Phylogeny and taxonomy of the food-borne pathogen Clostridium botulinum and its neurotoxins

Journal of Applied Microbiology 1998, 84, 5–17

A REVIEWPhylogeny and taxonomy of the food-borne pathogenClostridium botulinum and its neurotoxins

M.D. Collins and A.K. EastDepartment of Microbiology, BBSRC Institute of Food Research, Reading, UK

6078/01/97: received 29 January 1997, revised 28 April 1997 and accepted 30 April 1997

1. Summary, 5 4. Phylogenetic relationships of the botulinum2. Cl. botulinum organisms and phenotypic divisions, toxins, 9

5 5. Phylogenetic relationships of nontoxic3. Genotypic division within Cl. botulinum non-haemagglutinating proteins, 11

3.1 DNA–DNA hybridization, 7 6. Taxonomic considerations, 143.2 16S rRNA sequencing, 7 7. References, 15

—–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

1. SUMMARY covalently bound progenitor toxin complex of two or moreprotein components. Information on the evolutionary his-

Until recently, all clostridia producing neurotoxins able totories of the various non-toxic progenitor proteins is currently

cause paralysis symptomatic of botulism were deemed to belimited, although there is evidence of gene recombination. In

Clostridium botulinum. Defining Cl. botulinum on the basisparticular, chimera-like or mosaic non-toxic-non-haemag-

of this single phenotypic trait has resulted in the speciesglutinins (NTNH) genes in group I Cl. botulinum have been

encompassing metabolically very diverse organisms, and fourdescribed, and it is now apparent that the phylogeny of the

distinct phenotypic groups are recognized within this taxonNTNHs is not going to ‘mirror’ that of botulinal neurotoxins,

(designated groups I-IV). Nucleic acid hybridization and 16Salthough their genes are physically contiguous. In this article,

ribosomal RNA sequencing studies have revealed the pres-the current state of knowledge of the phylogenetics of the

ence of four phylogenetically distinct lineages within thespecies Cl. botulinum and its neurotoxins is reviewed, and a

species, which correlate with these phenotypic divisions. Inview is presented that a nomenclature based rigidly on BoNT

addition to marked phenotypic and genotypic heterogeneityproduction is no longer tenable.

between groups, the taxonomy of the species is further com-plicated by the existence of strains which are closely related, ifnot genetically identifiable, to members of each Cl. botulinum

2. CL. BOTULINUM ORGANISMS ANDgroup, but are non-toxigenic. Furthermore, strains of speciesPHENOTYPIC DIVISIONother than Cl. botulinum (viz. Cl. baratii, Cl. butyricum) have

been found which express botulinum neurotoxin (BoNT). In 1897, van Ermengem reported the isolation of an anaerobic,Great advances have been made in recent years in elu- spore-forming, toxin-producing bacillus from the remains of

cidating the nucleotide sequences of genes encoding the vari- a salted ham that had caused severe neuroparalytic illnessesous BoNT antigenic types (A through to G). Genealogical and three fatalities among 34 musicians in Belgium in 1895.trees derived from BoNTs show marked discordance with He named the organism Bacillus botulinus. As subsequentthose depicting ‘natural’ relationships inferred from 16S incidents of botulism were investigated, it was found that therRNA and phenotypic clusters, and strong evidence exists causative organisms had assorted physiological characteris-for BoNT gene transfer between some groups of Cl. botulinum tics, and sometimes the toxins had different serological speci-(e.g. groups I and II), and with non-botulinum species. Bot- ficities (e.g. Leuchs 1910; Bengston 1922; Seddon 1922;ulinum neurotoxin is produced by Cl. botulinum as a non- Landemann 1904; Gunnison and Meyer 1929; Meyer and

Gunnisson 1929). As the organisms invariably consisted ofCorrespondence to: Dr M.D. Collins, Department of Microbiology, BBSRCanaerobic spore-forming rods, they were later assigned to theInstitute of Food Research, Earley Gate, Whiteknights Road, Reading,

RG6 6BZ, UK. genus Clostridium. Today, Cl. botulinum is the taxonomic

© 1998 The Society for Applied Microbiology

6 M.D. COLLINS AND A.K. EAST

designation used for all organisms known to produce bot- organisms have an optimum temperature of 37 °C and pro-duce spores with high heat resistance (112 °C/1·23; tem-ulinum neurotoxin (BoNT) and thereby having the capability

of causing botulism in humans and animals (Prevot 1953). perature/D-value): group II for strains of type E and non-proteolytic/saccharolytic strains of type B and F; organismsStrains of Cl.botulinum are traditionally classified into seven

types (A through to G), depending on the serological speci- are psychrotrophic, growing optimally at 30 °C or less, andhave spores with low heat resistance (80 °C/0·6–1·25): groupficity of the neurotoxin produced (Hatheway 1990, 1992).

Toxin type is determined by neutralization of biological III organisms produce either type C or D toxin and aregenerally non-proteolytic; strains of this group grow opti-activity with specific antitoxin reagents. Although most Cl.

botulinum organisms produce a single BoNT, it is now recog- mally at 40 °C and have an intermediate spore resistance toheat (104 °C/0·1–0·9): group IV strains produce type G toxin,nized that some strains produce mixtures of two toxin types

(viz. AF (Gimenez and Cirrarelli 1970b); AB (Poumeyrol are asaccharolytic and differ from other groups in not pro-ducing lipase; organisms grow optimally at 37 °C and theiret al. 1983; Cordoba et al. 1995), and BF (Hatheway and

McCroskey 1989)) and that many type A strains harbour spore heat resistance is similar to that of group III. Types A,B, E and F are mainly involved in botulism in man whereassilent/cryptic type B neurotoxin genes (Franciosa et al. 1994;

Cordoba et al. 1995). types C and D are responsible for botulism in animals(Table 2). Although the isolation of type G organisms fromDefining Cl. botulinum solely on the basis of BoNT pro-

duction has resulted in the species encompassing a range of autopsy specimens has been reported, evidence that botulismwas the cause of death has not been demonstrated.metabolically very diverse micro-organisms (Hatheway 1990,

1992). Holdeman and Brooks (1970) divided Cl. botulinum From a purely taxonomic viewpoint, it is recognized thata nomenclature based on BoNT production is unsatisfactorytypes A through F into three metabolic groups. Strains of type

G, originally isolated from soil and described by Gimenez and and that in most circumstances, the different metabolicgroups of Cl. botulinum would be assigned to different species.Ciccarelli (1970a), were subsequently placed in a fourth group

by Smith and Hobbs (1974). The composition of the four In view of its utility to medical microbiologists, and in orderto avoid possible confusion, however, there has been angroups (designated I to IV) within Cl. botulinum (Table 1) are

as follows: group I for strains of type A, proteolytic strains understandable reluctance to initiate a major change innomenclature. The taxonomy of Cl. botulinum has been fur-of type B and F, and dual toxin types AB, AF and BF;

Table 1 Phenotypic differences between organisms capable of producing botulinum neurotoxina

—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Characteristic Groups—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

I II III IV Cl. butyricum Cl. baratii—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––Toxin types A, B, F B, E, F C, D G E FProteolysis + − − + − −Liquefaction of gelatin + + + + − −Fermentation of:

Glucose + + + − + +Fructose 2 + 2 − + +Mannose − + + − + +Maltose 2 + 2 − + +Sucrose − + − − + +Trehalose − + − − + −

Lipase + + + − − −Metabolic acidsb A, iB, B, iv A, B A, P, B A, iB, B, iV, PA A, B A, BOptimal growth temperature 35–40 °C 18–25 °C 40 °C 37 °C 30–37 °C 30–45 °CMinimum growth temperature 10+ °C 3·3 °C 15 °C 10 °CSpore heat resistance 112 °C 80 °C 104 °C 104 °C

(temperature/D-value) 1·23 0·6–1·25 0·1–0·9 0·8–1·12—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––a For biochemical reactions: +, all strains are positive; −, all strains are negative; 2, some strains are positive and some are negative.For temperature values, none listed if not readily available in the literature. Data from Hatheway (1990, 1992).b Metabolic acids: A, acetic; P, propionic; B, butyric; iB, isobutyric; iV, isovaleric; PP, phenylpropionic (hydrocinnamic); PA,phenylacetic.

© 1998 The Society for Applied Microbiology, Journal of Applied Microbiology 84, 5–17

PHYLOGENETICS OF CL. BOTULINUM AND ITS NEUROTOXINS 7

Table 2 Diseases and types of Clostridium botulinum—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Toxigenic typesa Species mainly affected Commonest vehicles—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––A Man (also wound and infant botulism), chickens Home-canned vegetables, fruits, meat and fish

(‘limberneck’)B Man (also wound and infant botulism), horses and cattle Prepared meats, especially pork productsCa Aquatic birds (Western duck sickness) Rotting vegetation of alkaline marshes, invertebratesCb Cattle (Midland cattle disease), horses (forage poisoning) Toxic food, carrion, pork liverD Cattle (lamziekte) CarrionE Man, fish, aquatic birds Marine products and fish productsF Man (also infant botulism) Meat productsG Unknown Soil—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––a Ca strains produce C1, a neurotoxin, and C2, which is not a neurotoxin but is lethal in mice, ducks and geese; congestion andhaemorrhage in the lung and dripping of the nares are observed in birds injected with C2 toxin, but paralytic signs of botulism are absent. Cb

strains produce only C2 toxin.

ther complicated in recent years by reports that some strains two genetic groups of toxigenic organisms are intermediatelyrelated to each other, as well as to Cl. novyi (types A and B)of Cl. butyricum and Cl. baratii produce neurotoxins that

cross-react with BoNT types E (Aureli et al. 1986; McCro- and Cl. haemolyticum (Nakamura et al. 1983). Within groupIV clostridia, all Cl. botulinum type G strains examined toskey et al. 1986) and F (Hall et al. 1985; McCroskey et al.

1991), respectively. Thus, six phenotypically distinct groups date are highly related to each other and represent a singlegenomic species (Suen et al. 1988b). Some non-toxigenicof clostridia are now known to be capable of producing bot-

ulinum neurotoxin. strains designated Cl. subterminale and Cl. hastiforme are alsomembers of this group (×70% relatedness) (Suen et al.1988b). However, some other phenotypically similar non-

3. GENOTYPIC DIVISION WITHIN CL.toxigenic strains are intermediately related (including Cl.

BOTULINUMsubterminale type strain) and correspond to closely related butnevertheless different species (Suen et al. 1988b). DNA-

3.1 DNA–DNA hybridizationDNA pairing studies have shown that BoNT synthesizingstrains of Cl. baratii and Cl. butyricum are genetically highlyThe genetic relatedness of Cl. botulinum groups I to IV and

their non-toxigenic counterparts has been the subject of sev- related to non-toxigenic reference strains (including typestrains) of their respective species (Suen et al. 1988b). Theseeral chromosomal DNA-DNA pairing studies (Lee and Rie-

mann 1970a,b; Nakamura et al. 1977, 1983; Suen et al. 1988b). toxigenic strains are genetically remote from group I-IV Cl.botulinum strains (Suen et al. 1988b).In general, all strains, regardless of toxin types within groups

I (i.e. types A, B, F) or II (i.e. types B, E, F), are highly related,while strains belonging to different phenotypic groups exhibit

3.2 16S rRNA sequencinglow relatedness, even though some produce the same toxintype (e.g. type B strains of group I and II). Non-toxigenic Although nucleic acid hybridization studies, as long ago as the

1970s, demonstrated considerable genotypic heterogeneityvariants of phenotypically similar organisms generally showhigh or intermediate relatedness to their toxigenic counter- within Cl. botulinum (Lee and Reimann 1970a,b; Nakamura

et al. 1977, 1983), it is only with the advent of 16S rRNAparts. For example, some non-toxigenic proteolytic strainsidentifiable as Cl. sporogenes exhibit high relatedness (×70%) gene sequencing that the full extent of the genealogical diver-

sity within this species-complex has become apparent (Hut-to group I Cl. botulinum, whilst others (including the typestrain of Cl. sporogenes), although closely related, represent son et al. 1993). 16S rRNA gene sequence analysis is the ‘gold-

standard’ method for determining phylogenetic relationships,different genomic species (Lee and Riemann 1970b; Wu etal. 1972; Nakamura et al. 1977). A similar situation is found and this approach has revolutionized knowledge of microbial

evolution and systematics. The phylogenetics of the genuswith non-toxigenic strains which otherwise phenotypicallyresemble non-proteolytic Cl. botulinum. Many of these strains Clostridium has been systematically investigated in recent

years and it is now evident that the genus is not a mono-genotypically clearly correspond to Group II Cl. botulinumbut do not produce BoNT. Some genetic diversity exists phyletic group but actually consists of more than 20 genera

and probably embraces several families (e.g. Collins et al.among group III strains (Nakamura et al. 1983). At least



1994). 16S rRNA gene sequencing (Collins et al. 1994) has similarity) and, together with their non-toxigenic counter-parts, form a single phylogenetic unit. The closest phylo-shown that Cl. botulinum groups I-IV and other clostridial

species producing BoNT are members of a clade designated genetic relative of group I Cl. botulinum and Cl. sporogenesis Cl. oceanicum. Similarly, non-proteolytic or group II Cl.Clostridium cluster I which corresponds to the genus Clo-

stridium sensu stricto. A tree depicting the phylogenetic inter- botulinum types B, E and F (and non-toxigenic variants)form a distinct line quite separate from other saccharolyticrelationships of the various Cl. botulinum types and some

cluster I Clostridium species is shown in Fig. 1. It is evident clostridia and are phylogenetically far removed from group ICl. botulinum. 16S rRNA sequencing shows that group IIIfrom the tree that the phylogenetic subdivisions of Cl. bot-

ulinum based on 16S rRNA directly ‘mirror’ the subdivisions Cl. botulinum types C and D are phylogenetically closelyrelated to each other (approximately 99% 16S rRNA(I-IV) based on phenotype. Within rRNA Clostridium cluster

I, organisms of Cl. botulinum phenotypic groups I-IV form sequence similarity) and represent a third lineage. Althoughtypes C and D organisms are highly related, the reportedfour distinct phylogenetic lineages (Hutson et al. 1993). All

strains, regardless of toxin type, within proteolytic or group approximately 1% sequence divergence between two strainsexamined, strongly indicates species heterogeneity withinI Cl. botulinum (i.e. types A, B and F) are highly related to

each other (approximately 99 · 6–100% 16S rRNA sequence group III organisms. Clostridium novyi has been shown to be

Fig. 1 Dendrogram showing thephylogenetic position of Clostridiumbotulinum groups I-IV within cluster IClostridium based on 16S rRNA genesequences. The tree was constructedusing the neighbour-joining methodand is based on a comparison of acontinuous stretch ofapproximately 1340 bases. Bootstrapproportions of confidence (−90%)based on 200 replicates are given at thebranching points



a close phylogenetic relative, albeit a different species, of resemble each other, and their similarity to tetanus toxin(tetanospasmin, TeNT), which also exhibits zinc endo-group III Cl. botulinum. Surprisingly, two thermophilic clo-

stridia, Cl. thermopalmarium and Cl. thermobutyricum, possess peptidase activity, is striking (Elmore et al. 1995). Figure 2shows two trees depicting the phylogenetic interrelationshipsa distant but significant (as indicated by high bootstrap value)

phylogenetic affinity with group III Cl. botulinum. Group IV of the L and H chains of the various BoNT types. It is evidentthat the BoNT and 16S rRNA derived trees are incompatible,Cl. botulinum is phylogenetically quite separate from the other

three Cl. botulinum groups. Cl. subterminale is phylo- and that major topological differences reinforce the dis-cordance between toxin type and ‘natural’ (genotypic) divi-genetically a close relative of Cl. botulinum type G, with

some strains being genealogically indistinguishable. With the sions within the Cl. botulinum species complex. The mostsignificant finding to emerge from comparative phylogeneticpossible exception of a poorly supported association with Cl.

estertheticum, Cl. botulinum type G and Cl. subterminale show analyses is the strong likelihood of BoNT gene trans-missibility within the Cl. botulinum species complex, and alsono close affinity with any other clostridial species (Collins et

al. 1994). Clostridium hastiforme, which phenotypically with non-botulinum species. In particular, the BoNT/Esof Cl. botulinum and Cl. butyricum are very highly relatedresembles Cl. botulinum type G and C.subterminale, is phylo-

genetically far removed from these species and has recently (approximately 97% amino acid identity), as are BoNT/Bsof proteolytic (group I) and non-proteolytic (group II) Cl.been shown to be a member of the genus Tissierella (Farrow

et al. 1995). 16S rRNA sequencing has also shown that BoNT botulinum (approximately 93% amino acid identity). Suchexceptionally high levels of sequence relatedness could besynthesizing strains of Cl. baratii and Cl. butyricum are genea-

logically indistinguishable from non-toxigenic strains of their indicative of recent lateral gene transfer between these geno-types. Similarly, it is evident from 16S rRNA sequence analy-respective species (Hutson et al. 1993) consistent with DNA-

DNA reassociation studies (Suen et al. 1988a). Clostridium sis that BoNT/F producing organisms (viz. group I andII Cl. botulinum and Cl. baratii) form three quite separatetetani, which produces tetanus toxin (TeNT), although also

a member of Clostridium cluster I, is phylogenetically distinct phylogenetic lineages (Fig. 1). However, the type F toxins ofthese species are of common ancestry, forming a genea-from the four Cl. botulinum groups, including toxigenic Cl.

baratii and Cl. butyricum. logically tight grouping (Fig. 2). The type F toxins fromgroup I and II Cl. botulinum are more closely related to eachother (approximately 87% amino acid identity) than either

4. PHYLOGENETIC RELATIONSHIPS OFare to Cl. baratii (approximately 71–74% amino acid identity).

BOTULINUM TOXINSIt is also apparent from the treeing analysis that the threetype F toxins are more loosely associated with each otherBotulinum neurotoxin is among the most toxic of all naturally

occurring substances. Estimates of the amount of type A than are the type E toxins of Cl. botulinum and Cl. butyricum.Although there is little doubt that the three type F toxinstoxin needed to cause death in man vary between 0·1 and

1·0 mg, corresponding to between 300 and 30 000 mouse are of common ancestry, if BoNT gene transfer has in factoccurred between these three species (i.e. Cl. botulinumintraperitoneal LD50 doses (Schantz and Sugiyama 1974).

The active botulinum neurotoxins have molecular masses groups I and II and Cl. baratii), the significantly lower identi-ties indicate that such an event occurred earlier in evol-of about 150 kDa, each consisting of a heavy (H) chain

(approximately 100 kDa) and a light (L) chain (approximately utionary history (assuming a similar rate of nucleotide changein the genes). It is evident from comparative sequence analysis50 kDa) linked by a disulphide bridge. The toxins are initially

synthesized as single chain polypeptides which are enzy- that type G toxin, although quite separate from that of typeB, nevertheless exhibits a loose but statistically significantmatically cleaved to form the active dichain structure. The

various BoNTs, which all have a common zinc-binding motif (bootstrap value 100%) phylogenetic association (in muchthe same way as BoNT/E and BoNT/F exhibit a distant butin their L chains, represent a distinct group of zinc-dependent

proteases (Schiavo et al. 1992). The BoNT endopeptidases significant evolutionary affinity). Similarly, BoNT/C andBoNT/D share a loose phylogenetic association. Recently,specifically cleave proteins involved in docking and fusion

of synaptic vesicles, thereby blocking neuroexocytosis and some strains of group III Cl. botulinum have been shownto contain genes encoding a hybrid or mosaic neurotoxinneurotransmitter release. Types A and E cleave SNAP-25

(Binz et al. 1994); types B, D, F and G degrade VAMP/syn- composed of part of BoNT/C and BoNT/D (Moriishi et al.1996a,b).aptobrevin (Schiavo et al. 1992, 1994; Yamasaki et al. 1994),

and type C1 degrades HPC-1/syntaxin (Blasi et al. 1993). In view of the high sequence and structural similaritiesbetween BoNT and TeNT, it is highly likely that these toxinsThe past few years have seen the publication of several

BoNT gene sequences, so that complete sequences for all Cl. arose from a single ancestral form (referred to as ancestral-NT). The gene encoding such an ancestral-NT may havebotulinum types (A through G) are now known (Table 3).

Comparative analysis has shown that all BoNT types closely originated in a species distinct from Cl. botulinum and Cl.



Table 3 Published BoNT genesequences

—–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Organism BoNT type Reference—–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––Clostridium botulinum group I BoNT/A Thompson et al. (1990);

Binz et al. (1990b)Cl. botulinum group I BoNT/A ‘infant’ Willems et al. (1993)Cl. botulinum group I BoNT/B Whelan et al. (1992)Cl. botulinum group II BoNT/B Hutson et al. (1994)Cl. botulinum group III BoNT/C Hauser et al. (1990)Cl. botulinum group III BoNT/D Binz et al. (1990a)Cl. botulinum group III BoNT/C-D ‘hybrid’ Moriishi et al. (1996a,b)Cl. botulinum group II BoNT/E Poulet et al. (1992);

Whelan et al. (1992)Cl. botulinum group I BoNT/F Elmore et al. (1995)Cl. botulinum group II BoNT/F East et al. (1992)Cl. botulinum group IV BoNT/G Campbell et al. (1993)Cl. botulinum type A(B) group I BoNT/B ‘silent’ Hutson et al. (1996)Cl. butyricum BoNT/E Poulet et al. (1992)Cl. baratii BoNT/F Thompson et al. (1993)—–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Fig. 2 Dendrograms showing the phylogenetic relationships of the L and H BoNT chains. The trees were based on complete aminoacid sequences and constructed by using the neighbour-joining method. Bootstrap proportions of confidence (based on 200 replicates) aregiven at branching points

tetani, or, more likely, in an ancestor of a recognized Cl. organisms, are depicted in Figs 3 and 4, respectively. Severalpoints are evident from the two ancestral schemes shown. (i)botulinum organism. The NTs probably diverged from the

ancestral-NT gene in such an organism as a consequence of The transfer of ancestral-NT to multiple lines viz. Cl. tetaniand Cl. botulinum group III, and Cl. botulinum group I (theits partition into three or four clostridial subpopulations. The

occurrence of the ancestral-NT gene in either group I or II latter dependent on the origin of ancestral-NT in group IICl. botulinum) Figs 3 and 4. (ii) Two distinct BoNT sublinesorganisms most readily explains the dispersion of the various

NT types. Two speculative evolutionary scenarios for the (designated BoNT/’B-G’ and BoNT/’E-F’) would have hadto evolve from the ancestral-NT to account for the knownancestral origins of recognized Cl. botulinum types (including

dual toxin types), based on current phylogenetic knowledge distribution of BoNT types in groups I or II Cl. botulinumand their phylogenetic affinities. In view of the significantand the possible occurrence of ancestral-NT in group I and II



Fig. 3 Scheme depicting the ancestral relationships of recognized clostridial NTs based on ancestral-NT originating in group IClostridium botulinum

phylogenetic association between types B and G toxins, it is (Fig. 3), then dual toxin types (AB, AF and BF) probablyarose by BoNT gene transfer between group I organismshighly likely that the gene encoding the latter in group IV

Cl. botulinum originated from this BoNT/‘B-G’ subline. (iii) (there is strong evidence of recombination involving genesencoding progenitor complex proteins in group I organismsSequence divergence values are consistent with Cl. baratii

acquiring the gene encoding type F toxin relatively early (i.e. (see below)).before group II Cl. botulinum acquired BoNT/F from groupI in Fig. 3 or vice versa in Fig. 4). (iv) Assuming a similar rate

5. PHYLOGENETIC RELATIONSHIPS OFof nucleotide change (and there is no evidence contrary to

NON-TOXIC NON-HAEMAGGLUTINATINGthis for BoNT/E and F), Cl. butyricum acquired the BoNT/E

PROTEINSgene relatively recently (in comparison to Cl. baratii) in evo-lutionary terms. (v) Both ancestral trees are entirely consistent Botulinum neurotoxin is produced by Cl. botulinum as a non-

covalently bound complex of two or more proteinwith the known distribution of BoNT types A, B and F, andcombinations thereof, in group I Cl. botulinum. For example, components. The nature of the complex, which can include

haemagglutinating proteins (HA) as well as a non-toxic non-in Fig. 4, the occurrence of group I Cl. botulinum types Band F are consistent with non-toxigenic Cl. sporogenes-like haemagglutinating protein (NTNH), ranges in size from the

M-(300 kDa, all serotypes except G, no haemagglutinatingorganisms acquiring these genes from group II Cl. botulinum(or type A organisms acquiring the genes and subsequently activity), to the L-(500 kDa, serotypes A, B, C, D and G)

and finally, the LL-(900 kDa, serotype A) forms. The non-loosing the gene encoding BoNT/A). Similarly, if toxigenicCl. botulinum type A acquired these genes (i.e. types B and toxic proteins enhance toxin potency and are thought to

increase the stability of BoNT in the gastro-intestinal tract.F) from group II Cl. botulinum, then the occurrence of groupI dual toxin types (AB, AF and BF) becomes readily explic- The genes encoding these non-toxin proteins are clustered

with that encoding BoNT. For example, the NTNH gene isable. By contrast, if ancestral-NT originated in group I



Fig. 4 Speculative scheme depicting the ancestral relationships of recognized clostridial NTs based on ancestral-NT originating ingroup II Clostridium botulinum

located immediately upstream of that encoding BoNT in all lutionary patterns of the BoNT and NTNH genes are incom-patible. In particular, there is evidence of chimera-liketoxin types, and genes for other components of complexes

have been shown to be clustered upstream of the NTNH NTNH gene sequences. Hutson et al. (1996) recently char-acterized two BoNT gene clusters encoded by neurotoxigenicgene for toxin types A, B and C. Knowledge of the structural

organization of the cluster of genes encoding BoNT-complex type A Cl. botulinum strain CDC667 containing a ‘silent’BoNT/B gene (designated type A (B)). It was found that theand associated proteins is currently far from complete.

However, five distinct gene cluster arrangements are cur- NTNH adjacent to the defective BoNT/B gene was ‘mosaic’,in that the 5? and 3?-regions of the gene showed high hom-rently described (Fig. 5).

The evolutionary histories of the various progenitor toxin ology (approximately 99% amino acid identity) with cor-responding regions of proteolytic type B NTNH, while acomponents have not been systematically investigated, but

because their genes are proximal to that encoding BoNT, it central region of about 470 amino acids was identical withthe equivalent region of the NTNH encoded in the type Ais not unreasonable to expect that their genealogy would

mirror that of the BoNTs. Although comparative sequence gene cluster. A subsequent study by East et al. (1996) hasshown such chimera-like NTNH genes are common-placeinformation is presently very limited, preliminary phylo-

genetic analyses indicate that this is not the case for the within group I Cl. botulinum organisms. Figure 6 shows exam-ples of pairwise analyses performed on NTNH sequences ofNTNHs. Of the various Cl. botulinum organisms, the

NTNHs of group I Cl. botulinum have been the most studied, proteolytic Cl. botulinum strains, and illustrates that for somepairs of sequences, there is considerable variation in thewith full gene sequences known for Cl. botulinum types A, B,

F and A(B) (type A strain harbouring silent type BoNT/B degree of homology along their lengths. Figure 7 shows threegenealogical trees constructed from different regions of thegene). It is evident that for group I Cl. botulinum, the evo-



Fig. 5 Diagrammatic representation of the arrangement of genes encoding BoNT- complex and associated proteins. Figure summarizedfrom East and Collins (1994); Hauser et al. (1994); East et al. (1996); Bhandari et al. (1996)

Fig. 6 Pairwise analysis showing amino acid identity within the NTNHs encoded by strains of proteolytic Clostridium botulinum. Forstrain CDC667, 667 ‘A’ refers to the NTNH encoded upstream of BoNT/A gene and 667 ‘B’ refers to that upstream of the silent BoNT/Bgene. Other strains: A1 � 62 A; Bp proteolytic type B � NCTC 7273; A2 � infant type A Kyoto-F; Fp � Langeland, 91–100%; 81–90%; 71–80%; 61–70%; ¾60%. Ž, 91–100%; ,, 81–90%; �>> , 71–80%; �7 , 61–70%; �, ¾60%

NTNHs. Major differences between the trees reinforce the gene may be a ‘hot spot’ for such events within the BoNTcomplex cluster of genes. However, irrespective of the pos-incongruence in the pattern of observed relationships along

the length of the NTNHs. Such marked discordance is sible mechanisms, it is evident that the genealogy of theNTNHs does not mirror that of the BoNTs. This being theindicative of recombination and suggests that the NTNH



Fig. 7 Trees revealing the discordancein relationships along the length ofNTNHs: (i) residues 1–550; (ii) residues551–1020; (iii) residues 1021-C-terminus.NTNH sequences from the followingwere compared: proteolytic Cl.botulinum type A strains 62 A (A1) andKyoto-F (A2), proteolytic type Bstrain NCTC 7273 (Bp), type A(B)strain CDC 667: ‘A’ cluster (667 ‘A’) and‘B’ cluster (667 ‘B’), and non-proteolyticCl. botulinum type F strain Eklund 202F(Fnp). The numbers on the tree indicatebootstrap values for branch points;only values of ×90 are shown

case, it also seems likely that the phylogenetics of other rejected the name Cl. parabotulinum and proposed that allorganisms producing botulinum neurotoxin, and therebyassociated genes will also differ markedly from those of the

BoNTs. More comparative sequence data are clearly required having the capability of causing botulism in humans oranimals, be designated Cl. botulinum. Distinction of varianton a large number of strains (including non-botulinum species

such as Cl. baratii), and on the full complement of associated strains based on proteolytic activity that produce type Aand B toxins was made by appending var. ovolyticus or var.genes, before a complete understanding of the phylogenetics

of the botulinum neurotoxins and non-toxic complex proteins nonovolyticus to the name (e.g. Cl. botulinum B var. non-ovolyticus). Although this nomenclature (Prevot 1953) isis to be realized.retained to this day, there is an increasing awareness that theemphasis placed on BoNT production is resulting in so many

6. TAXONOMIC CONSIDERATIONSscientific inconsistencies that this may no longer be tenable.For example, since the recommendation of Prevot in 1953,van Ermengem (1896, 1897) was the first to establish that

botulism is caused by a toxin consumed in food and isolated several other toxin types have been described, such as type G,an organism with grossly different characteristics. In addition,the organism responsible. Soon afterwards, it became appar-

ent that the organisms capable of causing botulism were the present nomenclature of Cl. botulinum has been irrevo-cably complicated by the discovery of neurotoxigenic organ-physiologically very diverse. Indeed, as early as 1910, Leuchs

noted that the van Ermengem’s organism (Ellizelles strain) isms genetically identifiable as Cl. baratii and Cl. butyricum(Hall et al. 1985; Aureli et al. 1986; McCroskey et al. 1986,had a lower optimum growth temperature than isolates

(Darmstadt) from a subsequent outbreak of botulism reported 1991). As outlined earlier, genomic DNA-DNA pairing stud-ies have shown BoNT-producing strains of Cl. baratii andby Landemann (1904). Although the van Ermengem and

Landemann strains are no longer available, it is evident from Cl. butyricum possess ×70% relatedness with their respectivenon-toxigenic counterparts and as such, conform to the inter-the early literature that the former corresponds to non-pro-

teolytic (group II) type B and the latter to proteolytic (group nationally recommended definition of a genomic species(Wayne et al. 1987). However, by producing BoNT, theseI) type A. Additional toxin and physiological types were

described as other organisms were recovered from incidents strains are according to international nomenclature Cl. bot-ulinum! If the latter (Prevot 1953) is adhered to, then shouldof botulism in animals and man, and from the environment.

For example, in the 1920s, organisms causing botulism in not a further two phenotypic divisions (group V and VI)within Cl. botulinum be established? Recent advances in ourchickens in the United States (Bengston 1924) and cattle in

Australia (Seddon 1922) were due to type C whereas an understanding of the genealogical relationships within Cl.botulinum and its toxins have added to the scientific incon-outbreak of botulism in cattle in South Africa was caused

by type D (Meyer and Gunnison 1929). Bengtson (1922) sistencies of the nomenclature recommended by Prevot(1953). In particular, 16S rRNA sequencing has revealedproposed making a distinction between non-proteolytic and

proteolytic organisms by designating two species names, Cl. exceptional evolutionary distances between Cl. botulinumgroups I-IV and the phylogenetic absurdity of classifyingbotulinum and Cl. parabotulinum, respectively. Prevot (1953)



these in a single species (Hutson et al. 1993; Collins et al. neurotoxin-producing isolates of Cl. baratii and Cl. butyricumto retain their correct species name by becoming Cl. baratii1994). Indeed, in terms of genetic distance, the divergences

between the four Cl. botulinum groups are in some cases var. BoNT/F and Cl. butyricum var. BoNT/E, respectively.Although some genotypic heterogeneity would undoubtedlysignificantly greater than that found between some species

of different Gram-positive genera (e.g. Bacillus subtilis and remain in some of the four species (e.g. group III), this systemwould go a long way towards resolving the gross phylogeneticStaphylococcus aureus). Doubts about the wisdom of defining

a species solely on the basis of toxin production also stem absurdities of the present classification. It is important toemphasize that such a revised classification would also cor-from a growing awareness of the strong likelihood of toxin

gene transmissibility. For example, genes coding for toxin relate with the four universally recognized phenotypes andtherefore present no problems for clinical and veterinarycomplexes are associated with phages in types C and D (group

III), and possibly type E, and on a plasmid in type G Cl. diagnostic laboratories.botulinum (Hauser et al. 1992; Zhou et al. 1995). It is notknown if genes for toxin are also associated with transmissible

7. REFERENCESgenetic elements in groups I and II (types A, B and F).However, it is now clear from comparative sequence analysis Aureli, P., Fenecia, B., Pasolini, M., Bianfranceschi, L.M.,that BoNTs have been laterally transferred between divi- McCroskey, L.M. and Hatheway, C.L. (1986) Two cases of typesions/rRNA lineages (e.g. BoNT/E of group II Cl. botulinum E infant botulism caused by neurotoxigenic Clostridium butyricumand Cl. butyricum; BoNT/B and F of groups I and II Cl. in Italy. Journal of Infectious Diseases 154, 207–211.botulinum) (see Figs 3 and 4). Further complications with Bengston, I.A. (1922) Preliminary note on a toxin producing ana-

erobe isolated from the larvae of Lucilia caesar. Public Healthdefining Cl. botulinum on the basis of toxigenicity arise fromReports (USA) 37, 164–170.the instability of this characteristic for many strains. Non-

Bengtson, I.A. (1924) Studies on organisms concerned as causativetoxigenic strains genotypically identifiable with the four Cl.factors in botulism. Hygiene Laboratory Bulletin136. Washingtonbotulinum groups are well documented. The nomenclaturalDC: US Public Health Service.status of these organisms is very confused. Should a non-

Bhandari, M., Campbell, K.D., Collins, M.D. and East, A.K. (1997)toxigenic isolate derived from a toxigenic strain retain theMolecular characterization of genes encoding the botulinum

species name of its parent? Not so, according to the prevailing neurotoxin complex in Clostridium botulinum (Clostridium argen-nomenclature of Cl. botulinum! In which case, what should tinense) type G. Current Microbiology, in press.such strains be designated? Similarly, how does one properly Binz, T., Blasi, J., Yamasaki, S. et al. (1994) Proteolysis of SNAP-identify an organism that has lost its toxigenicity just prior 25 by types E and A botulinal neurotoxins. Journal of Biologicalto isolation? Even more problematic is the occurrence of non- Chemistry 269, 1617–1620.

Binz, T., Kurazono, H., Popoff, M.R. et al. (1990a) Nucleotidetoxigenic organisms which contain cryptic/silent genes orsequence of the gene encoding Clostridium botulinum neurotoxinsequences coding for BoNT (Franciosa et al. 1994). As func-type D. Nucleic Acids Research 18, 5556.tional toxin is not expressed, these strains are presumably, by

Binz, T., Kurazono, H., Wille, M., Frevert, J., Wernars, K. anddefinition, not Cl. botulinum!Niemann, H. (1990b) The complete sequence of botulinumIn view of the above complexities and contradictions, aneurotoxin type A and comparison with other clostridial neu-nomenclature rigidly based on the production of neurotoxinrotoxins. Journal of Biological Chemistry 265, 9153–9158.

is probably no longer viable. In view of the clinical and Blasi, J., Chapman, E.R., Yamasaki, S., Binz, T., Niemann, H. andveterinary importance of Cl. botulinum, it is important that Jahn, R. (1993) Botulinum neurotoxin C1 blocks neurotransmitterany future nomenclature should not only take into account release by means of cleaving HPB-1/syntaxin. EMBO Journalphenotypic and genotypic data, but also, botulinum neuro- 12, 4821–4828.toxin production. A sensible solution would be to designate Campbell, K.D., Collins, M.D. and East, A.K. (1993) Nucleotide

sequence of the gene coding for Clostridium botulinum (Clostridiumthe four phenotypic/phylogenetic groups (I-IV) separateargentinense) type G neurotoxin: genealogical comparison withspecies. A precedent has been set for a change of this kindother clostridial neurotoxins. Biochimica et Biophysica Acta 1216,due to the earlier proposal of Cl. argentinense for Cl. botulinum487–491.type G including some non-toxigenic strains previously

Collins, M.D., Lawson, P.A., Willems, A. et al. (1994) The phy-designated Cl. subterminale and Cl. hastiforme (Suen et al.logeny of the genus Clostridium. Proposal of five new genera1988). Furthermore, the production of neurotoxin by strainsand eleven new species combinations. International Journal of

could be indicated by some term such as var. (e.g. var. Systematic Bacteriology 44, 812–816.BoNT/A, B, C, D, E, F or G) after the species epithet. Cordoba, J.J., Collins, M.D. and East, A.K. (1995) Studies onStrains which are genotypically identifiable as members of the gene encoding botulinum neurotoxin type A of Clostridiumthese species but are, for instance, unable to express or have botulinum from a variety of sources. Systematic and Applied Micro-lost the capability to produce neurotoxin, would lack the var. biology 18, 12–22.

East, A.K., Bhandari, M., Stacey, J.M., Campbell, K.D. and Collins,BoNT designation. This system would also allow botulinum



M.D. (1996) Organization and phylogenetic interrelationships of Hauser, D., Eklund, M.W., Boquet, P. and Popoff, M.R. (1994)Organization of the botulinum neurotoxin C1 gene and its associ-genes encoding components of the botulinum toxin complex in

proteolytic Clostridium botulinum types A, B and F: evidence of ated non-toxic protein genes in Clostridium botulinum C 468.Molecular and General Genetics 243, 631–640.chimeric sequences in the gene encoding the non-toxic-non-

hemagglutinin component. International Journal of Systematic Hauser, D., Eklund, M.W., Kurazono, H. et al. (1990) Nucleotidesequence of Clostridium botulinum C1 neurotoxin. Nucleic AcidsBacteriology 46, 1105–1112.

East, A.K. and Collins, M.D. (1994) Conserved structure of genes Research 18, 4924.Hauser, D., Gilbert, M., Bouquet, P. and Popoff, M.R. (1992)encoding components of botulinum neurotoxin complex M and

the sequence of the gene coding for the nontoxic component in Plasmid localisation of a type E botulinal neurotoxin gene hom-ologue in toxigenic Clostridium butyricum strains, and absence ofnon-proteolytic Clostridium botulinum type F. Current Micro-

biology 29, 69–77. this gene in non-toxigenic C.butyricum. FEMS Microbiology Let-ters 99, 251–256.East, A.K., Richardson, P.T., Allaway, D., Collins, M.D., Roberts,

T.A. and Thompson, D.A. (1992) Sequence of the gene encoding Holdeman, L.V. and Brooks, J.B. (1970) Variation among strains ofClostridium botulinum and related clostridia. In Proceedings of thetype F neurotoxin of Clostridium botulinum. FEMS Microbiology

Letters 96, 225–230. First U.S.-Japan Conference of Toxic Microorganisms ed. Herzberg,M. pp. 278–286. Washington DC: US Government PrintingElmore, M.J., Hutson, R.A., Collins, M.D., Bodsworth, N.J.,

Whelan, S.M. and Minton, N.P. (1995) Nucleotide sequence of Office.Hutson, R.A., Collins, M.D., East, A.K. and Thompson, D.E.the gene coding for proteolytic (group I) Clostridium botulinum

type F neurotoxin: genealogical comparison with other clostridial (1994) Nucleotide sequence of the gene coding for non-proteolyticClostridium botulinum type B neurotoxin: comparision with otherneurotoxins. Systematic and Applied Microbiology 18, 23–31.

Farrow, J.A.E., Lawson, P.A., Hippe, H., Gauglitz, U. and Collins, clostridial neurotoxins. Current Microbiology 28, 101–110.Hutson, R.A., Thompson, D.E., Lawson, P.A., Schocken-Itturino,M.D. (1995) Phylogenetic evidence that the Gram-negative non-

sporulating bacterium Tissierella (Bacteroides) praeacuta is a mem- R.P., Bottger, E.C. and Collins, M.D. (1993) Genetic inter-relationships of proteolytic Clostridium botulinum types A, B, Fber of the Clostridium subphylum of the gram-positive bacteria

and description of Tissierella creatinini sp. nov. International Jour- and other members of the Clostridium botulinum complex asrevealed by small-subunit rRNA gene sequences. Antonie vannal of Systematic Bacteriology 45, 436–440.

Franciosa, G., Ferreira, J.L. and Hatheway, C.L. (1994) Detection Leeuwenhoek 64, 273–283.Hutson, R.A., Zhou, Y., Collins, M.D., Johnson, E.A., Hatheway,of type A, B, and E botulism neurotoxin genes in Clostridium

botulinum and other Clostridium species by PCR: evidence of C.L. and Sugiyama, H. (1996) Genetic characterization of Clo-stridium botulinum type A containing silent type B neurotoxinunexpressed type B toxin genes in type A toxigenic organisms.

Journal of Clinical Microbiology 32, 1911–1917. gene sequences. Journal of Biological Chemistry 271, 10786–10792.Landemann, G. (1904) Ueber die Ursache der DarmstaedterGimenez, D.F. and Ciccarelli, A.S. (1970a) Studies on strain 84 of

Clostridium botulinum. Zentralblatt fuer Bakteriologie Para- Bohnenvergiftung. Hygiene Rundschrift 10, 449–452.Lee, W.H. and Riemann, H. (1970a) Correlation of toxic and non-sitenkunde Infektionskrankheiten und Hygiene, Abt. 1 Orig. Reihe

A 215, 212–220. toxic strains of Clostridium botulinum by DNA composition andhomology. Journal of General Microbiology 60, 117–123.Gimenez, D.F. and Ciccarelli, A.S. (1970b) Another type of Clo-

stridium botulinum. Zentralblatt fuer Bakteriologie Parasitenkunde Lee, W.H. and Riemann, H. (1970b) The genetic relatedness ofproteolytic Clostridium botulinum strains. Journal of GeneralInfektionskrankheiten und Hygiene, Abt. 1 Orig. Reihe A 215, 221–

224. Microbiology 64, 85–90.Leuchs, J. (1910) Beitraege zur Kenntnis des Toxins and AntitoxinsGunnison, J.B. and Meyer, K.F. (1929) Cultural study of an inter-

national collection of Clostridium botulinum and parabotulinum. des Bacillus botulinus. Zeitschrift Hygiene Infektionskrankheiten 65,55–84.Journal of Infectious Diseases 45, 119–134.

Hall, J.D., McCroskey, L.M., Pincomb, B.J. and Hatheway, C.L. McCroskey, L.M., Hatheway, C.L., Fenicia, L., Pasolini, B. andAurelia, P. (1986) Characterization of an organism that produces(1985) Isolation of an organism resembling Clostridium barati

which produces type F botulinum toxin from an infant with type E botulinal toxin but which resembles Clostridium botulinumfrom the feces of an infant with type E botulism. Journal ofbotulism. Journal of Clinical Microbiology 21, 654–655.

Hatheway, C.L. (1990) Toxigenic clostridia. Clinical Microbiology Clinical Microbiology 23, 201–202.McCroskey, L.M., Hatheway, C.L., Woodruff, B.A., Greenberg,Reviews 3, 66–98.

Hatheway, C.L. (1992) Clostridium botulinum and other clostridia J.A. and Jurgenson, P. (1991) Type F botulism due to neuro-toxigenic Clostridium barati from an unknown source in an adult.that produce botulinum neurotoxin, In Clostridium botulinum—

Ecology and Control in Foods ed. Hauschild, A.H.W. and Dodds Journal of Clinical Microbiology 29, 2618–2620.Meyer, K.F. and Gunnison, J.B. (1929) South African culturesK.L. p. 3–20. New York: Marcel Dekker.

Hatheway, C.L. and McCroskey, L.M. (1989) Unusual neuro- of Clostridium botulinum and parabotulinum. XXXVIL (with adescription of Cl. botulinum type D, n. sp.). Journal of Infectioustoxigenic clostridia recovered from human fecal specimens in the

investigation of botulism. In Proceedings of the 5th International Diseases 45, 106–118.Moriishi, K., Koura, M., Abe, N. et al. (1996a) Mosaic structuresSymposium on Microbial Ecology: Recent Advances in Microbial

Ecology ed. Hattori, T., Ishida, Y., Maruyama, Y., Morita, R.Y. of neurotoxins produced from Clostridium botulinum types C andD organisms. Biochimica et Biophysica Acta 1307, 123–126.and Uchida, A. pp. 477–481. Japan Scientific Societies Press.



Moriishi, K., Koura, M., Fujii, N. et al. (1996b) Molecular cloning geneous group composed of all strains of Clostridium botulinumtoxin type G and some nontoxigenic strains previously identifiedof the gene encoding the mosaic neurotoxin, composed of partsas Clostridium subterminale or Clostridium hastiforme. Internationalof botulinum neurotoxin types C1 and D, and PCR detection ofJournal of Systematic Bacteriology 38, 375–381.this gene from Clostridium botulinum type C organism. Applied

Thompson, D.E., Brehm, J.K., Oultram, J.D. et al. (1990) Theand Environmental Microbiology 92, 662–667.complete amino acid sequence of Clostridium botulinum type ANakamura, S., Kimura, I., Yamakawa, K. and Nishida, S. (1983)neurotoxin, deduced by nucleotide sequence analysis of theTaxonomic relationships among Clostridium novyi types A and B,encoding gene. European Journal of Biochemistry 189, 73–81.Clostridium haemolyticum and Clostridium botulinum type C. Jour-

Thompson, D.E., Hutson, R.A., East, A.K., Allaway, D., Collins,nal of General Microbiology 129, 1473–1479.M.D. and Richardson, P.T. (1993) Nucleotide sequence of theNakamura, S., Okado, I., Nakashio, S. and Nishida, S. (1977) Clo-gene coding for Clostridium barati type F neurotoxin: comparisonstridium sporogenes isolates and their relationship to C. botulinumwith other clostridial neurotoxins. FEMS Microbiology Lettersbased on deoxyribonucleic acid reassociation. Journal of General108, 175–182.Microbiology 100, 395–401.

van Ermengem, E. (1896) Untersuchungen uber Falle von Fle-Poulet, S., Hauser, D., Quanz, D., Niemann, H. and Popoff, M.R.ischvergiftung mit Symptomen von Botulismus. Zentralblatt fuer(1992) Sequence of the botulinal neurotoxin derived from Clo-Bakteriologie Parasitenkunde Infektionskrankheiten und Hygiene I,stridium botulinum type E (strain Beluga) and Clostridium butyricumAbt. Orig. 19, 442–444.(strain ATCC 43181 and ATCC 43755). Biochemical and Biophy-

van Ermengem, E. (1897) Ueber einen neuen anaeroben Bacillussical Research Communications 183, 107–113.und seine Beiziehungen zum Botulismus. Zeitschrift HygienePoumeyrol, M., Billon, J., DeLille, F., Haas, C., Marmonier, A.Infektionskrankheiten 26, 1–5.and Sebald, M. (1983) Intoxication botulique mortelle due a une

Wayne, L.G., Brenner, D.J., Colwell, R.R. et al. (1987) Report ofsuoche de Clostridium botulinum de type AB. Malady and Infectionthe ad hoc committee on reconciliation of approaches to bacterial13, 750–754.systematics. International Journal of Systematic Bacteriology 37,Prevot, A.R. (1953) Rapport d’introduction du president du sous-463–464.comite Clostridium pour l’unification de la nonmenclature des

Whelan, S.M., Elmore, M.J., Bodsworth, N.J., Atkinson, T. andtypes toxigeniques de C. botulinum. International Bulletin of Bac-Minton, N.P. (1992) The complete amino acid sequence of theterial Nomenclature 3, 120–123.Clostridium botulinum type E neurotoxin, deduced by nucleotide

Schantz, E.J. and Sugiyama, H. (1974) The toxins of Clostridiumsequence analysis of the encoding gene. European Journal of Bio-

botulinum. Essays in Toxicology 5, 99–199.chemistry 204, 657–667.

Schiavo, G., Benfenati, F., Poulain, B. et al. (1992) Tetanus andWhelan, S.M., Elmore, M.J., Bodsworth, N.J., Brehm, J.K., Atkin-

botulinum-B neurotoxins block neurotransmitter release by pro-son, T. and Minton, N.P. (1992) Molecular cloning of the Clo-

teolytic cleavage of synaptobrevin. Nature 359, 832–835. stridium botulinum structural gene encoding the type B neurotoxinSchiavo, G., Malizio, C., Trimble, W.S. et al. (1994) Botulinum and determination of its entire nucleotide sequence. Applied and

G neurotoxin cleaves VAMP/synaptobrevin at a single Ala-Ala Environmental Microbiology 58, 2345–2354.peptide bond. Journal of Biological Chemistry 269, 20213–20216. Willems, A., East, A.K., Lawson, P.A. and Collins, M.D. (1993)

Schiavo, G., Rossetto, O., Santucci, A., DasGupta, B.R. and Mon- Sequence of the gene coding for the neurotoxin of Clostridumtecucco, C. (1992) Botulinum neurotoxins are zinc proteins. Jour- botulinum type A associated with infant botulism: comparisonnal of Biological Chemistry 267, 23479–23483. with other clostridial neurotoxins. Research Microbiology 44, 547–

Seddon, H.R. (1922) The specific identify of Bacillus parabotulinus. 556.Journal of Comparative Pathology and Therapy 35, 275–280. Wu, J.I.J., Riemann, H. and Lee, W.H. (1972) Thermal stability of

Smith, L.D.S. and Hobbs, G. (1974) Genus III. Clostridium Praz- the deoxyribonucleic acid hybrids between the proteolytic strainsmowski 1880, 23. In Bergey’s Manual of Determinative Bacteri- of Clostridium botulinum and Clostridium sporogenes. Canadianology, ed. Buchanan, R.E. and Gibbons, N.E. 8th edn, pp. 551– Journal Microbiology 18, 97–99.572. Baltimore: The Williams and Wilkins Co. Yamasaki, S., A.Baumeister, T., Binz, J. et al. (1994) Cleavage of

Suen, J.C., Hatheway, C.L., Steigerwalt, A.G. and Brenner, D.J. members of the synaptobrevin/VAMP family by types D and F(1988a) Genetic confirmation of identities of neurotoxigenic Clo- botulinal neurotoxins and tetanus toxin. Journal of Biologicalstridium baratii and Clostridium butyricum implicated as agents of Chemistry 269, 12764–12772.infant botulism. Journal of Clinical Microbiology 26, 2191–2192. Zhou, Y., Sugiyama, H., Nakano, H. and Johnson, E.A. (1995) The

Suen, J.C., Hatheway, C.L., Steigerwalt, A.G. and Brenner, D.J. genes for the Clostridium botulinum type G toxin are on a plasmid.Infection and Immunity 63, 2087–2091.(1988b) Clostridium argentinense sp. nov. a genetically homo-


Phylogeny and taxonomy of the food-borne pathogen Clostridium botulinum and its neurotoxins

Documents

Transcript of Phylogeny and taxonomy of the food-borne pathogen Clostridium botulinum and its neurotoxins