Concise enantioselective synthesis of the ten-membered lactone cephalosporolide G and its C-3 epimer

Article abstract of DOI:10.1002/chem.200901735

A short and highly stereo-selective sequence for the first enantioselective total synthesis of the naturally occurring 10-membered lactone, which was obtained in only eight steps, was reported. The macrolactone ring was carried out by the use of a high-yielding pyridinium chlorochromate (PCC)-mediated oxidative cleavage of a bicyclic intermediate, generated in a domino sequence from a p-peroxyquinol. The synthesis was started with (-)-rhododendrol, which was obtained by enzymatic resolution of the racemic derivative. Phenol (R)-5 was submitted to an oxidative dearomatisation process with singlet oxygen, generated from Oxone in the presence of NaHCO₃. The treatment of compound peroxyquinol with para-toluene sulfonic acid followed by Triton B gave, in one step and 49% yield, the tricyclic epoxide bicyclic derivative. A similar route was employed for the synthesis of the C-3 diastereoisomer of the natural product, which was obtained in only 7 steps and 15.2% overall yield.

Full text of DOI:10.1002/chem.200901735

DOI: 10.1002/chem.200901735
Concise Enantioselective Synthesis of the Ten-Membered Lactone
Cephalosporolide G and Its C-3 Epimer
Silvia Barradas, Antonio Urbano,* and M. Carmen CarreÇo*[a]
Among the naturally occurring lactones of ring sizes in
termediates in the synthesis of natural decanolides requires
a large number of hydroxy group protections and deprotec-
tions, making the synthetic sequences lengthy and very ex-
pensive.[8] For these reasons, there is a need for methodolo-
gies that allow for high-yielding construction of the macro-
lactone ring and new strategies to develop short, atom-eco-
nomic approaches to naturally occurring decanolides.
Cephalosporolide G (1; Scheme 2) was isolated in 1995
from the fungus Cephalosporium aphidicola,[9a] the absolute
configuration and specific rotation of which were not report-
the range of eight to eleven,[1] the ten-membered lactones,
also called decanolides, are the most abundant and biologi-
cally active.[2] This family of natural products displays a wide
range of biological properties, such as antibacterial and anti-
fungal activity and the inhibition of cholesterol biosynthesi-
s.[2a] Besides the macrolactone core, they share a methyl
group at C-9, but differ in the number and nature of oxygen
functionalities and the degree of unsaturation (Scheme 1).
Scheme 1. Examples of naturally occurring ten-membered lactones.
Concerning synthetic studies, the main challenge has been
the efficient construction of the medium-sized ring, since
cyclisations of acyclic precursors are difficult due to enthal-
py and entropy factors.[3] The two major approaches in-
volved the intramolecular lactonisation of activated hydrox-
yacids,[4] mainly by using the method developed by Yamagu-
chi, which generally proceeded in moderate yields[5] and the
ring-closing metathesis[6] of acyclic olefins, which is still lim-
ited in scope because the control of E/Z stereochemistry of
the double bond generated is difficult and because of the
low reactivity of the acyclic precursor when dense function-
ality close to the reaction centre exists.[7] Moreover, the use
of the chiral pool for the asymmetric preparation of key in-
Scheme 2. Retrosynthetic analysis of 1.
ed by the authors. Until very recently, none of the members
of the cephalosporolides,[9] a subfamily within the decano-
lides, had been synthesised. In 2008, Krishna and Sreeshai-
lam reported the first total synthesis of 4-OMe-cephalospor-
olide C (Scheme 1) from l-malic acid and (2R)-2,3-O-cyclo-
hexylideneglyceraldehyde in a lengthy 20-step reaction se-
quence, including three hydroxy group protection–deprotec-
tion steps, and a Yamaguchi macrolactonisation (48% yield)
for the key construction of the 10-membered ring.[10]
Herein, we describe the first enantioselective synthesis of
the natural decanolide 1 and its C-3 epimer in only eight
and seven steps, respectively, starting from the natural
phenol (R)-rhododendrol (5). Our strategy features three
highly selective oxidations to introduce the oxygenated func-
tionality at the C-1, C-4 and C-6 positions of the natural
product and the whole carbon skeleton is already present in
the starting material. Moreover, the key formation of the
[a] S. Barradas, Dr. A. Urbano, Prof. Dr. M. C. CarreÇo
Departamento de Quꢀmica Orgꢁnica (C-I)
Universidad Autꢂnoma de Madrid
Cantoblanco, 28049-Madrid (Spain)
Fax : (+34)91-4973966
E-mail: antonio.urbano@uam.es
carmen.carrenno@uam.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901735.
9286
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9286 – 9289

COMMUNICATION
macrolactone ring is carried out by the use of a high-yield-
ing pyridinium chlorochromate (PCC)-mediated oxidative
cleavage of a bicyclic intermediate, generated in a domino
sequence from a p-peroxyquinol.
derivative.[11] We submitted phenol (R)-5 to an oxidative
dearomatisation process with singlet oxygen, generated from
Oxone in the presence of NaHCO3,[12] to afford 4, in 65%
yield. The treatment of compound 4 with para-toluene sul-
fonic acid (p-TsOH; 0.12 equiv) followed by Triton B
(0.24 equiv) gave, in one step and 49% yield, the tricyclic
epoxide 3. The formation of 3 occurred in an efficient cata-
lytic tandem process[13] beginning with the acid-catalysed
conjugate addition of the secondary OH to one of the
double bonds of the cyclohexadienone moiety of 4, followed
by the base-catalysed epoxidation of the other double bond
of the molecule by the hydroperoxy group acting as an in-
tramolecular epoxidation reagent. The whole sequence oc-
curred in a highly chemo- and diastereoselective manner
leading to the exclusive formation of 3.
The original plan to access 1 included the reduction of the
carbonyl group of 3 to the corresponding S-carbinol to in-
stall the correct stereochemistry present at C-3 in the natu-
ral product (Scheme 3). Nevertheless, the reduction of
ketone 3 with NaBH4 exclusively afforded the alcohol (R)-6
in 96% yield. Reaction of 3 with diisobutylaluminum hy-
dride (DIBAL-H) or the bulky hydride l-Selectride also oc-
curred in a highly stereoselective manner to afford the same
R-configured carbinol 6 as the only detected isomer. The
origin of this stereoselectivity could be found in the struc-
ture of the cis-fused bicyclic derivative 3, in which only the
upper convex face of the carbonyl group is available for the
hydride approach (Scheme 3).
The retrosynthetic analysis of 1 from 5 is shown in
Scheme 2. Decanolide 1 could be synthesised by silyl depro-
tection and regioselective reductive opening of epoxide 2,
which could be prepared from the bicyclic derivative 3 by
stereoselective carbonyl reduction and a controlled oxida-
tive cleavage–ring-expansion process to generate the macro-
lactone. Compound 3, in turn, could be formed by a stereo-
controlled conjugate cyclisation/intramolecular epoxidation
of the peroxyquinol 4, which is easily available by oxidative
dearomatisation of 5. All of the carbon atoms in the final
target 1 can be recognised in the p-alkyl-substituted phenol
5.
As depicted in Scheme 3, we started the synthesis with
(À)-rhododendrol (R)-5 (>99% enantiomeric excess (ee)),
which was obtained by enzymatic resolution of the racemic
In view of this result, we decided to continue the synthesis
of the C-3 epimer of the natural cephalosporolide G to vali-
date our initially proposed strategy. After protection of the
secondary carbinol of compound 6 as the tert-butyldimethyl-
silyl (TBDMS) derivative
7 (TBDMSOTf, 2,6-lutidine,
CH2Cl2, 95%, OTf=trifluoromethanesulfonate), we under-
took the key step, which was the formation of the 6-keto 10-
membered lactone moiety by an oxidative cleavage of the
À
C-4a C-8a bond of the tricyclic epoxide 7 (Scheme 3). Ini-
tially, we made several attempts with oxidants, such as
RuCl3/NaIO4,[14] phenyliodonium
ACTHNUTRGEN(UNG III) diacetate (PIDA)/I2
[15]
or PIDA/I2 followed by meta-chloroperbenzoic acid
(mCPBA)/BF3·OEt2/pyridine (Pyr),[16] previously used for
similar substrates. Nevertheless, these methods gave com-
plex reaction mixtures in which the products of oxidation at
different positions could be detected showing an evident
lack of selectivity. To our delight, we found that PCC in the
presence of NaOAc (RT, 1 h), cleanly and selectively oxi-
dised compound 7 to the 6-keto decanolide 8, with a re-
markable 68% yield after chromatographic purification.
The initial reaction of the tertiary alcohol at C-4a of 7 with
PCC would form the corresponding chromate intermediate
7A, which could then initiate an intramolecular oxidation at
À
C-8a and fragmentation of the C-4a C-8a single bond, with
Scheme 3. Total synthesis of 3-epi-cephalosporolide
G (10) (7 steps,
15.2% overall yield). a) Oxone, NaHCO3, H2O/CH3CN, RT, 1 h, 65%;
b) i) p-TsOH (0.12 equiv), CHCl3, À208C, 3.5 h; ii) triton B (0.24 equiv),
CHCl3, RT, 4 h, 49%; c) NaBH4, EtOH, RT, 1.5 h, 96%; d) TBDMSOTf,
2,6-lutidine, CH2Cl2, 08C to RT, 30 min, 95%; e) PCC, NaOAc, CH2Cl2,
RT, 1 h, 68%; f) Pyr·HF, CH3CN, 08C to RT, 2 h, 81%; g) Al (Hg), THF/
EtOH/H2O, RT, 95 %.
concomitant oxidation of the carbinol at C-8a to the corre-
sponding ketone, to afford the 10-membered ring lactol 7B.
Finally, elimination of the ipso proton and reduction of the
chromium unit would give the required 6-keto 10-membered
lactone 8.
Chem. Eur. J. 2009, 15, 9286 – 9289
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9287

M. C. CarreÇo, A. Urbano, and S. Barradas
With decanolide 8 in hand, we recovered the free carbinol
at C-3 by treatment with HF·Pyr to afford alcohol 9 in 81%
yield. Finally, regioselective opening of the epoxide ring of 9
with sodium amalgam[17] gave, in 95% yield, compound 10,
which is the C-3 epimer of the naturally occurring decano-
lide 1. The ten-membered lactone structure, as well as the
correct configuration of the stereogenic centres of 10, was
demonstrated by X-ray analysis.[18] Thus, we have described
the total synthesis of 10 in only 7 steps from (R)-rhododen-
drol (5), in 15.2% overall yield.
Having established the viability of our approach to the
highly functionalised ten-membered lactone core of cephalo-
sporolides, we turned our attention to the preparation of a
precursor of the natural product 1 with the correct S abso-
lute configuration at the C-3 position. Taking into account
the observed stereochemical course in the reduction of the
epoxy ketone 3 (Scheme 3), we decided to evaluate the be-
haviour, in the reduction step, of a similar bicyclic ketone,
such as 12 (Scheme 4), which lacks the epoxide ring. The
synthesis of 12 started with treatment of 5 with Oxone in
the presence of NaHCO3, followed by the in situ addition of
a reductant such as Na2S2O3, affording the p-quinol 11 in
53% yield. The efficient differentiation of the two diastereo-
topic faces and the two diastereotopic double bonds of the
cyclohexadienone moiety of 11 could be achieved by a ste-
reoselective (95:5 dr) conjugate addition of the secondary
OH of the hydroxybutyl chain at C-4 to one of the double
bonds of 11, promoted by a catalytic amount of p-TsOH.
The bicyclic, cis-fused a,b-unsaturated ketone 12 was thus
isolated in 79% yield. In this case, the addition of a small
hydride source, such as DIBAL-H, to ketone 12 afforded an
82:18 mixture of alcohols (R)-13 and (S)-14; this is the first
time that attack on the lower face has been observed, albeit
in a very poor ratio. At this point, we reasoned that the ad-
dition of a bulky Lewis acid, prior to the addition of the hy-
dride, could modify this stereochemical behaviour because
the initial coordination to the carbonyl group and/or the
free OH would preferentially take place from the less en-
cumbered upper face of ketone 12, thus hindering the hy-
dride approach and possibly favouring the desired lower-
face hydride attack (Scheme 4). Indeed, when we submitted
ketone 12 to reduction with NaBH4 in the presence of
CeCl3·7H2O, a 55:45 mixture of carbinols (R)-13 and (S)-14
was obtained, the required lower-face attack still being the
minor one. Pleasingly, the reaction of compound 12 with
DIBAL-H in the presence of the exceptionally bulky Lewis
acidic reagent methyl aluminium bis(2,6-di-tert-butyl-4-
methylphenoxide) (MAD)[19] gave rise to a 30:70 mixture of
alcohols (R)-13 and (S)-14, from which the desired diaste-
reoisomer (S)-14 could be isolated in pure form with a re-
markable 65% yield after chromatographic separation. The
exact structure of the bicyclic carbinol (S)-14 could be se-
cured by X-ray analysis.[18] It is worth mentioning that when
we tried the reduction of epoxy ketone 3 (Scheme 3) in the
presence of CeCl3·7H2O or MAD, again carbinol (R)-6, re-
sulting from upper-face attack, could be detected as the only
reaction product.
Having installed the correct stereochemistry at C-3, we
continued the synthetic plan to access 1 as depicted in
Scheme 4. Firstly, we protected the secondary alcohol of
compound 14 as the TBDMS derivative 15 (TBDMSCl, imi-
dazole, DMAP, CH2Cl2, RT, 3 h, DMAP=4-dimethylamino-
pyridine) in 96% yield. Next, we treated the bicyclic deriva-
tive 15 with mCPBA to afford the tricyclic epoxide 16 selec-
tively, as the only diastereomer in 93% yield, after epoxida-
tion exclusively on the upper face of the double bond of 15.
With the appropriate functionality for the natural product
introduced, we undertook the key step: the generation of
the decanolide structure. Thus, treatment of the tricyclic ep-
oxide 16 with PCC in the presence of NaOAc gave rise to
the highly functionalised lactone 2 in an excellent 80%
yield, after an efficient oxidative cleavage–ring-expansion
process. The correct structure of decanolide 2 was confirmed
by X-ray analysis.[18] Finally, deprotection of the TBDMS
group of 2 (HF·Pyr, CH3CN, 08C to RT, 81%) followed by
regioselective reductive opening of the epoxide of 17 (Al
(Hg), THF/EtOH/H2O, RT, 82%) afforded 1, with the
Scheme 4. Total synthesis of cephalosporolide G (1) (8 steps, 12.9% over-
all yield). a) i) Oxone, NaHCO3, H2O/CH3CN, RT, 1 h; ii) Na2S2O3, RT,
5 min, 53%; b) p-TsOH (0.12 equiv), CHCl3, À208C, 3.5 h, 79%;
c) MAD, 08C, 1 h, then DIBAL-H, À788C, 14 h, 65%; d) TBDMSCl,
imidazole, DMAP, CH2Cl2, RT, 3 h, 96%; e) mCPBA, CH2Cl2, RT, 14 h,
93%; f) PCC, NaOAc, CH2Cl2, RT, 2.5 h, 80%; g) HF·Pyr, CH3CN, 08C
to RT, 1.5 h, 81%; h) Al (Hg), THF/EtOH/H2O, RT, 82 %.
9288
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 9286 – 9289

Cephalosporolide G
COMMUNICATION
Kreiter, K. Lehr, M. Leyendecker, W. S. Sheldrik, R. Exner, Chem.
Ber. 1991, 124, 3–12.
[4] For a general review, see: A. Parenty, X. Moreau, J.-M. Campagne,
Chem. Rev. 2006, 106, 911–939.
3S,4R,9R absolute configuration at the stereogenic centres,
showing identical NMR spectroscopy data to those de-
scribed for the natural product.[9a]
In summary, we have described a short and highly stereo-
selective sequence for the first enantioselective total synthe-
sis of the naturally occurring 10-membered lactone 1, which
was obtained in only 8 steps and 12.9% overall yield start-
ing from (R)-rhododendrol. Several key features of this ap-
proach are noteworthy. The starting material already pos-
sesses all of the carbon atoms present in the final target and
only one protection–deprotection step is necessary. On the
other hand, three selective oxidations allowed the introduc-
tion of the oxygenated functionality present in the natural
product and allowed the efficient construction of the ten-
membered lactone ring of the final target: an oxidative dear-
omatisation of the starting phenol, a hydroxy-directed epox-
idation of an olefin intermediate, and a PCC-mediated oxi-
dative cleavage–ring-expansion process of an angular hy-
droxy bicyclic derivative. A similar route was employed for
the synthesis of the C-3 diastereoisomer of the natural prod-
uct, which was obtained in only 7 steps and 15.2% overall
yield. These syntheses of 1 and the C-3 epimer 10 compare
favourably with the most efficient approaches to any
member of the naturally occurring ten-membered lactones
with similar complexity and, to the best of our knowledge,
are the shortest reported to date for the construction of
highly functionalised decanolides.
[5] 32–45% yield; for examples, see: a) M. B. Andrus, T.-L. Shih, J.
Org. Chem. 1996, 61, 8780–8785; b) P. Radha Krishna, P. V. Nara-
simha Reddy, Tetrahedron Lett. 2006, 47, 7473–7476; c) Y. Kobaya-
shi, S. Yoshida, M. Asano, A. Takeuchi, H. P. Acharya, J. Org.
Chem. 2007, 72, 1707–1716; d) P. Gupta, P. Kumar, Eur. J. Org.
Chem. 2008, 1195–1202.
[6] For recent reviews, see: a) A. Gradillas, J. Pꢅrez-Castells, Angew.
Chem. 2006, 118, 6232–6247; Angew. Chem. Int. Ed. 2006, 45, 6086–
6101; b) K. C. Nicolaou, P. G. Bulger, D. Sarlah, Angew. Chem. 2005,
117, 4564–4601; Angew. Chem. Int. Ed. 2005, 44, 4490–4527 K;
c) R. H. Grubbs, Tetrahedron 2004, 60, 7117–7140; d) S. J. Connon,
S. Blechert, Angew. Chem. 2003, 115, 1944–1968; Angew. Chem. Int.
Ed. 2003, 42, 1900–1923; e) A. Fꢆrstner, Angew. Chem. 2000, 112,
3140–3172; Angew. Chem. Int. Ed. 2000, 39, 3012–3043.
[7] D. K. Mohapatra, D. K. Ramesh, M. A. Giardello, M. S. Chorghade,
M. K. Gurjar, R. H. Grubbs, Tetrahedron Lett. 2007, 48, 2621–2625,
and references therein.
[8] See refs. [2b], [2c] and [5d], and the following: a) C. V. Ramana,
T. P. Khaladkar, S. Chatterjee, M. K. Gurjar, J. Org. Chem. 2008, 73,
3817–3822; b) J. Liu, L. Zhang, J. He, L. He, B. Ma, X. Pana, X.
She, Tetrahedron: Asymmetry 2008, 19, 906–911; c) M. Matsuda, T.
Yamazaki, K.-I. Fuhshuku, T. Sugai, Tetrahedron 2007, 63, 8752–
8760; d) S. Ghosh, R. V. Rao, Tetrahedron Lett. 2007, 48, 6937–
6940; e) P. R. Krishna, M. Narsingam, Synthesis 2007, 3627–3634.
[9] a) A. Farocq, J. Gordon, J. R. Hanson, J. A. Takahashi, Phytochemis-
try 1995, 38, 557–558; b) M. J. Ackland, J. R. Hanson, P. B. Hitch-
cock, A. H. Ratcliffe, J. Chem. Soc. Perkin Trans. 1 1985, 843–847;
c) V. Rukachaisirikul, S. Pramjit, Ch. Pakawatchai, M. Isaka, S. Su-
pothina, J. Nat. Prod. 2004, 67, 1953–1955.
[10] P. R. Krishna, A. Sreeshailam, Synlett 2008, 2795–2798.
[11] Y. Yuasa, S. Shibuya, Y. Yuasa, Synth. Commun. 2003, 33, 1469–
1475.
[12] M. C. CarreÇo, M. Gonzꢁlez-Lꢂpez, A. Urbano, Angew. Chem.
2006, 118, 2803–2807; Angew. Chem. Int. Ed. 2006, 45, 2737–2741.
[13] S. Barradas, M. C. CarreÇo, M. Gonzꢁlez-Lꢂpez, A. Latorre, A.
Urbano, Org. Lett. 2007, 9, 5019–5022.
Acknowledgements
We thank MICINN (grant no. CTQ2008-04691) for financial support.
S.B. wishes to thank MEC for a fellowship.
[14] a) H. M. C. Ferraz, L. S. Longo, Jr., J. Org. Chem. 2007, 72, 2945–
2950; b) H. M. C. Ferraz, L. S. Longo, Jr., Org. Lett. 2003, 5, 1337–
1339.
[15] a) G. H. Posner, M. A. Hatcher, W. A. Maio, Org. Lett. 2005, 7,
4301–4303; b) M. A. Hatcher, K. Borstnik, G. H. Posner, Tetrahe-
dron Lett. 2003, 44, 5407–5409.
Keywords: epoxidation · lactones · oxidative cleavage ·
oxidative dearomatization · total synthesis
[16] P. Wipf, W. Li, J. Org. Chem. 1999, 64, 4576–4577.
[17] M. Honzumi, K. Ogasawara, Tetrahedron Lett. 2002, 43, 1047–1049.
[18] CCDC-725048, 725049 and 725050 for 2, 10 and 14, respectively,
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from the Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/
cif.
[19] a) T. Ooi, K. Maruoka, Rev. Heteroat. Chem. 1998, 18, 61–85; b) K.
Maruoka, Y. Araki, H. Yamamoto, J. Am. Chem. Soc. 1988, 110,
2650–2652.
[1] a) G. Rousseau, Tetrahedron 1995, 51, 2777–2849; b) L. S. Lon-
go, Jr., F. I. Bombonato, H. M. C. Ferraz, Quim. Nova 2007, 30,
415–424; c) I. Shiina, Chem. Rev. 2007, 107, 239–273; d) H. M. C.
Ferraz, F. I. Bombonato, M. K. Sano, L. S. Longo, Jr., Quim. Nova
2008, 31, 885–900.
[2] For reviews, see: a) G. Drꢄger, A. Kirschning, R. Thiericke, M.
Zerlin, Nat. Prod. Rep. 1996, 13, 365–375; b) M. C. Ferraz, F. I.
Bombonato, L. S. Longo, Jr., Synthesis 2007, 3261–3285; c) V. B.
Riatto, R. A. Pilli, M. M. Victor, Tetrahedron 2008, 64, 2279–2300.
[3] a) G. Illuminati, L. Mandolini, Acc. Chem. Res. 1981, 14, 95–102;
b) L. Mandolini, Adv. Phys. Org. Chem. 1986, 22, 1–111; c) C. G.
Received: June 23, 2009
Published online: August 5, 2009
Chem. Eur. J. 2009, 15, 9286 – 9289
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9289