Development of a solid-phase approach to the natural product class of Ahp-containing cyclodepsipeptides

Article abstract of DOI:10.1002/ejoc.201101757

The 3-amino-6-hydroxy-2-piperidone (Ahp) containing cyclodepsipeptides are an interesting class of natural products that inhibit S1 (trypsin and chymotrypsin-like) serine protease in a reversible, noncovalent manner, turning them into potential chemical t

Full text of DOI:10.1002/ejoc.201101757

FULL PAPER
DOI: 10.1002/ejoc.201101757
Development of a Solid-Phase Approach to the Natural Product Class of
Ahp-Containing Cyclodepsipeptides
Sara C. Stolze,[a] Michael Meltzer,[a] Michael Ehrmann,[a] and Markus Kaiser*[a]
Keywords: Solid-phase synthesis / Natural products / Peptides / Masked glutamic aldehyde
The 3-amino-6-hydroxy-2-piperidone (Ahp) containing
cyclodepsipeptides are an interesting class of natural prod-
ucts that inhibit S1 (trypsin and chymotrypsin-like) serine
protease in a reversible, noncovalent manner, turning them
into potential chemical tools for protease research. Their sys-
tematic use in chemical biology is however hampered by
their tedious solution-phase chemical synthesis. To overcome
this limitation, we report a solid-phase approach to Ahp
cyclodepsipeptides that is based on the use of a masked
glutamic aldehyde moiety as a general Ahp precursor mole-
cule. As a proof-of-concept, we therefore recently reported
the solid-phase synthesis of Symplocamide A. Here, we want
to give a full account on the development and application of
the masked glutamic aldehyde moiety as well as the optimi-
zation of the solid-phase synthesis, which allowed the suc-
cessful synthesis of the natural product Symplocamide A.
Introduction
Cyanobacteria of freshwater or marine origin have been
recognized as proliferative producers of biologically active
secondary metabolites.[1] From different strains of cyano-
bacteria such as Microcystis, Anabaena, Oscillatoria, Nos-
toc, Lyngbya, or Symploca spp, an interesting class of natu-
ral products known as 3-amino-6-hydroxy-2-piperidone
(Ahp) containing cyclodepsipeptides have been isolated
(Figure 1). Ahp cyclopeptides are characterized by unique
chemical structures and potent bioactivities. In fact, Ahp
cyclodepsipeptides show selective, reversible, and most in-
triguingly noncovalent inhibition of S1 (trypsin- and chymo-
trypsin-like) serine proteases.[2]
Their noncovalent inhibition mode is of particular inter-
est for medicinal chemistry as well as chemical biology ap-
plications.[2h] Structural analysis of co-complexes of Ahp
cyclodepsipeptides obtained from natural product isolation
with S1 serine proteases have shed light on the underlying
molecular basis of this favorable inhibition mode. A crystal-
lographic study of a complex of the Ahp cyclodepsipeptide
A90720A and the serine protease trypsin elucidated that
Ahp cyclodepsipeptides bind in a substrate-like manner,
raising the question on how Ahp cyclodepsipeptides pre-
vent their own proteolytic processing. In one hypothesis,
the overall fold of the Ahp cyclodepsipeptides that critically
relies on the Ahp moiety results in an increased hydrolysis
activation barrier.[3] In this model, Ahp cyclodepsipeptides
Figure 1. Conserved structural features of Ahp cyclodepsipeptides.
Ahp cyclodepsipeptides are built up from a hexadepsipeptide
macrocycle that includes the eponymous 3-amino-6-hydroxy-2-pip-
eridone (Ahp) residue and one N-methylated phenylalanine deriva-
tive. In most known Ahp cyclodepsipeptides, a threonine moiety is
involved in the formation of the ester linkage; however, also deriva-
tives with a hydroxyproline residue on this position are known. In
addition, an exocyclic side chain is attached to the threonine or
hydroxyproline moiety.
mimic the mode of action of proteinaceous protease inhibi-
tors such as canonical serine protease inhibitors.[4] The
structural analysis of a complex of the Ahp cyclodepsipept-
ide Scyptolin A and the serine protease elastase however
indicated additional factors that contribute to the potent
inhibition: in this co-complex structure, the authors noted
that the hydroxy moiety of the Ahp residue is accommodat-
ing the position of the “catalytic” water molecule of the
hydrolysis reaction.[5] The authors suggest that this might
lead to an impairment of the hydrolysis reaction. Besides,
also steric hindrance of the movement of the catalytic triad
of the protease during the hydrolysis reaction has been sug-
gested as an important factor,[6] although the role of active
site dynamics on catalysis is still under debate.[7] In order
to gain deeper insights into the structural basis of inhibition
as well as for exploring the medicinal potential of Ahp
cyclodepsipeptides, the availability of structural analogs of
[a] Zentrum für Medizinische Biotechnologie,
Universität Duisburg-Essen,
Universitätsstr. 2, 45117 Essen, Germany
Fax: +49-201-183-4982
E-mail: markus.kaiser@uni-due.de
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201101757.
1616
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1616–1625

Solid-Phase Approach to Ahp-Containing Cyclodepsipeptides
Ahp cyclodepsipeptides for structure–activity studies would trypsin inhibitor Symplocamide A (Figure 2). The present
be highly desirable. So far, this approach is however ham- work thereby represents a full account on our previously
pered due to a lack of suitable synthetic methods for rapid published work on its synthesis.[9]
access to derivatives for structure–activity studies.
To date, Ahp cyclodepsipeptides are usually obtained by
natural product isolation, resulting however in only minute
amounts of substance. In addition, total syntheses of the
Results and Discussion
Ahp cyclodepsipeptides Somamide A and Micropeptin T-
Synthetic Strategy
20 through extensive solution-phase total synthesis have
been reported.[8] To gain a more convenient access to Ahp
In the total syntheses of the Ahp cyclodepsipeptides
cyclodepsipeptides and to non-natural analogs for struc- Micropeptin T-20 and Somamide A, Yokokawa and co-
ture–activity studies, we therefore aimed to establish a solid- workers demonstrated that the Ahp hemiaminal forms
phase-based synthetic route to this natural product class. spontaneously from a glutamic aldehyde macrolactam that
To illustrate the potential of the developed methodology, was generated at a late stage of the synthesis.[8] Conse-
we investigated a solid-phase-based synthetic route to the quently, we envisioned that Ahp cyclodepsipeptides could
Ahp cyclodepsipeptide natural product and potent chymo- be obtained upon solid-phase synthesis of a cyclodepsipept-
ide with a masked glutamic aldehyde residue that is un-
masked at the end of the synthesis. Such an approach
seemed feasible, as the Meldal group demonstrated that the
oxidative cleavage of double bonds by a dihydroxylation–
glycol cleavage protocol represents a powerful approach for
the generation of aldehydes on solid phase.[10] As double
bonds are stable to acidic as well as basic reaction condi-
tions, such a masked aldehyde residue would be compatible
with Fmoc/tBu as well as Merrifield solid-phase peptide
synthesis conditions. Furthermore, a substituted double
bond would allow linkage of the masked glutamic aldehyde
residue to solid support through its side chain, thereby en-
abling solid-phase manipulation of the N-terminal as well
as the C-terminal part of the masked aldehyde residue and
liberation of the glutamic aldehyde moiety upon resin cleav-
age (Scheme 1).
Figure 2. Chemical structure of the target Ahp cyclodepsipeptide
Symplocamide A.
Scheme 1. (a) Proposition of a homoallylglycine derivative as a masked glutamic aldehyde residue for generation of the Ahp residue and
(b) its potential application to the solid-phase synthesis of Ahp cyclodepsipeptides.
Eur. J. Org. Chem. 2012, 1616–1625
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1617

S. C. Stolze, M. Meltzer, M. Ehrmann, M. Kaiser
FULL PAPER
For linkage of the masked glutamic aldehyde residue to as this moiety can be removed under mild and orthogonal
solid support, a building block with a terminal carboxylic conditions during solid-phase synthesis. Consequently, this
acid moiety at its side chain for facile coupling to the resin protecting group change called for selective hydrogenation
was chosen. Moreover, in order to allow solid-phase peptide of the benzyl ester without concurrent reduction of the
synthesis as well as an on-resin macrolactamization, suitable double bond of the α,β-unsaturated carboxylic acid ester.
protecting groups for the N-terminal amino and C-terminal To achieve this, the hydrogenation conditions of Coleman
carboxyl group were required. We therefore started our et al. were employed.[14] These conditions use Et3SiH as a
studies on the solid-phase synthesis of Ahp cyclodepsipept- hydrogen source and PdII acetate as the corresponding cata-
ides with the development of a suitably masked glutamic lyst and delivered desired free acid 9 in 62% yield. In our
aldehyde residue. In a second step, we then employed this hands, however, yields were dependent of the reaction batch
building block for a solid-phase synthesis of an Ahp cyclo- size. In particular, large batch reactions led to lower yields.
depsipeptide derivative.
Despite this drawback, the free acid was subsequently
transformed into allyl ester intermediate 2 by allylation of
the corresponding cesium salt with allyl bromide as de-
scribed by Kunz et al. in 76% yield.[15]
Synthesis of a Masked Glutamic Aldehyde Building Block
We first evaluated α,β-unsaturated carboxylic acid 1 with
N-terminal Fmoc and C-terminal allyl ester protecting
groups as a masked glutamic aldehyde building block for
solid-phase synthesis (Scheme 2). We thought that this
building block could be obtained from intermediate 2,
which could be derived from corresponding glutamic alde-
hyde 3 through a Wittig reaction. As glutamic aldehydes,
however, tend to form hemiaminals, which significantly
hampers the efficiency of Wittig reactions,[11] we decided to
fully protect the amine functionality with two Boc protect-
ing groups.[12] This would allow the use of the commercially
available N-Boc glutamic acid benzyl ester 4 as a starting
material for the synthesis.
With this plan in mind, we initiated the synthesis of the
masked glutamic aldehyde building block, converting first
N-Boc glutamic acid benzyl ester 4 into methyl ester 5 by
methylation with methyl iodide (Scheme 3). Introduction of
the second Boc protecting group on the N-terminal amino
group was achieved by following Apella’s protocol.[13] Con-
sequently, 5 was treated with Boc2O and DMAP as a cata-
lyst, delivering intermediate 6 in 98% yield. Chemoselective
reduction of the methyl ester without concurrent reduction
of the benzyl ester was performed with DIBAL-H as the
reducing agent in toluene at –78 °C. Despite these mild re-
action conditions, desired aldehyde 7 was however isolated
in only 27% yield; alongside, reduction of the benzyl ester
as well as overreduction to the alcohol was observed, ex-
plaining the low yield in this reaction step. Subsequent Wit-
tig olefination of aldehyde 7 led to α,β-unsaturated tert-but-
yl carboxylic acid ester 8 in 84% yield. At this stage, we
went to replace the benzyl ester protecting group with an
Scheme 3. Synthesis of the bis-Boc protected masked glutamic al-
dehyde intermediate 2. Reagents and conditions: (a) MeI, K2CO3,
acetone, reflux, 4 h (quant). (b) Boc2O, DMAP, CH3CN, r.t., 20 h
(98%). (c) DIBAL-H, toluene, –78 °C, 2 h (27%). (d) 1. Ph3P,
ClCH2COOtBu, 2 m NaOH/toluene (1:1), 1 h, r.t.; 2. 7, ylide,
CH2Cl2, r.t., overnight (84%). (e) Pd(OAc)2, Et3SiH, Et3N,
CH2Cl2, r.t., overnight (62%). (f) 1. Cs2CO3, MeOH; 2. allyl brom-
allyl ester moiety. We chose an allyl ester protecting group, ide, DMF, r.t., overnight (76%).
Scheme 2. Retrosynthesis of a masked glutamic aldehyde building block featuring a N-terminal Fmoc protecting group and an α,β-
unsaturated carboxylic acid moiety for solid-phase synthesis.
1618
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1616–1625

Solid-Phase Approach to Ahp-Containing Cyclodepsipeptides
Although we successfully obtained intermediate 2, the
With 2 in hand, desired Fmoc-protected building block
overall yields were rather low, mainly due to inefficient re- 1 for solid-phase synthesis should be available by an ex-
duction to aldehyde 7. We therefore sought an alternative change of the N-terminal protecting group and cleavage of
synthetic approach to 2 that circumvented this inefficient the tert-butyl ester (Scheme 5). Cleavage of the Boc and
step. To this end, commercially available N-Boc glutamic tert-butyl ester protecting groups with TFA in dichloro-
acid benzyl ester 4 was first transformed into glutamic methane swiftly led to 17. However, synthesis of 1 proved to
alcohol 10 by using Shiori’s protocol (Scheme 4).[8b] Conse- be difficult and reaction yields were rather irreproducible,
quently, conversion of the acid functionality to a mixed an- ranging from 49% in the best case to no product formation
hydride, followed by reduction with sodium borohydride led in the worst case.
to 10 in 84% yield. The hydroxy function of 10 was then
protected as a tert-butyldimethylsilyl (TBDMS) ether; re-
sulting intermediate 11 was again blocked completely by
introduction of a second Boc group. The benzyl ester pro-
tecting group of bis-Boc-protected intermediate 12 was re-
moved by hydrogenation to enable the required exchange of
the protecting groups at the acid functionality. To achieve
this, a modified hydrogenation protocol was used and, un-
der the standard conditions involving the use of methanol
or ethanol as solvent, resulted in partial cleavage of the silyl
ether moiety. Use of ethyl acetate during hydrogenation
however allowed efficient benzyl cleavage, thereby generat-
ing free acid 13 in quantitative yield.[16] The allyl protecting
group was introduced again by using Kunz’s conditions,
Scheme 5. Attempted synthesis of Fmoc-protected masked glut-
amic aldehyde building block 1. Reagents and conditions: (a) TFA/
CH2Cl2 (1:1), r.t., 2 h (quant); (b) FmocCl, NaHCO3 (sat.), pH Ͼ
8, H2O/dioxane (1:1), 40 °C, overnight (max. 49%).
Even more devastating, we noticed that if this building
yielding allyl ester 14 in 95% yield. Subsequent selective block was coupled to the solid phase through its side chain
removal of the silyl ether moiety however proved to be more carboxyl moiety, subsequent Fmoc cleavage led to unde-
difficult than expected, as standard TBAF-mediated silyl sired side products, preventing its use in solid-phase synthe-
cleavage resulted in the formation of an undesired lactone. sis. We deduced that the low yield during installation of the
An alternative method consisting of the use of phos- Fmoc group on the amino group as well as its unexpected
phomolybdic acid immobilized on silica gel was more ef- reactivity on solid phase is most probably a result of the
ficient but resulted in the cleavage of one of the Boc groups α,β-unsaturated carboxyl moiety that renders this com-
at the N-terminal amino group.[17] Desired alcohol 15 was pound prone to nucleophilic attack, for example from the
finally obtained by a very mild, CuII-mediated cleavage re- amino group.
action.[18] Dess–Martin oxidation by using standard condi-
We therefore decided to replace the conjugated double
tions led to aldehyde intermediate 16, which after a final bond of the α,β-unsaturated carboxyl moiety with a masked
Wittig reaction, led to desired intermediate 2 in 94% yield. glutamic aldehyde building block in which the double bond
Scheme 4. Alternative synthesis of bis-Boc protected intermediate 2. Reagents and conditions: (a) 1. EtOCOCl, Et3N, THF, –15 °C, 1 h;
2. NaBH4, THF/H2O (1:1), –15 °C, 1 h, then r.t., 6 h (84%). (b) TBDMSCl, imidazole, CH2Cl2, 0 °C to r.t., overnight (95%). (c) Boc2O,
DMAP, CH3CN, r.t., 24 h (81% based on recovered starting material). (d) 10% Pd/C, H2, EtOAc, r.t., 2.5 h (99%). (e) 1. Cs2CO3, MeOH;
2. allyl bromide, DMF, r.t., overnight (95%). (f) CuCl2·2H2O, acetone/H2O (95:5), reflux, 2 h (92%). (g) Dess–Martin-periodinane,
CH2Cl2, r.t., 1 h (84%). (h) 1. Ph3P, ClCH2COOtBu, 2 m NaOH/toluene (1:1), 1 h, r.t.; 2. 16, ylide, CH2Cl2, r.t., overnight (94%).
Eur. J. Org. Chem. 2012, 1616–1625
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1619

S. C. Stolze, M. Meltzer, M. Ehrmann, M. Kaiser
FULL PAPER
was isolated. Moreover, we envisioned that the previously methyl polystyrene resin was acylated with acetic anhydride
difficult exchange of the two N-terminal Boc groups for an and subjected to the oxidative cleavage cocktail, and the
Fmoc group could become obsolete, if a resin would be resulting reaction products were analyzed, revealing insuf-
used that is stable both to the acidic and basic conditions. ficient stability of the resin to the cleavage conditions. To
This would allow the use of Boc as well as Fmoc chemistry overcome these problems, two resins based on a PEG core
during solid-phase synthesis. Consequently, we changed our (NovaPEG amino resin, aminoPEGA resin) instead of a
attention to the synthesis of masked glutamic aldehyde polystyrene core were next tested. Both resins were also acy-
building block 18 that featured both a bis-Boc protected lated and then subjected to the same cleavage conditions.
amino group as well as an isolated double bond in the side This time, fewer impurities were observed with both resins,
chain residue (Scheme 6a). Indeed, this compound should demonstrating that in principle both resins could be used
be obtainable by a Wittig reaction from aldehyde intermedi- for the subsequent solid-phase synthesis. For practical
ate 16, which we were able to obtain from the commercially reasons, we therefore chose to use the NovaPEG resin for
available starting material N-Boc glutamic acid benzyl ester our future synthesis experiments.
4 (Scheme 4).
We then started with a solid-phase synthesis of a Sym-
plocamide A model compound (19, Scheme 7). This deriva-
tive contained a “standard” N-methylphenylalanine residue
instead of the 3-bromo-4-methoxy-N-methyl tyrosine moi-
ety of Symplocamide A. Moreover, the glutamine moiety
was incorporated with a trityl protecting group on its side
chain amide residue as we used the standard Fmoc-
Gln(Trt)-OH amino acid during solid-phase synthesis. We
aimed to perform a full deprotection of the Ahp cyclodepsi-
peptide after oxidative cleavage from the resin.
Thus, the synthesis started with coupling of building
block 18 to a NovaPEG amino resin using DIC/HOBt acti-
vation. This was followed by a capping step. The Boc
groups were removed using 50% TFA in DCM followed
by neutralization with 10% triethylamine in DCM. Then, a
standard solid-phase peptide synthesis protocol consisting
of HBTU/HOBt activation was used to couple Fmoc-pro-
tected citrulline. Removal of the Fmoc protecting group
with 20% piperidine in DMF was followed by coupling of
Fmoc-protected threonine by using again HBTU/HOBt ac-
tivation. Of note, the hydroxy group of the threonine side
chain was incorporated without a protecting group. Again,
the Fmoc protecting group was cleaved and the next amino
acid, Fmoc-Gln(Trt)-OH, was coupled. Fmoc cleavage and
subsequent coupling of butyric acid with HBTU/HOBt ac-
tivation finished the elongation of the N-terminal end of
the Ahp cyclodepsipeptide. Next, the esterification of the
hydroxy group of the threonine moiety with Fmoc-Val-OH
was performed. In order to achieve an efficient reaction,
quadrupole couplings with Fmoc-Val-OH and diisoprop-
ylcarbodiimide as the coupling agent and DMAP as a cat-
alyst were performed for 2 h (per coupling) to drive the re-
action to completion. After subsequent Fmoc cleavage,
again with 20% piperidine in DMF, the Fmoc-protected N-
methyl amino acid phenylalanine was coupled after HBTU/
HOBt activation to the resin. Fmoc cleavage then set the
Scheme 6. (a) Retrosynthesis to the alternative masked glutamic al-
dehyde building block 18 that features an N-terminal bis-Boc pro-
tecting group and an isolated double bond. This building block
should be available via aldehyde 16 for which a synthesis from the
commercially available starting material N-Boc glutamic acid
benzyl ester 4 has been established. (b) Synthesis of 18. Reagents
and conditions: (a) 1. BrPh3P(CH2)3CO2H, LHMDS, THF, r.t.,
55 min; 2. 16, THF, 0 °C to r.t., overnight (64%).
Consequently, Wittig reaction with triphenylphosphon-
ium butyric acid and aldehyde intermediate 16 led to the
desired masked glutamic aldehyde building block 18 (with
a Z/E ratio of 9:1) in 64% yield.
Development of a Solid-Phase Synthesis for Ahp
Cyclodepsipeptides
With 18 in hand, we commenced with the solid-phase- stage for the coupling of Fmoc-Ile-OH. In order to assure
based approach to Ahp cyclodepsipeptides according to the efficient coupling to the N-methyl group, the strong activa-
envisioned strategy (Scheme 1). As the planned solid-phase ting reagent PyBrOP was used instead of the HBTU/HOBt
synthesis features a rather unusual oxidative cleavage as a reagent pair. Cleavage of the Fmoc group, followed by
key step, we first investigated the oxidative cleavage reaction cleavage of the allyl ester by a protocol established by Vaz
with different resin types. As reagents for cleavage, we used et al. was used to generate a linear precursor molecule ready
sodium periodate, osmium tetroxide, and DABCO to sup- for subsequent on-resin macrolactamization.[19] The cycliza-
press the formation of hydroxymethyl ketones as established tion reaction was then performed with PyBOP and HOBt
by the Meldal group.[10b] Thus, first “standard” amino- as coupling reagents and a long reaction time of 24 h. Oxi-
1620
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1616–1625

Solid-Phase Approach to Ahp-Containing Cyclodepsipeptides
Scheme 7. Solid-phase peptide synthesis of Symplocamide A model Ahp cyclodepsipeptide 19. Reagents and conditions: (a) 1. Amino-
NovaPEG resin (0.66 mmol/g), 18 (2.4 equiv.), HOBt, DIC, CH2Cl2/DMF (9:1), r.t., 24 h; 2. CH2Cl2/DIEA/Ac2O (3:1:1), r.t., 3 h.
(b) 1. TFA/CH2Cl2 (1:1), r.t., 1 h; 2. Et3N/CH2Cl2 (1:9), r.t., 2ϫ10 min; 3. Fmoc-Cit-OH, HOBt, HBTU, DIEA, DMF, r.t., 5 h (30% by
Fmoc determination, 2 steps). (c) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. Fmoc-Thr-OH, HOBt, HBTU, DIEA, DMF, r.t., 2 h.
(d) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. Fmoc-Gln(Trt)-OH, HOBt, HBTU, DIEA, DMF, r.t., 2 h (88% by Fmoc determination,
2 steps). (e) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. butyric acid, HOBt, HBTU, DIEA, DMF, r.t., 2 h. (f) Fmoc-Val-OH, DIC,
DMAP, CH2Cl2/DMF (9:1), r.t., 4ϫ2 h (79% by Fmoc determination, 2 steps). (g) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. Fmoc-
NMe-Phe-OH, HOBt, HBTU, DIEA, DMF, r.t., 2 h. (h) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. Fmoc-Ile-OH, PyBrOP, DIEA,
DMF, r.t., 24 h (60% by Fmoc determination, 2 steps). (i) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. Pd(PPh3)4, morpholine, CH2Cl2,
r.t., 30 min; 3. PyBOP, HOBt, DIEA, DMF, r.t., 24 h. (j) NaIO4, OsO4 (0.1 m in tBuOH), DABCO, H2O/THF (1:1), r.t., 20 h (1.2 overall
yield%).
dative cleavage was then achieved with sodium periodate, In general, the reaction conditions for removing an amide
osmium tetroxide, and DABCO, following essentially the trityl group are very harsh and require the use of 95% TFA.
protocol established by the Meldal group, which after When we submitted peptide 19 to these conditions, we ob-
HPLC purification delivered desired Ahp cyclodepsipeptide served rapid and complete decomposition of the Ahp cyclo-
19 in 1.2% overall yield.
depsipeptide, most probably due to side reactions on the
With this Ahp cyclodepsipeptide in hand, we next tested Ahp moiety. These findings indicate that Ahp cyclodepsi-
the cleavage of the trityl group from the glutamine residue. peptides are not stable to strong acidic conditions, calling
Eur. J. Org. Chem. 2012, 1616–1625
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1621

S. C. Stolze, M. Meltzer, M. Ehrmann, M. Kaiser
FULL PAPER
Scheme 8. Solid-phase peptide synthesis of Symplocamide A. Reagents and conditions: (a) 1. AminoNovaPEG resin (0.66 mmol/g), 16
(2.5 equiv.), HOBt, HBTU, DIEA, CH2Cl2/DMF (9:1), r.t., 24 h; 2. CH2Cl2/DIEA/Ac2O (3:1:1), r.t., 2ϫ2 h. (b) 1. TFA/CH2Cl2 (1:1),
r.t., 1 h; 2. Et3N/CH2Cl2 (1:9), r.t., 2ϫ10 min; 3. Fmoc-Cit-OH, HOBt, HBTU, DIEA, DMF, r.t., 4 h (36%, by Fmoc determination, 2
steps). (c) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. Fmoc-Thr-OH, HOBT, HBTU, DIEA, DMF, r.t., 5 h. (d) 1. Piperidine/DMF (1:4),
r.t., 2ϫ15 min; 2. Boc-Gln-OH, HOBT, HBTU, DIEA, DMF, 4 h. (e) 1. TFA/CH2Cl2 (1:1), r.t., 1 h; 2. Et3N/CH2Cl2 (1:9), r.t., 2ϫ10 min;
3. butyric acid, HOBT, HBTU, DIEA, DMF, r.t., 4 h. (f) Fmoc-Val-OH, DIC, DMAP, CH2Cl2/DMF (9:1), 4ϫ2 h, (92% by Fmoc
determination, 4 couplings). (g) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. Boc-mTyr(3-Br, 4-OMe)-OH, HOBT, HBTU, DIEA, DMF,
4.5 h. (h) 1. TFA/CH2Cl2 (1:1), r.t., 1 h; 2. Et3N/CH2Cl2 (1:9), r.t., 2ϫ10 min; 3. Fmoc-Ile-OH, PyBrOP, DIEA, DMF, 24 h (73%, by
Fmoc determination, 2 couplings). (i) 1. Piperidine/DMF (1:4), r.t., 2ϫ15 min; 2. Pd(PPh3)4, morpholine, CH2Cl2, r.t., 30 min; 3. PyBOP,
HOBt, DIEA, DMF, r.t., 24 h. (j) NaIO4, OsO4 (0.1 m in tBuOH), DABCO, H2O/THF (1:1), r.t., 20 h (3.0% overall yield).
for a synthesis route in which such protecting groups are obtained for the synthetic compound corresponded well to
avoided or are removed before formation of the Ahp moi- the data published for the natural product.[2g,9]
ety.
After showing that the masked glutamic aldehyde ap-
proach enables the generation of Ahp cyclodepsipeptides, Inhibition Properties of Symplocamide-Based Ahp
we continued our investigations with the synthesis of the Cyclodepsipeptides
natural product Symplocamide A (Scheme 8). In order
to prevent the cleavage of the trityl protecting group, we Ahp cyclodepsipeptides are known inhibitors of S1 serine
used Boc-Gln-OH instead of Fmoc-Gln(Trt)-OH as an proteases. For Symplocamide A, a potent inhibition of the
amino acid building block. Moreover, in this case, the serine protease chymotrypsin was reported, as expected
Symplocamide A-specific amino acid Boc-N-methyl 3- from the large, hydrophobic citrulline moiety that – accord-
bromo-4-methoxyphenylalanine was used instead of Fmoc- ing to the binding mode that has been elucidated with other
N-methylphenylalanine. This amino acid derivative was pre- analogs of the Ahp cyclodepsipeptides – should bind to the
pared by solution-phase chemistry by using standard meth- S1 pocket. We therefore tested the inhibition potency of our
ods.
synthetically prepared Symplocamide A analog – and as a
For the solid-phase synthesis of Symplocamide A, some further proof for its successful synthesis – we observed a
reaction conditions were slightly changed (Scheme 8). For potent Ki value of 0.32Ϯ0.09 μm, which is in good accord-
example, for loading the resin with the first amino acid de- ance with that reported (ref.[2g] IC50 = 0.38Ϯ0.08 μm).
rivative, that is, masked glutamic aldehyde building block
In addition, it was suggested that Symplocamide A might
18, HBTU and HOBt were used this time, resulting in a also act as an inhibitor of the proteasomal β5 subsite that
loading efficiency of 36%. However, most of the reaction displays chymotrypsin-like proteolytic activity.[2g] Conse-
conditions were kept as previously established and oxidative quently, we performed a biochemical inhibition assay by
cleavage from the resin delivered the crude natural product using standard conditions but unfortunately could not de-
in fair purity (see the Supporting Information, Figure 1, for tect any significant inhibition.[9] We furthermore tested the
the corresponding HPLC chromatogram). Subsequent puri- inhibitory potential of Ahp cyclodepsipeptide 19, which
fication by HPLC led to the desired natural product sym- also should represent a potent chymotrypsin inhibitor. In-
plocamide A in 3% overall yield. The spectroscopic data deed, 19 displayed potent chymotrypsin inhibition with a
1622
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1616–1625

Solid-Phase Approach to Ahp-Containing Cyclodepsipeptides
Na]+, 224 (100) [M – Boc]+. HRMS (EI): calcd. for C17H25O5N+
[M + H]+ 324.1805; found 324.1810.
Ki value of 1.09Ϯ0.11 μm, ranking this compound almost
equipotent to the natural product.
(S)-Benzyl-2-(tert-butoxycarbonylamino)-5-(tert-butyldimethylsilyl-
oxy)pentanoate (11): Compound 10 (3.92 g, 12.1 mmol) was dis-
solved in DCM (80 mL) and cooled to 0 °C. Imidazole (2.00 g,
30.3 mmol, 1.2 equiv.) and tert-butyldimethylsilyl chloride (2.19 g,
14.5 mmol, 2.5 equiv.) were added, and the reaction mixture was
stirred at 0 °C for 1 h and warmed to room temperature overnight.
Water was added to the reaction mixture, the phases were sepa-
rated, and the aqueous phase was extracted with ethyl acetate
(3ϫ80 mL). The combined organic extracts were washed with 1 m
aq. KHSO4 solution (150 mL) and aq. saturated NaHCO3 solution
(150 mL) and dried with Na2SO4. After evaporation of the solvent
to dryness, the crude product was purified by flash column
chromatography (10% ethyl acetate in cyclohexane) to afford 11
Conclusions
In summary, we have developed a first solid-phase ap-
proach towards the synthesis of natural products from the
Ahp cyclodepsipeptide class and exemplified this approach
by the synthesis of the natural product Symplocamide A.
Towards this end, the use of a masked glutamic aldehyde
building block as a precursor for the Ahp moiety repre-
sented a key element in the synthetic strategy. The oxidative
cleavage conditions reported herein are – to the best of our
knowledge – used for the first time to generate an aldehyde (5.05 g, 11.5 mmol, 95%) as a yellow oil. TLC (10% ethyl acetate
in cyclohexane): Rf = 0.35. [α]2D0 = –5.8 (c = 1.13, CHCl3). 1H NMR
(400 MHz, CDCl3): δ = 7.32–7.35 (m, 5 H), 5.13–5.19 (m, 2 H),
residue and simultaneously achieve cleavage from the solid
support. The synthesis of the Ahp precursor was achieved
4.31–4.36 (m, 1 H), 3.58–3.61 (t, J = 6.05 Hz, 2 H), 1.94–1.95 (m,
from commercially available N-Boc glutamic acid benzyl es-
1 H), 1.76–1.85 (m, 1 H), 1.51–1.56 (m, 2 H), 1.43 (s, 9 H), 0.88 (s,
ter and in high overall yields. The solid-phase synthesis con-
9 H), 0.03 (s, 6 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 172.6,
tains some unusual transformations that required the use
of nonstandard reaction conditions, in particular the solid-
155.4, 135.5, 128.5, 128.3, 128.1, 79.6, 66.8, 62.2, 53.3, 29.0, 28.3,
25.9, 18.3, –5.4 ppm. LC–MS: tR = 13.31 min (C18), m/z (%) = 460
phase esterification of the threonine residue as well as an
on-resin macrolactamization. We anticipate that the devel-
oped synthetic approach should present a general strategy
for the synthesis of Ahp cyclodepsipeptides, thereby paving
the way for the synthesis of tailor-made Ahp cyclodepsi-
peptides. This should enable more facile use of these inhibi-
tors as chemical biology tools in protease research.
(40) [M + Na]+, 338 (100) [M – Boc]+ HRMS (EI): calcd. for
.
C23H40O5NSi+ [M + H]+ 438.2670; found 438.2670.
(S)-Benzyl-2-[bis(tert-butoxycarbonyl)amino]-5-(tert-butyldimethyl-
silyloxy)-pentanoate (12): Compound 11 (4.94 g, 11.3 mmol) was
dissolved in acetonitrile (80 mL). After the addition of 4-(dimeth-
ylamino)pyridine (2.07 g, 16.9 mmol, 1.5 equiv.) at room tempera-
ture, the mixture was stirred for 10 min. Di-tert-butyl dicarbonate
(13.0 mL, 56.5 mmol, 5 equiv.) was added, and the reaction mixture
was stirred overnight at room temperature, thereby turning from a
pale yellow color to bright red. The reaction was quenched by the
addition of water (80 mL), and the aqueous phase was extracted
with ethyl acetate (3ϫ80 mL). To accelerate the sluggish phase sep-
aration, brine was added in small portions. The combined organic
extracts were washed with 1 m aq. HCl (150 mL) and brine
(4 ϫ150 mL). After evaporation of the solvents to dryness, the
crude mixture of product and unreacted starting material was puri-
fied by flash column chromatography (5% ethyl acetate in cyclo-
hexane) to yield 12 (3.88 g (7.2 mmol, 64%, 81% based on reco-
vered starting material) as a colorless oil along with unreacted
starting material 11. TLC (5% ethyl acetate in cyclohexane): Rf =
0.26. [α]2D0 = –20.2 (c = 1.26, CHCl3). 1H NMR (400 MHz, CDCl3):
δ = 7.31–7.33 (m, 5 H), 5.11–5.18 (m, 2 H), 4.88–4.92 (m, 1 H),
3.61–3.65 (dt, J = 6.40, 1.38 Hz, 2 H), 2.12–2.14 (m, 1 H), 1.91–
2.01 (m, 1 H), 1.54–1.61 (m, 2 H), 1.44 (s, 18 H), 0.88 (s, 9 H), 0.03
(s, 6 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 170.7, 152.2,
135.7, 128.4, 128.0, 127.9, 82.9, 66.7, 62.6, 58.1, 29.6, 27.9, 26.0,
25.9, 18.3, –5.3 ppm. LC–MS: tR = 11.29 min (C4), m/z (%) = 1097
(100) [2M + Na]+, 438 (35) [M – Boc]+. HRMS (EI): calcd. for
Experimental Section
Chemical Synthesis of Selected Compounds
(S)-Benzyl-2-(tert-butoxycarbonylamino)-5-hydroxypentanoate (10):
Boc-(S)-glutamic acid benzyl ester (4, 5.00 g, 14.8 mmol) was dis-
solved in anhydrous THF (60 mL) under an argon atmosphere and
cooled to –15 °C in an ice-acetone cooling bath. A white precipitate
formed upon successive addition of triethylamine (6.2 mL,
44.5 mmol, 3 equiv.) and ethyl chloroformate (4.2 mL, 44.5 mmol,
3 equiv.). The reaction mixture was maintained at –15 °C for 1 h.
After this time, sodium borohydride (2.24 g, 59.3 mmol, 4 equiv.)
dissolved in water (60 mL) was slowly added at –15 °C resulting in
the dissolution of the precipitate. After stirring for another 1 h at
–15 °C, the reaction mixture was warmed to room temperature. A
TLC check after 5 h indicated the complete consumption of the
starting material. The reaction mixture was then quenched with 1 m
aq. KHSO4 solution (100 mL), and the aqueous phase was ex-
tracted with ethyl acetate (3ϫ100 mL). The combined organic ex-
tracts were washed with brine (150 mL) and dried with Na2SO4.
After evaporation to dryness, the crude material was purified by
C56H94O14N2NaSi2 [2M + Na]+ 1097.6136; found 1097.6141.
+
flash column chromatography (50% ethyl acetate in cyclohexane) (S)-2-[Bis(tert-butoxycarbonyl)amino]-5-(tert-butyldimethylsilyloxy)-
to afford 10 (4.03 g, 12.5 mmol, 84%) as a pale yellow oil. TLC
(50% ethyl acetate in cyclohexane): Rf = 0.36. [α]2D0 = –2.3 (c =
0.98, CHCl3) {ref.[8b] [α]2D4 = –3.9 (c = 1.0, CHCl3)}. 1H NMR
pentanoic acid (13): Compound 12 (5.85 g, 10.9 mmol) was dis-
solved in ethyl acetate (80 mL) and transferred to a two-necked
round-bottomed flask equipped with a rubber septum and a stop-
(400 MHz, CDCl3): δ = 7.32–7.34 (m, 5 H), 5.10–5.19 (m, 2 H), cock under an argon atmosphere. Pd/C (10%) (2.93 g, 5 wt.-%) was
4.29–4.34 (m, 1 H), 3.58 (t, J = 6.20 Hz, 2 H), 1.93–2.00 (m, 1 H),
added at room temperature in small portions through a funnel,
after each portion the funnel was flushed with ethyl acetate (70 mL
1.84–1.91 (m, 1 H), 1.66–1.76 (m, 1 H), 1.53–1.60 (m, 2 H), 1.41
(s, 9 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 172.6, 141.0, in total). After the complete addition, the flask was equipped with
135.3, 128.5, 128.4, 128.2, 79.8, 65.0, 61.8, 53.2, 29.2, 28.2,
25.4 ppm. LC–MS: tR = 8.90 min (C18), m/z (%) = 346 (30) [M +
a hydrogenation balloon and carefully flushed by opening the stop-
cock from time to time. The hydrogenation was carried out for
Eur. J. Org. Chem. 2012, 1616–1625
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1623

S. C. Stolze, M. Meltzer, M. Ehrmann, M. Kaiser
FULL PAPER
2.5 h at room temperature, then the reaction mixture was filtered
through a pad of Celite, which was rinsed thoroughly with ethyl
acetate (4ϫ100 mL). Evaporation to dryness yielded 13 (4.84 g,
10.8 mmol, 99%) as a white solid. [α]2D0 = –19.8 (c = 0.91, CHCl3).
rial. The reaction mixture was quenched by the addition of aq.
saturated K2CO3 solution (60 mL), the phases were allowed to
separate, and the aqueous layer was extracted with ethyl acetate
(3 ϫ 60 mL). The combined organic extracts were washed with
1H NMR (400 MHz, CDCl3): δ = 4.91–4.95 (m, 1 H), 3.62–3.65 brine (150 mL) and dried with Na2SO4. After evaporation to dry-
(dt, J = 6.30, 1.74 Hz, 2 H), 2.11–2.20 (m, 1 H), 1.91–2.00 (m, 1 ness, the crude product was purified by flash column chromatog-
H), 1.54–1.59 (m, 2 H), 1.50 (s, 18 H), 0.89 (s, 9 H), 0.04 (s, 6 H)
ppm. 13C NMR (100 MHz, CDCl3): δ = 177.1, 152.0, 83.2, 62.5,
57.8, 29.5, 27.9, 26.1, 25.9, 18.3, –5.3 ppm. LC–MS: tR = 11.31 min
(C18), m/z (%) = 917 (100) [2M + Na]+, 347 (20) [M – Boc]+.
raphy (17 % ethyl acetate in cyclohexane) to afford 16 (1.34 g,
3.6 mmol, 84%) as a colorless oil, which was directly used for the
next step. TLC (17 % ethyl acetate in cyclohexane): Rf = 0.23.
1
[α]2D0 = –26.2 (c = 1.26, CHCl3). H NMR (400 MHz, CDCl3): δ =
HRMS (EI): calcd. for C42H82O14N2NaSi2 [2M + Na]+ 917.5197;
9.75 (s, 1 H), 5.83–5.92 (m, 1 H), 5.27–5.32 (dd, J = 17.20, 1.50 Hz,
1 H), 5.18–5.21 (dd, J = 10.50, 1.40 Hz, 1 H), 4.86–4.90 (m, 1 H),
4.58–4.60 (m, 2 H), 2.48–2.59 (m, 3 H), 2.13–2.20 (m, 1 H), 1.47
(s, 18 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 200.8, 169.9,
152.0, 131.7, 118.2, 83.4, 65.7, 57.4, 40.4, 27.9, 22.3 ppm. LC–MS:
tR = 11.11 min (C18), m/z (%) = 394 (80) [M + Na]+, 294 (90) [M –
Boc + Na]+. HRMS (EI): calcd. for C18H30O7N+ [M + H]+
372.2017; found 372.2018.
+
found 917.5191.
(S)-Allyl-2-[bis(tert-butoxycarbonyl)amino]-5-(tert-butyldimethylsil-
yloxy)-pentanoate (14): Compound 13 (4.61 g, 10.3 mmol) and
cesium carbonate (3.36 g, 10.3 mmol, 1 equiv.) were dissolved in
anhydrous methanol (40 mL). The solvent was evaporated, and the
remaining solid was coevaporated twice with toluene (40 mL) to
obtain the cesium salt of 13. The cesium salt was suspended in
DMF (80 mL) and allyl bromide (17.4 mL, 206 mmol, 20 equiv.)
was added slowly at room temperature. The reaction mixture was
stirred overnight at room temperature, the solvent was evaporated,
and the residue was coevaporated twice with toluene (40 mL) to
remove remaining DMF. The crude material was then purified by
flash chromatography (10% ethyl acetate in cyclohexane) to afford
14 (4.75 g, 9.7 mmol, 95%) as a pale yellow oil. TLC (5% ethyl
acetate in cyclohexane): Rf = 0.13. [α]2D0 = –20.1 (c = 1.21, CHCl3).
(S)-9-(Allyloxy)-8-[bis(tert-butoxycarbonyl)amino]-9-oxonon-4-enoic
Acid (18): 3-(Carboxypropyl)triphenylphosphonium bromide
(1.92 g, 4.5 mmol, 1.5 equiv.) was suspended in anhydrous THF
(20 mL) under an argon atmosphere. A solution of lithium bis(tri-
methylsilyl) amide (1 m in THF, 9.7 mL, 9.7 mmol, 3.25 equiv.) was
added within 40 min to the suspension at room temperature, which
was stirred for additional 15 min at room temperature. The re-
sulting dark red solution was cooled to 0 °C and 16 (1.11 g,
1H NMR (400 MHz, CDCl3): δ = 5.81–5.93 (m, 1 H), 5.28–5.32 3.0 mmol) dissolved in THF (20 mL) was added over 40 min. The
(dd, J = 17.20, 1.55 Hz, 1 H), 5.18–5.21 (dd, J = 10.50, 1.45 Hz, 1
H), 4.86–4.88 (m, 1 H), 4.58–4.60 (m, 2 H), 3.60–3.64 (dt, J = 6.40,
1.80 Hz, 2 H), 2.12–2.21 (m, 1 H), 1.89–1.96 (m, 1 H), 1.54–1.60
reaction mixture was warmed slowly to room temperature, and af-
ter 4 h, a TLC check indicated the completion of the reaction. 2 m
HCl was added to the reaction mixture until an acidic pH was
(m, 2 H), 1.48 (s, 9 H), 0.87 (s, 9 H), 0.03 (s, 6 H) ppm. 13C NMR obtained. The aqueous phase was extracted with ethyl acetate
(100 MHz, CDCl3): δ = 170.5, 152.1, 131.9, 117.9, 82.9, 65.5, 62.6,
58.1, 29.6, 27.9, 26.2, 25.9, 18.3, –5.3 ppm. LC–MS: tR
14.37 min, m/z (%) = 510 (100) [M + Na]+, 410 (80) [M – Boc +
(3ϫ40 mL), and the combined extracts were dried with Na2SO4.
After evaporation of the solvents to dryness, the crude mixture was
purified by flash column chromatography (50% ethyl acetate in cy-
=
Na]+. HRMS (EI): calcd. for C24H45O7NNaSi+ [M + Na]+ clohexane) to afford 18 (824 mg, 1.9 mmol, 64%) in a 9:1 Z/E ratio
510.2858; found 510.2860.
as a yellow oil. TLC (50% ethyl acetate in cyclohexane): Rf = 0.35.
1H NMR (500 MHz, CDCl3): δ = 5.83–5.91 (m, 1 H), 5.43–5.44 (t,
J = 5.20 Hz 1 H), 5.37–5.42 (dt, J = 11.10, 5.90 Hz, 1 H), 5.27–
5.31 (d, J = 16.20 Hz, 1 H), 5.17–5.19 (d, J = 10.50 Hz, 1 H), 4.83–
4.86 (m, 1 H), 4.57–4.58 (d, J = 5.35 Hz, 2 H), 2.28–2.40 (m, 4 H),
2.11–2.19 (m, 2 H), 2.02–2.06 (m, 1 H), 1.89–1.95 (m, 1 H), 1.47
(s, 18 H) ppm. 13C NMR (126 MHz, CDCl3): δ = 178.9, 170.5,
152.1, 131.8, 130.3, 130.0, 128.9, 128.4, 117.9, 83.0, 65.6, 57.8, 33.9,
29.7, 29.5, 29.1, 27.9 ppm. LC–MS: tR = 10.73 min (C18), m/z (%)
= 486 (45) [M + 2Na]+, 464 (45) [M + Na]+, 242 (100) [M –
2Boc]+. HRMS (EI): calcd. for C22H36O8N+ [M + H]+ 442.2435;
found 442.2436.
(S)-Allyl-2-[bis(tert-butoxycarbonyl)amino]-5-hydroxypentanoate
(15): Compound 14 (4.63 g, 9.5 mmol) was dissolved in a solution
of 5% water in acetone (100 mL) at room temperature. Copper(II)
chloride dihydrate (81 mg, 0.5 mmol, 0.05 equiv.) was added, and
the reaction mixture was heated to reflux. A TLC check after 1 h
of refluxing indicated the complete conversion of the starting mate-
rial. The solvent was evaporated to dryness, and the crude product
was purified by flash column chromatography (33% ethyl acetate
in cyclohexane) to afford 15 (3.27 g, 8.7 mmol, 92%) as a colorless
oil. TLC (33% ethyl acetate in cyclohexane): Rf = 0.26. [α]2D0
=
1
–30.2 (c = 1.01, CHCl3). H NMR (400 MHz, CDCl3): δ = 5.82–
5.91 (m, 1 H), 5.26–5.31 (dd, J = 17.20, 1.50 Hz, 1 H), 5.17–5.20 Inhibition assays: The biochemical inhibition assays have been per-
(dd, J = 10.50, 1.40 Hz, 1 H), 4.85–4.89 (m, 1 H), 4.57–4.59 (m, 2
H), 3.62–3.65 (t, J = 6.40 Hz, 2 H), 2.17–2.26 (m, 1 H), 1.87–1.97
(m, 1 H), 1.57–1.64 (m, 2 H), 1.46 (s, 18 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 170.5, 152.1, 131.7, 118.0, 83.1, 65.6, 62.1,
57.9, 29.3, 27.9, 26.1 ppm. LC–MS: tR = 9.90 min (C18), m/z (%) =
396 (100) [M + Na]+, 295 (80) [M – Boc + Na]+, 173 (70) [M –
2Boc]+. HRMS (EI): calcd. for C18H31O7NNa+ [M + Na]+
396.1990; found 396.1993.
formed following the protocols reported in ref.[20]
Supporting Information (see footnote on the first page of this arti-
cle): General information on the synthesis conditions and experi-
mental protocols for the solution-phase synthesis of compounds 1–
2, 5–9, and 17; HPLC chromatogram of the crude resin cleavage
product of Symplocamide A.
Acknowledgment
(S)-Allyl-2-[bis(tert-butoxycarbonyl)amino]-5-oxopentanoate (16):
Compound 15 (1.60 g, 4.3 mmol) was dissolved in anhydrous DCM
under an argon atmosphere. Dess–Martin periodinane (1.91 g, The research that led to these results has received funding from the
4.5 mmol, 1.05 equiv.) was added at room temperature, and the re-
action mixture was stirred at room temperature for 1.5 h, when a
TLC check indicated the complete conversion of the starting mate-
European Research Council under the European Union’s Seventh
Framework Programme (FP7/2007-2013/ERC grant agreement no.
258413) (M. K.).
1624
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1616–1625

Solid-Phase Approach to Ahp-Containing Cyclodepsipeptides
[9] S. C. Stolze, M. Meltzer, M. Ehrmann, M. Kaiser, Chem. Com-
mun. 2010, 46, 8857–8859.
[10] a) T. E. Nielsen, S. T. Le Quement, M. Meldal, Org. Lett. 2007,
9, 2469–2472; b) T. E. Nielsen, M. Meldal, Org. Lett. 2005, 7,
2695–2698; c) S. T. Le Quement, T. E. Nielsen, M. Meldal, J.
Comb. Chem. 2007, 9, 1060–1072.
[1] M. Namikoshi, K. L. Rinehart, J. Ind. Microbiol. Biotechnol.
1996, 17, 373–384.
[2] a) J. F. Blom, B. Bister, D. Bischoff, G. Nicholson, G. Jung,
R. D. Süssmuth, F. Jüttner, J. Nat. Prod. 2003, 66, 431–434; b)
U. Matern, L. Oberer, R. A. Falchetto, M. Erhard, W. A.
König, M. Herdman, J. Weckesser, Phytochemistry 2001, 58,
1087–1095; c) S. Mundt, E. N. Zainuddin, R. Mentel, V. Wray,
R. Jansen, M. Nimtz, M. Lalk, J. Nat. Prod. 2007, 70, 1084–
1088; d) K. Taori, S. Matthew, J. R. Rocca, V. J. Paul, H.
Luesch, J. Nat. Prod. 2007, 70, 1593–1600; e) E. von Elert, L.
Oberer, P. Merkel, T. Huhn, J. F. Blom, J. Nat. Prod. 2005, 68,
1324–1327; f) E. Zafrir, S. Carmeli, J. Nat. Prod. 2009, 73, 352–
358; g) R. G. Linington, D. J. Edwards, C. F. Shuman, K. L.
McPhail, T. Matainaho, W. H. Gerwick, J. Nat. Prod. 2007, 71,
22–27; h) G. Radau, Curr. Enz. Inhib. 2005, 1, 295–307.
[3] A. Y. Lee, T. A. Smitka, R. Bonjouklian, J. Clardy, Chem. Biol.
1994, 1, 113–117.
[11]
[12]
J. P. Zhu, Y. X. Jia, J. Org. Chem. 2006, 71, 7826–7834.
a) V. S. Martin, G. Kokotos, J. M. Padron, T. Martin, W. A.
Gibbons, J. Org. Chem. 1998, 63, 3741–3744; b) J. M. Padrón,
G. Kokotos, T. Martín, T. Markidis, W. A. Gibbons, V. c. S.
Martín, Tetrahedron: Asymmetry 1998, 9, 3381–3394.
E. A. Englund, H. N. Gopi, D. H. Appella, Org. Lett. 2004, 6,
859–859.
[13]
[14]
R. S. Coleman, J. A. Shah, Synthesis 1999, 1399,1400.
[15] H. Kunz, H. Waldmann, C. Unverzagt, Int. J. Pept. Prot. Res.
1985, 26, 493–497.
[16] H. Sajiki, T. Ikawa, K. Hattori, K. Hirota, Chem. Commun.
2003, 654–655.
[4] W. Bode, R. Huber, Eur. J. Biochem. 1992, 204, 433–451.
[5] a) U. Matern, C. Schleberger, S. Jelakovic, J. Weckesser, G. E.
Schulz, Chem. Biol. 2003, 10, 997–1001; b) M. A. McDonough,
C. J. Schofield, Chem. Biol. 2003, 10, 898–900.
[17] G. D. Kishore Kumar, S. Baskaran, J. Org. Chem. 2005, 70,
4520–4523.
[18] Z. P. Tan, L. Wang, J. B. Wang, Chin. Chem. Lett. 2000, 11,
753–756.
[6] A. B. Brauer, J. D. McBride, G. Kelly, S. J. Matthews, R. J. Lea-
therbarrow, Bioorg. Med. Chem. 2007, 15, 4618–4628.
[7] a) E. Zakharova, M. P. Horvath, D. P. Goldenberg, Proc. Natl.
Acad. Sci. USA 2009, 106, 11034–11039; b) Y. Zhou, Y. Zhang,
Chem. Commun. 2011, 47, 1577–1579.
[8] a) F. Yokokawa, A. Inaizumi, T. Shioiri, Tetrahedron Lett.
2001, 42, 5903–5908; b) F. Yokokawa, A. Inaizumi, T. Shioiri,
Tetrahedron 2005, 61, 1459–1480; c) F. Yokokawa, T. Shioiri,
Tetrahedron Lett. 2002, 43, 8673–8677.
[19] E. Vaz, L. Brunsveld, Org. Lett. 2006, 8, 4199–4202.
[20] a) M. Meltzer, S. Hasenbein, P. Hauske, N. Kucz, M.
Merdanovic, S. Grau, A. Beil, D. Jones, T. Krojer, T. Clausen,
M. Ehrmann, M. Kaiser, Angew. Chem. 2008, 120, 1352; An-
gew. Chem. Int. Ed. 2008, 47, 1332–1334; b) P. Hauske, M.
Meltzer, C. Ottmann, T. Krojer, T. Clausen, M. Ehrmann, M.
Kaiser, Bioorg. Med. Chem. 2009, 17, 2920–2924.
Received: December 6, 2011
Published Online: January 23, 2012
Eur. J. Org. Chem. 2012, 1616–1625
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1625