Total synthesis of cyclotheonamide C using a tandem backbone-extension- coupling methodology

Article abstract of DOI:10.1002/anie.200802005

Mastering unusual amino acids: The first total synthesis of the potent protease inhibitor cyclotheonamide C (see structure representation) has been achieved. The key features of the synthesis are a three-component tandem procedure to create a masked α-keto-β-arginine within a peptide chain, and the control of the extended conjugation of a vinylogous α,β-dehydrotyrosine. (Chemical Equation Presented)

Full text of DOI:10.1002/anie.200802005

Communications
DOI: 10.1002/anie.200802005
Natural Products Synthesis
Total Synthesis of Cyclotheonamide C using a Tandem Backbone-
Extension–Coupling Methodology**
StØphane P. Roche, Sophie Faure, and David J. Aitken*
The cyclotheonamides (Ct) are a family of nine macrocyclic
pentapeptides isolated from the sponges Theonella swinhoei
and ircinia.[1] A common structural feature of these com-
pounds is a highly electrophilic a-keto-b-arginine (k-Arg)
moiety, which is responsible for their potency as inhibitors of
serine proteases of the trypsin family,[1,2,3b] including throm-
bin, for which IC50 values are in the 3–80 nm range.[1] The
challenge of the total synthesis of these com-
pounds has inspired several groups, leading to
successful preparations of the first two mem-
bers of the family, CtA and CtB.[3] In each of
these syntheses, the k-Arg moiety was prepared
using a multistep sequence as a discrete,
protected amino acid building block in the
first instance, then introduced conventionally
into the peptide chain assembly. The reactive
carbonyl function was masked as a protected
secondary alcohol, being unveiled by deprotec-
tion and oxidation in the last stages of the
synthesis. A more recent report on the total
synthesis of CtE2 and CtE3 constitutes an
approach for the first total synthesis of CtC,[1b] a unique
member of the Ct family in that it possesses a fully conjugated
vinylogous dehydrotyrosine (V-DTyr) structural feature,
which adds to the synthetic challenge of this compound.
Our synthetic plan is outlined in Figure 1. Macrocyclisa-
tion and deprotection constitute the final stages of an
approach which targets a key advanced linear pentapeptide.
Figure 1. Retrosynthetic analysis on cyclotheonamide C, identifying the key linear
pentapeptide advanced intermediate structure. PG=protecting group
alternative approach to the construction of the
k-Arg-containing fragment, since an a-Arg-
derived component was homologated to k-Arg
at the same time as the next amino acid residue was coupled.[4]
An inconvenience in this approach is arguably the premature
appearance of the free k-Arg carbonyl function. Nevertheless,
the attractive, tandem one-pot homologation–coupling
approach seemed to us an expedient strategy for the synthesis
of cyclotheonamides. We sought a procedure which could
combine the rapidity of construction with the provision of a
protecting feature, allowing creation of a masked k-Arg
within a peptide sequence. Herein, we describe such an
This intermediate features an a-silyloxyhomoarginine serving
as a masked k-Arg, which we envisaged creating using
Nemoto and co-workersꢀ masked acyl cyanide (MAC)
reaction.[5] This mild, three-component procedure combines
an aldehyde, an amine and a-(tert-butyldimethylsilyloxy)ma-
lononitrile 2 to afford an a-silyloxycarboxamide under mild
basic conditions. This reaction should allow the creation of an
a-hydroxy-b-amino carboxamide system,[6] the central feature
in our linear peptide target, from readily available a-amino-
acid-derived reactants (Scheme 1). Reputedly, the procedure
is not completely diastereoselective at the newly-formed
[*] Prof. Dr. D. J. Aitken
UniversitØ Paris-Sud 11, Laboratoire de Synth›se Organique &
MØthodologie
Institut de Chimie MolØculaire et des MatØriaux d’Orsay (CNRS
UMR 8182)
15 rue Georges Clemenceau, 91405 Orsay cedex (France)
Fax: (+33)1-6915-6278
E-mail: david.aitken@u-psud.fr
Dr. S. P. Roche, Dr. S. Faure
UniversitØ Blaise Pascal—Clermont-Ferrand 2
Laboratoire de Synth›se et Etudes de Syst›mes à IntØrÞt Biologique
(CNRS UMR 6504)
24 avenue des Landais, 63177 Aubi›re cedex (France)
[**] We thank B. LØgeret for mass spectral analyses and the Minist›re de
la Recherche for a PhD grant to S.P.R.
Scheme 1. The MAC reaction as a tandem backbone-extension–cou-
pling procedure for the creation of a masked a-keto-b-amino acid
within a peptide chain.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200802005.
6840
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6840 –6842

Angewandte
Chemie
stereogenic centre, but this is irrelevent for our purposes, as
this carbon centre is destined to become a ketone.[6] An
additional desirable feature of the MAC approach is that the
mild reaction conditions should allow use of an a-argininal
derived reactant with minimal risk of epimerisation.[7,8] The
pivotal part of our synthesis was therefore the one-pot
combination of a dipeptide amine for the southern part and
an argininal derivative for the northern part, to create the k-
Arg feature within a complex peptide structure.
Two aldehyde components were synthesised from a stock
of the readily available Weinreb amide 3 (Scheme 2).
Argininal derivative 4 was obtained by installation of the
Scheme 3. Preparation of the amine component 10. Reagents and
conditions: a) LiAlH4, THF, 08C, 1 h, 79%; b) Ph3P=CHCO2tBu,
activated MnO2, CH2Cl2, 408C, 36 h, 58%; c) Et2NH/CH3CN 1:2, 08C
to RT, 1 h, 89%.
need to isolate the intermediate aldehyde, which underwent
degradation upon any isolation attempts. Conversion of 9 into
the free amine 10 was achieved simply through treatment with
diethylamine.
With the aldehyde and amine components in hand, we
turned our attention to the key MAC reactions for peptide
construction (Scheme 4). Targeting tripeptide 11 in the first
combination, we examined various reaction conditions for the
three-component reaction between argininal derivative 4,
amine 10 and reagent 2. Satisfying results (53% yield) were
obtained for a chilled two-day reaction in the presence of 4-
pyrrolidinopyridine as base. N-Terminal deprotection of 11
and coupling with dipeptide 5 provided the key linear
pentapeptide 12. In a more spectacular fashion, tripeptide
aldehyde 6, amine 10 and reagent 2 were combined using
MAC reaction conditions to afford the linear pentapeptide 12
in a single reaction, with an acceptable 24% yield. This highly
convergent operation allowed the successful creation of the
masked k-Arg feature within a complex linear pentapeptide
and represents the most elaborate use of such methodology to
date.
Simultaneous deprotection of N-and C-termini of penta-
peptide 12 was achieved using trifluoroacetic acid, to furnish
the free pentapeptide for the macrocyclisation step. Activa-
tion of the fully conjugated “push-pull” V-DTyr C-terminus
proved a challenge, but after some experimentation, the
uronium coupling reagent TBTU, together with a stoichio-
metric amount of HOBt in a DMF/CH2Cl2 mixture, provided
the target macrocyclic peptide 13 in 69% yield. With the end
of our synthesis now in sight, it was time to liberate the k-Arg
moiety. Frustratingly, the action of TBAF on 13 effected
complete deprotection of both alcohol and phenol functions,
in further testimony to the “push-pull” effect of the extended
conjugated system. Subsequent oxidation attempts on sam-
ples with an unprotected phenol led to extensive degradation
of material. To our gratification, selective deprotection of the
secondary alcohol TBS ether of 13 in the presence of the
phenyl TIPS ether was achieved by the action of 1% aqueous
HCl in ethanol at 308C.[11] Dess–Martin periodinane oxida-
Scheme 2. Preparation of the aldehyde components 4 and 6. Reagents
and conditions: a) TFA/CH2Cl2 1:1, 08C, 1 h; b) FmocCl, Na2CO3,
H2O/dioxane, 08C to RT, 12 h, 97% (2 steps); c) LiAlH4, THF, À308C,
1.5 h, 59%; d) EDCI, DMAP, 5, CH2Cl2, 08C to RT, 18 h; e) LiAlH4,
THF, À108C, 1 h, 48% (2 steps). Boc=tert-butyloxycarbonyl, TFA=tri-
fluoroacetic acid, Fmoc=9-fluorenylmethyloxycarbonyl, EDCI=1-(3-
dimethylaminopropyl)-3-ethyl carbodiimide hydrochloride, DMAP=4-
dimethylaminopyridine.
base-labile Fmoc protecting group on Na followed by chemo-
selective reduction of the amide using LiAlH4 at low temper-
ature. The tripeptide aldehyde 6 was obtained by deprotec-
tion of 3 followed by EDCI/HOBt coupling with dipeptide 5,
then selective reduction using LiAlH4. Aldehydes 4 and 6
prepared in this manner were enantiomerically pure and
required minimal workup before being engaged in subse-
quent reactions.
The requisite amine component, dipeptide 10, was
constructed stereoselectively as follows (Scheme 3). Z-Dehy-
drotyrosine dipeptide 7, obtained by adaptation of an
established procedure,[9] was reduced to give alcohol 8 in
79% yield, then Taylor and co-workersꢀ one pot oxidation–
vinylogation procedure[10] using activated manganese(IV)
oxide and tert-butyloxycarbonylmethylene triphenylphos-
phorane gave 9, which bears the vinylogous dehydrotyrosine
feature (V-DTyr) with the E configuration exclusively for the
new double bond. This procedure conveniently avoids the
Angew. Chem. Int. Ed. 2008, 47, 6840 –6842
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6841

Communications
block, and thus shortens the overall sequence compared to
previous syntheses of CtA and CtB.
Received: April 29, 2008
Published online: July 25, 2008
Keywords: macrocycles · multicomponent reactions ·
.
natural products · peptides · total synthesis
[1] a) N. Fusetani, S. Matsunaga, J. Am. Chem. Soc. 1990, 112, 7053 –
7054; b) Y. Nakao, S. Matsunaga, N. Fusetani, Bioorg. Med.
Chem. 1995, 3, 1115 – 1122; c) Y. Nakao, N. Oku, S. Matsunaga,
N. Fusetani, J. Nat. Prod. 1998, 61, 667 – 670; d) Y. Murakami, M.
Takei, K. Shindo, C. Kitazume, J. Tanaka, T. Higa, H. Fukama-
chi, J. Nat. Prod. 2002, 65, 259 – 261.
[2] V. Ganesh, A. Y. Lee, J. Clardy, A. Tulinsky, Protein Sci. 1996, 5,
825 – 835.
[3] a) M. Hagihara, S. L. Schreiber, J. Am. Chem. Soc. 1992, 114,
6570 – 6571; b) B. E. Maryanoff, X. Qiu, K. P. Padmanabhan, A.
Tulinsky, H. R. Almond, P. Andrade-Gordon, M. N. Greco, J. A.
Kauffman, K. C. Nicolaou, A. Liu, P. H. Brungs, N. Fusetani,
Proc. Natl. Acad. Sci. USA 1993, 90, 8048 – 8052; c) P. Wipf, H.
Kim, J. Org. Chem. 1993, 58, 5592 – 5594; d) J. Deng, Y. Hamada,
T. Shioiri, S. Matsunaga, N. Fusetani, Angew. Chem. 1994, 106,
1811 – 1813; Angew. Chem. Int. Ed. Engl. 1994, 33, 1729 – 1731;
e) B. E. Maryanoff, M. N. Greco, H.-C. Zhang, P. Andrade-
Gordon, J. A. Kauffman, K. C. Nicolaou, A. Liu, P. H. Brungs, J.
Am. Chem. Soc. 1995, 117, 1225 – 1239; f) H. M. Bastiaans, J. L.
van der Baan, C. J. Ottenheijm, Tetrahedron Lett. 1995, 36,
5963 – 5966; g) J. Deng, Y. Hamada, T. Shioiri, Tetrahedron
Lett. 1996, 37, 2261 – 2264; h) H. M. Bastiaans, J. L. van der
Baan, C. J. Ottenheijm, J. Org. Chem. 1997, 62, 3880 – 3889.
[4] a) H. H. Wasserman, R. Zhang, Tetrahedron Lett. 2002, 43,
3743 – 3746; b) H. H. Wasserman, R. Zhang, Tetrahedron 2002,
58, 6277 – 6283.
[5] H. Nemoto, Y. Kubota, Y. Yamamoto, J. Org. Chem. 1990, 55,
4515 – 4516.
[6] H. Nemoto, R. Ma, I. Suzuki, M. Shibuya, Org. Lett. 2000, 2,
4245 – 4247.
[7] In the syntheses of CtA and CtB (ref. [3]), the masked k-Arg
building blocks are usually elaborated with argininal derivatives
which do not always benefit from mild reaction conditions and
are therefore vulnerable towards epimerisation.
Scheme 4. Multicomponent assembly of advanced peptides leading to
CtC. Reagents and conditions: a) 2, 4-PP, THF, À258C (24 h) then 08C
(24 h), 53%; b) 2, 4-PP, CH3CN, 08C, 15 h, 24%; c) Et2NH/CH3CN
1:2, 08C to RT, 1 h; d) 5, EDCI, HOBt, CH2Cl2, 08C to RT, 18 h, 79%
(2 steps); e) TFA/CH2Cl2 1:1, 08C, 1.5 h; f) TBTU, HOBt, CH2Cl2/DMF
2:1, 0.005m, 08C to RT, 40 h, 69% (2 steps); g) HCl/EtOH, 308C, 2 h,
35%; h) DMP, CH3CN, 608C, 1 h, 71%; i) HF·pyridine, anisole, RT,
12 h, 51%. TBTU=O-(benzotriazol-1-yl)-N,N,N’,N’-tetramethyluro-
nium tetrafluoroborate, DMP=Dess–Martin periodinane, 4-PP=4-pyr-
rolidinopyridine, HOBt=1-hydroxybenzotriazole.
tion was performed smoothly to provide a derivative of CtC
which was protected only on the dehydrotyrosine and
arginine side chains. Final deprotection of these moieties
was achieved in 51% yield using pyridine hydrofluoride in the
presence of anisole to furnish CtC.[12]
[8] This expectation was confirmed experimentally.
A model
reaction combining argininal derivative 3, reagent 2, and d-
Phe-OMe was shown to proceed without epimerization. See
Supporting Information for details.
[9] D. J. Aitken, S. Faure, S. Roche, Tetrahedron Lett. 2003, 44,
8827 – 8830.
In conclusion, we have successfully achieved the first total
synthesis of CtC, using an unconventional approach for the
assembly of the advanced linear pentapeptide. This multi-
component construction is epimerisation-free and obviates
the need for the upstream elaboration of a k-Arg building
[10] X. Wei, R. J. K. Taylor, J. Org. Chem. 2000, 65, 616 – 620.
[11] R. F. Cunico, L. Bedell, J. Org. Chem. 1980, 45, 4798 – 4801.
[12] NMR and MS data were consistent with those obtained for the
natural product. Copies of the spectra were kindly made
available to us by Prof. N. Fusetani.
6842 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6840 –6842