Total synthesis without protection: Three-step synthesis of optically active clavicipitic acids by a biomimetic route

Article abstract of DOI:10.1002/ejoc.200300603

A three-step synthesis of a mixture of optically active cis- and frans-clavicipitic acids 6, which are ergot alkaloids, was achieved, starting from 4-bromoindole (7) and dl-serine (dl-2). This short synthesis was made possible by omitting the protection and deprotection steps from the synthetic route. The key step was the spontaneous cyclization of 4-vinyltryptophan (10) formed from the Heck reaction of 4-bromotryptophan (8) with 2-methyl-3-buten-2- ol (9) in aqueous media. During this investigation, we also found that the palladium-catalyzed reaction of 8 with 9 showed an interesting pH dependence; under strongly basic conditions, the Heck reaction occurred to give a C ⁴-vinylated product 10, whereas an N-allylated product 19b was formed under neutral or weakly basic conditions. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200300603

FULL PAPER
Total Synthesis without Protection: Three-Step Synthesis of Optically Active
Clavicipitic Acids by a Biomimetic Route
Yuusaku Yokoyama,*[a] Hidemasa Hikawa,[a] Masaharu Mitsuhashi,[a] Aki Uyama,[a]
Yasuhiro Hiroki,[a] and Yasuoki Murakami[a]
Keywords: Heck reaction / Tryptophan / Protecting groups / Clavicipitic acids / Total synthesis
A three-step synthesis of a mixture of optically active cis-
and trans-clavicipitic acids 6, which are ergot alkaloids, was
achieved, starting from 4-bromoindole (7) and dl-serine (dl-
During this investigation, we also found that the palladium-
catalyzed reaction of 8 with 9 showed an interesting pH de-
pendence; under strongly basic conditions, the Heck reaction
2). This short synthesis was made possible by omitting the occurred to give a C4-vinylated product 10, whereas an N-
protection and deprotection steps from the synthetic route.
allylated product 19b was formed under neutral or weakly
The key step was the spontaneous cyclization of 4-vinyltryp- basic conditions.
tophan (10) formed from the Heck reaction of 4-bromotrypto- ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
phan (8) with 2-methyl-3-buten-2-ol (9) in aqueous media.
Germany, 2004)
Introduction
In biological systems, the syntheses of the compounds
essential for life are accomplished efficiently and cleanly via
enzymatic catalysis. Therefore, one of the most attractive
goals for a synthetic chemist is to develop a synthetic path-
way similar to the biosynthetic one, but without the use of
enzymes. One of the most important aspects of creating a
biomimetic synthesis is that the use of protecting groups
should be avoided. In this paper, we report[1] the biomimetic
synthesis of optically active clavicipitic acid (6), which is an
ergot alkaloid, biosynthesized from indole (1) and -serine
(-2) via -tryptophan (3) and γ,γ-dimethylallyltryptophan
(4, DMAT)[2] (Scheme 1). The synthetic outline is as follows
(Scheme 2): 1) two-step synthesis of enantiomerically pure
4-bromotryptophan (S-8) from 4-bromoindole (7) and dl-
serine (dl-2); 2) Heck reaction of S-8 with 2-methyl-3-buten-
2-ol (9), followed by spontaneous cyclization of the re-
sulting 4-vinyltryptophan (10). The outstanding features of
this route are that it requires only three steps, and that it
uses the reaction of free amino acids without protecting
groups. Although numerous chemical transformations of
amino acids have been reported,[3] most of them have been
performed with protected amino acids,[4] in order to avoid
side reactions.
Scheme 1. Biosynthetic pathway of the ergot alkaloids
[a]
Faculty of Pharmaceutical Sciences, Toho University,
2-2-1, Miyama, Funabashi, Chiba, 274Ϫ8510, Japan
Fax: (internat.) ϩ81-47-472-1595
E-mail: yokoyama@phar.toho-u.ac.jp
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Scheme 2. Three-step synthesis of clavicipitic acid (6)
1244
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200300603
Eur. J. Org. Chem. 2004, 1244Ϫ1253

Three-Step Synthesis of Optically Active Clavicipitic Acids by a Biomimetic Route
FULL PAPER
merically pure R-12b was obtained as a methyl ester (R-12c)
(Ͼ 99% ee) in 47% yield (Scheme 3, Route A). Thus, enantio-
merically pure S-8 was obtained in 36% overall yield from
7. ODS-silica gel, however, is too expensive to be used in
large-scale synthesis. Therefore, in laboratory scale reac-
tions (ca. 1 g scale), we used inexpensive cellulose gel (Cel-
lulofine GH-25) and polystyrene resin (Diaion SP-207)
for the purification of S-8, and this resulted in a reasonable
isolated yield (30%).
Results
Synthesis of Enantiomerically Pure (S)-4-Bromotryptophan
(S-8)
Following Snyder’s procedure[5] for the direct synthesis of
N-acetyltryptophan (dl-12a) from indole (1) and N-acetyl-
dehydroalanine (11), we heated a mixture of 4-bromindole
(7) and 11 in AcOH in the presence of Ac2O at 80 °C for
1.5 h, and, after esterification of the resulting acid 12b with
(CH3)3SiCHN2, the product was isolated as the methyl ester
12c (Table 1). The 36%[6] yield of 12c (run 1) was much
lower than the yield of 58% reported by Snyder,[5] and was
unsatisfactory. However, when dl-serine (2) was used in-
stead of 11, the yield of 12c increased to 58% (run 2). The
yield was further improved to 63% by the use of N-acetyl-
serine (dl-13) (run 3). Although the yield was almost satis-
factory for synthetic use, 13 is much more expensive than
dl-2. We thought that the lower yield in run 2 compared to
run 3 might be due to decomposition of 2 prior to the for-
mation of dl-N-acetylserine (dl-13). Therefore, dl-13 was
prepared in situ by reaction with Ac2O at a lower tempera-
ture (45 °C, 5 h) prior to the addition of 4-bromoindole (7).
The yield from this reaction was an improved 73% (run 4).
Table 1. Tryptophan synthesis by the reaction of 4-bromoindole (7)
with serine derivatives
Scheme 3. Synthesis of optically active 4-bromotryptophan (S)-8
Optically active S-8 was also synthesized by another
route, using conventional protectionϪdeprotection method-
ology (Scheme 3, Route B). We have already reported the
synthesis[9] in three steps from 4-bromoindole of almost en-
antiomerically pure (95% ee), fully protected N-Boc-1-tosyl-
4-bromotryptophan methyl ester (S-14a). S-14a was detosy-
lated by treatment with Mg/MeOH, and the resulting ester
14b was hydrolyzed with 4% aq. KOH in dioxane. This was
followed by heating with 50% AcOH at 80 °C for 3 h to
give S-8 in 87% overall yield from S-14a, with slight racem-
ization (92% ee). All spectral and physical data for the prod-
ucts from the two routes were completely identical.
Reaction of (S)-4-Bromotryptophan (S-8) and 2-Methyl-3-
buten-2-ol (9)
We have already reported[9] the four-step synthesis of op-
The N-acetyl group can be used in a kinetic resolution to
obtain optically active 4-bromotryptophan (S-8); treatment
of N-acetylamino acids with acylase for this purpose is well
established[7] as a practical and economical method in am- tically active cis- and trans-clavicipitic acids (6) from (S)-N-
ino acid synthesis. N-Acetyl-4-bromotryptophan (dl-12b)[8] Boc-1-tosyl-4-bromotryptophan methyl ester (S-14a) via 4-
was treated with commercially available Aspergillus acylase vinyltryptophan (15a) (Scheme 4). The key step of this
at pH 7.5 for 3 days at 37 °C. Although the reaction pro- route was the one-pot transformation of 15a to 17 by de-
ceeded smoothly, purification of the resulting S-8 was diffi- protection of the Boc group and subsequent cyclization of
cult, because of its high polarity. Various purification meth- the amine 16. Although this synthesis is similar to bio-
ods, including the use of acidic or basic ion-exchange resins, synthesis of 6 (Scheme 1) and is efficient compared to an-
dialysis membranes, and isoelectric precipitation, failed to other reported synthesis,[10] three out of the four steps in
separate 8 from the inorganic salts and enzymes. Finally, this route are deprotection steps. We considered that if the
we found that octadecylsilyl-silica (ODS-silica) gel column Heck reaction were possible with unprotected 4-bromotryp-
chromatography was very effective for the purification of 8 tophan (8), spontaneous cyclization of the resulting 4-vi-
on a small scale (100 mg), giving enantiomerically pure S-8 nylated product 10 should occur, to produce 6 in one pot
(Ͼ 99% ee) in 49% yield as the unprotected form; enantio- (Scheme 2).
Eur. J. Org. Chem. 2004, 1244Ϫ1253
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1245

Y. Yokoyama, H. Hikawa, M. Mitsuhashi, A. Uyama, Y. Hiroki, Y. Murakami
FULL PAPER
Scheme 4. Synthesis of clavicipitic acids (6) by protected route
At first, we attempted the reaction in organic solvents; a occurred as expected, simultaneously with deprotection of
mixture of 4-bromotryptophan (8), Pd(OAc)2, PPh3, and 9 the Boc group, to give a mixture of cis- and trans-6 in one
was heated in a polar organic solvent such as dioxane or pot, in 76% yield.
DMF. However, these reactions gave complex mixtures, as
These results indicated that if the reaction of the free am-
the free amino acid (8) was scarcely soluble in the solvents ino acid 8 could be made to work, then the clavicipitic acids
used. Next, the reaction was carried out in aqueous solu- 6 might be obtained in a one-pot synthesis. Consequently,
tion using a phase-transfer catalyst, according to the the Heck reaction of 8 was attempted again under different
method of Jeffery.[11] The reaction of dl-8 with 9 in the pres- reaction conditions. According to Bumagin,[13] the pal-
ence of K2CO3, PPh3, and nBu4NBr in H2O at 90 °C for ladium-catalyzed carbonylation of p-iodoaniline could be
5 h yielded an unknown product, which gave a negative made to work by changing the base from AcONa to K2CO3
ninhydrin test, along with recovered starting material (8, in aqueous DMF solution; therefore, we carried out the re-
28%). We then changed the phosphane ligand to the water- action using K2CO3 as a base. Surprisingly, the reaction
soluble phosphane ligand TPPTS,[12] instead of using a proceeded smoothly to give the desired product 10 [M ϭ
phase-transfer catalyst. Disappointingly, the sodium salt of K] in 91% yield (Table 2, run 3), which was stable enough to
the N-allylated product 19b was obtained in 34% yield as be isolated as its potassium salt by ODS-silica gel column
the sole product, along with starting material (8, 38%), after chromatography. The structure was confirmed after conver-
separation by ODS-silica gel column chromatography sion into 15c by BOC-protection of the amine and esterifi-
(Scheme 5). This reaction was repeated with -tryptophan cation with TMSCHN2, by comparison with an authentic
(3) under the same reaction conditions, and the N-allylated sample of 15c (see Exp. Sect.). We attempted this reaction
product 19a was produced in 71% yield.
using a variety of bases. NaHCO3 gave a complex mixture
(run 1), and Na2CO3 gave 51% of the desired product (run
2). A strong base (NaOH) gave 91% of 10 [M ϭ Na] (run
4), and a weak organic base (Et3N) gave 55% of the desired
product along with tryptophan (3, 38%) and recovered
starting material 8 (8%) (run 5). These results clearly reveal
that the strength of base used affects the preference as to
which is the reactive site in the starting molecule.
Table 2. Results of Heck reaction of 4-bromotryptophan (8) with
1,1-dimethylallyl alcohol (9)
Scheme 5. Palladium-catalyzed N-allylation of 4-bromotryptophan
(dl-8) with 1,1-dimethylallyl alcohol (9)
These results clearly show that the amino group must be
protected. Therefore, we carried out the reaction of dl-N-
Boc-4-bromotryptophan (dl-18) with 9 in the presence of a
catalytic amount of Pd(OAc)2, TPPTS, and NaHCO3 (3
equiv.) at 100 °C for 5 h in H2O, and the expected 4-vi-
nylated product dl-15b was formed in 70% yield (Scheme 4).
Since spontaneous cyclization of 16 had already been
achieved by deprotection of the Boc group of (S)-15a fol-
lowed by neutralization, the sodium salt (dl-15b) was
treated with 50% aq. AcOH at 80 °C for 3 h. Cyclization
1246
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1244Ϫ1253

Three-Step Synthesis of Optically Active Clavicipitic Acids by a Biomimetic Route
FULL PAPER
We had obtained the pure potassium salt of 10 [M ϭ K] formed[16] into enantiomerically pure R-20, which is an im-
in good yield. Treatment of 10 with 50% AcOH (50 °C, 3 h) portant intermediate for the synthesis of ergot alkaloids,
resulted in a smooth cyclization, giving 6 in 78% yield as a such as chanoclavine-I (21)[17a] and costaclavine (22)[17b]
isomeric mixture. Next, we attempted the one-pot synthesis (Scheme 6).
of optically active 6. After optically active S-8 (92% ee)
underwent the Heck reaction, the reaction mixture was
acidified with 50% aq. AcOH and warmed to 50 °C for
3 h. The one-pot reaction proceeded smoothly to give an
isomeric mixture of the natural products 6[14] in 61% yield
(Scheme 2). After esterification with (CH3)3SiCHN2, each
isomer was isolated as its methyl ester by silica gel column
chromatography, and the optical purities of these com-
pounds were determined by HPLC. Surprisingly, this one-
pot process did not cause any racemization, in spite of the
strong basic conditions used in the Heck reaction (K2CO3
in H2O at 130 °C for 2 h).
Discussion
Scheme 6. Synthesis of optically active ergot alkaloids
Synthesis of Optically Active 4-Bromotryptophan (S-8)
There have been several reports[15] on the direct synthesis
Scheme 7 shows a plausible mechanism for the formation
of tryptophan (3) from indole (1) and serine (2) or its ana- of N-acetyl-4-bromotryptophan (12b) from 4-bromoindole
logues. The aims of those reports were to mimic the bio- (7) and serine (2). Since reaction with -serine (-2) resulted
synthesis. However, the yield and enantiomeric excess from in complete racemization of 12b, direct nucleophilic substi-
the methods described above were far from satisfactory (3% tution at the hydroxyl carbon can be excluded. The reaction
yield, 35% ee) and this procedure cannot be used for practi- seems to be a simple acid-catalyzed Michael addition of 7
cal synthesis. In the present study, we have succeeded in to N-acetyldehydroalanine (11), which might be formed in
developing a practical synthesis of enantiomerically pure situ by the reaction of serine with Ac2O, via N-acetylserine
(S)-8 starting from 4-bromoindole (7) and dl-serine (dl-2) (13).[18] Although tryptophan synthesis by Michael addition
(Table 1 and Route A in Scheme 3). Although S-8 could be of indole to a dehydroamino acid is already known,[19] in
synthesized by another route in higher overall yield (67%) this reported case, special reaction conditions had to be de-
by protecting all the reactive functional groups (route B in signed in order to activate the double bond, as it was deacti-
Scheme 3), it should be emphasized that the former route vated by the electron-donating acetamide group. It also has
required only two steps and did not involve any expensive been reported that Michael addition of indole (1) to a reac-
reagents. In the latter route, four out of the six steps were tive conjugated ketone such as methyl vinyl ketone required
protection and deprotection procedures, and several expens- careful control of acidity to prevent side reactions such as
ive reagents (an equimolar amount of Pd(OAc)2, rhodium dimerization or polymerization.[20] Johnson[21] reported that
catalyst, and optically active ligand) were required. Further- the mixed anhydrides of acrylic acid were reactive enough
more, (R)-N-acetyl-4-bromotryptophan (R-12b), which was to undergo conjugate-addition, giving indole-3-propionic
obtained as a by-product (Scheme 3, Route A), was trans- acid. Snyder[5] also reported that N-acetyldehydroalanine
Scheme 7. Plausible mechanism for the reaction of N-acetyl-4-bromotryptophan (dl-12b)
Eur. J. Org. Chem. 2004, 1244Ϫ1253
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1247

Y. Yokoyama, H. Hikawa, M. Mitsuhashi, A. Uyama, Y. Hiroki, Y. Murakami
FULL PAPER
(11) did not react with indole in the absence of acetic anhy- highly activated palladium catalyst[26c] has been required for
dride. Our present study revealed that two equivalents of the reaction to proceed in an organic solvent. Our results
Ac2O relative to 2 were required to make the reaction pro- reveal that N-allylation is also accelerated in water. The in-
ceed, while one equivalent of Ac2O was enough to make termediates for N-allylation and C4-vinylation (Heck reac-
the reaction proceed with N-acetylserine (13). After the first tion) were known to be the π-allyl palladium complex 25
equivalent of Ac2O reacts with 2, the second equivalent re- and σ-complex 26, respectively (Scheme 8). In order to ex-
acts with the carboxyl group of 13, causing spontaneous plain this selectivity, Pd0 should react selectively with allyl
formation of the oxazolone 23 and dehydration to form the alcohol to form 25[27] under weakly basic conditions,
conjugated oxazolone 24,[22] a species which might be a key whereas under more strongly basic conditions, 26 should be
intermediate in this reaction. Since the double bond of 24 formed before 25. It is not clear at present why the forma-
should be more reactive than that of 11, bearing in mind the tion of the π-allyl palladium complex 25 and σ-complex 26
strong electron-withdrawing effect of the oxazolone group, could be pH dependent in water. We are currently investiga-
Michael addition to this compound would occur smoothly. ting the reaction of aromatic bromides containing an amino
The complete racemization of 12b when -2 was used sug- group, such as bromoanilines and bromobenzylamine.
gests that the asymmetric center was converted into an sp2
carbon during the reaction. A more detailed investigation
of the mechanism of this reaction based on reactions con-
ducted under various conditions, and its application to the
syntheses of other tryptophan derivatives will be reported
in a later paper.
Synthesis of Optically Active Clavicipitic Acids (6)
The one-pot transformation of free 4-bromotryptophan
(S-8) into the clavicipitic acids (6), as shown in Scheme 2,
involves three important new features: 1) the sequential re-
action of unprotected amino acid functional groups; 2) a
controlled chemoselective reaction that only occurs in aque-
ous solution, and 3) no racemization, in spite of the
strongly basic conditions in the Heck reaction.
Scheme 8. Intermediates for the palladium-catalyzed reaction of 4-
bromotryptophan (8)
There have been few reports of consecutive reactions of
free amino acids, such as that which led to 6 in one pot via
the spontaneous cylcization of 10. For example, Casal-
nuovo[23] reported the palladium-catalyzed alkynylation of
2-iodotyrosine using TPPTS in aqueous media, leading in-
itially to the expected alkyne, which then cyclized in situ to
give the benzofuran derivative. Another example involves
the participation of unprotected functional groups of amino
acids, as reported by Brewster.[24] He described the deamin-
ation of free optically active amino acids to give the
hydroxy acid with complete net retention, due to anchim-
eric assistance by the carboxyl group. Our present work has
demonstrated that unprotected functional groups can be ef-
fectively utilized in a one-pot reaction and that the design
of a very short synthesis may be possible by omitting the
protection and deprotection steps.
The third interesting feature was the absence of the race-
mization of amino acids. During the Heck reaction, racem-
ization did not occur, in spite of the strongly basic con-
ditions (3 equiv. of K2CO3 at 130 °C, for 5 h). The results
prompted us to investigate the racemization of amino acids
under basic conditions, and the extent of racemization in
water was compared to that in organic solvent. The exper-
imental details already have been reported.[28] From this
study, we found that free amino acids in water possessed
interesting properties. Free amino acids were more stable in
water than in polar organic solvents and they also did not
racemize under basic conditions in water.
The second feature is a palladium-catalyzed chemoselec-
tive reaction of 4-bromotryptophan (8). Changing the pH
affected which site in the molecule was reactive; N-al-
Conclusion
We have succeeded in developing a three-step synthesis
lylation occurred under weakly basic conditions of optically active clavicipitic acids (6) following a route
(Scheme 5), while the Heck reaction occurred selectively un- very similar to the biosynthetic pathway. This route: 1) is
der strongly basic conditions (Table 2). It should be em- the first practical synthesis of tryptophan from indole and
phasized that this reaction only occurred when water was serine, which mimics tryptophan biosynthesis; 2) provides a
used as the solvent. It is very interesting that pH plays a unique one-pot reaction sequence leading to the clavicipitic
critical role in the chemoselectivity of palladium-catalyzed acids; 3) allows a chemoselective palladium-catalyzed reac-
reactions in aqueous media. It is well known that the Heck tion (i.e. either N-allylation or the Heck reaction) of free 4-
reaction[25] occurs faster in water than in organic solvents, bromotryptophan in aqueous solution, with the choice of
whereas palladium-catalyzed N-allylation with allyl alcohol reactive site in the starting material being determined by
in water has not been reported. A Lewis acid[26a,26b] or the pH; and 4) makes use of the fact that water suppresses
1248
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1244Ϫ1253

Three-Step Synthesis of Optically Active Clavicipitic Acids by a Biomimetic Route
FULL PAPER
acetone, 3:1) to give 124 mg of 12c (36% yield) of a pale yellow
the racemization and decomposition of free amino acids
under strongly basic conditions.
solid, which was recrystallized from EtOAc/hexane to give white
1
needles, m.p. 197Ϫ198 °C. H NMR (CDCl3, 400 MHz): δ ϭ 1.89
We believe that the reactions of unprotected amino acids
would provide the possibility, not only of significantly
shortening synthetic routes through omission of the protec-
tion and deprotection steps,[29] but also of creating new syn-
thetic reactions under aqueous conditions. The present
investigation might be one of the most successful examples
in such methodology.
(s, 3 H) , 3.45 (dd, J ϭ 15.0, 8.0 Hz, 1 H), 3.63 (dd, J ϭ 15.0,
5.4 Hz, 1 H), 3.71 (s, 3 H), 4.95 (ddd, J ϭ 8.0, 8.0, 5.4 Hz, 1 H),
6.07 (br. d, J ϭ 8.0 Hz, 1 H), 6.99 (t, J ϭ 8.0 Hz, 1 H), 7.09 (d,
J ϭ 2.4 Hz, 1 H), 7.26 (d, J ϭ 8.0 Hz, 1 H), 7.27 (d, J ϭ 8.0 Hz,
1 H), 8.34 (br. s, 1 H) ppm. IR (KBr): ν˜ ϭ 3379, 3317, 1735 cmϪ1
.
FAB-MS: m/z ϭ 339 [Mϩ ϩ H] (70%), 341 [Mϩ ϩ H ϩ 2] (65%),
119 (base peak). C14H15N2O3Br (339.18): calcd. C 49.58, H 4.46,
N 8.26; found C 49.56, H 4.47, N 8.30.
Reaction of 7 with dl-N-Acetylserine (dl-13): A mixture of 7 (50 mg,
0.26 mmol), dl-N-acetylserine (dl-13) (68 mg, 0.52 mmol), and
Ac2O (0.01 mL, 1.1 mmol) in AcOH (0.9 mL) was heated at 80 °C
for 1.5 h under argon. The work-up and esterification procedure
was carried out as above to give crude dl-12c (66 mg) as a brown
solid that was purified as above to give 55 mg (63% yield) as pale
yellow solid.
Experimental Section
General Remarks: All reagents and solvents were obtained commer-
cially and used as received, unless otherwise indicated. Acylase
(from Aspergillus genus) was purchased from Tokyo Kasei Kogyo
Co., LTD (TCI), and was used without purification. All melting
points were determined on a Yanagimoto micro-melting hot stage
apparatus and are uncorrected. IR spectra were recorded as KBr
tablets (unless otherwise stated) on a JASCO FT/IR-230 spec-
trometer. NMR spectra were recorded on a JEOL GX-400
(400 MHz) spectrometer (unless otherwise stated) with tetrameth-
ylsilane as the internal reference. The following abbreviations are
used: s, singlet; d, doublet; dd, double doublet; t, triplet; q, quad-
ruplet; m, multiplet; br, broad; dif, diffuse; Ar, aromatic. EI Mass
spectra were measured with a JEOL JMS-01SG-2 or JMS-AM-II-
50 spectrometer using a direct inlet system. FAB Mass spectra were
measured with a JEOL JMS-600H spectrometer using a direct inlet
system. ESI Mass spectra were measured with a Thermoquest
LCQ-10 spectrometer. Optical rotations were measured with the
JASCO DIP-1000. Separations were performed using Kieselgel 60
(70Ϫ230 mesh, Merck) for silica gel column chromatography,
Chromatorex (Fuji Silysia Chemical Ltd.) for ODS-silica gel col-
umn chromoatography, Diaion SP-207 (Mitsubishi Chemical) for
synthetic absorbent chromatography, Celurofine GH-25C (Bio-
chemicals) for cellulose column chromatography, and Kieselgel
GF 254 for thin layer chromatography (TLC). Synthetic resin, SP-
207, was extensively washed with MeOH and H2O before use.
Reaction of 7 with dl-Serine (dl-2): A mixture of dl-2 (165 mg,
1.5 mmol) and Ac2O (0.3 mL, 3.0 mmol) in AcOH (1.8 mL) was
heated at 45 °C for 5 h. 7 (106 mg, 0.54 mmol) was added to the
resulting clear solution, and the mixture was heated at 80 °C for
1.5 h. The work-up, esterification, and purification procedure was
carried out as above to give dl-12c (133 mg, 73% yield) as a pale
yellow solid.
dl-N-Acetyl-4-bromotryptophan (dl-12b): A mixture of dl-2 (2.64 g,
25 mmol) and Ac2O (5.0 mL, 49 mmol) in AcOH (15 mL) was
heated at 80 °C for 1.5 h. 7 (2.4 g, 12.3 mmol) was added to the
resulting clear solution, and the mixture was heated at 80 °C for
1.5 h. The reaction mixture was basified with 30% NaOH and
washed with benzene three times. The aqueous layer was acidified
with conc. HCl, and extracted with organic solvent (EtOAc/ben-
zene, 2:1) three times. The combined organic layers were washed
with H2O and brine and dried with MgSO4. After evaporation of
the solvent, the resulting brown solid (3.67 g) was subjected to neu-
tralized silica gel chromatography [300 g, benzene/EtOAc (1:5) Ǟ
EtOAc] to give 12b in the EtOAc fraction as a pale brown viscous
1
oil (2.81 g, 71%). H NMR ([D6]DMSO, 400 MHz): δ ϭ 1.78 (s, 3
Supporting Information: 1H NMR spectra of 6, (S)-8, 10, dl-12b,
dl-15b, 18, 19a, and 19b (see also the footnote on the first page of
this article).
H), 3.07 (dd, J ϭ 15.0, 8.0 Hz, 1 H), 3.54 (dd, J ϭ 15.0, 5.0 Hz, 1
H), 4.55 (ddd, J ϭ 10.0, 8.0, 5.0 Hz, 1 H), 6.96 (t, J ϭ 8.0 Hz, 1
H), 7.17 (d, J ϭ 8.0 Hz, 1 H), 7.23 (d, J ϭ 2.0 Hz, 1 H), 7.36 (d,
J ϭ 8.0 Hz, 1 H), 8.16 (d, J ϭ 7.8 Hz, 1 H), 11.19 (br. s, 1 H) ppm.
IR (KBr): ν˜ ϭ 3320, 1733, 1640 cmϪ1. EI-MS: m/z ϭ 324 [Mϩ]
(1.4%), 326 [Mϩ ϩ 2] (1.3%), 208 (base peak). C13H13N2O3Br
(325.16): calcd. C 48.02, H 4.03, N 8.62; found C 48.06, H 4.05,
N 8.52%.
dl-N-Acetyl-4-bromotryptophan Methyl Ester (dl-12c). Reaction of
4-Bromoindole (7) with N-Acetyldehydroalanine (11): A mixture of
4-bromoindole (7) (199 mg, 1.0 mmol), N-acetyldehydroalanine
(11) (259 mg, 2.0 mmol), and Ac2O (0.40 mL, 4 mmol) in AcOH
(1.2 mL) was heated at 80 °C for 1.5 h under argon. The mixture
was basified with 30% KOH in an ice-bath and washed with or-
ganic solvent (EtOAc/benzene, 1:1). The aqueous layer was acidi-
Kinetic Resolution of dl-N-Acetyl-4-bromotryptophan (dl-12b): A
solution of dl-12b (151 mg, 0.46 mmol), acylase (61.1 mg), and
fied with conc. HCl and extracted three times with EtOAc. The CoCl2·6H2O (6.5 mg, 0.02 mmol) in 20 m NaH2PO4 buffer (pH
combined organic layers were washed with H2O and brine, and
dried with MgSO4. After evaporation of solvent and azeotropic
removal of AcOH with benzene, the resulting brown solid was dis-
solved in EtOAc (4 mL) and MeOH (2 mL). 2.0  Trimethylsilyldi-
7.51) (30 mL) was shaken at 37 °C for 48 h. The reaction mixture
was acidified with 5% HCl and washed three times with benzene-
EtOAc (1:1). The combined organic layers were washed with H2O,
followed by drying with MgSO4. Evaporation of the solvent gave
azomethane (TMSCHN2, 5.5 mmol, 2.7 mL) in hexane was added crude R-12b (78.3 mg), which was esterified with 2.0  TMSCHN2
to this solution, and the resulting mixture was kept for 30 min at
room temperature. Then, AcOH was added to quench the reaction,
and the solution was diluted with water. The aqueous layer was
and purified by silica gel chromatography, as described for the syn-
thesis of dl-12c, to give R-12c (74 mg, 47%), with an optical purity
of Ͼ 99% ee by HPLC analysis using a chiral column [SUMIPAX
extracted three times with EtOAc, and the combined organic layers OA-4600 (SCAS), hexane/iPrOH/AcOH, 100:20:1]; m.p. 198Ϫ202
were washed with saturated aqueous NaHCO3 and brine, and dried
°C, [α]2D2 ϭ Ϫ16.2 (CHCl3, c ϭ 0.26). All spectroscopic data were
with MgSO4. After evaporation of the solvent, the resulting brown identical with dl-12c. C14H14N2O3Br (339.18): calcd. C 49.58, H
oil (313 mg) was subjected to silica gel chromatography (benzene/
4.46, N 8.26; found C 49.64, H 4.45, N 8.24.
Eur. J. Org. Chem. 2004, 1244Ϫ1253
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1249

Y. Yokoyama, H. Hikawa, M. Mitsuhashi, A. Uyama, Y. Hiroki, Y. Murakami
FULL PAPER
The aqueous layer was neutralized with 30% NaOH, the solution
was reduced to ca. 5 mL by evaporation under reduced pressure
and then subjected to ODS-silica gel column chromatography
(20 g). After removal of inorganic salts by eluting with H2O, S-8
was obtained at the ratio of H2O/MeOH (3:1) as a pale brown solid
(68.3 mg, 49%), which was Ͼ 99% ee by HPLC analysis using a
MeOH solution (5.8 mL) was added. The mixture was kept at room
temperature for 1.5 h. After this time, the mixture was poured into
ice-water, and acidified with AcOH. The solution was extracted
with EtOAc three times, and the combined organic layers were
washed with water and dried with MgSO4. After evaporation of
the solvent, the (S)-4-bromo-N-(tert-butoxycarbonyl)tryptophan
chiral column [Chirabiotic
T
(Astec), EtOH/H2O/AcOH, (S-18) was obtained as a white solid (472 mg). It was dissolved in
100:100:1]; m.p. 262Ϫ263 °C, [α]2D2 ϭ Ϫ64.1 (AcOH, c ϭ 0.304).
50% aqueous AcOH (31 mL) and heated at 80 °C for 3 h. Evapor-
1H NMR ([D6]DMSO, 400 MHz): δ ϭ 2.91 (dd, J ϭ 15.2, 10.0 Hz, ation of the solvent gave pure S-8 as a white solid (319 mg, 90%
1 H), 3.54 (dd, J ϭ 15.2, 4.4 Hz, 1 H), 3.73 (dd, J ϭ 10.0, 4.4 Hz, overall yield from the ester), with an optical purity of 92% ee, as
1 H), 6.96 (dd, J ϭ 7.6, 7.6 Hz, 1 H), 7.16 (d, J ϭ 8.4 Hz, 1 H),
7.29 (d, J ϭ 2.0 Hz, 1 H), 7.35 (d, J ϭ 8.4 Hz, 1 H) ppm. IR (KBr):
ν˜ ϭ 3413, 1653 cmϪ1. ESI-MS: m/z ϭ 283 (Mϩ ϩ H), 285 [(Mϩ
ϩ2)ϩH]. C11H11N2O2Br (283.12): calcd. C 46.66, H 3.92, N 9.89;
found C 46.40, H 4.08, N 9.63.
determined by HPLC.
Palladium Catalyzed N-Allylation of 4-Bromotryptophan (8) with 2-
Methyl-3-buten-2-ol (9): A mixture of dl-8 (50 mg, 0.177 mmol), 9
(400 µL, 5.6 mmol), Pd(OAc)2 (4.4 mg, 0.02 mmol), TPPTS
(20 mg, 0.035 mmol), and AcONa (25 mg, 0.38 mmol) in H2O
(1.5 mL) was heated at 120 °C for 4.5 h in a sealed tube. The sol-
vent was removed under reduced pressure, and the resulting residue
was dissolved in a small amount of AcOH and subjected to ODS-
silica gel column chromatography (10 g). A solvent gradient, start-
ing with H2O and with increasing proportions of MeOH, was used
as eluent. When the eluent was H2O/MeOH (3:1), the starting ma-
terial 8 (19 mg, 35% recovery) was eluted. The second fraction
(H2O/MeOH, 2:1) gave crude dl-4-bromo-N-(3-methyl-2-buten-1-
yl)tryptophan (19b), which was washed with CH3CN to give a col-
orless solid (22 mg, 34% yield). m.p. 186Ϫ188 °C. 1H NMR
(CD3OD, 400 MHz): δ ϭ 1.54 (s, 3 H), 1.67 (s, 3 H), 3.30 (m, 1
H), 3.48 (d, J ϭ 7.0 Hz, 2 H), 3.75 (dd, J ϭ 15.0, 6.0 Hz, 1 H),
4.04 (dd, J ϭ 9.0, 6.0 Hz, 1 H), 5.05 (t, J ϭ 7.0 Hz, 1 H), 6.97 (t,
J ϭ 8.0 Hz, 1 H), 7.19 (d, J ϭ 8.0 Hz, 1 H), 7.26 (s, 1 H), 7.35 (d,
J ϭ 8.0 Hz, 1 H) ppm. IR (Nujol): ν˜ ϭ 3208, 1626 cmϪ1. FAB-
MS: m/z ϭ 351 [Mϩ] (80%), 353 [Mϩ ϩ2] (85%), 69 (base peak).
C16H18BrN2NaO2·2.1H2O (411.06): calcd. C 46.64, H 5.67, N 6.80;
found C 46.23, H 5.45, N 6.99.
Large-Scale Preparation of (S)-4-Bromotryptophan (S-8): After dl-
12b (832 mg, 2.56 mmol) was dissolved in 1.0  NaOH (2 mL),
the solution was diluted with 20 m NaH2PO4 (pH 7.66, 160 mL).
Acylase (86.6 mg, 2598 units) and CoCl2·6H2O (19.8 mg,
0.07 mmol) were added to this solution. The resulting pale red solu-
tion (pH 7.5) was shaken at 37 °C for 72 h. The reaction mixture
was acidified with conc. HCl and washed with benzene/EtOAc (1:1)
three times. The aqueous layer was neutralized with 30% NaOH
and reduced to 50 mL, then subjected to cellulose column chroma-
tography (Cellulofine GH-25-C, 350 mL), eluting with H2O. The
fraction containing S-8 was concentrated to 50 mL and subjected
to SP-207 column chromatography (200 mL), eluting with H2O/
EtOH (3:1) to give pure S-8 as a pale yellow solid (302 mg, 42%).
The combined organic layers were washed with H2O and dried with
MgSO4. Evaporation of the solvent gave crude R-12b (415 mg),
which was esterified with 2.0  TMSCHN2 and purified by silica
gel chromatography, as described for the synthesis of dl-12b, to give
R-12b (392 mg, 45%), with an optical purity of Ͼ 99% ee by HPLC
analysis using a chiral column [SUMIPAX OA-4600 (SCAS), hex-
ane/iPrOH/AcOH, 100:20:1].
Palladium Catalyzed N-Allylation of Tryptophan (3) with 2-Methyl-
3-buten-2-ol (9): A mixture of 3 (97 mg, 0.49 mmol), 9 (1.1 mL
11 mmol), Pd(OAc)2 (11 mg, 0.049 mmol), TPPTS (56 mg,
0.098 mmol), and AcONa (80 mg, 0.98 mmol) in H2O (2.0 mL) was
heated at 120 °C for 5 h in a sealed tube. The work-up and isolation
procedure was carried out as above to give pure N-(3-methyl-2-
buten-1-yl)tryptophan (19a) (93 mg, 71%) as colorless needles. m.p.
224Ϫ225 °C. 1H NMR (CD3OD, 400 MHz): δ ϭ 1.53 (s, 3 H), 1.68
(s, 3 H), 3.21 (dd, J ϭ 16.0, 6.0 Hz, 1 H), 3.44Ϫ3.54 (m, 3 H), 3.84
(dd, J ϭ 9.0, 4.0 Hz, 1 H), 5.03 (t, J ϭ 7.0 Hz, 1 H), 7.04 (t, J ϭ
7.0 Hz, 1 H), 7.12 (t, J ϭ 7.0 Hz, 1 H), 7.21 (s, 1 H), 7.36 (d, J ϭ
7.0 Hz, 1 H), 7.67 (d, J ϭ 7.0 Hz, 1 H) ppm. IR (KBr): ν˜ ϭ 3280,
Alternative Synthesis of (S)-4-Bromotryptophan (S-8) from (S)-4-
Bromo-N-(tert-butoxycarbonyl)-1-tosyltryptophan Methyl Ester (S-
14a)
(S)-4-Bromo-N-(tert-butoxycarbonyl)tryptophan Methyl Ester (14b):
Mg ribbon (1.50 g, 62 mmol) was added to MeOH (19 mL) at 0
°C. After hydrogen gas began to be evolved, a solution of S-14a[9]
(95% ee, 623 mg, 1.13 mmol) in MeOH (15 mL) was added at the
same temperature, and the resulting mixture was vigorously stirred
until the starting material disappeared (20 min). Then the reaction
mixture was quenched with saturated NH4Cl, and extracted with
EtOAc three times. The combined organic layers were washed with
saturated NaHCO3 and brine, and dried with MgSO4. After evap-
oration of the solvent, the resulting residue was subjected to silica
gel column chromatography to give pure (S)-4-bromo-N-(tert-bu-
toxycarbonyl)tryptophan methyl ester (14b, 431 mg, 96%) as a col-
orless solid. The optical purity was 95% by HPLC analysis using a
chiral column [CHIRALPAK AS (Daicel), hexane/iPrOH, 3:1];
m.p. 135Ϫ139 °C. [α]D20 ϭ Ϫ11.0 (CHCl3, c ϭ 2.2). 1H NMR
(CD3OD, 400 MHz): δ ϭ 1.32 (s, 9 H), 3.04 (dd, J ϭ 15.0, 10.0 Hz,
1 H), 3.46 (dd, J ϭ 15.0, 5.0 Hz, 1 H), 3.60 (s, 3 H), 4.30Ϫ4.40 (m,
1 H), 6.97 (t, J ϭ 7.5 Hz, 1 H), 7.18 (d, J ϭ 7.5 Hz, 1 H), 7.24Ϫ7.28
(m, 2 H), 7.37 (d, J ϭ 7.5 Hz, 1 H) ppm. IR (KBr): ν˜ ϭ 3333,
1730. 1689 cmϪ1. FAB-MS: m/z ϭ 396 [Mϩ] (20%), 398 [Mϩ ϩ2]
(18%), 210 (base peak). C17H21N2O4Br (397.26): calcd. C 51.40, H
5.33, N 7.05; found C 51.29, H 5.13, N 7.03.
1607 cmϪ1
.
FAB-MS: m/z
ϭ
273 [Mϩ] (base peak).
C16H20N2O2·H2O (290.36): calcd. C 66.19, H 7.64, N 9.65; found
C 66.20, H 7.23, N 9.61.
dl-4-Bromo-N-(tert-butoxycarbonyl)tryptophan (18): A mixture of
dl-4-bromo-N-(tert-butoxycarbonyl)tryptophan
methyl
ester
(40 mg, 0.10 mmol) and KOH (50 mg, 0.84 mmol) in MeOH
(0.45 mL) and dioxane (0.6 mL) was stirred for 1.5 h at room tem-
perature. The reaction mixture was poured into the ice water, acidi-
fied with AcOH and extracted with EtOAc three times. The com-
bined organic layers were washed with water and dried with
MgSO4. After evaporation of the solvent, dl-4-bromo-N-(tert-bu-
toxycarbonyl)tryptophan (18) was obtained as a colorless amor-
phous powder (38 mg, 99%). 1H NMR ([D6]DMSO, 100 °C,
400 MHz): δ ϭ 1.28(s, 9 H), 3.10 (dd, J ϭ 15.0, 10.0 Hz, 1 H),
3.54(dd, J ϭ 15.0, 5.0 Hz, 1 H), 4.34 (ddd, J ϭ 10.0, 8.0, 5.0 Hz,
1 H), 6.37 (br. s, 1 H), 6.93 (t, J ϭ 7.5 Hz, 1 H), 7.14 (d, J ϭ
7.5 Hz, 1 H), 7.21 (d, J ϭ 2.0 Hz, 1 H), 7.34 (d, J ϭ 7.5 Hz, 1 H),
(S)-4-Bromotryptophan (S-8): The above ester 14b (497 mg,
1.25 mmol) was dissolved in dioxane (7.8 mL), and a 10% KOH/
1250
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1244Ϫ1253

Three-Step Synthesis of Optically Active Clavicipitic Acids by a Biomimetic Route
10.9 (br. s, 1 H) ppm. IR (KBr): ν˜ ϭ 3418, 1696 cmϪ1. FAB-MS: 2 H), 7.15 (s, 1 H), 7.24 (dd, J ϭ 7.5, 1.5 Hz, 1 H), 7.44 (d, J ϭ
FULL PAPER
m/z ϭ 382 [Mϩ] (11%), 384 [Mϩ ϩ 2] (9%), 57 (base peak).
16.0 Hz, 1 H) ppm. IR (KBr): ν˜ ϭ 3418, 1625 cmϪ1. FAB-MS:
C16H19N2O4·0.3H2O (388.65): calcd. C 49.45, H 5.08, N 7.21; m/z ϭ 311 [Mϩ ϩ Na] (25%), 177 (base peak).
found C 49.23, H 4.71, N 7.00.
(S)-4-(3-Hydroxy-3-methyl-1-buten-1-yl)tryptophan Potassium Salt
dl-4-(3-Hydroxy-3-methyl-1-buten-1-yl)-N-(tert-butoxycarbonyl)-
tryptophan, Sodium Salt (dl-15b): A mixture of the above acid 18
(62 mg, 0.16 mmol), Pd(OAc)2 (3.4 mg, 0.015 mmol), TPPTS
(18 mg, 0.031 mmol), NaHCO3 (39.4 mg, 0.47 mmol), and 2-
methyl-3-buten-2-ol (9) (0.72 mL, 6.9 mmol) in H2O (0.72 mL) was
heated at 100 °C for 5 h. Then, the reaction mixture was directly
subjected to ODS-silica gel column chromatography to remove in-
organic salts by eluting with H2O. Then, elution with MeOH gave
the crude vinylated product (dl-15b, 55 mg), which was purified
using preparative TLC (CH2Cl2/MeOH, 5:1) to give pure dl-15b
as pale yellow solid (47 mg, 70%). m.p. 164Ϫ171 °C. 1H NMR
([D6]DMSO, 100 °C, 400 MHz): δ ϭ 1.00Ϫ1.40 (m, 15 H), 2.70
(m, 1 H), 3.46 (dd, J ϭ 15.0, 3.0 Hz, 1 H), 3.95 (dt, J ϭ 9.0, 3.0 Hz,
1 H), 6.12 (d, J ϭ 16.0 Hz, 1 H), 6.92 (dd, J ϭ 8.0, 2.0 Hz, 1 H),
6.95 (t, J ϭ 8.0 Hz, 1 H), 7.05 (br. s, 1 H), 7.17 (dd, J ϭ 8.0, 2.0 Hz,
1 H), 7.48 (d, J ϭ 16.0 Hz, 1 H), 10.32 (br. s, 1 H) ppm. IR (KBr):
ν˜ ϭ 3410, 1607 cmϪ1. FAB-MS: m/z ϭ 411 [Mϩ] (95%, base peak).
C21H27N2O5Na·2H2O (446.47): calcd. C 56.49, H 7.00, N 6.27;
found C 56.48, H 6.85, N 6.15.
(S-10, M ؍ K): A mixture of S-8 (39.2 mg, 0.14 mmol, 92% ee),
Pd(OAc)2 (3.1 mg, 0.014 mmol), TPPTS (15.9 mg, 0.028 mmol),
K2CO3 (27.3 mg, 0.20 mmol), and 2-methyl-3-buten-2-ol (9, 600
µL, 6.25 mmol) in H2O (0.6 mL) was heated at 130 °C for 7 h in a
sealed tube. Isolation was carried out as above to give (S)-4-(3-
Hydroxy-3-methyl-1-buten-1-yl)tryptophan potassium salt (S-10,
M ϭ K) as a pale yellow amorphous solid (40.2 mg, 89%), whose
optical purity was 91% ee, as analyzed by HPLC after tert-butoxy-
carbonylation and esterification (see below). All spectroscopic data
were identical with dl-10, [M ϭ K] (see above) [α]2D0 ϭ Ϫ75.4
(MeOH, c ϭ 1.18). C16H19KN2O3·2.1H2O (364.27): calcd. C 52.76,
H 6.37, N 7.69; found C 52.59, H 6.23, N 7.68.
Determination of Optical Purity of S-10 [M ؍ K]: A solution of
Boc2O (4.6 mg, 0.021 mmol) in 1,4-dioxane (50 µL) was added,
with vigorous stirring, to a solution of S-10 [M ϭ K] (2.0 mg,
0.0061 mmol) and 1  aqueous NaOH (25 µL) in 1,4-dioxane (50
µL) and H2O (50 µL) at room temperature. After 1 h, the reaction
was quenched by pouring into ice-water, and the mixture was ex-
tracted with EtOAc three times. The combined organic layers were
washed with saturated aqueous NaHCO3 and brine, and dried with
MgSO4. After evaporation of the solvent, a diethyl ether solution
of CH2N2 (1.5 mL) was added to the resulting residue (8.9 mg) at
0 °C, and the mixture was allowed to stand for 15 min. After re-
moval of the solvent, the product was purified by preparative TLC
to a give pale yellow amorphous solid (2.1 mg, 85%), whose optical
purity was 91% ee, as determined by HPLC analysis (CHIRAL-
PAK AS, hexcane/iPrOH, 7:1). All spectroscopic data were ident-
ical with authentic dl-4-(3-hydroxy-3-methyl-1-buten-1-yl)-N-(tert-
butoxycarbonyl)tryptophan methyl ester (dl-15c) (see below).
dl-Clavicipitic Acids (6) from dl-15b: A solution of dl-15b (14.1 mg,
0.0343 mmol) in 50% aqueous AcOH (3.0 mL) was heated at 80
°C for 3 h. After the starting material had disappeared (3 h), as
monitored by HPLC [Supelcosil LC-ABZ (SUPELCO), 20 m
NaHPO4/CH3CN, 5:1], the reaction mixture was directly subjected
to ODS-silica gel column chromatography using gradient elution
from H2O to MeOH. A mixture of cis- and trans-clavicipitic acids
(dl-6) was obtained at the fraction of H2O/MeOH (2:1), as pale
yellow solid (7.0 mg, 76%), which was recrystallized from H2O/
MeOH to give pale yellow prisms. decomp. 278Ϫ282 °C. 1H NMR
(CD3OD, 400 MHz): δ ϭ 1.85 (s, [cis], 3 H), 1.90 (s, [cis], 3 H) 1.92
(s, [trans], 3 H), 1.93 (s, [trans], 3 H), 3.13 (dd, J ϭ 20.0, 14.0 Hz
[trans], 1 H), 3.35 (dd, J ϭ 20.0, 14.0 Hz [cis], 1 H), 3.72 (dd, J ϭ
14.0, 2.0 Hz [cis], 1 H), 3.76 (dd, J ϭ 14.0, 2.0 Hz [trans], 1 H),
3.95 (br. d, J ϭ 14 Hz [trans], 1 H), 4.19 (dd, J ϭ 15.0, 2.0 Hz [cis],
1 H), 5.30 (br. d, J ϭ 11.0 Hz [cis], 1 H), 5.49 (br. d, J ϭ 11.0 Hz
[trans], 1 H), 5.55 (br. d, J ϭ 11.0 Hz [trans], 1 H), 5.85 (d, J ϭ
11.0 Hz [cis], 1 H), 6.79 (d, J ϭ 9.0 Hz [cis], 1 H), 6.82 (d, J ϭ
9.0 Hz [trans], 1 H), 7.03 (t, J ϭ 9.0 Hz [cis], 1 H), 7.07 (t, J ϭ
9.0 Hz [trans], 1 H), 7.13 (s, [cis], 1 H), 7.17 (s, [trans], 1 H), 7.26
(d, J ϭ 9.0 Hz [cis], 1 H), 7.29 (d, J ϭ 9.0 Hz [trans], 1 H) ppm.
IR (KBr): ν˜ ϭ 3700Ϫ2800, 1625 cmϪ1. EI-MS: m/z ϭ 270 [Mϩ]
(54%),44 (base peak). C16H18N2O2·H2O (288.24): calcd. C 66.65,
H 6.99, N 9.72; found C 66.98, H 6.76, N 9.01.
dl-N-(tert-Butoxycarbonyl)-4-(3-hydroxy-3-methyl-1-buten-1-yl)-
tryptophan Methyl Ester (dl-15c): Mg ribbon (239 mg, 9.84 mmol)
was added to MeOH (6.8 mL) and the mixture was stirred at room
temperature. After hydrogen gas began to be evolved, a solution
of dl-4-(3-hydroxy-3-methyl-1-buten-1-yl)-N-(tert-butoxycarbonyl)-
1-tosyltryptophan methyl ester[9] (dl-15a, 201 mg, 0.362 mmol) in
MeOH (15 mL) was added. Further Mg (225 mg and 230 mg) was-
added after 30 min and 1 h, respectively, and the reaction was
stirred for a further 1 h at ambient temperature. After this time,
the reaction mixture was quenched with saturated aqueous NH4Cl
and extracted with EtOAc three times. The combined organic layers
were washed with saturated NaHCO3 and brine and dried with
MgSO4. After evaporation of the solvent, the resulting residue
(160 mg) was subjected to silica gel column chromatography (ben-
zene/EtOAc, 4:1) to give the dl-15c (131 mg, 90%) as colorless solid,
which was recrystallized from benzene/hexane to give colorless
prisms, m.p. 74Ϫ84 °C. 1H NMR (C6D6, 400 MHz): δ ϭ 1.33 (s, 9
H), 1.50 (s, 3 H), 1.53 (s, 3 H), 3.03 (s, 3 H), 3.09 (dd, J ϭ 14.0,
8.0 Hz, 1 H), 3.24Ϫ3.34 (m, 1 H), 4.92 (dd, J ϭ 18.0, 8.0 Hz, 1 H),
5.16 (d, J ϭ 8.0 Hz, 1 H), 6.37 (s, 1 H), 6.38 (d, J ϭ 16.0 Hz, 1 H),
6.87 (br. s, 1 H), 6.95 (d, J ϭ 8.0 Hz, 1 H), 7.24 (d, J ϭ 8.0 Hz, 1
H), 7.69 (d, J ϭ 16.0 Hz, 1 H) ppm. IR (KBr): ν˜: ν˜ ϭ 3391, 1694
dl-4-(3-Hydroxy-3-methyl-1-buten-1-yl)tryptophan Potassium Salt
(10) [M ؍ K]: A mixture of dl-8 (13.2 mg, 0.047 mmol), Pd(OAc)2
(1.3 mg, 0.0058 mmol), TPPTS (5.2 mg, 0.0091 mmol), K2CO3
(9.5 mg, 0.0687 mmol), and 2-methyl-3-buten-2-ol (9, 100 µL,
0.597 mmol) in H2O (0.2 mL) was heated at 130 °C for 6 h in a
sealed tube. After the starting material had disappeared (3 h), as
monitored by HPLC [Supelcosil LC-ABZ (SUPELCO), 20 m
Na2HPO4/CH3CN, 5:1], the reaction mixture was directly subjected
to ODS-silica gel column chromatography using gradient elution
from H2O to MeOH. dl-4-(3-Hydroxy-3-methyl-1-buten-1-yl)tryp-
tophan potassium salt (dl-10, M ϭ K) was obtained at the fraction
cmϪ1
. FAB-MS: m/z ϭ
402 [Mϩ] (18%), 69 (base peak).
C22H30N2O5 (402.48): calcd. C 65.65, H 7.51, N 6.96; found C
65.57, H 7.57, N 6.77.
of H2O/MeOH (2:1), as a pale yellow amorphous solid (13.8 mg,
1
91%). H NMR (CD3OD, 400 MHz): δ ϭ 1.39 (s, 3 H), 1.40 (s, 3 dl-Calvicipitic Acids (6) from dl-4-(3-Hydroxy-3-methyl-1-buten-1-
H), 2.86 (dd, J ϭ 16.0, 12.0 Hz, 1 H), 3.68 (d, J ϭ 12.0 Hz, 1 H),
yl)tryptophan Potassium Salt (dl-10, M ؍ K): A solution of dl-10
3.74 (d, J ϭ 16.0 Hz, 1 H), 6.19 (d, J ϭ 16.0 Hz, 1 H), 7.0Ϫ7.1 (m, [M ϭ K] (13.8 mg, 0.0423 mmol) in 50% aqueous AcOH (2.0 mL)
Eur. J. Org. Chem. 2004, 1244Ϫ1253
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1251

Y. Yokoyama, H. Hikawa, M. Mitsuhashi, A. Uyama, Y. Hiroki, Y. Murakami
FULL PAPER
ination of this experiment showed that in fact, the reaction gave
N-acetyl-4-bromotryptophan (12b); see ref. [2].
K. Irie, A. Ishida, T. Nakamura, T. Ohishi, Chem. Pharm. Bull.
1984, 32, 2126Ϫ2139.
Pure acid dl-12b was obtained after chromatography with neu-
tralized silica gel (hexane/EtOAc, 1:2) in 71% yield (see Exp.
Sect.).
was warmed at 50 °C for 3 h. After the starting material had disap-
peared (3 h), as monitored by HPLC [Supelcosil LC-ABZ (SUP-
ELCO), 20 m NaHPO4/CH3CN, 5:1], the reaction mixture was
directly subjected to ODS-silica gel column chromatography using
gradient elution from H2O to MeOH. A mixture of cis- and trans-
clavicipitic acids (dl-6) was obtained at the fraction of H2O/MeOH
(2:1), as a pale yellow solid (8.9 mg, 78%).
[7]
[8]
[9]
Y. Yokoyama, T. Matsumoto, Y. Murakami, J. Org. Chem.
1995, 60, 1486Ϫ1487.
One-pot Synthesis of Optically Active Clavicipitic Acids (6) from
[10] [10a]
H. Shinohara, T. Fukuda, M. Iwao, Tetrahedron 1999, 55,
(S)-4-Bromotryptophan (S-8):
A mixture of S-8 (26.1 mg,
10989Ϫ11000. [10b] M. Iwao, F. Ishibashi, Tetrahedron 1997, 53,
51Ϫ58.
0.092 mmol, 92% ee), Pd(OAc)2 (2.2 mg, 0.0098 mmol), TPPTS
(10.3 mg, 0.018 mmol), K2CO3 (19.2 mg, 0.14 mmol), and 2-
methyl-3-buten-2-ol (9, 400 µL, 4.15 mmol) in H2O (0.4 mL) was
heated at 130 °C for 8 h in a sealed tube. After this time, the reac-
tion mixture was allowed to cool to room temperature, 60% aque-
ous AcOH (1.6 mL) was added, and the resulting mixture was
warmed at 60 °C for 2 h. After this time, the reaction mixture was
directly subjected to ODS-silica gel column chromatography by
gradient elution from H2O to MeOH. A mixture of cis and trans
clavicipitic acids (S-6) was obtained at the fraction of H2O/MeOH
(2:1), as a pale yellow solid (15.1 mg, 61%). The optical purity of
each isomer was analyzed by HPLC after esterification (see below),
d.p. 253Ϫ260 °C (ref.[14] m.p. 262). C16H18N2O2·0.4H2O (277.54):
calcd. C 69.24, H 6.83, N 10.09; found C 69.45, H 6.98, N 9.73.
All spectroscopic data were identical with dl-6.
[10c]
M. Somei, S. Hamamoto, K. Nakagawa, F. Yam-
[10d]
ada, T. Ohta, Heterocycles 1994, 37, 719Ϫ24.
A. Boyles,
D. E. Nichols, J. Org. Chem. 1988, 53, 5128Ϫ30. [10e] M. Mats-
umoto, H. Kobayashi, N. Watanabe, Heterocycles 1987, 26,
[10f]
1197Ϫ202.
A. P. Kozikowski, M. N. Greco, J. Org. Chem.
1984, 49, 2310Ϫ2314. [10g] H. Muratake, T. Takahashi, M. Nat-
sume, Heterocycles 1983, 20, 1963Ϫ8.
[11]
[12]
T. Jeffery, Tetrahedron Lett. 1994, 35, 3051Ϫ3054.
J. S. Merola, H. Stetlzer, in: Aqueous-Phase Organometaallic
Catalysis: Concepts and Application) (Eds.: B. Cornils, W. A.
Hermann), Wiley-VCH: Weinheim, 1998; pp. 58Ϫ89.
N. A. Bumagin, K. V. Nikitin, I. P. Beletskaya, J. Organomet.
Chem. 1988, 358, 563Ϫ565.
[13]
[14]
Clavicipitic acids were isolated as an isomeric mixture, and the
melting point of the mixture has been reported: C.-W. Robbers,
H. Otsuka, H. G. Floss, J. Org. Chem. 1980, 45, 1117Ϫ1121.
[15] [15a]
D. E. Metzler, M. Ikawa, E. E. Snell, J. Am. Chem. Soc.
Determination of Optical Purity of cis and trans-Clavicipitic Acids
(6): A methanolic solution of the above isomeric mixture (23.1 mg,
0.086 mmol) was esterified with an ethereal solution of CH2N2.
After evaporation of the solvent, each isomer was isolated by pre-
parative TLC to give the pure cis- (1.7 mg) and trans- (2.8 mg) iso-
mers as pale yellow needles. The optical purities of the isomers
were 92% (cis) and 93% (trans), based on HPLC analysis (CHIR-
ALPAK AS, hexane/iPrOH, 50:1). trans-Clavicipitic acid methyl es-
ter; m.p. 144Ϫ148 °C (ref.[9] 158Ϫ160 °C). [α]2D2 ϭ Ϫ116.3 (EtOH,
c ϭ 0.14). [ref.[9] [α]D22 ϭ Ϫ129.1 (EtOH, c ϭ 0.72), Ͼ 99% ee]. All
spectroscopic data were identical with reported trans-clavicipitic
acid methyl ester.[9] cis-Clavicipitic acid methyl ester; m.p. 132Ϫ134
°C (ref.[9] 144Ϫ144.5 °C). [α]D22 ϭ Ϫ178.1 (EtOH, c ϭ 0.09). [ref.[2]
[α]2D2 ϭ Ϫ195.3 (EtOH, c ϭ 0.39), Ͼ 99% ee]. All spectroscopic
data were identical with reported cis-clavicipitc acid methyl ester.[9]
[15b]
1954, 76, 648Ϫ652.
W. Weiner, J. Winkler, S. C. Zimmer-
man, W. Czarnik, R. Breslow, J. Am. Chem. Soc. 1985, 107,
[15c]
4093Ϫ4094.
Y. Murakami, Y. Hisaeda, T. Miyajima, H.
Sakata, J. Kikuchi, Chem. Lett. 1993, 645Ϫ648 and references
are cited therein.
[16]
Y. Yokoyama, K. Osanai, M. Mitsuhasi, K. Kondo, Y. Murak-
ami, Heterocycles 2002, 55, 653Ϫ659.
[17] [17a]
Y. Yokoyama, K. Kondo, M. Mitsuhashi, Y. Murakami,
[17b]
Tetrahedron Lett. 1996, 37, 9309Ϫ9312.
K. Osanai, Y.
Yokoyama, K. Kondo, Y. Murakami, Chem. Pharm. Bull. 1999,
47, 1587Ϫ1590.
[18]
We observed the formation of 11 by heating serine (2) or N-
acetylserine (13) with Ac2O in AcOH, followed by the direct
evaporation of the solvent and Ac2O. A similar reaction has
also been reported: M. Nakagawa, Y. Torisawa, T. Hosoka,
K. Tanabe, F. Tavet, M. Aikawa, T. Hino, Heterocycles 1993,
35, 1167Ϫ1170.
[19] [19a]
C. Balsamini, G. Diamantini, A. Duranti, G. Spadoni, A.
Tontini, Synthesis 1995, 370Ϫ372. [19b] G. Spadoni, C. Balsam-
ini, A. Bedini, E. Duranti, A. Tontini, J. Heterocyclic Chem.
1992, 29, 305Ϫ309. [19c] G. Tarizia, C. Balsamini, G. Spandoni,
E. Duranti, Synthesis 1988, 514Ϫ517.
Acknowledgments
This work was supported by a Grant-in-Aid for Scientific Research
[(C)-No 13672237] from the Ministry of Education, Culture,
Sports, Science, and Technology of Japan.
[20] [20a]
J. S. Yadav, S. Abraham, B. V. S. Rekky, G. Sabitha, Syn-
[20b]
thesis 2001, 2165Ϫ2169.
bayashi, Adv. Synth. Catal. 2001, 343, 174Ϫ176.
K. Manabe, N. Aoyama, S. Ko-
[20c]
P. Har-
rington, M. A. Kerr, Can. J. Chem. 1998, 76, 1256Ϫ1265.
H. Johnson, D. G. Crosby, J. Org. Chem. 1960, 25, 569Ϫ570.
Snyder also proposed the same intermediate 24, see ref.[5]
A. L. Casalnuovo, J. C. Calabrese, J. Am. Chem. Soc. 1990,
112, 4324Ϫ4330.
P. Brewster, F. Hiron, C. K. Ingold, P. A. D. S. Rao, Nature
1950, 166, 179Ϫ180.
Comprehensive review of the Heck reaction in aqueous media;
I. P. Beletskaya, A. V. Cheprakow, chapter 6, in: Organic syn-
thesis in Water (Ed.: P. A. Grieco), Blackie Academic and Pro-
fessional, London, 1998; pp. 141Ϫ213.
[21]
[22]
[23]
[1]
Part of this work has previously been published as a communi-
cation; see: Y. Yokoyama, H. Hikawa, M. Mitsuhashi, A. Uy-
ama, Y. Murakami, Tetrahedron Letters 1999, 40, 7803Ϫ7806.
[24]
[25]
[2]
H. G. Floss, Tetrahedron 1976, 32, 873Ϫ912.
[3]
G. M. Coppola, H. F. Schuster, Asymmetric Synthesis. Con-
struction of Chiral Molecules Using Amino Acids, John Wiley &
Sons: New York, 1987.
G. C. Barrett, in: Chemistry and Biochemistry of Amino Acids
(Eds.: G. C. Barrett), Chapman and Hall: London, 1985; pp.
354Ϫ375.
H. R. Snyder, J. A. MacDonald, J. Am. Chem. Soc. 1955, 77,
1257Ϫ1259.
[4]
[26] [26a]
S.-C. Yong, C-W. Hung, J. Org. Chem. 1999, 64,
5000Ϫ5001. [26b] M Sakamoto, I. Shimizu, A. Yamamoto, Bull.
Chem. Soc. Jpn. 1996, 69, 1065Ϫ1078. [26c] F. Ozawa, H. Okam-
oto, S. Kawagishi, S. Yamamoto, T. Minami, M. Yoshifuji, J.
Am. Chem. Soc. 2002, 124, 10968Ϫ10969.
[5]
[6]
Although we have reported that the reaction of 4-bromoindole
(7) with N-acetyldehydroalanine (11) did not proceed, reexam-
1252
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1244Ϫ1253

Three-Step Synthesis of Optically Active Clavicipitic Acids by a Biomimetic Route
FULL PAPER
[27]
[29]
Since this reaction did not take place in the absence of a pal-
ladium catalyst, it was clear that the reaction did not proceed
by an SN2 or SN2Ј mechanism, but rather via a π-allyl-pal-
ladium complex 25 (Scheme 8).
Somei and Yamada reported the synthesis of dl-6,7-secoagro-
clavine and dl-aurantioclavine without using any protective
[29a]
groups.
M. Somei, F. Yamada, Chem. Pharm. Bull. 1984,
[29b]
32, 5064Ϫ5065.
mei, Chem. Pharm. Bull. 1985, 33, 2162Ϫ2163.
Received September 26, 2003
F. Yamada, Y. Makita, T. Suzuki, M. So-
[28]
H. Yokoyama, H. Hikawa, Y. Murakami, J. Chem. Soc., Perkin
1 2001, 1431Ϫ1434.
Eur. J. Org. Chem. 2004, 1244Ϫ1253
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1253