Biological Evaluation and X-ray Co-crystal Structures of Cyclohexylpyrrolidine Ligands for Trypanothione Reductase, an Enzyme from the Redox Metabolism of Trypanosoma

Article abstract of DOI:10.1002/cmdc.201800067

The tropical diseases human African trypanosomiasis, Chagas disease, and the various forms of leishmaniasis are caused by parasites of the family of trypanosomatids. These protozoa possess a unique redox metabolism based on trypanothione and trypanothione reductase (TR), making TR a promising drug target. We report the optimization of properties and potency of cyclohexylpyrrolidine inhibitors of TR by structure-based design. The best inhibitors were freely soluble and showed competitive inhibition constants (K_i) against Trypanosoma (T.) brucei TR and T. cruzi TR and in vitro activities (half-maximal inhibitory concentration, IC₅₀) against these parasites in the low micromolar range, with high selectivity against human glutathione reductase. X-ray co-crystal structures confirmed the binding of the ligands to the hydrophobic wall of the “mepacrine binding site” with the new, solubility-providing vectors oriented toward the surface of the large active site.

Full text of DOI:10.1002/cmdc.201800067

DOI: 10.1002/cmdc.201800067
Full Papers
Biological Evaluation and X-ray Co-crystal Structures of
Cyclohexylpyrrolidine Ligands for Trypanothione
Reductase, an Enzyme from the Redox Metabolism of
Trypanosoma
Raoul De Gasparo,[a] Elke Brodbeck-Persch,[a] Steve Bryson,[b, c] Nina B. Hentzen,[a]
Marcel Kaiser,*[d, e] Emil F. Pai,*[b, c] R. Luise Krauth-Siegel,*[f] and FranÅois Diederich*[a]
The tropical diseases human African trypanosomiasis, Chagas
disease, and the various forms of leishmaniasis are caused by
parasites of the family of trypanosomatids. These protozoa
possess a unique redox metabolism based on trypanothione
and trypanothione reductase (TR), making TR a promising drug
target. We report the optimization of properties and potency
of cyclohexylpyrrolidine inhibitors of TR by structure-based
design. The best inhibitors were freely soluble and showed
competitive inhibition constants (Ki) against Trypanosoma (T.)
brucei TR and T. cruzi TR and in vitro activities (half-maximal in-
hibitory concentration, IC50) against these parasites in the low
micromolar range, with high selectivity against human gluta-
thione reductase. X-ray co-crystal structures confirmed the
binding of the ligands to the hydrophobic wall of the “mepa-
crine binding site” with the new, solubility-providing vectors
oriented toward the surface of the large active site.
Introduction
Human African trypanosomiasis (HAT, also known as sleeping
sickness), Chagas disease, and the various forms of leishmania-
sis are caused by parasites of the family of trypanosomatids,
namely Trypanosoma (T.) brucei ssp., T. cruzi, and Leishmania
(L.) spp., respectively. These tropical diseases endanger a popu-
lation of around 500 million people, with an estimated 50000
deaths occurring annually.[1] Most affected are the poorest re-
gions of sub-Saharan Africa, central and South America, India,
and Asia, where control, prevention, and treatment of the dis-
eases are more challenging due to economic and social condi-
tions.[1,2] Currently used drugs have major drawbacks, such as
prolonged and complicated administration, and severe side ef-
fects, worsened by the emergence of resistance, highlighting
the necessity for new, safe, affordable, and orally administered
drugs.[1a,2,3] To date, only a limited number of new active sub-
stances against trypanosomatids are in late-stage develop-
ment.[2,3] Trypanosomatids possess a unique redox metabolism
based on the dithiol trypanothione and the enzyme trypano-
thione reductase (TR) that differs distinctively from the gluta-
thione and glutathione reductase (GR)-based metabolism in
mammals.[4] This peculiarity provides promising targets for the
development of selective drugs. The two most widely investi-
gated target enzymes of the trypanothione-based redox me-
tabolism are TR and trypanothione synthetase.[5] TR is a homo-
dimeric flavoenzyme that catalyzes the NADPH-dependent re-
duction of trypanothione disulfide (TS2, 1, Figure 1) to trypano-
thione.[4a] Trypanothione is used for various essential processes
by the parasites: maintenance of the intracellular reducing en-
vironment, synthesis of DNA precursors, defense against reac-
tive oxygen species, and detoxification of heavy metals.[4a] In-
hibition of ꢀ90% of the TR activity in T. brucei has been
shown to cause increased sensitivity to oxidative stress, culmi-
nating in the death of the parasite.[6]
[a] R. De Gasparo, Dr. E. Brodbeck-Persch, N. B. Hentzen, Prof. Dr. F. Diederich
Laboratorium fꢀr Organische Chemie, ETH Zꢀrich, Vladimir-Prelog-Weg 3,
8093 Zꢀrich (Switzerland)
E-mail: diederich@org.chem.ethz.ch
[b] Dr. S. Bryson, Prof. Dr. E. F. Pai
Departments of Biochemistry, Medical Biophysics, and Molecular Genetics,
University of Toronto, Medical Sciences Building, #5358, 1 King’s College
Circle, Toronto, ON, M5S 1A8 (Canada)
[c] Dr. S. Bryson, Prof. Dr. E. F. Pai
The Campbell Family Institute for Cancer Research, University Health Net-
work, 101 College Street, Toronto, ON, M5G 1L7 (Canada)
E-mail: pai@hera.med.utoronto.ca
[d] Dr. M. Kaiser
Swiss Tropical and Public Health Institute, Socinstrasse 57, 4002 Basel (Swit-
zerland)
E-mail: marcel.kaiser@swisstph.ch
[e] Dr. M. Kaiser
In 2014, we reported the design, biological evaluation, and
analysis of the binding mode of small TR inhibitors derived
from the 1-[1-(benzo[b]thien-2-yl)cyclohexyl]piperidine (BTCP,
2, Figure 1) structure.[7] BTCP (2) was previously discovered by
Fairlamb and co-workers to be a competitive inhibitor of
T. brucei TR with a competitive inhibition constant (Ki) of
1 mm.[8] Co-crystal structures of inhibitor 3 with the TR of
University of Basel, Petersplatz 1, 4003 Basel (Switzerland)
[f] Prof. Dr. R. L. Krauth-Siegel
Biochemie-Zentrum Heidelberg (BZH), Universitꢁt Heidelberg, Im Neuen-
heimer Feld 328, 69120 Heidelberg (Germany)
E-mail: luise.krauth-siegel@bzh.uni-heidelberg.de
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under:
https://doi.org/10.1002/cmdc.201800067.
ChemMedChem 2018, 13, 1 – 12
1
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Figure 1. Molecular structures of TR substrate trypanothione disulfide (TS2,
1), TR inhibitors 1-[1-(benzo[b]thien-2-yl)cyclohexyl]piperidine (BTCP, 2), and
BTCP analogue 3. The molecular weights (MW) are indicated.[7,8]
T. brucei and T. cruzi have been solved, providing a basis for ra-
tional structure-based design aimed at improving the potency
and physical properties of TR inhibitors containing this scaf-
fold.[7,9] The co-crystal structures[7] showed that inhibitor 3
binds in both enzymes to the so-called mepacrine-binding
site[10] and that two different binding modes can be populated
in the spacious cavity of TR (Figure 2a and Section S1 in the
Supporting Information (SI)).[7,11] The key interactions observed
in both binding modes are Coulombic interactions between
the protonated pyrrolidine nitrogen and the side chain of
Glu18, CꢁH···p interactions of the cyclohexyl and pyrrolidine
moieties with Trp21, and further dispersive contacts to Leu17
(Figures S1 and S2 in Section S1 in the SI). Both active sites of
the homodimeric enzyme are occupied by 3 with slight differ-
ences in the binding mode. In one active site of T. brucei TR
(monomer B, see Figure S1b in Section S1 in the SI), the thia-
zole ring establishes S···p and dispersive S···S interactions[12]
with the side chain of Met113, whereas the indole moiety in-
teracts with Met113 (SCꢁH···p) and the amide bond between
Gly112 and Met113 through amide···p stacking.[13] In the other
active site, the indole ring is turned by 1808, thereby establish-
ing a short CꢁH···p interaction (d(C···C)=3.6 ꢁ) with HꢁCa of
Met113 (monomer A, see Figure S1a in Section S1 in the SI).[7]
In T. cruzi TR, the replacement of Gly112 (T. brucei TR) with
the larger, highly solvated side chain of Glu112[14] is assumed
to cause the rotation of the bond between the thiazole ring
and the attached cyclohexyl C(sp3) atom by 1808, leading to
the indole ring pointing away from Met113 toward Ile106 and
to the formation of a short OꢁH···p interaction with the side
chain of Ser109 (Figure S2 in Section S1 in the SI).[7]
Figure 2. a) Overlay of the co-crystal structures of inhibitor 3 bound to
T. brucei (PDB ID: 4NEV, 2.5 ꢁ resolution) and T. cruzi TR (4NEW, 2.8 ꢁ)[7] show-
ing the two observed orientations of the indolyl-thiazole moiety. The surface
and binding residues of T. brucei TR are shown. A close-up of the overlay is
provided in Figure S3 in Section S1 in the SI. b) Predicted binding mode of
bis-aryl-substituted thiophene 4 in the active site of T. brucei TR (PDB ID:
4NEV). Color codes: gray, Cenzyme (T. brucei TR); green, Cligand 3 (bound to
T. brucei TR); yellow, Cligand 3 (bound to T. cruzi); violet, Cligand 4; blue, N; red, O;
yellow, S.
The aim of the present study was to further optimize low-
molecular-weight inhibitors of TR derived from the BTCP scaf-
fold, ultimately seeking enhanced inhibitory potencies and op-
timized properties for projected in vivo studies. The BTCP li-
gands synthesized to date have only limited water solubility,
which in the previous study prevented the biological evalua-
tion of almost half of the ligand series and rendered co-crystal-
lization experiments highly challenging.[7] Herein, we report
the synthesis and biological evaluation of three new BTCP-
based TR inhibitor series with the goal to provide ligands with
improved water solubility and potency (Scheme 1). New X-ray
co-crystal structures were obtained, which hold great promise
for guiding future rational design work.
&
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
2
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Scheme 1. Design strategies for the improvement of TR inhibitor 3.
Results and Discussion
Ligand design
The design strategies for optimization of inhibitor 3[7] are de-
picted in Scheme 1. We hoped for improved biological activity
by combining the two different binding modes observed in
the co-crystal structures of 3 with T. brucei and T. cruzi TR, thus
targeting BTCP derivatives 4–11 with diarylated thiophene
cores (Figure 3). Overlays of the predicted binding of 4 (Fig-
ure 2b), obtained using the modeling software MOLOC,[15] with
the co-crystal structures of 3 supported this hypothesis (Fig-
ure S4 in Section S2 in the SI).
To increase the water solubility, two different strategies
aiming at decreasing the calculated logD7.4 values (clogD at
pH 7.4; value for 3: 3.9) were employed. In the first approach,
the lipophilic cyclohexyl moiety was replaced by an oxetanyl
moiety,[16] leading to ligands 12–15 (Figure 4a). Modeling
using MOLOC suggested that the polar O atom of the oxetane
ring would favorably point toward solvent. Density functional
theory (DFT) calculations at the B3LYP/6-31G(d) level of theory
(in water with polarizable continuum solvent model) using
Gaussian09[17] showed that the alignment of the pyrrolidine
and heteroarene moieties in 3 was conserved upon introduc-
tion of the oxetanyl ring (Figure S5 in Section S3.1 in the SI).
In the second approach, water-solubilizing morpholine, N-
methylpiperazine, piperazine, or piperidine moieties were at-
tached via a short C2 linker to the N atom of the indole moiety
Figure 4. Molecular structures of a) oxetane derivatives 12–15 and b) N-sub-
stituted indole derivatives 16–19. Calculated molecular properties (clogD7.4
TPSA, and pKa) are listed in Table S1 in Section S4.1 in the SI.
,
of 3, yielding the targeted compounds 16–19, which at physio-
logical pH should be fully (piperidine) or partially protonated
(Figure 4b). The calculated topological polar surface area
(TPSA,[18] see Table S1 in Section S4.1 in the SI for details on the
calculations) was maintained below 80 ꢁ2 to enhance the prob-
ability of the compounds to cross the blood–brain barrier in
order to be potentially active in combatting the central nerv-
ous system stage of HAT.[19]
Figure 3. Molecular structures of the diaryl-substituted thiophenes 4–11. The dihedral angles S1ꢁC1ꢁC2ꢁC3 and C4ꢁC5ꢁC6ꢁC7 of 4 are defined.
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
3
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Scheme 3. Synthesis of oxetane derivatives 12–15: a) Ti(OEt)4, THF, 508C,
4 h, 42%; b) 28, nBuLi, THF, ꢁ788C, 30 min, then 27, THF, ꢁ78 to 258C,
45 min, 60%; c) 4m HCl in dioxane, MeOH, 08C, 1 min, quant.; d) iPr2NEt,
CH3CN, 908C, 1.5–6 days, 60–90%.
Scheme 2. Synthesis of diaryl-substituted thiophene derivatives 4–11: a) 1H-
benzo[d][1,2,3]-triazole, tetrahydropyran, 258C, then addition of (4,5-dibro-
mothien-2-yl)lithium, tetrahydropyran, ꢁ508C, 45%; b) arylboronates,
[PdCl2(dppf)], Cs2CO3, 1,4-dioxane/H2O (10:1), 68–858C, 2–23 h, 35–92%;
c) N-bromosuccinimide, DMF, 0 to 258C, 2 h, 67%; d) 21, 1H-benzo[d][1,2,3]-
triazole, THF, 258C, then addition of (4,5-diarylthien-2-yl)lithium, THF, ꢁ788C,
65%; e) KOH, THF/MeOH (10:1), 408C, 45 min, 99%. dppf=1,1’-bis(diphenyl-
phosphino)ferrocene, DMF=N,N-dimethylformamide, TIPS=triisopropylsilyl,
THF=tetrahydrofuran.
ic sequence was used, starting with Suzuki coupling between
commercially available 3-bromothiophene (22) and phenylbor-
onic acid.
A regioselective bromination[22] at position 2 of thiophene 23
using N-bromosuccinimide, followed by a second Suzuki cou-
pling, allowed the introduction of a triisopropylsilyl (TIPS)-pro-
tected indole to give intermediate 24 (Scheme 2). The transfor-
mation of enamine 21 with 1H-benzo[d][1,2,3]triazole, followed
by substitution with lithiated 24, and subsequent KOH-mediated
deprotection of the TIPS group afforded target compound 11.
The oxetane derivatives 12–15 were synthesized by a proce-
dure described by Hamzik and Brubaker (Scheme 3).[23] Con-
densation of 3-oxetanone (25) with tert-butylsulfinamide (26)
using titanium(IV) ethoxide as a dehydrating agent gave
imine 27.[23] Addition of lithiated benzo[d]thiophene (28) to
imine 27 provided compound 29 and subsequent acidic de-
protection gave the ammonium salt 30. Finally, cyclization with
Synthesis
The synthetic route to the diaryl-substituted thiophenes 4–10
is based on the preparation of the key dibromide intermediate
20, obtained by attack of 1H-benzo[d][1,2,3]triazole onto en-
amine 21 to give an N,N-acetal, followed by nucleophilic sub-
stitution with a lithiated 2,3-dibromothiophene (Scheme 2).[20]
Subsequent Suzuki cross-couplings[7,21] provided the target
structures. For the preparation of ligand 11, a different synthet-
Scheme 4. Synthesis of N-substituted indole derivatives 16–19 and 31: a) NaH, TsOEt, DMF, 0 to 458C, 2.5 h, 86%; b) NaH, 32, DMF, 0 to 458C, 4 h, 94%;
c) 1. TBDMSOTf, 2,6-lutidine, CH2Cl2, 258C, 1–1.5 h, 2. K2CO3, THF/MeOH (2:5), 258C, 1–1.5 h, 31–46%; d) NaH, TsCl, DMF, 0 to 258C, 19 h, 96%; e) NaH, Cs2CO3,
34–36, toluene, 1008C, 14–18 h, 38–68%. Ts=p-toluenesulfonyl, TBDMSOTf=tert-butyldimethylsilyl trifluoromethanesulfonate.
&
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
4
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
the respective a,w-dihalogenated alkyls yielded the target li-
gands 12–15.
weaker inhibition, and the solubility of ligand 6 in the assay
buffer was too low to allow inhibition measurements.
The synthesis of the N-substituted indole derivatives 16–19
and 31 started from 3 (Scheme 4).[7] Ligand 19 and control
compound 31 were obtained by nucleophilic substitution of
the deprotonated indole 3 with either ethyl p-toluenesulfo-
nate, to give 31, or tosylate 32 to give piperidine 19 after Boc
deprotection. The other derivatives 16–18 were prepared via
tosyl intermediate 33. Reaction of alcohols 34, 35, and 36 with
33 led to target ligands 16 and 17, and piperazine 18 (after
Boc cleavage of 37), respectively.[24]
A possible reason for the lack of improved affinity upon in-
troduction of a second aryl substituent on the thiophene ring
was provided by a conformational analysis of the unbound and
bound ligand. Upon binding to the enzyme, the ligand adopts
an unfavorable conformation that is higher in energy than that
of the free ligand. Therefore, the energy gained by the interac-
tions of the additional aryl ring (as compared with 3) is cancel-
led by the cost of this conformational change. A comparison of
the bound structure obtained by modeling using MOLOC with
the structure of the free ligand optimized by DFT calculations
(at the B3LYP/6-31G(d) level of theory in water with polarizable
continuum solvent model) revealed that the torsion angles be-
tween the two aryl substituents and the thiophene moiety
were substantially smaller in the binding mode predicted by
modeling (Figure S6 in Section S3.2 in the SI). Indeed, the dihe-
dral angles S1ꢁC1ꢁC2ꢁC3 and C4ꢁC5ꢁC6ꢁC7 were predicted
to be 288 and 398, respectively, in the bound conformation,
whereas in the free state, dihedral angles of 478 and 458, re-
spectively were preferred (see Figure 3 for the definition of the
dihedral angles). Optimized binding is therefore only achieved
for higher-energy conformations of the ligands.
Enzymatic assays
All ligands were tested in enzymatic assays against T. cruzi and
T. brucei TR (see Section S5.1 in the SI for experimental de-
tails).[7,25] Surprisingly, none of the ligands of the diaryl-substi-
tuted thiophene series showed improved binding affinity
against either T. cruzi or T. brucei TR compared to inhibitor 3
(Table 1). Ligands 4, 5, 10, and 11 displayed inhibitory potencies
against TR from both species that were similar to that of 3.
The ligands were either pure competitive inhibitors or
mixed-type. Thus, no relevant dependency on the diaryl substi-
tution pattern on the thiophene ring was observed. Com-
pounds 7–9 with smaller diaryl substituents displayed a
The substitution of an oxetane for the cyclohexyl ring in 12–
15 led to improved water solubility, allowing the members of
Table 1. Inhibition of T. cruzi and T. brucei TR by diaryl-substituted thiophene derivatives 4–11.
T. cruzi TR
T. brucei TR
Compd
Ar1
–
Ar2
Inh. [%][a]
51
Ki [mm][b]
4.0ꢂ0.5
Inh. [%][a]
19
Ki [mm][b]
12ꢂ1
3[7]
20[c]
(42)
4
5
7
8
9
43
41
23
15
10
4.8ꢂ0.8
6.3ꢂ1.3
n.d.
23
14
11
11
10
35[c]
(78)
n.d.
n.d.
n.d.
n.d.
n.d.
10[c]
(56)
10
11
38
41
4.3ꢂ0.6
5.6ꢂ0.7
29
14
25ꢂ9
(133)
[a] Percent inhibition by 12 mm inhibitor in the presence of 40 mm TS2, which corresponds to ꢃ3–4-fold KM for T. brucei TR and ꢃ2-fold KM for T. cruzi
TR.[7,25,26] The reported values are the means of at least two measurements. [b] At least two different inhibitor concentrations measured, unless otherwise
stated. The Ki’ value is given in parentheses if mixed-mode inhibition was observed. [c] Only one inhibitor concentration was used to determine the inhibi-
tion constants due to low solubility of the compounds in the assay buffer; the inhibitor constants were derived from the linear plot; n.d. =not deter-
mined.
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
5
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
this series to be tested without the solubility issues encoun-
tered with the previous series (see Section S5.4 for further de-
tails). Under the assay conditions applied ([inhibitor]=[TS2]=
40 mm; see Table S2 in Section S5.2 in the SI), however, no in-
hibition against either T. cruzi or T. brucei TR was measured.
Nevertheless, these negative findings provided some inter-
esting insight into the energetic contributions of the cyclohex-
ylpyrrolidine moiety to the binding of the BTCP analogues. The
presence of the s-electron-withdrawing oxetanyl O atom in b–
position to the pyrrolidine N atom decreases the calculated pKa
value of the tertiary amine center by two units (Table S1 in Sec-
tion S4.1 in the SI).[16,27] Under the assay conditions applied
(pH 7.5, Section S5.1 in the SI), these amines are no longer fully
protonated, leading to the loss of ion-pairing interactions with
Glu18. This complete disappearance in affinity was not expect-
ed based on previous mutation studies. Indeed, mutation of
Glu18 to Ala18 led to only a threefold decrease in the inhibito-
ry activity of 3.[7] Therefore, we assumed that the CꢁH···p inter-
actions of the protonated pyrrolidine with Trp21 and accompa-
nying cation–p interactions[12a,28] are substantial contributors to
ligand complexation. If the pyrrolidine N atom is not protonat-
ed, the neighboring CꢁH bonds are much less polarized, result-
ing in much weaker CꢁH···p interactions and in the absence of
cation–p interactions. Furthermore, the smaller oxetanyl ring
might lead to less favorable desolvation upon complexation
than the larger cyclohexyl ring.
TR (Table 2). Although the ligand efficiency (LE) decreases from
3 (LE, T. cruzi =0.29 kcalmolꢁ1; LE, T. brucei=0.27 kcalmolꢁ1) to
18 (LE, T. cruzi=0.23 kcalmolꢁ1; LE, T. brucei=0.22 kcalmolꢁ1),
this new compound provides an improved lead for further de-
velopment. The introduction of ethylmorpholine in 16 or ethyl
in 31 reduced the biological activity.
X-ray co-crystal structures: binding mode of the N-substitut-
ed indole series
Co-crystal structures of ligands 18 (PDB ID: 6BTL) and 19 (PDB
ID: 6BU7) in complex with T. brucei TR were solved at 2.8 and
2.7 ꢁ resolution, respectively. Binding occurs in a similar fash-
ion to the accommodation of 3 in the active site of T. brucei TR
(PDB ID: 4NEV)[7] with the indolyl-thiazole cores adopting iden-
tical orientations (Figure 5a). The newly introduced water-solu-
bility-providing substituents orient toward the periphery of the
active site. Both active sites (monomer A and monomer B) in
the two homodimeric complexes appear fully occupied, but
with differing electron densities and temperature factors (B-fac-
tors). The latter are correlated to mobility. In the main text,
only the complexes with lower temperature factors (mono-
mer B) are discussed.
The ligand orientation differs slightly in the four active sites
of the two homodimeric complexes (for an overlay, see Fig-
ure S9 in Section S6.3 in the SI). The previously investigated in-
teractions of ligand 3 with T. brucei TR[7] are conserved in the
two new co-crystal structures (Figures 5b and 6). The pyrroli-
dine moiety establishes a Coulombic interaction with Glu18
(d(Npyrrolidine···OGlu18)=4.0 and 4.1 ꢁ) and CꢁH···p interactions
with Trp21 (d(CH2···CTrp21)=3.7 ꢁ), whereas the thiazole ring is
involved in orthogonal S···p and dispersive S···S interactions
with Met113 (d(S···SMet113)=4.2 and 4.1 ꢁ). The indole moiety is
The water solubility of all the members of the N-substituted
indole series with a six-membered heteroalicyclic substituent
was also significantly improved (see Section S5.4 in the SI for
further details). Ligands 17–19 showed stronger, fully competi-
tive, inhibition of TR of both species in comparison with 3, with
inhibitor 18 being the best of the series with a twofold lower Ki
value for T. cruzi TR and a threefold lower Ki value for T. brucei
Table 2. Inhibition of T. cruzi and T. brucei TR by N-substituted indole analogues 16–19 and 31.
T. cruzi TR
T. brucei TR
Compd
R
Inh. [%][a]
79
Ki [mm][b]
4.0ꢂ0.5
Inh. [%][a]
43
Ki [mm][b]
12ꢂ1
3[7]
16
17
18
38
78
88
n.d.
34
60
80
n.d.
3.2ꢂ0.3
2.1ꢂ0.2
6.5ꢂ0.6
3.8ꢂ0.3
19
31
84
27
2.9ꢂ0.3
68
14
6.4ꢂ0.7
n.d.
n.d.
[a] Percent inhibition by 40 mm inhibitor in the presence of 40 mm TS2, which corresponds to ꢃ3–4-fold KM for T. brucei TR and ꢃ2-fold KM for T. cruzi
TR.[7,25,26] The reported values are the means of at least two measurements. [b] At least two different inhibitor concentrations measured, unless otherwise
stated; n.d.=not determined.
&
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
involved in CꢁH···p interactions with the side chain of Met113
(d(C···CaMet113)=4.1 and 3.7 ꢁ).
In the other active sites (corresponding to monomers A) of
the homodimeric TR complexes of 18 and 19, a defined water
molecule solvates the side chain of Glu18 (Figures S10 and S11
in Sections S6.4 and S6.5 in the SI). In these active sites, a dif-
ferent major conformation is assigned to the cyclohexyl ring of
the ligands but the broader electron density maps suggest
that the cyclohexyl ring adopts at least two different chair con-
formations. A study on BTCP (2) revealed that the equilibrium
between these two cyclohexyl conformations is solvent depen-
dent and that the Gibbs energy between them varies between
0.26 and ꢁ0.51 kcalmolꢁ1 in different solvents.[29] Nevertheless,
the pyrrolidine N atoms are at a reasonable distance to the car-
boxylate of Glu18 (d(N···OGlu18)=4.3 and 4.6 ꢁ) allowing for
Coulombic interactions also in this conformation.
The indole and thiazole ring planes deviate from co-planarity
with dihedral angles between ꢁ48 and 228 (for definition of
the dihedral angle and example of the ligand with an angle of
228, see Figure S11 in Section S6.5 in the SI). These values are
in agreement with the dihedral angle of 138 measured in the
small-molecule crystal structure of 17 (Figure S14 in Sec-
tion S7.1 in the SI; CCDC code 1816440) and match the results
of the torsion angle scan performed using Gaussian09[17] (ꢄ
1 kcalmolꢁ1, DFT:B3LYP/6-31G(d) level of theory in water with
polarizable continuum solvent model; see Figure S7 in Sec-
tion S3.3 in the SI).
Additionally, the ethyl linkers of the newly introduced sub-
stituents are in close proximity to Asp116 (d(CH2···OAsp116)=3.2–
3.6 ꢁ) in the two active sites of both complexes with 18 and
19, possibly establishing CꢁH···O interactions (Figures 5b and
6, and Figures S10a and S11 in Sections S6.4 and S6.5 in the
SI). This new interaction might explain the improved inhibition
of T. brucei TR observed for 18 and 19 relative to 3.
The flexibility of the newly introduced substituents com-
pared with the core of the ligand is highlighted by the differ-
ent conformations observed in the co-crystal structures
(Figure 7). In the electron density maps drawn at a lower cut-
off level (0.5s), all the atoms of the inhibitor are covered,
whereas at a higher level (1s) only the core is contained in the
map (Figures 5b and 6). Accordingly, the B-factors for the
ligand atoms of the TR complex with 19 are significantly
higher for the ethylpiperidine substituent relative to the rest of
the molecule (Figure S12 in Section S6.6 in the SI).
Figure 5. Co-crystal structure of 18 in complex with T. brucei TR (PDB ID:
6BTL, 2.8 ꢁ resolution). a) Overlay of the binding modes of inhibitors 3
(yellow structure, PDB ID: 4NEV, monomer A)[7] and 18 (orange structure,
PDB ID: 6BTL, monomer B) in T. brucei TR showing the similarity in binding.
b) Binding mode of 18 in T. brucei TR (monomer B). Color code: gray, Cenzyme
yellow, Cligand 3; orange, Cligand 18; blue, N; red, O; yellow, S. The 2FoꢁFc elec-
tron density map is drawn at 1s and 0.5s as black and dark green meshes,
respectively; distances are given in ꢁ.
;
Finally, the overlay of the co-crystal structure of 18 with the
active site of T. cruzi TR suggests, that in T. cruzi TR an addition-
al Coulombic interaction with Glu112, which replaces Gly112 in
T. brucei TR, might be established (d(H2N+···OGlu112)=3.1 ꢁ; Fig-
ure S13 in Section S6.7 in the SI).
Selectivity, antiparasitic activities, and cytotoxicity
Selectivity against human GR (hGR) was studied for the ligands
that showed relevant activity against TR (Table S3 in Sec-
tion S5.3 in the SI). From the proposed binding mode, we ex-
pected the compounds to be inactive against hGR due to the
replacement of four residues involved in the binding of our li-
Figure 6. Co-crystal structure of 19 in complex with T. brucei TR (PDB ID:
6BU7, 2.7 ꢁ resolution, monomer B). Color code: gray, Cenzyme; orange, Cligand
blue, N; red, O; yellow, S. The 2FoꢁFc electron density map is drawn at 1s
and 0.5s as black and dark green meshes, respectively; distances are given
in ꢁ.
;
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
7
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
used ([inhibitor]=40 mm; [glutathione disulfide (GSSG), the
substrate of hGR]=62 mm, corresponding approximately to the
KM;[31] see Section S5.1 in the SI for details). This is a considera-
ble improvement in selectivity because previous BTCP ana-
logues displayed hGR inhibition under identical conditions,
with compound 3 inhibiting 12% of the hGR activity at 40 mm
inhibitor and 60 mm substrate concentrations.[7]
Selectivity against hGR has been previously observed, and
electrostatic complementarity plays an important role in ach-
ieving this important target differentiation.[9,30b,32] The substrate
of TR, TS2 (1), has an overall +1 charge and features electro-
static complementarity to the disulfide substrate binding site
of TR with overall negative electrostatic potential. In contrast,
GSSG is a dianion and its binding site in hGR is characterized
by an overall positive electrostatic potential created by a clus-
ter of arginine residues.[30b] The new ligands 16–19 actually
contain two basic sites, the protonated pyrrolidine and the
protonated/partially protonated heteroalicyclic moiety of the
new substituent (see pKa values in Table S1 in Section S4.1 in
the SI). These potentially doubly positively charged ligands do
not bind into the strongly positively polarized active site of
hGR, as reported for other positively charged TR inhibi-
tors.[9,32b,33]
Figure 7. Overlay of the four ligands observed in the two complexes of 18
and 19 with homodimeric TR, highlighting the different conformations of
the ethylpiperazine/ethylpiperidine moieties (orange) compared with the
core of the inhibitor (gray). a) Overlay by superimposition of the four active
sites. b) Overlay by superimposition of the central thiazole moiety.
gands to T. brucei TR (Glu18, Trp21, Ser109, and Met113) by
other residues in hGR (Ala34, Arg37, Ile113, and Asn117).[30]
Compounds 4, 5, 10, and 11, incorporating a diaryl-substituted
thiophene motif, showed significantly higher inhibition of hGR
than 3. A detailed kinetic analysis of 5 revealed noncompeti-
tive inhibition of hGR with a Ki value of 40 mm. Inhibitors 4, 5,
10, and 11 displayed mixed-mode inhibition of T. brucei TR
(Table 1). Thus, it is possible that they exert a similar mecha-
nism of inhibition on hGR. Ligands 16–19, on the other hand,
showed purely competitive inhibition of TR. According to the
two co-crystal structures, these ligands bind in the region of
the active site where the sequences of TR and hGR differ most.
Gratifyingly, 16–19 did not inhibit hGR under the conditions
The in vitro activities of selected ligands of the three series
were studied against the trypanosomatids T. brucei rhodesiense,
T. cruzi, L. donovani, as well as the malarial parasite Plasmodium
(P.) falciparum, and mammalian L6 cells (Table 3). With the ex-
ception of the oxetanyl ligands 12–15, all derivatives showed
higher cell growth inhibition for T. brucei rhodesiense than for
T. cruzi and L. donovani (Table 3). In general, the oxetane series
12–15 was mostly inactive toward all cell types studied, with
the majority of the IC50 values being higher than 100 mm. The
lower activity against T. cruzi relative to T. brucei rhodesiense
presumably arose from a limited penetration into the host
cells. With the exception of the oxetanes, all of the ligands also
inhibited the proliferation of P. falciparum in the single-digit
Table 3. In vitro activities against T. brucei rhodesiense (T.b.r.), T. cruzi (T.c.), L. donovani (L.d.), P. falciparum NF54 (P.f.), mammalian L6 cells, and selectivity
index (S.I.)=L6/T.b.r.
Compd
IC50 [mm][a]
S.I.
T.b.r.
T.c.
L.d.
P.f.
7.8
0.3
0.6
3.1
8.4
7.9
1.2
0.6
L6
(L6/T.b.r.)
3[7]
4
5
7
8
9
10
11
12
13
14
15
16
17
18
19
3.5
4.0
1.4
3.4
3.5
11.3
1.5
4.0
130
131
98.4
281
3.9
19.0
5.7
7.8
10.3
36.4
55.1
6.5
n.d.
14.5
13.8
20.5
48.8
114
21.5
26.4
145
28.9
5.7
4.0
8
1.4
3
3
9
6.5
3
1.6
1
1
1
1
15
4
5
6
9.5
32.0
74.0
4.6
6.0
6.5
63.6
31.4
52.7
168
124
26.6
95.6
70.4
55.7
137
102
90.2
271
58.0
9.5
9.2
11.6
32.6
80.8
261
>174
151
7.6
0.7
0.4
0.4
2.2
1.9
2.0
>200
107.8
>200
21.0
26.6
[a] Values are the means of two independent assays; n.d. =not determined.
&
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
8
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
micromolar to sub-micromolar range. The trypanothione-based
metabolism is not found in P. falciparum, which indicates that
our class of ligands inhibits additional targets. We had previ-
ously observed particularly high cell-based inhibition of P. falci-
parum.[7,9,34] The structural similarities with the antimalarial
drug chloroquine, namely the extended aromatic system and
the presence of basic secondary and/or tertiary amines, leads
us to propose that our ligands might have a similar mode of
action to the drug. Chloroquine is known to accumulate in the
acidic food vacuole due to its basicity, where it inhibits the for-
mation of hemozoin, leading to oxidative stress and death of
the parasite.[35]
with the goal of reaching sub-micromolar affinities for T. brucei
and T. cruzi TR.
Finally, we note that all known inhibitors of T. brucei and
T. cruzi TR with solved co-crystal structures bind to the mepa-
crine site.[7,10,36] It should be noted, however, that other ligands
displayed alternative binding modes in co-crystal structures
with L. infantum.[37]
Experimental Section
General details and synthetic procedures: The synthesis and
characterization of ligand 18 and intermediates 33 and 37 are re-
ported below. General details and all experimental and analytical
data for other new compounds are described in the Supporting In-
With the exception of the oxetane series, the tested com-
pounds displayed some selectivity against L6 myoblast cells,
with selectivity indices (S.I.=L6/T.b.r.) of up to 15. The relatively
low selectivity might arise due to an alternative target in mam-
malian cells.
1
formation, which also contains atom numberings used to assign H
and 13C NMR signals. Compound 3 was prepared following a previ-
ously reported procedure.[7] The purity of all final compounds was
determined to be ꢀ95% by elemental analysis and/or by high-
pressure liquid chromatography (HPLC). Calculated properties (pKa,
TPSA, and clogD), DFT calculations, and biological assays are re-
ported in the Supporting Information. Crystallographic representa-
tions were prepared using PyMOL.[39]
Conclusions
Using previous structural information of small BTCP analogues
bound to TR, we designed three new series of ligands with the
goals of improving water solubility and potency of this class of
inhibitors. The diaryl-substituted thiophene series (ligands 4–
11), combining the different modes of binding previously ob-
served for 3 with T. brucei and T. cruzi TR, did not lead to the
expected improvement in potency, presumably due to the
high-energy conformation of the ligands required to bind to
the enzyme. Water solubility was successfully improved with
the oxetane series (ligands 12–15) but target affinity under the
applied assay conditions ([inhibitor]=[TS2]=40 mm) was lost.
We explain this by the electron-withdrawing nature of the oxe-
tane O atom, which reduces the basicity and degree of proto-
nation of the pyrrolidine N atom in the b-position. The lack of
protonation eliminates ion pairing with Glu18 and cation–p in-
teractions with the indole ring of Trp21, and weakens the Cꢁ
H···p interactions with this ring due to decreased polarization
of the pyrrolidine CꢁH bonds. The N-substituted indole series
(ligands 16–19 and 31) resulted in improved water solubility
and low micromolar inhibition constants for both T. cruzi and
T. brucei TR. Piperazine 18 is the strongest inhibitor of the
series with a twofold and threefold improvement in the Ki
values for T. cruzi and T. brucei, respectively, relative to com-
pound 3. Furthermore, ligand 18 showed in vitro activities in
the single-digit micromolar range against T. brucei (IC50 =
1.9 mm) and in the low micromolar range against T. cruzi (IC50 =
21.0 mm). The binding mode of the initial hit 3[7] was conserved
in the co-crystal structures of T. brucei TR with 18 and 19. This
is quite remarkable considering the large active site of TR that
allows for different binding modes of small molecules[7] as well
as the binding of multiple ligands within the large cavi-
ty.[36a,37b,c,38] Analysis of the crystal structures showed that new
interactions with the carboxylate of Asp116 are the probable
reason for the higher affinity (low Ki value) for T. brucei TR. By
virtue of the knowledge gained from this structure–activity re-
lationship study and the confirmation of the binding mode, it
should be possible to further improve this class of inhibitors,
1-(4-Methylbenzene-1-sulfonyl)-5-{5-[1-(pyrrolidin-1-yl)cyclohex-
yl]-1,3-thiazol-2-yl}-1H-indole (33): A solution of 3[7] (500 mg,
1.42 mmol) in DMF (5.0 mL) was treated portionwise at 08C with
95% NaH (68 mg, 2.84 mmol), stirred for 30 min, and treated drop-
wise with
a solution of p-toluenesulfonyl chloride (380 mg,
1.99 mmol) in DMF (2.5 mL). The mixture was stirred for 19 h at
258C, diluted with EtOAc (20 mL), saturated aqueous NH4Cl
(10 mL), and water (10 mL). The phases were separated, and the
aqueous phase was extracted with EtOAc (3ꢂ20 mL). The com-
bined organic phases were washed with water (3ꢂ100 mL) and
brine (100 mL), dried over MgSO4, and filtered. Evaporation afford-
ed 33 (688 mg, 96%) as white crystals, which were used without
further purification. Rf =0.21 (hexane/EtOAc/Et3N 60:40:1); m.p. 82–
1
838C; H NMR (400 MHz, CD2Cl2): d=1.37–1.51 (m, 4H; Hax-C(3’,5’),
H2C(4’)), 1.58–1.65 (m, 4H; H2C(3’’,4’’)), 1.66–1.76 (m, 2H; Heq-
C(3’,5’)), 1.92–2.07 (m, 4H; H2C(2’,6’)), 2.34 (s, 3H; CH3), 2.52–2.62
(m, 4H; H2C(2’’,5’’)), 6.73 (dd, J=3.6, 0.8 Hz, 1H; H-C(3’’’)), 7.26
(br.d, J=8.4 Hz, 2H; H-C(3iv,5iv)), 7.55 (s, 1H; H-C(4)), 7.61 (d, J=
3.7 Hz, 1H; H-C(2’’’)), 7.78 (br.d, J=8.4 Hz, 2H; H-C(2iv,6iv)), 7.90 (dd,
J=8.7, 1.8 Hz, 1H; H-C(6’’’)), 8.01 (dt, J=8.7, 0.8 Hz, 1H; H-C(7’’’)),
8.11 ppm (dd, J=1.8, 0.7 Hz, 1H; H-C(4’’’)); 13C NMR (101 MHz,
CD2Cl2): d=21.87 (CH3), 22.77 (C(3’,5’)), 24.16 (C(3’’,4’’)), 26.41
(C(4’)), 38.20 (C(2’,6’)), 45.40 (C(2’’,5’’)), 58.83 (C(1’)), 109.87 (C(3’’’)),
114.36 (C(7’’’)), 119.66 (C(4’’’)), 123.52 (C(6’’’)), 127.31 (C(2iv,6iv)),
128.05 (C(2’’’)), 130.30 (C(5’’)), 130.53 (C(3iv,5iv)), 131.80 (C(3a’’’)),
135.50 (C(1iv)), 135.95 (C(7a’’’)), 140.52 (weak br., C(5)), 141.20 (C(4)),
146.13 (C(4iv)), 166.49 ppm (C(2)); IR (ATR): n˜ =2930, 2854, 1596,
1508, 1493, 1467, 1444, 1425, 1372, 1283, 1239, 1167, 1124, 1092,
1017, 994, 883, 811, 777, 766, 720, 703, 668, 620 cmꢁ1; HRMS (ESI):
m/z (%): calcd for C28H32N3O2S2: 506.1930 [M+H]+; found:
506.1928 (8); calcd for C24H23N2O2S2: 435.1195 [MꢁC4H8N]+; found:
435.1193 (100).
tert-Butyl 4-[2-(5-{5-[1-(pyrrolidin-1-yl)cyclohexyl]-1,3-thiazol-2-
yl}-1H-indol-1-yl)ethyl]piperazine-1-carboxylate (37): A solution
of tert-butyl 4-(2-hydroxyethyl)piperazine-1-carboxylate (36,
364 mg, 1.58 mmol) in toluene (13 mL) under an argon atmosphere
was cooled to 08C, treated with Cs2CO3 (515 mg, 1.58 mmol) and
portionwise with 95% NaH (38 mg, 1.58 mmol), stirred for 30 min
at 258C, then treated with N-tosylated indole 33 (400 mg,
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
9
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
0.79 mmol), and stirred at 1008C for 18 h. The mixture was diluted
with water (10 mL) and the solvent partially evaporated. The result-
ing mixture was diluted with EtOAc (20 mL) and water (20 mL). The
phases were separated, and the aqueous layer was extracted with
EtOAc (3ꢂ20 mL). The combined organic layers were washed with
brine (40 mL), dried over MgSO4, filtered, and evaporated. MPLC
(SiO2; hexane/EtOAc/Et3N gradient 100:0:1!30:70:1) afforded 37
(304 mg, 68%) as a brown oil. Mixed fractions of detosylated start-
ing material 3 and product were obtained: Rf =0.18 (hexane/
EtOAc/Et3N 30:70:1); 1H NMR (400 MHz, CD2Cl2): d=1.44 (s, 9H;
C(CH3)3), 1.44–1.51 (m, 4H; Hax-C(3iv,5iv), H2C(4iv)), 1.64 (br.quintet,
J=3.2 Hz, 4H; H2C(3v,4v)), 1.67–1.77 (m, 2H; Heq-C(3iv,5iv)), 1.98–2.06
(m, 4H; H2C(2iv,6iv)), 2.40 (t, J=5.0 Hz, 4H; H2C(3,5)), 2.57–2.64 (m,
4H; H2C(2v,5v)), 2.76 (t, J=6.5 Hz, 2H; H2C(1’)), 3.34–3.41 (m, 4H;
H2C(2,6)), 4.25 (t, J=6.5 Hz, 2H; H2C(2’)), 6.54 (dd, J=3.1, 0.8 Hz,
1H; H-C(3’’)), 7.23 (d, J=3.2 Hz, 1H; H-C(2’’)), 7.40 (dt, J=8.6,
0.8 Hz, 1H; H-C(7’’)), 7.53 (s, 1H; H-C(4’’’)), 7.81 (dd, J=8.6, 1.7 Hz,
1H; H-C(6’’)), 8.17 ppm (dd, J=1.7, 0.6 Hz, 1H; H-C(4’’)); 13C NMR
(101 MHz, CD2Cl2): d=22.83 (C(3iv,5iv)), 24.18 (C(3v,4v)), 26.47 (C(4iv)),
28.66 (C(CH3)3), 38.26 (C(2iv,6iv)), 43.87 (weak br., C(2,6)), 44.90
(C(2’)), 45.43 (C(2v,5v)), 53.79 (C(3,5)), 58.36 (C(1’)), 58.80 (C(1iv)),
79.79 (C(CH3)3), 102.50 (C(3’’)), 110.27 (C(7’’)), 119.53 (C(4’’)), 120.56
(C(6’’)), 126.51 (C(5’’)), 129.25 (C(3a’’)), 130.03 (C(2’’)), 137.46
(C(7a’’)), 139.19 (C(5’’’)), 140.89 (C(4’’’)), 155.04 (C=O), 168.21 ppm
(C(2’’’)); IR (ATR): n˜ =2932, 2857, 2812, 1688, 1613, 1509, 1453,
1420, 1364, 1339, 1289, 1262, 1244, 1195, 1166, 1123, 1092, 1037,
1003, 899, 865, 827, 798, 761, 718, 653, 617 cmꢁ1; HRMS (ESI): m/z
(%): calcd for C32H46N5O2S: 564.3367 [M+H]+; found: 564.3361
(14); calcd for C28H37N4O2S: 493.2632 [MꢁC4H8N]+; found: 493.2628
(100).
C27H38N5S: 464.2842 [M+H]+; found: 464.2842 (1); calcd for
C23H29N4S: 393.2107 [MꢁC4H8N]+; found: 393.2104 (100).
T. brucei TR co-crystallization, X-ray data collection, structure de-
termination and refinement: The concentration of the proteins
for crystallization was 10 mgmLꢁ1 in Tris buffer (20 mm, pH 8). In-
hibitors 18 and 19 were dissolved in DMSO (10 mm) and added in
1 mL aliquots to the protein solution (95 mL), resulting in a final
concentration of 0.5 mm. Adding the inhibitor caused a small
amount of precipitate to form, which was removed by centrifuga-
tion. Both protein–inhibitor complexes were crystallized by the
hanging drop vapor diffusion method, mixing protein (2 mL) with
well solution (2 mL, 0.1m HEPES, pH 7.5, 2.0m (NH4)2SO4). The crys-
tals were dipped in well solution plus 20% glycerol and flash-
frozen at ꢁ1808C for data collection. Diffraction data for the
T. brucei TR complexes were collected at beamline 23-ID-D of the
Advanced Photon Source Facility, Argonne National Laboratories
(Lemont, IL) and reduced using the software packages XDS[40] and
HKL2000.[41] The structures were determined by molecular replace-
ment (Phaser)[42] using TR from T. brucei (PDB ID: 2WOI)[36b] as the
search model. The atomic models were refined using Phenix.[43]
Data collection and refinement statistics are listed in Table S4 in
Section S6.1 in the SI. Atomic coordinates and structure factors
were deposited in the PDB with the accession codes 6BTL and
6BU7 for the T. brucei TR complexes with 18 and 19, respectively.
Acknowledgements
The authors are grateful to the ETH Research Council (ETH-01 13-
2) for the generous support of the work done at the ETH. Re-
search at the University of Heidelberg was supported by the
Deutsche Forschungsgemeinschaft (DFG KR 1242/8-1) and at the
University of Toronto (UofT) by the National Science and Engi-
neering Research Council of Canada (RGPIN-2015–04877) and the
Canadian Research Chairs Program. We are thankful to Dr. Bruno
Bernet (ETH) for proofreading the Experimental Section, Michael
Solar (ETH) and Dr. Nils Trapp (ETH) for recording and determin-
ing the crystal structure of 17, and Cꢂdric Schaack (ETH) as well
as Geoffrey Schwertz (ETH) for proofreading the entire manu-
script and helpful discussions. We thank Natalie Dirdjaja (BZH)
for the preparation of trypanothione disulfide and the enzymes,
and Alejandro E. Leroux (BZH) for support during the kinetic
analyses and valuable discussions. We thank Monica Cal (Swiss
TPH), Romina Rocchetti (Swiss TPH) and Sonja Keller-Mꢁrki (Swiss
TPH) for the parasite assay results. We acknowledge the help of
Annie Cunningham (UofT) and Bryan Eger (UofT) with protein
crystallization and data collection and reduction, respectively.
Part of the research was done using beamline 23-ID-D at the Ad-
vanced Photon Source, a US Department of Energy (DOE) Office
of Science User Facility operated for the DOE Office of Science by
Argonne National Laboratory under Contract No. DE-AC02-
06CH11357. GM/CA@APS has been funded in whole or in part by
the National Cancer Institute (ACB-12002) and the National Insti-
tute of General Medical Sciences (AGM-12006).
1-[2-(Piperazin-1-yl)ethyl]-5-{5-[1-(pyrrolidin-1-yl)cyclohexyl]-1,3-
thiazol-2-yl}-1H-indole (18): A solution of 37 (90 mg, 0.16 mmol)
in CH2Cl2 (3 mL) was treated at 08C with 2,6-lutidine (0.37 mL,
3.19 mmol), and dropwise with tert-butyldimethylsilyl trifluorome-
thanesulfonate (0.18 mL, 0.80 mmol), stirred for 1 h at 258C, then
treated with saturated aqueous NH4Cl solution (10 mL). The phases
were separated, and the aqueous layer was extracted with CH2Cl2
(3ꢂ10 mL). The combined organic layers were dried over MgSO4,
filtered, and evaporated. A solution of the residue in THF/MeOH
(2:5, 6 mL) was treated with aqueous K2CO3 (1m, 1.80 mL), stirred
for 1 h, then treated with saturated aqueous NH4Cl solution
(10 mL), and extracted with CH2Cl2 (4ꢂ20 mL). The combined or-
ganic layers were washed with brine (30 mL), dried over MgSO4, fil-
tered, and evaporated. Short flash chromatography (SiO2; CH2Cl2/
MeOH gradient 95:5!80:20), followed by recrystallization from
CH2Cl2/cyclohexane afforded 18 (33 mg, 31%) as an off-white
powder. Rf =0.02–0.14
(CH2Cl2/MeOH
90:10);
m.p. 1168C
(decomp.); 1H NMR (400 MHz, CD2Cl2): d=1.35–1.52 (m, 4H; Hax-
C(3iv,5iv), H2C(4iv)), 1.62–1.81 (m, 6H; Heq-C(3iv,5iv), H2C(3v,4v)), 2.06–
2.21 (m, 4H; H2C(2iv,6iv)), 2.53–2.64 (m, 4H; H2C(2’’’,6’’’)), 2.72–2.86
(m, 6H; H2C(2’’), H2C(2v,5v)), 2.90–3.06 (m, 4H; H2C(3’’’,5’’’)), 4.24 (t,
J=6.3 Hz, 2H; H2C(1’’)), 6.55 (d, J=3.2 Hz, 1H; H-C(3’)), 7.22 (d, J=
3.2 Hz, 1H; H-C(2’)), 7.40 (d, J=8.7 Hz, 1H; H-C(7’)), 7.60 (s, 1H; H-
C(4)), 7.81 (dd, J=8.6, 1.8 Hz, 1H; H-C(6’)), 8.18 ppm (d, J=1.7 Hz,
1H; H-C(4’)); 13C NMR (101 MHz, CD2Cl2): d=22.88 (C(3iv,5iv)), 24.04
(C(3v,4v)), 26.02 (C(4iv)), 37.55 (C(2iv,6iv)), 44.89 (C(1’’)), 45.08
(C(2’’’,6’’’)), 46.10 (C(2v,5v)), 52.52 (C(3’’’,5’’’)), 58.29 (C(2’’)), 60.94
(weak br., C(1iv)), 102.68 (C(3’)), 110.31 (C(7’)), 119.76 (C(4’)), 120.63
(C(6’)), 126.20 (C(5’)), 129.28 (C(3a’)), 130.14 (C(2’)), 136.81 (weak
br., C(5)), 137.58 (C(7a’)), 142.35 (br., C(4)), 169.52 ppm (weak, C(2));
IR (ATR): n˜ =3364, 2929, 2852, 2475, 1612, 1510, 1488, 1449, 1397,
1339, 1281, 1261, 1201, 1134, 1091, 1048, 1029, 999, 883, 854, 803,
760, 719, 653, 637, 618 cmꢁ1; HRMS (MALDI–ESI): m/z (%): calcd for
Conflict of interest
The authors declare no conflict of interest.
&
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
10
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[21] R. Pereira, B. Iglesias, A. R. de Lera, Tetrahedron 2001, 57, 7871–7881.
[22] a) S. Gronowitz, N. Gjçs, R. M. Kellogg, H. Wynberg, J. Org. Chem. 1967,
32, 463–464; b) X.-H. Xu, A. Azuma, A. Kusuda, E. Tokunaga, N. Shibata,
Eur. J. Org. Chem. 2012, 1504–1508.
[23] P. J. Hamzik, J. D. Brubaker, Org. Lett. 2010, 12, 1116–1119.
[24] a) M. A. Eissenstat, J. D. Weaver III, Tetrahedron Lett. 1995, 36, 2029–
2032; b) S. B. Sobolov, J. Sun, B. A. Cooper, Tetrahedron Lett. 1998, 39,
5685–5688; c) J. R. Denton, Synthesis 2010, 775–782.
[25] M. C. Jockers-Scherꢃbl, R. H. Schirmer, R. L. Krauth-Siegel, Eur. J. Bio-
chem. 1989, 180, 267–272.
[26] F. Lizzi, G. Veronesi, F. Belluti, C. Bergamini, A. Lꢄpez-Sꢅnchez, M. Kaiser,
R. Brun, R. L. Krauth-Siegel, D. G. Hall, L. Rivas, M. L. Bolognesi, J. Med.
Chem. 2012, 55, 10490–10500.
[27] M. Morgenthaler, E. Schweizer, A. Hoffmann-Rçder, F. Benini, R. E.
Martin, G. Jaeschke, B. Wagner, H. Fischer, S. Bendels, D. Zimmerli, J.
Schneider, F. Diederich, M. Kansy, K. Mꢃller, ChemMedChem 2007, 2,
1100–1115.
[28] D. A. Dougherty, Acc. Chem. Res. 2013, 46, 885–893.
[29] F. Zenone, J. M. Kamenka, Eur. J. Med. Chem. 1991, 26, 677–686.
[30] a) T. Aboagye-Kwarteng, K. Smith, A. H. Fairlamb, Mol. Microbiol. 1992,
6, 3089–3099; b) C. S. Bond, Y. Zhang, M. Berriman, M. L. Cunningham,
A. H. Fairlamb, W. N. Hunter, Structure 1999, 7, 81–89.
[31] A. Nordhoff, U. S. Bꢃcheler, D. Werner, R. H. Schirmer, Biochemistry 1993,
32, 4060–4066.
[32] a) G. B. Henderson, A. H. Fairlamb, P. Ulrich, A. Cerami, Biochemistry
1987, 26, 3023–3027; b) C. H. Faerman, S. N. Savvides, C. Strickland,
M. A. Breidenbach, J. A. Ponasik, B. Ganem, D. Ripoll, R. L. Krauth-Siegel,
P. A. Karplus, Bioorg. Med. Chem. 1996, 4, 1247–1253.
[33] R. Fernandez-Gomez, M. Moutiez, M. Aumercier, G. Bethegnies, M.
Luyckx, A. Ouaissi, A. Tartar, C. Sergheraert, Int. J. Antimicrob. Agents
1995, 6, 111 – 118.
[34] a) B. Stump, C. Eberle, M. Kaiser, R. Brun, R. L. Krauth-Siegel, F. Diederich,
Org. Biomol. Chem. 2008, 6, 3935–3947; b) B. Stump, C. Eberle, W. B.
Schweizer, M. Kaiser, R. Brun, R. L. Krauth-Siegel, D. Lentz, F. Diederich,
ChemBioChem 2009, 10, 79–83.
[35] a) H. Ginsburg, E. Nissani, M. Krugliak, Biochem. Pharmacol. 1989, 38,
2645–2654; b) S. Kumar, M. Guha, V. Choubey, P. Maity, U. Bandyopad-
hyay, Life Sci. 2007, 80, 813–828.
[36] a) A. Saravanamuthu, T. J. Vickers, C. S. Bond, M. R. Peterson, W. N.
Hunter, A. H. Fairlamb, J. Biol. Chem. 2004, 279, 29493–29500; b) S. Pat-
terson, M. S. Alphey, D. C. Jones, E. J. Shanks, I. P. Street, J. A. Frearson,
P. G. Wyatt, I. H. Gilbert, A. H. Fairlamb, J. Med. Chem. 2011, 54, 6514–
6530.
[37] a) A. Ilari, P. Baiocco, L. Messori, A. Fiorillo, A. Boffi, M. Gramiccia, T. Di
Muccio, G. Colotti, Amino Acids 2012, 42, 803–811; b) P. Baiocco, G.
Poce, S. Alfonso, M. Cocozza, G. C. Porretta, G. Colotti, M. Biava, F.
Moraca, M. Botta, V. Yardley, A. Fiorillo, A. Lantella, F. Malatesta, A. Ilari,
ChemMedChem 2013, 8, 1175–1183; c) F. Saccoliti, G. Angiulli, G. Pupo,
L. Pescatori, V. N. Madia, A. Messore, G. Colotti, A. Fiorillo, L. Scipione,
M. Gramiccia, T. Di Muccio, R. Di Santo, R. Costi, A. Ilari, J. Enzyme Inhib.
Med. Chem. 2017, 32, 304–310.
[38] S. Bonse, C. Santelli-Rouvier, J. Barbe, R. L. Krauth-Siegel, J. Med. Chem.
1999, 42, 5448–5454.
[39] The PyMOL Molecular Graphics System, Version 1.7.6.5, Schrçdinger LLC,
2015.
Keywords: antiprotozoal agents · molecular recognition ·
neglected diseases · trypanothione reductase · X-ray co-crystal
structures
[1] a) K. Stuart, R. Brun, S. Croft, A. Fairlamb, R. E. Gꢃrtler, J. McKerrow, S.
Reed, R. Tarleton, J. Clin. Invest. 2008, 118, 1301–1310; b) Fourth WHO
report on neglected tropical disease, World Health Organization, 2017,
ISBN 978-92-4-156544-8; c) T. Vos, et al., Lancet 2016, 388, 1545–1602.
[2] M. P. Barrett, S. L. Croft, Br. Med. Bull. 2012, 104, 175–196.
[3] a) G. Eperon, M. Balasegaram, J. Potet, C. Mowbray, O. Valverde, F. Chap-
puis, Expert Rev. Anti-Infect. Ther. 2014, 12, 1407–1417; b) M. C. Field, D.
Horn, A. H. Fairlamb, M. A. J. Ferguson, D. W. Gray, K. D. Read, M. De
Rycker, L. S. Torrie, P. G. Wyatt, S. Wyllie, I. H. Gilbert, Nat. Rev. Microbiol.
2017, 15, 217–231.
[4] a) A. H. Fairlamb, A. Cerami, Annu. Rev. Microbiol. 1992, 46, 695–729;
b) R. L. Krauth-Siegel, H. Bauer, R. H. Schirmer, Angew. Chem. Int. Ed.
2005, 44, 690–715; Angew. Chem. 2005, 117, 698–724.
[5] a) L. S. C. Bernardes, C. L. Zani, I. Carvalho, Curr. Med. Chem. 2013, 20,
2673–2696; b) A. E. Leroux, R. L. Krauth-Siegel, Mol. Biochem. Parasitol.
2016, 206, 67–74.
[6] S. Krieger, W. Schwarz, M. R. Ariyanayagam, A. H. Fairlamb, R. L. Krauth-
Siegel, C. Clayton, Mol. Microbiol. 2000, 35, 542–552.
[7] E. Persch, S. Bryson, N. K. Todoroff, C. Eberle, J. Thelemann, N. Dirdjaja,
M. Kaiser, M. Weber, H. Derbani, R. Brun, G. Schneider, E. F. Pai, R. L.
Krauth-Siegel, F. Diederich, ChemMedChem 2014, 9, 1880–1891.
[8] S. Patterson, D. C. Jones, E. J. Shanks, J. A. Frearson, I. H. Gilbert, P. G.
Wyatt, A. H. Fairlamb, ChemMedChem 2009, 4, 1341–1353.
[9] For previously reported TR ligands, see; a) C. Eberle, J. A. Burkhard, B.
Stump, M. Kaiser, R. Brun, R. L. Krauth-Siegel, F. Diederich, ChemMed-
Chem 2009, 4, 2034–2044; b) C. Eberle, B. S. Lauber, D. Fankhauser, M.
Kaiser, R. Brun, R. L. Krauth-Siegel, F. Diederich, ChemMedChem 2011, 6,
292–301.
[10] E. M. Jacoby, I. Schlichting, C. B. Lantwin, W. Kabsch, R. L. Krauth-Siegel,
Proteins Struct. Funct. Genet. 1996, 24, 73–80.
[11] E. Persch, O. Dumele, F. Diederich, Angew. Chem. Int. Ed. 2015, 54,
3290–3327; Angew. Chem. 2015, 127, 3341–3382.
[12] a) L. M. Salonen, M. Ellermann, F. Diederich, Angew. Chem. Int. Ed. 2011,
50, 4808–4842; Angew. Chem. 2011, 123, 4908–4944; b) B. R. Beno, K.-
S. Yeung, M. D. Bartberger, L. D. Pennington, N. A. Meanwell, J. Med.
Chem. 2015, 58, 4383–4438.
[13] a) M. Harder, B. Kuhn, F. Diederich, ChemMedChem 2013, 8, 397–404;
b) M. Giroud, J. Ivkovic, M. Martignoni, M. Fleuti, N. Trapp, W. Haap, A.
Kuglstatter, J. Benz, B. Kuhn, T. Schirmeister, F. Diederich, ChemMedChem
2017, 12, 257–270.
[14] To simplify the comparison, T. brucei TR residue numbering is used also
for T. cruzi TR throughout the manuscript.
[15] P. R. Gerber, K. Mꢃller, J. Comput.-Aided Mol. Des. 1995, 9, 251–268.
[16] G. Wuitschik, E. M. Carreira, B. Wagner, H. Fischer, I. Parrilla, F. Schuler,
M. Rogers-Evans, K. Mꢃller, J. Med. Chem. 2010, 53, 3227–3246.
[17] Gaussian09 (Revision A.02), M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Pe-
tersson, H. Nakatsuji, X. Li, M. Caricato, A. Marenich, J. Bloino, B. G. Jane-
sko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov,
J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J.
Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzew-
ski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F.
Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov,
T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C.
Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C.
Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas,
J. B. Foresman, D. J. Fox, Gaussian, Inc., Wallingford CT, 2016.
[18] P. Ertl, B. Rohde, P. Selzer, J. Med. Chem. 2000, 43, 3714–3717.
[19] K. M. Mahar Doan, J. E. Humphreys, L. O. Webster, S. A. Wring, L. J.
Shampine, C. J. Serabjit-Singh, K. K. Adkison, J. W. Polli, J. Pharmacol.
Exp. Ther. 2002, 303, 1029–1037.
[40] W. Kabsch, Acta Crystallogr. Sect. D 2010, 66, 125–132.
[41] Z. Otwinowski, W. Minor in Methods in Enzymology, Vol. 276 (Eds.: C. W.
Carter, Jr., R. M. Sweet), Academic Press, New York, 1997, pp. 307–326.
[42] A. J. McCoy, R. W. Grosse-Kunstleve, P. D. Adams, M. D. Winn, L. C. Storo-
ni, R. J. Read, J. Appl. Crystallogr. 2007, 40, 658–674.
[43] P. D. Adams, P. V. Afonine, G. Bunkꢄczi, V. B. Chen, I. W. Davis, N. Echols,
J. J. Headd, L.-W. Hung, G. J. Kapral, R. W. Grosse-Kunstleve, A. J. McCoy,
N. W. Moriarty, R. Oeffner, R. J. Read, D. C. Richardson, J. S. Richardson,
T. C. Terwilliger, P. H. Zwart, Acta Crystallogr. Sect. D 2010, 66, 213–221.
Manuscript received: January 31, 2018
[20] a) A. R. Katritzky, Z. Najzarek, Z. Dega-Szafran, Synthesis 1989, 66–69;
b) A. R. Katritzky, H. Yang, S. K. Singh, J. Org. Chem. 2005, 70, 286–290.
Accepted manuscript online: && &&, 0000
Version of record online: && &&, 0000
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
11
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
R. De Gasparo, E. Brodbeck-Persch,
S. Bryson, N. B. Hentzen, M. Kaiser,*
E. F. Pai,* R. L. Krauth-Siegel,*
F. Diederich*
Improve and preserve: Three new
ligand series based on 1-[1-(benzo[b]-
thien-2-yl)cyclohexyl]piperidine, a
known trypanothione reductase (TR) in-
hibitor scaffold, were developed
&& – &&
through structure-based design to im-
prove the water solubility and affinity of
this class of compounds. Their biological
activity was investigated in target- and
cell-based assays. Two X-ray co-crystal
structures confirmed the predicted
binding mode, with ligand anchoring at
the mepacrine site of the large, solvent-
exposed active site.
Biological Evaluation and X-ray Co-
crystal Structures of
Cyclohexylpyrrolidine Ligands for
Trypanothione Reductase, an Enzyme
from the Redox Metabolism of
Trypanosoma
&
ChemMedChem 2018, 13, 1 – 12
www.chemmedchem.org
12
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!