Cyclic dipeptides: Catalyst/promoter-free, rapid and environmentally benign cyclization of free amino acids

Article abstract of DOI:10.1039/c1gc15043j

"The best catalyst is no catalyst." With growing public concern over global warming and the amount of greenhouse gases, it is important to reduce the amount of chemicals and eliminate waste, to obtain better results in a simple, selective, safe, and environmentally benign fashion compared to conventional tedious chemical synthesis. Herein, we disclose an environmentally benign, rapid, catalyst/promoter/coupling reagent-free cyclization procedure of free amino acids to furnish exclusively cyclic dipeptides (2,5- diketopiperazines, DKPs) in excellent or even quantitative yield, along with their solid state self-assembling properties. This process is extremely simple and highly efficient with little or no traditional synthetic skills and without any chromatographic purification. Synthesis of structurally diverse DKPs has been achieved with a dramatic decrease in the reaction time, the amount/number of solvents used, a significant increase in the yield and nearly complete elimination of waste. As a result, this is an excellent example for the environmentally benign, clean and green chemistry concept. The most exciting outcome of our investigation is an unusual case of chiral self-recognition encountered upon the cyclization of rac-pipecolic acid, which resulted in the formation of the meso-product exclusively.

Full text of DOI:10.1039/c1gc15043j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 1203
www.rsc.org/greenchem
PAPER
Cyclic dipeptides: catalyst/promoter-free, rapid and environmentally benign
cyclization of free amino acids†
Nonappa,* Kari Ahonen, Manu Lahtinen* and Erkki Kolehmainen*
Received 12th January 2011, Accepted 21st February 2011
DOI: 10.1039/c1gc15043j
“
The best catalyst is no catalyst.” With growing public concern over global warming and the
amount of greenhouse gases, it is important to reduce the amount of chemicals and eliminate
waste, to obtain better results in a simple, selective, safe, and environmentally benign fashion
compared to conventional tedious chemical synthesis. Herein, we disclose an environmentally
benign, rapid, catalyst/promoter/coupling reagent-free cyclization procedure of free amino acids
to furnish exclusively cyclic dipeptides (2,5-diketopiperazines, DKPs) in excellent or even
quantitative yield, along with their solid state self-assembling properties. This process is extremely
simple and highly efficient with little or no traditional synthetic skills and without any
chromatographic purification. Synthesis of structurally diverse DKPs has been achieved with a
dramatic decrease in the reaction time, the amount/number of solvents used, a significant increase
in the yield and nearly complete elimination of waste. As a result, this is an excellent example for
the environmentally benign, clean and green chemistry concept. The most exciting outcome of our
investigation is an unusual case of chiral self-recognition encountered upon the cyclization of
rac-pipecolic acid, which resulted in the formation of the meso-product exclusively.
1
0
Introduction
and bradykinin antagonists. It has been shown that DKPs are
involved in quorum sensing (cell to cell communication in Vibrio
spp.), neuroprotection, and can be used as blood–brain bar-
rier (BBB) shuttles. Furthermore, the self-assembly of DKPs
derived from aromatic amino acids as building blocks for bio-
Bioactive cyclodipeptides [(termed as 2,5-diketopiperazines
11
12
(
DKPs), 2,5-dioxopiperazines or dipeptide anhydrides)] are
13
ubiquitous in nature, and they represent the simplest and
1,2
first completely characterized small peptides. This class of
molecules are potential drug candidates because of their
14
nanostructures/peptide nanotubes, has recently been reported.
Because of their rigidity and restricted conformational freedom,
they provide a unique opportunity to design suitable receptors
and drug candidates, which can be used as models to shed
light on more complex peptides and proteins. The crystalline
nature of DKPs has proved to be useful to elucidate the primary
3
4
5
6
antibacterial, anticancer, herbicidal, and antiviral properties.
Recent progress in the field of molecular biology, medicinal
chemistry, neuroscience, and nanoscience has uncovered the
wide spectrum of biological activities of DKPs, such as in-
hibitors of topoisomerases, collagenase-1, chitinase, tubulin
depolymerization, plasminogen activator inhibitor-1 (PAI-1),
15
7
8
9
16
structure or analyze the sequence of polypeptides and proteins.
This property of DKPs has attracted both experimental and
theoretical chemists. Very recently, Bud eˇ sˇ ´ı nsk y´ et al. reported
combined experimental X-ray studies and computational calcu-
lations of cyclopeptides with diverse functionalities. Blanco
et al. evaluated discrimination in the recognition of self-
assembled complexes of DKPs using density functional theory
(DFT) calculations. It has been suggested that microbial pep-
tides are derived partly or completely by enzymatic condensation
and ring expansion of diketopiperazines. Recent findings using
hydrothermal conditions also showed the importance of DKPs
towards the origin and evolution of life in primordial earth.
Department of Chemistry, FI-40014, University of Jyv a¨ skyl a¨ , Finland.
E-mail: nonappa@jyu.fi, manu.k.lahtinen@jyu.fi, erkki.t.kolehmainen@
jyu.fi; Tel: + 358 14 260 2653
17
†
Electronic supplementary information (ESI) available: Supplementary
18
information including experimental procedures along with MW graphs
1
13
for all the compounds studied, H, C NMR, ESI-MS data, elemental
analysis, melting points, optical rotation, single crystal X-ray data
collection procedure, tables containing crystallographic details of (1,
19
9
, 11–13, bond length, bond angles, dihedral angles and hydrogen
20
bonding interactions), CCDCs 783550–783553 contain the supplemen-
tary crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c1gc15043j
A careful analysis of structurally diverse natural products
clearly suggests that a number of medicinally important natural
products contain DKP moieties embedded with them (Fig. 1).
21
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1203–1209 | 1203

View Article Online
24
for some time. While working on this topic, we envisioned
that a microwave reaction with simple amino acids without any
catalyst, promoter or base would be interesting. We speculated
that such a reaction may result in a complex mixture of
products. In order to test our hypothesis, glycine was chosen
as a model compound in our initial studies. Solution of glycine
in water, N,N-dimethlyformamide (DMF) and solid glycine
were subjected to controlled (temperature, pressure and power)
microwave (MW) irradiation (ESI†). The reaction mixture in
DMF turned into a clear solution after 5 min and transformed
from colorless to a pale yellow solution. When the reaction had
ceased, a pale yellow precipitate was formed. Analysis of the
filtered product revealed the formation of 2,5-diketopiperazine
Fig. 1 Representative examples of natural products containing DKP
moieties.
(
DKP) 1 (Fig. 2). It is noteworthy that no other side product
was obtained and the compound required no chromatographic
separation or extraction. The product obtained underwent
It is noteworthy that the cyclization of amino acids leading
to DKP formation continue to be major reactions in natural
product synthesis, and still the methods suffer from protec-
tion/deprotection and the use of a base or an acid and a low
spontaneous crystallization in D O (NMR sample) and the
2
structure was proved unambiguously using single crystal X-ray
diffraction (Fig. 4). On the other hand the reaction carried out
with water as a solvent turned into a clear solution and unreacted
starting material was recovered. However, it was not possible
to attain the requisite temperature for the solid alone and as
a result, the starting material remained unchanged. Inspired
by this observation we decided to use DMF as a solvent and
continued our investigation with other optically active amino
acids. A series of compounds were screened and, interestingly,
all amino acids underwent cyclization and furnished their
respective DKP derivatives (Fig. 2). Carrying out a selective
reaction when multiple functional groups are present is a
challenge. Therefore, in order to test our methodology tyrosine
was subjected for MW reaction under similar experimental
conditions. After completion of the reaction a brown precipitate
was obtained, which after filtration followed by recrystallization
from methanol resulted in 6 as a colorless solid. It is noteworthy
that the compound 6 and its analogues are reported to bind
m-opioid receptors and are active against tumour cell lines at
22
yield. There is a steady increase in the isolation of DKPs
from natural sources. Studies on DKPs started as early as
the 1940s and several synthetic methods have been reported
in the literature via which both symmetric and unsymmetrical
2,23
DKPs can be obtained. So far there have been no reports
on the condensation of free secondary amino acids, such
as proline and substituted proline derivatives without using
any catalyst or promoter, to yield pure DKPs in quantitative
yield.
It is often said that “the best catalyst is no catalyst”.
Herein, we disclose a rapid catalyst- and promoter-free cy-
clization/condensation of free amino acids using controlled
microwave irradiation, which furnished exclusively DKPs in an
excellent yield without involving any chromatographic purifica-
tion or extraction. The most important and interesting finding
in our discovery is that the rapid cyclization of secondary amino
acids, such as proline, substituted proline and pipecolic acid
derivatives provided a one step synthesis of tri- or pentacyclic
DKPs and their structures have been ascertained unambiguously
using single crystal X-ray diffraction. One of the most fascinat-
ing findings of our investigation is the chiral self-recognition of
rac-pipecolic acid, which resulted in the optically inactive meso-
product in spite of several possible products. Specific examples
showing a dramatic decrease in the amount/number of solvents,
number of steps, reaction time, cost and a significant increase
in the yield, the nearly complete elimination of waste and other
purification steps compared to that of conventional synthetic
methods, are presented. This is one of the excellent examples for
the environmentally benign, green, clean and quick chemistry
approach to the synthesis of bioactive small molecules without
any tedious purification methods. The fundamental concepts
of environmentally benign synthesis, the application of this
methodology and the solid state self-assembling properties of
DKPs are highlighted in this paper.
25
mM concentration. The above results are highly interesting
and resulted in a new direction toward the synthesis of DKPs
in a green, clean and rapid approach, which is highly efficient in
terms of time, purity, yield and labor. However, the cyclization
of secondary amino acids, such as proline, will be even more
exciting and the condensation of substituted prolines are equally
challenging because of possible side reactions. DKPs derived
from proline and substituted prolines are frequently found
to be an integral part of many natural products, such as
26
Spirotryoprostratin A (Fig. 1). This prompted us to investigate
the cyclization of secondary amino acids. When proline was
subjected to the similar reaction conditions, a pale yellow
solution was obtained, which upon solvent evaporation followed
by addition of acetonitrile, resulted in a colorless crystalline
solid. Structural characterization and systematic spectral anal-
ysis revealed the formation of racemic DKPs. It is noteworthy
that there exist both experimental and theoretical investigations
towards understanding the racemization of amino acids under
pyrolytic conditions. Extensive mechanistic studies for various
amino acids including the racemization of proline and its
derivatives under various reaction conditions have been reported
Results and Discussion
27
The condensation of a free carboxylic group with an amine
in the literature. Such a phenomenon depends on multiple
experimental factors. A number of experiments conducted to
(
intermolecular) under microwave conditions has been known
204 | Green Chem., 2011, 13, 1203–1209
1
This journal is © The Royal Society of Chemistry 2011

View Article Online
Fig. 2 Representative graphs of the controlled MW reaction of compound 9; temperature (red); pressure (blue); power (green) and the examples
a
b
c
studied. rac-mixture 2 = 3; rac-mixture 7 = 8; starting material remained unreacted.
understand prebiotic chemistry resulted in the formation of
racemic products.
sycamore cell wall was composed of a glycopeptide where the
peptide part was a cyclic dimer of 4-hydroxy proline; (ii) this
20
28
Nevertheless, a detailed literature search suggested that
compounds like 9 will be of immense interest because of the
following reasons; (i) Heath and Northcote reported that the
compound was an intermediate in the synthesis of a microtubule
inhibitor, where the synthesis involves multiple steps, reagents,
protection/deprotection and tedious purifications. Inspired by
29
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1203–1209 | 1205

View Article Online
these observations we continued our investigation towards the
one-step synthesis of a target molecule starting from trans-4-
hydroxy-L-proline ((2S,4R)-4-hydroxy proline). To our pleasant
surprise the reaction mixture yielded a pale yellow solid upon
solvent evaporation and recrystallization in methanol furnished
the desired compound 9 in quantitative yield (Fig. 2 and 3). This
clearly indicates how one can think beyond the conventional
approach.
has been ascertained unambiguously using single crystal X-ray
analysis (Fig. 4). It is noteworthy that there have been several
efforts to utilize microwave irradiation for the condensation of
amino acids, but the procedure still involves multiple reagents,
a promoter, a base, ionic liquids and tedious chromatographic
30
purification. Houston et al. reported a series of cyclodipetides
as chitinase inhibitors and the authors concluded that the
cyclo(Pro-Gly) substructure is enough for binding, allowing
29
9
The synthesis reported earlier (Fig. 3) for the compound
utilizes 2 different starting materials (more expensive), more
modification of the non-proline residue side chain. Inspired
9
by this report, we decided to prepare cyclo(Pro-Gly) as a model
compound to test our speculations starting from L-Pro-Gly-OH.
Indeed, the reaction mixture turned to a clear colorless solution
and upon cooling to room temperature, the crystalline solid was
obtained. The spectral characterization and successful single
crystal X-ray analysis unambiguously proved the formation of
compound 12 (Fig. 4).
than 6 different solvents (~2.0 L), catalysts, chromatographic pu-
rification, a longer reaction time (days) and multiple side/waste
products (5), whereas our approach is extremely simple with a
relatively cheap starting material and one solvent (0.04 L) with a
dramatic decrease in the reaction time (25 min) and only water as
a side product at the end of the reaction. The structure of 9 was
proved unambiguously using a single crystal X-ray structure for
the first time (Fig. 4). The determination of the optical rotation
Chiral self-recognition
29
was in agreement with data reported in the literature.
A large number of biochemical reactions in living systems
depend on chiral recognition, “a process in which discrimination
between enantiomers or enantiotopic ligands achieved by well
31a
defined reagents or catalysts (enzymes or synthetic)”. Chiral
self-recognition is a process in which a chiral compound reacts
31b,c
preferentially with one of its enantiomers.
Complete self-
recognition will result in exclusive formation of either a racemic
mixture of enantiomeric products or an achiral meso product.
However, when there is no chiral self-recognition, the reaction
products will be a mixture of rac : meso (1 : 1) products in equal
amounts. This is of great importance in organic synthesis and
the principle is often compared to that of non-linear effects in
asymmetric catalysis. In the course of our investigation towards
the synthesis of DKPs derived from pipecolic acid, an unusual
case of chiral self-recognition was encountered.
Racemic pipecolic (piperidine-2-carboxylic) acid under MW
reaction conditions resulted in a colorless solid. Recrystal-
lization from methanol resulted in quality crystals and X-ray
single crystal analysis revealed the formation of the meso-
product exclusively, showing chiral self-recognition (Fig. 4).
Theoretically, a racemic starting material (1 : 1 mixture of L-
and D-pipecolic acid) may undergo cyclization and result in the
formation of a racemic mixture of DKPs or the meso product
Fig. 3 A comparison of the traditional approach and our approach
towards the synthesis of compound 9.
(Fig. 5). Although the exact reason behind self-recognition in
pipecolic acid is not known, we speculate that this may be due
to the conformational stability of the DKP or to the more
energetically favoured interaction between enantiomers than
within one single enantiomer. What is more important to note
here is the possible synthetic scheme outlined in Fig. 5b. The
synthesis of 13 is possible only using optically pure enantiomers,
with two different starting materials and multiple steps and
catalysts, and will result in tedious purification and waste. This
is one of the most fascinating outcomes of this investigation as
it is almost impossible to obtain compound 13 exclusively using
routine organic synthetic procedures starting from a racemic
mixture.
Fig. 4 The molecular structures and labelling schemes of compounds
1
*
, 9 and 11–13. Symmetry operator for 1 * = -x + 2, -y, -z + 1 and 13:
= -x, -y + 1, -z + 1. The hydrogens are numbered according to their
host atoms. The thermal ellipsoids are drawn with 50% probability.
The strategy was extended for the construction of a pen-
tacyclic system by taking (S)-(-)-indoline-2-carboxylic acid
as starting material (Fig. 2). This reaction also furnished
compound 11 in quantitative yield, the structure of which
Single crystal X-ray studies
The understanding and unambiguous determination of the
molecular structure of small molecules are of utmost importance
1
206 | Green Chem., 2011, 13, 1203–1209
This journal is © The Royal Society of Chemistry 2011

View Article Online
drawn through carbons C(3) and C(6) forming a shallow boat
◦
conformation (dihedral angle N(2)–C(3)–C(4)–N(5) = 28.6 ),
so that both keto-groups (O(1) and O(2)) are situated above
the main plane (via atoms C(1)–C(4)–C(8)–C(11)), whereas
the H(3) and H(6) hydrogens are located below the plane, as
are the hydroxyl groups O(3) and O(4). Dipeptide molecules
are ordered in parallel rows along the b-axis, forming wavy
layers along the c-axis at the same time, with every second
dipeptide facing up, whereas the other points down. An extensive
hydrogen bonding network is present in the crystal lattice, as
both the keto- and hydroxyl groups participate in 3-dimensional
hydrogen bonding with neighbouring water molecules (Fig.
6a), and due to this conventional hydrogen bonding between
adjacent dipeptides cannot be observed, instead only weak
Fig. 5 (a) A reaction scheme showing the possible products from the
cyclization of rac-pipecolic acid and the selective formation of 13 (dashed
lines represent mirror planes); (b) a possible synthetic route to 13 using
conventional synthetic methods.
interactions occur between the C(3)–H(3) ◊ ◊ ◊ O(2)#5 and C(7)–
H(7A) ◊ ◊ ◊ O(3)#6 atoms having d(H ◊ ◊ ◊ A) distances of 2.35 and
2.53 A˚ , respectively. Moreover water molecules are hydrogen
˚
bonded to each other having d(H ◊ ◊ ◊ A) distances of 1.93(3) A
◦
with a contact angle of 173(3) , and they act as hydrogen
and are equally challenging. Single crystal X-ray structure
determination provides a unique opportunity to study and
extract useful information on molecular structure in the solid
state. Attempts were made to crystallize all the samples, while
compounds 1, 9, 11–13 underwent crystallization resulting in
quality single crystals, the others resulted in amorphous solids
and their structures have been characterized by NMR (ESI†).
The crystallographic data of compounds 1, 9, 11–13, the selected
bond lengths, angles together, dihedral angles and hydrogen
bond distances (ESI, Tables S1–S6†) along with the labelling and
the molecular structure of the dipeptides (Fig. 4) are presented.
Compound 1 crystallizes in the monoclinic space group
bonding links both between dipeptide molecules situated within
a single layer as well as between two dipeptide layers (ESI,
Table S6†).
Racemic compound 11 underwent crystallization upon addi-
tion of methanol and resulted in the monoclinic space group
P2 /c (No 14) having a single molecule in an asymmetric unit
1
(Fig. 4). In contrast to compound 9, the dipeptide molecule is
now slightly more folded across the axis between the C(3) to
◦
C(6) atoms (dihedral angle N(2)–C(3)–C(4)–N(5) = 39.2 ) and
due to the aromatic ring, the molecule is forced to be nearly
planar across the folding axis. The CSD search reveals one
33
entry (COSGEX) having the same molecular composition as
P2
1
/c (No. 14) having a half molecule in an asymmetric unit
11 but representing the enantiopure L-form that crystallizes in
(
ESI, Fig. S5†), therefore being structurally equivalent to that
the orthorhombic space group P2
1
P2
1
P2 and has nearly the
1
32
3
3
˚
˚
reported in the literature (CSD: DIKPIP02), although the
crystallographic agreement factors (R-values) are clearly better
in our case (ESI, Table S4†). The six-membered diketopiperazine
same unit cell volume (1339.1 A ) as 11 (1342.7 A ). Also the
molecular conformations of both the racemic and enantiopure
crystal structures are very similar (ESI, Fig. S2†) having only
a small conformational difference in the folding axis (via C(3)
to C(6)). Therefore the enantiopure L-form (COSGEX) has a
◦
is virtually planar, having a dihedral angle of -178.4(1) for
O(1)–C(2)–N(3)–C(4)#1. The molecules form infinite chains
parallel to the (101) plane and individual molecules are located
on top of each other along the a-axis (ESI, Fig. S1†). In these
chains, dipeptide molecules are hydrogen bonded to each other
via O(1) and N(3) (and their symmetry equivalents) atoms
◦
slightly larger dihedral angle of 42.5 (N(2)–C(3)–C(4)–N(5)).
In the case of compound 11, both enantiomers are packed
partially facing each other, forming rows of paired molecules
along a- and b-axes (Fig. 6b). These pairs are packed along
◦
˚
showing a d(H ◊ ◊ ◊ A) distance of 1.958(16) A. Moreover, the
the c-axis having a nearly 90 horizontal rotation on each pair
adjacent and top dipeptide chains are connected to each other
of facing dipeptides (ESI, Fig. S3†). This enables weak p–p
by weaker hydrogen bonding via H(4a)–O(1) and H(4b)–O(1)
atoms, having distances of 2.64 and 2.51 A, respectively. The
interactions to occur between the aromatic groups of the two
overlapping molecules that have ~90 longitudinal rotation in
◦
˚
selected bond distances and angles are collected in Tables S2
and S3 (ESI†). Crystallization of compound 9 in methanol
furnished single crystals as its dihydrate in the orthorhombic
their orientation, as well as between an aromatic group and
the six-membered ring of the diketopiperazine core of two
facing enantiomers. In the enantiopure structure, the packing
scheme is somewhat different as molecules are now packed in
rows along the b-axis, forming a wavelike formation along the
c-axis. Moreover, dipeptides are not facing each other while
space group P2
1
2
1
1
2 (No. 19), having a single molecule in the
asymmetric unit with two molecules of water (Fig. 6a). Although
the absolute configuration cannot be determined reliably as
Mo-radiation has been used in the structure determination
without a heavier scatterer (meaningless Flack parameter =
◦
packed along the a-axis and having a longitudinal ~90 turn
on every second molecule, albeit molecules are also flipped on
each rotation. In compound 11 conventional hydrogen bonding
cannot be observed in the packing scheme. However, dipeptide
molecules are interconnected to each other by weak C–H ◊ ◊ ◊ O
interactions, of which d(H ◊ ◊ ◊ A) distances vary from 2.38 to
-
0.1(11)), the structure is expected to be a pure enantiomer of
the L-form, clearly suggesting that the enantiomerically pure
starting material (4-hydroxy-L-proline) did not undergo any
racemization under the reaction conditions. The six-membered
diketopiperazine core is slightly folded upwards via an axis
˚
2.58 A.
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1203–1209 | 1207

View Article Online
Fig. 6 The molecular packing (a) of 9 along the a- (top) and b-axis (middle) and the hydrogen bonding scheme (bottom) (some water and dipeptide
molecules have been omitted for clarity); (b) the packing of 11 along the a- (left) and b-axis (right); (c) the packing of 12 along the a- (top) and b-axis
(
bottom). The hydrogen bonding between the N(5) and O(1) atoms are shown by dashed green lines; (d) the molecular packing of compound 13
along the a- (top) and b-axis (bottom) (hydrogen atoms have been omitted for clarity).
Compound 12 crystallizes in the orthorhombic space group
P2 (No. 19) having also a single molecule in an asymmetric
Compound 13 crystallizes in the monoclinic space group
P21/c (No. 14), having a half molecule in an asymmetric unit
1
2
1
2
1
unit. Although the absolute configuration cannot be determined
unambiguously, the structure is expected to be the L-form.
The structural parameters for the L-form can be found in the
(Fig. 6d). The six-membered diketopiperazine core is virtually
◦
◦
planar having dihedral angles of -0.5 and 175 for O(1)–C(1)–
C(2)–N(2)#1 and O(1)–C(1)–N(2)–C(6)#1, respectively. In ad-
dition, the dipeptide is in a chair-conformation having a dihedral
34
CSD (entry code LPROGL03), having nearly equal unit cell
3
3
◦
˚
˚
volume 740 A compared to that of 12 (721 A ). The molecules
are packed in infinite hydrogen bonded parallel zigzag rows
along the b-axis, and at the same time the chains overlap along
the a-axis (Fig. 6c). Furthermore, the rows going through the
angle of -60.5 for O(1)–C(1)–C(2)–C(3). The molecules are
overlapping along the a-axis and form rows along the b-axis.
The dipeptide molecules in every other row are pointing right
whereas they point to the left in the other ones (Fig. 6d). The
molecules are connected to each other only by weak hydrogen
bonds, which are visible between the carbonyl oxygens and the
C–H hydrogens, H6a and H6b belonging to the two nearest
molecules located on the top and bottom layers (ESI, Fig. S5†)
◦
asymmetric unit along the a-axis are tilted nearly 90 to each
other when viewed along the b-axis. The hydrogen bonding
occurs only between the H(5) and O(1) atoms of adjacent
˚
dipeptides, having a d(H ◊ ◊ ◊ A) distance of 1.96(2) A with an
◦
˚
angle of 175(3) . In addition, weak hydrogen bonding C–
with d(H ◊ ◊ ◊ A) distances of 2.35 and 2.49 A, respectively.
H ◊ ◊ ◊ O C occurs from the H(3) and H6a hydrogens to the O1
˚
and O2 carbonyls, respectively, with distances of about 2.5 A.
Conclusions
The packing scheme of the known L-enantiomer is very similar,
as the infinite hydrogen bond networks forming zigzags run
along the b-axis and having the same contact atoms for hydrogen
bonding.
In summary, we have shown how free amino acids can
be selectively cyclized under microwave conditions to afford
cyclodipeptides (2,5-diketopiperazines, DKPs). This process
1
208 | Green Chem., 2011, 13, 1203–1209
This journal is © The Royal Society of Chemistry 2011

View Article Online
is extremely simple, environmentally benign and involves no
catalysts/promoters or coupling reagents. The entire process
involves no chromatographic purification and as a result, the
solvent waste is minimal or even completely eliminated. We
also demonstrated the potential application of this process for
the synthesis of a natural product, which otherwise involves
multiple steps and sensitive reaction conditions using traditional
synthetic methods. An unusual case of chiral self-recognition
was encountered in the case of rac-pipecolic acid. We have shown
how one can think beyond traditional synthetic methods and we
believe that our demonstration will be highly useful for synthetic,
bioorganic, medicinal, and materials chemists, and biologists. As
there are no catalysts or promoters involved, it will be of great
interest to carry out reactions of molecules having a variety of
functional groups. Though a few amino acids resulted in racemic
DKPs, others yielded enantiomerically pure forms indicating
that the process of racemization depends on the stability of
amino acids and the reaction conditions. We strongly believe
that the application of our method offers a unique opportunity
to solve problems in natural product synthesis, chemical biology,
and the design of biomaterials.
9 D. R. Houston, B. Synstad, V. G. H. Eijsink, M. J. R. Stark, I. M.
Eggleston and D. M. F. van Aalten, J. Med. Chem., 2004, 47, 5713–
720.
0 B. Nicholson, G. K. Lloyd, B. R. Miller, M. A. Palladino, Y. Kiso
and Y. Hayashi, Anti-Cancer Drugs, 2006, 17, 25–31.
11 K. E. Klose, J. Bacteriol., 2006, 188, 2025–2026.
2 A. I. Faden, S. M. Knoblach, I. Cernak, L. Fan, R. Vink, G. L.
Araldi, S. T. Fricke, L. R. Bryan and A. P. Kozikowski, J. Cereb.
Blood Flow Metab., 2003, 23, 342–354.
13 M. Teixido, E. Zurita, M. Malakoutikhah, T. Tarrago and E. Giralt,
J. Am. Chem. Soc., 2007, 129, 11802–11813.
4 L. Adler-Abramovich, D. Aronov, P. Beker, M. Yevin, S. Stempler,
5
1
1
1
L. Buzhansky, G. Rosenman and E. Gazit, Nat. Nanotechnol., 2009,
4
, 849–854.
15 (a) P. M. Fischer, J. Pept. Sci., 2003, 9, 9–35; (b) A. S. M. Ressurrelc a˜ o,
R. Delatouche, C. Gennari and U. Plarulli, Eur. J. Org. Chem., 2010,
2
17–228.
1
6 S. Yamada, Y. Yokoyama and T. A. Shioiri, Experientia, 1976, 32,
398–399.
17 M. Bud
eˇ sˇ ´ı nsk y´ , I. Cisarov a´ , J. Podlaha, F. Borremans, J. C. Martins,
M. Waroquier and E. Pauwels, Acta Crystallogr., Sect. B: Struct.
Sci., 2010, B66, 662–677.
1
8 F. Blanco, I. Alkorta, I. Rozas and J. Elguero, J. Phys. Org. Chem.,
2010, 23, 1155–1172.
1
2
9 B. W. Bycroft, Nature, 1969, 224, 595–597.
0 C. Huber, W. Eisenreich, S. Hecht and G. W a¨ chtersh a¨ user, Science,
2
003, 301, 938–940.
2
2
2
1 B. Clark, R. Capon, E. Lacey, S. Tennata and J. H. Gill, J. Nat. Prod.,
005, 68, 1661–1664.
2 J. Kim, J. A. Ashenhurst and M. Movassaghi, Science, 2009, 324,
38–241.
3 C. J. Dinsmore and D. C. Beshore, Tetrahedron, 2002, 58, 3297–
312.
24 P. Lidstr o¨ m, J. Tierney, B. Wathey and J. Westman, Tetrahedron,
001, 57, 9225–9283.
2
Acknowledgements
2
We thank the rector of the University of Jyv a¨ skyl a¨ and the
Academy of Finland (Decision n:o 127006, December 12th
3
2
008) for the financial support towards the post-doctoral
2
fellowship. The Nanoscience centre for microwave facilities,
X-ray diffractions facilities in the department of chemistry,
University of Jyv a¨ skyl a¨ are acknowledged. Spec Lab techni-
cian Reijo Kauppinen for NMR measurement and Spec lab
technician Mirja Lahtiper a¨ for mass spectral measurements are
acknowledged.
2
5 G. Kilian, H. Jamie, S. C. A. Brauns, K. Dyason and P. J. Milne,
Pharmazie, 2005, 60, 305–309.
6 B. M. Trost and D. T. Stiles, Org. Lett., 2007, 9, 2763–2766.
2
27 (a) P. M. Shou and J. L. Bada, Naturwissenschaften, 1980, 67, 37–38;
(
b) H. Kuroda, S. Kubo, N. Chino, T. Kimura and S. Sakakibara,
Int. J. Pept. Protein Res., 1992, 40, 114–118; (c) R. Nagaraj and
P. Balaram, Tetrahedron, 1981, 37, 2001–2005; (d) S. Yamada, C.
Hongo, R. Yoshioka and I. Chibata, J. Org. Chem., 1983, 48, 843–
8
46.
Notes and references
2
8 M. F. Heath and D. H. Northcote, Biochem. J., 1973, 135, 327–329.
1
2
3
C. Prasad, Peptides, 1995, 15, 151–164.
29 F. N. Shirtoa, H. T. Nagasawa and J. A. Elberling, J. Med. Chem.,
R. B. Corey, J. Am. Chem. Soc., 1938, 60, 1568–1604.
S. Hirano, S. Ichikawa and A. Matsuda, Bioorg. Med. Chem., 2008,
1977, 20, 1176–1181.
30 M. Jainta, M. Nieger and S. Br a¨ se, Eur. J. Org. Chem., 2008, 5418–
5424.
31 (a) E. L. Eliel, S. H. Wilen and L. N. Mander, in Stereochemistry of
Organic Compounds, John Wiley & Sons, Inc, New York, 1994; (b) S.
K. Pradhan and K. R. Thakker, Tetrahedron Lett., 1987, 28, 1813–
1816; (c) E. Touboul and G. Dana, Tetrahedron, 1975, 31, 1925–
1931.
32 T. R. Sarangarajan, K. Panchanatheswaran, J. N. Low and C.
Glidewell, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2005,
C61, o118–o121.
1
6, 428–436.
4
E. Van Der Merwe, D. Huang, D. Petersen, G. Kilian, P. J. Milne, M.
Van de Venter and C. Frost, Peptides, 2008, 29, 1305–1311.
R. R. King and L. A. Calhoun, Phytochemistry, 2009, 70, 833–841.
S. Sinha, R. Srivastava, E. D. Clercq and R. K. Singh, Nucleosides,
Nucleotides Nucleic Acids, 2004, 23, 1815–1824.
5
6
7
P. A. Charlton, R. W. Faint, F. Bent, J. Bryans, I. Chicarelli-Robinson,
I. Mackie, S. Machin and P. Bevan, Thromb. Haemost., 1996, 75, 808–
8
12.
8
A. K. Szardenings, V. Antonenko, D. A. Campbell, N. D. Francisco,
S. Ida, L. Shi, N. Sharkov, D. Tien, Y. Wang and M. Navre, J. Am.
Chem. Soc., 1999, 42, 1348–1357.
33 M. Jainta, M. Nieger and S. Brase, J. Mol. Struct., 2009, 921, 85–88.
34 D. Hendea, S. Laschat, A. Baro and W. Frey, Helv. Chim. Acta, 2006,
89, 1894–1909.
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1203–1209 | 1209