Intraannular savige-fontana reaction: One-step conversion of one class of monocyclic peptides into another class of bicyclic peptides

Article abstract of DOI:10.1002/chem.200701088

Cyclisation and cross-linking strategies are important for the synthesis of cyclic and bicyclic peptides. These macrolactams are of great interest due to their increased biological activity compared to linear analogues. Herein, we describe the synthesis of a cyclic peptide containing an Hpi toxicophore, reminiscent of phakellistatins and omphalotins. The first intraannular cross-linking of such a peptide is then presented: using neat TFA to catalyse a Savige-Fontana tryptathionylation, the Hpi-containing peptide is converted to a bicyclic amatoxin analogue. As such, this methodology represents an efficient cyclisation method for cross-linking peptides and exposes a heretofore unrealised relationship between two different classes of peptide natural products. This finding increases the degree of potential chemical space for library generation.

Full text of DOI:10.1002/chem.200701088

DOI: 10.1002/chem.200701088
Intraannular Savige–Fontana Reaction: One-Step Conversion of One Class of
Monocyclic Peptides into Another Class of Bicyclic Peptides
Jonathan P. May and David M. Perrin*[a]
Abstract: Cyclisation and cross-linking
strategies are important for the synthe-
sis of cyclic and bicyclic peptides.
These macrolactams are of great inter-
est due to their increased biological ac-
tivity compared to linear analogues.
Herein, we describe the synthesis of a
cyclic peptide containing an Hpi toxi-
cophore, reminiscent of phakellistatins
and omphalotins. The first intraannular
cross-linking of such a peptide is then
presented: using neat TFA to catalyse
such, this methodology represents an
efficient cyclisation method for cross-
linking peptides and exposes a hereto-
fore unrealised relationship between
two different classes of peptide natural
products. This finding increases the
degree of potential chemical space for
library generation.
a
Savige–Fontana tryptathionylation,
the Hpi-containing peptide is convert-
ed to a bicyclic amatoxin analogue. As
Keywords: cross-linking · indole ·
natural products
peptides
·
oxidation
·
Introduction
tin binds to RNA pol II,[13] whilst phalloidin binds to F-actin
(Figure 1).[14] A characteristic structural feature of the ama-
toxin and phallotoxin peptide families is the tryptathionine
linkage that effectively cross-links the main chain peptide
cycle, and which has been synthetically introduced by at
least three different approaches as part of several ongoing
investigations.
Omphalotins[1] and certain phakellistatins[2] are monocyclic
peptides with antifungal and antimitotic activities, which
were isolated from various marine organisms. These prod-
ucts, along with many others,[3–9] present a characteristic 3a-
hydroxypyrroloACHTREUNG[2,3-b]indoline (Hpi) motif, the elaboration
of which has been the subject of several synthetic investiga-
tions, including post-synthetic oxidation of the indole within
the macrolactam. For example, phakellistatin 3 (Figure 1), a
cyclic heptapeptide with anticancer activity isolated from
Phakellia carteri marine sponge[10] was synthesised from pha-
kellistatin 13 via a photo-oxidation of tryptophan. However,
yields were low and separation of the diastereomeric prod-
ucts was problematic.[2]
Amatoxins and phallotoxins comprise two classes of
chemically related bicyclic peptides (octa- and heptapeptides
respectively), which are biosynthesised by the mushroom
Amanita phalloides.[11,12] Both toxins show remarkable bio-
logical potency and bind to protein folds on their respective
targets with subnanomolar dissociation constants; a-amani-
Tryptathionine was first reported to have been introduced
by formation of cysteine sulfenyl chloride and subsequent
thiolation at the indole 2-position.[15] While this approach
was successfully employed in solution phase syntheses, there
has been but a single report of a solid-phase methodology
towards phalloidin using this chemistry to form the trypta-
thionine.[16] Moreover, this approach is laborious as it in-
volves extensive orthogonal protection strategies in conjunc-
tion with non-linear (bidirectional) peptide synthesis. Most
recently, Shuresko and Lokey, in their synthesis of phalloi-
din, reported an elegant approach to tryptathionylation, co-
incidental with I2-mediated detritylation of cysteine.[17] An
alternative and well established route to a tryptathionine is
via the Savige–Fontana reaction[18–20] in which, under acidic
conditions, the cysteine thiol reacts with the Hpi moiety at
the N-terminus of the linear peptide to afford the trypta-
thionine cross-link (Scheme 1). Concomitant with trypta-
thionine formation is the liberation of the N-terminus for
continued peptide coupling, or in the cases of amanitin and
phalloidin, macrolactamisation.
[a] Dr. J. P. May, Prof. D. M. Perrin
Department of Chemistry, University of British Columbia
2036 Main Mall, Vancouver, BC V6T 1Z1 (Canada)
Fax : (+1)604-822-2847
E-mail: dperrin@chem.ubc.ca
A priori there appears to be no biotic relationship be-
tween the Hpi-containing phakellistatins and omphalotins,
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
3404
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3404 –3409

FULL PAPER
following Hpi incorporation,
acid treatment cleaves the pro-
tecting groups and initiates
tryptophanylation of cysteine.
This method grows the peptide
from C to N terminus in a con-
ventional manner using simple
protecting group chemistry,
which is suited to solid-phase
techniques. To the best of our
knowledge, the Savige–Fonta-
na reaction has never been de-
scribed for cases where Hpi is
contained internally within the
peptide. In fact, Zanotti et al.
reported on an attempted
Savige–Fontana reaction with
an N-acetylated Hpi; no tryp-
tathionine formation was ob-
served, and hence, it was con-
cluded that the N-acylated Hpi
was resistant to acid catalysed
ring opening and tryptathiony-
Figure 1. Omphalotin D, phakellistatin 3, a-amanitin and phalloidin.
lation.[25,26] In contrast to that
finding, here we describe the
first synthesis of an intraannu-
lar tryptathionine linkage derived from an N-acylated Hpi
moiety incorporated internally within a peptide chain (see
below). In our study, an Hpi-containing tripeptide is pre-
pared and purified as a single diastereomer prior to incorpo-
ration into longer peptides for subsequent cyclisation.[27]
A
solid-phase approach was used to generate an octapeptide
with the Hpi moiety present in the middle of the peptide.
Mild cleavage conditions and N–C backbone coupling yield-
ed an octapeptide not dissimilar to a derivative of phakellis-
tatin 3. Acid treatment affords high yielding conversion to
the desired tryptathionine cross-linked bicyclic peptide.
Scheme 1. Savige–Fontana reaction in the construction of tryptathionine
linkages, where R1 =Boc or Tr, R2 =Tr.
and the tryptathionine-containing amatoxins and phallotox-
ins. Nevertheless, since an Hpi can be converted to a trypta-
thionine bridge via the Savige–Fontana reaction, a potential
synthetic relationship exists between both classes of natural
products. Such a relationship might be exploited in the crea-
tion of “superfamilies” of synthetically related libraries of
cyclic Hpi-containing peptides, which could be converted at
will to libraries of bicyclic peptides containing tryptathio-
nine bridges. The realisation of such a strategy is predicated
on: a) the ability to cleanly introduce an Hpi within the
macrocyclic peptide and b) that this Hpi would react intra-
annularly under acidic conditions with a suitably placed cys-
teine via the Savige–Fontana reaction, to afford a tryptathio-
nine bridge.
Results and Discussion
Pure H-Hpi-Gly-OMe dipeptide was prepared from Tr-Trp-
Gly-OMe via dimethyldioxirane (DMDO) oxidation and
subsequent detritylation, as previously reported.[27] Coupling
of the syn–cis diastereomer to Fmoc-Ile-OH was carried out
using
(benzotriazol-1-yloxy)-tripyrrolidino-phosphonium
hexafluorophosphate (PyBOP) to generate a fully protected
tripeptide 4 (Scheme 2). For coupling to a growing peptide
chain, selective deprotection of the methyl ester was ach-
ieved by using Me3SnOH.[28] This left the base-labile Fmoc
moiety untouched. Following purification, [syn–cis]-Fmoc-
Ile-Hpi-Gly-OH 5 was ready for solid-phase synthesis.
Despite the potential link between macrolactams contain-
ing an Hpi and those containing a tryptathionine, to date
the Savige–Fontana reaction has only ever been described in
cases where the Hpi moiety is found as the N-terminal resi-
due in a solution-phase peptide synthesis.[21–24] Immediately
The solid-phase chemistry was carried out using a 2-chlor-
otrityl resin due to its high acid lability (Scheme 3).[29]
Fmoc-protected amino acids were used for the synthesis of
the desired octapeptide (relevant side chains protected with
trityl). Glycine was chosen as the C-terminal residue (at po-
Chem. Eur. J. 2008, 14, 3404 –3409
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3405

D. M. Perrin and J. P. May
just 1.2 equiv of tripeptide and
HBTU, for a longer period of
1 h. Hence, the resin-tethered
octapeptide 7 was produced in
this way.
Selective
acid-mediated
cleavage from the resin with
hexafluoroisopropanol (HFIP)
released the linear octapeptide
8 with trityl protection on the
cysteine and asparagine resi-
dues.[30] Macrocyclisation was
performed using PyBOP in
DMF/DIPEA for 18 h with a
yield of 40–60% for macrocyc-
lisation and an overall yield of
10–16% based upon the resin
loading, which is reasonable
for an octapeptide solid-phase
synthesis followed by a macro-
cyclic N to C coupling of this
type. Hence a cyclic peptide 9
containing an Hpi moiety was
synthesised. TFA treatment of
this compound dispatched the
remaining trityl groups and in-
duced the Savige–Fontana re-
action to afford a tryptathio-
nine linkage in place of the
cysteine and Hpi residues. This
reaction proceeded in good
yield and generated essentially
one major product as observed
by HPLC (see Supporting In-
formation). Characterisation of
this compound confirmed that
this was indeed the desired bi-
cyclic amatoxin derivative 10
(Pro2-Ile3-S-deoxo-amanina-
mide).[31–33] The yield is of par-
ticular note, despite the rela-
tively harsh conditions, HPLC
traces show almost 90% purity
in the crude mixture.
Scheme 2. Synthesis of [syn–cis]-Fmoc-Ile-Hpi-Gly-OH (5). i) DCC, HOBt, NEt3, CH2Cl2, RT, 18 h; ii)
DMDO/acetone, CH2Cl2, ꢀ708C, 1 h (70%); iii) HFIP, CH2Cl2, RT, 15 min (2 diastereomers 70%); iv) Fmoc-
Ile-OH, PyBOP, DIPEA, DMF, RT, 5 h (40%); v) Me3SnOH, CH2ClCH2Cl, 808C, 12 h (92%).
Scheme 3. Synthesis of Pro2-Ile3-S-deoxo-amaninamide (10). i) deprotection (a) and coupling (b) conditions
were used in all cases except for addition of the tripeptide 5 where coupling conditions (c) were used: a) Piper-
idine, DMF, RT, 10 min; b) Fmoc-AA-OH (4 equiv), HBTU (4 equiv), DIPEA, DMF, RT, 20 min; c) Fmoc-
Ile-Hpi-Gly-OH, HBTU, DIPEA, DMF, RT, 1 h; ii) HFIP/CH2Cl2 1:4, RT, 10 min; iii) PyBOP, DIPEA, DMF,
RT, 18 h (10–16% from 6); iv) TFA, RT, 5 h (59%).
As the anti–cis form of 9 did
not cleanly undergo tryptathio-
nylation, the reactivity of both
sition 6 of amatoxins) to obviate any epimerisation, possible
with other amino acids during cleavage or macrolactamisa-
tion. The resin was functionalised with Fmoc-Gly-OH in
N,N-diisopropyldiethylamine (DIPEA)/CH2Cl2 for 2 h. De-
protection was effected with piperidine/DMF (20%) for 10
minutes and all couplings were performed with Fmoc-AA-
OH (4 equiv), 2-(1H-benzatriazole-1-yl)-1,1,3,3-tetramethy-
luronium hexafluorophosphate (HBTU) (4 equiv) in DMF/
DIPEA (0.5%) for 20 min. The only exception to this was
coupling of the Hpi tripeptide 5 which was performed with
Hpi diastereomers was investigated in the context of Fmoc-
Gly-Hpi-Gly-Cys(Tr)-OMe to verify that thioetherification
would ensue with related substrates, irrespective of Hpi ste-
reochemistry (syn–cis or anti–cis). Hence, [anti–cis]-Fmoc-
Gly-Hpi-Gly-OMe was prepared as previously reported.[27]
Following selective deprotection of the C-terminus and cou-
pling of H-Cys(Tr)-OMe with HBTU, [anti-cis]-Fmoc-Gly-
Hpi-Gly-Cys(Tr)-OMe (12) was afforded in good yield
(Scheme 4). This linear tetrapeptide was then treated with
neat TFA at room temperature for 4 h to yield the desired
3406
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3404 –3409

Monocyclic Peptides
FULL PAPER
distinct products with greatly
differing CD spectra, which
are due to an epimerisation of
Ile to d-allo-Ile and not atro-
pisomers,[33] as previously dis-
cussed for amanitin.[31,32] Of
note, similar conformational
concerns have been raised with
regards
to
phalloidin.[16]
Herein, we created an amatox-
in by deliberately choosing a
linear peptide synthetic route
with Gly at the C terminus for
macrolactamisation, thus obvi-
ating any possibility of epimer-
isation. Once cyclised, acidifi-
cation afforded the tryptathio-
nine bridge within a single bi-
cyclic amanitin derivative ex-
hibiting identical MS/HPLC
characteristics to Pro2-Ile3-S-
deoxo-amaninamide.[33]
Scheme 4. Synthesis of [anti–cis]-Fmoc-Gly-Hpi-Gly-Cys(Tr)-OMe (13). i) Fmoc-Gly-OH, PyBOP, DIPEA,
DMF, RT, 8 h; ii) Me3SnOH, CH2ClCH2Cl, 808C, 12 h; iii) H-Cys(Tr)-OMe, HBTU, DIPEA, DMF, RT, 3 h (2
steps 76%); iv) TFA, RT, 4 h (42%).
Of note, Shuresko and
Lokey published a formidable
cyclic tryptathionine 13. UV absorbance of this product
showed an overlap of the Fmoc group with the characteristic
absorption of the tryptathionine. A small amount of this was
treated with piperidine to remove the protecting group and
a clean tryptathionine UV absorbance spectrum was ob-
served (lmax =290 nm, see Supporting Information). This
finding suggests that the steric bulk of the Ile, and/or the
conformational nature of the [anti–cis]-Hpi within a macro-
cyclic ring, prevented clean tryptathionylation in the case of
the macrolactam. Although both N-glycylated Hpi anti–cis
and syn–cis diastereomers, cleanly underwent tryptathiony-
lation, the extent to which various steric/conformational ef-
fects might limit the generality of this reaction was not fur-
ther investigated.
approach to the intraannular oxidative thioetherification of
tryptophan using I2 in DMF. Their findings complement the
results presented here. In contrast to their elegant work, the
results herein allow access to several classes of natural prod-
ucts (omphalotins, phakellistatins, amatoxins and phallotox-
ins) from minor modifications to the same library of com-
pounds (phallotoxins have a deletion at Asx relative to ama-
toxins). This would appear to be significant in order to
access a large area of chemical space with a directed combi-
natorial library approach. With this methodology one can
imagine constructing a library of phakellistatins and related
peptides that can be easily transformed into synthetic ama-
toxins, phallotoxins and related peptides, all from essentially
the same linear resin-bound library of hepta- and octapepti-
des. The generally high yielding conversions we observe
could be of particular importance in producing libraries of
small quantities of these bicyclic peptides where high yields
are necessary for sufficient activity to be observed in a sub-
sequent screen. The construction of such libraries is current-
ly being examined and will be reported in due course.
Conclusion
The salient points of this work are as follows: This work ex-
pands the reactivity of the Savige–Fontana reaction to intra-
annular systems, which did not appear possible according to
earlier reports. As such, a circularly permuted approach to
synthesizing tryptathionine bridged peptides is now feasible,
in accordance with a synthetic strategy that will generally
allow for macrolactamisation at almost any chosen amino-
acid pair. Consequently, one avoids coupling at sterically
constrained amino acids (e.g. Ile) that are prone to epimeri-
sation. Indeed, the standard linear approach to synthesizing
Experimental Section
General experimental details: All reagents were obtained from commer-
cial sources and used without further purification unless otherwise stated.
All amino acids were l-amino-acids unless otherwise stated. Thin layer
chromatography was performed on Merck 60 F254 silica plates and
viewed under UV radiation (254 nm), or staining with ninhydrin with
gentle heating. Silicycle silica gel (230–400 mesh) was used for flash
column chromatography. Further purification of peptides was performed
using an Agilent 1100 HPLC. HPLC was carried out using a reverse-
phase C18 column (8.5”15”1.5 mm) with gradients combining buffer A
and B. Buffer A=(H2O + 0.1% TFA); Buffer B=(MeCN + 0.05%
amanitin starts with an octapeptide that undergoes
a
Savige–Fontana cyclisation to install the tryptathionine and
unmasks the terminal NH2 of the tryptathionine, followed
by macrocyclisation between Trp and Ile. In our hands, this
approach has consistently provided two chromatographically
Chem. Eur. J. 2008, 14, 3404 –3409
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3407

D. M. Perrin and J. P. May
TFA). For all HPLC peptide purifications a photodiode array detector
was used between 200 and 400 nm. 1H NMR and 13C NMR were per-
formed at 300/400/600 MHz and 75/100/150 MHz, respectively. Chemical
shifts for all spectra were reported in parts per million and referenced to
the residual protic solvent peak. Mass spectrometry data was acquired on
a Waters-Micromass ZQ system, using positive or negative ionisation
mode in MeOH or MeCN.
10 mg. This residue was dissolved in dry DMF and PyBOP and DIPEA
were added. The mixture was stirred for 18 h at room temperature. Sol-
vent was removed in vacuo and the mixture was purified by silica gel
column chromatography (CH2Cl2/MeOH 0!10%). Pure fractions were
combined to yield a white solid (3.0 mg, 16% based upon resin loading).
Rf =0.5 (CH2Cl2/MeOH 9:1); 1H NMR (600 MHz, [D6]DMSO): d=9.52
(s, 1H, NHindole), 8.78 (d, J=6.6 Hz, 1H, CONHCys), 8.72 (dd, J=4.6,
3.3 Hz, 1H, CONHGly), 8.65 (d, J=6.6 Hz, 1H, CONHIle), 8.51 (dd, J=
8.0, 3.5 Hz, 1H, CONHGly), 8.01 (s, 1H, NHAsn), 7.36–7.10 (m, 33H,
ArHTr,indole, CONHAsn,Ile), 7.04 (t, J=7.7 Hz, 1H, ArHindole), 6.64 (t, J=
7.7 Hz, 1H, ArHindole), 6.46 (d, J=7.7 Hz, 1H, ArHindole), 5.97 (s, 1H,
OH), 5.22 (s, 1H, CH8a), 4.43–4.30 (m, 2H, CHaIle,aAsn), 4.12 (dd, J=17.0,
8.0 Hz, 1H, CHaGly), 4.03 (dd, J=8.6, 4.6 Hz, 1H, CHaPro), 3.97 (dd, J=
10.1, 8.8 Hz, 1H, CHaIle), 3.81 (dd, J=11.3, 5.5 Hz, 1H, CHaHpi), 3.60 (dd,
J=15.1, 3.3 Hz, 1H, CHaGly), 3.43–3.32 (m, 2H, CHaCys,aGly), 3.19 (dd, J=
17.0, 3.5 Hz, 1H, CHaGly), 2.90 (t, J=11.4 Hz, 1H, CHbCys), 2.75 (t, J=
12.3 Hz, 1H, CHbAsn), 2.65–2.47 (m, 3H, CHdPro,bCys,bAsn), 2.43 (dd, J=12.1,
5.5 Hz, 1H, CHbHpi), 2.28 (dd, J=12.1, 11.3 Hz, 1H, CHbHpi), 2.00–1.88
(m, 2H, CHbIle,bPro), 1.78–1.72 (m, 1H, CHdPro), 1.66–1.58 (m, 1H, CHbIle),
1.54–1.46 (m, 1H, CHbPro), 1.41–1.34 (m, 1H, CHgIle), 1.24–1.05 (m, 2H,
CHgIle,gPro), 1.05 (d, J=7.0 Hz, 3H, CH3g’Ile), 1.02–0.72 (m, 2H, CHgPro,gIle),
0.84–0.69 (m, 6H, CH3g’Ile,dIle), 0.64–0.55 (m, 1H, CHgIle), 0.35 ppm (t, J=
7.4 Hz, 3H, CH3dIle); 13C NMR (150 MHz, [D6]DMSO): d=172.1, 171.5,
170.1, 169.8, 169.6, 168.9, 168.6, 168.3, 166.8, 151.7, 144.5, 144.4, 130.0,
129.1, 129.0, 128.7, 128.0, 127.3, 126.7, 126.3, 123.8, 117.8, 108.7, 85.9,
81.8, 69.3, 65.9, 60.4, 54.9, 54.0, 46.7, 46.0, 43.4, 42.2, 41.4, 40.8, 40.1, 35.5,
34.2, 30.8, 29.0, 24.9, 24.3, 24.1, 15.4, 13.8, 9.4, 8.0 ppm; ES+/MS: m/z:
1364.9 [M+Na]+; HRMS (ES+): m/z: calcd for C77H84N10O10SNa:
1363.5990 [M+Na]+, found 1363.5992.
ACHTREUNG[syn–cis]-Fmoc-Ile-Hpi-Gly-OMe (4): PyBOP (0.24 g, 0.5 mmol), [syn–
cis]-H-Hpi-Gly-OMe (0.12 g, 0.4 mmol), and DIPEA (80 mL, 0.5 mmol)
was added to a solution of Fmoc-Ile-OH (0.17 g, 0.5 mmol) in dry DMF
(20 mL). The mixture was stirred for 18 h at room temperature. On com-
pletion of the reaction the mixture was evaporated to dryness and puri-
fied by flash chromatography on silica gel (hexane/EtOAc 1:1) to yield
pure product as a white foam (0.10 g, 40%). Rf =0.5 (CH2Cl2/MeOH
9:1); 1H NMR (300 MHz, CDCl3): d=7.72 (d, J=7.1 Hz, 2H, ArHFmoc),
7.61–7.51 (m, 2H, ArHFmoc), 7.40–7.19 (m, 6H, NHCO, ArH4, ArHFmoc),
7.05 (t, J=7.6 Hz, 1H, ArH5), 6.71 (t, J=7.6 Hz, 1H, ArH6), 6.51 (d, J=
7.6 Hz, 1H, ArH7), 6.01–5.89 (m, 2H, NHCO, CH8a), 4.59 (d, J=8.8 Hz,
1H, CH2), 4.47 (dd, J=10.5, 7.3 Hz, 1H, CHFmoc), 4.35–4.05 (m, 4H,
CHFmoc, CHFmoc, CHaGly, CHaIle), 3.93 (dd, J=18.2, 4.7 Hz, 1H, CHaGly),
3.73 (s, 3H, CHOMe), 2.35 (d, J=14.0 Hz, 1H, CH3), 2.27 (dd, J=14.0,
8.8 Hz, 1H, CH3), 1.93–1.70 (m, 1H, CHbIle), 1.64–1.51 (m, 1H, CHg),
1.15 (m, 1H, CHgIle), 0.96–0.81 ppm (m, 6H, CHg’Ile, CHdIle); 13C NMR
(75 MHz, CDCl3): d=173.2 (CCONH), 172.9 (CCONH), 169.4 (CCO), 156.6
(COCONH), 146.4 (C7a), 143.6 (CFmoc), 141.2 (CFmoc), 129.6 (C3b), 129.6
(CH6), 127.7 (CHFmoc), 126.9 (CHFmoc), 124.9 (CHFmoc), 122.3 (CH4), 119.9
(CHFmoc), 119.2 (CH5), 109.9 (CH7), 88.9 (C3a), 87.2 (C8a), 67.3 (CH2Fmoc),
61.6 (CH2), 56.6 (CHFmoc), 52.4 (CH3OMe), 46.9 (CHaIle), 41.4 (CHaGly),
3
g
40.2 (CH2 ), 38.0 (CHbIle), 24.7 (CH2 ), 14.7 (CH3g’), 10.5 ppm (CH3d);
ES+/MS: m/z: 649.3 [M+Na]+; HRMS (ES+): m/z: calcd for
C35H39N4O7: 627.2819 [M+H)+]+, found 627.2796.
Amatoxin 10: TFA (5 mL) was added to 9 (3.0 mg, 0.002 mmol) and
stirred for 5 h at room temperature. The solvent was evaporated in vacuo
and the residue was redissolved in water before evaporating again. This
was repeated three times. The residue was dissolved in MeOH/H2O, fil-
tered with a 0.5 mm syringe filter then purified by HPLC. Pure product
was combined to yield a white solid residue (1.1 mg, 59%). This sample
ACHTREUNG[syn–cis]-Fmoc-Ile-Hpi-Gly-OH (5): [syn–cis]-Fmoc-Ile-Hpi-Gly-OMe
(0.10 g, 0.2 mmol) and trimethyltin hydroxide (0.14 g, 0.8 mmol) were
weighed into a dry flask and 1,2-dichloroethane (10 mL) was added. The
mixture was stirred overnight at 768C. TLC (CH2Cl2/MeOH 9:1) showed
completion of the reaction and the mixture was concentrated in vacuo.
Purification was with silica gel column chromatography (CH2Cl2/MeOH
9:1) and fractions were combined to yield pure product (0.90 g, 92%).
Rf =0.1 (CH2Cl2/MeOH 9:1); 1H NMR (400 MHz, CDCl3): d=7.75 (d,
J=7.5 Hz, 2H, ArHFmoc), 7.61 (dd, J=7.1, 4.4 Hz, 2H, ArHFmoc), 7.38
(dd, J=7.1, 6.9 Hz, 2H, ArHFmoc), 7.34–7.21 (m, 4H, NHCO, ArH4,
ArHFmoc), 7.08 (t, J=7.6 Hz, 1H, ArH5), 6.82 (d, J=9.9 Hz, 1H, NHCO),
6.76 (t, J=7.4 Hz, 1H, ArH6), 6.58 (d, J=7.9 Hz, 1H, ArH7), 5.88 (s, 1H,
CH8a), 4.65 (d, J=8.5 Hz, 1H, CH2), 4.45 (dd, J=10.1, 7.6 Hz, 1H,
CHFmoc), 4.38–4.25 (m, 3H, CHFmoc, CHaGly, CHaIle), 4.20 (t, J=7.4 Hz,
1H, CHFmoc), 3.92 (dd, J=18.1, 3.1 Hz, 1H, CHaGly), 2.45 (d, J=14.2 Hz,
1H, CH3), 2.37 (dd, J=14.2, 8.8 Hz, 1H, CH3), 1.83–1.70 (m, 1H, CHbIle),
1.69–1.53 (m, 1H, CHg), 1.37–1.13 (m, 2H, OH, CHgIle), 0.96–0.79 ppm
(m, 6H, CHg’Ile, CHdIle); 13C NMR (100 MHz, CDCl3): d=174.6 (CCONH),
172.6 (CCONH), 171.4 (CCO), 156.9 (COCONH), 146.3 (C7a), 143.7 (CFmoc),
141.3 (CFmoc), 129.8 (C3b), 129.6 (CH6), 127.8 (CHFmoc), 127.1 (CHFmoc),
125.2 (CHFmoc), 122.5 (CH4), 120.1 (CHFmoc), 119.5 (CH5), 110.1 (CH7),
89.0 (C3a), 87.5 (C8a), 67.6 (CH2Fmoc), 62.1 (CH2), 56.8 (CHFmoc), 47.1
was co-eluted with
a
reference sample of Pro2-Ile3-S-deoxo-amanin-
amide,[33] which gave an identical retention time. Rf =0.5 (CHCl3/MeOH/
water 90:13:1); HRMS (ES+): m/z: calcd for C39H54N10O9SNa: 861.3696
[M+Na]+, found 861.3694.
AHCTRE[GNU anti–cis]-Fmoc-Gly-Hpi-Gly-Cys(Tr)-OMe (12): [anti–cis]-Fmoc-Gly-
Hpi-Gly-OMe (60 mg, 0.10 mmol) was heated with Me3SnOH (5 equiv)
in dichloroethane (10 mL) as previously described.[27] On completion the
reaction was evaporated to dryness and purified on a short silica plug.
The white residue was dissolved in DMF and H-Cys(Tr)-OMe (42 mg,
0.11 mmol), HBTU (42 mg, 0.11 mmol) and DIPEA (39 mL, 0.22 mmol)
were added. The reaction was stirred for 3 h and TLC (CH2Cl2/MeOH
9:1) showed consumption of starting material. The reaction was evaporat-
ed to dryness then redissolved in CH2Cl2 and washed with citric acid
(5%), sat. NaHCO3, sat. NaCl and dried over anhydrous Na2SO4. The
residue was purified by silica chromatography (CH2Cl2/MeOH 0!10%)
to give the title compound as a white solid (73 mg, 76%). Rf =0.6
(CH2Cl2/MeOH 9:1); 1H NMR (400 MHz, [D6]DMSO): d=8.50 (d, 1H,
J=7.6 Hz, NHCO), 8.00 (d, 1H, J=7.9 Hz, NHCO), 7.88 (d, 2H, J=
7.3 Hz, ArHFmoc), 7.69 (d, 2H, J=7.6 Hz, ArHFmoc), 7.50–7.17 (m, 21H,
ArHTr, ArHFmoc, NHCO, ArHindole), 7.09 (t, 1H, J=7.3 Hz, ArHindole), 6.69
(t, 1H, J=7.3 Hz, ArHindole), 6.57 (d, 1H, J=7.9 Hz, ArHindole), 4.91 (s,
1H, CHHpi8a), 4.32–4.15 (m, 4H, CH2Fmoc, CHaGly, CHaCys), 4.05–3.86 (m,
3H, CHFmoc, CHaGly, CHaHpi), 3.69–3.20 (m, 6H, CH3OMe, CH2aGly, NH8),
3
g
(CHaIle), 41.8 (CHaGly), 40.5 (CH2 ), 37.9 (CHbIle), 24.9 (CH2 ), 14.7
(CH3g’), 10.5 ppm (CH3d); ES+/MS: m/z: 635.2 [M+Na]+; HRMS (ES+):
m/z: calcd for C34H36N4O7Na: 635.2482 [M+Na]+, found 635.2485.
Compounds 6–7: 2-Chlorotrityl resin was functionalised with Fmoc-Gly-
OH in the presence of DIPEA. After thorough washing the Fmoc-Gly-
resin was dried in vacuo (resin loading was calculated to be
1.39 mmolgꢀ1). Fmoc deprotection was effected with piperidine (20%) in
DMF for 10 min. Coupling was effected with Fmoc-AA-OH (4 equiv),
HBTU (4 equiv), DMF/DIPEA (0.5%) agitating for 20 min (all standard
l-amino acids). The tripeptide 5 (1.1 equiv) was coupled with HBTU
(1.1 equiv), DMF/DIPEA (0.5%), shaking for 1 h.
2.59–2.29 ppm (m, 4H, CHbCys
,
CHbHpi); 13C NMR (100 MHz,
[D6]DMSO): d=171.9, 171.7, 170.2, 169.7, 157.9, 149.2, 145.5, 145.3,
142.1, 132.2, 131.0, 130.5, 129.5, 129.2, 128.5, 128.4, 128.1, 126.7, 125.2,
121.5, 119.6, 110.6, 84.7, 79.2, 67.9, 67.2, 53.6, 52.9, 48.1, 47.5, 46.5, 44.7,
41.2, 34.3 ppm; ES+/MS: m/z: 938.3 [M+Na]+.
N-(9H-Fluoren-9-ylmethoxy)carbonylglycyl(2-mercapto-tryptophanyl-
glycyl-cysteine cyclic sulfide) methyl ester (13): [anti–cis]-Fmoc-Gly-Hpi-
Gly-Cys(Tr)-OMe (40 mg, 0.044 mmol) was stirred with TFA (4 mL) for
4 h at room temperature. Methanol was added and the solvent removed
in vacuo, which was repeated three times. The residue was purified on
Cyclic peptide 9: The linear precursor was cleaved from the resin using
HFIP/CH2Cl2 1:4 (10 mL) with stirring for 10 min. Solvent was removed
by evaporation and the residue was dissolved in MeOH and filtered to
remove resin. The filtrate was dried to yield a crude pale yellow residue,
3408
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3404 –3409

Monocyclic Peptides
FULL PAPER
silica gel column chromatography (CH2Cl2/MeOH 0!15%) and purified
further by HPLC to yield pure product (12 mg, 42%). Rf =0.4 (CH2Cl2/
MeOH 9:1); 1H NMR (400 MHz, CDCl3): d=8.25 (d, J=5.1 Hz, 1H),
8.12–8.05 (brs, 1H), 8.00 (d, J=7.4 Hz, 1H), 7.80 (d, J=7.4 Hz, 2H),
7.75 (d, J=7.9 Hz, 1H), 7.73–7.66 (m, 2H), 7.43–7.27 (m, 5H), 7.15 (t,
J=7.9 Hz, 1H), 7.08 (t, J=7.4 Hz, 1H), 4.82–4.78 (m, 2H), 4.41 (d, J=
6.5 Hz, 2H), 4.28 (t, J=6.5 Hz, 1H), 4.19 (dd, J=15.3, 5.7 Hz, 1H), 3.97–
3.83 (m, 2H), 3.79 (s, 3H), 3.68–3.59 (m, 1H), 3.51–3.37 (m, 2H), 3.16–
3.08 ppm (m, 2H); 13C NMR (100 MHz, CDCl3): d=174.3, 171.4, 171.2,
170.9, 158.8, 144.8, 142.2, 139.1, 128.4, 128.1, 127.8, 125.3, 123.3, 120.5,
120.3, 119.6, 117.3, 113.1, 111.5, 67.8, 53.9, 53.2, 52.9, 49.5, 44.5, 37.3,
30.3 ppm; ES+/MS: m/z: 678.1 [M+Na]+.
[13] D. A. Bushnell, P. Cramer, R. D. Kornberg, Proc. Natl. Acad. Sci.
USA 2002, 99, 1218–1222.
[14] T. Wieland, V. M. Govindan, Febs. Lett. 1974, 46, 351–353.
[15] D. E. Tupper, I. A. Pullar, J. A. Clemens, J. Fairhurst, F. C. Risius,
G. H. Timms, S. Wedley, J. Med. Chem. 1993, 36, 912–918.
[16] M. O. Anderson, A. A. Shelat, R. K. Guy, J. Org. Chem. 2005, 70,
4578–4584.
[17] L. A. Schuresko, R. S. Lokey, Angew. Chem. 2007, 119, 3617–3619;
Angew. Chem. Int. Ed. 2007, 46, 3547–3549.
[18] W. E. Savige, Aust. J. Chem. 1975, 28, 2275–2287.
[19] W. E. Savige, A. Fontana, J. Chem. Soc. Chem. Commun. 1976, 600–
601.
[20] W. E. Savige, A. Fontana, Int. J. Pept. Protein Res. 1980, 15, 102–
112.
[21] G. Zanotti, C. Birr, T. Wieland, Int. J. Pept. Protein Res. 1981, 18,
162–168.
[22] G. Zanotti, C. Mohringer, T. Wieland, Int. J. Pept. Protein Res. 1987,
30, 450–459.
[23] G. Zanotti, T. Wieland, G. Dauria, L. Paolillo, E. Trivellone, Int. J.
Pept. Protein Res. 1990, 35, 263–270.
[24] G. Zanotti, L. Falcigno, M. Saviano, G. D’auria, B. M. Bruno, T.
Campanile, L. Paolillo, Chem. Eur. J. 2001, 7, 1479–1485.
[25] G. Zanotti, C. Birr, T. Wieland, Int. J. Pept. Protein Res. 1978, 12,
204–216.
Acknowledgements
J.P.M. received a post-doctoral fellowship from The Royal Society (UK);
D.M.P. received a Junior Career Award from the Michael Smith Founda-
tion for Health Research in B.C.; this work was supported by U.B.C.
start-up funds, funding from the Michael Smith Foundation for Health
Research in B.C., and funding from the Canadian Institutes of Health
Research.
[26] Refer to Supporting Information for our study on Ac-Hpi-Gly-
Cys(Tr)-OMe (14).
[27] J. P. May, P. Fournier, J. Pellicelli, B. O. Patrick, D. M. Perrin, J. Org.
Chem. 2005, 70, 8424–8430.
[28] K. C. Nicolaou, A. A. Estrada, M. Zak, S. H. Lee, B. S. Safina,
Angew. Chem. 2005, 117, 1402–1406; Angew. Chem. Int. Ed. 2005,
44, 1378–1382.
[1] E. Buchel, U. Martini, A. Mayer, H. Anke, O. Sterner, Tetrahedron
1998, 54, 5345–5352.
[2] K. L. Greenman, D. M. Hach, D. L. Van Vranken, Org. Lett. 2004,
6, 1713–1716.
[3] P. R. Hewitt, E. Cleator, S. V. Ley, Org. Biomol. Chem. 2004, 2,
2415–2417.
[4] J. M. Roe, R. A. B. Webster, A. Ganesan, Org. Lett. 2003, 5, 2825–
2827.
[29] P. Athanassopoulos, K. Barlos, D. Gatos, O. Hatzi, C. Tzavara, Tetra-
hedron Lett. 1995, 36, 5645–5648.
[5] T. M. Kamenecka, S. J. Danishefsky, Chem. Eur. J. 2001, 7, 41–63.
[6] J. M. Schkeryantz, J. C. G. Woo, P. Siliphaivanh, K. M. Depew, S. J.
Danishefsky, J. Am. Chem. Soc. 1999, 121, 11964–11975.
[7] E. Francis, R. Rahman, S. Safe, A. Taylor, J. Chem. Soc. Perkin
Trans. 1 1972, 470–472.
[8] Y. Fukui, M. Somei, Heterocycles 2001, 55, 2055–2057.
[9] W. X. Hong, L. J. Chen, C. L. Zhong, Z. J. Yao, Org. Lett. 2006, 8,
4919–4922.
[30] R. Bollhagen, M. Schmiedberger, K. Barlos, E. Grell, J. Chem. Soc.
Chem. Commun. 1994, 2559–2560.
[31] G. Zanotti, G. Petersen, T. Wieland, Int. J. Pept. Protein Res. 1992,
40, 551–558.
[32] W. Schmitt, G. Zanotti, T. Wieland, H. Kessler, J. Am. Chem. Soc.
1996, 118, 4380–4387.
[33] J. P. May, P. Fournier, B. O. Patrick, D. M. Perrin, Chem. Eur. J.
2008, 14, in press, DOI: 10.1002/chem.200701297.
[10] G. R. Pettit, R. Tan, D. L. Herald, R. L. Cerny, M. D. Williams, J.
Org. Chem. 1994, 59, 1593–1595.
[11] T. Wieland, H. Faulstich, Experientia 1991, 47, 1186–1193.
[12] T. Wieland, Naturwissenschaften 1977, 64, 303–309.
Received: July 16, 2007
Revised: November 10, 2007
Published online: February 18, 2008
Chem. Eur. J. 2008, 14, 3404 –3409
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3409