Allicin and derivates are cysteine protease inhibitors with antiparasitic activity

Article abstract of DOI:10.1016/j.bmcl.2010.07.062

Allicin and derivatives thereof inhibit the CAC1 cysteine proteases falcipain 2, rhodesain, cathepsin B and L in the low micromolar range. The structure-activity relationship revealed that only derivatives with primary carbon atom in vicinity to the thiosulfinate sulfur atom attacked by the active-site Cys residue are active against the target enzymes. Some compounds also show potent antiparasitic activity against Plasmodium falciparum and Trypanosoma brucei brucei.

Full text of DOI:10.1016/j.bmcl.2010.07.062

Bioorganic & Medicinal Chemistry Letters 20 (2010) 5541–5543
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Allicin and derivates are cysteine protease inhibitors with antiparasitic activity
Thilo Waag a, Christoph Gelhaus b, Jennifer Rath c, August Stich c, Matthias Leippe b, Tanja Schirmeister a,
*
a Institute of Pharmacy and Food Chemistry, University of Würzburg, Am Hubland, 97074 Würzburg, Germany
b Department of Zoophysiology, Zoological Institute, University of Kiel, Olshausenstr. 40, 24098 Kiel, Germany
c Department of Tropical Medicine, Medical Mission Institute, Salvatorstr. 7, 97074 Würzburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Allicin and derivatives thereof inhibit the CAC1 cysteine proteases falcipain 2, rhodesain, cathepsin B and
L in the low micromolar range. The structure–activity relationship revealed that only derivatives with
primary carbon atom in vicinity to the thiosulfinate sulfur atom attacked by the active-site Cys residue
are active against the target enzymes. Some compounds also show potent antiparasitic activity against
Plasmodium falciparum and Trypanosoma brucei brucei.
Received 25 June 2010
Revised 14 July 2010
Accepted 15 July 2010
Available online 6 August 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Allicin
Thiosulfinate
Cysteine protease
Inhibitor
Parasite
Allicin (S-allyl-2-propenyl thiosulfinate, 1, Scheme 1) is the
main biologically active component of garlic extracts. It is pro-
duced during the crushing of garlic by the interaction of alliin,
(Scheme 1), a non-protein amino acid, with the pyridoxal phos-
phate containing enzyme, alliinase.1 Allicin was found to inhibit
the growth of a wide range of bacteria, it also shows antifungal,
antiviral, and antiparasitic activity.2–4
Even though the antibacterial properties of allicin were already
detected in 1940s, the mechanism of action was not elucidated in
detail until 1998. NMR studies indicated that the particular struc-
ture of allicin makes it a good candidate for interaction with SH-
groups of proteins and other biologically active molecules. In order
to study the reaction of allicin with thiol containing proteins, the
O
O
S
1
HS
OH
S
OH
NH2
NH2
Scheme 2. Reaction of allicin with cysteine.5
This was already proven for the cytopathic effects of Entamoeba
histolytica: the inhibition of parasite cathepsin-like proteases con-
tributes to the antiparasitic activity of allicin.3 Thus, allicin proba-
bly also inhibits related proteases from other parasites. In the
present work we present synthesis of allicin and derivates thereof,
the testing of the compounds against cathepsin-like cysteine
proteases using fluorometric enzyme assays and results of micro-
bial growth assays.
The synthesis of the potential inhibitors (as racemates) starts
from symmetrical disulfides which were prepared from the corre-
sponding thiols by oxidation with catalytic amounts of iodine in
DMSO (Scheme 3).6
interaction of allicin with L-cysteine was used as a model. The iso-
lated product was the S-allyl-mercapto cysteine, not its sulfoxide
derivate (Scheme 2).5
It is reasonable to assume, that the broad-spectrum antimicro-
bial effects of allicin are due to inhibitory effects they may have on
various thiol-dependent enzymes, for example, cysteine proteases.
The thiosulfinates were synthesized by oxidation of the corre-
sponding disulfides with an organic peracid. Peracetic, perbenzoic,
perfuroic, perphtalic as well as percamphoric acid can be used.7
O
S
NH2
CO2H
O
S
alliinase
mCPBA
CH2Cl2
I2 (cat.)
DMSO
S
O
S
1
alliin
S
R
R
2
R SH
R
R
S
S
50 ºC, 1-2 h
40 ºC, 1 h
Scheme 1. Biosynthesis of allicin (1).
Scheme 3. Synthesis of allicin (1, R = allyl) and symmetrical derivatives 2–7
[R = n-propyl (2), benzyl (3), tert-butyl (4), cyclohexyl (5), 3-methylbuty-1-yl
(isopentyl) (6), n-hexyl (7)].
* Corresponding author. Tel.: +49 (0)9313185440.
E-mail address: schirmei@pzlc.uni-wuerzburg.de (T. Schirmeister).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.07.062

5542
T. Waag et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5541–5543
known to play essential roles in the pathogenicity of the proto-
SH
zoa.10 Selected compounds were also tested against the respective
parasites.
SO2Cl2,
CH2Cl2
pyridine,
CH2Cl2
O
S
O
S
The postulated enzyme inhibition mechanism (Schemes 2 and
513) involves a thiol-disulfide exchange reaction5 yielding the cys-
teine residue of the active site of the enzyme blocked as disulfide.
Due to subsequent disulfide exchange by DTT (dithiothreitol)
which is necessarily present in the assay medium the disulfide-
blocked enzyme can be reactivated yielding free enzyme and a
mixed disulfide containing the inhibitor’s thiol fragment and
DTT. This new disulfide then can react as a reversible inhibitor
which is well known for cysteine proteases.15 In such a case,
time-independent inhibition will be observed, that is, the degree
of inhibition does not change over time (linear progress curves,
Fig. 1, bottom). However, in some cases, namely for compounds
2, 3, and 7 time-dependent inhibition was observed, that is, inhibi-
tion potency increases with time. This is an indicator for irrevers-
ible inhibition, that is, the reactivation of the enzyme by DTT
may not be possible in such cases. Examples for both, time-depen-
dent and time-independent inhibition are shown in Figure 1.
Additionally, allyldisulfide (starting material for synthesis of
allicin (1)) was tested against the enzymes. However, no decrease
S
S
S
Cl
-15 ºC, 1h
-20 ºC, 2h
(8)
Scheme 4. Synthesis of mixed thiosulfinate (8).
OMe
OMe
HS
S
or
S
RT, 30 min
pyridine,
S
O
S
or
S
CH2Cl2
H2N
HS
S
NH2
Scheme 5. Model reaction of thiosulfinate (2) with thiols.
meta-Chloro perbenzoic acid (mCPBA) was the reagent of choice in
the described work.
in enzyme activity was observed at 20 lM, showing that the thio-
sulfinate moiety is necessary for inhibition.
For the synthesis of the mixed thiosulfinate (8) dibenzyl disul-
fide was treated with sulfuryl chloride to yield benzyl sulfinyl
chloride (Scheme 4).8 Reaction with tert-butyl mercaptan yielded
the mixed thiosulfinate (8) as single regioisomer.
The inhibitory effects of allicin and its derivates were tested
against the parasite enzymes falcipain 2 from Plasmodium falcipa-
rum, the protozoa causing malaria, and rhodesain from Trypano-
soma brucei rhodesiense, causing African sleeping sickness, and
the mammalian cysteine proteases cathepsin B and L. These en-
zymes belong to the cysteine protease clan A, family C1 and have
similar three-dimensional structures.9 The parasite proteases are
Table 1 summarizes the inhibition constants of the various cys-
teine proteases, and the IC50-values for their antiparasitic activity.
With the exception of compounds 4 (tert-butyl), 5 (cyclohexyl),
and the mixed thiosulfinate 8 (benzyl/tert-butyl) the synthesized
thiosulfinates inhibited the proteases in the low micromolar range.
The most active inhibitors were allicin (1) itself and derivatives 6
and 7 with larger lipophilic groups. These groups could address
the hydrophobic S2 pockets of the enzymes. The inactivity or only
weak activity of compounds 4, 5 and 8 indicates that the inhibition
reaction (Scheme 2 and 513) may be sterically hindered if a second-
340
320
300
280
260
240
220
200
180
160
140
0.4
0.2
k
0
0
2
4
6
8
10
0
20
40
60
I (µM)
80
100
time (min)
380
360
340
320
300
280
260
240
220
200
180
160
140
20
18
16
14
12
10
8
6
4
2
0
2
4
6
8
10
0
20
40
60
I (µM)
80
100
time (min)
Figure 1. Inhibition of cathepsin L by compd 2 (top), left: progress curves at 0, 10, 20, 40, 50, 60, 80, 100 lM; right: kobs versus [I] diagram for determination of second-order
rate constant of inhibition k2nd. Inhibition of cathepsin L by compd (1) (bottom), left: progress curves at 0, 10, 20, 40, 50, 60, 80, 100
lM; right: Dixon plot for determination of
Ki.

T. Waag et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5541–5543
5543
Table 1
Biological activities of allicin and derivatives14
Compds
Cathepsin B
Inhibition
Cathepsin L
Inhibition
Falcipain 2
Inhibition
Rhodesain
Inhibition
P. falciparum
Inhibition
T. b. brucei
Inhibition
IC50
Ki,
l
M
Ki,
lM
Ki,
lM
Ki,
lM
IC50
,
lM
, lM
Allicin (1)
2
8.6 0.29
15.6 3.6
9.3 0.39
11.3 2.4
1.04 0.08
26.4 2.3
5.31 0.83
4.44 0.85
5.21 0.96
78.3 23.3
13.8 0.06
>40
(6664 400 MÀ1 minÀ1
31.7 0.30
)
(10,761 680 MÀ1 minÀ1
4.5 1.2
)
(7150 1950 MÀ1 minÀ1
2.31 0.07
)
b
b
b
3
7.8 2.2
30.3 12.6
n.d.a
(48,556 1120 MÀ1 minÀ1
>100 (ca. 10)a
>100 (ca. 35)a
3.04 0.47
)
(7800 1300 MÀ1 minÀ1
>100 (ca. 10)a
>100 (ca. 30)a
0.36 0.05
)
b
b
b
4
5
6
7
n.i.a
n.i.a
72.5 14.7
34.4 13.3
54.7 22.5
10.9 1.61
>40
>100 (ca. 30)a
18.9 1.03
5.94 1.30
>100 (ca. 20)a
3.4 1.1
n.d.a
n.d.a
10.4 2.45
1.80 0.05
1.00 0.25
3.08 0.11
(44,099 3110 MÀ1 minÀ1
n.i.b
)
(83,456 19,089 MÀ1 minÀ1
>100 (ca. 20)a
)
b
8
n.i.b
>100 (ca. 15)b
52.9 11.3
11.9 4.33
a
% Inhibition at 100
Time-dependent inhibition, second-order rate constant k2nd
l
M; n.i. no inhibition at 100
lM, n.d. not determined.
b
.
ary (compound 5) or tertiary (compounds 4, 8) carbon atom is in
vicinity to the sulfur atom of the thiosulfinate which is attacked
by the active-site cysteine. Comparison of allicin (1) with its satu-
rated derivative 2 shows that allicin is more active. This might be
due to the reactivity of the allyl group of allicin which is known
to undergo various fragmentation reactions yielding several reac-
tive intermediates.11
The compounds do not inhibit one of the tested enzymes selec-
tively, and also no selectivity between cathepsin L-like enzymes
(falcipain 2, rhodesain, cathepsin L) and cathepsin B is observed.
A slight preference for cathepsin L-like enzymes is found for inhib-
itor 6 only. The enzyme-inhibiting potencies of the compounds
apparently contribute to their antiparasitic activity, as two of the
most potent falcipain and rhodesain inhibitors (1, 7) display potent
antiplasmodial and antitrypanosomal activity.
in performing the enzyme testings, and finally we thank our tech-
nicians and coworkers for bravely tolerating the garlic and mercap-
tan odours.
Supplementary data
Supplementary data (synthesis, analytical data of the com-
pounds) associated with this article can be found, in the online ver-
sion, at doi:10.1016/j.bmcl.2010.07.062.
References and notes
1. Stoll, A.; Seebeck, E. Adv. Enzymol. 1951, 11, 377.
2. Ankri, S.; Mirelman, D. Microb. Infect. 1999, 2, 125.
3. Ankri, S.; Miron, T.; Rabinkov, A.; Wilchek, M.; Mirelman, D. Antimicrob. Agents
Chemother. 1997, 41, 2286.
However, with the exception of these two compounds, no clear
correlation between antiprotease and antiparasitic activity is
found. This indicates that the compounds may also address other
targets (other proteases, e.g., falcipain 3, TbCatB,12 or non-protease
thiol-dependent enzymes, e.g., dehydrogenases) than the tested
cysteine proteases, and that other properties such as cell perme-
ability of the compounds may be important. Since the dihexyl
derivative 7 is more stable in solution than allicin, which is decom-
posed in DMSO solution within several days due to its allyl groups,
compound 7 may be a good starting point for designing allicin
derivatives with enhanced affinities. The affinities could probably
be optimized by moieties which better fit into the primed and
non-primed substrate binding pockets of the target enzymes.
4. Coppi, A.; Cabinian, M.; Mirelman, D.; Sinnis, P. Antimicrob. Agents Chemother.
2006, 50, 1731.
5. Rabinkov, A.; Miron, T.; Konstantinovski, L.; Wilchek, M.; Mirelman, D.; Weiner,
L. Biochim. Biophys. Acta 1998, 1379, 233.
6. Lowe, O. G. J. Org. Chem. 1975, 40, 2096.
7. Small, L. D.; Bailey, J. H.; Cavallito, C. J. J. Am. Chem. Soc. 1947, 69, 1710.
8. Youn, J.; Herrmann, R. Tetrahedron Lett. 1986, 27, 1493.
9. Barrett, A. J.; Rawlings, N. D. Biol. Chem. 2001, 382, 727.
10. McKerrow, J. H.; Rosenthal, P. J.; Swenerton, R.; Doyle, P. Curr. Opin. Infect. Dis.
2008, 21, 668.
11. Block, E.; Ahmad, S.; Catalfamo, J. L.; Jain, M. K.; Apitz-Castro, R. J. Am. Chem.
Soc. 1986, 108, 7045.
12. Kerr, I. D.; Wu, P.; Marion-Tsukamaki, R.; Mackey, Z. B.; Brinen, L. S. PLOS Negl.
Trop. Dis. 2010, 46, e701.
13. In order to elucidate whether, besides allicin (Scheme 2),5 other thiosulfinates
react with low molecular weight thiols and thiol-dependent enzymes via a thiol-
disulfide exchange we reacted thiosulfinate (2) with 4-methoxythiophenol and
o-amino thiophenol, respectively. The products were isolated and analyzed by
means of NMR spectroscopy. Indeed, the corresponding mixed disulfides, and
not the mixed thiosulfinates emerged from these reactions (Scheme 5).
14. Enzyme assays and parasite growth assays were performed as described
previously: (a) Breuning, A.; Degel, B.; Schulz, F.; Büchold, C.; Stempka, M.;
Machon, U.; Heppner, S.; Gelhaus, C.; Leippe, M.; Ley, M.; Kisker, C.; Rath, J.;
Stich, A.; Gut, J.; Rosenthal, P. J.; Schmuck, C.; Schirmeister, T. J. Med. Chem.
2010, 53, 1951; (b) Evers, A.; Heppner, S.; Leippe, M.; Gelhaus, C. Biol. Chem.
2008, 389, 1523.
Acknowledgements
Financial support by the Deutsche Forschungsgemeinschaft
(DFG) is gratefully acknowledged (SFB630). We thank Heidrun
Ließegang and Astrid Evers, both at Kiel University, Germany, for
assistance in performing in vitro testing against P. falciparum, and
Cornelia Heindl (University of Würzburg, Germany) for assistance
15. Otto, H.-H.; Schirmeister, T. Chem. Rev. 1997, 97, 133–171.