Solid-phase synthesis of aza-kahalalide f analogues: (2/r, 3r)-2-amino-3-azidobutanoic acid as precursor of the aza-threonine

Article abstract of DOI:10.1002/ejoc.200901345

The solid-phase synthesis of six novel analogues of Kahalal-ide F (KF), a natural product currently undergoing Phase II clinical trials, is reported. In all these compounds, amides were used as isosteres for the depsi bond. For two of these compounds, we performed an efficient synthesis of N-Fmoc-protected (2R, 3R)-2-amino-3-azidobutanoic acid, precursor of the aza-threonine. This is the first example of the solid-phase reduction of an azide group in the preparation of aza-Thr-containing peptides in the solid phase.

Full text of DOI:10.1002/ejoc.200901345

FULL PAPER
DOI: 10.1002/ejoc.200901345
Solid-Phase Synthesis of Aza-Kahalalide F Analogues:
(2R,3R)-2-Amino-3-azidobutanoic Acid as Precursor of the Aza-Threonine
Irene Izzo,*[a,b] Gerardo A. Acosta,[b,c] Judit Tulla-Puche,[b,c] Tommaso Cupido,[b,c]
Maria Jesus Martin-Lopez,[d] Carmen Cuevas,[d] and Fernando Albericio*[b,c,e]
Keywords: Peptides / Depsipeptides / Solid-phase synthesis / Natural products / Antitumor agents / Azides
The solid-phase synthesis of six novel analogues of Kahalal- protected (2R,3R)-2-amino-3-azidobutanoic acid, precursor
ide F (KF), a natural product currently undergoing Phase II
clinical trials, is reported. In all these compounds, amides
were used as isosteres for the depsi bond. For two of these
compounds, we performed an efficient synthesis of N-Fmoc-
of the aza-threonine. This is the first example of the solid-
phase reduction of an azide group in the preparation of aza-
Thr-containing peptides in the solid phase.
served.[6] However, the mechanism of cell death induction
remains to be elucidated and KF is currently in phase II
clinical trials.
Introduction
Kahalalide F (KF, 1, Figure 1) is a cyclodepsipeptide iso-
lated from the marine mollusc Elysia rufescens and its algae
diet Bryopsis pennata.[1,2] Preclinical studies have shown
that KF has potent cytotoxic activity in vitro against cell
lines from solid tumors, several with strong multidrug resis-
tance, including prostate, breast, and colon carcinomas,
neuroblastoma, and osteosarcoma.[3] In particular, oncosis
in human prostate and breast cancer cells has been ob-
served.[4] In contrast, nontumoral cell lines are five to forty
times less sensitive to KF.[4] Phase I clinical trials in adult
patients with advanced solid tumors have identified the re-
commended dose for phase II clinical studies, which is lim-
ited by a transient elevation of aminotransferases.[5] KF is
a COMPARE (National Cancer Institute)-negative com-
pound: among antitumor drugs it shows a unique mecha-
nism of action. This mechanism is mostly unknown but
may be related to the hydrophobic nature of the compound.
Lysosomes are the intracellular targets of KF and, in fact,
in HeLa cervical cancer cells and monkey COS-1 fibro-
blasts changes in lysosomal membranes have been ob-
Figure 1. Structure of Kahalalide F.
The head-to-side-chain cyclodepsipeptide structure of
KF is a complex structure with 13 amino acids, including a
rare (Z)-didehydro-(R)-aminobutyric acid (ZDhb), and an
aliphatic acid [5-methylhexanoic acid (5MeHex)] at the N-
terminus.[1,2] The cycle is formed between the carboxylic
acid of -Val(1) and the hydroxy group of -allo-Thr(6).
Stepwise solid-phase synthesis[7] and a convergent method-
ology[8] for the preparation of KF have been reported. Re-
cently, structure–activity relationship studies on 132 novel
KF analogues revealed that the biological activity of KF is
highly sensitive to backbone stereotopic modification but
not to modification of the length of the side-chain.[9]
In a recent study by our laboratory addressing the search
for new analogues of the natural marine product thiocoral-
ine,[10] a two-fold symmetric potent antitumor antibiotic
isolated from Micromonospora sp. L-13-ACM2-092, the in-
troduction of N-Me-amides, as synthetic surrogate for the
depsi moiety, afforded derivatives with comparable activi-
[a] Department of Chemistry, University of Salerno,
via Ponte Don Melillo, 84084 Fisciano, Salerno, Italy
Fax: +39-089 969603,
E-mail: iizzo@unisa.it
[b] Institute for Research in Biomedicine, Barcelona Science Park,
Baldiri Reixac 10, 08028 Barcelona, Spain
Fax: +34-934037126,
E-mail: albericio@ irbbarcelona.org
[c] CIBER-BBN, Networking Centre on Bioengineering, Biomate-
rials and Nanomedicine,
Barcelona Science Park, 08028 Barcelona, Spain
[d] Pharma Mar, S.A,
28770 Colmenar Viejo, Madrid, Spain
[e] Department of Organic Chemistry, University of Barcelona
Marti i Franques 11, 08028 Barcelona, Spain
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200901345.
2536
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2536–2543

Solid-Phase Synthesis of Aza-Kahalalide F Analogues
ties and increased serum stabilities. The ester/carboxamide mational rigidity.[11] All the analogues contained (4S)-
substitution was thus considered for other depsipeptides. MeHex instead of 5-MeHex because in previous studies its
Therefore, with the aim of finding new KF analogues with presence had been shown to enhance the efficacy of KF
improved biological activity and to evaluate the effects of analogues against breast and prostate xenografts.[9,12] First,
amide and N-Me-amide bonds as isosteres of the depsi the analogue containing (2S)-3-N-methyl-2,3-diaminopr-
bond, herein we report the solid-phase synthesis of five opanoic acid (3NMe-Dap-OH in 2) in place of -allo-Thr-
novel aza analogues of KF (2–6, Figure 2). The introduc- OH was prepared to validate the synthetic strategy by using
tion of N-Me-amides allows the hydrogen-bonding map of the most economic  derivative. Then, analogues incorpo-
the molecule to be preserved. Moreover, N-alkyl amino ac- rating (2R)-3-N-methyl-2,3-diaminopropanoic acid (3NMe-
ids are quite common in bioactive natural peptides and have -Dap-OH in 3), (2R)-2,3-diaminopropanoic acid (-Dap-
a considerable influence on pharmacological parameters: OH in 4), (2R,3R)-3-N-methyl-2,3-diaminobutanoic acid
membrane permeability, proteolitic stability, and confor- (3NMe--Dab-OH in 5), and (2R,3R)-2,3-diaminobutanoic
acid (-Dab-OH in 6) were prepared. A key feature of the
synthetic process for the preparation of analogues 5 and 6
was the use of Fmoc-protected (2R,3R)-2-amino-3-azido-
butanoic acid (7, Figure 2) as precursor of the aza-threo-
nine. Herein we also report an alternative efficient synthesis
for the known compound 7.[13]
Results and Discussion
The analogues 2–6 were synthesized following a strategy
similar to that used for the first solid-phase synthesis of
KF,[7] with modifications for the introduction of the di-
amino acids and their further N-methylation.
For the analogues 2–4 (Scheme 1), the approach involved
the incorporation of Fmoc-Val-OH followed by the presyn-
thesized dipeptide Alloc-Phe-(Z)-Dhb-OH on the amino
group of the aza-serine after construction of the linear pep-
tide sequence on a 2-chlorotrityl chloride resin (ClTrt-Cl
resin)[14] using the Fmoc/tBu strategy. When required, the
N-methyl group was introduced onto the o-NBS-protected
Figure 2. Novel analogues of Kahalalide F (2–6) and (2R,3R)-2-
amino-3-azidobutanoic acid (7).
Scheme 1. Synthetic strategies for the solid-phase synthesis of KF analogues 2–4. Reagents and conditions: (a) Fmoc--Val-OH, DIEA,
DCM; (b) piperidine/DMF (1:4); (c) Fmoc-aa-OH or (4S)MeHex acid, HATU, DIEA, DMF; (d) [Pd(PPh3)4], PhSiH3, DCM; (e) o-NBS-
Cl, DIEA, DCM; (f) PPh3, MeOH, DIAD, THF; (g) HOCH2CH2SH, DBU, DMF; (h) Alloc-Phe-(Z)Dhb-OH, HATU, DIEA, DMF;
(i) TFA/DCM (1:99); (l) DIPCDI, HOBt, DIEA, DCM; (m) TFA/H2O (19:1). Alloc, allyloxycarbonyl; Boc, tert-butoxycarbonyl; Dap,
diaminopropionic acid; Dhb, didehydro-(R)-aminobutyric acid; DIEA, N,N-diisopropylethylamine; DIPCDI, N,NЈ-diisopropylcarbodiim-
ide; Fmoc, 9-fluorenylmethoxycarbonyl; HATU, 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium hexafluorophos-
phate 3-oxide; HOBt: 1-hydroxybenzotriazole; o-NBS, o-nosyl; TFA, trifluoroacetic acid.
Eur. J. Org. Chem. 2010, 2536–2543
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2537

I. Izzo, F. Albericio et al.
FULL PAPER
Figure 3. Structure of dimer 15.
amine under Mitsunobu conditions.[11] The peptide was
elongated by using HATU and DIPEA as coupling rea-
gents. In all cases, single coupling with 4 equiv. of the
Fmoc-amino acid and HATU and 8 equiv. of DIEA in
DMF gave negative ninhydrin or chloranil tests.
The cyclization and final deprotection reactions were
performed in solution: The use of DIPCDI/HOBt/DIEA
prevented epimerization of -Val(4) in the troublesome lac-
tamization.[9]
This approach afforded the crude cyclopeptides 2 and 4
in good yields and purity (74 and 89%, respectively).[15]
These compounds were then purified by using a semipre-
parative column to afford the pure compounds.[15] Surpris-
ingly, in the case of the analogue 3,[16] whereas the linear
precursor 12 (Scheme 1) was prepared with a purity of
Scheme 2. Synthesis of Fmoc-protected (2R,3R)-2-amino-3-azido-
butanoic acid (7). Reagents and conditions: (a) BOP-Cl, tert-butyl
carbazate, DIEA; (b) MsCl, TEA, CH2Cl2, 65% for two steps; (c)
NaN3, DMF, 65 °C, 64%; (d) 6  HCl, 110 °C; (e) Fmoc-OSu,
NaHCO3, dioxane/H2O (1:1), 30% for two steps. BOP-Cl, bis(2-
oxo-3-oxazolidinyl)phosphinic chloride.
Ͼ98%,[15] in the last steps of the synthesis, the cyclization groups and reaction with Fmoc-OSu afforded the protected
and deprotection of the side-chain protecting groups, a azide in five steps and 12% overall yield (the strategy was
crude product containing a mixture of two compounds (47 not optimized).
and 32%, respectively) was obtained.[15] These compounds
The linear peptide 19 (Scheme 3) was prepared with a
were separated by using a semipreparative column to afford purity of 92%. In this case elongation of the peptide was
two pure compounds. By MALDI-TOF MS and 1H NMR, performed by using DIPCDI/HOAt as the coupling reagent
TOCSY, and COSY analysis, the two compounds were in the presence of the azide group. In fact, partial β-elimi-
identified as compound 3 and the dimer 15 (Figure 3).
nation of the azide group, which gave a dehydro amino acid,
For the preparation of the analogues 5 and 6, the -aza- was observed in the presence of HATU and DIPEA and
allo-threonine was required. α,β-Diamino acids are com- milder conditions for the coupling reactions were required.
mon in natural products, as free amino acids, in peptides In general, in all cases, single coupling with 3 equiv. of
and β-lactam antibiotics. They can be prepared from β-hy- Fmoc-amino acid, DIPCDI, and HOAt in DMF gave nega-
droxy amino acids by azide displacement of the hydroxy tive ninhydrin or chloranil tests. It is remarkable that only
group followed by azide reduction.[17] This is a particularly 2 equiv. of the azido acid 7 were necessary for its coupling
attractive methodology for the synthesis of amines because reaction.
both the “chiral pool” and contemporary asymmetric meth-
Once the linear precursor had been obtained, the feasi-
odologies have made available a vast array of potential en- bility of reducing the azide group on the ClTrt-Cl resin was
antiomerically enriched azide substrates.[18] In this work, evaluated. In a first attempt, the Staudinger reaction[20] was
the Fmoc-protected (2R,3R)-2-amino-3-azidobutanoic acid considered the most attractive method as a result of its
(7), prepared in solution, was introduced into the solid- mildness. In this reaction, triphenyl- or trimethylphosphane
phase and after sequence elongation was completed and be- in THF was used to convert the azides into iminophospor-
fore cleavage from the resin the azide group was reduced on anes, which were then easily hydrolyzed to afford the amine.
the solid phase. This strategy allowed us to avoid a formal Some examples of azide reduction on solid phases[21] by
protection of the amino group. Compound 7 was synthe- using the Staudinger reaction have been reported in the lit-
sized as shown in Scheme 2 and studies to reduce the azide erature. However, to the best of our knowledge, this is the
group, incorporated into the linear sequence on the solid first time that this reaction has been used for the prepara-
phase, were realized.[19]
tion of aza-Thr-containing peptides on a solid phase.
The first step, namely the synthesis of iminophosphorane
Boc--threonine (16), which has two stereogenic centers
and is commercially available, was used as the starting ma- 20, proved to be quite slow and even by using a higher
terial. The well-known stereospecific sequence, mesylation/ concentration of triphenylphosphane (0.3  in THF), the
azide nucleophilic substitution, allowed the introduction of reaction was almost complete only after 48 h.[15] Because
the desired configuration and, at the same time, the amino- long reaction times were considered to be impractical, the
masked group at C-3. Full deprotection of the functional reaction was performed by using a 0.5  solution of
2538
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2536–2543

Solid-Phase Synthesis of Aza-Kahalalide F Analogues
Scheme 3. Solid-phase reduction of the azide group for the synthesis of KF analogues 5 and 6. Reagents and conditions: (a) piperidine/
DMF (2:8); (b) Fmoc-aa-OH or (4S)MeHex acid, DIPCDI, HOAt, DMF; (c) PPh3, THF; (d) THF, H2O, 75 °C; (e) SnCl2, PhSH, DIEA,
THF.
P(CH3)3 in THF. Under these conditions the synthesis of gents have been used for the reduction of azides, in terms
iminophosphorane is generally quicker.[21] In this case, how- of their practical applicability, most of these methods have
ever, the reaction proved to be equally slow and a complex certain disadvantages, especially for a solid support.[18]
reaction mixture was observed after 12 h. Furthermore, the
Bartra and Vilarrasa and co-workers reported that tin(II)
hydrolysis of 20 proved to be unexpectedly very difficult. complexes, prepared by treatment of SnCl2 or Sn(SR)2 with
Several attempts were made to hydrolyze 20 by using dif- appropriate amounts of RSH and tertiary amines, are ef-
ferent ratios of THF/H2O (4:1, 10:1, 1:4, 0:1), basic condi- ficient reducing agents for the transformation of azides to
tions (pyridine or NH4OH),[22] and microwave irradia- amines.[27] Some examples have been reported[28] in which
tion.[23] In all cases, no amine 21 was obtained for short these conditions have been applied to azide reduction in the
reaction times whereas for longer reaction times decomposi- solid phase. However, to the best of our knowledge, they
tion or the formation of a complex mixture was observed. have not been used for peptide synthesis on the less robust
Recently, some researchers reported the slow hydrolysis of linker of the chlorotrityl resin. To obtain the amine 21
iminophosphoranes, which was addressed by the addition (Scheme 3), we added a freshly prepared solution of 0.2 
of acid.[24] However, this option was not compatible with SnCl2, 0.8  PhSH, and 1.0  DIEA in THF to a suspen-
our solid-phase system.
sion of the peptidyl resin in THF. After 1 h the reaction
The steric hindrance and low accessibility of the high was not complete and after four treatments (4ϫ3 h) the
lipophilic peptide chain could explain the troubles with this reaction proceeded no further and 66% of amine 21 and
step, as has also been pointed out by Vaultier et al., who 34% of the starting material were detected by HPLC analy-
reported different reactivities of iminophosphoranes de- sis.[15] With regard to the difficulty of this step, the moder-
pending on the character of the carbon (primary, second- ate yield was considered a good compromise and given that
ary, or tertiary) bonded to the functional group.[25] In a the reaction is relatively clean, the remaining steps for the
totally different context, Diederichsen and co-workers very synthesis of the KF analogues 5 and 6, shown in Scheme 4,
recently reported that undesired azide reduction in a pep- were performed directly on the mixture. The methyl group,
tide depends on the position of the azide-containing moiety when required, was introduced into the o-nosyl-protected
in the peptide.[26] Moreover, even though a variety of rea- amine under Mitsunobu conditions but a longer reaction
Scheme 4. Solid-phase synthesis of KF analogues 5 and 6. Reagents and conditions: (a) o-NBS-Cl, DIEA, DCM; (b) PPh3, MeOH,
DIAD, THF; (c) HOCH2CH2SH, DBU, DMF; (d) Fmoc-aa-OH or Alloc-Phe-(Z)Dhb-OH, HATU, DIEA, DMF; (e) piperidine/DMF
(2:8); (f) [Pd(PPh3)4], PhSiH3, DCM; (g) TFA/DCM (1:99); (h) DIPCDI, HOBt, DIEA, DCM; (i) TFA/H2O (19:1).
Eur. J. Org. Chem. 2010, 2536–2543
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2539

I. Izzo, F. Albericio et al.
FULL PAPER
the Fmoc group and washings were carried out as described above.
The Alloc group was removed with [Pd(PPh3)4] (18 mg,
0.015 mmol, 0.1 equiv.) in the presence of PhSiH3 (190 µL,
1.5 mmol, 10 equiv.) under Ar followed by extensive washing with
CH2Cl2 (3ϫ3 mL), 0.02  sodium diethyldithiodicarbamate in
DMF (3ϫ3 mL), DMF (3ϫ3 mL), and CH2Cl2 (3ϫ3 mL). An
aliquot of the peptidyl resin was treated with 1% TFA in CH2Cl2
and HPLC analysis of the crude 9 obtained after evaporation
showed a purity of 94%. HPLC: tR = 5.6 min, CH3CN (+ 0.036%
TFA) in H2O (+ 0.045% TFA): from 50 to 100; column: SunFireTM
C18, 3.5 µm, 4.6ϫ100 mm; detection 220 nm. MS (ES): calcd. for
C65H118N12O15 1307.7; found 1308.9 [M + H]+. At this point the
resin was divided into two equal portions.
time and more treatments were necessary to complete each
step. The products were purified by using a semipreparative
column to afford the two pure compounds.[15] This ap-
proach afforded the crude cyclopeptides 5[16] and 6 in satis-
factory yields (23 and 59%, respectively) and purity.
Conclusions
Herein we have described the synthesis of six novel KF
analogues of Kahalalide F (2–6 and 15) containing amide
and N-Me-amide bonds in place of the depsi bond. Elong-
ation of the peptide sequence and N-methylation, when re-
quired, were carried out in the solid phase. The aza-Thr
derivatives, which are the most challenging derivatives from
a synthetic point of view, were prepared by using (2R,3R)-
2-amino-3-azidobutanoic acid as the aza-Thr precursor. Re-
duction of the azide to the amino function was performed
in the solid phase. Although the scope of this reduction can
Synthesis of 3: A solution of o-NBS-Cl (51 mg, 0.23 mmol) and
2,4,6-collidine (0.050 µL, 0.38 mmol) in CH2Cl2 was added to the
resin and the mixture was stirred for 60 min. This reaction was
repeated under the same conditions three times. After filtration and
washing with CH2Cl2 (1ϫ0.5 min), DMF (3ϫ0.5 min), and
CH2Cl2 (3ϫ0.5 min), a solution of PPh3 (101 mg, 0.38 mmol) and
MeOH (31 µL, 0.77 mmol) in THF was added to the resin and
be sequence-dependent, this method may be useful for the stirred for 1 min and then a solution of DIAD (76 µL, 0.38 mmol)
in THF was added to the peptidyl resin. After stirring the resin for
20 min, it was washed with CH2Cl2 (1ϫ0.5 min), DMF
(2ϫ0.5 min), and CH2Cl2 (3ϫ0.5 min). This reaction was repeated
under the same conditions three times. After treatment (2ϫ15 min)
with DBU (115 µL, 0.77 mmol) and 2-mercaptoethanol (108 µL,
1.54 mmol) in DMF (1.5 mL), the resin was washed with DMF
(3ϫ0.5 min), CH2Cl2 (3ϫ0.5 min), and DMF (3ϫ0.5 min).
Fmoc-Val-OH (130 mg, 0.77 mmol, 5 equiv.) and Alloc-Phe-(Z)-
Dhb-OH (205 mg, 0.62 mmol, 4 equiv.) were added sequentially
using HATU (117 mg, for 0.31 mmol and 4 equiv.; 146 mg, for
0.38 mmol and 5 equiv.) and DIEA (105 µL, for 0.62 mmol and
8 equiv.; 130 µL, for 0.77 mmol and 10 equiv.) in DMF (1.0 mL).
After 90 min of coupling, the chloranil and ninhydrin tests were
negative. Removal of the Fmoc and Alloc groups and washings
were carried out as reported before. The protected peptide was
cleaved from the resin by TFA/CH2Cl2 (1:99) (8ϫ2 s), the filtrate
was collected in H2O (70 mL) and H2O was partially removed un-
der reduced pressure. A light-yellow precipitate appeared in the re-
maining water. Acetonitrile was then added to dissolve this solid
preparation of other aza-Thr-containing peptides. Aza-
Kahalalide F showed stability in human serum for 72 h.
Biological tests on tumor cell lines are currently in progress
and will be reported in due course.
Experimental Section
General: Details of materials, methods, HPLC profiles and com-
plete NMR spectroscopic characterization of compounds 6 and 7
are reported in the Supporting Information.
Synthesis of 3 and 4
Synthesis of Resin 9: Cl-TrtCl resin (0.2 g, 1.54 mmol/g) was placed
in a 10 mL polypropylene syringe fitted with a polyethylene filter
disc. The resin was then washed with CH2Cl2 (5ϫ0.5 min) and a
solution of Fmoc--Val-OH (67 mg, 0.2 mmol) and DIEA
(0.050 mL, 0.31 mmol) in CH2Cl2 (1.0 mL) was added. The mixture
was then stirred for 10 min. Further DIEA was then added
(0.158 mL, 0.93 mmol) and the mixture was stirred for 50 min. The until the solution turned clear. The mixture was then lyophilized
reaction was completed by the addition of MeOH (0.250 mL) after
stirring for 10 min. The Fmoc--Val-O-TrtCl resin was subjected
to washings/treatments with CH2Cl2 (3ϫ0.5 min) and DMF
(3ϫ0.5 min) and was then incubated three times with 20% piperi-
dine/DMF (3 mL, v/v) on a shaker platform for 2 min and
2ϫ10 min followed by extensive washing with DMF (3ϫ3 mL)
and CH2Cl2 (3ϫ3 mL). The loading was 0.77 mmol/g, as calcu-
lated by Fmoc determination.
to give 125 mg (0.071 mmol) of the linear protected peptide 12 with
a purity of Ͼ98%, as shown by HPLC [tR = 6.9 min, CH3CN (+
0.036% TFA) in H2O (+ 0.045% TFA): from 40 to 100, at room
temperature; tR = 5.5 min, CH3CN (+ 0.036% TFA) in H2O (+
0.045% TFA): from 50 to 100 at 40 °C; column: SunFireTM C18,
3.5 µm, 4.6ϫ100 mm; detection 220 nm].
The protected peptide (125 mg, 0.071 mmol) was dissolved in
CH2Cl2 (71 mL, 1 m). HOBt (43 mg, 0.284 mmol, 4 equiv.) dis-
solved in the minimum volume of DMF, DIEA (28 µL,
0.163 mmol, 2.3 equiv.), and DIPCDI (44 µL, 0.284 mmol,
Fmoc--allo-Ile-OH (217 mg, 0.62 mmol, 4 equiv.), Fmoc--
Dap(Alloc)-OH (253 mg, 0.62 mmol, 4 equiv.), Fmoc--allo-Ile-
OH (217 mg, 0.62 mmol, 4 equiv.), Fmoc-Orn(Boc)-OH (280 mg, 4 equiv.) were progressively added. The mixture was stirred on a
0.62 mmol, 4 equiv.), Fmoc--Pro-OH (207 mg, 0.62 mmol, shaker platform overnight. Then the solvent was removed by evap-
4 equiv.), and Fmoc--Val-OH (209 mg, 0.62 mmol, 4 equiv.),
oration under reduced pressure. The protected cyclic peptide was
Fmoc-Val-OH (261 mg, 0.77 mmol, 5 equiv.), Fmoc-Thr(tBu)-OH dissolved in TFA/H2O (19:1; 11.4:0.6 mL) and the mixture was
(0.306 mg, 0.77 mmol, 5 equiv.), Fmoc--Val-OH (209 mg, stirred for 1 h. Most of the solvent was removed by evaporation
0.62 mmol, 4 equiv.), and (4S)-MeHex-OH (80 mg, 0.62 mmol,
4 equiv.) were added sequentially to the H--Val-O-TrtCl resin de-
scribed above using HATU (234 mg, for 0.62 mmol and 4 equiv.;
293 mg, for 0.77 mmol and 5 equiv.) and DIEA (210 µL, for
1.24 mmol and 8 equiv.; 261 µL, for 1.54 mmol and 10 equiv.) in
DMF (1.0 mL). In all the cases, after 90 min of coupling, the ninhy-
drin or chloranil test, when required, were negative. Removal of
under reduced pressure. The residue was then added dropwise to
cold diethyl ether. A white precipitate appeared that was centri-
fuged and washed with cold diethyl ether three times, dissolved in
CH3CN/H2O (1:1), and lyophilized to afford 114 mg of the crude
product. HPLC showed a mixture containing two main products
[tR
= 16.0 min (3, 47%) and 26.0 min (15, 32%); CH3CN
(+ 0.036% TFA) in H2O (+ 0.045% TFA): from 30 to 60; column:
2540
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2536–2543

Solid-Phase Synthesis of Aza-Kahalalide F Analogues
Symmetry C18, 5 µm, 7.8ϫ100 mm, detection at 220 nm]. 66 mg of Synthesis of 7: A 6  HCl solution (34 mL) was added to 18 (1.6 g,
the crude product was purified by HPLC under the same condi-
tions to afford 5.6 mg of pure 3 and 8 mg of pure 15.
4.5 mmol). The mixture was initially warmed to 65 °C and the
starting material slowly dissolved. The temperature was then in-
creased to 110 °C. The reaction was checked by HPLC and after
24 h was stopped, concentrating the reaction mixture under re-
duced pressure. This crude product was used in the next step with-
out further purification. HPLC: tR = 2.3 min, CH3CN (+ 0.036%
TFA) in H2O (+ 0.045% TFA): from 0 to 100; column: XBridgeTM
C18, 5 µm, 4.6ϫ150 mm; detection 220 nm]. MS (ES): calcd. for
C4H8N4O2 144.0; found 144.9 [M + H]+.
3: MS (MALDI-TOF): calcd. for C75H125N15O15 1475.95; found
1476.95 [M + H]+, 1499.02 [M + Na]+, 1515.0 [M + K]+.
15: MS (MALDI-TOF): calcd. for C150H250N30O30 2951.90; found
2952.91 [M + H]+, 2974.89 [M + Na]+, 2990.86 [M + K]+.
Synthesis of 4: Sequence elongation, cyclization, and deprotection
of the side-chains were realized as described. After deprotection,
no precipitation was observed. The crude (206 mg) was recovered
after lyophilization and analyzed by HPLC [tR = 4.5 min, CH3CN
(+ 0.036% TFA) in H2O (+ 0.045% TFA): from 30 to 100; column:
SunFireTM C18, 3.5 µm, 4.6ϫ100 mm; detection at 220 nm]. 20 mg
of the crude product was purified by HPLC [tR = 15.3 min CH3CN
(+ 0.036% TFA) in H2O (+ 0.045% TFA): from 30 to 60; column:
Symmetry C18, 5 µm, 7.8ϫ100 mm, detection at 220 nm] to afford
2.0 mg of pure 4.
Fmoc-OSu (1.95 g, 5.8 mmol) and NaHCO3 (2.4 g, 2.8 mmol) were
added to a solution of the crude product (4.5 mmol) in a mixture
of dioxane/H2O (1:1, 38 mL) at 0 °C. The mixture was allowed to
react at room temperature for 5 h under controlled pH (8–9). H2O
(30 mL) was added to the mixture, which was extracted twice with
EtOAc (80 mL each). Water was acidified to give pH 3–4 and ex-
tracted with EtOAc (3ϫ200 mL). The combined organic phases
were washed with H2O (2ϫ100 mL) and brine (100 mL), dried
(Na2SO4), and concentrated in vacuo to give 7 (0.477 g, 30% for
two steps) as a white amorphous solid. This product was quite pure,
as shown by HPLC analysis [tR = 12.2 min, CH3CN (plus 0.036%
TFA) in H2O (plus 0.045% TFA): from 0 to 100; column:
XBridgeTM C18, 5 µm, 4.6ϫ150 mm; detection 220 nm].
4: MS (MALDI-TOF): calcd. for C75H125N15O15 1475.95; found
1476.95 [M + H]+, 1499.02 [M + Na]+, 1515.0 [M + K]+.
Synthesis of Fmoc-Protected (2R,3R)-2-Amino-3-azidobutanoic
Acid 7
1
Synthesis of 17: tert-Butyl carbazate (1.8 g, 13.7 mmol), BOP-Cl
(4.6 g, 17.9 mmol), and DIEA (4.8 mL, 27.4 mmol) were added to
a solution of Boc--threonine (16; 3.0 g, 13.7 mmol) in dry CH2Cl2
(70 mL) at 0 °C over 10 min. The mixture was warmed to room
temperature and after 6 h was concentrated, dissolved in EtOAc,
and washed with a solution of 1  NaHCO3, 1  H2SO4, and a
saturated aqueous solution of NaCl. The combined organic phases
were concentrated and dried (MgSO4) to give the crude product
(4.50 g), which was used without further purification in the next
step. TEA (3.8 mL, 2.70 mmol) and MsCl (1.9 mL, 2.43 mmol)
were added to the crude product (4.5 g, 13.5 mmol) in dry CH2Cl2
(110 mL) at 0 °C over 10 min. The mixture was warmed to room
temperature and after 1 h a saturated aqueous solution of NaCl
was added and the mixture was extracted with CH2Cl2
(2ϫ100 mL). The combined organic phases were concentrated and
dried (MgSO4) to give the crude product, which was purified by
flash chromatography (Hex/EtOAc from 70:30 to 20:80) to give 17
as a white amorphous solid (3.63 g, 65% for two steps).
[α]D = –57.5 (c = 0.3, CHCl3). H NMR (400 MHz, CDCl3): δ =
7.76 (m, 2 H, ArH), 7.59 (m, 2 H, ArH), 7.40 (m, 2 H, ArH), 7.31
(m, 2 H, ArH), 5.60 (d, J = 8.3 Hz, 1 H, NH), 4.51 (dd, J = 8.0,
3.2 Hz, 1 H, 2-H), 4.43 (m, 2 H, CH2-Fmoc), 4.23 (t-like, J =
6.8 Hz, 1 H, CH-Fmoc), 3.95 (m, 1 H, 3-H), 1.42 (d, J = 6.6 Hz,
3 H, 4-H) ppm. 13C NMR (100 MHz, CDCl3): δ = 173.0, 155.9,
143.6, 143.5, 141.0 (ϫ2), 127.8 (ϫ2), 127.1 (ϫ2), 125.0 (ϫ2), 120.0
(ϫ2), 67.4, 58.6, 57.6, 47.1, 15.6 ppm. MS (ES): calcd. for
C19H18N4O4 366.37; found 367.2 [M + H]+.
Synthesis of 5 and 6
Synthesis of 19: Cl-TrtCl resin (0.300 g, 1.54 mmol/g) was placed
in a 10 mL polypropylene syringe fitted with a polyethylene filter
disc. The resin was then washed with CH2Cl2 (5ϫ0.5 min) and
a solution of Fmoc--Val-OH (100 mg, 0.269 mmol) and DIEA
(0.078 mL, 0.462 mmol) in CH2Cl2 (0.5 mL) was added. The mix-
ture was then stirred for 10 min. Further DIEA was then added
(0.235 mL, 1.39 mmol) and the mixture was stirred for 50 min.
MeOH (0.24 mL) was added and the reaction was completed after
stirring for 10 min. The Fmoc--Val-O-TrtCl resin was subjected to
the following washings/treatments with CH2Cl2 (3ϫ0.5 min) and
DMF (3ϫ0.5 min), and the resin was then incubated three times
with 20% piperidine/DMF (3 mL, v/v) on a shaker platform for
2 min and 2ϫ7 min followed by extensive washing with DMF
(3ϫ3 mL) and CH2Cl2 (3ϫ3 mL). The loading was 0.84 mmol/g,
as calculated by Fmoc determination. Fmoc--allo-Ile-OH
(268 mg, 0.76 mmol, 3 equiv.), 7 (185 mg, 0.50 mmol, 2 equiv.),
Fmoc--allo-Ile-OH (273 mg, 0.76 mmol, 3 equiv.), Fmoc-
Orn(Boc)-OH (344 mg, 0.76 mmol, 3 equiv.), Fmoc--Pro-OH
(256 mg, 0.76 mmol, 3 equiv.), and Fmoc--Val-OH (258 mg,
0.76 mmol, 3 equiv.), Fmoc-Val-OH (258, mg, 0.76 mmol, 3 equiv.),
Fmoc-Thr(tBu)-OH (302 mg, 0.77 mmol, 3 equiv.), Fmoc--Val-
OH (258 mg, 0.76 mmol, 3 equiv.), and (4S)-MeHex-OH (99 mg,
1
[α]D = –10.9 (c = 1.0, CHCl3). H NMR (400 MHz, CDCl3): δ =
8.11 (br. s, 1 H, NH), 6.39 (br. s, 1 H, NH), 5.47 (br. d, J = 8.6 Hz,
1 H, NH), 5.29 (m, 1 H, 3-H), 4.48 (m, 1 H, 2-H), 3.03 (br. s, 3 H,
CH3SO2), 1.52–1.44 [m, 21 H, overlapped, C(CH3)3 and 4-H] ppm.
13C NMR (100 MHz, CDCl3): δ = 168.9, 155.7, 149.7, 86.4, 82.2,
78.2, 57.0, 38.3, 28.0 (ϫ6), 18.0 ppm. MS (ES): calcd. for
C15H29N3O8S 411.2; found 412.2 [M + H]+.
Synthesis of 18: NaN3 (3.6 g, 55.4 mmol), was added to a solution
of 17 (5.7 g, 13.8 mmol) in dry DMF (180 mL) at room tempera-
ture. The mixture was heated at 65 °C for 24 h and after the ad-
dition of water (150 mL) was extracted with hexane/Et2O (1:1,
300ϫ3). The combined organic phases were washed with water
(3ϫ150 mL) and
(3ϫ150 mL) and then were dried (Na2SO4) and concentrated in
vacuo to give crude 18 (3.0 g, 64%) as a white amorphous solid. 0.76 mmol, 3 equiv.) were added sequentially to the H--Val-O-
a saturated aqueous solution of NaCl
This product was shown to be quite pure by HPLC analysis and
was used in the next step without further purification. HPLC: tR
TrtCl resin described above using DIPCDI (117 µL, for 0.76 mmol
and 3 equiv.; 78 µL, for 0.50 mmol and 2 equiv.), HOAt (103 mg,
= 10.9 min, CH3CN (+ 0.036% TFA) in H2O (+ 0.045% TFA): for 0.76 mmol and 3 equiv.; 69 mg, for 0.50 mmol and 2 equiv.) in
from 0 to 100; column: XBridgeTM C18, 5 µm, 4.6ϫ150 mm; detec-
DMF (1.2 mL). In all cases, after 90 min of coupling, the ninhydrin
tion 220 nm]. MS (ES): calcd. for C14H26N6O5 358.39; found 359.2 and chloranil tests were negative. Removal of the Fmoc group and
[M + H]+.
washings were carried out as described above. An aliquot of the
Eur. J. Org. Chem. 2010, 2536–2543
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2541

I. Izzo, F. Albericio et al.
FULL PAPER
peptidyl resin was treated with 1% TFA in CH2Cl2 and HPLC
analysis of the crude product obtained after evaporation showed a
protected peptide was cleaved from the resin by TFA/CH2Cl2 (1:99;
8ϫ2 min). The filtrate was collected in H2O (130 mL) and the H2O
purity of 91% [tR = 6.1 min CH3CN (+ 0.036% TFA) in H2O was partially removed under reduced pressure. A light-yellow pre-
(+ 0.045% TFA): from 80 to 100; column: SunFireTM C18, 3.5 µm,
4.6ϫ100 mm, detection 220 nm]. MS (ES): calcd. C66H118N14O15
1346.89; found 1347.5 [M + H]+.
cipitate appeared in the remaining water. Acetonitrile was then
added to dissolve this solid until the solution turned clear. Lyophili-
zation afforded 112 mg of the linear protected peptide 26 with a
purity of 57% as shown by HPLC [tR = 1.9 min, CH3CN (+
0.036% TFA) in H2O (+ 0.045% TFA): from 80 to 100; column:
SunFireTM C18, 3.5 µm, 4.6ϫ100 mm; detection 220 nm]. Cycliza-
tion and deprotection of the side-chains were realized as described
above. The protected cyclic peptide was dissolved in TFA/H2O
(19:1; 11.4:0.6 mL) and the mixture was stirred for 1.5 h. Most of
the solvent was removed by evaporation under reduced pressure
and then the residue was added dropwise to cold diethyl ether. A
white precipitate appeared which was centrifuged and washed with
cold diethyl ether three times, dissolved in CH3CN/H2O (1:1), and
lyophilized to afford 90 mg of crude product which had a purity of
Synthesis of 21. Bartra and Vilarrasa Reduction: A freshly prepared
solution containing 0.2  of SnCl2, 0.8  of PhSH, and 1  of
DIEA (6.3 mL) in dry THF was added to a suspension of the pepti-
dyl resin 19 in dry THF (180 µL). The suspension was stirred on a
shaker platform for 1 h. This treatment was repeated under the
same conditions three times. An aliquot of the peptidyl resin was
treated with 1% TFA in CH2Cl2 and HPLC analysis of the crude
product obtained after evaporation showed a purity of 66%. HPLC
[tR = 7.5 min CH3CN (+ 0.036% TFA) in H2O (+ 0.045% TFA):
from 5 to 100; column: SunFireTM C18, 3.5 µm, 4.6ϫ100 mm, de-
tection 220 nm]. MS (ES): calcd. for C66H120N12O15 1321.7; found
1322.9 [M + H]+.
59%, as shown by HPLC analysis [tR
= 5.0 min, CH3CN
(+ 0.036% TFA) in H2O (+ 0.045% TFA): from 30 to 100; column:
SunFireTM C18, 3.5 µm, 4.6ϫ100 mm; detection at 220 nm. 50 mg
of the crude product was purified by HPLC [tR = 5.0, CH3CN
(+ 0.036% TFA) in H2O (+ 0.045% TFA): from 30 to100; column:
Symmetry C18, 5 µm, 7.8ϫ100 mm, detection at 220 nm] to afford
7.9 mg of 6.
Synthesis of 5: Methylation, sequence elongation, and cleavage
from the resin were performed as described above with some excep-
tions on two-thirds of the peptidyl resin 21. This time nosylation
and methylation proved to be more difficult and the reactions were
repeated four and five times, respectively. The protected peptide
was cleaved from the resin by TFA/CH2Cl2 (1:99) (8ϫ2 min), the
filtrate was collected in H2O (135 mL), and the H2O was partially
removed under reduced pressure. A light-yellow precipitate ap-
peared in the remaining water. Acetonitrile was then added to dis-
solve this solid until the solution turned clear. The mixture was
then lyophilized to give a crude product (218 mg) containing the
title compound [25%, as shown by HPLC analysis: tR = 7.2 min,
CH3CN (+ 0.036% TFA) in H2O (+ 0.045% TFA): from 30 to 100
at 40 °C; column: SunFireTM C18, 3.5 µm, 4.6ϫ100 mm; detection
220 nm]. Cyclization and deprotection of the side-chains were real-
ized as described above. The protected cyclic peptide was dissolved
in TFA/H2O (19:1; 22.8:1.2 mL) and the mixture was stirred for
1.5 h. Most of the solvent was removed by evaporation under re-
duced pressure and then the residue was added dropwise to cold
diethyl ether. A white precipitate appeared and was centrifuged and
washed with cold diethyl ether three times, dissolved in CH3CN/
H2O (1:1), and lyophilized to afford 206 mg of the crude product,
which was then analyzed by HPLC [5: tR = 6.0 min, CH3CN
(+ 0.036% TFA) in H2O (+ 0.045% TFA): from 40 to 60; column:
SunFireTM C18, 3.5 µm, 4.6ϫ100 mm; detection at 220 nm]. 57 mg
of the crude product was purified by HPLC [tR = 6.0 CH3CN
(+ 0.036% TFA) in H2O (+ 0.045% TFA): from 45 to 60, column:
Symmetry C18, 5 µm, 7.8ϫ100 mm, detection at 220 nm] to afford
5.8 mg of 5.
6: MS (MALDI-TOF): calcd. for C75H125N15O15 1475.95; found
1499.10 [M + Na]+. A complete description of the chemical shifts
and NMR spectroscopic characterization for compound 6 are re-
ported in the Supporting Information.
Supporting Information (see also the footnote on the first page of
this article): General procedures, detailed experimental for the
azide reduction reaction, and characterization data (HPLC and
NMR).
Acknowledgments
Financial support by the Università degli Studi di Salerno and the
Provincia di Salerno (grant Progetto di mobilità internazionale,
Protocollo di Intesa dell’Università degli Studi di Salerno con la
Provincia di Salerno) to I. I. and by the Centro de Investigación
Científica y Tecnológica (CICYT) (CTQ2009-07758), the Genera-
litat de Catalunya (2009SGR 1024), the Institute for Research in
Biomedicine, and the Barcelona Science Park is gratefully acknowl-
edged.
[1] M. T. Hamann, P. J. Scheuer, J. Am. Chem. Soc. 1993, 115,
5825–5826.
[2] M. T. Hamann, C. S. Otto, P. J. Scheuer, J. Org. Chem. 1996,
5: MS (MALDI-TOF): calcd. for C76H127N15O15 1489.96; found
1513.02 [M + Na]+, 1529.0 [M + K]+. A complete description of
the chemical shifts and NMR spectroscopic characterization data
for compound 5 are reported in the Supporting Information.
61, 6594–6660.
[3] a) G. T. Faircloth, W. Grant, B. Smith, J. Supko, A. Brown, A.
Geldof, J. Jimeno, Proc. Am. Assoc. Cancer Res. 2000, 41, 600;
b) L. Shao, L. Weissbach, G. T. Faircloth, B. A. Chabner, F. J.
Hornicek, Proc. Am. Assoc. Cancer Res. 2001, 42, 203; c) G. T.
Faircloth, B. Smith, W. Grant, J. Jimeno, L. García-Grávalos,
K. Scotto, A. Shtil, Proc. Am. Assoc. Cancer Res. 2001, 42,
213.
[4] Y. Suárez, L. González, A. Cuadrado, M. Berciano, M.
Lafarga, A. Muñoz, Mol. Cancer Ther. 2003, 2, 863–872.
[5] B. Pardo, L. Paz-Ares, J. Tabernero, E. Ciruelos, M. Garcia,
R. Salazar, A. Lopez, M. Blanco, A. Nieto, J. Jimeno, M. A.
Izquierdo, J. M. Trigo, Clin. Cancer Res. 2008, 14,1116–1123.
[6] M. García-Rocha, P. Bonay, J. Avila, Cancer Lett. 1996, 99,
43–50.
Synthesis of 6: Sequence elongation was performed on one-third of
the peptidyl resin 21 as described before. Fmoc-Val-OH (114 mg,
0.34 mmol, 4 equiv.) and Alloc-Phe-(Z)DhbOH (112 mg,
0.34 mmol, 4 equiv.) were added sequentially using HATU
(129 mg, 0.34 mmol, 4 equiv.) and DIEA (114 µL, 0.67 mmol,
4 equiv.) in DMF (1.0 mL). After 90 min the ninhydrin test was
negative. The Alloc group was removed with [Pd(PPh3)4] (10 mg,
0.0084 mmol, 0.1 equiv.) in the presence of PhSiH3 (91 µL,
0.84 mmol, 10 equiv.) under Ar followed by extensive washing with
CH2Cl2 (3ϫ3 mL), 0.02  sodium diethyldithiodicarbamate in
DMF (3ϫ3 mL), DMF (3ϫ3 mL), and CH2Cl2 (3ϫ3 mL). The
[7] a) A. Lopez-Macia, J. C. Jimenez, M. Royo, E. Giralt, F. Alber-
icio, J. Am. Chem. Soc. 2001, 123, 11398–11401; b) P. E. Lopez,
2542
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 2536–2543

Solid-Phase Synthesis of Aza-Kahalalide F Analogues
A. Isidro-Llobet, C. Gracia, L. J. Cruz, A. Garcia-Granados,
A. Parra, M. Alvarez, F. Albericio, Tetrahedron Lett. 2005, 46,
7737–7741.
C. Gracia, A. Isidro-Llobet, L. J. Cruz, G. A. Acosta, M. Alva-
rez, C. Cuevas, E. Giralt, F. Albericio, J. Org. Chem. 2006, 71,
7196–7204.
J. C. Jiménez, A. López-Macià, C. Gracia, S. Varón, M. Carr-
ascal, J. M. Caba, M. Royo, A. M. Francesch, C. Cuevas, E.
Giralt, F. Albericio, J. Med. Chem. 2008, 51, 4920–4931.
J. Tulla-Puche, E. Marcucci, E. Prats-Alfonso, N. Bayo-Puxan,
F. Albericio, J. Med. Chem. 2009, 52, 834–839.
E. Biron, J. Chatterjee, H. Kessler, J. Pept. Sci. 2006, 12, 213–
219, and references cited therein.
[19] Attempts to reduce the azide in solution by hydrogenation fol-
lowed by protection of the amine group afforded scarce overall
yield of the diprotected aza-threonine.
[20] H. Staudinger, J. Meyer, Helv. Chim. Acta 1919, 2, 635–646.
[21] For some examples, see: a) F. Debaene, N. Winssinger, Org.
Lett. 2003, 5, 4445–4447; b) M. R. Tremblay, J. Simard, D.
Poirier, Biorg. Med. Chem. Lett. 1999, 9, 2827–2832; c) R.
Lang, L. Yan, J. Loebach, M. Ge, Y. Uozumi, K. Sekanina, N.
Horan, J. Gildersleeve, C. Thompson, A. Smith, K. Biswas,
W. C. Still, D. Kahne, Science 1996, 274, 1520–1522.
[22] a) J. Rebek Jr., S. H. Shaber, Y.-K. Shue, J.-C. Gehret, S. Zim-
merman, J. Org. Chem. 1984, 49, 5164–5174; b) K. Baum, P. T.
Berkowitz, V. Grakauskas, T. G. Archibald, J. Org. Chem.
1983, 48,2953–2956.
[8]
[9]
[10]
[11]
[12]
a) G. T. Faircloth, M. Elices, H. Sasak, P. M. Aviles, C. Cuevas,
PCT Int. Appl. WO 2004035613, A2 20040429, 2004 [Chem.
Abstr. 2004, 140, 357671]; b) F. Albericio, A. Fernandez, E.
Giralt, C. Gracia, P. Lopez, S. Varon, C. Cuevas, A. Lopez-
Macia, A. Francesch, J. C. Jimenez, M. Royo, PCT Int. Appl.
WO 2005023846, A1 20050317, AN 2005:239012, 2005 [Chem.
Abstr. 2005, 142, 298335].
[23] X. Wang, S. G. Nelson, D. P. Curran, Tetrahedron 2007, 63,
6141–6145.
[24] S. Pothukanur, Z. Pianowski, N. Winssinger, Eur. J. Org. Chem.
2008, 3141–3148.
[25] M. Vaultier, N. Knouzi, R. Carrie, Tetrahedron Lett. 1983, 24,
763–764.
[26] P. E. Schneggenburger, B. Worbs, U. Diederichsen, J. Pept. Sci.
2010, 16, 10–14.
[13] U. Schmidt, K. Mundinger, B. Riedl, G. Haas, R. Lau, Synthe-
sis 1992, 1201–1202.
[14] K. Barlos, D. Gatos, W. Schäfer, Angew. Chem. Int. Ed. Engl.
1991, 30, 590–593.
[15] See the Supporting Information.
[16] For cyclic peptides 3 and 5 containing N-methyl amino acids,
no rotameric conformations were observed by NMR analysis.
[17] Y. Luo, M. A. Blaskovich, G. A. Lajoie, J. Org. Chem. 1999,
64, 6106–6111, and references cited therein.
[18] S. Brase, C. Gil, K. Knepper, V. Zimmermann, Angew. Chem.
Int. Ed. 2005, 44, 5188–5240.
[27] M. Bartra, P. Romea, F. Urpí, J. Vilarrasa, Tetrahedron 1990,
46, 587–594.
[28] For some examples, see: a) E. K. Kick, J. A. Ellman, J. Med.
Chem. 1995, 38, 1427–1430; b) J.-M. Kim, Y. Bi, S. J. Paikoff,
P. G. Schultz, Tetrahedron Lett. 1996, 37, 5305–5308; c) Z.
Zhang, T. Carter, E. Fan, Tetrahedron Lett. 2003, 44, 3063–
3066; d) R. Maltais, C. Mercier, F. Labrie, D. Poirier, Mol.
Diversity 2005, 9, 67–79.
Received: November 22, 2009
Published Online: March 19, 2010
Eur. J. Org. Chem. 2010, 2536–2543
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2543