smith, 1977). Clostridium botulinum isolation was carried out in enrichment tubes with cooked meat medium (Difco), and onto plates of reinforced clostridial agar (Oxoid), incubated them at 35 ºC for 5 days under anaerobiosis with the Gas-Pak system (BBL). The analysis of 100 samples of poultry litter, showed that in 5 (five) of them, type C botulinic toxin was detected. From these samples, in 4 (four) of them was also isolated C. botulinum type C. From the 100 samples, 5% (5), was contamined by toxin and in 9% (9) spores were demonstrated, showing the risk of supplying this type of feeding stuff to the animals without any treatment in order to avoid losses by botulism.

Financial Support: CNPq.

Isolation of Clostridium botulinum type C from an outbreak of Botulism in chickens

R.P. Schocken-Iturrino*, J.O.B. Sorbara, F.Ávila, A.A. Medeiros.
Fac. Ciências Agrárias e Veterinárias - Jaboticabal-UNESP, Brazil

An outbreak of type C botulism involving about 1,000 fourty-two-day-old broiler chickens in Sao Paulo inlands. Birds presented clinical signs ranged from complete flaccid paralysis of limbs and neck to drowsiness and loss of mobility caused by wing and leg weakness. The feathers became ruffled and were easily removed. Eyelid paralysis was present. The birds did not eat or drink water and died 3 to 12 hours after the symptoms were observed. The material was analized in our laboratory in Jaboticabal, one half ml serum of each bird brought to the lab. (eight) was inoculated intraperitoneally into each of four mice. Crops, gizzards and intestinal contents as well as pieces of liver, feces and feed, were extracted separately at 4 ^oC with 0.2M gelatin-phosphate buffer( pH 6.5) and tested for detection of toxin. Samples of 5g of feed and washings of crop, gizzard and intestine content, besides liver pieces and feces were examined for the presence of C. botulinum by enrichment in cooked meat medium (Difco), submitted to a thermal shock at 80 ^oC for 10 minutes and cooled with tap water, then incubated at 35 ^oC for 5 days, after which were centrifuged and tested for toxin. Gram` smears and biochemicals tests of colonies grown up onto reinforced clostridial agar plates (Oxoid) incubated in anaerobics jars with the gas-pak system were done. Type C botulinic toxin was demonstrated in all birds serum and in crop, gizzard, intestinal contents and in liver of some birds. Extracts of these samples caused lethal toxemia to mice in 24 h. Type C antitoxin neutralised specifically the toxin. Poultry litter was also positive, we believe the direct source of toxin were carcass, not collected from the sheds to avoid stress due to the high population 23 birds by m².

Financial Support: CNPq.

Development and characterisation of a mouse model displaying resistance against the colonisation of the gut by the enteropathogen Clostridium difficile.

V. Thomas¹, V. Rochet², S. Bulteau¹*, H. Boureau¹, A. Collignon¹, J. Doré² and P. Bourlioux^1
1Faculté de Pharmacie, Laboratoire de Microbiologie, Tour E1, 3° étage, rue J-B Clément, F92296 Chatenay-Malabry.
²LEPSD, INRA, F78352 Jouy-en-Josas Cédex.

Clostridium difficile (Cd) is a major nosocomial pathogen causing antibiotically induced diarrhea and pseudomembranous colitis. It takes advantage of the disturbance of the endogenous gut flora to colonise the colon and deliver its toxins. In order to investigate the potential mecanisms responsible for the barrier effect of normal flora against Cd, we developped a mouse model. The caecal flora from Cd-resistant hamster was successfully simplified and implanted into an axenic C3H mouse thereby producing a Cd-resistant mouse. The three dominant strains constituting the minimal barrier flora were isolated and identified by biochemical means on the one hand: fermentation profiles, enzymatic potential and total protein profiles, and by genetic techniques on the other hand: DNA/DNA hybridization, G+C percentage determination and 16S rRNA sequencing. These three bacteria are Clostridium indolis, C. cocleatum and C. fusiformis. Up to now, the latter had remained unclassified and we refer to it as C. fusiformis because of its shape. We have designed specific oligonucleotide probes based on the 16S rRNA sequences of the barrier flora and Cd, and we will also present our preliminary results on in situ hybridisation on faeces and caecum from our trixenic mouse model. This technique together with electronic microscopy should give us insights into the interactions between the gut, the barrier flora and Cd.

Measurement of the Effect of Clostridum chauvoei- and Clostridium septicum- Toxins Isolated in North-east Mexico on Distinct Cell Cultures

A. Wong-González*, F. Roth, K. Jansen, H.S.H. Seifert and H. Böhnel
Institute for Crop and Animal Production in the Tropics, Georg-August-University, Göttingen, Germany

Clostridium (C.) septicum, one of the causal organisms of classic malignant oedema in farm animals is closely related to Clostridum (C.) chauvoei, which is the causative agent of blackleg. These two species have similar characteristics as far as results from biochemical methods and gas chromatography are concerned. C. septicum however produces a lethal cytotoxic toxin which is not usually found in supernatants of C. chauvoei cultures.
A total of 267 samples, collected from sick or dead animals in the fields from Northeastern Mexico, were bacteriologically analysed and differentiated by the gas chromatography technique. For the differentiation of the isolated bacteria by gas chromatography their metabolic products and the long-chained fatty acids of the cell wall were compared. 67 strains could be identified as clostridia. 12 of these strains were characterised as identical to reference strains. The 55 remaining strains were shown to be pathogens. From these 55 strains, 17 belong to the group of C .chauvoei/C .septicum, 12 to the group of C .sordellii/C. bifermentans, 17 were identified as C. botulinum, 3 as C. haemolyticum, 2 as C. novyi A, 1 as C. perfringens and 3 as local pathogen field strains.
Studies on the effect of culture supernatants of the 17 strains of the C. chauvoei/septicum group on cell cultures of the lines EBL, 3T3, BHK21-BSR/PK5/88; CHO-K1 and MDCK were performed. The objective was to obtain further data for identification as the results from gas chromatography do not allow exact differentiation between C. chauvoei and C.septicum. The strains discussed in the following paragraphs are here still grouped according to the results from gas chromatography.
The strains 809 (Clostridium ssp.), 974 and 1200 (C. chauvoei) and 802, 853, 945 and 970 (C. septicum) show less than 12 cytotoxic units per ml supernatant (AU/ml). In supernatants of the strains 1199, 1241 and 1242 (C. chauvoei) and strain 955 (C. septicum), 100-250 AU/ml were found. High levels of cytotoxin of around 1700 AU/ml were reached in tests with strains 996 and 1178 (C. chauvoei) and 958, 892 and 924 (C. septicum).
The results were obtained in tests with BHK21-BSR/PK5/88 cells as this had proved to be the most sensitive cell line, closely followed by 3T3 and CHO-K1 cells. MDCK cells were of little sensitivity. These tendencies were confirmed with cytotoxins from all 17 strains investigated, regardless of their classification by gas chromatography.
Differences in cytotoxicity between the 17 strains were reproducible and suggested a differentiation between C. chauvoei and C. septicum different than indicated by gas chromatography.

Last Revised: 28 September 1998/Dr. Frank Gessler, E-Mail: fgessle@gwdg.de