Total synthesis, assignment of the absolute stereochemistry, and structure-activity relationship studies of subglutinols A and B

Article abstract of DOI:10.1002/asia.201000147

Immunosuppressive drugs are used to prevent rejection of transplanted organs and treat autoimmune diseases. Clinically approved immunosuppressive drugs possess undesirable side effects, including acute neurological toxicity, chronic nephrotoxicity, and osteoporosis. As a result, considerable efforts have been devoted to the identification of immunosuppressive natural products that lack cytotoxicity and undesirable side effects on bone structure. Subglutinols A (1a) and B (1b) are diterpene pyrones isolated from Fusarium subglutinans. Compounds 1a and 1b are equipotent in the mixed lymphocyte reaction assay and thymocyte proliferation assay (IC₅₀=0.1 μm). Owing to the lack of toxicity, 1a and 1b are expected to be promising new immunosuppressive drugs. Herein, we detail our efforts that have culminated in a stereoselective synthesis of 1a and 1b from the (S)-(+)-5-methyl-Wieland-Miescher ketone and determined their absolute stereochemistries. We also present initial biological data to show the great potential of 1a as an immunosuppressive drug with dose-dependent osteogenic activity.

Full text of DOI:10.1002/asia.201000147

FULL PAPERS
DOI: 10.1002/asia.201000147
Total Synthesis, Assignment of the Absolute Stereochemistry, and Structure-
Activity Relationship Studies of Subglutinols A and B
Hyoungsu Kim,[a] Joseph B. Baker,[a] Yongho Park,[a] Hyung-Bae Park,[b]
Patrick D. DeArmond,[a] Seong Hwan Kim,[c] Michael C. Fitzgerald,[a]
Dong-Sup Lee,[b] and Jiyong Hong*[a]
Abstract: Immunosuppressive drugs
are used to prevent rejection of trans-
planted organs and treat autoimmune
diseases. Clinically approved immuno-
suppressive drugs possess undesirable
side effects, including acute neurologi-
cal toxicity, chronic nephrotoxicity, and
osteoporosis. As a result, considerable
efforts have been devoted to the iden-
tification of immunosuppressive natural
products that lack cytotoxicity and un-
desirable side effects on bone structure.
Subglutinols A (1a) and B (1b) are di-
terpene pyrones isolated from Fusari-
um subglutinans. Compounds 1a and
1b are equipotent in the mixed lym-
phocyte reaction assay and thymocyte
Owing to the lack of toxicity, 1a and
1b are expected to be promising new
immunosuppressive drugs. Herein, we
detail our efforts that have culminated
in a stereoselective synthesis of 1a and
1b from the (S)-(+)-5-methyl-Wie-
land–Miescher ketone and determined
their absolute stereochemistries. We
also present initial biological data to
show the great potential of 1a as an
immunosuppressive drug with dose-de-
pendent osteogenic activity.
proliferation
assay
(IC50 =0.1 mm).
Keywords: immunosuppressive
agents · nucleophilic substitution ·
reductive deoxygenation · tandem
reactions · total synthesis
Introduction
munosuppressive drugs (e.g., cyclosporin A, FK506) possess
undesirable side effects, including acute neurological toxici-
ty, chronic nephrotoxicity, and osteoporosis. As a result, con-
siderable efforts have been devoted to the identification of
immunosuppressive natural products that lack cytotoxicity
and undesirable side effects on bone structure.[2]
Subglutinols A (1a) and B (1b) are diterpenoid pyrones
isolated by Clardy and co-workers from Fusarium subglutin-
ans, an endophytic fungus from the vine Tripterygium wilfor-
dii.[3] The structure and relative stereochemistry of 1a and
1b were determined by extensive NMR spectroscopic analy-
sis and X-ray diffraction analysis, but the absolute configura-
tion of the natural products has not been established. Com-
pounds 1a and 1b are equipotent in the mixed lymphocyte
reaction (MLR) assay and thymocyte proliferation (TP)
assay (IC50 =0.1 mm).[3] Cyclosporin A is roughly as potent in
the MLR assay and 104 times more potent in the TP assay.
Owing to the lack of toxicity, 1a and 1b are expected to be
promising new immunosuppressive drugs.
Immunosuppressive drugs are used to suppress unwanted
immune responses. Specifically, they are used to prevent re-
jection of transplanted organs and treat autoimmune diseas-
es, including rheumatoid arthritis, multiple sclerosis, and in-
sulin-dependent type-1 diabetes.[1] Clinically approved im-
[a] Dr. H. Kim, J. B. Baker, Y. Park, P. D. DeArmond,
Prof. Dr. M. C. Fitzgerald, Prof. Dr. J. Hong
Department of Chemistry
Duke University
Durham, NC 27708 (USA)
Fax : (+1)919-660-1605
E-mail: jiyong.hong@duke.edu
[b] H.-B. Park, Prof. Dr. D.-S. Lee
Laboratory of Immunology
Cancer Research Institute
College of Medicine
Seoul National University
Seoul 110-799 (Korea)
There have been reports on structurally related and bio-
logically interesting diterpenoid a- or g-pyrones.[4–8] Dani-
shefsky and co-workers reported the elegant synthesis of
(Æ)-sesquicillin.[9] In addition, the enantioselective synthesis
of (À)-candelalides A–C,[10] (À)-nalanthalide,[11] and (+)-ses-
quicillin[12] was completed by Katoh and co-workers.
[c] Dr. S. H. Kim
Laboratory of Chemical Genomics
Korea Research Institute of Chemical Technology
Daejeon 305-600 (Korea)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201000147.
1902
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1902 – 1910

SN2’ reaction) provided 4b as a
single diastereomer. The stereo-
chemical outcome observed in
the tandem reaction can be ra-
tionalized on the basis that the
unfavorable 1,3-diaxial interac-
tion of the C12 allyl substituent
and the C17 methyl group in
conformation 6A is larger than
that of the hydrogen and
methyl group in conformation
6B, thus preferentially afford-
ing the 2,3-trans-2,5-trans-tetra-
hydrofuran 4b. The configura-
tion of the newly formed C12
stereocenter of 4b was con-
firmed by a single-crystal X-ray
diffraction analysis.[18] To the
Recently, we disclosed a stereoselective synthesis of 1a
and 1b from the (S)-(+)-5-methyl-Wieland–Miescher
ketone and determined their absolute stereochemistry.[13]
We also presented initial biological data to show the great
potential of 1a as an immunosuppressive drug with dose-de-
pendent osteogenic activity. Herein, we detail our efforts
that have culminated in the stereoselective synthesis of 1a
and 1b. In addition, we report initial structure-activity rela-
tionship (SAR) studies and a preliminary binding study of
1a with cyclophilin A (CypA).
Results and Discussion
Scheme 1 summarizes our approach for the stereoselective
synthesis of subglutinols A (1a) and B (1b) from the
common intermediate 7. The strategy underlying our syn-
thetic plan for 4a was to apply BF3·OEt2-promoted deoxyge-
nation of the g-hydroxy ketone 5 followed by stereoselec-
tive reduction of the oxocarbenium ion intermediate to
afford 4a. For the synthesis of 4b, a novel tandem proce-
dure of olefin cross-metathesis (CM) of 7 with allyl chloride
followed by intramolecular SN2’ cyclization of the hydroxy
alkene 6 would stereoselectively provide the key intermedi-
ate 4b in the synthesis of 1b. Final coupling of the decalins
2a and 2b and the a-pyrone moiety would complete the
synthesis of 1a and 1b, respectively.
Scheme 1. Retrosynthesis of subglutinols A (1a) and B (1b).
best of our knowledge, the tandem CM/SN2’ reaction has
never been reported for the stereoselective synthesis of tet-
rahydrofurans and other heterocycles.[19,20] In fact, few ap-
proaches to the stereoselective synthesis of tetrahydrofurans
and tetrahydropyrans involve intramolecular SN2’ reactions,
perhaps because of the low nucleophilicity of oxygen and a
less well-defined transition state.[17c] Recently, we extended
this tandem CM/SN2’ reaction to the stereoselective synthe-
sis of 4-hydroxy-2,6-cis-tetrahydropyrans and applied the re-
action in a protecting-group-free synthesis of (Æ)-diospong-
Synthesis of the 2,3-trans-2,5-trans-tetrahydrofuran 4b for
subglutinol B (1b): As shown in Scheme 2, the synthesis of
the key intermediate 4b to subglutinol B (1b) began with an
olefin cross-metathesis reaction[14] of 7,[15] the common inter-
mediate readily prepared from the enantiomerically pure
(S)-(+)-5-methyl-Wieland–Miescher ketone.[16] Treatment of
7 with allyl chloride in the presence of Grubbs second-gen-
eration catalyst and subsequent intramolecular SN2’ reac-
tion[17] of the corresponding hydroxy alkene 6 (tandem CM/
AHCTUNGTRENNUNG
in A.[21]
Chem. Asian J. 2010, 5, 1902 – 1910
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1903

J. Hong et al.
FULL PAPERS
Scheme 4. BF3·OEt2-promoted reductive deoxygenation of cyclic hemike-
tals: a) O3, EtOAc, À788C, 5 min; PPh3, 258C, 6 h, 100%; b) for 9a:
i) CH2=CHMgBr, THF, À78 to 08C, 7 h; ii) MnO2, CH2Cl2, 258C, 10 h,
28% for 2 steps; for 9b: i) PCC, NaOAc, CH2Cl2, 4 ꢁ MS, 48C, 12 h,
71%; ii) (E)-TMSCH=CHLi, THF, À788C, 3.5 h, 21% (BRSM);
c) BF3·OEt2, CH2Cl2/CH3CN 3:1, NaBH3CN, À788C, for 11a: 40 min,
76%; for 11b: 20 min, 61%. PCC=pyridinium chlorochromate. TMS=
trimethylsilyl.
Scheme 2. Synthesis of the 2,3-trans-2,5-trans-tetrahydrofuran 4b for sub-
glutinol B (1b): a) allyl chloride, Grubbs second-generation catalyst
(20 mol%), CH2Cl2, reflux, 30 h, 53% (76% based on recovered starting
material (BRSM).
9a. As previously described, we expected that BF3·OEt2-
promoted deoxygenation of the g-hydroxy ketone 9a fol-
lowed by reduction of the oxocarbenium ion intermediate
10A with a reducing agent (Et3SiH or NaBH3CN) would
preferentially provide the 2,3-trans-2,5-cis-tetrahydrofuran
11a by addition of the hydride from the direction opposite
to the C17 methyl group (Scheme 4). However, the reaction
conditions for reductive deoxygenation (BF3·OEt2,
NaBH3CN, À788C, 10 min) gave the 2-ethyltetrahydrofuran
11a instead of the desired 2-ethenyltetrahydrofuran 4a
through 1,4-addition of a hydride to 10A followed by 1,2-
addition of a hydride. Nevertheless, careful analysis of
1H NMR spectral data revealed that the configuration of
C12 in 11a had the desired S configuration. Encouraged by
the stereochemical outcome, we expected that blocking 1,4-
addition of a hydride by increasing steric hindrance at C14
of the oxocarbenium ion intermediate would stereoselec-
tively provide the desired 2-ethenyltetrahydrofuran 4a. Dis-
appointingly, the reductive deoxygenation reaction of 9b ex-
clusively produced 11b by 1,4-addition of a hydride, which
led us to turn our attention to alkyne derivatives of 9a.
As shown in Scheme 5, g-hydroxy alkynones 5, 13b, and
13c were prepared from the lactol 8 by addition of lithium
acetylides to 8 followed by MnO2 oxidation of the resulting
propargylic alcohols 12b–d. Compound 13a was prepared
from 12c by desilylation followed by MnO2 oxidation of the
resulting propargylic alcohol 12a.
Synthesis of the 2,3-trans-2,5-cis-tetrahydrofuran 4a for sub-
glutinol A (1a): With completion of the 2,3-trans-2,5-trans-
tetrahydrofuran 4b for subglutinol B (1b), we turned our at-
tention to the stereoselective synthesis of the 2,3-trans-2,5-
cis-tetrahydrofuran 4a, the key intermediate to subglutino-
l A (1a).
The strategy underlying our synthetic plan for D was to
apply BF3·OEt2-promoted deoxygenation of the cyclic hemi-
ketal B followed by stereoselective reduction of the oxocar-
benium ion intermediate C to afford D (Scheme 3).[22] We
anticipated a hydride to be added to the oxocarbenium ion
intermediate C from the direction opposite to the C17
methyl group to stereoselectively provide the 2,3-trans-2,5-
cis-tetrahydrofuran D for the synthesis of 1a.
Compounds 5 and 13a–c were subjected to the conditions
for the BF3·OEt2-promoted reductive deoxygenation reac-
tion (Table 1). Neither a hydrogen atom nor TMS group at
C14 completely blocked the 1,4-addition of a hydride
(Table 1, entries 1 and 2). However, we were pleased to find
that the TES or TIPS group effectively blocked the 1,4-addi-
tion of hydride to afford 14c and 14d as single diastereo-
mers, respectively (entries 3 and 4). Both steric and elec-
tronic factors seem to affect the reaction outcome.
Scheme 3. An approach to 2,3-trans-2,5-cis-tetrahydrofuran 4a in the syn-
thesis of subglutinol A (1a).
Ozonolysis of 7 afforded the lactol 8 in quantitative yield
(Scheme 4). Addition of CH2=CHMgBr to 8 followed by
MnO2 oxidation proceeded to provide the g-hydroxy ketone
1904
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1902 – 1910

Total Synthesis of Subglutinols A and B
First generation synthesis of subglutinols A and B and es-
tablishment of their absolute stereochemistry: With the key
intermediates 4a and 4b in hand, we turned our attention to
the installation of the a-pyrone moiety.
Compounds 4a and 4b were converted to the appropri-
ately functionalized aldehydes 19a and 19b by following the
procedures established by Danishefsky and Katoh
(Scheme 7).[9–11] The initial attempt to install the isopropyli-
Scheme 5. Preparation of g-hydroxy alkynones 5 and 13a–c: For 13a:
a) i) (triethylsilyl)acetylene, nBuLi, THF, À788C, 10 min; then 8, À78 to
08C, 2 h, 61%; ii) TBAF, THF, 258C, 30 min, 78%; b) MnO2, CH2Cl2,
258C, 6 h, 66%; for 13b, 13c, and 5: a) R-acetylene, nBuLi, THF, À788C,
10 min; then 8, À78 to 08C, 2–5 h; b) MnO2, CH2Cl2, 258C, 1–6 h, 13b:
76% for 2 steps, 13c: 58% for 2 steps, 5: 85% for 2 steps. TBAF=tetra-
butylammonium fluoride, TES=triethylsilyl, TIPS=triisopropylsilyl.
Table 1. Outcome of the reductive deoxygenation reaction.
Scheme 7. Synthesis of appropriately functionalized aldehyde 19a and
19b: a) aq HCl, THF, 608C, 3 h, 15a: 91%, 15b: 98%; b) NaH, ethylfor-
mate, THF, reflux, 1 h; c) ethyl vinyl ether, PPTS, 258C, 4 h, 16a: 67%
for 2 steps, 16b: 85% for 2 steps; d) NaBH4, EtOH, 08C, 30 min;
e) MsCl, Et3N, CH2Cl2, 0 to 258C, 30 min, 17a: 81% for 2 steps, 17b:
87% for 2 steps; f) NaBH4, EtOH, 08C, 1 h, 3a: 95%, 3b: 88%;
g) nBu3SnCH2I, KH, [18]crown-6, THF, 08C, 30 min, 18a: 85%, 18b:
88%; h) nBuLi, hexanes, À50 to 258C, 1.5 h, 2a: 70%, 2b: 71%; i) Dess–
Martin periodinane, NaHCO3, CH2Cl2, 258C, 1 h, 19a: 97%, 19b: 86%.
PPTS=pyridinium p-toluenesulfonate.
Entry
Substrate
Ratio of 14:11[a]
Yield [%][b]
1
2
3
4
13a
13b
13c
5
1:1
8:1
14c only
14d only
29
67
86
91
[a] Determined by integration of the 1H NMR spectrum of the crude
product. [b] Yield of isolated 14.
Deprotection of the TIPS group of 14d by treatment with
TBAF followed by partial reduction of 14a with Lindlarꢂs
catalyst gave the key intermediate 4a to subglutinol A (1a)
(Scheme 6). The configuration of the newly formed C12 ste-
reocenter of 4a was confirmed by single-crystal X-ray dif-
fraction analysis.[23]
dene group by CM reaction of 4b with 2-methylpropene fol-
lowed by deprotection of the ketal group under various con-
ditions resulted in the epimerization at C12 (a/b 3:1) and
led us to attempt the installation of the isopropylidene
group at a later stage of the synthesis. Acid-catalyzed ketal-
deprotection of 4a and 4b afforded the corresponding
ketone 15a and 15b, respectively, without the epimerization
at C12. Formylation of 15a and 15b followed by ethoxyeth-
yl-protection smoothly proceeded to give 16a and 16b, re-
spectively. Reduction, dehydration, reduction, and stannyl-
methylation set the stage for the stereocontrolled [2,3]-
Wittig rearrangement. Upon treatment of 18a and 18b with
nBuLi, Wittig rearrangement smoothly proceeded to pro-
vide 2a and 2b (diastereomeric ratio>10:1). Dess–Martin
oxidation of the rearrangement products 2a and 2b afforded
the decalin aldehydes 19a and 19b, in 97 and 86%, respec-
tively.
First, we chose a direct coupling of the decalin moiety E
and the a-pyrone moiety F to complete the synthesis of 1a
and 1b (Scheme 8). We anticipated that the installation of
the a-pyrone moiety could be achieved by a direct coupling
of the metallated a-pyrone moiety F with the activated dec-
Scheme 6. Synthesis of 2,3-trans-2,5-cis-tetrahydrofuran 4a for the synthe-
sis of subglutinol A (1a): a) TBAF, THF, 258C, 1 h, 97%; b) H2, Lindlarꢂs
catalyst (10 wt.%), EtOAc/pyridine (30:1), 258C, 3 h, 99%.
Chem. Asian J. 2010, 5, 1902 – 1910
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1905

J. Hong et al.
FULL PAPERS
Scheme 8. Attempts for direct coupling of decalin and a-pyrone.
alin moiety E. Unfortunately, arising from the steric conges-
tion around the decalin moiety, our efforts for direct cou-
pling reactions resulted in no reaction, which led us to adopt
Katohꢂs procedures.[10,11]
The coupling reaction of 19b with the 3-lithio-a-pyrone
(generated by treatment of 20 with nBuLi)[10,11] followed by
a Barton–McCombie deoxygenation reaction successfully
provided 22 (Scheme 9). Olefin cross-metathesis of 22 with
2-methylpropene in the presence of Grubbs second-genera-
tion catalyst provided 23. However, our attempts to convert
23 to subglutinol B (1b) under various conditions (NaOH,
HCl, BCl3, BBr3, or TMSI) resulted in either no reaction or
complex reaction mixtures. Since the a-pyrone moiety is an
equivalent to vinylogous carbonate, methyl-deprotection of
23 might be challenging under mild conditions. We envi-
sioned that addition of the 3-lithio-g-pyrone (generated by
treatment of 24 with nBuLi),[10,11] a vinylogous ester, fol-
lowed by hydrolysis and subsequent tautomerization would
provide the desired a-pyrone moiety. Addition of the 3-
lithio-g-pyrone to 19b, Barton–McCombie deoxygenation,
and olefin cross-metathesis provided 27b (Scheme 9). As we
expected, final hydrolysis of 27b under basic conditions
(NaOH, THF/MeOH, reflux, 8 h) followed by tautomeriza-
tion of the g-pyrone to the a-pyrone provided 1b. Com-
pound 19a was also converted to subglutinol A (1a) by fol-
lowing the procedures described in Scheme 9. Our synthetic
compounds 1a and 1b were identical in all respects with the
authentic natural products. Comparison of synthetic 1a and
1b ([a]2D5) with natural 1a and 1b enabled the assignment of
the absolute stereochemistries of 1a and 1b as 12S and 12R,
respectively.
Scheme 9. First-generation synthesis of subglutinols A (1a) and B (1b):
a) 20, nBuLi, THF, À788C, 5 min, then 19b, À788C, 30 min, 47%; b) CS2,
NaHMDS, THF, À788C, 1 h, then MeI, À788C, 30 min; c) nBu3SnH, cat.
AIBN, toluene, reflux, 2 h, 57% for two steps; d) 2-methylpropene,
Grubbs second-generation catalyst (30 mol%), CH2Cl2, 508C, 36 h, 96%;
e) 24, nBuLi, THF, À788C, 5 min, then 19, À788C, 30 min, 25a: 68%,
25b: 82%; f) CS2, NaHMDS, À788C, 1 h, then MeI, À788C, 30 min;
g) nBu3SnH, cat. AIBN, toluene, reflux, 2 h, 26a: 65% for two steps,
26b: 83% for two steps; h) 2-methylpropene, Grubbs second-generation
catalyst (20 mol%), CH2Cl2, 508C, 36 h, 27a: 79%, 27b: 83%; i) 2n
NaOH, THF/MeOH (2:1), reflux, 8 h, 1a: 80%, 1b: 96%. AIBN=azo-
A second-generation, more straightforward and efficient
synthesis of subglutinols A and B: Even though the coupling
of the decalin aldehydes 19a and 19b and the lithiated g-
pyrone 24 by following the procedures established by Katoh
and co-workers successfully completed the synthesis of 1a
and 1b, we thought that the synthetic route was not straight-
forward and efficient for the generation of a variety of ana-
logues for biological studies. To establish a more straightfor-
ward and efficient method for the installation of the a-
pyrone, we extensively investigated the scope of the regio-
and stereoselective intermolecular SN2’ reaction of a propio-
nate moiety. With respect to the stereoselectivity, we expect-
ed that the propionate group should be added from the a-
face opposite to the axially oriented C18 methyl group,
which is analogous to the intramolecular sigmatropic rear-
AHCTUNGTREGbNNUN isisobutyronitrile, NaHMDS=sodium hexamethyl disilazide.
rangement reactions reported by Danishefsky and
Katoh.[9–11] However, intermolecular SN2’ alkylation at the
sterically hindered neopentyl C4-position was expected to
be challenging.
After an extensive search of reaction conditions, we were
delighted to find that conversion of 3b to the phosphate
28b followed by CuI-mediated intermolecular SN2’ addition
of 29 to 28b in the presence of CuI·2LiCl provided 30b as a
single diastereomer in good regioselectivity (SN2’/SN2 5:1;
1906
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1902 – 1910

Total Synthesis of Subglutinols A and B
Table 2, entry 5).[24–26] The improvement in regioselectivity
(SN2’/SN2) was observed by raising reaction temperature or
increasing the Et2O/THF ratio.[27] When the reaction was
conducted solely in Et2O, the reaction mixture showed a low
solubility.
Table 2. CuI-mediated intermolecular SN2’ alkylation of phosphate 28b
with [2-(1,3-dioxolan-2-yl)ethyl]magnesium bromide.
Entry
Solvent
T [8C]
Yield [%][a]
Ratio of
30b/30b’[b]
1
2
3
4
5
6
THF
THF
THF
Et2O/THF 1:1
Et2O/THF 3:1
Et2O/THF 3:1
À30
0
25
25
25
50
89
100
82
1:3
1:1
2:1
3:1
5:1
1:1
Scheme 10. Second-generation synthesis of subglutinols A (1a) and B
(1b): a) ClP(O)(OEt)2, Et3N, DMAP, CH2Cl2, 0 to 258C, 30 min, 28a:
AHCTUNGTRENNUNG
92%, 28b: 97%; b) 29, CuI·2LiCl, Et2O/THF, 258C, 10 min, then phos-
phate 28, 258C, 30 min, 30a: 64%, 30b: 80%; c) Jones reagent, acetone,
258C, 30 min; d) MeI, K2CO3, DMF, 258C, 2 h, then NaOMe, MeOH,
258C, 1 h, 31a: 34% (44% BRSM) for two steps, 31b: 45% (57%
BRSM) for 2 steps; e) 2-methylpropene, Grubbs second-generation cata-
lyst (15 mol%), CH2Cl2, 508C, 72 h, 32a: 91%, 32b: 85%; f) LDA, THF,
À788C, 30 min, then 33, À788C, 30 min; g) Dess–Martin periodinane,
NaHCO3, CH2Cl2, 258C, 30 min, 34a: 53% for 2 steps, 34b: 84% for
2 steps; h) MeI, CaCO3, CH3CN/H2O 3:1, 258C, 24 h; i) DBU, benzene,
reflux, 1 h, 1a: 53% for 2 steps, 1b: 54% for 2 steps. DBU=1,8-
91
95
ND[c]
[a] Combined yield of isolated 30b and 30b’. [b] Determined by integra-
tion of the 1H NMR spectrum of the crude product. [c] Not determined.
Oxidation of 30a and 30b, methyl ester formation, and
olefin cross-metathesis with 2-methylpropene smoothly pro-
ceeded to provide 32a and 32b (Scheme 10). Aldol reaction
of 32a and 32b with 33[28] followed by Dess–Martin oxida-
tion set the stage for final cyclization to the a-pyrone.[9] De-
protection of 1,3-dithaine in 34a and 34b and subsequent
DBU-mediated cyclization to give the a-pyrone completed
the synthesis of 1a and 1b.[9,29]
diazabicyclo
LDA=lithium diisopropylamide.
[5.4.0]undec-7-ene,
DMAP=4-dimethylaminopyridine,
Dose-dependent osteogenic activity of subglutinol A (1a):
While clinically approved immunosuppressive drugs (e.g.,
cyclosporin A, FK506) are anabolic on bone at low concen-
trations as they increase osteoblast differentiation and bone
mass,[32] they elicit an opposite and catabolic response at
clinically relevant higher concentrations causing undesirable
side effects to bone structure (dose-dependent biphasic ef-
fects), including osteopenia, osteoporosis, and increased inci-
dence of bone fractures.[33] As a result, considerable effort
has been devoted to the identification of immunosuppres-
sive drugs that lack the undesirable biphasic effects and pro-
mote bone formation in a dose-dependent manner. Such
drugs with dose-dependent osteogenic activity might help
reduce bone-associated side effects and be clinically useful
for bone tissue transplantation.
Initial structure-activity relationship studies: To identify the
structural requirements of the natural products for their im-
munosuppressive activity, we evaluated subglutinol A (1a)
and related synthetic intermediates (Table 3).[30] MLR assay
to assess immunosuppressive activity of 1a and the closely
related synthetic intermediates (3a, 25a, 26a, and 27a) re-
vealed that 1a was indeed more potent than cyclosporin A.
In addition, 26a and 27a exhibited useful levels of immuno-
suppressive activity. On the other hand, 3a and 25a were
completely inactive in the MLR assay. The initial SAR
study underscores that the a- or g-pyrone moiety plays an
important role in immunosuppressive activity. The a- or g-
pyrone moiety might serve as an electrophile to form a co-
valent linkage to amino acid residues (e.g., Cys) in the
active site of molecular target(s).[31]
The effect of 1a on the bone morphogenetic protein-2
(BMP-2)-induced commitment of murine pluripotent mes-
enchymal precursor C2C12 cells into osteoblasts was deter-
mined by the expression level of alkaline phosphatase
(ALP), an early phase marker of osteoblast differentia-
Chem. Asian J. 2010, 5, 1902 – 1910
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1907

J. Hong et al.
FULL PAPERS
Table 3. Immunosuppressive activity of 1a, 3a, 25a, 26a, and 27a.
Binding study of subglutinol A (1a) with cyclophilin A
(CypA): To determine if 1a and cyclopsorin A have the
same molecular target, CypA, the SUPREX technique,
which has previously been used for the analysis of CypA–cy-
closporin A binding,[36] was used to determine the affinity of
1a for CypA (Figure 2). The SUPREX analysis of CypA in
Compound
IC50 [nm]
1a
25
3a
25a
26a
inactive
inactive
838
27a
109
cyclosporin A
89
tion.[34] Up to the highest concentration examined (20 mm),
1a increased the expression/activity of ALP in a dose-de-
pendent manner (Figure 1a). The dose-dependent induction
of ALP expression/activity by 1a shows the great clinical po-
tential of 1a as an immunosuppressive drug without undesir-
able side effects to bone structure.
Figure 2. SUPREX curves obtained for subglutinol A (1a) and DMSO.
*
*
=CypA+subglutinol A (1a), =CypA+DMSO.
the presence of 1a and in the control sample, which con-
tained no 1a and just 10% DMSO, showed no evidence of
binding (i.e., SUPREX curves were essentially identical).
1=2
SUPREX
The absence of a significant shift (>0.3m) in the C
value (i.e., the denaturant concentration at the transition
midpoint) in the presence of 1a suggests that the dissocia-
tion constant for the 1a binding interaction is >~20 mm. We
1=2
SUPREX
have previously shown that cyclosporin A shifts the C
value of CypA ~1.5m units to ~3m under experimental con-
ditions similar to those used here.[36] The results of this bind-
ing study implied that 1a and 1b likely elicit their immuno-
suppressive activities through a mechanism distinctive from
cyclosporin A, which suggests that the mode of action
should be explored in greater detail.
Conclusions
Figure 1. Effect of 1a on a) ALP protein level/activity in C2C12 cells,
b) mRNA expression level of AP-1 family transcription factors, and
c) protein expression level of AP-1 family transcription factors.
In summary, we completed the stereoselective synthesis of
subglutinols A (1a) and B (1b) from the readily available
(S)-(+)-5-methyl-Wieland–Miesher ketone. We applied an
unprecedented tandem CM/intramolecular SN2’ reaction to
the synthesis of 1b. BF3·OEt2-promoted deoxygenation of
the cyclic hemiketal followed by stereoselective reduction of
the oxocarbenium ion intermediate was explored for the
synthesis of 1a. In addition, we demonstrated that the CuI-
mediated intermolecular SN2’ reaction is a straightforward
and efficient method for the installation of the a-pyrone
moiety. Alternatively, the tautomerization of the g-pyrone
to the a-pyrone could be utilized for the installation of the
a-pyrone moiety of 1a and 1b. Comparison of synthetic 1a
and 1b ([a]2D5) with natural 1a and 1b enabled the assign-
ment of the absolute stereochemistry of the natural products
(1a and 1b) as 12S and 12R, respectively. The synthetic
Since the expression of Fra-2, an AP-1 family transcrip-
tion factor, has been shown to be critical in osteoblast differ-
entiation induced by cyclosporin A,[35] We examined the
effect of 1a on mRNA expression of Fra-2 and other AP-1
family transcription factors. As shown in Figure 1b, with the
exception of c-Fos, 1a dramatically induced the expression
of Fra-1, Fra-2 c-Jun, and Jun D at a transcript level. In ad-
dition, Western blot analysis showed that the protein expres-
sion level of Fra-1, Fra-2 c-Jun, and Jun D in the nucleus
was increased by 1a (Figure 1c). These data imply that AP-1
family transcription factors could be one of the key regula-
tors for the anabolic activity of 1a.
1908
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1902 – 1910

Total Synthesis of Subglutinols A and B
ture of SN2’-adduct 30a and SN2-adduct 30a’, which was purified by
column chromatography (silica gel, hexanes/EtOAc 10:1) to afford SN2’-
adduct 30a (67.8 mg, 64%) and SN2-adduct 30a’ (12.8 mg, 12%) as color-
less oils.
strategies would provide an access to more potent and less
cytotoxic analogues and tools for the study of their molecu-
lar mechanism of action. In addition, we presented prelimi-
nary SAR studies and initial biological data to show the sig-
nificant potential of 1a as an immunosuppressive drug with
dose-dependent osteogenic activity. We also reported a
binding study of 1a to CypA to provide insights into the
mechanism of 1a and 1b by which 1a and 1b elicit their im-
munosuppressive activities. Further studies to identify the
mode of action of 1a and 1b are currently in progress.
SN2’-adduct 30a: Rf =0.32 (hexanes/EtOAc 6:1); [a]2D4:9 =À44.5 (c=0.17
in CHCl3); 1H NMR (400 MHz, CDCl3): d=5.95 (ddd, J=17.2, 10.4,
6.8 Hz, 1H), 5.21 (ddd, J=17.2, 1.6, 1.6 Hz, 1H), 5.05 (ddd, J=10.4, 1.6,
1.6 Hz, 1H), 4.82 (dd, J=4.4, 4.4 Hz, 1H), 4.69 (dd, J=2.0, 2.0 Hz, 1H),
4.54 (dd, J=2.0, 2.0 Hz, 1H), 4.36–4.42 (m, 1H), 3.81–3.97 (m, 4H), 3.08
(dd, J=11.6, 3.6 Hz, 1H), 2.15 (ddd, J=13.6, 5.2, 1.6 Hz, 1H), 2.08 (ddd,
J=13.6, 13.6, 5.6 Hz, 1H), 1.36–1.85 (m, 13H), 1.21 (dd, J=10.0, 2.8 Hz,
1H), 0.93 (s, 3H), 0.79 ppm (d, J=0.8 Hz, 1H); 13C NMR (100 MHz,
CDCl3): d=147.7, 140.0, 115.0, 110.6, 104.7, 87.5, 78.3, 64.8, 58.1, 46.8,
45.3, 44.2, 38.4, 34.8, 32.5, 30.5, 25.0, 24.6, 22.3, 20.6, 16.8 ppm; HRMS
(EI): m/z: calcd for C22H34O3: 346.2508 [M]+; found: 346.2511.
Experimental Section
SN2-adduct 30a’: Rf =0.42 (hexanes/EtOAc 1:2); 1H NMR (400 MHz,
CDCl3): d=5.98 (ddd, J=17.2, 10.4, 6.4 Hz, 1H), 5.21 (ddd, J=17.2, 1.6,
1.6 Hz, 1H), 5.16 (s, 1H), 5.06 (ddd, J=10.4, 1.6, 1.6 Hz, 1H), 4.83 (dd,
J=4.8, 4.8 Hz, 1H), 4.39–4.44 (m, 1H), 3.81–3.97 (m, 4H), 3.12 (dd, J=
12.0, 3.6 Hz, 1H), 1.98 (dd, J=8.8, 5.2 Hz, 2H), 1. 89 (dd, J=7.2, 7.2 Hz,
2H), 1.81 (dddd, J=11.2, 3.6, 3.6, 3.6 Hz, 1H), 1.65–1.77 (m, 3H), 1.44–
1.63 (m, 7H), 1.26 (ddd, J=12.8, 12.8, 4.0 Hz, 1H), 1.24 (dd, J=13.2,
2.4 Hz, 1H), 0.92 (s, 3H), 0.82 ppm (d, J=1.2 Hz, 3H); 13C NMR
(100 MHz, CDCl3): d=140.1, 134.6, 133.1, 114.9, 104.5, 87.9, 78.5, 64.8,
50.3, 46.1, 44.2, 38.5, 36.9, 35.4, 33.4, 29.1, 23.5, 22.4, 22.2, 20.9, 16.2 ppm.
Compound 4b: Allyl chloride (1.23 mL, 15.1 mmol) and Grubbs second-
generation catalyst (213.7 mg, 0.252 mmol) were added to a solution of
olefin 7 (1.412 g, 5.04 mmol) in CH2Cl2 (50 mL, 0.10m) at 258C. The re-
sulting mixture was refluxed for 1.5 h. An addition of allyl chloride
(0.412 mL, 5.04 mmol) and Grubbs second-generation catalyst (213.7 mg,
0.252 mmol) was repeated three times every 1.5 h. The reaction mixture
was refluxed for a further 24 h, cooled to 258C, and concentrated in
vacuo. The residue was purified by column chromatography (silica gel;
hexanes/EtOAc, 10:1 to 6:1) to afford 2,3-trans-2,5-trans-tetrahydrofuran
4b as a white solid (781.3 mg, 53%, 76% based on recovered starting
material). Rf =0.42 (hexanes/EtOAc 7:1); [a]2D5:9 =À25.9 (c=1.01 in
Compound 30b: A 5:1 mixture of SN2’-adduct 30b and SN2-adduct 30b’
was purified by column chromatography (silica gel, hexanes/EtOAc 10:1)
to afford SN2’-adduct 30b (34.5 mg, 80%) and SN2-adduct 30b’ (6.7 mg,
15%) as colorless oils.
1
CHCl3); H NMR (400 MHz, CDCl3): d=5.89 (ddd, J=16.8, 10.0, 6.4 Hz,
1H), 5.18 (dd, J=16.8, 1.2 Hz, 1H), 5.08 (dd, J=10.0, 1.2 Hz, 1H), 4.49
(ddd, J=6.8, 6.8, 6.8 Hz, 1H), 3.89–3.96 (m, 3H), 3.80–3.86 (m, 1H),
3.21–3.26 (m, 1H), 1.88 (dd, J=11.6, 3.2 Hz, 1H), 1.80–1.85 (m, 1H),
1.44–1.69 (m, 9H), 1.24–1.33 (m, 2H), 1.08 (s, 3H), 0.87 ppm (s, 3H);
13C NMR (100 MHz, CDCl3): d=140.0, 113.9, 113.3, 85.9, 78.2, 65.4, 64.8,
48.5, 47.5, 45.2, 43.9, 30.5, 29.9, 23.2, 22.9, 22.1, 18.4, 15.5 ppm; IR (neat):
SN2’-adduct 30b: Rf =0.39 (hexanes/EtOAc 6:1); [a]2D8:0 =À39.4 (c=1.00
in CHCl3); 1H NMR (400 MHz, CDCl3): d=5.88 (ddd, J=16.2, 10.4,
6.8 Hz, 1H), 5.19 (ddd, J=17.2, 2.0, 2.0 Hz, 1H), 5.04 (ddd, J=10.4, 1.2,
1.2 Hz, 1H), 4.81 (dd, J=4.4, 4.4 Hz, 1H), 4.69 (dd, J=2.4, 2.4 Hz, 1H),
4.54 (dd, J=2.0, 2.0 Hz, 1H), 4.48 (ddd, J=8.4, 6.8, 6.8 Hz, 1H), 3.80–
3.97 (m, 4H), 3.16 (dd, J=11.6, 3.6 Hz, 1H), 2.15 (ddd, J=13.6, 4.8,
2.0 Hz, 1H), 2.02–2.11 (m, 1H), 1.90 (dd, J=11.2, 6.8 Hz, 1H), 1.78–1.83
(m, 1H), 1.50–1.75 (m, 6H), 1.37–1.48 (m, 4H), 1.18–1.27 (m, 2H), 0.93
(s, 3H), 0.83 ppm (d, J=0.8 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=
147.6, 139.8, 113.9, 110.6, 104.7, 86.1, 78.1, 64.8, 58.0, 47.7, 44.96, 44.85,
38.3, 34.7, 32.5, 30.5, 25.0, 24.5, 22.4, 20.6, 15.5 ppm; IR (neat): n˜ =2933,
1730 cmÀ1; HRMS (EI): m/z: calcd for C22H34O3: 346.2508 [M]+; found:
346.2513.
SN2-adduct 30b’: Rf =0.45 (hexanes/EtOAc 1:2); 1H NMR (400 MHz,
CDCl3): d=5.89 (ddd, J=17.2, 10.4, 6.8 Hz, 1H), 5.20 (ddd, J=17.2, 1.2,
1.2 Hz, 1H), 5.15 (s, 1H), 5.02 (ddd, J=10.4, 1.2, 1.2 Hz, 1H), 4.82 (dd,
J=4.8, 4.8 Hz, 1H), 4.52 (ddd, J=8.4, 6.8, 6.8 Hz, 1H), 3.80–4.00 (m,
4H), 3.20 (dd, J=12.0, 3.2 Hz, 1H), 1.86–1.99 (m, 5H), 1.19–1.29 (m,
3H), 0.92 (s, 3H), 0.85 ppm (d, J=1.2 Hz, 3H); 13C NMR (100 MHz,
CDCl3): d=139.8, 134.6, 133.1, 113.9, 104.5, 86.6, 78.2, 64.8, 49.9, 47.1,
45.0, 38.4, 36.9, 35.3, 33.3, 29.0, 23.5, 22.5, 22.2, 20.8, 15.1 ppm.
n˜ =2934, 2872, 1455, 1380, 1174, 1102, 1027, 952, 902 cmÀ1
; HRMS
(FAB): m/z: calcd for C18H29O3: 293.2117 [M+H]+; found: 293.2111.
Compound 14d: Et3SiH (0.47 mL, 2.95 mmol) was added to a stirred so-
lution of g-hydroxy TIPS-alkynone 5 (454.9 mg, 0.983 mL) in CH2Cl2
(40 mL, 0.025m) at À788C. The resulting mixture was stirred at the same
temperature for 20 min, and then BF3·Et2O (0.37 mL, 2.95 mmol) was
added. The reaction mixture was allowed to warm slowly to À208C for
2 h, and was then quenched with a saturated aqueous NaHCO3 solution,
and the resulting mixture was stirred vigorously at 258C for 20 min. The
layers were separated, and the aqueous layer was extracted with CH2Cl2
(1ꢃ20 mL). The combined organic layers were washed with brine, dried
over anhydrous Na2SO4, and concentrated in vacuo. The residue was pu-
rified by column chromatography (silica gel, hexanes/EtOAc 10:1) to
afford 2,3-trans-2,5-cis-tetrahydrofuran 14d (398.4 mg, 91%) as a color-
less oil. Rf =0.38 (hexanes/EtOAc 10:1); [a]2D5:4 =À16.0 (c=1.02 in
CHCl3); 1H NMR (400 MHz, CDCl3): d=4.63 (dd, J=10.0, 3.2 Hz, 1H),
3.89–3.96 (m, 3H), 3.80–3.86 (m, 1H), 3.06 (dd, J=12.0, 3.6 Hz, 1H),
1.70–1.90 (m, 3H), 1.46–1.73 (m, 10H), 1.09 (s, 3H), 1.07–1.12 (m, 21H),
1.00 ppm (s, 3H); 13C NMR (100 MHz, CDCl3): d=113.1, 108.6, 87.4,
85.6, 66.2, 65.2, 64.6, 49.0, 48.5, 44.4, 43.7, 30.4, 29.7, 22.97, 22.90, 21.9,
18.5, 18.2, 15.6, 11.1 ppm; IR (neat): n˜ =2940, 2863, 1462 cmÀ1; HRMS
(FAB): m/z: calcd for C27H45O3Si: 445.3138 [MÀH]+; found: 445.3136.
Compound 30a: Et2O (10 mL) was added to a yellow/green solution of
CuI·2LiCl (2.44 mL, 0.5m in THF, 1.22 mmol). The resulting mixture was
stirred at 258C for 10 min before 29 (2.44 mL, 0.5m in THF, 1.22 mmol)
was added dropwise. The resulting dark-green solution was stirred until
turned to dark black, then phosphate 28a (121.4 mg, 0.305 mmol) in
Et2O (5 mL) was added dropwise. The reaction mixture was stirred at the
same temperature for 30 min before being quenched with saturated aque-
ous NH4Cl and NH4OH and diluted with Et2O. The resulting cloudy solu-
tion was stirred vigorously until the aqueous layer turned to clean light
blue. The layers were separated and the aqueous layer was extracted
with Et2O. The combined organic layers were washed with brine, dried
over anhydrous Na2SO4, and concentrated in vacuo to afford a ~5:1 mix-
Acknowledgements
We are grateful to M. Dickens and Dr. D. Pham for the X-ray crystal
structure determination and to Professors D. Kim and S. Craig for helpful
discussions. This work was supported by Duke University and Duke
Chemistry Undergraduate Summer Research Program. J.B.B. was sup-
ported by the Pharmacological Sciences Training Grant (GM007105).
Y.P. gratefully acknowledges the North Carolina Section of the American
Chemical Society for the Howie James Scholarship. S.H.K. was supported
by the Korea Healthcare technology R&D Project, Ministry of Health,
Welfare & Family Affairs, Republic of Korea (A080216). We acknowl-
edge the NCBC (grant no. 2008-IDG-1010) for funding of NMR spectro-
scopic instrumentation.
Chem. Asian J. 2010, 5, 1902 – 1910
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
1909

J. Hong et al.
FULL PAPERS
[1] L. A. J. O’Neill, Nat. Rev. Drug Discovery 2006, 5, 549–563.
[2] J. Mann, Nat. Prod. Rep. 2001, 18, 417–430.
[3] J. C. Lee, E. Lobkovsky, N. B. Pliam, G. Strobel, J. Clardy, J. Org.
Chem. 1995, 60, 7076–7077.
[4] a) R. Uchida, R. Imasato, Y. Yamaguchi, R. Masuma, K. Shinomi,
H. Tomoda, S. Omura, J. Antibiot. 2005, 58, 397–404; b) B. Engel,
G. Erkel, T. Anke, O. Sterner, J. Antibiot. 1998, 51, 518–521.
[5] S. B. Singh, D. L. Zink, A. W. Dombrowski, G. Dezeny, G. F. Bills,
J. P. Felix, R. S. Slaughter, M. A. Goetz, Org. Lett. 2001, 3, 247–250.
[6] M. A. Goetz, D. L. Zink, G. Denzeny, A. Dombrowski, J. D. Polish-
ook, J. P. Felix, R. S. Slaughter, S. B. Singh, Tetrahedron Lett. 2001,
42, 1255–1257.
[7] S. Gupta, S. B. Krasnoff, J. A. A. Renwick, D. W. Roberts, J. R.
Steiner, J. Clardy, J. Org. Chem. 1993, 58, 1062–1067.
[8] H. Kikuchi, T. Hoshi, M. Kitayama, M. Sekiya, Y. Katou, K. Ueda,
Y. Kubohara, H. Sato, M. Shimazu, S. Kurata, Y. Oshima, Tetrahe-
dron 2009, 65, 469–477.
[9] F. Zhang, S. J. Danishefsky, Angew. Chem. 2002, 114, 1492–1495;
Angew. Chem. Int. Ed. 2002, 41, 1434–1437.
[10] a) K. Watanabe, K. Iwasaki, T. Abe, M. Inoue, K. Ohkubo, T.
Suzuki, T. Katoh, Org. Lett. 2005, 7, 3745–3748; b) T. Oguchi, K.
Watanabe, K. Ohkubo, H. Abe, T. Katoh, Chem. Eur. J. 2009, 15,
2826–2845.
[11] T. Abe, K. Iwasaki, M. Inoue, T. Suzuki, K. Watanabe, T. Katoh,
Tetrahedron Lett. 2006, 47, 3251–3255.
[12] T. Oguchi, K. Watanabe, H. Abe, T. Katoh, Heterocycles 2010, 80,
229–250.
[13] H. Kim, J. B. Baker, S.-U. Lee, Y. Park, K. L. Bolduc, H.-B. Park,
M. G. Dickens, D.-S. Lee, Y. Kim, S. H. Kim, J. Hong, J. Am. Chem.
Soc. 2009, 131, 3192–3194.
[14] a) M. Scholl, S. Ding, C. W. Lee, R. H. Grubbs, Org. Lett. 1999, 1,
953–956; b) A. K. Chatterjee, R. H. Grubbs, Org. Lett. 1999, 1,
1751–1753; c) A. K. Chatterjee, D. P. Sanders, R. H. Grubbs, Org.
Lett. 2002, 4, 1939–1942; d) A. K. Chatterjee, R. H. Grubbs, Angew.
Chem. 2002, 114, 3303–3306; Angew. Chem. Int. Ed. 2002, 41, 3171–
3174; e) A. K. Chatterjee, T. L. Choi, D. P. Sanders, R. H. Grubbs J.
Am. Chem. Soc. 2003, 125, 11360–11370; f) S. J. Connon, S. Ble-
chert, Angew. Chem. 2003, 115, 1944–1968; Angew. Chem. Int. Ed.
2003, 42, 1900–1923.
[22] a) H. Kim, C. M. Wooten, Y. Park, J. Hong, Org. Lett. 2007, 9,
3965–3968; b) H. Yoda, M. Mizutani, K. Takabe, Heterocycles 1998,
48, 679–686.
[23] Crystal structure data for 14a (CCDC 687309 contains the supple-
mentary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre at www.ccdc.cam.ac.uk/data_request/cif): C18H26O3; Mr =
290.39; colorless cubic crystal; ca. 0.38ꢃ0.35ꢃ0.26 mm3; T=296 K;
graphite-monochromated MoKa radiation (l=0.71073 ꢁ); ortho-
rhombic, space group P212121; a=8.1782(3), b=11.0055(4), c=
17.9809(7) ꢁ; V=1618.37(10) ꢁ3; 1calcd =1.192 MgmÀ3
; Z=4; m=
0.079 mmÀ1; 2qmax =57.688; 37374 reflections collected of which 2909
unique reflections were used (Rint =0.0314), R1=0.0470 (for 2415 re-
flections with I>2s(I)), wR2 =0.1355 for 191 parameters. Largest
diffraction peak/hole=0.218/À0.226 eꢁÀ3
.
[24] a) A. Yanagisawa, N. Nomura, H. Yamamoto, Tetrahedron 1994, 50,
6017–6028; b) A. Yanagisawa, Y. Noritake, N. Nomura, H. Yama-
moto, Synlett 1991, 251–253.
[25] CuI-mediated intermolecular SN2’ alkylation of phosphate 28b or
the corresponding allyl halide (Cl or Br) with Zn-homoenolate was
not successful due to low yield (<20%) and regioselectivity (<1:1).
For the Zn-homoenolate, see: E. Nakamura, S. Aoki, K. Sekiya, H.
Oahino, I. Kuwajima, J. Am. Chem. Soc. 1987, 109, 8056–8066.
[26] When CuCN·2LiCl was used instead of CuI·2LiCl, the SN2’ alkyla-
tion reaction did not provide 30b. With a catalytic or stoichiometric
amount of CuBr·SMe2, 30b’ was formed as the major product.
[27] For an extensive discussion about the effects of reaction tempera-
ture and solvent on CuI-mediated Grignard reactions, see: a) J. E.
Blꢄckvall, M. Sellen, J. Chem. Soc. Chem. Commun. 1987, 827–829;
b) J. E. Bꢄckvall, M. Sellen, B. Grant, J. Am. Chem. Soc. 1990, 112,
6615–6621.
[28] Compound 14 was prepared from the known alcohol by Parikh–
Doering oxidation (SO3·pyridine, DMSO, Et3N, CH2Cl2). For prepa-
ration of the known alcohol, see: A. Nickon, A. D. Rodriguez, V.
Shirhatti, R. Ganguly, J. Org. Chem. 1985, 50, 4218–4226.
[29] It is noteworthy that when we used 1,3-dioxolane instead of 1,3-di-
thiane as a carbonyl protecting group, the epimerization of C12 was
observed during acid- or Pd-mediated deprotection of the 1,3-dioxo-
lane group.
[15] A. Ardon-Jimenez, T. G. Halsall, J. Chem. Soc. Perkin Trans. 1 1978,
1461–1470.
[30] It should be noted that we observed the instability of subglutinol B
(1b) and disappearance of the C3–C19 exo-methylene group in the
1H NMR spectrum upon prolonged exposure to solvents, including
CD3OD, CHCl3, and DMSO.
[31] N. Kudo, B. Wolff, T. Sekimoto, E. P. Schreiner, Y. Yoneda, M. Ya-
nagida, S. Horinouchi, M. Yoshida, Exp. Cell Res. 1998, 242, 540–
547.
[32] a) P. Orcel, J. Bielakoff, D. Modrowski, Miravet, L. M. C. de Verne-
joul, J. Bone Miner. Res. 1989, 4, 387–391; b) L. Tang, S. Ebara, S.
Kawasaki, S. Wakabayashi, T. Nikaido, K. Takaoka, Cell Biol. Int.
2002, 26, 75–84; c) A. Aroldi, A. Tarantino, G. Montagnino, B.
Cesana, C. Cocucci, C. Ponticelli, Transplantation 1997, 63, 380–386.
[33] a) S. Epstein, E. Shane, J. P. Bilezikian, Curr. Opin. Rheumatol.
1995, 7, 255–261; b) C. Movsowitz, S. Epstein, M. Fallon, F. Ismail,
S. Thomas, Endocrinology 1988, 123, 2571–2577.
[34] T. Katagiri, A. Yamaguchi, M. Komaki, E. Abe, N. Takahashi, T.
Ikeda, V. Rosen, J. M. Wozney, A. Fujisawa-Sehara, T. Suda, J. Cell
Biol. 1994, 127, 1755–1766.
[35] a) H. Yeo, L. H. Beck, J. M. McDonald, M. Zayzafoon, Bone 2007,
40, 1502–1516; b) L. R. McCabe, C. Banerjee, R. Kundu, R. J. Har-
rison, P. R. Dobner, J. L. Stein, J. B. Lian, G. S. Stein, Endocrinology
1996, 137, 4398–4408.
[16] H. Hagiwara, H. Uda, J. Org. Chem. 1988, 53, 2308–2311.
[17] a) J. Lee, J. Hong, J. Org. Chem. 2004, 69, 6433–6440; b) D. Kim,
W. J. Choi, J. Y. Hong, I. Y. Park, Y. B. Kim, Tetrahedron Lett. 1996,
37, 1433–1434; For a review on the intramolecular SN2’ reaction,
see: c) L. A. Paquette, C. J. M. Stirling, Tetrahedron 1992, 48, 7383–
7423.
[18] Crystal structure data for 4b (CCDC 686833 contains the supple-
mentary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre at www.ccdc.cam.ac.uk/data_request/cif): C18H28O3; Mr =
292.40; colorless plate crystal; ca. 0.50ꢃ0.43ꢃ0.21 mm3; T=298 K;
graphite-monochromated MoKa radiation (l=0.71073 ꢁ); ortho-
rhombic, space group P212121; a=8.5896(3), b=10.1995(3), c=
37.4208(12) ꢁ; V=3278.42(18) ꢁ3; 1calcd =1.185 MgmÀ3; Z=8; m=
0.079 mmÀ1; 2qmax =50.828; 61472 reflections collected of which 4270
unique reflections were used (Rint =0.0355), R1=0.0406 (for 3549 re-
flections with I>2s(I)), wR2 =0.1066 for 402 parameters. Largest
diffraction peak/hole=0.314/À0.156 eꢁÀ3
.
[19] a) P. Li, T. Wang, T. Emge, K. J. Zhao, J. Am. Chem. Soc. 1998, 120,
7391–7392; b) P. Li, J. Yang, K J. Zhao, J. Org. Chem. 1999, 64,
2259–2263.
[36] M. Z. Wang, J. T. Shetty, B. A. Howard, M. J. Campa, E. F. Patz, Jr.,
M. C. Fitzgerald, Anal. Chem. 2004, 76, 4343–4348.
[20] For a recent example of a tandem reaction associated with CM, see:
J.-R. Chen, C.-F. Li, X.-L. An, J.-J. Zhang, X.-Y. Zhu, W.-J. Xiao,
Angew. Chem. 2008, 120, 2523–2526; Angew. Chem. Int. Ed. 2008,
47, 2489–2492 and references therein.
Received: February 23, 2010
[21] K. Lee, H. Kim, J. Hong, Org. Lett. 2009, 11, 5202–5205.
Published online: June 16, 2010
1910
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 1902 – 1910