Metal-free, regioselective triazole ligations that deliver locked cis peptide mimetics

Full text of DOI:10.1002/anie.200806390

Communications
DOI: 10.1002/anie.200806390
Biocompatible Ligations
Metal-Free, Regioselective Triazole Ligations that Deliver Locked cis
Peptide Mimetics**
Ahsanullah, Peter Schmieder, Ronald Kꢀhne, and Jꢁrg Rademann*
Chemical ligation reactions have become popular during
recent years as they enable formation of chemoselective
linkages under mild, often aqueous reaction conditions, and
with high yields.[1] Chemical ligations are especially useful if
they can couple complex functionalized molecules without
protecting groups under physiological conditions or in the
presence of living cells. Reversible ligations have been used
successfully in fragment-based screening and drug discov-
ery.[2] A ligation strategy furnishing conformationally locked,
biologically active molecules would be an exciting extension
of current methodology. Such a strategy could be especially
valuable for the structure-based design and synthesis of
inhibitors of protein–protein interactions and peptidyl iso-
merases.[3]
Our aim was to establish the direct incorporation of
1,5-disubstituted triazole rings into peptides starting from
standard amino acid building blocks. A reaction yielding
these products would allow straightforward access to locked
cis peptide mimetics, or “clack peptides” (Scheme 1).[4] The
5-peptidyl-(1H-1,2,3-triazol-1-yl) peptides thus obtained
should allow the study and exploitation of the effects of cis
peptide geometry, either for open-chain conformations (O) or
for peptide turns (T1 and T2) induced by the cis peptide bond
(Scheme 1).[5] Moreover, the 5-peptidyl-1H-1,2,3-triazoles
(NH triazoles) that are unsubstituted in the 1-position,
which have not been prepared synthetically until now, might
serve as proteolytically stable bioisosters of the C termini of
peptides and proteins (Scheme 1, bottom).
1,2,3-Triazoles are attractive ligation products as they are
thermodynamically and physiologically stable. Accordingly,
they are found in several orally administered drugs.[6,7]
1,3-Dipolar cycloadditions that deliver 1,2,3-triazoles from
alkynes and azides were first reported by Huisgen.[8] These
additions became especially popular after they were found to
proceed regioselectively with a base and with copper(I) as
Scheme 1. Replacement of a peptide amide bond with 1,5-disubsti-
tuted 1,2,3-triazole, furnishing a 5-peptidyl-1H-1,2,3-triazolyl peptide as
a “locked cis peptide mimetic” or “clack peptide”. It is expected that
cis peptide mimetics would populate either open-chain conformations
(O) or turn structures (e.g. T1, T2). 5-Peptidyl-1H-1,2,3-triazoles (“NH-
triazoles”) are expected to constitute an enzymatically stable C termi-
nus with a H-bonding donor and acceptor that is similar to the wild
type.
catalyst, yielding 1,4-disubstituted 1,2,3-triazoles.[9] In con-
strast, 1,5-disubstituted 1,2,3-triazoles can be formed regio-
selectively from alkynes and azides using a RuII catalyst.[10]
Unfortunately, the published methods for regioselective
triazole ligations, both for the 1,4- and 1,5-isomers, rely on
the use of heavy metal salts, thus excluding applications in the
presence of living cells. Therefore, the development of a
metal-free, and thus biocompatible, regioselective triazole
ligation method was considered as highly attractive target.[11]
We envisioned that 5-peptidyl-(1H-1,2,3-triazol-1-yl) pep-
tides should be accessible by solid-phase synthesis starting
from the polymer-supported phosphoranylidene acetate 1
(Scheme 2). Racemization-free C-acylations of polymer-
bound phosphoranylidene acetates with Fmoc-protected
[*] M. Phil. Ahsanullah, Dr. P. Schmieder, Dr. R. Kꢀhne,
Prof. Dr. J. Rademann
Leibniz Institute for Molecular Pharmacology (FMP)
Robert-Rꢁssle-Strasse 10, 13125 Berlin (Germany)
Fax: (+49)30-9406-2981
E-mail: rademann@fmp-berlin.de
M. Phil. Ahsanullah, Prof. Dr. J. Rademann
Institute for Chemistry and Biochemistry, Free University Berlin
Takustrasse 3, 14195 Berlin (Germany)
[**] We gratefully acknowledge the Higher Education Commission
Pakistan and DAAD for a stipend granted to A. and the DFG
(RA895/2, FOR 806, and SFB 765). J.R. thanks the Fonds der
Chemischen Industrie for continuous support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200806390.
5042
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 5042 –5045

Angewandte
Chemie
reactions have been calculated, thus explaining the observed
1,5-selectivity by interaction of the olefin HOMO with the
azide LUMO.[14c]
This reaction mode suggests a negative effect of azide
electron density on the reaction rate. Thus we investigated the
scope of this reaction by using azides with different electron
densities (Scheme 2, Table 1). Azides substituted with elec-
tron-withdrawing substitutents, such as sulfonyl and acyl,
reacted rapidly in dichloromethane and yielded the 1,5-
disubstitued triazolyl sulfonamides, and carboxamides,
respectively. 4-Nitrobenzoyl-azide (6) led to the 1-acyl-1,2,3-
triazole 16. The aromatic azide 4-azidobenzoic acid (7)
required prolonged reaction times to yield the 1-aryl-5-
peptidyl triazole 17 with high purity and yield. Employing
trifluoromethanesulfonyl azide (TfN3; 5) as dipole directly
furnished 5-peptidyl-1H-1,2,3-triazole 18 after aqueous work-
up. Aliphatic azides, such as benzylazide (8), methyl-2-azido-
acetate (9), and 2-azido-acetamide (10), did not react in
dichloromethane, even after several days. More polar sol-
vents, such as tetrahydrofuran or DMF, however, enabled
efficient reactions at slightly elevated temperatures, cleanly
furnishing triazoles 19–21. The observed reactivity trends of
azides and the effect of solvent polarity are in accordance with
a polar transition state and with a LUMO-controlled 1,3-
dipolar cycloaddition reaction.
The polymer-supported peptidyl phosphorane 3 also has
the potential for direct access to 5-peptidyl-(1,2,3-triazol-1-yl)
peptides. To verify this hypothesis, 2-azido acids were
synthesized from unprotected amino acids by a diazo transfer
reaction employing trifluoromethanesulfonyl azide 5.[15] The
N-terminal azidopeptide amides 11–13 were prepared on
Rink resin using the respective 2-azido acids in the final
coupling step, whereas N-terminal azidopeptide acid 14 was
prepared on 2-chlorotrityl resin. Azidopeptides 11–14 were
reacted in excess with the peptidylphosphorane under the
conditions defined for electron-rich azides, and the desired
peptidyl triazolyl peptides 22–25 were isolated in good yields
by reverse-phase HPLC.
Scheme 2. Reaction conditions: a) Fmoc-Leu-OH and MSNT, lutidine
in CH2Cl2; or: BTFFH, DIPEA, DMF (R=TMSE or tert-butyl), 14 h;
b) 20% piperidine/DMF; c) Fmoc-Phe-OH, DIC, HOBT, DMF, 2 h;
d) 20% Ac2O/DMF; e) TAS-F/DMF or TFA/CH2Cl2 (R=TMSE or tert-
butyl); f) 4–14, CH2Cl2, or THF. MSNT=1-(mesitylene-2-sulfonyl)-3-
nitro-1H-1,2,4-triazole, BTFFH=bis(tetramethylen)fluoroformamidi-
nium hexafluorophosphate, DIPEA=N,N-diisopropylethylamine,
TMSE=trimethylsilylethyl, HOBt=1-hydroxybenzotriazole, TAS-F=
tris(dimethylamino)sulfonium difluorotrimethylsilicate, TFA=tri-
fluoroacetic acid.
amino acids have been recently shown to yield Fmoc amino
acyl phosphoranylidene acetates 2.[12,13] Following Fmoc
cleavage of 2, the peptide could be elongated via the free
amino group by standard peptide chemistry. Deprotection of
the trimethylsilylethyl (TMSE) or tert-butyl ester group
furnished the decarboxylated peptidyl phosphorane 3 as the
only product.
All the new products were purified by column chroma-
tography if necessary, and were characterized by HR-MS and
by full assignment of their 1H and 13C NMR spectra. The
potential of peptidyl triazolyl peptides to form stable
conformations in solution was investigated with product 22
in DMSO using NMR spectroscopy. A complete assignment
Herein we report reactions of peptidyl phosphorane 3
with azides 4–14. At room temperature, polymer 3 reacted
smoothly with 4-toluenesulfonyl azide 4, furnishing the
5-peptidyl-(1H-1,2,3-triazol-1-yl) tosylate 15 in high yield.
The product was obtained in excellent purity solely by
washing the resin, followed by evaporation of the solvent.
The second product, triphenylphosphane oxide, remained
attached to the polymer. Neither epimerized byproducts nor
traces of the 1,4-substituted triazole were detected. Product
formation can be rationalized by a 1,3-dipolar cycloaddition
reaction by considering the phosphorous ylide 3a as phos-
phonium enolate in accordance with the resonance structure
3b. The cyclic intermediate can be formed either by a
concerted or a stepwise mechanism. To the best of our
knowledge, this reaction of peptidyl phosphoranes has not
been reported to date, and is also the first account of a dipolar
cycloaddition of phosphoranes on a polymeric support.
Earlier work has reported 1,3-dipolar cycloadditions with
electron-rich olefins,[14] and the transition states of such
1
of all the H and 13C resonances was obtained from a set of
two-dimensional spectra. Based on this assignment, distances
were extracted from a 2D-ROESY experiment. More than 30
distances could be obtained in this way, and they were used as
distance restraints in a molecular dynamics simulation. Sets of
structures were calculated by a simulated annealing proce-
dure. An overlay of ten calculated structures indicated
consistently a turn-like bend of the triazole-flanking amino
acid residues (Figure 1), which was characterized by the
inward orientation of the À1 amide carbonyl group and the
+ 1 amide NH group (numbering in N- to C-terminal
direction). Therefore, the preferred conformation of peptidyl
triazol-1-yl peptide 22 in DMSO solution most closely
resembled the T2 structure in Scheme 1. (Precise rotational
angles and their corresponding standard deviations are
Angew. Chem. Int. Ed. 2009, 48, 5042 –5045
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5043

Communications
Table 1: (Continued)
Table 1: Selected triazole ligation products.
Entry Product
Purity[c] [%]
Yield [%]
Purity[c] [%]
Entry
Product
Yield [%]
10
74[b]
97
1
2
3
88[a]
96
95[a]
74[d]
11
73[b]
98
[a] Reaction in CH2Cl2 at RT. [b] Reaction in THF at 608C. [c] HPLC
purities were determined at 220 nm with UV/Vis spectroscopy. Com-
pounds 15 and 17–21 were more than 80% pure in crude products after
trituration with acetonitrile. [d] Purity of the crude product; compound
16 degraded partially to 18 during purification or storage.
80[a]
92
4
5
88[a]
92[d]
62[b]
88
6
7
70[b]
91
95
72[b]
8
68[b]
82
96
Figure 1. The solution structure of 5-peptidyl-triazol-1-yl peptide 22 as
determined in DMSO by ROESY NMR spectroscopy. Measured aver-
age distances were used as distance constraints in a molecular
dynamics simulation. An overlay of 10 structures obtained by simu-
lated annealing is shown. All the structures display a bent-like turn
structure. Amino acids flanking the triazole motif are relatively rigid,
with the carbonyl of the À1 amino acid and the NH of the +1 amino
acid pointing inwards in all ten structures.
9
72[b]
5044 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 5042 –5045

Angewandte
Chemie
summarized in the Supporting Information). More detailed
studies concerning the effects of elongated peptide tails, of
amino acid stereochemistry, and cyclization of 1,5-triazolyl
peptides are under way and will deliver a more comprehen-
sive picture of the potential of this compound class in
conformation control.
In summary, we have presented the first metal-free
regioselective triazole ligation. The reaction has a broad
scope and, most relevantly, allows full integration into peptide
synthesis, thus avoiding the use of tediously prepared amino
acid alkynes. The new triazole ligation can be employed to
yield products with carefully controlled conformations.
Details of structural investigations into the ligation products
will be reported in due course.
Keywords: 1,3-dipolar cycloaddition · peptide mimetics ·
phosphoranes · structural biology · triazoles
.
[1] a) C. P. R. Hackenberger, D. Schwarzer, Angew. Chem. 2008,
120, 10182 – 10228; Angew. Chem. Int. Ed. 2008, 47, 10030 –
10074; b) S. B. K. Kent, Chem. Soc. Rev. 2009, 38, 338 – 351;
c) M. Kꢁhn, R. Breinbauer, Angew. Chem. 2004, 116, 3168 –
3178; Angew. Chem. Int. Ed. 2004, 43, 3106 – 3116; d) J.
Rademann, Angew. Chem. 2004, 116, 4654 – 4656; Angew.
Chem. Int. Ed. 2004, 43, 4554 – 4556.
[2] M. Schmidt, A. Isidro-Llobet, A. El-Dahshan, M. Lisurek, J.
Tan, R. Hilgenfeld, J. Rademann, Angew. Chem. 2008, 120,
3319 – 3323; Angew. Chem. Int. Ed. 2008, 47, 3275 – 3278.
[3] V. Uryga-Polowy, D. Kosslick, C. Freund, J. Rademann, Chem-
BioChem 2008, 9, 2452 – 2462.
[4] As the term “click peptides” has been adopted for 1,4-
disubstitued triazolyl peptides, we suggest calling 5-peptidyl-
(1H-1,2,3-triazol-1-yl) peptides that are cis peptide mimetics as
“clack peptides”.
Experimental Section
1
Synthetic procedures and analytical data (HRMS; H, 13C NMR) of
all new compounds are given in the Supporting Information.
Synthesis of a 5-peptidyl-1H-1,2,3-triazole: 18: An excess of freshly
prepared trifluoromethanesulfonyl azide 5 (3 equiv) in CH2Cl2 (5 mL)
was added to phosphorane resin 3 (200 mg, 0.254 mmol), and the
mixture was stirred for 5 h at room temperature. The polymer support
was separated by filtration and washed with CH2Cl2. The solvent were
removed under vacuum and the crude product was dissolved in
acetonitrile. Water (1 mL) was added, and the mixture was lyophilized
to deliver 18 as a pale yellow solid (99 mg, 88%). 1H NMR (300 MHz,
CD3CN): d=0.83, 0.88 (2d, J= 6.1, 6.7, 6H, 2CH3, Leu), 1.46–1.63 (m,
3H, CH, CH2, Leu), 1.86 (s, 3H, CH3, acetyl), 2.86–3.11 (m, 2H, CH2,
Phe), 4.23–4.32 (m, 1H, CH, Phe), 4.55–4.62 (m, 1H, CH, Leu), 6.93–
7.49 ppm (m, 6H, arom.). 13C NMR: (75.5 MHz, CDCl3): d=22.0, 23.2,
25.0, 30.7, 38.5, 46.1, 54.8, 117.7, 122.0, 125.9, 129.0, 129.8, 136.4, 171.2,
177.6 ppm. HRMS (ESI): m/z calcd for C18H25N5O2 [M +H]+:
344.20865; found: 344.20845.
[5] a) M. M. Lenman, S. L. Ingham, D. Gani, Chem. Commun. 1996,
85 – 87; b) Y. Hitotsuyanagi, S. Motegi, H. Fukaya, K. Takeya, J.
Org. Chem. 2002, 67, 3266 – 3271; c) A. Tam, U. Arnold, M. B.
Soellner, R. T. Raines, J. Am. Chem. Soc. 2007, 129, 12670 –
12671; d) S. J. Coats, J. S. Link, D. Gauthier, D. J. Hlasta, Org.
Lett. 2005, 7, 1469 – 1472. e) A. Paul, H. Bittermann, P. Gmeiner,
Tetrahedron 2006, 62, 8919 – 8927. Others have used tetrazoles as
cis peptide mimetics: f) R. J. Nachman, J. Zabrocki, J. Olczak,
H. J. Williams, G. Moyna, A. I. Scott, G. M. Coast, Peptides 2002,
23, 709 – 716.
[6] For triazoles in bioactive compounds, see the world drug index
(WDI): http://scientific.thomsonreuters.com/products/wdi/.
[7] For a review concerning applications of regioselective triazole
formations, see: M. Meldal, C. W. Tornøe, Chem. Rev. 2008, 108,
2952 – 3015.
Synthesis of
a 5-peptidyl-(1H-1,2,3-triazol-1-yl)-peptide: 22:
[8] R. Huisgen, Proc. Chem. Soc. London 1961, 357 – 369.
[9] a) C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002,
67, 3057 – 3064; b) V. V. Rostovtsev, L. G. Green, V. V. Fokin,
K. B. Sharpless, Angew. Chem. 2002, 114, 2708 – 2711; Angew.
Chem. Int. Ed. 2002, 41, 2596 – 2599.
[10] L. Zhang, X. Chen, P. Xue, H. H. Y. Sun, I. D. Williams, K. B.
Sharpless, V. V. Fokin, G. Jia, J. Am. Chem. Soc. 2005, 127,
15998 – 15999.
[11] Strained cycloalkynes have been reported to yield triazoles
without heavy metal catalysis. Regioselectivity is, however, lost
in these special cases, which require considerable synthetic
effort: a) J. M. Baskin, J. A. Prescher, S. T. Laughlin, N. J. Agard,
P. V. Chang, I. A. Miller, A. Lo, J. A. Codelli, C. R. Bertozzi,
Proc. Natl. Acad. Sci. USA 2007, 104, 16793 – 16797; b) J. A.
Codelli, J. M. Baskin, N. J. Agard, C. R. Bertozzi, J. Am. Chem.
Soc. 2008, 130, 11486 – 11493.
[12] S. Weik, J. Rademann, Angew. Chem. 2003, 115, 2595 – 2598;
Angew. Chem. Int. Ed. 2003, 42, 2491 – 2494.
[13] A. El-Dahshan, S. Weik, J. Rademann, Org. Lett. 2007, 9, 12670 –
12671.
[14] a) K. N. Houk, J. Sims, C. R. Watts, L. J. Luskus, J. Am. Chem.
Soc. 1973, 95, 7301 – 7315; b) G. R. Harvey, J. Org. Chem. 1966,
31, 1587 – 1590; c) P. Ykman, G. Mathys, G. LꢂAbbꢃ, G. Smets, J.
Org. Chem. 1972, 37, 3213 – 3216; d) L. Cafici, T. Pirali, F.
Condorelli, E. Del Grosso, A. Massarotti, G. Sorba, P. L.
Canonico, G. C. Tron, A. A. Genazzani, J. Comb. Chem. 2008,
10, 732 – 740.
Phosphorane resin 3 (100 mg, 0.127 mmol) was pre-swollen in THF
(1 mL) in a glass vial, and azidopeptide 11 (1.5 equiv, 0.19 mmol),
dissolved in THF (1 mL), was added to the vial. The mixture was
heated overnight at 608C in sealed glass vial. After cooling to room
temperature, the polymer support was filtered off and washed with
THF and CH2Cl2. Solvents were removed under vacuum, and 22 was
isolated by preparative reverse-phase HPLC to remove remains of
the azidopeptide reagent. Compound 22 was obtained as a white
lyophilisate (48 mg, 68%). 1H NMR: (300 MHz, [D6]DMSO): d =
0.82, 0.85 (2d, J = 6.1 Hz, 6H, CH3, Leu), 1.35–1.52 (m, 2H, Leu),
1.60 (d, 3H, J = 7.3 Hz, CH3, Ala), 1.65–1.67 (m, 1H, CbH, Leu), 1.73
(s, 3H, CH3, acetyl), 2.75–3.02 (m, 4H, 2CH2, Phe), 1.53–1.87 (m, 3H,
CH2, CH, Leu), 2.50–3.31 (m, 4H, 2CH2, Phe), 4.39–4.49 (m, 2H,
2CH, Phe), 5.04–5.12 (m, 1H, CaH, Leu), 5.43 (q, J = 7.3, 1H, CH,
Ala), 7.02 (d, J = 7.4, 2H, NH2), 7.11–7.24 (m, 10H, arom., Phe), 7.55
(s, 1H, CH, triazole), 8.01, 8.04, 8.53 ppm (3d, J = 7.9, 8.3, 7.9, 3H,
3NH). 13C NMR: (75.5 MHz, [D6]DMSO): d = 17.7, 21.6, 22.3, 22.6,
24.2, 37.2, 41.4, 43.2, 53.7, 54.3, 56.7, 126.1, 127.9, 129.0, 131.6, 137.5,
139.7, 167.5, 169.1, 171.1, 172.2 ppm. HRMS (ESI): m/z calcd for
C30H39N7O4 [M+H]+: 562.31418; found: 562.31419.
The structure of compound 22 was calculated using repeated
simulated annealing with a heating phase of 2000 fs up to a
temperature of 1000 K followed by an exponential cooling for
10000 fs up to 0 K, and using NMR-derived range constraints with a
force constant of 200 kcalmolÀ1. In all structures presented in
Figure 1, the range constraints were violated less than 0.2 ꢀ. All
calculations were carried out using SYBYL7.3 (Tripos Inc.).
[15] J. T. Lundquist IV, J. C. Pelletier, Org. Lett. 2001, 3, 781 – 783.
Received: December 31, 2008
Revised: February 13, 2008
Published online: May 26, 2009
Angew. Chem. Int. Ed. 2009, 48, 5042 –5045
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5045