Cinnamides as selective small-molecule inhibitors of a cellular model of breast cancer stem cells

Article abstract of DOI:10.1016/j.bmcl.2013.01.025

A high-throughput screen (HTS) was conducted against stably propagated cancer stem cell (CSC)-enriched populations using a library of 300,718 compounds from the National Institutes of Health (NIH) Molecular Libraries Small Molecule Repository (MLSMR). A cinnamide analog displayed greater than 20-fold selective inhibition of the breast CSC-like cell line (HMLE-sh-Ecad) over the isogenic control cell line (HMLE-sh-eGFP). Herein, we report structure-activity relationships of this class of cinnamides for selective lethality towards CSC-enriched populations.

Full text of DOI:10.1016/j.bmcl.2013.01.025

Bioorganic & Medicinal Chemistry Letters 23 (2013) 1834–1838
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Cinnamides as selective small-molecule inhibitors of a cellular model
of breast cancer stem cells
Andrew R. Germain a, Leigh C. Carmody a, Partha P. Nag a, Barbara Morgan a, Lynn VerPlank a,
Cristina Fernandez a, Etienne Donckele a, Yuxiong Feng b,c, Jose R. Perez a, Sivaraman Dandapani a,
Michelle Palmer a, Eric S. Lander a,c, Piyush B. Gupta b,c, Stuart L. Schreiber a,d, Benito Munoz a,
⇑
a Chemical Biology Platform and Probe Development Center, Broad Institute of MIT and Harvard, 7 Cambridge Center, Cambridge, MA 02142, USA
b Whitehead Institute for Biomedical Research, Cambridge, MA 02142, USA
c Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02142, USA
d Howard Hughes Medical Institute, Broad Institute of Harvard and MIT, 7 Cambridge Center, Cambridge, MA 02142, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A high-throughput screen (HTS) was conducted against stably propagated cancer stem cell (CSC)-
enriched populations using a library of 300,718 compounds from the National Institutes of Health
(NIH) Molecular Libraries Small Molecule Repository (MLSMR). A cinnamide analog displayed greater
than 20-fold selective inhibition of the breast CSC-like cell line (HMLE_sh_Ecad) over the isogenic control
cell line (HMLE_sh_eGFP). Herein, we report structure–activity relationships of this class of cinnamides
for selective lethality towards CSC-enriched populations.
Received 29 August 2012
Revised 2 January 2013
Accepted 7 January 2013
Available online 16 January 2013
Keywords:
Cancer stem cells
Ó 2013 Published by Elsevier Ltd.
Breast cancer
Molecular Libraries Probe Production Center
Network (MLPCN)
Epithelial-to-mesenchymal
transdifferentiation (EMT)
There is a growing consensus that many cancers, including
breast cancer, contain a small population of cancer stem-like cells
(CSCs) that are responsible for the initiation and propagation of
cancers.1 CSCs appear to be resistant to many current treatments
and may be the basis for all too frequent relapse.2 Since conven-
tional anti-cancer drugs do not specifically target CSCs, there exists
a need to find small molecules that are selectively toxic to CSCs.
Cell viability assays conducted directly on the bulk population of
tumor cells have not identified compounds selectively lethal to
CSCs since the latter constitute a small portion of the total popula-
tion of cancer cells.3 In theory, a phenotypic high-throughput
screen (HTS) conducted directly on CSCs with libraries of small
molecules would lead to useful starting points for developing ther-
apeutics against CSCs. Unfortunately, even after significant efforts,
isolation, characterization and culturing of CSCs is not yet
practical.4
has been developed.3,5 The assay used a stable cell line that was
induced to undergo epithelial-to-mesenchymal transition (EMT),
through the introduction of a short hairpin RNA targeting the
CDH1 gene, as a method to stably propagate CSC-enriched popula-
tions. The availability of essentially isogenic control cell line
(HMLE_sh_GFP), which is a mammary epithelial cell line which
has not gone through EMT, for the secondary validation assay of
the primary screen, minimized the probability of finding promiscu-
ous compounds that are not selective for CSCs but rather target ge-
netic differences between screened cell lines.
Compounds identified using this strategy were shown to selec-
tively eradicate CSCs in heterogeneous cancer cell populations that
were not experimentally induced into EMT.5
Recently, an in vitro assay measuring cell viability using surro-
gate CSC-like populations that can be maintained with high purity
O
N
S
N
H
⇑
Corresponding author.
E-mail addresses: benmz101@yahoo.com, bmunoz@broadinstitute.org (Benito
1
Munoz).
Figure 1. Prioritized cinnamide scaffold identified from HTS screen.
0960-894X/$ - see front matter Ó 2013 Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.bmcl.2013.01.025

A. R. Germain et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1834–1838
1835
O
1.
5
O
3
EtO
O
O
R2 NH2
R2
PPh3
H
OH
N
H
Benzene, 70°C
EDCI, DMAP
2. NaOH
iPr2NEt, DCM
R1
R1
4
R1
2
MeOH/CH2Cl2
Scheme 1. General reaction scheme of analog synthesis.
Table 1
Various amide core modifications
O
R
6
Selectivityc
a
Analog
R
EC50
(lM)
HMLE_sh_eGFP
HMLE_shEcad (control)
N
1
2.35
50.55
21.5
NA
S
S
N
H
6a
Inactiveb
Inactive
OH
N
6b
6c
21.10
13.33
Inactive
27.00
21.10 versus inactive
2.03
N
Me
O
N
S
N
H
N
6d
6e
6f
3.55
7.26
2.05
S
N
H
N
N
N
N
13.60
51.60
4.65
9.97
0.73
S
N
H
Inactive
5.97
51.60 versus inactive
N
H
N
H
6g
6h
1.28
0.95
N
N
H
Me
3.44
3.28
N
H
6i
59.65
123.3
2.07
NA
N
H
N
H
Me
Me
6j
Inactive
Inactive
Inactive
Inactive
N
6k
NA
Me
a
b
c
Average of at least three runs.
Inactive = No inhibition at highest dose (20
Selectivity = EC50 (HMLE_sh_eGFP)/EC50 (HMLE_shEcad).
lM).
Herein, we report the identification of cinnamides as a new
vided cinnamic acids 4 which upon coupling with various amines
afforded the cinnamide analogs 2.
class of small molecules that can selectively kill breast CSC-like
cells. These compounds should facilitate future studies of CSCs.
A high-throughput screen (HTS) of >300,000 compounds from
the NIH MLPCN collection was performed using the assay devel-
oped by Gupta, Lander and co-workers and all data was subse-
Results from cellular assays for the series are described in the
tables below. As can be seen in Table 1 none of the modifications
of the amide core on the eastern portion of the molecule was
tolerated. The free acid 6a was found to be inactive and methyla-
tion of the amide 6b results in a 10-fold decrease in potency. Intro-
duction of a carbonyl functionality in the thiazoline core 6c was
not tolerated. Replacement of the thiazoline with the correspond-
ing thiazole 6d or other five membered heteroaromatics 6e–g re-
sulted in a significant drop in potency and selectivity. The analog
with a larger aromatic group 6h resulted in a drop in potency
and selectivity. Alkyl amides 6i–k demonstrated poor or no po-
tency in cell based assays.
quently submitted to PubChem (AID 2127).6
A cinnamide
derivative 1 was one scaffold identified as selectively cytotoxic to
CSC-like cells and prioritized for further optimization (Fig. 1). The
structure–activity relationships of cinnamide analog 1 is the sub-
ject of this communication.7
Analogs of 2 were synthesized either from aldehydes 3 or com-
mercially available cinnamic acids 4, as illustrated in Scheme 1.
Treatment of 3 with Wittig reagent 5 followed by hydrolysis pro-

1836
A. R. Germain et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1834–1838
Table 2
Various olefin modifications
a
Analog
Structure
EC50
(l
M)
HMLE_sh_eGFP (control)
Selectivityc
9.91 versus inactive
NA
HMLE_shEcad
9.91
O
O
N
N
S
S
N
H
7
8
Inactiveb
Inactive
Me
Me
N
H
Inactive
Me
Me
O
N
S
S
N
H
9
Inactive
Inactive
Inactive
Inactive
NA
NA
Me
Me
CN
Me
O
N
O
N
N
H
10
Me
Me
N
S
Me
Me
N
H
11
12
Inactive
Inactive
Inactive
Inactive
NA
NA
Me
Me
S
N
O
N
H
O
S
N
H
13
Inactive
Inactive
NA
Me
Me
O
N
N
N
Me
S
N
H
14
15
Inactive
Inactive
Inactive
Inactive
Inactive
Inactive
NA
NA
NA
O
O
S
S
N
H
NH
O
N
H
16
N
Me
a
Average of at least three runs.
Inactive = No inhibition at highest dose (20
Selectivity = EC50 (HMLE_sh_eGFP)/EC50 (HMLE_shEcad).
b
c
lM).
The effect of modification of the olefin is illustrated in Table 2.
potency (EC50 = 9.91
lM), but complete selectivity over the control
As can be seen, modification of the olefin of the cinnamide core re-
sulted in a severe loss of activity. Changes in oxidation state (com-
cell line and a significant increase in PBS solubility (30.8 lM).
Other substitutions with methoxy groups resulted in compounds
with no activity 18b–e. Disubstitution with chloro-groups pro-
vided only inactive compounds 18f–h. Compounds with 2,4-disub-
stitution that contain either a larger alkoxy group 18i or i-propyl
group 18j at the 4-position were synthesized in an attempt to im-
prove the potency of the 2,4-dimethoxy compound while taking
advantage of its selectivity and solubility; however, these com-
pounds resulted in a loss of potency.
pounds
7 and 8) and substitution at the a- or b-position
(compounds 9, 10, and 11) failed to result in compounds with
activity. The cis-olefin 12 and cyclopropyl derivatives 13 were
found to be poorly active. Replacement of the cinnamic amide moi-
ety with different heteroaromatics such as indoles and benzofu-
rans did not result in compounds with useful activity against the
HMLE_shEcad cell line (compounds 14–16).
In addition to the unsubstituted phenyl 17a, we studied the ef-
fects of mono-substitution of chloro, methyl, methoxy, and cyano
at the ortho, meta, and para positions of the phenyl ring and these
results are summarized in Table 3. Removal of substituents from
the aromatic ring, 17a, resulted in a complete loss of activity. Sub-
stitution at the ortho-position or the meta-position of the phenyl
ring with several different substituents resulted in mostly inactive
compounds, while substitution at the para-position provided com-
pounds with moderate activity.
Based on the SAR describe above, we further investigated thiaz-
oline cinnamides with substitutions at the para-position and these
results are presented in Table 5. Two compounds that retained
activity and full inhibition are the 4-ethyl analog 19d and the 4-
i-butyl compound 19f. The ethyl analog 19d resulted in a slight in-
crease in potency, good improvement in selectivity and clear
enhancement in solubility (11.5 lM for 19d vs <1 lM for the hit
compound 1). The 4-i-butyl compound 19f resulted in a significant
increase in potency with a small loss of selectivity. Larger substit-
uents provided potent compounds, such as, 4-phenyl substitution
19g, which is slightly more potent and selective than 1; however,
it was observed to be a partial inhibitor possibly due to low solu-
The cellular data from analogs with a di- or tri-substituted phe-
nyl ring on the western portion of the molecule is described in Ta-
ble 4. 2,4-Dimethoxy substitution 18a resulted in a decrease in

A. R. Germain et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1834–1838
1837
Table 3
Table 4
SAR of monosubstituted phenyl cinnamides
SAR of di- and tri-substituted phenyl cinnamides
O
N
O
N
S
S
N
H
R
N
H
9
7
R
Selectivityc
a
Analog
R
EC50
(
l
M)
Analog
R
EC50as (
lM)
Selectivityc
HMLE_shEcad HMLE_sh_eGFP
(control)
HMLE_shEcad HMLE_sh_eGFP
(control)
OMe
17a
17b
H
2-Cl
Inactiveb
35.0
Inactive
Inactive
NA
9.91 versus
inactive
18a
18b
9.91
Inactiveb
Inactive
35.0 versus
inactive
1.44
0.73
NA
OMe
17c
17d
17e
17f
17g
3-Cl
4-Cl
2-Me
3-Me
4-Me
14.95
13.60
Inactive
Inactive
16.00
21.5
9.97
Inactive
Inactive
Inactive
OMe
Inactive
NA
NA
OMe
OMe
NA
16.00 versus
inactive
NA
18c
Inactive
Inactive
17h
17i
17j
2-
OMe
3-
OMe
4-
Inactive
62.15
93.3
OMe
OMe
96.20
Inactive
1.55
NA
OMe
18d
18e
Inactive
Inactive
Inactive
Inactive
NA
NA
OMe
OMe
Inactive
OMe
2-CN
3-CN
4-CN
17k
17l
17m
Inactive
Inactive
26.93
Inactive
Inactive
Inactive
NA
NA
OMe
OMe
Cl
26.93 versus
inactive
18f
Inactive
Inactive
Inactive
Inactive
NA
NA
a
b
c
Average of at least three runs.
Inactive = no inhibition at highest dose (20
Selectivity = EC50 (HMLE_sh_eGFP)/EC50 (HMLE_shEcad).
Cl
Cl
lM).
18g
Cl
Cl
bility. In an effort to overcome the low solubility, related pyridine
analog 19h was synthesized. Although full inhibition was observed
with 19h, loss of selectivity prohibits further advancement.
Based on the SAR described above, the 4-ethyl substituted ana-
log 19d was designated as a probe with the NIH Molecular Li-
braries Program (ML243) due to its optimal potency, excellent
selectivity, and solubility profile.
18h
18i
18j
Inactive
Inactive
28.43
Inactive
Inactive
Inactive
NA
NA
Cl
OMe
Me
Me
O
OMe
The solubility of 19d was experimentally determined to be
11.5 lM in phosphate-buffered saline (PBS, pH 7.4, 23 °C) solution
28.43 versus
inactive
Me
Me
(Table 6). Plasma protein binding (PPB) was determined to be
96.6% bound in human plasma. The probe is stable in human plas-
ma, with approximately 97.1% remaining after a 5-h incubation
period. The compound was found to be stable in glutathione
(GSH) with 82% remaining after 48 h. The stability of 19d in PBS
(0.1% DMSO) was measured. More than 75% was present after
48 h of incubation.
Compound 19d was screened against a panel of 68 different tar-
gets that are commonly used in drug discovery for lead profiling.
The assays were done by Ricerca Biosciences, LLC and include tar-
gets from various areas such as, GPCRs, ion channels, and trans-
porters. Compound 19d was only active in one of the 68 assays,
where 19d blocked a radio-labeled ligand to the Adenosine A2A
a
b
c
Average of at least three runs.
Inactive = no inhibition at highest dose (20 lM).
Selectivity = EC50 (HMLE_sh_eGFP)/EC50 (HMLE_shEcad).
quest. Studies are currently underway to determine the ability of
this compound to inhibit other breast cancer stem cell-like cell
lines along with tumor cell lines with naturally high populations
of cancer stems cells. Studies are also underway in combination
with other scaffolds identified in this screen7 utilizing gene expres-
sion analysis to understand the mode of action of these compounds
and to identify potential targets for selectivity toward cancer stem
cells.
Receptor by 51% at 10 l
M.8 As the assay used to identify this scaf-
fold was phenotypic, the activity against Adenosine A2A must be
considered as these compounds are used in efforts to identify a po-
tential target.
The current work evaluated over 300,000 compounds of the
NIH-MLPCN collection to identify agents capable of selectively
inhibiting a cell-line model of breast cancer stem cells. After triage
through a series of counter screens and an orthogonal assay, the 4-
i-propyl cinnamide 1 was selected for further SAR investigation. As
a result of these synthetic studies, compound 19d emerged as a po-
tent and selective inhibitor. Compound 19d inhibited CSC-like cells
Acknowledgments
The authors are grateful to Ms. Patti Aha and Ms. Lori Thomae
for their assistance in manuscript preparation. This work was
funded by the NIH-MLPCN program (1 U54 HG005032-1 awarded
to S.L.S.).
Supplementary data
Supplementary data (experimental protocols for cellular assays
and for the preparation of compounds) associated with this article
can be found, in the online version, at http://dx.doi.org/10.1016/
j.bmcl.2013.01.025.
with an IC50 of 2.00 lM and shows 32-fold selectivity against the
control cell line. This compound has been registered with NIH
Molecular Libraries Program (ML243) and is available upon re-

1838
A. R. Germain et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1834–1838
Table 5
Further SAR of para-substituted phenyl cinnamides
O
N
S
N
H
8
R
Selectivityc
a
Analog
R
EC50 (lM)
HMLE_shEcad
HMLE_sh_eGFP (control)
1
i-Pr
F
CF3
NMe2
Et
t-Bu
i-Bu
Ph
2.35
50.55
21.5
NA
NA
19a
19b
19c
19d
19e
19f
19g
Inactiveb
Inactive
27.4
Inactive
Inactive
Inactive
63.98
20.55
2.17
27.4 versus inactive
32.0
NA
4.54
53.1
2.00
Inactive
0.478
0.65
34.6
N
19h
3.85
5.65
1.47
a
b
c
Average of at least three runs.
Inactive = no inhibition at highest dose (20
Selectivity = EC50 (HMLE_sh_eGFP)/EC50 (HMLE_shEcad).
lM).
Table 6
Physicochemical characterization of ML243
O
N
S
N
H
Me
19d (ML243)
PBS solubility
11.5
PPB (human)
96.6%
Plasma stability (human)
97.1%
PBS stability (48 h)
75%
GSH stability (48 h)
82%
Lead profiling screen
1/68 Targetsa
l
M
a
See Supplementary data for further details.
4. Dontu, G.; Abdallah, W. M.; Foley, J. M.; Jackson, K. W.; Clarke, M. F.; Kawamura,
M. J.; Wicha, M. S. Genes Dev. 2003, 17, 1253.
References and notes
5. Gupta, P. B.; Onder, T. T.; Jiang, G.; Tao, K.; Kuperwasser, C.; Weinberg, R. A.;
Lander, E. S. Cell 2009, 138, 645.
1. Nguyen, L. V.; Vanner, R.; Dirks, P.; Eaves, C. J. Nat. Rev. Cancer 2012, 12, 133. and
references therein.
6. (a) Carmody, L. C.; Germain, A. R.; VerPlank, L.; Nag, P. P.; Munoz, B.; Perez, J. R.;
Palmer, M. A. J. Biomol. Screen. 2012, 17, 1204; (b) All data for this project can be
found at: http://pubchem.ncbi.nlm.nih.gov/assay/assay.cgi?aid=2721.
7. Germain, A. R.; Carmody, L. C.; Morgan, B.; Fernandez, C.; Forbeck, E.; Lewis, T.
A.; Nag, P. P.; Ting, A.; Verplank, L.; Feng, F.; Perez, J. R.; Dandapani, S.; Palmer,
M.; Lander, E. S.; Gupta, P. B.; Schreiber, S. L.; Munoz, B. Bioorg. Med. Chem. Lett.
2012, 22, 3571.
2. (a) Bao, S.; Wu, Q.; McLendon, R. E.; Hao, Y.; Shi, Q.; Hjelmeland, A. B.; Dewhirst,
M. W.; Bigner, D. D.; Rich, J. N. Nature 2006, 444, 756; (b) Diehn, M.; Cho, R. W.;
Lobo, N. A.; Kalisky, T.; Dorie, M. J.; Kulp, A. N.; Qian, D.; Lam, J. S.; Ailles, L. E.;
Wong, M. Nature 2009, 458, 780; (c) Diehn, M.; Clarke, M. F. J. Natl. Cancer Inst.
2006, 98, 1755; (d) Eyler, C. E.; Rich, J. N. J. Clin. Oncol. 2008, 17, 2839; (e) Li, X.;
Lewis, M. T.; Huang, J.; Gutierrez, C.; Osborne, C. K.; Wu, M. F.; Hilsenbeck, S. G.;
Pavlick, A.; Zhang, X.; Chamness, G. C. J. Natl. Cancer Inst. 2008, 100, 672.
3. Mani, S. A.; Guo, W.; Liao, M. J.; Eaton, E. N.; Ayyanan, A.; Zhou, A. Y.; Brooks, M.;
Reinhard, F.; Zhang, C. C.; Shipitsin, M.; Campbell, L. L.; Polyak, K.; Brisken, C.;
Yang, J.; Weinberg, R. A. Cell 2008, 133, 704.
8. See Supplementary data.