SEARCH

SEARCH BY CITATION

Keywords:

  • Clostridial neurotoxins;
  • SNAP-25;
  • SNAP-23;
  • Proteolysis;
  • Protease binding assay

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

Abstract : Tetanus toxin and the seven serologically distinct botulinal neurotoxins (BoNT/A to BoNT/G) abrogate synaptic transmission at nerve endings through the action of their light chains (L chains), which proteolytically cleave VAMP (vesicle-associated membrane protein)/synaptobrevin, SNAP-25 (synaptosome-associated protein of 25 kDa), or syntaxin. BoNT/C was reported to proteolyze both syntaxin and SNAP-25. Here, we demonstrate that cleavage of SNAP-25 occurs between Arg198 and Ala199, depends on the presence of regions Asn93 to Glu145 and Ile156 to Met202, and requires about 1,000-fold higher L chain concentrations in comparison with BoNT/A and BoNT/E. Analyses of the BoNT/A and BoNT/E cleavage sites revealed that changes in the carboxyl-terminal residues, in contrast with changes in the amino-terminal residues, drastically impair proteolysis. A proteolytically inactive BoNT/A L chain mutant failed to bind to VAMP/synaptobrevin and syntaxin, but formed a stable complex (KD = 1.9 × 10-7M) with SNAP-25. The minimal essential domain of SNAP-25 required for cleavage by BoNT/A involves the segment Met146-Gln197, and binding was optimal only with full-length SNAP-25. Proteolysis by BoNT/E required the presence of the domain Ile156-Asp186. Murine SNAP-23 was cleaved by BoNT/E and, to a reduced extent, by BoNT/A, whereas human SNAP-23 was resistant to all clostridial L chains. Lys185Asp or Pro182Arg mutations of human SNAP-23 induced susceptibility toward BoNT/E or toward both BoNT/A and BoNT/E, respectively.

Clostridia produce several powerful neurotoxins ; tetanus toxin and botulinal neurotoxins BoNT/A to BoNT/G cause the clinical manifestations of tetanus and botulism in a large variety of animal species and humans. The toxins are synthesized as single-chain polypeptides with molar masses of ~150 kDa. On lysis of the bacteria and activation by proteolytic cleavage, the light chains (L chains ; 50 kDa) remain disulfide bound to the heavy chains (H chains ; 100 kDa). The extreme neurotoxicity is largely ascribed to the H chains, which bind to neuronal receptors that internalize the holotoxins, and translocate the L chains into the cytosol. Here, the L chains block fusion of synaptic vesicles with the presynaptic membrane (Simpson, 1989).

The genes of the eight known clostridial neurotoxins have been cloned and characterized (for review, see Niemann et al., 1994). The L chains contain a Zn2+-binding motif, His-Glu-X-X-His, also found in an increasing number of zinc-dependent metalloproteases (Jongeneel et al., 1989). Soon after demonstration of cleavage of VAMP (vesicle-associated membrane protein)/synaptobrevin (Trimble et al., 1988) by tetanus toxin and BoNT/B at the same peptide bond (Link et al., 1992 ; Schiavo et al., 1992), substrates and scissile bonds were identified for all other botulinum serotypes. These studies revealed that BoNT/D, BoNT/F, and BoNT/G also hydrolyze synaptobrevin, although each at a different peptide bond. BoNT/A and BoNT/E cleave SNAP-25 (synaptosome-associated protein of 25 kDa), again at distinct sites close to the C-terminus (for review, see Montecucco and Schiavo, 1994 ; Niemann et al., 1994), and BoNT/C proteolyzes syntaxin (Blasi et al., 1993b ; Schiavo et al., 1995) and SNAP-25 (Foran et al., 1996 ; Williamson et al., 1996).

When docked synaptic vesicles fuse at release sites of the nerve terminal with the presynaptic membrane, the three substrates syntaxin, SNAP-25, and VAMP/synaptobrevin are thought to associate into low-energy ternary complexes involving subdomains with pronounced heptad symmetry of hydrophobic residues (Hayashi et al., 1994). The energy released during formation of the complex is thought to drive the actual fusion reaction (Jahn and Hanson, 1998 ; Weber et al., 1998). Related complexes, composed of isoforms of the three synaptic proteins, are crucial for the fusion of transport vesicles in nonneuronal cells (Ferro-Novick and Jahn, 1994 ; Rothman, 1994). Some of these vesicle-mediated transport steps can be blocked through the action of clostridial neurotoxins (Galli et al., 1994 ; Ikonen et al., 1995).

The L chains of clostridial neurotoxins form a class of proteases with a unique substrate specificity. They cleave only one of several identical peptide bonds in their respective target molecules and fail to proteolyze short peptides spanning the individual cleavage regions (Shone et al., 1993 ; Yamasaki et al., 1994a). Based on these observations, it was proposed that the proteases detect specific sequence motifs present in all target molecules. Searches for such common structural features in the three substrates led Rossetto and colleagues (1994) to suggest that a helical substrate recognition motif [soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) motif] present in multiple copies in synaptobrevin, syntaxin, and SNAP-25 mediates specific recognition by clostridial neurotoxins. Studies using mutational analysis in synaptobrevin demonstrated that these motifs are involved in target recognition. These studies also indicated that all five synaptobrevin-specific neurotoxins recognize the target in distinct ways (Pellizzari et al., 1996, 1997). Recently, Washbourne et al. (1997) stated that efficient cleavage of SNAP-25 by BoNT/A and E also required the presence of at least one copy of such common recognition motifs.

In this study, we investigated the molecular basis of interaction between BoNT/A, C, and E with SNAP-25 and two nonneuronal SNAP-25 isoforms, human SNAP-23 (hSNAP-23) (Ravichandran et al., 1996) and mouse SNAP-23 (mSNAP-23) (also termed syndet ; Araki et al., 1997 ; Wang et al., 1997). By using a proteolytically inactive BoNT/A L chain, we established, for the first time, a toxin L chain/substrate binding assay. Specific deletion and point mutants were generated from SNAP-25 and its isoforms to analyze the contribution of subdomains or individual residues to binding and proteolysis. The data are compatible with a substrate-induced conformational fit. Furthermore, we show that single amino acid replacements generate hSNAP-23 derivatives that can be cleaved by BoNT/A and E.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

Plasmid constructions

DNA fragments encoding individual L chains were obtained by PCR, using purified bacteriophage-specific (BoNT/C) or total bacterial DNA [BoNT/A (strain 62A) and E (Beluga)], respectively, and were cloned in pQE3 (Qiagen, Hilden, Germany). Mutants of SNAP-25- and hSNAP-23-specific plasmids were constructed by PCR in pGEX-KG (Guan and Dixon, 1991), pGEX-2T (Pharmacia, Freiburg, Germany), pQE3, or pSP72 (Promega, Heidelberg, Germany).

Expression and purification of recombinant proteins

The E. coli strain M15pREP4 (Qiagen) was transfected with the individual L chain-encoding plasmids, and proteins were produced and purified according to the manufacturer's instructions except that BoNT/C and BoNT/E L chains were induced for 3 h of incubation at 21 or 30°C, respectively. Fractions containing the L chains were dialyzed against toxin assay buffer (150 mM K glutamate, 10 mM HEPES-KOH, pH 7.2), frozen in liquid nitrogen, and kept at -70°C.

Glutathione-S-transferase (GST), GST—SNAP-25 variants, GST-syntaxin Ia, and GST-synaptobrevin 2 were affinity purified on glutathione (GT)-Sepharose (Pharmacia) according to Guan and Dixon (1991) and finally dialyzed against toxin assay buffer.

In vitro transcription and translation

BoNT/A L chain (Glu224Gln)-, SNAP-25-, hSNAP-23-, or mSNAP-23-specific mRNAs were synthesized in vitro from suitable transcription plasmids (Binz et al., 1994 ; Wang et al., 1997) linearized downstream from the coding regions. Translations were performed in reticulocyte lysate (Promega), using 0.25 μg mRNA in the presence of [35S]methionine (24 μCi, 1,200 Ci/mmol ; ICN Biomedicals, Irvine, CA, U.S.A.) in a total volume of 25 μl.

Toxin treatment

A cleavage assay contained 1 μl of the translation mixture of [35S]methionine-labeled variants of SNAP-25, hSNAP-23, or mSNAP-23 and the respective L chain and was incubated for 60 min at 37°C in a total volume of 10 μl of toxin assay buffer. E. coli expressed GST-SNAP-25-His6 variants (3 μM final concentrations) were incubated in a total volume of 100 μl of toxin assay buffer containing L chains. Aliquots (15 μl) were withdrawn at specified time intervals. Reactions were stopped by the addition of 15 μl of double-concentrated sample buffer [120 mM Tris-HCl, pH 6.75, 10% (vol/vol) β-mercaptoethanol, 4% (wt/vol) sodium dedecyl sulfate (SDS), 20% (wt/vol) glycerol, 0.014% (wt/vol) bromphenol blue]. Samples were boiled for 3 min and subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) as described by Laemmli (1970), using 12.5 or 15% gels. Proteins were visualized by staining with Coomassie Blue or by fluorography and quantified with a Sharp JX-325 high-resolution scanner by using an ImageMaster TM 1-D program (version 1.10 ; Pharmacia). Alternatively, radiolabeled samples were analyzed with a BAS-1500 phosphoimager (Fuji Photo Film, Japan).

For determination of the BoNT/C cleavage site in SNAP-25, 32 μg of SNAP-25-His6 was incubated for 16 h at 37°C with recombinant L chain (250 nM final concentration) in a total volume of 100 μl of toxin assay buffer. Cleavage products were isolated by reversed-phase chromatography, using a Nucleosil 5-mm C8 column (250 × 4 mm) from Macherey and Nagel (Düren, Germany). N-terminal amino acid sequences were determined on a Model 473A protein sequencer from Applied Biosystems (Foster City, CA, U.S.A.).

Toxin binding assay

Various GST-fusion proteins (0.1 nmol each) prebound to 10 μl of GT-Sepharose beads were suspended in 190 μl of toxin assay buffer containing 0.02% Triton X-100. Beads were then incubated for 30 min at 4°C with identical amounts of radio-labeled BoNT/A (Glu224Gln) L chain, as generated by in vitro transcription/translation. The beads were collected by centrifugation. Unbound material was recovered from the supernatant by trichloracetic acid precipitation. Beads were washed three times each with 50 bed volumes of the same buffer. The washed pellet fraction was boiled in SDS sample buffer and analyzed together with the supernatant fraction by SDS-PAGE and fluorography.

For determination of the BoNT/A-SNAP-25 dissociation constant (KD), GST-SNAP-25 was incubated in the presence of various concentrations of BoNT/A(Glu224Gln) L chain ranging from 156 to 2,500 nM. The amount of bound L chain was quantified by laser densitometric scanning after SDS-PAGE and Coomassie Blue staining and corrected for nonspecific binding by subtracting the value obtained for GST. The KD value was calculated by Scatchard plot analysis.

Identification of the BoNT/C scissile bond in SNAP-25

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

Foran et al. (1996) showed that BoNT/C, in addition to proteolyzing syntaxin, also cleaves SNAP-25 in a position similar to that of BoNT/A. The precise site of proteolysis, however, was not determined. To identify the scissile bond, recombinant SNAP-25-His6 was digested with recombinant L chains of BoNT/C and, as a control, BoNT/A. The products of the two reactions were analyzed by reversed-phase HPLC, and material underlying individual peak fractions was subjected to amino acid sequencing (Fig. 1). In agreement with previous results (Binz et al., 1994), the first peak fraction of the BoNT/A reaction yielded the sequence RATKMLGSGVP. The corresponding fragment obtained in the BoNT/C reaction eluted slightly faster than the BoNT/A product and gave the sequence ATKMLGSGVP (Fig. 1, bottom panel). In both instances, material eluting after 29 min carried the authentic N-terminus of SNAP-25. We conclude that BoNT/C hydrolyzes the Arg198-Ala199 bond and thus cleaves in a novel position in comparison with BoNT/A or E.

image

Figure 1. Identification of the BoNT/C scissile bond in SNAP-25. Aliquots of HPLC-purified SNAP-25-His6 (32 μg each) were incubated for 16 h at 37°C with BoNT/C or BoNT/A L chains (each 250 nM final concentration). Samples were analyzed by reversed-phase HPLC using a linear gradient of 10-90% acetonitrile in 0.1% aqueous trifluoroacetic acid starting after 5 min. Amino acid sequences specify the N-terminal sequences of peptides underlying individual peak fractions.

Download figure to PowerPoint

Mutational analyses of the BoNT/A, C, and E scissile bonds

The amino acid positions around the cleavage site of proteases are designated -P3-P2-P1-P1′-P2′-P3′-, where P1-P1′ represents the actual site of proteolysis. The determination of the BoNT/C cleavage site in SNAP-25 completes the list of susceptible peptide bonds that may be cleaved by clostridial neurotoxin proteases. A comparison of these sites indicates the great diversity of the residues tolerated by this class of metalloproteases. To gain additional insights into the particular requirements for BoNT/A and E, we replaced the residues of the P1 and P1′ positions by hydrophobic, other polar, negatively charged, or positively charged residues and analyzed the susceptibility of the resulting mutants (Table 1). Of five substitutions introduced at the P1 position of the BoNT/A cleavage site, only threonine caused a slight but reproducible reduction of cleavage. Next, Arg198 of the P1′ position was systematically mutated to residues of different character. Except for a change to tyrosine, all other mutations, even the conservative exchange to lysine, dramatically reduced the sensitivity of SNAP-25 to BoNT/A (Table 1A). Similar results were obtained with BoNT/E cleavage site mutants (Table 1B). Again, substitutions in the P1′ position impaired cleavage. We conclude that in accordance with previous studies regarding the BoNT/A and BoNT/B cleavage sites (Shone and Roberts, 1994 ; Schmidt and Bostian, 1997), the properties of the P1 residue appear less critical for cleavage, whereas specific amino acid residues must be present in the P1′ positions. In this respect, BoNT/C differs from the other neurotoxins : five of the seven substitutions introduced into the P1 position affected proteolysis drastically (Table 1A).

Table 1. Mutational analysis of the BoNT/A, C, and E cleavage sites in SNAP-25Radiolabeled substrates were generated by in vitro transcription/translation and incubated with various concentrations of BoNT/A, C, or E L chain in toxin assay buffer. After 1 h of incubation at 37°C, samples were analyzed by SDS-PAGE and fluorography. Percentage of cleavage was quantified by laser densitometric scanning. Data are mean values of three independent experiments and represent the concentration required to reach 50% proteolysis. The amino acid single letter code was used.
A.Mutant  BoNT/A (nM)   BoNT/C (μM)
  BoNT/A BoNT/C E194 A N Q R A T K201 0.2 0.5
  M197 0.2 Not tested
  S197 0.2 Not tested
  T197 0.3 Not tested
  E197 0.2 Not tested
  K197 0.2 Not tested
  Y198 0.6 0.5
  A198 60 5.3
  S198 100 1.8
  T198 2,500 >6.0
  D198 8,000 >6.0
  E198 No cleavage >6.0
K198  100 0.5
B. Mutant  BoNT/E (nM)  
   BoNT/E Q177 I D R I M E K184 0.2 
   V181 0.5 
   F181 100 
   G181 25 
   A181 2.5 
   S181 10 
   N181 60 

Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

As revealed by our own and previously published data, however, proteolytic susceptibility or resistance of a potential substrate against neurotoxin proteases cannot be explained solely on the basis of residues present in the scissile bond. Recently, two isoforms of SNAP-25 were identified in human B lymphocytes (hSNAP-23 ; Ravichandran et al., 1996) and in murine 3T3-L1 adipocytes (mSNAP-23, also termed syndet ; Araki et al., 1997 ; Wang et al., 1997). These nonneuronal isoforms and SNAP-25 share significant sequence identity and the amino acid residues present in the potential scissile bonds are largely conserved (Fig. 2A). It has been reported that hSNAP-23 was resistant to the action of BoNT/E (Macaulay et al., 1997). In a similar manner, mSNAP-23 was shown to resist moderate concentrations of BoNT/A and C (Chen et al., 1997) or BoNT/E (Macaulay et al., 1997). In contrast, Washbourne et al. (1997) reported that mSNAP-23 was cleaved by BoNT/E. To reexamine this issue, we incubated in vitro translated hSNAP-23 and mSNAP-23 with recombinant L chains of BoNT/A, C, and E (Fig. 2B). hSNAP-23 was not proteolyzed by any of the three L chains even when concentrations of up to 5 μM were applied. In contrast, mSNAP-23 was cleaved by BoNT/E and, much less efficiently, by BoNT/A. Demonstration of successful cleavage required, however, prolonged incubation times and elevated toxin concentrations.

image

Figure 2. Susceptibility of various SNAP-25 isoforms to BoNT/A, BoNT/C, and BoNT/E. A : Alignment of the C-terminal regions of SNAP-25, hSNAP-23, and mSNAP-23. Peptide bonds susceptible to clostridial neurotoxins are boxed. Colons (:) identify nonconservative amino acid substitutions. B : Radiolabeled SNAP-25, hSNAP-23, and mSNAP-23 generated by in vitro transcription/translation were incubated with various recombinant botulinal L chains. Concentrations used were BoNT/A (1 nM), BoNT/C (500 nM), and BoNT/E (1nM) for SNAP-25 ; 5 μM each of BoNT/A, BoNT/C, and BoNT/E for hSNAP-23 ; and BoNT/A (5 μM) BoNT/C (5 μM), and BoNT/E (0.4 μM) for mSNAP-23. Incubation was for 60 min at 37°C for SNAP-25 or 3 h for hSNAP-23 and mSNAP-23. Cleavage was analyzed by SDS-PAGE and fluorography.

Download figure to PowerPoint

It remains to be shown, therefore, whether proteolysis of mSNAP-23 by BoNT/A and E would be detectable in vivo. In this context, it is noteworthy that the L chains of clostridial neurotoxins display a remarkable stability of several days in injected cells (Erdal et al., 1995) and, therefore, detrimental long-term in vivo effects of the L chains cannot be excluded. Furthermore, BoNT/C was reported to be almost as active as BoNT/A in mouse spinal cord cell cultures (Williamson et al., 1996), whereas it was ~1,000-fold less active in vitro. In conclusion, it is rather speculative to extrapolate from the in vitro cleavage rates on the in vivo activities of particular L chains.

Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

To define those subdomains of SNAP-25 in which individual amino acid substitutions could alter the susceptibility toward the three proteases, we next generated a set of N- and C-terminally truncated substrate molecules. The results are presented in Fig. 3. SNAP-25(93-206) was as susceptible to cleavage by all three toxins as full-length SNAP-25. BoNT/A and E proteolyzed GST-SNAP-25(146-206) as efficiently as SNAP-25(1-206), whereas BoNT/C proteolysis was significantly decreased. The removal of additional 10 N-terminal amino acids [to yield mutant SNAP-25(156-206)] diminished cleavage by both BoNT/A and E, whereas cleavage by BoNT/C remained unaltered. SNAP-25(167-206) showed a further reduction in BoNT/C cleavage and a drastic reduction in BoNT/E cleavage, and BoNT/A proteolysis was no longer detectable. However, application of a higher BoNT/A L chain concentration (100 nM final concentration) resulted in 50% cleavage of this mutant after 60 min (data not shown). It is unlikely that the increased resistance of truncated SNAP-25 derivatives is merely due to steric hindrance imposed by the presence of the N-terminally located GST, as such hindrance should affect the three L chains to the same extent.

image

Figure 3. Identification of the minimal essential domains of SNAP-25 required for cleavage by BoNT/A, C, and E. The cleavage rates of BoNT/A, C, and E were determined with GST-SNAP-25-His6 or GST-SNAP-25 mutants, respectively, as detailed in Materials and Methods. Toxin concentrations used were 1 nM (BoNT/A ; A), 500 nM (BoNT/C ; B), and 1 nM (BoNT/E ; C). Aliquots of the reaction mixtures were taken at the time points indicated and analyzed by 12.5% SDS-PAGE and Coomassie Blue staining. Amounts of proteins were quantified by laser densitometric scanning. Data are mean ± SD values of four to six independent experiments. GST-SNAP-25(1-206) (□) ; GST-SNAP-25(93-206) (○) ; GST-SNAP-25(146-206) (X) ; GST-SNAP-25(156-206) (•) ; GST-SNAP-25(167-206) (▪).

Download figure to PowerPoint

In parallel experiments we examined the impact of C-terminal deletions of SNAP-25 on cleavage by the three L chains. SNAP-25 derivatives retaining as few as five, four, and six residues on the C-terminal side of the respective scissile bonds, were cleaved by BoNT/A, C, and E as efficiently as full-length SNAP-25 (data not shown).

Together, these data indicate that interaction of BoNT/A, C, and E with SNAP-25 depends on distinct domains ; BoNT/A requires the presence of the domain Met146-Met202. For BoNT/E, the minimal segment that allows efficient cleavage encompasses the segment Ile156-Asp186, whereas BoNT/C interaction depends on a larger substrate structure involving regions Asn93 to Glu145 and Ile156 to Met202.

Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

Before cleavage, toxin and substrate must form an enzyme-substrate complex involving contacts at one or multiple regions in both molecules. We investigated, therefore, whether it was possible to dissect the overall cleavage process into individual substeps involving assembly into the enzyme-substrate complex and a subsequent cleavage step. To prevent proteolysis of individual SNAP-25 derivatives during assembly, we replaced Glu24 of the catalytic zinc-binding domain His223-Glu-Leu-Ile-His227 by Gln224. As shown previously, this mutant lacks proteolytic activity (Blasi et al., 1993a), probably because the δ-carboxyl group of Glu224 is essential for stabilizing a water molecule in one of the tetrahedral positions around the catalytic zinc ion. As the mutated protein retains the zinc ion (Yamasaki et al., 1994b), the structural changes should be minimal and the mutated L chain should be a good candidate to study its binding to GST-SNAP-25 derivatives.

Radiolabeled BoNT/A(Glu224Gln) L chain was generated by in vitro translation and was incubated with GST-SNAP-25 immobilized on GT-Sepharose beads. Bound (pellet fraction) and unbound material (supernatant fraction), were analyzed by SDS-PAGE and fluorography (Fig. 4A). The concentration of the bound GST-fusion protein was controlled by SDS-PAGE and Coomassie Blue staining (Fig. 4B). Specific binding of the BoNT/A L chain to GST-SNAP-25 was indeed observed. No binding was obtained with GT-Sepharose beads carrying GST alone, GST-syntaxin(1-267), or GST-synaptobrevin(1-116). This experimental approach was subsequently used to map those subdomains of SNAP-25 that were involved in the binding of the L chain. A deletion of the 25 N-terminal residues of SNAP-25 reduced binding of the L chain only slightly. A further deletion up to Thr46 reduced binding about threefold. SNAP-25(146-206), lacking the N-terminal three-fourths of the molecule, still retained ~10% binding capacity of full-length SNAP-25. Complete loss of binding occurred on deletion of an additional 10 amino acids yielding SNAP-25(156-206). It is interesting that a SNAP-25 derivative containing the entire N-terminal sequence up to the BoNT/E scissile bond [SNAP-25(1-180)] failed to trap the mutant L chain, indicating that the region between the BoNT/E and BoNT/A scissile bonds is also essential for toxin binding. This is corroborated by the finding that SNAP-25(1-197), the product of the BoNT/A reaction, exhibited 65% binding capacity of full-length SNAP-25. In summary, the minimal essential fragment of SNAP-25 capable of binding to the BoNT/A L chain requires the presence of the region Met146-Gln197.

image

Figure 4. Binding of the BoNT/A L chain to SNAP-25 depends on the segment Met146-Gly155. A : Radiolabeled BoNT/A(Glu224Gln) L chain was incubated with GT-Sepharose beads precharged with GST, GST-syntaxin, GST-synaptobrevin, and GST-SNAP-25-His6 fusion proteins. Numbers specify amino acid residues of SNAP-25. Supernatant (S) and washed pellet (P) fractions were analyzed by SDS-PAGE and fluorography. Binding (presence of radiolabeled L chain in the pellet fraction) can be demonstrated as long as the capture molecule contains the segment Met146-Gly155 and the region between the BoNT/E and BoNT/A scissile bonds. B : Analyses of pellet fractions by SDS-PAGE and Coomassie Blue staining to control for the amounts of capture molecules. Lane 1, GST ; lane 2, GST-syntaxin ; lane 3, GST-synaptobrevin ; lane 4, GST-SNAP-25(1-206)-His6 ; lane 5, GST-SNAP-25(14-206) ; lane 6, GST-SNAP-25(25-206) ; lane 7, GST-SNAP-25(36-206) ; lane 8, GST-SNAP-25(46-206) ; lane 9, GST-SNAP-25(93-206)-His6 ; lane 10, GST-SNAP(146-206)-His6 ; lane 11, GST-SNAP-25(156-206)-His6 ; lane 12, GST-SNAP-25(167-206)-His6 ; lane 13, GST-SNAP-25(1-197) ; lane 14, GST-SNAP-25(1-180).

Download figure to PowerPoint

Binding experiments, using the in vitro translated radiolabeled L chain and GST-SNAP-25, provided a rapid approach for mapping the various interacting domains. A drawback of this approach, however, was the relative inefficiency of binding. Additional binding studies indicated that maximal binding of the L chain occurred within 30 min of incubation, and overnight incubations did not bring an improvement. In a separate set of experiments, we next used an inactive E. coli-derived BoNT/A L chain. By using this recombinant L chain, binding to SNAP-25 was saturable yielding an L chain to SNAP-25 ratio of ~1:1, as evidenced by densitometric scanning of Coomassie Blue-stained gels (Fig. 5A). The dissociation constant of such complexes was determined by Scatchard plot analyses yielding a KD of 1.9 × 10-7M (Fig. 5B).

image

Figure 5. Scatchard plot analysis of BoNT/A L chain binding to SNAP-25. A : GST or GST-SNAP-25 immobilized on GT-Sepharose beads was incubated with the indicated amounts of recombinant BoNT/A(Glu224Gln) L chain. Bound L chain was determined as above by SDS-PAGE and quantified by laser densitometric scanning. Binding to SNAP-25 was saturable, yielding an L chain to SNAP-25 ratio of ~1 : 1. Bands labeled by the asterisk are N-terminal degradation products of GST-SNAP-25. B : The molar ratio of bound to free L chain was plotted versus the amount of bound L chain. Data are mean values of three independent experiments. The dissociation constant KD was 1.9 × 10-7M.

Download figure to PowerPoint

Mutational analysis of the region Met146-Gly155

The above binding studies showed that the best binding to BoNT/A L chain occurred already with full-length SNAP-25. Binding gradually decreased on N-terminal deletions and was no longer detectable with derivatives lacking the Met146-Gly155 segment. Related copies of this segment have been detected in each of the three substrates and their impact on the cleavage reaction has been studied for some of the clostridial neurotoxin L chains (Rossetto et al., 1994 ; Pellizzari et al., 1996, 1997). To clarify the role of the Met146-Gly155 segment in greater detail, we constructed mutants in which negatively charged amino acid residues were replaced by uncharged polar residues or in which the entire motif was deleted. As shown in Fig. 6, a replacement of Asp147 and Glu148 by Asn and Gln had only a minor impact on cleavage by BoNT/A and E. In a similar manner, a Glu151Thr replacement (as present in hSNAP-23 and mSNAP-23) had no effect on the cleavage rate (data not shown). The removal of the entire region from SNAP-25(93-206) reduced the rate of proteolysis to levels that were obtained with SNAP-25(156-206). We conclude that the presence of the region Met146-Gly155 is indeed essential for efficient toxin-substrate interaction. The observation that the negatively charged amino acids may be replaced by related polar residues without changing the cleavage rate excludes the possibility that this segment interacts with the BoNT/A L chain via ionic forces. It remains to be shown whether the polar residues are in fact engaged in a direct interaction with the L chain or whether they are part of a sensor that controls folding of the substrate into a particular secondary structure.

image

Figure 6. Charged residues of the segment Met146-Gly155 have no impact on cleavage. The cleavage rates of BoNT/A, C, and E were determined with GST-SNAP-25(93-206) as a control and several variants thereof containing mutations in the SNARE motif as detailed in Materials and Methods. Toxin concentrations used were 1 nM (BoNT/A ; A), 500 nM (BoNT/C ; B), and 1 nM (BoNT/E ; C). Aliquots of the individual reaction mixtures were taken at the time points indicated and analyzed as in Fig. 3. GST-SNAP-25(93-206) (○) ; GST-SNAP-25(93-206)Asp147Asn, Glu148Gln (□) ; GST-SNAP-25(93-206)▵145-153 (•).

Download figure to PowerPoint

Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

The SNAP-25 segment Met146-Gln197 represents a minimal length substrate for both BoNT/A and E. Only a limited number of amino acid changes are found in the corresponding regions of hSNAP-23 and mSNAP-23 (Fig. 2A). Such residues could mark direct interaction sites with the toxin L chains. Alternatively, they could have a profound influence on the secondary structure of the substrate and could, therefore, influence the toxin target interaction indirectly. To analyze the role of particular residues in greater detail, we replaced some of them in hSNAP-23 by those found in rat SNAP-25.

A Thr158Glu mutation yielded a product that remained resistant to all three botulinal L chains (Table 2). In SNAP-25, an aspartate residue is found in the P2 position of the BoNT/E cleavage site, whereas the poorly susceptible mSNAP-23 carries a glutamine and the toxin-resistant hSNAP-23 contains a lysine in this position (Fig. 2A). As indicated in Table 2, a Lys185Asp replacement in hSNAP-23 had no effect on the reactions with BoNT/A or C but generated an effectively cleaved substrate for BoNT/E. Indeed, hSNAP-23(Lys185Asp) proved to be a better substrate for BoNT/E than mSNAP-23 (Table 2). This finding is in support of a previous study, in which mutation of the P2 position of the BoNT/A cleavage site in SNAP-25 drastically reduced the cleavage rate (Schmidt and Bostian, 1997). Next, we assessed the impact of an Ile173Met substitution. This residue was chosen for mutagenesis, because a similar Ile46Met replacement in rat synaptobrevin 1 dramatically increased cleavage by BoNT/D (Yamasaki et al., 1994a). Furthermore, Ile173 and Ile46 of hSNAP-23 and rat synaptobrevin 1, respectively, reside in identical positions with respect to the corresponding cleavage sites for BoNT/E and D. This mutant remained resistant against BoNT/E but, interestingly, showed an increased sensitivity toward BoNT/A. A Pro182Arg mutation yielded a hSNAP-23 derivative that, in contrast to wild-type hSNAP-23, could be cleaved by BoNT/A and E. Likewise, introduction of a proline residue into the equivalent position of SNAP-25 [SNAP-25(Arg176Pro)] drastically reduced cleavage by BoNT/A and E (Table 2). Together, these data demonstrate that the presence of a proline residue within the minimal essential interactive domain deteriorates the substrate quality of SNAP-25 derivatives.

Table 2. Effect of point mutations in SNAP-25 and hSNAP-23 on cleavage with BoNT/A, C, and ERadiolabeled substrates were generated by in vitro transcription/translation. SNAP-25 and its mutant were treated with BoNT/A and E (each at 0.3 nM final concentration) and BoNT/C (0.5 μM). For mSNAP-23, hSNAP-23, and its mutants, toxin L chains were applied at 1 μM final concentration.
MutantBoNT/ABoNT/CBoNT/E
  1. aExperiments done with 125 nM final concentration of BoNT/E L chain. All incubations were performed for 60 min at 37°C and cleavage was analyzed by SDS-PAGE and fluorography. Data represent mean ± SD values of four independent experiments and are expressed as percentages of cleavage.

SNAP-2549.7 ± 8.036.1 ± 6.048.8 ± 6.5
SNAP-25 (Arg176Pro) 24.0 ± 5.539.8 ± 6.615.4 ± 3.2
hSNAP-23No cleavageNo cleavageNo cleavage
hSNAP-23 (Thr158Glu) No cleavageNo cleavageNo cleavage
hSNAP-23 (Ile173Met) 0.9 ± 0.2No cleavageNo cleavage
hSNAP-23 (Pro182Arg) 7.0 ± 1.0No cleavage12.7 ± 0.6
hSNAP-23 (Lys185Asp) No cleavageNo cleavage100 ± 2.2 79.7 ± 8.3 a
mSNAP-233.6 ± 0.8No cleavage100 ± 2.2 38.2 ± 5.0 a

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

We analyzed structural features of SNAP-25 and of its nonneuronal murine and human SNAP-23 isoforms that influence proteolytic cleavage by botulinum neurotoxin L chains of serotypes A, C, and E. The following several discoveries were made that may provide additional insights into the complex mechanism of the unique substrate recognition by clostridial L chain proteases : (1) The three proteases cleave the identical substrate. Each of them, however, shows distinct requirements regarding the minimal essential domains for cleavage. (2) The L chains of BoNT/A and E require specific residues around the scissile bond in the P2 and P1' positions, whereas residues in the P1 position are less crucial. For BoNT/C, the nature of the residue in the P1 position is also important. (3) Using a proteolytically inactive BoNT/A L chain, we developed an in vitro binding assay that allowed us to distinguish substrate domains that are important for enzyme-substrate complex formation and for subsequent proteolysis. Our studies indicate that the actual cleavage reaction constitutes the rate-limiting step. The minimal essential domains required for optimal cleavage of SNAP-25 by BoNT/A, C, and E extend from Met146 to Met202, Asn93 Met202, and Met146 to to Asp186, respectively. (4) Within these domains, single amino acid substitutions may drastically alter the susceptibility toward individual L chains. The closely related nonneuronal isoform mSNAP-23 is cleaved by BoNT/E and by high concentrations of BoNT/A but not by BoNT/C. In contrast, hSNAP-23 is resistant against all three L chains. Single amino acid substitutions partially restore cleavage of hSNAP-23. (5) Together, our findings are compatible with the possibility that the clostridial L chains turn into active proteases only on contact with their cognate substrate.

Before we discuss the impact of SNAP-25 subdomains on binding and cleavage in detail, we should recall that in the cell the SNARE proteins SNAP-25, synaptobrevin, and syntaxin exist in at least two distinct conformational states, only one of which allows proteolytic degradation by clostridial neurotoxin L chains. The individual monomeric SNAREs have little secondary structure under physiological conditions. This follows from both NMR analyses of a synthetic synaptobrevin polypeptide (Cornille et al., 1994) and circular dichroic spectroscopy of SNAP-25 (Fasshauer et al., 1997). These unprotected SNAREs can be attacked by clostridial L chains (Hayashi et al., 1994). On contact with other SNARE proteins, synaptobrevin, SNAP-25, and syntaxin assemble spontaneously into various low-energy heterodimeric or heterotrimeric complexes. Formation of such complexes is driven by coiled-coil formation (Hayashi et al., 1994 ; Fasshauer et al., 1997) and paralleled by the gain of resistance against clostridial neurotoxins (Hayashi et al., 1994). Such complexes may be dissociated at the expense of ATP through the joint action of α-SNAP [α-soluble N-ethylmaleimide-sensitive fusion protein (NSF) attachment protein] and NSF yielding again toxin-sensitive SNARE monomers (Hayashi et al., 1995). According to our current view of membrane fusion (Jahn and Hanson, 1998 ; Weber et al., 1998), the conformational energy stored in the unfolded SNARE monomers is liberated during assembly into SNARE complexes and used to pull the vesicle and target membranes toward each other. On contact with a neurotoxin L chain, however, such conformational energy could also drive conformational changes in either the substrate alone, or, perhaps more likely, in both the substrate and the L chain, resulting in the formation of stable enzyme-substrate complexes.

Within the minimal interactive domains of SNAP-25 determined for BoNT/A, C, and E, the individual cleavage site region, encompassing the scissile bond itself and ~20 residues located N-terminally thereof, constitutes a most critical subdomain influencing both substrate binding and cleavage. As demonstrated here with a proteolytically inactive BoNT/A L chain, this region clearly contributes to binding, as the BoNT/A cleavage product (ending with Gln197) retains 65% binding activity of full-length SNAP-25 and the BoNT/E product (ending with Arg180) fails to bind. Efficient binding, however, does not necessarily imply efficient cleavage. This follows from the observation that the SNAP-25(Arg198Glu) mutant binds to the BoNT/A L chain with a similar efficiency as wild-type SNAP-25 (T. Binz, unpublished observation) without being subsequently cleaved (Table 1). Our mutational analyses of the cleavage sites of BoNT/A and E further support the theory that the nature and the size of residues present in the P1' and in the P2 position are crucial for cleavage, whereas those in the P1 position are not. Similar findings have been reported by Schmidt and Bostian (1997) who showed, in addition, that residues of the P3' and P4 positions have little impact on cleavage. BoNT/C differs from the other two proteases, as changes in P1 influence cleavage drastically (Table 1).

Comparative binding and cleavage analyses of N-terminal deletion mutants of SNAP-25 brought the surprising insight that BoNT/A required the entire SNAP-25 molecule for optimal binding, whereas cleavage was not affected unless these deletions affected the minimal size substrate represented by SNAP-25(Met146-Met202). Binding to full-length SNAP-25 was saturable, yielding a 1 : 1 ratio and a KD of 1.9 × 10-7M. Even a deletion of the 13 N-terminal residues of SNAP-25, however, reduced binding already to ~60% of full-length SNAP-25. In comparison, binding to SNAP-25(Met146-Met202) was ~10% without affecting the rate of proteolysis. These findings suggest that substrate recognition and binding occur faster than the actual cleavage reaction. A deletion of an additional 10 residues, to yield SNAP-25(Ile156-Gly206), reduced binding below detectable levels and drastically reduced cleavage by BoNT/A, but much less dramatically by BoNT/E. For BoNT/C, a more extended substrate structure, encompassing residues Asn93 to Met202 was required to establish optimal cleavage. It is surprising that a deletion of the segment Glu145 to Ser154 from this substrate affected cleavage only slightly.

What is the function of the segment Glu145 to Ser154 and of related motifs in the other SNAREs ? These motifs are highly conserved during evolution, suggesting that they could serve a common, as yet unknown, function in the individual SNAREs. Recently, Washbourne et al. (1997) reported that a deletion of the entire N-terminal domain including the Glu145-Ser154 domain resulted in a dramatic loss of cleavage by reduced BoNT/A and E holotoxins. The authors concluded that this segment was indeed essential for toxin action. These findings agree with our observation that cleavage by BoNT/A and E was optimal only with substrates containing this segment. In our study, however, effective cleavage by the recombinant BoNT/E L chain continued also with substrates lacking this segment. The reason for this discrepancy is unclear. One approach to clarify the role of this segment could involve studies on potentially inhibitory properties of synthetic peptides representing this Glu145-Ser154 segment. Unfortunately, a corresponding peptide was found to be insoluble (Rossetto et al., 1994). Studies with related segments in synaptobrevin had demonstrated that the presence of the negatively charged residues was obligatory for cleavage (Pellizzari et al., 1996, 1997). To our surprise, however, substitutions of these acidic amino acid residues in SNAP-25 by the corresponding noncharged amides affected cleavage only slightly. Considering that BoNT/A, C, and E share the property of tolerating the uncharged amino acids with BoNT/D (Pellizzari et al., 1997), we may conclude that the clostridial neurotoxins are specialized proteases that have evolved from one ancestral gene, recognizing originally the same repetitive motif in their individual substrates. During evolution, however, each neurotoxin may have developed its individual dependency on particular residues within this motif. Following these lines of arguments, BoNT/E and C show the least dependency ; substrates lacking the Glu145-Ser154 motif are cleaved only slightly worse or even equally well, respectively (Fig. 6).

At least for BoNT/A, the cleavage reaction appears to constitute the rate-limiting step. This suggests that on a first contact, SNAP-25 induces a conformational change of the BoNT/A L chain, as has been reported for the interaction between tetanus toxin and its substrate synaptobrevin (Cornille et al., 1997). This could explain why hSNAP-23 and mSNAP-23 are poor substrates. Even after mutational correction of the SNARE motifs and despite the presence of largely conserved cleavage sites, mSNAP-23 was cleaved by high concentrations of BoNT/E only and, much less efficiently, by BoNT/A, whereas hSNAP-23 remained resistant against all three L chains. In keeping with these findings, binding of mSNAP-23 to BoNT/A was barely detectable and no binding of hSNAP-23 was observed. This is remarkable, as hSNAP-23 and SNAP-25 share 63% identical residues within the domain mapped to be essential for optimal cleavage by BoNT/A. Unfortunately, circular dichroism spectroscopy of BoNT/A-SNAP-25 complexes did not reveal a change in α-helicity. However, such a difference would probably not be detected, if new helices were formed at the expense of previously existing helices. An Arg176Pro substitution in SNAP-25 drastically reduced the cleavability by BoNT/A and E but not by BoNT/C. Likewise, the reverse mutation in hSNAP-23 restored cleavability by BoNT/A and E (Table 2). The simplest interpretation of these findings would imply that Arg176 of SNAP-25 is required to interact with the two L chains. An alternative hypothesis would suggest that introduction of helix-breaking residues prevents formation of the Michaelis complex. Clearly, this issue demands more efforts and mutants and more sophisticated techniques such as stopped-flow analyses. At any rate, the generation of toxin-sensitive SNAP-23 mutants could provide potentially valuable tools to study the role of these SNAREs in apical transport in polarized cells (Galli et al., 1998).

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Identification of the BoNT/C scissile bond in SNAP-25
  6. Murine and human SNAP-25 isoforms display distinct sensitivities to cleavage by botulinal L chains
  7. Effects of N- and C-terminal deletions of SNAP-25 on cleavage by BoNT/A, C, and E
  8. Binding of the BoNT/A L chain to SNAP-25 depends on the presence of the region Met146-Gln197
  9. Single amino acid replacements in hSNAP-23 confer susceptibility to BoNT/A and E
  10. DISCUSSION
  11. Acknowledgements

We thank Drs. Richard H. Scheller (Stanford) and William Trimble (Toronto) for plasmids encoding synaptobrevin and syntaxin, respectively. Clones for mSNAP-23 and hSNAP-23 were kindly provided by Drs. Giulia Baldini (New York) and Thierry Galli (Paris), respectively. This study was supported by the Deutsche Forschungsgemeinschaft (IIB2-Bi 660/1-1) and Fonds der Chemischen Industrie. V.V.V. and K.Y. were supported by fellowships from the Alexander von Humboldt Foundation.

  • 1
    Araki S., Tamori Y., Kawanishi M., Shinoda H., Masugi J., Mori H., Niki T., Okasawa H., Kubota T., Kasuga M. (1997) Inhibition of the binding of SNAP-23 to syntaxin by Munc 18c.Biochem. Biophys. Res. Commun. 234,257262.
  • 2
    Binz T., Blasi J., Yamasaki S., Baumeister A., Link E., Südhof T.C., Jahn R., Niemann H. (1994) Proteolysis of SNAP-25 by types E and A botulinal neurotoxins.J. Biol. Chem. 269,16171620.
  • 3
    Blasi J., Chapman E.R., Link E., Binz T., Yamasaki S., De Camilli P., Südhof T.C., Niemann H., Jahn R. (1993a) Botulinum neurotoxin A selectively cleaves the synaptic protein SNAP-25.Nature 365,160163.
  • 4
    Blasi J., Chapman E.R., Yamasaki S., Binz T., Niemann H., Jahn R. (1993b) Botulinum neurotoxin C1 blocks neurotransmitter release by means of cleaving HPC-1/syntaxin.EMBO J. 12,48214828.
  • 5
    Chen F., Foran P., Shone C.C., Foster K.A., Melling J., Dolly J.O. (1997) Botulinum neurotoxin B inhibits insulin-stimulated glucose uptake into 3T3-L1 adipocytes and cleaves cellubrevin unlike type A toxin which failed to proteolyze the SNAP-23 present.Biochemistry 36,57195728.
  • 6
    Cornille F., Goudreau N., Ficheux D., Niemann H., Roques B.P. (1994) Solid-phase synthesis, conformational analysis and in vitro cleavage of synthetic human synaptobrevin II 1-93 by tetanus toxin L chain.Eur. J. Biochem. 222,173181.
  • 7
    Cornille F., Martin L., Lenoir C., Cussac D., Roques B.P., Fournie-Zaluski M.C. (1997) Cooperative exosite-dependent cleavage of synaptobrevin by tetanus toxin light chain.J. Biol. Chem. 272,34593464.
  • 8
    Erdal E., Bartels F., Binscheck T., Erdmann G., Frevert J., Kistner A., Weller U., Wever J., Bigalke H. (1995) Processing of tetanus and botulinum A neurotoxins in isolated chromaffin cells.Naunyn Schmiedebergs Arch. Pharmacol. 351,6778.
  • 9
    Fasshauer D., Bruns D., Shen B., Jahn R., Brn̈ger A.T. (1997) A structural change occurs upon binding of syntaxin to SNAP-25.J. Biol. Chem. 272,45824590.
  • 10
    Ferro-Novick S. & Jahn R. (1994) Vesicle fusion from yeast to man.Nature 370,191193.
  • 11
    Foran P., Lawrence G.W., Shone C.C., Foster K.A., Dolly J.O. (1996) Botulinum neurotoxin C1 cleaves both syntaxin and SNAP-25 in intact and permeabilized chromaffin cells : correlation with its blockade of catecholamine release.Biochemistry 25,26302636.
  • 12
    Galli T., Chilcote T., Mundigl O., Binz T., Niemann H., De Camilli P. (1994) Tetanus toxin-mediated cleavage of cellubrevin impairs exocytosis of transferrin receptor-containing vesicles in CHO cells.J. Cell Biol. 125,10151024.
  • 13
    Galli T., Zahraoui A., Vaidyanathan V.V., Raposo G., Tian J.M., Karin M., Niemann H., Louvard D. (1998) A novel tetanus neurotoxin-insensitive vesicle-associated membrane protein in SNARE complexes of the apical plasma membrane of epithelial cells.Mol. Biol. Cell 9,14371448.
  • 14
    Guan K.L. & Dixon J.E. (1991) Eukaryotic proteins expressed in Escherichia coli : an improved thrombin cleavage and purification procedure of fusion proteins with glutathione S-transferase. Anal. Biochem. 192,262267.
  • 15
    Hayashi T., McMahon H., Yamasaki S., Binz T., Hata Y., Südhof T.C., Niemann H. (1994) Synaptic vesicle membrane fusion complex : action of clostridial neurotoxins on assembly.EMBO J. 13,50515061.
  • 16
    Hayashi T., Yamasaki S., Nauenburg S., Binz T., Niemann H. (1995) Diassembly of the reconstituted synaptic vesicle membrane fusion complex in vitro.EMBO J. 14,23172325.
  • 17
    Ikonen E., Tagaya M., Ullrich O., Montecucco C., Simons K. (1995) Different requirements for NSF, SNAP, and rab proteins in apical and basolateral transport in MDCK cells.Cell 81,571580.
  • 18
    Jahn R. & Hanson P.I. (1998) SNAREs line up in new environment.Nature 393,1415.
  • 19
    Jongeneel C.V., Bouvier J., Bairoch A. (1989) A unique signature identifies a family of zinc-dependent proteases.FEBS Lett. 242,211214.
  • 20
    Laemmli U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4.Nature 227,680685.
  • 21
    Link E., Edelmann L., Chou J.H., Binz T., Yamasaki S., Eisel U., Baumert M., Südhof T.C., Niemann H., Jahn R. (1992) Tetanus toxin action : inhibition of neurotransmitter release linked to synaptobrevin proteolysis.Biochem. Biophys. Res. Commun. 189,10171023.
  • 22
    Macaulay S.L., Rea S., Gough K.H., Ward C.W., James D.E. (1997) Botulinum E toxin light chain does not cleave SNAP-23 and only partially impairs insulin stimulation of GLUT4 translocation in 3T3-L1 cells.Biochem. Biophys. Res. Commun. 237,388393.
  • 23
    Montecucco C. & Schiavo G. (1994) Mechanism of action of tetanus and botulinum neurotoxins.Mol. Microbiol. 13,18.
  • 24
    Niemann H., Blasi J., Jahn R. (1994) Clostridial neurotoxins : new tools for dissecting exocytosis.Trends Cell Biol. 4,179185.
  • 25
    Pellizzari R., Rossetto O., Lozzi L., Giovedi S., Johnson E., Shone C.C., Montecucco C. (1996) Structural determinants of the specificity for synaptic vesicle-associated membrane protein/synaptobrevin of tetanus and botulinum type B and G neurotoxins.J. Biol. Chem. 271,2035320358.
  • 26
    Pellizzari R., Mason S., Shone C.C., Montecucco C. (1997) The interaction of synaptic vesicle-associated membrane protein/synaptobrevin with botulinum neurotoxins D and F.FEBS Lett. 409,339342.
  • 27
    Ravichandran V., Chawla A., Roche P.A. (1996) Identification of a novel syntaxin- and synaptobrevin/VAMP-binding protein, SNAP-23, expressed in noneuronal tissues.J. Biol. Chem. 271,1330013303.
  • 28
    Rossetto O., Schiavo G., Montecucco C., Poulain B., Deloye F., Lozzi L., Shone C.C. (1994) SNARE motif and neurotoxins.Nature 372,415416.
  • 29
    Rothmann J.E. (1994) Mechanisms of intracellular protein transport.Nature 372,5563.
  • 30
    Schiavo G., Benfenati F., Poulain B., Rossetto O., Polverino de Laureto P., Das Gupta B.R., Montecucco C. (1992) Tetanus and botulinum-B neurotoxins block neurotransmitter release by proteolytic cleavage of synaptobrevin.Nature 359,832835.
  • 31
    Schiavo G., Shone C.C., Bennett M.K., Scheller R.H., Montecucco C. (1995) Botulinum neurotoxin type C cleaves a single Lys-Ala bond within the carboxyl-terminal region of syntaxins.J. Biol. Chem. 270,1056610570.
  • 32
    Schmidt J.J. & Bostian K.A. (1997) Endoproteinase activity of type A botulinum neurotoxin : substrate requirements and activation by serum albumin.J. Protein Chem. 16,1926.
  • 33
    Shone C.C. & Roberts A.K. (1994) Peptide substrate specificity and properties of the zinc-endopeptidase activity of botulinum type B neurotoxin.Eur. J. Biochem. 225,263270.
  • 34
    Shone C.C., Quinn C.P., Wait R., Hallis B., Fooks S.G., Hambleton P. (1993) Proteolytic cleavage of synthetic fragments of vesicle-associated membrane protein, isoform-2 by botulinum type B neurotoxin.Eur. J. Biochem. 217,965971.
  • 35
    Simpson L.L. , ed (1989) Botulinum Neurotoxin and Tetanus Toxin. Academic Press, San Diego.
  • 36
    Trimble W.S., Cowan D.M., Scheller R.H. (1988) VAMP-1 : a synaptic vesicle-associated integral membrane protein.Proc. Natl. Acad. Sci. USA 85,45384542.
  • 37
    Wang G., Witkin W.J., Hao G., Bankaitis V.A., Scherer P.E., Baldini G. (1997) Syndet is a novel SNAP-25 related protein expressed in many tissues.J. Cell Sci. 110,505513.
  • 38
    Washbourne P., Pellizzari R., Baldini G., Wilson M.C., Montecucco C. (1997) Botulinum neurotoxin types A and E require the SNARE motif in SNAP-25 for proteolysis.FEBS Lett. 418,15.
  • 39
    Weber T., Zemelman B.V., McNew J.A., Westermann B., Gmachl M., Parlati F., Söllner T.H., Rothmann J.E. (1998) SNAREpins : minimal machinery for membrane fusion.Cell 92,759772.
  • 40
    Williamson L.C., Halpern J.L., Montecucco C., Brown J.E., Neale E.A. (1996) Clostridial neurotoxins and substrate proteolysis in intact neurons.J. Biol. Chem. 271,76947699.
  • 41
    Yamasaki S., Baumeister A., Binz T., Blasi J., Link E., Cornille F., Roques B., Fykse E.M., Südhof T.C., Jahn R., Niemann H. (1994a) Cleavage of members of the synaptobrevin/VAMP family by types D and F botulinal neurotoxins and tetanus toxin.J. Biol. Chem. 269,1276412772.
  • 42
    Yamasaki S., Hu Y., Binz T., Kalkuhl A., Kurazono H., Tamura T., Jahn R., Kandel E., Niemann H. (1994b) Synaptobrevin/vesicle-associated membrane protein (VAMP) of Aplysia californica : structure and proteolysis by tetanus toxin and botulinal neurotoxins type D and F.Proc. Natl. Acad. Sci. USA 91,46884692.