Peptide Mechanosynthesis by Direct Coupling of N-Protected α-Amino Acids with Amino Esters

Article abstract of DOI:10.1002/ejoc.201600617

In view of developing alternatives to classical peptide synthesis strategies that suffer from low efficacy and negative environmental impact, the reactivity of N-protected α-amino acids, amino esters, and N-ethyl-N′-(3-dimethylaminopropyl)carbodiimide was

Full text of DOI:10.1002/ejoc.201600617

DOI: 10.1002/ejoc.201600617
Communication
Peptide Synthesis
Peptide Mechanosynthesis by Direct Coupling of N-Protected
α-Amino Acids with Amino Esters
Vincent Porte,[a] Marion Thioloy,[a] Titouan Pigoux,[a] Thomas-Xavier Métro*[a]
Jean Martinez,[a] and Frédéric Lamaty*[a]
Abstract: In view of developing alternatives to classical pept- wide range of dipeptides that could be produced up to the
ide synthesis strategies that suffer from low efficacy and nega-
tive environmental impact, the reactivity of N-protected α-
amino acids, amino esters, and N-ethyl-N′-(3-dimethylamino-
propyl)carbodiimide was studied under liquid-assisted grinding
(LAG) conditions. The optimal reaction conditions enabled the
intensive and environmentally benign mechanosynthesis of a
multigram scale. The critical influence of the nature of the liquid
additive on the course of the reaction was revealed. Combined
with solvent-free tert-butoxycarbonyl deprotection steps, this
approach was applied to the synthesis of tri- and tetrapeptides,
thus opening the way to the synthesis of much longer peptides.
Results and Discussion
Introduction
Continuing our efforts to provide the organic chemist with envi-
ronmentally friendly and easily scalable methods,[9] we envi-
sioned to study the mechanically mediated coupling of non-
activated α-amino acids with amino ester salts under elevated
milling-load conditions to implement an intensive and environ-
mentally sound peptide synthesis approach. Preliminary screen-
ing of coupling agents [e.g., N-ethyl-N′-(3-dimethylaminoprop-
yl)carbodiimide (EDC), (1-cyano-2-ethoxy-2-oxoethylidenamino-
Peptides are a promising class of drug candidates owing to
their high efficiency, selectivity, and biodegradability.[1] Unfor-
tunately, their classical ways of production in solution or on
solid support require huge amounts of volatile and toxic or-
ganic solvents (e.g., DMF, CH2Cl2, and 1-methylpyrrolidin-2-one)
and bases (e.g., Et3N, N,N-diisopropylethylamine, and piper-
idine).[2] To lower the environmental impact of peptide synthe-
sis, several strategies have been developed, including those in-
volving the use of water as the solvent,[3] the use of microwave
heating under solvent-free conditions,[4] the design of coupling
reagents with improved safety profile,[5] and the use of continu-
ous-flow conditions.[6] In addition to these strategies, we devel-
oped an original approach based on the use of ball-milling
technology.[7] Ball milling in organic synthesis has regained in-
terest in recent years owing to the possibility, among others, to
avoid the use of problematic organic solvents, to reduce the
time of the reaction, and to isolate otherwise elusive reaction
intermediates.[8] Our latest developments in the utilization of
ball-milling technology applied to the synthesis of peptides was
based on the utilization of activated N-protected α-amino es-
ters (e.g., N-carboxyanhydrides or hydroxysuccinimide esters),[7]
but the low commercial availability of these chemicals ham-
pered the scope of the approach.
oxy)dimethylaminomorpholinocarbenium
hexafluorophos-
phate (COMU®), N,N′-diisopropylcarbodiimide (DIC), 1,1′-carb-
onyldiimidazole (CDI)] revealed the superiority of EDC, which
was confirmed by the work of Štrukil and co-workers.[10] Indeed,
they utilized EDC·HCl to produce dipeptides in a ball mill by
coupling α-amino acids with α-amino esters. Unfortunately, this
approach was limited to the synthesis of four easily accessible
dipeptides on a small scale (<100 mg of isolated dipeptide) and
low milling load (ML: <83 mg mL–1).[11] In addition, the reaction
required an extended length of time (3 h) and the use of corro-
sive and toxic 4-(dimethylamino)pyridine (DMAP) as the base
and explosive nitromethane as the liquid additive. Similarly, Du-
angkamol and co-workers described the synthesis of dipeptides
on a small scale (ca. 100 mg) by grinding N-protected α-amino
acids with cyanuric chloride and CH2Cl2.[12] Yet, both of these
chemicals present highly undesirable safety profiles that are not
compatible with an efficient and sustainable strategy for pept-
ide synthesis.[13] For our part, we selected the difficult coupling
of two valine residues as a first model reaction in our quest for
optimal reaction conditions. Thus, pTosOH·H-Val-OBn (pTosOH =
p-toluenesulfonic acid), Z-Val-OH (1.2 equiv.) (Z = benzyloxy-
carbonyl), NaH2PO4 (4.0 equiv.), and Oxyma (ethyl hydroxy-
iminocyanoacetate, 1.2 equiv.) were introduced in a stainless
steel (SS) grinding jar with one polytetrafluoroethylene (PTFE)
ball and, if needed, a liquid additive (Figure 1).
[a] Institut des Biomolécules Max Mousseron (IBMM), UMR 5247, Université de
Montpellier, CNRS, ENSCM,
Place Eugène Bataillon,
34095 Montpellier cedex 5, France
E-mail: thomas-xavier.metro@umontpellier.fr
frederic.lamaty@umontpellier.fr
http://www.greenchem.univ-montp2.fr
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under http://dx.doi.org/10.1002/ejoc.201600617.
Eur. J. Org. Chem. 2016, 3505–3508
3505
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
of chemical homogeneity. Without any liquid additive, the aver-
age conversion was low (60 %) and the standard deviation was
very high (38 %), which indicated very low homogeneity (Fig-
ure 1). The utilization of a liquid additive was then envisaged
to improve the homogeneity of the mixture. Liquid additives
have been widely used in recent years,[16] and yet the influence
of their chemical nature on the course of a reaction has been
studied in a very limited number of cases.[17] First, the addition
of apolar, aprotic dodecane (400 μL) during the coupling led to
a low conversion (55 %, Figure 1). The use of a larger volume
of dodecane or a ball of higher density (i.e., SS) did not improve
significantly the course of the reaction. Then, highly polar and
protic liquids such as water, glycerol, and polyethylene glycol
(PEG)-300 were envisaged to promote the reaction, but unfortu-
nately, our attempts were met with no success. Polar and apro-
tic liquid additives were then evaluated: DMF furnished both
an excellent average conversion (96 %) and a standard devia-
tion (2 %). Unfortunately, DMF is a highly problematic sol-
vent,[18] for which other solvents with preferable safety profiles
have been identified.[19] Two of them, γ-valerolactone (GVL) and
EtOAc, were envisaged under our conditions. To our delight,
100 % average conversion was obtained upon using GVL as the
liquid additive along with a SS ball (Figure 1). Similarly, excellent
average conversion (93 %) and standard deviation (3 %) were
obtained upon using EtOAc as the liquid additive. Unlike GVL,
the low flash point of EtOAc (–3 °C) led us to perform this
reaction with a PTFE ball under an inert atmosphere (N2) to
prevent the formation of sparks (see above). These results are
a clear indication that liquid-assisted grinding (LAG) is a very
powerful strategy for the development of environmentally
sound organic syntheses, as it allows merging of the best of a
solvent-free synthesis (high speed of reaction) with the best of a
solution-based synthesis (absence of mass-transfer limitations)
while avoiding their respective drawbacks (mass-transfer limita-
Figure 1. Liquid additive influence on average conversion and chemical
homogeneity during the mechanosynthesis of Z-Val-Val-OBn.
Oxyma was chosen as the epimerization suppressant owing
to its advantageous safety profile.[14,15] The use of a ball made
of PTFE instead of SS was decided to avoid sparks that could
arise from shocks between the metallic ball and the jar, and
this allowed the scope of the reagents to be extended to those
with very low flash points. The mixture was grinded for 2 min
in a vibrating ball mill (vbm), EDC (1.2 equiv.) was added, and
the jar was closed and agitated for another 2 min. To evaluate
the chemical homogeneity inside the jar, the mixture was sam-
pled in three different spots. Each of these samples was sub-
jected to HPLC analysis to determine the average conversion. tions for solvent-free synthesis; slow kinetics and toxic solvents
The standard deviation was calculated to serve as an indicator for solution-based synthesis).
Table 1. Examples of dipeptide synthesis by mechanochemistry.
Entry
N-Protected amino acid
Amino ester
Liquid additive
Dipeptide
Yield[a] [%]
1
2
3
4
5
6
7
8
Z-Val-OH
pTosOH·H-Val-OBn
HCl·H-Leu-OMe
HCl·H-Gly-OMe
HCl·H-Phe-OMe
pTosOH·H-Leu-OBn
pTosOH·H-Leu-OBn
HCl·H-Ala-OMe
HCl·H-Ile-OMe
GVL
GVL
Z-Val-Val-OBn
73
Boc-Phe-OH
Boc-Phe-OH
Boc-Phe-OH
Boc-Leu-OH·H2O
Boc-Leu-OH·H2O
Boc-Ala-OH
Boc-Ile-OH·0.5H2O
Z-Aib-OH
Z-Aib-OH
Boc-Phe-Leu-OMe
Boc-Phe-Gly-OMe
Boc-Phe-Phe-OMe
Boc-Leu-Leu-OBn
Boc-Leu-Leu-OBn
Boc-Ala-Ala-OMe
Boc-Ile-Ile-OMe
Z-Aib-Val-OMe
62[b]
91[c] (66)[d]
77[c]
70 (88)[d]
69[e]
63
EtOAc
EtOAc
EtOAc
PEG-300
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
77
9
HCl·H-Val-OMe
89[f]
10
11
12
HCl·H-Aib-OMe
DBS·H-Pro-OtBu[g]
HCl·H-Leu-OMe
Z-Aib-Aib-OMe
Z-Aib-Pro-OtBu
Boc-Arg(NO2)-Leu-OMe
73
Z-Aib-OH
Boc-Arg(NO2)-OH
90[h]
88[c]
[a] Yield of isolated product. ML = 105 mg mL–1. [b] >99 % de. [c] EDC (1.7 equiv.), 15 min agitation. [d] ML = 432.5 mg mL–1, 25 mL PTFE reactor, 20 mm
PTFE ball, 5 min agitation. [e] 60 min agitation. [f] EDC (1.8 equiv.), 20 min agitation. [g] DBS = dibenzenesulfonamide. [h] EDC (1.8 equiv.), 10 min agitation.
Eur. J. Org. Chem. 2016, 3505–3508
www.eurjoc.org
3506
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Table 2. Tri- and tetrapeptides mechanosynthesis.
Entry
Boc amino acid
Amino ester
Product
Yield[a] [%]
1
2
3
4
5
6
7
8
Boc-Leu-OH·H2O
Boc-Arg(Tos)-OH
Boc-Glu(OCy)-OH
Boc-Trp-OH
Boc-Ala-OH
Boc-Phe-OH
HCl·H-(Leu)2-OBn
HCl·H-(Leu)2-OBn
HCl·H-(Leu)2-OBn
HCl·H-(Leu)2-OBn
HCl·H-Phe-Gly-OMe
HCl·H-Phe-Gly-OMe
HCl·H-(Leu)3-OBn
HCl·H-(Leu)3-OBn
Boc-(Leu)3-OBn
65
65
51
47
52
Boc-Arg(Tos)-(Leu)2-OBn
Boc-Glu(OCy)-(Leu)2-OBn
Boc-Trp-(Leu)2-OBn
Boc-Ala-Phe-Gly-OMe
Boc-Phe-Phe-Gly-OMe
Boc-Pro-(Leu)3-OBn
Boc-(Leu)4-OBn
57
Boc-Pro-OH
Boc-Leu-OH·H2O
74
91[b]
[a] Yield of isolated product. [b] EDC (1.8 equiv.), 25 min agitation.
Upon using the best conditions (GVL along with a SS ball), tion of diketopiperazines was not detected by LC–MS analysis
Z-Val-Val-OBn was isolated in 73 % yield after only 10 min ball of the crude materials. Finally, upon treating the tripeptide
milling, classical acid–base extractions, followed by chromatog- HCl·H-(Leu)3-OBn with Boc-Pro-OH and Boc-Leu-OH·H2O, the
raphy on silica gel to eliminate GVL (B.p. 207–208 °C) (Table 1, tetrapeptides Boc-Pro-(Leu)3-OBn and Boc-(Leu)4-OBn were pro-
entry 1). Similarly, Boc-Phe-Leu-OMe was produced in 62 % duced in excellent yields of 74 and 91 %, respectively (Table 2,
yield with >99 % diastereomeric excess, as determined by HPLC entries 7 and 8), which clearly indicated the high potential of
analysis on a chiral stationary phase (Table 1, entry 2).
this approach to produce even longer peptides.
Once again, the use of GVL made purification on column
chromatography inevitable. In contrast, the use of EtOAc for
the coupling of Boc-Phe-OH with HCl·H-Gly-OMe furnished Boc-
Phe-Gly-OMe in a very good yield of 91 % without requiring
purification on silica gel (Table 1, entry 3). Upon performing the
reaction at a much higher milling load (ML = 432.5 mg mL–1),
this reaction furnished 1.5 g of Boc-Phe-Gly-OMe in 66 % yield
(Table 1, entry 3). The same strategy enabled the production of
2.2 g of Boc-Leu-Leu-OBn in 88 %, providing enough material
to envisage synthesis of much longer peptides by using this
approach (Table 1, entry 5). Other nonpolar dipeptides such as
Boc-Ala-Ala-OMe and Boc-Ile-Ile-OMe were isolated in yields of
63 and 77 %, respectively (Table 1, entries 7 and 8). Further-
more, the production of Z-Aib-Val-OMe and Z-Aib-Aib-OMe in
yields of 89 and 73 %, respectively, showed the efficiency of
this approach, even for highly hindered substrates (Table 1, en-
tries 9 and 10). These reaction conditions are more sustainable
and much simpler to implement than previous strategies util-
ized for coupling two Aib residues.[20] A secondary amine and
a highly polar amino acid were also used as substrates, and Z-
Aib-Pro-OtBu and Boc-Arg(NO2)-Leu-OMe were isolated in
yields of 90 and 88 %, respectively (Table 1, entries 11 and 12).
Of note, the excess amounts of reactants required for the syn-
thesis of all the peptides described herein are generally much
lower than those in classical strategies. To produce longer pept-
ides, Boc-Leu-Leu-OBn and Boc-Phe-Gly-OMe were quantita-
tively deprotected by gaseous HCl treatment without solvent.[7]
Then, treatment of HCl·H-(Leu)2-OBn with Boc-Leu-OH·H2O and
Boc-Arg(Tos)-OH furnished the corresponding tripeptides in a
satisfying yields of 65 % (Table 2, entries 1 and 2). Similarly, Boc-
Glu(OCy)-(Leu)2-OBn (Cy = cyclohexyl) and Boc-Trp-(Leu)2-OBn
were isolated, yet in moderate yields of 51 and 47 %, respec-
tively (Table 2, entries 3 and 4). Notably, the undesirable forma-
Conclusion
In summary, liquid-assisted grinding enabled the intensive, scal-
able, and environmentally benign production of various di-
peptides by direct coupling of non-activated N-protected α-
amino acids with amino esters mediated by EDC. The success
of this strategy relied on the association of ball milling, a judi-
ciously selected liquid additive, and reagents with reduced
safety profiles. Combined with solvent-free deprotection steps,
tri- and tetrapeptides could also be synthesized in good to ex-
cellent yields. Further applications of this liquid-assisted grind-
ing strategy to the synthesis of longer peptides are currently
underway.
Acknowledgments
The authors thank the Centre National de la Recherche Scientif-
ique (CNRS) and Université de Montpellier for financial support.
François Métro is warmly acknowledged for producing graphi-
cal abstract artwork.
Keywords: Mechanochemistry · Ball milling · Green
chemistry · Amino acids · Peptides
[1] K. Fosgerau, T. Hoffmann, Drug Discovery Today 2015, 20, 122.
[2] Y. E. Jad, G. A. Acosta, S. N. Khattab, B. G. de la Torre, T. Govender, H. G.
Kruger, A. El-Faham, F. Albericio, Org. Biomol. Chem. 2015, 13, 2393.
[3] a) C. M. Gabriel, M. Keener, F. Gallou, B. H. Lipshutz, Org. Lett. 2015, 17,
3968; b) R. De Marco, A. Tolomelli, A. Greco, L. Gentilucci, ACS Sustainable
Chem. Eng. 2013, 1, 566.
[4] A. Mahindra, N. Patel, N. Bagra, R. Jain, RSC Adv. 2014, 4, 3065.
Eur. J. Org. Chem. 2016, 3505–3508
www.eurjoc.org
3507
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[5] a) D. Dev, N. B. Palakurthy, K. Thalluri, J. Chandra, B. Mandal, J. Org. Chem.
2014, 79, 5420; b) A. El-Faham, R. S. Funosas, R. Prohens, F. Albericio,
Chem. Eur. J. 2009, 15, 9404.
[13]
CH2Cl2 is classified as toxic and carcinogenic, and cyanuric chloride is
classified as toxic, corrosive, and a skin sensitizer according to Regulation
(EC) No 1272/2008.
[6] S. Fuse, Y. Mifune, T. Takahashi, Angew. Chem. Int. Ed. 2014, 53, 851;
Angew. Chem. 2014, 126, 870.
[7] a) J. Bonnamour, T.-X. Métro, J. Martinez, F. Lamaty, Green Chem. 2013,
15, 1116; b) V. Declerck, P. Nun, J. Martinez, F. Lamaty, Angew. Chem. Int.
Ed. 2009, 48, 9318; Angew. Chem. 2009, 121, 9482.
[8] a) S. L. James, C. J. Adams, C. Bolm, D. Braga, P. Collier, T. Friščić, F. Gre-
pioni, K. D. M. Harris, G. Hyett, W. Jones, A. Krebs, J. Mack, L. Maini, A. G.
Orpen, I. P. Parkin, W. C. Shearouse, J. W. Steed, D. C. Waddell, Chem. Soc.
Rev. 2012, 41, 413; b) V. Štrukil, D. Gracin, O. V. Magdysyuk, R. E. Dinne-
bier, T. Friščić, Angew. Chem. Int. Ed. 2015, 54, 8440; Angew. Chem. 2015,
127, 8560.
[9] a) T.-X. Métro, X. J. Salom-Roig, M. Reverte, J. Martinez, F. Lamaty, Green
Chem. 2015, 17, 204; b) T.-X. Métro, J. Bonnamour, T. Reidon, A. Duprez,
J. Sarpoulet, J. Martinez, F. Lamaty, Chem. Eur. J. 2015, 21, 12787; c) A.
Beillard, E. Golliard, V. Gillet, X. Bantreil, T.-X. Métro, J. Martinez, F. Lamaty,
Chem. Eur. J. 2015, 21, 17614; d) E. Colacino, P. Nun, F. M. Colacino, J.
Martinez, F. Lamaty, Tetrahedron 2008, 64, 5569.
[14]
[15]
Oxyma is not a hazardous substance according to Regulation (EC) No.
1272/2008.
R. Subirós-Funosas, R. Prohens, R. Barbas, A. El-Faham, F. Albericio, Chem.
Eur. J. 2009, 15, 9394.
G. A. Bowmaker, Chem. Commun. 2013, 49, 334.
a) T. Friščić, S. L. Childs, S. A. A. Rizvi, W. Jones, CrystEngComm 2009, 11,
418; b) T. Friščić, A. V. Trask, W. D. S. Motherwell, W. Jones, Cryst. Growth
Des. 2008, 8, 1605; c) M. Tireli, M. Juribašić Kulcsár, N. Cindro, D. Gracin,
N. Biliškov, M. Borovina, M. Ćurić, I. Halasz, K. Užarević, Chem. Commun.
2015, 51, 8058.
[16]
[17]
[18] DMF is classified as presenting reproductive toxicity according to Regula-
tion (EC) No 1272/2008.
[19] D. Prat, A. Wells, J. Hayler, H. Sneddon, C. R. McElroy, S. Abou-Shehada,
P. J. Dunn, Green Chem. 2016, 18, 288, and references cited therein.
[20] a) K. G. Jastrzabek, R. Subiros-Funosas, F. Albericio, B. Kolesinska, Z. J.
Kaminski, J. Org. Chem. 2011, 76, 4506; b) J. Clayden, A. Castellanos, J.
Solà, G. A. Morris, Angew. Chem. Int. Ed. 2009, 48, 5962; Angew. Chem.
2009, 121, 6076; c) P. Rossi, F. Felluga, P. Tecilla, F. Formaggio, M. Crisma,
C. Toniolo, P. Scrimin, J. Am. Chem. Soc. 1999, 121, 6948.
[10] a) V. Štrukil, B. Bartolec, T. Portada, I. Đilović, I. Halasz, D. Margetić, Chem.
Commun. 2012, 48, 12100; b) D. Tan, V. Štrukil, C. Mottillo, T. Friščić,
Chem. Commun. 2014, 50, 5248.
[11] Milling load (ML) is defined as the sum of the mass of the reactants per
free volume in the jar.
[12] C. Duangkamol, S. Jaita, S. Wangngae, W. Phakhodee, M. Pattarawarapan,
RSC Adv. 2015, 5, 52624.
Received: May 19, 2016
Published Online: June 27, 2016
Eur. J. Org. Chem. 2016, 3505–3508
www.eurjoc.org
3508
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim