A macrocyclic coumarin-containing tripeptide via CuAAC chemistry

Article abstract of DOI:10.1039/b906762k

A Cu-catalysed macrocyclisation was performed to obtain a macrocyclic coumarin-containing tripeptide for use in thrombin activity measurements.

Full text of DOI:10.1039/b906762k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A macrocyclic coumarin-containing tripeptide via CuAAC chemistryw
Sander S. van Berkel, Bas van der Lee, Floris L. van Delft and Floris P. J. T. Rutjes*
Received (in Cambridge, UK) 3rd April 2009, Accepted 12th May 2009
First published as an Advance Article on the web 9th June 2009
DOI: 10.1039/b906762k
A Cu-catalysed macrocyclisation was performed to obtain a
macrocyclic coumarin-containing tripeptide for use in thrombin
activity measurements.
cyclisation using auxiliaries8 have been reported. With the
introduction of the copper-catalysed azide–alkyne cyclo-
addition (CuAAC) reaction, a versatile and mild coupling
method was established which rapidly found application
in various fields of chemistry.9 In addition, the CuAAC
reaction had a significant impact on the synthesis of new
peptidomimetics.10 Studies have shown that a 1,2,3-triazole
can be a suitable mimic for an amide bond in peptides,11
moreover, the obtained 1,4-connected 1,2,3-triazole ring can
serve as a b-turn mimic.12 Recently, the CuAAC has also been
applied as a tool for macrocyclisation both in solution13 and
on solid phase.14 We therefore envisioned that the CuAAC
could serve as an efficient cyclisation method for the synthesis
of macrocycles 2a/b.
Thrombosis is the basis of diseases such as cerebral and
coronary infarction and pulmonary embolism and thus can
be considered the major cause of death in the western world.
A method for measuring the effect of newly developed
thrombin inhibitors on the haemostatic system is the so-called
thrombin generation test (TGT) devised by Hemker et al.1
Upon selective hydrolysis of
developing thrombin, a fluorophore is released thereby
providing direct measure for the thrombin activity.
a fluorogenic peptide by
a
Substrates used in the thrombin generation test are generally
linear and flexible tripeptides bearing a C-terminal fluoro-
phore (e.g. 7-amino-4-methylcoumarin, AMC).2 An example
of such a peptide is H-Gly-Pro-Arg-AMC (1, Fig. 1). This
peptide is, however, hydrolysed extremely rapidly by thrombin
(FIIa),2 while a slow-reacting substrate is required for accurate
and prolonged measurements. Moreover, the released AMC
has a moderate solubility in aqueous solutions and thus
interferes with the accurate assessment of enzyme activities
as a result of fluorescent quenching.3 In addition, AMC is
able to act as a competitive inhibitor of thrombin at high
concentrations.4
Retrosynthetic analysis of cyclo-[-Gly-Pro-Arg-AMC-
[triazole]-spacer-] analogues 2 leads to the linear peptides 3
(Scheme 1), containing the requisite alkyne and azide
functionalities for the proposed CuAAC macrocyclisation.
Optimal convergence of the route is obtained by splitting the
linear peptide into an azidocoumarin-functionalised arginine
residue 4 and an alkyne-containing Gly-Pro peptide 5. The
alkyne tail can be coupled to the Gly-Pro peptide unit either
via an amide or carbamate type linkage (Scheme 2).
The arginine residue is decorated with 7-amino-4-azidomethyl-
coumarin 13, which in turn was thought to be derived from
N-acetyl-3-aminophenol.
We envisaged a fluorophore-containing cyclic structure
(i.e. 2, Fig. 1) as a compelling strategy to (a) overcome the
poor solubility of AMC in physiological media and (b) give
rise to improved kinetic parameters. Constraining the
linear peptide H-Gly-Pro-Arg-AMC in a bulky macrocyclic
conformation is expected to hamper the entry into the
thrombin active site, thus decreasing the affinity of the
substrate for the binding pocket of thrombin resulting in
Synthesis of synthon 5a commenced with coupling of
3-butyn-1-yl chloroformate to unprotected glycine (Scheme 2),
which after acid–base extraction was obtained in excellent
yield (95%). Next, coupling of L-proline methyl ester to 6a was
performed under standard peptide coupling conditions, to
afford compound 8 in a reasonable yield of 64%. Finally,
ester hydrolysis gave the desired synthon 5a in good yield
(70%). For the synthesis of synthon 5b (Scheme 2), peptide
coupling of glycine methyl ester with hexynoic acid was
followed by hydrolysis of methyl ester 7, resulting in
compound 6b in good yield. A second peptide coupling was
performed to link proline tert-butyl ester to fragment 6b,
affording smooth formation of dipeptide 9 (97% yield). The
favourable kinetic parameters (i.e. high KM and low kcat
)
suitable for the TGT. Moreover, precipitation and undesired
competitive inhibition of AMC upon thrombin hydrolysis will
not occur as a result of the triazole linkage of AMC to the
tripeptide backbone.
The main difficulty in the synthesis of cyclic peptides is the
macrocyclisation.5 For penta- and especially tetrapeptides the
cyclisation of all L-configured linear precursors is notoriously
difficult. Various methods for macrocyclisation, such as
lactonisation or lactamisation,6 ring-closing metathesis7 or
Institute for Molecules and Materials, Radboud University Nijmegen,
Heyendaalseweg 135, NL-6525 AJ Nijmegen, The Netherlands.
E-mail: F.Rutjes@science.ru.nl; Fax: (+31) 24 365 3393;
Tel: (+31) 24 365 3202
w Electronic supplementary information (ESI) available: Synthetic
procedure for compounds 1–16, click conditions, biological evalua-
tion. See DOI: 10.1039/b906762k
Fig. 1 Thrombin (FIIa) hydrolysis of linear (1) and cyclic (2) peptide.
ꢀc
This journal is The Royal Society of Chemistry 2009
4272 | Chem. Commun., 2009, 4272–4274

View Article Online
Scheme 3 Reagents and conditions: (a) Et2O, Et3N, r.t., 18 h, for 10a:
Cbz–Cl (51%), for 10b: Ac2O (99%) and for 10c: EtO2C–Cl (54%);
(b) typically: 60–70% H2SO4, r.t. to 45 1C, 2–5 d (starting form 10a:
0%, for 11b: 20% and for 11c: 70%); (c) various conditions (0%);
(d) conc. HCl, 2-propanol, reflux, 16 h (72%); (e) NaN3 (5 equiv.),
MeCN–acetone, r.t., 48 h (95%); (f) Boc-Arg-OH, POCl3, pyridine,
À15 1C to r.t., 1 h (41%); (g) TFA, CH2Cl2, r.t. 16 h (86%).
solution of H2SO4. Cbz-protected aminophenol (10a) did not
give the desired product under these conditions.
Scheme 1 Retrosynthesis of macrocycles 2.
Conversely, N-ethoxycarbonyl-protected aminophenol 10b
produced coumarin 11b in a satisfactory 70% yield, while
acetyl-protected aminophenol 10c gave only 20% of coumarin
11c. Deprotection of 11b applying various reaction conditions
afforded mostly mixtures of unidentifiable compounds.
Formation of ACMC 12 was then pursued by deprotection
of coumarin 11c with HCl in 2-propanol to produce 12 in good
yield (72%). Finally, nucleophilic displacement of the chloride
by an azide using NaN3 in acetone–acetonitrile produced key
intermediate 13 in excellent yield (95%).
Next, the fluorescent probe AAMC (13) was coupled to
Boc-protected arginine using phosphoryl chloride in pyridine
at À15 1C leading to Boc-Arg-AAMC (14). Purification of the
product by counter-current chromatography provided
Boc-Arg-AAMC (14) in a reasonable yield of 41%.
Subsequent Boc-removal with TFA afforded synthon 4.
Scheme 2 Reagents and conditions: (a) NaOH (aq) pH 9.5–10.5, r.t.,
72 h (95%); (b) hexynoic acid, EDC, DMAP, DMF, 0 1C to r.t., 18 h
(97%); (c) NaOH (2M), dioxane, r.t., 16 h (70%); (d) H-Pro-OMe,
DCC, NMM, EtOAc, 0 1C to r.t., 16 h (64%); (e) H-Pro-OtBu, EDC,
HOBt, CH2Cl2, 0 1C to r.t., 18 h (97%); (f) NaOH (2M), dioxane, r.t.
18 h (77%); (g) TFA, CH2Cl2, r.t., 18 h (99%).
last step involved acid-catalysed hydrolysis of the tert-butyl
ester to afford target compound 5b in excellent yield (99%).
The fluorophore 7-amino-4-azidomethylcoumarin (AAMC,
13), required for the synthesis of H-Arg-AAMC (4), is not
commercially available so that a synthetic route was devised.
Starting from 3-aminophenol both Lewis acid-catalysed and
microwave-assisted Pechmann condensations were explored in
order to obtain aminochloromethylcoumarin 12 (ACMC).
A protective group was required since direct Pechmann
condensation of 3-aminophenol under a variety of conditions
turned out to be fruitless. Therefore, the amine functionality of
3-aminophenol was protected with different protecting groups
in reasonable to good yields (Scheme 3). The N-protected
aminophenols 10a–c were subsequently subjected to the
Scheme 4 Reagents and conditions: (a) EDC, HOBt, DMAP, DMF,
0 1C, r.t., 18 h (3a (24%), 3b (54%)); (b) Boc2O (4 eq), THF, Et3N, r.t.,
16 h (50%); (c) CuBr/TMEDA, THF, 40 1C, 7 h (23%); (d) HCl in
EtOAc, r.t., 16 h (99%).
Pechmann condensation, typically applying
a 60–70%
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 4272–4274 | 4273

View Article Online
Having key intermediates 4 and 5a/b in hand, coupling of
these fragments was required to obtain linear peptides 3a and
3b (Scheme 4). To this end, amide formation was performed
utilizing EDC, HOBt and DMAP. Coupling of 5a to 4 gave
the linear carbamate 3a in moderate yield (24%). Fortunately,
coupling of 5b with 4 proceeded significantly better so
that linear amide 3b was produced in a yield of 54% after
purification by counter-current chromatography.
improvement of the kinetic parameters may be achieved by
changing the Pro-Gly sequence and/or the alkyl spacer-length
of the cyclic peptide. Moreover, other challenging strategies to
construct the cyclic peptide involve coupling of the azide and
alkyne fragments via CuAAC, and/or by invoking strain-
promoted (Cu-free) 1,3-dipolar cycloaddition approaches,
followed by an intramolecular lactamization reaction.
These are all strategies currently under investigation in our
laboratories.
Copper-catalysed macrocyclisation of 3a or 3b was pursued
applying modified literature procedures. Since we anticipated
that elevated temperatures could well result in decomposition
of compounds 3a and 3b, our attention was focused on
intramolecular azide–alkyne cycloaddition at relatively low
temperatures15 and under high dilution (o0.001 M).
Unfortunately, the different conditions applied on either 3a
or 3b (see ESIw) failed to produce sufficient quantities of
the desired macrocycle for adequate purification and identifi-
cation. Since Cu strongly coordinates to the basic guanidine
moiety, and thus hampers purification and possibly also
cyclisation, compound 3b was Boc-protected at the guanidine
prior to macrocyclisation (Scheme 4). Addition of four
equivalents of di-tert-butyl dicarbonate resulted in the forma-
tion of three products (i.e. mono-, di- and trisubstituted
guanidine), of which the bis-Boc-protected compound 15
appeared to be the major product (50%).
The authors thank Prof. H. C. Hemker and Dr R. Wagenvoord
(Synapse BV, Maastricht, The Netherlands) for their contri-
bution and useful discussions. These investigations were
supported with financial aid from the Netherlands Technology
Foundation (STW).
Notes and references
1 H. C. Hemker, P. Giesen, R. Al Dieri, V. Regnault, E. de Smedt,
R. Wagenvoord, T. Lecompte and S. Beguin, Pathophysiol.
Haemost. Thromb., 2003, 33, 4.
2 M. K. Ramjee, Anal. Biochem., 2000, 277, 11.
3 H. D. Shine, L. Hertz, J. de Vellis and B. Haber, Neurochem. Res.,
1981, 6, 453.
4 S. S. van Berkel, Thrombin Generation: Molecules and Tools,
PhD-Thesis, Radboud University Nijmegen, 2008.
5 J. S. Davies, Amino Acids, Pept. Proteins, 2003, 34, 149.
6 (a) Y. Shao, W. Lu and S. B. H. Kent, Tetrahedron Lett., 1998, 39,
3911; (b) L. Zhang and J. P. Tam, J. Am. Chem. Soc., 1999, 121,
3311; (c) J. T. Lundquist IV and J. C. Pelletier, Org. Lett., 2002, 4,
3219.
7 (a) S. J. Miller, H. E. Blackwell and R. H. Grubbs, J. Am. Chem.
Soc., 1996, 118, 9606; (b) U. Kazmaier, C. Hebach, A. Watzke,
S. Maier, H. Mues and V. Huch, Org. Biomol. Chem., 2005, 3, 136;
(c) N. Ghalit, J. Kemmink, H. W. Hilbers, C. Versluis, D. T. S.
Rijkers and R. M. J. Liskamp, Org. Biomol. Chem., 2007, 5, 924;
(d) J. Illsinghe, C. Xing Guo, R. Garland, A. Ahmed, B. van
Lierop, J. Elaridi, W. R. Jackson and A. J. Robinson, Chem.
Commun., 2009, 295.
8 (a) W. D. F. Meutermans, S. W. Golding, G. T. Bourne,
L. P. Miranda, M. J. Dooley, P. F. Alewood and M. L. Smythe,
J. Am. Chem. Soc., 1999, 121, 9790; (b) V. M. F. Cardona,
O. Hartley and P. Botti, J. Pept. Res., 2003, 61, 152;
(c) H. Bieraugel, H. E. Schoemaker, H. Hiemstra and J. H. van
Maarseveen, Org. Biomol. Chem., 2003, 1, 1830.
9 For reviews, see: (a) J. E. Moses and A. D. Moorhouse, Chem. Soc.
Rev., 2007, 36, 1249; (b) A. J. Dirks, J. J. L. M. Cornelissen,
F. L. van Delft, J. C. M. van Hest, R. J. M. Nolte, A. E. Rowan
and F. P. J. T. Rutjes, QSAR Comb. Sci., 2007, 26, 1200;
(c) M. Meldal and C. W. Tornøe, Chem. Rev., 2008, 108, 2952.
10 Y. L. Angell and K. Burgess, Chem. Soc. Rev., 2007, 36, 1674.
11 (a) H. C. Kolb and K. B. Sharpless, Drug Discovery Today, 2003, 8,
1128; (b) W. S. Horne, C. D. Stout and M. R. Ghadiri, J. Am.
Chem. Soc., 2003, 125, 9372; (c) W. S. Horne, M. K. Yadav,
C. D. Stout and M. R. Ghadiri, J. Am. Chem. Soc., 2004, 126,
15366.
12 (a) K. Oh and Z. Guan, Chem. Commun., 2006, 3069;
(b) Y. L. Angell and K. Burgess, J. Org. Chem., 2005, 70, 9595.
13 (a) J. H. van Maarseveen, W. S. Horne and M. R. Ghadiri, Org.
Lett., 2005, 7, 4503; (b) V. D. Bock, R. Perciaccante, T. P. Jansen,
H. Hiemstra and J. H. van Maarseveen, Org. Lett., 2006, 8, 919;
(c) V. D. Bock, D. Speijer, H. Hiemstra and J. H. van Maarseveen,
Org. Biomol. Chem., 2007, 5, 971.
14 (a) S. Punna, J. Kuzelka, Q. Wang and M. G. Finn, Angew. Chem.,
Int. Ed., 2005, 44, 2215; (b) R. A. Turner, A. G. Oliver and
R. S. Lokey, Org. Lett., 2007, 9, 5011; (c) V. Goncalves,
B. Gautier, A. Regazzetti, P. Coric. S. Bouaziz, C. Garbay,
M. Vidal and N. Inguimbert, Bioorg. Med. Chem. Lett., 2007,
17, 5590.
Gratifyingly, subjecting 15 to Cu-catalysed ring-closing
conditions (i.e. CuBr/TMEDA, THF, 40 1C, 7 h) resulted in
the formation of the macrocyclic fluorophore-containing
tripeptide 16 in an encouraging 23% yield.
Next, the kinetic parameters of both the linear and cyclic
peptide were determined by constructing Michaelis–Menten
and Lineweaver–Burk plots from which the binding constant
(KM) and the hydrolysis rate (kcat) were determined (see ESIw).
Comparing the binding affinity of the linear (1) and cyclic
peptide (2b) showed a significant reduction (KM = 746.9 mM
and 3693.9 mM, respectively). Calculations of kcat showed only
a slight decrease for the cyclic peptide 2b as compared to the
linear peptide 1 (kcat = 23.3 sÀ1 and 29.3 sÀ1, respectively), but
nevertheless remained high. Visual analysis of the turbidity of
the samples after hydrolysis by thrombin, as a measure for the
sedimentation of AMC, showed a clear difference between the
cyclic and linear peptide. While the samples containing high
concentrations of linear peptide 1 (4800 mM) clearly became
turbid, no turbidity at these concentrations was observed for
cyclic peptide 2b, thus indicating an improved solubility of
AMC as a result of its linkage to the hydrophilic peptide
backbone.
In conclusion, we have successfully demonstrated that
Cu-catalysed macrocyclisation of an azide- and alkyne-
containing linear peptide resulted in a cyclic fluorogenic
tripeptide. Biological evaluation of both the linear and cyclic
peptide demonstrated that the latter had improved kinetic
properties. The macrocyclic (bulky) shape of the cyclic peptide
indeed resulted in a five-fold increase of the binding constant
(KM) as compared to the linear peptide, but the hydrolysis rate
(kcat) remained high. It clearly shows, however, the feasibility
of this unprecedented cyclisation strategy to improve the
kinetic properties of the fluorescent substrate. Further
15 T.-S. Hu, R. Tannert, H.-D. Arndt and H. Waldmann, Chem.
Commun., 2007, 3942.
ꢀc
This journal is The Royal Society of Chemistry 2009
4274 | Chem. Commun., 2009, 4272–4274