Asymmetric total synthesis of fredericamycin a: An intramolecular cycloaddition pathway

Article abstract of DOI:10.1002/chem.200500443

The asymmetric total synthesis of the potent antitumor antibiotic fredericamycin A ((S)-1) was achieved by the intramolecular [4+2] cycloaddition of the silylene-protected styrene derivative (S)-7 followed by the aromatic Pummerer-type reaction of the sul

Full text of DOI:10.1002/chem.200500443

DOI: 10.1002/chem.200500443
Asymmetric Total Synthesis of Fredericamycin A: An Intramolecular
Cycloaddition Pathway
Shuji Akai, Toshiaki Tsujino, Nobuhisa Fukuda, Kiyosei Iio, Yoshifumi Takeda,
Ken-ichi Kawaguchi, Tadaatsu Naka, Kazuhiro Higuchi, Emi Akiyama,
Hiromichi Fujioka, and Yasuyuki Kita*[a]
Abstract: The asymmetric total synthe-
sis of the potent antitumor antibiotic
fredericamycin A ((S)-1) was achieved
by the intramolecular [4+2] cycloaddi-
tion of the silylene-protected styrene
derivative (S)-7 followed by the aro-
matic Pummerer-type reaction of the
sulfoxide (S)-5. Although we had al-
ready succeeded in the total synthesis
of racemic 1 by the same approach,
synthesis of its asymmetric version was
more complicated than we had expect-
ed due to the difficulties involved in
constructing the quaternary carbon
center and the tendency of this center
to undergo facile racemization. Race-
mization of this center during the in-
stallation of the acetylene moiety on
the dione (R)-8 was the most serious
aspect. Systematic studies of its DE-
ring analogue (R)-25 revealed that rac-
emization of the quaternary carbon
center proceeded by a retro-aldol–aldol
reaction of the initial adduct, (1R)-39a-
Li, and that the degree of racemization
was dependent on the reaction temper-
ature. The racemization process could
be completely depressed by keeping
the reaction temperature at ꢀ788C.
The construction of the stereogenic
quaternary carbon center was achieved
by the lipase-catalyzed desymmetriza-
tion of the prochiral 1,3-diol 9a bearing
the DEF-ring moiety. These studies en-
abled us to attain the asymmetric total
synthesis of (S)-1 while completely re-
taining the chiral integrity created by
the enzymatic reactions.
Keywords: antitumor
agents
·
asymmetric synthesis · enzymatic
desymmetrization · intramolecular
cycloaddition
·
Pummerer-type
reaction · racemization
Introduction
that 1 inhibits both topoisomerases I and II.[2d] Another im-
portant characteristic of 1 is its unique and complicated
structure, which consists of two sets of peri-hydroxy tricyclic
aromatic moieties connected through a stereogenic spiro
quaternary carbon atom, which is made chiral by the pres-
ence of a single methoxy group at the farthest position on
the A-ring. However, its absolute configuration could not be
determined by either X-ray crystallography or spectroscopic
studies (Figure 1).
Fredericamycin A (1) was isolated from a new strain of
Streptomyces griseus at the Frederick Cancer Research
Center in Frederick, MD, in 1981. It exhibits potent antitu-
mor activity against several in vivo tumor models, such as
P388 leukemia, B16 melanoma, and CD8F mammary, and
does not show mutagenicity in the Ames test.[1,2] It has been
suggested that the biological properties of 1 originate from
its inhibition of RNA and protein syntheses, preferentially
to DNA synthesis, and later studies have also demonstrated
Due to its prospective biological activities, the unusual
spiro structure, and the difficulty in determining its absolute
stereochemistry, numerous synthetic studies on the total syn-
[a] Dr. S. Akai, Dr. T. Tsujino, N. Fukuda, Dr. K. Iio, Dr. Y. Takeda,
K. Kawaguchi, Dr. T. Naka, Dr. K. Higuchi, E. Akiyama,
Dr. H. Fujioka, Prof. Dr. Y. Kita
Graduate School of Pharmaceutical Sciences
Osaka University, 1–6, Yamadaoka
Suita, Osaka 565–0871 (Japan)
Fax: ( +81)6-6879-8229
E-mail: kita@phs.osaka-u.ac.jp
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Figure 1. Structure of fredericamycin A (1).
6286
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6286 – 6297

FULL PAPER
thesis of 1 and its derivatives have been conducted. These
studies have, thus far, resulted in five total syntheses of race-
mic 1,[3–7] and the first synthesis of the optically pure natural
1.[8,9] The last accomplishment by Bogerꢁs group in this area
required the optical resolution of a fully protected racemic
compound by using a chiral HPLC column at the final stage,
and therefore the absolute configuration of 1 was not deter-
mined by this group. The majority of these total syntheses,
and the related model studies, involved the construction of
the spiro C- or D-ring during the final stage of the synthesis;
however, the lack of sufficient methods that can distinguish
between the enantiotopic faces of the highly symmetrical
ABC-plane has been the most probable obstacle in the ap-
plication of these methods to asymmetric synthesis. To date,
no group besides our own (vide infra),[10–13] has succeeded in
the asymmetric construction of the optically active spiro
system of 1, and even the methods that could be potentially
applicable to the asymmetric synthesis of 1 are few.[9h,14]
Boger and his colleagues reported that the L1210 cytotox-
ic activities of 1 and its enantiomer were similar, and that a
racemic ABCDE-ring analogue was less potent.[8a] However,
the biological activities of both enantiomers of partial struc-
tures or analogues of 1 has not been evaluated.
Recently, we achieved[10,11] the asymmetric total synthesis
of natural 1, by the regioselective intermolecular [4+2] cy-
cloaddition[15,16] of the dienolate anion 2 (the AB-ring unit)
to the optically pure CDEF-ring unit (S)-3, and thereby de-
termined the absolute stereochemistry of the natural prod-
uct as S for the first time. The stereogenic quaternary
carbon center of (S)-3 was created by the stereospecific re-
arrangement of the optically pure benzofuzed trans-epoxy
acylate 4.[17,18] A similar reaction of (S)-3 with the regioiso-
mer of 2, which has a methoxy group at a different position
on the A-ring, produced the enantiomer of natural 1.
At the same time, we also completed the total synthesis
of racemic 1 by the second route shown in Scheme 1.[12a] The
key steps in this route include the intramolecular cycloaddi-
tion reaction[12b–e] of the silylene-protected styrene derivative
(ꢁ)-7 to construct the hexacyclic carbon framework found
in (ꢁ)-6, and the aromatic Pummerer-type reaction[19] of the
sulfoxide (ꢁ)-5, which acts to substitute the B-ring sulfinyl
group with a hydroxyl group.
Provided we could prepare the pivotal dione intermediate
(R)-8 with a high optical purity, the latter route was believed
to be the best route towards a asymmetric total synthesis of
(S)-1. However, our challenge to make (S)-1 presented a
few significant problems; namely, the difficult construction
of the quaternary carbon center of (R)-8, its facile racemiza-
tion, and the difficulties involved in detecting this racemiza-
tion.
We have been investigating the asymmetric synthesis of 1
by two approaches as shown in Scheme 1, each of which fea-
tures: 1) the preparation of a pivotal intermediate, early in
the synthesis, containing the optically active quaternary
carbon center with defined stereochemistry and 2) the com-
pletion of the total synthesis with retention of configuration.
Another important subject in our studies was the develop-
ment of effective protocols for the enantiodivergent synthe-
sis of both enantiomers of 1 and its various analogues.
After quite intensive studies, we have finally developed a
useful method for the preparation of (R)-8 by the lipase-cat-
alyzed desymmetrization of a prochiral diol 9 (R=CH2OH).
We have also elucidated the racemization mechanism and
discovered a means to prevent it. Thereby, the asymmetric
Scheme 1. Our two approaches for the asymmetric total syntheses of fredericamycin A (1).
Chem. Eur. J. 2005, 11, 6286 – 6297
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6287

Y. Kita et al.
total synthesis of natural 1 has been accomplished, and we
describe here the full details of this synthesis.
products 19, which have a greater degree of stability against
racemization under various conditions (Scheme 3).[24] As a
model study, we have applied this method to the diol 20,
Results and Discussion
Preparation of the optically active dione intermediate (R)-
8: To pursue the asymmetric total synthesis of 1 according
to Scheme 1, the first objective of our group was to prepare
the pivotal optically active dione (R)-8. Although we could
obtain (R)-8 with 96–98% ee by the separation of a 1:1 mix-
ture of the two diastereomers of the RAMP-hydrazone 11
by column chromatography, prepared by the acetylation of
the hydrazone 10,[12a] the separation process was too labori-
ous to produce a satisfactory amount of (R)-8 (Scheme 2).
Trials to improve the diastereoselectivity of the reaction of
10 to 11 were also fruitless.[20]
Scheme 3. Lipase-catalyzed desymmetrization of prochiral 1,3-diols 15.
which contains the DE-ring moiety, to afford (R)-21 with
96% ee.[24c] The conversion of (R)-21 to the pivotal DE-ring
dione intermediate (R)-25 was achieved with a good overall
yield (Scheme 4).[13]
Scheme 2. Initial trials for the preparation of (R)-8.
The hydrolase-catalyzed desymmetrization of the prochi-
ral disubstituted malonates has been recognized as a power-
ful and environmentally conscious method for the produc-
tion of optically active compounds with a stereogenic qua-
ternary carbon center.[21,22] However, our studies on the de-
symmetrization of the diester 12[12d] ended in the spontane-
ous decarboxylation of the monoester 13 to give only 14
(Scheme 2).
Scheme 4. Preparation of (R)-25 by the lipase-catalyzed desymmetriza-
tion of the prochical diol 20. a) TBSCl, pyridine, DMF; b) K2CO3,
MeOH; c) Dess–Martin periodinane, MeCN; d) NaClO2, NaH2PO4, 2-
methyl-2-butene, tBuOH/H2O; e) BF3·Et2O, CH2Cl2; f) MeLi, HMPA,
THF.
Therefore, our attention shifted to the lipase-catalyzed
enantiotopic selective acylation of the prochiral 2,2-disubsti-
tuted propan-1,3-diol 9a (R=CH2OH) in organic solvents.
Although the existing method using the most popular acyl
donor, vinyl acetate 16, has often suffered from the facile
racemization of the products 17 by intramolecular acyl
group migration,[23] we have recently developed a novel acyl
donor, 1-ethoxyvinyl 2-furoate 18a; this acyl donor provides
With these promising results in hand, we intended to
apply this desymmetrization protocol to the prochiral 1,3-
diol 9a,[25] followed by the conversion of the product to the
pivotal dione intermediate (S)-8. However, the reaction con-
ducted under similar conditions (iPr2O at 408C) was very
sluggish due to the very poor solubility of 9a, and provided
the
product
with
unsatisfactory
enantioselectivity
6288
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6286 – 6297

Total Synthesis of Fredericamycin A
FULL PAPER
(43% ee).[26] The use of other less polar solvents, such as cy-
clohexane, heptane, toluene, and tBuOMe also resulted in
slow reaction rates and moderate enantioselectivities (22–
54% ee). Although polar solvents such as MeCN and THF
were found to improve the solubility of 9a, the enantioselec-
tivities of these reactions were disappointing (up to
28% ee). We finally found that the use of a mixture of iPr2O
and MeCN produced better results. The enantioselectivity
was affected by the ratio of these solvents, among which the
use of a 10:1 mixture of iPr2O and MeCN at 408C for eight
days provided the best optical purity of (R)-26 (83% ee,
57% yield).[27] In this reaction, 9a was recovered in 40%
yield. The optical purity of (R)-26 (83% ee) was further im-
proved by its kinetic resolution with ethoxyvinyl butylate
18b (Scheme 5) and Pseudomonas aeruginosa lipase
did not proceed, we adopted an alternative method to pro-
duce (R)-8; this method involved the methylation of the al-
dehyde (R)-29, followed by the oxidation of the resultant
diol (8R)-32 (Scheme 6). The difference between the behav-
ior of the DEF-ring compounds compared with the DE-ring
analogues may be attributed to the significantly greater
steric hindrance produced by the two methoxy groups on
the E- and F-rings, relative to that produced by the single
methoxy group of the DE-ring analogues.
Scheme 6. Preparation of (R)-8 from (R)-26. a) TBSOTf, Na2CO3, THF;
b) K2CO3, MeOH; c) Dess–Martin periodinane, MeCN; d) NaClO2,
NaH2PO4, 2-methyl-2-butene, tBuOH, H2O; e) MeLi, THF/HMPA;
f) MeLi, THF; g) Bu4NF, THF; h) Swern oxidation.
Scheme 5. Preparation of (R)-26 by lipase-catalyzed reactions. a) 18a,
Candida rugosa lipase (Meito MY), iPr2O/MeCN/H2O (1000:100:1),
408C; b) 18b, Pseudomonas aeruginosa lipase (Toyobo-LIP), iPr2O,
408C.
Racemization of the quaternary carbon center and the solu-
tion to this problem: Next, we examined the conversion of
(R)-8 (95% ee) to the hexacyclic compound (S)-38 by a
series of transformations similar to the racemate synthesis
(Scheme 7).[12a] Thus, (R)-8, the alkyne 33, the acyl chloride
34, and LiN(TMS)2 were mixed in THF at ꢀ788C, and the
mixture was warmed to room temperature to give a 6:1 mix-
ture of two diastereomers (8R)-35. This mixture was then
treated with LiN(TMS)2 (3 equiv) in THF to induce the
acyl-group migration to the methyl ketone terminus, and the
subsequent Moffatt oxidation of the alcohol afforded a
single product, the trione (S)-36, which was converted to the
cobalt complex( S)-37. The treatment of (S)-37 with
Me2SiCl2, Et3N, and chloranil in refluxing toluene was one
of the key steps in the direct construction of the hexacyclic
skeleton (S)-6; the procedure involved the intramolecular
[4+2] cycloaddition of the silylene derivative (S)-7, followed
by oxidative aromatization.[12b–e] The application of our aro-
matic Pummerer-type reaction[19] to the sulfoxide (S)-5 using
1-ethoxyvinyl chloroacetate 18c replaced the B-ring sulfinyl
group with the acyloxy group via the O,S-acetal intermedi-
ate A. However, the optical purity of the desired product
(S)-38 was reduced to 82% ee (determined by chiral HPLC
analysis). We found it difficult to ascertain at which step in
the synthesis racemization of the quaternary carbon center
(Toyobo LIP). The latter reagent combination accelerated
the enantioselective esterification of (S)-26 to afford (R)-26
(97% ee, 60% yield) and the diester 27 (34% yield). Quan-
titative recycling of 27 to 9a by alkaline hydrolysis increased
the yield of (R)-26 (97% ee) to 86% (based on the con-
sumed amount of 9a), and the repeated experiment afford-
ed a satisfactory amount of (R)-26 with 95–97% ee. The ab-
solute stereochemistry of 26 was tentatively assigned as R
based on the structural similarity of 9a to 20; this was later
confirmed (see, Figure 2 later and its explanation).
The derivatization of (R)-26 (95% ee) to (R)-8 (95% ee)
was achieved in 96% overall yield with some modifications
to the above-mentioned conversion of the DE-ring analogue
(R)-20. These modifications were based on the following ex-
perimental results. Firstly, silylation of the hydroxyl group of
(R)-26 with organic bases, such as Et3N and pyridine result-
ed in partial racemization of the product (about a 20% ee
decrease) by intramolecular acyl-group migration, whereas
the use of inorganic solid bases such as Na2CO3 completely
suppressed the racemization. Secondly, as the conversion of
the carboxylic acid (R)-30 to the methyl ketone with MeLi
Chem. Eur. J. 2005, 11, 6286 – 6297
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6289

Y. Kita et al.
Scheme 7. First trials to convert (R)-8 to (S)-38. a) 33 (1.0 equiv), 34 (4 equiv), LiN(TMS)2 (2.0 equiv), THF; b) LiN(TMS)2, (3 equiv), THF; c) Dess–
Martin periodinane, MeCN; d) [Co2(CO)8], CH2Cl2; e) BCl3, CH2Cl2; f) Me2SiCl2, Et3N, chloranil, toluene; g) (tBu)2Si(OTf)2, Et3N, DMF; h) mCPBA,
CH2Cl2; i) 18c, cat. pTsOH, toluene.
was occurring for the following two reasons. Firstly, we
could not find an accurate method to determine the optical
purity of the synthetic intermediates between (R)-8 and (S)-
38. Secondly, any racemization of the intermediate com-
pounds was hard to detect as most of the intermediates only
contain a single stereogenic carbon center.
A similar racemization process also occurred during our
concurrent study towards the asymmetric synthesis of the
ABCDE-ring analogue of 1 from the dione (R)-25.[13] Thus,
the treatment of (R)-25 (96% ee) with lithium phenylthio-
acetylide at ꢀ788C, followed by the warming of the result-
ing reaction mixture to room temperature, produced the
adduct 39b (28% yield) as the only identifiable product
(Scheme 8). In this case, the optical purity of 39b could be
determined by HPLC analysis by using a chiral column. To
our surprise, this compound had been produced with 0% ee.
On the other hand, when we carried out the same reaction
of (R)-25 (90% ee) at ꢀ788C, followed by quenching at the
same temperature after 30 min, the product (1R)-39a, ex -
pected to be a diastereomer of 39b, was isolated without
any loss of chiral integrity. Having been intrigued by these
results, we investigated the effect of temperature on the
degree of racemization and ratio of these two products in an
attempt to clarify the racemization mechanism.
Thus, (R)-25 (90% ee) was treated with lithium phenyl-
thioacetylide at ꢀ788C for 30 min to ensure complete con-
sumption of (R)-25; then the reaction mixture was warmed
to the temperature indicated in Table 1 and stirred for a fur-
ther 30 min before being quenched with an aqueous solution
of NH4Cl. The results of these experiments (Table 1) provid-
ed us with useful information about the relation of the
quenching temperature to both the optical purity and the di-
astereoselectivity of the products: In the range from ꢀ60 to
ꢀ208C, partial racemization and/or isomerization took place
to give a mixture of (1R)-39a[28] and (1R)-39b[28] (entries 2–
6). However, at higher temperatures, the ratio of 39a to 39b
decreased with the decreasing optical purity of both 39a and
39b, and at 08C or above only 39b was obtained as a race-
mate (entries 7 and 8). In addition, the treatment of 39a
(90% ee) with pyridine (1.2 equiv) in THF at room tempera-
ture overnight produced racemic 39b as a single product
[Eq. (1)].
Scheme 8. Model studies examining the installation of an ethynyl group
on (R)-25.
6290
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6286 – 6297

Total Synthesis of Fredericamycin A
FULL PAPER
Table 1. The effect of temperature on the reaction of (R)-25 with lithium phenylthioacetylide.
with an aqueous solution of
NH4Cl provided
a separable
mixture of (8R)-35a[28] and its
diastereomer (8R)-35b.[28] This
reaction was found to be repro-
ducible, producing 35a and 35b
in a ratio of 19–21:1. (8R)-35a
and (8R)-35b were isolated in
yields of 41–57% and 2–3%,
respectively. Although we could
not find any effective method
to determine their optical
purity, the major product (8R)-
35a was expected to have main-
tained its chiral integrity,
whereas the minor product
(8R)-35b might have under-
gone some degree of racemiza-
tion. Therefore, only (8R)-35a
Entry
Quenching Temp.
(1R)-39a and (1R)-39b
ee [%] of 39a[c] ee [%] of 39b[c]
[8C][a]
Ratio of 39a/39b[b]
Total yield [%][b]
1
2
3
4
5
6
7
8
ꢀ78
ꢀ60
ꢀ50
ꢀ40
ꢀ30
ꢀ20
0
>99/<1[d]
ꢂ96/ꢃ4
91/9
83/17
51/49
90
–
43
44
46
46
45
42
31
32
90
90
90
84
37
–
–
64
51
40
29
2
14/86
<1/>99[d]
<1/>99[d]
20
–
0
[a] After a mixture of (R)-25 (90% ee), 33 (1.0 equiv), and nBuLi (1.1 equiv) was stirred at ꢀ788C for 30 min,
the reaction mixture was warmed to the given temperature, stirred for 30 min, and then quenched. [b] Based
on the isolated yield of 39a,b after silica-gel chromatography. [c] Determined by HPLC using a Daicel Chiral-
1
cel OJ column. [d] Formation of the isomer was not detected by H NMR spectroscopy or by HPLC.
These results provided a plausible racemization mecha-
nism [Eq. (2)]. According to this mechanism, 39a was ini-
tially formed under kinetic control, and upon warming of
the reaction mixture this compound gradually underwent
isomerization to the thermodynamically more stable com-
pound 39b, with racemization occurring by the retro-aldol–
aldol process.[29] Therefore, it was concluded that the reac-
tion below ꢀ608C inevitably produced the product with
complete retention of absolute configuration.
Scheme 9. One-pot preparation of (1R)-40a from (R)-25. a) 33
(1.0 equiv), nBuLi (1.1 equiv), HMPA/THF, ꢀ788C; b) 34, DMAP,
CH2Cl2; c) 33 (1.0 equiv), 34 (4 equiv), LiN(TMS)2 (2.0 equiv), THF,
ꢀ788C.
In our previous communication, the ester (1R)-40a was
obtained by the acylation of (1R)-39a (96% ee) with the A-
ring acid chloride 34.[13] Later we disclosed that by mixing
(R)-25 (96% ee), 33 (1 equiv), 34 (4 equiv), and LiN(TMS)2
(2 equiv) together in THF/HMPA (HMPA=hexamethyl-
phosphoramide), two reactions consecutively proceeded,
even at ꢀ788C, to directly produce (1R)-40a (56% yield) as
a single diastereomer with perfect retention of absolute con-
figuration (Scheme 9).[30]
was subjected to the subsequent transformation to give the
hexacyclic product (R)-38. The optical purity of (R)-38 was
determined to be 95% ee, and therefore, this proved that
the chiral integrity of the intermediate compounds had been
completely retained throughout the transformation steps.
The absolute stereochemistry of (S)-6 was confirmed by
the similarity of its CD spectrum to that of a known related
compound (S)-42[10a] (Figure 2).
The final transformation of (R)-38 to natural fredericamy-
cin A (1) was achieved without any difficulty (Scheme 11).
Thus, the demethylation of the F-ring of (R)-38, followed by
oxidation afforded the aldehyde (R)-44 in 91% yield. The
Wittig reaction of (R)-44 gave (R)-45 as a ꢂ10:1 mixture of
the E,E- and E,Z-side chain isomers; compound (R)-45 was
The asymmetric total synthesis of natural 1: Based on the
promising results obtained during the synthesis of the DE-
ring analogues, we restarted our studies towards the asym-
metric total synthesis of natural 1 (Scheme 10). Thus, (R)-8
(95% ee) was treated with 33 (1.0 equiv), 34 (4 equiv), and
LiN(TMS)2 (2.0 equiv) in THF at ꢀ788C for 1–1.5 h.
Quenching the reaction mixture at the same temperature
Chem. Eur. J. 2005, 11, 6286 – 6297
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6291

Y. Kita et al.
Scheme 10. Improved conversion of (R)-8 to (R)-38. a) 33 (1.0 equiv), 34 (4 equiv), LiN(TMS)2 (2.0 equiv), THF, ꢀ788C; b) LiN(TMS)2, (3 equiv), THF;
c) Dess–Martin periodinane, MeCN; d) [Co2(CO)8], CH2Cl2; e) BCl3, CH2Cl2; f) Me2SiCl2, Et3N, chloranil, toluene, 1008C; g) (tBu)2Si(OTf)2, Et3N,
DMF; h) mCPBA, CH2Cl2; i) 18c, cat. pTsOH, toluene.
Scheme 11. Completion of the asymmetric total synthesis of (S)-1.
a) (tBu)2Si(OTf)2, Et3N, MeI, DMF; b) SeO2, dioxane; c) (MeCH=
CHCH2)Ph3P+Brꢀ, nBuLi, THF; d) BBr3, CH2Cl2; e) air, THF/H2O;
f) HPLC separation.
Conclusion
We have demonstrated herein the synthetic utility of our in-
tramolecular cycloaddition pathway as a method for the
asymmetric total synthesis of natural 1. This follows our pre-
vious report on the enantiodivergent synthesis of both enan-
Figure 2. CD spectrum of (S)-6 (thin line) and (S)-42 (thick line) in
iPrOH.
tiomers of the ABCDE-ring analogue by a similar intramo-
lecular cycloaddition pathway.[13] This protocol features the
subjected to deprotection with BBr3 followed by autoxida-
tion to afford a mixture of (S)-1 and its E,Z-isomers. Sepa-
ration of the minor impurity by HPLC (Jasco Megapak SIL
NH2–10, 1”25 cm, CHCl3/hexane/acetic acid 800:200:1,
flow rate 5 mLminꢀ1)[10a] afforded pure (S)-1 (35% yield
from (R)-44). This product was identical in all respects
(1H NMR, IR, UV, and CD spectroscopy, m.p., HRMS,
HPLC, and TLC) with the authentic natural fredericamy-
cin A as well as the synthetic sample obtained by our inter-
molecular cyclization route.[10a]
following advantages: 1) the enantiodivergent preparation
of either enantiomer can be achieved with perfect retention
configuration and 2) this methodology is applicable to vari-
ous analogues by changing the A-ring moiety[12a,e] or the
DE(F)-ring diols, and provides an effective alternative to
our intermolecular cycloaddition pathway.[10] It is worth
noting that among the large number of synthetic studies to-
wards the asymmetric total synthesis of natural 1 as well as
its analogues,[9] successful examples have been limited to
only our two routes (Scheme 1).
6292
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6286 – 6297

Total Synthesis of Fredericamycin A
FULL PAPER
During this and our previous studies,[13,29] we have dis-
closed the latent risk of racemization of the stereogenic qua-
ternary carbon center of the compounds, such as 39, bearing
an a,a-disubstituted b-hydroxycarbonyl moiety. The racemi-
zation took place under basic conditions by the retro-aldol–
aldol process; however, these issues were solved by the
choice of the proper reaction conditions. In the reported
syntheses of racemic 1[3,8] and its partial structures,[31] the
base-initiated aldol reactions have often been employed to
construct the spiro CD-ring system. Special caution will be a
requisite for the application of these reactions to the asym-
metric synthesis.
(m, 1H), 2.50 (s, 3H), 2.94–3.18 (m, 3H), 3.94 (s, 3H), 3.98–4.00 (m, 1H),
4.14 (s, 3H), 4.61 (d, J=11.0 Hz, 1H), 4.74 (d, J=11.0 Hz, 1H), 6.49 (dd,
J=1.5, 3.5 Hz, 1H), 6.96 (s, 1H), 7.10 (dd, J=0.5, 3.5 Hz, 1H), 7.26 (s,
1H), 7.57 ppm (dd, J=0.5, 1.5 Hz, 1H); 13C NMR (75 MHz, CDCl3): d=
23.7, 30.2, 32.3, 53.7, 55.2, 63.9, 67.1, 67.4, 111.1, 111.8, 112.7, 117.6, 118.0,
133.2, 142.8, 144.4, 146.4, 148.9, 149.7, 154.3, 158.7, 158.8 ppm; HRMS
(FAB): m/z: calcd for C22H23NO6: 398.1604 [M+H]+; found: 398.1598.
8-Butyryloxymethyl-8-(2-furoyloxy)methyl-1,9-dimethoxy-3-methyl-6,7-
dihydro-8H-cyclopenta[g]isoquinoline (27): IR (KBr): n˜ =1732, 1628,
1
1612, 1566 cmꢀ1; H NMR (300 MHz, CDCl3): d=0.80 (t, J=7.5 Hz, 3H),
1.50 (q, J=7.5 Hz, 2H), 2.13–2.21 (m, 2H), 2.16 (t, J=7.5 Hz, 2H), 2.42
(s, 3H), 2.98–3.03 (m, 2H), 3.81 (s, 3H), 4.06 (s, 3H), 4.41 (d, J=11.0 Hz,
1H), 4.54 (d, J=11.0 Hz, 1H), 4.57 (d, J=11.0 Hz, 1H), 4.73 (d, J=
11.0 Hz, 1H), 6.38 (dd, J=1.5, 3.5 Hz, 1H), 6.88 (s, 1H), 6.97 (d, J=
3.5 Hz, 1H), 7.17 (s, 1H), 7.47 ppm (brs, 1H); 13C NMR (68 MHz,
CDCl3): d=13.7, 18.4, 23.7, 30.5, 32.6, 36.2, 53.0, 53.7, 63.5, 67.4, 68.0,
111.2, 111.7, 112.7, 117.0, 117.9, 132.0, 142.7, 144.3, 146.3, 148.6, 149.2,
154.7, 158.4, 158.8, 173.4 ppm; HRMS (FAB): m/z: calcd for C26H30NO7:
468.2022 [M+H]+; found: 468.2003.
Experimental Section
General aspects: Melting points (m.p.) are not corrected. 1H NMR spec-
tra were measured at 270–500 MHz with either Me4Si or CHCl3 (d=
7.26 ppm) as an internal standard. 13C NMR spectra were measured at
67.8–150 MHz with CDCl3 (d=77.0 ppm) as an internal standard. IR
spectra were recorded by diffuse reflectance measurement of samples dis-
persed in KBr powder. Chiral HPLC analyses were carried out using
Daicel CHIRALCEL OD (250 mm”4.6 mm), CHIRALCEL OJ
(250 mm”4.6 mm), and CHIRALPAK AD (250 mm”4.6 mm) columns
with a hexane/iPrOH mixture as the eluent. Flash column chromatogra-
phy was carried out using silica gel BW-300 (200–400 mesh, Fuji Silysia
Chemicals (Japan)). Pressure glass reactors (Taiatsu Techno (Japan))
were used for the intramolecular [4+2] cycloaddition reactions. Starting
materials: 18a,[24c] 18b,[24e] 18c,[32] 33,[33] 34,[12a] and crotyltriphenylphos-
phonium bromide[34] were prepared according to the reported methods.
The preparation of 9a and the conversion of (R)-25–(1R)-40a are de-
scribed in the Supporting Information. Candida rugosa lipase and Pseu-
domonas aeruginosa lipase were gifts from Meito Sangyo (Japan) and
Toyobo (Japan), respectively, and were dried (1 mmHg, RT, overnight)
prior to use. All other materials were commercially available. Yields
refer to the isolated material with ꢂ95% purity as determined by
1H NMR spectroscopic analysis, unless otherwise noted.
Hydrolysis of 27: A mixture of 27 (96 mg), KOH (2n, 0.13 mL), MeOH
(1.3 mL), and CH2Cl2 (0.80 mL) was stirred at 08C for 5 h. After this
time, the reaction mixture was quenched with saturated aqueous NH4Cl,
and the product extracted with EtOAc. The combined organic layers
were washed with brine, dried (Na2SO4), and concentrated in vacuo. Puri-
fication of the residue by flash column chromatography (hexanes/EtOAc
1:1!EtOAc) provided 9a (61 mg, 98% yield).
(R)-8-(tert-Butyldimethylsilyloxy)methyl-8-hydroxymethyl-1,9-dime-
thoxy-3-methyl-6,7-dihydro-8H-cyclopenta[g]isoquinoline
((R)-28):
Under a nitrogen atmosphere, Na2CO3 (130 mg, 1.25 mmol) and TBSOTf
(tert-butyldimethylsilyl triflate, 0.30 mL, 1.25 mmol) were added to a so-
lution of (R)-26 (95% ee, 100 mg, 0.25 mmol) in anhydrous THF
(13 mL), and the reaction mixture was stirred for 20 min. After this time,
iced water was added, and the product was extracted with diethyl ether.
The combined organic layers were washed with brine, dried (Na2SO4),
and concentrated in vacuo. The residue was roughly purified by filtration
through a short pad of silica gel (hexanes/EtOAc 3:1) to give (R)-8-(tert-
butyldimethylsilyloxy)methyl-8-(2-furoyloxy)methyl-1,9-dimethoxy-3-
methyl-6,7-dihydro-8H-cyclopenta[g]isoquinoline as a colorless oil. Al-
though this product contained a small amount of TBSOH (tert-butyldi-
methylsilyl alcohol), it was used for the following reaction without fur-
ther purification. A sample of the pure material was obtained by prepara-
tive TLC (hexanes/EtOAc 3:1). IR (KBr): n˜ =1730, 1628, 1612,
Lipase-catalyzed desymmetrization of 9a:
A mixture of 9a (0.30 g,
0.99 mmol), Candida rugosa lipase (Meito MY, 0.60 g), and 18a (0.54 g,
2.97 mmol) was stirred in a mixture of iPr2O (20 mL), MeCN (2.0 mL),
and H2O (0.020 mL) at 408C for eight days. After this time, the reaction
mixture was filtered through a Celite pad. Water (20 mL) was added to
the resulting filtrate, which was stirred vigorously for a further 2 h. The
product was then extracted with EtOAc, and the combined organic
layers were washed with brine, dried (Na2SO4), and concentrated in
vacuo. Purification of the residue by flash column chromatography (hex-
anes/EtOAc, 2:1!1:2) afforded both (R)-26 (0.22 g, 57% yield) and 9a
(0.120 g, 40% yield). The optical purity of (R)-26 was determined to be
83% ee by chiral HPLC analysis (Daicel CHIRALCEL OD-H, hexanes/
iPrOH 98:2, flow rate 1.0 mLminꢀ1, 388C). Retention time: 43.7 min for
(S)-26 and 46.9 min for (R)-26.
1566 cmꢀ1
;
1H NMR (300 MHz, CDCl3): d=ꢀ0.03 (s, 3H), 0.03 (s, 3H),
0.84 (s, 9H), 2.18–2.27 (m, 1H), 2.31–2.41 (m, 1H), 2.49 (s, 3H), 3.02–
3.07 (m, 2H), 3.74 (d, J=9.5 Hz, 1H), 3.87 (s, 3H), 4.13 (s, 3H), 4.17 (d,
J=9.5 Hz, 1H), 4.72 (d, J=11.0 Hz, 1H), 4.82 (d, J=11.0 Hz, 1H), 6.40–
6.41 (m, 1H), 6.93 (d, J=3.0 Hz, 1H), 6.94 (s, 1H), 7.22 (s, 1H),
7.51 ppm (d, J=1.0 Hz, 1H); 13C NMR (67.8 MHz, CDCl3): d=ꢀ5.6,
ꢀ5.5, 18.2, 23.6, 25.8, 30.7, 32.2, 53.7, 55.3, 63.3, 66.7, 69.0, 111.3, 111.6,
112.8, 116.9, 117.6, 133.4, 142.6, 144.6, 146.2, 148.3, 150.2, 154.5, 158.8,
158.9 ppm; HRMS (FAB): m/z: calcd for C28H38NO6Si: 512.2468
[M+H]+; found: 512.2471.
Potassium carbonate (70 mg, 0.50 mmol) was added to a solution of the
above crude product in MeOH (13 mL), and the reaction mixture was
stirred at RT for 3 h. After this time, the reaction mixture was concen-
trated in vacuo, and then purified by flash column chromatography (hex-
anes/EtOAc, 5:1) to afford (R)-28 (105 mg, quant from (R)-26) as a col-
orless oil. [a]3D0 =+45.6 (c=0.88 in CHCl3); IR (KBr): n˜ =3600–3200,
Kinetic resolution of (R)-26 (83% ee): A mixture of (R)-26 (83% ee,
0.22 g, 0.56 mmol), Pseudomonas aeruginosa lipase (Toyobo LIP, 0.45 g),
and 18b (0.54 g, 3.4 mmol) was stirred in iPr2O (23 mL) at 408C for 6 h.
After this time, the reaction mixture was filtered through a Celite pad,
and water (20 mL) was added to the resulting filtrate, which was stirred
vigorously for a further 2 h. The product was then extracted with EtOAc,
and the combined organic layers were washed with brine, dried
(Na2SO4), and concentrated in vacuo. Purification of the residue by flash
column chromatography (hexanes/EtOAc 2:1) afforded both (R)-26
(97% ee, 134 mg, 60% yield) and 27 (96 mg, 34% yield) as colorless oils.
1
1628, 1612, 1564 cmꢀ1; H NMR (270 MHz, CDCl3): d=0.02 (s, 3H), 0.06
(s, 3H), 0.87 (s, 9H), 2.16–2.33 (m, 2H), 2.45 (s, 3H), 2.95–3.00 (m, 2H),
3.43 (brs, 1H), 3.67 (d, J=9.5 Hz, 1H), 3.76–3.85 (m, 1H), 3.84 (s, 3H),
4.10 (s, 3H), 4.21 (d, J=10.5 Hz, 1H), 4.26 (d, J=9.5 Hz, 1H), 6.87 (s,
1H), 7.16 ppm (s, 1H); 13C NMR (75 MHz, CDCl3): d=ꢀ5.7, ꢀ5.6, 18.0,
23.5, 25.7, 30.1, 31.4, 53.5, 56.5, 63.6, 68.1, 68.3, 110.9, 112.5, 117.1, 133.9,
142.4, 148.2, 149.7, 153.8, 158.4 ppm; HRMS (FAB): m/z: calcd for
C23H36NO4Si: 418.2413 [M+H]+; found: 418.2422.
(R)-8-(2-Furoyloxy)methyl-8-hydroxymethyl-1,9-dimethoxy-3-methyl-6,7-
dihydro-8H-cyclopenta[g]isoquinoline ((R)-26): [a]2D5 =ꢀ152 (c=0.50 in
CHCl3) for 97% ee; IR (KBr): n˜ =3600–3200, 1720, 1628, 1612,
(R)-8-(tert-Butyldimethylsilyloxy)methyl-8-formyl-1,9-dimethoxy-3-
methyl-6,7-dihydro-8H-cyclopenta[g]isoquinoline ((R)-29): Under a ni-
1566 cmꢀ1 1H NMR (300 MHz, CDCl3): d=2.06–2.18 (m, 1H), 2.21–2.30
;
Chem. Eur. J. 2005, 11, 6286 – 6297
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6293

Y. Kita et al.
trogen atmosphere, Dess–Martin periodinane (0.40 g, 0.96 mmol) was
added to an ice-cooled solution of (R)-28 (195 mg, 0.47 mmol) in MeCN
(10 mL), and the mixture was stirred at RT for 4 h. After this time, satu-
rated aqueous Na2S2O3 was added at 08C, and the product was extracted
with EtOAc. The combined organic layers were washed with brine, dried
(Na2SO4), and concentrated in vacuo. Purification of the residue by flash
column chromatography (hexanes/EtOAc 5:1) produced (R)-29 (192 mg,
99% yield) as a colorless oil. [a]3D0 =ꢀ22.2 (c=1.1 in CHCl3); IR (KBr):
(160 mg, 0.49 mmol) in anhydrous CH2Cl2 (3.0 mL) was added at ꢀ788C,
and the reaction mixture was stirred for 2 h. Finally, Et3N (2.8 mL,
17.6 mmol) was added at ꢀ788C, and the reaction mixture was warmed
to RT and stirred for a further 45 min. The reaction mixture was then
quenched with saturated aqueous NH4Cl, and the product was extracted
with CH2Cl2. The combined organic layers were washed with brine, dried
(Na2SO4), and concentrated in vacuo. Purification of the residue by flash
column chromatography (hexanes/EtOAc 4:1) produced (R)-8 (152 mg,
quant) as a colorless oil. The optical purity of this product was deter-
mined to be 95% ee by chiral HPLC analysis (Daicel CHIRALPAK AD-
H, hexanes/iPrOH 99:1, flow rate 0.8 mLminꢀ1, 258C). Retention time:
15.3 min for (R)-8 and 21.8 min for (S)-8. [a]2D6 =+303 (c=1.0 in CHCl3);
n=1726, 1628, 1612, 1568 cmꢀ1
;
1H NMR (270 MHz, CDCl3): d=ꢀ0.02
(s, 3H), 0.04 (s, 3H), 0.83 (s, 9H), 2.23–2.31 (m, 1H), 2.48 (s, 3H), 2.48–
2.59 (m, 1H), 3.04–3.13 (m, 2H), 3.75 (d, J=10.0 Hz, 1H), 3.80 (s, 3H),
4.12 (s, 3H), 4.52 (d, J=10.0 Hz, 1H), 6.93 (s, 1H), 7.21 (s, 1H),
10.00 ppm (s, 1H); 13C NMR (67.8 MHz, CDCl3): d=ꢀ5.5, ꢀ5.4, 18.2,
23.7, 25.8, 30.0, 31.0, 53.7, 63.5, 64.7, 65.0, 110.9, 112.7, 117.2, 132.5, 142.9,
148.8, 149.4, 153.5, 158.6, 202.2 ppm; HRMS (FAB): m/z: calcd for
C23H34NO4Si: 416.2257 [M+H]+; found: 416.2273.
IR (KBr): n˜ =1726, 1701, 1628, 1570 cmꢀ1 1H NMR (300 MHz, CDCl3):
;
d=2.20 (s, 3H), 2.33 (ddd, J=5.0, 8.5, 13.0 Hz, 1H), 2.50 (s, 3H), 2.75
(ddd, J=6.0, 7.0, 13.0 Hz, 1H), 3.06–3.19 (m, 2H), 3.85 (s, 3H), 4.13 (s,
3H), 6.96 (s, 1H), 7.28 (s, 1H), 10.04 ppm (s, 1H); 13C NMR (67.8 MHz,
CDCl3): d=23.7, 26.7, 30.9, 31.6, 53.7, 63.7, 74.2, 111.0, 112.7, 117.5,
130.1, 143.4, 148.3, 149.5, 154.0, 158.7, 197.1, 206.7 ppm; HRMS (FAB):
m/z: calcd for C18H20NO4: 314.1392 [M+H]+; found: 314.1397.
(8R)-8-(1-Hydroxyethyl)-8-hydroxymethyl-1,9-dimethoxy-3-methyl-6,7-di-
hydro-8H-cyclopenta[g]isoquinoline ((8R)-32): Under a nitrogen atmos-
phere, MeLi (1.0m in Et2O, 1.0 mL, 1.0 mmol) was added to a solution of
(R)-29 (0.22 g, 0.52 mmol) in anhydrous THF (15 mL) at ꢀ788C, and the
reaction mixture was stirred at ꢀ788C for 1.5 h. After this time, saturated
aqueous NH4Cl was added at ꢀ788C, and the reaction mixture was
warmed to RT with vigorous stirring. The product was then extracted
with EtOAc, and the combined organic layers were washed with brine,
dried (Na2SO4), and concentrated in vacuo. Purification of the residue by
flash column chromatography (hexanes/EtOAc 4:1) produced (8R)-8-
(tert-butyldimethylsilyloxy)methyl-8-(1-hydroxyethyl)-1,9-dimethoxy-3-
methyl-6,7-dihydro-8H-cyclopenta[g]isoquinoline (0.22 g, 96% yield) as a
colorless oil and as a 1:1 mixture of two diastereomers. [a]3D0 =+62.6 (c=
(8R)-1-(8-Acetyl-1,9-dimethoxy-3-methyl-6,7-dihydro-8H-cyclopenta[g]-
isoquinolin-8-yl)-3-phenylthio-2-propynyl-2-benzyloxy-4,5-dimethoxy-
benzoate ((8R)-35a) and its diastereomer ((8R)-35b): Under a nitrogen
atmosphere, lithium bis(trimethylsilyl)amide (1.0m in THF, 0.60 mL,
0.60 mmol) was added at ꢀ788C to
a solution of (R)-8 (95 mg,
0.30 mmol), 33 (1.0m in THF, 0.30 mL, 0.30 mmol), and 34 (prepared in
situ from 2-benzyloxy-4,5-dimethoxybenzoic acid (0.34 g, 1.2 mmol)) in
anhydrous THF (3.0 mL). The reaction mixture was stirred at ꢀ788C for
1.5 h, and then quenched by the addition of saturated aqueous NH4Cl.
This solution was warmed to RT with vigorous stirring, and the product
was extracted with EtOAc. The combined organic layers were washed
with brine, dried (Na2SO4), and concentrated in vacuo. Purification of the
residue by flash column chromatography (hexanes/EtOAc 3:1!benzene/
diethyl ether 9:1) produced both (8R)-35a (124 mg, 57% yield) and
(8R)-35b (6 mg, 3% yield) as pale yellow gums.
1.0 in CHCl3); IR (KBr): n˜ =3600–3200, 1626, 1612, 1562 cmꢀ1 1H NMR
;
(270 MHz, CDCl3): d=ꢀ0.04 (s, 3/2H), 0.03 (s, 3/2H), 0.06 (s, 3/2H),
0.10 (s, 3/2H), 0.80 (s, 9/2H), 0.90 (s, 9/2H), 0.93 (d, J=6.5 Hz, 3/2H),
1.10 (d, J=6.5 Hz, 3/2H), 2.03–2.22 (m, 2H), 2.46 (s, 3H), 2.93–3.05 (m,
2H), 3.28 (brd, J=6.5 Hz, 1/2H), 3.66 (d, J=9.5 Hz, 1/2H), 3.82–3.88
(m, 1/2H), 3.85 (s, 3/2H), 3.89 (s, 3/2H), 4.10 (s, 3/2H), 4.11 (s, 3/2H),
4.07–4.15 (m, 1H), 4.49–4.52 (m, 1H), 4.79 (q, J=6.5 Hz, 1/2H), 6.90 (s,
1H), 7.18 ppm (s, 1H); 13C NMR (67.8 MHz, CDCl3): d=ꢀ5.7, ꢀ5.6,
ꢀ5.5, 18.0, 18.1, 18.7, 19.2, 23.6, 25.7, 26.8, 30.7, 30.9, 31.4, 53.5, 53.6, 58.9,
61.2, 63.3, 63.4, 67.4, 70.8, 71.0, 73.3, 110.8, 111.0, 112.5, 117.0, 117.1,
133.7, 134.3, 142.2, 142.3, 148.0, 148.2, 149.3, 150.8, 153.3, 153.8, 158.49,
158.50 ppm; HRMS (FAB): m/z: calcd for C24H38NO4Si: 432.2570
[M+H]+; found: 432.2569.
(8R)-35a (less polar): [a]2D5 =+173 (c=1.0 in CHCl3); IR (KBr): n˜ =2183,
1727, 1711, 1626, 1611, 1566 cmꢀ1 1H NMR (270 MHz, CDCl3): d=2.19
;
(s, 3H), 2.45 (s, 3H), 2.57 (ddd, J=7.0, 9.0, 14.0 Hz, 1H), 2.85 (ddd, J=
5.0, 9.0, 14.0 Hz, 1H,), 3.15 (ddd, J=5.0, 9.0, 14.5 Hz, 1H), 3.34 (ddd, J=
7.0, 9.0, 14.5 Hz, 1H), 3.56 (s, 3H), 3.73 (s, 3H), 3.83 (s, 3H), 4.06 (s,
3H), 4.91 (d, J=12.5 Hz, 1H), 4.97 (d, J=12.5 Hz, 1H), 6.35 (s, 1H),
6.71 (s, 1H), 6.78 (s, 1H), 6.85 (s, 1H), 7.14–7.19 (m, 2H), 7.23–7.35 (m,
7H), 7.40–7.45 ppm (m, 2H); 13C NMR (67.8 MHz, CDCl3): d=23.7,
26.4, 31.9, 32.2, 53.7, 55.9, 56.0, 63.3, 67.9, 68.2, 72.1, 73.0, 77.2, 96.2,
100.1, 111.0, 111.3, 112.4, 113.3, 116.8, 126.1, 126.3, 127.0, 127.6, 128.3,
129.1, 131.5, 132.4, 136.6, 142.5, 143.1, 149.1, 149.6, 153.0, 154.4, 154.5,
159.0, 163.5, 205.5 ppm; HRMS (FAB): m/z: calcd for C42H40NO8S:
718.2474 [M+H]+; found: 718.2444.
Under a nitrogen atmosphere, Bu4NF (1.0m in THF, 1.3 mL, 1.3 mmol)
was added to a solution of the above product (0.22 g, 0.50 mmol) in anhy-
drous THF (15 mL) at ꢀ788C, and the reaction mixture was stirred at
ꢀ788C for 10 min. After this time, the reaction mixture was gradually
warmed to RT. Water was then added and the product was extracted
with EtOAc. The combined organic layers were washed with brine, dried
(Na2SO4), and concentrated in vacuo. Purification of the residue by flash
column chromatography (hexanes/EtOAc 1:1) produced (8R)-32 (160 mg,
(8R)-35b (more polar): The IR, 1H NMR, and 13C NMR spectroscopic
data of this product compared well with the data for racemic 35b.[12a]
(S)-{1-[1,9-Dimethoxy-3-methyl-8-(3-phenylthio-2-propioloyl)-6,7-dihy-
dro-8H-cyclopenta[g]isoquinolin-8-yl]-3-(2-hydroxy-4,5-dimethoxyphen-
yl)-1,3-propanedione}hexacarbonyldicobalt ((S)-37): Under a nitrogen at-
mosphere, lithium bis(trimethylsilyl)amide (1.0m in THF, 0.51 mL,
0.51 mmol) was added to a solution of (8R)-35a (123 mg, 0.17 mmol) in
anhydrous toluene (3.2 mL) at ꢀ408C, and the reaction mixture was stir-
red for 4 h. After this time, the reaction mixture was quenched by the ad-
dition of saturated aqueous NH4Cl at ꢀ408C, and the product was ex-
tracted with EtOAc. The combined organic layers were washed with
brine, dried (Na2SO4), and concentrated in vacuo. Purification of the resi-
due by flash column chromatography (hexanes/EtOAc 2:1) produced
(R)-3-(2-benzyloxy-4,5-dimethoxy)phenyl-1-[8-(1-hydroxy-3-phenylthio-2-
propynyl)-1,9-dimethoxy-3-methyl-6,7-dihydro-8H-cyclopenta[g]isoquino-
quant) as colorless oil and as a 1:1 mixture of two diastereomers. [a]D24
+18.8 (c=1.0 in CHCl3); IR (KBr): n˜ =3700–3100, 1626, 1562 cmꢀ1
=
;
1H NMR (270 MHz, CDCl3): d=1.02 (d, J=6.5 Hz, 3/2H), 1.14 (d, J=
6.5 Hz, 3/2H), 1.91–2.04 (m, 1H), 2.12–2.22 (m, 1/2H), 2.32–2.43 (m,
1/2H), 2.49 (s, 3H), 2.99–3.06 (m, 2H), 3.73 (d, J=11.0 Hz, 1/2H), 3.80
(d, J=11.0 Hz, 1/2H), 3.89 (s, 3/2H), 3.95 (s, 3/2H), 4.05 (d, J=11.0 Hz,
1/2H), 4.14 (s, 3H), 4.14–4.23 (m, 1/2H), 4.17 (d, J=11.0 Hz, 1/2H), 4.70
(q, J=6.5 Hz, 1/2H), 6.94 (s, 1H), 7.22 (s, 1/2H), 7.24 ppm (s, 1/2H);
13C NMR (67.8 MHz, CDCl3): d=18.5, 18.8, 23.6, 27.6, 29.7, 30.3, 31.1,
36.6, 53.69, 53.72, 59.2, 60.7, 63.66, 63.70, 66.9, 69.6, 70.8, 71.8, 110.7,
110.8, 112.49, 112.51, 117.4, 117.5, 132.9, 135.2, 142.3, 142.4, 148.3, 148.5,
149.4, 150.3, 153.48, 153.5, 158.42, 158.44 ppm; HRMS (FAB): m/z: calcd
for C18H24NO4: 318.1705 [M+H]+; found: 318.1689.
lin-8-yl]-1,3-propanedione (123 mg, quant) as a pale yellow gum. [a]D27
ꢀ125 (c=1.0 in CHCl3); IR (KBr): n˜ =3500–3300 (br), 2190–2170, 1611,
1566 cmꢀ1 1H NMR (270 MHz, CDCl3): d=2.34–2.45, (m, 2H), 2.49 (s,
=
(R)-8-Acetyl-8-formyl-1,9-dimethoxy-3-methyl-6,7-dihydro-8H-cyclopen-
ta[g]isoquinoline ((R)-8): Under
(0.94 mL, 11.8 mmol) was added to
a
nitrogen atmosphere, DMSO
solution of oxalyl chloride
;
a
3H), 2.85–3.05 (m, 2H), 3.79 (s, 3H), 3.85 (s, 3H), 3.89 (s, 3H), 4.04 (s,
3H), 4.87 (d, J=11.5 Hz, 1H), 4.90 (d, J=11.5 Hz, 1H), 5.02 (d, J=
9.5 Hz, 1H), 5.10 (d, J=9.5 Hz, 1H), 6.41 (s, 1H), 6.66 (s, 1H), 6.78–6.85
(0.56 mL, 5.9 mmol) in anhydrous CH2Cl2 (3.0 mL), and the reaction mix-
ture was stirred at ꢀ788C for 1 h. After this time, a solution of (R)-32
6294
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6286 – 6297

Total Synthesis of Fredericamycin A
FULL PAPER
(m, 2H), 6.80 (s, 1H), 6.82 (s, 1H), 6.92–7.01 (m, 1H), 6.93 (s, 1H), 7.18–
7.24 (m, 3H), 7.30–7.36 (m, 3H), 7.49 (s, 1H), 16.32 ppm (s, 1H);
13C NMR (67.8 MHz, CDCl3): d=23.7, 30.7, 36.5, 53.6, 56.0, 56.2, 63.1,
67.7, 68.3, 71.2, 71.7, 77.2, 98.1, 98.7, 99.5, 111.8, 111.7, 112.5, 114.7, 117.3,
125.3, 125.8, 127.2, 128.1, 128.47, 128.53, 131.9, 132.1, 135.9, 142.7, 143.0,
148.8, 149.8, 152.7, 153.5, 153.6, 158.7, 178.3, 198.2 ppm; HRMS (FAB):
m/z: calcd for C42H40NO8S: 718.2474 [M+H]+; found: 718.2444.
column chromatography (CH2Cl2!CH2Cl2/EtOAc 100:1) afforded (S)-6
(3.3 mg, 28% yield) as a yellow gum. [a]2D0 =+14.9 (c=1.2 in CHCl3);
CD (c=1.3”10ꢀ5 molLꢀ1 in iPrOH) [q]m20ax =+4.7”104 (258), ꢀ1.5”104
(280), +3.6”104 (300), ꢀ1.2”104 (381), +1.7”104 cm2 dmolꢀ1 (475 nm);
IR (KBr): n˜ =1730, 1703, 1630, 1603, 1571 cmꢀ1 1H NMR (500 MHz,
;
CDCl3): d=1.15 (s, 9H), 1.19 (s, 9H), 1.93–2.03 (m, 1H), 2.32–2.38 (m,
1H), 2.45 (s, 3H), 3.08–3.16 (m, 1H), 3.19–3.27 (m, 1H), 3.44 (s, 3H),
3.94 (s, 3H), 4.00 (s, 3H), 4.01 (s, 3H), 6.90 (s, 1H), 6.93–6.97 (m, 1H),
6.95 (s, 1H), 7.09 (t, J=7.5 Hz, 2H), 7.16 (d, J=7.5 Hz, 2H), 7.20 ppm (s,
1H); 13C NMR (125 MHz, CDCl3): d=21.2, 23.7, 26.1, 26.2, 32.2, 35.8,
53.3, 56.4, 61.6, 62.3, 66.1, 103.9, 111.1, 113.0, 113.8, 116.9, 120.5, 125.3,
128.5, 128.7, 134.3, 134.6, 140.9, 141.0, 143.1, 148.4, 150.0, 150.9, 151.1,
152.5, 155.3, 159.0, 198.3, 198.8 ppm; HRMS (FAB): m/z: calcd for
C43H46NO8SSi: 764.2714 [M+H]+; found: 764.2715.
Dess–Martin periodinane (31 mg, 0.074 mmol) was added to an ice-
cooled solution of the above product (35 mg, 0.049 mmol) in MeCN
(2.8 mL), and the reaction mixture was stirred for 2 h. After this time, ad-
ditional Dess–Martin periodinane (10 mg, 0.025 mmol) was added, and
the reaction mixture was stirred at RT for a further 1 h. The reaction
mixture was then quenched by the addition of saturated aqueous
Na2S2O3 at 08C, and the product was extracted with EtOAc. The com-
bined organic layers were washed with brine, dried (Na2SO4), and con-
centrated in vacuo to afford (S)-3-(2-benzyloxy-4,5-dimethoxyphenyl)-1-
[1,9-dimethoxy-3-methyl-8-(3-phenylthio-2-propioloyl)-7,8-dihydro-6H-
cyclopenta[g]isoquinolin-8-yl]-1,3-propanedione ((S)-36). Due to its insta-
bility, this product was used for the following reaction without further pu-
rification.
(S)-8,9-[Di(tert-butyl)silylenedioxy]-1’,5,6,9’-tetramethoxy-3’-methyl-4-
phenylsulfinyl-6’,7’-dihydrospiro[2H-benz[f]indene-2,8’-[8’H]cyclopent-
[g]isoquinoline]-1,3-dione ((S)-5): mCPBA (meta-chloroperoxybenzoic
acid, 80% purity, 1.7 mg, 0.0080 mmol) was added to a solution of (S)-6
(5.0 mg, 0.0065 mmol) in CH2Cl2 (1.7 mL) at ꢀ788C, and the reaction
mixture was stirred for 1 h at ꢀ608C. After this time, saturated aqueous
Na2S2O3 and saturated aqueous NaHCO3 were added at 08C, and the
product was extracted with CH2Cl2. The combined organic layers were
washed with brine, dried (Na2SO4), and concentrated in vacuo. Purifica-
tion of the residue by flash column chromatography (hexanes/EtOAc
1:1) afforded (S)-5 (3.3 mg, 65% yield) as a yellow gum. This product
was found to consist of two diastereomers in ratio of approximately 3:2.
Under an argon atmosphere, octacarbonyldicobalt (21 mg, 0.054 mmol)
was added to the above crude product in anhydrous CH2Cl2 (1.4 mL).
The reaction mixture was stirred at RT for 10 min and then concentrated
in vacuo. Purification of the crude residue was achieved by flash column
chromatography (hexanes/EtOAc 2:1) to afford (S)-{3-(2-benzyloxy-4,5-
dimethoxyphenyl)-1-[1,9-dimethoxy-3-methyl-8-(3-phenylthio-2-propiolo-
yl)-6,7-dihydro-8H-cyclopenta[g]isoquinolin-8-yl]-1,3-propanedione}hexa-
carbonyldicobalt (37 mg). Although this product was not pure, it was
used for the following reaction without further purification due to its in-
stability.
IR (KBr): n˜ =1740, 1709, 1630, 1605, 1568 cmꢀ1 1H NMR (300 MHz,
;
CDCl3): d=1.15, 1.16, 1.17, 1.18 (4s, 18H), 2.43, 2.46 (2s, 3H), 2.35–2.65
(m, 2H), 3.10 (s, 9/5H), 3.25–3.40 (m, 2H), 3.54 (brs, 12/5H), 3.80 (s,
9/5H), 3.96, 4.00, 4.01, 4.05 (4s, 6H), 6.90, 6.93, 6.96, 6.98 (4s, 2H), 7.22–
7.44 (m, 4H), 7.74 (d, J=7.0 Hz, 4/5H), 7.86 ppm (d, J=7.5 Hz, 6/5H);
13C NMR (125 MHz, CDCl3): d=21.0, 21.1, 21.3, 21.4, 23.65, 23.70, 26.07,
26.12, 26.17, 32.27, 32.35, 35.95, 36.04, 53.33, 53.38, 56.34, 56.37, 60.8,
61.1, 62.0, 62.8, 65.9, 66.0, 103.6, 104.1, 111.0, 111.2, 112.9, 113.0, 113.2,
116.8, 117.0, 125.6, 126.2, 128.2, 128.3, 129.0, 129.1, 134.0, 134.1, 143.2,
143.3, 145.2, 145.7, 148.5, 148.7, 149.9, 150.2, 151.16, 151.2, 152.8, 154.3,
155.3, 155.5, 158.8, 159.0, 197.6, 197.7 ppm; HRMS (FAB): m/z: calcd for
C43H46NO9SiS: 780.2662 [M+H]+; found: 780.2667.
Under an argon atmosphere, BCl3 (0.10m in CH2Cl2, 0.10 mL,
0.010 mmol) was added drop wise to an ice-cooled solution of the above
product (13 mg) in CH2Cl2 (4.0 mL), and the reaction mixture was stirred
at 08C for 5 min. BCl3 (0.10m in CH2Cl2, 0.20 mL, 0.020 mmol) was then
added, and the reaction mixture was stirred at 08C for another 25 min.
After this time, the reaction mixture was poured into a mixture of ice
and water. The product was extracted with CH2Cl2, and the combined or-
ganic layers were washed with brine, dried (Na2SO4), and concentrated in
vacuo. Purification of the residue by flash column chromatography (hex-
anes/EtOAc 3:1) produced (S)-37 as a dark green gum and as a mixture
of keto and enol forms (6.0 mg, 39% yield from (8R)-35a). IR (KBr):
n˜ =2099, 2064, 2039, 1717, 1628, 1615 cmꢀ1; HRMS (FAB): m/z: calcd for
C35H32NO8S: 626.1849 [M+HꢀCo2(CO)6]+; found: 626.1837. 1H and
13C NMR spectra were complicated due to the presence of tautomers.
1H NMR (300 MHz, CDCl3) for the major isomer: d=2.42–2.55 (m, 1H),
2.49 (s, 3H), 3.06–3.20 (m, 3H), 3.79 (s, 3H), 3.90 (s, 3H), 3.95 (s, 3H),
4.04 (s, 3H), 4.27 (d, J=15.5 Hz, 1H,), 4.51 (d, J=15.5 Hz, 1H), 6.41 (s,
1H), 6.96 (s, 1H), 7.31 (s, 1H), 7.35–7.50 (m, 5H), 7.58 (s, 1H),
12.40 ppm (s, 1H).
(R)-4-Chloroacetoxy-8,9-[di(tert-butyl)silylenedioxy]-1’,5,6,9’-tetrameth-
oxy-3’-methyl-6’,7’-dihydrospiro[2H-benz[f]indene-2,8’-[8’H]cyclopent[g]-
isoquinoline]-1,3-dione ((R)-38): Under a nitrogen atmosphere, a solu-
tion of 18c (2.0 mL, 0.12 mmol) in anhydrous toluene (0.05 mL) and a so-
lution of anhydrous pTsOH (para-toluenesulfonic acid, 0.03 mg,
0.0002 mmol) in anhydrous toluene (0.05 mL) were added, in this order,
to (S)-5 (1.2 mg, 0.0015 mmol). The reaction flask was then set in an oil
bath, which had been pre-heated to 1108C, and stirred for 20 min. After
cooling, saturated aqueous NaHCO3 solution was added to the reaction
mixture, and the product was extracted with CH2Cl2. The combined or-
ganic layers were washed with brine, dried (Na2SO4), and concentrated in
vacuo. Purification of the residue by preparative TLC (hexanes/EtOAc
4:1) afforded (R)-38 (1.0 mg, 60% yield) as a yellow gum. The optical
purity of (R)-38 was determined to be 95% ee by chiral HPLC analysis
Some typical 1H NMR (300 MHz, CDCl3) data for minor isomers: d=
2.03 (s, 3H), 3.62 (s, 3H), 3.89 (s, 3H), 6.27 (s, 1H), 6.72 (s, 1H), 12.03 (s,
1H), 15.61 ppm (s, 1H).
(Daicel CHIRALCEL OD (hexanes/iPrOH 95:5, flow rate 1.0 mLminꢀ1
,
(S)-8,9-[Di(tert-butyl)silylenedioxy]-1’,5,6,9’-tetramethoxy-3’-methyl-4-
phenylthio-6’,7’-dihydrospiro[2H-benz[f]indene-2,8’-[8’H]cyclopent[g]iso-
208C). Retention time: 8.6 min for (S)-38 and 9.9 min for (R)-38. [a]D20
=
ꢀ23.5 (c=0.91 in CHCl3); IR (KBr): n˜ =1736, 1707, 1605, 1572 cmꢀ1
;
quinoline]-1,3-dione ((S)-6): Under
a nitrogen atmosphere, Et3N
1H NMR (300 MHz, CDCl3): d=1.12 (s, 9H), 1.17 (s, 9H), 2.47 (s, 3H),
2.47–2.60 (m, 2H), 3.36–3.44 (m, 2H), 3.51 (brs, 3H), 3.89 (s, 3H), 4.01
(s, 3H), 4.02 (s, 3H), 4.46 (d, J=15.5 Hz, 1H), 4.51 (d, J=15.5 Hz, 1H),
6.95 (s, 2H), 7.30 ppm (s, 1H); HRMS (FAB): m/z: calcd for
C39H4335ClNO10Si: 748.2344 [M+H]+; found: 748.2319.
(0.017 mL, 0.12 mmol) and Me2SiCl2 (0.074 mL, 0.058 mmol) were added
successively to a solution of (S)-37 (14 mg, 0.015 mmol) in anhydrous tol-
uene (2.0 mL). A pressure glass reactor was used, and the reaction mix-
ture was stirred at RT for 30 min, after which time chloranil (15 mg,
0.062 mmol) was added. The reactor was then sealed and heated at
1008C for 6 h. After cooling, the reaction mixture was concentrated in
vacuo and the resulting residue dissolved in anhydrous DMF (1.2 mL).
Et3N (0.077 mL, 0.54 mmol) and (tBu)2Si(OTf)2 (0.040 mL, 0.11 mmol)
were then added, and the reaction mixture was stirred at RT for 5 h.
After this time, iced water and saturated aqueous NaHCO3 solution were
added to the reaction mixture and the product was extracted with diethyl
ether. The combined organic layers were washed with brine, dried
(Na2SO4), and concentrated in vacuo. Purification of the residue by flash
(R)-4-Chloroacetoxy-8,9-[di(tert-butyl)silylenedioxy]-5,6,9’-trimethoxy-3’-
methyl-6’,7’-dihydrospiro[2H-benz[f]indene-2,8’-8’H-cyclopent[g]isoqui-
noline]-1,1’(2’H),3-trione ((R)-43): Under a nitrogen atmosphere, Et3N
(17 mL, 0.12 mmol), (tBu)2Si(OTf)2 (22 mL, 0.060 mmol), and MeI (4 mL,
0.060 mmol) were added, in this order, to a solution of (R)-38 (4.4 mg,
5.9 mmol) in anhydrous DMF (0.3 mL). The reaction mixture was then
stirred at RT for 28 h. After this time, water was added, and the product
was extracted with EtOAc. The combined organic layers were washed
Chem. Eur. J. 2005, 11, 6286 – 6297
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6295

Y. Kita et al.
with brine, dried (Na2SO4), and concentrated in vacuo. Purification of the
residue by flash column chromatography (hexanes/EtOAc 1:1) afforded
(R)-43 (4.3 mg, 99% yield) as a yellow gum. [a]2D3 =ꢀ27.5 (c=0.86 in
retson, T. T. Wei, K. M. Byrne, R. F. Geoghegan, Jr., R. J. White, J.
Antibiot. 1981, 34, 1389–1401; b) R. Misra, R. C. Pandey, J. V. Sil-
verton, J. Am. Chem. Soc. 1982, 104, 4478–4479; c) K. M. Byrne,
B. D. Hilton, R. J. White, R. Misra, R. C. Pandey, Biochemistry 1985,
24, 478–486; d) R. Misra, R. C. Pandey, B. D. Hilton, P. P. Roller,
J. V. Silverton, J. Antibiot. 1987, 40, 786–802; for biosynthesis see:
e) K. M. Byrne, B. D. Hilton, R. J. White, R. Misra, R. C. Pandey,
Biochemistry 1985, 24, 478–486.
CHCl3); IR (KBr): n˜ =1790, 1736, 1707, 1643, 1605, 1580 cmꢀ1 1H NMR
;
(300 MHz, CDCl3): d=1.12 (s, 9H), 1.16 (s, 9H), 2.24 (s, 3H), 2.45–2.59
(m, 2H), 3.27–3.43 (m, 2H), 3.58 (brs, 3H), 3.89 (s, 3H), 4.02 (s, 3H),
4.45 (d, J=15.5 Hz 1H), 4.52 (d, J=15.5 Hz, 1H), 6.18 (s, 1H), 6.94 (s,
1H), 7.12 (s, 1H), 9.12 ppm (brs, 1H); HRMS (FAB): m/z: calcd for
C38H4135ClNO10Si: 734.2188 [M+H]+; found: 734.2186.
[2] For studies on biological activity see: a) D. J. Warnick-Pickle, K. M.
Byrne, R. C. Pandey, R. J. White, J. Antibiot. 1981, 34, 1402–1407;
b) B. D. Hilton, R. Misra, J. L. Zweier, Biochemistry 1986, 25, 5533–
5539; c) R. Misra, J. Antibiot. 1988, 41, 976–981; d) M. D. Latham,
C. K. King, P. Gorycki, T. L. Macdonald, W. E. Ross, Cancer Chemo-
ther. Pharmacol. 1989, 24, 167–171; e) N. S. Dalal, X. Shi, Biochem-
istry 1989, 28, 748–750.
[3] a) T. R. Kelly, N. Ohashi, R. J. Armstrong-Chong, S. H. Bell, J. Am.
Chem. Soc. 1986, 108, 7100–7101; b) T. R. Kelly, S. H. Bell, N.
Ohashi, R. J. Armstrong-Chong, J. Am. Chem. Soc. 1988, 110, 6471–
6480.
[4] a) D. L. J. Clive, Y. Tao, A. Khodabocus, Y.-J. Wu, A. G. Angoh,
S. M. Bennett, C. N. Boddy, L. Bordeleau, D. Kellner, G. Kleiner,
D. S. Middleton, C. J. Nichols, S. R. Richardson, P. G. Vernon, J.
Chem. Soc. Chem. Commun. 1992, 1489–1490; b) D. L. J. Clive, Y.
Tao, A. Khodabocus, Y.-J. Wu, A. G. Angoh, S. M. Bennett, C. N.
Boddy, L. Bordeleau, D. Kellner, G. Kleiner, D. S. Middleton, C. J.
Nichols, S. R. Richardson, P. G. Vernon, J. Am. Chem. Soc. 1994,
116, 11275–11286; c) D. L. J. Clive, Y. Tao, A. Khodabocus, Y.-J.
Wu, A. G. Angoh, S. M. Bennett, C. N. Boddy, L. Bordeleau, M.
Cantin, D. R. Cheshire, D. Kellner, G. Kleiner, D. S. Middleton, C. J.
Nichols, S. R. Richardson, L. Set, P. G. Vernon, Stud. Nat. Prod.
Chem. 1995, 16, 27–74.
[5] a) A. V. Rama Rao, A. K. Singh, B. V. Rao, K. M. Reddy, Tetrahe-
dron Lett. 1993, 34, 2665–2668; b) A. V. Rama Rao, A. K. Singh,
B. V. Rao, K. M. Reddy, Heterocycles 1994, 37, 1893–1912.
[6] L. Saint-Jalmes, C. Lila, J. Z. Xu, L. Moreau, B. Pfeiffer, G. Eck, L.
Pelsez, C. Rolando, M. Julia, Bull. Soc. Chim. Fr. 1993, 130, 447–
449.
[7] J. A. Wendt, P. J. Gauvreau, R. D. Bach, J. Am. Chem. Soc. 1994,
116, 9921–9926.
[8] a) D. L. Boger, O. Hüter, K. Mbiya, M. Zhang, J. Am. Chem. Soc.
1995, 117, 11839–11849; b) D. L. Boger, J. Heterocycl. Chem. 1996,
33, 1519–1531.
[9] The synthetic studies on the partial structure of 1 have been sum-
marized in reference [8a]. For recent studies see: a) M. Braun, H.
Sacha, D. Galle, S. Baskaran, Pure Appl. Chem. 1996, 68, 561–564;
b) P. A. Evans, T. A. Brandt, Tetrahedron Lett. 1996, 37, 1367–1370;
c) P. T. Perumal, M. Venugopal, T. P. Velusamy, Indian J. Chem.
Sect. B 1996, 35, 242–243; d) D. L. J. Clive, X. Kong, C. C. Paul, Tet-
rahedron 1996, 52, 6085–6116; e) S. Baskaran, E. Nagy, M. Braun,
Liebigs Ann./Receuil 1997, 311–312; f) R. R. Sudini, U. R. Khire,
R. K. Pandey, P. Kumar, Indian J. Chem. Sect. B 2000, 39, 738–745;
g) D. Bouyssi, G. Balme, Synlett 2001, 1191–1193; h) W. H. Moser, J.
Zhang, C. S. Lecher, T. L. Frazier, M. Pink, Org. Lett. 2002, 4, 1981–
1984.
(R)-4-Chloroacetoxy-8,9-[di(tert-butyl)silylenedioxy]-3’-formyl-5,6,9’-tri-
methoxy-6’,7’-dihydrospiro[2H-benz[f]indene-2,8’-8’H-cyclopent[g]isoqui-
noline]-1,1’(2’H),3-trione ((R)-44): Under a nitrogen atmosphere, a mix-
ture of (R)-43 (4.3 mg, 5.9 mmol) and SeO2 (2.6 mg, 23 mmol) in anhy-
drous dioxane (0.4 mL) was stirred under reflux for 1 h. After cooling,
the reaction mixture was concentrated in vacuo. Purification of the resi-
due by flash column chromatography (hexanes/EtOAc 1:1) afforded (R)-
44 (4.0 mg, 91% yield) as a yellow gum. [a]2D4 =ꢀ19.1 (c=0.80 in CHCl3);
IR (KBr): n˜ =1788, 1734, 1707, 1659, 1605, 1580 cmꢀ1
;
1H NMR
(300 MHz, CDCl3): d=1.12 (s, 9H), 1.17 (s, 9H), 2.52–2.67 (m, 2H),
3.34–3.48 (m, 2H), 3.62 (brs, 3H), 3.89 (s, 3H), 4.02 (s, 3H), 4.46 (d, J=
15.5 Hz, 1H), 4.51 (d, J=15.5 Hz, 1H,), 6.96 (s, 1H), 7.04 (s, 1H), 7.42 (s,
1H), 8.70 (brs, 1H), 9.53 ppm (s, 1H); HRMS (FAB): m/z: calcd for
C38H3935ClNO11Si: 748.1981 [M+H]+; found: 748.1976.
(S)-Fredericamycin A (1): Under a nitrogen atmosphere, nBuLi (1.6m so-
lution in hexane, 0.051 mL, 0.082 mmol) was added to an ice-cooled sus-
pension of crotyltriphenylphosphonium bromide (43 mg, 0.11 mmol) in
anhydrous THF (2.4 mL), and the reaction mixture was stirred at 08C for
15 min. This mixture was then added to a solution of (R)-44 (4.0 mg,
5.3 mmol) in anhydrous THF (0.5 mL) at ꢀ788C. The reaction mixture
was stirred at ꢀ788C for 30 min, warmed to RT, and then stirred for 1 h.
After this time, saturated aqueous NH4Cl was added, and the product
was extracted with EtOAc. The combined organic layers were washed
with brine, dried (Na2SO4), and concentrated in vacuo to afford a mixture
of (R)-45 and a few partially deprotected products.
Under a nitrogen atmosphere, the above crude mixture was dissolved in
anhydrous CH2Cl2 (1.0 mL), and the solution was cooled to ꢀ788C. BBr3
(1.0m solution in CH2Cl2, 0.10 mL, 0.10 mmol) was added, and the reac-
tion mixture was stirred at ꢀ788C for 4 h. The reaction mixture was then
warmed to RT and stirred overnight. After this time, water was added to
the reaction mixture, and the product was extracted with CH2Cl2. The
combined organic layers were washed with brine, dried (Na2SO4), and
concentrated in vacuo. THF (5 mL) and water (1 mL) were added to the
resulting residue, and the reaction mixture was stirred, open to the air, at
RT for 24 h. After this time, the product was extracted with CH2Cl2, and
the combined organic layers were washed successively with brine, dried
(Na2SO4), and concentrated in vacuo. Purification of the residue by prep-
arative TLC (CHCl3/MeOH/AcOH 90:10:1) afforded a mixture of 1 and
its E,Z isomer. Further purification of this mixture by HPLC (JASCO
Megapac SIL NH2–10, CHCl3/MeOH/AcOH 800:200:1) afforded
1
(1.0 mg, 35% from (R)-44) as a red solid. This product was found to be
identical (m.p., 1H NMR, IR, UV, and CD spectroscopy, HPLC, TLC)
with an authentic sample of natural fredericamycin A. HRMS (FAB):
m/z: calcd for C30H22NO9: 540.1294 [M+H]+; found: 540.1274.
[10] a) Y. Kita, K. Higuchi, Y. Yoshida, K. Iio, S. Kitagaki, K. Ueda, S.
Akai, H. Fujioka, J. Am. Chem. Soc. 2001, 123, 3214–3222; for
model studies see: b) Y. Kita, S. Kitagaki, R. Imai, S. Okamoto, S.
Mihara, Y. Yoshida, S. Akai, H. Fujioka, Tetrahedron Lett. 1996, 37,
1817–1820; c) K. Iio, N. G. Ramesh, A. Okajima, K. Higuchi, H. Fu-
jioka, S. Akai, Y. Kita, J. Org. Chem. 2000, 65, 89–95; d) K. Iio, A.
Okajima, Y. Takeda, K. Kawaguchi, B. A. Whelan, S. Akai, Y. Kita,
ARKIVOC 2003, 8, 144–162.
Acknowledgements
We thank Akiko Okajima and Seiko Morishita for their technical assis-
tance. This work was supported by Grants-in-Aid for Scientific Research
(S and C) from the Ministry of Education, Culture, Sports, Science and
Technology (Japan). Meito Sangyo (Japan) and Toyobo (Japan) are
thanked for the generous gift of the lipases, and also Dr. Hiroshi Hasega-
wa (SS Pharmaceutical (Japan)) for generously providing us with an au-
thentic sample of fredericamycin A.
[11] For reviews see: a) Y. Kita, S. Akai, H. Fujioka, Yuki Gosei Kagaku
Kyokaishi 1998, 56, 963–974; b) Y. Kita, Yakugaku Zasshi 2002, 122,
1011–1035.
[12] a) Y. Kita, K. Iio, K. Kawaguchi, N. Fukuda, Y. Takeda, H. Ueno, R.
Okunaka, K. Higuchi, T. Tsujino, H. Fujioka, S. Akai, Chem. Eur. J.
2000, 6, 3897–3905; for model studies, see: b) Y. Kita, R. Okunaka,
T. Honda, M. Shindo, O. Tamura, Tetrahedron Lett. 1989, 30, 3995–
[1] For isolation and structure elucidation see: a) R. C. Pandey, M. W.
Toussaint, R. M. Stroshane, C. C. Kalita, A. A. Aszalos, A. L. Gar-
6296
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2005, 11, 6286 – 6297

Total Synthesis of Fredericamycin A
FULL PAPER
3998; c) Y. Kita, R. Okunaka, T. Honda, M. Kondo, O. Tamura, Y.
Tamura, Chem. Pharm. Bull. 1991, 39, 2106–2114; d) Y. Kita, H.
Ueno, S. Kitagaki, K. Kobayashi, K. Iio, S. Akai, J. Chem. Soc.
Chem. Commun. 1994, 701–702; e) S. Akai, K. Iio, Y. Takeda, H.
Ueno, Y. Kita, Synlett 1997, 310–312.
Bull. 1988, 36, 4337–4344; c) G. Egri, J. Bꢂlint, R. Peredi, E. Fogas-
sy, L. Novꢂk, L. Poppe, J. Mol. Catal. B 2000, 10, 583–596.
[24] For a review see: a) S. Akai, Yakugaku Zasshi 2003, 123, 919–931.
For typical papers see: b) S. Akai, T. Naka, T. Fujita, Y. Takebe, Y.
Kita, Chem. Commun. 2000, 1461–1462; c) S. Akai, T. Naka, T.
Fujita, Y. Takebe, T. Tsujino, Y. Kita, J. Org. Chem. 2002, 67, 411–
419; d) S. Akai, T. Tsujino, E. Akiyama, K. Tanimoto, T. Naka, Y.
Kita, J. Org. Chem. 2004, 69, 2478–2486. See also: e) S. Akai, T.
Naka, Y. Takebe, Y. Kita, Chem. Pharm. Bull. 2000, 48, 1519–1523.
[25] For the preparation of 9a see Supporting Information.
[26] Other Candida lipases (Amano AY, Meito OF, Roche Chirazyme
L3, and Roche Chirazyme L5) also catalyzed the desymmetrization
reaction but with similar or decreased enantioselectivities and chem-
ical yields. The Pseudomonas lipases were not reactive.
[13] For
a preliminary study on the asymmetric synthesis of the
ABCDE-ring analogues see: S. Akai, T. Tsujino, N. Fukuda, K. Iio,
Y. Takeda, K. Kawaguchi, T. Naka, K. Higuchi, Y. Kita, Org. Lett.
2001, 3, 4015–4018.
[14] a) M. Toyota, S. Terashima, Tetrahedron Lett. 1989, 30, 829–832;
b) W. Trypke, A. Steigel, M. Braun, Synlett 1992, 827–829.
[15] Y. Tamura, M. Sasho, K. Nakagawa, T. Tsugoshi, Y. Kita, J. Org.
Chem. 1984, 49, 473–478.
[16] For reviews see: a) Y. Tamura, Y. Kita, Yuki Gosei Kagaku Kyo-
kaishi 1988, 46, 205–217; b) Y. Kita, Y. Takeda, Kagaku to Kogyo
(Osaka, Jpn.) 1997, 71, 298–309; c) M. Kirihara, Y. Kita, Heterocy-
cles 1997, 46, 705–726.
[27] The optical purities of (R)-36 for each ratio of iPr2O to MeCN were
as follows: 68% ee (5:1), 67% ee (1:1), 44% ee (1:5), and 28% ee
(1:10).
[17] For a review see: a) H. Fujioka, Y. Yoshida, Y. Kita, Yuki Gosei
Kagaku Kyokaishi 2003, 61, 133–143. For typical papers see: b) H.
Fujioka, S. Kitagaki, R. Imai, M. Kondo, S. Okamoto, Y. Yoshida, S.
Akai, Y. Kita, Tetrahedron Lett. 1995, 36, 3219–3222; c) Y. Kita, S.
Kitagaki, R. Imai, S. Okamoto, S. Mihara, Y. Yoshida, S. Akai, H.
Fujioka, Tetrahedron Lett. 1996, 37, 1817–1820; d) Y. Kita, S. Kita-
gaki, Y. Yoshida, S. Mihara, D.-F. Fang, H. Fujioka, Tetrahedron
Lett. 1997, 38, 1061–1064; e) Y. Kita, S. Kitagaki, Y. Yoshida, S.
Mihara, D.-F. Fang, M. Kondo, S. Okamoto, R. Imai, S. Akai, H. Fu-
jioka, J. Org. Chem. 1997, 62, 4991–4997; f) Y. Kita, Y. Yoshida, S.
Kitagaki, S. Mihara, D.-F. Fang, A. Furukawa, K. Higuchi, H. Fujio-
ka, Tetrahedron 1999, 55, 4979–4998; g) Y. Kita, A. Furukawa, J.
Futamura, K. Higuchi, K. Ueda, H. Fujioka, Tetrahedron Lett. 2000,
41, 2133–2136.
[18] For comprehensive reviews on the construction of stereogenic qua-
ternary carbon centers see: a) K. Fuji, Chem. Rev. 1993, 93, 2037–
2066; b) E. J. Corey, A. Guzman-Perez, Angew. Chem. 1998, 110,
402–415; Angew. Chem. Int. Ed. 1998, 37, 388–401; c) J. Christoff-
ers, A. Mann, Angew. Chem. 2001, 113, 4725–4732; Angew. Chem.
Int. Ed. 2001, 40, 4591–4597; d) I. Denissova, L. Barriault, Tetrahe-
dron 2003, 59, 10105–10146.
[28] As we have not determined the absolute stereochemistry at the C1’
position, it has been tentatively assigned for convenience (see
schemes).
[29] We have also observed a similar racemization of the stereogenic
quaternary a-carbon center in the b-hydroxyketone see referen-
ce [10a]; in addition, a similar racemization of a b-hydroxynitrile
has been reported see: T. Kimura, N. Yamamoto, Y. Suzuki, K.
Kawano, Y. Norimine, K. Ito, S. Nagato, Y. Iimura, M. Yonaga, J.
Org. Chem. 2002, 67, 6228–6231.
[30] The conversion of (1R)-40a to the trione (S)-41 was carried out as
reported see reference [13]. The same sequential treatment of (ꢁ)-
39b produced (ꢁ)-41 (52% overall yield), which provided us with
additional evidence for the diastereomeric relationship between
(1R)-39a and (ꢁ)-39b (Scheme 12).
[19] a) Y. Kita, Y. Takeda, K. Iio, K. Yokogawa, K. Takahashi, S. Akai,
Tetrahedron Lett. 1996, 37, 7545–7548; b) S. Akai, Y. Takeda, K. Iio,
K. Takahashi, N. Fukuda, Y. Kita, J. Org. Chem. 1997, 62, 5526–
5536; c) Y. Kita, Y. Takeda, M. Matsugi, K. Iio, K. Gotanda, K.
Murata, S. Akai, Angew. Chem. 1997, 109, 1525–1527; Angew.
Chem. Int. Ed. Engl. 1997, 36, 1529–1531.
[20] For a recent review on the chemistry of the RAMP- ((R)-1-amino-2-
(methoxymethyl)pyrrolidine) and SAMP- ((S)-1-amino-2-(methoxy-
methyl)pyrrolidine) hydrozones see: A. Job, C. F. Janeck, W. Bet-
tray, R. Peters, D. Enders, Tetrahedron 2002, 58, 2253–2329.
[21] For some examples see: a) A. Fadel, L. Vandromme, Tetrahedron:
Asymmetry 1999, 10, 1153–1162; b) T. Honda, T. Koizumi, Y. Ko-
matsuzaki, R. Yamashita, K. Kanai, H. Nagase, Tetrahedron: Asym-
metry 1999, 10, 2703–2712; c) F.-R. Alexandre, S. Legoupy, F. Huet,
Tetrahedron 2000, 56, 3921–3926; d) A. Salgado, T. Huybrechts, A.
Eeckhaut, J. van der Eycken, Z. Szakonyi, F. Fülçp, A. Tkachev, N.
De Kimpe, Tetrahedron 2001, 57, 2781–2786; e) B. L. Kedrowski, J.
Org. Chem. 2003, 68, 5403–5406.
Scheme 12. The conversion of (1R)-39a and (ꢁ)-39b to (S)-41 and (ꢁ)-
41, respectively a) LiN(TMS)2, THF; b) Moffatt oxidation; c) 34, DMAP,
CH2Cl2.
[31] For examples see references [9a,b,e,h]; see also, S. V. Kessar, R.
Vohra, N. P. Kaur, K. N. Singh, P. Singh, J. Chem. Soc. Chem.
Commun. 1994, 1327–1328.
[32] N. Shibata, M. Matsugi, N. Kawano, S. Fukui, C. Fujimori, K. Go-
tanda, K. Murata, Y. Kita, Tetrahedron: Asymmetry 1997, 8, 303–
310.
[22] For related reviews on the enzymatic desymmetrization of symmet-
ric compounds see: a) E. Schoffers, A. Golebiowski, C. R. Johnson,
Tetrahedron 1996, 52, 3769–3826; b) C. R. Johnson, Acc. Chem. Res.
1998, 31, 333–341.
[33] a) A. Herunsalee, M. Isobe, T. Goto, Tetrahedron 1991, 47, 3727–
3736; b) P. A. Magriotis, J. T. Brown, Org. Synth. 1993, 72, 252–263.
[34] F. Bohlmann, H.-J. Mannhardt, Chem. Ber. 1956, 89, 1307–1315.
[23] a) A. Fadel, P. Arzel, Tetrahedron: Asymmetry 1997, 8, 283–291. See
also: b) H. Suemune, T. Harabe, Z.-F. Xie, K. Sakai, Chem. Pharm.
Received: April 20, 2005
Published online: August 1, 2005
Chem. Eur. J. 2005, 11, 6286 – 6297
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6297