Examination of α-exosite inhibitors against Botulinum neurotoxin A protease through structure-activity relationship studies of chicoric acid

Article abstract of DOI:10.1016/j.bmcl.2017.10.021

Botulinum neurotoxins (BoNT) are among the most toxic known substances and currently there are no effective treatments for intraneuronal BoNT intoxication. Chicoric acid (ChA) was previously reported as a BoNT/A inhibitor that binds to the enzyme's α-exosite. Herein, we report the synthesis and structure-activity relationships (SARs) of a series of ChA derivatives, which revealed essential binding interactions between ChA and BoNT/A. Moreover, several ChA-based inhibitors with improved potency against the BoNT/A were discovered.

Full text of DOI:10.1016/j.bmcl.2017.10.021

Bioorganic & Medicinal Chemistry Letters 27 (2017) 4956–4959
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Examination of
a-exosite inhibitors against Botulinum neurotoxin A
protease through structure-activity relationship studies of chicoric acid
Song Xue a,b, Hajime Seki a,b, Marek Remes a,b,d, Peter Šilhár a,b, Kim Janda a,b,c,
⇑
a Department of Chemistry and The Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, United States
b Department of Immunology and The Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, United States
c Worm Institute for Research Medicine (WIRM), The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, 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 (BoNT) are among the most toxic known substances and currently there are no
Received 28 July 2017
Revised 9 October 2017
Accepted 10 October 2017
Available online 12 October 2017
effective treatments for intraneuronal BoNT intoxication. Chicoric acid (ChA) was previously reported
as a BoNT/A inhibitor that binds to the enzyme’s
a-exosite. Herein, we report the synthesis and struc-
ture-activity relationships (SARs) of a series of ChA derivatives, which revealed essential binding interac-
tions between ChA and BoNT/A. Moreover, several ChA-based inhibitors with improved potency against
the BoNT/A were discovered.
Keywords:
Botulinum neurotoxins
Chicoric acid
Ó 2017 Elsevier Ltd. All rights reserved.
Structure-activity relationship
Botulinum neurotoxins (BoNT), which are produced by the
Gram-positive bacterium Clostridium botulinum, are among the
most lethal known human poisons. The most potent stereotype,
BoNT/A, exhibits an intravenous lethal dose of 1–2 ng/kg in
humans.1 Accordingly, BoNT/A is classified as a bioterror threat
due to its tremendous toxicity and ease of production. Despite
these concerns and the lack of effective countermeasures in the
instance of overdose, BoNT/A is also widely used as both a cosmetic
and therapeutic.2
The lethality of BoNT/A results from intoxication of peripheral
neurons, which is mediated through its heavy chain (HC) and light
chain (LC).3 The HC ensures the toxin passes the digestive system,
enters circulation, and reaches peripheral neuromuscular junc-
tions, where it is recognized by receptors that mediate endocytosis
of the holotoxin.4 Once translocated into the cytosol, the released
LC, a Zn2+ dependant endopeptidase, specifically binds and cleaves
synaptosomal-associated protein of 25 kDa (SNAP-25). Cleavage of
SNAP-25 irreversibly impairs the membrane fusion machinery
required for the exocytosis of acetylcholine at neuromuscular junc-
tions. Acetylcholine is essential for neuromuscular transmission;
thus, BoNT/A intoxication of nerve endings results in flaccid paral-
ysis and potentially asphyxiation, when paralysis occurs in the res-
piratory system.4
Unfortunately, no effective cure has been developed for BoNT/A
intoxication. Available treatments are simply supportive, and
patients suffer from long hospital stays requiring mechanical respi-
ration.5 While an antibody-based antitoxin can be administered
immediately following BoNT/A exposure, the antitoxin is not effec-
tive once the toxin has been internalized into neuronal cells (<12 h
post exposure).6 Therefore, strategies to antagonize BoNT/A intra-
neuronally are urgently needed. Small molecule inhibitors offer
the sole opportunity for a post-intoxication, intraneuronal therapy.
Earlier, we reported the natural product chicoric acid (ChA) as a
non-competitive, partial inhibitor of BoNT/A LC with an IC50 = 5.9
mM (Fig. 1A).7 While the majority of previously reported BoNT/A
inhibitors bind the enzyme’s active site, ChA binds to the
a-exosite,
an allosteric region.8 Our study revealed that the
a-exosite plays an
integral role in BoNT/A catalytic activity and stability,9 and is
therefore targetable for inhibitor development. In a subsequent
study, an i-Pr ester analog of ChA (ChA i-Pr ester) demonstrated
a lower IC50 value of 2.7 mM with complete inhibition under satu-
rating conditions (Fig. 1B).10 Kinetic analysis of ChA and ChA i-Pr
ester used in combination revealed that the two compounds were
mutually exclusive, as parallel curves were observed in the Yone-
tani-Theorell plot (Fig. 1C).11 In other words, ChA and ChA i-Pr
ester were found to bind at the same site of BoNT/A LC. Impor-
tantly, this study also demonstrated that synthetic modifications
to the ChA scaffold were tolerated by the enzyme.
⇑
Corresponding author at: Department of Chemistry and The Skaggs Institute for
Chemical Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, La
Jolla, CA 92037, United States.
E-mail address: kdjanda@scripps.edu (K. Janda).
Current addresses: (a) Central European Institute of Technology, Brno University
d
of Technology, Technicka 11 3058/10, CZ-616 00 Brno, Czech Republic. (b) Depart-
ment of Chemistry and Biochemistry, Mendel University in Brno, Zemedelska 1, CZ- 9
613 00 Brno, Czech Republic.
https://doi.org/10.1016/j.bmcl.2017.10.021
0960-894X/Ó 2017 Elsevier Ltd. All rights reserved.

S. Xue et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 4956–4959
4957
Fig. 1. Structure of Chicoric Acid (ChA) and its i-Pr ester analog (ChA i-Pr ester) (A). Inhibition curves of ChA and ChA i-Pr ester (B). Yonetani-Theorell plot of ChA and ChA i-Pr
ester.
Though the kinetic parameters and binding site for ChA inhibi-
tion have been revealed, a BoNT/A LC – ChA co-crystal structure
remains elusive and thus, the specific binding interactions between
the enzyme and small molecule remain unknown. To better under-
stand ChA’s mechanism of binding, as well as to develop more
potent inhibitors, we synthesized a series of ChA derivatives for
structure-activity relationship (SAR) studies.
The chemical structure of ChA is defined by two caffeic acid
motifs linked by tartaric acid. From our results with ChA i-Pr ester,
we hypothesized that hydrophobic ester modifications of the
Scheme 1. Synthesis of ChA derivatives with various tartaric ester linkers. Reagents and conditions: (a) Ac2O, pyr.; (b) SOCl2, benzene; (c) pyr., DMAP, DCM; (d) 2 M HCl,
acetone; (e) pyr., DMAP, DCM; (f) 2 M HCl, acetone.

4958
S. Xue et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 4956–4959
Table 1
SAR of tartaric acid linker substitutions.a
Compound
R
IC50 (mM)
Maximum Inhibition (%)
100
Compound
R
IC50 (mM)
Maximum Inhibition (%)
100
1
3.2 0.6
9
3.2 0.6
O
O
2
3
4
2.4 0.5
1.7 0.2
0.13 0.03
100
100
75
10
11
12
0.56 0.10
0.39 0.05
0.14 0.02
100
100
100
MeO2C
O
O
5
0.49 0.08
96
13
0.13 0.01
100
6
7
1.1 0.2
100
100
14
15
0.05 0.01
0.30 0.04
100
100
0.41 0.21
O
O
O
O
8
0.45 0.11
100
16
3.3 0.9
100
a
See the Supporting Information for detailed syntheses.
tartaric acid linker may improve ChA’s inhibitory potency. Thus,
we first explored a series of ChA derivatives with various linkers,
We first altered the caffeic acid substructure by truncating the
motif to a protocatechuic acid ester, resulting in compounds 17
and 18. Analog 17 showed moderate inhibition of the BoNT/A LC
with an IC50 > 20 mM, suggesting that the phenols maintain some
LC binding ability, but the truncated protocatechuic acid ester
may be too short for the phenols to fully engage the interacting
LC residues. Analog 17 was further modified to include a hydrophi-
lic amine group, resulting in 18. Consistent with our previous
results, this compound exhibited no inhibition. We hypothesize
that this hydrophilic modification disrupted the aforementioned
hydrophobic effect, for which the hydrophobic tartaric esters were
optimized. To explore how the caffeic acid phenols interact with
the LC, analogs 19–24 were synthesized (Fig. 2, see the Supporting
Information for detailed syntheses). First, the parent 3,4-dihydrox-
yphenyl group was reduced to single para- or meta- hydroxy sub-
stitution to afford compounds 19 and 20. Both compounds showed
weak inhibition with IC50 values = 9.5 and >20 mM, respectively.
These results indicate that all four phenols are necessary for
BoNT/A LC inhibition, of which the two 4-position phenols are
the most important. As a comparison, the four phenols of 5 were
acetylated to afford analog 21, which was inactive against BoNT/
A LC as expected. Additionally, compounds 22–24, in which the
phenols were replaced with nitro, aldehyde or alcohol substituents,
also exhibited no BoNT/A LC inhibition. As many of these deriva-
tives exhibit hydrogen bond accepting properties (OAc, NO2 and
CHO, in particular), we posit that the four phenols are essential
hydrogen bond donors in ChA–BoNT/A LC binding interaction.
including cycloalkyl-, aryl-, or alkyl-diesters and
a diamide
(Scheme 1). The synthesized compounds were examined for inhibi-
tion of BoNT/A LC activity by LC–MS assay with the 66-mer SNAP-
25 substrate, as described in our previous reports.12 The structures
and IC50 values are shown in Table 1.
The results revealed that modification of the tartaric acid linker
to hydrophobic esters was tolerated. Importantly, all synthesized
derivatives maintained inhibition of BoNT/A LC. Not only did most
of the analogs have lower IC50 values, they also showed complete
inhibition of the enzyme under saturating conditions, in compar-
ison to the partial inhibition of BoNT/A LC by the parent, ChA. These
findings indicate that hydrophobic linkers containing the dihydrox-
yphenylacrylate moiety are tantamount for effective BoNT/A LC
inhibition. To explain this trend, we hypothesize that a hydrophobic
region exists in the
a-exosite, which translates to better binding
affinities for the more hydrophobic ester-based inhibitors.
Given that the tartaric acid linker was readily modified without
compromising BoNT/A LC inhibition, we next hypothesized that
the caffeic acid motif, and particularly the four phenol groups, were
critical for ChA binding to BoNT/A LC. To investigate the caffeic acid
substructure, we prioritized analog 5, with an IC50 = 0.49 0.08
lM, as our lead compound for the following studies. A total of 8
analogs with caffeic acid modifications were synthesized to
elucidate the SAR between the caffeic acid motif and BoNT/A LC
inhibitory potency (Fig. 2).

S. Xue et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 4956–4959
4959
Fig. 2. SAR of caffeic acid modifications.
In sum, by synthesizing a series of ChA derivatives and analyz-
References
ing their BoNT/A LC inhibitory activity, we revealed that the two
caffeic acid motifs are critical for ChA binding to the LC. The inter-
action between the inhibitors and the LC may occur through
hydrogen bond formation involving the phenol groups of ChA as
donors. Moreover, substitution of the tartaric acid linker was toler-
ated and hydrophobic ester modifications were favorable for inhi-
bition. Overall, analogs 1–16 demonstrated improved BoNT/A LC
inhibitory potency in both IC50 value and maximum inhibition effi-
ciency, in comparison to the ChA parent. As a result, analogs 1–16
may provide leads for further crystallographic as well as therapeu-
tic studies.
1. Arnon SS. Botulinum toxin as a biological weapon: medical and public health
management (vol 285, pg 1059, 2001). JAMA. 2001;285:2081–2085.
2. Hackett R, Kam PC. Botulinum toxin: pharmacology and clinical developments:
a literature review. Med Chem. 2007;3:333–345.
3. Rossetto O, Schiavo G, Montecucco C, et al. Snare motif and neurotoxins. Nature.
1994;372:415–416.
4. Montecucco C, Molgo J. Botulinal neurotoxins: revival of an old killer. Curr Opin
Pharmacol. 2005;5:274–279.
5. Sobel J. Botulism. Clin Infect Dis. 2005;41:1167–1173.
6. Sobel J. Diagnosis and treatment of botulism: a century later, clinical suspicion
remains the cornerstone. Clin Infect Dis. 2009;48:1674–1675.
7. Silhar P, Capkova K, Salzameda NT, Barbieri JT, Hixon MS, Janda KD. Botulinum
neurotoxin A protease: discovery of natural product exosite inhibitors. J Am
Chem Soc. 2010;132:2868–2869.
8. Breidenbach MA, Brunger AT. Substrate recognition strategy for botulinum
neurotoxin serotype A. Nature. 2004;432:925–929.
Acknowledgments
9. Xue S, Javor S, Hixon MS, Janda KD. Probing BoNT/A protease exosites:
implications for inhibitor design and light chain longevity. Biochemistry.
2014;53:6820–6824.
We acknowledge the National Institutes of Health (AI117878
and AI119564) for funding. This is manuscript #29551 from The
Scripps Research Institute.
10. Seki H, Xue S, Hixon MS, et al. Toward the discovery of dual inhibitors for
botulinum neurotoxin A: concomitant targeting of endocytosis and light chain
protease activity. Chem Commun (Camb). 2015;51:6226–6229.
11. Yonetani T. The Yonetani-Theorell graphical method for examining overlapping
subsites of enzyme active centers. Methods Enzymol. 1982;87:500–509.
A. Supplementary data
ˇ
12. Capková K, Hixon MS, McAllister LA, Janda KD. Toward the discovery of potent
inhibitors of botulinum neurotoxin A: development of a robust LC MS based
assay operational from low to subnanomolar enzyme concentrations. Chem
Commun (Camb). 2008;3525–3527.
Supplementary data (synthesis of the compounds and charac-
terizations) associated with this article can be found, in the online
version, at https://doi.org/10.1016/j.bmcl.2017.10.021.