Spirolactones of tyrosine: Synthesis and reaction with nucleophiles

Article abstract of DOI:10.1039/a801803k

Several quinonoidal spirolactones (1-oxaspiro[4,5]deca-6,9-diene-2,8-diones) 12 have been synthesized chemically or electrochemically by oxidation of the corresponding tyrosine phenols 3. The spirolactones are of considerable value in peptide chemistry, b

Full text of DOI:10.1039/a801803k

View Article Online / Journal Homepage / Table of Contents for this issue
Spirolactones of tyrosine: synthesis and reaction with nucleophiles
Antun Hutinec, Athanassios Ziogas, Medhat El-Mobayed and Anton Rieker*
Institute of Organic Chemistry, University of Tübingen, Auf der Morgenstelle 18,
D-72076 Tübingen, Germany
Several quinonoidal spirolactones (1-oxaspiro[4,5]deca-6,9-diene-2,8-diones) 12 have been synthesized
chemically or electrochemically by oxidation of the corresponding tyrosine phenols 3. The spirolactones
are of considerable value in peptide chemistry, because they react as active esters with amino acid esters
to give dipeptides. In the latter, the side chain of the tyrosine moiety is present as a quinol (4-hydroxy-
cyclohexa-2,5-dien-1-one) system, which can be reduced to the genuine tyrosyl dipeptide. Unsubstituted
spirolactones also react as active esters but give dipeptides only in modest yields.
Introduction
Results and discussion
Spirolactones of tyrosines have played a significant role in
organic chemistry for several years. They are intermediates in
the synthesis of some alkaloids and antibiotics such as arano-
rosin A1–4 or anticapsin B.5 However, they may also serve as
Synthesis of tert-butyl-substituted tyrosine derivatives
Two different methods have been used to synthesize tert-butyl-
substituted tyrosine derivatives. The first one proceeds via a
formylated amino malonic acid ester and was first described by
Teuber et al.8 (Scheme 1). As starting phenol they used 2,6-di-
O
O
OH
OH
CO2Et
O
O
O
H
C
NHR2
CO2Et
H
O
O
CO2H
CH2
C
(6′R)
C6H13
CH2
R1
N
H
OH
EtO2C
CO2Et
NH2
CH3 CH3
NHR2
2
1a R1 = Cl
1b R1 = Br
A
B
1. NaOH
2. HCl
synthons useful in peptide chemistry, as is shown in this
paper.
3. –CO2
OH
OH
The key-step in their synthesis is the oxidative cyclization of
-tyrosine and its derivatives 3 (see Scheme 1 for synthesis).
Since tyrosines are redox-active phenols, they can be oxid-
ized chemically or electrochemically, to phenoxy radicals or
phenoxenium cations,6 depending on the applied oxidation
potential and the extant proton concentration. The relevant
oxidation potentials can be determined from cyclic voltam-
mograms (CVs) of 3. In this context, the phenoxenium ions
are the desired species, because they should react intramolecu-
larly with the nucleophilic carboxy group of the side chain to
give the spirolactones 12 (Scheme 2) with an autoprotected
C-terminus and side-chain of the original tyrosine. These spiro-
lactones were indeed formed in very high yields when the
intermediate phenoxenium ions were stabilized by tert-butyl
groups in both ortho-positions of the oxylium oxygen. Other
reasons to use tert-butyl substituted tyrosines were the higher
stability of the tert-butylated spirolactones 12 themselves, as
well as the less pronounced adsorption of the corresponding
tyrosines at the anode in the case of the electrochemical syn-
thesis. The unsubstituted spirolactones could be synthesized
only in modest yields (up to 40%), even in the chemical oxid-
ation with phenyliodine() bis(trifluoroacetate) (PIFA), where
no adsorption effects were interfering. The oxidation of the
ring-unsubstituted tyrosine derivatives with N-bromosuccin-
imide (NBS) gave the 7,9-dibrominated spirolactones,7 whereas
the tert-butylated species furnished 12 without bromination.
The spirolactones 12, as expected, proved to be active esters
capable of coupling with nucleophiles like amines or amino
acid esters to form tyrosine amides or dipeptides.
Aniline
for R2 = For
CH2
CH
R2HN CO2H
CH2
C H
H2N CO2H
4
3
Compound R2
3a
3b
3c
3d
3e
Ac
Bz
Fmoc
For
Z
Scheme 1
tert-butyl-4-chloromethylphenol 1a which was coupled with
diethyl 2-(formylamino)malonate to give the phenolic malonate
2. After hydrolysis and decarboxylation the N-protected di-tert-
butyltyrosine 3d (R2 = For, formyl) was obtained. The formyl
group was cleaved by heating 3d in aniline to give 4. We used
the brominated phenol 1b instead of 1a, since 1a could not be
prepared in good yields.
The synthesis of 4 was optimized, and the yields of the
respective steps could be raised as follows: the coupling with
diethyl 2-(formylamino)malonate to give 2 (R2 = For) from 81
to 92%, the hydrolysis of the ester groups in 2 from 68 to 89%,
J. Chem. Soc., Perkin Trans. 1, 1998
2201

View Article Online
OH and unprotected carboxy group, as in 3a and 3d, give oxid-
ation peaks at 1100–1200 mV; a separate peak for the oxidation
of the carboxy group cannot be found up to 1600 mV. On the
other hand, compound 3c reveals two oxidation peaks, one at
1210 mV due to phenolic oxidation, and one at 1560 mV which
might be attributed to the oxidation of the carboxy group. The
N- and side-chain-protected tyrosine 710 also reveals a peak at
1540 mV (Fig. 1b), however, this may be due to phenol ether
oxidation, and does not give unequivocal proof of the oxid-
ation of the carboxy group at this potential; the totally pro-
tected tyrosine 810 shows a phenol ether oxidation at 1380 mV.
Hence, we measured the CV of acetic acid under the same
conditions and found an oxidation peak at 1580 mV in good
agreement with that of 3c. The fact that 3a and 3d do not show
an oxidation peak in this area is not well understood. The peak
may be shifted to higher potentials; for 3a a shoulder was
observed at 1.9 V in the end-rise of the current of the support-
ing electrolyte oxidation. These results are in agreement with
those obtained with other phenols,11 also showing that the
tyrosines 3 are first oxidized in one-electron transfers to the
cation radicals 9 (Scheme 2), which deprotonate to the neutral
phenoxyl radicals 11. These are further oxidized at the given
potential to form the phenoxenium ions 10. Thus, the oxidation
peaks at 1100–1200 mV correspond to overall ECE reactions of
3 to 10.
In the presence of an excess of a base (NH4OH or potassium
tert-butoxide), the phenolic and carboxylic functions of the
N-protected amino acid exist as anions, i.e. phenolate and carb-
oxylate, respectively. CV Measurements in this medium give
results different from those in neutral solutions. The oxidation
peak is now shifted to Ϫ460 to Ϫ640 mV and is accompanied
by a re-reduction peak (Ϫ560 to Ϫ740 mV). Thus, for example,
the oxidation peak of the phenolate of 5 can be located at Ϫ537
mV and the corresponding re-reduction peak at Ϫ597 mV
(details given in Fig. 1c). There is no dependence of the peak
potential on the scan rate ν, ∆E = Epox Ϫ Epred = 60 mV, and iox/
ired ~ 1, which means that a reversible one-electron transfer
occurs with E0 = (Epox ϩ Epred)/2 = Ϫ567 mV between phenol-
ate 5 , the electrode, and phenoxyl radical 5 . Similarly, the
phenolates 3Ϫ are oxidized to the phenoxyl radicals in the
presence of a base, but here the one-electron transfer is quasi-
reversible, as can be seen from the increasing ∆E with increas-
ing scan rate ν. For the case of 3a the relevant data are given in
Table S1 of the Supplementary Material.† The E0 of 3aϪ is
Ϫ691 mV and therefore more negative than that of 5Ϫ. This
seems to be a result of the presence of the carboxylate anion in
the latter.
Fig. 1 Cyclic voltammograms in 0.1  (Et)4NBF4–MeCN. a: 5, ν = 500
mV sϪ1, c = 1 × 10Ϫ3 mol lϪ1; b: 7, ν = 200 mV sϪ1, c = 1 × 10Ϫ3 mol lϪ1
c: 5, ν = 200 mV sϪ1, c = 2 × 10Ϫ4 mol lϪ1, c (KOBut) = 4 × 10Ϫ4 mol lϪ1
;
.
the decarboxylation to give 3 from 87 to 93% and the cleavage
of the formyl group in 3d from 83 to 96%. The overall yield of 4
could therefore be increased from 40 to 73%.
With the second method described by Cohen et al.9 we syn-
thesized N-benzoyltyrosine 3b (R2 = Bz) in adequate yields.
Electrochemical oxidation
Cyclic voltammetry. By means of cyclic voltammetry it is
possible to get important information on the redox behaviour
of individual parts of the tyrosine molecule with simultaneous
masking of all other functions.
Ϫ
ؒ
The CVs of N-protected tyrosine esters, like 510 (Fig. 1a),
OH
OH
CH2
CH
CH2
CH
The oxidation of the carboxylate, on the other hand, occurs
nearly at the same potential as that of the carboxy group itself
(i.e. at about 1500 to 1600 mV), which is due to the strong
acidity of the latter producing the carboxylate in equilibrium
(pKa of the unsubstituted tyrosine = 2.2). This is also demon-
AcNH
CO2Me
CF3CONH
CO2Me
5
6
ox
strated by the fact that the Ep of acetic acid (see above) in
OCH2C6H5
OMe
MeCN does not change by adding KOBut to the solution.
Preparative electrochemistry. As a consequence of the above
potential considerations, for a preparative-scale electrolysis,
N-protected tyrosines 3 were anodically oxidized to the phen-
oxenium ions 10 at potentials of 1300–1400 mV under neutral
conditions or in the presence of weak bases such as lutidine.
Then, the nucleophilic carboxy group of the side chain indeed
added to the para-position of the ions 10 to give the spiro-
lactones 12 (Scheme 2). Unfortunately, the electrochemical
oxidation was affected by strong adsorption effects; very often,
a fast decrease in the current was observed, caused by coating
of the electrode leading to a poor conversion of the tyrosines.
CH2
CH2
CH
CH
BocNH
CO2H
ZNH
CO2Me
7
8
show completely irreversible oxidation peaks at 1200–1250 mV
(vs. Ag/10Ϫ2  Agϩ/MeCN). These peaks correspond to the
oxidation of the phenolic OH-group, and their potentials are
characteristic of 2,6-di-tert-butyl-4-alkylphenols.11 The corre-
sponding oxidation peak of a non-tert-butylated tyrosine, like
610 is shifted to more positive potentials (1330 mV), which is
reasonable because the tert-butyl groups exert a weak electron-
donating effect. Similarly, the tyrosines 3 with a free phenolic
† This data is available as supplementary material (SUPPL. NO. 57389,
1 p.). For details of the Supplementary Publications Scheme see
‘Instructions for Authors’, J. Chem. Soc., Perkin Trans. 1, available via
the RSC Web page (http://www.rsc.org/authors).
2202
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Table 1 Oxidation of tyrosine derivatives 3a
H
OH
+O
12
a
b
c
d
e
f
R1
But
But
But
But
But
Br
Br
I
R2
Oxidant
Yield (%)
R1
R1
R1
R1
Ac
Bz
NBS
Ana
NBS
An
NBS
An
NBS
An
NBS
An
NBS
An
NBS
An
NBS
An
PIFA
An
97
44
100
28
100
64
93
40
92
85
97
54
97
72
98
66
–e–
+e–
CH2
CH
CH2
CH
CO2H
Fmoc
For
Z
R2HN
CO2H
R2HN
3
9
+H+
–H+
Boc
Z
+
O
g
h
i
O
R1
R1
R1
R1
Z
+e–
–e–
H
Boc
Z
76b
18
j
H
PIFA
An
85c
17
CH2
CH
CH2
CH
R2HN
R2HN
CO2H
a Anodic oxidation. b Determined by NMR spectroscopy; after
chromatography 23%. c Determined by NMR spectroscopy; after
chromatography 38%.
CO2H
11
10a
ating side reactions made it desirable to test new oxidants for
the transformation of N-protected--tyrosines 3 (R1 = H) into
the corresponding dienone lactones 12. Among those oxidants,
phenyliodine() diacetate (PIDA) and phenyliodine() bis-
(trifluoroacetate) (PIFA) proved to be the most effective
ones.1,2,15,16 These high valent iodoso derivatives, in a small
excess under argon, indeed converted N-protected tyrosines
into the corresponding lactones.1 The yields of the isolated
ring-unsubstituted spirocyclic tyrosines were again very
modest. Analytical TLC showed that originally only one
product was formed and the NMR spectra of the crude product
indicated that up to 85% of the desired compound was present.
Chromatography on SiO2, however, was ineffective for purific-
ation, and only 23–38% of pure 12 could be isolated.
O
O
+
8
R1
R1
9
R1
R1
7
6
10
–H+
5
1
O
4
OH
2
3
R2HN
O
R2HN
O
12
10b
Scheme 2
Better yields could be achieved using graphite instead of
platinum as the working electrode.12
Spirolactones of N-protected 3,5-di-tert-butyltyrosine deriv-
atives 3‡ (R1 = But, R2 = Ac, Bz, Fmoc, For, Z) could thus be
synthesized anodically in acetonitrile in 28 to 85% yields (Table
1). In the case of ring-unsubstituted tyrosines 3‡ (R1 = H),
however, the yields of isolable spirolactones 12 were relatively
modest (17–18%).13,14 This is presumably due to the instability
of the species under the work-up conditions. Thus, the conver-
sion of Z-tyrosine in 0.1  NaClO4–MeCN at a graphite elec-
trode into the spirolactone was nearly quantitative (analytical
TLC) but after purification only 17% of the desired product
could be isolated.
Spirolactones as active esters
As has been shown above, the oxidation of N-protected tyrosine
derivatives leads to intramolecular cyclization. The resulting
spirolactone 12 may be regarded as a tyrosine in which the
C-terminus and the phenolic side chain are intramolecularly
protected, needing no external protective reagent. One could
think of cleaving the N-protecting group in 12 and using the
free amino group for peptide coupling. While trying to remove
the N-protecting group from 3-(N-Fmoc)-7,9-di-tert-butyl-1-
oxaspiro[4,5]deca-6,9-diene-2,8-dione 12c with 15% piperidine–
CH2Cl2, we obtained after a few minutes the piperidine amide
13 (Scheme 3), the result of a simultaneous cleavage of the
Fmoc group and ring opening of the spirocycle.
Up to now it was not possible to remove the N-protecting
groups in 12 without cleavage of the lactone ring. On the other
hand, the above mentioned reaction of 12c to give 13 led us to
the idea to react these spirocyclic lactones with amino acid
esters to form a peptide bond. In the resulting dipeptides,
the original tyrosine side chain would be present as a quinol.
This strategy was indeed found to be successful, and treating
equimolar amounts of 3-(N-formylamino)-7,9-di-tert-butyl-1-
oxaspiro[4,5]deca-6,9-diene-2,8-dione 12d with the amino acid
ester NH2-Ala-OMe in CH2Cl2, for example, gave the dipeptide
15a (R1 = But, R2 = For, R3 = Me, R4 = Me) after 60 h.
Chemical oxidation
Spirolactones 12 can also be synthesized using chemical oxid-
ants. These must be able to produce phenoxenium ions 10 in
a formal two-electron oxidation process. Thus, the choice of
the oxidant is very important, because its oxidizing strength
indirectly determines the yield of the spirolactones.
Schmir, Cohen and Witkop7 reported a study on the action
of bromine or N-bromosuccinimide (NBS) on derivatives of
tyrosine and simpler analogues. In contrast to the anodic oxid-
ation, the formation of a spirolactone with NBS was almost
quantitative within a few minutes, however, the phenolic side
chain was brominated too. On the other hand, the oxidation of
3,5-di-tert-butylated tyrosine derivatives with NBS did not
produce a dibrominated lactone but gave the desired 3,5-di-tert-
butylated species 12 in excellent yields (Table 1). The bromin-
The oxidative intramolecular esterification to form the lac-
tone 12 can therefore be regarded as an activation process of
the carboxy group. Apparently, nucleophilic attack of the
amino group of the ester at the carbonyl group of the lactone
takes place, finally leading to coupling of both amino acids.
‡ For the numbering see Scheme 2, for the type of substituents Table 1.
J. Chem. Soc., Perkin Trans. 1, 1998
2203

View Article Online
Table 2 Coupling conditions and yields of 15
O
Amino acid
ester
12
Compd.
15
Reaction
conditions
Yield
(%)
R1
R2
R3
R4
O
O
a
b
c
But For
But For
Me
H
Me
Et
Et
CH2Cl2, 96 h, 20 ЊC
CH2Cl2, 72 h, 20 ЊC
CHCl3, 80 h, 60 ЊC
cat. TsOH
CHCl3, 120 h, 60 ЊC
CHCl3, 80 h, 60 ЊC
cat. TsOH
CHCl3, 96 h, 60 ЊC
CHCl3, 80 h, 60 ЊC
cat. TsOH
CHCl3, 80 h, 60 ЊC
cat. TsOH
60
63
67
Fmoc-HN
12c
But
But
Z
Bui
15% Piperidine–CH2Cl2
c
d
Z
Bui
Bui
Et
Et
37
51
But Fmoc
O
d
e
But Fmoc
Bui
Bui
Et
Et
15
14
Br
H
Z
Z
CH2
O
Bui
Et
7
OH
O
+
f
H2N
C
N
O
N
catalyst. Basic catalysts, such as N-methyl morpholine, were not
effective.
As already mentioned, the reaction of the spirolactones with
amino acid esters leads to a dipeptide in which the phenolic part
of the tyrosine has a quinonoidal structure. This may hamper
the application, although such non-natural amino acids might
be of interest for screening tests. The rearomatization to the
normal tyrosine form should, however, also be achievable by
reducing the quinonoidal dipeptide.
13
Scheme 3
Presumably, an unstable amino hemiketal 14 is formed as an
intermediate which is converted to the stable dipeptide 15 with
simultaneous ring opening (Scheme 4).
We did not work out such procedures in detail. However, we
found that reduction of 15 to the tyrosyl dipeptides 16 can be
achieved in 52–75% yield by the acetonide of ascorbic acid in
CHCl3 using ZnCl2 as catalyst (Scheme 5). De-tert-butylation
O
R1
R1
H2N
+
CH CO2R4
R3
H2C
R2HN
O
OH
O
O
Vitamin C
ZnCl2
12
OH
CH2
CH
H2C
CH
O
O
R1 NH
C
O
NH CH CO2R3
R2
R1 NH
C
NH CH CO2R3
R2
R1
R1
R1
R1
O
15
16
H2C
CH
OH
H2C
R2HN
O
O–
NH2
Compound R1
R2
R3
Yield of 16 (%)
R2HN
C
NH CH CO2R4
R3
+
CH CO2R4
a
c
For
Z
Me 75
Et 52
O
Me
R3
CH2CH(Me)2
14
15
Scheme 4
Scheme 5
It was observed that sterically unhindered amino acid esters
react faster with the spirolactones 12 but, on the other hand,
they also polymerize faster. Therefore, an excess of the amino
acid ester is required for the reaction. Higher temperatures
reduce the reaction time but also increase the polymerization
rate of the ester. The reaction conditions and coupling yields
are collected in Table 2.
It was expected that proton donors would further activate the
spiroester carbonyl group and thus increase the yield of the
dipeptide. However, only a catalytic amount of acid might be
useful, in order not to block the amino group of the amino
ester. The polymeric acid Amberlyst 15, a macroporous sulfonic
acid resin on a polystyrene base, was first tested, because it
should provide non-hydrated protons. Unfortunately, it proved
to be unsuitable for this reaction: even after 100 h no dipeptide
could be detected. The amino acid ester, however, disappears
during the reaction without reacting with the spirolactone.
Presumably, it is adsorbed on the porous catalyst. Presently,
p-toluenesulfonic acid is the best catalyst for this reaction, and
yields of the dipeptides could be increased 2–3 times with this
of 16c with AlCl3 in toluene (3–22 h at 25 ЊC) or in nitro-
methane–benzene (4–5 h, 25–50 ЊC) led only to decomposition.
Structure determination
The structure of the synthesized compounds has been deter-
mined by 1H NMR, 13C NMR, IR and mass spectroscopy.
NMR Spectroscopy is a particularly good tool for character-
izing the present reaction products, which are quinolide com-
pounds on the one hand and amino acids or dipeptides on the
other. A series of compounds with these structural units has
been measured earlier,17,18,19 and serve as basic molecules for
chemical shift considerations with the help of increment sys-
tems. Thus, for instance, it was possible to assign the fragments
-Ala-OMe, -Gly-OEt and -Leu-OEt in the spectra. Because of
the presence of the asymmetric α-carbon atom of the amino
acid moiety, the signals of equivalent nuclei may split due to
magnetic non-equivalence (hereafter abbreviated as MNE).
1
Thus, the H and 13C NMR signals of the tert-butyl groups of
the quinolide ring, as well as the 13C signals of the olefinic
2204
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
carbons of the rings may be doubled. The protons at the posi-
tions meta to the carbonyl group generally cause an AB split-
ting pattern, because, in addition to MNE, they couple with
4J ≈ 2–3 Hz. The same is true for the geminal protons of the
CH2 moiety in the Z group or in the tyrosine side chain
(1J ≈ 12–14 Hz).
The synthesis of the tert-butyl-substituted tyrosine deriv-
atives according to Scheme 1 leads to compounds with
,-configuration. As a consequence, the corresponding spiro-
lactones obtained on electrochemical or chemical oxidation are
also in the ,-configuration. Since the reaction between the
spirolactones and -amino acid esters does not lead to racemiz-
ation of the latter, two diastereomeric quinolide dipeptides are
present after the reaction causing a doubling or broadening
of most of the NMR signals. The interpretation of the 1H
NMR spectra is therefore difficult because many signals overlap
and/or coincide. Thus, in the Experimental section, only the
positions of the multiplets are presented.
acetonitrile and 153.6 ml of acetate buffer, pH = 4.6) a solution
of NBS (800 mg, 4.5 mmol) in 54 ml of 20% acetonitrile–
acetate buffer was added. A crystalline substance separated
within a few minutes. The reaction mixture was chilled for 2 h
and the product was collected, washed with n-hexane and dried.
If necessary, the product was recrystallized from acetonitrile–
water or n-hexane–EtOAc. For exceptions see the detailed
procedures.
Electrochemical oxidation. General procedure 3
The preparative anodic oxidation was performed in an
undivided cell, i.e. a cylindrical glass vessel with a lateral tube
to take up the reference electrode. The anode was made of
graphite and the cathode was a cylindrical platinum-net (90%
Pt and 10% Ir). For the potential-controlled electrolysis the
reference electrode was Ag/Agϩ (Ag/0.01  Agϩ in acetonitrile)
with a potential of 0.35 V vs. SCE.21 A 0.1  NaClO4 solution
in acetonitrile was used as the sole supporting electrolyte (SE),
and the produced protons were neutralized by an excess of 2,6-
lutidine (2,6-dimethylpyridine).
The oxidation potential was set to 1300–1400 mV to oxidize
the phenolic OH-group, and the concentration of the substrate
in the solution amounted to about 0.01 mol lϪ1. At the end of
the reaction, the mixture was evaporated to one half its original
volume and diluted with 400 ml of water. The product was
extracted with diethyl ether (3 × 100 ml) and washed with
water. The organic layer was dried over MgSO4, and evaporated
to yield the crude products which were recrystallized or purified
by silica gel chromatography.
Conclusions
Tyrosine derivatives can be oxidized chemically or electro-
chemically in a two-electron oxidation via a free or incipient
phenoxenium ion to form spirolactones. These compounds are
not only precursors for the synthesis of some antibiotics, such
as aranorosin or anticapsin, but also synthons useful in peptide
synthesis. They react as active esters with amines and amino
acid esters to give quinonoidal amides or dipeptides that can be
reduced to tyrosyl amides or tyrosyl dipeptides. This procedure
constitutes a new method to introduce a tyrosine, preferentially
with a non-natural side chain, at the N-terminus of amino acids
or oligopeptides. The tert-butyl substituted derivatives can then
be used as amino acid spin labels to draw conclusions on the
configuration and conformation as well as the tertiary structure
of the labelled compound, e.g. a peptide.20
Quinoidal tyrosyldipeptides 15 from spirolactones 12. General
procedure 4
To a stirred solution of the spirolactone 12 (0.69 mmol) in
CH2Cl2 or CHCl3 (30 ml) the amino acid ester (1.16 mmol) was
added. The mixture was stirred for several hours under reflux
(see Table 2). The solvent was removed, the residue washed with
n-hexane and dried. The product was purified by column
chromatography on silica gel with CH2Cl2–methanol to yield
the desired quinolide dipeptides. Sometimes, catalytic amounts
(5–10 mg) of p-toluenesulfonic acid were used to catalyse the
reaction. For exceptions see the detailed procedures.
Experimental
Melting points were determined on a Mettler FP61 melting
point apparatus and are uncorrected. IR Spectra were recorded
on a Perkin-Elmer IR-281 (KBr disks). Low resolution mass
spectra were obtained on a Varian Matt 711 (200 ЊC, 70 eV or
50 ЊC/FD). NMR Spectra were recorded on a Bruker AC 250
(250 MHz for 1H and 62.9 MHz for 13C) spectrometer in CDCl3
unless otherwise noted. Chemical shifts (δ) are given in ppm
and the coupling constants (J) in Hz. Anhydrous solvents were
freshly distilled from either P2O5 or magnesium. For analytical
TLC, Merck silica gel 60 F-254 plates were used. Cyclovoltam-
metric experiments were conducted on a Bruker polarograph E
350 with a Wenking voltage scanner and a MVS 87 Houston
Instrument Omnigraphic 2000 xy-printer in MeCN vs. Ag/
10Ϫ2  Agϩ/MeCN with (Et)4NBF4 or NaClO4 as electrolytes.
Although no problems were encountered, it should be noted that
perchlorates in general may be explosive and should be handled
with due care.
3-Acetylamino-7,9-tert-butyl-1-oxaspiro[4,5]deca-6,9-diene-2,8-
dione 12a
General procedure 2: the reaction of Ac-Tyr(3,5-But2)-OH 3a
(510 mg, 1.50 mmol) and NBS (800 mg, 4.5 mmol) gave the
product 12a (485 mg, 97%) as a white solid, mp 179 ЊC (Found:
C, 68.5; H, 8.3; N, 4.2. C19H27NO4 requires C, 68.4; H, 8.2; N,
4.2%); Rf 0.23 (n-hexane–EtOAc, 1:1); δH(250 MHz; CDCl3)
1.20 (18H, s, But), 2.06 (3H, s, Ac), 2.34 (1H, dd, J ≈ 12, J ≈ 13,
CCH2CH), 2.68 (1H, dd, J ≈ 9, J ≈ 13, CCH2CH), 4.85 (1H,
ddd, J ≈ 6.5, J ≈ 9, J ≈ 12, CH2CHNHR), 6.46 (1H, d, J ≈ 6.5,
NH), 6.50 (1H, d, J ≈ 3, Hquin), 6.52 (1H, d, J ≈ 3, Hquin);
δC(62.9 MHz; CDCl3) 22.84, 29.25; 34.95, 35.04 (MNE); 40.09,
49.70, 78.25; 135.23, 137.61 (MNE); 148.09, 148.63 (MNE);
170.60, 174.33, 185.32; m/z (EI) 333 (Mϩ).
Oxidation with PIFA. General procedure 1
General procedure 3: the reaction of Ac-Tyr(3,5-But2)-OH 3a
(2.0 g, 6.0 mmol) in 200 ml SE at a potential of Eox = 1300 mV
yielded 12a (875 mg, 44%) after 10 h.
To a stirred solution of 1.80–2.20 mmol of N-protected tyrosine
3 in 30 ml of absolute acetonitrile at Ϫ5 ЊC a few drops of
pyridine and 2.0–2.45 mmol of phenyliodoso bis(trifluoro-
acetate) (PIFA) were added under argon. The reaction mixture
was stirred for 20 min at Ϫ5 ЊC. Afterwards, the mixture was
diluted with 20 ml of water and extracted with CH2Cl2. The
combined organic layers were dried over MgSO4 and concen-
trated to give a greenish oil which was chromatographed on
SiO2 (50% EtOAc–n-hexane) to give the spirocyclic product as
a white solid.
3-Benzoylamino-7,9-di-tert-butyl-1-oxaspiro[4,5]deca-6,9-diene-
2,8-dione 12b
General procedure 2: the reaction of Bz-Tyr(3,5-But2)-OH 3b
(200 mg, 0.5 mmol) and NBS (237 mg, 1.33 mmol) gave the
product 12b (197 mg, 100%) as a white solid, mp 215–245 ЊC
(decomp.) (Found: C, 72.69; H, 7.52; N, 3.46. C24H29NO4
requires C, 72.89; H, 7.39; N, 3.54%); Rf 0.46 (n-hexane–
CH2Cl2, 1:1); δH(250 MHz; CDCl3) 1.22 (9H, s, But), 1.24 (9H,
s, But), 2.48 (1H, t, J ≈ 13, CCH2CH), 2.85 (1H, dd, J ≈ 9,
J ≈ 13, CCH2CH), 5.00–5.12 (1H, m, CH2CHNHR), 6.55 (1H,
Oxidation with NBS. General procedure 2
To a solution of N-protected tyrosine derivatives 3 (1.50 mmol)
in 192 ml of 20% acetonitrile–acetate buffer (38.4 ml of
J. Chem. Soc., Perkin Trans. 1, 1998
2205

View Article Online
d, J ≈ 3, Hquin), 6.58 (1H, d, J ≈ 3, Hquin), 6.97 (1H, d, J ≈ 8,
NH), 7.39–7.93 (5H, m, Bz); δC(62.9 MHz; CDCl3) 29.2; 35.1,
35.2 (MNE); 40.5, 50.1, 78.4, 127.1, 128.7, 132.4, 132.7; 135.1,
137.6 (MNE); 146.7, 148.1 (MNE); 167.6, 174.4, 185.3; m/z (EI)
395 (Mϩ).
J ≈ 10, J ≈ 13, CCH2CH), 4.62 (1H, t, J ≈ 10, CH2CHNHR),
7.59 (1H, d, J ≈ 3, Hquin), 7.76 (1H, d, J ≈ 3, Hquin); δC(62.9
MHz; CDCl3) 28.36, 36.50, 50.08, 79.37, 80.44; 123.13, 123.98
(MNE); 148.13, 148.95 (MNE); 156.13, 173.51, 183.99; m/z
(FD) 437 (Mϩ).
General procedure 2: the reaction of Boc-Tyr(3,5-Br2)-OH 3f
(100 mg, 0.22 mmol) and NBS (42.7 mg, 0.27 mmol) yielded 12f
(93.3 mg, 97%), mp 187–189 ЊC.
General procedure 3: the reaction of Bz-Tyr(3,5-But2)-OH 3b
(600 mg, 1.51 mmol) in 200 ml SE at a potential of Eox = 1400
mV gave 12b (166 mg, 28%) after 5 h.
General procedure 3: the reaction of Boc-Tyr(3,5-Br2)-OH 3f
(600 mg, 1.37 mmol) in 150 ml SE at a potential of Eox = 1300
mV yielded 12f (323 mg, 54%) after 6 h.
3-(Fluoren-9-ylmethoxycarbonylamino)-7,9-di-tert-butyl-1-oxa-
spiro[4,5]deca-6,9-diene-2,8-dione 12c
General procedure 2: the reaction of Fmoc-Tyr(3,5-But2)-OH
3c (3.0 g, 5.8 mmol) in buffer (745 ml) and NBS (2.70 g, 15.2
mmol) in buffer (210 ml) yielded white crystals (2.98 g, 100%),
mp 115–117 ЊC; δC(250 MHz, CD3COCD3) 1.24 (18H, s, But),
2.62 (2H, dd, J ≈ 10, J ≈ 3, CCH2CH), 4.25 [1H, t, J ≈ 7,
CH2CH(C6H4)2], 4.40 (2H, d, J ≈ 7, OCH2CH), 4.86–4.95 (1H,
m, CH2CHNHR), 6.70 (1H, d, J ≈ 3, Hquin), 7.00 (1H, d, J ≈ 3,
Hquin), 7.19 (1H, d, J ≈ 8, NH), 7.32–7.38 (4H, m, Harom-Fmoc),
3-Benzyloxycarbonylamino-7,9-dibromo-1-oxaspiro[4,5]deca-
6,9-diene-2,8-dione 12g7
General procedure 2: the reaction of Z-Tyr-OH 3j (5.0 g, 15.8
mmol) in buffer (2.0 l) and NBS (8.4 g, 47.2 mmol) in buffer
(580 ml) yielded 12g (7.44 g, 97%) as a white solid, mp 215 ЊC
(Found: C, 43.41; H, 2.72; N, 2.63; Br, 33.98. C17H13Br2NO5
requires C, 43.34; H, 2.78; N, 2.97; Br, 33.92%); Rf 0.64
(CH2Cl2–EtOAc, 10:3); δH(250 MHz; CD3COCD3) 2.76 (1H,
dd, J ≈ 11, J ≈ 14, CCH2CH), 2.99 (1H, dd, J ≈ 11, J ≈ 14,
CCH2CH), 4.90 (1H, q, J ≈ 10.5, CH2CHNHR), 5.13 (2H, s,
C6H5CH2O), 7.33–7.37 (5H, m, Harom-Z), 7.65 (1H, d, J ≈ 3,
Hquin), 7.95 (1H, d, J ≈ 3, Hquin); δC(62.9 MHz; CD3COCD3)
36.4, 50.5, 67.3, 79.5; 122.5, 123.3 (MNE); 128.6, 128.7, 129.2,
137.5; 147.9, 148.7 (MNE); 156.7, 171.9, 173.3; m/z (FD) 473
(Mϩ ϩ 2H).
7.69 (2H, d, J ≈ 7, Harom-Fmoc), 7.86 (2H, d, J ≈ 7, Harom
-
Fmoc); δC(62.9 MHz, CD3COCD3) 29.5; 35.3, 35.5 (MNE);
38.6, 47.9, 51.2, 67.3, 77.3, 120.7, 125.9, 127.8, 128.5; 138.9,
140.2 (MNE); 142.0, 144.8; 147.2, 147.9 (MNE); 156.7, 174.3,
186.3; m/z (EI) 513 (Mϩ).
General procedure 3: the reaction of Fmoc-Tyr(3,5-But2)-OH
3c (600 mg, 1.17 mmol) in 100 ml SE at a potential of Eox = 1400
mV gave after 4 h the desired product 12c (382 mg, 64%).
General procedure 3: the reaction of Z-Tyr(3,5-Br2)-OH 3g
(700 mg, 1.48 mmol) in 150 ml SE at a potential of Eox = 1400
mV gave 12g (504 mg, 72%) after 6 h.
3-Formylamino-7,9-di-tert-butyl-1-oxaspiro[4,5]deca-6,9-diene-
2,8-dione 12d
General procedure 2: the reaction of For-Tyr(3,5-But2)-OH 3d
(500 mg, 1.56 mmol) and NBS (833 mg, 4.68 mmol) gave 12d
(460 mg, 93%), mp 121–123 ЊC (Found: C, 67.62; H, 7.45; N,
4.32. C18H25NO4 requires C, 67.69; H, 7.89; N, 4.39%); Rf 0.34
(CH2Cl2–MeOH, 20:1); δH(250 MHz; CDCl3) 1.22 (18H, s,
But), 2.36 (1H, dd, J ≈ 12, J ≈ 13, CCH2CH), 2.78 (1H, dd,
J ≈ 9, J ≈ 13, CCH2CH), 4.87 (1H, ddd, J ≈ 6, J ≈ 9, J ≈ 13,
CH2CHNHR), 6.33 (1H, d, J ≈ 6, NH), 6.49 (1H, d, J ≈ 3,
Hquin), 6.51 (1H, d, J ≈ 3, Hquin), 8.29 (1H, s, CHO); δC(62.9
MHz; CDCl3) 29.26; 35.38, 35.48 (MNE); 39.56, 48.69, 78.75;
134.96, 137.63 (MNE); 148.41, 148.99 (MNE); 162.05, 174.65,
186.32; m/z (EI) 319 (Mϩ).
3-Benzyloxycarbonylamino-7,9-diiodo-1-oxaspiro[4,5]deca-6,9-
diene-2,8-dione 12h
General procedure 2: Z-Tyr(3,5-I2)-OH 3h (1.0 g, 1.76 mmol)
and NBS (1.0 g, 5.62 mmol) yielded white solid 12h (975 mg,
98%), mp 155–157 ЊC; δH(250 MHz; CDCl3) 2.72 (1H, dd,
J ≈ 10, J ≈ 13.5, CCH2CH), 2.97 (1H, dd, J ≈ 10, J ≈ 13.5,
CCH2CH), 4.84–4.95 (1H, m, CH2CHNHR), 5.13 (2H, s, CH2-
Z), 7.37 (5H, br s, Harom-Z), 7.95 (1H, d, J ≈ 3, Hquin), 8.24 (1H,
d, J ≈ 3, Hquin); m/z (FD) 565 (Mϩ).
General procedure 3: the reaction of Z-Tyr(3,5-I2)-OH 3h
(300 mg, 0.53 mmol) in 175 ml SE at a potential of Eox = 1400
mV gave 12h (197 mg, 66%) after 3 h. Compound 3h was syn-
thesized in 68% yield by mixing a solution of iodine and sodium
iodide (in water) with a solution of Z-Tyr-OH in 20% aqueous
ethylamine.22
General procedure 3: the reaction of For-Tyr(3,5-But2)-OH
3d (1.0 g, 3.1 mmol) in 160 ml SE at a potential of Eox = 1400
mV gave 12d (400 mg, 40%) after 6 h.
3-Benzyloxycarbonylamino-7,9-di-tert-butyl-1-oxaspiro[4,5]-
deca-6,9-diene-2,8-dione 12e
3-tert-Butoxycarbonylamino-1-oxaspiro[4,5]deca-6,9-diene-2,8-
dione 12i4
General procedure 2: the reaction of Z-Tyr(3,5-But2)-OH 3e
(1.0 g, 2.33 mmol) and NBS (1.248 g, 7.01 mmol) yielded 12e
(910 mg, 92%), mp 177–179 ЊC (Found: C, 70.19; H, 7.55; N,
3.47. C25H31NO5 requires C, 70.56; H, 7.34; N, 3.29%); Rf 0.46
(n-hexane–EtOAc, 2:1); δH(250 MHz; CDCl3) 1.14 (18H, s,
But), 2.32 (1H, t, J ≈ 13, CCH2CH), 2.56 (1H, dd, J ≈ 9, J ≈ 13,
CCH2CH), 4.63 (1H, m, J ≈ 9, J ≈ 13, CH2CHNHR), 5.12 (2H,
d, J ≈ 12, C6H5CH2O), 5.55 (1H, d, J ≈ 7, NH), 6.45 (2H, s,
Hquin), 7.26 (5H, s, Hquin-Z); δC(62.9 MHz; CDCl3) 29.26;
34.95, 35.03 (MNE); 40.14, 50.98, 67.58, 77.87, 128.20, 128.42,
128.62, 135.78; 135.28, 137.69 (MNE); 148.11, 148.62 (MNE);
155.99, 173.76, 185.38; m/z (EI) 426 (Mϩ ϩ H).
General procedure 1: the reaction of Boc-Tyr-OH 3i (618 mg,
2.19 mmol) and PIFA (1.05 g, 2.45 mmol) yielded 12i (140 mg,
23%) [76% as the crude product (by NMR spectroscopy)], mp
184–185 ЊC (Found: C, 59.89; H, 5.91; N, 5.14. C14H17NO5
requires C, 60.20; H, 6.14; N, 5.02%); Rf 0.51 (n-hexane–EtOAc,
1:1); δH(250 MHz; CDCl3) 1.39 (9H, s, But), 2.43 (1H,
dd, J ≈ 12, J ≈ 13, CCH2CH), 2.65 (1H, dd, J ≈ 9, J ≈ 13,
CCH2CH), 4.48–4.58 (1H, m, CH2CHNHR), 5.36 (1H, d,
J ≈ 7, NH), 6.19–6.26 (2H, m, MNE, Hquin), 6.83 (2H, d, J ≈ 9,
Hquin); δC(62.9 MHz; CDCl3) 28.25, 38.68, 50.24, 76.08, 81.20;
129.12, 129.26 (MNE); 144.23, 146.18 (MNE); 155.98, 173.51,
183.99; m/z (FD) 280 (Mϩ ϩ H).
General procedure 3: the reaction of Z-Tyr(3,5-But2)-OH 3e
(900 mg, 2.1 mmol) in 200 ml SE at a potential of Eox = 1400
mV gave 12e (761 mg, 85%) after 6 h.
General procedure 3: the reaction of Boc-Tyr-OH 3i (600
mg, 2.13 mmol) in 250 ml SE at a potential of Eox = 1400 mV
yielded 12i (107 mg, 18%) after 4 h.
3-tert-Butoxycarbonylamino-7,9-dibromo-1-oxaspiro[4,5]deca-
6,9-diene-2,8-dione 12f
General procedure 2: the reaction of Boc-Tyr-OH 3i (1.0 g, 3.55
mmol) and NBS (1.896 g, 10.65 mmol) gave a white solid 12f
(1.38 g, 89%), mp 188–189 ЊC; δH(250 MHz; CDCl3) 1.46 (9H, s,
But), 2.53 (1H, dd, J ≈ 10, J ≈ 13, CCH2CH), 2.72 (1H, dd,
3-Benzyloxycarbonylamino-1-oxaspiro[4,5]deca-6,9-diene-2,8-
dione 12j1,2,4
General procedure 1: the reaction of Z-Tyr-OH 3j (567 mg, 1.80
mmol) and PIFA (857 mg, 2.0 mmol) yielded white solid 12j
(216 mg, 38%) [85% as the crude product (by NMR spec-
2206
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
troscopy)], mp 46 ЊC (Found: C, 65.28; H, 4.89; N, 4.33.
C17H15NO5 requires C, 65.17; H, 4.82; N, 4.47%); Rf 0.45
(n-hexane–EtOAc, 1:1); δH(250 MHz; CDCl3) 2.38 (1H, t,
J ≈ 12, CCH2CH), 2.56 (1H, dd, J ≈ 9, J ≈ 13, CCH2CH), 4.51–
4.61 (1H, m, CH2CHNHR), 5.03 (2H, br s, C6H5CH2O), 5.79
(1H, d, J ≈ 7, NH), 6.14–6.18 (2H, m, Hquin, MNE), 6.69–6.78
(2H, m, Hquin, MNE), 7.25 (5H, s, Harom); δC(67.8 MHz; CDCl3)
38.20, 50.50, 67.65, 76.16, 128.22, 128.50, 128.67, 129.13,
129.73, 135.69; 144.21, 146.11 (MNE); 155.99, 173.44, 184.06;
m/z (FD) 313 (Mϩ).
J ≈ 4, CH2CH(CH3)2]; 1.11 (9H, s, But), 1.12 (9H, s, But),
(MNE); 1.15–1.21 (3H, m, CO2CH2CH3), 1.45–1.62 [3H, m,
CHCH2CH(CH3)2, CH2CH(CH3)2], 1.76 (1H, t, J ≈ 8, J ≈ 14,
CCH2CH), 2.25 (1H, ddd, J ≈ 3, J ≈ 4, J ≈ 14, CCH2CH), 3.52–
3.69 (1H, m, NHCHCO2Et), 4.06–4.12 (2H, m, CO2CH2CH3),
4.41–4.50 (1H, m, CH2CHNHR), 4.47 (1H, s, OH), 5.02–5.04
(2H, m, C6H5CH2O), 5.61 [1H, d, J ≈ 8, NH(Tyr)], 5.72–5.82
[1H, m, NH(Tyr)];§ 6.46 (1H, d, J ≈ 3, Hquin), 6.58 (1H, d, J ≈ 3,
Hquin) MNE; 6.92 [1H, d, J ≈ 9, NH(Leu)], 7.24–7.26 (5H, m,
Harom Z); δC(62.9 MHz; CD3OD) 14.10; 21.91, 22.74 (MNE);
24.87; 29.32, 29.35 (MNE); 34.64, 34.71 (MNE); 41.23, 44.34,
51.07, 61.45, 67.37, 68.55, 72.28, 128.16, 128.29, 128.54; 135.87,
140.99 (MNE); 141.76; 146.13, 146.51 (MNE); 155.92, 171.70,
172.79, 186.14; m/z (FD) 584 (Mϩ).
General procedure 3: the reaction of Z-Tyr-OH 3j (700 mg,
2.22 mmol) in 250 ml SE at a potential of Eox = 1400 mV gave
12j (116 mg, 17%) after 3 h.
4-(2-Amino-3-oxo-3-piperidinopropyl)-2,6-di-tert-butyl-4-
hydroxycyclohexa-2,5-dienone 13
N-[2-(Fluoren-9-ylmethoxycarbonylamino)-3-(3,5-di-tert-butyl-
1-hydroxy-4-oxocyclohexa-2,5-dien-1-yl)propionyl] leucine ethyl
ester 15d
The spirolactone 12c (810 mg, 1.58 mmol) was dissolved in 30
ml of 15% piperidine–CH2Cl2. After 5 min the reaction was
completed and the solvent removed. The crude product was
chromatographed (silica gel, CH2Cl2–methanol, 20:1) to give
13 (487 mg, 1.29 mmol); νmax/cmϪ1 3500–3300, 3060, 2950,
1700, 1635, 1450, 1360; δH(250 MHz; CDCl3) 1.17 (9H, s, But),
1.20 (9H, s, But) MNE; 1.43–1.50 (8H, m, piperid.), 1.83–1.90
(1H, m, CCH2CH), 2.10–2.17 (1H, m, CCH2CH), 4.04 (1H, d,
J ≈ 10, CH2CHNH2), 5.27 (1H, s, OH), 6.44–6.51 (1H, d, J ≈ 3,
Hquin), 6.84–6.92 (1H, d, J ≈ 3, Hquin); δC(62.9 MHz; CDCl3)
24.3, 26.1; 29.4, 29.5 (MNE); 34.5, 34.7 (MNE); 44.5, 45.9,
49.8, 69.6; 142.2, 144.1 (MNE); 145.5, 146.1 (MNE); 171.1,
186.7.
General procedure 4: 12c (100.6 mg, 0.2 mmol) and H-Leu-OEt
(50 mg, 0.31 mmol) in 20 ml of CHCl3 in the presence of 3.5 mg
(0.2 mmol) p-toluenesulfonic acid as catalyst at 60 ЊC yielded
colourless oily 15d (68 mg, 51%) after 80 h; Rf 0.23 (CH2Cl2–
MeOH, 50:1); δH(250 MHz; CD3OD) 0.91–0.95 [6H, m,
CH2CH(CH3)2]; 1.19 (9H, s, But), 1.20 (9H, s, But) (MNE);
1.24–1.32 (3H, m, CO2CH2CH3); 1.55–1.79 [3H, m, CHCH2-
CH(CH3)2, CH2CH(CH3)2], 1.73–1.90 (1H, m, CCH2CH),
2.25–2.36 (1H, m, CCH2CH); 3.54–3.55 (1H, m, NHCH-
CO2Et), 4.11–4.20 [3H, m, CO2CH2CH3, (C6H4)2CHCH2],
4.36–4.45 (2H, m, C6H5CH2O), 4.46–4.59 (1H, m, CH2CH-
NHR), 5.67 [1H, d, J ≈ 9, NH(Tyr)]; 6.53 (1H, d, J ≈ 3, Hquin),
6.63–6.74 (1H, m, Hquin) MNE, 6.78–6.87 [1H, m, NH(Leu)],
N-[2-Formylamino-3-(3,5-di-tert-butyl-1-hydroxy-4-oxocyclo-
hexa-2,5-dien-1-yl)propionyl] alanine methyl ester 15a
7.25–7.41 (4H, m, Harom-Fmoc), 7.57 (2H, d, J ≈ 5, Harom
-
Fmoc), 7.75 (2H, d, J ≈ 7, Harom-Fmoc); m/z (FD) 672 (Mϩ).
General procedure 4: 12d (220 mg, 0.69 mmol) and H-Ala-OMe
(120 mg, 1.16 mmol) in 30 ml of CH2Cl2 gave 15a (176 mg,
60%) as a white solid at 20 ЊC after 96 h. No catalyst was used.
Mp 145 ЊC (Found: C, 62.48; H, 8.29; N, 6.74. C22H34N2O6
requires C, 62.54; H, 8.11; N, 6.63%); Rf 0.37 (CH2Cl2–MeOH,
15:1); δH(250 MHz; CD3OD) 1.21 (9H, s, But), 1.22 (9H, s, But)
MNE; 1.30 (3H, d, J ≈ 7, MeOCOCHCH3), 1.90 (1H, dd, J ≈ 8,
J ≈ 14, CCH2CH), 2.23–2.37 (1H, m, CCH2CH), 3.70 (3H, s,
CO2CH3), 4.37 (1H, q, J ≈ 7, CH2CHNHR), 4.51–4.64 (1H, m,
NHCHCO2Me), 6.53–6.61 (2H, m, Hquin, MNE), 8.10 (1H, s,
CHO); δC(62.9 MHz; CD3OD) 14.0, 27.5, 33.2, 39.8, 42.0, 48.2,
50.2, 66.7; 141.1, 141.7 (MNE); 144.5, 144.9 (MNE); 170.7,
172.0, 175.0, 185.2; m/z (FD) 422 (Mϩ).
N-[2-Benzyloxycarbonylamino-3-(3,5-dibromo-1-hydroxy-4-
oxocyclohexa-2,5-dien-1-yl)propionyl] leucine ethyl ester 15e
General procedure 4: 12g (100 mg, 0.212 mmol) and H-Leu-
OEt (40.5 mg, 0.255 mmol) in 8 ml of CHCl3 in the presence
of 3.65 mg (0.212 mmol) p-toluenesulfonic acid as catalyst at
60 ЊC yielded 15e (19 mg, 14%) after 68 h; Rf 0.49 (CH2Cl2–
MeOH, 20:1); δH(250 MHz; CD3OD) 0.78–1.00 [6H, m,
CH2CH(CH3)2], 1.21–1.28 (3H, m, CO2CH2CH3), 1.47–1.57
[2H, m, CHCH2CH(CH3)2], 1.58–1.70 [1H, m, CH2CH(CH3)2],
1.72–2.26 (2H, m, CCH2CH), 4.09–4.18 (2H, m, CO2CH2CH3),
4.39–4.51 (1H, m, NHCHCO2Et), 4.64–4.74 (1H, m, CH2CH-
NHR), 5.11–5.15 (2H, m, C6H5CH2O), 7.32–7.34 (5H, m,
Harom-Z), 7.36–7.39 (2H, m, Hquin); m/z (FD) 630 (Mϩ).
N-[2-Formylamino-3-(3,5-di-tert-butyl-1-hydroxy-4-oxocyclo-
hexa-2,5-dien-1-yl)propionyl] glycine ethyl ester 15b
General procedure 4: 12d (50 mg, 0.15 mmol) and H-Gly-OEt
(20 mg, 0.19 mmol) in 30 ml of CH2Cl2 gave 15b (40 mg, 63%)
at 20 ЊC after 72 h as a colourless oil. No catalyst was used.
(Found: C, 62.48; H, 8.29; N, 6.74. C22H34N2O6 requires C,
62.54; H, 8.11; N, 6.63%); Rf 0.37 (CH2Cl2–MeOH, 15:1);
δH(250 MHz; CDCl3) 0.93 (3H, t, J ≈ 4, CO2CH2CH3); 1.15
(9H, s, But), 1.16 (9H, s, But) MNE; 1.82–1.95 (1H, m,
CCH2CH), 2.31–2.41 (1H, m, CCH2CH), 3.90 (2H, d, J ≈ 3,
NHCH2CO2), 3.99 (1H, s, OH), 4.11–4.28 [4H, m, NH(Tyr),
NH(Gly), CO2CH2CH3], 4.50–4.70 (1H, m, CH2CHNHR),
6.56 (2H, s, Hquin), 8.03 (1H, s, CHO); δC(62.9 MHz; CD3OD)
14.4, 29.8; 34.6, 34.7 (MNE); 41.4, 44.6, 48.6, 61.4, 68.7; 143.6,
144.3 (MNE); 145.7, 145.8 (MNE); 161.5, 169.6, 170.2, 183.6;
m/z (FD) 422 (Mϩ).
N-[2-Benzyloxycarbonyl-3-(1-hydroxy-4-oxocyclohexa-2,5-dien-
1-yl)propionyl] leucine ethyl ester 15f
General procedure 4: 12j (100 mg, 0.319 mmol) and -Leu-OEt
(66 mg, 0.415 mmol) in 50 ml of CHCl3 in the presence of 5.5
mg (0.319 mmol) p-toluenesulfonic acid as catalyst at 60 ЊC
gave 15f (11 mg, 7%) after 80 h; Rf 0.39 (CH2Cl2–MeOH, 30:1);
δH(250 MHz; CDCl3) 0.87–0.97 [6H, m, CH2CH(CH3)2], 1.20–
1.31 (3H, m, CO2CH2CH3), 1.45–1.72 [3H, m, CHCH2-
CH(CH3)2, CH2CH(CH3)2], 1.90–2.08 (1H, m, CCH2CH),
2.37–2.43 (1H, m, CCH2CH), 3.85 (1H, s, OH), 4.09–4.22 (2H,
m, CO2CH2CH3), 4.48–4.52 (1H, m, NHCHCO2Et), 4.63–4.74
(1H, m, CH2CHNHR), 5.10 (2H, s, C6H5CH2O), 6.61–6.62
(2H, m, Hquin), 7.04–7.05 (2H, m, Hquin), 7.35 (5H, s, Harom-Z);
m/z (FD) 454 (Mϩ Ϫ 2 Ϫ OH).
N-[2-Benzyloxycarbonylamino-3-(3,5-di-tert-butyl-1-hydroxy-4-
oxocyclohexa-2,5-dien-1-yl)propionyl] leucine ethyl ester 15c
General procedure 4: 12e (300 mg, 0.704 mmol) and H-Leu-
OEt (156 mg, 0.980 mmol) in 50 ml of CHCl3 in the presence of
12 mg (0.704 mmol) p-toluenesulfonic acid as catalyst at 60 ЊC
yielded 15c (275 mg, 67%) after 80 h as a colourless oil; Rf 0.31
(CH2Cl2–MeOH, 50:1); δH(250 MHz; CD3OD) 0.84 [6H, d,
Methyl 2-[3-(3,5-di-tert-butyl-4-hydroxyphenyl)-2-(formyl-
amino)propionylamino]propanoate 16a by reduction of 15a
Dipeptide 15a (20.0 mg, 0.047 mmol), (ϩ)-5,6-O-isopropyl-
§ Doubling of the NMR signals due to the presence of two
diastereomers.
J. Chem. Soc., Perkin Trans. 1, 1998
2207

View Article Online
idene--ascorbic acid (21.0 mg, 0.095 mmol), CHCl3 (15 ml)
and anhydrous ZnCl2 (ca. 5 mg) were refluxed for ca. 30 h.
Separation by TLC (silica gel, CH2Cl2–MeOH, 20:1) yielded
16a (14 mg, 75%), identical to an authentic sample prepared by
classical coupling of 3d and Ala-OMeؒHCl with dicyclohexyl-
carbodiimide in the presence of N-methylmorpholine and
hydroxybenzotriazole, mp 93–95 ЊC; Rf 0.52; νmax/cmϪ1 3630,
3300, 3260, 3060, 2960, 2910, 2870, 1740, 1690, 1665, 1540,
1500, 1455, 1380, 1365; m/z (FD) 406 (Mϩ). The by-product
was the methyl ether of 15a {N-[2-formylamino-3-(3,5-di-tert-
butyl-1-methoxy-4-oxocyclohexa-2,5-dien-1-yl)propionyl] alan-
ine methyl ester} (3 mg, 16%); Rf 0.46; m/z (FD) 435 (Mϩ Ϫ 1).
References
1 P. Wipf, Y. Kim and P. C. Fritch, J. Org. Chem., 1993, 58, 7195.
2 P. Wipf and Y. Kim, J. Org. Chem., 1993, 58, 1649.
3 A. V. Rama Rao, M. K. Gurjar and P. A. Sharma, Tetrahedron Lett.,
1991, 32, 6613.
4 A. McKillop, L. McLaren, R. J. K. Taylor, R. J. Watson and
N. Lewis, Synlett, 1992, 201.
5 H. Hara, T. Inoue, H. Nakamura, M. Endoh and O. Hoshino,
Tetrahedron Lett., 1992, 33, 6491.
6 A. Rieker, R. Beißwenger and K. Regier, Tetrahedron, 1991, 47, 645;
A. Rieker and B. Speiser, J. Electroanal. Chem., 1979, 102, 373.
7 G. L. Schmir, L. A. Cohen and B. Witkop, J. Am. Chem. Soc., 1959,
81, 2229.
8 H. J. Teuber, H. Krause and V. Berariu, Liebigs Ann. Chem., 1978,
757.
9 L. A. Cohen and W. M. Jones, J. Am. Chem. Soc., 1962, 84,
Ethyl 2-[2-benzyloxycarbonylamino-3-(3,5-di-tert-butyl-4-
hydroxyphenyl)propionylamino]-4-methylpentanoate 16c by
reduction of 15c
1629.
10 A. Ziogas, PhD Thesis, University of Tübingen, 1993.
11 B. Speiser and A. Rieker, J. Chem. Res. (M), 1977, 3601.
12 E. L. Dreher, PhD Thesis, University of Tübingen, 1979.
13 M. El-Mobayed, PhD Thesis, University of Tübingen, 1981.
14 A. I. Scott, P. A. Dodsen, F. Capra and M. B. Meyers, J. Am. Chem.
Soc., 1963, 85, 3702.
15 T. Inoue, K. Naitoh, S. Kosemura, I. Umezawa, N. Serizawa,
N. Mori and H. Itokawa, Heterocycles, 1983, 20, 397; Y. Tamura,
T. Yakura, J. Haruta and Y. Kita, J. Org. Chem., 1987, 52,
3927.
16 A. Pelter, A. Hussain, G. Smith and R. S. Ward, Tetrahedron, 1997,
53, 3879.
17 A. Rieker, J. Bracht, E. L. Dreher and H. P. Schneider, in Houben-
Weyl-Müller, Georg Thieme, Stuttgart, 1979, vol. 7, Part 3b, p. 523.
18 M. H. Khalifa, G. Jung and A. Rieker, Angew. Chem., 1980, 92, 739;
M. H. Khalifa, G. Jung and A. Rieker, Liebigs Ann. Chem., 1982,
1068.
A solution of 15c (60.0 mg, 0.102 mmol) in 50 ml of CHCl3 was
treated with (ϩ)-5,6-O-isopropylidene--ascorbic acid (44.5
mg, 0.205 mmol) and anhydrous ZnCl2 (10 mg). This mixture
was heated to 60 ЊC for 50 h, filtered and the solvent was
removed under reduced pressure. The crude product was chrom-
atographed on silica gel with n-hexane–EtOAc to give 16c (30
mg, 52%); Rf 0.36 (n-hexane–EtOAc, 8:2) 0.36; δH(250 MHz;
CDCl3) 0.77–0.79 [6H, m, CH2CH(CH3)2], 1.14–1.22 (3H, m,
CO2CH2CH3), 1.33 (18H, s, But), 1.32–1.53 [3H, m, CHCH2-
CH(CH3)2, CH2CH(CH3)2], 2.87–2.95 (2H, m, CCH2CH),
4.00–4.12 (2H, m, CO2CH2CH3), 4.32–4.46 (2H, m, NHCH-
CO2Et, CH2CHNHR); 5.01 (1H, s, C6H5CH2O), 5.02 (1H, s,
C6H5CH2O);¶ 5.06 (1H, s, OH), 5.21–5.31 [1H, m, NH(Leu)];
6.12 [1H, d, J ≈ 8, NH(Tyr)], 6.22 [1H, d, J ≈ 8, NH(Tyr)];¶ 6.89
(1H, s, Harom), 6.91 (1H, s, Harom);¶ 7.21–7.29 (5H, m, Harom-Z);
δC(62.9 MHz; CD3OD) 14.11, 22.10, 22.59, 24.82, 30.29, 34.28,
38.14, 41.61, 51.00, 56.16, 61.28, 67.02; 125.66, 125.86 (MNE);
126.86, 128.14; 127.94, 128.53 (MNE); 136.23, 152.82, 155.94,
170.81, 172.44; m/z (FD) 568 (Mϩ).
19 A. Rieker and S. Berger, Org. Magn. Reson., 1972, 4, 857.
20 A. Hutinec, A. Ziogas and A. Rieker, Amino Acids, 1996, 11,
345.
21 K. Regier, PhD Thesis, University of Tübingen, 1989.
22 J. H. Barnes, E. T. Borrows, J. Elks, B. A. Hems and A. G. Long,
J. Chem. Soc., 1950, 2824.
Acknowledgements
We gratefully acknowledge the DFG for support of this work.
Paper 8/01803K
Received 4th March 1998
Accepted 30th April 1998
¶ Doubling of the NMR signals due to the presence of 2 diastereomers.
2208
J. Chem. Soc., Perkin Trans. 1, 1998