Parasite-based screening and proteome profiling reveal orlistat, an FDA-approved drug, as a potential anti Trypanosoma brucei agent

Article abstract of DOI:10.1002/chem.201200482

Trypanosoma brucei is a parasite that causes African sleeping sickness in humans and nagana in livestock and is transmitted by the tsetse fly. There is an urgent need for the development of new drugs against African trypanosomiasis due to the lack of vaccines and effective drugs. Orlistat (also called tetrahydrolipstatin or THL) is an FDA-approved antiobesity drug targeting primarily the pancreatic and gastric lipases within the gastrointestinal tract. It shows potential activities against tumors, mycobacteria, and parasites. Herein, we report the synthesis and evaluation of an expanded set of orlistat-like compounds, some of which showed highly potent trypanocidal activities in both the bloodstream form (BSF) and the procyclic form (PCF) of T. brucei. Subsequent in situ parasite-based proteome profiling was carried out to elucidate potential cellular targets of the drug in both forms. Some newly identified targets were further validated by the labeling of recombinantly expressed enzymes in Escherichia coli lysates. Bioimaging experiments with a selected compound were carried out to study the cellular uptake of the drug in T. brucei. Results indicated that orlistat is much more efficiently taken up by the BSF than the PCF of T. brucei and has clear effects on the morphology of mitochondria, glycosomes, and the endoplasmic reticulum in both BSF and PCF cells. These results support specific effects of orlistat on these organelles and correlate well with our in situ proteome profiling. Given the economic challenges of de novo drug development for neglected diseases, we hope that our findings will stimulate further research towards the conversion of orlistat-like compounds into new trypanocidal drugs. Go fishing for parasite targets: Orlistat-like small-molecule probes have been tested to show potent trypanocidal activities against Trypanosoma brucei. Subsequent in situ proteomic profiling has identified and validated selected cellular targets (see figure). Given the economic challenges of de novo drug development for neglected diseases, these findings indicate orlistat should be considered further as a new trypanocidal drug. Copyright

Full text of DOI:10.1002/chem.201200482

FULL PAPER
DOI: 10.1002/chem.201200482
Parasite-Based Screening and Proteome Profiling Reveal Orlistat, an
FDA-Approved Drug, as a Potential Anti Trypanosoma brucei Agent[**]
Peng-Yu Yang,[a] Min Wang,[b] Kai Liu,[b] Mun Hong Ngai,[a] Omar Sheriff,[b]
Martin J. Lear,[a] Siu Kwan Sze,[c] Cynthia Y. He,*[b] and Shao Q. Yao*[a]
Abstract: Trypanosoma brucei is a para-
site that causes African sleeping sick-
ness in humans and nagana in livestock
and is transmitted by the tsetse fly.
There is an urgent need for the devel-
opment of new drugs against African
trypanosomiasis due to the lack of vac-
cines and effective drugs. Orlistat (also
called tetrahydrolipstatin or THL) is
an FDA-approved antiobesity drug tar-
geting primarily the pancreatic and gas-
tric lipases within the gastrointestinal
tract. It shows potential activities
against tumors, mycobacteria, and par-
asites. Herein, we report the synthesis
and evaluation of an expanded set of
orlistat-like compounds, some of which
showed highly potent trypanocidal ac-
tivities in both the bloodstream form
(BSF) and the procyclic form (PCF) of
T. brucei. Subsequent in situ parasite-
based proteome profiling was carried
out to elucidate potential cellular tar-
gets of the drug in both forms. Some
newly identified targets were further
validated by the labeling of recombi-
nantly expressed enzymes in Escheri-
chia coli lysates. Bioimaging experi-
ments with a selected compound were
carried out to study the cellular uptake
of the drug in T. brucei. Results indicat-
ed that orlistat is much more efficiently
taken up by the BSF than the PCF of
T. brucei and has clear effects on the
morphology of mitochondria, glyco-
somes, and the endoplasmic reticulum
in both BSF and PCF cells. These re-
sults support specific effects of orlistat
on these organelles and correlate well
with our in situ proteome profiling.
Given the economic challenges of de
novo drug development for neglected
diseases, we hope that our findings will
stimulate further research towards the
conversion of orlistat-like compounds
into new trypanocidal drugs.
Keywords: antiparasite agents
·
drug discovery · orlistat · proteo-
mics · trypanosomiasis
Introduction
terized proliferative stages: the bloodstream form (BSF) in
the blood and tissue fluids of the mammal hosts and the
procyclic form (PCF) in the midgut of the insect vector, the
tsetse fly. For the parasite to survive in different hosts, con-
siderable adaptations are required in the parasite morpholo-
gy, surface composition, and metabolism, including lipid and
energy metabolism. Limited therapeutic compounds are
available for human and animal trypanosomiasis, as well as
diseases caused by other related parasites (for example,
Chagas disease caused by T. cruzi and leishmaniasis caused
by Leishmania spp.). Most of the drugs currently used are
decades old.[1–3] Resistance and toxicity to current therapies
make treatment increasingly problematic, and thus the de-
velopment of new drugs is an important priority. One rapid
and cost-effective strategy for the discovery of new trypano-
cidal drugs is to explore existing drugs that possess well-
characterized pharmacokinetic and safety profiles and that
are already approved for other uses.[4] This “piggy-backing”
approach has historically been proven to be successful in
bringing new therapies to the antiparasite drug discovery
pipeline. A recent example is miltefosine, initially developed
for breast cancer but now used for treating visceral leishma-
niasis.[5] Orlistat (Figure 1; also called tetrahydrolipstatin
and referred to as THL hereafter), a derivative of the natu-
ral product lipstatin, is a US Food and Drug Administration
(FDA) approved antiobesity drug targeting primarily the
pancreatic and gastric lipases within the gastrointestinal
Human African trypanosomiasis (HAT; commonly known
as sleeping sickness), caused by the protozoan parasite Try-
panosoma brucei, is responsible for 50000–70000 deaths
each year within sub-Saharan Africa.[1] Recently, due to im-
migrant carriers or infections from contaminated transfu-
sions and organ transplants, the disease has been reported in
areas where it normally does not exist. The trypanosomes
undergo a complex life cycle that includes two well-charac-
[a] P.-Y. Yang, M. H. Ngai, Dr. M. J. Lear, Prof. Dr. S. Q. Yao
Department of Chemistry
National University of Singapore
3 Science Drive 3, Singapore 117543 (Singapore)
Fax : (+65)6779-1691
E-mail: chmyaosq@nus.edu.sg
[b] M. Wang, K. Liu, O. Sheriff, Dr. C. Y. He
Department of Biological Sciences
National University of Singapore
14 Science Drive 4, Singapore 117543 (Singapore)
E-mail: dbshyc@nus.edu.sg
[c] Dr. S. K. Sze
School of Biological Sciences
Nanyang Technological University
60 Nanyang Drive, Singapore 637551 (Singapore)
[**] FDA: US Food and Drug Administration.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201200482.
Chem. Eur. J. 2012, 18, 8403 – 8413
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8403

Figure 1. Chemical structures of the compounds used in this study and an overall workflow of our parasite-based screening, proteome profiling, and
target identification by using THL-like probes. a) Chemical structures of THL, lipstatin (a natural product from which THL was derived), and three
well-known antiparasitic compounds. Cbz: benzyloxycarbonyl. b) Synthetic scheme for the THL-like probes (1a–1g) used in this study. TBS: tert-butyldi-
methylsilyl; TMAL: tandem Mukaiyama aldol lactonization; AA: amino acid; DIAD: diisopropylazodicarboxylate; THF: tetrahydrofuran. c) Overall
workflow of parasite-based screening, proteome profiling, bioimaging, and pull-down/LC-MS/MS analysis for target identification. Reaction of the
alkyne handle on the probe with rhodamine–azide through click chemistry allowed visualization of target proteins by in-gel fluorescence screening and
fluorescence microscopy. Target identification was done by pull-down/LC-MS/MS experiments against the probe-labeled parasite proteome “clicked”
with biotin–azide.[26] SDS: sodium dodecylsulfate.
tract.[6] Its potential activities against tumors (targeting fatty
acid synthase or FAS[7–10]), mycobacteria, and parasites have
recently been reported.[11–14] Studies have not been carried
out with this drug, however, to understand the kinetics and
molecular basis of its trypanocidal activities or to obtain its
detailed cellular and distribution profiles in live parasites.
Inspired by the concept of activity-based protein profiling
(ABPP),[15–25] we recently made use of natural-product-like
small-molecule probes to successfully identify putative cellu-
lar targets of THL in human liver cancer cells.[26–28] In this
approach, a terminal alkyne (see 1a in Figure 1) was intro-
duced into the parental THL structure to provide the neces-
sary functionality for target identification through CuI-cata-
lyzed 1,3-dipolar cycloaddition (click chemistry),[29,30] with-
out interfering with the native biological properties of the
drug. Herein, we report the chemical synthesis of new THL-
like probes, as well as the first in situ proteome-wide profil-
ing of these probes in BSF and PCF parasites. We found
that the THL-like compounds consistently inhibited BSF
proliferation more efficiently than PCF proliferation. Subse-
quent in situ profiling identified both common and unique
putative THL-binding targets in BSF and PCF parasites,
some of which were further validated by using in vitro label-
ing assays. Cellular imaging with selected THL probes in
8404
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8403 – 8413

Screening and Proteome Profiling of a Potential Anti Trypanosoma brucei Agent
FULL PAPER
combination with bioorthogonal click chemistry and azide-
containing fluorophores enabled us to visually monitor cel-
lular uptake of these probes, as well as parasite-morphology
changes caused by the drug treatments. From these studies,
we concluded that THL and THL-like compounds had spe-
cific, strong antiparasitic effects on T. brucei BSF cells,
partly due to the efficient uptake of the drugs by these para-
sites. With the already well-established pharmacokinetic and
safety profiles of THL as an FDA-approved drug, we con-
cluded that THL should be seriously considered as a strong
candidate for further antiparasite drug development.
Further screening assays were carried out against both BSF
and PCF parasites with 1a–1g and with the key intermedi-
ate 2 (identical to 1a except that the N-formylated amino
ester moiety was removed). Three known trypanocidal com-
pounds were used as positive controls: Cbz-Phe-Ala-CHN2
(abbreviated as DMK) and N-Mpip-Phe-Hph-VSPh (also
known as K11777; N-Mpip: N-methylpiperazine; Hph: ho-
mophenylalanine; VSPh: phenyl vinylsulfone), which are
cysteine protease inhibitors, and the broad-spectrum lipase
inhibitor methoxyarachidonyl fluorophosphate (MAFP; Fig-
ure 1a).[31–35] Both DMK and K11777 are irreversible inhibi-
tors of parasite cathepsin L and cathepsin B like proteases
and have shown trypanocidal activities in the nanomolar/mi-
cromolar range.[31–34] Notably, K11777 has passed safety stud-
ies in rodents, dogs, and primates for the treatment of
Chagas disease caused by T. cruzi. It is one of the most ad-
vanced trypanocidal agents known to-date and is currently
in late-stage preclinical development.[36] Dimethylsulfoxide
(DMSO) was included in the assays as a negative control.
All trypanocidal assays were initially carried out in triplicate
with a single concentration of the drug compounds (that is,
125 nm against BSF parasites and 5000 nm against PCF para-
sites). Selected compounds were subsequently used for de-
tailed dose-dependent studies. As shown in Figure 2a, THL
and analogues 1a, 1d, and 1e (highlighted in red) were sig-
nificantly more potent than all three positive controls, espe-
cially against BSF T. brucei. It is also interesting to note that
THL and its analogues may have other cellular targets in
addition to parasite lipases, because MAFP (a broad-spec-
trum lipase inhibitor) displayed one of the lowest trypanoci-
dal activities amongst all the compounds tested. The N-for-
mylated amino ester moiety appeared to be essential for try-
panocidal activity, as demonstrated by the lack of activity of
the hydroxy lactone intermediate 2. Similarly, a drop in the
trypanocidal activity (by approximately threefold) was ob-
served if the amino acid group in this moiety was changed
from glycine to b-glycine (compare compounds 1c and 1 f).
The stereochemistry of the Ca position in the N-formylated
amino ester moiety of THL seemed to be important for the
activity, because 1a was shown to be much more active than
the corresponding epimer 1g. When the trypanocidal activi-
ties of all compounds were compared in both the BSF and
PCF of T. brucei, it became evident that these drugs were
much more effective in killing the BSF parasites than the
PCF parasites (Figure 2a); a significantly higher dosage of
the compounds (5000 nm) was needed in order to achieve
comparable killing effects in the PCF parasites. Further
dose-dependent studies were carried out with selected com-
pounds (THL and 1a) against both the BSF and PCF para-
sites (Figure 2b). The results again showed that the prolifer-
ation of both life-cycle stages of the trypanosomes was effi-
ciently inhibited by both THL (ED50 =(49.1Æ1.7) nm for
BSF and (1.55Æ3.65) mm for PCF) and 1a (ED50 =(46.2Æ
3.7) nm for BSF and (1.45Æ3.56) mm for PCF). The effective
trypanocidal profile immediately places THL as one of the
most potent drugs known against T. brucei. This, together
with its already well-characterized pharmacokinetic and
Results and Discussion
Synthesis of THL-like probes 1a–1g: The THL-like probes
(1a–1g) used in the current study are shown in Figure 1. 1a
(also named as THL-R) is a previously reported com-
pound.[26] It is structurally identical to THL except for a ter-
À
ꢀ
minal C C to C C bond replacement in the right-hand ali-
phatic chain. According to our previous studies, this modifi-
cation preserves the compoundꢁs biological activities whilst
allowing subsequent in situ proteome profiling and bioimag-
ing experiments.[26,27] Compounds 1b–1g retain the key
structural features of 1a by keeping the two key aliphatic
chains intact, so as to maximize the potential lipase- and
synthase-binding properties. The N-formylated amino ester
moiety in these compounds, which is derived from l-leucine
in 1a (and THL), was strategically substituted with different
natural and unnatural amino acids in order to evaluate their
ability to discriminate different cellular targets in the para-
sites. Similarly to 1a, compounds 1b–1g were conveniently
synthesized from the key intermediate 2 with various N-for-
mylated amino acids under Mitsunobu conditions; these re-
actions gave rise to the corresponding configurationally in-
verted products in a single step with good yields (46–89%).
Trypanocidal activities of THL analogues in T. brucei: With
compounds 1a–1g synthesized, we set out to address two
key questions: 1) How does the introduction of a terminal
alkyne handle or a variation at the amino ester position of
the THL scaffold affect its trypanocidal activities? 2) What
might be the potential cellular targets of THL in the PCF
and BSF trypanosomes? We first measured the antiparasite
activity of 1a and THL against the BSF of T. brucei, which
is the mammal-infecting form. All growth/inhibition studies
were performed by using the viacount assay on a Guava
Easycyte system (Millipore, USA), which allowed automat-
ed, quantitative analysis of the killing kinetics of these com-
pounds (see Figure S2 in the Supporting Information). Both
compounds killed cultured parasites with comparable poten-
cy in a dose-dependent manner (that is, approximately 80%
killing against the BSF at 125 nm). These results confirmed
that the replacement of a terminal ethyl group by an ethynyl
group was imperceptible and that the potential cellular tar-
gets identified from our subsequent in situ proteomic profil-
ing with 1a were likely targets of THL itself (see below).
Chem. Eur. J. 2012, 18, 8403 – 8413
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8405

C. Y. He, S. Q. Yao et al.
taining THL-like analogues. We
had previously shown that the
introduction of an alkyne
handle in THL makes it possi-
ble to perform in situ pro-
teome-wide profiling and iden-
tification of potential drug-
binding targets through down-
stream conjugation of the pro-
tein/drug complex to reporter
tags.[26] Although 1a, 1d, and
1e all showed similar trypanoci-
dal
profiles
against
BSF
T. brucei, we chose 1a for the
subsequent target identification
because this compound is the
closest structural mimic to
THL. We first compared the in
situ proteome reactivity profiles
of all of the THL probes, as
well as 2, against the cellular
protein targets in live T. brucei
(Figure 2c). Each probe (0.1–
10 mm) was directly added to
the cell-culture medium in
which the parasites were grown,
either alone or in the presence
of competing THL (over a con-
centration range of 10–50 mm;
see Figure S3 in the Supporting
Information). After two hours,
the parasites were washed (to
remove excessive probes), ho-
mogenized, incubated with
rhodamine–azide under click-
Figure 2. Trypanocidal activities and in situ proteome profiling of THL-derived compounds. a) Comparison of
the trypanocidal effects of 1a–1g against BSF parasites (white/highlighted bars, with 125 nm compound con-
centration) and PCF parasites (gray bars; with 5000 nm compound concentration), respectively, after 24 h.
Probes with the highest trypanocidal activity against BSF are highlighted in red (1a, 1d, and 1e). The BSF kill-
ing profile of THL under identical conditions is highlighted in blue. Results represent the average (Æ the stan-
dard deviation) of three independent experiments. b) ED50 curves of THL and 1a against BSF and PCF. c) In
situ proteome profiling of THL analogues against BSF (left) and PCF (right). THL (50 mm), MAFP (50 mm),
or cerulenin (a fatty acid synthase inhibitor; 100 mm) were added to selected lanes in competition experiments.
Probe-labeled proteins were detected by click conjugation to a rhodamine–azide reporter tag, followed by
SDS polyacrylamide gel electrophoresis (PAGE) and in-gel fluorescence scanning, as previously reported.[26]
safety profiles (as an FDA-approved drug),[4,6] bodes well
for the argument that THL should be considered as one of
the most promising trypanocidal drug candidates. It should
be noted that most of the trypanocidal drugs currently
under development focus on another highly similar parasitic
protozoan, T. cruzi, which causes Chagas disease and infect-
ed at least 12 million people worldwide.[2,3] Due to the high
similarity between T. cruzi and T. brucei, it is reasonable to
assume that THL and its analogues may also be effective
antiparasite agents against T. cruzi.
chemistry conditions, separated by SDS-PAGE, and ana-
lyzed by in-gel fluorescence scanning (Figure 2c). In gener-
al, most probes showed comparable in situ proteome reac-
tivity profiles, which indicated similar cellular targets in live
parasites. On the other hand, the in situ proteome reactivity
profiles with BSF and PCF parasites showed clear differen-
ces (compare the two gels in Figure 2c), a result that points
to the existence of both common and unique protein targets
for THL in the two different proliferating forms of the para-
site. Compound 2 only weakly labeled a handful of proteins,
which is likely a reflection of its insignificant trypanocidal
activity, as well as of the need for full retention of the THL
structure in future drug development. Most of the labeled
bands in both the BSF and PCF profiling gels were THL
sensitive; that is, their labeling was inhibited by the presence
of a high-concentration of competing THL. Similar trends
were observed if competing MAFP and cerulenin, both of
which are serine/cysteine-targeting irreversible inhibitors,
were added. We therefore concluded that most of the la-
beled proteins were likely specific cellular targets of THL
and possess active serine/cysteine residues. Next, we per-
formed large-scale pull-down analyses to identify THL-tar-
Comparative in situ proteome profiling of T. brucei and pu-
tative target identification: The results above indicated that
THL and analogues 1a, 1d, and 1e possessed highly potent
trypanocidal activities, especially against the BSF T. brucei,
which lives in the mammalian host. The trypanocidal activity
of THL may involve other mechanisms in addition to target-
ing parasite lipases (THL is a well-known lipase inhibitor)
because THL and the analogues killed T. brucei more effi-
ciently than the general lipase inhibitor MAFP.[37] To further
investigate the killing mechanism of THL in T. brucei, we
took advantage of the unique properties of our alkyne-con-
8406
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8403 – 8413

Screening and Proteome Profiling of a Potential Anti Trypanosoma brucei Agent
FULL PAPER
geted cellular proteins in both BSF and PCF parasites. Pro-
tein bands specifically labeled by 1a in both BSF and PCF
parasites were enriched (after click-chemistry conjugation
with biotin–azide) by avidin–agarose beads, separated by
SDS-PAGE, and visualized by silver staining (Figure S4 in
the Supporting Information). The entire lane from each
pull-down experiment was excised into multiple gel slices
and each gel slice was individually processed for in-gel tryp-
sin digestion. BSF and PCF parasites treated with DMSO
were used as negative controls for the pull-down experi-
ments and the subsequent proteome analyses by LC-MS/MS
analysis. Two independent experiments identified a total of
160 and 161 proteins in the PCF and BSF samples, respec-
tively (Figure 3). Of these, 68 proteins were observed in
both forms. 29 proteins for PCF parasites and 45 for BSF
parasites (or 18 and 27%, respectively) were hypothetical
proteins without known biological functions. Details of the
protein identification, including their score values, peptide
matches, and protein masses obtained from data analysis,
are listed in Tables S1–S3 in the Supporting Information.
The most prevalent proteins identified (as determined by
spectral counts) in both the PCF and BSF of T. brucei were
metabolic enzymes that are involved in the glycolytic path-
way, amino acid metabolism, fatty acid metabolism, and
lipid biosynthesis. Also detected were cytoskeletal proteins
that are abundantly present in the cells. Other proteins, in-
cluding ribosomal proteins, transporters, various isoforms of
histone proteins, nucleic acid binding proteins, variable sur-
face glycoproteins, and peroxidases, were also identified. We
further grouped the identified proteins according to their
known subcellular localizations (Figure 3b), either theoreti-
cally predicted or experimentally confirmed (based on liter-
ature data mining). Remarkably, 31–34% of all putative tar-
gets are mitochondrial proteins, 14–15% are glycosomal
proteins, and 12% are endoplasmic reticulum (ER) proteins,
results that suggest a major effect of THL on these organ-
elles in T. brucei. Among the proteins identified, some were
inevitably nonspecific protein binders of THL because of
their ꢂstickyꢁ nature and their high endogenous expression
level; these offer little potential therapeutic value. Conse-
quently, we focused our attention on other ꢂhitsꢁ that 1) pos-
sess nucleophilic serine/cysteine residues in their predicted
active sites, because they are more likely to be specific tar-
gets of THL, and 2) are known as potential diagnostic mark-
ers, drug targets, or possible vaccine candidates.[38] A delib-
erate effort was made to categorize these potential THL tar-
gets into both unique and common targets in BSF and PCF
parasites. As shown in Table 1, many proteins identified
only in the BSF pull-down analysis are proteins involved in
lipid and fatty acid biosynthesis, including glycosylphospha-
tidylinositol-specific phospholipase C (GPI-PLC), inositol
phosphorylceramide synthesis (SLS1), phosphatidyltransfer-
ase (PIS), essential neutral sphingomyelinase (TbnSMase),
choline/ethanolamine phosphotransferase (CEPT), GPI
transamidase component GAA1 (TbGAA1), GPI-anchor
transamidase subunit 8 (GPI8), and glycosylphosphatidyli-
nositol anchor (TbGPI16). Although it is reasonable to
assume that these lipid-related proteins are true cellular tar-
gets of THL in T. brucei, given this drugꢁs well-characterized
molecular targets in other biological systems,[6–12] it is inter-
esting to note that they evaded our pull-down/target identifi-
cation analysis in the PCF of T. brucei, even though RNA-
sequencing data suggested expression of these proteins in
PCF as well. We also identified proteins such as calpain,
cathepsin L-like cysteine peptidase (CP), cytoskeleton-asso-
ciated protein CAP5.5 (CAP5.5), metacaspase (MCA1), and
protein disulfide isomerase (BS2), all of which possess
active cysteine residues and had previously been identified
as targets of THL and b-lactones.[21,23–26,28] Sterol 14-alpha-
demethylase (CYP51), a membrane-associated enzyme, was
Figure 3. Classification and validation of identified proteins. Functional classifications (a) and the known subcellular localization (b) of identified proteins
in BSF parasites (PCF data are shown in parentheses). c) Target validation experiments on five recombinantly expressed ꢂhitsꢁ. The labeling experiment
was done with 1a against the proteins expressed in the whole E. coli proteome lysates. Equal amounts of protein lysates and 1a were used in each lane.
The labeled lysates were subjected to click chemistry with rhodamine–azide, SDS-PAGE analysis, and in-gel fluorescence scanning (shown in gray scale).
A GST-expressing lysate (lane 1) was used as a negative control: no labeling of GST was observed. For the five “hits”, all of the respective lanes showed
a strongly labeled, fluorescent band at the expected molecular weight. Control experiments showing equal protein loading in each lane are provided in
Figure S5 in the Supporting Information (Coomassie gel and dot-blot with anti-GST antibody).
Chem. Eur. J. 2012, 18, 8403 – 8413
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8407

C. Y. He, S. Q. Yao et al.
Table 1. Representative proteins identified with 1a in T. brucei.[a]
Entry
T. brucei gene
Protein name
Location
Detection
Lipid metabolism:
1
2
3
4
5
6
7
8
Tb927.8.6390
lysophospholipase, alpha/beta hydrolase (TbLysoPLA)
phospholipase A1 (PLA1)
phospholipase A2 like protein
n.a.
n.a.
n.a.
membrane
membrane
membrane
ER
ER
M
membrane
n.a.
n.a.
membrane
n.a.
ER
both
PCF
BSF
BSF
both
BSF
BSF
BSF
BSF
BSF
BSF
BSF
BSF
PCF
BSF
both
Tb927.1.4830
Tb09.211.3650
Tb927.2.6000
Tb11.01.6800
Tb09.211.1030
Tb09.160.0530
Tb927.5.3710
Tb927.10.8900
Tb10.100.0100
Tb927.10.13860
Tb927.4.1920
Tb10.70.2420
Tb11.01.4790
Tb927.6.1820
Tb09.211.2730
glycosylphosphatidylinositol-specific phospholipase C (GPI-PLC)[b]
1-acyl-sn-glycerol-3-phosphate acyltransferase
inositol phosphorylceramide synthase (SLS1)[b]
phosphatidyltransferase (PIS)[c]
essential neutral sphingomyelinase (TbnSMase)[b]
choline/ethanolamine phosphotransferase (CEPT)[b]
GPI transamidase component GAA1 (TbGAA1)[b]
GPI-anchor transmidase subunit 8 (GPI8)[b]
glycosylphosphatidylinositol (GPI) anchor (TbGPI16)[b]
GPI inositol deacylase precursor (GPIdeAc)
phospholipid:diacylglycerol acyltransferase-like protein
dolichyl pyrophosphate phosphatase[b]
9
10
11
12
13
14
15
16
gim5A
G
Fatty acid metabolism:
17
18
19
Tb09.160.2770
Tb09.160.2810
Tb927.8.7100
Proteolysis:
fatty acyl CoA synthetase (ASC1)[b]
fatty acyl CoA synthetase (ASC3)[b]
acetyl-CoA carboxylase
membrane
membrane
M
BSF
both
both
20
21
22
23
24
25
Tb927.8.8330
Tb927.6.1000
Tb927.4.3950
Tb11.02.0730
Tb927.10.1040
Tb927.10.8230
calpain, cysteine peptidase[c]
n.a.
lysosome
n.a.
M
n.a.
both
both
PCF
BSF
both
BSF
cysteine peptidase, clan CA, family C1, cathepsin L like (CP)[c]
cysteine peptidase, clan CA, family C2, cytoskeleton-associated protein CAP5.5 (CAP5.5)[c]
metacaspase, cysteine peptidase, clan CD, family C13 (MCA1)[c]
serine carboxypeptidase III precursor, serine peptidase, clan SC, family S10 (CBP1)
protein disulfide isomerase, bloodstream-specific protein 2 precursor (BS2)[b]
cytoplasm
Miscellaneous proteins:
26
27
28
29
30
Tb927.10.7090
Tb927.4.2450
Tb11.02.4080
Tb09.v1.0380
Tb927.6.4280
alternative oxidase (AOX)[b]
M
BSF
BSF
both
PCF
both
thioredoxin[b]
n.a.
n.a.
n.a.
G
sterol 14-alpha-demethylase (CYP51)[c]
spermidine synthase (SpSyn)[c]
glyceraldehyde 3-phosphate dehydrogenase (GAPDH)[c]
[a] G: glycosomal; M: mitochondrial; n.a.: not available. [b] Sensitive to RNA interference. [c] Putative drug target. Further validation experiments were
attempted for nine “hit” proteins (entries 1, 6–8, 16, 21, 23, 25 and 30). Five proteins (entries 1, 8, 23, 25, and 30) were successfully cloned/expressed/la-
beled. The other four proteins (entries 6, 7, 16, and 21) could not be expressed as soluble proteins and therefore were not pursued further.
identified in both the BSF and PCF pull-down analyses. This
protein has recently been validated as a potential target for
antitrypanosome therapy.[39,40] Gratifyingly, we identified
two fatty acyl CoA synthetases (namely, ASC1 and ASC3),
which are the parasitic homologues of mammalian fatty acid
synthetases and are known targets of THL.[7,10,26] Their cel-
lular activities have recently been shown to be important for
BSF parasites in RNA interference (RNAi) based experi-
ments.[41,42] Nevertheless, owing to the highly complex cellu-
lar environment and the intrinsic limitations of affinity pull-
down assays and mass spectrometry, false positives/nonspe-
cific protein binders could be minimized but not eliminated
entirely from our results. To confirm our pull-down/LC-MS/
MS results, we selected nine ꢂhitsꢁ of THL from Table 1
based on a number of criteria (presence of nucleophilic
active-site residues, molecular weight match, MS confidence,
etc.) and attempted to clone/express them in E. coli as gluta-
thione S-transferase (GST) fusions. Five proteins that were
successfully expressed in soluble forms were BS2, TbLyso-
PLA, MCA1, TbnSMase, and GAPDH. Bacteria overex-
pressing these proteins, together with bacteria overexpress-
ing GST only (as a negative control), were homogenized, la-
beled with 1a, and subsequently analyzed by in-gel fluores-
cence scanning following click chemistry with rhodamine–
azide (Figure 3c). All five fusion proteins were successfully
labeled by 1a specifically, even in the presence of other en-
dogenous bacterial proteins. GST alone, however, was not
labeled by 1a (Figure 3c; see Figure S5 in the Supporting
Information for control experiments). Taken together, these
results indicate that all five T. brucei proteins were likely
true cellular targets of THL. Other ꢂhitsꢁ identified from the
pull-down experiments were also possible targets of THL
and remain to be validated by using similar approaches.
Nevertheless, our parasite-based screening, proteome profil-
ing, target identification, and validation results suggest that
THL acts as a trypanocidal drug in T. brucei by targeting es-
sential parasitic proteins involved in lipid and fatty acid met-
abolic pathways, including lipases and fatty acyl CoA syn-
thetases, and likely other proteins possessing active serine/
cysteine residues.
Cellular uptake and morphological changes of T. brucei
upon drug treatment: There has been enormous interest in
the development of small-molecule-based imaging probes
8408
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8403 – 8413

Screening and Proteome Profiling of a Potential Anti Trypanosoma brucei Agent
FULL PAPER
(for example, activity-based probes) capable of detecting en-
zymatic activities within intact cells and in vivo.[43–46] Al-
though our THL-like probes were not expected to be target-
specific imaging probes, we realized immediately that 1a,
which is a THL mimic containing an extremely conservative
lipoproteins, which carry phospholipids and cholesteryl
esters, than PCF parasites.[47] It is therefore intriguing to
speculate that our THL-like probes (such as 1a), given their
lipid-like chemical structure, may also enter the BSF para-
sites effectively through receptor-mediated endocytosis
pathways. The same reason may also explain why THL and
compounds 1a–1g were much more potent trypanocidal
agents in BSF parasites than in PCF parasites. It should be
highlighted that, with the unique properties of these probes
in imaging experiments (cell permeability, “clickable” fluo-
rescence tagging), they allow convenient, rapid, and highly
sensitive monitoring of drug action in cells and may be fur-
ther developed into novel probes for positron emission to-
mography (PET) by clicking with reporter tags containing
suitable radioisotopes (such as 18F).[48] We further analyzed
the morphological changes of different subcellular organ-
elles in BSF and PCF T. brucei upon treatment with 1a
(125 nm and 2.5 mm for BSF and PCF, respectively, for 24 h).
Anti-TbBiP, anti-TbTrypanopain, and anti-TbSKL antibod-
ies were used to fluorescently stain the ER, lysosomes, and
glycosomes of the parasites, respectively. MitoTracker was
À
ꢀ
C C to C C structural modification and possessing full try-
panocidal activities, may serve as a useful imaging probe to
monitor the cellular uptake and, potentially, the subcellular
localization of THL, as well as to detect endogenous “THL-
responsive” targets in both the BSF and PCF of T. brucei by
fluorescence microscopy. From our present studies, THL
and its analogues killed BSF parasites much more effective-
ly than PCF parasites. In situ proteome profiling suggested
that THL targeted similar pathways/organelles in these two
different forms. Although it is possible that THL achieved
much more efficient killing of BSF samples by selectively in-
hibiting target(s) uniquely present in BSF (the investigation
of this possibility is ongoing), we wondered whether the dis-
parity in trypanocidal activity might also be due to differen-
tial cellular uptakes of this drug in the two forms of the par-
asite. To test this, live parasites were treated with probe 1a,
fixed, and treated in situ with rhodamine–azide, then
imaged. Immunofluorescence experiments were carried out,
if applicable, to determine the subcellular localization of the
drug, as well as subsequent morphological changes on the
parasites (Figure 4; Figures S6 and S7 in the Supporting In-
formation). PCF parasites displayed significantly poorer cel-
lular uptake of the drug; at least a 10 mm concentration of
1a was needed in order for it to be sufficiently detected
inside the parasites. On the other hand, the cellular uptake
of the drug could be readily detected in BSF parasites even
with only 1 mm 1a (Figure S6 in the Supporting Informa-
tion). Previous studies had suggested that BSF parasites ex-
hibit much more rapid rates of endocytosis of low-density
used as
a mitochondrion fluorescence marker. DAPI
(2 mgmLÀ1) was used to stain the kinetoplast and nuclear
DNA of the parasites. As shown in Figure 4, the normal net-
work of the mitochondrion visualized with MitoTracker col-
lapsed into a flattened sheet with a series of fenestrations
upon drug treatment. The ER and glycosomes were barely
visible compared with those in the control cells (treated
with DMSO only). These observations confirm that THL
has clear effects on the morphology of the mitochondria,
glycosomes, and ER in both BSF and PCF cells, thereby
supporting specific effects of THL on these organelles and
correlating well with our in situ proteome profiling.
Figure 4. Morphological changes in T. brucei upon treatment with 1a for 24 h. Immunofluorescence was carried out with the T. brucei parasites for visual-
ization of intracellular compartments. MitoTracker Red CMXRos (Invitrogen) was used to stain the mitochondrion (left; pseudocolored in red). Immu-
nofluorescence staining (pseudocolored in green) was used to label glycosomes (middle), the ER (right), and lysosomes (Figure S6 in the Supporting In-
formation) by using the primary antibodies anti-TbSKL, anti-TbBiP and anti-TbTrypanopain, respectively. This was followed by fluoroscein isothiocya-
nate conjugated antirabbit immunoglobulin G secondary antibody and nuclei (4’,6-diamidino-2-phenylindole (DAPI); in blue) stainings. All images were
acquired under the same settings. The scale bar represents 10 mm.
Chem. Eur. J. 2012, 18, 8403 – 8413
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8409

C. Y. He, S. Q. Yao et al.
Conclusion
Experimental Section
Synthesis of THL analogues: Compounds 1b–1g were prepared by using
the same synthetic procedures previously described for 1a,[26] as de-
scribed in the Supporting Information.
Trypanosoma brucei is a parasite that causes African sleep-
ing sickness in humans and nagana in livestock and is trans-
mitted by the tsetse fly. There is an urgent need for the de-
velopment of new drugs against African trypanosomiasis
due to the lack of vaccines and effective drugs. THL, as an
FDA-approved antiobesity drug, was previously shown to
possess potential activities against tumors, mycobacteria,
and parasites, but the molecular mechanism has not been
documented. In the current study, we have successfully syn-
thesized and evaluated an expanded set of THL-like probes
for their trypanocidal activities against both the bloodstream
form (BSF) and procyclic form (PCF) of the protozoan par-
asite T. brucei. Our results showed THL and its analogues
1a, 1d, and 1e possessed one of the most potent trypanoci-
dal profiles known against T. brucei to date, especially
against the BSF parasites. Subsequent in situ proteomic
profiling of the THL-like probes enabled us to identify, for
the first time, putative cellular targets of THL in both BSF
and PCF parasites, which indicated organelle-specific target-
ing of the drug. Some of the newly identified targets were
further validated by labeling of recombinantly expressed en-
zymes in E. coli lysates. Our results suggest that THL acts as
a potent trypanocidal drug in T. brucei by targeting essential
parasitic proteins involved in lipid and fatty acid metabolic
pathways, including lipases, fatty acyl CoA synthetases, and
other proteins possessing active serine/cysteine residues.
With the unique design and properties (for example, cell
permeability and clickability) of these THL-like probes,
they allowed convenient, rapid, and highly sensitive imag-
ing-based experiments to be carried out with live parasites
for the study of drug uptake and parasite-morphology
changes. Our results confirmed much more efficient cellular
uptake of THL-like compounds by BSF cells than PCF par-
asites. Clearly, in order to obtain a better understanding of
the possible mechanisms and pharmacological effects of
THL in T. brucei, additional studies will be required for the
validation of the other THL-binding targets identified from
our present study and to further understand the biological
consequences of some of these already-validated targets. We
are also testing the trypanocidal activities of these com-
pounds in other related parasites such as T. cruzi, and the
results will be reported in due course. Notwithstanding,
from our current results, we conclude that THL is one of
the most potent trypanocidal compounds known to date.
This, together with its already well-characterized pharmaco-
kinetic and safety profiles as an FDA-approved drug, sup-
ports the notion that THL should be considered as one of
the most promising trypanocidal drug candidates for further
development.
(S)-(S)-1-[(2S,3S)-3-(hex-5-yn-1-yl)-4-oxooxetan-2-yl]tridecan-2-yl 2-for-
mamidopropanoate (1b): Representative Mitsunobu reaction procedure
for the b-lactone:[26,28] Briefly, b-lactone 2 (18 mg, 0.05 mmol), triphenyl-
phosphine (20 mg, 0.075 mmol), PPh3 (79 mg, 0.3 mmol), and N-formyl-l-
alanine (23 mg, 0.2 mmol) were placed in a round-bottomed flask and
azeotroped under vacuum with xylene (0.5 mL) for 30 min. Addition of
dry THF (1 mL) was followed by cooling of the mixture to 08C. DIAD
(15 mL, 0.075 mmol) was then added through a syringe. The mixture was
stirred at 08C for 10 min and then allowed to warm to ambient tempera-
ture. The reaction was monitored by TLC. After the reaction was com-
plete, the mixture was concentrated in vacuo and purified by flash chro-
matography (20% EtOAc/hexane) to yield 1b (20 mg, 89%) as a color-
less oil; 1H NMR (500 MHz, CDCl3): d=8.19 (s, 1H), 6.16 (brs, 1H),
5.07 (m, 1H), 4.64 (quin, J=7.6 Hz, 1H), 4.32 (dt, J=8.9, 4.4 Hz, 1H),
3.23 (dt, J=7.6, 3.8 Hz, 1H), 2.72 (dt, J=6.3, 2.5 Hz, 2H), 2.16 (dt, J=
14.5, 4.4 Hz, 1H), 2.01 (dt, J=14.5, 4.4 Hz, 1H), 1.95 (t, J=2.5 Hz, 1H),
1.54–1.62 (m, 8H), 1.45 (d, J=7.0 Hz, 3H), 1.25 (brs, 18H), 0.88 ppm (t,
J=7.6 Hz, 3H); 13C NMR (125 MHz, CDCl3): d=172.1, 170.4, 160.4,
83.8, 74.8, 72.9, 68.8, 60.4, 56.9, 47.1, 38.9, 34.2, 31.9, 29.6, 29.5, 29.4,
29.31. 29.27, 27.9, 27.1, 25.7, 25.1, 22.7, 21.0, 18.3, 18.1, 14.2, 14.1 ppm;
ESI-MS: m/z calcd for C26H43NO5 [M+H]+:450.3; found: 450.3.
(S)-1-[(2S,3S)-3-(hex-5-yn-1-yl)-4-oxooxetan-2-yl]tridecan-2-yl 2-forma-
midoacetate (1c): Prepared according to the representative Mitsunobu
reaction procedure by using b-lactone 2 (26 mg, 0.075 mmol), triphenyl-
phosphine (30 mg, 0.113 mmol), N-formyl-glycine (31 mg, 0.3 mmol), and
DIAD (23 mL, 0.113 mmol) in THF (1.0 mL). Purification by flash chro-
matography on SiO2 (20% EtOAc/hexane) gave 1c (24 mg, 73%) as
a colorless oil; 1H NMR (500 MHz, CDCl3): d=8.24 (s, 1H), 6.23 (brs,
1H), 4.34 (dt, J=8.8, 4.4 Hz, 1H), 4.10 (dd, J=18.3, 5.7 Hz, 1H), 4.02
(dd, J=18.3, 5.7 Hz, 1H), 3.22 (dt, J=7.6, 3.8 Hz, 1H), 2.20 (dt, J=6.3,
2.5 Hz, 2H), 2.18 (t, J=8.2 Hz, 1H), 2.01 (dt, J=14.1, 4.4 Hz, 1H), 1.95
(t, J=2.5 Hz, 1H), 1.73–1.85 (m, 4H), 1.52–1.59 (m, 8H), 1.24 (brs,
18H), 0.86 ppm (t, J=6.9 Hz, 3H); 13C NMR (125 MHz, CDCl3): d=
170.5, 169.2, 161.0, 83.8, 74.9, 72.9, 68.8, 56.8, 40.1, 38.8, 34.0, 31.8, 29.54,
29.48, 29.37, 29.27, 29.24, 27.9, 27.0, 25.7, 25.1, 22.6, 18.0, 14.0 ppm; ESI-
MS: m/z calcd for C25H41NO5 [M+H]+: 436.3; found: 436.3.
(2S)-(S)-1-[(2S,3S)-3-(hex-5-yn-1-yl)-4-oxooxetan-2-yl]tridecan-2-yl
2-
formamido-3-methylpentanoate (1d): Prepared according to the repre-
sentative Mitsunobu reaction procedure by using b-lactone 2 (28 mg,
0.08 mmol), triphenylphosphine (31 mg, 0.12 mmol), N-formyl-l-isoleu-
cine (51 mg, 0.32 mmol), and DIAD (24 mL, 0.12 mmol) in THF
(1.0 mL). Purification by flash chromatography on SiO2 (20% EtOAc/
hexane) gave 1d (21 mg, 54%) as a colorless oil; 1H NMR (500 MHz,
CDCl3): d=8.23 (s, 1H), 6.10 (brd, J=8.2 Hz, 1H), 5.02 (m, 1H), 4.64
(dd, J=8.9, 5.1 Hz, 1H), 4.29 (dt, J=8.9, 5.1 Hz, 1H), 3.24 (dt, J=7.6,
4.5 Hz, 1H), 2.21 (dt, J=6.3, 2.5 Hz, 2H), 2.18 (t, J=7.6 Hz, 1H), 2.01
(dt, J=10.1, 5.0 Hz, 1H), 1.95 (t, J=2.5 Hz, 1H), 1.55–1.85 (m, 12H),
1.26 (brs, 17H), 0.96 (d, J=6.9 Hz, 3H), 0.94 (t, J=7.6 Hz, 3H),
0.87 ppm (t, J=7.6 Hz, 3H); 13C NMR (125 MHz, CDCl3): d=170.9,
170.5, 160.7, 83.7, 74.4, 72.8, 68.7, 56.9, 55.4, 38.6, 37.6, 33.9, 31.9, 29.6,
29.5, 29.4, 29.29, 29.28, 27.9, 27.1, 25.7, 25.1, 24.9, 22.6, 18.1, 15.6, 14.1,
11.5 ppm; ESI-MS: m/z calcd for C29H46NO5 [M+H]+: 492.4; found:
492.4.
(S)-(S)-1-[(2S,3S)-3-(hex-5-yn-1-yl)-4-oxooxetan-2-yl]tridecan-2-yl
2-
formamido-3-phenylpropanoate (1e): Prepared according to the repre-
sentative Mitsunobu reaction procedure by using b-lactone 2 (18 mg,
0.05 mmol), triphenylphosphine (20 mg, 0.075 mmol), N-formyl-l-phenyl-
alanine (19 mg, 0.1 mmol), and DIAD (15 mL, 0.075 mmol) in THF
(1.0 mL). Purification by flash chromatography on SiO2 (20% EtOAc/
hexane) gave 1e (12 mg, 46%) as white solids; 1H NMR (500 MHz,
CDCl3): d=8.17 (s, 1H), 7.26–7.32 (m, 3H), 7.17–7.19 (m, 2H), 6.10
(brd, J=6.9 Hz, 1H), 4.99 (m, 1H), 4.90 (dd, J=13.9, 7.3 Hz, 1H), 4.16–
4.24 (m, 1H), 3.08–3.20 (m, 3H), 2.21 (dt, J=6.3, 2.5 Hz, 2H), 2.18 (t, J=
8410
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8403 – 8413

Screening and Proteome Profiling of a Potential Anti Trypanosoma brucei Agent
FULL PAPER
7.6 Hz, 1H), 2.01 (dt, J=10.1, 5.0 Hz, 1H), 1.95 (t, J=2.5 Hz, 1H), 1.55–
1.85 (m, 6H), 1.26 (brs, 20H), 0.88 ppm (t, J=6.95 Hz, 3H); 13C NMR
(125 MHz, CDCl3): d=170.8, 170.5, 160.5, 139.6, 129.3, 129.2, 128.7,
128.6, 127.3 (2), 83.9, 74.5, 72.9, 68.7, 56.8, 52.1, 38.5, 37.8, 33.9, 31.9, 29.6,
29.5, 29.4, 29.3 (2), 27.9, 27.1, 25.7, 25.1, 24.6, 22.6, 18.1, 14.1 ppm; ESI-
MS: m/z calcd for C32H47NO5 [M+H]+: 525.3; found: 525.4.
saline (PBS) and resuspended in PBS (100 mL). Cells were homogenized
by sonication and diluted to approximately 1 mgmLÀ1 with PBS. To ini-
tiate the click-chemistry reaction, a freshly premixed solution (20 mL)
containing rhodamine–azide (100 mm final concentration), tris(2-carbox-
yethyl)phosphine (TCEP; 1 mm final concentration), tris[(1-benzyl-1H-
1,2,3-triazol-4-yl)methyl]amine (TBTA; 100 mm final concentration), and
CuSO4 (1 mm final concentration) was added. The reaction was incubated
at 108C for 2 h with gentle mixing. Termination of the reaction was ach-
ieved by addition of prechilled acetone (0.5 mL). The resulting solution
was then placed at À208C for 30 min, and this was followed by centrifu-
gation (13000 rpm for 10 min) at 48C. The supernatant was discarded
and the precipitated protein pellets were washed with prechilled metha-
nol (2ꢃ200 mL), air-dried for 10 min, resuspended in 1ꢃ standard reduc-
ing SDS-loading buffer (25 mL), and then heated for 10 min at 958C. Fi-
nally, the protein samples (ꢁ20 mg per lane) were separated on a 10%
SDS-PAGE gel and analyzed by in-gel fluorescence scanning with a Ty-
phoon 9410 Variable-Mode Imager scanner (GE Amersham).
(S)-1-[(2S,3S)-3-(hex-5-yn-1-yl)-4-oxooxetan-2-yl]tridecan-2-yl 3-forma-
midopropanoate (1 f): Prepared according to the representative Mitsuno-
bu reaction procedure by using b-lactone 2 (35 mg, 0.1 mmol), triphenyl-
phosphine (40 mg, 0.15 mmol), N-formyl-b-glycine (23 mg, 0.2 mmol),
and DIAD (30 mL, 0.15 mmol) in THF (2.0 mL). Purification by flash
chromatography on SiO2 (20% EtOAc/hexane) gave 1 f (30 mg, 67%) as
a colorless oil; 1H NMR (500 MHz, CDCl3): d=8.15 (s, 1H), 6.24 (brs,
1H), 5.09 (dt, J=6.3, 3.2 Hz, 1H), 4.36 (dt, J=8.8, 4.4 Hz, 1H), 3.57
(quin, J=5.6 Hz, 1H), 3.22 (dt, J=6.9, 3.8 Hz, 1H), 2.55 (dd, J=6.3,
5.0 Hz, 2H), 2.21 (dt, J=6.9, 2.5 Hz, 2H), 2.10 (dt, J=15.1, 8.2 Hz, 1H),
2.00 (dt, J=15.1, 3.8 Hz, 1H), 1.96 (t, J=2.5 Hz, 1H), 1.53–1.84 (m,
10H), 1.25 (brs, 18H), 0.87 ppm (t, J=7.0 Hz, 3H); 13C NMR (125 MHz,
CDCl3): d=172.0, 170.4, 161.2, 83.8, 76.0, 72.0, 68.8, 56.9, 39.0, 34.3, 33.4,
31.9, 29.6, 29.5, 29.4, 29.29, 29.26, 27.9, 27.0, 25.7, 25.2, 22.6, 18.0,
14.1 ppm; ESI-MS: m/z calcd for C26H43NO5 [M+H]+: 450.3; found:
450.3.
Pull-down and mass spectrometry identification: For proteomic experi-
ments, BSF and PCF trypanosomes (ꢁ2ꢃ109 cells, ꢁ5 mg each), labeled
in Cunninghamꢁs medium (1ꢃ107 cellsmLÀ1) with 1a or DMSO (negative
control to rule out the background of unspecific protein binding on
avidin–agarose beads), were harvested, washed, and homogenized in
PBS. Copper-catalyzed azide–alkyne cycloaddition reagents were added
at the same concentrations as those described above, except that biotin–
azide was substituted for rhodamine–azide. Acetone-precipitated and
methanol-washed protein pellets were solubilized in PBS containing
0.1% (w/v) SDS by brief sonication. Insoluble materials were precipitat-
ed by centrifugation (13000 g for 10 min) at 48C. The supernatants were
then incubated with gentle shaking at 48C overnight with Neutravidin
agarose beads (50 mLmgÀ1 protein; product no. 29204, Thermo Scientific,
USA), which have been prewashed twice with PBS. The bead/complexes
were washed extensively four times with 1% (w/v) SDS in PBS, three
times with PBS, and twice with ammonium bicarbonate (250 mm). Elu-
tion of bound proteins from the beads was then performed by using the
boiling buffer (200 mm tris(hydroxymethyl)aminomethane (pH 6.8),
400 mm 1-dithiothreitol, 8% (w/v) SDS). Gel bands were isolated,
washed, and tryptically digested, then analyzed by nano-LC-MS/MS as
described previously.[26,28,49]
(R)-(S)-1-[(2S,3S)-3-(hex-5-yn-1-yl)-4-oxooxetan-2-yl]tridecan-2-yl
2-
formamido-4-methylpentanoate (1g): Prepared according to the repre-
sentative Mitsunobu reaction procedure by using b-lactone 2 (18 mg,
0.05 mmol), triphenylphosphine (20 mg, 0.075 mmol), N-formyl-d-leucine
(16 mg, 0.1 mmol), and DIAD (15 mL, 0.075 mmol) in THF (1.0 mL). Pu-
rification by flash chromatography on SiO2 (20% EtOAc/hexane) gave
1g (18 mg, 73%) as a colorless oil; 1H NMR (500 MHz, CDCl3): d=8.20
(s, 1H), 5.93 (brd, J=8.2 Hz, 1H), 5.02 (m, 1H), 4.63 (dt, J=8.8, 3.0 Hz,
1H), 4.37 (m, 1H), 3.24 (dt, J=7.6, 3.8 Hz, 1H), 2.20 (m, 2H), 2.16 (t,
J=7.4 Hz, 1H), 2.01 (dt, J=14.1, 4.4 Hz, 1H), 1.92 (t, J=2.5 Hz, 1H),
1.50–1.84 (m, 10H), 1.25 (brs, 17H), 0.96 (d, J=5.7 Hz, 6H), 0.87 ppm (t,
J=7.6 Hz, 3H); 13C NMR (125 MHz, CDCl3): d=172.2, 170.7, 160.7,
83.9, 74.4, 72.5, 68.7, 60.4, 56.8, 49.7, 41.3, 38.5, 36.6, 33.8, 31.9, 29.6, 29.5,
29.4, 29.3, 29.2, 28.0, 27.1, 25.7, 25.2, 24.9, 24.7, 22.8, 22.6, 21.8, 18.1, 14.2,
14.1 ppm; ESI-MS: m/z calcd for C29H46NO5 [M+H]+: 492.4; found:
492.4.
Target validation by recombinant protein expression in E. coli lysates:
Five hits were successfully expressed in soluble form from E. coli as GST
fusion proteins. Target genes were amplified from the corresponding ge-
nomes by the following PCR cloning primers:
Trypanosome strains and cell culture: T. brucei procyclic parasites YTAT
1.1 were grown at 288C and 5% CO2 in Cunninghamꢁs medium supple-
mented with 15% heat-inactivated fetal bovine serum (FBS). T. brucei
BSF parasites were grown at 378C and 5% CO2 in HMI-9 medium sup-
plemented with 20% heat-inactivated FBS.
BS2 fwd
CGGGATCCATGCGCGCTATTTTTCTTGTTGC
Guava ViaCount assay: Parasite number and percentage viability were
determined in a 96-well plate format by using the Guava ViaCount assay
on a Guava PCA-96 system (Guava Technologies, USA) according to the
manufacturerꢁs instructions. Briefly, BSF and PCF trypanosomes were
harvested in exponential growth phase and adjusted to a concentration of
1ꢃ105 cellsmLÀ1 in complete growth medium. Diluted trypanosomes
were dispensed manually with a multichannel pipette (BrandTech, USA).
Compounds to be tested were serially diluted in DMSO with a multichan-
nel pipette. Positive and negative controls consisted of wells containing
THL, two cysteine protease inhibitors DMK and K11777, a broad-spec-
trum lipase inhibitor MAFP, or 1% DMSO alone. After 24 h of incuba-
tion, the cell density and viability were evaluated by using the ViaCount
assay. ED50 values were calculated by sigmoid curve fitting with Graph-
Pad Prism 5.0 software (San Diego, USA). All data were collected in
triplicate.
BS2 rev
CCGCTCGAGCTACAGATCTTGTTTATCAACATTGC
LysoPLA fwd
5’-CGGGATCCATGTTTGGCACGCCGGTTG
LysoPLA rev
5’-CCGCTCGAGTTACGATTTCGATGAAGGTCCG
MCA1 fwd
CGGGATCCATGTCGTCTGAAAAAGTCGCTG
MCA1 rev
CCGCTCGAGTTATTTTCCACCAAGAGGGGTG
TbnSMase fwd
CGGGATCCATGGCTGCGGAGATAACTGTTC
TbnSMase rev
In situ proteome profiling and in-gel fluorescence scanning: T. brucei par-
asites were transferred into 6-well plates (PCF, 2 mL at ꢁ1ꢃ
ACGCGTCGACTCATTTACAACCATTACCTTTATGCATTTC
GAPDH fwd
107 cellsmLÀ1
)
or 25 mL cell-culture flasks (BSF, 10 mL at ꢁ2ꢃ
CGGGATCCATGACTATTAAAGTTGGCATCAATGG
106 cellsmLÀ1) and incubated with THL analogues for 2 h at culture tem-
perature with or without a competing inhibitor (THL, MAFP, or cerule-
nin). All compounds were solubilized in DMSO. To avoid adverse effects
on parasite growth, the final DMSO concentration in the assay never ex-
ceeded 1% in the cultivation medium. Medium containing 1% DMSO
was used as a negative control. After incubation, the parasite cells were
pelleted at 2000 rpm for 10 min, washed twice with phosphate-buffered
GAPDH rev
CCGCTCGAGTTATAGCTTTGCTGCACGGTCC
PCR products were identified on agarose gels, and gel bands were isolat-
ed and extracted with a Qiagen gel extraction kit. Concentrations of
DNA were measured with
a NanoDrop spectrophotometer 1000
(Thermo). Purified PCR product (100 nmol) and vector pGEX-6P-
Chem. Eur. J. 2012, 18, 8403 – 8413
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8411

C. Y. He, S. Q. Yao et al.
1 (30 nmol) were used for an in vitro recombination reaction with T4
DNA ligase (NEB). After transformation in competent TOP10 E. coli,
cells were plated on Luria–Bertani (LB) agar plates containing
100 mgmLÀ1 ampicillin. Clones of transformed cells were selected and
grown in ampicillin-containing LB medium. Cells were harvested, and
plasmids were isolated. The expression clone was transformed in compe-
tent BL21 E. coli cells and selected on LB agar plates containing
100 mgmLÀ1 ampicillin. The validity of the clones was confirmed by plas-
mid sequence analysis. Recombinant clones were grown in ampicillin-
containing LB medium, and target-gene expression was induced with
0.2 mm isopropyl-b-d-thiogalactopyranoside. The cell culture (10 mL) of
the bacterial cell pellets was washed with PBS, resuspended in lysis
buffer (1 mL; 0.2% SDS in PBS), and lyzed by sonication. The protein
samples were adjusted to a final concentration of 0.5 mgmLÀ1 by dilution
in PBS prior to in vitro labeling (labeling conditions: 1a (5 mm), 30 min
of incubation at room temperature, followed by the click reaction with
rhodamine–azide as described above). Bacterial lysates over-expressing
GST were similarly labeled as a negative control.
[1] P. P. Simarro, J. Jannin, P. Cattand, PLoS Med. 2008, 5, e55.
[2] R. Pink, A. Hudson, M.-A. Mouries, M. Bendig, Nat. Rev. Drug Dis-
covery 2005, 4, 727–740.
[3] A. R. Renslo, J. H. McKerrow, Nat. Chem. Biol. 2006, 2, 701–710.
[4] C. Chong, D. J. Sullivan Jr., Nature 2007, 448, 645–646.
[5] S. Sundar, N. Engl. J. Med. 2002, 347, 1739–1746.
[6] US Food and Drug Administration, THL (marketed as Alli and
Xenical) information (online); available from: http://www.fda.gov/
Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatient-
sandProviders/ucm180076.htm (Accessed: November 15, 2010).
[7] S. J. Kridel, F. Axelrod, N. Rozenkrantz, J. W. Smith, Cancer Res.
2004, 64, 2070–2075.
[8] J. L. Little, F. B. Wheeler, D. R. Fels, C. Koumenis, S. J. Kridel,
Cancer Res. 2007, 67, 1262–1269.
[9] C. W. Pemble, L. C. Johnson, S. J. Kridel, W. T. Lowther, Nat. Struct.
Mol. Biol. 2007, 14, 704–709.
[10] L. M. Knowles, C. Yang, A. Osterman, J. W. Smith, J. Biol. Chem.
2008, 283, 31378–31384.
Fluorescence microscopy: For drug-uptake analysis, trypanosomes (1ꢃ
105 cellsmLÀ1 for both forms) were incubated in growth medium contain-
ing different concentrations of 1a at the culture temperature and 5%
CO2 for 2 h. Medium containing 1% DMSO was used as a negative con-
trol. The parasites were then washed twice with PBS, fixed with 4% par-
aformaldehyde in PBS for 15 min at room temperature, washed with PBS
(2ꢃ5 min with gentle agitation), and then sedimented to poly-l-lysine-
coated coverslips. Fixed cells were permeabilized with 0.25% Triton-
X 100 in PBS for 15 min at room temperature and washed with PBS (2ꢃ
5 min with gentle agitation). The cells were blocked with 3% bovine
serum albumin (BSA) in PBS for 30 min at room temperature and
washed with PBS (2ꢃ5 min with gentle agitation). The cells were then
treated with a freshly premixed click-chemistry reaction solution (rhoda-
mine–azide (10 mm final concentration from a 10 mm stock solution in
DMSO), TCEP (1 mm final concentration from a 50 mm freshly prepared
stock solution in deionized water), TBTA (100 mm final concentration
from a 10 mm stock solution in DMSO), and CuSO4 (1 mm final concen-
tration from a 100 mm freshly prepared stock solution in deionized
water)) in PBS for 1 h at room temperature. The cells were washed with
PBS (1ꢃ5 min with gentle agitation) and cold methanol (1ꢃ5 min with
gentle agitation), then with 1% Tween-20 and 0.5 mm ethylenediaminete-
traacetate in PBS (3ꢃ2 min with gentle agitation), and with PBS (2ꢃ
5 min with gentle agitation). The cells were then incubated in PBS con-
taining 2 mgmLÀ1 of DAPI for 15 min at room temperature to stain the
kinetoplast and nuclear DNA and were washed with PBS (2ꢃ5 min with
gentle agitation) and finally with deionized water (1ꢃ5 min with gentle
agitation) before being mounted onto the Fluoromount G (Emsdiasum,
USA). Images were acquired by using an Observer Z1 microscope
(Zeiss, Germany) equipped with a 63ꢃ NA1.4 objective and a CoolSNAP
HQ2 CCD camera (Photometrics, USA) or an LSM 510 META micro-
scope (Zeiss, Germany) equipped with an EC Plan-Neofluar 100ꢃ NA1.3
objective. For immunofluorescence analysis, cells treated with 1a for 24 h
were attached to coverslips, fixed with 4% paraformaldehyde, blocked
with 3% BSA in PBS, and then incubated with the corresponding anti-
bodies. Anti-TbBiP, anti-TbTrypanopain, and anti-TbSKL[50] were used to
label the ER, the lysosome, and the glycosome, respectively. DAPI
(2 mgmLÀ1) was used to stain the kinetoplast and nuclear DNA. Mito-
Tracker Red CMXRos (Invitrogen, USA) was used to stain the mito-
chondrion.[49]
[11] K. L. Low, J. Bacteriol. 2009, 191, 5037–5043.
[12] S. K. Parker, R. M. Barkley, J. G. Rino, M. L. Vasil, PLoS One 2009,
4, e4281.
[13] P. K. Crellin, J. P. Vivian, J. Scoble, F. M. Chow, N. P. West, R. Bram-
mananth, N. I. Proellocks, A. Shahine, J. Le Nours, M. C. J. Wilce,
W. J. Britton, R. L. Coppel, J. Rossjohn, T. Beddoe, J. Biol. Chem.
2010, 285, 30050–30060.
[14] Z. B. Mackey, Chem. Biol. Drug Des. 2006, 67, 355–363.
[15] M. J. Evans, B. F. Cravatt, Chem. Rev. 2006, 106, 3279–3301.
´
[16] M. Fonovic, M. Bogyo, Expert Rev. Proteomics 2008, 5, 721–730.
[17] M. Uttamchandani, J. Li, H. Sun, S. Q. Yao, ChemBioChem 2008, 9,
667–675.
[18] E. E. Carlson, ACS Chem. Biol. 2010, 5, 639–653.
[19] W. P. Heal, T. H. T. Dang, E. W. Tate, Chem. Soc. Rev. 2011, 40,
246–257.
[20] T. Bçttcher, M. Pitscheider, S. A. Sieber, Angew. Chem. 2010, 122,
2740–2759; Angew. Chem. Int. Ed. 2010, 49, 2680–2698.
[21] a) T. Bçttcher, S. A. Sieber, Angew. Chem. 2008, 120, 4677–4680;
Angew. Chem. Int. Ed. 2008, 47, 4600–4603; b) T. Bçttcher, S. A.
Sieber, J. Am. Chem. Soc. 2008, 130, 14400–14401.
[22] a) P.-Y. Yang, M. Wang, L. Li, H. Wu, C. Y. He, S. Q. Yao, Chem.
Eur. J. 2012, 18, 6528–6541; b) H. Shi, C.-J. Zhang, G. Y. J. Chen,
S. Q. Yao, J. Am. Chem. Soc. 2012, 134, 3001–3014; c) P.-Y. Yang,
M. Wang, C. Y. He, S. Q. Yao, Chem. Commun. 2012, 48, 835–837;
d) H. Shi, X.-M. Cheng, S. K. Sze, S. Q. Yao, Chem. Commun. 2011,
47, 11306–11308; e) H. Wu, J. Ge, P.-Y. Yang, J. Wang, M. Uttam-
chandani, S. Q. Yao, J. Am. Chem. Soc. 2011, 133, 1946–1954; f) K.
Liu, P.-Y. Yang, Z. Na, S. Q. Yao, Angew. Chem. 2011, 123, 6908–
6913; Angew. Chem. Int. Ed. 2011, 50, 6776–6781; g) H. Shi, M. Ut-
tamchandani, S. Q. Yao, Chem. Asian J. 2011, 6, 2803–2815; h) K.
Liu, H. Shi, H. Xiao, A. G. L. Chong, X. Bi, Y. T. Chang, K. Tan,
R. Y. Yada, S. Q. Yao, Angew. Chem. Int. Ed. 2009, 48, 8293–8297;
i) J. Wang, M. Uttamchandani, J. Li, M. Hu, S. Q. Yao, Chem.
Commun. 2006, 3783–3785; j) S. Chattopadhaya, E. W. S. Chan,
S. Q. Yao, Tetrahedron Lett. 2005, 46, 4053–4056; k) E. W. S. Chan,
S. Chattopadhaya, R. C. Panicker, X. Huang, S. Q. Yao, J. Am.
Chem. Soc. 2004, 126, 14435–14446.
[23] Z. Wang, C. Gu, T. Colby, T. Shindo, R. Balamurugan, H. Wald-
mann, M. Kaiser, R. A. L. van der Hoorn, Nat. Chem. Biol. 2008, 4,
557–563.
[24] F. J. Dekker, O. Rocks, N. Vartak, S. Menninger, C. Hedberg, R. Ba-
lamurugan, S. Wetzel, S. Renner, M. Gerauer, B. Schçlermann, M.
Rusch, J. W. Kramer, D. Rauh, G. W. Coates, L. Brunsveld, P. I. H.
Bastiaens, H. Waldmann, Nat. Chem. Biol. 2010, 6, 449–456.
[25] M. Rusch, T. J. Zimmermann, M. Burger, F. J. Dekker, K. Gormer,
G. Triola, A. Brockmeyer, P. Janning, T. BÅttcher, S. A. Sieber, I. R.
Vetter, C. Hedberg, H. Waldmann, Angew. Chem. 2011, 123, 10012–
10016; Angew. Chem. Int. Ed. 2011, 50, 9838–9842.
Acknowledgements
We thank the Competitive Research Program (R143-000-218-281) from
the National Research Foundation (NRF) of Singapore for generous
funding support. C.Y.H is a research fellow funded by the Singapore Na-
tional Research Foundation. We also wish to thank Professor James
Bangs for the anti-TbBiP and anti-trypanopain antibodies, and Professor
Suresh Subramani for the anti-SKL antibody.
[26] P.-Y. Yang, K. Liu, M. H. Ngai, M. J. Lear, M. R. Wenk, S. Q. Yao, J.
Am. Chem. Soc. 2010, 132, 656–666.
8412
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8403 – 8413

Screening and Proteome Profiling of a Potential Anti Trypanosoma brucei Agent
FULL PAPER
[27] M. H. Ngai, P.-Y. Yang, K. Liu, Y. Shen, S. Q. Yao, M. J. Lear,
Chem. Commun. 2010, 46, 8335–8837.
[28] P.-Y. Yang, K. Liu, C. Zhang, G. Y. J. Chen, Y. Shen, M. H. Ngai,
M. J. Lear, S. Q. Yao, Chem. Asian J. 2011, 6, 2762–2775.
[29] E. M. Sletten, C. R. Bertozzi, Angew. Chem. 2009, 121, 7108–7133;
Angew. Chem. Int. Ed. 2009, 48, 6974–6998.
[30] L. I. Willems, W. A. van der Lin, N. Li, K.-Y. LI, N. Liu, S. Hoogen-
doorn, G. A. van der Marel, B. I. Florea, H. S. Overkleeft, Acc.
Chem. Res. 2011, 44, 718–729.
[31] S. Scory, Y. D. Stierhof, C. R. Caffrey, D. Steverding, Kintoplastid
Bio. Dis. 2007, 6, 2.
[32] O. V. Nikolskaia, J. Clin. Invest. 2006, 116, 2739–2747.
[33] Z. B. Mackey, T. C. OꢁBrien, D. C. Greenbaum, R. B. Blank, J. H.
McKerrow, J. Biol. Chem. 2004, 279, 48426–48433.
[34] P. S. Doyle, Y. M. Zhou, J. C. Engel, J. H. McKerrow, Antimicrob.
Agents Chemother. 2007, 51, 3932–3939.
[35] a) E. L. Ridgley, L. Ruben, Mol. Biochem. Parasitol. 2001, 114, 29–
40; b) MAFP readily hydrolyzes in 24 h under our assay conditions,
and its actual effective concentration is therefore much lowered rel-
ative to that shown in Figure 2.
[36] K. Brak, P. S. Doyle, J. H. McKerrow, J. A. Ellman, J. Am. Chem.
Soc. 2008, 130, 6404–6410.
[40] G. I. Lepesheva, H.-W. Park, T. Y. Hargrove, B. Vanhollebeke, Z.
Wawrzak, J. M. Harp, M. Sundaramoorthy, W. D. Nes, E. Pays, M.
Chaudhuri, F. Villalta, M. R. Waterman, J. Biol. Chem. 2010, 285,
1773–1780.
[41] Y. S. Morita, K. S. Paul, P. T. Englund, Science 2000, 288, 140–143.
[42] L. Riviꢄre, P. Moreau, S. Allmann, M. Hahn, M. Biran, N. Plazolles,
J.-M. Franconi, M. Boshart, F. Bringaud, Proc. Natl. Acad. Sci. USA
2009, 106, 12694–12699.
[43] A. Baruch, D. A. Jeffery, M. Bogyo, Trends Cell Biol. 2004, 14, 29–
35.
[44] G. Blum, G. von Degenfeld, M. J. Merchant, H. M. Blau, M. Bogyo,
Nat. Chem. Biol. 2007, 3, 668–677.
[45] A. W. Puri, M. Bogyo, ACS Chem. Biol. 2009, 4, 603–616.
[46] a) M. Hu, L. Li, H. Wu, Y. Su, P.-Y. Yang, M. Uttamchandani, Q.-H.
Xu, S. Q. Yao, J. Am. Chem. Soc. 2011, 133, 12009–12020; b) Y.
Loh, H. Shi, M. Hu, S. Q. Yao, Chem. Commun. 2010, 46, 8407–
8409; c) J. Li, S. Q. Yao, Org. Lett. 2009, 11, 405–408.
[47] I. Coppens, P. J. Courtoy, Annu. Rev. Microbiol. 2000, 54, 129–156.
[48] O. Jacobson, X. Chen, Curr. Top. Med. Chem. 2010, 10, 1048–1059.
[49] H. Wu, J. Ge, S. Q. Yao, Angew. Chem. 2010, 122, 6678–6682;
Angew. Chem. Int. Ed. 2010, 49, 6528–6532.
[50] a) J. D. Bangs, L. Uyetake, M. J. Brickman, A. E. Balber, J. C. Boot-
hroyd, J. Cell Sci. 1993, 105, 1101–1113; b) R. F. Peck, A. M. Shi-
flett, K. J. Schwartz, A. McCann, S. L. Hajduk, J. D. Bangs, Mol. Mi-
crobiol. 2008, 68, 933–946; c) G. A. Keller, S. Krisans, S. J. Gould,
J. M. Sommer, C. C. Wang, W. Schliebs, W. Kunau, S. Brody, S. Sub-
ramani, J. Cell Biol. 1991, 114, 893–904.
[37] M. A. Albert, J. R. Haanstra, V. Hannacert, J. Van Roy, J. Biol.
Chem. 2005, 280, 28306–28315.
[38] S. Alsford, D. J. Turner, S. O. Obado, A. Sanchez-Flores, L. Glover,
M. Berriman, C. Hertz-Fowler, D. Horn, Genome Res. 2011, 21,
915–924.
[39] G. I. Lepesheva, R. D. Ott, T. Y. Hargrove, Y. Y. Kleshchenko, I.
Schuster, W. D. Nes, G. C. Hill, F. Villalta, M. R. Waterman, Chem.
Biol. 2007, 14, 1283–1293.
Received: February 14, 2012
Revised: March 22, 2012
Published online: June 1, 2012
Chem. Eur. J. 2012, 18, 8403 – 8413
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8413