Pharmacophore-guided lead optimization: The rational design of a non-zinc coordinating, sub-micromolar inhibitor of the botulinum neurotoxin serotype a metalloprotease

Article abstract of DOI:10.1016/j.bmcl.2009.01.111

Botulinum neurotoxins, responsible for the neuroparalytic syndrome botulism, are the deadliest of known biological toxins. The work described in this study was based on a three-zone pharmacophore model for botulinum neurotoxin serotype A light chain inhib

Full text of DOI:10.1016/j.bmcl.2009.01.111

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5811–5813
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pharmacophore-guided lead optimization: The rational design of a non-zinc
coordinating, sub-micromolar inhibitor of the botulinum
neurotoxin serotype a metalloprotease
b
c
a
a
d
e
J. C. Burnett a, , C. Wang , J. E. Nuss , T. L. Nguyen , A. R. Hermone , J. J. Schmidt , R. Gussio ,
*
P. Wipf b, , S. Bavari
c,
*
*
a SAIC Frederick, Inc., Target Structure-Based Drug Discovery Group, Frederick, National Cancer Institute at Frederick, P.O. Box B, Frederick, MD 21702, United States
b Combinatorial Chemistry Center, University of Pittsburgh, 219 Parkman Avenue, Pittsburgh, PA 15260, United States
c Department of Immunology, Target Identification, and Translational Research, Division of Bacteriology, United States Army Medical Research Institute of Infectious Diseases,
1425 Porter Street, Frederick, MD 21702, United States
d Department of Cell Biology and Biochemistry, Division of Biochemistry, United States Army Medical Research Institute of Infectious Diseases, 1425 Porter Street,
Frederick, MD 21702, United States
e Information Technology Branch, Developmental Therapeutics Program, National Cancer Institute at Frederick, P.O. Box B, Frederick, MD 21702, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Botulinum neurotoxins, responsible for the neuroparalytic syndrome botulism, are the deadliest of
known biological toxins. The work described in this study was based on a three-zone pharmacophore
model for botulinum neurotoxin serotype A light chain inhibition. Specifically, the pharmacophore
defined a separation between the overlaps of several different, non-zinc(II)-coordinating small molecule
chemotypes, enabling the design and synthesis of a new structural hybrid possessing a Ki = 600 nM
( 100 nM).
Received 11 December 2008
Accepted 7 January 2009
Available online 13 February 2009
Keywords:
Inhibitor design
Inhibitor synthesis
Botulinum neurotoxin
Metalloprotease
Biothreat agent
Pharmacophore
Drug design
Ó 2009 Published by Elsevier Ltd.
Botulinum neurotoxins (BoNTs) are secreted by anaerobic,
spore forming bacteria within the genus Clostridium.1 Structurally,
the toxins are composed of a 100 kDa heavy chain (HC) and a
50 kDa light chain (LC) that are covalently linked through a disul-
fide bridge.2,3 The HC component binds to pre-synaptic membrane
receptors at neuromuscular junctions and facilitates toxin internal-
ization.1 The LC component, which is liberated from the HC via the
intracellular reduction of the disulfide linker, is a zinc (Zn(II))
metalloprotease that cleaves components of the SNARE (soluble
N-ethylmaleimide-sensitive fusion protein attachment protein
receptor) protein complex (composed of SNAP-25 (synaptosomal-
associated protein of 25 kDa), VAMP (vesicle-associated membrane
protein, also referred to as synaptobrevin), and syntaxin).4,5 Of the
seven known BoNT serotypes (classified A–G), A and E cleave
SNAP-25,6,7 serotypes B, D, F, and G cleave VAMP,8–11 and serotype
C1 cleaves both SNAP-25 and syntaxin.12 Botulinum neurotoxin
mediated proteolysis of SNARE proteins inhibits the exocytosis of
acetylcholine into neuromuscular junctions.1 This in turn results
in life threatening flaccid paralysis.
Due to the lethality of BoNTs (e.g., the lethal intravenous dose of
BoNT serotype A in humans is 1 ng kgÀ1)1 and their potential use as
bioweapons,13,14 there is an urgent need for the identification and
development of small molecule, non-peptidic, inhibitors (SMNPIs)
to counter the toxins’ metalloprotease activity—post intoxication;
currently available antitoxin vaccines13 are not effective after neu-
ronal internalization of the toxin. At this time, the majority of such
research has focused on the BoNT serotype A LC (BoNT/A LC), as it
induces long-term neuroparalysis15 (i.e., months) and has been
responsible for the majority of reported BoNT poisonings.13 Yet,
of the described SMNPIs of the BoNT/A LC,16–26 only Zn(II)-coordi-
nating o,p-dichlorocinnamic hydroxamate is reported to possess an
IC50 < 0.5
HPLC-based assay, it was significantly less potent, with an
l
M;16 however, when this compound was tested in our
* Corresponding authors. Tel.: +1 804 225 0527; fax: +1 804 827 3664 (J.C.B.);
tel.: +1 412 624 8606; fax: +1 412 624 0787 (P.W.); tel.: +1 301 619 4246; fax: +1
301 619 2348 (S.B.).
E-mail addresses: burnett.james@mail.nih.gov (J.C. Burnett), pwipf@pitt.edu
(P. Wipf), sina.bavari@us.army.mil (S. Bavari).
IC50 > 29 l
M.17 Moreover, in a subsequent publication by Eubanks
et al.,21 this molecule was reported to be neurotoxic and displayed
poor in vivo activity.
0960-894X/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2009.01.111

5812
J. C. Burnett et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5811–5813
In contrast to SMNPIs incorporating a relatively unselective
ven the fact that we have been unable to improve the potency of
our lead for therapeutic development, NSC 240898,17 solely within
pharmacophore Zones 1 and 2 (Fig. 1b). Specifically, we recently
reported the synthesis of fourteen congeners of this SMNPI that
were generated to explore the possibility of improving its affinity
for the BoNT/A LC substrate binding cleft.28 However, only one
derivative, possessing a conservative sulfur replacement of the par-
ent SMNPI’s ether oxygen, provided equipotent inhibition.28 In
contrast, the substitution of functional groups on the NSC
240898 central aromatic ring, the incorporation of additional het-
eroatoms in the arene moiety, and the replacement and/or rear-
rangement of the inhibitor’s two amidine functions, all resulted
in less potent derivatives.28 Thus, in light of the fact that we have
been unable to synthetically optimize inhibitory potency within
Zones 1 and 2 for this scaffold, our optimization strategy was al-
tered to focus on the incorporation of a Zone 3 component.
Zn(II)-coordinating moiety, such as a hydroxamate,16,20,24,26 our re-
search has focused on the pharmacophore-based identification and
development of non-Zn(II)-coordinating SMNPIs.17–19,25 In particu-
lar, we have used three-dimensional search queries derived from
our gas phase pharmacophore for BoNT/A LC inhibition. This model
relies solely on comparisons of the structures of the small mole-
cules and their accompanying SAR (structure activity relation-
ships), and has resulted in the identification of a variety of non-
Zn(II)-coordinating SMNPIs, including 4-amino-7-chloroquinoline
(4,7-ACQ) conjugates18,19,25, multi-heterocyclic dicationic com-
pounds17,19, and a rigid diazachrysene chemotype.27 Moreover, as
novel SMNPI chemotypes are identified, they are used to further
refine the model, thereby improving its ability to guide SMNPI
identification and optimization. As an example, we recently re-
ported a refined 3-Zone gas-phase pharmacophore (Fig. 1a) that
was generated based on the separation of different SMNPI chemo-
type overlays.27 In the same study, we validated this pharmaco-
phore via the discovery a new terephthalamide-based SMNPI
(NSC 104999).27
The design of an analog of NSC 240898 incorporating a Zone 3
component relied on the previous hypothesis that 4,7-ACQ-based
SMNPI Q2-15 (IC50 = 11.3 lM ( 4.7 lM), and 60% inhibition at 20
uM conc.19) occupies pharmacophore Zones 2 and 3 (Fig. 1c).27
Hence, by overlaying these two SMNPIs (i.e., NSC 240898 and
Q2-15) within the context of the 3-Zone pharmacophore, a poten-
tial ‘hybrid’ 3-Zone inhibitor retaining the entire structure of NSC
240898, but with a 4,7-ACQ moiety tethered via a propyl linker,
is a plausible synthetic candidate (Fig. 1d). However, because
NSC 240898 is not symmetrical, and meets the criteria of pharma-
cophore Zones 1 and 2 in two orientations (Fig. 1b), two designs
were possible (Fig. 1d). In the first, a 4,7-ACQ substructure is teth-
ered to the amidine substituent located on the 6 position of the
NSC 240898 indole component. In the second, the 4,7-ACQ is teth-
ered to the amidine located on the 4 position of NSC 240898’s ter-
minal phenyl component.
The identification of a third pharmacophore zone for inhibitor
occupancy is of significant importance to our research program gi-
A straightforward method to simultaneously prepare designs 1
and 2 (Fig. 1d) was the preparation of a regioisomer mixture
(Scheme 1). In brief, 1 equiv of NSC 240898 (the synthesis of which
has been described29) was coupled with 4, which was prepared via
the nucleophilic substitution of 3 with propane-1,3-diamine. The
reaction was carried out for 30 min in a monomode microwave
reactor at 150 °C, and RP-HPLC purification afforded a 4:1 mixture
of TFA salts 1 and 2.
In vitro testing of 1 and 2 in our HPLC-based assay17,19,30–34 re-
sulted in 95% BoNT/A LC inhibition (at 20 lM conc. SMNPI). A more
Figure 1. The 3-Zone pharmacophore for BoNT/A LC inhibition provided the design
strategy for regioisomers 1 and 2. (a) The 3-Zone pharmacophore. Zone 1 (black text
and shapes) consists of a cationic component (black oval) and a planar component
(black rectangle). This Zone may also contain a component that links it to Zone 2.
Zone 2 (red shapes) consists of a cationic component (red oval), a planar component
(red rectangle), and may also possess a component that links it to Zone 1. Note,
while either Zone 1 or Zone 2 may possess the linker component, the Zones cannot
simultaneously possess linker components. Zone 3 (blue rectangle) consists of a
hydrophobic component that may be aromatic or aliphatic. (b) The previously
described SMNPI NSC 24089817 (cyan carbons, red oxygen, and blue nitrogens) fits
equally well within Zones 1 and 2 of the pharmacophore with two orientations. (c)
The previously described SMNPI Q2-1519 (magenta carbons, blue nitrogens, and
light green chlorines) occupies Zone 2 and Zone 3 of the pharmacophore.27 (d) 3-
Zone regioisomer designs 1 and 2. In each case the Zone 3 component of Q2-15
(magenta carbons) is coupled to the Zone 2 amidine component of NSC 240898
(cyan carbons) via a propyl bridge. Since NSC 240898 occupies Zones 1 and 2 in two
different orientations, two regioisomeric designs are possible.
Scheme 1. Synthesis of regioisomers 1 and 2. aReagents and conditions: (a) 120–
130 °C, 6–8 h. (b) PhSH, EtOH,
lW, 150 °C, 30 min. (c) RP-HPLC: MeOH:H2O (0.1%
TFA), 30:70 to 100:0.

J. C. Burnett et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5811–5813
5813
2. Lacy, D. B.; Tepp, W.; Cohen, A. C.; DasGupta, B. R.; Stevens, R. C. Nat. Struct. Biol.
1998, 5, 898.
3. Turton, K.; Chaddock, J. A.; Acharya, K. R. Trends Biochem. Sci. 2002, 27, 552.
4. Singh, B. R. Nat. Struct. Biol. 2000, 7, 617.
5. Foran, P. G.; Davletov, B.; Meunier, F. A. Trends Mol. Med. 2003, 9, 291.
6. Blasi, J.; Chapman, E. R.; Link, E.; Binz, T.; Yamasaki, S.; De Camilli, P.; Sudhof, T.
C.; Niemann, H.; Jahn, R. Nature 1993, 365, 160.
7. Binz, T.; Blasi, J.; Yamasaki, S.; Baumeister, A.; Link, E.; Sudhof, T. C.; Jahn, R.;
Niemann, H. J. Biol. Chem. 1994, 269, 1617.
detailed analysis of the inhibition kinetics revealed that the regioi-
somers possess a Ki = 600 nM ( 100 nM). To the best of our knowl-
edge, this is the most potent inhibition of the BoNT/A LC provided
by a non-Zn(II)-coordinating SMNPI reported to date. Moreover,
this result further validates the 3-Zone pharmacophore for BoNT/
A LC inhibition (Fig. 1a), and demonstrates the usefulness of this
model for guiding the development of more potent SMNPIs. Specif-
ically, the utility of the pharmacophore is evidenced by the fact
that coupling NSC 240898, a 2-Zone inhibitor (Fig. 1b) possessing
8. Schiavo, G.; Benfenati, F.; Poulain, B.; Rossetto, O.; Polverino de Laureto, P.;
DasGupta, B. R.; Montecucco, C. Nature 1992, 359, 832.
9. Schiavo, G.; Malizio, C.; Trimble, W. S.; Polverino de Laureto, P.; Milan, G.;
Sugiyama, H.; Johnson, E. A.; Montecucco, C. J. Biol. Chem. 1994, 269, 20213.
10. Schiavo, G.; Rossetto, O.; Catsicas, S.; Polverino de Laureto, P.; DasGupta, B. R.;
Benfenati, F.; Montecucco, C. J. Biol. Chem. 1993, 268, 23784.
11. Schiavo, G.; Shone, C. C.; Rossetto, O.; Alexander, F. C. H.; Montecucco, C. J. Biol.
Chem. 1993, 268, 11516.
12. Blasi, J.; Chapman, E. R.; Yamasaki, S.; Binz, T.; Niemann, H.; Jahn, R. Embo. J.
1993, 12, 4821.
13. Arnon, S. S.; Schechter, R.; Inglesby, T. V.; Henderson, D. A.; Bartlett, J. G.;
Ascher, M. S.; Eitzen, E.; Fine, A. D.; Hauer, J.; Layton, M.; Lillibridge, S.;
Osterholm, M. T.; O’Toole, T.; Parker, G.; Perl, T. M.; Russell, P. K.; Swerdlow, D.
L.; Tonat, K. Jama 2001, 285, 1059.
a Ki = 10
l
M17 (IC50 = 3.0
second, weaker SMNPI, that is, Q2-15 (Fig. 1c)
M ( 4.7 M)), provides a 3-Zone hybrid (composed
lM ( 1.4 l
M)28) with the Zone 3 compo-
nent of
a
(IC50 = 11.3
l
l
of a regioisomer mixture (Fig. 1d)) that is 16.7-fold more potent
than NSC 240898.
Future studies will focus on: 1) determining which, if either of
the regioisomers (i.e., 1 or 2) is more potent, 2) defining a compre-
hensive SAR for optimizing Zone 3 occupancy, and 3) determining
the optimal tether for linking the Zone 2 and 3 components.
In conclusion, we have used a 3-Zone pharmacophore for BoNT/
A LC inhibition to guide the design of the first non-Zn(II)-coordi-
14. Paddle, B. M. J. Appl. Toxicol. 2003, 23, 139.
15. Eleopra, R.; Tugnoli, V.; Rossetto, O.; De Grandis, D.; Montecucco, C. Neurosci.
Lett. 1998, 256, 135.
16. Boldt, G. E.; Kennedy, J. P.; Janda, K. D. Org. Lett. 2006, 8, 1729.
17. Burnett, J. C.; Ruthel, G.; Stegmann, C. M.; Panchal, R. G.; Nguyen, T. L.;
Hermone, A. R.; Stafford, R. G.; Lane, D. J.; Kenny, T. A.; McGrath, C. F.; Wipf, P.;
Stahl, A. M.; Schmidt, J. J.; Gussio, R.; Brunger, A. T.; Bavari, S. J. Biol. Chem. 2007,
282, 5004.
18. Burnett, J. C.; Opsenica, D.; Sriraghavan, K.; Panchal, R. G.; Ruthel, G.; Hermone,
A. R.; Nguyen, T. L.; Kenny, T. A.; Lane, D. J.; McGrath, C. F.; Schmidt, J. J.;
Vennerstrom, J. L.; Gussio, R.; Solaja, B. A.; Bavari, S. J. Med. Chem. 2007, 50,
2127.
19. Burnett, J. C.; Schmidt, J. J.; Stafford, R. G.; Panchal, R. G.; Nguyen, T. L.;
Hermone, A. R.; Vennerstrom, J. L.; McGrath, C. F.; Lane, D. J.; Sausville, E. A.;
Zaharevitz, D. W.; Gussio, R.; Bavari, S. Biochem. Biophys. Res. Commun. 2003,
310, 84.
nating SMNPI of the BoNT/A LC that is active in the sub-lM range.
This entailed coupling a lead for therapeutic development, NSC
240898, with a 4,7-ACQ component via a propyl linker. The valida-
tion of the 3-Zone pharmacophore, supported by the increased po-
tency of the 4:1 mixture of regioisomers 1 and 2, represents a new
paradigm for guiding the synthetic optimization of non-Zn(II)-
coordinating SMNPIs of the BoNT/A LC.
Acknowledgments
20. Capkova, K.; Yoneda, Y.; Dickerson, T. J.; Janda, K. D. Bioorg. Med. Chem. Lett.
2007, 17, 6463.
21. Eubanks, L. M.; Hixon, M. S.; Jin, W.; Hong, S.; Clancy, C. M.; Tepp, W. H.;
Baldwin, M. R.; Malizio, C. J.; Goodnough, M. C.; Barbieri, J. T.; Johnson, E. A.;
Boger, D. L.; Dickerson, T. J.; Janda, K. D. Proc. Natl. Acad. Sci. U S A 2007, 104,
2602.
22. Johnson, S. L.; Chen, L. H.; Harbach, R.; Sabet, M.; Savinov, A.; Cotton, N. J. H.;
Strongin, A.; Guiney, D.; Pellecchia, M. Chem. Biol. Drug Des. 2008, 71, 131.
23. Merino, I.; Thompson, J. D.; Millard, C. B.; Schmidt, J. J.; Pang, Y. P. Bioorg. Med.
Chem. 2006, 14, 3583.
24. Park, J. G.; Sill, P. C.; Makiyi, E. F.; Garcia-Sosa, A. T.; Millard, C. B.; Schmidt, J. J.;
Pang, Y. P. Bioorg. Med. Chem. 2006, 14, 395.
25. Solaja, B. A.; Opsenica, D.; Smith, K. S.; Milhous, W. K.; Terzic, N.; Opsenica, I.;
Burnett, J. C.; Nuss, J.; Gussio, R.; Bavari, S. J. Med. Chem. 2008, 51, 4388.
26. Tang, J.; Park, J. G.; Millard, C. B.; Schmidt, J. J.; Pang, Y. P. PLoS ONE 2007, 2,
e761.
27. Hermone, A. R.; Burnett, J. C.; Nuss, J. E.; Tressler, L. E.; Nguyen, T. L.; Solaja, B.
A.; Vennerstrom, J. L.; Schmidt, J. J.; Wipf, P.; Bavari, S.; Gussio, R. Chem. Med.
Chem. 2008, 3, 1905.
This research was funded by Defense Threat Reduction Agency
project 3.10024_06_RD_B and Agreement Y3CM 100505 (MRMC
and NCI, National Institutes of Health). Additionally, this project
has been funded in whole or in part with federal funds from the
National Cancer Institute, National Institutes of Health, under con-
tract N01-CO-12400. The content of this publication does not nec-
essarily reflect that views or policies of the Department of Health
and Human Services or the U.S. Army, nor does mention of trade
names, commercial products, or organizations imply endorsement
by the U.S. Army or the U.S. Government. This research was sup-
ported [in part] by the Developmental Therapeutics Program in
the Division of Cancer Treatment and Diagnosis of the National
Cancer Institute.
28. Wang, C. W.; Widom, J.; Petronijevic, F.; Burnett, J. C.; Nuss, J. E.; Bavari, S.;
Gussio, R.; Wipf, P. Heterocylces 2009, 79, 487.
Supplementary data
29. Lazo, J. S.; Nunes, R.; Skoko, J. J.; de Oliveira, P. E.; Vogt, A.; Wipf, P. Bioorg. Med.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.01.111.
Chem. 2006, 14, 5643.
30. Schmidt, J. J.; Bostian, K. A. J. Protein Chem. 1997, 16, 19.
31. Schmidt, J. J.; Bostian, K. A. J. Protein Chem. 1995, 14, 703.
32. Schmidt, J. J.; Stafford, R. G. Appl. Environ. Microbiol. 2003, 69, 297.
33. Schmidt, J. J.; Stafford, R. G. FEBS Lett. 2002, 532, 423.
34. Schmidt, J. J.; Stafford, R. G.; Bostian, K. A. FEBS Lett. 1998, 435, 61.
References and notes
1. Brunger, A. T.; Jin, R.; Breidenbach, M. A. Cell Mol. Life Sci. 2008, 65, 2296.