Evaluation of phosphoramidon and three synthetic phosphonates for inhibition of botulinum neurotoxin B catalytic activity

Article abstract of DOI:10.1002/(SICI)1099-1263(199912)19:1+3.0.CO;2-M

Three putative metalloprotease inhibitors were synthesized and tested for their ability to inhibit the catalytic activity of botulinum neurotoxin B light chain (BoNT/B LC). The compounds were designed to emulate the naturally occurring metalloprotease inhibitor phosphoramidon, which has been reported to be a weak antagonist of BoNT/B action. All three analogs contained the dipeptide Phe-Glu in place of Leu-Trp of phosphoramidon and possessed a phenyl, ethyl or methyl group in place of the rhamnose sugar of the parent compound. The inhibitors were evaluated in a cell-free assay based on the detection of a fluorescent product following cleavage of a 50-mer synaptobrevin peptide ([Pya⁸⁸] S 39-88) by BoNT/B LC. This peptide corresponds to the hydrophilic core of synaptobrevin-2 and contains a fluorescent analog L-pyrenylalanine (Pya) in place of Tyr⁸⁸. Cleavage of [Pya⁸⁸] S 39-88 by BoNT/B LC gives rise to fragments of 38 and 12 amino acid residues. Quantification of BoNT/B-mediated substrate cleavage was achieved by separating the 12-mer fragment (FETSAAKLKRK-Pya) that contains the C-terminal fluorophore and measuring fluorescence at 377 nm. The results indicate that the phenyl-substituted synthetic compound ICD 2821 was slightly more active than phosphoramidon, but analogs with methyl or ethyl substitutions were relatively inactive. These findings suggest that phosphonate monoesters may be useful for providing insights into the structural requirement of BoNT/B protease inhibitors.

Full text of DOI:10.1002/(SICI)1099-1263(199912)19:1+3.0.CO;2-M

JOURNAL OF APPLIED TOXICOLOGY
J. Appl. Toxicol. 19, S5–S11 (1999)
Evaluation of Phosphoramidon and Three
Synthetic Phosphonates for Inhibition of
Botulinum Neurotoxin B Catalytic Activity
†‡
1
2
3
3
Michael Adler, * James D. Nicholson, David F. Starks, Charles T. Kane, Fabrice
4
Cornille and Brennie E. Hackley, Jr.
1
Neurotoxicology Branch, Pharmacology Division, US Army Medical Research Institute of Chemical Defense, Aberdeen
Proving Ground MD, 21010, USA
2
4
3
Astrox Corp., Rockville, MD 20850, Starks Associates, Inc., Buffalo, NY 14213, USA
D e´ partement de Pharmacochimie Mol e´ culaire et Structurale, Universit e´ Ren e´ Descartes, U266 INSERM CNRS UMR
8
600, Paris Cedex 06, France
Key words: phosphoramidon; phosphonate monoesters, zinc metalloprotease; botulinum neurotoxin B; synaptobrevin-2
Three putative metalloprotease inhibitors were synthesized and tested for their ability to inhibit the catalytic
activity of botulinum neurotoxin B light chain (BoNT/B LC). The compounds were designed to emulate
the naturally occurring metalloprotease inhibitor phosphoramidon, which has been reported to be a weak
antagonist of BoNT/B action. All three analogs contained the dipeptide Phe-Glu in place of Leu-Trp of
phosphoramidon and possessed a phenyl, ethyl or methyl group in place of the rhamnose sugar of the
parent compound. The inhibitors were evaluated in a cell-free assay based on the detection of a fluorescent
88
product following cleavage of a 50-mer synaptobrevin peptide ([Pya ] S 39–88) by BoNT/B LC. This
peptide corresponds to the hydrophilic core of synaptobrevin-2 and contains a fluorescent analog L-
88
88
pyrenylalanine (Pya) in place of Tyr . Cleavage of [Pya ] S 39–88 by BoNT/B LC gives rise to fragments
of 38 and 12 amino acid residues. Quantification of BoNT/B-mediated substrate cleavage was achieved by
separating the 12-mer fragment (FETSAAKLKRK-Pya) that contains the C-terminal fluorophore and
measuring fluorescence at 377 nm. The results indicate that the phenyl-substituted synthetic compound ICD
2821 was slightly more active than phosphoramidon, but analogs with methyl or ethyl substitutions were
relatively inactive. These findings suggest that phosphonate monoesters may be useful for providing
insights into the structural requirement of BoNT/B protease inhibitors.
3
failure and death. The pathophysiology of BoNT
INTRODUCTION
intoxication is thought to be a consequence of cleavage
of one of three proteins that are required for the
docking of synaptic vesicles with active zones in the
nerve terminal: synaptobrevin, syntaxin and SNAP-
25. Cleavage of these proteins results from the zinc
The botulinum neurotoxins (BoNTs) comprise a group
of seven dichain protein toxins, A–G, secreted by
the bacterium Clostridium botulinum. Each serotype
1
4
–6
consists of a ca. 100-kDa heavy chain with two func-
tional domains: a C-terminal domain for binding to
ectoacceptors on cholinergic nerve endings and an N-
terminal domain for the cytosolic translocation of the
toxin. The heavy chain is associated with a ca. 50 kDa
light chain (LC) that is responsible for inhibition of
metalloprotease activity of the BoNT LC. Botulinum
neurotoxin light chains possess a characteristic HEXXH
sequence that is a universal feature of zinc metallopro-
teases. The LCs bind zinc in a 1:1 stoichiometric ratio,
require zinc for proteolytic activity and can be inhibited
by exposure to chelators such as N,N,N′N′-tetrakis(2-
2
5
acetylcholine release.
pyridylmethyl) ethylenediamine (TPEN).
Exposure to BoNT leads to the potentially fatal
condition of botulism. The symptoms of intoxication
generally begin with a gradual onset of weakness in
muscles innervated by the cranial nerves, manifested
as diplopia, dysphagia and dysarthria. Symptoms pro-
gress to generalized muscle weakness and difficulty in
breathing that in severe cases culminate in respiratory
The present effort represents a rational approach
towards the design of selective metalloprotease inhibi-
tors for BoNT/B as a basis for subsequent development
of effective therapeutic agents. The inhibitors were
designed to emulate the structure of phosphoramidon,
a naturally occurring metalloprotease inhibitor with
demonstrated, albeit weak, anti-BoNT activity at the
7
mouse neuromuscular junction. Two essential features
of phosphoramidon were incorporated in the design of
the synthetic compounds: a phosphonate functional
group for interacting with the active site zinc and a
*
Correspondence to: M. Adler, Pharmacology Division, US Army
Medical Research Institute of Chemical Defense, 3100 Ricketts Point
Rd, Aberdeen Proving Ground, MD 21010-5400, USA.
8
†
The opinions or assertions contained herein are the private views
dipeptide core for active site recognition. The dipep-
of the authors and are not to be construed as official or as reflecting
the views of the Army or the Department of Defense.
tide Phe-Glu was selected because point mutation stud-
77
ies suggested that Phe plays a key role in the suscep-
‡
This article is a US Government work and is in the public domain
in the United States.
tibility of human synaptobrevin to cleavage by BoNT/B
Received 26 January 1999
Revised 4 May 1999
Published in 1999 by John Wiley & Sons, Ltd.
Accepted 4 May 1999

S6
M. ADLER ET AL.
9
LC. The compounds (Fig. 1) were tested for inhibitory
activity by monitoring cleavage of a fluorescent peptide
that encompasses residues 39–88 of human synaptobre-
by transfer hydrogenation with 10% palladium on
carbon, employing ammonium formate as the source
1
2
of hydrogen, to give 9c as a hydrated diammonium
salt (9c: Analysis calculated for C H N O P · 0.8
88
vin-2 ([Pya ] S 39–88). Two of the analogs (ICD
820-ethoxy and ICD 2822-methoxy) were found to
be relatively inactive. The phenoxy analog (ICD 2821),
however, was more potent than phosphoramidon. These
results suggest that phosphonate monoesters may be
useful for establishing structure–activity requirements
of BoNT/B LC inhibitors.
2
0
29
4
8
2
H O: C, 48.15; H, 6.18; N, 11.23. Found: C, 48.15,
2
48.05; H, 5.77, 5.82; N, 10.61, 10.58).
The identity of the compounds was confirmed
1
31
31
by H-NMR, P-NMR [9a: P-NMR (D O), ␦ 6.56
2
3
1
31
(s, 1P); 9b: P-NMR (D O), ␦ 7.74 (s, 1P); 9c: P-
NMR (D O), ␦ 2.67 (s, 1P)], and (mass spectrum
2
2
analysis [9a: electrospray MS, m/z 401 (M-H); 9b:
electrospray MS, m/z 387 (M-H); 9c: electrospray MS,
1
3
m/z 449 (M-H)].
EXPERIMENTAL
Synthesis of analogs
Evaluation of inhibitory activity
A recently described fluorescence assay was used for
14
Preparation of phosphoramidon analogs is difficult
because of the synthetic challenges associated with
the phosphorus-linked glycopeptide. However, it was
reported that replacement of the rhamnose moiety in
phosphoramidon by small alkoxy groups yielded metal-
loprotease inhibitors with potencies similar to that of
investigating the putative BoNT/B LC inhibitors. The
assay is based on measuring reductions in the rate of
cleavage of a 50 amino acid peptide that spans residues
39–88 of human synaptobrevin-2 and contains a
modified tyrosine fluorophore (pyrenylalanine, Pya) on
8
8
residue 88 ([Pya ] S 39–88). The approach used in
the present study was similar to that reported pre-
8
the parent compound. Such rhamnose-free phos-
1
4,15
phoramidon analogs provided more feasible targets and
were therefore selected for synthesis in the current
study. The essential modification of the phosphorami-
don molecule was the replacement of Leu-Trp by Phe-
Glu so that the dipeptide sequence of the analogs
would correspond to residues adjacent to the BoNT/B
viously.
Briefly, the reactions were carried out at
22° C in a 100-␮l volume. Pure BoNT/B LC (60 nM)
was preincubated with 50 ␮M ZnSO , 1 mM dithiothre-
4
itol and 20 mM HEPES (pH 7.1) for 30 min followed
8
8
by a 30-min incubation with [Pya ] S 39–88 (10 ␮M).
The above LC concentration and incubation time are
within the linear range of the assay. The reaction
4
14
are cleavage site of synaptobrevin-2. The synthetic
scheme for the preparation of the methoxy and ethoxy
target compounds (9a and 9b) is analogous to the
methodology described recently by Bertenshaw et al.
was stopped by addition of 0.9 ml of 0.1% trifluo-
roacetic acid (TFA, v/v) in 72% methanol (v/v). The
8
88
blank contained HEPES and [Pya ] S 39–88 but lacked
and is outlined in Fig. 2. Preparation of the phenoxy
analog (9c) required a different protection–deprotection
strategy and is illustrated in Fig. 3.
BoNT/B LC and ZnSO . Stock solutions of the phos-
4
phoramidon analogs were dissolved in 20 mM HEPES
at pH 7.1 and added to the preincubation mixture.
8
8
Peptides 3a and 3b were prepared by coupling
commercially available t-Boc-l-phenylalanine (1)
with l-glutamic acid, dimethyl ester (2a) or dibenzyl
ester (2b), respectively, using 2-(1H-benzotriazol-
During incubation, BoNT/B LC cleaves [Pya ] S
76 77
39–88 between Gln and Phe , resulting in frag-
ments of 38 and 12 residues, with the latter contain-
8
8
ing the Pya fluorophore. The 12-mer [Pya ] S 77–
88 fluorescent cleavage product was separated from
the intact substrate and the 38-mer non-fluorescent
fragment using C18 Sep-Pak Vac matrix columns
10
1
-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate.
Coupling was followed by removal of the t-butoxycar-
bonyl protecting group with trifluoracetic acid (TFA),
then addition of p-toluenesulfonic acid to give 4a and
8
8
(Waters Corp., Milford, MA). Elution of [Pya ]
S 77–88 was accomplished by use of 65% methanol/
11
4
b as tosylate salts. Chlorophosphates 7a and 7b
1
4
were prepared by addition of phenol (6a) to ethyl
dichlorophosphate (5a) and methyl dichlorophosphate
0.1% TFA solution. Soleihac et al. reported that over
99% of the fluorescent 12-mer cleavage product is
selectively eluted under similar experimental con-
(5b), respectively. Intermediates 8a and 8b were then
prepared by phosphorylation of 4a with chlorophos-
phates 7a and 7b, respectively. The protected phos-
phorylated dipeptides were hydrolyzed with 1.5 M
aqueous LiOH to give 9a and 9b as hydrated tri-
lithium salts containing excess LiCl (9a: Analysis
calculated for C H Li N O P · 1.7 LiCl · 2.75 H O:
ditions. Fractional inhibition was calculated at ␭ex =
343 nm/␭em = 377 nm using the formula 1−(I−B)/
(C−B), where I and C represent fluorescence in the
presence and absence of inhibitor, respectively, and B
is the fluorescence of the blank solution.
16
20
3
2
8
2
C, 35.47; H, 4.74. Found: C, 35.49, 35.40; H, 4.70,
.72; 9b: Analysis calculated for C H Li N O P · 0.55
Materials
4
1
5
18
3
2
8
CH CN · 1.5 LiCl · 1.9 H O: C, 36.73; H, 4.49; N,
Methanol (Optima spectrophotometric grade) was
obtained from Fisher Scientific Co. (Pittsburgh, PA).
Trifluoroacetic acid (peptide synthesis grade) was
obtained from Millipore, Inc. (Beverly, MA). Pure
BoNT/B LC was prepared by Drs Eric Johnson and
Michael Goodnough (Food Research Institute, Madison,
3
2
6
6
.78. Found: C, 36.78, 36.68; H, 4.53, 4.54; N, 6.82,
.78).
Chlorophosphate (7c) was prepared by addition of
benzyl alcohol (6b) to phenyl dichlorophosphate (5c).
Intermediate 8c was then prepared by phosphorylation
of 4b with chlorophosphate (7c). This tribenzyl-
protected phosphorylated dipeptide was debenzylated
WI). Ultrapure ZnSO and HEPES were acquired from
Sigma-Aldrich Corp. (St. Louis, MO) and TPEN was
4
Published in 1999 by John Wiley & Sons, Ltd.
J. Appl. Toxicol. 19, S5–S11 (1999)

PHOSPHONATE MONOESTERS ON BoNT/B PROTEASE ACTIVITY
S7
Figure 1. Structural formulae of ICD 2820, ICD 2821, ICD 2822 and phosphoramidon. The synthetic inhibitors are dipeptides
containing Phe-Glu in which the N-terminal nitrogen of Phe is bound to a phosphorus atom and the substituents are attached
to one of the phosphonate oxygens. The synthetic compounds are identical except for the nature of the substituent, which
consists of an ethyl (ICD 2820), phenyl (ICD 2821) or methyl (ICD 2822) group. Phosphoramidon differs from the synthetic
compounds in the composition of the dipeptide (Leu-Trp in place of Phe-Glu) and by the presence of a rhamnose sugar.
Published in 1999 by John Wiley & Sons, Ltd.
J. Appl. Toxicol. 19, S5–S11 (1999)

S8
M. ADLER ET AL.
Figure 2. Scheme for synthesis of ICD 2820 (l-glutamic acid, N-(N-(ethoxyhydroxyphosphinyl)-l-phenylalanyl)-, trilithium salt) and
ICD 2822 (l-glutamic acid, N-[N-(methoxyhydroxyphosphinyl)-l-phenylalanyl]-, trilithium salt).
purchased from Calbiochem (San Diego, CA). All other
chemicals were reagent grade or higher.
Fig. 4. As is clear from the convergence of the fluor-
88
escence spectra, cleavage of [Pya ] S 39–88 was not
altered significantly by addition of zinc in the range
1
0 ␮M–1 mM. The observation that proteolytic activity
does not require exogenous zinc is consistent with the
high affinity of BoNT LC for the active site zinc.
RESULTS
5
Increasing the zinc concentration to 5 mM, however,
8
8
Effect of ZnSO on BoNT/B-mediated substrate
4
led to a marked inhibition of [Pya ] S 39–88 cleavage.
Inhibition by high zinc concentrations has been
observed in previous studies but its mechanism remains
cleavage
In previous in vitro assays, it has often been unclear
whether exogenous zinc was required to achieve opti-
mal catalytic rates for BoNT LC. To address this
1
6,17
unclear.
Although supplemental zinc was clearly
not required for BoNT/B LC activity, 50 ␮M ZnSO4
88
was routinely included in the incubation mixture to
question, we examined the cleavage of [Pya ] S 39–
avoid complications from potential chelator actions of
8
8 in the absence of added zinc and in the presence
1
8
of 10 ␮M–5 mM ZnSO . The results are illustrated in
test agents or buffers.
4
Published in 1999 by John Wiley & Sons, Ltd.
J. Appl. Toxicol. 19, S5–S11 (1999)

PHOSPHONATE MONOESTERS ON BoNT/B PROTEASE ACTIVITY
S9
Figure 3. Scheme for synthesis of ICD 2821 (l-glutamic acid, N-[N-(phenoxyhydroxyphosphinyl)-l-phenylalanyl]-, diammonium salt).
Effect of TPEN on BoNT/B-mediated substrate
cleavage
Effect of phosphoramidon and synthetic analogs
88
on [Pya ] S 39–88 cleavage
To examine the consequence of removing zinc that is
tightly bound to the active site of BoNT/B LC, experi-
ments similar to those described above were carried
out in the presence of TPEN, a chelator with high
Phosphoramidon and three synthetic phosphonate monoes-
ter compounds were examined for their ability to inhibit
88
cleavage of [Pya ] S 39–88. Figure 6 shows concen-
tration–response data for the effects of phosphoramidon
and synthetic phosphonate monoester compounds on
18
affinity for transition metals such as zinc. Addition
of TPEN to the reaction mixture was expected to
88
BoNT/B LC-mediated cleavage of [Pya ] S 39–88. As
with TPEN, the compounds were incubated with BoNT/B
LC for 30 min prior to addition of substrate. Phosphoram-
idon was found to be a weak inhibitor of BoNT/B LC,
requiring 4 mM to produce a detectable inhibition and
14.7 Ϯ 1.2 mM (mean Ϯ SEM) to achieve half-maximal
88
inhibit [Pya ] S 39–88 proteolysis by chelating zinc
that is bound to the BoNT/B LC active site. A marked
inhibition of BoNT/B catalytic activity was indeed
observed, as illustrated in Fig. 5. In the presence of
TPEN, there was a concentration-dependent decrease in
fluorescence intensity, indicating inhibition of BoNT/B-
mediated proteolysis. The TPEN concentrations of 20,
88
inhibition of [Pya ] S 39–88 catalysis. Synthetic com-
pound ICD 2821 was more potent than phosphoramidon,
with an ic50 of 8.1 Ϯ 1.7 mM. This difference was
significant by the Student’s t-test (P Ͻ 0.05). The analogs
with either an ethyl (ICD 2820) or a methyl (ICD 2822)
substitution displayed little inhibitory activity up to the
highest concentration tested (12 mM).
3
0, 40 and 50 ␮M reduced the fluorescence intensity
at 377 nm to 77, 68, 53 and 18% of control, respect-
ively (Fig. 5). These results are consistent with the
5
requirement of zinc for BoNT/B-mediated proteolysis
and serve to validate the assay for unknown inhibitors.
Published in 1999 by John Wiley & Sons, Ltd.
J. Appl. Toxicol. 19, S5–S11 (1999)

S10
M. ADLER ET AL.
88
4
Figure 4. Spectra of the 12-mer [Pya ] S 77–88 fluorescent peptide (FETSAAKLKRK-Pya) in the absence and presence of ZnSO .
8
8
Production of [Pya ] S 77–88 from BoNT/B LC-mediated cleavage of the 50-mer parent peptide was independent of exogenous
zinc up to 1 mM but underwent marked inhibition after addition of 5 mM ZnSO
4
.
88
Figure 5. Fluorescence spectra of the 12-mer [Pya ] S 77–88 cleavage product in the absence and presence of the metal chelator
TPEN. The progressive reduction in substrate cleavage with increasing TPEN concentration was attributed to removal of zinc
bound to the active site of BoNT/B LC. Zinc sulfate was omitted from the incubation media in this experiment.
7
muscles to BoNT/B; and the small size of phos-
phoramidon makes this compound an attractive candi-
date for drug development. When examined in the
current assay, which measures inhibition of the metallo-
protease activity of BoNT/B LC directly, phosphorami-
don was found to be a relatively poor inhibitor, effec-
tive only in the millimolar range.
In designing inhibitors, it was assumed that the low
potency of phosphoramidon may reflect its inability to
fit optimally into the active site of BoNT/B LC,
because its dipeptide composition (Leu-Trp) differs
markedly from Phe-Glu, which is adjacent to the cleav-
age site of the substrate, synaptobrevin-2. A potential
solution was to synthesize compounds of comparable
DISCUSSION
The results of the present investigation suggest that it
is possible to inhibit the catalytic activity of BoNT/B
LC by using small dipetide metalloprotease inhibitors.
The synthetic compounds ICD 2820, ICD 2821 and
ICD 2822 were modeled on the structure of phos-
phoramidon. This naturally occurring metalloprotease
inhibitor was selected as the starting point for the
current synthesis for two compelling reasons: phos-
phoramidon was demonstrated to delay the onset of
paralysis following exposure of mouse diaphragm
Published in 1999 by John Wiley & Sons, Ltd.
J. Appl. Toxicol. 19, S5–S11 (1999)

PHOSPHONATE MONOESTERS ON BoNT/B PROTEASE ACTIVITY
S11
tion is highly unfavorable: instead, a methyl, ethyl or
phenyl group was attached to this position based on
reports that such substitutions led to little loss of
8
activity in matrix metalloproteases.
Of the three analogs tested in the present study, those
with small alkyl substituents were found to be relatively
inactive even at the highest concentration tested (12 mM),
whereas the compound with a phenyl substituent was
more potent than phosphoramidon (Fig. 6), suggesting that
an aromatic or bulky group may be important for activity.
These results, coupled with the high potency displayed
by the isocoumarin analog ICD 1578 against BoNT/B
15
LC, suggest the existence of an aromatic binding pocket
near the active site of BoNT/B LC. The relatively low
potency of ICD 2821 in spite of its correct dipeptide
sequence (Figure 1) suggests that the complementary sites
for the dipeptide may not be accessible in the native
Figure 6. Concentration–response data for the inhibition of
BoNT/B LC activity by phosphoramidon (᭹), ICD 2820 (̄), ICD
19
toxin. Future synthetic efforts exploiting the features of
the first-generation lead compounds should give rise to
more specific and more potent inhibitors of BoNT/B LC.
The initial albeit moderate success of this approach sug-
gests that a pharmacological treatment for BoNT intoxi-
cation is a realistic goal.
2
821 (̆) and ICD 2822 (v). The symbols represent means of
four determinations. The curves for phosphoramidon and ICD
2
821 were fit by non-linear regression with ic50 values of
4.7 Ϯ 1.2 and 8.1 Ϯ 1.7 mM, respectively. The fits to ICD 2820
1
and ICD 2822 data are arbitrary.
size with a similar phosphonate functional group to
that found in phosphoramidon but with a dipeptide
sequence optimized to interact with the active site of
BoNT/B LC. Coupling of a rhamnose sugar was not
attempted in the synthetic compounds because the reac-
Acknowledgement
This work was supported in part by the US Army Medical Research
and Development Command under Contract No. DAMD17-93-C-
3003 awarded to D.F.S. and C.T.K.
REFERENCES
1
. L. L. Simpson, Peripheral actions of the botulinum toxins.
In Botulinum Neurotoxin and Tetanus Toxin, ed. by L. L.
Simpson, pp. 153–178, Academic Press, New York (1989).
. J. O. Dolly, A. C. Ashton, C. McInnes, J. D. F. Wadsworth,
B. Poulain, L. Tauc, C. C. Shone and J. Melling, Clues to the
multi-phasic inhibitory action of botulinum neurotoxins on
release of transmitters. J. Physiol. (Paris) 84, 237–246 (1990).
. E. Habermann and F. Dreyer, Clostridial neurotoxins: hand-
ling and action at the cellular and molecular level. Curr.
Top. Microbiol. Immunol. 129, 93–179 (1986).
11. W. J. Hoekstra, S. S. Sunder and R. J. Cregge, Large-
scale synthesis of anticoagulant decapeptide MDL 28050.
Tetrahedron 48, 307–318 (1992).
12. M. K. Anwer and A. F. Spatola, An advantageous method
for the rapid removal of hydrogenolysable protecting
groups under ambient conditions; synthesis of leucine-
enkephalin. Synthesis 929–932 (1980).
13. D. F. Starks, C. T. Kane, J. D. Nicholson, B. E. Hackley and
M. Adler, Analogs of phosphoramidon as metalloprotease
inhibitors for botulinum toxin serotype B. Proc. 1996 Medi-
cal Defense Bioscience Review 3, APG 1608–1615,
ADA 321842. USAMRICD, MD (1996).
14. J.-M. Soleihac, F. Cornille, L. Martin, C. Lenoir, M.-C. Four-
nie-Zaluski and B. P. Roques, A sensitive and rapid fluor-
escence-based assay for determination of tetanus toxin
peptidase activity. Anal. Biochem. 241, 120–127 (1996).
15. M. Adler, J. D. Nicholson, F. Cornille and B. E. Hackley,
Jr., Efficacy of a novel metalloprotease inhibitor on botu-
linum neurotoxin B activity. FEBS Lett. 429, 234–238
(1998).
16. P. Foran, C. C. Shone and J. O. Dolly, Differences in
the protease activities of tetanus and botulinum B toxins
revealed by the cleavage of vesicle-associated membrane
protein and various sized fragments. Biochemistry 33,
15365–15374 (1994).
17. J. L. Nowakowski, B. C. Courtney, Q. A. Bing and M. Adler,
Production of an expression system for a synaptobrevin
fragment to monitor cleavage by botulinum neurotoxin B.
J. Prot. Chem. 17, 453–462 (1998).
18. G. Anderegg, E. Hubmann, N. G. Podder and F. Wenk, Die
thermodynamik der metallokomplexbildung mit bis-, tris
und tertrakis [(2-pyridyl)methyl]-aminen. Helv. Chim. Acta
60, 123–140 (1977).
19. F. Cornille, L. Martin, C. Lenoir, D. Cussac, B. P. Roques
and M.-C. Fournie-Zaluski, Cooperative exosite-dependent
cleavage of synaptobrevin by tetanus toxin light chain. J.
Biol. Chem. 272, 3459–3464 (1997).
2
3
4
. G. Schiavo, F. Benfenati, B. Poulaine, O. Rossetto, P. Pol-
verino de Laureto, B. R. DasGupta and C. Montecucco,
Tetanus and botulinum
B
neurotoxins block neuro-
transmitter release by proteolytic cleavage of synaptobre-
vin. Nature 359, 832–835 (1992).
5
. G. Schiavo, O. Rossetto, F. Tonello and C. Montecucco,
Intracellular targets and metalloprotease activity of tetanus
and botulinum neurotoxins. Curr. Top. Microbiol. Immu-
nol. 195, 257–274 (1995).
. W. Volknandt, The synaptic vescicle and its targets. Neuro-
science 64, 277–300 (1995).
. S. S. Deshpande, R. E. Sheridan and M. Adler, A study of
zinc-dependent metalloendopeptidase inhibitors as phar-
macological antagonists in botulinum neurotoxin poison-
ing. Toxicon 33, 551–557 (1995).
. S. R. Bertenshaw, R. S. Rogers, M. K. Stern, B. H. Norman,
W. M. Moore, G. M. Jerome, L. M. Branson, J. F. McDon-
ald, E. G. McMahon and M. A. Palomo, Phosphorus-con-
taining inhibitors of endothelin converting enzyme: effects
of the electronic nature of phosphorus on inhibitor
potency. J. Med. Chem. 36, 173–176 (1993).
6
7
8
9
. C. C. Shone and A. K. Roberts, Peptide substrate specificity
and properties of the zinc-endopeptidase activity of botuli-
num type B neurotoxin. Eur. J. Biochem. 225, 263–270 (1994).
1
0. J. Dudash, Jr., J. Jiang, S. C. Mayer and M. M. Joulli e´ ,
Comparative study of selected coupling reagents in dipep-
tide synthesis. Synth Comm. 23, 349–356 (1993).
Published in 1999 by John Wiley & Sons, Ltd.
J. Appl. Toxicol. 19, S5–S11 (1999)