Molecular construction of Clostridium botulinum type F progenitor toxin

Mapping of the antibody-binding regions on the HN-domain (residues 449-859) of botulinum neurotoxin A with antitoxin antibodies from four host species. Full profile of the continuous antigenic regions of the H-chain of botulinum neurotoxin A

Previously, we mapped the antibody (Ab) and T-cell recognition regions on the HC domain (residues 855-1296) of the 848-residue heavy (H) chain of botulinum neurotoxin A (BoNT/A). We have mapped here the HN-domain (residues 449-859) regions that bind protective anti-BoNT/A Abs raised in four different species. We synthesized, purified, and characterized 29 19-residue peptides that spanned the entire HN and overlapped consecutively by 5 residues, and also region L218-231 around the L-chain's substrate-binding site. Human, horse, mouse, and chicken anti-BoNT/A Abs did not bind to the L-peptide but recognized similar HN regions within peptides 519-537/533-551/547-565/561-579 (with slight left- or right-shifts), 743-761, 785-803, and 813-831/827-845 overlap. Recognition of other peptides that bound lower Ab levels showed similarities and also some differences. Peptide 463-481, strongly immunodominant with horse antisera, did not bind human, mouse, and chicken Abs. However, peptide 449-467 bound Abs in these three antisera, and the region may have shifted to the right (peptide 463-481) with horse Abs. The overlap 659-677/673-691 reacted strongly with human Abs whereas with mouse and chicken antisera, only peptide 673-691 showed low reactivity. Horse antisera had no detectable Ab binding to region(s) 659-691. The Ab-recognition regions on the H chain occupy surface locations in BoNT/A three-dimensional structure, but the great part of the surface is not immunogenic. Regions recognized by the protective antisera of the four different species are prime candidates for inclusion in synthetic vaccine designs.

Inactivation of botulinum and tetanus toxins by chelators

Purified type A botulinum toxin of about 10(6) mouse 50% lethal doses per ml was greater than 99.9% inactivated when incubated at pH 7.4 for 30 min at 37 degrees C in 20 mM 1,10-phenanthroline (PTL) or 2,2'-dipyridyl (DPD) and was 96% inactivated when incubated in 70 mM 8-hydroxyquinoline-5-sulfonic acid (HQL), but was not affected when incubated in 200 mM EDTA. When used as a representative of the chelating agents, PTL inactivated greater than or equal to 99.9% of toxicity in the culture filtrate of C. botulinum type A, B, and E strains. Highly purified tetanus toxin at 2.5 x 10(5) 50% lethal doses per ml lost all toxicity in 40 mM PTL or 150 mM DPD but was not detectably affected by 100 mM HQL (the highest concentration possible). Toxin inactivation by 20 mM PTL was completely blocked when the PTL was prereacted with an equimolar amount of Zn2+ and significantly reduced when it was preincubated with one-third its molar amount of Fe2+. DPD at 20 mM had little toxin-inactivating potency when preincubated with an equimolar amount of Zn2+ and only some of this potency when preincubated with an equimolar amount of Fe2+. Toxicity was not recovered by adding Zn2+ or Fe2+ to PTL-treated toxin. Neutron activation analysis of type A toxin showed that for each toxin molecule present, there was 1 atom of Fe, 0.4 atom of Zn, and 22 to 55 atoms each of Ca and Mg. The biological activity of botulinum toxin seems to depend on a metal component, which is likely to be Fe.

The use of monoclonal antibodies to analyze the structure of Clostridium botulinum type E derivative toxin

Six monoclonal antibodies against Clostridium botulinum type E derivative toxin were prepared. Three of the five binding to the heavy chain neutralized the derivative toxin; the other one binding to the light chain did not. Immunoblotting analysis with the monoclonal antibodies showed that the fragment obtained by tryptic digestion consisted of the light chain and part of the heavy chain (H-1 fragment) linked together by a disulfide bond(s) and that the antigenic determinants common between type E and F derivative toxins were located on both the heavy and light chains. The fragment induced by chymotrypsin treatment, like the tryptic fragment, bound to four monoclonal antibodies. The mild tryptic treatment and reduction resulted in separation of the chymotryptic fragment into two smaller fragments corresponding to the light chain and H-1 fragment. These results indicate that H-1 fragment contains the amino-terminal portion of the heavy chain. The monoclonal antibody neutralizing the toxin and probably recognizing the epitope on the carboxyl-terminal portion (H-2 fragment) of the heavy chain effectively competed for binding of 125I-labeled derivative toxin to synaptosomes. Of the two monoclonal antibodies neutralizing the toxin and recognizing the epitopes on H-1 fragment, one partially inhibited binding, but the other did not. This suggests that the binding of 125I-labeled derivative toxin depends mainly on the carboxyl-terminal region of the heavy chain and that interference with binding is not the only means of toxin neutralization.

Binding to mouse brain synaptosomes of Clostridium botulinum type E derivative toxin before and after tryptic activation

Clostridium botulinum type E125 I-labelled derivative toxin, both the partially active (intact) and the fully active (nicked) forms, bound to mouse brain synaptosomes. They gave the same KD and Bmax values in binding to synaptosomes, although the nicked form possessed a mouse lethal potency about 30 times higher than that of the intact form. These results may indicate that the binding site of the derivative toxin is not modified by tryptic activation and that the binding to synaptosomes is independent of the blockade of acetylcholine release.

Evidence that botulinum C2 toxin has two dissimilar components

Botulinum C2 toxin produced by most toxigenic and nontoxigenic strains of Clostridium botulinum types C and D contains two distinct protein components, and these were separated by ion-exchange chromatography and gel filtration. Neither of these components manifested the original toxicity, but the original toxicity was restored when the two components were mixed together and trypsinized. This indicates that C2 toxin consists of two dissimilar protein components and that the cooperation of the two is required to elicit toxicity.

Oral toxicities of Clostridium botulinum type C and D toxins of different molecular sizes

Clostridium botulinum type C progenitor toxins of different molecule sizes, C-L (16S) and C-M (12S), were purified from cultures of strains 573, Stockholm, and CB-19. C-L toxin showed some hemaggglutinin activity, whereas C-M toxin did not. Neither C-L nor C-M toxin was activated upon trypsinization. Molecular dissociation of purified type C-L and C-M toxins into toxic and nontoxic components was demonstrated by sucrose density gradient ultracentrifugation and diethylaminoethyl-Sephadex chromatography at pH 8.0. The molecular construction of type C progenitor toxin appears to be analogous to that reported for botulinum toxins of other types. C-L and D-L toxins showed higher oral toxicities to mice than did C-M or D-M toxin. Such higher oral toxicities were ascribed to the higher stabilities of these toxins in gastric and intestinal juices.

Interaction of botulinum type A, B and E derivative toxins with synaptosomes of rat brain

Clostridium botulimum 125I-labelled derivative toxin immediately bound to rat synaptosomes. Of the two fragments of type B derivative toxin, the large-molecular-weight fragment (fragment I) inhibited the binding of labelled type B derivative toxin to synaptosomes in the same manner as unlabelled type B toxin did. The inhibition by the small-molecular-weight fragment (fragment II) was less than that by fragment I. These findings suggest that type B toxin binds to synaptosomes mainly with some part of fragment I. The binding of labelled type A and E derivative toxins was inhibited by either of the unlabelled type A or E derivative toxins, but not by type B derivative toxin. It is concluded that synaptosomes of rat brain possess relatively specific binding sites for botulinum toxin types.

Development of antitoxin with each of two complementary fragments of Clostridium botulinum type B derivative toxin

Two fragments with molecular weights of 111,000 (fragment I) and 59,000 (fragment II) were separated from each other by gel filtration of dithiothreitol and urea-treated, trypsinized derivative toxin (molecular weight, 170,000) of the proteolytic Okra strain of Clostridium botulinum type B on a column of Sephadex G-200 (superfine) with a buffer containing dithiothreitol and urea. Upon removal of dithiothreitol and urea by dialysis, the two fragments reassembled to reconstruct the derivative toxin molecule. Both fragments were immunogenic, and both anti-fragments neutralized type B toxin. The neutralizing activities of both anti-fragment I and anti-fragment II were, however, lower than that of the anti-derivative toxin, suggesting that the molecular integrity of derivative toxin is essential for sufficient production of the neutralizing antibody. The immunological difference found between type B toxin from a proteolytic strain and that from a nonproteolytic strain was ascribed to the antigenic difference of fragment I.

Clostridium botulinum type D toxin: purification, molecular structure, and some immunological properties

Clostridium botulinum type D progenitor toxin was purified. The addition of ribonucleic acid to the whole culture helped initial acid precipitation of the toxin. As with type B, both L (16S) and M toxins (12S) obtained from a hemagglutinin-positive strain, whereas M toxin only was produced by a hemagglutinin-negative strain. M toxin (molecular weight, 300,000) consisted of one molecule each of a toxic (molecular weight, 170,000) and a nontoxic component (molecular weight, 130,000); L toxin consisted of both components plus hemagglutinin. The specific toxicity of M toxin was 5 X 10(8) mean lethal doses per mg of N; that of L toxin was 2.4 X 10(8) mean lethal doses per mg of N. These toxins were fully or nearly fully active, but in un-nicked form. Trypsinization caused nicking in the toxic component, forming a molecule made up of two peptide chains with molecular weights of 110,000 and 60,000; there was little or no increase in toxicity. The toxic component of type D was not antigenically related to that of type C, whereas the nontoxic component was antigenically indistinguishable from that of type C. The toxicities of both L nad M toxins of the hemagglutinin-positive strain were increased twofold by trypsinization. Neither toxin contained the C2 toxic factor elaborated by C and D strain.

Activation of botulinum toxins in the absence of nicking

The derivative toxins purified from cultures of proteolytic strains of Clostridium botulinum types A and F were found to have been only partially nicked but were fully activated. Trypsinization of C. botulinum type B derivative toxin at pH 6.0 resulted in simultaneous activation and nicking, whereas at pH 4.5, activation preceded nicking. The toxin was split by trypsin at pH 6.0 into two fragments with molecular weights of 112, ooo and 57,000. The toxin contained at least three trypsin-sensitive peptide bonds, one of which was more sensitive than the others at pH 6.0. These results indicate that activation of botulinum toxins by trypsin or endogenous protease (s) is not a direct result of nicking.

Correlation between oral toxicity and in vitro stability of Clostridium botulinum type A and B toxins of different molecular sizes

The in vitro sensitivity to acid and pepsin differed markedly among Clostridium botulinum type A and B toxins of different molecular sizes. The larger the molecular size of the toxin, the higher the resistance to these agents. Tye B derivative toxin was rapidly inactivated, but the progenitor toxins resisted in vitro exposure to rat intestinal juice. The molecular dissociation of the progenitor toxins did not occur in rat intestinal juice of pH 7.0, but did occur in a buffer solution of the same pH. The oral toxicity may depend mostly on the stability of toxin molecules in the stomach and, to a less extent, in the intestine. The present results seem to justify the conclusion that C. botulinum type A and B progenitor toxins with molecular sizes larger than 16S are more potent oral toxins than 12S progenitor toxins.

Oral toxicities of Clostridium botulinum toxins in response to molecular size

Clostridium botulinum type A, B, and F toxins of different molecular sizes were fed to mice to compare the oral toxicities. The progenitor toxin, a complex of a toxic and nontoxic component, of any type was higher in oral toxicity to mice than the dissociated toxic component or the derivative toxin. The former may no doubt play a more important role in the pathogenesis of food-borne botulism. The higher oral toxicity possessed by the progenitor toxin, including the exceptionally high one found with type B-L toxin, can be explained solely by the protection afforded by the nontoxic component attached to the toxic component. The possibility of the highest oral toxicity of type B-L toxin to humans is discussed.

Isolation and molecular size of Clostridium botulinum type C toxin

A procedure is described for the purification of hemagglutinin-free Clostridium botulinum type C toxin. The toxin was purified approximately 1,000-fold from the original culture supernatant in an overall yield of 60% to a final specific toxicity of 4.4 x 10(7) minimal lethal doses/mg of protein. The toxin had a molecular weight of 141,000 and consisted of a heavy and a light chain. The molecular weights of the subunits were approximately 98,000 and 53,000. When comparing the molecular size and composition of type C toxin to that of botulinum toxins of different types, some common features may be suggested; i.e., the toxin has a molecular weight between 141,000 to 160,000 and is comprised of a heavy and a light chain linked by disulfide bonds (or bond).