Induction of chirality: Experimental evidence of atropisomerism in azapeptides

Article abstract of DOI:10.1039/c2cc31161e

Methylation of the peptide bond in model azadipeptides leads to the E configuration and hence to atropisomerism due to a restricted rotation around the N-N axis. This journal is

Full text of DOI:10.1039/c2cc31161e

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 5772–5774
www.rsc.org/chemcomm
COMMUNICATION
Induction of chirality: experimental evidence of atropisomerism in
azapeptidesw
Philipp A. Ottersbach,a Gregor Schnakenburgb and Michael Gu
¨
tschow*a
Received 16th February 2012, Accepted 17th April 2012
DOI: 10.1039/c2cc31161e
Methylation of the peptide bond in model azadipeptides leads to
the E configuration and hence to atropisomerism due to a
restricted rotation around the N–N axis.
Table 1 Preparation of 6–9 and structures of 6–13
Azapeptides are peptides in which the CaH unit of one or more
amino acid(s) has been replaced by a nitrogen atom. This class of
compounds has attracted much attention as bioactive
peptidomimetics.1 The carbon–nitrogen replacement results in
unique conformational preferences of the peptide. Hence, the
analysis of the secondary structure of such molecules has gained
relevance since the adoption of a specific conformation of
(aza)peptides is crucial for biological recognition. Several
attempts have been made to characterize conformational properties
of azapeptides by theoretical techniques.2 However, considering
their importance, the valuable crystallographic and spectroscopic
investigations on azapeptides are rather limited.3
Compound
R1
R2
R3
2
3
4
5
6
7
8
9
10
11
12
13
H
H
Me
Me
H
H
Me
H
Me
H
Me
H
Me
H
Me
Me
Et
CONH2
CONH2
CONH2
CONH2
CONH2
CONH2
CONH2
CONH2
H
H
Me
Me
H
Me
Me
Me
In this work, structurally reduced azadipeptides with one
amino acid connected to one aza-amino acid were examined
by different NMR and crystallographic methods. Due to the
exchange of the C-terminal CaH by N, the resulting carbazic
acid is unstable, and the carboxylic group has to be modified.
Accordingly, we envisaged azadipeptide amides, bearing a
primary amide instead of a carboxylic acid function. The
simplest conceivable stable azadipeptide, H-Gly-azaGly-NH2,
was equipped with a protecting group and thus, the achiral
azadipeptide amide, Cbz-Gly-azaGly-NH2 (6), was regarded as
the parent molecule for this investigation (Table 1). In the
course of the structural analysis of 6 and its derivatives,
we observed an unexpected occurrence of diastereotopicity
depending on the substitution pattern of azadipeptide amides.
Thus, a systematic N-methylation of 6 was performed (6–9).
For these scaffold modifications, different synthetic entries had
to be explored.
H
Me
Me
Reaction of 1 with 2-methylsemicarbazide (3), prepared from
N-methylhydrazine and potassium cyanate, yielded the
N2-methylated azadipeptide Cbz-Gly-azaAla-NH2 (7). To
obtain Cbz-Gly-azaSar-NH2 (8), an N1-methylated isomer of
7, a synthetic entry to 1-methylsemicarbazide (4) was needed.
A facile access, superior to reported methods,4 was established
(see ESIw). Next, 1 was coupled with 1,2-dimethylsemicarbazide
(5), prepared from potassium cyanate and 1,2-dimethyl-
hydrazine dihydrochloride, to produce Cbz-Gly[CCONMe]-
azaAla-NH2 (9).
1
We obtained unexpected results by analyzing the H NMR
spectra (500 MHz) of azadipeptide amides; all spectra were
recorded in DMSO-d6. The Gly methylene protons of
Cbz-Gly[CCONMe]-azaAla-NH2 (9) showed an individual
1H NMR signal with a geminal coupling, in addition to
coupling with the neighboring carbamate proton. The ABX
system (2J = 17.4 Hz, 3J = 5.7 Hz) indicated diastereotopicity,
although the molecule does not contain any asymmetric
carbon atom. In the case of Cbz-Gly-azaSar-NH2 (8), the
Gly methylene protons appeared as two broad signals, show-
ing their chemical non-equivalence. However, the spectra of
Cbz-Gly-azaGly-NH2 (6) and Cbz-Gly-azaAla-NH2 (7) displayed
Cbz-Gly-azaGly-NH2 (6) was easily obtained by mixed
anhydride coupling of Cbz-Gly-OH (1) with semicarbazide (2).
a Pharmaceutical Institute, Pharmaceutical Chemistry I,
University of Bonn, An der Immenburg 4, 53121 Bonn, Germany.
E-mail: guetschow@uni-bonn.de
b Institute of Inorganic Chemistry, University of Bonn,
Gerhard-Domagk-Strasse 1, 53121 Bonn, Germany
w Electronic supplementary information (ESI) available: Experimental
procedures, HPLC experiments, electronic structure calculations.
CCDC 837002 and 837003. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c2cc31161e
c
5772 Chem. Commun., 2012, 48, 5772–5774
This journal is The Royal Society of Chemistry 2012

View Online
of the peptidic CO–N1Me bond in 9 leads to an orientation of
the Gly methylene group towards the N2 atom. The molecular
status of respective azadipeptide amides 8 and 9 results in
steric interactions of both N1 substituents with the N2
substituent(s) and, thus, in an enhanced rotational barrier of
the N–N bond.
a doublet (3J E 6 Hz) for the equivalent protons of the CH2
group. Thus, the 1H NMR signal of the Gly fragment was
employed as a facile and suitable sensor of chirality to allow
for comparative analyses of azadipeptide amides and related
molecules. In continuation of our study, it was intended to
reveal the underlying cause of chirality.
It might be taken into consideration that the nitrogen N2 of
compounds 6–9 adopts an sp3 hybridization state as it has been
observed for the acylated azapipecolic amide motif in peptides.5
Restricted pyramidal inversion of that nitrogen could then
produce two enantiomers and would thus be the source of
chirality. However, as expected, our crystallographic data
indicate the sp2 hybridization state of N2 in azadipeptide amides
7 and 9 (for bond distances and angles, see CIF file, ESIw).
As reported, 1,2-dibenzoyl-1,2-dipropanoylhydrazine, exhi-
biting the bis-acylated hydrazine moiety also present in aza-
dipeptide amides, displayed geminal coupling of the methylene
protons, and slow rotation around the N–N axis was suggested
to be responsible for the chirality.6 Rotational barriers of the
N–N bond were also studied theoretically by NBO approaches,
using diformylhydrazine as a model compound, but conclusions
were impaired due to the complexity of this molecule.2e
Thus, we considered azadipeptide amides to be atropochiral
molecules with a hindered single-bond rotation. Atropisomer-
ism7 is a source of chirality that has been increasingly recog-
nized as a concept in drug development, since chirality of small
molecules results in different pharmacological effects of the
distinct enantiomers.8
To enforce this rotation, a variable temperature (VT) NMR
(500 MHz) experiment of 9 was performed. No coalescence of
the Gly methylene protons’ resonances occurred and well-
1
separated signals remained even at 396 K. Hence, VT H NMR
spectra of 9 at 300 MHz were recorded for line-shape analysis to
determine the rotational barrier. A value of 117 kJ molÀ1 indicates
the atropisomerism of compound 9 (see ESIw).
As mentioned, 7 showed chemical equivalence of the Gly
methylene protons, in contrast to 8, although both molecules
bear one methylated nitrogen. The Z configuration of the
CO–N1H bond in 7 leads to an extended structure, in which a
free rotation around the N–N bond is possible (Fig. 1). The
hindered rotation of 8 can again be explained by the E configu-
ration of the CO–N1Me bond and the resulting steric interactions
of the N2 carbamoyl group with both N1 substituents. From the
VT NMR (500 MHz) experiment, a coalescence of the two broad
signals, each belonging to one Gly methylene proton of 8, was
achieved at 318 K. Line-shape analysis provided a rotational
barrier of 73 kJ molÀ1 (see ESIw).
As atropisomerism was diagnosed for the azadipeptide
amides 9, it was intended to identify minimal structural
requirements and further amino acid derivatives (10–13;
Table 1) were synthesized (see ESIw) and inspected. As
expected for hydrazide 10, chemical equivalence of the Gly
methylene protons was observed in the 1H NMR spectrum. In
the related N1,N2-bis-methylated hydrazide 11 with the E
configuration of the CO–N1Me bond, steric interactions of
the N1 substituents with the N2 methyl group should be
possible. However, no diastereotopicity was found for 11,
implicating that the rotational barrier is not sufficiently high.
The exchange of CONH2 (in 8) by methyl (in 11) allowed for a
facilitated rotation around the N–N bond, most probably
because of the changed geometry of the substituents at the
N2 atom depending on its hybridization state. Introduction of
Our observation that Cbz-Gly-azaGly-NH2 (6) displayed no
diastereotopicity according to 1H NMR, but the N1,N2-bis-
methylated analogue Cbz-Gly[CCONMe]-azaAla-NH2 (9)
did, is in agreement with the sterically demanding substitution
of the latter compound. Cbz-Gly-azaAla-NH2 (7) showed
chemical equivalence of the Gly methylene protons, in
contrast to Cbz-Gly-azaSar-NH2 (8), although both molecules
bear one methylated nitrogen. Thus, the configuration of the
CO–N bonds in these peptides was inspected.
Crystal structures have shown the Z,Z configuration of
diformylhydrazine.9 Accordingly, we postulate the CO–N1
and the N2–CO bonds in the studied azadipeptides to be Z
configured,10 provided that the nitrogen atoms are not methylated
(R1 = H and R2 = H, respectively). Moreover, the Z
configuration of the CO–N1 bond in Cbz-Gly-azaAla-NH2
(7) was proved by X-ray crystallography (see ESIw). In the case
of N-methyl substitution of the CO–N1 and N2–CO bond
(R1 = Me and R2 = Me, respectively), it is assumed that these
are mainly E configured,11 as evidenced by NOE correlations
between the methyl group at N2 and one glycine methylene
proton of 9 in DMSO-d6 (see ESIw). The crystal structures of
Cbz-Gly-azaAla-NH2 (7) and Cbz-Gly[CCONMe]-azaAla-NH2
(9) further confirm this assumption.
The major E configuration of methylated CO–N bonds is in
agreement with an examination of aza-amino acids by using
ab initio MO and DFT methods.2b The authors calculated
a
stabilized E configuration of the methyl derivatives
Ac-azaAla-NHMe and Ac-azaSar-NHMe, in comparison with
Ac-azaGly-NHMe. Moreover, carbohydrazides preferentially
adopt the E configuration of the CO–N bond when the
nitrogen atoms are fully substituted.12 The E configuration
Fig. 1 Azadipeptide amides with free (6, 7) and restricted (8, 9; aS
enantiomers depicted) rotation around the N–N bond.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 5772–5774 5773

View Online
a sterically more demanding second methyl group at N2 led to
compound 12, which did not display a geminal coupling of the
Gly methylene protons. As the N2 substituents in 12 are
identical (R2 = R3 = Me), no atropisomers can be formed
as a matter of principle. Further enlargement of the substitu-
tion was achieved in 13 with an ethyl and a methyl group
(R2 a R3) at the terminal N2 nitrogen. This ordinary carbo-
hydrazide already showed diastereotopicity. A VT NMR
(300 MHz) experiment with compound 13 was performed,
yielding a rotational barrier of 87 kJ molÀ1 (see ESIw).
The minimum energy conformations of selected compounds
(6–9, 11, 13) were computed at the non-local density func-
tional level of theory, and the Gibbs free energy barriers of
rotation around the N–N axis were calculated. Two rotational
barriers were obtained in the cases of 9, 13 and 8. The energies
of the corresponding transition states are 121 and 124 kJ molÀ1
(9), 90 and 92 kJ molÀ1 (13) and 57 and 90 kJ molÀ1 (8). These
values reflect the decreasing conformational restriction with
respect to the N–N rotation as determined for 9, 13 and 8 in
the VT NMR experiments. Our NMR results regarding the
chemical equivalence of the Gly methylene protons of 6, 7 and
11 are also in agreement with the calculations. No barrier of
rotation in the case of 6 and only one (48 kJ molÀ1) in the case
of 7 was found. Two rotational barriers for 11 (42 and
62 kJ molÀ1) were determined. However, the energy barrier
of 42 kJ molÀ1 is not high enough to introduce diastereotopicity
at 303 K.
The authors are grateful to C. Schmidt and S. Terhart-
Krabbe for assistance. G.S. thanks Prof. A. C. Filippou for
support. This work has been supported by the German
Research Foundation (GRK 804).
Notes and references
¨
1 (a) O. D. Ekici, Z. Z. Li, A. J. Campbell, K. E. James, J. L. Asgian,
J. Mikolajczyk, G. S. Salvesen, R. Ganesan, S. Jelakovic,
M. G. Grutter and J. C. Powers, J. Med. Chem., 2006, 49, 5728;
¨
(b) R. Loser, M. Frizler, K. Schilling and M. Gutschow, Angew.
¨
¨
Chem., Int. Ed., 2008, 47, 4331; (c) C. Proulx and W. D. Lubell,
Org. Lett., 2010, 12, 2916; (d) M. Frizler, F. Lohr, M. Lulsdorff
¨
and M. Gutschow, Chem.–Eur. J., 2011, 17, 11419; (e) C. Proulx,
¨
D. Sabatino, R. Hopewell, J. Spiegel, Y. G. Ramos and
W. D. Lubell, Future Med. Chem., 2011, 3, 1139; (f) D. Sabatino,
C. Proulx, P. Pohankova, H. Ong and W. D. Lubell, J. Am. Chem.
Soc., 2011, 133, 12493.
2 (a) M. Thormann and H.-J. Hofmann, THEOCHEM, 1999,
469, 63; (b) H.-J. Lee, J.-W. Song, Y.-S. Choi, H.-M. Park and
K.-B. Lee, J. Am. Chem. Soc., 2002, 124, 11881; (c) H.-J. Lee,
J.-H. Kim, H.-J. Jung, K.-Y. Kim, E.-J. Kim, Y.-S. Choi and
C.-J. Yoon, J. Comput. Chem., 2004, 25, 169; (d) C. H. Reynolds
and R. E. Hormann, J. Am. Chem. Soc., 1996, 118, 9395;
(e) H.-J. Lee, M.-H. Lee, Y.-S. Choi, H.-M. Park and K.-B. Lee,
THEOCHEM, 2003, 631, 101.
3 (a) R. Vanderesse, V. Grand, D. Limal, A. Vicherat, M. Marraud,
C. Didierjean and A. Aubry, J. Am. Chem. Soc., 1998, 120, 9444;
(b) C. Hemmerlin, M. T. Cung and G. Boussard, Tetrahedron
Lett., 2001, 42, 5009; (c) H.-J. Lee, K.-H. Choi, I.-A. Ahn, S. Ro,
H. G. Jang, Y.-S. Choi and K.-B. Lee, J. Mol. Struct., 2001,
569, 43; (d) S. Ro, H.-J. Lee, I.-A. Ahn, D.-K. Shin, K.-B. Lee,
C.-J. Yoon and Y.-S. Choi, Bioorg. Med. Chem., 2001, 9, 1837;
(e) R. L. Broadrup, B. Wang and W. P. Malachowski, Tetra-
hedron, 2005, 61, 10277.
4 (a) M. O. Forster and W. B. Saville, J. Chem. Soc. Abstr., 1920,
117, 753; (b) K. A. Jensen, Acta Chem. Scand. Ser. B, 1977, 31, 145.
5 C. Didierjean, A. Aubry, F. Wyckaert and G. Boussard, J. Pept.
Res., 2000, 55, 308.
As compound 9 forms two atropenantiomers according to
line-shape analysis and theoretical calculations, introduction
of a further chiral element leads to diastereomers. Accordingly,
we have prepared Cbz-L-Phe-Gly[CCONMe]-azaAla-NH2
(14) and (1S,3S,4R-menthyloxycarbonyl)-Gly[CCONMe]-
azaAla-NH2 (15). Moreover, the Gly fragment of 9 was
replaced by L-Phe to obtain Cbz-L-Phe[CCONMe]-azaAla-
NH2 (16). In fact, these three peptides displayed a doubling of
1H and 13C NMR resonances (see ESIw).
To prove atropisomerism in azadipeptides by a further
technique, chiral HPLC of Cbz-Gly[CCONMe]-azaAla-NH2
(9) was performed. The existence of two different atropenan-
tiomers could be concluded from the chromatogram, whereas
achiral HPLC gave only one peak. As expected, the related
but, due to two identical substituents, non-chiral compound 12
showed only one peak under chiral as well as achiral HPLC
conditions. Cbz-L-Phe[CCONMe]-azaAla-NH2 (16), as a pair
of atropdiastereomers, gave two peaks in chiral and achiral
chromatography (see ESIw).
6 (a) M. P. Coogan and S. C. Passey, J. Chem. Soc., Perkin Trans. 2,
2000, 2060. For seminal works on rotational barriers of hydrazine
derivatives, see; (b) G. J. Bishop, B. J. Price and I. O. Sutherland,
Chem. Commun., 1967, 672; (c) J. R. Fletcher and I. O. Sutherland,
J. Chem. Soc. D, 1969, 706.
7 Atropisomers exhibit rotational barriers higher than about 95 kJ molÀ1
at room temperature. (a) M. Oki, Top. Stereochem., 1983, 14, 1;
(b) C. Wolf, Dynamic Stereochemistry of Chiral Compounds, RSC
Publishing, Cambridge, UK, 2008.
8 (a) J. Clayden, W. J. Moran, P. J. Edwards and S. R. LaPlante,
Angew. Chem., Int. Ed., 2009, 48, 6398; (b) B. Paul, G. L. Butterfoss,
M. G. Boswell, P. D. Renfrew, F. G. Yeung, N. H. Shah, C. Wolf,
R. Bonneau and K. Kirshenbaum, J. Am. Chem. Soc., 2011,
133, 10910; (c) S. R. LaPlante, P. J. Edwards, L. D. Fader,
A. Jakalian and O. Hucke, ChemMedChem, 2011, 6, 505;
(d) S. R. LaPlante, L. D. Fader, K. R. Fandrick, D. R. Fadrick,
O. Hucke, R. Kemper, S. P. F. Miller and P. J. Edwards, J. Med.
Chem., 2011, 54, 7005; (e) H. Takahashi, S. Wakamatsu, H. Tabata,
T. Oshitari, A. Harada, K. Inoue and H. Natsugari, Org. Lett., 2011,
In summary, we have synthesized and analyzed structurally
reduced model azapeptides. Diastereotopicity was applied as a
sensor of chirality to reveal atropisomerism which originates
from a restricted rotation around the N–N bond. The E configu-
ration of the CO–NMe bond in such peptides causes the hindered
single-bond rotation. The stereochemical stability of atrop-
isomers increases with the size of the aza-amino acid residue.
The experimental results of this study may sensitize other
researchers in the field of drug discovery or peptide engineering
to the phenomenon of atropisomerism in azapeptides.
13, 760; (f) H. Steinmetz, K. Gerth, R. Jansen, N. Schlager, R. Dehn,
¨
¨
S. Reinecke, A. Kirschning and R. Muller, Angew. Chem., Int. Ed.,
2011, 50, 532; (g) E. W. D. Burke, G. A. Morris, M. A. Vincent,
I. H. Hillier and J. Clayden, Org. Biomol. Chem., 2012, 10, 716.
9 (a) H. Hope and T. Ottersen, Acta Crystallogr., Sect. B, 1978,
34, 3623; (b) K. Tanaka, Acta Crystallogr., Sect. B, 1978, 34, 2487;
(c) M. Eisenstein, Acta Crystallogr., Sect. B, 1979, 35, 2614.
10 H.-J. Lee, H. J. Jung, J. H. Kim, H.-M. Park and K.-B. Lee, Chem.
Phys., 2003, 294, 201.
11 J. Chatterjee, C. Gilon, A. Hoffman and H. Kessler, Acc. Chem.
Res., 2008, 41, 1331.
12 E. Licandro and D. Perdicchia, Eur. J. Org. Chem., 2004, 665.
c
5774 Chem. Commun., 2012, 48, 5772–5774
This journal is The Royal Society of Chemistry 2012