Syntheses of triostin a antibiotic and nucleobase-functionalized analogs as new DNA binders

Article abstract of DOI:10.1002/ejoc.200900530

A total synthesis of the natural product triostin A, wherein the N-methylated depsipeptide scaffold is constructed by solution-phase peptide chemistry followed by disulfide formation and macrocyclization, is described, Finally, the quinoxal.in.es were att

Full text of DOI:10.1002/ejoc.200900530

FULL PAPER
DOI: 10.1002/ejoc.200900530
Syntheses of Triostin A Antibiotic and Nucleobase-Functionalized Analogs as
New DNA Binders
Anmol Kumar Ray[a] and Ulf Diederichsen*[a]
Keywords: Amino acids / Cyclic depsipeptide / DNA recognition / Nucleobases / Triostin A
A total synthesis of the natural product triostin A, wherein
the N-methylated depsipeptide scaffold is constructed by
solution-phase peptide chemistry followed by disulfide for-
mation and macrocyclization, is described. Finally, the quin-
oxalines were attached to provide the DNA bisintercalator.
Analogs of triostin A were obtained by the successive func-
tionalization of the cyclic depsipeptide with pyrimidine or
purine recognition units. The attachment of functional units
was achieved by the orthogonal protection of the respective
side chain amino functionalities. The nucleobase-function-
alized triostin analogs have the potential to recognize
double-stranded DNA by hydrogen bonding. The interaction
with DNA was investigated by UV spectroscopy and fluores-
cence intercalator displacement.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
hydrogen-bonding pattern. The triostin A backbone organi-
zation is further advantageous as an abasic site recognition
motif. Abasic sites are DNA lesions obtained by depurin-
ation and are also known intermediates in the DNA repair
process.[11–14] A nucleobase attached to the triostin depsi-
peptide scaffold could replace the lost nucleobase, whereas
the other recognition unit might function as an efficient in-
tercalator.
Introduction
The natural product triostin A isolated from Streptomy-
ces aureus belongs to the family of quinoxaline antibiot-
ics.[1–4] They are potent antitumor agents inhibiting RNA
synthesis by specific binding of double-stranded DNA (ds-
DNA) through bisintercalation. A structural characteristic
of triostin A (1) is a rigid, disulfide-bridged, bicyclic, depsi-
peptide scaffold, which preorganizes two quinoxaline inter-
calating units (Figure 1).[5,6] The aromatic groups are ori-
ented in parallel at a distance of 10.5 Å. This is a perfect
orientation of two intercalators to interact with two adja-
cent DNA base pairs. Indeed, the specificity of triostin A
for a C-G sequence of DNA is based on the defined orien-
tation of the quinoxalines – which interact by π-stacking
with the DNA bases – and the hydrogen bonding between
the peptide scaffold and ds-DNA.[7] Our interest in triostin
A is based on the unique characteristic of the depsipeptide
scaffold to provide recognition units in a preorganized ori-
entation.[8–10] Besides bisintercalation, the distance of
10.5 Å for the parallel-oriented recognition units generally
allows for DNA recognition by hydrogen bonding with the
nucleobases in the major or minor groove. The major
groove – presenting the Hoogsteen base-pair site – should
especially be a target for specific differentiation between the
canonical nucleobase pairs. Considering the rigid backbone
of triostin A, every third DNA base pair would be ad-
dressed provided that the recognition unit has the required
[a] Institut für Organische und Biomolekulare Chemie, Georg-
August-Universität Göttingen,
Tammannstr. 2, 37077 Göttingen, Germany
Fax: +49-551-39-22944
E-mail: udieder@gwdg.de
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900530.
Figure 1. Structure of triostin A (1) and nucleobase-substituted an-
alogs 2–5 based on the triostin A backbone.
Eur. J. Org. Chem. 2009, 4801–4809
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4801

A. Kumar Ray, U. Diederichsen
FULL PAPER
Synthetic access to the bicyclic triostin A depsipeptide chains (Scheme 1). The Tce group increased the solubility
backbone opens the possibility of specifically attaching of all the depsipeptides in organic solvents and enabled the
various recognition units and subsequently making various isolation of the free amines by extraction with ethyl acetate.
backbone modifications. A key intermediate would be an We coupled amino acids Z--Ser-OTce (10) and Fmoc--
orthogonally amine-protected bicyclic depsipeptide, which Ser-OTce (11) with N-Boc-N-Me-Val-OH to yield the esters
could be successively functionalized using amide formation. 12 and 13, respectively.[16] We deprotected these depsipep-
Since larger quantities of triostin A analogs were required tides using 4  HCl in dioxane and then coupled them with
for cocrystallization experiments with ds-DNA, the synthe- N–Me–Boc--Cys(Acm)-OH.[22,23] We obtained tridepsi-
sis of triostin A and nucleobase analogs was approached by peptides 14 and 15, respectively, by activation with 1-(3-
solution-phase peptide synthesis.
dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride
Triostin A (1) is an octadepsipeptide symmetrically com- (EDCI·HCl) and 1-hydroxy-7-azabenzotriazole (HOAt).[24]
posed of two tetrapeptides containing amino acids N- We used acetamidomethyl (Acm) side chain protection for
methyl--cysteine, -serine, -alanine, and N-methyl--val- cysteine to provide deprotection and oxidation in a single
ine. The tetrapeptides are macrocyclized forming two ester step at a later stage. We obtained tetradepsipeptides 16 and
linkages between the serine side chain and the C-terminus 17 after the Boc-deprotection of the respective tridepsipep-
of the opposite tetrapeptide. The disulfide-bridged depsi- tides with 4  HCl in dioxane followed by EDCI/HOAt-
peptide is provided by oxidation; serine amino groups are mediated coupling with Boc--Ala-OH. We selectively de-
orthogonally protected to allow for attachment of different protected tetradepsipeptide 16 to hydrochloride 6 and
recognition units. The N-methylation of four amino acids oligomer 17 to the free carboxylic acid 7 using 4  HCl in
renders amide formation more difficult. Two solution-phase dioxane and zinc powder in 90% HOAc, respectively.
syntheses of triostin A by Olsen et al. and Shin et al. are
We accomplished the coupling of fragments 6 and 7 to
known to date, both following a strategy of linking the tet- yield the linear octadepsipeptide 18 by standard peptide
rapeptides followed by macrocyclization and oxidation or coupling with EDCI/HOAt in DCM in good yield
vice versa.[15,16] A differentiation of the amino groups for (Scheme 2). The reductive cleavage of the Tce ester 18 was
functionalization was not required in these triostin A syn- followed by the iodine-mediated deprotection of the Acm
theses as triostin A is symmetrical. In addition, an elegant groups and spontaneous disulfide bond formation under
solid-phase protocol was recently provided.[17] Our ap- high dilution. The terminal ends of resulting depsipeptide
proach to triostin A is driven by the desire to access deriva- 20 were sufficiently preorganized for macrocyclization. We
tives with two different DNA recognition functionalities. obtained the key 21 by acidic N-terminal deprotection fol-
This approach requires orthogonal protection of the amino lowed by ring closure yielding the bicyclic octadepsipeptide
groups of the triostin A scaffold and was used previously in a respectable 58% yield using diisopropylcarbodiimide
to synthesize triostin A backbone analogs, like TANDEM (DIC)/HOAt activation. The orthogonal protecting groups
(des-N-tetramethyltriostin) and des-N-(tetramethyl)aza- at the amino groups allowed the synthesis of substrates
triostin.[8–10,18,19]
presenting two different recognition units. Therefore, be-
The solution-phase peptide synthesis of triostin A is de- sides triostin A, various nucleobase-substituted building
scribed. Furthermore, four triostin derivatives 2–5 were pre- blocks were accessible based on the bicyclic scaffold 21 after
pared containing nucleobases as recognition units in dif- selective Fmoc and Z deprotection.
ferent combinations. Their interaction with ds-DNA was
evaluated by spectroscopic methods.
We accomplished the synthesis of the natural product tri-
ostin A (1) with two quinoxaline moieties by Z-group de-
protection followed by coupling of 2-quinoxaline carboxylic
acid mediated by DIC/HOAt activation, Fmoc-deprotec-
tion, and coupling of the second quinoxaline unit under
identical conditions. The successive deprotection and cou-
pling procedure is not the most beneficial strategy for tri-
ostin A itself, as indicated by a 15% yield over the last four
steps. Nevertheless, the main focus of this study was set on
the triostin A derivatives with mixed functionalities.
We performed the functionalization of the triostin A
scaffold 21 with two different recognition units by subse-
quent deprotection and coupling. We prepared the required
nucleobase-substituted acetic acids with suitable protecting
groups accordingly to the synthesis of nucleo amino acids
used for PNA oligomers.[25] We selected benzyloxycarbonyl
(Z) and phenoxy (PhO) protecting groups to increase the
Results and Discussion
We obtained the disulfide-bridged depsipeptide scaffold
of triostin A by the fusion of two tetrapeptides, disulfide
formation, and macrocyclization. Since the tetrapeptides
were identical in sequence but differed in their serine pro-
tecting groups, we decided to include the ester linkage in
the tetrapeptide and use amide bond formation for linkage
and macrocyclization.
Synthesis of Orthogonally Protected Triostin A Backbone
and Functionalization with Various Recognition Units
We built suitably protected tetradepsipeptides 6 and 7 solubility in organic solvents and to prevent nucleophilic
stepwise in solution. Starting with Z--Ser(tBu)-OH (8) and reactivity of the nucleobase amino groups. We obtained the
Fmoc--Ser(tBu)-OH (9),[19–21] we prepared the trichloroe- first triostin A derivative 2 by coupling methylene guanine
thanol (Tce) esters and subsequently deprotected the side instead of the second quinoxaline unit. We obtained further
4802
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 4801–4809

Syntheses of Triostin A Antibiotic
Scheme 1. Synthesis of tetradepsipeptides 6 and 7.
Scheme 2. Macrocyclization, oxidation, and functionalization to yield the natural product triostin A (1) and its nucleobase analogs 2–5.
derivatives by the linkage of two nucleobase recognition firmed the integrity of the respective derivatives by mass
units. For the synthesis of triostin A analogs 3–5 presenting spectrometry and NMR spectroscopy. The NMR spectro-
two different recognition units, we removed the Z protecting scopic data of the synthesized triostin A (1) were identical
group of 21 with TFA/thioanisole and attached the first nu- with the data previously reported.[26]
cleobase moiety to the free amino group by treatment with
DIC and HOAt (Scheme 2). We removed the Fmoc group
subsequently by addition of 20% piperidine in DMF. We
coupled the second nucleobase moiety furnishing triostin A
analogs 3–5. We deprotected these analogs with a mixture
DNA Interaction of Nucleobase-Containing
Triostin A Analogs
of thioanisole and TFA in order to remove the nucleobase
We investigated the DNA-binding potential of the new
protecting groups and purified them by HPLC. We con- triostin A derivatives using a well-defined ds-DNA decamer
Eur. J. Org. Chem. 2009, 4801–4809
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4803

A. Kumar Ray, U. Diederichsen
FULL PAPER
with an adeninyl-hairpin structure given by the sequence competing intercalators leads to a decrease of fluorescence
5ЈCGTAGCGTAC AAAAA GTACGCTACG3Ј (22). In a intensity. Since there is a CG specificity of bisintercalating
second DNA hairpin structure (5ЈCGTAXCGTAC triostin A and considering nearest neighbor exclusion of
AAAAA GTACGCTACG3Ј, 23) we exchanged one gua- intercalator positions, we expected a substitution of about
nosine nucleotide for an abasic site (X) in order to also in- 50% of the ethidium bromide molecules. Indeed, by the ad-
vestigate the abasic site binding potential of the synthetic dition of successive amounts of triostin A to DNA 22 satu-
triostin analogs 2–5.
rated with ethidium bromide we obtained a value of about
50% fluorescence intensity (Figure 3). The A-T triostin an-
alog 4 provided similar ethidium displacement results,
whereas the other nucleobase-modified triostin analogs 2,
3, and 5 showed significantly lower ethidium displacement
pointing to a binding mode that differs from bisintercal-
ation.
Temperature-Dependent UV Spectroscopy
We determined the double strand stability of hairpin
DNA 22 by temperature-dependent UV spectroscopy indi-
cating the cooperative destacking of nucleobase pairs with
increasing temperature. For the double strand 22, we deter-
mined a stability of Tm = 64 °C (1.5 µ) (Figure 2). Adding
five equivalents of triostin A significantly increased the sta-
bility to Tm = 72 °C indicating the contribution of two in-
tercalating quinoxalines and stabilization based on back-
bone recognition. The influence of the triostin analogs 2–5
on the stability of hairpin DNA 22 was rather weak. We
found a slight increase in DNA duplex stability in the pres-
ence of the Q-G analog 2 (Tm = 65 °C) or the A-G analog
5 (Tm = 65 °C). We detected stabilities comparable to or
slightly lower than the self association of DNA 22 in case
the C-G analog 3 (Tm = 62 °C) or the T-A analog 4 (Tm
=
63 °C). From the thermal UV profiles of analogs 2–5 with
DNA 22, we could draw no conclusions about the possible
interacting modes except that they differ from triostin A
bisintercalation. The interaction of triostin A and analogs
2–5 with DNA 23 containing the abasic site provided a sim-
ilar result (Supporting Information). Triostin A stabilized
the hairpin DNA 23 (Tm = 40 °C, 1.5 µ to Tm = 50 °C),
whereas the derivatives 2 (Tm = 41 °C), 3 (Tm = 43 °C), 4
(Tm = 43 °C), and 5 (Tm = 43 °C) only provided minor du-
plex stabilization. None of the triostin analogs displayed
especially favored recognition of an abasic site.
Figure 3. Ethidium bromide fluorescence intercalator displacement
with DNA hairpin sequence 5ЈCGTAGCGTAC AAAAA
GTACGCTACG3Ј (22) and triostin A (1) and 22 with the respec-
tive nucleobase analogs 2–5.
Comparable FID experiments with DNA 23 provided a
lower level of displacement for triostin A of about 25%,
indicating the influence of the abasic site (Figure 4). As for
DNA analog 23, the A-T triostin analog 4 showed a similar
displacement. Nevertheless, at higher concentrations, we
reached further displacement of approximately 40%. The
A-G analog 5 also attained a level of displacement compar-
able to triostin A, at least at higher concentrations. The
Figure 2. Temperature-dependent UV spectra of DNA hairpin
5ЈCGTAGCGTAC AAAAA GTACGCTACG3Ј (22) and 22 (2 m,
1.5 µ, HEPES, NaCl 10 m, pH 7.0) with triostin A (1) and re-
spective nucleobase analogs 2–5 (7.5 µ).
Fluorescence Intercalator Displacement (FID)
Figure 4. Ethidium bromide fluorescence intercalator displacement
with DNA hairpin sequence 5ЈCGTAXCGTAC AAAAA
GTACGCTACG3Ј (23) and triostin A (1) and 23 with the respec-
In a FID assay, the fluorescence of DNA-bound ethid-
ium bromide is detected.[27,28] Starting with ds-DNA satu-
rated with intercalating ethidium bromide, the addition of tive nucleobase analogs 2–5.
4804
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 4801–4809

Syntheses of Triostin A Antibiotic
FID Experiment: ds-DNA 22 or 23 was added to ethidium bromide
(5 equiv., one equiv. per two base pairs of the hairpin loop) in 2 m
HEPES buffer (10 m aq. NaCl at pH 7.0, 10 mm cuvette, 500 µL
of total volume). Triostin A or an analog was added in portions
(2 µL of 0.8 m solution in DMSO), and the sample was incubated
at 25 °C for 30 min before the emission spectrum was measured
(λexcitation = 545 nm, λemission = 595 nm).
FID of the other derivatives was negligible. Overall, the
FID experiments indicated differences in binding of triostin
A and its analogs but did not provide conclusions on the
mode of interaction. Cocrystallization of the triostin ana-
logs with ds-DNA is currently being attempted.
N-Z-
D
-Ser{N-Fmoc-
D
-Ser[N-Boc-
-Cys(Acm)-N-Me-L
L
-Ala-N-Me-L-Cys(Acm)-N-Me-
-Val}-OTce (18): To a
Conclusions
L
-Val]-
L
-Ala-N-Me-
L
solution of 7 (1.5 g, 1.88 mmol) in DCM (5 mL) at 0 °C, HOAt
(308 mg, 2.26 mmol, 1.2 equiv.) and EDCI (434 mg, 2.26 mmol,
1.2 equiv.) were added sequentially. The mixture was stirred at 0 °C
for 30 min followed by the addition of a solution of 6 (1.5 g,
2.02 mmol, 1.07 equiv.) in DCM (12 mL). The reaction mixture
was stirred at 0 °C for 9 h under argon before being poured into
cold aq. HCl (1 ⁿ, 200 mL). The aq. phase was extracted with ethyl
acetate (2ϫ200 mL). The combined organic phases were washed
with NaHCO3 (5%, 200 mL) and saturated NaCl (120 mL), dried
(Na2SO4), and concentrated in vacuo. Flash chromatography
(SiO2, 4ϫ15 cm, MeOH/ethyl acetate, 1:19) yielded 18 (1.95 g,
1.28 mmol, 68%) as a white foam. Rf (ethyl acetate/MeOH, 20:1)
= 0.62. M.p. 104 °C. [α]2D0 = –108 (MeOH, c = 0.08). UV (MeOH):
A total synthesis of the natural product triostin A and
analogs carrying nucleobases instead of the quinoxalines as
recognition units is presented. The key intermediate al-
lowing variable functionalization of the triostin backbone
is the orthogonally protected, disulfide-bridged cyclodepsi-
peptide. This derivative is used as a template, allowing the
attachment of all kinds of functionalities to be organized at
a 10.5 Å distance. The nucleobase-containing triostin ana-
logs were prepared with the intention to create new ds-
DNA binding motifs. UV and FID experiments provide evi-
dence for the recognition that differs from that of triostin
A bisintercalation.
λmax = 266 nm. IR (KBr): ν = 3417, 2967, 2953, 1727, 1648, 1526,
˜
1
1437, 1377, 1255, 1170, 1058, 739, 575 cm–1. H NMR (600 MHz,
CDCl3, 25 °C): δ = 0.73–0.78 (m, 6 H, Val-CH3), 0.95–0.99 (m, 6
H, Val-CH3), 1.27–1.30 (m, 6 H, Ala-CH3), 1.41 (s, 6.5 H, tBu-
rot), 1.42 (s, 2.5 H, tBu-rot), 1.95 (s, 3 H, Acm-CH3-rot), 2.05 (s,
3 H, Acm-CH3-rot), 2.13–2.20 (m, 2 H, Val-Hβ), 2.79–2.83 (m, 2
H, Cys-Hβ), 2.87 (s, 3 H, Val-NCH3/Cys-NCH3), 2.88 (s, 3 H, Val-
NCH3/Cys-NCH3), 2.91 (s, 3 H, Val-NCH3/Cys-NCH3), 2.92 (s, 3
H, Val-NCH3/Cys-NCH3), 3.02–3.14 (m, 2 H, Cys-Hβ), 4.19–4.23
(m, 2 H, Fmoc-CH), 4.33–4.36 (m, 4 H, Acm-CH2), 4.44–4.92 (m,
13 H, Tce-CH2, Ala-Hα, Fmoc-CH2, Val-Hα, Ser-Hα, Ser-Hβ),
5.09–5.12 (m, 2 H, Z-CH2), 5.35–5.38 (m, 1 H, NH), 5.56–5.70 (m,
2 H, Cys-Hα), 6.52–6.59 (m, 1 H, NH), 6.71–6.78 (m, 1 H, NH),
7.26–7.38 (m, 9 H, Fmoc-H3, Fmoc-H4, Z-Ph), 7.57–7.61 (m, 2 H,
Fmoc-H2), 7.73–7.74 (m, 2 H, Fmoc-H5) ppm. 13C NMR
(125 MHz, CDCl3, 25 °C): δ = 18.2 (Ala-CH3), 18.6, 18.7, 18.9,
19.8, 19.9, 20.0 (Val-CH3, Ala-CH3), 23.0, 23.1 (Acm-CH3), 26.9
(Val-CHβ), 28.3 [C(CH3)3], 30.0, 30.3, 30.4, 30.5 (Val-NCH3, Cys-
NCH3, Cys-CH2β), 41.8 (Acm-CH2), 46.7 (Ala-CHα), 47.0 (Ala-
CHα), 52.6 (Fmoc-CH), 53.4, 53.5 (Ser-CHα, Cys-CHα), 61.9 (Val-
CHα), 64.4 (Ser-CH2β), 67.2 (Fmoc-CH2), 67.9 (Z-CH2), 74.6 (Tce-
CH2), 79.9 [C(CH3)3], 94.2 (Tce-CCl3), 120.0 (Fmoc-C5), 125.1,
125.2 (Fmoc-C2), 127.1, 127.7, 128.0 (Fmoc-C3/Z-Ph), 128.2
(Fmoc-C4), 128.6 (Fmoc-C3/Z-Ph), 136.1 (Z-Phipso), 141.3 (Fmoc-
C6), 143.7 (Fmoc-C1), 155.1, 156.1, 156.6 (Boc-CO, Fmoc-CO, Z-
CO), 167.8, 170.1, 170.6 (Ser-CO, Val-CO, Cys-CO, Ala-CO, Acm-
CO) ppm. HRMS (ESI): calcd. for C68H94Cl3N10O19S2 [M + H]+
1523.5198; found 1523.5207.
Experimental Section
General Remarks: Solvents were used in the highest grade available.
DCM was distilled from calcium hydride prior to use. DMF was
purchased dry and stored over molecular sieves (4 Å). Commer-
cially available reagents were of analytical grade and used without
further purification. Melting points were obtained with a Büchi
501 Dr. Tottoli apparatus. Optical rotations were determined with
a Perkin–Elmer 241 polarimeter. IR spectra were recorded with a
Perkin–Elmer 1600 Series FT-IR spectrometer using KBr pellets.
NMR spectra were recorded with Varian INOVA-600 and INOVA-
300 instruments. Chemical shifts are referenced to the residual sol-
vent peaks. ESI-MS data were measured with a LCQ Finnigan
spectrometer. HRMS data were determined with a Bruker APEX-
Q IV 7T spectrometer. HPLC was performed with a Pharmacia
Äkta basic system using YMC JЈsphere ODS-H80, RP-C18 col-
umns for both analytical samples (250ϫ4.6 mm, 5 µ, 120 Å,
1 mLmin–1) and preparative runs (250ϫ20 mm, 5 µ, 120 Å,
10 mLmin–1); eluent A: water/TFA (0.1%); eluent B: acetonitrile/
water, 4:1, TFA (0.1%). Analytical thin-layer chromatography was
performed by using Merck silica gel 60, F254-precoated aluminum
plates. UV light (254 nm) or dyeing with ninhydrin (5% in ethanol)
was used for detection.
UV Melting Experiment: To ds-DNA 5ЈCGTAGCGTAC AAAAA
GTACGCTACG3Ј
(22)
or
5ЈCGTAXCGTAC
AAAAA
GTACGCTACG3Ј (23) at 1.5 µ in buffer (500 µL of 2 m HEPES
buffer in 10 m aq. NaCl at pH 7.0 in a 10 mm cuvette) triostin or
a triostin analog (7.5 µ, 5 equiv. in DMSO added in 2 µL from a
stock solution) was added. Measurements of temperature-depend-
ent UV spectra were performed with a JASCO V-550 UV/Vis spec-
trometer equipped with JASCO ETC-505S/ETC-505T temperature
controller. The determination of the oligomer concentration was
based on the absorption at 260 nm measured at 20 °C. Samples
were placed in a quartz cell of 1 cm path length, and the data were
collected at eight different wavelengths (240–300 nm) according to
the following temperature protocol with a regular flow of nitrogen:
20 °C Ǟ 80 °C (10 min) Ǟ 80 °C (3 min) Ǟ 0 °C (26 min) Ǟ 0 °C
(10 min) Ǟ 80 °C (180 min) Ǟ 0 °C (180 min) Ǟ 28 °C (5 min).
Z-
Val]-
D
-Ser{N-Fmoc-
-Ala-N-Me-
D
-Ser[N-Boc-
L
-Ala-N-Me-
L-Cys(Acm)-N-Me-L-
L
L-Cys(Acm)-N-Me-L
-Val}-OH (19): A solution of
18 (1.9 g, 1.25 mmol) in 90% aq. HOAc (65 mL) was cooled to
0 °C followed by the addition of zinc powder (4.09 g, 62.5 mmol,
50 equiv.). The suspension was stirred at 0 °C for 4 h, filtered, and
the filtrate was concentrated in vacuo. The residue was treated with
cold HCl (1 , 150 mL) and extracted with CHCl3 (3ϫ150 mL).
The combined organic phases were washed with saturated aq. NaCl
(150 mL), dried (Na2SO4), and concentrated in vacuo. After purifi-
cation by flash chromatography (SiO2, 2ϫ16 cm, MeOH/ethyl ace-
tate/HOAc, 3:16:1) and coevaporation with toluene (3ϫ25 mL), 19
(1.23 g, 0.88 mmol, 70%) was obtained as a white foam. Rf (ethyl
acetate/MeOH/HOAc, 10:2:1) = 0.30. M.p. 209 °C. [α]2D0 = –97
Eur. J. Org. Chem. 2009, 4801–4809
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4805

A. Kumar Ray, U. Diederichsen
FULL PAPER
(MeOH, c = 0.04). UV (MeOH): λmax = 261 nm. IR (KBr): ν =
(125 MHz, [D6]DMSO, 25 °C): δ = 16.5, 17.2, 17.3, 18.5, 18.8, 19.0,
19.5, 19.6, 19.7, 19.8 (Val-CH3, Ala-CH3), 26.6 (Val-CHβ), 26.9
˜
3440, 2925, 2362, 1646, 1464, 1381, 1262, 1060, 670 cm–1. 1H NMR
3
(600 MHz, [D6]DMSO, 25 °C): δ = 0.75 (d, JH,H = 7 Hz, 4.5 H, (Val-CHβ), 28.1 [C(CH3)3], 30.2, 30.7, 31.0 (Val-NCH3/Cys-NCH3,
Val-CH3), 0.84–0.86 (m, 1.5 H, Val-CH3), 0.89–0.91 (m, 4.5 H, Val- Cys-CH2β), 46.1, 46.4, 46.6 (Fmoc, Ala-CHα), 53.8 (Cys-CHα),
3
3
CH3), 0.94 (d, JH,H = 7 Hz, 1.5 H, Val-CH3), 1.11 (d, JH,H
=
62.3 (Ser-CHα), 63.0 (Ser-CHα), 63.4 (Val-CHα), 65.6 (Ser-CH2β),
65.7 (Z-CH2-rot), 66.1 (Fmoc-CH2), 78.0 [C(CH3)3], 120.1 (Fmoc),
3
7 Hz, 0.7 H, Ala-CH3), 1.14 (d, JH,H = 7 Hz, 3 H, Ala-CH3), 1.19
(d, JH,H = 7 Hz, 2.3 H, Ala-CH3), 1.35 (s, 2 H, tBu-rot), 1.37 (s, 125.1 (Fmoc-C2), 127.0 (Fmoc), 127.6, 127.7, 127.8, 128.3 (Fmoc-
3
7 H, tBu-rot), 1.82 (s, 6 H, Acm-CH3), 2.14–2.19 (m, 2 H, Val-Hβ),
2.57–2.63 (m, 0.5 H, Cys-Hβ), 2.69–2.71 (m, 1.5 H, Cys-Hβ), 2.72–
2.75 (m, 3 H, Val-NCH3/Cys-NCH3), 2.85–2.88 (m, 9 H, Val-
C3, Fmoc, Z-Ph), 136.6 (Z-Phipso), 136.7 (Fmoc), 140.7 (Fmoc),
143.6 (Fmoc), 143.7 (Fmoc), 154.8, 155.9 (Boc-CO, Fmoc-CO, Z-
CO), 169.2, 169.3, 169.4, 169.6, 170.6, 172.0 (Ser-CO, Val-CO, Cys-
NCH3/Cys-NCH3), 2.96–3.02 (m, 1.5 H, Cys-Hβ), 3.06–3.10 (m, CO, Ala-CO) ppm. HRMS (ESI): calcd. for C60H81N5O17S2 [M +
0.5 H, Cys-Hβ), 3.98–4.01 (m, 1 H, Ser-Hα), 4.02–4.06 (m, 2 H, Na]+ 1249.5155; found 1249.5149.
Acm-CH2, Ser-Hβ), 4.15–4.24 (m, 3 H, Acm-CH2, Ser-Hβ), 4.25–
N-Z-
D-Ser-L-Ala-N-Me-L-Cys-N-Me-L-Val-N-Fmoc-D-Ser-L-Ala-
4.33 (m, 5 H, Fmoc-CH, Fmoc-CH2, Acm-CH2, Ser-Hβ), 4.34–
4.37 (m, 3 H, Ala-Hα, Ser-Hα, Val-Hα), 4.42–4.45 (m, 2 H, Val-
N-Me-L-Cys-N-Me-
L
-Val) (Serine-hydroxy)-Dilactone Disulfide
(21): A solution of 20 (540 mg, 430 µmol) in dioxane (1.1 mL) was
treated with HCl (4 , 3.2 mL) in dioxane at 0 °C. The mixture was
stirred at 0 °C for 30 min and at 25 °C for 30 min before the vola-
tiles were removed in vacuo. The residual HCl was removed by
adding Et2O (30 mL) to the hydrochloride salt followed by solvent
removal in vacuo. The residue was dissolved in DCM (400 mL),
and the resulting solution was treated sequentially with HOAt
(350 mg, 2.58 mmol) and EDCI (495 mg, 2.58 mmol). The reaction
mixture was stirred at 0 °C for 12 h before being poured into HCl
(1 , 117 mL). The organic phase was extracted with ethyl acetate
(2ϫ234 mL), and the combined organic phases were washed with
NaHCO3 (5%, 234 mL) and saturated aq. NaCl (70 mL). The or-
ganic phase was dried (Na2SO4), filtered, and concentrated in
vacuo. Flash chromatography (SiO2, 4ϫ15 cm, ethyl acetate) af-
forded 21 (282 mg, 250 µmol, 58%) as a white solid. Rf (ethyl ace-
3
3
Hα, Ser-Hβ), 4.52 (d, JH,H = 10 Hz, 1 H, Ser-NH), 4.56 (d, JH,H
= 10 Hz, 1 H, Ser-NH), 4.68–4.71 (m, 0.2 H, Ala-Hα), 4.74–4.79
(m, 0.8 H, Ala-Hα), 5.03 (s, 0.5 H, Z-CH2), 5.05 (s, 1.5 H, Z-CH2),
5.39–5.44 (m, 0.5 H, Cys-Hα), 5.46–5.51 (m, 1.5 H, Cys-Hα), 6.49–
6.62 (m, 1 H, Ser-NH), 6.80–6.89 (m, 1 H, Ser-NH), 7.31–7.36 (m,
7 H, Fmoc, Z-Ph), 7.40–7.42 (m, 2 H, Fmoc), 7.63–7.72 (m, 3 H,
3
Fmoc, Ala-NH), 7.87–7.89 (d, JH,H = 7 Hz, 2 H, Fmoc), 8.27–
8.39 (m, 3 H, Ala-NH, Acm-NH) ppm. 13C NMR (150 MHz, [D6]-
DMSO, 25 °C): δ = 17.3 (Ala-CH3), 17.7 (Ala-CH3), 18.7 (Val-
CH3), 19.7 (Val-CH3), 19.8 (Val-CH3), 22.5 (Acm-CH3), 26.9 (Val-
CHβ), 27.0 (Val-CHβ), 28.1 (Cys-CH2β), 29.0 [C(CH3)3], 29.1
(Cys-CH2β), 30.0, 30.1, 31.5, 31.7 (Val-NCH3/Cys-NCH3), 40.1
(Acm-CH2), 45.2 (Ala-CHα), 46.4 (Ala-CHα), 46.5 (Fmoc-CH),
52.5 (Cys-CHα), 52.7 (Cys-CHα), 52.8 (Ser-CHα), 53.4 (Ser-CHα),
62.3 (Val-CHα), 62.5 (Val-CHα), 63.5 (Ser-CH2β), 64.0 (Ser-CH2β),
65.6 (Z-CH2), 66.0 (Fmoc-CH2), 78.0 [C(CH3)3], 120.0 (Fmoc-C5),
125.2 (Fmoc-C2), 127.0, 127.6, 127.7, 127.8, 128.3, (Fmoc, Z-Ph),
136.8 (Z-Phipso), 140.6 (Fmoc), 143.7 (Fmoc), 154.8, 155.8, 155.9
(Boc-CO, Fmoc-CO, Z-CO), 167.6, 169.3, 169.5, 169.6, 169.7,
170.6, 172.0 (Ser-CO, Val-CO, Cys-CO, Ala-CO, Acm-CO) ppm.
HRMS (ESI): calcd. for C66H93N10O19S2 [M + H]+ 1393.6054;
found 1393.6055.
tate) = 0.5. [α]2D0 = –119 (MeOH, c = 0.05). UV (MeOH): λmax
257, 264 nm. IR (KBr): ν = 3423, 2926, 2371, 1956, 1647, 1519,
=
˜
1463, 1402, 1256, 1172, 1054, 741, 523, 416 cm–1 1H NMR
.
(600 MHz, [D6]DMSO, 25 °C): δ = 0.85 (d, 3JH,H = 7 Hz, 3 H, Val-
CH3), 0.88 (d, 3JH,H = 6 Hz, 3 H, Val-CH3), 0.94 (d, 3JH,H = 7 Hz,
6 H, Val-CH3), 1.24–1.29 (m, 6 H, Ala-CH3), 2.23–2.35 (br. m, 4
H, Val-Hβ, Cys-Hβ), 2.78 (s, 3 H, Val-NCH3/Cys-NCH3), 2.79 (s,
3 H, Val-NCH3/Cys-NCH3), 3.14 (s, 3 H, Val-NCH3/Cys-NCH3),
3.16 (s, 3 H, Val-NCH3/Cys-NCH3), 3.48–3.52 (m, 2 H, Cys-Hβ),
4.17–4.24 (m, 3 H, Ser-Hβ), 4.25–4.30 (m, 2 H, Fmoc-CH, Ser-
Hβ), 4.30–4.38 (m, 4 H, Fmoc-CH2, Ser-Hα), 4.39–4.46 (m, 2 H,
Ala-Hα), 4.66–4.70 (m, 2 H, Val-Hα), 5.06–5.15 (m, 2 H, Z-CH2),
6.08–6.15 (br. m, 2 H, Cys-Hα), 7.31–7.44 (m, 9 H, Z-Ph, Fmoc),
7.59–7.62 (s, Ala-NH, 1 H, br.), 7.66–7.73 (m, 5 H, Fmoc, Ala-
Z-
D-Ser[N-Fmoc-
D
-Ser(N-Boc-
L-Ala-N-Me-L-Cys-N-Me-L-Val)-L-
Ala-N-Me-L-Cys-N-Me-L-Val]-OH Disulfide (20): To a solution of
iodine (2.17 g, 8.5 mmol, 10 equiv.) in DCM/MeOH (770 mL, 9:1)
a solution of 19 (1.18 g, 0.85 mmol) in DCM (350 mL) was added
dropwise, and the mixture was stirred at 25 °C for 6 h. The reaction
mixture was cooled to 0 °C, and Na2S2O3 (5%) was added slowly
until the excess iodine was discharged, and the color disappeared.
The organic phase was washed with aq. HCl (1 , 120 mL) and
saturated aq. NaCl (100 mL), dried (Na2SO4), filtered, and concen-
trated in vacuo. Flash chromatography (SiO2, 2ϫ18 cm, MeOH/
ethyl acetate, 3:17, 0.3% HOAc) followed by coevaporation with
toluene (3ϫ25 mL) furnished 20 (590 mg, 470 µmol, 55%) as a
pale yellow solid. Rf (ethyl acetate/MeOH, 5:1, 0.5 % HOAc) =
0.34. M.p. 281–283 °C. [α]2D0 = –119 (MeOH, c = 0.05). UV
NH, Ser-NH), 7.88 (d, JH,H = 8 Hz, 2 H, Fmoc-H5) ppm. 13C
3
NMR (150 MHz, [D6]DMSO, 25 °C): δ = 15.7 (Ala-CH3), 15.8
(Ala-CH3), 19.5 (Val-CH3), 19.6 (Val-CH3), 19.7 (Val-CH3), 19.8
(Val-CH3), 26.6 (Val-CHβ), 30.1 (Val-NCH3/Cys-NCH3), 31.0
(Val-NCH3/Cys-NCH3), 31.1 (Val-NCH3/Cys-NCH3), 42.8 (Cys-
CH2β), 46.4 (Ala-CHα), 46.6 (Fmoc-CH), 53.7 (Ser-CHα), 56.4
(Cys-CHα), 63.0 (Val-CHα), 64.9 (Ser-CH2β, Fmoc-CH2), 66.0 (Z-
CH2), 119.9 (Fmoc), 120.1 (Fmoc), 121.2 (Fmoc), 124.9 (Fmoc),
126.9, 126.9, 127.2, 127.6, 127.6, 127.8, 128.2, 128.8 (Fmoc, Z-Ph),
136.6 (Z-Phipso), 140.7 (Fmoc), 143.5, 143.6 (Fmoc), 168.5 (Z-CO),
169.5 (Fmoc-CO), 169.6, 169.8, 169.9 (Ser-CO, Val-CO, Cys-CO),
171.0 (Ala-CO) ppm. HRMS (ESI): calcd. for C55H71N8O14S2 [M
+ H]+ 1131.4525; found 1131.4529.
(MeOH): λmax = 257, 264 nm. IR (KBr): ν = 3423, 2926, 2371,
˜
1956, 1647, 1519, 1463, 1402, 1256, 1172, 1054, 741, 523, 416 cm–1.
1H NMR (600 MHz, [D6]DMSO, 25 °C): δ = 0.75–0.76 (m, 6 H,
Val-CH3), 0.91–0.93 (m, 6 H, Val-CH3), 1.15–1.16 (m, 3 H, Ala-
3
CH3), 1.24 (d, JH,H = 7 Hz, 3 H, Ala-CH3), 1.37 (s, 9 H, tBu),
2.17–2.23 (m, 2 H, Val-Hβ), 2.82 (m, 2 H, Cys-Hβ), 2.87–2.97 (m, General Method of Preparation of Nucleobase-Substituted Triostin
12 H, Val-NCH3/Cys-NCH3), 3.18–3.21 (m, 2 H, Cys-Hβ), 4.25– A Analogs. Step 1: Depsipeptide 21 was treated with thioanisole
4.52 (m, 13 H, Ala-Hα, Val-Hα, Fmoc-CH, Fmoc-CH2, Ser-Hα, (14.6 µL per µmol of 21) and TFA (146 µL per µmol of 21). The
Ser-Hβ), 5.03–5.04 (m, 2 H, Z-CH2), 5.54–5.62 (m, 0.7 H, Cys-
Hα), 5.68–5.71 (m, 1.3 H, Cys-Hα), 7.32–7.35 (m, 8 H, Fmoc, Z- hydrochloride salt was formed by the repeated addition of HCl in
Ph, Ser-NH), 7.40–7.43 (m, 3 H, Fmoc, Ser-NH), 7.69–7.70 (m, 2 dioxane (2 , 3ϫ146 µL per µmol of 21) and subsequent evapora-
H, Fmoc), 7.83–7.90 (m, 3 H, Fmoc/NH) ppm. 13C NMR tion. Step 2: The resulting residue was dissolved in DCM/DMF
mixture was stirred for 10 h at 20 °C before being evaporated. The
4806
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 4801–4809

Syntheses of Triostin A Antibiotic
(9:1, 200 µL per µmol of 21) at 0 °C and, whilst stirring, was treated
sequentially with HOAt (2 equiv.), NMM (3 equiv.), DIC
(10 equiv.), and the nucleobase acetic acid derivatives R1-COOH
(2 equiv.) under argon. After being stirred for 1 h, the mixture was
reacted at room temperature for 72 h, diluted with ethyl acetate
(700 µL per µmol of 21), washed with water (350 µL per µmol of
21), and concentrated in vacuo. Step 3: To the resulting residue,
20% piperidine in DMF (75 µL per µmol of 21) was added, and
the mixture was stirred for 20 min at 25 °C before the volatiles were
removed in vacuo. Again, 20% piperidine in DMF (75 µL per µmol
of 21) was added, the mixture was stirred for another 15 min, and
the solvents were evaporated before the residue was coevaporated
subsequently with DMF (3ϫ146 µL per µmol of 21) and toluene
(2ϫ146 µL per µmol of 21). The resulting solid was washed with
Et2O (2ϫ146 µL per µmol of 21) and dried thoroughly. Step 4:
The resulting residue was dissolved in DCM/DMF (9:1, 200 µL per
µmol of 21) at 0 °C and, whilst stirring, was treated sequentially
with HOAt (2 equiv.), NMM (3 equiv.), DIC (10 equiv.), and the
nucleobase amino acid derivative R2-COOH (2 equiv.) under ar-
gon. After being stirred for 1 h, the mixture was reacted at room
temperature for 72 h, diluted with ethyl acetate (700 µL per µmol
of 21), washed with water (350 µL per µmol of 21), and concen-
trated in vacuo. Step 5: The residue was treated with thioanisole
(11 µL per µmol of 21) and TFA (110 µL per µmol of 21) and
stirred for 10 h at 25 °C, before being concentrated in vacuo.
47.2 (Ala-CHα), 51.6 (Ser-CHα), 53.7 (Cys-CHα), 61.9 (Val-CHα),
64.8 (Ser-CH2β), 129.5 (quin-C8), 129.6 (quin-C6), 131.1 (quin-
C5), 132.0 (quin-C8), 140.3 (quin-C9), 142.6 (quin-C10), 143.6
(quin-C3), 143.9 (quin-C2), 163.7, 167.9, 169.3, 170.4, 172.8 (Val-
CO, Ala-CO, Cys-CO, quin-CO) ppm. HRMS (ESI): calcd. for
C50H63N12O12S2 [M + H]+ 1087.4124; found 1087.4123.
[N-(Quinoxaline-9-carbonyl)-
D
-Ser-
L
-Ala-N-Me-
L
-Cys-N-Me-
L-
Val]-[N-(guanin-9-yl acetyl)- -Ser-
D
L
-Ala-N-Me-
L
-Cys-N-Me-L-Val]
(Serine Hydroxy)-Dilactone Disulfide (2): R1-COOH = 2-quinox-
alinecarboxylic acid, R2-COOH = [2-amino-6-(benzyloxy)purin-9-
yl]acetic acid. The resulting residue obtained, after step 5 of the
general method of preparation, from the starting 21 (20 mg,
17.7 µmol) was dissolved in acetonitrile/water, filtered, and purified
by HPLC (20–90% eluent B in 30 min, tR = 16.4 min) to afford the
pale yellow solid 2 (3.1 mg, 16 %). 1H NMR (600 MHz, [D6]-
3
DMSO, 25 °C): δ = 0.84 (d, JH,H = 7 Hz, 3 H, Val-CH3), 0.94 (d,
3JH,H = 7 Hz, 3 H, Val-CH3), 0.96 (d, 3JH,H = 7 Hz, 3 H, Val-CH3),
3
3
0.99 (d, JH,H = 7 Hz, 3 H, Val-CH3), 1.27 (d, JH,H = 7 Hz, 3 H,
3
Ala-CH3), 1.37 (d, JH,H = 7 Hz, 3 H, Ala-CH3), 2.19–2.25 (m, 1
H, Val-Hβ), 2.31 (d, JH,H = 14 Hz, 2 H, Cys-Hβ), 2.37–2.40 (m,
2
1 H, Val-Hβ), 2.80 (s, 3 H, Val-NCH3/Cys-NCH3), 2.84 (s, 3 H,
Val-NCH3/Cys-NCH3), 3.1 (s, 3 H, Val-NCH3/Cys-NCH3), 3.33 (s,
3 H, Val-NCH3/Cys-NCH3), 3.52–3.58 (m, 2 H, Cys-Hβ), 4.09 (dd,
2
3JH,H = 4, JH,H = 11 Hz, 1 H, Ser-Hβ), 4.41–4.48 (m, 2 H, Ser-
Hβ) 4.50–4.56 (m, 2 H, Ala-Hα), 4.61–4.63 (m, 1 H, Ser-Hα), 4.72
3
2
(d, JH,H = 11 Hz, 1 H, Val-Hα), 4.78 (d, JH,H = 17 Hz, 1 H,
Triostin A (1): R1-COOH = 2-quinoxalinecarboxylic acid, R2-
COOH = 2-quinoxalinecarboxylic acid. The resulting residue after
step 4 of the general procedure, from the starting 21 (20 mg,
17.7 µmol) was dissolved in acetonitrile/water, filtered, and purified
by HPLC (35–100% eluent B in 30 min, tR = 17.4 min) to afford 1
3
2
acetyl-CH2), 4.84 (d, JH,H = 11 Hz, 1 H, Val-Hα), 4.85 (d, JH,H
= 17 Hz, 1 H, acetyl-CH2), 4.90–4.93 (m, 1 H, Ser-Hα), 6.13–6.16
(m, 2 H, Cys-Hα), 6.34 (br. s, 2 H, G-NH2), 7.67 (br. s, 1 H, G-
CH8), 7.74 (dd, JH,H = 13, JH,H = 6 Hz, Ser-Hβ), 7.94–8.07 (m,
4 H, quin-H5, H6, H7, H8), 8.12 (d, JH,H = 9 Hz, 1 H Ser-NH),
2
3
3
1
(2.8 mg, 15%) as a pale yellow solid. H NMR (600 MHz, CDCl3,
3
8.19–8.28 (m, 2 H, Ala-NH), 8.56 (d, JH,H = 9 Hz, 1 H, Ser-NH),
3
25 °C): δ = (major conformer) = 0.67 (d, JH,H = 7 Hz, 1 H, Ala-
9.56 (s, 1 H, quin-H3), 10.58 (br. s, 1 H, G-NH) ppm. 13C NMR
(125 MHz, [D6]DMSO, 25 °C): δ = 15.5 (Ala-CH3), 15.8 (Ala-CH3)
19.5 (Val-CH3), 19.6 (Val-CH3), 19.7 (Val-CH3), 26.6 (Val-CHβ),
27.6 (Val-CHβ), 30.1 (Val-NCH3/Cys-NCH3), 30.2 (Val-NCH3/
Cys-NCH3), 30.7 (Val-NCH3/Cys-NCH3), 31.2 (Val-NCH3/Cys-
NCH3), 43.7 (Cys-CH2β), 45.1 (acetyl-CH2), 46.6 (Ser-CH2β), 46.8
(Ala-CHα), 50.9 (Ser-CHα), 51.4 (Ser-CHα), 56.2 (Cys-CHα), 56.4
(Cys-CHα), 62.8 (Val-CHα), 62.9 (Val-CHα), 64.7 (Ser-CH2β),
123.8 (G-C3), 126.0 (quin-C5, C6, C7, C8, minor conformer), 128.7
(quin-C5), 131.1 (quin-C7, C8), 131.7 (quin-C7, C8), 132.3 (quin-
C6), 135.2 (quin-C9), 139.4 (quin-C10), 143.1 (quin-C2), 143.6
(quin-C3), 153.6 (G-C4, C6), 156.3 (G-C2), 162.8, 167.0, 167.8,
168.2, 169.5, 169.6, 169.7, 171.1, 171.2 (Ser-CO, Val-CO, Cys-CO,
Ala-CO, acetyl-CO) ppm. HRMS (ESI): calcd. for C48H64N15O13S2
[M + H]+ 1122.2788; found 1122.2786.
CH3), 0.70 (d, 3JH,H = 7 Hz, 5 H, Ala-CH3), 1.03 (d, 3JH,H = 7 Hz,
3
6 H, Val-CH3), 1.1 (d, JH,H = 7 Hz, 6 H, Val-CH3), 2.29–2.34 (m,
2 H, Val-Hβ), 3.02 (s, 6 H, Val-NCH3/Cys-NCH3), 3.30 (s, 6 H,
3
3
Val-NCH3/Cys-NCH3), 3.31 (dd, JH,H = 7, JH,H = 16 Hz, 2 H,
3
3
Cys-Hβ), 4.40 (dd, JH,H = 8, JH,H = 15 Hz, 2 H, Cys-Hβ), 4.24
(d, 3JH,H = 10 Hz, 2 H, Val-Hα), 4.56 (dd, 3JH,H = 1, 3JH,H = 11 Hz,
3
3
2 H, Ser-Hβ), 4.70 (dd, JH,H = 6, JH,H = 11 Hz, 2 H, Ser-Hβ),
5.00–5.02 (m, 2 H, Ala-Hα), 5.04–5.06 (m, 2 H, Ser-Hα), 5.70 (m,
2 H, Cys-Hα), 7.83–8.20 (m, 8 H, quin-H5, H6, H7, H8), 8.38 (d,
3JH,H = 9 Hz, 2 H, Ala-NH), 8.96 (d, 3JH,H = 9 Hz, 2 H, Ser-NH),
9.58 (s, 2 H, quin-H3) ppm. 13C NMR (125 MHz, CDCl3, 25 °C):
δ = 17.4 (Ala-CH3), 20.0 (Val-CH3), 20.6 (Val-CH3), 29.7 (Val-
CHβ), 30.0 (Val-NCH3/Cys-NCH3), 32.4 (Val-NCH3/Cys-NCH3),
39.8 (Cys-CH2β), 44.8 (Ala-CHα), 53.4 (Ser-CHα), 53.8 (Cys-
CHα), 64.6 (Ser-CH2β), 65.3 (Val-CHα), 129.5 (quin-C8), 129.7
(quin-C6), 131.4 (quin-C5), 132.1 (quin-C8), 140.2 (quin-C9), 142.6
[N-(Cytosin-1-ylacetyl)-
D-Ser-
L
-Ala-N-Me-
L
-Cys-N-Me-
L-Val]-[N-
(quin-C10), 142.7 (quin-C3), 143.7 (quin-C2), 163.9, 168.4, 170.2, (guanin-9-yl acetyl)-
D
-Ser-
L
-Ala-N-Me-
L
-Cys-N-Me-L
-Val] (Serine
170.6, 172.9 (Val-CO, Ser-CO, Ala-CO, Cys-CO, quin-CO) ppm. Hydroxy)-Dilactone Disulfide (3): R1-COOH = [(4-N-benzyloxycar-
1H NMR (600 MHz, CDCl3, 25 °C): δ (minor conformer) = 0.85 bonyl)cytosin-1-yl]acetic acid,[25] R2-COOH = [2-amino-6-(ben-
3
3
(d, JH,H = 7 Hz, 6 H, Val-CH3), 1.07 (d, JH,H = 7 Hz, 6 H, Val-
zyloxy)purin-9-yl]acetic acid. The resulting residue obtained, after
3
CH3), 1.43 (d, JH,H = 7 Hz, 6 H, Ala-CH3), 2.29–2.34 (m, 2 H, step 5 of the general method of preparation, from the starting 21
Val-Hβ), 2.96 (s, 6 H, Val-NCH3/Cys-NCH3), 3.07 (s, 6 H, Val- (20 mg, 17.7 µmol) was dissolved in acetonitrile/water, filtered, and
3
NCH3/Cys-NCH3), 3.27–3.28 (m, 4 H, Cys-Hβ), 4.46 (dd, JH,H
=
purified by HPLC (0–50% eluent B in 30 min, tR = 20 min) to
8, 3JH,H = 11 Hz, 2 H, Ser-Hβ), 4.59 (dd, 3JH,H = 2, 3JH,H = 12 Hz, afford a white solid 3 (3.4 mg, 18%). 1H NMR (600 MHz, [D6]-
3
2 H, Ser-Hβ), 4.74–4.77 (m, 2 H, Ala-Hα), 4.92–4.94 (m, 2 H, Ser- DMSO, 25 °C): δ = 0.79 (d, JH,H = 7 Hz, 3 H, Val-CH3), 0.85 (d,
Hα), 5.20 (d, JH,H = 10 Hz, 2 H, Val-Hα), 6.79 (m, 2 H, Cys-Hα), 3JH,H = 7 Hz, 3 H, Val-CH3), 0.88 (d, 3JH,H = 7 Hz, 3 H, Val-CH3),
3
3
3
3
7.26 (d, JH,H = 1 Hz, 2 H, Ala-NH), 7.83–8.20 (m, 8 H, quin-H5,
0.97 (d, JH,H = 7 Hz, 3 H, Val-CH3), 1.30 (d, JH,H = 7 Hz, 3 H,
3
3
H6, H7, H8), 8.96 (d, JH,H = 7 Hz, 2 H, Ser-NH), 9.64 (s, 2 H, Ala-CH3), 1.31 (d, JH,H = 7 Hz, 3 H, Ala-CH3), 2.09 (br. s, 1 H,
quin-H3) ppm. 13C NMR (125 MHz, CDCl3, 25 °C): δ = 17.8 (Ala-
CH3), 18.4 (Val-CH3), 20.4 (Val-CH3), 27.4 (Val-CHβ), 30.1 (Val-
NCH3/Cys-NCH3), 31.1 (Val-NCH3/Cys-NCH3), 40.4 (Cys-CH2β),
Val-Hβ), 2.27–2.32 (m, 2 H, Cys-Hβ), 2.35–2.41 (m, 1 H, Val-Hβ),
2.77 (s, 3 H, Val-NCH3/Cys-NCH3), 2.79 (s, 3 H, Val-NCH3/Cys-
NCH3), 3.07 (s, 3 H, Val-NCH3/Cys-NCH3), 3.20 (s, 3 H, Val-
Eur. J. Org. Chem. 2009, 4801–4809
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4807

A. Kumar Ray, U. Diederichsen
FULL PAPER
2
NCH3/Cys-NCH3), 3.51 (d, JH,H = 12 Hz, 1 H, Cys-Hβ), 3.53 (d,
C2), 156.0 (T-C4), 165.9, 167.3, 168.2, 168.6, 168.6, 168.8, 170.9,
170.9, 171.2, 172.1, 172.4 (Ser-CO, Val-CO, Cys-CO, Ala-CO, ace-
tyl-CO, COOH) ppm. HRMS (ESI): calcd. for C44H61N15O14S2 [M
3JH,H = 12 Hz, 1 H, Cys-Hβ), 4.09 (dd, J = 4, JH,H = 11 Hz, 2
3
2
H, Ser-Hβ), 4.43–4.47 (m, 3 H, Ser-Hβ, acetyl-CH2), 4.48–4.50 (m,
2 H, Ala-Hα), 4.58–4.59 (m, 1 H, Ser-Hα), 4.63–4.64 (m, 1 H, Ser- + H]+ 1116.4352; found 1116.4351.
3
Hα), 4.68–4.70 (m, 1 H, acetyl-CH2), 4.72 (d, JH,H = 11 Hz, 2 H,
[N-(Adenin-9-ylacetyl)-
D
-Ser-
L
L
-Ala-N-Me-
L
-Cys-N-Me-
L-Val]-[N-
2
2
Val-Hα), 4.82 (d, JH,H = 17 Hz, 1 H, acetyl-CH2), 4.87 (d, JH,H
(guanin-9-yl acetyl)- -Ser-
D
-Ala-N-Me-L-Cys-N-Me-L-Val] (Serine-
= 17 Hz, 1 H, acetyl-CH2), 6.00 (d, 3JH,H = 7 Hz, 1 H, C-H5), 6.09
Hydroxy)-Dilactone Disulfide (5): R1-COOH = (N6-(Z)-adenin-9-
yl)acetic acid, R2-COOH = [2-amino-6-(benzyloxy)purin-9-yl]ace-
tic acid. The resulting residue obtained, after step 5 of the general
procedure, starting with 21 (20 mg, 17.7 µmol), was dissolved in
acetonitrile/water, filtered, and purified by HPLC (10–70% eluent
B in 30 min, tR = 13.2 min) to afford 5 as a white solid (3.86 mg,
3
3
3
(dd, JH,H = 3, JH,H = 11 Hz, 1 H, Cys-Hα), 6.19 (dd, JH,H = 4,
3JH,H = 11 Hz, 1 H, Cys-Hα), 6.53 (br. s, 2 H, G-NH2), 7.56 (d,
3JH,H = 6 Hz, 1 H, Ala-NH), 7.76 (br. s, 1 H, G-CH8), 7.82 (d,
3JH,H = 7 Hz, 1 H, Ala-NH), 7.86 (d, JH,H = 7 Hz, 1 H, C-H6),
3
3
8.26 (d, JH,H = 8 Hz, 1 H, Ser-NH), 8.50 (br. s, 1 H, Ser-NH),
11.18 (br. s, 1 H, G-NH) ppm. 13C NMR (125 MHz, [D6]DMSO,
25 °C): δ = 15.7 (Ala-CH3), 15.9 (Ala-CH3), 19.4 (Val-CH3), 19.5
(Val-CH3), 19.6 (Val-CH3), 19.7 (Val-CH3), 26.6 (Cys-CH2β, Val-
CHβ), 26.8 (Cys-CH2β, Val-CHβ), 30.1 (Val-NCH3/Cys-NCH3),
30.8 (Val-NCH3/Cys-NCH3), 42.5 (Cys-CH2β), 45.9 (acetyl-CH2),
46.4 (Ala-CHα), 46.5 (Ala-CHα), 51.4 (Ser-CHα), 51.5 (Ser-CHα),
51.9, 55.8 (Cys-CHα), 56.0 (Cys-CHα), 62.8 (Val-CHα, Val-CHβ),
62.9 (Val-CHα, Val-CHβ), 64.3 (Ser-CH2β), 64.4 (Ser-CH2β), 93.5
(C-C5), 115.8 (G-C5), 138.3 (C-C6), 151.5 (G-C8), 153.8 (G-C6),
156.9 (G-C2), 157.9 (C-C2), 158.2 (G-C4), 167.0 (C-C4), 167.4,
167.5, 167.7, 169.4, 169.5, 169.6, 169.8, 170.9, 171.3 (Ser-CO, Val-
CO, Cys-CO, Ala-CO, acetyl-CO) ppm. HRMS (ESI): calcd. for
C45H65N16O14S2 [M + H]+ 1117.2387; found 1117.2385.
1
3
19%). H NMR (600 MHz, [D6]DMSO, 25 °C): δ = 0.85 (d, JH,H
= 6 Hz, 3 H, Val-CH3), 0.86 (d, 3JH,H = 6 Hz, 3 H, Val-CH3), 0.95
3
3
(d, JH,H = 3 Hz, 3 H, Val-CH3), 0.96 (d, JH,H = 3 Hz, 3 H, Val-
3
CH3), 1.34 (d, JH,H = 7 Hz, 6 H, Ala-CH3), 2.28–2.30 (m, 2 H,
Val-Hβ), 2.79 (s, 6 H, Val-NCH3/Cys-NCH3), 3.12 (s, 3 H, Val-
NCH3/Cys-NCH3), 3.16 (s, 3 H, Val-NCH3/Cys-NCH3), 3.50 (d,
3
3JH,H = 12 Hz, 2 H, Cys-Hβ), 3.52 (d, JH,H = 12 Hz, 2 H, Cys-
3
2
Hβ), 4.11 (d, JH,H = 4, JH,H = 11 Hz, 2 H, Ser-Hβ), 4.41–4.45
(m, 2 H, Ala-Hα), 4.41–4.45 (m, 2 H, Ser-Hβ), 4.63–4.65 (m, 2 H,
3
3
Ser-Hα), 4.71 (d, JH,H = 10 Hz, 1 H, Val-Hα), 4.73 (d, JH,H
=
2
10 Hz, 1 H, Val-Hα), 4.83 (d, JH,H = 17 Hz, 1 H, acetyl-CH2),
2
2
4.91 (d, JH,H = 17 Hz, 1 H, acetyl-CH2), 5.08 (d, JH,H = 17 Hz,
2
1 H, acetyl-CH2), 5.19 (d, JH,H = 17 Hz, 1 H, acetyl-CH2), 6.10
3
2
3
(dd, JH,H = 3, JH,H = 11 Hz, 1 H, Cys-Hα), 6.14 (dd, JH,H = 3,
2JH,H = 11 Hz, 1 H, Cys-Hα), 6.46 (br. s, 2 H, G-NH2), 7.78 (d,
3JH,H = 6 Hz, 1 H, Ala-NH), 7.79 (d, 3JH,H = 6 Hz, 1 H, Ala-NH),
7.86 (br. s, 1 H, G-CH8), 8.13–8.18 (br., 2 H, A-NH2), 8.27 (s, 1
H, A-H2/H8), 8.28 (s, 1 H, A-H2/H8), 8.26–8.30 (1 H, Ser-NH),
8.56 (d, 3JH,H = 9 Hz, 1 H, Ser-NH), 10.73 (br. s, 1 H, G-NH) ppm.
13C NMR (125 MHz, [D6]DMSO, 25 °C): δ = 15.8 (Ala-CH3), 19.4
(Val-CH3), 19.5 (Val-CH3), 19.6 (Val-CH3), 23.2, 26.6 (Val-CHβ),
30.1 (Val-NCH3/Cys-NCH3), 30.8 (Val-NCH3/Cys-NCH3), 30.9
(Val-NCH3/Cys-NCH3), 42.6 (Cys-CH2β), 45.2 (acetyl-CH2), 45.5
(acetyl-CH2), 46.6 (Ala-CHα), 51.4 (Ser-CHα), 51.6 (Ser-CHα),
56.2 (Cys-CHα), 62.9 (Val-CHα), 63.1 (Val-CHα), 64.6 (Ser-CH2β),
64.7 (Ser-CH2β), 117.9 (G-C5), 121.7 (A-C5), 138.0 (G-C8), 143.2
(A-C2/C8), 149.2 (A-C2/C8), 151.2 (A-C6, G-C6), 153.8 (G-C2),
156.2 (A-C4), 158.2 (G-C4), 166.8, 166.9, 167.8, 167.9, 169.5, 169.8,
169.8, 171.0 (Ser-CO, Val-CO, Cys-CO, Ala-CO, acetyl-CO) ppm.
HRMS (ESI): calcd. for C44H61N18O13S2 [M + H]+ 1141.4416;
found 1141.4414.
[N-(Thymin-1-ylacetyl)-
D
-Ser-
L
-Ala-N-Me-
L
-Cys-N-Me-
L-Val]-[N-
(adenin-9-yl acetyl)- -Ser-
D
L-Ala-N-Me-
L-Cys-N-Me-L
-Val] (Serine-
Hydroxy)-Dilactone Disulfide (4): R1-COOH = (thymin-1-yl)acetic
acid, R2-COOH = [N6-(Z)-adenin-9-yl]acetic acid. The resulting
residue obtained, after step 5 of the general method of preparation
starting with 21 (20 mg, 17.7 µmol), was dissolved in acetonitrile/
water, filtered, and purified by HPLC (0–70% eluent B in 30 min,
1
tR = 20 min) to afford the white solid 4 (4.12 mg, 21%). H NMR
(600 MHz, CD3CN, 25 °C): δ = 0.83 (d, JH,H = 7 Hz, 3 H, Val-
3
CH3), 0.84 (d, 3JH,H = 7 Hz, 3 H, Val-CH3), 0.95 (d, 3JH,H = 7 Hz,
3 H, Val-CH3), 0.97 (d, 3JH,H = 7 Hz, 3 H, Val-CH3), 1.22 (d, 3JH,H
= 7 Hz, 3 H, Ala-CH3), 1.38 (d, 3JH,H = 7 Hz, 3 H, Ala-CH3), 1.89
(d, 3JH,H = 1 Hz, 3 H, Thy-CH3), 1.98–2.02 (m, 1 H, Val-Hβ), 2.05–
2.08 (m, 1 H, Val-Hβ), 2.23–2.27 (m, 2 H, Cys-Hβ), 2.77 (s, 3 H,
Val-NCH3/Cys-NCH3), 2.79 (s, 3 H, Val-NCH3/Cys-NCH3), 2.87
(s, 3 H, Val-NCH3/Cys-NCH3), 3.09 (s, 3 H, Val-NCH3/Cys-
3
2
NCH3), 3.47–3.53 (m, 2 H, Cys-Hβ), 4.11 (dd, JH,H = 4, JH,H
=
12 Hz, 1 H, Ser-Hβ), 4.20 (dd, 3JH,H = 4, 3JH,H = 12 Hz, 1 H, Ser-
Supporting Information (see also the footnote on the first page of
this article): Temperature-dependent UV data for the interaction of
triostin A and respective analogs with DNA 23. Experimental and
analytical details for the synthesis of 10–17, 6, and 7. 1H NMR
spectra for 18–21 and 1–5.
Hβ), 4.36 (s, 2 H, T-CH2), 4.44–4.49 (m, 2 H, Ala-Hα), 4.56–4.61
3
(m, 1 H, Ser-Hα), 4.56–4.61 (m, 2 H, Ser-Hβ), 4.67 (d, JH,H
=
=
3
11 Hz, 1 H, Val-Hα), 4.68–4.69 (m, 1 H, Ser-Hα), 4.74 (d, JH,H
2
11 Hz, 1 H, Val-Hα), 4.86 (d, JH,H = 17 Hz, 1 H, A-CH2), 5.12
(d, 2JH,H = 17 Hz, 1 H, A-CH2), 6.03 (dd, 3JH,H = 4, 3JH,H = 11 Hz,
3
3
1 H, Cys-Hα), 6.15 (dd, JH,H = 4, JH,H = 11 Hz, 1 H, Cys-Hα),
3
6.62 (br. s, 2 H, A-NH2), 7.04 (d, JH,H = 7 Hz, 1 H, Ala-NH),
7.06 (d, JH,H = 7 Hz, 1 H, Ser-NH), 7.12–7.13 (br. s, 1 H, Ser-
Acknowledgments
3
3
4
NH), 7.18 (d, JH,H = 6 Hz, 1 H, Ala-NH), 7.36 (d, JH,H = 1 Hz,
1 H, Thy-H6), 8.19 (s, 1 H, A-H2/H8), 8.33 (s, 1 H, A-H2/H8),
11.96 (br. s, 1 H, Thy-NH) ppm. 13C NMR (125 MHz, CD3CN,
25 °C): δ = 12.3 (Thy-CH3), 16.4 (Ala-CH3), 16.9 (Ala-CH3), 20.0
(Val-CH3), 20.1 (Val-CH3), 20.3 (Val-CH3), 28.1 (Val-CHβ), 28.3
(Val-CHβ), 30.9 (Val-NCH3/Cys-NCH3), 31.1 (Val-NCH3/Cys-
NCH3), 31.5 (Val-NCH3/Cys-NCH3), 31.6 (Val-NCH3/Cys-
NCH3), 47.7 (Ala-CHα), 47.8 (Ala-CHα), 48.3 (A-CH2), 52.5 (Ser-
CHα), 52.6 (Ser-CHα), 52.9 (Ser-CH2β), 53.4, (T-CH2), 57.0 (Cys-
CHα), 57.2 (Cys-CHα), 64.3 (Val-CHα), 64.4 (Val-CHα), 65.0 (Ser-
CH2β), 66.2 (Ser-CH2β), 111.9 (A-C3), 128.0 (T-C5), 129.1 (T-C6),
130.4 (A-C8), 131.2 (A-C4), 142.8 (A-C6), 150.9 (T-C2), 153.7 (A-
We gratefully acknowledge the financial support of the Deutsche
Forschungsgemeinschaft (DFG) (Di 542 7-1).
[1] a) J. Shoji, K. Katagiri, J. Antibiot. 1961, 14, 335–339; b) H.
Otsuka, J. Shoji, Tetrahedron 1965, 21, 2931–2938.
[2] a) H. Otsuka, J. Shoji, K. Kawano, Y. Kyogoku, J. Antibiot.
1976, 29, 107–110; b) M. Shin, K. Inouye, H. Otsuka, Bull.
Chem. Soc. Jpn. 1978, 51, 1501–1506.
[3] a) S. Dawson, J. P. Malkinson, D. Paumier, M. Searcey, Nat.
Prod. Rep. 2007, 24, 109–126; b) D. G. Martin, S. A. Mizsak,
C. Biles, J. C. Stewart, L. Baczynskyj, P. A. Meulman, J. Anti-
biot. 1975, 28, 332–336.
4808
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 4801–4809

Syntheses of Triostin A Antibiotic
[4]
[16] M. Shin, K. Inouye, H. Otsuka, Bull. Chem. Soc. Jpn. 1984,
57, 2203–2210.
[17] J. T. Puche, E. Marcucci, M. Fermin, N. B. Puxan, F. Albericio,
Chem. Eur. J. 2008, 14, 4475–4478.
[18] D. L. Boger, J. K. Lee, J. Org. Chem. 2000, 65, 5996–6000.
[19] J. P. Malkinson, M. K. Anim, M. Zloh, M. Searcey, A. J.
Hampshire, K. R. Fox, J. Org. Chem. 2005, 70, 7654–7661.
[20] G. Asche, H. Kunz, H. Nar, H. Koppen, H. Briem, K. H. Pook,
P. W. Schiller, N. N. Chung, C. Lemieux, F. Esser, J. Pept. Res.
1998, 51, 323–336.
[21] J. M. Caba, I. M. Rodriguez, I. Manzanares, E. Giralt, F. Al-
bericio, J. Org. Chem. 2001, 66, 7568–7574.
[22] D. L. Boger, S. Ichikawa, W. C. Tse, M. P. Hedrick, Q. Jin, J.
Am. Chem. Soc. 2001, 123, 561–568.
[23] N. B. Puxan, A. Fernández, J. T. Puche, E. Riego, M. Álvarez,
F. Albericio, Int. J. Pept. Res. Ther. 2007, 13, 295–306.
[24] L. A. Carpino, J. Am. Chem. Soc. 1993, 115, 4397–4398.
[25] K. L. Dueholm, M. Egholm, C. Behrens, L. Christensen, H. F.
Hansen, T. Vulpius, K. H. Petersen, R. H. Berg, P. E. Nielsen,
O. Buchardt, J. Org. Chem. 1994, 59, 5767–5773.
[26] Y. Kyogoku, N. Higuchi, M. Watanabe, K. Kawano, Biochemis-
try 1981, 20, 1959–1970.
a) A. Dell, D. H. Williams, H. R. Morris, G. A. Smith, J.
Feeney, G. C. K. Roberts, J. Am. Chem. Soc. 1975, 97, 2497–
2502; b) A. P. Praseuth, M. B. Praseuth, H. Oguri, H. Oikawa,
K. Watanabe, C. C. C. Wang, Biotechnol. Prog. 2008, 24, 134–
139; c) A. P. Praseuth, C. C. C. Wang, K. Watanabe, K. Hotta,
H. Oguri, H. Oikawa, Biotechnol. Prog. 2008, 24, 1226–1231.
F. K. Katagiri, T. Yoshida, K. Sato, Antibiotics (Ed.: J. W. Cor-
coran, F. E. Hahn), Springer-Verlag, Heidelberg 1975, vol. 3,
p. 234 and references cited therein.
[5]
[6]
[7]
F. Takusagawa, J. Antibiot. 1985, 38, 1596–1604.
A. H.-J. Wang, G. Ughetto, G. J. Quigley, T. Hakoshima, G. A.
van der Marel, J. H. van Boom, A. Rich, Science 1984, 225,
1115–1121.
G. M. Sheldrick, A. Heine, K. Schmidt-Bäse, E. Pohl, P. G.
Jones, E. Paulus, M. J. Waring, Acta Crystallogr., Sect. B 1995,
51, 987–999.
[8]
[9]
K. B. Lorenz, U. Diederichsen, J. Org. Chem. 2004, 69, 3917–
3927.
[10] B. Dietrich, U. Diederichsen, Eur. J. Org. Chem. 2005, 147–
153.
[11] T. Lindahl, B. Nyberg, Biochemistry 1972, 11, 3610–3618.
[12] T. A. Kunkel, Proc. Natl. Acad. Sci. USA 1984, 81, 1494–1498.
[13] T. Lindahl, R. D. Wood, Science 1999, 286, 1897–1905.
[14] S. Boiteux, M. Guillet, DNA Repair 2004, 3, 1–12.
[15] P. K. Chakravarty, R. K. Olsen, Tetrahedron Lett. 1978, 19,
1613–1616.
[27] Y. H. Shim, P. B. Arimondo, A. Laigle, A. Garbesi, S. Lavielle,
Org. Biomol. Chem. 2004, 2, 915–921.
[28] a) A. D. Richards, A. Rodgers, Chem. Soc. Rev. 2007, 36, 471–
483; b) C. B. Spillane, J. A. Smith, J. L. Morgan, F. R. Keene,
J. Biol. Inorg. Chem. 2007, 12, 819–824.
Received: May 13, 2009
Published Online: August 17, 2009
Eur. J. Org. Chem. 2009, 4801–4809
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4809