Click-based synthesis and proteomic profiling of lipstatin analogues

Article abstract of DOI:10.1039/c0cc01276a

Using click chemistry to enable both structural diversity and proteome profiling within a natural product derived library, two out of nineteen lipstatin analogues showed similar activity to Orlistat against fatty acid synthase (FAS), but with an improved

Full text of DOI:10.1039/c0cc01276a

Chemical Communications
www.rsc.org/chemcomm
Volume 46 | Number 44 | 28 November 2010 | Pages 8293–8488
ISSN 1359-7345
COMMUNICATION
Yao and Lear et al.
Click-based synthesis and proteomic profiling of lipstatin analogues
1359-7345(2010)46:44;1-C

View Article Online
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Click-based synthesis and proteomic profiling of lipstatin analoguesw
Mun H. Ngai,a Peng-Yu Yang,a Kai Liu,b Yuan Shen,b Markus R. Wenk,bc Shao Q. Yao*ab
and Martin J. Lear*a
Received 7th May 2010, Accepted 10th June 2010
DOI: 10.1039/c0cc01276a
Using click chemistry to enable both structural diversity and
proteome profiling within a natural product derived library,
two out of nineteen lipstatin analogues showed similar activity
to Orlistat against fatty acid synthase (FAS), but with an
improved ability to induce tumour cell death.
combination of a rapid synthesis of THL-like compounds
(using click chemistry) and a cell-based chemical proteomic
approach to enable the direct activity-based protein profiling
(ABPP) of potential cellular targets (on and off) of THL,
1 (Fig. 1).5
The design of our lipstatin-derived library is based on the
general structure of THL (1) and the success of our THL-R
alkyne-probe 2 (Fig. 1).4 The b-lactone pharmacophore and
amino ester would thus be preserved (to maintain sufficient
binding/activity to FAS) and a terminal alkyne handle
introduced to the right aliphatic side chain (for identification
of cellular targets via the downstream conjugation to reporter
tags).6,7 A left alkyne-terminated aliphatic side chain would
then be diversified through a triazole ring (cf. 3), via click
chemistry, with a variety of aliphatic/aromatic azides.8 Here,
we anticipated the known functional group compatibility of
click conditions would tolerate a relatively reactive b-lactone
component.6 We were, however, uncertain whether the
triazole moiety in 3 would engage in favourable or unfavourable
interactions with FAS, thereby influencing the potency and
specificity.
Tetrahydrolipstatin (THL, 1; Fig. 1) is an FDA-approved
antiobesity drug called Orlistat, which works primarily on
pancreatic and gastric lipases within the gastrointestinal (GI)
tract.1 Among other lipstatins, THL (1) has recently shown
promising activity against off-targets such as human fatty acid
synthase (FAS), an enzyme essential for the survival of tumour
cells.2 There are, however, a number of challenges that prevent
1 from being considered as a potential antitumour drug; it has
poor solubility and bioavailability,3 and, most importantly,
lacks sufficient potency and specificity.4 Recognising the need
to address these impending issues, we decided to search
for better lipstatin analogues that might possess similar
(or improved) potency against FAS, and at the same time
affect fewer cellular off-targets.4 Herein, we describe the
identification of two such compounds, made possible by the
Our synthesis parallels the elegant work of Romo, Smith
and co-workers, who adopted olefin metathesis to diversify an
alkene lipstatin analogue.3 Herein, we targeted a di-alkyne
lipstatin analogue (9) as our chemical progenitor (Scheme 1).
Starting from 4-pentynal 4, Keck allylation9 with allyl-
tributyltin (in the presence of a titanium-(S)-BINOL complex)
afforded the chiral homoallylic alcohol (R)-5 in an excellent ee
of 497% (cf. modified Mosher esters analysis; Scheme S1 and
Fig. S1w). The alcohol 5 was then TBS-protected and the olefin
oxidatively cleaved with potassium osmate and sodium periodate
to afford the aldehyde 6.10 The key ZnCl2-mediated, tandem
Mukaiyama aldol-lactonisation (TMAL)3 between the
b-hydroxy aldehyde 6 and the thiopyridyl ketene acetal 7
(prepared in four steps from 6-bromohexanoic acid;
Scheme S2w) afforded the trans-b-lactone 8 as separable 9 : 1
anti/syn diastereomers in 39% yield after desilylation.
1H NMR analysis of the b-lactone ring protons showed a
characteristic coupling constant of 3.8 Hz for a trans-b-lactone
Fig. 1 Click-based design of lipstatin analogues 3.
3
(3Jcis = 6.5 Hz and Jtrans = 4–4.5 Hz).11 Formation of the
b-lactone was further confirmed by the presence of a strong IR
absorption band at 1818 cmÀ1. Subsequent Mitsunobu
reaction of the readily separable, major anti-hydroxy-b-lactone
8 with N-formyl-L-leucine afforded the desired di-alkyne
b-lactone 9 in 66% yield. Next, several click chemistry
conditions were examined;12 it was found that triazole formation
was highly sensitive to the catalyst, additives and solvents used
(Supporting Informationw). Upon optimisation, we found the
a Department of Chemistry, and Medicinal Chemistry Program of the
Life Sciences Institute, National University of Singapore,
3 Science Drive 3, Singapore 117543.
E-mail: Martin.Lear@nus.edu.sg; Tel: (+65) 65163998
b Department of Biological Sciences, National University of Singapore,
14 Science Drive 4, Singapore 117543.
E-mail: chmyaosq@nus.edu.sg; Fax: (+65) 67791691;
Tel: (+65) 65162925
c Department of Biochemistry, Yong Loo Lin School of Medicine,
National University of Singapore, Singapore 117597
w Electronic supplementary information (ESI) available: Experimental
details. See DOI: 10.1039/c0cc01276a
i
use of CuI (0.4 equiv.) and Pr2NEt (0.8 equiv.) in DCE/H2O
(1 : 1) gave the desired triazole-adduct with complete
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 8335–8337 | 8335

View Article Online
(highlighted in blue), comparable to that seen with THL (1)
and THL-R (2) (highlighted in red). In contrast, other library
members (i.e. those derived from aliphatic, PEG, and carbonyl
azides), including the intermediate 9, showed a significantly reduced
anti-proliferative activity. We next carried out comparative analysis
with two of the four hits (3k and 3s) by measuring their ability to
induce phosphorylation of eIf2a in prostate cancer PC-3 cells
(Fig. 2C); PC-3 cells treated with different amounts of each
compound showed similarly elevated eIf2a phosphorylation.
Lastly, we tested the compounds against the invasive human breast
cancer MCF-7 cells (Fig. 2D); similar degrees of caspase-8
activation were observed (at either 5 or 25 mM). Collectively, all
these lines of evidence indicate that both 3k and 3s are at least as
potent as THL (1) and THL-R (2) as potential antitumour agents.
We next compared the in situ proteome reactivity profiles of
the hits against THL-R (2; a close mimic of Orlistat) to directly
reveal the potential range of cellular protein targets. Ideally,
for example,
a hit compound should predominantly
(i.e. specifically) target FAS and no other protein. Live HepG2
cells were thus directly treated with each of the four THL
analogues (3k, 3m, 3r and 3s; 20 mM final concentration),
either alone or in the presence of the THL competitor 1 (over a
concentration range of 50–200 mM).3 The cells were
subsequently washed (to remove excessive probes), homogenised,
incubated with rhodamine-azide under click-chemistry
conditions,4 separated by SDS-PAGE gel, and analysed by
in-gel fluorescence scanning (Fig. 2E; left gel).
All hit compounds (except 3s, in which the FAS band was
significantly weakened) labeled FAS (the 265 kDa band) as
efficiently as THL-R (2), indicating they likely possess similar
potency as THL (1) in targeting FAS. The proteins involved
were indicated to be Orlistat-specific since competition studies
with THL (1) (center gel in Fig. 2E) or hydroxylamine
displayed diminished band-labeling levels (see Supporting
Informationw). Similar proteome labeling profiles, albeit with
significantly higher background levels, were also obtained
when cell lysates were used (right gel in Fig. 2E). This under-
scores the importance of in situ labeling as a prerequisite for
accurate and specific identification of true cellular targets.
In summary, a small natural product-based library of
lipstatin analogues (i.e. 3a–s) was synthesised and tested for
cellular activity and protein specificity. The synthesis of the
library showcases the chemical compatibility of click
conditions to tether various aliphatic/aromatic azides with a
relatively labile b-lactone lipstatin core (9). It is noted that,
though click chemistry has previously been used in many other
biological applications,7 our current report is one of the few
examples7c where it has been successfully implemented to
facilitate the rapid lead identification of natural product-like
compounds. The introduction of a triazole ring into the
aliphatic side chain of the analogues 3 appeared to be tolerated
in many cases by FAS (a validated THL enzyme target). Our
current approach highlights the dual use of click chemistry to
not only diversify a compound library in an expedient manner,
but also allow the rapid activity-based protein profiling
(ABPP) and identification of cellular targets of hit compounds.
From this study, we have tentatively identified four candidates
(3k/m/r/s) to possess similar anti-proliferative activity against
HepG2 cells as compared to Orlistat (THL, 1).
Scheme 1 Synthesis of b-lactone 9 and lipstatin analogues 3.
consumption of the alkyne in high purity and within 24 h.
A total of 19 aromatic and aliphatic azides (10a–s) with different
electronic properties were used (cf. R1 below Scheme 1) to give
the TMS-protected click compounds, 11a–s. Subsequent
C-desilylation of the crude click adducts with AgNO3/H2O gave
the desired THL analogue library, 3a–s.13
With nineteen THL analogues (3a–s) in hand, we next studied
their biological activities (Fig. 2). On the basis of established
lipstatin biology,3,4 three cellular assays were evaluated for cell
proliferation, eIf2a phosphorylation and caspase-8 activation.
For the proliferation assay, a human hepatocellular liver
carcinoma cell line (HepG2) was selected. As shown in
Fig. 2A, the 19 THL compounds, as well as the wildtype THL
(1), THL-R (2) and library progenitor 9, showed differential
degrees of activity. The azides 10 alone did not show any cellular
activity even up to 50 mM (see 10k in Fig. 2A as a representative
example). Four compounds from the library (Fig. 2B),
derived from two sulfonamide-containing azides (3k and 3m)
and two aromatic azides (3r and 3s), respectively, displayed the
highest activity in blocking the proliferation of HepG2 cells
ꢀc
This journal is The Royal Society of Chemistry 2010
8336 | Chem. Commun., 2010, 46, 8335–8337

View Article Online
Fig. 2 (A) Dose-dependent inhibition of HepG2 cell proliferation by THL (1), THL-R (2), and triazole analogues 3a–s using XTT assay.4
Data represent the average s.d. for three trials. (B) Hits identified from screening. (C) Western blot analysis of eIf2a phosphorylation in PC-3 cells
upon treatment with 2 and 3k/s. GAPDH was used as a loading control. (D) Activation of caspase-8 in MCF-7 cells treated with the indicated
concentrations of 2 and 3k/s. (E) Proteome profiling of HepG2 cells using 1, 2 and 3k/m/r/s. Both in situ (live cell) and in vitro (whole-cell) lysate
labelings were carried out. In situ proteome labeling of HepG2 cells, with or without Orlistat (over a concentration range of 50–200 mM), followed
by click chemistry with rhodamine-azide, SDS-PAGE analysis, and in-gel fluorescence scanning (for in situ dose-dependent profile and in-gel
hydroxylamine treatment profiles, see Fig. S2).
We thank the National University of Singapore for
Graduate Scholarships (to M.H.N., P.Y. and K.L.) and the
National Research Foundation, Competitive Research
Program (NRF-G-CRP 2007-04) for generous funding.
3 (a) G. Ma, M. Zancanella, Y. Oyola, R. D. Richardson,
J. W. Smith and D. Romo, Org. Lett., 2006, 8, 4497–4500;
(b) R. D. Richardson, G. Ma, Y. Oyola, M. Zancanella,
L. M. Knowel, P. Cieplak, D. Romo and J. W. Smith, J. Med.
Chem., 2008, 51, 5285–5296.
4 P.-Y. Yang, K. Liu, M. H. Ngai, M. J. Lear, M. R. Wenk and
S. Q. Yao, J. Am. Chem. Soc., 2010, 132, 656–666.
5 (a) M. Uttamchandani, J. Li, H. Sun and S. Q. Yao,
ChemBioChem, 2008, 9, 667–675; (b) M. J. Evans and
B. F. Cravatt, Chem. Rev., 2006, 106, 3279–3301.
Notes and references
1 R. Guerciolini, Int. J. Obes. Relat. Metab. Disord., 1997,
21(Suppl 3), S12–23.
2 (a) S. J. Kridel, F. Axelrod, N. Rozenkrantz and J. W. Smith, Cancer
Res., 2004, 64, 2070–2075; (b) C. W. Pemble, L. C. Johnson,
S. J. Kridel and W. T. Lowther, Nat. Struct. Mol. Biol., 2007, 14,
704–709; (c) J. L. Little, F. B. Wheeler, D. R. Fels, C. Koumenis and
S. J. Kridel, Cancer Res., 2007, 67, 1262–1269; (d) L. M. Knowles,
F. Axelrod, C. D. Browne and J. W. Smith, J. Biol. Chem., 2004, 279,
30540–30545; (e) L. M. Knowles, C. Yang, A. Osterman and
J. W. Smith, J. Biol. Chem., 2008, 283, 31378–31384;
(f) J. A. Menendez, L. Vellon and R. Lupu, Ann. Oncol., 2005, 16,
1253–1267; (g) J. L. Little, F. B. Wheeler, C. Koumenis and
S. J. Kridel, Mol. Cancer Ther., 2008, 7, 3816–3824; (h) S. J. Kridel,
W. T. Lowther and C. W. Pemble, Expert Opin. Invest. Drugs, 2007,
16, 1817–1829; (i) J. A. Menendez, L. Vellon and R. Lupu, Exp. Biol.
Med., 2005, 230, 151–154; (j) M. A. Carvalho, K. G. Zecchin,
F. Seguin, D. C. Bastos, M. Agostini, A. L. C. A. Rangel,
S. S. Veiga, H. F. Raposo, H. C. F. Oliveira, M. Loda,
R. D. Coletta and E. Graner, Int. J. Cancer, 2008, 123, 2557–2565;
(k) J. A. Menendez, L. Vellon and R. Lupu, Int. J. Gynecol. Cancer,
2006, 16, 219–221; (l) C. D. Browne, E. J. Hindmarsh and
J. W. Smith, FASEB J., 2006, 20, 2027–2035; (m) F. Cheng,
Q. Wang, M. Chen, F. A. Quiocho and J. Ma, Proteins: Struct.,
Funct., Bioinf., 2008, 70, 1228–1234.
6 (a) E. M. Sletten and C. R. Bertozzi, Angew. Chem., Int. Ed., 2009,
48, 6974–6998; (b) K. A. Kalesh, P.-Y. Yang, R. Srinivasan and
S. Q. Yao, QSAR Comb. Sci., 2007, 26, 1135–1144.
7 (a) A. Brik, C.-Y. Wu and C.-H. Wong, Org. Biomol. Chem., 2006,
4, 1446–1457; (b) K. A. Kalesh, H. Shi, J. Ge and S. Q. Yao,
Org. Biomol. Chem., 2010, 8, 1749–1762; (c) M. Morten Meldal
and C. W. Tornøe, Chem. Rev., 2008, 108, 2952–3015.
8 R. Srinivasan, L. P. Tan, H. Wu, P.-Y. Yang, K. A. Kalesh and
S. Q. Yao, Org. Biomol. Chem., 2009, 7, 1821–1828.
9 (a) G. E. Keck, K. H. Tarbet and L. S. Geraci, J. Am. Chem. Soc.,
1993, 115, 8467–8468; (b) R. Garcia-Fandino, M. J. Aldegunde,
E. M. Codesido, L. Castedo and J. R. Granja, J. Org. Chem., 2005,
70, 8281–8290.
10 Y. Mori, Y. Kohchi, M. Suzuki, S. Carmeli, R. E. Moore and
G. M. L. Patterson, J. Org. Chem., 1991, 56, 631–637.
11 (a) W. Adam and V. O. Navasalgado, J. Org. Chem., 1995,
60, 578–584; (b) A. Pommier and J. M. Pons, Synthesis, 1993,
441–459.
12 R. Srinivasan, J. Li, S. L. Ng, K. A. Kalesh and S. Q. Yao,
Nat. Protoc., 2007, 2, 2655–2664.
13 A. Carpita, L. Mannocci and R. Rossi, Eur. J. Org. Chem., 2005,
1859–1864.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 8335–8337 | 8337