Biomimetic total synthesis of meiogynin A, an inhibitor of Bcl-xL and bak interaction

Article abstract of DOI:10.1021/jo101088h

A short, convergent, and selective synthesis of meiogynin A, an inhibitor of the antiapoptotic protein Bcl-xL, has been performed. This synthesis, based on a biomimetic approach, allowed the determination of its absolute configuration. Three isomers of me

Full text of DOI:10.1021/jo101088h

pubs.acs.org/joc
original substituted cis-decalin carbon skeleton with a carboxy-
Biomimetic Total Synthesis of Meiogynin A,
an Inhibitor of Bcl-xL and Bak Interaction
lic acid at the ring junction and five asymmetric centers. Com-
pound 2, the epimer at C-1 of meiogynin A, was also isolated in
the active fraction as a minor compound but exhibited a less
potent affinity for Bcl-xL. The biological properties and unique
structure of meiogynin A (1) led us to consider its total
synthesis, not only to determine its absolute configuration
but also to have access to various analogues for SAR studies.
We hypothesized (Figure 1) that the biosynthesis of com-
pounds 1 and 2 may involve a Diels-Alder reaction between
a racemic bisabolatriene acid unit 3 and a chiral zingiberene
unit 43,4 either in an endo (compound 1) or in an exo mode
(compound 2). That was supported by the identification of
R-bisabolol, a putative precursor of 3 and 4, by GC-MS
analysis of the essential oil of the bark of M. cylindrocarpa.2
In order to validate this biomimetic pathway, and to assess
unambiguously the absolute configuration of the five asym-
metric centers, we have performed a short, convergent, and
selective total synthesis of meiogynin A 1 based on a final
Diels-Alder reaction between triene 3 and four dienophiles
(4R,10R), (4R,10S), (4S,10S), and (4S,10R) 4, potential pre-
cursors of meiogynin A and diastereomers. The selective
synthesis of triene 3, with a good control of the E,E config-
uration of the double bonds, was performed in six steps from
commercial trans-1,4-cyclohexanedimethanol 5 (Scheme 1).
After mono TBS-alcohol protection,5 the remaining free
alcohol was easily oxidized to the aldehyde 6 using Dess-
Martin periodinane. A Seyferth-Gilbert homologation was
then carried out using the Bestmann-Ohira reagent6 to give
the alkyne 7 in a very good yield (94%).
In order to settle the trisubstituted double bond, a Neigishi
carboalumination reaction7 was performed on compound 7,
using Lipshutz’s conditions.8 With iodine as the electrophile,
the desired 8-E compound was obtained along with 8% of its
inseparable 8-Z isomer (77% yield). When using NBS, no Z
isomer could be detected in the crude mixture, but the
reaction was not reproducible and yields were much lower
(from 20 to 58%). Direct Suzuki coupling of compound 8
with pinacol boronate ester 12 elaborated from the commer-
cial alkyne 11 using Schwartz’s catalyst9 led to the corre-
sponding triene 9. Unfortunately, that compound was
unstable to purification conditions, which meant that a
tandem carboalumination-coupling reaction10 could not be
envisaged, in our case, because of the fragileness of this
derivative.
ꢀ
Dalia Fomekong Fotsop, Fanny Roussi,* Aurelie Leverrier,
ꢀ
Anne Breteche, and Franc-oise Gueritte
ꢀ
ꢀ
Centre de recherche de Gif, Institut de Chimie des Substances
Naturelles, CNRS, UPR 2301, 1, avenue de la terrasse,
91198 Gif sur Yvette, France
fanny.roussi@icsn.cnrs-gif.fr
Received June 4, 2010
A short, convergent, and selective synthesis of meiogynin
A, an inhibitor of the antiapoptotic protein Bcl-xL, has
been performed. This synthesis, based on a biomimetic
approach, allowed the determination of its absolute con-
figuration. Three isomers of meiogynin A have also been
elaborated. One of these was found to be three times more
potent than the natural compound.
Apoptosis is the programmed cell death (PCD) used by
multicellular organisms to regulate tissue homeostasis through
the elimination of useless or potentially harmful cells. Dis-
ruption of this process is implicated in many kinds of
diseases, such as cancer. The Bcl-2 family of proteins governs
the commitment to PCD at the mitochondrion (the intrinsic
apoptotic pathway). It is divided into antiapoptotic such as
Bcl-2 or Bcl-xL and proapoptotic proteins such as Bax, Bad,
or Bim.1 In an effort to find potent inhibitors of the anti-
apoptotic protein Bcl-xL, we have recently isolated a new
dimeric sesquiterpenoid by a bioassay-guided screening of
Malaysian plants extracts.2 This compound, meiogynin A (1)
(Figure 1), isolated from the bark of Meiogyne cylindrocarpa,
acts as an antagonist of the Bcl-xL/Bak association. Only its
relative configuration could be determined, and the relative
configuration of the methyl group at C-100 was proposed
using conformational analysis. Meiogynin A possesses an
(3) Hagiwara, H.; Okabe, T.; Ono, H.; Kamat, V. P.; Hoshi, T.; Suzuki,
T.; Ando, M. J. Chem. Soc., Perkin Trans. 1 2002, 895–900.
(4) Breeden, D. C.; Coates, R. M. Tetrahedron 1994, 50, 11123–11132.
(5) Rieger, J. M.; Brown, M. L.; Sullivan, G. W.; Linden, J.; Macdonald,
T. L. J. Med. Chem. 2001, 44, 531–539.
€
(6) Muller, S.; Liepold, B.; Roth, G. J.; Bestmann, H. J. Synlett 1996,
521–522.
(7) (a) Van Horn, D. E.; Negishi, E. J. Am. Chem. Soc. 1978, 100, 2252–
2254. (b) Negishi, E.; Kondakov, D. Y.; Choueiry, D.; Kasai, K.; Takahashi,
T. J. Am. Chem. Soc. 1996, 118, 9577–9588.
(1) (a) Lowe, S. W.; Cepero, E.; Evan, G. Nature 2004, 432, 307–315.
(b) Letai, A. G. Nature Rev. Cancer 2008, 8, 121–132.
(2) Litaudon, M.; Bousserouel, H.; Awang, K.; Nosjean, O.; Martin,
ꢀ
M.-T.; Tran Huu Dau, M. E.; Hadi, H. A.; Boutin, J. A.; Sevenet, T.;
ꢀ
Gueritte, F. J. Nat. Prod. 2009, 72, 480–483.
(8) Lipshutz, B. H.; Butler, T.; Lower, A. J. Am. Chem. Soc. 2006, 128,
15396–15398.
(9) Pereira, S.; Srebnik, M. Organometallics 1995, 14, 3127–3128.
(10) See, for example: Lipshutz, B. H.; Butler, T.; Lower, A.; Servesko, J.
Org. Lett. 2007, 3737–3740.
7412 J. Org. Chem. 2010, 75, 7412–7415
Published on Web 10/07/2010
DOI: 10.1021/jo101088h
r
2010 American Chemical Society

Fotsop et al.
JOCNote
SCHEME 1. Synthesis of Triene 3
FIGURE 1. Biomimetic hypothesis.
The TBDMS ether 8 was thus selectively oxidized to the
acid 10 using a modified one-pot procedure with CrO3-
H5IO6.11 The minor Z isomer 10 could be partially removed
at this step (ratio E/Z 95/5).
The vinyl iodide thus obtained was engaged in a Suzuki
coupling with pinacol boronate ester 12. The best conditions
were found to be as follows: tetrakis(triphenylphosphine)
palladium, K3PO4, THF/H2O at 25 °C for 16 h. Triene 3 was
unstable when purified by conventional methods (normal- or
reversed-phase column chromatography), but thanks to its
acid function, it could be obtained pure by an acid-base
extraction. Compound 3 was also light sensitive, and per-
forming the Suzuki reaction in the dark increased yield from
62% to 100% after acid-base extraction. Expected triene 3
was obtained with an overall yield of 24% for 6 steps.
By analogy with the previous results of Hagiwara3 and
Nicolaou,12 a synthesis of the four potential chiral dieno-
philes was envisaged from (R)- and (S)-citronellal via the
formation of the cyclohexenone intermediates 14 and 18 by a
Robinson annulation (Scheme 2). The Robinson annula-
tion was first carried out in a nonselective manner to give 14
and 18 as a mixture of diastereomers.3 A carbonylation13 was
performed on their enol triflates and led to the formation
of the expected dienophiles 15, 16 and 19, 20. The two mix-
tures of diastereomers were separated at that stage14 to
yield four pure compounds. These compounds were prone
to aromatization at room temperature but could be stored
SCHEME 2. Synthesis of Chiral Dienophiles 15, 16, 19, and 20
(11) Zhang, S.; Xu, L.; Trudell, M. L. Synthesis 2005, 16, 1757–1760.
(12) (a) Nicolaou, K. C.; Sarlah, D.; Shaw, D. M. Angew. Chem., Int. Ed.
2007, 46, 4708–4711. (b) Nicolaou, K. C.; Wu, T. R.; Sarlah, D.; Shaw,
D. M.; Rowcliffe, E.; Burton, D. R. J. Am. Chem. Soc. 2008, 130, 11114–
11121.
(13) Freskos, J. N.; Laneman, S. A.; Reilly, M. L.; Ripin, D. H. Tetra-
hedron Lett. 1994, 35, 835–838.
(14) Conditions of separation of dienophiles: Hypercarb column; eluant:
water þ 0.1% formic acid/MeOH þ 0.1% formic acid 10/90.
for weeks under argon at -20 °C. In parallel, the synthesis of
cyclohexenones 21 and 22 was performed, according to
Nicolaou’s procedure (Scheme 3).12 A proline-derived
J. Org. Chem. Vol. 75, No. 21, 2010 7413

Fotsop et al.
JOCNote
SCHEME 3. Synthesis of Chiral Dienophiles 15 and 20
SCHEME 4. Final Cycloaddition and Synthesis of Compounds
23-26
FIGURE 2. Meiogynin A 24 and its isomers.
once, in an hetero-Diels-Alder reaction, and as a diene,19 we
were pleased to see that compounds 15, 16, 19, and 20 react
solely as dienophiles. In addition, the chemo, regio, and
facial selectivities were total as only the less hindered diene of
3 reacts with these dienophiles, anti to their lateral chain. The
observed diastereoselectivities of the DA were also very
satisfying as the expected products were obtained as major
compounds along with only 9% of their exo isomers (in the
same proportion as in the active fraction of the plant extract).
In addition, 5% of adducts 27 arising from a cycloaddition
between dienophiles and remaining E,Z-triene 3 were
formed.20 The endo isomers 23-26 could be isolated by
preparative HPLC (Figure 2).21
catalyst15 and 3,4-dihydroxybenzoate as a cocatalyst
were used for the enamine-mediated Michael addition of
(S)-citronellal 13 and (R)-citronellal 17 to methyl vinyl
ketone and was followed by an intramolecular aldol con-
densation of the resulting ketoaldehyde product.16 Their
subsequent transformation into acids 15 and 20 allowed
the unambiguous identification of each of the four pre-
vious synthesized dienophiles 15, 16, 19, and 20.17
The final Diels-Alder (DA) reactions between triene 3
and pure dienophiles 15, 16, 19, and 20 were carried out using
2-bromobenzeneboronic acid as an organocatalyst as de-
scribed by Hall et al. (Scheme 4).18 In our case, the reaction
had to be performed at 50 °C, since no cycloadducts were
observed at room temperature. The reaction was completed
within 72 h with 30 mol % of catalyst (because of the two acid
functions, the reaction was much slower with a smaller load
of catalyst). Cycloadducts were obtained in reasonable yields
ranging from 60 to 74% (Scheme 4). As, to our knowledge,
methyl 1,3 cyclohexadiene-2-carboxylate has only been used
The spectroscopic data (1H and 13C NMR) of these
products showed noticeable differences in chemical shifts
for H100 (0.77 ppm for natural meiogynin A and compounds
24 and 26 and 0.79 ppm for compounds 23 and 25), C800
(15.5 ppm for natural meiogynin A and 24 and 26 and 16.1 ppm
for 23 and 25), C6 (25.4 ppm and 27.6 ppm), and C7 (36.1 ppm
and 36.5 ppm). These data and the comparison of optical
rotations for compounds 24 and 26, allowed us to determine
compound 24 (1S,5R,7R,10R,100S) to be meiogynin A. Never-
theless, having unambiguously established the absolute config-
uration of natural meiogynin A by its total synthesis, we were
surprised to observe that its optical rotation ([R]D -200; c =
0.1 in CH2Cl2) was considerably different from that which we
reported previously for the same compound obtained from its
natural source ([R]D -281; c = 0.12 in CH2Cl2).2 However,
after further, careful purification of the natural compound, its
optical rotation was now practically identical ([R]D -195; c =
0.1 in CH2Cl2) to the synthetic product, thereby confirming our
configurational assignment.
The binding affinity of compounds 23-26 was evaluated
on Bcl-xL by competition against the fluorescent-tagged
BH3 domain of the protein Bak, as described22 and compared
(15) Enders, D.; Kipphart, H.; Gerdes, P.; BrenaValle, L. J.; Bhushan, V.
Bull. Soc. Chim. Belg. 1988, 97, 691–704.
(16) In our case, the observed de (determined by HPLC) for the enamine-
mediated Michael addition of (S)-citronellal 13 and (R)-citronellal 17 to methyl
vinyl ketone were disappointing (60% de from 13 and 56% de from 17).
(17) The catalyst derived from (R)-proline should give (10S,4R)-14 and
(10R,4R)-18 as major isomers, precursors of dienophiles 16 and 19.
(18) Al-Zoubi, R. M.; Marion, O.; Hall, D. G. Angew. Chem., Int. Ed.
2008, 47, 2876–2879.
(20) Ratios of these compounds were determined by HPLC.
(21) Conditions of separation of endo/exo adducts: Sunfire column; eluant:
MeOH þ 0.1% TFA/ ACN þ 0.1% TFA/ H2O þ 0.1% TFA 10/70/20.
(22) Quian, J.; Voorbach, M. J.; Huth, J. R.; Coen, M. L.; Zhang, H.; Ng,
S.-C.; Comess, K. M.; Petros, A. M.; Rosenberg, S. H.; Warrior, U.; Burns,
D. J. Anal. Biochem. 2004, 328, 131–138.
(19) Boger, D. L.; Patel, M J. Org. Chem. 1984, 49, 4099–4101.
7414 J. Org. Chem. Vol. 75, No. 21, 2010

Fotsop et al.
JOCNote
TABLE 1. Binding Affinities of Natural Meiogynin A and Compounds
23-26 for the Antiapoptotic Protein Bcl-xL
TFA/H2O þ 0.1% TFA 10/70/20), tR 20.3 min: [R]25 -200
D
1
(c 0.1, CH2Cl2); H NMR (500 MHz, CDCl3) δ 5.65 (d, J =
entry
compd
IC50a,c (μM)
Kib,c (μM)
10.1 Hz, 1H), 5.59 (d, J = 10.1 Hz, 1H), 5.17 (br. s, 1H), 5.12 (t,
J = 7.0 Hz, 1H), 4.96 (d, J = 10.6 Hz, 1H), 3.18 (m, 1H), 2.56
(m, 1H), 2.35 (m, 1H), 2.26 (t, J = 12.1 Hz, 1H), 2.02 (m, 1H),
1.99 (m, 1H), 1.95 (m, 1H), 1.93 (m, 1H), 1.91 (m, 1H), 1.87 (m,
1H), 1.81 (m, 1H), 1.79 (m, 1H), 1.78 (m, 1H), 1.71 (s, 3H), 1.67
(m,1H), 1.66 (s, 3H), 1.65 (s, 3H), 1.63 (s, 3H), 1.57 (m, 1H), 1.51
(m, 1H), 1.50 (m, 1H), 1.42 (m, 1H), 1.33 (m, 1H), 1.26 (m, 1H),
1.19 (m, 1H), 1.13 (m, 1H), 0.77 (d, J = 6.6 Hz, 3H); 13C NMR
(75 MHz, CDCl3) δ 182.2, 180.8, 140.1, 133.7, 133.4, 131.2,
129.8, 124.7, 123.2, 120.1, 48.6, 46.3, 43.2, 42.8, 36.3, 36.1, 34.1,
31.6, 30.7, 30.1, 29.3, 28.6, 27.9, 25.9, 25.7, 25.4, 23.3, 17.6, 15.5,
14.6; IR (film): 2921, 2853, 1702 cm-1; MS (ESþ, CH2Cl2/
MeOH) m/z = 491.0 [M þ H]þ; HRMS (ESþ) calcd for
C30H44O4Na 491.3137, found 491.3153.
1
2
3
4
5
1
23
24
25
26
10.7 ( 1.6
129 ( 17
11.6 ( 0.5
2.9 ( 0.7
52 ( 12
8.3 ( 1.2
100 ( 13
9 ( 0.4
2.3 ( 0.5
40 ( 9
aIC50 is the concentration of compound inducing 50% of inhibition of
the binding of the labeled reference compound Bak to Bcl-xL. bKi is the
concentration corresponding to 50% of inhibition of the binding of the
labeled reference compound Bak to Bcl-xL and corrected for experi-
mental conditions according to Cheng and Prussof.23 cValues are
reported as the means of two independent determinations.
with that of a pure sample of natural meiogynin A. The activity
of compound 24 was similar to that of natural meiogynin 1, and
interestingly, one of its diastereomers (compound 25, with an
opposite configuration at C100) was three times more potent
(Table 1).
In conclusion, we have performed a short (11 steps), con-
vergent, and selective synthesis of meiogynin A (1) based on a
biomimetic hypothesis with only one protection step. We
also have elaborated three isomers 23, 25, and 26. An in vitro
affinity displacement assay revealed that compound 25
exhibited a more potent affinity for Bcl-xL than the natural
product, showing how important it is to control the configura-
tion of all the asymmetric centers. Pharmacomodulations on
these interesting compounds are currently underway in our
laboratory.
(1S,5R,7R,10R,100R)-25. From triene 3 (40 mg, 0.17 mmol)
and dienophile 19 (20 mg, 0.08 mmol) in dichloromethane (0.3
mL) was obtained desired compound 25 (along with 9% of the
exo adduct and 5% of 27) after purification (24.5 mg, 62%).
Compound 25 was obtained pure after preparative HPLC
(Sunfire column, eluant: MeOH þ 0.1% TFA/CH3CN þ
0.1% TFA/H2O þ 0.1% TFA 10/70/20), tR 20.2 min: [R]25
D
1
-220 (c 0.1, CH2Cl2); H NMR (500 MHz, CDCl3) δ 5.64 (d,
J = 10.2 Hz, 1H), 5.55 (d, J = 10.2 Hz, 1H), 5.12 (br. s, 1H), 5.05
(t, J = 7.0 Hz, 1H), 4.92 (d, J = 10.0 Hz, 1H), 3.13 (m, 1H), 2.51
(m, 1H), 2.26 (m, 1H), 2.21 (t, J = 12.0 Hz, 1H), 1.97 (m, 1H),
1.94 (m, 1H), 1.90 (m, 1H), 1.88 (m, 1H), 1.86 (m, 1H), 1.84 (m,
1H), 1.80 (m, 1H), 1.77 (m, 1H), 1.74 (m,1H), 1.65 (s, 3H), 1.61
(s, 3H), 1.60 (s, 3H), 1.59 (m,1H), 1.57 (s, 3H), 1.55 (m, 1H), 1.43
(m, 1H), 1.40 (m, 1H), 1.38 (m, 1H), 1.31 (m, 1H), 1.26 (m, 1H),
1.19 (m, 1H), 1.13 (m, 1H), 0.79 (d, J = 6.6 Hz, 3H); 13C NMR
(125 MHz, CDCl3) δ 182.5, 181.2, 139.9, 133.2, 132.1, 131.0,
130.1, 124.6, 123.0, 119.9, 48.5, 46.1, 43.1, 42.8, 36.5, 36.4, 33.5,
31.4, 30.5, 29.8, 29.2, 28.4, 27.8, 27.6, 25.8, 25.5, 23.2, 17.5, 16.1,
Experimental Section
General Procedure for the Cycloaddition Reactions. Triene 3
(2.00 equiv), dienophiles 15, 16, 19, or 20 (1.00 equiv), and
2-bromobenzeneboronic acid (0.3 equiv) were dissolved in wet
dichloromethane and stirred, in the dark, under argon, at 50 °C
for 3 days. The solvent was removed in vacuo, and the crude
mixture was purified by flash chromatography on a reversed-
phase column (Versapack C18 cartridge) to afford the expected
endo adduct along with 9% of its exo isomer and 5% of 27. The
endo/exo adducts were separated by preparative HPLC.
(1S,5R,7R,10R,100S)-24. From triene 3 (38 mg, 0.16 mmol)
and dienophile 16 (19 mg, 0.08 mmol) in dichloromethane (0.3
mL) was obtained desired compound 24 (along with 9% of the
exo adduct and 5% of 27) after purification (22.5 mg, 60%).
Compound 24 was obtained pure after preparative HPLC (Sun-
fire column, eluant: MeOH þ 0.1% TFA/CH3CN þ 0.1%
14.5; IR (film): 2921, 2849, 1700 cm-1; MS (ESþ, CH2Cl2
/
MeOH) m/z = 491.0 [M þ H]þ; HRMS (ESþ) calcd for
C30H44O4Na 491.3137, found 491.3155.
Acknowledgment. We thank Prof. Khalijah Awang (Depart-
ment of Chemistry, University of Malaya, Pantai Valley,
Kuala Lumpur, Malaysia), Dr. B. Pfeiffer, Dr. O. Nosjean
(Institut de Recherche Servier), Dr. M. Litaudon, and
Dr. R. Dodd (ICSN) for helpful discussions. We also thank
CNRS for a grant for one of us (D.F.F.).
Supporting Information Available: Complete experimental
procedures, 1H and 13C spectra for all new compounds, and HPLC
chromatograms for final compounds 23-26. This material is
available free of charge via the Internet at http://pubs.acs.org.
(23) Cheng, Y. C.; Prussof, W. H. Biochem. Pharmacol. 1973, 22,
3099–3108.
J. Org. Chem. Vol. 75, No. 21, 2010 7415