New applications of PhI(OAc)2 in synthesis: Total synthesis and SAR development of potent antitumor natural product psymberin/irciniastatin A

Article abstract of DOI:10.1055/s-0029-1216926

A novel PhI(OAc)2-mediated oxidative cyclization reaction is discovered for the synthesis of a-oxy N-acyl aminals and hemiaminals in good yields from readily synthesized N-acyl enamines. This methodology represents a cascade process to construct the core structure of the pederin family of natural products. The total synthesis of psymberin, a member of the pederin family, is accomplished using this ring-closure reaction as the key step. This new method is further showcased in the preparation of advanced psymberin analogues. The biological data of these analogues are presented. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0029-1216926

PAPER
2855
New Applications of PhI(OAc)2 in Synthesis: Total Synthesis and SAR
Development of Potent Antitumor Natural Product Psymberin/Irciniastatin A
Total
S
ynthesis of
i
Psym
b
n
erin g Shao,* Xianhai Huang,* Anandan Palani, Robert Aslanian, Alexei Buevich, John Piwinski, Robert Huryk,
Cynthia Seidel-Dugan
Department of Chemical Research, Schering-Plough Research Institute, Kenilworth, NJ 07033, USA
Fax +1(908)7407664; E-mail: xianhai.huang@spcorp.com; E-mail: ning.shao@spcorp.com
Received 30 May 2009
synthesis of these natural products and has posed signifi-
Abstract: A novel PhI(OAc)2-mediated oxidative cyclization reac-
tion is discovered for the synthesis of a-oxy N-acyl aminals and
hemiaminals in good yields from readily synthesized N-acyl enam-
ines. This methodology represents a cascade process to construct
cant synthetic challenge as those syntheses usually require
several chemical transformations. Consequently, there are
several synthetic methods available to synthesize N-acyl
the core structure of the pederin family of natural products. The total aminals and hemiaminals. The generally reported meth-
ods are stepwise processes and involve the introduction of
synthesis of psymberin, a member of the pederin family, is accom-
plished using this ring-closure reaction as the key step. This new
method is further showcased in the preparation of advanced psym-
more synthetically challenging N-acyl aminal, and usually
berin analogues. The biological data of these analogues are present-
the a-oxy functionality prior to the introduction of the
require further modification of functional groups such as
ed.
nitrile, a-oxy carboxylic acid and aldehyde, N-acyl imi-
date, acyloxy acetal, or N-acyl amino acid.5
1
2
3
4
Introduction
Methodology Development
Total Synthesis of Psymberin
Synthesis of Psymberin Analogues: The Discovery of C11-
Deoxypsymberin
In light of these synthetic studies, we feel that a more con-
vergent method to prepare this functional unit is highly
desirable. Herein, we report a one-step oxidative prepara-
tion of a-alkyloxy or -acyloxy N-acyl aminals and hemi-
aminals from readily synthesized N-acyl enamines (1)
which enables the simultaneous introduction of the a-oxy
group and the N-acyl aminal and hemiaminal. We decided
to study the possibility of employing an electron-transfer
process to generate radical cation 2¢ from N-acyl enamine
1 with the hope that this process would differentiate the
electrophilicity of the two vinyl carbons (Scheme 1).
Thus, through this process both monocyclic (6 and 7) and
bicyclic a-oxy N-acyl aminal (8) functional units can be
accessed.
5
Conclusion
Key words: (diacetoxyiodo)benzene, psymberin, oxidative cycli-
zation, total synthesis, antitumor
1
Introduction
Nature has produced countless natural products with im-
portant biological activity, and many of these have be-
come the source of new therapeutic agents.1 The pederin
family is one class of natural products that has very potent
antitumor activity, and some members have advanced to
clinical trial.2 Due to their potent biological activity and
structural complexity, the total synthesis of these natural
products has attracted significant attention from the syn-
thetic chemistry community. An analysis of the structures
of the 36 members of the pederin class2 reveals that each
member possesses the same core structure, the cyclic
a-oxy N-acyl aminal unit, a functional group that can also
be found in non-pederin family compounds such as
zampanolide3 and lucilactaene.4 There are primarily two
forms of this unit in these natural products: monocyclic
and bicyclic a-oxy N-acyl aminal (Figure 1). The a-alkyl-
oxy or -acyloxy N-acyl aminals are seemingly unusual
substrates due to their presumed chemical instability.
However, they are the main functional groups responsible
for the activity of these natural products. The construction
of the a-oxy N-acyl aminal is usually the focal point of the
OMe
OH
OR
monocyclic core:
R'
bicyclic core:
OMe
OH
OH
O
MeO
OH
O
H
H
N
H
MeO
H
N
H
O
O
O
OMe
H
OMe
O
O
O
Mycalamide A R = H
Mycalamide B R = Me
(protein synthesis inhibitor)
Pedrin (ribosome binding agent)
Theopedrin E (cytotoxin) R' = H
monocyclic hemiaminal
O
O
OH
bicyclic hemiaminal
O
O
O
CO2Me
N
H
H
H
H
H
O
O
O
N
H
HO
Lucilactaene (cell cycle inhibitor)
Zampanolide (cytotoxin)
SYNTHESIS 2009, No. 17, pp 2855–2872
x
x.
x
x
.
2
0
0
9
Advanced online publication: 07.08.2009
DOI: 10.1055/s-0029-1216926; Art ID: C02009SS
© Georg Thieme Verlag Stuttgart · New York
Figure 1 a-Oxy N-acyl aminal and hemiaminal containing natural
products

2856
N. Shao et al.
PAPER
O
O
was known to react with electron-rich aromatic systems to
generate radical cations and has been used in numerous
oxidation processes.8 To our delight, the first reaction
with 2.2 equivalents of PhI(OAc)2 and 50 equivalents of
methanol in trifluoroethanol (TFE) at 0 °C gave the de-
sired product 6a in 60% yield as a 3.5:1 mixture of the
anti- and syn-diastereomers, along with three other prod-
ucts 6b–d which were isolated in 15, 8, and 10% yield, re-
spectively (Scheme 2).
O
1 R2OH
[O]
NH
NH
NH
R1
R1
R1
SET
OR2
2'
2''
1
O
R1
O
[O]
SET
2 R3OH
NH
R3O
NH
+
R1
OR2
OR2
3
2
R1 = alkyl, R2, R3 = H, alkyl, acyl
In order to improve the yield of the desired product 6a, we
screened a series of different reaction conditions
(Table 1). To minimize the formation of product 6c, de-
rived from the attack of solvent, various solvents were in-
vestigated. The solvent played a very import role in the
success of the reaction. Tetrahydrofuran, dioxane, N,N-
dimethylformamide, dimethylsulfoxide, and tert-butanol
were not good solvents for this reaction. Acetonitrile
(Table 1, entry 1) was an acceptable solvent and gave the
product in 45% yield. With dichloromethane as solvent,
6a and 6b were isolated in 35 and 30% yield, respectively
(Table 1, entry 2), and some additional unidentified more
polar by-products were observed. The amount of nucleo-
phile clearly affected the product distribution although the
desired 6a remained the major product (Table 1, entries
3–5). When less methanol was used (4.4 equiv), the for-
mation of open-chain product 6d was minimized, but the
amount of 6b and 6c increased to a total of 50%. When a
large excess of methanol was used (400 equiv), 6d was
isolated in 35% yield, while 6b and 6c were not formed.
Finally, the best conditions (Table 1, entries 6, 7) were
achieved using 20 equivalents of methanol with non-nu-
cleophilic hexafluoroisopropanol (HFI) as the solvent.
Under these conditions, the desired product 6a was ob-
tained in 82% yield with 6b as the only major by-product
which was readily separable by silica gel column chroma-
tography. To minimize the formation of 6b, more elec-
NHAc
OH
NHAc
OH
NHAc
O
O
OMe
intra- and intermolecular
tandem oxidative process
7
4
6
NHAc
OR
O
NHAc
O
OH
O
O
5
8
Scheme 1 Proposed oxidative synthesis of a-oxy N-acyl aminal and
hemiaminal
2
Methodology Development
In order to test this proposal, the simple substrate 4 was
prepared as shown in Scheme 2. Vinyl iodide 12 was pre-
pared from alcohol 11 in three steps and coupled with
acetamide using Buchwald’s chemistry6 to give a mixture
of Z/E-isomers (1:2) in good yield when the reaction was
carried out in toluene. When the reaction was performed
in DMF, the major product obtained was the bisacetamide
coupled compound. When CuTC7 was employed as the
coupling reagent, the reaction was not successful. Pure
E- and Z-enamides [(E)-4 and (Z)-4] were obtained after
deprotection of the TBS group and flash chromatography.
Iodide 12 could also be prepared from alcohol 9 in four
steps (Scheme 2). In this case, the major product was the
Z-isomer when diiodide 10 was reduced to iodide 12 with
Zn-Cu.
9
tron-deficient oxidant PhI(OTFA)2 (BTI) was used as the
oxidant. The reaction proceeded smoothly to give the de-
sired product 6a in good yield (Table 1, entries 8, 9) when
acetonitrile and HFI were used as the solvent. Although
the overall yield was slightly lower than when PhI(OAc)2
was used as the oxidant, the amount of product incorpo-
rating trifluoroacetate was indeed reduced to a negligible
amount. Noteworthy, the ratio of anti-6a to syn-6a
changed to 1:1 simply by using this more electron-defi-
cient oxidant. When sterically hindered PhI(OCOt-Bu)2
was used as the oxidant in HFI, the reaction went smooth-
ly to give the desired product in excellent yield, and the ra-
tio of anti-6a to syn-6a improved to 6.5:1 (Table 1, entry
10), however, still with 10% of the pivalate-attacked prod-
uct isolated. Finally, we found out that the reaction pro-
ceeded equally well with 1.2 equivalents of PhI(OAc)2
which reduced the amount of 6b to minimum (Table 1,
entry 7). Slow addition of PhI(OAc)2 (2.2 equiv) with a
syringe pump worked well (Table 1, entry 11), but could
not totally suppress the formation of 6b. We also looked
into oxidants other than hypervalent iodine reagents, such
as Cu(II)-2-ethylhexanoate and ferrocenium hexafluoro-
With 4 in hand, we first tried the reaction of (E)-4 in the
presence of (diacetoxyiodo)benzene [PhI(OAc)2] which
I
HO
OH
1. TBSCl, Et3N, 50%
2. SO3⋅py, 87%
TBSO
I
3. Ph3P, CHI3, KOt-Bu, 78%
10
9
Zn-Cu, 80%
1. AIBN, Bu3SnH, THF, 85%
2. I2, CH2Cl2, 99%
3. TBSOTf, lutidine, 95%
I
HO
TBSO
11
12
NHAc
+
1. AcNH2, CuI, Cs2CO3, PhMe,
dimethyl ethylenediamine
2. TBAF, 85% for two steps
OH
OH
NHAc
(E)-4:(Z)-4 = 2:1
Scheme 2 Preparation of precursor 4
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

PAPER
Total Synthesis of Psymberin
2857
Table 1 Optimization of Reaction Conditions for the Oxidative Formation of N-Acyl Aminal
NHAc
OMe
NHAc
O
NHAc
OAc
NHAc
PhI(OAc)2 (2.2 equiv),
MeOH (50 equiv),
NHAc
7
HO MeO
O
O
OH
O1
6
OMe
2
3
+
+
+
5
4
TFE, 0 °C
CF3
6d 10%
6c 8%
6b 15%
6a 60%
(E)-4
Entry
Oxidanta
MeOH [equiv]
Solvent
MeCN
CH2Cl2
TFE
Yield of 6a [%] (dr)c
45 (2.5:1)
35 (1:1)
Yield 6b/6c/6dd [%]
–
1
2
PhI(OAc)2
PhI(OAc)2
PhI(OAc)2
PhI(OAc)2
PhI(OAc)2
PhI(OAc)2
PhI(OAc)2
50
20
30/0/<5
25/25/0
0/0/35
3
4.4
400
20
45 (4:1)
4
TFE
52 (4:1)
5
TFE
68 (3.5:1)
82 (4:1)
15/10/<5
10/0/<5
5/0/<5
6
20
HFI
b
7
20
HFI
77 (4:1)
8
PhI(OTFA)2
PhI(OTFA)2
PhI(OCOt-Bu)2
PhI(OAc)2
20
HFI
66 (1.1)
<5e/0/<5
<5e/0/<5
10f/0/<5
10/0/<5
9
20
MeCN
HFI
64 (1.2:1)
85 (6.5:1)
80 (4:1)
10
11
20
20
HFIg
a Oxidant used: 2.2 equiv.
b Oxidant used: 1.2 equiv.
c Isolated yield, dr = diastereomeric ratio of anti/syn-isomers which is determined by 2D NMR spectroscopy.
d Based on the NMR spectrum of crude product and isolated yield, relative stereochemistry is not determined.
e Where OAc = OCOCF3.
f Where OAc = OCOt-Bu.
g PhI(OAc)2 in HFI was slowly added with a syringe pump.
phosphonate,10 but those reagents were not effective in this reaction to give the desired products in very good
this reaction. When NIS or I2 was used as the oxidant, only yields (Table 2, entries 2, 3, 6 and 9). When there was no
a complex mixture was produced. For these cyclization added nucleophile, the acetate group from PhI(OAc)2 act-
reactions, solvent seemed to play an important role in ed as nucleophile and gave the desired O-acyl N-acyl am-
terms of the diastereoselectivity; the ratio of the anti- and inal 6b in excellent yield (Table 2, entry 4) which
syn-isomers was 1:1 in dichloromethane, in contrast to provides an efficient way to generate this type of sub-
2.5:1 and 4:1, respectively, when acetonitrile and HFI strate. Moreover, water could act as a nucleophile as well
were used as the solvent (Table 1, entries 1, 2, and 6).
to give N-acyl heminal 18 in good yield as illustrated in
entry 7 (Table 2). Surprisingly, this type of hemiaminal
was stable and could be readily purified with silica gel
flash-column chromatography. Considering the abun-
dance of N-acyl hemiaminal containing natural products,
this method will be very useful for the synthesis of this
type of structural unit. When the starting materials were
Boc- or Cbz-derived enamines, the reaction proceeded
smoothly to give the cyclized product (entries 10, 11) al-
though the more electron-rich Boc-aminal product was
less stable upon purification. The reaction also worked for
the formation of 5-membered rings although the amount
of acetate-derived product increased (entry 12). When
bulky nucleophiles were used (entries 13, 14), the acetate
was isolated as the only product. The diastereoselectivity
of the five-member ring formation was surprisingly good
with only one diastereomer being isolated.
With the reaction conditions optimized, we next tested the
scope of this reaction (Table 2). Different substrates were
prepared using the same procedure as for 4. Surprisingly,
the geometry of the enamine did not affect the outcome of
the reaction at all. When (Z)-4 was used as the starting ma-
terial (Table 2, entry 1), the reaction proceeded smoothly
to give 6a in excellent yield with a similar diastereoselec-
tivity (anti/syn = 3:1) (Table 2, entry 1). Based on this re-
sult, we used other substrates without separation of the N-
acyl enamine isomers. As shown in Table 2, different acyl
enamines such as benzyl and cyclopropyl acyl enamines
15 and 19 reacted efficiently to give the desired tetrahy-
dropyran products in excellent yields. The phenyl ring and
the cyclopropyl groups were not affected under the reac-
tion conditions. Nucleophiles including long-chain prima-
ry alcohols and secondary alcohols worked equally well in
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

2858
N. Shao et al.
PAPER
Table 2 Oxidative Formation of 2-(N-Acyl Aminal) Tetrahydro-
pyrans
Table 2 Oxidative Formation of 2-(N-Acyl Aminal) Tetrahydro-
pyrans (continued)
Entry
1
Substratea
Product (yield [%], dr)b
Entry
Substratea
Product (yield [%], dr)b
NHAc
O
HO
O
NHAc
O
2
Ph
7
OMe
Ph
HN
HN
O
12
HO
R
(Z)-4 (MeOH, 20 equiv)
6a (82, 3:1)
NHAc
NHAc
27 (87, >10:1)
R = OMe/R = OAc = 3:2
26 (MeOH, 20 equiv)
HO
O
O
2
3
4
O
O
Ph
Ph
HN
HN
4 (n-BuOH, 20 equiv)
13 (82, 4:1)
13
14
HO
O
NHAc
NHAc
OAc
HO
O
O
26 (n-BuOH, 20 equiv)
28 (82, >10:1)
O
O
14 (77, 4:1)
4 (IPA, 20 equiv)
Ph
Ph
HN
HN
NHAc
NHAc
O
HO
HO
O
OAc
OAc
28 (80, >10:1)
26 (IPA, 20 equiv)
6b (92, 4:1)
4 (no added nucleophile)
a PhI(OAc)2 (2.2 equiv) was used as the oxidant in all cases.
b Isolated yield, relative stereochemistry of the major isomer is
shown, dr = diastereomeric ratio of 2,7-anti/2,7-syn-isomers, the ratio
and configuration is determined by NMR spectroscopy.
c Estimated yield by NMR spectroscopy, isolation was difficult due to
instability.
O
O
Ph
Ph
Ph
Ph
HN
O
HN
5
6
7
8
9
HO
OMe
15 (MeOH, 20 equiv)
O
16 (86, 3:1)
O
To further demonstrate the usefulness of this new reac-
tion, secondary and tertiary alcohols were also prepared
and tested as substrates for these cyclizations (Table 3,
compounds 29, 31, 33, 35). As listed in Table 3, both sec-
ondary and tertiary alcohols worked very well to give the
desired products in good yields. For secondary alcohols,
the stereochemical outcome was consistent with primary
alcohols. When PhI(OAc)2 was used as the oxidant, the
2,7-anti/2,7-syn ratio was about 4:1, and the ratio became
1:1 when BTI was used as the oxidant. In both cases, the
stereoselectivity at C2 and C6 was the same with a 2,6-cis/
2,6-trans ratio of 1:1. When a tertiary N-acyl enamine
(37) was used as the precursor (Table 3, entry 7), the reac-
tion was not effective and resulted in a complex mixture.
This result suggests that the secondary enamine is essen-
tial for the success for the reaction.
Ph
HN
HN
HO
O
O
15 (IPA, 20 equiv)
17 (75, 3:1)
O
O
Ph
HN
HN
HO
O
OH
15 (H2O, 20 equiv)
18 (65, 3.5:1)
O
O
HN
HN
O
HO
OMe
19 (MeOH, 20 equiv)
20 (85, 3.5:1)
O
O
The diastereoselectivity of the reaction was determined by
NMR analysis of compound 30 (Table 3). For illustration
purposes, the stereochemistry of the C2 carbon was drawn
in the S-configuration (Figure 2). At low temperature
(–50 °C) in all four isomers the J coupling between H7
and H2 protons was in the range of 1.5–3.5 Hz and the
coupling between H2 and H3 axial protons in the range of
9–11 Hz, which showed that the H2 and H7 protons were
preferably in gauche conformation and H2 was preferably
in axial orientation relative to the six-membered ring.
Taking into account two relatively strong NOEs between
the H7 and H3 equatorial protons and between the NH and
H3 axial protons in isomers I and III, the relative stereo-
chemistry of C7 was assigned as S (Figure 2). In isomers
II and IV, there was no NOE between the NH and H3 ax-
ial protons, whereas the NOE between H7 and H3 equato-
HN
HN
HO
O
O
19 (IPA, 20 equiv)
21 (75, 4:1)
NHBoc
NHBoc
HO
O
OMe
10
11
23 (60)c
22 (MeOH, 20 equiv)
NHCbz
NHCbz
HO
O
OMe
24 (MeOH, 20 equiv)
25 (83, 4:1)
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

PAPER
Total Synthesis of Psymberin
2859
H
Table 3 Oxidative Formation of 2-(N-Acyl Aminal) 2,6-Disubsti-
tuted Tetrahydropyrans
Ac
3a H6
H3a
Me
OMe
H7
H
N
O
O
AcHN
MeO
3e H6
Substratea
Product (yield [%], dr)b
H3e Me
H2
H
Entry
1
H2
H7
II
I
NHAc
NHAc
OH
2
O
6
7
H
OMe
Ac
3a H6
H3a Me
N
H
OMe
H7
O
O
MeO
AcHN
29
30 (81, 4.6:1:1:4.6)
3e H6
H3e Me
H2
H
H7
O
O
H2
III
IV
HN
HN
2
3
4
5
OH
OH
OH
O
Figure 2 Stereochemistry determination for cyclic N-acyl aminals
(arrows indicate NOEs used in stereochemical analysis)
OMe
31
33
32 (89, 4:1:1:4)
O
O
3
Total Synthesis of Psymberin
Ph
Ph
Ph
Ph
Ph
HN
HN
O
OMe
With the methodology developed, we were ready to uti-
lize it in the total synthesis of natural products. In 2004, a
new member of the pederin family, psymberin /irciniasta-
tin A (38) was identified by the laboratories of Crews and
Pettit, respectively.11 The absolute stereochemistry at C4
was assigned via the total synthesis by De Brabander’s
group in 2005, irciniastatin A and psymberin turned out to
be the same compound.12 The exquisite molecular struc-
ture of psymberin combined with its potent biological
activity11 have attracted much attention from the synthetic
organic chemistry community. After the first total synthe-
sis, a formal synthesis by Williams’ group involving a
spirodiepoxide opening reaction was reported.13 More re-
cently, the Smith group published a stereocontrolled total
synthesis using a late-stage Curtius rearrangement.14 In
our retrosynthetic analysis, a late-stage PhI(OAc)2-medi-
ated tandem oxidative cylization–addition reaction was
designed to construct the core tetrahydropyran ring with
the challenging acylaminal moiety installed simulta-
neously (39 → 38). Thus, the natural product could be po-
tentially synthesized from three equally complex pieces
through a CuI-mediated coupling reaction to form the N7-
C8 bond and a substrate-controlled Mukaiyama aldol re-
action to connect C14 to C15 (Scheme 3).
34 (82, 3:1:1:3)
O
O
Ph
HN
HN
O
OMe
34 (40, 1:1:1:1)
33 (MeCN, BTI)
O
O
Ph
HN
OH
HN
O
OMe
33 (HFI, BTI)
34 (62, 1:1:1.2:0.8)
O
O
Ph
HN
HN
OH
6
7
O
OMe
36 (85, 4.3:1c)
35
N(Me)Ac
OH
NAc
O
OMe
37
complex reaction
The dihydroisocoumarine unit 4315 and psymberamide
side chain 4016 were synthesized as previously reported.
For the synthesis of the central piece (Scheme 4), we first
tried an enantioselective Masamune aldol reaction17 be-
tween alkyne aldehyde 45 and enol ether 44, but only the
allene product was obtained as judged by NMR studies.
The allene might be derived from isomerization of the
alkyne during the reaction. On the other hand, commer-
cially available 3-benzyloxy-propionaldehyde underwent
a highly enantioselective Masamune aldol addition with
44 to give the secondary alcohol as a single enantiomer
(er > 50:1) with the desired R-configuration as confirmed
by the synthesis of Mosher’s ester 53. The secondary al-
cohol was subsequently protected with a TBS group to
give 47. Treatment of 47 with TMSCH2Li in pentane18
gave ketone 48 in one step which was converted into enol
ether 49 by treatment with TMSOTf/Et3N. Direct Masam-
une aldol reaction between 3-benzyloxy-propionaldehyde
a PhI(OAc)2 (2.2. equiv) was used as oxidant in HFI in all cases unless
otherwise indicated.
b Isolated yield, relative stereochemistry of the 2,6-trans-2,7-anti-iso-
mer is shown, dr = diastereomeric ratio of 2,6-trans-2,7-anti-/2,6-
trans-2,7-syn-/2,6-cis-2,7-syn-/2,6-cis-2,7-anti-isomers. The ratio
and configuration is determined by NMR spectroscopy.
c Diastereomeric ratio of 2,7-anti-/2,7-syn-isomers which is deter-
mined by NMR, stereochemistry of the major isomer is shown.
rial protons was still strong, thus suggesting R-
configuration at C7. Relative stereochemistry of the C7
and C6 carbons was then established based on the 1,3-di-
axial NOEs between H2 and methyl protons in I and IV
and between the H2 and H6 protons in isomers II and III.
These NOEs were consistent with R-configuration of the
C6 carbon in isomers I and IV, and S-stereochemistry in
II and III (Figure 2).
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

2860
N. Shao et al.
PAPER
enamide formation
OMe
O
OMe
OMe
O
11
5
4
OH
1
9
8
8
oxidative
cyclization
N
H
7
N
H
OR
OH
O
OR
HO
13
15
17
HO
16
14
15
RO
21
OH
25
O
OR
23
O
Psymberin (38)
OH
O
OR
O
39
OTBS
I
8
OR
O
I
42
HO
OMe
O
14
15
RO
H
TIPSO
NH2
OR
O
OTPS
O
O
40
OR
O
aldol reaction
O
TIPSO
41
43
Scheme 3 Retrosynthetic analysis of psymberin
BH3⋅THF, ligand 52,
EtCN, 82%
the lactone more electron-deficient and prone to ring
opening.
OH
TMSO
EtO
O
•
H
H
+
O
TMS
Chelation-controlled reduction of ketone 57 provided a
secondary alcohol 58 at C13 with excellent diastereose-
lectivity (dr = 15:1) which was transformed to 59 (E/Z =
5:1) via bis-acetylation at C13 and C15, hydrogenation,
oxidation, and Takai vinyl iodide formation.20 The stereo-
chemistry of 58 was confirmed by acetonide formation to
give 61 and NMR analysis. Amide 40 and vinyl iodide 59
were coupled using copper(I) iodide under Buchwald’s
conditions to give protected N-acyl enamine 60, and the
major E-isomer was separated at this stage. Upon treat-
ment of (E)-60 with sodium methoxide in methanol, the
two acetyl groups on C13 and C15 were cleaved and both
left unprotected. We assumed that the C15 hydroxy group
might not compete with the C13 hydroxy group in the key
ring-closure reaction due to formation of an unfavorable
ring size. When the crude product underwent the oxida-
tive cyclization reaction, a very complex mixture was ob-
tained and no desired product could be isolated. To
determine the cause of this poor result, we decided to car-
ry out some advanced model studies. In the first model
(64, Scheme 6), the psymberamide side chain was truncat-
ed to a simple acetamide and the isocoumarin side chain
was not installed. Ketone 48 was reduced to an alcohol
and protected with an acetyl group, followed by conver-
sion into vinyl iodide 63. Following Buchwald’s protocol,
the vinyl iodide was coupled with acetamide to give the
enamide which was deprotected to give precursor 64.
Compound 64 was treated with PhI(OAc)2 at 0 °C for 1.5
hours, and an 80% yield of the desired cyclization product
OEt
46
45
44
1. BH3⋅THF, 52, BnO(CH2)2CHO, 95%
2. 2,6-lutidine, TBSOTf, CH2Cl2, 99%
er > 50:1
BnO
TMSCH2Li, 95%
OTBS
TMSOTf,
100%
BnO
O
OTBS
BnO
TMSO
OTBS
O
OEt
47
49
48
MeO2C
BocHN
HO2C
p-TolO2SHN
HO2C
H2N
1. TFA
2. TsCl
3. 2 N KOH
1. Boc2O, 60 °C, 4 d
2. MeI, K2CO3, 4 d
then MeI, Me4NOH
60 °C, 4 d
60% overall
MeO
MeO
HO
OMe
OMe
OH
50
51
Ligand 52
MeO CF3
O
TMSO
BnO
O
BnO
O
O
3-benzyloxy
propionaldehyde
54
OEt
53
Scheme 4 Synthesis of the central piece
and enol ether 54 followed with TBS protection did pro-
vide product 48, but no enantioselectivity was observed in
the presence of ligand 52.
To continue the synthesis, enol ether 49 was coupled with
aldehyde 43 in the presence of BF3·OEt219 to give 57 in a
satisfying 91% yield and 5:1 ratio favoring the desired
isomer (Scheme 5). By contrast, the aldol reaction of 49
with substrate 55, bearing two phenol acetate groups in-
stead of TIPS, gave four isomers under identical reactions
conditions. This might be caused by b-elimination to open
the lactone ring in 55 followed by reclosure of the ring
since the electron-withdrawing bisacetate groups made
65 was isolated with
1.5:2.2:1.0:2.1.
a diastereomeric ratio of
In the second model (66, Scheme 7), the psymberamide
side chain was truncated to a simple acetamide while
maintaining the rest of the molecule the same as the natu-
ral product. After coupling of vinyl iodide 59 with acet-
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

PAPER
Total Synthesis of Psymberin
2861
BnO
OTBS
BF3⋅OEt2,
–78 °C
BnO
TMSO
OTBS
RO
O
+
O
O
50%
AcO
RO
O
OH
O
49
55 R = Ac
43 R = TIPS
56
mixture of isomers
AcO
O
43, BF3⋅OEt2,
–78 °C, 91%
BnO
HO
OTBS
BnO
OTBS
13
1. Ac2O, py, 78%
O
2. H2, Pd/C, 100%
3. Dess–Martin
periodinane, 95%
TIPSO
TIPSO
catecholborane, 92%
OH
O
4. CrCl2, CHI3, 90%
OH
O
TIPSO
O
TIPSO
O
58
57
11
OTBS
I
OMe
O
13
AcO
OTBS
N
H
40, CuI,
95%
TIPSO
15
OTPS
AcO
1. NaOMe
2. PhI(OAc)2
21
X
OAc
O
TIPSO
TIPSO
O
OAc
O
59, E/Z = 5:1
TIPSO
O
60, E/Z = 5:1
BnO
OTBS
BnO
OH
HO
O
O
1. dimethoxy acetonide
2. TBAF, 57% two steps
TIPSO
HO
OH
O
O
TIPSO
O
61
58
HO
O
Scheme 5 First attempt at the oxidative cyclization reaction
amide followed by deprotection of the diacetate, the PhI(OAc)2-mediated cyclization reaction at 0 °C, the re-
product was purified by silica gel column chromatogra- action was slow and there was no progress at this temper-
phy. After careful NMR analysis, we noticed that the aro- ature. After warming to room temperature overnight and
matic proton at 6.4 ppm disappeared and shifted to 5.6 stirring for an additional eight hours all the starting mate-
ppm which suggested that one of the TIPS groups on the rial was consumed. Two major products (67 and epi-67)
phenol was cleaved by sodium methoxide in methanol. were isolated after protecting groups’ removal with TBAF
This might be the reason why the first attempt failed, since and proved to be the desired acylaminal products, one of
the electron-rich free phenol could be oxidized in the pres- which matched the majority of the chemical shifts in the
ence of PhI(OAc)2. To circumvent this problem, an elec- proton NMR spectrum of psymberin.
tron-withdrawing acetyl group was selectively introduced
on the phenol (66). When compound 66 was subjected to
attempted the cyclization in the real system. Precursor 68
Encouraged by the success of these model studies, we re-
BnO
OTBS
BnO
AcO
OTBS
OTBS
1. H2/Pd/C
2. Dess–Martin [O]
1. NaBH4
2. Ac2O, py
I
O
AcO
3. CrCl2, CHI3, THF,
82% 3 steps
85% 2 steps
48
63
62
1. acetamide, CuI
2. NaOMe, 85%
O
OMe
O
OTBS
OTBS
PhI(OAc)2, HFIP, 80%
N
H
N
H
O
HO
65
64
4 isomers, ratio = 1.5:2.2:1.0:2.1
Scheme 6 First model study of the cyclization reaction
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

2862
N. Shao et al.
PAPER
O
OTBS
OTBS
I
N
H
AcO
HO
1. CuI, 95%
2. NaOMe
TIPSO
AcO
3. Ac2O, py,
90% 2 steps
OAc
OH
O
O
66
TIPSO
O
TIPSO
O
O
59
1. PhI(OAc)2
2. TBAF, 70%
O
OMe
O
OMe
OH
OH
N
H
N
H
O
HO
+
HO
OH
epi-67
O
OH
O
OH
O
OH
O
67
Scheme 7 Second model study of the cyclization reaction
was obtained from compound 60 in two steps following ucts disappeared. The structures of 69 and 70 were deter-
the same sequence as the second model study (Scheme 8). mined by NMR analysis. Both compounds resulted from
When compound 68 was subjected to the cylization con- the PhI(OAc)2-mediated amide cyclization to the terminal
ditions, the reaction did indeed proceed. However, after double bond (Scheme 8, 71 and 72) which was further
the products were carefully isolated, two major products confirmed by a model study.16
were identified as 69 and 70 in 40 and 20% yield, respec-
tively. In the proton NMR spectrum, the methoxy groups
at C4 and the terminal double bond in both isolated prod-
At this point, it became obvious that the terminal double
bond needed to be properly masked before the cylization.
OMe
O
OMe
O
OTBS
OTBS
N
H
N
H
OTPS
HO
OTPS
AcO
1. NaOMe, MeOH
AcO
2. Ac2O, py,
90% 2 steps
TIPSO
OH
O
OAc
O
TIPSO
O
O
TIPSO
68, E/Z = 5:1
60, E/Z = 5:1
PhI(OAc)2, MeOH,
HFI
OTBDPS
N
O
OTBDPS
O
MeO
AcO
O
N
OTBS
OTBS
HO
HO
+
AcO
O
OH
O
OH
TIPSO
O
TIPSO
O
70 20%
69 40%
68
69
70
major
~ 40%
minor
~ 20%
– H+
– Me+
Ph
I
AcO
N
O
H
H
N
O
O
N
O
TBDPSO
Me
O
O
TBDPSO
Me
TBDPSO
Me
Me
Me
Me
71
72
72'
ROH
+ PhI + AcO–
possible reaction pathway
Scheme 8 A novel PhI(OAc)2-mediated cyclization reaction
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

PAPER
Total Synthesis of Psymberin
2863
1. BH3, THF, 88%
2. BnBr, NaH, 92%
3. TBAF, 99%
separated at this step. The enantiomers at C1 were not re-
solved since there would be conversion to a double bond
at a later stage.
1. 2-propenylMgBr,
CuI, 82%
OMe
O
OTBS
OTBS
2. Me3OBF4, proton
sponge, 95%
4. ClCOCOCl, DMSO,
Et3N, 94%
74
73
To improve the diastereoselectivity at C5, we attempted a
chelation-controlled cyanohydrin formation reaction fol-
lowed by a Mitsunobu reaction to invert the stereochem-
istry (Scheme 10). The addition was indeed selective to
give a 10:1 ratio in favor of the opposite C5 stereoisomer
(77) using MgBr2·OEt2 as the chelating agent. However,
the Mitsunobu reaction did not invert the stereochemistry,
presumably due to a neighboring group participation pro-
cess (79).
OMe
O
1. TMSCN, AlCl3, 87%
2. TPSCl, Et3N
OMe
BnO
5
1
NH2
O
BnO
3. MeCONH2, PdCl2, 53%
(pure isomer) over two steps
OTPS
75
76
Scheme 9 Synthesis of benzyloxy-masked side chain
We chose to use a benzyloxy group as the masking group
which could be converted into a terminal double bond by
Grieco’s protocol.21 We proceeded with the synthesis of
amide 76 (Scheme 9). Regioselective epoxide opening of
73 with isopropenylmagnesium bromide gave a second-
ary alcohol which was protected as a methyl ether with
Me3OBF4 to give 74. Ether 74 was converted into 75 in
four steps via hydroboration, benzylation, deprotection of
the TBS group, and Swern oxidation. Aldehyde 75 under-
went cyanohydrin formation (dr = 2:1), and the free alco-
hol was protected as a TPS ether. The nitrile group was
hydrolyzed22 in the presence of palladium(II) chloride and
acetamide to give amide 76 with the isomers being easily
With amide 76 in hand, it was coupled with vinyl iodide
59 followed by protecting-group manipulations to give
compound 80, which was subjected to PhI(OAc)2 oxida-
tion (Scheme 11). The reaction went slowly and the mix-
ture was stirred at room temperature for 60 hours until
TLC showed the disappearance of the starting material.
Since a chiral center was created with the terminal benzyl-
oxy group, a mixture of eight diastereomers was produced
in this step; however, isolation of each compound was not
necessary. The desired products were isolated in a total
yield of 72% consisting two pairs of major diastereomers
(30% each, C8,C9 = S,S and C8,C9 = R,R) and two pairs
of minor isomers (6% each). The mixture was treated with
Ac2O to protect the C15 hydroxy group and hydrogenated
to remove the benzyl group to give alcohol 81 which was
transformed to create a terminal double bond using
Grieco’s protocol.21 The acylaminal survived these condi-
tions and no isomerization occurred, which was con-
firmed by carrying a single diastereomer through the
process. The protected alkenes were then treated with
TBAF at 50 °C. Again, the acylaminal was very stable and
survived the elevated temperature under basic conditions.
Two major products (1:1 ratio) were isolated by extensive
preparative TLC separation to give the final products
BnO
BnO
BnO
OMe
AcOH
ADDP
80%
OMe
TMSCN
MgBr2⋅OEt2
OMe
CN
O
CN
95%
OH
75
78
OAc
77
10:1
+
BnO
BnO
OMe
Me
O
CN
CN
OAc
minor
desired
78'
AcOH
79
Scheme 10 Attempted diastereoselective cyanohydrin formation
OTBS
11
OMe
O
I
BnO
OTBS
AcO
13
N
H
OTBDPS
HO
TIPSO
15
1. 76, CuI, 95%
2. NaOMe, MeOH
21
OAc
AcO
O
3. Ac2O, py, 95%
over 2 steps
OH
80, E/Z = 5:1
O
TIPSO
O
59, E/Z = 5:1
TIPSO
O
1. PhI(OAc)2, 72%
2. Ac2O, py, DMAP
3. H2, Pd/C, 95%
over two steps
OMe
O
OMe
OMe
O
OMe
O
HO
OTBS
OH
9
8
N
N
H
H
OTBDPS
O
OH
1. o-O2NC6H4SeCN
2. H2O2, THF
AcO
15
HO
3. TBAF, 67%
over three steps
OAc
OH
O
O
TIPSO
O
OH
O
81 C8, C9 = S,S
epi-81 C8, C9 = R,R
Psymberin 38
epi-38 C8, C9 = R,R
Scheme 11 The completion of the total synthesis of psymberin
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

2864
N. Shao et al.
PAPER
psymberin 38 and its epimer (epi-38). The spectral data ble bond plays an important role in psymberin’s activity.
(1H, 13C, optical rotation, MS) of compound 38 matched We then used a phenyl group to mimic the electronic ef-
exactly with those reported of natural psymberin.
fect of the double bond with no substituents at C4 and C5;
however, compound 86 and its epimer 88 showed dramat-
ically decreased activity (>10000 nM). On the other hand,
when we installed substituents on the phenyl-substituted
side chain, compound 87 regained its cytotoxicity (32
nM), and interestingly, its epimer 89 also displayed mod-
4
Synthesis of Psymberin Analogues: The Dis-
covery of C11-Deoxypsymberin
To further demonstrate the scope of the PhI(OAc)2-medi- erate activity (615 nM). These results suggest that the sub-
ated oxidative cyclization reaction in complex molecule stitution on the side chain (C4 and C5) plays an important
synthesis, we turned our attention to the synthesis of ad- role in psymberin activity.
vanced psymberin analogues. This will also serve the pur-
Table 4 Anti-Cancer Activity of Side–Chain-Modified Psymberin
Analoguesa
pose of testing the biological activity of these analogues.
We initially focused our effort on the modification of the
O
OMe
O
OMe
side chain and left the dihydroisocoumarin part untouched
so that we could apply the oxidative cyclization strategy
starting from a common intermediate. Notably, enamides
82 and 83 cyclized smoothly via the PhI(OAc)2-mediated
cyclization reaction to give the corresponding cyclization
products 86 and 87 and their epimers 88 and 89 after glo-
bal deprotection with tetra-n-butylammonium fluoride
(Scheme 12).
OH
OH
9
8
9
8
R
N
H
R
N
H
O
O
HO
HO
OH
OH
O
O
67, 86, 87, 90
epi-67, 88, 89
IC50 [nM]
>10000
OH
O
OH
O
Compd
R
O
O
OMe
67
Me
9
8
OTBS
OTBS
R
N
H
R
N
H
86
87
>10000
32 1
HO
O
15
AcO
AcO
PhI(OAc)2,
75–85%
OMe
OH
OH
O
O
TIPSO
O
82, 83
TIPSO
O
OH
OMe
84a, 85a, C8, C9 = S,S
84b, 85b, C8, C9 = R,R
90
260 36
HO
Me
TBAF, 90–95%
OH
O
OMe
O
OMe
O
epi-67
>10000
>10000
OH
OH
R
N
H
R
N
H
O
88
+
HO
HO
OMe
OH
O
OH
O
89
615 15
OH
O
OH
O
OH
88,89, C8, C9 = R,R
86,87, C8, C9 = S,S
a The CellTiter-Glo Luminescent Cell Viability Assay (Promega,
Technical bulletin 288) was employed in this study. IC50 data are the
mean value of three experiments with statistical significance calculat-
ed. Value for psymberin in this assay is 0.42 nM.
Ph
Ph
82, 84a, 84b, R =
86, 88, R =
OMe
OMe
Ph
83, 85a, 85b, R =
Ph
87, 89, R =
OTBDPS
OH
Scheme 12 The synthesis of side-chain-modified psymberin
We next turned our attention to modifying the tetrahydro-
pyran ring core of psymberin. We noticed that irciniasta-
tin B (C11-substituted with =O) consistently showed ten
times stronger cell-growth inhibition than irciniastatin A
(psymberin, C11-substituted with –OH) in most of the
cancer cell lines tested.11b Based on this observation, we
synthesized the C11-deoxypsymberin 91 using a similar
synthetic sequence as for natural psymberin starting with
the preparation of central piece 92 (Scheme 13).23 Again,
the key PhI(OAc)2-mediated cyclization worked well (94,
95) and gave the desired products in good yields. The four
diastereomers of the final products (91, 91a–c) were care-
With these advanced psymberin analogues in hand, we
chose to use the readily available HOP62 human lung can-
cer cell line for testing of the analogues (Table 4). When
the side chain was truncated to a methyl group, both 67
and its epimer epi-67 showed much reduced cytotoxic ac-
tivity. When the terminal double bond was converted into
a terminal hydroxy group, compound 90 (prepared by
deprotection from compound 81) retained a good level of
cytotoxicity (260 nM), but was 650-fold weaker com-
pared to psymberin (0.42 nM). This suggests that the dou-
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

PAPER
Total Synthesis of Psymberin
2865
11
I
13
AcO
BnO
TMSO
15
TIPSO
BnO
Br
21
OAc
O
92
TIPSO
O
93, E/Z = 5:1
OMe
O
OMe
O
OMe
BnO
HO
9
8
N
H
N
H
OTPS
HO
OTPS
O
AcO
15
1. PhI(OAc)2, 89%
2. Ac2O, py, DMAP
AcO
OH
O
OAc
O
3. H2, Pd/C, 85%
over two steps
TIPSO
O
TIPSO
O
95 C8, C9 = S,S
epi-95 C8, C9 = R,R
94
1. o-O2NC6H4SeCN
2. H2O2, THF
3. TBAF, 60%
over three steps
O
OMe
O
OMe
9
8
9
8
OMe
O
OMe
O
N
H
N
H
O
O
N
H
OH
91a C8, C9 = R,R
91b C8, C9 = R,S
HO
O
OMe
OH
O
9
8
N
H
O
OH
O
91c C8, C9 = S,R
C11-Deoxypsymberin 91
Scheme 13 The synthesis of C11-deoxypsymberin and its epimers
fully isolated and subjected to biological testing. The ma- gratifyingly its epimers (91a–c) showed improved activi-
jor product with the configuration of the natural product ty compared to its corresponding psymberin epimers.
(compound 91) displayed extremely potent activity, and Compound 91 was consistently 3–10-fold more potent
Table 5 Anti-Cancer Activity of Psymberin, C11-Deoxypsymberin and Its Epimersa
O
OMe
O
OMe
O
OMe
O
OMe
S
S
S
N
H
R
R
R
N
H
N
H
S
R
N
H
O
O
O
O
91b
91a
91
91c
91
91a
n.d.
n.d.
n.d.
177
n.d.
n.d.
n.d.
n.d.
n.d.
91b
n.d.
n.d.
n.d.
46
91c
8.7
5.9
1.6
3.0
5.3
3.9
2.9
6.1
3.8
Psymberin 38
0.76
epi-38
6800
3800
2400
4600
4200
5200
3100
4800
n.d.
Cell line
ACHN
DU145
H226
Tissue type
kidney
prostate
lung
0.265
0.149
0.034
0.055
0.142
0.076
0.073
0.160
0.066
0.30
0.18
0.42
HOP62
MB231
MKN45
PC3
lung
n.d.
n.d.
n.d.
n.d.
n.d.
0.27
breast
0.28
gastric
prostate
colon
0.19
0.82
SW620
NHDF
0.84
normal
a The CellTiter-Glo Luminescent Cell Viability Assay (Promega, Technical bulletin 288) was employed in this study. The value shown is IC50
data in nM and the mean value of three experiments with statistical significance calculated; n.d. = not determined.
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

2866
N. Shao et al.
PAPER
than psymberin across all cancer cell lines tested with IC50 advanced analogues of psymberin were prepared using
values in the subnanomolar ranges while epimer 91c the oxidative cyclization reaction as the key step. During
showed excellent activity against all cell lines tested with this journey, we have identified potent C11-deoxypsym-
IC50 value in single digit nanomolar ranges (Table 5).
berin which is consistently more potent than psymberin.
These findings open the door for new opportunities in the
SAR development of the psymberin and pederin family.
To further determine whether the C11-deoxypsymberin
SAR could be translated to its analogues a hybrid struc-
ture 96 with the side chain containing a terminal phenyl
group was synthesized. The synthesis was straightforward
starting from epoxide 73 and the key cyclization reaction
was efficient (101 → 96) (Scheme 14). Indeed, com-
pounds 96 and its epimer epi-96 displayed very good po-
tency against HOP62 human lung cancer line. Compound
96 with natural psymberin stereochemistry had a subna-
nomolar IC50 value (0.133 nM) which was 3-fold more
All reactions were carried out under a nitrogen atmosphere with dry
solvents under anhydrous conditions, unless otherwise noted. Re-
agents were purchased at the highest commercial quality and used
without further purification, unless otherwise stated. Flash chroma-
tography (FC) was carried out with EM science Silica gel 60 (neu-
tral, 230–400 mesh). 1H NMR and 13C NMR spectra were recorded
on a Bruker Avance 500 NMR spectrometer with chemical shifts re-
ported in ppm relative to the residual deuterated solvent. LCMS was
potent than psymberin (0.42 nM). Even the corresponding recorded on Applied Biosystem API-150 apparatus. Key experi-
mental procedures as well as spectroscopic data of key products are
summarized in the following experimental section. All other exper-
epimer epi-96 showed good potency (IC50 = 165 nM).
These findings are very helpful to facilitate further psym-
imental procedures and spectroscopic data for the following com-
berin SAR studies and may be applicable to all members
of the pederin family of natural products.
pounds can be found in the supporting information: 4, 6, 10, 12–38,
40, 43, 46–49, 52, 57–70, 74–78, 80–101.
Oxidative Cyclization of N-Acyl Enamine; General Procedure
OMe
OMe
H
1. PhMgBr,
82%
1. TBAF,
99%
O
To a stirred solution of the substrate (0.2 mmol) in HFI (2 mL) un-
der argon at 0 °C, MeOH (4.0 mmol, 20 equiv) was added followed
by the dropwise addition of PhI(OAc)2 (146 mg, 0.44 mmol, 2.2
equiv) in HFI (1 mL). The reaction mixture was stirred at 0 °C for
1 h before it was diluted with CH2Cl2 (for hindered cases, the reac-
tions were stirred at r.t. until TLC indicated the completion of the
reaction). The diluted solution was passed through a short pad of sil-
ica gel and eluted with EtOAc. The solvent was removed under re-
duced pressure, and the residue was purified with silica gel flash
column chromatography (acetone–hexane) to give the desired prod-
uct as a mixture of diastereomers.
OTBS
OTBS
O
2. DMP,
94%
2. Me3OBF4,
95%
73
97
98
1. TMSCN, AlCl3,
99%
2. TPSCl
OMe
O
OMe
MeCONH2, PdCl2,
60% two steps
59 +
CN
NH2
OTPS
OTPS
99
100
1. 59, CuI, 92%
2. NaOMe, MeOH
3. Ac2O, py, 90%
over 2 steps
Synthesis of Ester 47
To a solution of ligand 52 (2.12 g, 5.4 mmol) in propylnitrile (40
mL) was added BH3·THF complex (5.4 mL, 5.4 mmol) at r.t. and
the mixture was stirred at 45 °C for 1 h before cooled to –78 °C.
Ketene acetal 44 (7.71 g, 41 mmol) was then added at –78 °C, fol-
lowed by the addition of aldehyde BnO(CH2)2CHO (4.43 g, 27
mmol) as a solution in propylnitrile (10 mL) over 4 h via a syringe
pump. The reaction was stirred for another 2 h before quenching
with pH 7 buffer (50 mL). The reaction mixture was warmed to r.t.
and diluted with EtOAc (100 mL). The organic solution was washed
with NaHCO3 (30 mL), dried over Na2SO4 and concentrated. The
residue was purified by FC (EtOAc–hexane, 1:10 to 1:2) to give a
mixture of TMS-protected hydroxy ester and free hydroxy ester.
The TMS-protected product was treated with HCl–Et2O (2.0 M, 2.0
equiv) at 0 °C for 1 h at r.t. The solution was directly concentrated
and purified by FC (EtOAc–hexane, 1:2) to give the alcohol (7.16
g, 95%).
OMe
O
OMe
OMe
O
9
8
N
H
N
H
1. PhI(OAc)2,
82%
OH
O
OTPS
HO
2. TBAF,
90%
15
HO
AcO
OH
O
OH
O
OH
O
101
TIPSO
O
C11-Deoxypsymberin derivative 96
+
O
OMe
R
N
H
R
O
epi-96 C8, C9 = R,R
To a solution of the alcohol (5.45 g, 19.4 mmol) in CH2Cl2 (30 mL)
was added 2,6-lutidine (4.66 mL, 40 mmol) at 0 °C. After stirring
for 5 min at 0 °C, TBSOTf (9.18 mL, 40 mmol) was added drop-
wise. The reaction mixture was allowed to stir at r.t. for 2 h. After
the reaction was complete (TLC), the solution was diluted with
EtOAc (150 mL). The organic layer was washed with NaHCO3 (50
mL), dried over Na2SO4 and concentrated. The residue was purified
by FC (EtOAc–hexane, 1:10) to give the TBS-protected product 47
(7.56 g, 99%) as an oil.
Scheme 14 The synthesis of the C11-deoxypsymberin analogue
5
Conclusion
A novel PhI(OAc)2-mediated cascade ring-closing reac-
tion was discovered to construct the core unit of the ped-
erin family of natural products. This methodology can be
utilized to synthesize both cyclic a-oxy N-acyl aminals
and hemiaminals. We successfully applied this chemistry
in the total synthesis of the natural product psymberin. To
further demonstrate the scope of this reaction, a series of
[a]D20 +41.85 (c 1.0, CH2Cl2).
1H NMR (CDCl3): d = 7.36–7.26 (m, 5 H), 4.48 (dd, J = 12.0, 15.5
Hz, 2 H), 4.14–4.03 (m, 3 H), 3.55–3.46 (m, 2 H), 1.80 (ddt, J = 3.0,
7.5, 14.5 Hz, 1 H), 1.66 (ddt, J = 5.0, 15.0, 15.5 Hz, 1 H), 1.22 (t,
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

PAPER
Total Synthesis of Psymberin
2867
J = 7.5 Hz, 3 H), 1.15 (s, 3 H), 1.09 (s, 3 H), 0.86 (s, 3 H), 0.03 (s,
3 H), 0.01 (s, 3 H).
1H NMR (CDCl3): d = 7.34–7.26 (m, 5 H), 6.30 (s, 1 H), 4.48 (d,
J = 12.0 Hz, 1 H), 4.44 (d, J = 12.0 Hz, 1 H), 4.36 (ddd, J = 2.5, 5.0,
12.5 Hz, 1 H), 4.31–4.27 (m, 1 H), 4.05 (dd, J = 2.5, 7.5 Hz, 1 H),
3.49 (dd, J = 5.5, 8.0 Hz, 2 H), 3.24 (d, J = 2.0 Hz, 1 H), 2.89 (dd,
J = 2.5, 16.0 Hz, 1 H), 2.78–2.71 (m, 2 H), 2.08 (s, 3 H), 1.83–1.69
(m, 2 H), 1.62–1.54 (m, 5 H), 1.35–1.25 (m, 6 H), 1.18–1.05 (m, 42
H), 0.86 (s, 9 H), 0.05 (s, 3 H), 0.04 (s, 3 H).
13C NMR (CDCl3): d = 177.2, 138.7, 128.5, 127.8, 127.7, 74.0,
73.0, 67.9, 60.6, 48.4, 34.2, 26.2, 22.1, 20.4, 18.5, 14.3, –3.8, –4.0.
MS (ES): m/z [M + H]+ calcd for C22H39O4Si: 395.26; found
395.24.
MS (ES): m/z [M + Na]+ calcd for C22H38NaO4Si: 417.24; found
417.34.
13C NMR (CDCl3): d = 217.4, 163.1, 159.0, 157.7, 141.4, 138.5,
128.6, 127.9, 118.3, 109.7, 78.3, 74.0, 73.2, 67.9, 67.5, 53.8, 42.9,
41.7, 34.7, 29.9, 26.2, 22.3, 20.2, 18.5, 18.3, 18.2, 17.9, 13.5, 13.4,
12.0, 9.6, –3.7, –3.8.
Synthesis of Ketone 48
To a solution of compound 47 (7.56 g, 19.2 mmol) in anhyd pentane
(50 mL) was added dropwise trimethylsilylmethyl lithium (58 mL,
58.2 mmol) at 0 °C. The mixture was stirred at 0 °C for 4 h with
TLC monitoring (Et2O–hexane, 1:10). After the reaction was com-
plete (TLC), anhyd MeOH (10 mL) was added to this suspension
and the resulting emulsion was stirred for another 1 h at r.t. The
mixture was diluted with Et2O–H2O (100 mL, 3:1). The aqueous
layer was extracted with Et2O (100 mL), dried over Na2SO4 and
concentrated. The residue was purified by FC (EtOAc–hexane,
1:10) to give methyl ketone 48 (5.81 g, 82%) as an oil.
MS (ES): m/z [M + H]+ calcd for C52H91O8Si3: 927.60; found
927.73.
MS (ES): m/z [M + Na]+ calcd for C52H90NaO8Si3: 949.58; found
949.68.
Synthesis of Alcohol 58
To a solution of compound 57 (1.15 g, 1.54 mmol) in anhyd THF
(30 mL) was added catecholborane (20 mL, 1.0 M in THF, 20
mmol) at –78 °C. After stirring at 0 °C for 15 h, the reaction was
quenched by anhyd MeOH (10 mL) and an aq solution of sodium
potassium tartrate (50 mL). The mixture was further stirred at r.t. for
1 h followed by a normal workup process with EtOAc. The residue
was purified by FC (EtOAc–hexane, 1:8) to give the desired product
58 (1.05 g, 92%) (de = 15:1).
[a]D20 +7.45 (c 1.0, CH2Cl2).
1H NMR (CDCl3): d = 7.35–7.27 (m, 5 H), 4.49 (d, J = 12.0 Hz, 1
H), 4.45 (d, J = 12.0 Hz, 1 H), 4.04 (dd, J = 2.5, 7.5 Hz, 1 H), 3.50–
3.48 (m, 2 H), 2.13 (s, 3 H), 1.78–1.72 (m, 1 H), 1.62–1.55 (m, 1 H),
1.10 (s, 3 H), 1.06 (s, 3 H), 0.87 (s, 9 H), 0.05 (s, 3 H), 0.03 (s, 3 H).
[a]D20 +44.37 (c 0.1, CH2Cl2).
13C NMR (CDCl3): d = 213.7, 138.7, 128.6, 127.7, 74.1, 73.1, 67.7,
53.5, 34.5, 27.1, 26.2, 22.1, 20.3, 18.5, –3.8.
1H NMR (CDCl3): d = 7.37–7.28 (m, 5 H), 6.30 (s, 1 H), 4.87 (br, 1
H), 4.51 (d, J = 12.0 Hz, 1 H), 4.48 (d, J = 12.0 Hz, 1 H), 4.35 (ddd,
J = 2.5, 5.5, 12.5 Hz, 1 H), 4.25 (br, 1 H), 4.14 (d, J = 10.0 Hz, 1 H),
4.06 (d, J = 9.5 Hz, 1 H), 3.66 (dd, J = 2.5, 7.5 Hz, 1 H), 3.60–3.52
(m, 2 H), 2.98 (dd, J = 3.0, 16.5 Hz, 1 H), 2.80 (dd, J = 12.0, 16.0
Hz, 1 H), 2.10 (s, 3 H), 2.11–2.09 (m, 1 H), 1.87–1.78 (m, 2 H),
1.66–1.59 (m, 1 H), 1.35–1.25 (m, 6 H), 1.17–1.11 (m, 39 H), 1.02
(s, 3 H), 0.87 (s, 9 H), 0.75 (s, 3 H), 0.10 (s, 3 H), 0.07 (s, 3 H).
MS (ES): m/z [M + H]+ calcd for C21H37O3Si: 365.25; found
365.36.
MS (ES): m/z [M + Na]+ calcd for C21H36NaO3Si: 387.23; found
387.09.
Synthesis of Enol Ether 49
13C NMR (CDCl3): d = 163.5, 158.9, 157.6, 141.8, 138.6, 128.6,
127.9, 118.4, 109.7, 81.3, 78.7, 77.9, 73.1, 72.7, 67.6, 43.1, 40.8,
35.0, 33.3, 29.9, 26.2, 23.5, 21.0, 18.4, 18.3, 18.2, 13.5, 13.4, 12.1,
9.9, –3.8, –4.1.
To a solution of compound 48 (1.34 g, 3.6 mmol) in CH2Cl2 (15
mL) at 0 °C was added Et3N (2 mL, 14.4 mmol) and then TMSOTf
(1.3 mL, 7.2 mmol). The reaction was allowed to stir at r.t. for 1 h.
The solution was diluted with hexane (75 mL), washed with aq
NaHCO3 (30 mL) and dried over Na2SO4. The solution was directly
passed through a short pad of silica gel plug and washed with
EtOAc–hexane (100 mL, 1:20). The combined organic solution was
concentrated to give compound 49 (1.58 g, quant.) as a light yellow
oil. Crude 49 was used for the next step without any further purifi-
cation.
MS (ES): m/z [M + H]+ calcd for C52H93O8Si3: 929.62; found
929.68.
MS (ES): m/z [M + Na]+ calcd for C52H92NaO8Si3: 951.60; found
951.66.
Compound 59
1H NMR (CDCl3): d = 7.33–7.25 (m, 5 H), 4.48 (s, 2 H), 4.11 (s, 1
H), 3.95 (s, 1 H), 3.87 (dd, J = 3.0, 8.0 Hz, 1 H), 3.55–3.45 (m, 2 H),
1.93–1.86 (m, 1 H), 1.61–1.54 (m, 1 H), 1.02 (s, 3 H), 0.94 (s, 3 H),
0.87 (s, 9 H), 0.18 (s, 9 H), 0.03 (s, 3 H), 0.01 (s, 3 H).
1H NMR (CDCl3): d = 6.52–3.46 (m, 1 H), 6.28 (s, 1 H), 6.06 (d,
J = 14.5 Hz, 1 H), 5.09–5.05 (m, 1 H), 4.91 (dd, J = 1.5, 10.5 Hz, 1
H), 4.07 (ddd, J = 2.5, 6.5, 12.0 Hz, 1 H), 3.50 (dd, J = 4.0, 7.0 Hz,
1 H), 2.90 (dd, J = 2.0, 16.0 Hz, 1 H), 2.63 (dd, J = 12.0, 16.0 Hz, 1
H), 2.38–2.32 (m, 1 H), 2.25–2.19 (m, 1 H), 2.12–2.10 (m, 1 H),
2.08 (s, 3 H), 2.07 (s, 3 H), 2.02 (s, 3 H), 1.99–1.96 (m, 1 H), 1.78–
1.72 (m, 1 H), 1.34–1.24 (m, 6 H), 1.15 (d, J = 7.0 Hz, 3 H), 1.12–
1.09 (m, 36 H), 0.89 (s, 9 H), 0.88 (s, 3 H), 0.85 (s, 3 H), 0.06 (s, 3
H), 0.05 (s, 3 H).
13C NMR (CDCl3): d = 165.0, 139.0, 128.5, 127.8, 127.6, 88.2,
73.2, 72.9, 68.6, 45.4, 33.8, 26.4, 24.7, 19.8, 18.6, 0.28, –3.8.
Synthesis of Ketone 57
A mixture of aldehyde 43 (910 mg, 1.7 mmol) and silyl enol ether
49 (1.58 g, 3.6 mmol) was azeotroped with toluene and dissolved in
anhyd CH2Cl2 (25 mL). The solution was cooled to –78 °C and
freshly distilled BF3·OEt2 (2.2 mL, 18 mmol) was added dropwise.
The reaction was allowed to stir at –78 °C overnight. The reaction
was quenched by adding sat. NaHCO3 (30 mL) before warming up
to r.t. The solution was extracted with EtOAc (80 mL). The com-
bined organic layer was washed with brine (20 mL) and dried over
Na2SO4. After concentration the residue was purified by FC (15%
EtOAc in hexane) to give the product (1.15 g, 91%) as a colorless
oil. The excess enol ether was recovered as the hydroxy ketone 57.
13C NMR (CDCl3): d = 171.0, 170.7, 162.8, 159.0, 157.7, 144.4,
141.0, 139.1, 118.3, 109.8, 109.7, 83.7, 78.1, 77.9, 77.1, 76.3, 74.4,
71.6, 43.4, 43.2, 40.2, 39.5, 32.7, 30.4, 26.3, 21.4, 21.3, 21.1, 20.9,
20.1, 18.5, 18.2, 18.1, 13.5, 13.3, 11.9, 9.2, –3.1, –3.9.
MS (ES): m/z [M + H]+ calcd for C50H90IO9Si2: 1045.50; found
1045.72.
Synthesis of Enamide 80a
An oven-dried sealed tube was charged with amide 76 (500 mg,
0.99 mmol), CuI (31 mg, 0.16 mmol), and Cs2CO3 (326 mg, 1
[a]D20 +28.59 (c 0.1, CH2Cl2).
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

2868
N. Shao et al.
PAPER
mmol) sequentially. Under argon, dimethyl ethylenediamine (0.035
mL, 0.32 mmol) was then added followed by a solution of vinyl io-
dide 59 (346 mg, 0.33 mmol) in toluene (5 mL). The tube was filled
with argon and quickly capped and sealed. The reaction was stirred
vigorously at 70 °C for 20 h. After the solution was cooled to r.t., it
was diluted with EtOAc (10 mL) and filtered off via a short pad of
silica gel and washed with EtOAc (50 mL). The combined organic
solution was concentrated and purified by FC (acetone–hexane,
1:20 to 1:10) to give enamide 80a (450 mg, 95%) as an oil. The ex-
cess amide was recovered. For characterization purpose, a small
amount of the E/Z-mixture was separated.
Synthesis of Compound 81a
Under an atmosphere of argon, to a solution of compound 80 (300
mg, 0.24 mmol) in HFI (4 mL) at 0 °C was added MeOH (0.2 mL)
and then dropwise a solution of PhI(OAc)2 (320 mg, 0.96 mmol) in
HFI (2 mL). The solution was stirred at r.t. for 70 h. The reaction
mixture was diluted with EtOAc (20 mL) and filtered through a
short pad of silica gel. The filtrate was concentrated and then puri-
fied by FC (acetone–hexane, 1:10 to 1:2) to give two crude fractions
which were rechromatographed with EtOAc–toluene (1:10 to 1:2)
to give 81a and epi-81a (90 mg each, 60% combined). Another
~15% of a side product was obtained as a mixture, which was not
fully characterized.
E-Isomer
1H NMR (CDCl3): d = 8.18 (d, J = 11.0 Hz, 1 H), 7.69–7.30 (m, 15
H), 6.62 (dd, J = 12.0, 13.5 Hz, 1 H), 6.29 (s, 1 H), 5.10 (br, 1 H),
5.03 (m, 1 H), 4.92 (d, J = 9.0 Hz, 1 H), 4.47 (s, 1 H), 4.45 (dd,
J = 5.0, 10.0 Hz, 1 H), 4.37 (dd, J = 1.5, 19.5 Hz, 1 H), 4.07 (ddd,
J = 2.5, 6.5, 12.5 Hz, 1 H), 3.42 (m, 2 H), 3.27 (m, 1 H), 3.19–3.08
(m, 4 H), 2.92 (dd, J = 2.5, 16.5 Hz, 1 H), 2.63 (dd, J = 12.5, 16.5
Hz, 1 H), 2.30 (m, 1 H), 2.15 (m, 2 H), 2.12–1.72 (m, 13 H), 1.60
(m, 1 H), 1.35–1.25 (m, 8 H), 1.18–1.10 (m, 49 H), 0.90–0.80 (m,
18 H), 0.02 (s, 3 H), –0.04 (s, 3 H).
Compound 81a
1H NMR (CDCl3): d = 7.69–7.24 (m, 15 H), 6.52, 6.51 (s, 1 H), 5.19
(m, 1 H), 4.43–4.38 (m, 3 H), 4.26 (br, 1 H), 4.08 (br, 1 H), 3.97 (br,
1 H), 3.72, 3.70 (s, 1 H), 3.52–3.48 (m, 2 H), 3.34 (s, 3 H), 3.33 (s,
3 H), 3.29 (m, 1 H), 3.22–3.00 (m, 6 H), 2.92–2.85 (m, 1 H), 2.31
(s, 3 H), 2.05 (s, 3 H), 2.04 (s, 3 H), 1.84–1.44 (m, 3 H), 1.37–1.31
(m, 3 H), 1.13–1.10 (m, 30 H), 0.90 (s, 9 H), 0.80 (s, 3 H), 0.77 (s,
3 H), 0.74, 0.71 (d, J = 6.5 Hz, 3 H), 0.04 (s, 6 H).
13C NMR (CDCl3): d = 172.4, 172.2, 168.5, 162.7, 157.0, 152.9,
142.3, 136.3, 135.9, 132.8, 130.3, 130.1, 128.4, 128.0, 127.9, 127.6,
127.5, 113.3, 82.3, 79.1, 76.2, 75.5, 72.9, 72.2, 56.2, 43.3, 34.5,
30.4, 30.0, 29.6, 27.3, 25.9, 21.0, 19.6, 18.1, 16.8, 14.3, 13.3, 12.0,
9.8, –4.4, –4.9.
13C NMR (CDCl3): d = 171.1, 170.8, 168.4, 163.0, 159.0, 157.7,
141.2, 136.3, 135.9, 130.5, 128.5, 128.3, 128.1, 127.8, 127.7, 127.6,
123.1, 118.4, 110.0, 109.7, 82.1, 78.2, 76.4, 75.6, 74.7, 73.1, 73.0,
71.8, 43.3, 39.4, 33.9, 30.5, 27.3, 26.4, 21.5, 21.4, 21.1, 20.4, 18.6,
18.3, 18.2, 16.9, 13.5, 13.4, 11.9, 9.1, –2.8, –3.9.
MS (FAB): m/z [M + Na]+ calcd for C71H109NNaO13Si3: 1290.7104;
found 1290.7100.
MS (FAB): m/z [M + Na]+ calcd for C81H129NNaO13Si4: 1458.8439;
found 1458.8439.
Synthesis of Alcohols 81 and epi-81
Synthesis of Compound 80
To a solution of compound 81a (48 mg, 0.038 mmol) in CH2Cl2 (1
mL) was added pyridine (0.30 mL), Ac2O (0.15 mL) and DMAP
(2.5 mg, 0.02 mmol) sequentially. The reaction mixture was stirred
at r.t. overnight. After quenching with aq NH4Cl, the aqueous layer
was extracted with EtOAc thoroughly. The combined organic layer
was washed with brine (10 mL), dried over Na2SO4 and concentrat-
ed. The residue was roughly purified by FC and directly used in the
next reaction step. To a solution of the acetate compound (20 mg,
0.015 mmol) in MeOH (2 mL) was charged Pd/C (5 mg), and the
flask was sealed with a rubber stopper. The inner atmosphere was
exchanged three times with H2 before the solution was allowed to
stir under H2 (double layer H2 balloon) overnight. The solution was
then filtered off a celite pad followed by washing with MeOH
(2 ꢀ 30 mL). The combined organic layer was concentrated and pu-
rified by FC to give alcohol 81 (17.3 mg, 95%). epi-81 was obtained
in the same sequence and could be isolated by PTLC. For character-
ization purposes both compounds were separated and analyzed by
NMR spectroscopy (epi-81-I, epi-81-II).
The compound 80a (450 mg, 0.31 mmol) was treated with a solu-
tion of NaOMe in MeOH (10 mL, 0.06 M) at r.t. for 5 h before
quenching by H2O. The mixture was then extracted with EtOAc
thoroughly. The organic layer was washed with brine (20 mL),
dried over Na2SO4 and concentrated. The crude product was dis-
solved in CH2Cl2 (8 mL). At 0 °C, pyridine (200 mL), DMAP (4.0
mg) and then Ac2O (100 mL) were added sequentially. The reaction
was allowed to stir at 0 °C for 1 h before quenching with aq NH4Cl
(10 mL). The aqueous layer was extracted with EtOAc (80 mL).
The combined organic layer was washed with NH4Cl (15 mL) and
brine (15 mL). After being dried and concentrated, the residue was
purified by FC (acetone–hexane, 1:20 to 1:6) to give 80 as a mixture
(368 mg, 95%) of four isomers (E/Z was separable on PTLC, hence,
a small part was separated for characterization purposes). The mix-
ture was carried on to the next reaction directly. NMR data of the
major E-isomer are given (NMR spectrum of the Z-isomer is at-
tached in the Supporting Information).
1H NMR (CDCl3): d = 8.19 (d, J = 10.5 Hz, 1 H), 7.69–7.26 (m, 15
H), 6.65 (dd, J = 11.5, 13.5 Hz, 1 H), 6.53 (s, 1 H), 5.00 (m, 1 H),
4.86 (s, 1 H), 4.47 (s, 1 H), 4.43–4.36 (m, 2 H), 4.26 (s, 1 H), 4.15
(d, J = 9.0 Hz, 1 H), 4.06 (d, J = 10.5 Hz, 1 H), 3.43 (m, 2 H), 3.26
(m, 1 H), 3.18 (m, 1 H), 3.14–3.08 (m, 4 H), 3.03 (d, J = 16.5 Hz, 1
H), 2.86 (dd, J = 12.5, 16.0 Hz, 1 H), 2.42 (m, 1 H), 2.33 (s, 3 H),
2.26 (m, 1 H), 2.05–1.99 (m, 4 H), 1.88 (m, 2 H), 1.62 (m, 1 H),
1.40–1.30 (m, 4 H), 1.15–1.10 (m, 29 H), 1.01 (s, 3 H), 0.89 (s, 9
H), 0.92 (d, J = 7.0 Hz, 3 H, isomer I), 0.86 (d, J = 7.0 Hz, 3 H, iso-
mer II), 0.82 (s, 3 H), 0.14 (s, 3 H), 0.05 (s, 3 H).
Compound 81
1H NMR (CDCl3): d = 7.68–7.33 (m, 10 H), 6.52 (s, 1 H), 5.39 (m,
1 H), 5.07 (dm, J = 10.0 Hz, 1 H), 4.49 (dd, J = 2.0, 10.0 Hz, 1 H),
4.19 (br, 1 H), 4.11 (m, 1 H), 3.55 (br, 1 H), 3.39, 3.38 (s, 3 H), 3.34
(dm, J = 12.5 Hz, 1 H), 3.23 (m, 2 H), 3.13 (dm, J = 16.5 Hz, 1 H),
2.99, 2.98 (s, 3 H), 2.68–2.61 (m, 2 H), 2.32 (s, 3 H), 2.14 (m, 1 H),
2.09 (m, 1 H), 2.02 (s, 3 H), 1.98, 1.97 (s, 3 H), 1.77–1.39 (m, 6 H),
1.37–1.30 (m, 3 H), 1.18 (d, J = 7.0 Hz, 3 H), 1.13–1.11 (m, 29 H),
0.88 (s, 9 H), 0.82 (s, 3 H), 0.77, 0.76 (s, 3 H), 0.66, 0.61 (d, J = 7.0
Hz, 3 H), 0.05 (s, 3 H), 0.04 (s, 3 H).
13C NMR (CDCl3): d = 168.6, 168.4, 168.3, 162.7, 157.0, 153.0,
142.2, 136.2, 135.8, 130.5, 130.4, 128.4, 128.2, 128.0, 127.6, 127.5,
115.0, 113.5, 84.9, 81.6, 78.8, 77.8, 76.2, 75.5, 73.0, 72.9, 72.4,
43.0, 34.9, 33.1, 30.3, 30.0, 29.7, 27.2, 26.2, 23.7, 21.1, 19.7, 18.4,
18.3, 18.2, 16.8, 13.3, 12.0, 9.99, –3.3, –3.9.
13C NMR (CDCl3): d = 172.3, 172.1, 170.5, 168.7, 162.3, 157.0,
153.0, 141.8, 136.3, 135.7, 132.9, 132.8, 130.4, 130.3, 130.1, 128.0,
127.9, 119.6, 119.5, 114.6, 113.4, 83.0, 82.5, 81.7, 79.2, 71.4, 68.1,
67.3, 57.6, 56.5, 33.5, 32.4, 30.0, 27.3, 25.9, 21.3, 21.1, 19.6, 18.1,
18.0, 17.5, 17.2, 13.3, 11.9, 11.8, –4.4, –5.0.
MS (FAB): m/z [M + Na]+ calcd for C70H107NNaO12Si3: 1260.6999;
found 1260.7005.
MS (FAB): m/z [M + Na]+ calcd for C66H105NNaO14Si3: 1242.6741;
found 1242.6745.
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

PAPER
Total Synthesis of Psymberin
2869
Compound epi-81-I
EtOAc (10 mL), the aqueous layer was separated and the pH was
adjusted to 6 with aq NaHSO4 solution. The aqueous layer was then
extracted with EtOAc (30 mL) thoroughly. The combined organic
layer was washed with brine (10 mL), dried and concentrated.
PTLC of the residue (MeOH–CH2Cl2, 1:20) gave epi-psymberin
(epi-38) as a light yellow glass.
1H NMR (CDCl3): d = 7.86 (d, J = 10.0 Hz, 1 H), 7.73–7.34 (m, 10
H), 6.52 (s, 1 H), 5.34 (m, 1 H), 4.90 (d, J = 10.0 Hz, 1 H), 4.47 (d,
J = 1.5 Hz, 1 H), 4.08 (m, 1 H), 3.65 (dm, J = 10.5 Hz, 1 H), 3.42
(dd, J = 5.0, 11.0 Hz, 1 H), 3.23 (s, 3 H), 3.27–3.10 (m, 4 H), 2.97
(s, 3 H), 2.66 (dd, J = 12.0, 16.0 Hz, 1 H), 2.34 (s, 3 H), 2.30 (m, 1
H), 2.01 (s, 3 H), 1.97 (s, 3 H), 1.78–1.42 (m, 8 H), 1.36–1.28 (m, 3
H), 1.16–1.10 (m, 31 H), 0.92 (s, 3 H), 0.87 (s, 12 H), 0.64 (d,
J = 6.5 Hz, 3 H), 0.05 (s, 3 H), 0.01 (s, 3 H).
[a]D20 –13.5 (c 0.1, MeOH).
1H NMR (CDCl3): d = 11.24 (s, 1 H), 7.37 (d, J = 9.5 Hz, 1 H), 6.28
(s, 1 H), 5.04 (dd, J = 2.5, 10.0 Hz, 1 H), 4.83 (s, 1 H), 4.80 (s, 1 H),
4.56 (ddd, J = 3.5, 9.0, 13.0 Hz, 1 H), 4.36 (d, J = 4.0 Hz, 1 H), 4.12
(m, 1 H), 3.80 (dm, J = 11.5 Hz, 1 H), 3.73 (m, 1 H), 3.49 (dd,
J = 5.0, 11.0 Hz, 1 H), 3.43 (s, 3 H), 3.35 (s, 3 H), 3.31 (d, J = 10.5
Hz, 1 H), 2.95 (dd, J = 4.0, 16.5 Hz, 1 H), 2.90 (dd, J = 12.0, 16.5
Hz, 1 H), 2.33 (dd, J = 9.0, 14.5 Hz, 1 H), 2.20 (m, 1 H), 2.04 (s, 3
H), 1.95 (m, 1 H), 1.77 (s, 3 H), 1.73 (m, 2 H), 1.65 (m, 1 H), 1.44
(m, 1 H), 1.13 (d, J = 7.5 Hz, 3 H), 0.96 (s, 3 H), 0.88 (s, 3 H).
MS (FAB): m/z [M + Na]+ calcd for C66H105NNaO14Si3: 1242.6741;
found 1242.6745.
Synthesis of Psymberin 38 and Its Epimer
Substrate 81 was azeotropically condensed with toluene three times
before use. A solution of the compound (25 mg, 0.02 mmol) in an-
hyd THF (1 mL) was charged with o-nitrophenyl selenocyanate (18
mg, 0.08 mmol) followed by Bu3P (0.025 mL, 0.1 mmol). The reac-
tion mixture should turn to dark red immediately. The solution was
allowed to stir at r.t. for 2 h. Then it was quenched by adding MeOH
(0.5 mL) followed by stirring for additional 15 min before it was
concentrated. The residue was purified by FC (EtOAc–hexane, 1:10
to 2:3) to give a mixture of selenated products. (The mixture result-
ed from the liability of phenol acetate and the TIPS group as a part
of them is taken off under the reaction conditions. However, this has
no sequence since the phenol is not affected.) The mixture was car-
ried to the next step directly. To a solution of the substrate at 0 °C
was added THF (2 mL) and then 30% aq H2O2 (0.2 mL). The mix-
ture was stirred at r.t. for 45 min and then at 50 °C for 1–2 h. The
reaction mixture was diluted with EtOAc (20 mL) and then washed
with Na2SO3 thoroughly. After drying and concentration, the resi-
due was dissolved in THF (1 mL), and to this solution was added
TBAF (0.1 M, 0.3 mL). The reaction mixture was stirred at 50 °C
for 20 h (TLC visualization with UV showed only one bright spot).
To the mixture was added H2O (1 mL) and EtOAc (20 mL), the
aqueous layer was separated and the pH was adjusted to 6 with aq
NaHSO4 solution dropwise (~5 mL). The aqueous layer was then
extracted with EtOAc (20 mL) thoroughly. The combined organic
layer was washed with brine (5 mL), dried and concentrated. PTLC
(MeOH–CH2Cl2, 1:20) of the residue gave psymberin 38 (8.1 mg,
67% over three steps) as a light yellow glass.
13C NMR (CDCl3): d = 172.8, 170.8, 162.5, 160.9, 142.1, 140.3,
113.5, 101.5, 86.5, 81.3, 80.8, 80.0, 74.8, 73.0, 71.3, 57.9, 56.5,
42.8, 39.4, 37.5, 32.3, 32.2, 28.4, 23.0, 22.4, 12.7, 10.8, 9.9.
MS (FAB): m/z [M + Na]+ calcd for C31H47NaO11: 632.30; found
632.23.
Synthesis of Compounds 86 and 88
An oven-dried, sealed tube was charged with 4-phenylbutanamide
(28 mg, 0.172 mmol), CuI (5 mg, 0.028 mmol), and Cs2CO3 (60 mg,
0.18 mmol) sequentially. Under argon, dimethyl ethylenediamine
(0.007 mL, 0.057 mmol) was then added followed by a solution of
vinyl iodide 59 (60 mg, 0.057 mmol) in toluene (1.5 mL). The tube
was filled with argon and quickly capped and sealed. The reaction
was stirred vigorously at 70 °C overnight. After the mixture was
cooled to r.t., it was diluted with EtOAc (30 mL), filtered off a short
pad of silica gel and washed with EtOAc (20 mL). The combined
organic layer was concentrated and purified by PTLC (EtOAc–hex-
ane, 1:10 to 1:3) to give the enamide 82a (55 mg, 90%) as an oil.
The isomeric mixture of 82a was carried through the next step with-
out separation of the two isomers.
Major trans-Isomer
1H NMR (CDCl3): d = 7.33 (d, J = 10.5 Hz, 1 H), 7.28–7.15 (m, 5
H), 6.76 (dd, J = 10.5, 14.0 Hz, 1 H), 6.29 (s, 1 H), 5.17 (m, 1 H),
5.08 (m, 1 H), 4.97 (m, 1 H), 4.11 (dd, J = 2.5, 6.0, 12.5 Hz, 1 H),
3.44 (m, 1 H), 2.87 (dd, J = 2.0, 16.0 Hz, 1 H), 2.66–2.63 (m, 3 H),
2.24–2.20 (m, 2 H), 2.08 (br, 6 H), 2.00 (s, 3 H), 1.99–1.96 (m, 2 H),
1.35–1.26 (m, 6 H), 1.14–1.10 (m, 36 H), 0.89–0.85 (m, 15 H), 0.03
(br, 6 H).
[a]D20 +19.5 (c 0.2, MeOH).
1H NMR (CDCl3): d = 11.13 (s, 1 H), 7.22 (br, 1 H), 7.12 (d,
J = 10.0 Hz, 1 H), 6.31 (s, 1 H), 5.44 (dd, J = 9.0, 10.0 Hz, 1 H),
4.81 (s, 1 H), 4.80 (s, 1 H), 4.54 (ddd, J = 4.0, 4.0, 12.5 Hz, 1 H),
4.43 (dd, J = 2.5, 3.0 Hz, 1 H), 4.40 (s, 1 H), 4.17 (br, 1 H), 3.94
(dm, J = 8.5 Hz, 1 H), 3.89 (m, 1 H), 3.75 (m, 1 H), 3.66 (dd,
J = 4.0, 10.5 Hz, 1 H), 3.53 (d, J = 11.0 Hz, 1 H), 3.38 (s, 3 H), 3.37
(s, 3 H), 2.87 (dd, J = 3.0, 16.0 Hz, 1 H), 2.80 (dd, J = 12.5, 16.5 Hz,
1 H), 2.38 (dd, J = 9.0, 14.5 Hz, 1 H), 2.18 (dd, J = 4.5, 14.5 Hz, 1
H), 2.06 (m, 1 H), 2.01 (s, 3 H), 1.88 (m, 1 H), 1.80 (m, 1 H), 1.75
(s, 3 H), 1.62 (m, 1 H), 1.09 (d, J = 7.5 Hz, 3 H), 0.96 (s, 3 H), 0.91
(s, 3 H).
MS (ES): m/z [M + H]+ calcd for C60H102NO10Si3: 1080.68; found
1081.17.
Compound 82a (40 mg, 0.037 mmol) was treated with a NaOMe so-
lution in MeOH (0.08 M, 1 mL) at r.t. for 5 h before quenching with
H2O (2 mL). The mixture was then extracted with EtOAc (30 mL)
thoroughly. The organic layer was washed with brine (10 mL),
dried over Na2SO4 and concentrated. The crude product was dis-
solved in CH2Cl2 (1 mL). At 0 °C, pyridine (30 mL), DMAP (1.0
mg) and then Ac2O (15 mL) were added sequentially. The reaction
was allowed to stir at 0 °C for 1 h before quenching with aq NH4Cl
(2 mL). The aqueous layer was extracted with EtOAc (30 mL). The
combined organic layer was washed with NH4Cl (10 mL) and brine
(10 mL). After drying and concentration, the residue was purified
by PTLC (EtOAc–hexane, 1:10 to 2:3) to give 82 as a mixture (40
mg) of two isomers (E/Z). The mixture was carried on to the next
step directly.
13C NMR (CDCl3) d = 173.8, 170.7, 162.5, 161.4, 142.2, 139.9,
113.6, 113.3, 101.7, 101.5, 82.2, 80.7, 79.6, 78.6, 74.0, 73.3, 71.6,
58.2, 56.5, 42.8, 39.0, 37.8, 32.3, 29.9, 28.6, 23.3, 22.9, 14.0, 10.7,
9.7.
MS (FAB): m/z [M + Na]+ calcd for C31H47Na O11: 632.30; found
632.23.
The same procedure as for the synthesis of psymberin 38 was per-
formed with compound epi-81 in order to obtain epi-38. The only
difference is that treatment with TBAF cannot remove the acetate
on the secondary alcohol in this case. After removal of the silyl
group by TBAF, the residue was finally dissolved in MeOH (1 mL),
and aq LiOH (1 N, 0.2 mL) was added. The reaction mixture was
stirred at r.t. overnight. To this solution was added H2O (2 mL) and
MS (ES): m/z [M + H]+ calcd for C49H81NO9Si2: 882.54; found
882.63.
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

2870
N. Shao et al.
PAPER
Under an atmosphere of argon, to a solution of compound 82 (40
mg, 0.0476 mmol) in HFI (0.5 mL) at 0 °C was added MeOH (0.04
mL) followed by dropwise addition of a solution of PhI(OAc)2 (32
mg, 0.096 mmol) in HFI (0.5 mL). The solution was stirred at r.t.
for 20 h. The reaction mixture was diluted with EtOAc (30 mL) and
filtered through a short pad of silica gel. The filtrate was concentrat-
ed and then purified by FC (acetone–hexane, 1:10 to 1:2) to give the
crude product which was treated with TBAF (0.1 M, 4 mL) at 50 °C
overnight. The residue was carefully purified with PTLC (MeOH–
CH2Cl2, 1:20) to yield the desired compound 86 (12 mg, 50%) and
the epi-isomer 88 (4 mg, 20%).
Compound 89
1H NMR (CDCl3): d = 11.28 (br, 1 H), 7.42 (d, J = 10.0 Hz, 1 H),
6.34 (s, 1 H), 5.11 (dd, J = 3.0, 10.0 Hz, 1 H), 4.59 (m, 1 H), 4.27
(d, J = 5.5 Hz, 1 H), 4.18–4.13 (m, 2 H), 3.84 (m, 1 H), 3.77 (m, 1
H), 3.52 (m, 1 H), 3.42 (s, 3 H), 3.34 (s, 3 H), 2.98–2.89 (m, 4 H),
2.09 (s, 3 H), 1.99–1.46 (m, 5 H), 1.17 (d, J = 7.0 Hz, 3 H), 1.00 (s,
3 H), 0.92 (s, 3 H).
13C NMR (CDCl3): d = 173.4, 162.7, 138.3, 130.1, 128.7, 126.8,
101.7, 86.6, 83.7, 81.7, 80.2, 77.2, 75.0, 73.1, 71.5, 58.4, 56.7, 43.0,
39.6, 35.9, 32.5, 28.6, 22.6, 12.9, 10.9, 10.0, 9.9
MS (ES): m/z [M + Na]+ calcd for C34H47NNaO11: 668.30; found
668.33.
Compound 86
[a]D20 +18.24 (c 0.1, MeOH).
Synthesis of C11-Deoxypsymberin 91 and Its Epimers
1H NMR (CDCl3): d = 11.24 (s, 1 H), 7.30–7.12 (m, 5 H), 6.29 (s, 1
H), 6.07 (br, 1 H), 5.85 (d, J = 11.0 Hz, 1 H), 5.35 (dd, J = 6.0, 9.5
Hz, 1 H), 4.52 (m, 1 H), 3.97 (m, 1 H), 3.91 (d, J = 10.0 Hz, 1 H),
3.85 (s, 1 H), 3.64 (dd, J = 3.5, 8.0 Hz, 1 H), 3.59 (d, J = 11.0 Hz, 1
H), 3.35 (s, 3 H), 2.98 (dd, J = 3.0, 11.5 Hz, 1 H), 2.83 (dd, J = 12.5,
16.5 Hz, 1 H), 2.62 (t, J = 7.0 Hz, 2 H), 2.38–2.22 (m, 2 H), 2.04 (s,
3 H), 2.01–1.55 (m, 7 H), 1.11 (d, J = 7.0 Hz, 3 H), 0.99 (s, 3 H),
0.83 (s, 3 H).
An oven-dried, sealed tube was charged amide 76 (410 mg, 0.81
mmol), CuI (26 mg, 0.135 mmol), and Cs2CO3 (264 mg, 0.81
mmol) sequentially. Under argon, dimethyl ethylenediamine (0.029
ml, 0.27 mmol) was then added followed by a solution of vinyl io-
dide 93 (260 mg, 0.27 mmol) in toluene (5 mL). The tube was filled
with argon and quickly capped and sealed. The reaction was stirred
vigorously at 70 °C for 20 h. After the mixture was cooled to r.t., it
was diluted with EtOAc (20 mL), filtered off a short pad of silica
gel and washed with EtOAc (80 mL). The combined organic layer
was concentrated and purified by FC (acetone–hexane, 1:20 to
1:10) to give enamide 94a (334 mg, 95%) as an oil. The excess
amide was recovered. The mixture was carried through to the next
step without separation of the isomers.
13C NMR (CDCl3): d = 174.5, 170.9, 162.7, 161.1, 141.6, 140.3,
128.9, 128.8, 126.5, 113.5, 102.4, 101.7, 82.9, 80.3, 79.9, 73.3,
72.2, 60.9, 56.6, 43.3, 38.6, 36.5, 32.8, 30.4, 28.8, 27.4, 14.6, 10.9,
10.1.
LCMS: m/z [M + H]+ calcd for C33H46NO9: 600.3; found 600.3.
MS (ES): m/z [M + Na]+ calcd for C75H115NNaO12Si3: 1328.76;
found 1328.77.
Compound epi-88
[a]D20 +25.10 (c 0.1, MeOH).
Compound 94a (265 mg, 0.20 mmol) was treated with a NaOMe so-
lution in MeOH (10 mL, 0.06 M) at r.t. for 5 h before quenching
with H2O (5 mL). The mixture was then extracted with EtOAc (50
mL) thoroughly. The organic layer was washed with brine (10 mL),
dried over Na2SO4 and concentrated. The crude product was dis-
solved in CH2Cl2 (8 mL). At 0 °C, pyridine (500 mL), DMAP (4.0
mg) and then Ac2O (250 mL) were added sequentially. The reaction
was allowed to stir at 0 °C for 1 h before quenching with aq NH4Cl
(2 mL). The aqueous layer was extracted with EtOAc (30 mL). The
combined organic layer was washed with NH4Cl (10 mL) and brine
(10 mL). After drying and concentration, the residue was purified
by FC (acetone–hexane, 1:20 to 1:6) to give 94 as a mixture (210
mg, 95%) of four isomers. The mixture was carried on to the next
reaction directly.
1H NMR (CDCl3): d = 11.21 (s, 1 H), 7.30–7.15 (m, 5 H), 6.31 (s, 1
H), 6.01 (d, J = 10.0 Hz, 1 H), 5.35 (br, 1 H), 5.11 (d, J = 2.5, 10.0
Hz, 1 H), 4.57 (dt, J = 4.5, 11.5 Hz, 1 H), 4.20 (m, 1 H), 4.09 (d,
J = 10.5 Hz, 1 H), 3.76 (d, J = 12.0 Hz, 1 H), 3.49 (dd, J = 5.0, 11.5
Hz, 1 H), 3.31 (s, 3 H), 2.94–2.85 (m, 2 H), 2.65 (t, J = 8.0 Hz, 2 H),
2.28 (t, J = 8.0 Hz, 2 H), 2.02 (s, 3 H), 2.00–1.94 (m, 2 H), 1.78–
1.59 (m, 5 H), 1.13 (d, J = 7.0 Hz, 3 H), 0.96 (s, 3 H), 0.86 (s, 3 H).
13C NMR (CDCl3): d = 174.2, 171.0, 162.6, 161.4, 141.7, 140.4,
128.9, 126.5, 113.7, 102.1, 101.7, 86.8, 81.4, 79.9, 75.0, 73.4, 56.5,
43.0, 39.6, 36.3, 35.6, 32.5, 32.3, 30.1, 28.7, 27.4, 22.6, 13.1, 11.0,
10.1.
LCMS: m/z [M + H]+ calcd for C33H46NO9: 600.3; found 600.3.
MS (ES): m/z [M + H]+ calcd for C64H94NO11Si2: 1108.64; found
1108.63.
Synthesis of Compounds 87 and 89
87 and 89 were prepared from 59 in the same way as described in
the synthesis of 86 and 88 is an overall combined 66% yield over
two steps (33% each with the isomers separated at the last step).
MS (ES): m/z [M + Na]+ calcd for C64H93NNaO11Si2: 1130.62;
found 1130.68.
Under an atmosphere of argon, to a solution of compound 94 (420
mg, 0.38 mmol) in HFI (4 mL) at 0 °C was added MeOH (0.46 mL)
followed by dropwise addition of a solution of PhI(OAc)2 (498 mg,
1.5 mmol) in HFI (2 mL). The solution was stirred at r.t. for 40 h.
The reaction mixture was diluted with (30 mL) and filtered through
a short pad of silica gel. The filtrate was concentrated and then pu-
rified by FC (EtOAc–toluene, 1:10 to 1:2) to give two crude frac-
tions, A (220 mg) and B (183 mg), which contained 95a, epi-95a,
95b, and epi-95b. These two fractions of mixtures were carried on
separately in the same way with following steps to give the final
four products with diastereomers separated at on-going steps. All
the stereochemistry was assigned according to the NMR spectra of
natural psymberin and epi-psymberin.
Compound 87
[a]D20 +21.32 (c 0.1, CH2Cl2).
1H NMR (CDCl3): d = 11.13 (s, 1 H), 7.10 (d, J = 10.0 Hz, 1 H),
6.30 (s, 1 H), 5.48 (dd, J = 8.5, 10.0 Hz, 1 H), 4.52 (ddd, J = 4.0, 4.5,
12.0 Hz, 1 H), 4.39 (br, 1 H), 4.37 (s, 1 H), 4.34 (d, J = 2.5 Hz, 1 H),
3.93 (m, 1 H), 3.89 (m, 1 H), 3.76 (m, 1 H), 3.68 (m, 1 H), 3.55 (d,
J = 10.0 Hz, 1 H), 3.39 (s, 3 H), 3.26 (s, 3 H), 2.91 (dd, J = 8.0, 14.0
Hz, 1 H), 2.86–2.78 (m, 3 H), 2.08 (m, 2 H), 2.01 (s, 3 H), 1.86–1.75
(m, 3 H), 1.09 (d, J = 7.0 Hz, 3 H), 0.99 (s, 3 H), 0.92 (s, 3 H).
13C NMR (CDCl3): d = 174.1, 170.9, 162.8, 161.3, 140.1, 138.7,
129.9, 128.7, 126.7, 113.6, 102.1, 101.8, 83.8, 82.3, 79.9, 78.7,
74.4, 73.6, 73.4, 71.8, 60.8, 58.7, 56.7, 43.1, 39.2, 36.3, 32.6, 30.2,
28.9, 23.5, 21.5, 14.6, 14.0, 10.9, 9.7.
To a solution of fraction A (containing compounds 95a and epi-95a,
0.038 mmol each) in CH2Cl2 (1 mL) was added pyridine (0.30 mL),
Ac2O (0.15 mL) and DMAP (2.5 mg, 0.02 mmol) sequentially. The
reaction mixture was stirred at r.t. overnight before quenching with
LCMS: m/z [M + H]+ calcd for C34H48NO11: 646.3; found 646.4.
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

PAPER
Total Synthesis of Psymberin
2871
aq NH4Cl (5 mL). The aqueous layer was extracted with EtOAc (30
mL) thoroughly. The combined organic layer was washed with
brine (10 mL), dried over Na2SO4 and concentrated. The residue
was purified by FC (EtOAc–toluene, 1:4), and then directly used in
the next step.
17.0 Hz, 1 H), 2.94 (dd, J = 12.0, 16.5 Hz, 1 H), 2.40 (dd, J = 9.0,
15.0 Hz, 1 H), 2.25 (dd, J = 3.5, 15.0 Hz, 1 H), 2.07 (s, 3 H), 1.83
(s, 3 H), 2.00 (m, 1 H), 1.80–1.20 (m, 6 H), 1.18 (d, J = 7.0 Hz, 3
H), 0.99 (s, 3 H), 0.92 (s, 3 H).
13C NMR (CDCl3): d = 173.0, 162.6, 142.5, 140.4, 113.6, 101.6,
88.2, 81.7, 81.0, 80.2, 79.1, 73.4, 71.5, 58.1, 56.6, 43.0, 38.4, 37.7,
33.1, 32.9, 28.7, 27.5, 23.8, 23.2, 19.3, 10.9, 10.0.
To a solution of the acetate compound (0.015 mmol) in MeOH (2
mL) was charged Pd/C (5 mg), and the flask was sealed with a rub-
ber stopper. The inner atmosphere was exchanged three times with
H2 before the solution was allowed to stir under H2 (double layer H2
balloon) overnight. The mixture was filtered off through a celite
pad followed by washing with MeOH (2 ꢀ 10 mL). The combined
organic layer was concentrated and purified by FC (EtOAc–toluene,
1:10 to 1:2) to give the alcohols 95 and epi-95. Isolated pure isomers
were converted into the final products via the following steps.
LCMS: m/z [M + Na]+ calcd for C31H47NNaO10: 616.3; found
616.3.
Compound 91b
Prepared from fraction B containing 95b and epi-95b following the
same procedure as for the preparation of 91 from fraction A contain-
ing 95a and epi-95a.
Substrate 95 (0.02 mmol) was azeotroped with toluene (3 × 10 mL),
dissolved in anhyd THF (1 mL) and charged with o-nitrophenyl se-
lenocyanate (18 mg, 0.08 mmol) followed by Bu3P (0.025 mL, 0.1
mmol). The mixture should turn to dark red immediately. The solu-
tion was allowed to stir at r.t. for 2 h. The reaction was quenched by
adding MeOH (0.5 mL), and the mixture was stirred for additional
15 min before it was concentrated. The residue was purified by FC
(acetone–hexane, 1:1) to give the desired selenated product. (Partial
deprotection of phenol acetate and TIPS group was observed. This
did not affect the subsequent reactions.)
[a]D20 +6.11 (c 0.2, CH2Cl2).
1H NMR (CDCl3): d = 11.17 (br, 1 H), 7.11 (d, J = 10.0 Hz, 1 H),
6.32 (s, 1 H), 5.34 (dd, J = 8.0, 9.5 Hz, 1 H), 4.83 (s, 1 H), 4.78 (s,
1 H), 4.59 (m, 1 H), 4.44 (d, J = 4.0 Hz, 1 H), 4.31 (br, 1 H), 4.10
(d, J = 9.5 Hz, 1 H), 3.77–3.72 (m, 2 H), 3.63 (d, J = 10.0 Hz, 1 H),
3.42 (s, 3 H), 3.41 (s, 3 H), 3.38–3.35 (m, 1 H), 2.97 (dd, J = 3.0,
16.5 Hz, 1 H), 2.85 (dd, J = 12.5, 17.0 Hz, 1 H), 2.28 (dd, J = 9.5,
14.5 Hz, 1 H), 2.06 (d, J = 14.5 Hz, 1 H), 2.00 (s, 3 H), 1.95–1.44
(m, 7 H), 1.75 (s, 3 H), 1.15 (d, J = 7.0 Hz, 3 H), 0.99 (s, 3 H), 0.91
(s, 3 H).
To a solution of the selenide at 0 °C was added THF (2 mL) and
then 30% aq H2O2 (0.2 mL). The mixture was stirred at r.t. for 45
min and then at 50 °C for 1–2 h. The reaction mixture was diluted
with EtOAc (20 mL) and then washed with a solution of Na2SO3
(2 ꢀ 5 mL) thoroughly. After drying and concentration, the residue
was dissolved in THF (1 mL), and TBAF (0.1 M, 0.3 mL) was add-
ed. The reaction mixture was stirred at 50 °C for 20 h (TLC visual-
ization with UV showed only one bright spot). To the mixture was
added H2O (1 mL) and EtOAc (20 mL), the aqueous layer was sep-
arated and the pH was adjusted to 6 with an aq NaHSO4 solution
dropwise (~5 mL). The aqueous layer was then extracted with
EtOAc (20 mL) thoroughly. The combined organic layer was
washed with brine (5 mL), dried and concentrated. PTLC (MeOH–
CH2Cl2, 1:20) of the residue gave deoxypsymberin 91 as a light yel-
low glass (7.0 mg, 60% over three steps).
13C NMR (CDCl3): d = 172.6, 171.0, 162.6, 161.6, 142.3, 140.2,
113.7, 113.6, 102.0, 101.7, 82.9, 81.0, 80.8, 80.7, 73.6, 72.7, 71.5,
57.9, 56.7, 43.0, 37.4, 33.5, 32.8, 31.8, 30.1, 28.7, 27.5, 23.2, 23.1,
22.3, 10.8, 9.8.
LCMS: m/z [M + H]+ calcd for C31H48NO10: 594.3; found 594.3.
Compound 91c
Prepared from fraction B containing 95b and epi-95b following the
same procedure as for the preparation of 91 from fraction A contain-
ing 95a and epi-95a.
[a]D20 +21.03 (c 0.1, CH2Cl2).
1H NMR (CDCl3): d = 11.17 (br, 1 H), 7.10 (d, J = 10.0 Hz, 1 H),
6.31 (s, 1 H), 5.09 (dd, J = 6.0, 10.0 Hz, 1 H), 4.81 (s, 1 H), 4.77 (s,
1 H), 4.55 (m, 1 H), 4.42 (d, J = 3.5 Hz, 1 H), 4.31 (br, 1 H), 4.14
(m, 1 H), 3.76 (ddd, J = 3.5, 4.0, 9.5 Hz, 1 H), 3.46 (m, 1 H), 3.41
(m, 1 H), 3.40 (s, 3 H), 3.35 (s, 3 H), 2.93 (dd, J = 3.5, 17.0 Hz, 1
H), 2.86 (dd, J = 7.5, 16.0 Hz, 1 H), 2.32 (dd, J = 9.5, 14.5 Hz, 1 H),
2.13 (dd, J = 3.5, 14.5 Hz, 1 H), 1.96 (s, 3 H), 1.94–1.37 (m, 7 H),
1.74 (s, 3 H), 1.12 (d, J = 7.0 Hz, 3 H), 0.95 (s, 3 H), 0.85 (s, 3 H).
Compound 91
[a]D20 +10.23 (c 0.2, CH2Cl2).
1H NMR (CDCl3): d = 11.11 (br, 1 H), 6.32 (s, 1 H), 5.44 (dd,
J = 7.5, 10.0 Hz, 1 H), 4.80 (br, 2 H), 4.50 (dt, J = 4.0, 12.0 Hz, 1
H), 4.45 (d, J = 3.0 Hz, 1 H), 3.99 (d, J = 9.5 Hz, 1 H), 3.78 (m, 2
H), 3.59 (d, J = 10.5 Hz, 1 H), 3.39 (s, 3 H), 3.36 (s, 3 H), 2.84 (dd,
J = 3.5, 16.5 Hz, 1 H), 2.78 (dd, J = 12.0, 16.5 Hz, 1 H), 2.39 (dd,
J = 9.0, 14.5 Hz, 1 H), 2.18 (dd, J = 3.5, 14.5 Hz, 1 H), 1.98 (s, 3 H),
1.75 (s, 3 H), 1.85 (m, 2 H), 1.72 (m, 2 H), 1.57–1.41 (m, 3 H), 1.09
(d, J = 7.0 Hz, 3 H), 0.93 (s, 6 H).
13C NMR (CDCl3): d = 172.4, 171.2, 162.6, 161.7, 142.3, 140.3,
113.9, 113.5, 101.9, 101.6, 88.1, 82.8, 80.7, 80.5, 79.6, 72.9, 71.7,
57.9, 57.0, 42.9, 38.8, 37.3, 33.4, 33.1, 28.6, 27.5, 23.4, 23.1, 19.5,
10.8, 9.8.
LCMS: m/z [M + H]+ calcd for C31H48NO10: 594.3; found 594.3.
13C NMR (CDCl3) d = 174.1, 171.0, 162.7, 162.0, 142.5, 140.0,
113.9, 113.4, 101.7, 101.6, 83.1, 81.1, 80.3, 78.8, 74.4, 73.4, 73.2,
60.9, 58.2, 56.7, 43.1, 37.9, 33.5, 32.8, 32.5, 28.8, 27.7, 23.1, 22.0,
21.6, 14.6, 10.9, 9.6.
Synthesis of C11-Deoxypsymberin 96 and epi-96
Compounds 96 and epi-96 were prepared in a combined 56% yield
over 5 steps (28% each with isomer separation at the last step) from
100 and 95 in the same way as described for the synthesis of 86 and
88.
LCMS: m/z [M + H]+ calcd for C31H48NO10: 594.3; found 594.3.
Compound 91a
Prepared from epi-95 following the same procedure as for the prep-
aration of 91 from 95.
Compound 96
1H NMR (CDCl3): d = 11.13 (br, 1 H), 7.27–7.16 (m, 5 H), 6.31 (s,
1 H), 5.48 (dd, J = 7.5, 10.0 Hz, 1 H), 4.50 (dt, J = 4.0, 11.5 Hz, 1
H), 4.38 (d, J = 3.0 Hz, 1 H), 3.98 (d, J = 11.0 Hz, 1 H), 3.79 (m, 2
H), 3.59 (d, J = 10.0 Hz, 1 H), 3.38 (s, 3 H), 3.27 (s, 3 H), 2.91 (dd,
J = 8.5, 14.0 Hz, 1 H), 2.88–2.75 (m, 3 H), 1.97 (s, 3 H), 1.87–1.80
(m, 2 H), 1.75–1.68 (m, 2 H), 1.58–1.43 (m, 3 H), 1.09 (d, J = 7.5
Hz, 3 H), 0.94 (s, 3 H), 0.93 (s, 3 H).
[a]D20 +22.66 (c 0.1, CH2Cl2).
1H NMR (CDCl3): d = 11.24 (br, 1 H), 7.43 (d, J = 10.0 Hz, 1 H),
6.27 (s, 1 H), 5.02 (dd, J = 2.5, 10.0 Hz, 1 H), 4.83 (s, 1 H), 4.81 (s,
1 H), 4.55 (m, 1 H), 4.37 (d, J = 4.5 Hz, 1 H), 4.18 (m, 1 H), 3.79
(m, 2 H), 3.49 (s, 3 H), 3.46 (m, 1 H), 3.40 (s, 3 H), 3.00 (dd, J = 4.0,
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York

2872
N. Shao et al.
PAPER
13C NMR (CDCl3) d = 174.0, 171.0, 162.7, 161.8, 140.0, 138.9,
138.8, 130.0, 129.9, 129.8, 128.7, 128.6, 126.7, 113.8, 101.7, 83.8,
83.1, 80.1, 78.8, 74.2, 73.4, 73.3, 58.7, 58.2, 56.7, 43.1, 36.2, 33.5,
32.8, 32.3, 28.8, 27.7, 22.0, 21.7, 10.9, 9.7.
(6) Jiang, L.; Job, G. E.; Klapars, A.; Buchwald, S. L. Org. Lett.
2003, 5, 3667.
(7) Shen, R.; Lin, C.-T.; Porco, J. A. Jr. J. Am. Chem. Soc. 2002,
124, 5650.
(8) (a) Stang, P. J.; Zhdankin, V. V. Chem. Rev. 1996, 96, 1123
and references therein. (b) Moriarty, R. M.; Chany, C. J. II;
Kosmeder, J. W. II; Du Bois, J. e-EROS Encyclopedia of
Reagents for Organic Synthesis [Online]; Paquette, L. A.;
Crich, D.; Fuchs, P. L.; Molander, G., Eds.; John Wiley &
Sons Ltd.: Hoboken, 2006, and references therein.
(9) (a) Moriarty, R. M.; Kosmeder, J. W. II. e-EROS
Encyclopedia of Reagents for Organic Synthesis [Online];
Paquette, L. A.; Crich, D.; Fuchs, P. L.; Molander, G., Eds.;
John Wiley & Sons Ltd.: Hoboken, 2001. (b) Correa, A.;
Tellitu, I.; Dominguez, E.; SanMartin, R. J. Org. Chem.
2006, 71, 8316.
(10) Baran, P. S.; Richter, J. M.; Lin, D. W. Angew. Chem. Int.
Ed. 2005, 44, 609 and references therein; Angew. Chem.
2005, 117, 615.
(11) (a) Cichewicz, R. H.; Valeriote, F. A.; Crews, P. Org. Lett.
2004, 6, 1951; and references cited therein. (b) Pettit, G. R.;
Xu, J. P.; Chapuis, J. C.; Pettit, R. K.; Tackett, L. P.; Doubek,
D. L.; Hooper, J. N. A.; Schmidt, J. M. J. Med. Chem. 2004,
47, 1149.
LCMS: m/z [M + H]+ calcd for C34H48NO10: 630.3; found 630.3.
Compound epi-96
1H NMR (CDCl3): d = 11.25 (br, 1 H), 7.44 (d, J = 10.0 Hz, 1 H),
7.29–7.19 (m, 5 H), 6.28 (s, 1 H), 5.05 (dd, J = 2.5, 10.0 Hz, 1 H),
4.53 (ddd, J = 3.5, 5.0, 12.5 Hz, 1 H), 4.26 (d, J = 5.0 Hz, 1 H), 4.13
(m, 1 H), 3.73 (m, 2 H), 3.40 (d, J = 10.5, Hz, 1 H), 3.35 (s, 3 H),
3.28 (s, 3 H), 2.93–2.81 (m, 4 H), 1.98 (s, 3 H), 1.94 (m, 1 H), 1.69–
1.43 (m, 6 H), 1.11 (d, J = 7.5 Hz, 3 H), 0.92 (s, 3 H), 0.85 (s, 3 H).
13C NMR (CDCl3): d = 173.4, 171.1, 162.5, 161.5, 140.3, 138.5,
130.1, 128.7, 126.8, 113.8, 102.0, 101.6, 88.3, 83.7, 81.8, 80.0,
79.1, 73.6, 71.5, 58.5, 56.6, 43.0, 38.3, 35.9, 33.1, 32.8, 28.7, 27.5,
23.8, 19.4, 10.9, 10.0.
LCMS: m/z [M + Na]+ calcd for C34H47NNaO10: 652.3; found
652.4.
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synthesis.
(12) Jiang, X.; Garcia-Fortanet, J.; De Brabander, J. K. J. Am.
Chem. Soc. 2005, 127, 11254; and references cited therein.
(13) Ning, S.; Kiren, S.; Williams, L. J. Org. Lett. 2007, 9, 1093.
(14) Smith, A. B. III.; Jurica, J. A.; Walsh, S. P. Org. Lett. 2008,
10, 5625.
Acknowledgment
We thank Drs. Ismail Kola and Robert Bishop for their support of
the SPRI postdoctoral program.
(15) Huang, X.; Shao, N.; Palani, A.; Aslanian, R.; Buevich, A.
Org. Lett. 2007, 9, 2597.
(16) Huang, X.; Shao, N.; Palani, A.; Aslanian, R.; Buevich, A.;
Seidel-Dugan, C.; Huryk, R. Tetrahedron Lett. 2008, 49,
3592.
(17) Parmee, E. R.; Tempkin, O.; Masamune, S. J. Am. Chem.
Soc. 1991, 113, 9365.
(18) Mulzer, J.; Mantoulidis, A.; Ohler, E. J. Org. Chem. 2000,
65, 7456.
(19) Evans, D. A.; Hoveyda, A. H. J. Org. Chem. 1990, 55, 5190.
(20) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986,
108, 7408.
References
(1) Simmons, T. L.; Andrianasolo, E.; McPhail, K.; Flatt, P.;
Gerwick, W. H. Mol. Cancer Ther. 2005, 4, 333.
(2) Narquizian, R.; Kocienski, P. J. The Pederin Family of
Antitumor Agents: Structures, Synthesis and Biological
Activity, In The Role of Natural Products in Drug Discovery;
Mulzer, J.; Bohlmann, R., Eds.; Springer: New York, 2000,
25–56; Ernst Schering Research Foundation Workshop 32.
(3) Ding, F.; Jennings, M. P. J. Org. Chem. 2008, 73, 5965.
(4) Coleman, R. S.; Walczak, M. C.; Campbell, E. L. J. Am.
Chem. Soc. 2005, 127, 16038.
(21) Grieco, P. A.; Takigawa, T.; Schillinger, W. J. J. Org. Chem.
1980, 45, 2247.
(5) (a) Smith, A. B. III; Safonov, I. G.; Corbett, R. M. J. Am.
Chem. Soc. 2002, 124, 11102. (b) Kagawa, N.; Ihara, M.;
Toyota, M. Org. Lett. 2006, 8, 875. (c) Petri, A. F.; Bayer,
A.; Maier, M. E. Angew. Chem. Int. Ed. 2004, 43, 5821;
Angew. Chem. 2004, 116, 5945. (d) Troast, D. M.; Porco,
J. A. Jr. Org. Lett. 2002, 4, 991 and references therein.
(e) Wan, S.; Green, M. E.; Park, J.-H.; Floreancig, P. E. Org.
Lett. 2007, 9, 5385 and references therein. (f) Kiren, S.;
Ning, S.; Williams, L. J. Tetrahedron Lett. 2007, 48, 7456.
(22) Maffioli, S. I.; Marzorati, E.; Marazzi, A. Org. Lett. 2005, 7,
5237.
(23) Huang, X.; Shao, N.; Huryk, R.; Palani, A.; Aslanian, R.;
Seidel-Dugan, C. Org. Lett. 2009, 11, 867.
Synthesis 2009, No. 17, 2855–2872 © Thieme Stuttgart · New York