Chirality holds the key for potent inhibition of the botulinum neurotoxin serotype a protease

Article abstract of DOI:10.1021/ol902820z

(Chemical Equetion Presentation) Botulinum neurotoxin serotype A (BoNT/A) is the most toxic protein known to man and also a bioterrorism agent. As defined by our previous research targeting the etiological agent responsible for BoNT/A intoxication, a protease, we now report on the asymmetric synthesis of four new BoNT/A inhibitors; the most potent of this series is roughly 2-fold more active than the best small molecule inhibitor currently known.

Full text of DOI:10.1021/ol902820z

ORGANIC
LETTERS
2010
Vol. 12, No. 4
756-759
Chirality Holds the Key for Potent
Inhibition of the Botulinum Neurotoxin
Serotype A Protease
†
†
‡
§
ˇ
G. Neil Stowe, Peter Silha´r, Mark S. Hixon, Nicholas R. Silvaggi, Karen
N. Allen,§,* Scott T. Moe,| Alan R. Jacobson,|,* Joseph T. Barbieri,⊥ and Kim
D. Janda†,*
Departments of Chemistry and Immunology, The Skaggs Institute for Chemical Biology
and Worm Institute of Research and Medicine (WIRM), The Scripps Research Institute,
10550 North Torrey Pines Road, La Jolla, California 92037, Takeda San Diego, Inc.,
10410 Science Center DriVe, San Diego, California 92121, Department of Chemistry,
Boston UniVersity, Boston, Massachusetts 02215, Absolute Science, Inc., P.O. Box
382366, Cambridge, Massachusetts 02238, and Departments of Microbiology and
Molecular Genetics, Medical College of Wisconsin, 8701 Watertown Plank Road,
Milwaukee, Wisconsin 53226
drkallen@bu.edu; arj@absolutescience.com; kdjanda@scripps.edu
Received December 8, 2009
ABSTRACT
Botulinum neurotoxin serotype A (BoNT/A) is the most toxic protein known to man and also a bioterrorism agent. As defined by our previous
research targeting the etiological agent responsible for BoNT/A intoxication, a protease, we now report on the asymmetric synthesis of four
new BoNT/A inhibitors; the most potent of this series is roughly 2-fold more active than the best small molecule inhibitor currently known.
The seven serotypes (A-G) of botulinum neurotoxins
(BoNTs) are proteins produced by bacteria of the genus
Clostridium.1 BoNTs are synthesized as ∼150-kDa proteins
consisting of a 100-kDa heavy chain (HC) and a 50-kDa
light chain (LC) linked by disulfide bonds.2 The HC is
responsible for translocation and binding, while the LC is a
zinc-dependent endopeptidase that cleaves soluble N-ethyl-
maleimide-sensitive fusion proteins (SNARE) located at
nerve endings.3 Cleavage of SNARE proteins by the BoNT
LC leads to flaccid paralysis and subsequent death due to
† The Scripps Research Institute.
‡ Takeda San Diego.
(1) Roxas-Duncan, V.; Enyedy, I.; Montgomery, V. A.; Eccard, V. S.;
Carrington, M. A.; Lai, H.; Gul, N.; Yang, D. C. H.; Smith, L. A.
Antimicrob. Agents Chemother. 2009, 53, 3478–3486.
§ Boston University.
| Absolute Science.
(2) Silvaggi, N. R.; Boldt, G. E.; Hixon, M. S.; Kennedy, J. P.; Tzipori,
S.; Janda, K. D.; Allen, K. N. Chem. Biol. 2007, 14, 533–542.
⊥ Medical College of Wisconsin.
10.1021/ol902820z  2010 American Chemical Society
Published on Web 01/21/2010

inhibition of the release of acetylcholine from synaptic
terminals.4
functionality, a monoanionic bidentate chelator of zinc that
has been previously used for the inhibition of metallopro-
teinases,11 (2) a para-halogenated phenyl ring was needed,12
and (3) an ortho,para-dihalogenated ring was superior to
para-substition alone10 (Figure 1). Indeed, X-ray crystal-
lographic analysis of the bound inhibitors revealed hydroxyl
coordination of the hydroxamate to the Zn(II) ion in the
active site in addition to phenyl ring interaction with a
hydrophobic portion of the enzyme.2 Addition of the ortho
chlorine substituent created an additional favorable interac-
tion with an active site arginine, making this the most potent
inhibitor reported to date.2
While our initial research has been critical to lending
credence that small nonpeptidic molecules can be tailored
to the BoNT/A protease, it is readily apparent that more
potent inhibitors will be required if a therapeutic is the final
goal. As such, we sought to synthesize a series of molecules
containing additional chemical diversity for potential interac-
tion(s) with the enzyme active site. We envisioned the
preparation of hydroxy ethyl hydroxamates (R)- and (S)-4
and -5, which would not only extend within polar space of
the enzyme’s combining site but also explore the importance
of molecular chirality and its impact on enzyme inhibition.
Current treatment for BoNT intoxication employs an
antitoxin to sequester the toxin followed by removal of the
complex from the body. However, this treatment option is
dependent upon prompt diagnosis of intoxication and must
be administered rapidly before the toxin can enter the cell.5
The antitoxin also suffers from the risk of allergic reaction,6
long-term effects of unknown origin,7 and insufficient
available quantities for treatment of a large-scale attack.8
BoNT serotype A (BoNT/A) is considered to be the most
deadly BoNT serotype based on the following: First, and
most importantly, BoNT/A has an extremely long duration
of action, i.e., weeks to months. Second, there is no antidote
for BoNT/A intoxication, with severe intoxication cases
requiring mechanical ventilation due to respiratory paralysis.9
Third, BoNT/A, together with serotype B, is responsible for
∼1000 cases of human BoNT poisoning per year.1 Finally,
BoNT/A has a high potential for bioterrorism use due to its
extreme toxicity and ease of dissemination through the food
or water supply. Thus, the inhibition of BoNT/A LC is an
attractive target for nonpeptidic small molecules that can
target the enzyme within an intoxicated cell.
Scheme 1
.
(A) Structures of Hydroxamates (R)- and (S)-4 and
-5; (B) Retrosynthetic Analysis
Figure 1
inhibitors.
. Structure and IC50 values of previously identified BoNT/A
Previous research from one of our laboratories screened a
diverse library of small molecules and discovered several
inhibitors of BoNT/A, with trans-cinnamic hydroxamate 1
being our initial lead (Figure 1). Hydroxamate 2 stemmed
from structure-activity studies, and is one of the most potent
inhibitors known, having a Ki value of 300 ( 12 nM.10 From
these studies several lessons were learned, including the
following: (1) all structures required the hydroxamate
We viewed these structures as rational choices based on
active-site molecular modeling wherein a key goal was to
displace a nacent water molecule within the enzyme active
site, which on the basis of previous protease inhibitors
targeting active site water molecules could improve Ki up
to 100-fold.13 To access such structures required an asym-
metric synthetic strategy that would construct both enantio-
mers of the final hydroxamate products from a single, readily
(3) 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–
163.
(4) Hambleton, P. J. Neurol. 1992, 239, 16–20.
(5) Capkova, K.; Salzameda, N. T.; Janda, K. D. Toxicon 2009, 54, 575–
582.
(6) Black, R. E.; Gunn, R. A. Am. J. Med 1980, 69, 567–570.
(7) Villar, R. G.; Elliott, S. P.; Davenport, K. M. Infect. Dis. Clin. North
Am. 2006, 20, 313–327.
(8) Arnon, S. S.; Schechter, R.; Inglesby, T. H.; Henderson, D. A.;
Bartlett, J. G.; Ascher, M. S.; Eitzen, E.; Fine, A. D.; Hauer, J. JAMA,
J. Am. Med. Assoc. 2001, 285, 1050–1079.
(11) (a) Skiles, J. W.; Gonnella, N. C.; Jeng, A. Y. Curr. Med. Chem.
2004, 22, 2911–2977. (b) Grant, S.; Easley, C.; Kirkpatrick, P. Nat. ReV.
Drug DiscoV. 2007, 6, 21–22.
(9) Hicks, R. P.; Hartell, M. G.; Nichols, D. A.; Bhattacharjee, A. K.;
Von Hamont, J. E.; Skillman, D. R. Curr. Med. Chem. 2005, 12, 667–690.
(10) Boldt, G. E.; Kennedy, J. P.; Janda, K. D. Org. Lett. 2006, 8, 1729–
1732.
(12) Capkova, K.; Yoneda, Y.; Dickerson, T. J.; Janda, K. D. Bioorg.
Med. Chem. Lett. 2007, 17, 6463–6466.
Org. Lett., Vol. 12, No. 4, 2010
757

available, starting material (Scheme 1). Herein, we report
the successful asymmetric synthesis and inhibitory evaluation
of hydroxamates (R)- and (S)-4 and -5 with BoNT/A
protease.
a similar chemoenzymatic approach could be successful due
to the similar nature of the two structures. We planned to
synthesize diesters 13 and 14 from readily available starting
materials as shown in Scheme 2.
Thus, our synthesis of diesters 13 and 14 started with the
synthesis of alkenes 9 and 10 using the aqueous Wittig
reaction of methyl bromoacetate with the appropriate alde-
hyde16 followed by Michael-type addition of dimethyl
malonate to yield triesters 11 and 12 in good yield.15
Selective triester decarboxylation17 yielded diesters 13 and
14, the substrates for enzymatic hydrolysis. Diester 13 was
hydrolyzed with chymotrypsin to give (R) acid/ester 15 with
high enantioselectivity and chemical yield. Hydrolysis of
o-,p-dichloro diester 14 using chymotrypsin also proceeded
with high enantioselectivity (>99%) and respectable yield
(45%) to give (R) acid/ester 16. The enantioselectivity and
absolute stereochemistry of these hydrolytic enzymatic
resolutions was determined by acid derivitization using (R)-
and (S)-1-(1-naphthyl)ethylamine followed by comparison
of the difference in chemical shift of the diastereomeric
methoxy ester protons 19 and 20 and 21 and 22 according
to the procedure of Hoye et al. (Scheme 2).18
Scheme 2
.
Synthesis of Acids 15 and 16 and Derivitization to
Determine ee and Stereochemistry
Scheme 3
.
Synthesis of Hydroxamate Enantiomers (R)- and
(S)-4 and -5
Our envisioned hydroxamate synthesis involved ring-
opening of the corresponding lactones (R)- and (S)-23 and
-24, which could be independently synthesized from acid/
esters 15 and 16 via chemoselective lactonization.14 Acid/
ester 15 was previously synthesized in high enantiomeric
excess (>98%) and yield (85%) by chymotrypsin-mediated
enantioselective hydrolysis of diester 13,15 and we reasoned
that the previously unreported hydrolysis of diester 14 using
With acids 15 and 16 in hand, we sought to prepare both
enantiomers of lactones 23 and 24 via chemoselective
lactonization.14 Thus, ester reduction with LiBH4 followed
by cyclization under acidic conditions yielded (S) lactones
23 and 24. Conversely, (R) lactones 23 and 24 were prepared
by carboxylic acid reduction with BH3·DMS followed by
(13) (a) Silvaggi, N. R.; Wilson, D.; Tzipori, S.; Allen, K. N. Biochem-
istry 2008, 47, 5736–5745. (b) Fu, Z.; Chen, S.; Baldwin, M. R.; Boldt,
G. E.; Crawford, A.; Janda, K. D.; Barbieri, J. T.; Kim, J. J. Biochemistry
2006, 45, 8903–8911. (c) Rich, D. H.; Sun, E. T. O. J. Med. Chem. 1980,
23, 27–33.
(14) (a) Huang, F. C.; Lee, L. F. H.; Mittal, R. S. D.; Ravikumar, P. R.;
Chan, J. A.; Sih, C. J.; Caspi, E.; Eck, C. R. J. Am. Chem. Soc. 1975, 97,
4144–4149. (b) Yu, M. S.; Lantos, I.; Peng, Z.; Yu, J.; Cacchio, T.
Tetrahedron Lett. 2000, 41, 5647–5651.
(16) El-Batta, A.; Jiang, C.; Zhao, W.; Anness, R.; Cooksy, A. L.;
Bergdahl, M. J. Org. Chem. 2007, 72, 5244–5259.
(17) Krapcho, A. P. Synthesis 1982, 805–808.
(15) (a) Chenevert, R.; Desjardins, M. Can. J. Chem. 1994, 72, 2312–
2317. (b) Chen, L.; Zaks, A.; Chackalamannil, S.; Dugar, S. J. Org. Chem.
1996, 61, 8342–8343.
(18) (a) Hoye, T. R.; Koltun, D. J. Am. Chem. Soc. 1998, 120, 4638–
4643. (b) Hoye, T. R.; Hamad, A. S.; Koltun, D. O.; Tennakoon, M. A.
Tetrahedon Lett. 2000, 41, 2289–2293.
758
Org. Lett., Vol. 12, No. 4, 2010

lactonization under acidic conditions. The lactones were then
smoothly converted to hydroxamates (R)- and (S)-4 and -5
employing hydroxylamine under basic conditions12 (Scheme
3).
BoNT/A inhibitor known. We did not determine Ki values
for hydroxamates (S)-4 and -5 since they were inferior
inhibitors as shown by IC50 values.
In summary, a concise asymmetric synthesis of four new
inhibitors of the highly toxic BoNT/A LC has been devised.
As discovered previously, the o-,p-dichloro hydroxamates
are more potent inhibitors than the simple p-chloro hydrox-
amates. We have also clearly shown that control of stereo-
chemistry is crucial for optimum inhibition. Finally, we note
that a crystal structure of hydroxamate (R)-4 within the
BoNT/A active site has been solved,20 supporting our
hypothesis that the hydroxyethyl moiety of (R)-4 and -5
replaces an active-site water molecule. Additional research
using a combination of crystallography, synthesis, and kinetic
analysis is envisioned to discover more potent inhibitors of
this unique metalloprotease and will be reported in due
course.
Table 1. IC50 and Ki Values for Hydroxamates (R)- and (S)-4
and -5
compd
IC50 (µM)
KI(µM)
(S)-4
(R)-4
(S)-5
(R)-5
36
8
21
1
n.d.
1.7 ( 0.3
n.d.
0.16 ( 0.02
Assays were conducted at 22.5°C, pH 7.4 in 40 mM HEPES buffer at
10 µM inhibitor concentration, 0.075 nM enzyme (BoNT/A) concentration
using SNAP-25 (66-mer) substrate. An entry of n.d. denotes value was not
determined.
Acknowledgment. This project has been funded with
federal funds from the National Institute of Allergy and
Infectious Diseases, National Institutes of Health, and the
Department of Health and Human Services under contract
nos. N01-AI30050 and AI080671. J.T.B. acknowledges
membership within and support from the Region V “Great
Lakes” RCE (NIH award 1-U54-AI-057153).
All hydroxamates were screened for activity against
BoNT/A using our LC/MS-based assay that employs a
truncated version of SNAP-25, the SNARE protein cleaved
by the BoNT/A LC.19 Our results clearly show the impor-
tance of chirality; in both cases, the (R) hydroxamate IC50
was less than the (S) hydroxamate. The Ki of (R) p-chloro
and o-,p-dichloro hydroxamates (R)-4 and (R)-5 was deter-
mined to be 1.7 ( 0.3 and 0.16 ( 0.02 µM, respectively
(Table 1), with inhibition being competitive in nature (data
not shown). Importantly, hydroxamate (R)-5 is roughly 2-fold
more effective than the best small molecule nonpeptidic
Supporting Information Available: Experimental syn-
thetic procedures and NMR and HRMS data for all previ-
ously unreported compounds. This information is available
free of charge via the Internet at http://pubs.acs.org.
OL902820Z
(19) Capkova, K.; Hixon, M. S.; McAllister, L. A.; Janda, K. D. Chem.
Commun. 2008, 3525–3527.
(20) Silvaggi, N. R.; Allen, K. N. The crystal structure and additional
kinetic findings will be published elsewhere.
Org. Lett., Vol. 12, No. 4, 2010
759