Superactivation of the botulinum neurotoxin serotype A light chain metalloprotease: A new wrinkle in botulinum neurotoxin

Article abstract of DOI:10.1021/ja057699z

Small molecules based upon a 2-acylguanidine-5-phenyl thiophene scaffold that can activate the light chain metalloprotease of botulinum neurotoxin serotype A (BoNT LC/A) by an apparent reduction in K_m are reported. On the basis of structure-act

Full text of DOI:10.1021/ja057699z

Published on Web 03/09/2006
Superactivation of the Botulinum Neurotoxin Serotype A Light Chain
Metalloprotease: A New Wrinkle in Botulinum Neurotoxin
Laura A. McAllister, Mark S. Hixon, Jack P. Kennedy, Tobin J. Dickerson, and Kim D. Janda*
Departments of Chemistry and Immunology, The Skaggs Institute for Chemical Biology, and Worm Institute for
Research and Medicine (WIRM), The Scripps Research Institute, 10550 North Torrey Pines Road,
La Jolla, California 92037
Received November 11, 2005; E-mail: kdjanda@scripps.edu
The seven neurotoxins of the anaerobic spore-forming bacterium
Clostridium botulinum (BoNTs A-G) are the most lethal poisons
known, with BoNT serotype A having a LD50 for a 70 kg human
of a mere 0.8 µg by inhalation.1 Historically associated with food
poisoning, these proteinaceous toxins inhibit the release of acetyl-
choline at neuromuscular junctions by cleavage of SNARE proteins,
resulting in progressive flaccid paralysis.2 It is as a result of this
potent neurotoxic activity that BoNTs are of substantial concern
as potential bioterrorist weapons.3 Paradoxically, the use of BoNT
in medicine has become increasingly popular in recent years, with
the most well-known application being its use in the cosmetic
treatment of glabellar lines (i.e., forehead wrinkles).4 Less well
Figure 1. 2-Acylguanidyl-5-phenyl thiophenes synthesized to examine
BoNT LC/A activation.
publicized is the emergence of BoNT as an excellent therapeutic
tool in the treatment of a variety of serious medical conditions.5
Multiple sclerosis, stroke, cerebral palsy, migraine, and backache
are just a few of the conditions for which BoNT therapy has proven
effective. However, repeated toxin exposure can lead to the
development of a significant immune response, resulting in a
considerable drawback associated with BoNT therapy.6 Tolerance
develops most rapidly when patients are treated frequently with
high doses of the toxin. We hypothesized that administration of
BoNT in combination with a molecule which can “activate” the
proteolytic activity of the toxin would allow the use of lower doses,
thus reducing the unintended adaptive immune response. Herein,
we disclose the discovery of the first small molecule activators of
BoNT A catalytic activity based upon a 2-acylguanidine-5-phenyl
thiophene scaffold that display structure-activity relationships
(SAR) indicative of discrete toxin binding and result in the greatest
protease activation reported to date.
During the course of our recent investigations into the identifica-
tion of BoNT A light chain metalloprotease (BoNT LC/A)
inhibitors, we found that modest inhibition could be obtained with
molecules as simple as arginine hydroxamic acid;7 in essence, this
compound is simply comprised of a zinc-binding moiety and a
guanidinyl group. Using this as a guide, we prepared a series of
compounds containing a 2-acylthiophene moiety as a zinc-binding
functionality, combined with acylguanidyl groups as an arginine
side chain mimetic. While under our standard FRET-based screen-
ing conditions7 no inhibition was observed, however, we were
surprised to discover that one compound (1a) consistently produced
a 2-fold enhancement of BoNT LC/A catalytic activity (Table 1).
Gratifyingly, comparable activation was also observed with the
native substrate SNAP-25(141-206),7 indicating that the activation
was not the result of a fluorescence artifact.
Table 1. Structure-Activity Relationship of
5-Phenyl-2-acylguanidyl Thiophene Activators of BoNT LC/A under
Standard Screening Conditions
max activation
compound
X
Y
(compound concn)
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
2
S
S
S
S
S
S
S
S
S
S
S
O
S
O
S
S
3-SEt
4-SEt
4-nPr
4-fold (500 µM)
3.5-fold (100 µM)
7-fold (200 µM)
3-fold (500 µM)
7-fold (20 µM)
4-fold (20 µM)
inactive
inactive
inactive
inactive
inactive
3-SMe
4-SnBu
3-SnBu
3-SnBn
3-OMe
2-SMe
2-SEt
H
3-SEt
inactive
inactive
inactive
inactive
3
4
5
6
3-SEt
3-SEt
inactive
derivatives 1k or 3. Substitution of the phenyl ring at the 2-position
also yielded inactive compounds (1i and 1j); however, 4-thioethyl
substituted analogue 1b was found to be a more potent activator
than lead compound 1a. On the basis of these data, we focused
our attention on the nature of substituents at the 3- and 4-positions.
4-Butylthioether 1e and 3-butylthioether 1f were both significantly
better activators than 1a or 1b, illustrating that increasing the length
of the aliphatic chain at the 3- and 4-positions enhances activity.
However, replacement of this aliphatic chain with an aromatic
substituent as in 1g abrogated activation. Compound 6 was
examined to ascertain the importance of the distance between the
guanidyl group and thiophene core; however, this modification
resulted in a complete loss of activity.
Using activator 1a as a lead, we explored the molecular
requirements for activation by synthesizing and testing compounds
1-6 (Figure 1) for their effect on BoNT LC/A activity. Interest-
ingly, the corresponding furan analogue 2, carboxylic acid 5, and
methyl ether substituted derivative 1h were nonactivating. In
addition, no activation was observed using phenyl or 2-pyridyl
Given the tremendous clinical promise of a BoNT activator
coupled with the lack of a fully characterized allosteric site on the
BoNT light chain,8 we examined BoNT activation by these
9
4176
J. AM. CHEM. SOC. 2006, 128, 4176-4177
10.1021/ja057699z CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Figure 3. BoNT LC/A activation profiles as a function of 1e and several
detergents: 1e (black circles), CTAB (blue circles), CPB (blue triangles),
Tween-20 (red circles), NLS (red triangles), CHAPS (red squares), and
deoxycholate (red diamonds). Data represent the average of duplicate runs.
Figure 2. BoNT LC/A activation in the presence of 100 µM 1b (red
triangles), 100 µM 1c (black circles), and 40 µM 1e (blue squares) as a
function of substrate concentration. Data represent the average of duplicate
measurements.
been demonstrated recently in a thorough study of the activation
and inhibition of T. lanuginosus lipase by detergents.11 Here, no
evidence of micelles or premicellar aggregates was found as judged
from dynamic light scattering and fluorescence experiments. Given
that the most highly activating compounds 1e and 1f show maximal
activity at lower concentrations than these cationic detergents, the
compilation of this with our previous SAR data suggests that these
compounds activate BoNT LC/A by discrete binding to the toxin
or its substrate, thereby facilitating catalytic activity.
In summary, we have unveiled a small molecule scaffold based
on 2-acyl guanidyl-5-phenyl thiophenes that strongly activates
BoNT LC/A catalytic activity through an apparent reduction in Km.
The activation profile and structure-activity relationship for
activation suggests the presence of a specific binding domain on
the enzyme and not a “detergent-like” mechanism. As the impor-
tance of BoNT in medicine continues to expand, adaptive immune
responses to the toxin must be addressed. The discovery and
optimization of small molecule activators may ultimately provide
a valuable method for minimizing BoNT dosage, thereby increasing
BoNT clinical efficacy.
compounds in greater detail in order to understand the mechanism
of this phenomenon. Kinetically, the catalytic activation provided
by compounds 1a-1f could result from modulation of kcat, Km, or
both. Enhancement in kcat only will produce equal activation at all
substrate concentrations, while enhancements in Km will produce
greater activation at low substrate concentration and no activation
at saturating substrate concentration. Enhancements in both kcat and
Km will produce the greatest activation at limiting substrate
concentration while asymptotically approaching a constant amount
of activation at saturating substrate concentration. Thus, varying
the substrate concentration at a fixed compound concentration
allows one to distinguish which mechanism is in effect. Examination
of the three most active compounds (1a, 1c, and 1e) revealed a
reduction in Km and little to no effect on kcat (Figure 2). Critically,
the 14-fold rate enhancement shown by 1e at limiting substrate
concentrations is the greatest activation reported for a protease.
Indeed, 2-fold activation of a protease has been previously referred
to as a state of “superactivation”.9
Two of the elements required for activation, the aliphatic
substituents on the phenyl ring and the positively charged 2-acyl
guanidyl group, suggest the structural motif of a detergent and,
therefore, rate enhancement from a nonspecific effect. To address
this issue, the impact on BoNT LC/A activity of six detergents
with varying electrostatic profiles was investigated. All of the
detergents tested were inhibitory above their critical micelle
concentration (CMC); however, at lower concentrations, varying
activation profiles became apparent (Figure 3). Neutral and zwit-
terionic detergents, such as Tween-20 and CHAPS, respectively,
had little effect on enzyme activity, while the anionic detergents
deoxycholate and N-lauryl sarcosine (NLS) were particularly
inhibitory. Alternatively, the cationic cetyl-based detergents cetyl-
trimethylammonium bromide (CTAB) and cetyl pyridium bromide
(CPB) were activating when present 5-fold below their 500 µM
CMC, but became strongly inhibitory as CMC was approached.
Activation of proteases by cationic detergents has been observed
and studied extensively with R-chymotrypsin.10 In contrast to our
studies, activation of R-chymotrypsin occurs only at detergent
concentrations above CMC and, therefore, is an interfacial mech-
anism of activation whereby enzyme and substrate interact at the
interface between detergent micelles and surrounding solvent.
Conversely, due to the observed enhancement below CMC, the
activation of BoNT LC/A by CTAB and CPB is likely due to
discrete binding of one or more detergent molecules to the enzyme.
Precedence for this sort of noninterfacial activation mechanism has
Acknowledgment. This work was supported by the NIH
(AI066507) and The Skaggs Institute for Chemical Biology.
Supporting Information Available: Full experimental procedures
and characterization for all compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Burnett, J. C.; Henchal, E. A.; Schmaljohn, A. L.; Bavari, S. Nat. ReV.
Drug DiscoVery 2005, 4, 281-297.
(2) Simpson, L. L. Annu. ReV. Pharmacol. Toxicol. 2004, 44, 167-193.
(3) Madsen, J. M. Clin. Lab. Med. 2001, 21, 593-605.
(4) Shukla, H. D.; Sharma, S. K. Crit. ReV. Microbiol. 2005, 31, 11-18.
(5) (a) Montecucco, C.; Molgo, J. Curr. Opin. Pharmacol. 2005, 5, 274-
279. (b) Munchau, A.; Bhatia, K. P. BMJ 2000, 320, 161-165.
(6) Brin, M. F. Muscle NerVe Suppl. 1997, 20, S146-S168.
(7) Boldt, G. E.; Kennedy, J. P.; Hixon, M. S.; McAllister, L. A.; Barbieri,
J. T.; Tzipori, S.; Janda, K. D., J. Comb. Chem. 2006, in press.
(8) Breidenbach, M. A.; Brunger, A. T. Nature 2004, 432, 925-929.
(9) Sarcar, S.; Jain, T. K.; Maitra, A. Biotechnol. Bioeng. 1992, 39, 474-
478.
(10) (a) Viparelli, P.; Alfani, F.; Cantarella, M. Biochem. J. 1999, 344, 765-
773. (b) Celej, M. S.; D’Andrea, M. G.; Campana, P. T.; Fidelio, G. D.;
Bianconi, M. L. Biochem. J. 2004, 378, 1059-1066.
(11) Mogensen, J. E.; Sehgal, P.; Otzen, D. E. Biochemistry 2005, 44,
1719-1730.
JA057699Z
9
J. AM. CHEM. SOC. VOL. 128, NO. 13, 2006 4177