First total synthesis of the irciniasulfonic acids

Article abstract of DOI:10.1055/s-2005-862387

A flexible and concise first total synthesis of the irciniasulfonic acids is reported.

Full text of DOI:10.1055/s-2005-862387

652
LETTER
First Total Synthesis of the Irciniasulfonic Acids
First
Total
S
ynthes
d
is of the Irci
r
niasulfon
i
ic
A
cid
a
s
n P. Dobbs,*a Alberto Venturelli,a,b Laura A. Butler,a Robert J. Parkera
a
Department of Chemistry, University of Exeter, Stocker Road, Exeter, EX4 4QD, UK
Fax +44(1392)263434; E-mail: A.Dobbs@exeter.ac.uk
b
Dipartimento di Scienze Farmaceutiche, Università degli Studi di Modena e Reggio Emilia, Via G. Campi n. 183, 41110 Modena, Italy
Received 22 September 2004
duction of these chains, or alternative esters, at a suitably
late stage.
Abstract: A flexible and concise first total synthesis of the ircinia-
sulfonic acids is reported.
We chose propylene oxide as a readily available starting
material, which is also available in both enantiomerically
pure forms. Initially working with the racemic epoxide,
nucleophilic ring-opening was achieved using hex-1-yne,
n-BuLi and either boron trifluoride etherate3 or diethyl-
aluminium chloride4 to give non-4-yn-2-ol (3). Although
the yield was lower using boron trifluoride, the regiose-
lectivity was excellent (100:0 for the desired product),
while it was 98:2 for diethylaluminium chloride, with the
two regioisomers being impossible to separate. The inter-
nal alkyne was readily migrated to the chain terminus us-
ing potassium hydride/1,3-diaminopropane5 (4, 66%) and
the alcohol protected as the TBS derivative (5, 86%).
Deprotonation of the acetylene and quenching with
(Boc)2O yielded the terminal t-butyl ester 6, in 81% yield.
Cuprate addition6 to the a,b-unsaturated alkynyl ester
yielded the trisubstituted alkene 7 as a single regioisomer
in 75% yield. The geometry was confirmed by NOE
experiments, with a 7.7% enhancement observed of the
methyl group on irradiation of the alkene proton. Depro-
tection of the TBS group occurred readily on treatment
with TBAF–acetic acid7 revealing the free hydroxyl group
for esterification. Our design strategy was such that it
Key words: total synthesis, irciniasulfonic acid, multidrug resis-
tance modulator
First reported in 2001,1 the irciniasulfonic acids, 1a–c, 2
(Figure 1) were isolated from the Japanese marine sponge
Ircina sp. and have been reported as potential multidrug
resistance (MDR) modulators, reversing MDR at 33 mg/
mL against P-gp overexpressing MDR tumor cells (KB/
VJ300) in the presence of 10 ng/mL of vincristine.1,2 To
date, no synthetic effort towards these compounds has
been reported.
In planning the total synthesis, we aimed to develop a ver-
satile route that would allow preparation of both the natu-
ral and unnatural enantiomers of ISA, with the flexibility
to be able to introduce many different side chains, in ad-
dition to the three naturally occurring chains. This will be
crucial in future structure–activity-based studies. It would
also be necessary to address the formation of the trisubsti-
tuted alkene and when precisely to add the sulfonic acid
moiety. Thus our strategy involved preparation of the fat-
ty acids separately from the main section of the molecule,
with a synthesis sufficiently convergent to allow the intro-
O
Irciniasulfonic acid (ISA, 1)
-
O3S
O
OR
O
1a R =
O
1b R =
O
1c R =
2
R = H Deacylirciniasulfonic acid
Figure 1 The irciniasulfonic acids
SYNLETT 2005, No. 4, pp 0652–0654
0
4.
0
3.
2
0
0
5
Advanced online publication: 22.02.2005
DOI: 10.1055/s-2005-862387; Art ID: D28704ST
© Georg Thieme Verlag Stuttgart · New York

LETTER
First Total Synthesis of the Irciniasulfonic Acids
653
would now be possible to form the irciniasulfonic acids followed by PCC oxidation gave the pre-requisite alde-
with any side chain, and not just the three fatty acid chains hyde 19 (58% for the two steps), which was coupled with
found in the original source. Model studies were initially the non-stabilised ylid 20 derived from bromoheptane, to
performed using hexanoyl chloride, which gave the ester give exclusively the cis-olefin in good yield (21, 68%).
8 in 69% for the two steps (Scheme 1).
Acetyl hydrolysis on 21 and then, oxidation and reaction
with oxalyl chloride for both this product and 16 gave the
required acid chlorides 17 and 22, which were used imme-
diately for ester formation without characterisation.
For the natural product, three different side chains were
required. The simplest was derived from commercially
available tricosanoic acid; reaction with oxalyl chloride
gave the acid chloride, which was immediately coupled It was at this late stage that our selective esterification
with deprotected 7 to give the fatty ester in 74% yield. The strategy came into play. It was now possible to cleave
remaining two natural side chains required longer prepa- selectively the t-butyl ester using 50% TFA in dichlo-
rations. The acid chloride 17 required to introduce chain romethane, leaving the long chain esters intact (24a–d,
1b was prepared using a double Wittig reaction approach. >95% yield, irrespective of long chain ester, Scheme 3).
Initially, efforts were focussed on forming a bis-phospho- Treatment with oxalyl chloride/DMF in dichloromethane
nium bromide salt 10 from 1,4-dibromobutane, with the gave the acid chloride, which was not isolated, but taken
aim of performing a double Wittig reaction;8 this proved up in toluene and the by-products removed in vacuo, be-
impossible, presumably due to the poor solubility of the fore adding isethionic acid (sodium salt) in toluene and
monophosphonium bromide, which precipitated from the heating the mixture under reflux.12 After purification, the
reaction mixture, not allowing the second substitution to natural products were obtained in 29–49% yield.13 Com-
occur. It was possible, however, to utilise the monophos- pound 25 was obtained as the first reported analogue of
phonium bromide salt 11 and to generate the non-stabi- the ISA’s.
lised ylid from this in quantitative yield using n-BuLi in
THF. Reaction with the aldehyde 13, formed via the
Swern oxidation of commercial 1-pentadecanol (12), gave
the cis-olefin in 73% yield (with spectroscopic evidence
showing >11:1 cis:trans ratio). Subsequent reaction with
PPh3 followed by n-BuLi/THF again gave the non-stabi-
lised ylid which, on this occasion, was reacted with the
Repeating the reaction sequence, starting from (R)- or (S)-
propylene oxide gives access to both the natural and
unnatural enantiomers of ISA, in yields comparable with
the racemic synthesis.
In summary, we herein disclose the first total synthesis of
the irciniasulfonic acids via a convergent and flexible ap-
proach, that allows for the synthesis of both enantiomers
of the natural product and for the incorporation of a range
of side chains, enabling optimisation of biological activi-
ty. The results of these studies will be reported in due
course.
lactol 15, generated either by DIBAL reduction of d-
valerolactone9 (45% yield for reduction) or by the addi-
tion of water to 2,3-dihydropyran10 (66% for addition) to
give the desired unsaturated alcohol 16 in 63% yield
(Scheme 2).11 A Wittig approach was also employed for
1c. Mono acetyl-protection of pentadecane-1,15-diol
O
OH
OH
KH
3
H2N
NH2
4
n-BuLi/Et2AlCl/toluene, –30 °C, 1.5 h
n-BuLi/BF3·OEt2, THF, –78 °C, 1 h
Yield: 76% Ratio 98:2
Yield: 56% Ratio 100:0
66%
TBS-Cl
Imidazole/DMAP
DMF
76%
O
OTBS
OTBS
1. n-BuLi
O
2. (Boc)2O
81% (over 2 steps)
6
5
CuBr2·SMe2, MeLi
THF, –78 °C, 2 h
75%
O
OTBS
R
O
Me
Me
1. TBAF, HOAc
2.
O
O
O
7
8
O
O
R = C5H11
R
Cl
69% (over 2 steps)
Scheme 1 Synthesis of the core structure of the irciniasulfonic acids
Synlett 2005, No. 4, 652–654 © Thieme Stuttgart · New York

654
A. P. Dobbs et al.
LETTER
PPh3
89%
1. n-BuLi/THF/0 °C
Br
PPh3
Br–
Br
( )12
Br
Br
9
O
2.
11
14, 73%
2.2 equiv PPh3
Reflux
( )13
H
1. PPh3
2. n-BuLi/THF/0 °C
13
X
3.
O
OH
15
(COCl)2, DMSO
NEt3, CH2Cl2, –78 °C
100%
PPh3
Br –
Ph3P
Br –
10
OH
( )13
12
HO
( )12
16, 63%
1. CrO3, H2SO4, H2O/acetone
2. (COCl)2, DMSO
Et3N, CH2Cl2, –78 °C
3. (COCl)2, DMFcat, THF, reflux
HO
( )13 OH
AcCl
O
O
( )12
Cl
PCC
17
AcO
OH
AcO
H
( )13
( )13
18
19, 58% (2 steps)
1. PPh3
Br
PPh3
2. n-BuLi
20
THF, 0 °C
O
1. LiOH/aq THF
AcO
Cl
( )13
( )13
2. PCC
3. (COCl)2, DMFcat
21, 68%
22
Scheme 2 Synthesis of the fatty acid side chains
OTBS
Me
References
O
(1) Kawakami, A.; Miyamoto, T.; Higuchi, R.; Uchiumi, T.;
Kuwano, M.; Van Soest, R. W. M. Tetrahedron Lett. 2001,
42, 3335.
(2) Nagle, D. G.; Zhou, Y.-D.; Mora, F. D.; Mohammed, K. A.;
Kim, Y.-P. Curr. Med. Chem. 2004, 11, 1725.
(3) Yamaguchi, M.; Nobayashi, Y.; Hirao, I. Tetrahedron 1984,
40, 4261.
(4) (a) Suzuki, T.; Saimoto, H.; Tomioka, H.; Oshima, K.;
Nozaki, H. Tetrahedron Lett. 1982, 23, 3597. (b)Matthews,
R. S.; Eickhoff, D. J. J. Org. Chem. 1985, 50, 3923.
(5) (a) Schweitzer, S.; Voss, G.; Gerlach, H. Liebigs Ann. Chem.
1994, 189. (b) Hamada, T.; Daikai, K.; Irie, R.; Katsuki, T.
Tetrahedron: Asymmetry 1995, 6, 2441.
(6) Vuligonda, V.; Thacher, S. M.; Chandraratna, R. A. S. J.
Med. Chem. 2001, 44, 2298.
(7) Bachi, M. D.; Bar-Ner, N.; Melman, A. J. Org. Chem. 1996,
61, 7116.
(8) (a) Dower, W. V.; Peter, K.; Vollhardt, C. Tetrahedron
1986, 42, 1873. (b) Spiteller, D.; Boland, W. J. Org. Chem.
2003, 68, 8743.
7
O
O
O
1. HF/pyridine
2. Acid chloride
Acid Chloride Yield
R
R
O
O
Me
23a Hexanoyl
23b Tricosanoyl
23c
23d
69%
72%
66%
62%
O
17
22
TFA/CH2Cl2
>95%
Me
O
O
24a–d >95%
OH
1. (COCl)2, DMFcat, CH2Cl2
O
2. , toluene, reflux
SO3 Na
HO
R
O
25
1a
1b
1c
49%
47%
43%
29%
Me
O
SO3 Na
O
Scheme 3
(9) Ley, S. V.; Santafianos, D.; Blaney, W. M.; Simmonds, M.
S. J. Tetrahedron Lett. 1987, 28, 221.
(10) Raederstorff, D.; Shu, A. Y. L.; Thompson, J. E.; Djerassi,
C. J. Org. Chem. 1987, 52, 2337.
(11) A related approach to (5Z,9Z)-dienoic acids has been
reported. See: Carballeira, N. M.; Emiliano, A.; Guzmán, A.
Chem. Phys. Lipids 1999, 100, 33.
(12) Watanabe, T.; Hayashi, K.; Yoshimatsu, S.; Sakai, K.;
Takeyama, S.; Takashima, K. J. Med. Chem. 1980, 23, 50.
(13) All compounds gave satisfactory analytical and
spectroscopic data. Compound data for the three natural
products in good agreement with the values given in ref. 1.
Acknowledgment
We thank the Marie-Curie Training Site scheme of the European
Union for support of Alberto Venturelli under Training Site HPMT-
CT-2000-00016 (PODEUM) and Pfizer (Summer Studentship to
Laura Butler) for funding. We are grateful to Dr. Tom Miyamoto for
supplying a copy of a 1H NMR spectrum of an authentic sample of
ISA (1a).
Synlett 2005, No. 4, 652–654 © Thieme Stuttgart · New York