Total synthesis of (+)-isatisine A

Article abstract of DOI:10.1021/jo101209y

The asymmetric total synthesis of (+)-isatisine A has been accomplished commencing with a Lewis acid-catalyzed cyclization of homochiral (S)-vinylcyclopropane diester and N-tosylindole-2-carboxaldehyde to construct the tetrahydrofuran ring. A palladium-ca

Full text of DOI:10.1021/jo101209y

pubs.acs.org/joc
Total Synthesis of (þ)-Isatisine A
Avedis Karadeolian and Michael A. Kerr*
Department of Chemistry, The University of Western Ontario, 1151 Richmond Street, London, Ontario,
Canada N6A 5B7
makerr@uwo.ca
Received June 21, 2010
The asymmetric total synthesis of (þ)-isatisine A has been accomplished commencing with a Lewis
acid-catalyzed cyclization of homochiral (S)-vinylcyclopropane diester and N-tosylindole-2-carbox-
aldehyde to construct the tetrahydrofuran ring. A palladium-catalyzed oxidative decarboxylation
was utilized to obtain the dihydrofuran required for the subsequent dihydroxylation reaction to
install the diol present on the tetrahydrofuran ring. The total synthesis was completed by an indole
oxidation and electrophilic aromatic substitution sequence to construct isatisine A acetonide, which
was then carried forward to the antipode of the natural product. The absolute configuration of the
natural enantiomer (-)-isatisine A was determined to be C2(S), C9(R), C10(S), C12(R), and C13(R).
Introduction
Isatisine A (1) (Figure 1) is a complex bisindole natural
product isolated from the leaves of Isatis indigotica Fort, a
shrub prevalent in the Anhui province of China.1 The roots
and leaves of I. indigotica have been used in traditional
Chinese medicine for the treatment of viral diseases including
influenza, viral pneumonia, mumps, and hepatitis.2
During the fractionation process, Isatis indigotica Fort leaf
extracts were eluted with acetone on silica gel, which yielded
isatisine A acetonide (2), initially believed to be the natural
product; however, it was likely an artifact of the isolation
process. Suspecting this, isatisine A (1) was prepared by
hydrolysis of 2 and matched with an HPLC trace of a crude
extract, leading to the supposition that 1 and not 2 was in fact
the biogenetic product. Although the relative stereochemistry
of 2 (and consequently 1) was confirmed by single crystal X-ray
diffraction studies of acetonide 2, the absolute stereochemistry
remained unknown. The biological activity of isatisine A (1)
could not be determined due to the small amount obtained
FIGURE 1. Structure of isatisine A (1) and its acetonide 2.
from hydrolysis of the acetonide 2; however, 2 has been shown
to possess anti-HIV-1IIIB activity with an EC50 = 37.8 μM.1 An
enantioselective total synthesis of the natural product from a
starting material with known absolute stereochemistry would
allow for the determination of the absolute configuration of
isatisine A and would also provide material for biological
testing. Our initial communication on the total synthesis of
(þ)-isatisine A (1) outlined our successful route, herein we
report our overall findings.3
Our interest in isatisine A stemmed from a natural product
previously synthesized in our group, namely mersicarpine (5).4
In the final stages of the total synthesis of mersicarpine, indole 3
(1) Liu, J.-F.; Jiang, Z.-Y.; Wang, R.-R.; Zheng, Y.-T.; Chen, J.-J.;
Zhang, X.-M.; Ma, Y.-B. Org. Lett. 2007, 9, 4127–4129.
(2) Zheng, H. Z.; Dong, Z. H.; Yu, Q. Modern Study of Traditional
Chinese Medicine; Xueyuan Press: Beijing, China, 1997; Vol. 1, pp 328-334.
(3) Karadeolian, A.; Kerr, M. A. Angew. Chem., Int. Ed. 2010, 49, 1133–1135.
(4) Magolan, J.; Carson, C. A.; Kerr, M. A. Org. Lett. 2008, 10, 1437–1440.
6830 J. Org. Chem. 2010, 75, 6830–6841
Published on Web 09/21/2010
DOI: 10.1021/jo101209y
r
2010 American Chemical Society

Karadeolian and Kerr
JOCArticle
SCHEME 1. Final Steps in the Synthesis of Mersicarpine
SCHEME 2. Proposed Late Stage Route to Isatisine A (1) from Advanced Indole Substrate 6
was oxidized to indoxyl 4 followed by imine formation to yield 5
(Scheme 1). When comparing intermediate 4 and isatisine A, the
common indoxyl moiety becomes apparent. Thus, we envisioned
that isatisine A could potentially be accessed through oxidation
of indole 6 to the indoxyl 7 (by a similar method to that used for
mersicarpine), followed by indole addition into this species
(Scheme 2).
Oxidation of the indole C2-C3 double bond has been well
documented, where the identities of the oxidation products
depend on the nature of the substituents on the N1, C2, and
C3 positions of the indole. Reagents that have been shown
to oxidize indole include (hexamethylphosphoramide)oxo-
diperoxomolybdenum(VI) (MoO5 HMPA),5 bis(acetyl-
3
acetonato)oxovanadium(IV),6 dimethyl dioxirane,7 singlet
oxygen,8 and m-CPBA.9 Important examples within these
studies, which are pertinent to our proposed route, are instances
where indole oxidation was followed by electrophilic aromatic
substitution (EAS) on a second indole molecule with the indoxyl
species (Figure 2). Initial reports of such processes came from
work by Sakamoto using MoO5 HMPA to oxidize 2-phenylin-
3
dole (eq 1, Figure 2).5c Speier has reported the synthesis of a simi-
lar dimeric species when oxidizing 2-methylindole with oxygen in
the presence of a vanadium catalyst (eq 2, Figure 2).7 Finally
Jimenez, working with oxodiperoxomolybdenum oxidants with
trialkylphosphine oxide ligands, has reported the desired EAS
in many instances (eqs 3 and 4, Figure 2).10 Jimenez’ studies
demonstrated that indoles containing N-alkyl subtituents under-
go the EAS reaction, while N-acyl indoles do not (eq 5, Figure 2).
The rationale used to explain this observation relies on the
FIGURE 2. Examples of indole oxidation with concomitant EAS
of a second indole molecule.
(5) (a) Kawasaki, T.; Chien, C. S.; Sakamoto, M. Chem. Lett. 1983, 855–858.
(b) Chien, C. S.; Suzuki, T.; Kawasaki, T.; Sakamoto, M. Chem. Pharm. Bull.
1984, 32, 3945–3951. (c) Chien, C. S.; Takanami, T.; Kawasaki, T.; Sakamoto,
M. Chem. Pharm. Bull. 1985, 33, 1843–1848. (d) Chien, C. S.; Hasegawa, A.;
Kawasaki, T.; Sakamoto, M. Chem. Pharm. Bull. 1986, 34, 1493–1496. (e) Wang,
Z.; Jimenez, L. S. Tetrahedron Lett. 1996, 37, 6049–6052.
(6) Balogh-Hergovich, E.; Speier, G. J. Mol. Catal. 1989, 57, L9–L12.
(7) (a) Zhang, X.; Foote, C. S. J. Am. Chem. Soc. 1993, 115, 8867–8868.
(b) Colandrea, V. J.; Rajaraman, S.; Jimenez, L. S. Org. Lett. 2003, 5, 785–787.
(8) (a) Carloni, P.; Eberson, L.; Greci, L.; Stipa, P.; Tosi, G. J. Chem.
Soc., Perkin Trans. 2 1991, 1779–1783. (b) Zhang, X. J.; Foote, C. S.; Khan,
S. I. J. Org. Chem. 1993, 58, 47–51. (c) Ling, K.-Q. Synth. Commun. 1995, 25,
3831–3835. (d) Ling, K. Q. Synth. Commun. 1996, 26, 149–152.
(9) Buller, M. J.; Cook, T. G.; Kobayashi, Y. Heterocycles 2007, 72, 163–166.
(10) Altinis Kiraz, C. I.; Emge, T. J.; Jimenez, L. S. J. Org. Chem. 2004,
69, 2200–2202.
formation of an iminium ion 18 upon oxidation of indole 14
(Scheme 3). This iminium intermediate can be attacked by a
second indole moiety to form dimer 15. An electron-withdraw-
ing substituent on the nitrogen, such as an acyl group, would
inhibit the formation of the iminium intermediate, thereby
preventing the dimerization process.
Results and Discussion
To demonstrate the feasibility of the indole oxidation and
EAS process toward the synthesis of isatisine A (1), two
J. Org. Chem. Vol. 75, No. 20, 2010 6831

Karadeolian and Kerr
JOCArticle
afforded indoxyl 26 as the methyl aminal in moderate yield
(Scheme 4). Stronger oxidants such as m-CPBA or in situ
generated DMDO did not furnish any indoxyl species.
Gratifyingly, EAS of indoxyl 26 with indole took place under
very mild reaction conditions and in high yield to produce
compound 28.14 Construction of the pyrroloindolone ring
was more difficult than anticipated; standard conditions for
direct lactam formation proved unsuccessful until a two-step
procedure involving hydrolysis of the ester to carboxylic acid
29 followed by EDC-induced cyclization was able to furnish
the model compound 21 in 66% yield over the two steps.
With the successful demonstration of an oxidation/EAS
sequence we focused our efforts on the construction of the
indole substrate required for the natural product itself.
As described earlier, our final disconnection involves
oxidation of indole 6 followed by an electrophilic aromatic
substitution onto indole. Scheme 5 shows a full retrosynthe-
sis of isatisine A. Indole 6 would arise from dihydroxylation
of an R,β-unsaturated ester, which would be obtained from a
compound such as 30. The unit of unsaturation would be
installed by removal of an ester and oxidation, either in a
multistep fashion or in one single transformation. Tetrahy-
drofuran 30 would be obtained by Johnson’s15 Lewis acid-
mediated cyclization of indole-2-carboxaldehyde 31 and a
suitable cyclopropane diester 32.
Three 2-substituted 1,1-cyclopropane diesters were se-
lected as possible substrates for the synthesis of isatisine A.
We first investigated the reactivity of cyclopropane 32a,16
which contains the primary alcohol present in the natural
product protected as the benzyl ether (Table 1). Cyclizations
with this cyclopropane and electron-rich indole 31a led to
decomposition of the indole with recovery of the cyclopro-
pane diester (entry 1). When an electron-poor indole 31b was
subjected to the same reaction conditions, a very low yield of
product was obtained as a 2:1 mixture of 2,5-cis:2,5-trans
isomers (entry 2). Attempts at increasing this yield by either
increasing the catalyst loading (not shown) or by raising the
reaction temperature led to decomposition of the reagents
with no product formation (entry 3). Attempts with Sn(OTf)2
also led to substrate decomposition under the reaction condi-
tions (entry 4).
FIGURE 3. Substrates for model study and model target com-
pound 21.
SCHEME 3. Rationale Explaining Dimerization Process
model studies were conducted to form the western half of 1
by this sequence (Figure 3). In our first study, pyrroloindo-
lone 19 was synthesized and carried onto model substrate 21.
In our second study, 2-substituted indole 20 was utilized to
construct model compound 21.
The oxidation and EAS sequence with pyrroloindolone 1911
was investigated with various oxidizing reagents (Figure 4).
Oxidation of 19 with in situ generated DMDO led to complete
cleavage of the indole C2-C3 bond, producing carboxylic acid
22 in high yield (Figure 4, eq 1).4 Attempts at preventing the over
oxidation by lowering the reaction temperature were unsuccess-
ful due to insolubility of substrate 19. Use of the milder Vedejs’
reagent (MoO5 HMPA Py)12 yielded a mixture of three com-
3
3
pounds over a week-long reaction time (Figure 4, eq 2). Our
desired indoxyl product 23 accompanied by its methyl aminal 25
were isolated in 17% and 7.6% yield, respectively, with the
major product of this reaction consisting of compound 24 in
25% yield. Oxidation of 19 with oxodiperoxomolybdenum
bis(triphenylphosphine oxide) complex led to exclusive forma-
tion of compound 24 in 48% yield over a shorter 16 h reaction
time (Figure 4, eq 3). We were able to further oxidize this species
to indoxyl 25 by treatment with IBX in DMSO. Although EAS
with indole was not attempted with indoxyl 25, this two-step
oxidation sequence was viewed as an alternative, should the
direct oxidation of an indole to the respective indoxyl be unsuc-
cessful during our efforts toward isatisine A.
We then turned to Yadav’s silylmethyl-substituted cyclo-
propane diester 32b.17 Yadav has demonstrated the success-
ful cyclization of N-Boc-indole-3-carboxaldehyde with this
substrate. The three aldehydes (entries 5-7) we subjected to
his reaction conditions did not yield any product. Perform-
ing the reaction at increased temperatures led to decomposi-
tion of the cyclopropane diester, as did the use of a stronger
Lewis acid (entries 8 and 9).
We were delighted to find that oxidation of 19 with DMDO
afforded the desired indoxyl 23 as the sole product (Figure 4,
eq 4). Treatment of 23 with methanesulfonyl chloride and
triethylamine in the presence of indole facilitated the EAS
process to yield a 2:1 mixture of the model compound 21 and
indoxyl starting material 23, respectively. This reaction was not
optimized since the purpose of this study was to test the
feasibility of the indole oxidation and EAS sequence.
To obviate some of the difficulties encountered in the
oxidation of pyrroloindolone 19, an alternative oxidation
substrate, indole 20 was synthesized though a modification
of a literature procedure beginning with o-nitrotoluene and
succinic anhydride.13 Oxidation of 20 with Vedejs’ reagent
(14) Ling, K.-Q. Synth. Commun. 1996, 26, 149–152.
(15) (a) Pohlhaus, P. D.; Johnson, J. S. J. Org. Chem. 2005, 70, 1057–
1059. (b) Pohlhaus, P. D.; Johnson, J. S. J. Am. Chem. Soc. 2005, 127, 16014–
16015. (c) Pohlhaus, P. D.; Sanders, S. D.; Parsons, A. T.; Li, W.; Johnson,
J. S. J. Am. Chem. Soc. 2008, 130, 8642–8650. (d) Parsons, A. T.; Campbell,
M. J.; Johnson, J. S. Org. Lett. 2008, 10, 2541–2544. (e) Parsons, A. T.;
Johnson, J. S. J. Am. Chem. Soc. 2009, 131, 3122–3123. (f) Campbell, M. J.;
Johnson, J. S. J. Am. Chem. Soc. 2009, 131, 10370–10371. (g) Sanders, S. D.;
Ruiz-Olalla, A.; Johnson, J. S. Chem. Commun. 2009, 7601–7601.
(16) Burgess, K.; Ke, C.-Y. Synthesis 1996, 12, 1463–1467.
(17) (a) Yadav, V. K.; Sriramurthy, V. Org. Lett. 2004, 6, 4495–4498. (b)
Yadav, V. K.; Sriramurthy, V. Angew. Chem., Int. Ed. 2004, 43, 2669–2671.
(c) Yadav, V. K.; Kumar, N. V.; Parvez, M. Chem. Commun. 2007, 2281–
2283. (d) Yadav, V. K.; Gupta, A. Tetrahedron Lett. 2008, 49, 3212–3215. (e)
Gupta, A.; Yadav, V. K. Tetrahedron Lett. 2006, 47, 8043–8047. (f) Yadav,
V. K.; Gupta, A. Tetrahedron Lett. 2009, 50, 4647–4650.
(11) Crenshaw, M. D.; Zimmer, H. J. Heterocycl. Chem. 1984, 21, 623–624.
(12) Vedejs, E. J. Am. Chem. Soc. 1974, 96, 5944–5946.
(13) Arcari, M.; Aveta, R.; Brandt, A.; Cecchetelli, L.; Corsi, G. B.;
Dirella, M. Gazz. Chim. Ital. 1991, 121, 499–504.
6832 J. Org. Chem. Vol. 75, No. 20, 2010

Karadeolian and Kerr
JOCArticle
FIGURE 4. Oxidation of pyrroloindolone 19 leading to model compound 21.
SCHEME 4. Synthesis of Model Compound 21 from N-H In-
dole 20
SCHEME 5. Retrosynthetic Analysis of Isatisine A
indole-2-carboxaldehydes containing electron-withdrawing
protecting groups on the nitrogen were good substrates for
the cyclization reaction with these vinylcyclopropane die-
sters (entries 12-14). Under SnCl4 catalysis a 5:2 mixture of
2,5-cis:2,5-trans isomers was obtained in good yield. An
attempt at increasing the diastereoselectivity of the reaction
by lowering the reaction temperature was unsuccessful; the
diastereoselectivity remained unchanged but the yield was
substantially lowered (entry 13).
With suitable substrates and reaction conditions in hand for
the tetrahydrofuran formation, we moved forward with our
synthetic efforts toward isatisine A. Treatment of cyclopropane
diester 32c and aldehyde 31b with catalytic SnCl4 delivered
tetrahydrofuran 30a in good yield and acceptable diastereos-
electivity (Scheme 6). Unfortunately, the diastereomers were
inseparable at this point, and were carried forward as a mix-
ture. Krapcho demethoxycarbonylation successfully removed
Cyclopropanes containing a vinyl substituent (32c,d) were
then screened with various aldehydes and reactions condi-
tions (entries 10-14). As seen before, electron-rich indoles
were not tolerated under the Lewis acidic conditions and
decomposed when used alongside vinylcyclopropane diester
32c. N-Boc-indole also decomposed due to cleavage of the
Boc group, which is known to occur under these reaction
conditions (entry 11).18 We were delighted to find that
(18) Bose, D. S.; Kumar, K. K.; Reddy, A. V. N. Synth. Commun. 2003,
33, 445–450.
J. Org. Chem. Vol. 75, No. 20, 2010 6833

Karadeolian and Kerr
JOCArticle
TABLE 1. Attempts at Cyclization Reaction To Form Tetrahydrofuran Core 30
entry
R1
R2
R3
conditions
outcome
1
2
3
4
5
6
7
8
9
10
11
12
13
14
CH2OBn
Me (32a)
Me
Me
Bn (31a)
Ts (31b)
Ts (31b)
Ts (31b)
H (31c)
Bn (31a)
Ts (31b)
Ts (31b)
Ts (31b)
Bn (31a)
Boc (31d)
Ts (31b)
Ts (31b)
Ts (31b)
SnCl4, DCE, 45 °C
SnCl4, DCE, 55 °C
SnCl4, toluene, 110 °C
Sn(OTf)2, toluene, 110 °C
Sn(OTf)2, CH2Cl2, rt
Sn(OTf)2, CH2Cl2, rt
Sn(OTf)2, CH2Cl2, rt
Sn(OTf)2, CH2Cl2, 42 °C
SnCl4, CH2Cl2, rt
SnCl4, CH2Cl2, rt
Sn(OTf)2, CH2Cl2, rt
SnCl4, CH2Cl2, rt
SnCl4, CH2Cl2, 0 °C
SnCl4, CH2Cl2, rt
aldehyde decomposition
25% yield, 2:1 (cis:trans)
decomposition
Me
decomposition
CH2TBDPS
Me (32b)
Me
Me
Me
Me
Me (32c)
Me
Me
Me
Allyl (32d)
recovered starting material
recovered starting material
recovered starting material
cyclopropane decomoposition
cyclopropane decomoposition
aldehyde decomposition
aldehyde decomposition
74% yield, 5:2 (cis:trans)
55% yield, 5:2 (cis:trans)
81% yield, 2.4:1 (cis:trans)
vinyl
vinyl
SCHEME 6. Initial Synthetic Attempt
upcoming dihydroxylation by installation of a leaving group
followed by elimination. Unfortunately, the requisite leaving
group could not be installed in our substrate in this manner.
Treatment of 37 with strong base led to instantaneous decom-
position of the material, even at low temperatures. Attempts at
generating the silyl-enol ether of 37 followed by bromination or
oxidation were also unsuccessful.
Rather than removing one of the esters followed by
oxidation, hydrolysis of one of the esters to its carboxylic
acid would yield a substrate that could undergo a radical
decarboxylation followed by trapping of the resulting radical
with a halide. The synthetic sequence to diester 43 is shown in
Scheme 7. Hydrolysis of 43 under various conditions was
unsuccessful due to the poor substrate solubility. It was at
this junction in our synthetic efforts toward isatisine A that
Tsuji’s oxidative decarboxylation of allyl β-keto esters19 was
brought to our attention (Scheme 8).20
To apply the palladium-catalyzed oxidative decarboxylation
reaction to the tetrahydrofuran system, an allyl ester would be
required. The previous study with the dimethyl ester substrate 43
(Scheme 7) showed hydrolysis to be difficult. A solution to this
issue was to incorporate the allyl ester into the cyclopropane
prior to tetrahydrofuran formation. Thus, cyclopropane 32d
was prepared, as a 5:1 diastereomeric mixture, in a two-step
sequence from the dimethyl ester by mono-saponification fol-
lowed by alkylation with allyl bromide. SnCl4-catalyzed cycliza-
tion with aldehyde 31b yielded a 2.4:1 mixture of 2,5-cis and 2,5-
trans isomers, respectively (Table 1, entry 14) (Scheme 9). An
inconsequential 3.5:1 mixture of diastereomers at the diester
carbon was also observed. Treatment of 30c with tris(dibenzyli-
deneacetone)dipalladium(0) in refluxing acetonitrile led to a
mixture of three compounds in nearly equal amounts. The three
isomers consisted of the 2,5-cis and 2,5-trans isomers of the
an ester group, yielding an inseparable 1:8 mixture of diastere-
omers 33 for both the 2,5-cis and 2,5-trans isomers. The mixture
of four diastereomers was subjected to a three-step protocol
involving dihydroxylation, oxidative cleavage, and reduction to
obtain a mixture of primary alcohols 34. Heating this mixture in
toluene in the presence of p-toluenesulfonic acid successfully
removed two of the diastereomers by lactone formation leav-
ing 35 and 36. Protection of the primary alcohols as silyl ethers
allowed for separation of the two remaining diastereomers by
flash column chromatography, providing 37 in 54% yield from
the diastereomeric mixture of 33. With substrate 37 in hand, we
attempted to insert the unit of unsaturation required for the
(19) (a) Shimizu, I.; Tsuji, J. J. Am. Chem. Soc. 1982, 104, 5844–5846. (b)
Tsuji, J.; Minami, I.; Shimizu, I. Chem. Lett. 1984, 1721–1724. (c) Minami, I.;
Nisar, M.; Yuhara, M.; Shimizu, I.; Tsuji, J. Synthesis 1987, 992–998. (d)
Mercier, C.; Mignani, G.; Aufrand, M.; Allmang, G. Tetrahedron Lett. 1991,
32, 1433–1436.
(20) We would like to thank Professor Jeffrey M. Manthorpe for this
suggestion.
6834 J. Org. Chem. Vol. 75, No. 20, 2010

Karadeolian and Kerr
JOCArticle
SCHEME 7. Synthesis of Diester 43
We attempted to carry the mixture of 44 forward to the
primary alcohol through the three-step sequence shown
previously. Dihydroxylation of the vinyl group was success-
ful, but the following oxidative cleavage led to decomposi-
tion of the material. Dihydroxylation of both olefins in 44
was also attempted without success. The incompatibility of
the vinyl substituent and the R,β-unsaturated ester forced us
to devise a route where the allyl ester could be installed after
conversion of the vinyl group to the primary alcohol.
SCHEME 8. Tsuji’s Palladium(0)-Catalyzed Oxidative Decar-
boxylation of β-Keto Allyl Esters
SCHEME 9. Palladium-Catalyzed Oxidative Decarboxylation
of Diester 32d
The use of a benzyl ester in place of the allyl ester on the
cyclopropane was explored whereby hydrogenation could re-
veal the carboxylic acid, which could then be allylated to yield
the requisite allyl ester. Benzyl methyl cyclopropane diester 32e
was prepared by using the same protocol as for the allyl methyl
cyclopropane diester 32d. By this time in our study of the Lewis
acid-catalyzed cyclization reaction of aldehydes and cyclopro-
pane diesters, we had begun employing Sn(OTf)2 as a preferred
Lewis acid. Three different aldehydes (methylcarbamate 31e,
benzylcarbamate 31f, and p-toluenesulfonamide 31b protected
indole-2-carboxadehydes) were subjected to the cyclization con-
ditions and carried forward through the sequence shown in
Scheme 10. Upon cyclization, a three-step procedure involving
dihydroxylation of the vinyl group and oxidative cleavage of the
diol followed by reduction of the resulting aldehyde yielded
primary alcohols 47. Protection of the primary alcohol led to
48 in which conversion of a benzyl ester to an allyl ester was
required to provide the substrate for oxidative decarboxylation.
Treatment of 48 with palladium on carbon under a hydrogen
atmosphere removed the benzyl group and the resulting car-
boxylic acid was allylated under Mitsunobu conditions to yield
esters 49. In the case of 48b, the benzyl carbamate was also
removed under the hydrogenation conditions and the lower
yield of allyl ester 49b is due to competitive lactamization of the
indole and the carboxylic acid under the reaction conditions.
Of the three allyl ester substrates, only 49c (Pg = Ts) was an
acceptable candidate for the oxidative decarboxylation reaction.
Substrate 49a did not furnish any olefin and was recovered in
high yield. This result may be explained by coordination of the
palladium species to the carbonyl of the carbamate, thus pre-
venting progression of the catalytic cycle. When 49b was sub-
jected to the oxidative decarboxylation conditions, an insepa-
rable mixture of products containing olefin isomers along with
proto-decarboxylation isomers was obtained. Substrate 49c on
the other hand furnished the desired olefin 50 in moderate but
acceptable yield, with the remaining mass balance consisting of
the trans isomer and the olefin regioisomer.
unsaturated ester along with the olefin isomer in which the
double bond lay between C2 and C3 of the tetrahydrofuran. The
observed ratio of products was later found to be due to the
selective β-hydride elimination of the 2,5-trans isomer producing
exclusively the C3-C4 double bond, whereas the 2,5-cis isomer
yielded roughly equal amounts of the double bond isomers.
Considering that diester 30c is almost a 2:1 mixture of cis and
trans isomers, this would account for the observed ratio of the
three olefin isomers.
During our optimization attempts for the oxidative decar-
boxylation reaction, we found acetonitrile to be the only suitable
solvent. Dimethylformamide showed initial promise with an
increased selectivity for the 2,5-cis C3-C4 olefin; however, the
reaction did not reach completion and upon scale-up, stopped at
less than 20% conversion. Attempts at changing the ligands on
the metal were also unsuccessful; amine ligands stopped the
reaction completely and phosphine ligands induced the compe-
titive allylation reaction rather than oxidative decarboxylation.
To confirm the relative stereochemistry of the substituents on
the tetrahydrofuran ring, a crystalline derivative of the major
diastereomer of 49c was obtained (Scheme 11). Removal of the
silyl ether followed by esterification with p-bromobenzoyl chlo-
ride furnished 51, which could be recrystallized from CH2Cl2
and hexanes to provide single crystals suitable for X-ray diffrac-
tion analysis. As expected, the major isomer consisted of the
J. Org. Chem. Vol. 75, No. 20, 2010 6835

Karadeolian and Kerr
JOCArticle
SCHEME 10. Successful Route to Dihydrofuran 50
SCHEME 11. Synthesis of Crystalline Derivative 51
only recovered starting material or decomposition was ob-
served. At this point we changed direction and attempted a
route involving oxidation of an N-H indole that would
presumably be more susceptible to the EAS reaction.
Beginning with primary alcohol 53, tosyl deprotection took
place in good yield to provide free indole 60 (Scheme 13).
Oxidation of 60 with m-CPBA9 yielded a mixture of aminals
61. Surprisingly, these intermediates were stable to silica gel and
were isolable for characterization. For the synthetic sequence,
the aminals were used as a crude mixture (after removal of the
m-chlorobenzoic acid byproduct by extraction with ethyl acet-
ate from a basic aqueous solution). Treatment of the aminals
with indole and CSA over a 42 h reaction period provided
isatisine A acetonide 2 in 50% yield from indole 60. Isatisine A
was obtained in 82% yield by hydrolysis of acetonide 2 with
acidic methanol. The specific rotation of synthetic isatisine A
acetonide 2 and isatisine A 1 was [R]25D þ271 and [R]25D þ274,
respectively. The specific rotation of isolated isatisine A acet-
onide is [R]14D -283, which is almost equal in magnitude but
opposite in direction to that of the synthetic material. Since we
know the absolute configuration of our product, which is
antipodal to the isolated compound, the absolute configuration
of the isolated natural product must be C2(S), C9(R), C10(S),
C12(R), and C13(R). A synthesis employing (R)-vinylcyclopro-
pane 32e would thus yield the natural enantiomer.
2,5-cis relative geometry of the substituents on the tetrahydro-
furan ring.
With the successful installation of an olefin in our tetrahy-
drofuran, we repeated the reaction sequence employing homo-
chiral vinylcyclopropane20 and used enantiopure material from
this point forward. At this junction, our initial aim was to
construct the lactam ring to form the pyrroloindolone, followed
by oxidation of the indole species and EAS onto indole. With
this in mind, dihydroxylation of 50 provided diol 52 in mod-
erate yield (Scheme 12). Acetonide formation occurred with
concomitant silyl deprotection to yield primary alcohol 53. The
TBS group was found to be very labile and was cleaved under
the mildest reaction conditions. Therefore, the primary alcohol
was protected again, this time as the TBDPS ether in nearly
quantitative yield. Removal of the tosyl group was achieved
under standard conditions, followed by DBU-induced lactam
formation to provide 55. Single crystals of 55 were obtained by
slow evaporation of CH2Cl2 from hexanes which were suitable
for X-ray crystal analysis to confirm the relative structure of 55.
Treatment of 55 with DMDO did not yield the indoxyl spe-
cies as we had seen for the model study; instead, epoxide 56 was
obtained in nearly quantitative yield and was stable at room
temperature, which is uncommon for such compounds.8b Ep-
oxide 56 was not stable to silica gel and was characterized with-
out purification. After many trials, we found that addition of
CSA to the DMDO solution after consumption of 55 led to a
mixture of two products: diol 57 and indoxyl 58. This mixture
was treated with IBX in refluxing ethyl acetate to obtain indoxyl
58 in 60% yield from pyrroloindolone 55. Just as oxidation of 55
was more complicated than in the model study, indoxyl 58
would not participate in the EAS reaction with indole to yield
TBDPS protected isatisine A acetonide 59. A variety of reaction
conditions were employed to promote the EAS reaction but
The indole oxidation and EAS cascade reaction requires
further discussion. Intermediate 62 (which was isolated in our
initial efforts) was not in fact the first product formed upon
treatment of 61 with indole and an acid catalyst. In fact, the
diastereomeric (and undesired) indole adduct 64 was the kinetic
product (formed almost instantly) and isomerized over time to
the desired 62, which in turn was converted to isatisine A
acetonide 2 and subsequently to isatisine A 1. This was mon-
1
itored by H NMR spectroscopy and is shown in Figure 5.
Aminals 61 were purified as the diastereomeric mixture and
their 1H NMR spectrum is shown in Figure 5 at t = 0 h. Within
15 min of adding 1 equiv of indole and CSA both aminal start-
ing materials were consumed and EAS diastereomer 64 was the
major product (2) with a small amount of diastereomer 62 (9).
After 1 h the amount of 62 had increased and after 12 h 62
become the major diastereomer. Furthermore, a small amount
of isatisine A acetonide 2 (b) was present after 12 h. By 36 h the
6836 J. Org. Chem. Vol. 75, No. 20, 2010

Karadeolian and Kerr
JOCArticle
SCHEME 12. Pyrroloindolone Route
SCHEME 13. Successful Route to (þ)-Isatisine A
reaction conditions. The amount of 62 decreases in concert
with the appearance of the isatisine A acetonide 2. Inspection
of simple molecular models indicates that lactam formation
via 64 would be difficult and so it seems that lactam forma-
tion from the desired 62 fortunately provides an irreversible
sink to the process. With 2 equiv of indole and CSA, the
progress of the reaction increased slightly.
A more detailed mechanistic picture is shown in Scheme
14. The acid catalyst facilitates formation of iminium ion 63,
a willing electrophile for the nucleophilic indole. While other
mechanistic possibilities are not unreasonable, we postulate
that the isomerizaiton of 64 to 62 may proceed via a gramine-
type fragmentation via intermediate 65. Reclosure of the
aniline onto the indole methide provides access to useful
quantities of 62, which in turn may be trapped as the desired
lactam 2.
Conclusion
In summary, we have completed the total synthesis of (þ)-
isatisine A (1) via its acetonide 2 in a 14-step process, with a
5.8% overall yield from homochiral cyclopropane diester
32e and aldehyde 31b using a Lewis acid-catalyzed cycliza-
tion reaction to construct the tetrahydrofuran ring. A palla-
dium-catalyzed oxidative decarboxylation reaction was used
to install a unit of unsaturation directly from the diester
species. Lastly, an oxidation and electrophilic aromatic
substitution reaction was employed to furnish acetonide 2
in a cascade process involving an equilibrium between EAS
substrate diastereomers which converge to acetonide 2.
Importantly, the absolute configuration of isatisine A has
been determined by our enantioselective synthesis of the
antipode of the natural product. Efforts are underway to
determine the biological activity of (þ)-isatisine A (1) along
with its acetonide 2.
major product was 2 with the other two intermediates still
present. After 72 h the reaction was nearly complete with less
than 5% of the material as 62.
Figure 6 shows, in graphical form, the progress of the
reaction beginning with the kinetic adduct 64. Its disappear-
ance is concomitant with the production of diastereomer 62,
the concentration of which peaks at about 10 h under our
J. Org. Chem. Vol. 75, No. 20, 2010 6837

Karadeolian and Kerr
JOCArticle
FIGURE 6. EAS Cascade at Various Time Intervals.
mass spectra (HRMS) were obtained at 70 eV. Toluene, THF,
ether, DMF, and methylene chloride were dried and deoxygenated
by passing the nitrogen-purged solvents through activated alumina
columns. All other reagents and solvents were used as purchased
from the supplier. The progress of reactions was followed by thin
layer chromatography (TLC) (silica gel 60 F254) and the devel-
oped plates stained with acidic anisaldehyde, phosphomolybdic
acid, or basic potassium permanganate. Flash column chromatog-
raphy was performed with silica gel (230-400 mesh).
Experimental Procedures. Compound 32e. To a stirring solu-
tion of dimethyl (S)-2-vinylcyclopropane 1,1-diester21 (32c)
(3.01 g, 16.34 mmol) (racemic material was prepared according
to literature procedures22) in methanol (12.5 mL) was added
12.5 mL of an aqueous 1.7 N NaOH solution. The resulting
solution was stirred for 1.5 h after which time TLC analysis
showed complete consumption of the starting material. The
reaction was poured into a separatory funnel containing 5%
aqueous HCl and extracted with EtOAc (3ꢀ). The combined
organic extracts were washed with brine, dried over MgSO4,
filtered, and concentrated in vacuo to yield 2.78 g of a clear
colorless oil. The resulting acid was dissolved in DMF (125 mL)
and K2CO3 (2.48 g, 17.97 mmol) was added followed by 2.15 mL
(17.97 mmol) of benzyl bromide. The solution was stirred under
argon for 48 h then added to H2O and extracted with Et2O (4ꢀ).
The combined organic extracts were washed with H2O (2ꢀ) and
brine, dried over MgSO4, and filtered. Concentration in vacuo
yielded 4.04 g (95% over 2 steps) of 32e as a clear colorless oil that
was used without further purification (Rf 0.44, 30% EtOAc/
1
hexanes); H NMR (400 MHz, CDCl3) δ 7.38-7.29 (m, 5H),
5.49-5.06 (m, 5H), 3.72 (s, 3H), 2.61 (q, J = 8.4 Hz, 1H), 1.74 (dd,
J = 4.8, 7.2 Hz, 1H), 1.60 (dd, J = 4.8, 9.2 Hz, 1H); 13C NMR
(100 MHz, CDCl3) δ169.3, 167.7, 135.5, 132.9, 128.5, 128.2, 127.8,
118.7, 67.1, 52.5, 35.9, 31.4, 20.6; IR (thin film, cm-1) νmax 3090,
3067, 2954, 1729, 1639, 1457, 1381, 1271, 1209, 1129, 991, 747, 698;
HRMS calcd for C15H16O4 260.1049, found 260.1056.
Compound 30f. Sn(OTf)2 (0.35 g, 0.84 mmol) was placed in a
flask equipped with a stir bar and purged with argon. Cyclo-
propane 32e (2.2 g, 8.45 mmol) and aldehyde 31b (3.5 g, 11.8
mmol) were dissolved in 9 mL of CH2Cl2 and added to the
flask.14c An additional 3 mL of CH2Cl2 was used to quantify the
transfer of cyclopropane and aldehyde. After being stirred for 11
h the reaction was passed through a small plug of silica gel and
the resulting solution was concentrated in vacuo. Purified by
flash column chromatography (5-13% EtOAc/hexanes)
yielded 4.2 g (89%) of 30f as a yellow foam. 1H NMR showed
30f to be a mixture of 4 diastereomers (1:0.23:0.07:0.04) with an
11:1, 2,5-cis: 2,5-trans ratio, respectively. The diastereomers at
the esters (3.5:1) are inconsequential since they will converge
FIGURE 5. EAS cascade reaction monitored by 1H NMR (circled
hydrogen indicated in the spectrum).
Experimental Section
General Considerations. Melting points were determined with
a Gallenkamp melting point apparatus and are uncorrected.
Infrared spectra were obtained as thin films on NaCl plates.
HPLCs were conducted with a dual λ-absorbance detector set to
256 nm. NMR experiments were conducted in CDCl3 (referenced
to 7.26 ppm for 1H and 77.0 for 13C) or d4-MeOD (referenced to
3.31 ppm for 1H or 49.15 ppm for 13C). Coupling constants (J) are
in Hz. The multiplicities of the signals are described by using the
following abbreviations: s = singlet, d = doublet, t = triplet, q =
quartet, p = pentet, m = multiplet, br = broad. High-resolution
(21) Carson, C. A.; Kerr, M. A. Angew. Chem., Int. Ed. 2006, 45, 6560–
6563.
(22) Kierstead, R. W.; Linstead, R. P.; Weedon, B. C. L. J. Chem. Soc.
1952, 74, 3610–3616.
6838 J. Org. Chem. Vol. 75, No. 20, 2010

Karadeolian and Kerr
JOCArticle
SCHEME 14. Mechanism of EAS Cascade
further on in the synthesis (Rf 0.26 and 0.34, 30% EtOAc/
hexanes); for 1H NMR (400 MHz, CDCl3) and 13C NMR
(100 MHz, CDCl3) see spectra in the Supporting Information;
HRMS calcd for C31H29NO7S 559.1651, found 559.1665.
Compound 47c. To the diastereomeric mixture of 30f (14.3 g,
25.5 mmol) dissolved in THF (175 mL), H2O (140 mL), and
acetone (35.5 mL) was added methanesulfonamide (2.42 g, 25.5
mmol), NMO (3.28 g, 28.0 mmol), and a large crystal of OsO4.
The reaction was monitored by TLC and upon completion
sodium sulfite was added to the reaction, which was then stirred
for a further 30 min. The reaction was then filtered through
Celite and the volatile components were removed in vacuo. The
remaining solution was added to brine and extracted with
EtOAc (4ꢀ), dried over MgSO4, filtered, and concentrated
under reduced pressure to yield a white foam (Rf 0.23 and
0.35, 70% EtOAc). The resulting foam was dissolved in THF
(180 mL) and H2O (180 mL), then cooled to 0 °C, and after
addition of NaIO4 (7.63 g, 35.7 mmol) the reaction flask was
removed from the ice bath and the solution was stirred at room
temperature. The reaction was monitored by TLC and upon
completion 300 mL of EtOH was added and the solution was
filtered through Celite rinsing the filter cake with 100 mL of
EtOH (Rf 0.55 streak, 70% EtOAc). NaBH4 (1.64 g, 43.4 mmol)
was added to the filtrate and the solution was stirred at room
temperature. TLC monitoring showed consumption of the
aldehyde after 45 min at which time the reaction was quenched
with dropwise addition of a saturated aqueous solution of
NH4Cl. The organic solvents were removed under reduced
pressure and the remaining solution was added to H2O and
extracted with EtOAc (4ꢀ). The combined organic extracts were
washed with brine, dried over MgSO4, filtered, and concen-
trated in vacuo. Purification by flash column chromatography
(30-50% EtOAc/hexanes) yielded 12.5 g of the diastereomeric
mixture of primary alcohols as a white foam (87% over 3 steps).
A small amount of the major isomer of 47c (shown above,
Rf 0.60, 70% EtOAc/hexanes) was isolated and used for char-
acterization; 1H NMR (600 MHz, CDCl3) δ 8.07 (d, J = 8.8 Hz,
1H), 7.70 (d, J = 8.4 Hz, 2H), 7.39 (d, J = 7.6 Hz, 1H), 7.27
(ddd, J = 1.6, 7.2, 7.2 Hz, 1H), 7.21-7.12 (m, 6H), 7.04-7.01
(m, 2H), 6.80 (s, 1H), 6.76 (s, 1H), 4.94 (d, J = 12.4 Hz, 1H), 4.57
(d, J = 12.4 Hz, 1H), 4.21 (dddd, J = 3.0, 4.8, 4.8, 11.4 Hz, 1H),
3.95-3.91 (m, 1H), 3.77 (s, 3H), 3.77-3.72 (m, 1H), 2.92 (dd,
J = 11.2, 13.2 Hz, 1H), 2.39 (dd, J = 5.2, 13.2 Hz, 1H), 2.27 (s,
3H), 2.04 (br s, 1H); 13C NMR (100 MHz, CDCl3) δ 170.6,
167.8, 144.7, 138.7, 136.9, 135.3, 134.7, 129.6, 129.1, 128.1,
128.0, 126.6, 124.8, 123.6, 121.1, 114.9, 111.6, 78.8, 77.4, 67.5,
66.1, 63.2, 53.1, 35.8, 21.4; IR (thin film, cm-1) νmax 3474, 3068,
3036, 2955, 2886, 1734, 1452, 1374, 1274, 1188, 1060, 735, 580;
HRMS calcd for C30H29NO8S 563.1614, found 563.1598.
Compound 48c. A mixture of alcohols 47c (12.5 g, 22.1 mmol)
was dissolved in CH2Cl2 (220 mL) followed by the successive
addition of TBSCl (4.67 g, 31.0 mmol) and imidazole (2.11 g,
31 mmol). The reaction was stirred for 12 h after which time it
was added to H2O and extracted with CH2Cl2 (2ꢀ). The
combined organic extracts was washed with brine, dried over
MgSO4, and filtered. Purification via flash column chromatog-
raphy (5% EtOAc/hexanes) yielded 14.2 g (94%) of the diaster-
eomeric mixture as a white foam. A small amount of the major
isomer of 48c (shown above, Rf 0.42, 30% EtOAc/hexanes) was
isolated and characterized; 1H NMR (600 MHz, CDCl3) δ 8.05
(d, J = 8.4 Hz, 1H), 7.67 (d, J = 8.4 Hz, 2H), 7.38 (d, J = 7.6 Hz,
1H), 7.24-7.05 (m, 10 H), 6.76 (s, 1H), 4.95 (d, J = 12.4 Hz,
1H), 4.68 (d, J = 12.4 Hz, 1H), 4.10 (dddd, J = 11.6, 4.0, 4.0, 4.0
Hz, 1H), 3.98 (dd, J = 11.2, 3.6 Hz, 1H), 3.81 (dd, J = 11.2, 4.0
Hz, 1H), 3.78 (s, 3H), 2.94 (t, J = 12.4 Hz, 1H), 2.34 (dd, J =
12.8, 4.4 Hz, 1H), 2.29 (s, 3H), 0.93 (s, 9H), 0.11 (s, 3H), 0.09 (s,
3H); 13C NMR (100 MHz, CDCl3) δ 170.8, 167.6, 144.6, 139.5,
137.1, 135.3, 135.0, 129.6, 129.5, 128.2, 128.1, 127.9, 126.6,
124.6, 123.6, 121.0, 114.9, 112.4, 79.0, 77.6, 67.4, 66.1, 63.4,
53.1, 35.6, 25.9, 21.5, 18.4, -5.3, -5.5; IR (thin film, cm-1) νmax
2956, 2930, 2858, 1736, 1598, 1472, 1451, 1175, 1091, 813, 703,
581; HRMS calcd for C36H43NO8SSi 677.2479, found 677.2473.
Compound 49c. To a diastereomeric mixture of 48c (14.1 g,
20.8 mmol) in THF (250 mL) was added 1.4 g of 10% Pd/C. The
reaction vessel was purged with hydrogen gas (4ꢀ) then placed
under 1 atm of hydrogen for 3 h after which time TLC analysis
showed consumption of the diester. The solution was filtered
through a pad of Celite and the Celite was rinsed with THF (200
mL). Triphenylphosphine (8.06 g, 30.7 mmol) and allyl alcohol
(2.1 mL, 30.7 mmol) were added to the solution along with a
magnetic stir bar. With stirring DIAD (6.0 mL, 30.7 mmol) was
added to the solution and stirring was continued for 50 min after
which time a further 0.3 equiv of all the reagents was added.
After 40 min TLC analysis showed consumption of the acid
substrate. Purification by flash column chromatography
(60-100% CH2Cl2/hexanes) yielded 12.9 g (99%) of a diaster-
eomeric mixture of products as a white foam. A small amount of
the major isomer of 49c was isolated and used for characteriza-
tion (Rf 0.38, 30% EtOAc/hexanes); 1H NMR (600 MHz,
CDCl3) δ 8.07 (dd, J = 8.0, 0.4 Hz, 1H), 7.68-7.65 (m, 2H),
J. Org. Chem. Vol. 75, No. 20, 2010 6839

Karadeolian and Kerr
JOCArticle
7.40-7.38 (m, 1H), 7.24 (ddd, J = 7.6, 7.6, 1.6 Hz, 1H), 7.17 (dd,
J = 7.6, 1.2 Hz, 1H), 7.14-7.12 (m, 2H), 7.03 (s, 1H), 6.75 (d,
J = 0.8 Hz, 1H), 5.55 (dddd, J = 16.4, 10.8, 6.0, 6.0 Hz, 1H),
5.04 (dq, J = 17.2, 1.2 Hz, 1H), 4.95 (dq, J = 10.4, 1.2 Hz, 1H),
4.33 (ddt, J = 12.8, 6.0, 1.2 Hz, 1H), 4.17 (ddt, J = 11.6, 6.0, 1.2
Hz, 1H), 4.11 (dq, J = 12.0, 4.0 Hz, 1H), 3.99 (dd, J = 11.2, 3.6
Hz, 1H), 3.87 (s, 3H), 3.83 (dd, J = 11.2, 3.6 Hz, 1H), 2.96 (dd,
J = 12.8, 11.6 Hz, 1H), 2.33 (dd, J = 13.2, 4.4 Hz, 1H), 2.28 (s,
3H), 0.95 (s, 9H), 0.12 (s, 3H), 0.11 (s, 3H); 13C NMR (100 MHz,
CDCl3) δ 170.9, 167.5, 144.6, 139.4, 137.1, 135.3, 131.0, 129.7,
129.5, 126.6, 124.5, 123.6, 121.0, 118.6, 114.8, 112.4, 78.9, 77.6,
66.5, 66.1, 63.4, 53.2, 35.8, 26.0, 21.5, 18.4, -5.3, -5.4 ; IR (thin
film, cm-1) νmax 2956, 2930, 2859, 1737, 1451, 1261, 1175, 1092,
914, 837, 749, 581; HRMS calcd for C32H41NO8SSi 627.2322,
found 627.2340; [R]25D þ201 (c 2.6, MeOH).
The reaction was stirred until it became black at which point
sodium sulfite was added to the reaction then the mixture was
stirred for a further 30 min. The solution was filtered through
Celite, added to water, and extracted with EtOAc (3ꢀ). The
combined organic extracts were washed with brine, dried over
MgSO4, filtered, and concentrated. Flash column chromato-
graphy (8-22% EtOAc/hexanes) yielded 2.15 g of 52 (49%) and
1.96 g of a 1:0.7 mixture of enoate starting material (in favor of
the desired isomer). The enoate mixture was resubjected to the
reaction conditions in acetone/H2O (16.5 mL, 10:1), with NMO
(0.5 g, 4.2 mmol), methanesulfonamide (0.4 g, 4.2 mmol), and
catalytic osmium tetraoxide to yield a further 0.58 g of 52 and
0.74 g of a 1:1 mixture of enoates. Overall yield after 2 cycles was
1
2.63 g (62%) (Rf 0.16, 30% EtOAc/hexanes); H NMR (400
MHz, CDCl3) δ 8.04 (d, J = 8.0 Hz, 1H), 7.59 (d, J = 8.4 Hz,
2H), 7.44 (d, J = 7.2 Hz, 1H), 7.28-7.12 (m, 5H), 7.73 (s, 1H),
4.58 (s, 1H), 4.50 (dd, J = 10.8, 8.8 Hz, 1H), 4.17-4.13 (m, 1H),
3.96-3.92 (m, 2H), 3.39 (s, 3H), 2.62 (d, 11.2 Hz, 1H), 2.30 (s,
3H), 0.98 (s, 9H), 0.19 (s, 3H), 0.16 (s, 3H); 13C NMR (100 MHz,
CDCl3) δ 171.4, 145.1, 138.6, 137.0, 134.9, 129.9, 129.7, 126.4,
124.6, 123.9, 121.1, 114.4, 111.8, 84.6, 83.2, 82.4, 73.3, 61.7, 53.3,
26.0, 51.5, 18.5, -5.3, -5.4 ; IR (thin film, cm-1) νmax 3477,
2954, 2929, 2857, 1735, 1452, 1371, 1268, 1229, 1175, 836, 764,
750, 581; HRMS calcd for C28H37NO8SSi 575.2009, found
575.1992; [R]25D þ182 (c 2.3, MeOH).
Compound 50. To the diastereomeric mixture of ally esters 49c
(2.36 g, 3.8 mmol) in acetonitrile (125 mL) was added Pd2dba3
(0.18 g, 0.19 mmol).20 The reaction was heated to 80 °C for 10 h
after which time it was filtered through a pad of Celite. Flash
column chromatography (5-20% EtOAc/hexanes) yielded 1.0
g of 50 as a white foam (49%, typically yield ranges from 45% to
55%). The remaining mass balance consists of 2,5-trans isomer
G and the undesired olefin isomer H. 50: 1H NMR (600 MHz,
CDCl3) δ 8.09 (d, J = 8.4 Hz, 1H), 7.91 (d, J = 8.0 Hz, 2H), 7.42
(d, 7.6 Hz, 1H), 7.31-7.27 (m, 1H), 7.21-7.17 (m, 4H), 6.83 (t,
J = 2.4 Hz, 1H), 6.46 (s, 1H), 5.09-5.03 (X of ABX, 1H), 3.70
(A/B of ABX, J = 10.0, 4.8 Hz, 1H), 3.67 (s, 3H), 3.50 (A/B of
ABX, J = 10.0, 7.2 Hz, 1H), 2.32 (s, 3H), 0.83 (s, 9H), -0.03 (s,
3H), -0.06 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 162.9, 144.6,
142.5, 136.8, 135.8, 133.7, 129.4, 128.8, 127.2, 125.0, 123.5,
121.3, 114.9, 110.3, 87.1, 78.1, 65.5, 51.9, 25.8, 21.5, 18.3,
-5.46; IR (thin film, cm-1) νmax 2955, 2929, 2858, 1727, 1451,
1375, 12691, 1177, 1150, 1121, 1091, 1057, 837, 782, 673, 581;
HRMS calcd for C28H35NO6SSi 541.1954, found 541.1958;
Compound 53. To a solution of 52 (0.61 g, 1.06 mmol) in
acetone (15 mL) were added 2,2-dimethoxypropane (0.9 mL, 7.5
mmol) and p-toluenesulfonic acid (0.21 g, 1.06 mmol). The
reaction was stirred for 16 h then a small amount of a saturated
aqueous solution of NaHCO3 was added to the reaction and the
acetone was removed under reduced pressure. The resulting
mixture was added to water, then extracted with EtOAc (3ꢀ),
and the combined organics were washed with brine, dried over
MgSO4, filtered, and concentrated. Purification by flash column
chromatography (15-40% EtOAc/hexanes) yielded 53 (0.45 g,
[R]25 þ131 (c 0.57, MeOH); Rf 0.40, 30% EtOAc/Hexanes.
D
G: 1H NMR (600 MHz, CDCl3) δ 8.12 (dd, J = 8.4, 0.8 Hz, 1H),
7.92 (d, J = 12.6 Hz, 2H), 7.41 (d, J = 8.0 Hz, 1H), 7.29 (ddd,
J = 8.4, 7.2, 1.2 Hz, 1H), 7.20-7.16 (m, 2H), 6.89 (dd, J = 8.4, 5
Hz, 1H), 6.46 (s, 1H), 4.99-4.94 (m, 1H), 3.90 (dd, J = 10.4, 4.8
Hz, 1H), 3.72 (dd, J = 10.0, 6.8 Hz, 1H), 3.68 (s, 3H), 2.32 (s,
3H), 0.91 (s, 9H), 0.11 (s, 3H), 0.09 (s, 3H); 13C NMR (100 MHz,
CDCl3) δ 162.8, 144.6, 142.7, 139.7, 137.0, 135.7, 134.3, 129.5,
128.8, 127.3, 125.0, 123.5, 121.2, 114.9, 109.8, 85.5, 78.3, 65.1,
51.9, 25.8, 21.5, 18.3, -5.4 (2C); IR (thin film) νmax 2954, 2929,
2885, 2857, 1726, 1598, 1472, 1375, 1257, 1175, 1150, 1121, 1091,
1
83%) as a white foam (Rf = 0.09, 30% EtOAc/hexanes); H
NMR (600 MHz, CDCl3) δ 8.09 (d, J = 8.4 Hz, 1H), 7.68
(apparent d, J = 8.4 Hz, 2H), 7.42 (d, J = 7.8 Hz, 1H), 7.27 (t,
J = 8.4 Hz, 1H), 7.19 (t, J = 7.8 Hz, 1H), 7.15 (apparent d, J =
8.4 Hz, 2H), 6.78 (s, 1H), 6.12 (s, 1H), 5.10 (d, J = 4.8 Hz, 1H),
4.32 (apparent q, J = 4.2 Hz, 1H), 4.06-3.95 (m, 2H), 3.32 (s,
3H), 2.42 (br s, 1H), 2.30 (s, 3H), 1.75 (s, 3H), 1.38 (s, 3H); 13
C
NMR (100 MHz, CDCl3) δ 170.7, 144.8, 137.4, 135.9, 135.0,
129.6, 129.3, 126.7, 125.0, 123.9, 121.0, 117.4, 115.3, 112.2, 94.8,
84.7, 84.3, 82.1, 62.6, 52.4, 27.4, 25.6, 21.5; IR (thin film, cm-1
)
max 3448, 2991, 2951, 1735, 1375, 1255, 1217, 1175, 1091, 914,
1
1055, 837, 815, 779, 749; Rf 0.45, 30% EtOAc/hexanes. H: H
ν
NMR (600 MHz, CDCl3) δ 8.04 (d, J = 8.4 Hz, 1H), 7.73 (d,
J = 8.0 Hz, 2H), 7.48 (d, J = 8.0 Hz, 1H), 7.32 (t, J = 8.0 Hz,
1H), 7.21 (t, J = 8.0 Hz, 1H), 7.13 (d, J = 8.0 Hz, 2H), 6.86 (s,
1H), 4.99-4.92 (m, X of ABX, 1H), 3.92 (A/B of ABX, J = 10.8,
7.2 Hz, 1H), 3.85 (A/B of ABX, J = 10.8, 7.2 Hz, 1H), 3.56 (s,
3H), 3.16 (dd, J = 14.8, 10.8 Hz, 1H), 3.00, (dd, J = 14.8, 8.8 Hz,
1H), 2.30 (s, 3H), 0.90 (s, 9H), 0.08 (s, 3H), 0.07 (s, 3H); 13C
NMR (100 MHz, CDCl3) δ 165.0, 157.1, 144.6, 137.0, 135.1,
129.4, 129.1, 127.2, 125.7, 123.7, 121.7, 115.8, 115.0, 107.6, 83.4,
77.2, 64.5, 51.0, 32.6, 25.9, 21.5, 18.3, - 5.2 (2C); IR (thin film,
cm-1) νmax 2953, 2930, 2900, 2886, 2858, 1706, 1375, 1258, 1190,
1176, 1123, 1093, 1056, 837, 813, 780, 750; HRMS calcd for
C28H35NO6SSi 541.1954, found 541.1809; Rf 0.40, 30% EtOAc/
hexanes.
Compound 52. (Note well, complete separation of the regioi-
someric double bond compounds was difficult and in most cases
a mixture of the two compounds was obtained from the previous
step and used in the dihydroxylation.) A 2:1 mixture of enoates
50 and H (6.21 g, 11.3 mmol total, 4.14 g, 7.5 mmol of desired
isomer) was dissolved in acetone/H2O (44 mL, 10:1) and to it
were added NMO (1.94 g, 16.6 mmol), methanesulfonamide
(1.25 g, 12.1 mmol), and a large crystal of osmium tetraoxide.
734; HRMS calcd for C25H27NO8S 501.1457, found 501.1476;
HPLC (Phenomenex Lux 3u Cellulose-2, 30% i-PrOH/hexanes,
1.0 mL/min, (() RT = 11.65, 17.67 min, (þ) RT = 17.63 min; ee
>98%; [R]25 þ106 (c 2.3, MeOH); mp (þ)-53 169-172 °C
D
(recrystallized from MeOH).
Compound 60. To compound 53 (0.45 g, 0.89 mmol) in MeOH
(15 mL) were added NH4Cl (0.21 g, 3.92 mmol) and magnesium
turnings (0.43 g, 17.69 mmol). The reaction was stirred and
monitored by TLC for completion. Upon consumption of the
starting material (Rf 0.51, 20% EtOAc/CH2Cl2) the solution
was poured into a separatory funnel containing a saturated
solution of NH4Cl and extracted with EtOAc (3ꢀ). The com-
bined organic extracts were washed with brine, dried over
MgSO4, filtered, and concentrated. Purification by flash column
chromatography (30% EtOAc/hexanes) yielded the titled com-
pound 60 (0.24 g, 77%) as a white foam (Rf 0.34, 20% EtOAc/
CH2Cl2); 1H NMR (600 MHz, CDCl3) δ 8.54 (br s, 1H), 7.55 (d,
J = 7.8 Hz, 1H), 7.33 (d, J = 8.4 Hz, 1H), 7.16 (dt, J = 7.2, 1.2
Hz, 1H), 7.07 (apparent t, J = 7.2 Hz, 1H), 6.51-6.50 (m, 1H),
5.26 (s, 1H), 5.07 (d, J = 3.6 Hz, 1H), 4.35 (apparent q, J = 4.2
Hz, 1H), 4.11-4.07 (m, 1H), 4.01-3.97 (m, 1H), 3.38 (s, 3H),
2.61 (br t, J = 6 Hz, 1H), 1.70 (s, 3H), 1.38 (s, 3H); 13C NMR
6840 J. Org. Chem. Vol. 75, No. 20, 2010

Karadeolian and Kerr
JOCArticle
(100 MHz, CDCl3) δ 171.2, 135.9, 131.9, 127.8, 122.1, 120.6,
119.7, 117.6, 110.9, 101.0, 93.3, 84.8, 84.7, 62.3, 52.7, 27.3, 25.2;
IR (thin film, cm-1) νmax 3395, 2992, 2961, 2887, 1734, 1663,
1457, 1437, 1377, 1214, 1118, 1085, 1047, 867, 751; HRMS calcd
7.59 (d, J = 7.8 Hz, 1H), 7.40 (t, J = 7.8 Hz, 1H), 7.22 (d, J = 8.4
Hz, 1H), 7.18 (s, 1H), 7.12 (t, J = 7.2 Hz, 1H), 7.08 (t, J = 7.2 Hz,
1H), 6.85 (d, J = 8.4 Hz, 1H), 6.78 (t, J = 7.2 Hz, 1H), 5.72 (s, 1H),
4.95 (d, J =3.0 Hz, 1H), 4.79 (s, 1H), 4.05 (q, J = 3.0 Hz, 1H), 3.88
for C18H21NO6 347.1369, found 347.1373; [R]25 þ59 (c 0.42,
(s, 1H), 3.52 (br s, 1H), 3.03 (s, 3H), 1.63 (s, 3H), 1.27 (s, 3H); 13
C
D
MeOH).
NMR (100 MHz, CDCl3) δ 197.3, 171.4, 159.8, 136.9, 136.7, 125.7,
125.2, 123.8, 122.0, 120.8, 120.3, 119.9, 119.1, 117.1, 112.2, 111.7,
Oxidation and EAS Sequence (2). To compound 60 (0.071 g,
0.20 mmol) in CH2Cl2 (9 mL) was added a solution of m-CPBA8
(0.11 g of 66% by mass, 0.41 mmol) in CH2Cl2 (9 mL) at room
temperature. The reaction was monitored by TLC and upon
consumption of the starting material (1.5 h) the mixture was added
to a separatory funnel containing a saturated solution of NaHCO3
and extracted with CH2Cl2 (3ꢀ). The combined organic extracts
were washed with brine, dried over MgSO4, filtered, and concen-
trated to yield 0.73 g of a yellow foam. This crude mixture contains
a 2:1 ratio of diastereomers 61 and was carried onto the next
reaction without any further purification (Rf major 0.50, minor
0.41, 50% EtOAc/hexanes). Indole (0.048 g, 0.41 mmol) and CSA
(0.094 g, 0.40 mmol) were added to the diastereomeric mixture of
intermediates 60 (0.73 g, crude) dissolved in CH2Cl2 (9 mL) and the
solution was stirred at room temperature. If the reaction was
stopped after 12 h indole addition product 62 could be isolated;
however, upon longer exposure (42 h) the acetonide of isatisine A
(2) was obtained as a light yellow solid (0.05 g, 50%) after purifica-
tion via flash column chromatograph (0.25-0.75% MeOH/CH2-
90.0, 89.6, 85.3, 83.4, 69.1, 62.1, 51.9, 27.4, 25.2; IR (thin film, cm-1
)
νmax 3389, 3058, 2990, 2934, 2249, 1733, 1699, 1618, 1486, 1376,
1249, 1103, 910, 734; HRMS calcd for C26H26N2O7 478.1740,
found 478.1735 (Rf 0.16, 50% EtOAc/hexanes). 2: 1H NMR (600
MHz, CD3OD) δ 8.00 (d, J = 8.4 Hz, 1H), 7.91 (d, J = 8.4 Hz,
1H), 7.76 (m, 1H), 7.63 (d, J = 7.8 Hz, 1H), 7.35 (d, J = 7.8 Hz,
1H), 7.31 (t, J = 7.2 Hz, 1H), 7.24 (s, 1H), 7.14 (t, J = 7.8 Hz, 1H),
7.07 (t, J = 7.8 Hz, 1H), 4.89 (s, 1H), 4.78 (d, J = 3.6 Hz, 1H), 4.15
(d of X of ABX, J = 3.6 Hz, 1H), 3.48 (A/B of ABX, J = 12, 4.8
Hz, 1H), 3.42 (A/B of ABX, J = 12, 4.8 Hz, 1H), 1.49 (s, 3H), 1.36
(s, 3H); 13CNMR (100 MHz, CD3OD) δ195.8, 171.5, 151.3, 139.2,
138.0, 127.4, 127.1, 126.3, 125.8, 124.4, 123.4, 121.4, 120.9, 119.6,
117.5, 113.1, 111.1, 99.5, 87.9, 87.2, 86.1, 76.4, 62.6, 27.4, 26.4;
HRMS calcd for C25H22N2O6 446.1478, found 446.1487; HPLC
(Phenomenex Lux 3u Cellulose-2, 35% i-PrOH/hexanes, 1.0 mL/
min, (() RT = 7.6, 11.04 min, (þ) RT = 11.02 min; ee = >98%;
[R]25D þ271 (c 1.6, MeOH) (Rf 0.34, 5% MeOH/CH2Cl2).
(þ)-Isatisine A (1). Acetonide 2 (28 mg, 0.063 mmol) was
dissolved in 1 N HCl in MeOH (3 mL) and the mixture was
stirred for 1.5 h at which point the solvent was removed. The
remaining yellow/orange residue was preadsorbed onto silica
and purified by flash column chromatography (1-2.5%
MeOH/CH2Cl2) to yield 21 mg of isatisine A (82%) as a light
yellow solid (Rf 0.51, 10% MeOH/CH2Cl2); 1H NMR (600
MHz, CDCl3) δ 7.98 (d, J = 7.8 Hz, 1H), 7.93 (d, J = 7.8 Hz,
1H), 7.76-7.73 (m, 1H), 7.63 (d, J = 7.8 Hz, 1H), 7.34 (d, J =
8.4 Hz, 1H), 7.31-7.28 (m, 2H), 7.13-7.10 (m, 1H), 7.05 (t, J =
7.8 Hz, 1H), 4.90 (s, 1H), 4.07 (d, J = 3.6 Hz, 1H), 3.86 (d of X of
ABX, J = 4.2 Hz, 1H), 3.41 (A/B of ABX, J = 11.4, 5.4 Hz, 1H),
3.34 (A/B of ABX, J = 11.4, 5.4 Hz, 1H); 13C NMR (100 MHz,
CDCl3) δ 196.8, 174.7, 151.9, 139.1, 138.0, 127.4, 127.0, 126.3,
126.1, 124.6, 123.2, 121.5, 120.7, 117.8, 112.9, 110.6, 90.0, 89.0,
84.7, 76.8, 74.4, 63.2; HRMS calcd for C22H18N2O6 406.1165,
found 406.1151; [R]25D þ274 (c 1.1, MeOH).
1
Cl2) (Rf 0.34, 5% MeOH/CH2Cl2). 61-major isomer: H NMR
(600 MHz, CDCl3) δ 7.59 (d, J = 7.8 Hz, 1H), 7.45 (t, J = 7.8 Hz,
1H), 6.91 (t, J = 7.8 Hz, 1H), 6.76 (d, J = 8.4 Hz, 1H), 5.39 (s, 1H),
4.83-4.81 (m, 1H), 4.49 (s, 1H), 4.27 (s, 2H), 3.93 (s, 3H), 3.82 (d,
J = 11.4 Hz, 1H), 1.53 (s, 3H), 1.34 (s, 3H); 13C NMR (100 MHz,
CDCl3) δ 197.7, 172.0, 159.3, 137.6, 125.4, 121.2, 120.4, 114.5,
113.4, 92.4, 85.9, 83.5, 83.3, 81.1, 77.2, 64.7, 53.0, 29.7, 26.6 (2C0s);
IR (thin film, cm-1) νmax 3332, 3001, 2984, 2936, 2853, 1718, 1617,
1487, 1312, 1270, 1255, 1174, 1097, 751; HRMS calcd for C18H19-
NO7 361.1162, found 361.1160; [R]25D -284 (c 0.30, MeOH) (Rf
0.50, 50% EtOAc/hexanes). 61-minor isomer: 1HNMR(400MHz,
CDCl3) δ 7.55 (d, J = 8.0 Hz, 1H), 7.48 (t, J = 7.6 Hz, 1H),
6.90-6.86 (m, 2H), 5.83 (s, 1H), 5.56 (s, 1H), 4.25-4.22 (m, 2H),
4.09 (s, 1H), 3.94-3.90 (m, 4H), 1.48 (s, 3H), 1.34 (s, 3H); 13C
NMR (100 MHz, CDCl3) δ 193.2, 168.2, 158.0, 138.1, 125.4, 120.7,
120.4, 113.9, 113.0, 91.5, 88.8, 84.7, 84.4, 82.1, 65.1, 52.8, 26.7, 26.4;
IR (thin film) νmax 3363, 2988, 2955, 2937, 1745, 1735, 1636, 1487,
1473, 1383, 1323, 1086, 924, 755; HRMS calcd for C18H19NO7
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council (NSERC) of Canada. We
gratefully acknowledge D. Hairsine of the University of
Western Ontario for MS analysis, Dr. Michael C. Jennings,
Aneta Borecki, and Jacquelyn Price of the University of
Western Ontario for X-ray analysis, and Dr. Adrian L.
Schwan of the University of Guelph for the use of his
polarimeter. A.K. is the recipient of an OGS scholarship.
We thank Dr. Jeffrey M. Manthorpe for helpful suggestions,
361.1162, found 361.1163; [R]25 -45 (c 0.31, MeOH) (Rf 0.41,
D
50% EtOAc/hexanes). 62:1H NMR (600 MHz, CD3OD) δ7.69(d,
J = 8.4 Hz, 1H), 7.53 (t, J = 8.4 Hz, 1H), 7.45 (d, J = 7.2 Hz, 1H),
7.38 (s, 1H), 7.32 (d, J = 8.4 Hz, 1H), 7.13 (d, J = 8.4 Hz, 1H), 7.06
(t, J = 7.8 Hz, 1H), 6.92 (t, J = 7.2 Hz, 1H), 6.77 (t, J = 7.2 Hz,
1H), 5.21 (s, 1H), 4.78 (proton under solvent peak), 4.03 (d of X of
ABX, J = 5.4 Hz, 1H), 3.70 (A/B of ABX, J = 12.0, 6.0 Hz, 1H),
3.53 (A/B of ABX, J = 12.0, 6.0 Hz, 1H), 3.09 (s, 3H), 1.59 (s, 3H),
1.23 (s, 3H); 13C NMR (100 MHz, CD3OD) δ 203.5, 174.0, 164.1,
139.1, 138.4, 126.4, 125.8, 125.3, 122.7, 122.2, 120.5, 119.6, 117.5,
113.7, 112.6, 110.8, 93.6, 92.9, 89.0, 86.8, 71.7, 63.3, 52.8, 28.6, 26.6;
IR (thin film, cm-1) νmax 3405, 2926, 2854, 1734, 1706, 1617, 1488,
1466, 1437, 1266, 1092, 1028, 740; HRMS calcd for C26H26N2O7
478.1740, found 478.1754 (Rf 0.07, 50% EtOAc/hexanes). 64: 1H
NMR (600 MHz, CDCl3) δ 8.42 (s, 1H), 7.88 (d, J = 8.4 Hz, 1H),
ꢀ
and Dr. Guillaume Belanger for helpful discussion.
Supporting Information Available: Complete experimental
procedures as well as 1H NMR and 13C NMR, IR, and MS data
for all new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 75, No. 20, 2010 6841