5(4H)-oxazolones as intermediates in the carbodiimide- and cyanamide-promoted peptide activations in aqueous solution

Article abstract of DOI:10.1002/anie.201207730

The early days: Although considered a species to be avoided in peptide chemistry, the intermediacy of 5(4H)-oxazolones is demonstrated to be essential for the formation of peptides through cyanamide and carbodiimide activation in aqueous solution (see sch

Full text of DOI:10.1002/anie.201207730

Angewandte
Chemie
DOI: 10.1002/anie.201207730
Amino Acids
5(4H)-Oxazolones as Intermediates in the Carbodiimide- and
Cyanamide-Promoted Peptide Activations in Aqueous Solution**
Grꢀgoire Danger,* Arthur Michaut, Manon Bucchi, Laurent Boiteau, Justine Canal,
Raphaꢁl Plasson, and Robert Pascal*
The formation of biopolymers in prebiotic environments is
still unresolved regarding nucleic acids and functional pep-
tides. Peptide function especially requires proper folding and
hence sufficient lengths. In 1969, Cavadore and Previero[1]
observed that the EDC-mediated (EDC = 1-ethyl, 3-(3-dime-
thylaminopropyl)carbodiimide hydrochloride) a-amino acid
polymerization in water is significantly improved when N-
acylated amino acids are introduced as initiators. Obviously,
EDC is highly unlikely to have been abiotically formed but
there are indications in the literature that underivatized
= =
carbodiimide HN C NH is involved as an intermediate in
reactions of cyanamide,[2] a prebiotically plausible reagent,
and dicyandiamide is reported to behave similarly as carbo-
diimides.[3] The behavior of N-acylamino acids as polymeri-
zation initiators was explained in the original work[1] by an
inhibition of activation owing to the greater acidity of free a-
amino acids (pKA ꢀ 2.3) compared to C-terminal carboxy
groups in peptides (pKA ꢀ 3.7). However, alternative explan-
ations could be proposed. For example, 1) the blockade of
elongation by diketopiperazine formation at the dipeptide
level,[4] or 2) an increased efficiency of the activation process
resulting from formation of the 5(4H)-oxazolone 2 (Sche-
me 1).[5a]
The formation of 5(4H)-oxazolones has exhaustively been
studied owing to its importance for the chiral integrity of
synthetic peptides,[7] a result of the fast proton exchange at the
a-carbon atom in the presence of bases. Peptide chemists tend
therefore to avoid the formation of 5(4H)-oxazolones during
Scheme 1. Overactivation through cyclization by fast intramolecular
conversion of activated N-acyl-a-amino acids into 5(4H)-oxazolones.
The presence of the amide oxygen nucleophile at a convenient position
for reaction allows the fast intramolecular[6] formation of a reactive
5(4H)-oxazolone from the instable activating agent adduct.
a-amino acid activation. But, in the context of prebiotic
peptide formation starting from racemic mixtures, epimeriza-
tion may, on the contrary, constitute a prerequisite for
symmetry breaking as a result of appropriate processes
involving an autocatalytic reproduction of chirality.[5] The
formation of 5(4H)-oxazolones is additionally likely to
increase reaction rates since it corresponds to the principle
of overactivation through cyclization (Scheme 1), thus
expressing that the presence of a conveniently positioned
intramolecular group can allow the formation of a highly
reactive intermediate which would not have been formed
intermolecularly.[8a] The cyanate-mediated activation of pep-
tides with a C-terminal aspartyl residue[9] proceeds similarly
by cyclization, and amino acid N-carboxyanhydrides are
involved when activated a-amino acids are coupled at the
N terminus.[8] Herein we report the first results of studies
undertaken with the aims of 1) clearly identifying the
mechanism of the EDC-promoted activation of N-acylamino
acids, 2) demonstrating the possibility of activating peptides
with cyanamide in a similar way, and 3) addressing the issue of
chirality in these processes suspected to proceed through
a chirally unstable 5(4H)-oxazolone.[7]
As a model of the activation of C-terminal residues in
peptides, the behavior of N-benzoyl-alanine (1a) and N-
acetyl-alanine (1b; 10 mm) in the presence of EDC (20 mm)
was monitored in buffered D2O (pD 5–7) by NMR spectro-
scopy. From Bz-Ala-OH (1a), the observation of an NMR
signal at d = 1.47 ppm (Figures 1B–D) is consistent with the
formation of an intermediate, which is predominantly present
in a deuterated form, and consistent with the observation of
a fast H/D isotope exchange at the a-carbon atom from a pure
sample of 2-phenyl-4-methyl-5(4H)-oxazolone (2a) under
[*] Dr. G. Danger, A. Michaut
Spectromꢀtries et Dynamique Molꢀculaire
Physique des Interactions Ioniques et Molꢀculaires, UMR 7345
CNRS—Aix-Marseille Universitꢀ
Centre Saint-Jꢀrꢁme—case 252, 13397 Marseille, Cedex 20 (France)
E-mail: gregoire.danger@univ-amu.fr
A. Michaut, M. Bucchi, Dr. L. Boiteau, J. Canal, Dr. R. Pascal
Institut des Biomolꢀcules Max Mousseron, UMR 5247
CNRS—Universitꢀ Montpellier 1 & Montpellier 2, CC17006
Place E. Bataillon, 34095 Montpellier (France)
E-mail: rpascal@univ-montp2.fr
Dr. R. Plasson
Department of Earth and Planetary Sciences, Harvard University
100 Edwin H. Lane Bvd, Cambridge, MA 02142 (USA)
[**] We are indebted to the interdisciplinary program of the CNRS
Planetary Environments and Origins of Life (EPOV) for support, and
to the COST Action CM0703 “Systems chemistry” for providing the
possibility of fruitful scientific exchanges during the realization of
this work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201207730.
Angew. Chem. Int. Ed. 2013, 52, 611 –614
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
611

.
Angewandte
Communications
reaction with nucleophiles. The presence of an amide oxygen
atom at a convenient position for reaction enables the
irreversible cyclization into 5(4H)-oxazolone in spite of
a nucleophilic power that is not likely to exceed that of the
leaving urea group of the O-acylisourea intermediate 3
(Scheme 2). 5(4H)-Oxazolones have been considered as
Scheme 2. Conversion of the O-acylisourea intermediate 3 of the
carbodiimide-promoted peptide activation into 5(4H)-oxazolone 2: the
strength of the amide oxygen nucleophile and that of urea oxygen
leaving group are likely to be similar.
Figure 1. Deuteration experiments in D2O solutions buffered with MES
(100 mm, pD 5.5). 1H NMR spectra (400 MHz) of methyl protons of
1a and 2a. A) Fast deuteration of 2a (10 mm) in MES buffer (after
10 min at RT). B) Activation of 1a (10 mm) with EDC (20 mm);
reaction progress after 10 min. C) Reaction progress after 65 min.
D) Reaction progress after 369 min.
potential intermediates of peptide activation,[11] the present
results indeed demonstrate that in diluted aqueous solution,
the corresponding pathway is predominant.
The nonreversible conversion into 5(4H)-oxazolone
increases the concentration of species able to react with
nucleophiles with rates which must not be very different from
that of the O-acylurea, and results in an increase in rate by
a factor of several orders of magnitude. The kinetic advantage
of a-acylamino acid derivatives is likely to be the source of the
efficiency in the formation of peptide chains in the experi-
ment reported by Cavadore and Previero.[1] We determined
the length of the peptides formed by EDC activation in a CO2/
similar reaction conditions (Figure 1A). Structure 2a was
confirmed after isolation of the intermediate by extraction
(see the Supporting Information). A similar behavior was
observed for Ac-Ala-OH (1b), except that H/D exchange was
1
slower so that the appearance of the H NMR signal for 2,4-
dimethyl-5(4H)-oxazolone (2b) and the subsequent deutera-
tion could be observed to occur independently (see Figure S2
in the Supporting Information). These observations establish
that a chirally unstable 5(4H)-oxazolone is transiently formed
when activating 1a and 1b with EDC. The exchange of
a proton at the C-terminal residue was also observed during
the activation of the dipeptide Ac-Tyr-Ala-OH at pH 6.5, thus
leading to a moderate excess in favor of the l,l-diastereomer
(see Figure S3 in the Supporting Information) and indicating
that elongation into longer peptides also proceeds through
5(4H)-oxazolones.
À
HCO3 buffer. Bz-Ala-OH (1a; 5 mm) and glycine (50 mm)
were reacted with ten portions of EDC (5 mm each) over
three days. The peptide products were separated using
a column with a cation-exchange resin and analyzed by
NMR spectroscopy to assess a mean value of about three for
the degree of polymerization (see Figure S4, in the Support-
ing Information). The presence of peptides Bz-Ala-(Gly)n-
OH (with n = 1 to 11) was confirmed by ESI/MS and MALDI/
MS (see Figures S5 and S6 in the Supporting Information). A
control experiment carried out in a similar way but in which
the Bz-Ala-OH initiator was omitted demonstrated the
absence of diketopiperazine which should have been
formed from any polymerization process in the absence of
an initiator.
The prebiotic activation of peptides according to a similar
scheme requires that a prebiotically available reagent could
replace EDC. The hydration of cyanamide into urea is a slow
reaction at neutral pH and is subject to acid catalysis.[12]
Evidence that the reaction involves a prior conversion into
carbodiimide has been reported.[2,13] To determine if cyana-
mide could act as a carbodiimide equivalent or precursor for
the activation of carboxy groups in peptides, 1a was reacted
for periods of time ranging from 48 hours to 30 days in D2O in
the presence of NH2CN. To increase the reaction rates,
samples were heated to 808C. An exchange of the a-hydrogen
atom was detected by 1H NMR spectroscopy (Figure 2),
whereas no reaction occurred in a control experiment in
To demonstrate that the process is driven by overactiva-
tion through cyclization, we monitored, by HPLC, the
progress of the coupling reaction of 1b (krel = 1, pKA =
3.72[10]) with glycine p-nitroanilide (H-Gly-pNA) in the
presence of EDC at room temperature (see the Supporting
Information). The initial reaction rate was compared with
those of carboxylic acids unable to react intramolecularly:
acetic acid (krel = ca. 0.1, pKA = 4.76[10]) and lactic acid (krel
=
ca. 0.01, pKA = 3.86[10]), with the latter bearing a hydroxy
substituent having electron-withdrawing properties similar to
that of the acetamido group of 1b. The approximate 100-fold
difference in kinetic rates (and the higher coupling yields)
demonstrates that the reaction of 1a takes place predom-
inantly via the 5(4H)-oxazolone (ca. 99%; Scheme 1). In
contrast, carboxylic acids lacking any possibility of assistance
by a neighboring group would proceed through the unstable
O-acylisourea intermediate 3 which rapidly reverts to the
reactants by expelling the carboxylate nucleophile and does
not accumulate to an extent sufficient enough to allow a fast
612
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 611 –614

Angewandte
Chemie
activated species from the location of their formation to that
of the self-organizing system.[15] For these reasons, cyanamide
could be considered a prototype of a prebiotic activating
agent which is able to transfer energy to an environment
distant from that of its formation, and could be exhausted by
very specific dissipative structures possibly using catalytic
pathways and potentially capable of growing at its expense.
This simple five-atom molecule has been identified in
interstellar mediums[16] in which it can also be converted
into carbodiimide.[17] We found no indication that it could be
delivered as such in meteorites but processes, such as the UV
or sunlight irradiation of NH4CN solutions, leading to its
abiotic formation on Earth have been proposed.[18] Cyan-
amide, most often studied in the field of prebiotic chemistry as
a peptide condensing agent,[19] turned out to be very useful for
the abiotic synthesis of nucleotides[20] and phosphorylation[21]
as well. The involvement of a 5(4H)-oxazolone during peptide
activation may allow the coupling of hindered a-amino
acids[22,23] such as the Ca-methylated amino acids which are
present in enantiomeric excess in certain meteorites.[23,24] The
issues of epimerization of acyl amino acids and of C-terminal
residues in peptides during the EDC- and cyanamide-
promoted coupling of peptides is under investigation in our
group. However, since considering it as a drawback for
peptide coupling is not relevant in prebiotic chemistry,[5]
epimerization, associated with stereoselectivity in the elon-
gation pathway, may lead to the formation of substantial
amounts of homochiral domains in peptides which may be
essential for their activity. Protometabolic peptide systems,[5]
in which cyanamide-promoted peptide formation is associ-
ated with depolymerization by hydrolytic processes, could
potentially enable it to reach states far from equilibrium with
respect to chirality.[25] Such states are ruled by a dynamic
kinetic stability[26] rather than evolving towards the equilib-
rium state. They may have contributed to symmetry breaking
in a way, thus presenting analogies to the APED model
proposed earlier.[25]
Figure 2. Activation of 10 mm 1a with 20 mm cyanamide in buffered
D2O (100 mm MES buffer, pD 5.5) at 808C for 72 h, monitored by
1H NMR spectroscopy (400 MHz). A) The dissymmetry of the methyl
doublet of 1a reveals the formation of a singlet at d=1.37 ppm
resulting from H/D exchange at Ca. B) Same reaction in the presence
of Gly (50 mm) after 72 h: the resonance at d=1.43 ppm, consistent
with the formation of 4a in a deuterated form and the symmetry of the
methyl doublet show that the 5(4H)-oxazolone reacts with glycine
faster than it is reverted into 1a.
which cyanamide was replaced by its hydrolysis product, urea.
The selective deuteration at a single site was confirmed by MS
analysis of 1a recovered from the reaction medium (see
Figure S7 in the Supporting Information).
A similar experiment was carried out in D2O with 1a in
the presence of 50 mm glycine and 20 mm cyanamide
(Figure 2), thus leading to the observation of a singlet at d =
1.43 ppm, which is consistent with the presence of an addi-
tional deuterated product in significant concentrations and
with the chemical shift determined independently for Bz-Ala-
Gly-OH (4a). The dipeptide 4a, formed in a 7.5% yield from
a 10 mm solution of 1a in a MES buffer at pH 5.5 in the
presence of 50 mm glycine and 33 mm cyanamide heated to
808C for 17 days, was subsequently identified by HPLC/ESI/
MS (see Figure S8 in the Supporting Information). No
peptide was observed from similar experiments carried out
at a higher pH value (6.5), which is consistent with the acid
catalysis of cyanamide reaction.[12]
These results show that cyanamide behaves as an activat-
ing agent for N-acyl a-amino acids and strongly suggest the
formation of a 5(4H)-oxazolone undergoing a fast exchange
of the proton bound to the a-carbon atom. They are
consistent with the fact that the reaction of cyanamide with
water and other nucleophiles proceeds through a carbodi-
imide[2,12,13] and is subject to acid catalysis.[12] This slow
reaction may at first glance be considered as a drawback, thus
rendering its participation to the formation of prebiotic
peptides unlikely. However, kinetic barriers have been
considered essential to hold chemical environments far from
equilibrium and therefore to bring about the development of
protometabolic pathways based on catalysis and autocataly-
sis.[14] Hydrolytic processes and side reactions must then
proceed with time scales consistent with the migration of
Received: September 25, 2012
Revised: October 18, 2012
Published online: November 21, 2012
Keywords: amino acids · chirality · NMR spectroscopy ·
.
peptides · reactive intermediates
[1] J.-C. Cavadore, A. Previero, Bull. Soc. Chim. Biol. 1969, 51,
1245 – 1253.
[2] S. V. Hill, A. Williams, J. L. Longridge, J. Chem. Soc. Perkin
Trans. 2 1984, 1009 – 1013.
[3] a) A. Brack, Pure Appl. Chem. 1993, 65, 1141 – 1151; b) A.
Brack, Chem. Biodiversity 2007, 4, 665 – 679.
[4] S. Pascual, D. M. Haddleton, D. M. Heywood, E. Khoshdel, Eur.
Polym. J. 2003, 39, 1559 – 1565.
[5] a) G. Danger, R. Plasson, R. Pascal, Chem. Soc. Rev. 2012, 41,
5416 – 5429; b) G. Danger, R. Plasson, R. Pascal, Astrobiology
2010, 10, 651 – 662.
[6] M. I. Page, W. P. Jencks, Proc. Natl. Acad. Sci. USA 1971, 68,
1678 – 1683.
[7] a) N. L. Benoiton, Biopolymers 1996, 40, 245 – 254; b) Chemistry
of Peptide Synthesis CRC Press, Boca Raton, FL, 2005.
Angew. Chem. Int. Ed. 2013, 52, 611 –614
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
613

.
Angewandte
Communications
[8] a) R. Pascal, L. Boiteau, A. Commeyras, Top. Curr. Chem. 2005,
259, 69 – 122; b) G. Danger, L. Boiteau, H. Cottet, R. Pascal, J.
Am. Chem. Soc. 2006, 128, 7412 – 7413.
[9] G. Danger, S. Charlot, L. Boiteau, R. Pascal, Amino Acids 2012,
42, 2331 – 2341.
[18] R. Lohrmann, J. Mol. Evol. 1972, 1, 263 – 269.
[19] a) C. Ponnamperuma, E. Peterson, Science 1965, 147, 1572 –
1574; b) D. W. Nooner, E. Sherwood, M. A. More, J. Oro, J.
Mol. Evol. 1977, 10, 211 – 220; c) J. R. Hawker, J. Oro, J. Mol.
Evol. 1981, 17, 285 – 294.
[10] G. Kortꢀm, W. Vogel, K. Andrussow, Dissociation Constants of
Organic Acids in Aqueous Solution, Butterworths, London,
1961.
[11] M. Goodman, W. J. McGahren, Tetrahedron 1967, 23, 2031 –
2050.
[20] a) M. W. Powner, B. Gerland, J. D. Sutherland, Nature 2009, 459,
239 – 242, and references therein; b) J. D. Sutherland, Cold
Spring Harb. Perspect. Biol. 2010, 2, a005439; c) M. W. Powner,
J. D. Sutherland, J. Szostak, J. Am. Chem. Soc. 2010, 132, 16677 –
16688.
[12] a) M. J. Sullivan, M. L. Kilpatrick, J. Am. Chem. Soc. 1945, 67,
1815 – 1823; b) M. L. Kilpatrick, J. Am. Chem. Soc. 1947, 69, 40 –
46.
[21] R. Lohrmann, L. E. Orgel, Science 1968, 161, 64 – 66.
[22] Z. Z. Brown, C. E. Schafmeister, J. Am. Chem. Soc. 2008, 130,
14382 – 14383.
[13] F. Tordini, A. Bencini, M. Bruschi, L. De Gioia, G. Zampella, P.
Fantucci, J. Phys. Chem. A 2003, 107, 1188 – 1196.
[14] A. Eschenmoser, Origins Life Evol. Biosphere 1994, 24, 389 –
423.
[15] a) R. Pascal, L. Boiteau, Philos. Trans. R. Soc. London Ser. B
2011, 366, 2949 – 2958; b) R. Pascal, J. Syst. Chem. 2012, 3, 3.
[16] P. Thaddeus, Philos. Trans. R. Soc. London Ser. B 2006, 361,
1681 – 1687.
[23] M. Crisma, A. Moretto, F. Formaggio, B. Kaptein, Q. B.
Broxterman, C. Toniolo, Angew. Chem. 2004, 116, 6863 – 6867;
Angew. Chem. Int. Ed. 2004, 43, 6695 – 6699.
[24] a) S. Pizzarello, E. Shock, Cold Spring Harb. Perspect. Biol. 2010,
2, a002105; b) A. S. Burton, J. C. Stern, J. E. Elsila, D. P. Glavin,
J. P. Dworkin, Chem. Soc. Rev. 2012, 41, 5459 – 5472.
[25] R. Plasson, H. Bersini, A. Commeyras, Proc. Natl. Acad. Sci.
USA 2004, 101, 16733 – 16738.
[17] F. Duvernay, T. Chiavassa, F. Borget, J.-P. Aycard, J. Am. Chem.
Soc. 2004, 126, 7772 – 7773.
[26] A. Pross, J. Syst. Chem. 2011, 2, 1.
614
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 611 –614