Unprecedented directed oxidative cross-coupling of sulfahydantoins with aldehydes via a radical sulfonate-sulfinate conversion

Article abstract of DOI:10.1039/c2nj40294g

An unexpected C-C cross-coupling radical oxidation involving aldehydes and glycine enolate equivalents such as activated mesyl-sulfahydantoins leading to β,β'-disubstituted aspartate semialdehydes (ASA) instead of the expected threonine analogues was obse

Full text of DOI:10.1039/c2nj40294g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2012, 36, 1560–1563
www.rsc.org/njc
LETTER
Unprecedented directed oxidative cross-coupling of sulfahydantoins with
aldehydes via a radical sulfonate–sulfinate conversionw
Yoann Aubin,a Evelina Colacino,a Djamel Bouchouk,ab Isabelle Chataigner,c
a
a
a
Marıa del Mar Sanchez, Jean Martinez and Georges Dewynter*
´
´
Received (in Montpellier, France) 17th April 2012, Accepted 30th May 2012
DOI: 10.1039/c2nj40294g
An unexpected C–C cross-coupling radical oxidation involving
aldehydes and glycine enolate equivalents such as activated
mesyl-sulfahydantoins leading to b,b0-disubstituted aspartate
semialdehydes (ASA) instead of the expected threonine analogues
was observed and various a-substituted non-proteinogenic amino acid
analogues were synthesized. A radical mechanism was envisaged and
supported by DFT calculations.
Non-proteinogenic amino acids and their derived peptides are
important components of biological systems and attractive
targets in synthetic chemistry because of the diverse range of
physiological and therapeutic activities they display.1,2 The
sulfahydantoin (3-oxo-1,2,5-thiadiazole-1,1-dioxide) ring, a highly
effective peptidomimetic scaffold where the non-hydrolyzable
sulfonamide functionality can be exploited as a valuable
candidate for the replacement of the amido group, is an
Scheme 1 Modulation of the reactivity of sulfahydantoin rings.
emerging class of heterocycles in this respect. The sulfonamide
functionality proved to be selective for the inhibition of
proteases,3,4 it constitutes the aglycone part in pseudonucleoside
analogues,5 or the substructure in constrained peptides.6 On
the other hand, the reactions of glycine derivatives, via their
corresponding enolates, are some of the most versatile ways to
introduce functional groups at the a-position of carbonyl and
create a large variety of modified amino acids.7 In this context,
the sulfahydantoin 1, activated in the N-5 position by an
electron withdrawing group, can be considered a glycine
enolate equivalent, able to promote the formation of new
C–C bonds under basic conditions. We have recently reported7
that threonine sulfahydantoin analogues 3 could be obtained
in good yields by an ionic highly diastereoselective aldolisation
reaction of Boc-activated sulfahydantoin 1 with aldehydes in
the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as a
base (pathway A, Scheme 1). We report herein a ‘serendipitous’
discovery: when the N-Boc protecting group was replaced by a
methanesulfonyl (mesyl, Ms) group, an unexpected product was
formed. Instead of the aldolisation product, the aspartate
semialdehyde (ASA) derivative 4 was unequivocally obtained,
provided that an aldehyde with an a-hydrogen such as 2 was
used (pathway B, Scheme 1). The assembly of two carbonyl
subunits by their a-carbon, which is undocumented to date,
afforded 1,4-carbonyl derivatives through a direct oxidative
cross-coupling reaction.
We speculated that formation of the 1,4-dicarbonyl derivative 4
would involve the spontaneous formation of radical intermediates,
instead of ionic species. The radical directed oxidative
cross-condensation reaction described here, named with the
acronym ꢀdocc, allowed conversion of a sulfonate into a
sulfinate group.z The radical sulfonate–sulfinate reduction was
postulated to follow an oxidative coupling involving enolates.
This original reaction led to a direct access to b,b0-disubstituted
aspartate semialdehyde (ASA) derivatives 4, which are important
synthetic and biosynthetic precursors, involved in bacterial amino
acid and peptidoglycan biosynthesis.8 The ionic functionalization
of the a-position of amino acids has been extensively described.
In contrast, examples involving a radical C–C bond formation
a
´
Institut des Biomolecules Max Mousseron (IBMM)
UMR 5247 CNRS – UM I-UM II Universite de Montpellier II,
´
Place E. Bataillon, 34095 Montpellier Cedex 5, France.
E-mail: dewynter@univ-montp2.fr; Fax: +33 (0)467 144866;
Tel: +33 (0)467 144287
b LCOA Groupe de Chimie Bio-Organique, Universite´ Badji Mokhtar,
Annaba, Algeria
c
´
Universite de Rouen, COBRA - UMR 6014 CNRS, rue Tesniere,
76728 Mont Saint-Aignan, France
`
w Electronic supplementary information (ESI) available: Experimental
procedures, 1H and 13C spectra and DFT calculations. See DOI:
10.1039/c2nj40294g
c
1560 New J. Chem., 2012, 36, 1560–1563
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
Scheme 3 Proposed reaction mechanism for radical ꢀdocc reaction.
Scheme 2 Synthesis of N-mesylated sulfahydantoin 8.
not present. This aliphatic quaternary carbon coupled with the
aldehydic hydrogen (2J = 22.9 Hz), the six hydrogens of
geminal methyl groups and the hydrogen at the a-position of
glycine (2J = 3.7 Hz), clearly indicating that the direct
oxidative coupling was realized between these two sites. The
a-carbon of glycine coupled with its geminal hydrogen atom
(1J = 146.3 Hz) and with the exchangeable proton at 4.88 ppm,
confirming that the sulfahydantoin ring was not mesylated.13
On the other hand, the sulfonate, which was liberated during
the reaction pathway, has been detected by mass spectrometry
in its reduced form as a sulfinate DBU salt, displaying a peak at
m/z = 79 (FAB in negative mode).
mechanism are quite rare,9 and in most cases duplication products
are obtained.10,11 To our knowledge, only one example in which
glycine is alkylated via a visible light-induced intermolecular
radical coupling reaction is reported in the literature.12 In order
to explore in detail the potential of our findings, we synthesized
large quantities of N-mesylated sulfahydantoin 8. The synthesis of
N-benzyl-sulfamoyl methyl glycinate 6 was carried out in three
steps as previously described,7 via a trans sulfamoylation-ring
closure pathway from chlorosulfonyl isocyanate (CSI) 5 and
methyl glycinate (Scheme 2). Although the intramolecular
cyclization of 6 leading to sulfahydantoin 7 could not be
improved in terms of yield (53%) in comparison with the
previously published procedure,7 scale-up of 7 (5 g) was
possible when acetonitrile was used as the co-solvent, instead
of t-BuOH alone. N-5 mesylated sulfahydantoin 8 was
obtained in the presence of methanesulfonyl chloride in very
good yield after purification (Scheme 2).
Starting from these experimental data and based on DFT
calculations, we proposed that condensation proceeded
through a homolytic mechanism (Scheme 3). The sulfonate
group would play the role of the necessary oxidant with the
particularity of being initially linked to the substrate. We
proposed the reduction of the sulphur atom from oxidation
state +4 (in 12) to +2 (in 13) through a radical-promoted
pathway according to a general scheme in agreement with the
cross-condensation observed experimentally, leading to 4
(Scheme 3). To date, the involvement of a sulfonyl group in
such oxidative coupling has not been reported. However, the
At the beginning our goal was directed towards the aldolization
reaction of N-Boc or N-Ms protected sulfahydantoins. Then,
isobutyraldehyde was chosen as a model substrate and the
reaction was carried out in the presence of two equivalents of
DBU, in dichloromethane at 0 1C (Scheme 1, pathways A and B)
under alkaline conditions. To our surprise, when the substrate
was the N-mesylated ring 8, full conversion of the starting
material was observed within five minutes (instead of one night)7
and a single product was obtained in 81% yield after purification.
However, the careful interpretation of the 1H-NMR, 13C-NMR,
coupled 13C-NMR, COSY clearly indicated that the expected
aldolization product was not formed, but a new compound
ꢀ
implication of the PhSO2 radical was highlighted in oxime
substitutions.14
As originally proposed by Rathke and Lindert for the
oxidative coupling of carboxylates,15 the first step of this
transformation involved enolization of 8. The intramolecular
rearrangement of enolate 9 to 10 was followed by radical
extrusion via an homolytic cleavage leading to a persistent9
a-carbon-centered anion radical 11 and a sulfonyl radical 12.
The radical sulfonate 12 was reduced by an intermolecular
single electron transfer (SET inter) in the presence of the
aldehyde, to afford sulfinic acid 13, which formed a salt with
the second DBU equivalent. The hydrogen-transfer reaction
between the methanesulfonyl radical 12 and the aldehyde
proceeded through the so-called polarity-reversal catalysis16
mediated by MeSO2ꢀ species.17 Moreover, it was demonstrated18
that the unpaired electron in 12 was centered mainly on sulphur in
an orbital predominantly of 3p character, with a pyramidal
geometry with respect to the sulphur atom. The radical intermediate
14 collapsed with 11 by a radical–radical coupling mechanism
1
was obtained. The H NMR spectrum13 showed a singlet at
9.36 ppm characteristic of an aldehyde proton. The benzylic
protons became anisochrones (two doublets at 4.76 ppm) as
the consequence of creation of a chiral center at the glycine
a-position, suggesting that a new C–C bond was formed.
Moreover, the presence of a broad singlet at 4.88 ppm
characteristic of an exchangeable proton indicated that the
N-5 nitrogen atom was not mesylated anymore. By analysing
the coupling constants nJ (n 4 1) measured in the coupled
13C-NMR spectrum,13 we determined the presence (at 50 ppm)
of an aliphatic quaternary carbon adjacent to the aldehyde,
indicating that the hydrogen atom of the isopropyl group was
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 1560–1563 1561

View Article Online
affording the observed ꢀdocc product 4 (Scheme 3). To get
more insight into the mechanistic details of this coupling
reaction, DFT calculations were undertaken, using the
GAUSSIAN 09 program, at the B3LYP/6-31+G(d,p) or
UB3LYP/6-31+G(d,p) (for radical and radical anion species)
level of theory.19,20 Stationary points for optimized geometries
were determined to have zero imaginary vibrational frequencies.
In the gas phase, optimization of the geometry of model enolate A,
bearing a methyl group instead of a benzyl group on the N-2
nitrogen atom for calculation cost reasons, turned out to be
very difficult. In spite of many efforts, it always led to the
transfer of the sulfonyl group from N-5 to C-4. This apparent
instability of the enolate anion led us to consider the mechanistic
pathway involving a transfer such as the one proposed from 9 to
10 depicted in Scheme 3. It is well known that anions are quite
sensitive to the polarity of the reaction medium. This effect was
thus considered using the polarizable continuum model (PCM).
When taking into account the effect of the dichloromethane
dielectric constant, it became possible to optimize the geometry
of the enolate anion A, which indeed proved to be less stable
than sulfinate B (by 32.75 kcal molꢁ1) (Scheme 4). The
reaction pathway from A to B was evaluated by localization
of the transition state (TS). The energy profile suggests a two
step mechanism, involving, first, the dissociation of the enolate
and, then, the attack of the sulfinate on the carbon atom of the
enolate (see ESIw). As expected, homolytic cleavage of the
C–O bond yielding radical anion C and radical sulfonate 12
was endothermic but the global process, modelled with
acetaldehyde D, is thermodynamically favorable since the
anion F, resulting from the coupling of radical anion C and
radical E, was more stable. In contrast to this pathway,
aldolization reaction was not favorable since it led to a less
stable alkoxide G (Scheme 4).
leading to 15 was demonstrated by NMR and mass spectro-
metry in separate experiments. It clearly proved the formation
of radical species from 10 by a SET process. N-Mesylated
sulfahydantoin 8 was suspended into a solution of DBU, in
dry and previously degassed toluene under an argon atmo-
sphere. The postulated radical intermediate 11 collapsed by a
radical–radical coupling mechanism, leading to the duplica-
tion product 15 (Scheme 3). The 1H NMR analysis of the
duplication product 15 showed an AB system centered at
4.71 ppm for the diastereotopic benzyl protons and a sharp,
well defined singlet at 5.01 ppm for the two magnetically
equivalent a-CH protons of the glycine residue, allowing us to
conclude that the dimer was in the meso form.13 We extended our
‘serendipitous’ discovery to other different substrates bearing a
mobile hydrogen in the a-position with respect to an electron
withdrawing group (EWG) and susceptible to give radicals, when
ꢀ
tested in the docc reaction (Table 1).
As a general trend, in the aldehyde series better results were
obtained when the a-position had a tertiary carbon substituted
by electron rich alkyl chains (Table 1, entries 2, 4 and 5),
leading to the expected ꢀdocc products in a very short time and
good yields. However, linear aldehydes or electron withdrawing
groups on the tertiary carbon (Table 1, entries 3 and 6 respectively)
were completely unreactive. This could be explained by taking into
account the electronic nature of the thiyl radical, particularly
electrophilic and able to react with relatively high electron density
centers. This behaviour was particularly enhanced by the ability of
sulphur to use d orbitals to accommodate negative charges. Linear
or branched ketones proved to be completely unreactive even after
24 hours or under prolonged heating (Table 1, entries 7–9). In
contrast complex mixtures were obtained under the same reaction
conditions with tert-butylacrylate or acrylonitrile maybe due
to their instability in the presence of radical species (Table 1,
entries 10 and 11). In order to succeed in the ꢀdocc coupling with
the refractory substrates we speculate that dichloromethane, a
By performing the same type of calculations on the carbamate
protected sulfahydantoins, which experimentally led to a classical
aldol reaction, we found that the coupling pathway was not possible
in this case since transfer of the carboxyl group from N-2 to C-4
would imply the formation of a carbene, a much less stable species
(see ESIw). The dimerization of the radical sulfahydantoin 11
Table 1 Selected data for the direct oxidative cross-coupling reaction
(ꢀdocc)
Entry R1
R2
R3
Product Yielda,b (%)
1
2
3
4
5
6
7
8
CH3
CH3
CH3
H
c-Hex
CH3e,f
CH3
CH3
CH3
H
H
H
H
H
H
H
4a
4b
4c
4d
4e
4f
80
69c
0
58
85d
0
0
0
0
n.d.
n.d.
CH3CH2
CH3CH2CH2
—
Ph
BocNHe,f
CH3
CH3
CH(CH3)2 4g
Ph
CH3
OCH3
CN
4h
4i
4l
4m
9
10
11
CH3CH2
CH3
CH3
CH3
CH3
a
b
Isolated yields. The diasteroisomeric ratio (dr) was determined by
1H NMR. The diasteroisomeric ratio was 2 : 1. The diasteroisomeric
c
d
e
ratio was 1 : 1. Optically pure S-enantiomer. The substrate was
f
synthetized according to previously reported procedures.22
Scheme 4 Energetic diagram for the radical pathway.
c
1562 New J. Chem., 2012, 36, 1560–1563
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
CH2Cl2 (3 times). The organic layer was dried over anhydrous
Na2SO4, filtered and concentrated under reduced pressure. The residue
was quickly purified by column chromatography on silica gel (3g)
(gradient: AcOEt/CH2Cl2 (v/v), 5/95–20/80) to afford the title compounds
4a–b,d–e.
1 R. O. Duthaler, Tetrahedron, 1994, 50, 1539.
2 R. M. Williams, Adv. Asymmetric Synth., 1995, 1, 45.
3 W. C. Groutas, R. Kuang, R. Venkataraman, J. B. Epp, S. Ruan
and O. Prakash, Biochemistry, 1997, 36, 4739.
4 R. Kuang, J. B. Epp, S. Ruan, H. Yu, P. Huang, S. He, J. Tu,
N. M. Schechter, J. Turbov, C. J. Froelich and W. C. Groutas,
J. Am. Chem. Soc., 1999, 121, 8128.
Scheme 5 ASA analogues 4 to access more complex derivatives.
hydrogen donor solvent, could be involved in the radical
cascade, inhibiting the reaction. Therefore 1,2-dichloroethane,
suitable for radical process studies, was selected as substitutive
solvent and all the experiments were repeated under the same
conditions as before. Unfortunately, the results were not
improved, while increasing the temperature at reflux proved
to be detrimental: a new product was detected in the crude
mixture obtained from a side SN2 reaction between the N-5
position of sulfahydantoin and the solvent. The potential
access to a variety of polyfunctional non-proteinogenic and
unnatural amino acids using ASA and its derivatives has
already been described.21 From this perspective, compounds 4 are
important synthetic intermediates, as the aldehyde moiety can be
functionalized leading to more complex structures. The reactivity of
compound 4a was tested in the reduction of aldehyde functions
with NaBH4 to afford functionalized 1,4-diol 16 and in the Pinnick
oxidation,22,23 leading to 1,4-ketoacid 17 in good yields (Scheme 5).
In conclusion, it has been possible to show that reactivity of
the sulfonamide unit was governed by the nature of the
protecting group, allowing otherwise ‘impossible transformations’
such as direct functionalization of the a-position of amino acids.
We described a serendipitous synthesis of disubstituted aspartate
semialdehydes via an oxidative cross condensation of glycine
enolate equivalents. The elucidation of the radical mechanism
was supported by DFT calculations. Optimization of the
procedure and other mechanistic studies are in progress, as
well as the extension of the new methodology to more complex
synthetic goals leading to quaternary amino acids, which are
often difficult to synthesize by ionic reactions.
5 (a) N. Aouf, G. Dewynter and J.-L. Montero, Tetrahedron Lett.,
1991, 32, 6545; (b) G. Dewynter, N. Aouf, Z. Regaına and
J.-L. Montero, Tetrahedron, 1996, 52, 993.
¨
6 (a) S. Boudjabi, G. Dewynter, N. Voyer, L. Toupet and
J.-L. Montero, Eur. J. Org. Chem., 1999, 2275; (b) G. Dewynter,
N. Aouf, M. Criton and J. L. Montero, Tetrahedron, 1993, 49, 65.
7 (a) D. Bouchouk, E. Colacino, L. Toupet, N. Aouf, J. Martinez
and G. Dewynter, Tetrahedron Lett., 2009, 50, 1100;
(b) U. Schollkopf, Top. Curr. Chem., 1983, 109, 65;
(c) D. Seebach, E. Juaristi, D. D. Miller, C. Schickli and
T. Weber, Helv. Chim. Acta, 1987, 70, 237.
8 P. Meffre, Amino Acids, 1999, 16, 251–272.
9 C. J. Easton, Radicals Organic Synthesis, ed. P. Renaud and
M. P. Sibi, Wiley VCH, Mannheim, 2001, vol. 2, pp. 505–522.
:
k, B. Colmenero and M. L. Quiroga,
10 C. Alvarez-Ibarra, A. G. Csa
´
J. Org. Chem., 1997, 62, 2478.
:
11 A. G. Csak and J. Plumet, Chem. Soc. Rev., 2001, 30, 313.
12 M. Schwarzberg, J. Sperling and D. Elad, J. Am. Chem. Soc., 1973,
95, 6418.
13 Refer to spectral data in the ESIw.
14 S. Kim and I. Y. Lee, Tetrahedron Lett., 1998, 39, 1587.
15 M. W. Rathke and A. Lindert, J. Am. Chem. Soc., 1971, 93, 4605.
16 B. P. Roberts, Chem. Soc. Rev., 1999, 28, 25.
´
17 D. Markovic, A. Varela-Alvarez, J. A. Sordo and P. Vogel, J. Am.
Chem. Soc., 2006, 128, 7782.
18 A. G. Davies, B. P. Roberts and B. R. Sanderson, J. Chem. Soc.,
Perkin Trans. 2, 1973, 626.
19 G. W T. M. J. Frisch, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci,
G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, M. Klene, J. E. Knox, J. B. Cross, V. Bakken,
C. Adamo, J. Jaramillo, C. Pomelli, J. W. Ochterski,
R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
Acknowledgements
The authors are grateful to Aurelien LEBRUN (UM2, Labo-
ratoire de Mesures Physiques Montpellier, France) for the
helpful discussions in the interpretation of NMR spectra. The
authors thank the CRIHAN (Saint Etienne du Rouvray,
France) for generous allocation of computer time.
¨
O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, Gaussian, Inc, Wallingford, CT, 2009.
20 This DFT method has been shown to be reliable for similar organic
systems. See for instance: (a) I. Chataigner, C. Panel, H. Gerard
´
and S. R. Piettre, Chem. Commun., 2007, 3288; (b) Y. Zhou and
W. H. Nelson, J. Phys. Chem. A, 2011, 115, 11566;
(c) K. Makarova, E. V. Rokhina, E. A. Golovina, H. Van As
and J. Virkutyte, J. Phys. Chem. A, 2012, 116, 443.
21 H. Lusch and H. C. Uzar, Tetrahedron: Asymmetry, 2000,
11, 4965.
22 B. S. Bal, W. E. Childers and H. W. Pinnick, Tetrahedron, 1981,
37, 2091.
23 S. Jiang, P. Li, C. C. Lai, J. A. Kelley and P. P. Roller, J. Org.
Chem., 2006, 71, 7307.
Notes and references
z Typical experimental procedure for the radical couplings of 8 with
aldehydes (Table 1). To a solution containing Ms-sulfahydantoin 8
(1.65 mmol, 502 mg) and the suitable aldehyde (1.65 mmol, 502 mg) in
CH2Cl2 (4 mL), DBU (3.63 mmol, 550 mg) was slowly added
(1.82 mmol, 130 mg) at room temperature. The reaction mixture was
stirred for 2 hours, then the reaction mixture was quenched with dilute
HCl 0.1% (1 time) at 0 1C. The aqueous layer was extracted with
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 1560–1563 1563