A nonpeptidic reverse-turn scaffold stabilized by urea-based dual intramolecular hydrogen bonding

Article abstract of DOI:10.1021/ol201247x

A novel nonpeptidic reverse-turn scaffold containing urea fragments that are connected by a conformationally constrained d-prolyl-cis-1,2- diaminocyclohexane (d-Pro-DACH) linker is reported. The scaffold adopts a well-defined reverse-turn conformation tha

Full text of DOI:10.1021/ol201247x

ORGANIC
LETTERS
2011
Vol. 13, No. 13
3486–3489
A Nonpeptidic Reverse-Turn Scaffold
Stabilized by Urea-Based Dual
Intramolecular Hydrogen Bonding
Amiya K. Medda,† Chul Min Park,†,§ Aram Jeon,† Hyunwoo Kim,† Jeong-Hun Sohn,*,‡
and Hee-Seung Lee*,†
Molecular-Level Interface Research Center, Department of Chemistry, KAIST,
Daejeon 305-701, Korea, and Chungnam National University, Daejeon 305-764, Korea
hee-seung_lee@kaist.ac.kr; sohnjh@cnu.ac.kr
Received May 10, 2011
ABSTRACT
A novel nonpeptidic reverse-turn scaffold containing urea fragments that are connected by a conformationally constrained D-prolyl-cis-1,2-
diaminocyclohexane (D-Pro-DACH) linker is reported. The scaffold adopts a well-defined reverse-turn conformation that is stabilized by dual
intramolecular hydrogen bonding in both solution and solid states.
Reverse turns are found abundantly in biologically
active peptides and globular proteins. They serve as mol-
ecular recognition sites in many biological events due to
their potent binding and exquisite selectivity.1 Numerous
efforts have been made to study reverse-turn scaffolds in
order to develop potent therapeutic agents2 and to identify
loop segments that can induce autonomous β-sheet fold-
ing.3 In addition, it has been demonstrated that short-length
peptide fragments containing reverse turns showed excellent
catalytic activity for asymmetric organic transformations.4
However, the use of peptidic reverse turns for these applica-
tions is often limited because natural peptides are not
resistant to enzymatic degradation and are conformation-
ally flexible in solution. Although many examples of non-
peptidic reverse-turn surrogates have been reported to
overcome the drawbacks,5 it is still challenging to find a
new type of nonpeptidic scaffold that can adopt a highly
populated reverse-turn conformation in solution. Herein,
we report a novel nonpeptidic reverse-turn scaffold, which
contains two urea fragments linked by a conformationally
† KAIST.
§ Korea Research Institute of Chemical Technology (KRICT).
‡ Chungnam National University.
(1) For selected reviews on reverse-turns, see: (a) Richardson, J. S.
Adv. Protein Chem. 1981, 34, 167. (b) Kabsch, W.; Sander, C. Biopoly-
mers 1983, 22, 2577. (c) Rose, G. D.; Gierasch, L. M.; Smith, J. A. Adv.
Protein Chem. 1985, 37, l. (d) Wilmot, C. M.; Thornton, J. M. J. Mol.
ꢀ
ꢀ
Biol. 1988, 203, 221. (e) Vass, E.; Miklos Hollosi, M.; Besson, F.; Buchet,
R. Chem. Rev. 2003, 103, 1917.
(2) (a) Giannis, A.; Kolter, T. Angew. Chem., Int. Ed. 1993, 32, 1244.
(b) Suat Kee, K.; Jois, S. D. S. Curr. Pharm. Des. 2003, 9, 1209.
(c) Patarroyo, M. E.; Patarroyo, M. A. Acc. Chem. Res. 2008, 41, 377.
(d) Hua, Q. Protein Cell 2010, 1, 537.
(3) (a) Gellman, S. H. Curr. Opin. Chem. Biol. 1998, 2, 717. (b) Searle,
M. S.; Ciani, B. Curr. Opin. Struct. Biol. 2004, 14, 458. (c) Hughes, R. M.;
Waters, M. L. Curr. Opin. Struct. Biol. 2006, 16, 514.
(4) (a) Miller, S. J. Acc. Chem. Res. 2004, 37, 601. (b) Gilbertson,
S. R.; Collibee, S. E.; Agarkov, A. J. Am. Chem. Soc. 2000, 122, 6522.
(b) Greenfield, S. J.; Agarkov, A.; Gilbertson, S. R. Org. Lett. 2003, 5,
3069. (c) Wu, F.-C.; Da, C.-S.; Du, Z. ꢀX.; Guo, Q.-P.; Li, W.-P.; Yi, L.;
Jia, Y.-N.; Ma, X. J. Org. Chem. 2009, 74, 4812.
(5) For selected reviews on reverse-turn mimetics, see: (a) Gante, J.
Angew. Chem., Int. Ed. 1994, 33, 1699. (b) Souers, A. J.; Ellman, J. A.
Tetrahedron 2001, 57, 7431. (c) Eguchi, M.; Kahn, M. Mini-Rev. Med.
Chem. 2002, 2, 447. (d) Trabocchi, A.; Menchi, G.; Guarna, F.;
Machetti, F.; Scarpi, D.; Guarna, A. Synlett 2006, 331. (e) Hershberger,
S. J.; Lee, S.-G.; Chmielewski, J. Curr. Top. Med. Chem. 2007, 7, 928.
(f) Hirschmann, R. F.; Nicolaou, K. C.; Angeles, A. R.; Chen, J. S.;
Smith, A. B., III. Acc. Chem. Res. 2009, 42, 1511. (g) Ross, N. T.; Katt,
W. P.; Hamilton, A. D. Philos. Trans. R. Soc. London, Ser. A 2010, 368,
989. (h) Smith, A. B., III; Charnley, A. K.; Hirschmann, R. Acc. Chem.
Res. 2011, 44, 180.
r
10.1021/ol201247x
Published on Web 06/08/2011
2011 American Chemical Society

constrained D-prolyl-cis-1,2-diaminocyclohexane (D-Pro-
DACH). The scaffold was envisioned to adopt a reverse-
turn conformation via dual intramolecular hydrogen bond-
ing in both solution and solid states.
Scheme 1. Summary of the Synthetic Procedures of Nonpeptidic
Reverse-Turn Candidates (1ꢀ6)a
Figure 1. Conventional peptide reverse-turn (β-turn) structure
(left) and nonpeptidic reverse-turn scaffold stabilized by urea-
based dual intramolecular hydrogen bonds (right).
Conventional reverse-turns (β-turns) are composed of
tetrapeptide sequences, in which the CR(1);CR(4) distance
˚
is e7 A, mostly with hydrogen bonding between CdO(1)
and N;H(4) that stabilizes the well-defined conformation
(Figure 1).6 To ensure a conformationally stable nonpep-
tidic reverse-turn scaffold, we employed urea functionality
to introduce simultaneous formation of 10- and 12-mem-
bered CdO---H;N hydrogen bonds.7 Due to the dual
hydrogen bonding, the scaffold B was expected to adopt a
more stable reverse-turn conformation, being populated
predominantly in solution, than the amide version A. In
this design, we chose a prolyl-1,2-diamine linker to connect
two urea functionalities.7b To find the best combination
between D-/L-proline and 1,2-diamine derivatives, the sub-
stitution pattern on 1,2-diamine was systematically scre-
ened from the unsubstituted one to the conformationally
rigid cis-1,2-DACH.
We prepared a series of reverse-turn candidates as sum-
marized in Scheme 1. Each starting material was converted
to the corresponding isocyanate which was then treated
with primary amines (CH3NH2 for 1ꢀ5 and t-BuNH2 for
6; see Supporting Information (SI) for details) to synthe-
size the bottom fragments A1ꢀA6. After removal of Boc
followed by coupling with Boc-Pro-OH, the resulting
amides were sequentially treated with TFA and isocya-
nates (benzyl for 1ꢀ5 and cyclohexyl for 6) to obtain the
desired turn candidates. Reference compounds, 7 (an
amide version of 1), 8a,8 and 8b9 (upper and bottom
fragments), were also prepared (see SI).
a All substituents (R1, R2, R3, R4) and the stereochemistries of
prolines are given in Table 1.
To evaluate the extent of the hydrogen bonding in 1ꢀ6,
we determined, at first, the temperature coefficients (ꢀΔδ/
ΔT) for the four NH protons by measuring their chemical
shifts in DMSO-d6 at temperatures ranging from 298 to
348 K (Table 1).10 In all reverse-turn candidates 1ꢀ6, the
values for NH(3) and NH(4) (ꢀΔδ/ΔT < 5) were much
smaller than those for NH(1) and NH(2) (ꢀΔδ/ΔT g 5),
whereas the values for the three NHs in the reference
amide 7 were >5 with NH(2) (5.4) ≈ NH(3) (5.5). The
results indicate that NH(1) and NH(2) are in solvent
accessible positions, and NH(3) and NH(4) are participat-
ing in intramolecular hydrogen bonding.10 Notably, the
values were not strongly influenced by the stereogenic
center of proline (1a vs 1b and 2a vs 2b), which is
different from other reverse-turn motifs such as Pro-
DADME;11 thus we considered only the D-Pro series for
further studies.
For NH(4), a significant change of the coefficients
(0.6ꢀ4.5) was observed upon variation of the substitution
pattern in the diamine linkers. A trend was seen that the
value increases as the flexibility of the diamine linker
increases: 5 (0.6) or 6 (1.6) < 4 (2.5) < 2a (2.9) < 1a
(3.1) < 3 (4.5). The combination of the urea functionality
with a conformationally constrained linker resulted in
remarkably low temperature coefficients for the NH(4) in
5 or 6. The data also show a nonequivalent hydrogen
bonding pattern between NH(3) and NH(4), in which the
(6) (a) Ball, J. B.; Hughes, R. A.; Allewood, P. F.; Andrews, P. R.
Tetrahedron 1993, 49, 3467. (b) Wilmot, C. M.; Thornton, J. M. J. Mol.
Biol. 1988, 203, 221.
(7) (a) Nowick, J. S. Acc. Chem. Res. 1999, 32, 287. (b) Fisk, J. D.;
Powell, D. R.; Gellman, S. H. J. Am. Chem. Soc. 2000, 122, 5443.
(c) Fischer, L.; Didierjean, C.; Jolibois, F.; Semetey, V.; Lozano, J. M.;
Briand, J. -P.; Marraud, M.; Poteau, R.; Guichard, G. Org. Biomol.
Chem. 2008, 6, 2596.
(8) Orito, K.; Miyazawa, M.; Kakamura, T.; Horibata, A.; Ushito,
H.; Nakasaki, H.; Yuguchi, M.; Yamashita, S.; Yamazaki, T.; Tokuda,
M. J. Org. Chem. 2006, 71, 5951.
(9) Artuso, E.; Degani, I.; Fochi, R.; Magistris, C. Synthesis 2007,
3497.
(10) (a) Kessler, H. Angew. Chem., Int. Ed. 1982, 22, 512. (b) Kemp,
D. S.; Bowen, B. R.; Muendel, C. C. J. Org. Chem. 1990, 55, 4650.
(c) Semetey, V.; Rognan, D.; Hemmerlin, C.; Graff, R.; Briand, J.-P.;
Marraud, M.; Guichard, G. Angew. Chem., Int. Ed. 2002, 41, 1893.
(11) Langenhan, J. M.; Guzei, I. A.; Gellman, S. H. Angew. Chem.,
Int. Ed. 2003, 42, 2402.
Org. Lett., Vol. 13, No. 13, 2011
3487

Table 1. Temperature Coefficients of 1ꢀ7 and Chemical Shifts of Selected Compounds
substituents
R2 R3
temperature coefficientsa
chemical shifts (ppm)b,c
Pro
R1
R4
Y
NH(1)
NH(2)
NH(3)
NH(4)
NH(1)
NH(2)
NH(3)
NH(4)
1a
1b
2a
2b
3
D-
L-
D-
L-
D-
D-
CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
NCH3
NCH3
NCH3
NCH3
NCH3
NCH3
7.5
7.0
7.1
7.8
7.8
8.1
5.2
5.1
5.1
5.3
5.9
5.8
3.9
3.4
3.2
3.8
4.3
4.6
3.1
3.5
2.9
2.8
4.5
2.5
CH3
H
CH2Ph
CH2Ph
H
H
H
H
4.80
6.94
5.18
4.45
4
CH3
CH3
4.77
6.00
5.71
4.53
(2.01)
4.72
(1.59)
6.51
(0.37)
5.49
(1.10)
4.63
5
6
D-
D-
D-
-(CH2)4-
-(CH2)4-
CH3
H
H
H
H
H
H
NCH3
N-tBu
CH2
8.1
6.3
7.1
6.6
5.0
5.4
4.7
3.8
5.5
0.6
1.6
(2.21)
4.20
(1.10)
6.56
(0.34)
5.31
(0.82)
4.30
(1.60)
(0.85)
(0.37)
(1.07)
7
H
8a
4.43
(2.19)
8b
4.08
4.08
(1.67)
(1.50)
a Temperature coefficients (ꢀΔδ/ΔT) were measured in DMSO-d6. b Chemical shifts of the amide NHs and urea NHs of 3ꢀ6 and 8a,b (2 mM in
CDCl3 at room temperature). c Values in parentheses denote Δδ (ppm) resulted from DMSO-titration experiments at 298 K.
latter is more sensitive to the linker flexibility than the
former.
Two NOE signals for NH(1)ꢀH(11) and NH(3)ꢀNH(4)
indicate that both urea fragments are in a transꢀtrans
conformation and, thus, the NHs are oriented opposite to
their urea carbonyl groups. A signal between H(14) and
NH(2) signifies the two urea fragments facing each other
for a proper alignment of hydrogen bonding donors
(NH(3) and NH(4)) and acceptor (C(9)dO). The correlation
between the H(17) and H(22) indicates the role of the cis-
cyclohexyldiamine ring, providing an appropriate tor-
sional angle for the reverse-turn conformation. Further-
more, the nonadjacent correlation between NH(4) and H(7)
unambiguously defines the stable turn conformation.13 In
The evidence of dual intramolecular hydrogen bonding
was strengthened by the DMSO-titration experiment.7c
For example, the chemical shifts of NH(3) and NH(4) in 5,
appearing more downfield than those in reference com-
pound 8b under nonaggregating conditions (2 mM in
CDCl3 at room temperature),12 were changed less signifi-
cantly than those of NH(1) and NH(2) in 5 or NHs in
reference compounds when DMSO-d6 was added gradu-
ally to the CDCl3 solution (Table 1). This result suggests
that both NH(3) and NH(4) are involved in intramolecular
hydrogen bonding. Both the temperature coefficient study
and DMSO-titration experiment indicated the existence of
dual intramolecular hydrogen bonding that is mostly
stable in 5.
To confirm the reverse-turn conformation of the scaf-
fold in both solution and solid state, we carried out NMR
and X-ray crystallographic studies. For the solution state
conformation, NOESY experiments for compounds 3ꢀ6
in CDCl3 at 4 ꢀC were performed after all the resonances
were assigned by COSY and TOCSY. The result for
compound 5, for example, is summarized in Figure 2.
Figure 2. Summary of NOE signals for 5 in CDCl3 (400 MHz,
mixing time 300 ms, in 2 mM solutions) at 4 ꢀC.
(12) The plot of concentration vs chemical shift for compound 5
showed that the concentration of 7.78 mM was the nonaggregation limit,
so 2 mM was chosen for the NMR studies.
3488
Org. Lett., Vol. 13, No. 13, 2011

a variational temperature study (from 323 to 223 K), both
NH(3) and NH(4) resonances were gradually shifted to
downfield as temperature went down, which supports
further that 5 adopts a stable reverse-turn conformation
via dual intramolecular hydrogen bonding (see SI). The
overall results from the NMR studies revealed that the
designed reverse-turn scaffold 5 is conformationally stable
in the solution state.
For the solid state confomational analysis of the turn
scaffold, compound 6,14 which showed similar temperature
coefficients and chemical shifts to those of 5, was crystallized
from a mixture of EtOHꢀCH2Cl2 (1:4) by slow evaporation
over a period of two days. The X-ray crystallographic
analysis apparently revealed the existence of dual intra-
˚
molecular hydrogen bonding (dN(1)
= 3.042 A,
O(3)
3 3 3
˚
—
—
= 157.5ꢀ and dN(2)H
= 3.022 A,
N(1)H
N(2)H
O(3)
O(3)
3 3 3
3 3 3
3 3 3
O(3) = 159.6ꢀ) (Figure 3).15,16 The backbone
Figure 3. Crystal structure of6.16 H-bondsshown as dashed lines.
torsion angles χ1 (C(17)ꢀN(4)ꢀC(13)ꢀC(12)) and χ2 (N(3)ꢀ
C(11)ꢀC(6)ꢀN(2)) are þ61.50ꢀ and þ56.42ꢀ, respectively.17
The two urea groups are slightly twisted by 17.7ꢀ, and the
central amide group, HN(3)ꢀCdO(2), is nearly perpendicu-
lar to the mean plane composed of four urea nitrogens
adopted in a reverse-turn conformation via 10,12-mem-
bered dual intramolecular hydrogen bonding in the solid
state,18 as in the solution state.19
˚
(79.24ꢀ). Additionally, the distance of 5.02 A between
C(4) and C(18) is within the criterion of a reverse turn
In summary, we have developed a novel nonpeptidic
reverse-turn scaffold containing urea fragments that are
connected by a conformationally constrained D-prolyl-cis-
1,2-diaminocyclohexane (D-Pro-DACH) linker. Owing to
the dual intramolecular hydrogen bonding, the scaffold
adopts a well-defined reverse-turn conformation, which is
more stable than its amide analogue, in both solution and
solid states. Our reverse-turn motif could serve as an
N-terminus-to-N-terminus linker for developing nonpep-
tidic parallel β-sheets and new reverse-turn mimetics.
˚
(7 A). This result indicates that the scaffold is well
(13) The NOE correlation was also observed in aqueous DMSO-d6
(see SI), whereas the signal for 2b at 5 mM in CDCl3 was not seen.
(14) The high crystalline nature of cyclohexyl and tert-butyl substi-
tuents in β NꢀO turns and helices was reported: Yang, D.; Zhang,
Y.-H.; Zhu, N.-Y. J. Am. Chem. Soc. 2002, 124, 9966.
(15) Crystal data for 6: C23H41N5O3, Mw = 435.32, monoclinic, space
˚
˚
˚
group P2(1), a = 10.1633(2) A, b = 9.9300(2) A, c = 12.1706(3) A, R =
3
˚
90.00ꢀ, β = 96.2770(10)ꢀ, γ = 90.00ꢀ, V = 1220.91(5) A , Z = 2, Fcalcd
=
1.185 g cmꢀ3, μ = 0.080 minꢀ1, T = 130 K, 12037 reflections measured,
independent: 5837 (Rint = 0.017). The structure was solved by direct
methods and refined by full-matrix least-squares on F2; R1 = 0.0357,
wR2 = 0.0904 [ I > 2σ(I)]; Flack(x) parameter = 0.0(8) maximal residual
Acknowledgment. H.-S.L. acknowledges the National
Research Foundation of Korea (NRF) Grant (Basic Re-
search Promotion Fund, 2011-0001318) funded by the
Korean Government (MEST). We thank Mr. Hack Soo
Shin for NMR measurements.
electron density: 0.47 eAꢀ3. CCDC 730130.
˚
(16) The compound numbering is arbitrary and different from those
in the NMR structure.
(17) Other dihedral angles in 6 such as φiþ2 and ψiþ2 that can be seen
in a natural β-turn are not analyzed since cis-1,2-DACH is not origi-
nated from a natural amino acid as per the usual nomenclature conven-
tion: ref 7c and references therein.
(18) For a recent example of an intramolecular H-bond for turn
mimics in the solid state, see: (a) Sacchetti, A.; Silvani, A.; Lesma, G.;
Pilati, T. J. Org. Chem. 2011, 76, 833. (b) Lesma, G.; Landoni, N.; Pilati,
T.; Sacchetti, A.; Silvani, A. J. Org. Chem. 2009, 74, 8098.
(19) An IR study of 6 also supported the existence of the hydrogen
bonding (see SI). In addition, the computational analysis is consistent
with the experimental results (see SI).
Supporting Information Available. Experimental details,
characterization of all new compounds, 1H and 13C NMR
spectra of 1ꢀ8, IR spetra of 6, NOESY spectra of 5,
computational analysis, and CIF file. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
Org. Lett., Vol. 13, No. 13, 2011
3489