Synthesis of spiro-1,2-dioxolanes and their activity against Plasmodium falciparum

Article abstract of DOI:10.1016/j.bmcl.2008.10.083

Artemisinin-derived compounds play an integral role in current malaria chemotherapy. Given the virtual certainty of emerging resistance, we have investigated spiro-1,2-dioxolanes as an alternative scaffold. The endoperoxide functionality was generated by the SnCl₄-mediated annulation of a bis-silylperoxide and an alkene. The first set of eight analogs gave EC₅₀ values of 50-150 nM against Plasmodium falciparum 3D7 and Dd2 strains, except for the carboxylic acid analog. A second series, synthesized by coupling a spiro-1,2-dioxolane carboxylic acid to four separate amines, afforded the most potent compound (EC₅₀ ～5 nM).

Full text of DOI:10.1016/j.bmcl.2008.10.083

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6521–6524
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of spiro-1,2-dioxolanes and their activity
against Plasmodium falciparum
Derek C. Martyn a, Armando P. Ramirez b, , Meaghan J. Beattie a, Joseph F. Cortese a, Vishal Patel a,c
,
Margaret A. Rush a,c, K. A. Woerpel b, Jon Clardy a,c,
*
a Broad Institute Infectious Diseases Initiative, 7 Cambridge Center, Cambridge, MA 02142, USA
b Department of Chemistry, University of California—Irvine, Irvine, CA 92697, USA
c Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA 02115, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Artemisinin-derived compounds play an integral role in current malaria chemotherapy. Given the virtual
certainty of emerging resistance, we have investigated spiro-1,2-dioxolanes as an alternative scaffold.
The endoperoxide functionality was generated by the SnCl4-mediated annulation of a bis-silylperoxide
and an alkene. The first set of eight analogs gave EC50 values of 50–150 nM against Plasmodium falciparum
3D7 and Dd2 strains, except for the carboxylic acid analog. A second series, synthesized by coupling a
spiro-1,2-dioxolane carboxylic acid to four separate amines, afforded the most potent compound (EC50
ꢀ5 nM).
Received 4 August 2008
Revised 8 October 2008
Accepted 10 October 2008
Available online 1 November 2008
Keywords:
Malaria
Plasmodium
Endoperoxide
Dioxolane
Ó 2008 Elsevier Ltd. All rights reserved.
Malaria is a major parasitic disease in many areas of Africa, Asia
and South America. While four species of Plasmodium infect hu-
mans, the majority of malaria cases are caused by Plasmodium fal-
ciparum, and its resistance to many contemporary antimalarials
has compounded the problem of disease control.1 Several of the
current combination therapies in clinical use for malaria treatment
rely on endoperoxide-based compounds.2 Artemisinin (1a, Fig. 1),
the parent compound from which the clinically relevant endoper-
oxide antimalarials 1b–d are derived, is a natural product isolated
from the Sweet Wormwood tree (Artemisia annua). As P. falciparum
has developed resistance to every other clinically deployed agent,
it seemed prudent to investigate alternative artemisinin-inspired
templates. Several groups have pursued the development of anti-
malarial endoperoxides produced by fully synthetic methods.3
The most well known is the trioxolane OZ277 (Figs. 1 and 2). Its
central trioxolane core resembles the endoperoxide of artemisi-
nin,4 and the amino side group was introduced to improve its phar-
macokinetic properties.3a
Figure 1. Artemisinin (1a), its semisynthetic derivatives (1b–d), and synthetic
trioxolane 2.
type compounds. A recent report6 described a similar approach
that afforded 1,2-dioxolanes with antiplasmodial activity. This let-
ter summarizes the synthesis and testing of a similar series of ana-
logs, in addition to the development of a synthetic pathway that
enables the incorporation of a wide range of auxiliary groups using
simple coupling methodology.
Crucial in the development of OZ277 was the observation that
trioxolanes 3 and 4 were completely inactive (Fig. 2), suggesting
that antimalarial activity decreases when the peroxide bond is
too exposed (as in 3) or sterically inaccessible (4).3a,7 For trioxolane
5, a compromise between stability and potency was met. Prepara-
tion of the dioxolane equivalents of 3 and 56 (6 and 7, Fig. 2) gave
In an attempt to discover low-cost, short synthetic routes to-
wards novel endoperoxides that display antiplasmodial activity,
we utilized a SnCl4-mediated annulation strategy5 to form a 1,2-
dioxolane core as an alternative to the trioxolane core of OZ277-
* Corresponding author. Tel.: +1 617 432 2845; fax: +1 617 738 3702.
E-mail address: jon_clardy@hms.harvard.edu (J. Clardy).
Present address: United States Patent and Trademark Office, Alexandria, VA
22313, USA.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.083

6522
D. C. Martyn et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6521–6524
Figure 2. Development of the active trioxolane and dioxolane scaffolds.
Scheme 1. Reagents: (i) H2O2, HCl; (ii) TES–Cl, Et3N (11a, 49%), (11b, 93%), (11c,
87%), (11d, 21%) (all yields over two steps); (iii) SnCl4 (8, 55%), (13a, 59%), (13b,
80%), (13c, 46%), (13d, 51%); (iv) Ac2O (14a, 51%); (v) MeOTf (14b, 43%); (vi) H5IO6,
RuCl3ÁH2O (14c, 57%).
Figure 3. Assay results for 8 and 9 against P. falciparum (EC50 values determined via
a 3H-hypoxanthine incorporation assay8).
contrasting results to that of the trioxolanes, with the more active
compound being the exposed dioxolane 6.
Table 1
Diastereomeric ratios and assay results for 13a–e and 14a–c
Our efforts have focused on compounds with accessible dioxo-
lane cores analogous to 6. A set of compounds that possessed a
variety of groups appended to the dioxolane core was tested
against P. falciparum (see Ref. 5 for synthetic details). The two most
active compounds, 8 and 9, are displayed in Fig. 3. The main struc-
tural feature of these compounds was the presence of alkyl and
cyclohexyl groups at the 3 and 5 positions of the dioxolane respec-
tively. This core structure served as the basis for the design and
synthesis of further analogs.
Synthesis of the first series of analogs is outlined in Scheme 1.
Cyclohexanones 10a–d were converted to the bis-silylperoxides
11a–d, which in turn were treated with alkene 12a in the presence
of SnCl4 to afford spiro-1,2-dioxolanes 8 and 13a–c. bis-Silylperox-
ide 11a was also treated with alkene 12b to afford the hydroxy-
ethyl spiro-1,2-dioxolane 13d, which was then separately
acetylated, methylated or oxidized to give 14a–c.
Compound
d.r.a
EC50 Pf 3D7 (nM)
EC50 Pf Dd2 (nM)
8
1:1
92
545
86
141
241
109
69
151
59
74
146
149
114
62
>1000
4
13a
8:1
>99:1
NA
1:1
1:1
1:1
1:1
NA
13b
13c
13d
14a
14b
14c
Artemisinin
>1000
4
a
Diastereomeric ratios were determined by GC–MS.
All compounds were tested in a DAPI-staining based in vitro
live/dead P. falciparum assay9 against parasite strains 3D7 and
Dd2. With the exception of the carboxylic acid 14c, all compounds
gave similar EC50 values (Table 1). The lower activity of 14c most
likely originates in its inability to cross cell membranes as a result
of its charged carboxylate group. The greater activity seen for
resynthesized 8 compared to the original material probably arises
from differences in the two parasite cell viability assays. Overall,
the assay results suggested that future series of analogs should fo-
cus on incorporating groups with more favorable pharmacokinet-
ics, analogous to the development of OZ277.
The synthesis of the second series of analogs is shown in
Scheme 2. Bis-silylperoxide 11d was treated with a TBDPS-pro-
tected hydroxyalkene, and the dioxolane was deprotected with
TBAF to afford 15. Oxidation gave carboxylic acid 16, which was
coupled to amines using PyBOP or EDCI/HOBT to afford amides
17a and 17b, the chloroquine (CQ) mimic 17c, and the OZ277
side-chain mimic 17d.
Dioxolanes 17a, 17b, and 17d were equipotent to the majority
of the first set of compounds (see Table 2), suggesting that the
presence of an amide has no detrimental effect upon activity,
Scheme 2. Reagents: (i) SnCl4; (ii) TBAF (42% over two steps); (iii) H5IO6, RuCl3ÁH2O
(96%); (iv) Amine, PyBOP, Et3N (17a, 72%), (17b, 65%); (v) EDCI, HOBT, then amine
(17c, 46%), (17d, 58%).

D. C. Martyn et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6521–6524
6523
Table 2
Table 4
Assay results for 17a–d
Assay results for 17c, 18, and 21 against three CQ-resistant P. falciparum strains
Compound
EC50 Pf 3D7 (nM)
EC50 Pf Dd2 (nM)
Compound
EC50 (nM)
FCB
17a
17b
17c
17d
100
81
5
77
57
6
7G8
106/1
17c
18
21
CQ
8
83
91
46
48
16
356
80
52
8
22
142
53
14
13
61
108
Artemisinin
than the parent compound, revealing that the high potency of 17c
results from the presence of both the endoperoxide and 7-chloro-
quinoline functionalities. Testing 18 in a 1:1 combination with CQ
gave a similar EC50 value against 3D7 to that of 17c, whereas the
higher EC50 value against Dd2 resulted from this strain’s resistance
to CQ. The 1:1 combination of 18 and 19 gave similar EC50 values
against both strains, further underlining the lower activity of 19
compared to CQ despite their analogous structures.
Scheme 3. Reagents: (i) EtNH2, PyBOP, Et3N (45%).
Dioxolanes 7, 8, 13a–13d, 14a, 14b, and 17a–d were subjected
to an assay that tests for a compound’s ability to inhibit heme crys-
tallization in a parasite-independent assay.11 None of these com-
pounds displayed activity compared to the positive controls
amodiaquine and CQ. Compounds 19 and 21 were also inactive
in this assay, which indicates that the 7-chloroquinoline moiety
possessed by these compounds and 17c does not inhibit heme
crystallization per se.
Scheme 4. Reagents: (i) BrCH2CO2H, EDCI (22%); (ii) Amine, K2CO3 (31%).
Compounds 17c, 18, and 21 were tested against three geograph-
ically distinct CQ-resistant P. falciparum strains to determine their
ability to overcome various resistance mechanisms (for the Pfcrt
mutations specific to each strain, see Ref. 12). Dioxolane 17c main-
tained potency against all three strains, as opposed to the observed
attenuation of CQ activity, particularly in strains 7G8 and FCB (Ta-
ble 4). Compound 18 displayed less activity than 17c, and it exhib-
ited the high degree of variability in EC50 values against different
strains that was shown against 3D7 and Dd2 (see Table 3). Finally,
the absence of an endoperoxide moiety in 21 contributed to de-
creased activity against these strains compared to 17c, underlining
the importance of the endoperoxide for activity.
In summary, we have utilized a SnCl4-mediated annulation to
synthesize compounds based on a spiro-1,2-dioxolane motif. Anti-
plasmodial activity appeared to be relatively independent of the
nature of the group appended to the 3 position, with the exception
of 17c, which possessed a 7-chloroquinoline group analogous to
that of CQ. This compound displayed potent activity against a
range of P. falciparum strains, including CQ-resistant ones, and
the combination of the endoperoxide and 7-chloroquinoline moie-
ties are required for this activity. Conversely, there is little evi-
dence that the 7-chloroquinoline functionality present in
compounds 17c, 19, and 21 acts in an analogous manner to CQ,
as these compounds did not inhibit heme crystallization, and the
potentiation of 17c activity must arise through another
mechanism.
and that most auxiliary groups have little effect on potency against
P. falciparum.
Dioxolane 17c was the most active compound found in this
study, and its potency is associated with a 7-chloroquinoline
group, analogous to that possessed by CQ. Walsh et al.10 recently
demonstrated that an artemisinin–quinine adduct displayed activ-
ity greater than that of artemisinin, quinine, or a 1:1 mixture of
artemisinin and quinine. The authors proposed that the adduct’s
increased potency may arise through its enhanced cellular uptake
over that of the individual components. Hydrolysis of the ‘mutual
prodrug’ affords the individual components and delivers them to
their site of action.
Two compounds were synthesized to determine whether the
high activity of 17c resulted predominantly from the 7-chloroquin-
oline moiety, or whether the endoperoxide plays an additional
role. Coupling of EtNH2 to acid 16 afforded 18, an analog of 17c that
lacks the 7-chloroquinoline group (Scheme 3). Additionally, a des-
peroxy analog of 17c was synthesized (Scheme 4). Coupling of
bromoacetic acid to 19 afforded 20, which gave pyrrolidine 21 by
the alkylation of spiroazadecane.
The assay results for dioxolane 18, amine 19 and pyrrolidine 21
reveal the origin of the high activity of 17c (Table 3). For compar-
ative purposes, CQ was tested as a control, and the expected differ-
ence in potency against 3D7 (CQ susceptible) and Dd2 (CQ
resistant) was observed. The individual components of 17c (18
and 19) and the desperoxy analog 21 were considerably less active
Acknowledgments
The authors thank the Medicines for Malaria Venture (MMV) for
financial support, and SPARC funding from the Broad Institute of
MIT and Harvard. A.R. and K.A.W. thank the National Science Foun-
dation (CHE-0315572) for support.
Table 3
Assay results for 18, 19, 21, and 18 in combination with CQ or 19
Compound
EC50 Pf 3D7 (nM)
EC50 Pf Dd2 (nM)
18
19
21
CQ
485
57
69
7
99
155
78
Supplementary data
66
18 + CQ
18 + 19
6
50
42
65
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.10.083.

6524
D. C. Martyn et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6521–6524
5. Ramirez, A.; Woerpel, K. A. Org. Lett. 2005, 7, 4617.
References and notes
6. Wang, X.; Dong, Y.; Wittlin, S.; Creek, D.; Chollet, J.; Charman, S. A.; Tomas,
J. S.; Scheurer, C.; Snyder, C.; Vennerstrom, J. L. J. Med. Chem. 2007, 50,
5840.
7. Dong, Y.; Chollet, J.; Matile, H.; Charman, S. A.; Chiu, F. C. K.; Charman, W. N.;
Scorneaux, B.; Urwyler, H.; Tomas, J. S.; Scheurer, C.; Snyder, C.; Dorn, A.; Wang,
X.; Karle, J. M.; Tang, Y.; Wittlin, S.; Brun, R.; Vennerstrom, J. L. J. Med. Chem.
2005, 48, 4953.
1. Daily, J. P. J. Clin. Pharmacol. 2006, 46, 1487.
2. Meshnick, S. R.; Taylor, T. E.; Kamchonwongpaisan, S. Microbiol. Rev. 1996, 60,
301.
3. (a) Vennerstrom, J. L.; Arbe-Barnes, S.; Brun, R.; Charman, S. A.; Chiu, F. C.;
Chollet, J.; Dong, Y.; Dorn, A.; Hunziker, D.; Matile, H.; McIntosh, K.;
Padmanilayam, M.; Santo, T. J.; Scheurer, C.; Scorneaux, B.; Tang, Y.; Urwyler,
H.; Wittlin, S.; Charman, W. N. Nature 2004, 430, 900; (b) Singh, C.; Sharma, U.;
Saxena, G.; Puri, S. K. Bioorg. Med. Chem. Lett. 2007, 17, 4097; (c) Singh, C.;
Kanchan, R.; Sharma, U.; Puri, S. K. J. Med. Chem. 2007, 50, 521; (d) Dong, Y.; Creek,
D.; Chollet, J.; Matile, H.; Charman, S. A.; Wittlin, S.; Wood, J. K.; Vennerstrom, J. L.
Antimicrob. Agents Chemother. 2007, 51, 3033; (e) Terzic, N.; Opsenica, D.; Milic,
D.; Tinant, B.; Smith, K. S.; Milhous, W. K.; Solaja, B. A. J. Med. Chem. 2007, 50,
5118; (f) O’Neill, P. M.; Rawe, S. L.; Borstnik, K.; Miller, A.; Ward, S. A.; Bray, P. G.;
Davies, J.; Oh, C. H.; Posner, G. H. ChemBioChem 2005, 6, 2048; (g) Benoit-Vical, F.;
Lelievre, J.; Berry, A.; Deymier, C.; Dechy-Cabaret, O.; Cazelles, J.; Loup, C.; Robert,
A.; Magnaval, J.-F.; Meunier, B. Antimicrob. Agents Chemother. 2007, 51, 1463.
4. Kaiser, M.; Wittlin, S.; Nehrbass-Stuedli, A.; Dong, Y.; Wang, X.; Hemphill, A.;
Matile, H.; Brun, R.; Vennerstrom, J. L. Antimicrob. Agents Chemother. 2007, 51,
2991.
8. Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.; Chulay, J. D. Antimicrob. Agents
Chemother. 1979, 16, 710.
9. Baniecki, M. L.; Wirth, D. F.; Clardy, J. Antimicrob. Agents Chemother. 2007, 51,
716.
10. Walsh, J. J.; Coughlan, D.; Heneghan, N.; Gaynor, C.; Bell, A. Bioorg. Med. Chem.
Lett. 2007, 17, 3599.
11. Rush, M. A.; Baniecki, M. L.; Mazitschek, R.; Cortese, J. F.; Wiegand, R.; Clardy, J.;
Wirth, D. F., in preparation.
12. Strain, source, Pfcrt mutations. Dd2, Indochina/Laos, M74I, N75E, K76T,
A220S, Q271E, N326S, I356T, R371I. 7G8, Brazil, C72S, M74N, K76T,
A220S, N326D, I356L. FCB, SE Asia, M74I, N75E, K76T, A220S, Q271E,
N326S, I356T, R371I. 106/1, Sudan, M74I, N75E, A220S, Q271E, N326S,
I356T, R371I.