A new strategy for solid-phase depsipeptide synthesis using recoverable building blocks

Article abstract of DOI:10.1021/ol047653v

(Chemical Equation Presented) The technique of choice for synthesis of small-scale depsipeptides is on a solid support. However, if expensive monomers have to be incorporated, solid-phase synthesis can quickly turn out to be unattractive because of its lo

Full text of DOI:10.1021/ol047653v

ORGANIC
LETTERS
2005
Vol. 7, No. 4
597-600
A New Strategy for Solid-Phase
Depsipeptide Synthesis Using
Recoverable Building Blocks
Fernando Albericio,*,† Klaus Burger,‡ Javier Ru´ız-Rodr´ıguez,† and Jan Spengler†,‡
Barcelona Biomedical Research Institute, Barcelona Science Park,
UniVersity of Barcelona, 08028 Barcelona, Spain, and Department of
Organic Chemistry, UniVersity Leipzig, 04103 Leipzig, Germany
albericio@pcb.ub.es
Received November 14, 2004
ABSTRACT
The technique of choice for synthesis of small-scale depsipeptides is on a solid support. However, if expensive monomers have to be incorporated,
solid-phase synthesis can quickly turn out to be unattractive because of its low atom economy. Herein, we describe a new type of recoverable
and reuseable
depsipeptides.
r-hydroxy acid building block for solid-phase synthesis and its application in the synthesis of a number of small cyclic
Depsipeptides are oligomers consisting of hydroxy and amino
acids linked together by ester and amide bonds. Naturally
occurring depsipeptides exhibit interesting biological activi-
ties, including antibacterial, antiviral, antifungal, antiinflam-
matory, and other properties. However, their greatest future
therapeutic potential will most probably be cancer treatment.1
Furthermore, depsipeptides can be branching units for
building dendrimers2 and possess the capacity to be nonviral
vectors for drug delivery.3 Finally, incorporation of hydroxy
acids into strategically positions on peptidic backbones is a
useful tool for evaluation of the role of individual hydrogen
bonds for secondary structure stability.4
every peptide sequence can be built with standard reaction
procedures. Every synthetic cycle is performed in a very short
time, compared to conventional solution synthesis. Since the
evaluation of physiological properties of synthetic depsipep-
tides is a newly emerging and rapidly growing research area,
solid-phase protocols for depsipeptide synthesis would be
highly attractive for creation of libraries in future. A literature
search for depsipeptide synthesis revealed that a broad range
of protocols are available, such as synthesis on solid phase,5
in solution,6 and combinations7 of these strategies. Ether,
ester, carbonate and silica protection of the hydroxylic group
(4) (a) Seebach, D.; Mahajan, Y. R.; Senthilkumar, R.; Rueping, M.;
Jaun, B. Chem. Commun. 2002, 1598-1599. (b) Aravinda, S.; Shamala,
N.; Das, C.; Balaram, P. Biopolymers 2002, 64, 255-267. (c) Haque, T.
S.; Little, J. C.; Gellman, S. H. J. Am. Chem. Soc. 1996, 118, 6975-6985.
(d) Shin, I.; Ting, A. Y.; Schultz, P. G. J. Am. Chem. Soc. 1997, 119,
12667-12668.
Solid-phase synthesis (SPS) is the strategy of choice for
small- and medium-scale peptide synthesis because almost
† University of Barcelona.
‡ University Leipzig.
(1) (a) Sarabia, F.; Chammaa, S.; Sa´nchez Ruiz, A.; Mart´ın Ortiz, L.;
Lo´pez Herrera, F. J. Curr. Med. Chem. 2004, 11, 1309-1332. (b) Ballard,
C. E.; Yu, H.; Wang, B. Curr. Med. Chem. 2002, 9, 471-498.
(2) Kress, J.; Rosner, A.; Hirsch, A. Chem. Eur. J. 2000, 6, 247-257.
(3) Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff, K. M.
Chem. ReV. 1999, 99, 3181-3198.
(5) (a) Kuisle, O.; Quin˜oa´, E.; Riguera, R. J. Org. Chem. 1999, 64, 8063-
8075. (b) Hermann, C.; Giammasi, C.; Geyer, A.; Maier, M. E. Tetrahedron
2001, 57, 8999-9010. (c) Boger, D. L.; Keim, H.; Oberhauser, B.; Schreiner,
E. P.; Foster, C. A. J. Am. Chem. Soc. 1999, 121, 6197-6205. (d) Lo´pez-
Macia`, A.; Jime´nez, J. C.; Royo, M.; Giralt, E.; Albericio, F. J. Am. Chem.
Soc. 2001, 123, 11398-11401.
10.1021/ol047653v CCC: $30.25
© 2005 American Chemical Society
Published on Web 01/20/2005

are employed.8 However, commercial sources of hydroxy
acids are limited and O-terminal protected hydroxy acids are
not available. Hydroxy acids as constituents of naturally
occurring depsipeptides are characterized by high structural
diversity. Apart from a few hydroxy analogues of proteino-
genic amino acids (lactic acid, hydroxyisovaleric acid, leucic
acid, isoleucic acid and phenyllactic acid), complex and
synthetically challenging architectures are common. Thus,
hydroxy acids are, in comparison to amino acids, rather
expensive monomers, often difficult to obtain. In this context,
the main drawback of solid-phase protocols is undoubtedly
the low atom economy compared to that of solution-phase
synthesis.9 To drive each reaction step to completion,
normally a 4- up to 10-fold excess of building blocks,
activation reagents and additives are added to the resin.
Normally, the added excess is neither recovered nor reused
in laboratory scale runs. Therefore, when highly expensive
hydroxy acid monomers are involved, solid-phase synthesis
can become unattractive because high costs overrule all other
advantages. In the following we report on a new solid-phase
protocol where the excess of the activated hydroxy acid
monomer can be easily recovered and reused.
Application of HFA-lactones in solid-phase synthesis would
overcome this disadvantage and connect synergistically two
straightforward techniques.
Scheme 1. Hexafluoroacetone as Bidentate Protecting/
Activating Reagent
In a first series of experiments we studied whether HFA-
hydroxy acids undergo nucleophilic ring opening with solid-
phase-bound N-terminal deprotected amino acids. Key
criteria for the value of the new approach are chiral integrity
of the products, yield, and reaction time. As model com-
pounds we choose HFA-D-mandelic acid 1a, because of its
high sensitivity to racemization, analogous to phenylglycine.
Four equivalents of 1a were reacted in different solvents with
H-Tyr(OtBu)-Rink-MBHA-resin. The progress of the reaction
was monitored with Kaiser’s ninhydrin test (check for free
NH2). Once a negative test is obtained, the resin was washed
with the corresponding solvent in order to remove all
hexafluoroacetone hydrate and the excess of starting material.
The product was cleaved from the resin with 95% TFA and
Hexafluoroacetone is a bidentate protecting/activating
reagent for R-functionalized carboxylic acids such as amino,
hydroxy and mercapto acids. In one reaction step, the
R-functionality and the adjacent carboxylic group undergo
a heterocyclization process yielding a lactone, in which the
carboxylic group is activated and the R-functionality is
protected. These five-membered heterocycles are stable
compounds and they can be prepared in multigram-scale.
The lactones represent activated esteres, which yield on
nucleophilic attack carboxylic acid derivatives already at
room temperature. Concomitantly, the R-functionality is
deblocked. This derivatization/deprotection proceeds with a
broad range of nucleophiles. Until recently, this strategy was
applied for solution synthesis of low-molecular-weight
derivatives such as esters, amides, peptides, and hydroxamic
acids. With respect to conventional protocols, in which
protection, activation, coupling and deprotection are per-
formed in separate steps, the HFA strategy saves two reaction
steps.10 However, if the product does not crystallize spon-
taneously, a separation from the hexafluoroacetone hydrate
formed as byproduct can be laborious and cause low yields.
1
lyophilized. HPLC-MS and H NMR confirmed that the
desired product 2a was formed. The 19F NMR spectra show
only the signal of trifluoroacetate, indicating that the OH-
group is deprotected. As expected, reaction time and degree
of racemization are solvent dependent. Dissociation of the
R-proton, which is more acidic in mandelic acid than in other
hydroxy acids because of the electron-withdrawing effect of
the phenyl group, is favored in polar solvents. In DMSO,
after 4.5 h reaction time, 18% of the diastereomer 2b was
detected. In DMF the reaction proceeds more slowly (24 h),
and again the tendency of racemization is high (11% 2b).11
On the other hand, in the apolar solvents DCM, chloroform
and toluene, despite long reaction times (up to 50 h) the
reaction proceeds practically without racemization (<1% 3b).
Finally, we found that THF is the solvent of choice. The
reaction time is short (5 h) and racemization can be neglected
(< 1%). The lactone is activated at the right degree to react
readily with amino groups but not with hydroxy groups under
solid-phase conditions, and therefore oligomerization was not
observed.12
(6) (a) Dutton, F. E.; Lee, B. H.; Johnson, S. S.; Coscarelli, E. M.; Lee,
P. H. J. Med. Chem. 2003, 46, 2057-2073. (b) Sefler, A. M.; Kozlowski,
M. C.; Guo, T.; Bartlett, P. A. J. Org. Chem. 1997, 62, 93-102. (c)
Scherkenbeck, J.; Plant, A.; Stieber, F.; Lo¨sel, P.; Dyker, H. Bioorg. Med.
Chem. Lett. 2002, 12, 1625-1628. (d) Chen, Y.; Bilban, M.; Foster, C. A.;
Boger, D. L. J. Am. Chem. Soc. 2002, 124, 5431-5440. (e) Krell, C. M.;
Seebach, D. Eur. J. Org. Chem. 2000, 1207-1218. (f) Joullie´, M. M.;
Portonovo, P.; Liang, B.; Richard, D. J. Tetrahedron Lett. 2000, 41, 9373-
9376.
(7) (a) Takahashi, T.; Nagamiya, H.; Doi, T.; Griffiths, P. G.; Bray, A.
M. J. Comb. Chem. 2003, 5, 414-428. (b) Ast, T.; Barron, E.; Kinne, L.;
Schmidt, M.; Germeroth, L.; Simmons, K.; Wenschuh, H. J. Peptide Res.
2001, 58, 1-11. (c) Lee, Y.; Silverman, R. B. Org. Lett. 2000, 2, 23, 3743-
3746.
In the next series of experiments we demonstrated that
the added excess of HFA-hydroxy acids can be recovered
and reused conveniently. Since carboxyl activation is an
(8) See the citations above.
(11) The di-depsipeptide 3b as reference was synthesized from HFA-
S-Man 1b. Both diastereomeres 3a and 3b have distinct retention times on
HPLC.
(12) Products resulting from nonselective multi-incorporation were
detected by addition of catalytic amounts of DMAP to the reaction mixture
and could be identified by HPLC-MS.
(9) Trost, B. M. Science 1991, 254 (5037), 1471-1477.
(10) (a) Burger, K.; Lange, T.; Rudolph, M. Heterocycles 2003, 59, 189-
198. (b) Radics, G.; Koksch, B.; El-Kousy, S. M.; Spengler, J.; Burger, K.
Synlett 2003, 12, 1826-1829. (c) Pumpor, K.; Windeisen, E.; Burger, K.
J. Heterocycl. Chem. 2003, 40, 435-442.
598
Org. Lett., Vol. 7, No. 4, 2005

Scheme 2. Solid-Phase Coupling of HFA-Hydroxy Acids
Scheme 3. Recovery and Reutilization of HFA-Hydroxy Acids
in a Solid-Phase Protocol
intrinsic property of the HFA compounds, no addition of
activation reagents and additives are required. H-Tyr(OtBu)-
Rink-MBHA-resin was treated with a 10-fold excess of HFA-
D-phenyllactic acid 1c in THF. After the reaction was
complete, the resin was filtered off, and the filtrate was
evaporated, redissolved and reused four more times. After
the fifth coupling cycle, 19F NMR spectroscopy showed the
signal of HFA-hydrate formed as byproduct in each coupling
turn. Compound 1c and HFA-hydrate show approximately
the same integration, indicating a consumption of ca. 50%
of the HFA compound 1c in agreement with the theoretically
estimated amount. The presence of HFA-hydrate do not
showed any negative influence on subsequent couplings. All
five samples of di-depsipeptide 2c showed a HPLC purity
of 98-99% with amounts of the diastereomer H-L-Phlac-
Tyr-NH2 (2d) less than 1%.13
Finally, the compatibility of the new method with the
Fmoc/tBu strategy on solid phase was demonstrated by
synthesizing five-membered cyclodepsipeptides 4a-c. The
linear precursors 3a-c were synthesized on the Wang
resin: the two hydroxy acid units were incorporated via HFA
derivatives according to the above-described method, and
the three Fmoc-AAs were esterificated with the solid-phase-
bound hydroxy groups using Steglich activation (DIC/
DMAP).14 Structural diversity was achieved by incorporation
of different hydroxy acids in the fourth position. After
cleavage from the resin, the linear precursors 3a-c were
obtained in purities between 57% and 92% (HPLC). After
HPLC purification, cyclization was performed under high
dilution conditions with DIC/HOAt/DIEA to give cyclodep-
sipeptides 4a-c in 28-31% yield and 95-98% purity
(HPLC).
In conclusion, HFA-hydroxy acids 1 are activated mono-
mers that work in solution as well as in solid phase. They
are excellently soluble in the most common organic solvents.
Their solutions are stable over a long time. Progress of
reactions can be monitored by 19F NMR spectroscopy.
Couplings proceed racemization-free in suitable solvents,
yielding products in a high purity. In solid-phase protocols,
added excess of HFA-hydroxy acids is easily recoverable
(13) Analogously to ref 11, the di-depsipeptide 3d as reference was
synthesized from HFA-S-Phlac 1d. Both diastereomeres 3c and 3d have
distinct retention times on HPLC.
and reuseable. From these findings we conclude that HFA-
hydroxy acids 1 are valuable coupling reagents for solid-
(14) Steglich, W.; Hoefle, G. Angew. Chem., Int. Ed. Engl. 1969, 8, 981.
Org. Lett., Vol. 7, No. 4, 2005
599

R-hydroxy acids in depsipeptides in laboratory scale or bulk
production.15
Scheme 4. Compatibility of the HFA Group with the Fmoc
Strategy: Synthesis of Cyclodepsipeptides
Acknowledgment. J.S. thanks the VW Foundation for
providing a grant for this project. This work was partially
supported by Generalitat de Catalunya [Grups Consolidats
(2001 SGR 0047) and CERBA], Ministerio de Ciencia y
Tecnolog´ıa (BQU 2003-00089), Pharma Mar Ltd., and
Barcelona Science Park.
Supporting Information Available: Experimental pro-
cedures and spectroscopic data. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL047653V
phase depsipeptide synthesis. Future possible applications
can be the incorporation of precious and/or laborious
(15) Bruckdorfer, T.; Marder, O.; Albericio, F. Curr. Pharm. Biotechnol.
2004, 5, 29-43.
600
Org. Lett., Vol. 7, No. 4, 2005