Synthesis of 1,5-triazole bridged vancomycin CDE-ring bicyclic mimics using RuAAC macrocyclization

Article abstract of DOI:10.1039/c3cc40628h

Herein we report the Ru(ii)-catalyzed double-macrocyclization of a hexapeptide to obtain a mimic of the bicyclic CDE-ring of vancomycin, followed by measurement of its binding affinity for small peptide ligands using ITC.

Full text of DOI:10.1039/c3cc40628h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Synthesis of 1,5-triazole bridged vancomycin CDE-ring
bicyclic mimics using RuAAC macrocyclization†
Jinqiang Zhang,a Johan Kemmink,a Dirk T. S. Rijkersa and Rob M. J. Liskamp*ab
Cite this: Chem. Commun., 2013,
49, 4498
Received 24th January 2013,
Accepted 27th March 2013
DOI: 10.1039/c3cc40628h
www.rsc.org/chemcomm
Herein we report the Ru(II)-catalyzed double-macrocyclization of a
hexapeptide to obtain a mimic of the bicyclic CDE-ring of vanco-
mycin, followed by measurement of its binding affinity for small
peptide ligands using ITC.
The glycopeptide antibiotic vancomycin is an outstanding example
showing that dramatic reduction of conformational freedom by
peptide side chain cyclization may lead to cavities or shell-like
structures, resulting in a very high binding affinity to its natural
ligand. It is a tremendous challenge to construct these peptide
based cavities by chemical synthesis and the native cavity as is
present in vancomycin is certainly not amenable to large-scale
synthesis and/or to the synthesis of a considerable number of
analogues.1 Therefore, alternative approaches for the construction
of molecules mimicking the cavity of vancomycin are very attractive.
In our on-going quest towards uncovering promising alter-
natives for the biaryl ether bridge in vancomycin2 and related
glycopeptide antibiotics, we recently found that the biaryl ether
bridge of the vancomycin DE-ring can be mimicked by either a
1,4- or a 1,5-disubstituted triazole functionality, accessible using
Cu(I)- or Ru(II)-catalyzed azide alkyne cycloaddition chemistry
(CuAAC3 or RuAAC4), respectively.5 Particularly, the lysine-
containing 1,5-triazole bridged DE-ring mimic, which has the
same ring size as the natural DE-ring, showed an excellent
structural resemblance to the corresponding part of vancomycin,
as judged by superimposition of this DE-ring mimic and a
molecular model of the X-ray structure of vancomycin in complex
with BLys-D-Ala-D-AlaB. These results were very encouraging and
implied the possibility of obtaining attractive mimics of the
Scheme 1 Design and retrosynthesis of the 1,5-triazole-bridged vancomycin
CDE-ring mimic 2.
bicyclic vancomycin CDE-ring system, in which the two biaryl
ether bridges are replaced by 1,5-triazole moieties (Scheme 1).
The AB-ring of vancomycin 16 was omitted in the proposed
mimic 2 and therefore the heptapeptide backbone was shortened
to a hexapeptide by deletion of the C-terminal aromatic amino
acid residue. The P1 and P5 positions were substituted by a lysine-
derived e-azido amino acid and are to be linked via a triazole
bridge originating from the central bisalkyne-functionalized
hydroxyphenylglycine residue at position P3. To complete the
hexapeptide sequence, D-leucine and L-alanine were incorporated
at positions P2 and P4, respectively, instead of the corresponding
amino acid residues in vancomycin, thereby focusing on the
construction of the bicylic mimic, while the crucial N-Me-D-leucine
residue for binding the ligand was retained at position P6.
Retrosynthetic analysis of the bicyclic vancomycin CDE-ring
a Medicinal Chemistry & Chemical Biology, Utrecht Institute for Pharmaceutical
Sciences, Department of Pharmaceutical Sciences, Faculty of Science,
Utrecht University, Universiteitsweg 99, 3584 CG Utrecht, The Netherlands.
E-mail: R.M.J.Liskamp@uu.nl; Fax: +31 30 253 6655; Tel: +31 30 253 7396/7307
b Chemical Biology & Medicinal Chemistry, School of Chemistry, University of
Glasgow, Joseph Black Building, University Avenue, Glasgow, G12 8QQ, Scotland,
UK. E-mail: Robert.Liskamp@glasgow.ac.uk
mimic
2 showed that the synthesis involves three stages
(Scheme 1). The first stage is synthesis of the required amino acid
building blocks 4–7 for assembly of linear hexapeptide 3, which is
functionalized with the required alkyne and azide functionalities.
Central and most challenging was the synthesis of the bisalkyne
hydroxyphenylglycine residue 5, which is the bicyclic bridging
residue. The second stage encompasses assembly of hexapeptide 3
by sequential peptide coupling steps. Finally, in the decisive third
stage, RuAAC macrocyclization of the linear hexapeptide 3 should
† Electronic supplementary information (ESI) available: Details of the synthetic
procedures and compound analysis, ITC experiments, copies of the 1H, 13C and
2D HSQC NMR spectra, and copies of the analytical HPLC and MALDI-TOF MS
data. See DOI: 10.1039/c3cc40628h
c
4498 Chem. Commun., 2013, 49, 4498--4500
This journal is The Royal Society of Chemistry 2013

View Article Online
Communication
ChemComm
afford the desired 1,5-triazole-bridged bicyclic hexapeptide 2. At this achieved by treatment with TBAF in THF furnishing the desired linear
point it seems apposite to emphasize that we were aiming for a hexapeptide 14 in an excellent yield (91%).
robust and convenient synthesis, which would give rise to consider-
The decisive last step was bicyclization of 14 via RuAAC
able quantities of (intermediate) building blocks as well as provid- chemistry. Since we have recently shown that the 1,5-triazole
ing, in principle, access to a number of analogues to design a biased bridged monocyclic tripeptides as DE-ring mimics of vanco-
compound library. Herein, we describe the optimized synthesis of mycin are excellently accessible by an intramolecular RuAAC
mimic 2 including its structural characterization and first results of macrocyclization reaction,5,11 the same reaction protocol was
its biological activity that is, binding affinity and in vitro activity.
applied for the bicyclization of hexapeptide 14 with 15 mol%
The required dipeptide segments 4 and 6 for hexapeptide 3 [Cp*RuCl]4 at a substrate concentration of 5 mM at 50 1C. It was
were obtained from a BOP-mediated coupling of the appropri- found that conversion was not complete after 24 h. However, to our
ate azido amino acids H-Lys(N3)-NHMe and Fmoc-D-Lys(N3)- delight, upon increasing the catalyst concentration to 30 mol%,
OH, accessible by a diazotransfer reaction,7 with Boc-D-Leu-OH bicyclization proceeded smoothly to completion within 24 h.
and H-Ala-OtBu, respectively, in an overall yield of 82 and 80%,
respectively (see Scheme S1, ESI†).
Analysis of the reaction mixture using HPLC showed the
presence of one major component. This compound was isolated
1
The alkyne-functionalized hydroxyphenylglycine derivative 5 was by preparative HPLC in 40% yield and characterized by H and
the key intermediate in the bicyclization of the linear peptide. After 1H-13C HSQC NMR analysis followed by MALDI-TOF MS, and was
several preliminary synthetic studies, it was found that compound 8 identified as the desired bicyclic compound 15 (see ESI†). Most
(Scheme 1, inset) was an ideal building block for 5 which upon importantly, both unique C–H moieties of the 1,5-disubstituted-
incorporation led to hexapeptide 3. The synthesis of 8 (see Scheme 1H-1,2,3-triazoles could be assigned and identified in the H-13C
1
S1, ESI†) started with the iodination of commercially available D- HSQC spectrum: dH 7.89 ppm/dC 134.3 ppm and dH 8.04 ppm/dC
hydroxyphenylglycine leading to a bis-ortho-iodide derivative in 85% 135.4 ppm, which was in agreement with the reported 13C shifts of
yield,8 which after conversion into its acetate (80%)9 was subjected 1,2,3-triazoles.12 Additional proof for the presence of a (bi)cyclic
to a Pd-catalyzed Sonogashira cross-coupling reaction to afford TIPS- structure of mimic 15 was obtained by treatment of the product
protected bisalkyne-functionalized hydroxyphenylglycine derivative with TCEP, as shown previously by Suga et al.,13 which is capable of
8 without requiring protection of the carboxylic acid moiety.
reducing azides to the corresponding amines. Upon exposure of
Assembly of the hexapeptide bicyclization precursor was now the product of the RuAAC macrocyclization, no change in mass or
carried out (Scheme 2). Coupling of acid 8 with the azido lysine HPLC retention time was observed indicating the absence of azide
containing dipeptide amine obtained by deprotection of 4 led to a functionalities, whereas the starting material 14 rapidly underwent
diastereoisomeric mixture of tripeptide 9, which could not be sepa- reduction of the azides as was apparent from the formation of a
rated at this stage of the synthesis, and therefore the diastereoiso- product (MS) with an observed loss of two N2-molecules, having
meric mixture was used in the next synthesis steps. After Boc-group also a different HPLC-retention time. After removal of the Boc-
removal, and an EDCI/HOBt-coupling of dipeptide acid Fmoc-D- group, the desired bicyclic compound 2 was obtained and isolated
Lys(N3)-Ala-OH (10), the major diastereoisomer of pentapeptide 11 using preparative HPLC in an attractive 55% yield.
was isolated using column chromatography in an overall yield of
An energy minimized molecular model of the bicyclic
56%.10 Next, removal of the Fmoc-group using dimethylamine was structure14 was superimposed on the crystal structure of
accompanied by the desired aminolysis of the phenolic acetate. Then, vancomycin-related balhimycin antibiotic,15 complexed with Lys-
coupling of Boc-N-Me-D-Leu-OH afforded hexapeptide 12 in 74% yield, D-Ala-D-Ala. An RMSD of 1.415 Å over nine atoms was calculated
which was followed by methylation to give fully protected hexapeptide (see Scheme S2 (ESI†) for atom numbering), and visual inspection
13 in 94% yield. Finally, removal of both TIPS-functionalities was showed a good resemblance between the bicyclic peptide backbone
and the CDE-ring part of vancomycin (Fig. 1).
To evaluate the properties of bicyclic vancomycin mimic 2 as a
CDE-ring mimic, isothermal microcalorimetry (ITC)16 was carried
out to determine the binding affinity of compound 2 with
Scheme 2 Assembly of the linear hexapeptide 14 followed by RuAAC macrocyclization
to afford 15. (a) TFA/DCM, 1 h, quant.; (b) 8, BOP, DIPEA, DCM, 3 h, 84%; (c) TFA/DCM, 1 h,
quant.; (d) 10, EDCI, HOBt, DIPEA, DCM, 4 h, 56%; (e) HN(CH3)2/THF, 1 h; (f) Boc-N-Me-D-
Leu-OH, EDCI, HOBt, DIPEA, DCM, 3 h, 74% (over two steps); (g) MeI, K2CO3, acetone, 24 h,
Fig. 1 Superimposition of balhimycin (in red) with bicyclic mimic 2.
Chem. Commun., 2013, 49, 4498--4500 4499
94%; (h) TBAF/THF, MeOH, 3h, 91%; (i) [Cp*RuCl]4, THF/MeOH, 50 1C, 24 h, 40%.
c
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
Table 1 The binding affinity as measured using ITCa
Notes and references
Compound
Ligand
Ka (MÀ1
)
1 (a) J. Xie, J. G. Pierce, R. C. James, A. Okano and D. L. Boger, J. Am.
Chem. Soc., 2011, 133, 13946; (b) J. Xie, A. Okano, J. G. Pierce,
R. C. James, S. Stamm, C. M. Crane and D. L. Boger, J. Am. Chem.
Soc., 2012, 134, 1284.
2 H. T. ten Brink, D. T. S. Rijkers, J. Kemmink, H. W. Hilbers and
R. M. J. Liskamp, Org. Biomol. Chem., 2004, 2, 2658; H. T. ten Brink,
D. T. S. Rijkers and R. M. J. Liskamp, J. Org. Chem., 2006, 71, 1817.
3 C. W. Tornøe, C. Christensen and M. Meldal, J. Org. Chem., 2002,
67, 3057; V. V. Rostovtsev, L. G. Green, V. V. Fokin and
K. B. Sharpless, Angew. Chem., Int. Ed., 2002, 41, 2596; M. Meldal
and C. W. Tornøe, Chem. Rev., 2008, 108, 2952.
1
1
1
1
2
2
2
2
Ac-D-Ala-OH
Ac-D-Ala-D-Ala-OH
Ac-Lys(Ac)-D-Ala-D-Ala-OH
Ac-Lys(Ac)-D-Ala-D-Lac-OH
Ac-D-Ala-OH
Ac-D-Ala-D-Ala-OH
Ac-Lys(Ac)-D-Ala-D-Ala-OH
Ac-Lys(Ac)-D-Ala-D-Lac-OH
(3.14 Æ 0.14) Â 103
(1.16 Æ 0.85) Â 104
(3.47 Æ 0.12) Â 105
(2.19 Æ 0.12) Â 103
(3.33 Æ 0.12) Â 103
(2.15 Æ 0.24) Â 103
(2.46 Æ 0.29) Â 103
(3.14 Æ 0.38) Â 103
a
Measured in a Na-citrate/citric acid buffer (0.02 M, pH 5.1).1b
4 L. Zhang, X. G. Chen, P. Xue, H. H. Y. Sun, I. D. Williams,
K. B. Sharpless, V. V. Fokin and G. C. Jia, J. Am. Chem. Soc., 2005,
127, 15998; M. M. Majireck and S. M. Weinreb, J. Org. Chem., 2006,
71, 8680; L. K. Rasmussen, B. C. Boren and V. V. Fokin, Org. Lett.,
2007, 9, 5337; B. C. Boren, S. Narayan, L. K. Rasmussen, L. Zhang,
H. T. Zhao, Z. Y. Lin, G. C. Jia and V. V. Fokin, J. Am. Chem. Soc.,
2008, 130, 8923 (correction B. C. Boren, S. Narayan, L. K.
Rasmussen, L. Zhang, H. T. Zhao, Z. Y. Lin, G. C. Jia and V. V. Fokin,
J. Am. Chem. Soc., 2008, 130, 14900).
Ac-D-Ala-D-Ala, which is the interacting part of the natural ligand of
vancomycin. As is shown in Table 1, mimic 2 was capable of
binding Ac-D-Ala-D-Ala-OH or a better model of the ligand, that is
Ac-Lys(Ac)-D-Ala-D-Ala-OH, with a (only) 5- to 100-fold lower affinity,
respectively, than vancomycin. Remarkably, and in contrast to
vancomycin, our mimic bound the Ac-D-Ala-D-Lac ligand, which is
present in the cell-wall precursor of vancomycin resistant bacteria,
with an even slightly higher affinity. The lower affinity of mimic 2
for Ac-Lys(Ac)-D-Ala-D-Ala may imply that bicyclic peptide 2 is still
not sufficiently rigid, possibly indicating the importance of the
AB-ring part in vancomycin, which may lead to additional stabili-
zation of the cavity-like structure. Since the binding affinity of
vancomycin and that of mimic 2 for Ac-D-Ala-OH were identical, it
became apparent that both compounds acted as a receptor for
small carboxylic acids17 and the measured values are in agreement
with those reported in the literature.18
5 J. Zhang, J. Kemmink, D. T. S. Rijkers and R. M. J. Liskamp, Org.
Lett., 2011, 13, 3438.
6 D. H. Williams and B. Bardsley, Angew. Chem., Int. Ed., 1999, 38, 1172;
¨
K. C. Nicolaou, C. N. C. Boddy, S. Brase and N. Winssinger, Angew.
Chem., Int. Ed., 1999, 38, 2096; B. K. Hubbard and C. T. Walsh, Angew.
Chem., Int. Ed., 2003, 42, 730.
7 E. D. Goddard-Borger and R. V. Stick, Org. Lett., 2007, 9, 3797.
8 Y. Yamada, A. Akiba, S. Arima, C. Okada, K. Yoshida, F. Itou, T. Kai,
T. Satou, K. Takeda and Y. Harigaya, Chem. Pharm. Bull., 2005, 53, 1277.
9 M. E. Tran-Huu-Dau, R. Wartchow, E. Winterfeldt and Y. S. Wong,
Chem.–Eur. J., 2001, 7, 2349.
10 During the synthesis of 9, the phenylglycine residue had partially
racemized
resulting
in
a
diastereoisomeric
ratio
of
RRS(desired):ÀSRS(epimerized) = 5.4 : 1, according to analytical HPLC,
À
however, it was not possible to separate these diastereoisomers.
Complete racemization of the phenylglycine would have led to two
equal peaks. After synthesis of 11, it was possible to separate the two
diastereoisomers, and in all likelihood the larger peak in the
analytical HPLC (corresponding also to the most intense spot on
TLC) originated from the larger peak in the diastereomeric mixture
Be this as it may, the ultimate challenge is to find biologically
active mimics of vancomycin. Therefore, the antibacterial
activity of the bicyclic mimic 2 was assessed using an in vitro
assay against the vancomycin sensitive bacterium Staphylococcus
aureus (ATCC 49230). Unfortunately, and somewhat surprisingly,
our mimic 2 did not show inhibition of bacterial growth even at
of 9, and is now the desired diastereoisomer (RS)RRS.
À
´
11 A. R. Kelly, J. Wei, S. Kesavan, J.-C. Marie, N. Windmon, D. W. Young
and L. A. Marcaurelle, Org. Lett., 2009, 11, 2257; L. A. Marcaurelle,
E. Comer, S. Dandapani, J. R. Duvall, B. Gerard, S. Kesavan,
M. D. Lee IV, H. Liu, J. T. Lowe, J.-C. Marie, C. A. Mulrooney,
B. A. Pandya, A. Rowley, T. D. Ryba, B.-C. Suh, J. Wei, D. W. Young,
L. B. Akella, N. T. Ross, Y.-L. Zhang, D. M. Fass, S. A. Reis,
W.-N. Zhao, S. J. Haggarty, M. Palmer and M. A. Foley, J. Am. Chem.
Soc., 2010, 132, 16962.
high concentrations, up to 400 mg mLÀ1
.
In conclusion, we successfully synthesized a bicyclic 1,5-
triazole bridged vancomycin CDE-ring mimic. For the first time,
RuAAC macrocyclization was employed for the synthesis of such
a relatively complex bicyclic peptidomimetic. Thus RuAAC
macrocyclization can be employed for such a system with
excellent intramolecular selectivity. Although molecular model-
ling of the bicyclic mimic 2 showed good structural resemblance
to vancomycin, despite the good binding affinity with the D-Ala-D-
Ala peptide sequences, no antibacterial activity was observed.
However, the bicyclic peptidomimic synthesized has a unique
and interesting molecular structure and represents a new struc-
turally close mimic of the vancomycin CDE-ring. With the
established flexible and robust synthesis of the linear peptide
as well as the RuAAC macrocyclization strategy, synthesis of a
small library of vancomycin mimics will be possible in which
amino acid residues at P2 and P4 are varied to incorporate
interaction sites for selective substrate binding.18b In addition,
variation of these highly pre-organized bicyclic peptides may
lead to other ligands capable of binding promising targets.
J. Zhang acknowledges the China Scholarship Council (CSC) for
a scholarship to perform his PhD studies in the Netherlands. We
thank Hans W. Hilbers for performing the MALDI-TOF analyses.
12 M. Empting, O. Avrutina, R. Meusinger, S. Fabritz, M. Reinwart,
M. Biesalski, S. Voigt, G. Buntkowsky and H. Kolmar, Angew. Chem.,
Int. Ed., 2011, 50, 5207; X. Creary, A. Anderson, C. Brophy, F. Crowell
and Z. Funk, J. Org. Chem., 2012, 77, 8756.
13 Y. Sako, J. Morimoto, H. Murakami and H. Suga, J. Am. Chem. Soc.,
2008, 130, 7232.
14 Modelling of the structures was accomplished using the Yasara
Structure 10.5.2.1 software package (www.yasara.org). The models
were energy minimized using a simulated annealing protocol
employing the Amber99 forcefield.
15 The crystal structure of balhimycin (a vancomycin-related glyco-
peptide antibiotic) in complex with BLys-D-Ala-D-AlaB (Protein
Data bank accession code: 1GO6) was used, see: G. Lehmann,
´
´
G. Bunkoczi, L. Vertesy and G. M. Sheldrick, J. Mol. Biol., 2002,
318, 723.
16 D. McPhail and A. Cooper, J. Chem. Soc., Faraday Trans., 1997,
93, 2283.
17 ITC control experiments in which the affinity of Ac-L-Ala-OH and
Ac-L-Ala-L-Ala-OH towards vancomycin was determined indicated
that these ligands did not interact. An indication that the Ka values
of Table 1 represent specific binding interactions.
18 (a) M. Bois-Choussy, L. Neuville, R. Beugelmans and J. Zhu, J. Org.
Chem., 1996, 61, 9309; (b) C. J. Arnush and R. J. Pieters, Eur. J. Org.
Chem., 2003, 3131.
c
4500 Chem. Commun., 2013, 49, 4498--4500
This journal is The Royal Society of Chemistry 2013