Cyclisation of Novel Amino Oxo Esters to Tetramic Acids - Density Functional Theory Study of the Reaction Mechanism

Article abstract of DOI:10.1002/ejoc.200300422

The synthesis of novel N-urethane-protected γ-methylanlino-β -oxo esters and their use as precursors for the preparation of N-methyltetramic acids is described. The presence of the bulky urethane protecting group on the nitrogen atom gives rise to rotational isomers detectable in the NMR spectra of the compounds, along with the keto/enol tautomerism. The mechanism of the cyclisation reaction of γ-amino-β-oxo esters to tetramic acids was studied theoretically by the B3LYP hybrid density functional method. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejoc.200300422

FULL PAPER
Cyclisation of Novel Amino Oxo Esters to Tetramic Acids ؊ Density
Functional Theory Study of the Reaction Mechanism
[
a]
[a]
[a]
[b]
Anastasia Detsi, Antreas Afantitis, Giorgos Athanasellis, John Markopoulos,
[a]
[c]
Olga Igglessi-Markopoulou,* and Chris-Kriton Skylaris*
Keywords: Acylation / Cyclization / Density functional theory / Medicinal chemistry
The synthesis of novel N-urethane-protected γ-methylamino-
β-oxo esters and their use as precursors for the preparation
of N-methyltetramic acids is described. The presence of the
bulky urethane protecting group on the nitrogen atom gives
rise to rotational isomers detectable in the NMR spectra of
the compounds, along with the keto/enol tautomerism. The
mechanism of the cyclisation reaction of γ-amino-β-oxo es-
ters to tetramic acids was studied theoretically by the B3LYP
hybrid density functional method.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
The pyrrolidine-2,4-dione ring system (tetramic acid) is
present in various naturally occurring compounds exhibit-
ing a wide range of biological and pharmaceutical activi-
ties.[ Representative examples are erythroskyrine [a my-
cotoxin containing an N-methyl-(S)-valine-derived acylte-
tramic acid terminus] and fuligorubin A[ [a pigment from
the slime mould Fuligo septica (L.) bearing a polyene chro-
mophore]. Recently, several natural products incorporating
the N-methyltetramic acid nucleus have been isolated. These
1]
[2]
3]
[4]
include aflastatin A, a specific inhibitor of aflatoxin pro-
duction by Asp. parasiticus, vancoresmycin,[5] a new anti-
biotic active against Gram-positive bacteria, melophlins A
and B, which exhibit anticancer activity, and melophlins
CϪO with antibacterial and antifungal activity, ancorino-
[
6]
[7]
[8]
side A and its Mg salt, and the potent antimycotic agent
cryptocin (Scheme 1).
[
9]
Since we have long been involved in the development of
a new methodology for the synthesis of N-acyl- and N-ure-
thane-protected tetramic acids through C-acylation/cyclis-
ation reactions of active methylene compounds with α-am-
ino acid activated derivatives as acylating agents,[ we fo-
cused on the extension of this synthetic route to the prep-
aration of N-methyltetramic acids, using the N-
hydroxysuccinimidyl ester of tert-butoxycarbonyl-N-meth-
ylglycine (Boc-sarcosine) as acylating agent. Removal of the
10]
[
a]
Laboratory of Organic Chemistry, Department of Chemical
Engineering, National Technical University of Athens,
Athens 15773, Greece
Scheme 1. Examples of natural products incorporating the N-meth-
yltetramic acid nucleus
[
[
b]
c]
Department of Chemistry, University of Athens,
Athens, Greece
Theory of Condensed Matter Group, Cavendish Laboratory
Madingley Road, Cambridge CB3 OHE, UK
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Boc-protecting group from the C-acylation compounds and
subsequent cyclisation of the corresponding hydrochloride
Eur. J. Org. Chem. 2003, 4593Ϫ4600
DOI: 10.1002/ejoc.200300422
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4593

O. Igglessi-Markopoulou, C.-K. Skylaris et al.
FULL PAPER
salts under basic conditions provided the desired N-methyl-
tetramic acids.
One interesting feature of this synthetic route is the iso-
lation of the C-acylation compounds, 4-[Boc(methyl)am-
ino]-3-hydroxybutenoates (Scheme 2), which belong to the
structural class of γ-amino-β-hydroxy acids and are novel
statine precursors. Statine (3S,4S)-4-amino-3-hydroxy-6-
methylheptanoic acid, and its analogues (Scheme 2) are
widely used in the design and synthesis of pseudopeptides
that inhibit aspartic acid proteases, such as renin, pepsin
Scheme 3
[
11]
and cathepsin D.
The N-methyl group as present in 4-
With this in mind, we performed the base-promoted C-
[
Boc(methyl)amino]-3-hydroxybutenoates can also be found acylation reactions of dimethyl and diethyl malonate and
in the statine analogue dolaisoleuine [(3R,4S,5S)-N,O-di- ethyl methylsulfonyl acetate with the N-hydroxysuccinim-
methylisostatine] (Scheme 2), a component of dolastatin 10, idyl ester of Boc-sarcosine (1) in order to investigate the
a marine natural product with cytotoxic and antineoplastic possibility of isolating the corresponding 2-substituted 4-
[
12]
activity.
The synthesis of N-methylstatine analogues is [Boc(methyl)amino]-3-hydroxy-2-butenoates, which, apart
therefore of particular interest, since they can be useful for from being novel statine analogues, could be used as precur-
structure-activity relationship studies. On the other hand, sors for the synthesis of the requisite 3-substituted N-meth-
these compounds can also be used as precursors for the yltetramic acids.
synthesis of 3-substituted N-methyltetramic acids.
The N-hydroxysuccinimidyl ester of Boc-sarcosine (1)
was prepared by standard experimental procedures.[ The
C-acylation reactions were performed under basic con-
ditions with NaH in THF (Scheme 4) and excess of the ac-
tive methylene compound. The corresponding 2-substituted
13]
4-[Boc(methyl)amino]-3-hydroxy-2-butenoates 2Ϫ4 were
isolated as viscous, colourless oils after flash column chro-
matography (see Exp. Sect.).
Removal of the Boc-protecting group proved to be essen-
tial for the cyclisation of the 2-substituted 4-[Boc(methyl)-
amino]-3-hydroxy-2-butenoates to the requisite 3-substi-
tuted N-methyltetramic acids. This deprotection was ef-
fected by treatment with saturated HCl in MeOH or EtOH
at room temperature, and the corresponding hydrochloride
salts 5Ϫ7 were isolated as hygroscopic, low-melting, white
solids after 0.5 h. Treatment of these salts with RONa/ROH
at room temperature for 24 h (monitored by TLC) resulted
in the isolation of 3-substituted N-methyltetramic acids
8Ϫ10 in good yields (70Ϫ90%).
Scheme 2
It is worth noting that, when analogous reactions were
performed with the N-hydroxysuccinimidyl esters of Boc-,
Z- and N-acetylglycine, the corresponding C-acylation
compounds could not be isolated and the N-substituted
tetramic acids were obtained as the only products directly
Furthermore, to investigate possible mechanisms for the
cyclisation reactions that produce tetramic acids, we have
examined the structures and energetics of critical points on
the potential energy surfaces of 2-(alkoxycarbonyl)-4-(alkyl-
amino)-3-hydroxy-2-butenoates by theoretical calculations
by the B3LYP Hybrid Density Functional Theory (DFT)
method in vacuo and in solvent environments.
[
9]
after workup (Scheme 5). This is due to the presence of a
proton directly attached to the nitrogen atom. This proton
is readily lost under basic conditions to initiate intramol-
ecular cyclisation, as we propose below in our DFT study
of the reaction mechanism. The lack of a proton on the
nitrogen atom blocks the cyclisation path, so the C-acyl-
ation products 2, 3 and 4 can be isolated.
Results and Discussion
1
13
The H and C NMR spectra of 2-substituted 4-[Boc-
(
methyl)amino]-3-hydroxy-2-butenoates 2Ϫ4 are of special
Experimental Results
interest. These molecules are representative examples of β-
The N-hydroxysuccinimidyl ester of Boc-sarcosine (1) dicarbonyl compounds capable of undergoing keto/enol
has proved to be a powerful acylating agent when used to tautomerism. Moreover, the presence of the bulky tert-bu-
acylate alkyl cyanoacetates under basic conditions, provid- toxycarbonyl protecting group gives rise to rotational iso-
ing the corresponding 4-[Boc(methyl)amino]-2-cyano-3- mers for both the keto and the enol tautomers (Scheme 6)
hydroxy-2-butenoates (Scheme 3) in good yields.[
13]
as a result of hindered rotation around the CϪN bond.
4594
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 4593Ϫ4600

Cyclisation of Novel Amino Oxo Esters to Tetramic Acids
FULL PAPER
Scheme 4. (i) NaH/THF, 1 h, room temp.; (ii) HCl/ROH, 0.5 h, room temp.; (iii) RONa/ROH, 24 h, room temp.
These rotamers are detectable at room temperature, as has
[14]
been previously reported for N,N-disubstituted amides
and N-substituted N-methylamino acid derivatives.[
12,15]
In CDCl solution, compounds 2Ϫ4 exist as mixtures of
3
keto and enol tautomers. The keto tautomer is detected by
the presence of a sharp singlet at δ ϭ 4.5Ϫ5.0 ppm, attribu-
table to the methine proton, whereas the two broad singlets
at low field ( δ ϭ 13Ϫ14 ppm) are indicative of the presence
of the strongly deshielded proton of the enol form partici-
pating in a hydrogen bond.
The presence of enol/keto tautomerism along with ro-
tational isomerism gives rise to relatively complex spectra
Scheme 5
1
for these compounds. In the H NMR spectrum (CDCl )
3
Scheme 6. Rotational isomers of 2-substituted 4-[Boc(methyl)amino]-3-hydroxy-2-butenoates
www.eurjoc.org
Eur. J. Org. Chem. 2003, 4593Ϫ4600
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4595

O. Igglessi-Markopoulou, C.-K. Skylaris et al.
FULL PAPER
of
methyl
4-[(tert-butoxycarbonyl)(methyl)amino]-3- the optimised geometries in order to verify that they were
hydroxy-2-(methoxycarbonyl)-2-butenoate (2), the simplest indeed equilibrium states (minima) or transition states
of the C-acylation compounds synthesized, we can clearly (saddle points). Energetics were obtained from single-point
observe two sets of signals, one for each tautomer, which calculations at the optimised structures with the larger
are split further into two sets of signals, one for each rot- pTZV basis set designed to improve the energies obtained
[
19]
amer.
with the pVDZ basis.
It was not necessary to augment
In this case the enol/keto ratio is 1.2 (based on inte- the pTZV basis with extra diffuse functions, as tests we car-
gration of the hydroxy and methine protons), so the enol ried out (augmentation with ‘‘Pople-style’’ diffuse func-
[
20]
tautomer slightly predominates. According to the literature, tions: s on hydrogen, s and p on other elements) resulted
the protons of the N-methyl group cis to the oxygen atom in changes in the relative energies of our anionic species of
(cisoid form) should resonate at higher field than those trans less than 1.0 kJ/mol while making our calculations con-
to the oxygen atom (transoid form) because the cis-methyl siderably more expensive. Solvation effects in THF (dielec-
groups experience greater shielding.[ These facts facili- tric constant ε ϭ 7.52) and in water (dielectric constant ε ϭ
tated the assignment of the signals of the spectrum and al- 78.4) were estimated by single-point energy calculations at
14]
[
21]
lowed the estimation of the transoid/cisoid ratio in the case the optimised geometries with the COSMO model.
of each tautomer. The results are summarized in Table 1.
Results Relating to the Cyclisation Mechanism Obtained by
the B3LYP Hybrid Density Functional Method
1
Table 1. Selected H NMR chemical shifts for methyl 4-[(tert-bu-
toxycarbonyl)(methyl)amino]-3-hydroxy-2-(methoxycarbonyl)-2-
We selected methyl 4-amino-3-hydroxy-2-(methoxycar-
bonyl)-2-butenoate (I, Scheme 7), the simplest molecule of
the class we are interested in, as a model for a detailed study
of the cyclisation reaction. From our calculations, the 3-
hydroxy-2-butenoate tautomeric form is the most stable of
the possible tautomeric forms for these compounds. The
same is true for the resulting tetramic acids, which therefore
appear as 3-substituted 4-hydroxypyrrolin-2-ones through-
out the text.
butenoate (2) (300 MHz, CDCl
3
)
Enol tautomer
transoid
form
Keto tautomer
cisoid
form
transoid
form
cisoid
form
NϪCH
transoid/cisoid
ϪCH
transoid/cisoid
3
2.87
1.1
4.18
1.1
2.85
4.12
2.91
1.2
4.34
1.2
2.89
4.26
2
Ϫ
It can clearly be observed that in both cases (enol and
keto tautomers) the transoid rotamer is slightly more abun-
dant than the cisoid form. This observation is consistent
with the fact that, since the equilibrium distributions of cis
and trans isomers of unsymmetrical N,NЈ-disubstituted am-
ides depend on both steric and electronic factors, the more
stable rotamers should be those in which the bulkier N-
substituents are cis to the carbonyl oxygen atoms (the
transoid forms in this case).
Scheme 7
Doubling of the corresponding signals of most of the car-
bon atoms in the C NMR spectra of compounds 2Ϫ4 was
also observed.
The cyclisation of I by intramolecular nucleophilic attack
of the nitrogen lone pair on the carbonyl group would pro-
duce structures such as IЈ. We found that structures of this
type are not stable neither as equilibrium states nor as tran-
sition states, and attempts to optimize them led back to the
reactant molecule I, a strong indication that this form of
cyclisation involving the neutral molecule I is not favoured.
In the Exp. Sect. of this paper it is mentioned that the
cyclisation reaction takes place in a solution of NaH in
THF when the N-substituents are a Boc group and an H
atom. If the N-substituents are a Boc group and a methyl
group, no cyclisation takes place. Only after the Boc group
is replaced by an H atom can cyclisation occur, again under
1
3
The hydrochloride salts 5Ϫ7 exist in keto/enol equilibria
1
13
in CDCl solution, as indicated by their H and C NMR
3
spectra (see Exp. Sect.). The 3-substituted N-methyltetramic
acids 8Ϫ10 exist in their enol forms in CDCl solution, as
3
there were no signals attributable to methine protons in
1
13
their H NMR spectra, whereas their C NMR spectra
3
lacked any resonance characteristic of the sp -CH methine
group expected for the keto tautomers.
Computational Details
We performed ab initio calculations by use of the highly basic conditions, this time brought about by use of
[
16]
B3LYP hybrid density functional method with the aid of a solution of MeONa in MeOH. These facts suggest that
[17]
NWChem 4.1 Computational Chemistry software. Equi- the presence of at least one proton attached to the nitrogen
librium and transition state structures were optimised with atom is a necessary condition for the intramolecular cyclis-
the pVDZ basis set, which provides highly accurate struc- ation reaction. It is therefore likely that the cyclisation is
[
18]
tures and at the same time is affordable in terms of com- initiated by the abstraction of this proton. The depro-
putational cost. Frequency calculations were performed at tonation reaction requires some other molecule such as a
4596
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 4593Ϫ4600

Cyclisation of Novel Amino Oxo Esters to Tetramic Acids
FULL PAPER
basic anion to bind the departing proton. In the case of structure is shown in Figure 1. On the product side we ob-
MeONa in methanol solution, a methoxide anion would be tained structure II-HOMe, the outcome of the depro-
the proton receptor. The reactant structure corresponding tonation reaction, also shown in Figure 1.
to this situation is the structure IϪOMe shown in Figure 1,
The sequence of structures IϪOMe, IϪOMe-TS and II-
in which a methoxide anion is hydrogen-bonded to one of HOMe shows that the deprotonation reaction is followed
the protons of the nitrogen atom of I. Starting from the by an intramolecular nucleophilic attack of the nitrogen
I؊OMe structure, we were able to locate the transition state lone pair on the carbonyl carbon atom, which results in the
I؊OMe-TS corresponding to the process of proton transfer formation of the five-membered heterocyclic ring present in
from the nitrogen atom to the oxygen atom of the methox- the II-HOMe structure. In summary, the reaction consists
ide anion. The optimized geometry of this transition state of two stages: (1) proton transfer from the nitrogen atom
Figure 1. Deprotonation and cyclisation of I and IV; a methoxide anion acts as the receiver of one of the protons initially attached to
the nitrogen atom; the optimized structures of the reactant, transition state and products are shown, including isosurface plots of their
highest occupied molecular orbital (HOMO) at 0.018 a.u.; the relative energy in kJ/mol of each structure in vacuo and in THF solvent
˚
(
in parentheses) is shown under its name; selected distances in A are also shown
Eur. J. Org. Chem. 2003, 4593Ϫ4600
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4597

O. Igglessi-Markopoulou, C.-K. Skylaris et al.
FULL PAPER
to the methoxide ion oxygen atom (structure IϪOMe-TS), mol in vacuo and Ϫ10.9 kJ/mol in THF, or during the
and (2) ring closure through formation of aqueous workup with the participation of an oxonium ion
a
ϩ
carbonϪnitrogen bond (structure II-HOMe). Figure 1 also H O :
3
shows the highest occupied molecular orbital (HOMO) of
each structure as an isosurface plot. Initially the HOMO is II-HOMe Ǟ III ϩ OMe ϩ HOMe
Ϫ
almost entirely localised on the methoxide anion, which be-
ϩ
3 2
II-HOMe ϩ H O Ǟ III ϩ 2 HOMe ϩ H O
ars the negative charge. As the reaction proceeds, the
HOMO vanishes from the methoxide anion and appears
along the path of the nucleophilic attack, so that the forma-
tion of the CϪN bond is concerted with the proton abstrac-
tion. The energy is increased by 11.6 kJ/mol (27.1 kJ/mol in
THF solvent) upon reaching the transition state and de-
creased by 21.8 kJ/mol (59.8 kJ/mol in THF solvent) when
moving from the transition state to the products.
ϩ
The energy we obtain for this reaction with H O is
3
Ϫ785.9 kJ/mol in vacuo and Ϫ250.9 kJ/mol in water. Simi-
larly, for IV-HOMe dissociation under basic conditions we
obtain an energy of 118.1 kJ/mol in vacuo and Ϫ19.5 kJ/
mol in THF, while under the acidic conditions of the aque-
ous workup we find an energy of Ϫ805.0 kJ/mol in vacuo
and Ϫ257.7 kJ/mol in water. Here the solvent also plays an
important role, particularly in the case of the dissociation
reactions of II-HOMe and IV-HOMe under basic con-
ditions, which are favoured energetically when the THF sol-
vent is taken into account.
The above cyclisation mechanism for the model molecule
I is also valid for its N-methyl analogue IV (Scheme 7).
Structures IVϪOMe, IV-OMe-TS and V-HOMe are ob-
tained (Figure 1), closely resembling structures I-OMe, I-
OMe-TS and II-HOMe in their geometrical parameters,
the main difference being that they each contain a methyl
group attached to the nitrogen atom. On going from the
reactant IV-OMe to the transition state IV-OMe-TS, the
energy is increased by 11.0 kJ/mol (31.0 kJ/mol in THF sol-
vent). It is then decreased by 10.6 kJ/mol (37.3 kJ/mol in
THF solvent) upon going from IV-OMe-TS to the product
V-HOMe. The fact that less energy is released in the cyclis-
ation reaction of IV than in the case of I can be attributed
to steric effects caused by the extra methyl group of IV.
The THF solvent affects the energetics of the cyclisation
of I-OMe and IV-OMe by stabilizing the equilibrium states
Ϫ and especially the products in the case of IV-OMe Ϫ
more than the transition states. This results in a more exo-
thermic reaction overall, albeit with a higher energy barrier.
Structures II-HOMe and V-HOMe are stable intermedi-
ates that will result in the formation of the desired tetramic
acids III and VI, respectively. This will require the expulsion
of one methoxy unit and the attainment of a planar ge-
ometry by the atoms participating in the ring, as shown in
the optimised structures in Figure 2. We must also note that
the methanol molecule that was hydrogen-bonded to the
nitrogen atom is also gone, since the nitrogen atom now has
a planar geometry.
Conclusions
We have synthesized novel statine analogues, namely 2-
substituted 4-[Boc(methyl)amino]-3-hydroxy-2-butenoates,
through C-acylation reactions between active methylene
compounds and the N-hydroxysuccinimidyl ester of Boc-
sarcosine. The NMR spectra of these compounds are rather
complicated because of the existence of two rotamers aris-
ing from hindered rotation around the CϪN bond and of
the two possible tautomeric forms (keto and enol) in
CDCl solution.
3
The new compounds were used as precursors for the syn-
thesis of 3-substituted N-methyltetramic acids. After re-
moval of the Boc protecting groups, the isolated hydrochlo-
ride salts were easily cyclised to the desired heterocyclic
compounds. This new approach to the synthesis of N-meth-
yltetramic acids provides a convenient and versatile way to
these biologically active species.
Moreover, the mechanism of the cyclisation reaction of 4-
amino-2-butenoates was studied theoretically by the B3LYP
hybrid density functional method in vacuo and in solvent
environments. We suggest that a plausible cyclisation path
would involve the formation of an anion by proton abstrac-
tion from the nitrogen atom. The presence of at least one
hydrogen atom attached to the nitrogen atom is therefore
necessary for cyclisation.
Experimental Section
Figure 2. Left: optimized structure of III; right: optimized structure
of VI; note that the nitrogen atom has a completely planar geo-
metry in both cases
General: Melting points were determined with a Gallenkamp
MFB-595 melting point apparatus; the results are given without
correction. The FT IR spectra were recorded with a Nicolet Magna
IR 560 instrument. The H and C NMR spectra were recorded
with a Varian Gemini 2000 spectrometer. The chemical shifts are
given relative to TMS; J values are given in Hz. TLC analyses:
1
13
We were not able to locate transition states for this last
stage in the formation of the tetramic acids. This could oc-
cur either by dissociation of II-HOMe, still under basic Silica gel Fluka F254 plates (0.2 mm layer thickness). Flash column
conditions, in which case the energy we obtain is 137.2 kJ/ chromatography: Merck Kieselgel 60 (230Ϫ400 mesh). Elemental
4598
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 4593Ϫ4600

Cyclisation of Novel Amino Oxo Esters to Tetramic Acids
FULL PAPER
analyses were obtained with a EuroVector EURO EA Elemental
Analyser.
1.43 (2 s, each 4.5 H, C(CH
3
)
3
], 2.86 (s, 3 H, NCH
3
), 3.19 (s, 3 H,
SO CH ), 4.22Ϫ4.29 (m, 4 H, COOCH
2
3
2
CH
3
1
and 4-H), 4.98 and
3
5.02 (2 br. s, 1 H, 2-H, keto form) ppm. C NMR (75 MHz,
N-Hydroxysuccinimidyl Ester of N-Boc-sarcosine (1): The ester was
CDCl
(CH
4), 62.6 (COOCH
C], 155.2 and 156.1 [(CH
form), 193.0 and 193.2 (C-3, keto form) ppm. IR (CHCl
3
): δ ϭ 13.67 and 13.74 (COOCH
], 35.2 and 35.5 (NCH ), 41.5 (SO
CH ), 63.3 and 63.4 (C-2, keto form), 80.7 and
COCO], 163.2 (C-1, keto
): ν˜ ϭ
3019 cm s (OH), 1737 m (CϭO, β-oxo ester, keto form), 1691 m
(CϭO, β-oxo ester, enol form and CϭO, urethane), 1328 s (SO ).
2
CH
3
), 27.95 and 28.06
prepared as described in ref.[
13]
[
3
)
3
3
2
CH
3
), 59.1 and 59.9 (C-
General Procedure for the C-Acylation of Active Methylene Com-
pounds with the N-Hydroxysuccinimidyl Ester of N-Boc-sarcosine 80.8 [(CH
2
3
3
)
3
3 3
)
(
1): The active methylene compound (10 mmol) was added drop-
wise at 0 °C to a mixture of sodium hydride (0.17 g, 7 mmol,
5Ϫ60% sodium hydride in oil) in anhydrous THF (20 mL) and
3
Ϫ1
5
2
the resulting mixture was stirred at room temperature for 1 h. The
N-hydroxysuccinimidyl ester of N-(tert-butoxycarbonyl)sarcosine
General Procedure for the Removal of the tert-Butoxycarbonyl Pro-
tecting Group: 2-Substituted 4-[Boc(methyl)amino]-3-hydroxy-2-
butenoate (3 mmol) was dissolved in 1 mL of a saturated solution
of HCl in MeOH (for compound 2) or EtOH (for compounds 3
and 4) and the mixture was stirred at room temperature for 0.5 h
(1 g, 3.5 mmol) was then added, and stirring was continued at room
temperature for 1 h. The solvent was evaporated in vacuo and the
residue was diluted with water. The aqueous layer was washed once
with diethyl ether, acidified with 10% hydrochloric acid and ex-
(
monitored by TLC). The solvent was evaporated and the residue
tracted with CH
2 2
Cl . The organic layer was separated and dried
was triturated with diethyl ether to give the corresponding hydro-
chloride salt as a highly hygroscopic white solid.
(
Na SO ), and the solvents were evaporated in vacuo. The oily resi-
2
4
due was purified by flash column chromatography (petroleum
ether/EtOAc, 8:2) to afford the corresponding 2-substituted 4-[Boc-
Methyl 3-Hydroxy-2-(methoxycarbonyl)-4-(methylamino)-2-buteno-
(
methyl)amino]-3-hydroxy-2-butenoates 2Ϫ5 as colourless, viscous
ate Hydrochloride (5): Yield 0.70 g, 98%. C
calcd. C 40.43, H 5.09, N 5.89; found C 40.65, H 5.30, N 5.64. H
NMR (300 MHz, CDCl ): δ ϭ 2.78,2,79 and 2.81 (3 br. s, 3 H,
NCH ), 3.73, 3.77, 3.81 and 3.86 [3 s, 6 H, (COOCH ], 4.27
pseudo-triplet, 1 H, 4-H), 4.42 (pseudo-triplet, 1 H, 4-H), 4.93 (s,
8 5
H14ClNO (239.6):
1
oils.
3
Methyl 4-[(tert-Butoxycarbonyl)(methyl)amino]-3-hydroxy-2-meth-
oxycarbonyl-2-butenoate (2): Yield 0.81 g, 76%; TLC (petroleum
3
3 2
)
(
ether/EtOAc, 8:2) R
H 6.98, N 4.62; found C 51.31, H 7.30, N 4.60. H NMR
300 MHz, CDCl ): δ ϭ 1.40 and 1.44 [2 s, each 4.5 H, C(CH ],
.85, 2.86, 2.89 and 2.91 (4 br. s, 3 H, NCH ), 3.73, 3.75, 3.79 and
.82 [4 s, 6 H, (COOCH ], 4.12, 4.18, 4.26 and 4.34 (4 br. s, 2 H,
-H), 4.58 (s, 0.5 H, 2-H, keto form), 14.03 (br. s, 0.6 H, OH, enol
f 7
ϭ 0.37. C13H21NO (303.3): calcd. C 51.48,
0
.5 H, 2-H, keto form), 9.55 (br. s, 1 H, NH), 9.74 (br. s, 0.9 H,
1
1
3
NH) ppm. C NMR (75 MHz, CDCl
CH ), 52.5, 52.7, 53.2 and 53.7 (COOCH
3
): δ ϭ 33.1 and 33.2 (N-
), 49.3 (C-4), 63.0 (C-2,
(
3
3 3
)
3
3
2
3
4
3
keto form), 100.2 (C-2, enol form), 164.2 and 164.8 (C-1, enol
3 2
)
form), 171.9 (C-1, keto form), 175.0 (C-3, enol form), 192.6 (C-3,
Ϫ1
2 2
keto form) ppm. IR (CH Cl
) : ν˜ ϭ 3054 cm m and 2986 m
13
form) ppm. C NMR (75 MHz, CDCl
(CH ], 35.4 and 35.5 (NϪCH ), 52.0, 52.5 and 53.2 (COOCH
7.4 and 58.1 (C-4), 61.8 and 62.2 (C-2, keto form), 80.2, 80.5 and
0.8 [(CH C], 98.1 and 98.5 (C-2, enol form), 155.4 and 155.7
COCO], 164.7, 164.8 and 167.1 (C-1, enol form), 172.0 (C-
, keto form), 181.6 and 182.0 (C-3, enol form), 195.6 and 195.7
3
): δ ϭ 27.9 and 28.1
ϩ
(
OH), 2685 br. m (NH
2
), 1736 m (CϭO, β-oxo ester, keto form),
[
3
)
3
3
3
),
1701 m (CϭO, β-oxo ester, enol form).
5
8
3
)
3
Ethyl 2-(Ethoxycarbonyl)-3-hydroxy-4-(methylamino)-2-butenoate
Hydrochloride (6): Yield 0.76 g, 95%. C10 18ClNO (267.7): calcd.
C 44.86, H 6.77, N 5.23; found C 44.49, H 7.15, N 4.94. H NMR
): ν˜ ϭ 3055 cm m, 2983 m (300 MHz, CDCl
): δ ϭ 1.27Ϫ1.35 [m, 6 H, (COOCH CH ], 2.79
OH), 1736 s (CϭO, β-oxo ester, keto form), 1698 s (CϭO, β-oxo (s, 3 H, NCH ), 4.20Ϫ4.38 [m, 6 H, (COOCH CH and NCH ),
4.75 (s, 0.5 H, 2-H, keto form), 9.80 and 9.90 (2 br. s, 2 H, NH
[(CH
3
)
3
H
5
1
1
(
(
Ϫ1
C-3, keto form) ppm. IR (CH
2
Cl
2
3
2
3 2
)
3
2
3
)
2
2
ester, enol form and CϭO, urethane).
2
)
),
),
1
3
ppm. C NMR (75 MHz, CDCl
3
): 13.8 and 13.9 (COOCH
2
CH
CH
3
Ethyl 4-[(tert-Butoxycarbonyl)(methyl)amino]-2-(ethoxycarbonyl)-3-
hydroxy-2-butenoate (3): Yield 0.93 g, 81%; TLC (petroleum ether/
3
6
1
3.0 and 33.1 (N-CH
3
), 49.1 (C-4), 61.4 and 62.5 (COOCH
2
3
3.0 and 63.6 (C-2, keto form), 100.9 (C-2, enol form), 163.7 and
64.6 (C-1, enol form), 171.4 (C-1, keto form), 174.2 (C-3, enol
EtOAc, 8:2). R
N 4.23; found C 54.10, H 8.05, N 4.61. H NMR (300 MHz,
CDCl ): δ ϭ 1.29 [t, J ϭ 7.2 Hz, 6 H, (COOCH CH ], 1.42 and
.46 (2 s, each 4.5 H, C(CH ], 2.86, 2.88, 2.91 and 2.93 (4 br. s,
H, NCH ), 4.13 (s, 1 H, 4-H), 4.19Ϫ4.35 [m, 5 H, (CO-
CH and 4-H], 4.55 (s, 0.5 H, 2-H, keto form), 13.97 and Ethyl 3-Hydroxy-4-(methylamino)-2-(methylsulfonyl)-2-butenoate
4.01 (2 br. s, each 0.2 H, OH enol form) ppm. C NMR (75 MHz, Hydrochloride (7): Yield 0.79 g, 97%. C
CDCl ): δ ϭ 13.8 and 13.9 (COOCH CH ), 28.0 and 28.2 [(CH ], C 35.10, H 5.89, N 5.12; found C 35.00, H 5.85, N 4.99. H NMR
5.4 and 35.5 (NCH ), 57.4 and 58.2 (C-4), 60.9, 61.5 and 61.7 (C- (300 MHz, CDCl /CD OD): δ ϭ 1.20Ϫ1.27 [m, (COOCH CH ],
, keto form), 62.3, 62.5 and 62.8 (COOCH CH ), 80.2, 80.5 and 2.64 and 2.66 (2 br. s, 3 H, NCH ), 3.16 (br. s, 3 H, SO CH ),
0.8 [(CH C], 98.8 and 98.3 (C-2, enol form), 155.5 and 156.2 4.15Ϫ4.27 [m, 5 H, (COOCH CH , 4-H and 2-H, keto form)
COCO], 164.5, 165.5 and 166.8 (C-1, enol form), 171.6 (C-
ppm. IR (KBr): ν˜ ϭ 2937 cm s, 2758 s, 2693 s (NH and OH),
, keto form), 181.2 (C-3, enol form), 195.9 (C-3, keto form) ppm. 1749 s (CϭO, β-oxo ester, keto form), 1341 s (SO ).
f 7
ϭ 0.60. C15H25NO (331.4): calcd. C 54.37, H 7.60,
1
Ϫ1
form), 192.4 (C-3, keto form) ppm. IR (CH
and 2976 m (OH), 2931 and 2870 br. m (NH
2
Cl
2
ϩ
: ν˜ ϭ 3052 cm
m
3
2
3 2
)
2
), 1733 s (CϭO, β-
1
3
3 3
)
oxo ester, keto form), 1695 m (CϭO, β-oxo ester, enol form).
3
OCH
1
2
3 2
)
13
8 5
H16ClNO S (273.7): calcd.
1
3
2
3
3 3
)
3
2
8
3
3
3
2
3 2
)
2
3
3
2
3
3
)
3
2
Ϫ1
3 2
)
3 3
[(CH )
1
2
Ϫ1
2 2
IR (CH Cl
): ν˜ ϭ 2982 cm m (OH), 1731 s (CϭO, β-oxo ester,
keto form), 1693 s (CϭO, β-oxo ester, enol form and CϭO, ure-
thane).
General Procedure for the Synthesis of 3-Substituted N-Methyltetra-
mic Acids: The hydrochloride salt (2 mmol) was added to a solution
of sodium alkoxide in alcohol (MeONa in dry MeOH for com-
pound 5 and EtONa in EtOH for compounds 6 and 7) prepared
Ethyl 4-[(tert-Butoxycarbonyl)(methyl)amino]-3-hydroxy-2-(methyl-
sulfonyl)-2-butenoate (4): Yield 0.80 g, 68%; TLC (petroleum ether/ from sodium (4 mmol) in alcohol (1 mL). The reaction mixture was
EtOAc, 8:2). R
f
ϭ 0.51. C13
H
23NO
7
S (337.4): calcd. C 46.29, H
stirred at room temperature for 24 h (monitored by TLC). The sol-
1
6.87, N 4.15; found C 45.97, H 7.21, N 4.50. H NMR (300 MHz, vent was evaporated in vacuo and the solid residue was dissolved
CDCl
3
): δ ϭ 1.29 [t, J ϭ 6.9 Hz, 6 H, (COOCH
2 3 2
CH ) ], 1.38 and
in water. The aqueous solution was acidified with 10% hydrochloric
Eur. J. Org. Chem. 2003, 4593Ϫ4600
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4599

O. Igglessi-Markopoulou, C.-K. Skylaris et al.
FULL PAPER
[
[
[
5]
6]
7]
acid and extracted with CH
2
Cl
2
. The organic layer was separated
C. Hopmann, M. Kurz, M. Bronstrup, J. Wink, D. LeBeller,
Tetrahedron Lett. 2002, 43, 435Ϫ438.
S. Aoki, K. Higuchi, Y. Ye, R. Satari, M. Kobayashi, Tetra-
hedron 2000, 56, 1833Ϫ1836.
and dried (Na
2 4
SO ), and the solvents were evaporated in vacuo to
afford the corresponding tetramic acids as solids.
3-(Methoxycarbonyl)-1-methyltetramic Acid (8): White solid. Yield
C.-Y. Wang, B.-G. Wang, S. Wiryowidagdo, V. Wray, R. vanSo-
est, K. G. Steube, H.-S. Guan, P. Proksch, R. Ebel, J. Nat.
Prod. 2003, 66, 51Ϫ56.
0.30 g, 89%; m.p.188Ϫ191 °C. C NO (171.05): calcd. C 49.12,
7
H
9
4
1
H 5.30, N 8.18; found C 49.48, H 5.67, N 7.94. H NMR
300 MHz, CDCl ): δ ϭ 2.98 (s, 3 H, NCH ), 3.91 (s, 3 H, CO-
OCH
), 3.97 (s, 2 H, 5-H) ppm. IR (KBr): ν˜ ϭ 2952 cmϪ1 m br.
OH), 1865 m, 1714 s, 1600 s (CϭO).
[
8]
E. Ohta, S. Ohta, S. Ikegami, Tetrahedron 2001, 57,
(
3
3
4699Ϫ4703.
3
[9]
J. Y. Li, G. Strobel, J. Harper, E. Lobkovsky, J. Clardy, Org.
Lett. 2000, 2, 767Ϫ770.
10] [10a]
(
[
A. Detsi, J. Markopoulos, O. Igglessi-Markopoulou, Chem.
3
0
-(Ethoxycarbonyl)-1-methyltetramic Acid (9): White solid. Yield
Commun. 1996, 1323Ϫ1324. [
10b]
M. Petroliagi, O. Igglessi-
.35 g, 94%; m.p. 183Ϫ186 °C (ref.[
22]
190Ϫ192°C). 1H NMR
): δ ϭ 1.37 (t, J ϭ 7.2 Hz, 3 H, COOCH CH ),
), 3.96 (s, 2 H, 5-H), 4.39 (q, J ϭ 7.2 Hz, 2 H,
Markopoulou, J. Chem. Soc., Perkin Trans. 1 1997, 3543Ϫ3548.
(300 MHz, CDCl
3
2
3
[10c]
G. Athanasellis, E. Gavrielatos, O. Igglessi-Markopoulou,
2
.97 (s, 3 H, NCH
3
Synlett 2001, 10, 1653Ϫ1655.
P. Castejon, A. Moyano, M. A. Pericas, A. Riera, Tetrahedron
1996, 52, 7063Ϫ7086.
Ϫ1
[11]
COOCH
2
CH
3
) ppm. IR (KBr): ν˜ ϭ 2982 cm m br. (OH), 1840
m, 1704 s, 1607 s (CϭO).
[
[
12]
13]
G. R. Pettit, S. B. Singh, J. K. Srirangam, F. Hogan-Pierson,
M. D. Williams, J. Org. Chem. 1994, 59, 1796Ϫ1800.
A. Detsi, E. Gavrielatos, M.-A. Adam, O. Igglessi-Markopou-
lou, J. Markopoulos, M. Theologitis, H. Reis, M. Papado-
poulos, Eur. J. Org. Chem. 2001, 4337Ϫ4342.
1
-Methyl-3-methylsulfonyltetramic Acid (10): Off-white solid. Yield
.29 g, 77%; m.p. 164Ϫ166 °C (from CH Cl /petroleum ether).
NO S (191.02): calcd. C 37.69, H 4.74, N 7.32; found C 37.78,
H 4.97, N 7.12. H NMR (300 MHz, CDCl
0
2
2
C
6
H
9
4
1
3
/CD
), 3.92 (s, 2 H, 5-H) ppm.
OD): δ ϭ 28.4 (NCH ), 42.8
3
OD): δ ϭ 2.83
[
[
14]
W. E. Stewart, T. H. Siddall, Chem. Rev. 1970, 70, 517Ϫ551.
(
3 2 3
s, 3 H, NCH ), 3.11 (s, 3 H, SO CH
15] [15a]
1
3C NMR (75 MHz, CDCl
S. Kano, Y. Yuasa, S. Shibuya, Heterocycles 1990, 31,
3
/CD
3
3
597Ϫ1600. [
15b]
1
M. Calmes, J. Daunis, R. Elyacoubi, R. Jacqu-
(
SO
IR (KBr): ν˜ ϭ 2928 cm m br. (OH), 1596 s (CϭO), 1302 s, 1136
s (SO ).
2 3
CH ), 52.1 (C-5), 105.5 (C-3), 165.7 (C-2), 175.6 (C-4) ppm.
[
15c]
ier, Tetrahedron: Asymmetry 1990, 1, 329Ϫ334.
E. Vedejs,
Ϫ1
C. Kongkitingam, J. Org. Chem. 2000, 65, 2309Ϫ2318.
[16] [16a]
[16b]
2
A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
P. J. Stev-
ens, F. A. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys.
Chem. 1994, 98, 11623.
Supporting Information (see also the footnote on the first page of
this article): Cartesian coordinates of optimised structures, energies
of single-point calculations.
[17]
R. J. Harrison, J. A. Nichols, T. P. Straatsma, M. Dupuis, E. J.
Bylaska, G. I. Fann, T. L. Windus, E. Apra, W. de Jong, S.
Hirata, M. T. Hackler, J. Anchell, D. Bernholdt, P. Borowski,
T. Clark, D. Clerc, H. Dachsel, M. Deegan, K. Dyall, D. El-
wood, H. Fruchtl, E. Glendening, M. Gutowski, K. Hirao, A.
Hess, J. Jaffe, B. Johnson, J. Ju, R. Kendall, R. Kobayashi, R.
Kutteh, Z. Lin, R. Littlefield, X. Long, B. Meng, T. Nakajima,
J. Nieplocha, S. Niu, M. Rosing, G. Sandrone, M. Stave, H.
Taylor, G. Thomas, J. van Lenthe, K. Wolinski, A. Wong, Z.
Zhang, NWChem, A Computational Chemistry Package for
Parallel Computers, version 4.1, Pacific Northwest National
Laboratory, Richland, Washington 99352-0999, USA, 2002.
A. Schäfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97,
Acknowledgments
We are grateful to Dr. Edoardo Apr a´ of the Pacific Northwest
National Laboratory for his help on issues relevant to the
NWChem software. A. D. is grateful to the Greek State Scholar-
ships Foundation for a postodoctoral fellowship (academic year
2001Ϫ2002). G. A. is grateful to the Committee of Research of the
National Technical University of Athens, Greece, for a doctoral
assistantship. We thank Ms. V. Skouridou (National Technical Uni-
versity of Athens, Biotechnology Laboratory) for recording the FT-
IR spectra.
[
[
[
18]
19]
20]
2
571.
A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100,
829.
5
T. Clark, J. Chandrasekhar, G. W. Spitznagel, P. V. Schleyer, J.
Comp. Chem. 1983, 4, 294.
[
[
[
1]
2]
3]
B. J. L. Royles, Chem. Rev. 1995, 95, 1981Ϫ2001.
[21]
B. H. Howard, H. Raistrick, Biochem. J. 1954, 57, 212.
A. Klamt, G. Schüürmann, J. Chem. Soc., Perkin Trans. 2
1993, 799.
22]
I. Casser, B. Steffan, W. Steglich, Angew. Chem. Int. Ed. Engl.
[
1
987, 26, 586Ϫ587.
T. P. Mulholland, R. Foster, D. B. Haydock, J. Chem. Soc.,
Perkin Trans. 1 1972, 2121Ϫ2128.
[
4]
H. Ikeda, N. Matsumori, M. Ono, A. Suzuki, A. Isogai, H.
Nagasawa, S. Sakuda, J. Org. Chem. 2000, 65, 438Ϫ444.
Received July 8, 2003
4600
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 4593Ϫ4600