Pro-apoptotic carboxamide analogues of natural fislatifolic acid targeting Mcl-1 and Bcl-2

Article abstract of DOI:10.1016/j.bmcl.2020.127003

A library of 26 novel carboxamides deriving from natural fislatifolic acid has been prepared. The synthetic strategy involved a bio-inspired Diels-Alder cycloaddition, followed by functionalisations of the carbonyl moiety. All the compounds were evaluated on Bcl-xL, Mcl-1 and Bcl-2 proteins. In this series of cyclohexenyl chalcone analogues, six compounds behaved as dual Bcl-xL/Mcl-1 inhibitors in micromolar range and one exhibited sub-micromolar affinities toward Mcl-1 and Bcl-2. The most potent compounds evaluated on A549 and MCF7 cancer cell lines showed moderate cytotoxicities.

Full text of DOI:10.1016/j.bmcl.2020.127003

Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pro-apoptotic carboxamide analogues of natural fislatifolic acid targeting
Mcl-1 and Bcl-2
Shelly Gapil Tiamasa,b, Florian Daressya,c, Alma Abou Samraa, Jérome Bignona,
Vincent Steinmetza, Marc Litaudona, Christophe Fourneaud, Kok Hoong Leonge, Azhar Ariffinb,
⁎
⁎
Khalijah Awangb, Sandy Desrata, , Fanny Roussia,
a Université Paris-Saclay, CNRS, Institut de Chimie des Substances Naturelles, UPR 2301, 91198 Gif-sur-Yvette, France
b Department of Chemistry, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia
c Université Paris-Saclay, UMR CNRS 8126, Institut Gustave Roussy, 114 rue Edouard-Vaillant, 94805 Villejuif Cedex, France
d Université Paris-Saclay, BioCIS, Faculté de Pharmacie de Châtenay-Malabry, 5 rue Jean-Baptiste Clément, 92296 Châtenay-Malabry, France
e Department of Pharmacy, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia
A R T I C L E I N F O
A B S T R A C T
Keywords:
A library of 26 novel carboxamides deriving from natural fislatifolic acid has been prepared. The synthetic
strategy involved a bio-inspired Diels-Alder cycloaddition, followed by functionalisations of the carbonyl moiety.
All the compounds were evaluated on Bcl-xL, Mcl-1 and Bcl-2 proteins. In this series of cyclohexenyl chalcone
analogues, six compounds behaved as dual Bcl-xL/Mcl-1 inhibitors in micromolar range and one exhibited sub-
micromolar affinities toward Mcl-1 and Bcl-2. The most potent compounds evaluated on A549 and MCF7 cancer
cell lines showed moderate cytotoxicities.
Natural product
Diels-Alder
Carboxamide
Pro-apoptotic
Mcl-1
Mcl-1 and Bcl-2 are pro-survival members of the Bcl-2-like proteins
family that play a key role in the mitochondrial apoptosis pathway. This
family of proteins is divided in three classes including multi-domain
pro-apoptotic proteins such as Bak and Bax; BH3-only pro-apototic
proteins like Bid, Bim, Bad and Noxa; and anti-apopotic proteins as Mcl-
1, Bcl-2 and Bcl-xL. These anti-apoptotic proteins are frequently upre-
gulated in cancer cells by inhibiting apoptosis after the sequestration of
BH3-only pro-apototic proteins.1 Therefore, the development of anti-
apopotic proteins antagonist has become a promising alternative for the
treatment of various cancers.2 In recent years several therapeutic agents
have been developed with different selectivity toward Mcl-1 and Bcl-2.
ABT-7373 and ABT-2634 (Fig 1), active at subnanomolar level on Bcl-xL
and Bcl-2 but not on Mcl-1, were found to cause severe thrombocyto-
penia.5 Moreover, resistance to ABT-737 has been observed due to up-
regulation of Mcl-1. A series of clinical candidate targeting Bcl-2 have
been discovered including ABT-1996 (Fig 1) launched on the market.
Some potent Mcl-1 inhibitors have also entered in clinical trial.7 Recent
studies revealed that combined inhibition of Mcl-1 and Bcl-2 may be a
relevant strategy for the treatment of cancers, and especially by de-
veloping Mcl-1/Bcl-2 dual inhibitor.8
Recently, a bioassay-guided purification of Fissistigma latifolium bark
extract using an in-vitro assay based on the modulation of Bcl-xL/Bak
and Mcl-1/Bid interactions led to the isolation of new cyclohexenyl
chalcones including fislatifolic acid 1 and fislatifolione 3 (Fig 2).9h
While the latter was not active on Bcl-2 family proteins, fislatifolic acid
1 revealed to be a good Bcl-xL/Mcl-1 dual inhibitor on micromolar
range. Later on, an asymmetric total synthesis of both enantiomers of
these natural myrcene-derived cyclohexenyl chalcones has been de-
veloped based on a bio-inspired asymmetric Diels-Alder cycloaddition
using a chiral Evans oxazolidinone.10f The affinity of the synthetic cy-
clohexyl chalcones 1 and 3 toward Bcl-2 family protein has been con-
firmed and the Weinreb amide intermediate 2 proved to be an excellent
Mcl-1/Bcl-2 dual inhibitor on submicromolar level (Fig. 2). This work
has shown that minor modifications on the carbonyl moiety could have
a significant impact on the activity of the compounds. These first results
encouraged us to explore pharmacomodulations on this part of the
molecule. In this study, we have developed a series of carboxamide
analogues of (+)- and (−)-fislatifolic acid 1. These compounds have
been evaluated by in-vitro affinity displacement assays based on the
modulation of Bcl-xL/Bak, Mcl-1/Bid, and Bcl-2/Bim interactions.
The synthesis of (+)- and (−)-fislatifolic acid 1 was achieved on 5 g
scale in 3 steps in 54-57% overall yield (Scheme 1). First, the
Our group has a longstanding interest in the isolation9 and synth-
esis10 of natural compounds that target anti-apopotic proteins.
⁎ Corresponding authors.
E-mail addresses: sandy.desrat@cnrs.fr (S. Desrat), fanny.roussi@cnrs.fr (F. Roussi).
https://doi.org/10.1016/j.bmcl.2020.127003
Received 18 December 2019; Received in revised form 30 January 2020; Accepted 1 February 2020
0960-894X/©2020ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:ShellyGapilTiamas,etal.,Bioorganic&MedicinalChemistryLetters,https://doi.org/10.1016/j.bmcl.2020.127003

S. Gapil Tiamas, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Fig. 1. Structures of ABT-737, ABT-263 and ABT-199.
to mimic other natural myrcene-derived cyclohexenyl chalcones like
ecarlottone9h or nicolaioidesin C.11 Thus, the use of β-alanine methyl
ester, methyl 4-aminobenzoate, and 4-anisidine led to carboxamide
analogues (+)- and (−)-4f-h in acceptable yields.
Saponification of ester function of (+)- and (−)-4f and 4g in pre-
sence of lithium hydroxide furnished the corresponding carboxylic
acids (+)- and (−)-4i and 4j in decent yields (Scheme 3). The release of
the phenol moiety of carboxamides (+)-4h was performed using stoe-
chiometric amount of boron tribromide (Scheme 4). Although de-
methylation was observed by proton NMR analysis, the expected pro-
duct (+)-4k was not formed. Indeed a cyclization of the 1,5-diene
occurred leading mainly to the bicylo compound (+)-8. This reaction
mediated by Brönsted or Lewis acid was studied more than 60 years by
Eschenmoser, Stork and others12 and is still employed nowadays.13 In
order to enrich our library of analogues, we decided to apply this cy-
clisation reaction on (+)- and (−)-fislatifolic acid 1. In presence of
boron trifluoride diethyl etherate, the cyclisation products (+)- and
(−)-9 could be isolated in 69-73% yield (Scheme 4). The corresponding
Weinreb amides (+)- and (−)-9 were then synthesized to compare
their affinity to Bcl-2 family proteins to the most active compound
(+)-2 of our series. The demethylated compounds (+)- and (−)-4k
were finally obtained by coupling between 4-aminophenol to (+)- and
(−)-fislatifolic acid 1 using the same conditions described above
(Scheme 5).
Fig. 2. From fislatifolic acid 1 toward the synthesis of carboxamides 4 as Mcl-1/
Bcl-2 dual inhibitors.
oxazolidinones (+)-6 and (−)-6 were prepared separately from com-
mercially available trans-cinnamic acid 5 and either (4S)- or (4R)-4-
benzyl-1,3-oxazolidin-2-one respectively. Then a dimethylaluminium
chloride-mediated Diels-Alder cycloaddition in presence of myrcene
gave the expected cycloadducts (+)-7 and (−)-7 in 82-85% yield with
a complete regio-, diastereo- and endo/exo-selectivity (> 95:5). Fi-
nally, (+)- and (−)-fislatifolic acid 1 were obtained after removal of
chiral auxiliary (Scheme 1).
All the products synthesized were evaluated by in-vitro affinity
displacement assays based on the modulation of Bcl-xL/Bak, Mcl-1/Bid,
and Bcl-2/Bim interactions (Table 1). Although less active, the car-
boxamides of methoxyamine, N-methylhydroxylamine and hydro-
xylamine (+)- and (−)-4a-c showed the same selectivity toward Mcl-1
and Bcl-2 than Weinreb amides (+)-2 and (−)-2. As previously no-
ticed, in this series the (+) enantiomers were found as or more active
than the (−) enantiomers. Compounds (+)- and (−)-4c even exhibited
sub-micromolar affinity to Bcl-2. On the contrary, both enantiomers of
compounds 4d and 4e revealed almost no activity on the three target
With carboxylic acid (+)- and (−)-1 in hands, various analogues
were prepared following the procedure used to obtain the Weinreb
amide intermediates (+)-2 and (−)-2 throughout the synthesis of
natural (+)- and (−)-fislatifolione 3.10f First, to understand the role of
the amides deriving from N,O-dimethylhydroxylamine in the interac-
tion with the target proteins and to modulate their affinity, various
Weinreb amide analogues (+)- and (−)-4a-e were prepared from
methoxyamine, N-methylhydroxylamine, hydroxylamine, ammonia,
and dimethylamine (Scheme 2). Larger groupments were then installed
2

S. Gapil Tiamas, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Scheme 1. Asymmetric synthesis of both
enantiomers of fislatifolic acid (+)- and
(−)-1: a) NEt3 (3 equiv.), trimethylacetyl
chloride (1.2 equiv.), THF, -40 °C, 2 h; b)
LiCl (1.2 equiv.), (4S)- or (4R)-4-benzyl-1,3-
oxazolidin-2-one (1.2 equiv.), r.t., 18 h; c)
myrcene (2 equiv.), Me2AlCl 1 M in hexanes
(2 equiv.), CH2Cl2, 0 °C to r.t., 16 h; d)
LiOH.H2O (4 equiv.), H2O2 30% (3 equiv.),
THF/H2O 2:1, r.t., 16 h.
Scheme 2. Synthesis of carboxamide derivatives (+)- and (−)-4a-h by peptidic coupling on (+)- and (−)-fislatifolic acid 1: a) HNR1R2 (1.1 equiv.), HATU (1.1
equiv.), N,N-diisopropylethylamine (2 equiv.), DMF, r.t., 2 h. (+) Enantiomers correspond to 1S,2S stereochemistry and (−) enantiomers to 1R,2R.
Scheme 3. Synthesis of carboxylic acid derivatives (+)- and (−)-4i and 4j by saponification of (+)- and (−)-4f and 4g respectively: a) LiOH.H2O (4 equiv.), THF/
H2O 2:1, r.t., 16 h. (+) Enantiomers correspond to 1S,2S stereochemistry and (−) enantiomers to 1R,2R.
3

S. Gapil Tiamas, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Scheme 4. Attempt of demthylation of car-
boxamide (+)-4h and cyclisation of 1,5-
diene of fislatifolic acid (+)- and (−)-1: a)
BBr3 1 M in CH2Cl2 (2 equiv.), CH2Cl2, -78
°C to r.t., 3 h; b) BF3.OEt2 (2 equiv.), CH2Cl2,
0 °C to r.t., 4 h; c) MeONHMe·HCl (1.1
equiv.), HATU (1.1 equiv.), N,N-diisopro-
pylethylamine (2 equiv.), DMF, r.t., 2 h. (+)
Enantiomers correspond to 1S,2S stereo-
chemistry and (−) enantiomers to 1R,2R.
Table 1
Biological evaluation of synthesized compounds on Bcl-xL/Bak, Mcl-1/Bid and
Bcl-2/Bim displacement assays.
Cpd.
Bcl-xL/Bak binding
Mcl-1/Bid binding
Bcl-2/Bim binding
affinitya Kib (μM)
affinitya Kib (μM)
affinitya Kib (μM)
(+)-1
5.7
0.7
0.6
4.0
0.5
0.6
0.1
> 23
> 23
(−)-1
9.6
8.6
Scheme 5. Synthesis of carboxamide derivatives (+)- and (−)-4k by peptidic
coupling on (+)- and (−)-fislatifolic acid 1: a) 4-Aminophenol (1.1 equiv.),
HATU (1.1 equiv.), N,N-diisopropylethylamine (2 equiv.), DMF, r.t., 2 h. (+)
Enantiomers correspond to 1S,2S stereochemistry and (−) enantiomers to
1R,2R.
(+)-2
> 23
> 23
> 23
> 23
> 23
> 23
19.0
0.9
0.3
0.1
(−)-2
> 33
16.6
> 33
7.2
> 23
2.4
(+)-4a
(−)-4a
(+)-4b
(−)-4b
(+)-4c
(−)-4c
(+)-4d
(−)-4d
(+)-4e
(−)-4e
(+)-4f
(−)-4f
(+)-4g
(−)-4g
(+)-4h
(−)-4h
(+)-4i
(−)-4i
(+)-4j
(−)-4j
(+)-4k
(−)-4k
(+)-9
1.7
0.1
0.7
0.2
0.3
0.1
0.1
2.7
0.6
0.6
0.4
0.4
2.1
9.2
2.3
1.0
7.3
0.6
17.4
0.6
8.2
0.7
> 23
> 23
20.1
> 33
> 33
5.4
> 23
> 23
> 23
> 23
3.1
proteins. Activity of the analogues with a longer chain was also studied.
While compound (+)-4f was found to be a Mcl-1/Bcl-2 dual inhibitor
with micromolar range affinities on both proteins, its enantiomer
(−)-4f expressed only a little affinity toward Mcl-1. Compounds 4g-i
and 4k were completely inactive against all three Bcl-2 family proteins.
In contrast, (+)-4j revealed to be Mcl-1 selective and its enantiomer
(−)-4j Bcl-2 selective. Finally, both enantiomers of the cyclized pro-
ducts 9 and 10 lost their activity in comparison with fislatifolic acid 1
and Weinreb amide 2.
1.9
0.3
0.3
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
8.3
> 33
5.5
2.2
15.0
> 33
> 33
> 33
> 33
> 33
> 33
8.5
1.0
> 23
> 23
> 23
> 23
> 23
> 23
> 23
> 23
2.2
0.4
0.3
In terms of cytotoxicity on A549 and MCF7 cancer cell lines, some of
the most potent compounds (+)-1, (+)-2, (+)-4a, (+)-4c and (+)-4f
according to in vitro protein assays revealed moderate cytotoxic activ-
ities (Table 2), in the same range as compound (+)-4k, not active on
Bcl-2 proteins. Thus, no direct correlation could be established between
protein affinity and cytotoxicity on cancer cell lines. This result in-
dicates that another mechanism could be involved in the cell death.
However, the Weinreb amide (+)-2 is still the most active analogue on
these two cancer cell lines with 31.2 and 42.1 μM on A549 and MCF7
respectively, thus approaching the result obtained with the cisplatin on
this latter cell line (43.2 μM).
> 33
> 33
> 33
9.0
0.3
0.1
> 23
> 23
> 23
> 23
> 23
> 23
1.5
(−)-9
> 33
18.9
15.0
5.2
(+)-10
(−)-10
Ref.c
2.4
3.8
1.2
1.2
Binding affinities were measured by fluorescence polarization14 after
a
competition between the ligand and a fluorescein-labeled peptide.
b
Ki is the concentration of the ligand corresponding to 50% of the binding of
the labelled reference compound, corrected for experimental conditions.15
In conclusion, a library of 26 novel carboxamides deriving from
synthetic fislatifolic acid (+)- and (−)-1 has been prepared based on
the results previously obtained. Their evaluation on the modulation of
Bcl-xL/Bak, Mcl-1/Bid, and Bcl-2/Bim interactions have shown the in-
cidence of tiny modifications on the carboxamide part and confirmed
than the (+) enantiomers were as or more active than the (–) en-
antiomers. In this series, six compounds behaved as dual Bcl-xL/Mcl-1
inhibitors in micromolar range but the previously prepared Weinreb
amide (+)-2 is still the most promising compound deriving from nat-
ural cyclohexenyl chalcones with sub-micromolar affinities toward Mcl-
1 and Bcl-2 but with moderate cytotoxicities on A549 and MCF7 cancer
cell lines.
c
Meiogynin A was used as reference compound.9h
Acknowledgments
This research was conducted within the frame of the International
French Malaysian Natural Product Laboratory (IFM-NatProLab) estab-
lished between CNRS-ICSN and University Malaya. We are grateful to
the Bright Spark Unit, University of Malaya, Malaysia and the French
Embassy in Malaysia for the scholarship of SGT. We also thank
Investissement d’Avenir grant of the Agence Nationale de la Recherche
(CEBA: ANR-10-LABX-25-01) for its financial support.
4

S. Gapil Tiamas, et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxxx
Table 2
g) Yamaguchi R, Lartigue L, Perkins G. Pharmacol Ther. 2019;195:13.
3. Oltersdorf T, Elmore SW, Shoemaker AR, et al. Nature. 2005;435:677.
4. Tse C, Shoemaker AR, Adickes J, et al. Cancer Res. 2008;68:3421.
5. a) Roberts AW, Seymour JF, Brown JR, et al. J Clin Oncol. 2012;30:488;
b) Gyulkhandanyan AV, Mutlu A, Allen DJ, et al. Throm Res. 2013;133:73.
6. Souers AJ, Leverson JD, Boghaert ER, et al. Nat Med. 2013;19:202.
7. Hird AW, Tron AE. Pharmacol Ther. 2019;198:59.
Cytotoxicity assays on A549 and MCF7 cell lines.
Cpd.
A549 IC50 (μM)a
MCF7 IC50 (μM)a
(+)-1
53.7
31.2
42.2
71.7
87.9
66.8
10.5
−
0.2
3.6
2.6
8.7
5.7
3.1
4.3
64.3
42.1
45.2
72.3
96.8
66.7
−
7.7
1.2
1.3
1.2
1.4
1.2
(+)-2
8. a) Klenke S, Akdeli N, Heukamp L, et al. BMC Cancer. 2019;19:243;
b) Weeden CE, Ah-Cann C, Holik AZ, et al. Oncogene. 2018;37:4475;
c) Moujalled DM, Pomilio G, Ghiurau C, et al. Leukemia. 2019;33:905;
d) Tanaka Y, Aikawa K, Nishida G, et al. J Med Chem. 2013;56:9635.
9. a) Fomekong Fotsop D, Roussi F, Le Callonec C, et al. Tetrahedron. 2008;64:2192;
b) Litaudon M, Bousserouel H, Awang K, et al. J Nat Prod. 2009;72:480;
c) Leverrier A, Tran Huu Dau ME, Retailleau P, et al. Org Lett. 2010;12:3638;
d) Leverrier A, Martin M-T, Servy C, et al. J Nat Prod. 2010;73:1121;
e) Leverrier A, Awang K, Guéritte F, et al. Phytochem. 2011;72:1443;
f) Apel C, Gény C, Dumontet V, et al. J Nat Prod. 2014;77:1430;
g) Azmi MN, Péresse T, Remeur C, et al. Fitoterapia. 2016;109:190;
h) Gény C, Abou Samra A, Retailleau P, et al. J Nat Prod. 2017;80:3179.
10. a) Fomekong Fotsop D, Roussi F, Leverrier A, et al. J Org Chem. 2010;75:7412;
b) Dardenne J, Desrat S, Guéritte F, et al. Eur J Org Chem. 2013:2116;
c) Desrat S, Remeur C, Gény C, et al. Chem Comm. 2014;50:8593;
d) Desrat S, Remeur C, Roussi F. Org Biomol Chem. 2015;13:5520;
e) Abou Samra A, Robert A, Gov C, et al. Eur J Med Chem. 2018;148:26;
f) Gapil Tiamas S, Audet F, Abou Samra A, et al. Eur J Org Chem. 2018;42:5830;
g) Remeur C, Desrat S, Gandon V, et al. Eur J Org Chem. 2018;42:5869.
11. Nguyen NT, Nguyen MTT, Nguyen HX, et al. J Nat Prod. 2017;80:141.
12. a) Eschenmoser A, Ruzicka L, Jeger O, et al. Helv Chim Acta. 1955;38:1890;
b) Storck G, Burgstahler A. J Am Chem Soc. 1955;77:5058.
(+)-4a
(+)-4c
(+)-4f
(+)-4k
Gefitinib
Cisplatin
Antimycin A
43.2
−
8.1
6.5
1.7
a
IC50 measures the drug concentrations required for 50% cell death.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2020.127003.
References
1. a) Reed JC. Curr Opin Oncol. 1995;7:541;
b) Letai AG. Nat Rev Cancer. 2008;8:121.
13. a) Castro MA, Miguel del Corral JM, Gordaliza M, et al. Synthesis. 2005;1:3202;
b) Hong S, Corey EJ. J Am Chem Soc. 2006;128:1346;
2. a) Recent reviews Cory S, Roberts AW, Colman PM, et al. Trends Cancer.
2016;2:443;
c) Fráter G, Schröder F. J Org Chem. 2007;72:1112;
b) Schenk RL, Strasser A, Dewson G. Biochem Biophys Res Commun. 2017;482:459;
c) Ashkenazi A, Fairbrother WJ, Leverson JD. Nat Rev Drug Discov. 2017:273;
d) Garner TP, Lopez A, Reyna DE, et al. Curr Opin Chem Biol. 2017;39:133;
e) Adams JM, Cory S. Cell Death Differ. 2018;25:27;
d) Tran Ngoc D, Albicker M, Schneider L, et al. Org Biomol Chem. 2010;8:1781.
14. Qian J, Voorbach MJ, Huth JR, et al. Anal Biochem. 2004;328:131.
15. Cheng Y, Prusoff WH. Biochem Pharmacol. 1973;22:3099.
f) Opferman JT, Kothari A. Cell Death Differ. 2018;25:37;
5