Efficient synthesis of bongkrekic acid. Three-component convergent strategy

Article abstract of DOI:10.1016/j.tetlet.2009.04.129

An efficient total synthesis of the apoptosis inhibitor bongkrekic acid was accomplished using a three-component convergent strategy involving a Kocienski-Julia olefination and a Suzuki-Miyaura coupling, in which the longest linear sequence was 18 steps and proceeded in 6.4% overall yield. The torquoselective olefination also contributed to the shortening of the synthesis.

Full text of DOI:10.1016/j.tetlet.2009.04.129

Tetrahedron Letters 50 (2009) 4164–4166
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Efficient synthesis of bongkrekic acid. Three-component convergent strategy
Yukiko Sato a, Yoshifumi Aso a, Mitsuru Shindo b,
*
a Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, 6-1, Kasugako-en, Kasuga 816-8580, Japan
b Institute for Materials Chemistry and Engineering, Kyushu University, 6-1, Kasugako-en, Kasuga 816-8580, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
An efficient total synthesis of the apoptosis inhibitor bongkrekic acid was accomplished using a three-
component convergent strategy involving a Kocienski-Julia olefination and a Suzuki–Miyaura coupling,
in which the longest linear sequence was 18 steps and proceeded in 6.4% overall yield. The torquoselec-
tive olefination also contributed to the shortening of the synthesis.
Received 2 April 2009
Revised 25 April 2009
Accepted 30 April 2009
Available online 5 May 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Natural product
Apoptosis inhibitor
Total synthesis
Kocienski-Julia olefination
Suzuki–Miyaura coupling
Torquoselective olefination
Ynolates
Bongkrekic acid (BKA) is a natural toxic antibiotic produced by
the bacterium Burkholderia cocovenenans.1 By binding to the ade-
nine nucleotide translocator (ANT)2 from inside mitochondria
and fixing its conformation in the m-state, BKA inhibits the mito-
chondrial permeability transition pore (MPT). It has therefore been
used as a pharmacological and biological reagent to modulate the
properties of the MPT or the ANT. Furthermore, BKA was also found
to inhibit mitochondria-induced apoptosis by preventing the for-
mation of MPT,3 which is linked to numerous phenomena, such
as inhibition of phosphatidylserine exposure,4 inhibition of chlo-
ride channels,5 and reduction of ischemic-induced neuronal
death.6 Although it is now an important tool as an apoptosis inhib-
itor, due to its limited availability from fermentation or chemical
synthesis,7 the bioactivity of BKA has not been extensively investi-
gated, especially its in vivo activity. In order to assess its use in and
potential contribution to apoptotic science, BKA and its analogues
will have to be synthesized on a large scale in pure form. BKA is a
heptaene-tricarboxylic fatty acid with two allylic chiral stereocen-
ters. The total synthesis of BKA was reported by Corey8 in 1984 and
by our group in 2004.9 Corey did not isolate BKA in pure form, ow-
ing to its instability; likewise our previous semi-convergent syn-
thesis was quite long (32 steps in the longest linear sequence). In
this Letter, we describe the second-generation synthesis of BKA,
including a three-component convergent strategy using a doubly
terminally functionalized middle segment.
In our previous synthesis, the final conversion, which required
16 steps after combining each segment, reduced the efficiency.
The key point of this new synthesis would be a reduction in the
number of steps after the three-component coupling. Our new syn-
thetic strategy is shown in Scheme 1. BKA was separated at the C3–
C4 and C12–C13 bonds into three segments, A, B, and C. Segments
OMe
HO2C
1
13
Kocienski-Julia
olefination
12
CO2H
4
3
HO2C
Suzuki-Miyaura coupling
Bongkrekic acid (BKA)
OMe
CO2R
RO2C
I
CHO
HO
Segment C
Segment A
ArO2S
(RO)2B
Segment B
* Corresponding author.
Scheme 1. The second-generation BKA synthesis: a three-component convergent
E-mail address: shindo@cm.kyushu-u.ac.jp (M. Shindo).
strategy.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.04.129

Y. Sato et al. / Tetrahedron Letters 50 (2009) 4164–4166
4165
pentyn-1-ol (5) in five steps, to afford 8 with excellent diastere-
oselectivity in good yield (Scheme 3). Reductive removal of the
chiral auxiliary, followed by protection of the alcohol
with a TBDPS group, gave 9, which was subjected to the Mitsun-
obu reaction, followed by oxidation, to provide the tetra-
zolylsulfonyl compound 10. After deprotection of the TBDPS
group with TBAF/AcOH and then Swern oxidation, the aldehyde
11 was subjected to the CrCl2-catalyzed borylalkenylation12 to
furnish the doubly terminally functionalized middle segment 13
(Segment B).
HO
a,b,c
d,e
OH
I
TBSO
1
2
CO2H
I
CO2MOM
I
f,g
TBSO
HO
Segment C has a multi-substituted conjugated diene coupled
with the MOM ester, a cis-alkene, and an asymmetric center. Alky-
nylation of the readily available (S)-glycidyl ether 14 with the lith-
ium acetylide of the MPM-protected propargyl alcohol, followed by
methylation, afforded 15, which was subjected to the Lindlar par-
tial reduction to give, after deprotection of the TBS group, the cis-
alkene 16 (Scheme 4). The alcohol was oxidized via the Swern pro-
tocol, followed by the Corey–Fuchs alkynylation13 to give the al-
kyne 17, after carboxylation of the terminal alkyne. The cis-
selective conjugate addition14 with a cuprate, and then successive
treatment with DIBAL and MnO2, afforded the requisite trisubsti-
tuted alkene 18. The next crucial step was the construction of
the unsaturated MOM ester 19. Attempts at the preparation of
the Wittig reagent 23 bearing a MOM ester were unsuccessful, pre-
sumably due to its instability. Since the direct formation of the
MOM ester would be favorable, we decided to use the torquoselec-
tive olefination via ynolates.15 The aldehyde was reacted with the
lithium ynolate 24, prepared from the treatment of ethyl 2,2-dib-
romopropanoate with t-BuLi,16 to give the unsaturated carboxyl-
ate, which was treated in situ with MOMCl to provide the
desired conjugated MOM ester in 75% yield as a single isomer. After
removal of the MPM group with DDQ, the resulting alcohol was
oxidized to 20 (Segment C).
4 (Segment A)
3
Scheme 2. Synthesis of Segment A. Reagents and conditions: (a) TBSCl, imidazole,
CH2Cl2, 0 °C then rt, 99%; (b) BuLi, ꢀ40 °C; (HCHO)n, THF, rt, 94%; (c) Red-Al, ether,
0 °C; I2, ꢀ50 °C then rt, 65%; (d) MnO2, benzene/CH2Cl2, rt; (e) NaClO2, NaH2PO4, 2-
methyl-2-butene, t-BuOH, H2O, THF, rt, 96% for two steps; (f) MOMCl, i-Pr2NEt, rt,
99%; (g) 3 M HCl, THF, rt, 84%.
B and C would be connected by the modified Julia olefination,10
and Segment A would be coupled with the resulting Segment B–
C via the Suzuki–Miyaura coupling.11 The oxidation state of Seg-
ment A should be high so as to decrease the number of final oxida-
tion steps leading to the dicarboxylic acid. Segment B, which has
the dual functions of alkenyl boronate for the Suzuki–Miyaura cou-
pling and arylsulfone for the Julia olefination, would be prepared
through asymmetric alkylation. The question was whether the
alkenyl boronate would be compatible with the Julia olefination.
Segment C would be prepared from the readily available enantio-
merically pure glycidol.
Scheme 2 shows the preparation of Segment A starting from 3-
butyn-1-ol (1), which was converted into (Z)-iodoalkene 2 by the
successive TBS protection, homologation of the lithium acetylide
with paraformaldehyde, and reductive alumination with Red-Al
followed by iodination. The alcohol of 2 was subjected to a step-
wise oxidation with MnO2 followed by NaClO2 to afford the car-
boxylic acid 3. After several attempts to protect the carboxylic
acid, the MOM ester was found to be suitable for acid-mediated
deprotection at the final stage, since under basic conditions, the
C2–C3 alkene conjugated to the C1-ester of BKA was readily isom-
erized to the E-isomer. The TBS group could be removed selectively
by treatment with 3 M HCl to provide 4.
With the three segments in hand, we addressed the final chal-
lenge of the three-component coupling. The Kocienski-Julia olef-
ination of Segment C (20) with Segment B (13) in the presence of
KHMDS successfully furnished the coupled Segment B–C (21)
with excellent E-selectivity in high yield. Next, 21 was coupled
with Segment
A (4) by the Suzuki–Miyaura coupling and
provided 22 having the entire BKA skeleton in excellent yield. Fi-
nally, straightforward oxidation of the primary alcohol by the
Jones reagent, followed by deprotection with 6 M HCl afforded
BKA.
The synthesis of Segment B commenced with the asymmetric
alkylation of the Evans oxazolidinone 7 with 6, prepared from 4-
OH
O
O
OH
OTBS
a,b,c,d,e
g,h,i
f
O
N
I
OTBS
OTBDPS
O
O
Bn
5
6
8
9
O
N
7
Bn
Ph
O
S
N
N
Ph
Ph
O
S
N
N
N
O
N
N
O
N
n
j,k
N
l,m
N
O
O
N
S
N
O
BCHCl2
O
O
B
O
OHC
OTBDPS
12
10
11
13 (Segment B)
Scheme 3. Synthesis of Segment B. Reagents and conditions: (a) TBSCl, imidazole, 4-DMAP, CH2Cl2, rt, 91%; (b) BuLi, THF, ꢀ78 °C; (HCHO)n, rt, 95%; (c) Red-Al, ether, 0 °C then
rt, 94%; (d) MsCl, Et3N, CH2Cl2, 0 °C then rt; (e) NaI, acetone, rt, 93% for two steps; (f) LDA, THF, 7, ꢀ78 °C then 6, ꢀ78 °C, allowed to ꢀ20 °C, 83%, >99% de; (g) LiAlH4, THF, 0 °C,
quant.; (h) TBDPSCl, imidazole, 4-DMAP, CH2Cl2, quant.; (i) 3 M HCl, THF, rt, 99%; (j) 1-phenyl-5-mercaptotetrazole, DEAD, Ph3P, THF, 0 °C, 97%; (k) (NH4)6Mo7O24ꢁ4H2O, 30%
H2O2, ethanol, 0 °C then rt, 90%; (l) TBAF, AcOH, THF, 96%; (m) (COCl)2, DMSO, Et3N, CH2Cl2, ꢀ78 °C; (n) CrCl2, LiI, Mn, TMSCl, 12, THF, 0 °C then rt, 66% for two steps.

4166
Y. Sato et al. / Tetrahedron Letters 50 (2009) 4164–4166
OMe
OMe
c,d
e,f,g,h
a,b
O
MPMO
MPMO
OH
OTBS
OTBS
15
i,j,k
16
14
CO2MOM
OMe
CHO
OMe
OMe
l
MPMO
MPMO
MPMO
CO2Et
19
18
17
OMe
MOMO2C
OMe
MOMO2C
CO2MOM
OMe
m,n
o
p
q,r
CO2MOM
BKA
OHC
O
B
HO
O
20 (Segment C)
21
22
Scheme 4. Synthesis of Segment C and completion of the total synthesis of BKA. Reagents and conditions: (a) 3-(4-methoxyphenylmethoxy)-1-propyne, BuLi, BF3ꢁOEt2, THF,
ꢀ78 °C, 64%; (b) NaH, MeI, THF, 0 °C then rt, quant.; (c) Lindlar catalyst, quinoline, hexane, H2, rt; (d) 3 M HCl, THF, rt, 93% for two steps; (e) (COCl)2, DMSO, Et3N, CH2Cl2,
ꢀ78 °C; (f) CBr4, Ph3P, CH2Cl2, 0 °C then rt, 77% for two steps; (g) BuLi, THF, ꢀ78 °C, 93%; (h) BuLi, ClCO2Et, THF, ꢀ78 °C then ꢀ45 °C, 89%; (i) MeLi, CuI, THF, ꢀ78 °C, 94%; (j)
DIBAL, THF, ꢀ78 °C, 96%; (k) MnO2, benzene, 50 °C; (l) CH3CBr2CO2Et, t-BuLi, THF, ꢀ78 °C then 0 °C; 18, rt; MOMCl, 0 °C then rt, 75% for two steps; (m) DDQ, CH2Cl2, H2O, rt,
93%; (n) MnO2, benzene, rt; (o) 13, KHMDS, THF, ꢀ78 °C; 20, ꢀ78 °C, 92% for two steps; (p) 4, (Ph3P)2PdCl2, Et3N, I2, MeOH, rt, 86%; (q) Jones reagent, acetone, 0 °C; (r) 6 M HCl,
THF, MeOH, rt, 34% for two steps.
rial Foundation, and the Program for Promotion of Basic and
Applied Research for Innovations in Bio-oriented Industry (BRAIN).
Me
Ph3P
CO2MOM
OLi
23
24
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2009.04.129.
The BKA produced herein was stable enough to be purified with
normal phase HPLC and the 1H and 13C NMR spectral data of the
synthetic BKA were identical to those of the natural BKA.17 The
References and notes
optical rotation of the synthetic BKA, ½a D23
ꢂ
ꢀ51.3 (c 1.5, CHCl3),
25
1. Isolation: (a) van Veen, A. G.; Mertens, W. K. Recl. Trav. Chim. Pays-Bas 1934, 53,
257; Structure: (b) de Bruijin, J.; Frost, D. J.; Nugteren, D. H.; Gaudemer, A.;
Lijmbach, G. W. M.; Cox, H. C.; Berends, W. Tetrahedron 1973, 29, 1541; The
absolute configuration: (c) Zylber, J.; Gaudemer, F.; Gaudemer, A. Experientia
1973, 29, 648.
2. Henderson, P. J. F.; Lardy, H. A. J. Biol. Chem. 1970, 245, 1319.
3. (a) Zamzami, N.; Marchetti, P.; Castedo, M.; Hirsch, T.; Susin, S. A.; Masse, B.;
Kroemer, G. FEBS Lett. 1996, 384, 53; For a review, see: (b) Green, D. R.; Reed, J.
C. Science 1998, 281, 1309.
was quite different from that of the natural product (½
aꢂ
+162.5
589
(the conditions were not described),1b in which even the sign is
opposite. To verify the specific rotation, the synthetic BKA was
esterified with diazomethane to convert it into the corresponding
trimethyl ester, which showed ½a D24
ꢂ
+81.5 (c 0.27, CHCl3), which
was identical with Corey’s8 (½a D23
ꢂ
+80) and our previous reports9
(½a 2D6
ꢂ
+80.0 (CHCl3)). From these results, we concluded that the
optical rotation described herein is correct, since our synthetic
BKA is chemically pure.
4. Chen, S.-P.; Yang, H.-L.; Her, G. M.; Lin, H.-Y.; Jeng, M.-F.; Wu, J.-L.; Hong, J.-R.
Virology 2006, 347, 379.
5. Malekova, L.; Kominkova, V.; Ferko, M.; Stefanik, P.; Krizanova, O.; Ziegelhöffer,
A.; Szewczyk, A.; Ondrias, K. Biochim. Biophys. Acta 2007, 1767, 31.
6. Muranyi, M.; Li, P.-A. Neurosci. Lett. 2005, 384, 277.
In conclusion, we have achieved an efficient convergent total
synthesis of (+)-BKA, using a torquoselective olefination and the
Kocienski-Julia olefination and the Suzuki–Miyaura coupling as
the segment-binding steps. It is noteworthy that after combining
the three segments, it took only two steps to complete the synthe-
sis, indicating the high efficiency of this synthesis to provide BKA
and its analogues. Furthermore, the torquoselective olefination
also contributes to the shortening of the synthesis. The longest lin-
ear sequence is only 18 steps and proceeds in 6.4% overall yield,
which is an improvement over our previous process (32 steps
and 0.6% overall yield). Actually, we have prepared enough BKA
to test for bioactivity. The biological evaluation of BKA and its ana-
logues, including the apoptotic inhibitory activity in vitro and re-
examination of the cytotoxicity in vivo, is now in progress in our
laboratory.
7. The cost of BKA is $875/1 mg from BIOMOLÒ
.
8. Corey, E. J.; Tramontano, A. J. Am. Chem. Soc. 1984, 106, 462.
9. Shindo, M.; Sugioka, T.; Umaba, Y.; Shishido, K. Tetrahedron Lett. 2004, 45, 8863.
10. For a review, see: Blakemore, P. R. J. Chem. Soc., Perkin Trans. 1 2002, 2563.
11. For a review, see: Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
12. Takai, K.; Shinomiya, N.; Kaihara, H.; Yoshida, N.; Moriwake, T.; Utimoto, K.
Synlett 1995, 963.
13. Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 36, 3769.
14. Corey, E. J.; Katzenellenbogen, J. A. J. Am. Chem. Soc. 1969, 91, 1851.
15. Shindo, M.; Mori, S. Synlett 2008, 2231.
16. Shindo, M.; Matsumoto, K.; Shishido, K. Org. Synth. 2007, 84, 11.
17. Analytical data of synthetic BKA: 1H NMR (600 MHz, CDCl3) d: 1.07 (3H, d,
J = 7.2 Hz), 1.88 (3H, s), 1.92 (1H, m), 1.94 (3H, s), 2.04 (2H, m), 2.13 (2H, m),
2.17 (1H, m), 2.27 (1H, ddd, J = 6.0, 8.4, 13.8 Hz), 2.36 (1H, m), 2.46 (1H, ddd,
J = 7.8, 8.4, 13.8 Hz), 3.21 (3H, s), 3.33 (1H, d, J = 16.2 Hz), 3.47 (1H, d,
J = 16.2 Hz), 4.33 (1H, dd, J = 4.8, 9.0 Hz), 5.33–5.43 (2H, m), 5.72 (1H, s), 5.74
(1H, dt, J = 6.6, 15.0 Hz), 6.01 (1H, dd, J = 8.4, 15.6 Hz), 6.05 (1H, dd, J = 10.2,
11.4 Hz), 6.30 (1H, dd, J = 10.8, 15.0 Hz), 6.34 (1H, d, J = 12.6 Hz), 7.44 (1H, d,
J = 15.6 Hz), 7.64 (1H, d, J = 12.6 Hz); 13C NMR (150 MHz, CDCl3) d: 11.8, 18.6,
20.5, 32.2, 32.6, 33.2, 38.8, 40.0, 40.7, 56.8, 80.0, 118.2, 124.1, 124.3, 124.7,
125.2, 125.3, 128.1, 131.4, 131.5, 134.6, 135.9, 145.5, 148.5, 148.9, 170.4, 174.7,
176.8; IR (Neat): 2930, 1697, 1684 cmꢀ1; MS (FAB) m/z 509 (M+Na); HRMS
Acknowledgments
This work was supported by Takeda Science Foundation, Re-
search for Promoting Technological Seeds JST, the Uehara Memo-
(FAB) calcd for C28H28NaO7: 509.2515 found 509.2509; ½a D23
ꢀ51.3 (c 1.5,
ꢂ
CHCl3).