Metal-catalysed tandem metathesis-hydrogenation reactions for the synthesis of carba analogues of cyclic peptides

Article abstract of DOI:10.1139/v05-100

Dicarba cyclic peptide analogues of the cystine-containing octapeptides octreotide. Lanreotide, and vapreotide, all known somatostatin analogues, have been synthesized via an on-resin homogeneous metal-catalysed sequence involving ruthenium-catalysed ring-closing metathesis followed by rhodium-catalysed hydrogenation.

Full text of DOI:10.1139/v05-100

875
Metal-catalysed tandem metathesis–hydrogenation
reactions for the synthesis of carba analogues of
cyclic peptides1
Amanda N. Whelan, Jomana Elaridi, Roger J. Mulder, Andrea J. Robinson, and
W. Roy Jackson
Abstract: Dicarba cyclic peptide analogues of the cystine-containing octapeptides octreotide, lanreotide, and vapreotide,
all known somatostatin analogues, have been synthesized via an on-resin homogeneous metal-catalysed sequence in-
volving ruthenium-catalysed ring-closing metathesis followed by rhodium-catalysed hydrogenation.
Key words: dicarba-analogues of cystine-containing peptides, octreotide, vapreotide, and lanreotide; tandem ring-closing
metathesis and hydrogenation.
Résumé : Des peptides cycliques dicarbonés analogues de la cystine contenant les octapeptides octréotide, lanréotide et
vapréotide, tous connus comme étant des analogues de la somatostatine, ont été synthétisés sur résine via une série de
catalyses homogènes impliquant une métathèse de fermeture de cycle catalysée par du ruthénium et une hydrogénation
catalysée par du rhodium.
Mots clés : analogues dicarbonés de la cystine contenant des peptides, octréotide, vapréotide et lanréotide; réaction tan-
dem de fermeture de cycle par métathèse et hydrogénation.
Whelan et al. 881
Introduction
the imaging of neuroendocrine tumours (9). It has also been
conjugated with 123I, 64Cu, and 99mTc (10). The 99mTc radio-
Cystine bridges are common structural motifs in naturally
occurring cyclic peptides. In some cases, as in spiruchostatin
A (1), the disulphide bridge acts as a reactive functional
group undergoing reduction within the cell to release metal-
chelating thiol groups (2, 3). In many other cases, however,
the cystine bridge serves only a skeletal, structural role,
maintaining secondary and tertiary structure and may be
replaced with a nonreducible structural mimic without sig-
nificantly affecting biological activity (4). Indeed, thioether
linkages (-S–CH2-) (5) and the all carbon (-CH2–CH2-)
bridge (6) have both been successfully implemented as
cystine (-S–S-) isosteres producing peptides with enhanced
biostability.
nuclide possesses a convenient physical half life (t1/2
=
6 min), an attractive radiation profile, is inexpensive, and
–
readily available as pertechnate (99mTcO4 ).3 Incorporation
of this isotope into cystine-containing peptides, however, is
often complicated by the need for a reductive step to gener-
ate Tc5+ for stable chelate formation (11). In the case of
octreotide, a concomitant reduction of the disulphide bridge
opens the cyclic peptide and results in the loss of receptor
binding affinity (12). This problem is not exclusive to
octreotide and also occurs during the Tc labelling of other
cyclic peptides containing cystine bridges, including the
octreotide analogues lanreotide 2a and vapreotide 3a. Given
the high desirability to label these peptides with technetium,
the replacement of the reductively labile disulphide bridge
with a carbon-based manifold has been investigated leading
to the dicarba analogues 1b, 2b, and 3b. This structural al-
teration would be inert to cellular reductases and also with-
The cyclic peptide octreotide 1a (7), a synthetic analogue
of the tetradecapeptide somatostatin, has one cystine bridge
that is remote from the receptor binding site. When labelled
with 111In, it is marketed as OctreoScan® (8) and used for
Received 14 December 2004. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 22 July 2005.
Dedicated to Professor Howard Alper in recognition of his major contribution to metal-catalysed reactions in organic synthesis.
A.N. Whelan. School of Chemistry, Monash University, Clayton, VIC 3800, Australia and CSIRO Molecular Science, Bag 10,
Clayton South, VIC 3169, Australia.
J. Elaridi and A.J. Robinson. School of Chemistry, Monash University, Clayton, VIC 3800, Australia.
R.J. Mulder. CSIRO Molecular Science, Bag 10, Clayton South, VIC 3169, Australia.
W.R. Jackson.2 Centre for Green Chemistry, Monash University, Clayton, VIC 3800, Australia.
1This article is part of a Special Issue dedicated to Professor Howard Alper.
2Corresponding author (e-mail: Roy.Jackson@sci.monash.edu.au).
399mTc has superior nuclear characteristics for imaging (t1/2 = 6.02 h and monoenergetic emission 141 keV), while 188Re physical
characteristics are ideal for therapy (t1/2 = 16.9 h and β-emission 2.11 MeV).
Can. J. Chem. 83: 875–881 (2005)
doi: 10.1139/V05-100
© 2005 NRC Canada

876
Can. J. Chem. Vol. 83, 2005
Scheme 1.
O
Ring Closing
Metathesis
Hydrogenation
H
N
H
N
N
H
H2N
O
O
NH
Grubbs catalysts (14) have previously been used to facili-
tate the formation of cyclic peptides via metathesis (15), but
only Vederas and co-workers (16), in a recent paper target-
ing oxytocin, have used bis(allyl glycine) containing pep-
tides to form dicarba mimics of the cystine bridge. We believe
that the protocol outlined in this paper demonstrates some
clear advantages over the previously published method. It is
also highly generic and can be applied to any peptide con-
taining a cystine bridge.
X
O
NH
X
O
H
N
H
N
HO
N
H
NH2
O
O
HO
HO
1 a X = S
Octreotide
1 b X = CH2 Dicarba octreotide analogue
Results and discussion
OH
Synthesis of linear peptides
The protocol described utilizes the non-proteinaceous
amino acid allyl glycine, which we have previously reported
can be conveniently generated in both enantiomeric forms,
via Rh(I)-catalysed hydrogenation of α-N-acyl dienamides
with excellent stereoselectivity (95% ee) and yield (90%)
(17). It is also commercially available, albeit at high cost.
The synthesis of dicarba-octreotide analogue 1b also re-
quires a second non-proteinaceous residue, Fmoc-protected
threoninol. This residue is readily formed by the reduction
of the protected amino acid threonine to the corresponding
alcohol, although it is also commercially available. The re-
sultant diol is functionalized as the p-carboxybenzaldehyde
acetal in preparation for attachment to the resin.
O
H
N
H
N
N
H
H2
N
O
O
NH
X
O
NH
X
O
O
H
N
H
N
H2N
N
H
NH2
O
O
HO
2 a X = S
Lanreotide
2
b X = CH2 Dicarba lanreotide analogue
The linear peptide precursor of the dicarba-octreotide ana-
logue was synthesized on Rink amide resin using a modified
solid-phase protocol described by Hsieh and co-workers
(18). The above described acetal is attached to the resin us-
ing the coupling reagents DIC and HOBt to give 4. The re-
maining sequence was constructed manually using standard
solid-phase peptide synthesis techniques, utilizing Fmoc-
protected amino acids and the coupling reagents HATU and
NMM (Scheme 2). Intermediates were carried through with-
out purification or characterization up to the octapeptide 5.
A sample of the linear peptide was obtained by cleavage
from the resin and was determined to be of >95% purity by
reverse-phase analytical chromatography and had the re-
quired molecular mass by mass spectral analysis.
OH
O
H
N
H
N
N
H
H2N
O
O
NH
X
O
NH
X
O
O
H
N
H
N
H2N
N
H
NH2
O
O
The resin-bound linear peptide precursors for the synthe-
sis of dicarba-analogues of lanreotide 2b and vapreotide 3b
were also prepared on Rink amide resin (Scheme 3). Use of
this resin allowed the direct generation of the required C-
terminal amide found in 2b and 3b upon acid cleavage from
the resin. All couplings were accomplished utilizing HATU
and NMM. After complete sequence construction, a small
sample was cleaved from the resin and the presence of the
desired peptide was confirmed by mass spectral analysis.
N
H
3 a X = S
Vapreotide
3 b X = CH2 Dicarba vapreotide analogue
stand chemical reducing agents, such as stannous chloride,
used in the preparation of Tc-labelled radiopharmaceuticals.
A highly generic, on-resin, tandem-homogeneous cataly-
sis strategy has been devised, which involves: (i) construc-
tion of the linear peptide sequence replacing cysteine
residues with allyl glycine; (ii) Ru(II)-catalysed metathesis
of the allyl groups to affect carbocycle formation; and (iii)
Rh(I)-catalysed homogeneous reduction of the newly formed
olefin to yield the desired sp3-hybridized bridge (13)
(Scheme 1).
Cyclisation by ring-closing metathesis of the linear
peptides
Ring-closing metathesis was carried out using the resin-
attached linear peptides 5–7 to eliminate any potential
problems arising from dimerization and (or) poor peptide sol-
ubility.
© 2005 NRC Canada

Whelan et al.
877
Scheme 2.
Scheme 3.
H
N
O
Fmoc
O
HN
O
O
O
O
H
N
4
R1
O
O
O
O
H
N
H
N
H
N
H
N
Fmoc
Deprotection 1. 20% Piperidine/DMF
coupling
2. Fmoc-Hag-OH, HATU, NMM
N
H
N
H
N
H
N
H
N
H
O
O
O
O
R2
O
N
O
H
N
Fmoc
N
H
O
O
R1 = 2-naphthyl
2 = CH(OtBu)CH3
R1 = Ph
2 = CH2(3-indolyl)
O
6
7
R
R
Stepwise elongation
O
Cl2(PCy3)2Ru=CHPh, DCM,
0.4 mol/L LiCl/DMF
HN
O
O
R1
O
O
O
O
H
N
H
N
H
N
H
N
H
N
H
N
Fmoc
Fmoc
N
H
N
H
N
H
N
H
N
H
N
H
O
O
O
O
O
O
O
N
O
NH
O
O
O
O
O
O
O
N
O
H
N
H
N
O
N
H
N
H
N
H
O
R2
O
O
5
12
13
R1 = Ph
R2 = CH2(3-indolyl)
Cl2(PCy3)2Ru=CHPh, DCM
0.4 mol/L LiCl/DMF
R
R
1 = 2-naphthyl
2 = CH(OtBu)CH3
1. Rh(PPh3)3Cl, H2 (60 psi),
10% MeOH:DCM
O
H
N
H
N
2. 20% piperidine/DMF
3. TFA cleavage
Fmoc
N
H
N
H
O
O
N
2b
3b
O
NH
O
O
O
O
H
N
H
N
On completion of the metathesis, an aliquot of peptide
was cleaved from the resin for analysis. Analytical LC–MS
analysis of the cleaved material revealed one major product
with molecular mass consistent with that of 9 and a pure
sample (>98%) was obtained in 46% yield after purification
by preparative HPLC.
O
N
N
H
O
H
O
O
O
O
H
N
8
O
1. Rh(PPh3)3Cl, H2 (60 psi),
10% MeOH:DCM
2. 20% piperidine/DMF
3. TFA cleavage
1b
O
H
N
H
N
Exposure of the fully protected allyl glycine containing
octreotide sequence 5 to the ruthenium benzylidene complex
Cl2(PCy3)2Ru=CHPh (Grubbs catalyst) (14) resulted in for-
mation of the carbocycle 8 in 65% yield4 (Scheme 2). It was
necessary to react the peptide whilst fully protected as it was
found that exposed amino and carboxylate groups prevented
metathesis, in agreement with previous observations with
other systems (19). To prevent aggregation of the resin-
tethered peptide during the metathesis reaction, it was vital
that a salt be added, in this case a solution of lithium chlo-
ride in DMF. Also essential for complete reaction were ex-
tended reaction times; reaction times less than 48 h yielded a
significant amount of starting material. Significantly, no
oligomerization or dimerization by acyclic diene metathesis
(ADMET) (20) was observed.
N
H
H2N
O
O
NH
O
NH
O
H
N
H
N
HO
N
H
NH2
O
O
HO
HO
9
The geometry of the newly formed olefinic bond of 9 was
investigated by NMR spectroscopy. The olefinic protons are
represented in the H NMR spectrum by a broad and ill-
1
4 Yields of peptides were determined after cleavage of the peptide from the resin and calculated using an average peptide loading of
0.71 mmol/g of resin.
© 2005 NRC Canada

878
Can. J. Chem. Vol. 83, 2005
1
Table 1. H NMR chemical shifts (ppm) of dicarba-octreotide
analogue 1b (500.13 MHz, DMSO-d6).
resolved signal, centred on 5.38 ppm, which integrates to
two hydrogens. COSY and TOCSY experiments correlated
this signal to both methylene groups adjacent to the newly
formed carbon–carbon double bond (2.20/2.36 and
2.27 ppm). Whilst coupling constants could not be accu-
NH
α-H
β-H
Others
D-Phe1
7.59
4.30
2.90, 2.75
H2, H6 7.62
H3, H5 7.35
H4 7.27
γ-H 4.05
H2, H6 7.81
H3, H5 7.60
H4 7.28
H2 8.51
H4 7.05
H5 7.50
H6 7.32
1
rately extracted from the one-dimensional H NMR spec-
trum or from the phase-sensitive COSY spectrum,
simulation5 of the NMR signals using the estimates of the
coupling constants from these spectra indicates that the
newly formed olefin is cis in geometry. The simulated spec-
trum that achieved very good similarity with the experimen-
tal data was obtained using a vicinal olefinic coupling
constant of 10.2 Hz, which is consistent with the cis geome-
try. Good agreement between the simulated and experimen-
tal spectra could not be achieved using values of 14–19 Hz,
which would be consistent with a trans double bond.
Cyclisation of the resin-attached dicarba-lanreotide 6 and
dicarba-vapreotide 7 precursors proceeded via the same
technique as described above (Scheme 3). The reactions did,
however, require longer reaction times, with both reactions
reaching completion after 96 h. In each case, mass spectral
analysis of the Fmoc- and resin-cleaved peptides showed
only the desired product and no remaining linear precursor
or higher oligomers. Preparative HPLC of the crude unsatu-
rated cyclic peptides 10 and 11 resulted in high purity mate-
rial albeit at reduced yield. The lanreotide 10 and vapreotide
11 analogues were subsequently isolated in 25% and 2%
yield with 98.6% and 99.2% purity, respectively.
Hag2
Phe3
8.09
7.80
4.29
4.61
1.05, 1.05
2.68, 2.80
D-Trp4
8.49
8.23
4.28
4.08
3.05, 2.82
1.61, 1.61
H7 6.95
Lys5
γ
-H 0.95, 0.95
δ-H 1.45, 1.45
ε -H 2.60, 2.60
γ-H 1.08
Thr6
7.74
7.66
7.15
4.02
4.22
3.58
4.88
1.58
3.82
Hag7
γ-H 4.10
Thr(ol)8
γ-H 0.95
δ-H 2.62
active sites. Rh(I)-catalysed hydrogenation in 10% methanol –
dichloromethane was investigated and this system was found
to effect quantitative reduction at room temperature under
mild hydrogen pressure (60 psi, 1 psi = 6.894 757 kPa). The
peptide 1b was purified using reverse-phase preparative
chromatography to give material of >98.9% purity in 26%
isolated yield and characterized using mass spectrometry
OH
R1
1
O
and H NMR spectroscopy.
H
N
H
N
The hydrogenation conditions established for the resin-
tethered octreotide analogue were then applied to the
olefinic dicarba analogues of lanreotide 12 and vapreotide
13 yielding the saturated carbocycles 2b and 3b as indicated
by MS and NMR (Scheme 3). Preparative HPLC again gave
significantly reduced mass recoveries of highly pure materi-
als, 4% and 20% yields, respectively, for 2b and 3b, with
purities of >97%.
N
H
H2N
O
O
NH
O
NH
O
O
H
N
H
N
H2N
N
H
NH2
R2
O
O
R
R
1 = 2-naphthyl
2 = CH(OH)CH3
R
R
1 = Ph
2 = CH2(3-indolyl)
10
11
Molecular modelling studies of carbocycle 1b (22) based
on X-ray diffraction data of octreotide (23) suggests that the
structural modification should not significantly perturb the
Phe7-D-Trp8-Lys9-Thr10 binding domain and hence somato-
statin receptor affinity. A detailed NMR study is in progress
Reduction of olefinic dicarba-peptides
Reduction of the newly installed olefinic bond is neces-
sary to complete the synthesis of the dicarba analogues of
octreotide 1b, lanreotide 2b, and vapreotide 3b. A homoge-
neous hydrogenation catalyst, the Rh(I)-based Wilkinson’s
catalyst, was chosen to facilitate the reduction of the resin-
bound material. This catalyst was chosen for its ease of use
and facile removal from resin-bound peptides.
1
and H chemical shifts for 1b and its unsaturated analogue 9
1
are summarized in Tables 1 and 2, respectively. H NMR
data for the dicarba-lanreotide 2b is shown in Table 3. The
preliminary data suggests the presence of a β-turn, in agree-
ment with the results of molecular modelling studies.
Competitive binding experiments of the six dicarba ana-
logues 1b–3b and 9–11 against 111In-DOTA-octreotate for
the sst2-expressing AR42J cell membranes is currently being
investigated. Further work will involve testing the binding
affinity of the new dicarba somatostatin analogues for other
somatostatin receptor sub-types and the effect of conjugating
bifunctional ligands on their affinity.
Reduction in dichloromethane alone, a solvent preferred
for efficient swelling of the Rink amide resin but non-ideal
for hydrogenation, was found to be exceptionally slow and
often incomplete. The addition of a coordinating solvent
subsequently enabled quantitative reduction (21). Traditional
hydrogenation solvents such as methanol have very poor
resin-swelling abilities, resulting in limited availability of re-
5 Simulation was performed using Bruker-Biospin’s NMR-Sim® v3.5. Spectral simulations were performed using the same acquisition and
processing parameters that were used to acquire the original data.
© 2005 NRC Canada

Whelan et al.
879
1
1
Table 2. H NMR chemical shifts (ppm) of unsaturated dicarba-
Table 3. H NMR chemical shifts (ppm) of dicarba-lanreotide
analogue 2b (500.13 MHz, DMSO-d6).
octreotide analogue 9 (500.13 MHz, DMSO-d6).
NH
α-H
β-H
Othersa
NH
α-H
β-H
Others
D-Phe1
7.37
4.41
2.71, 3.02
H2, H6
H3, H5
H4
D-β-Nal1
7.67
4.43
2.93, 3.12
H1 7.73
H3 7.43
H4 7.74
Hag2
Phe3
8.43
8.05
4.71
4.62
2.20, 2.36
2.76, 2.76
γ-H 5.38
H2, H6
H3, H5
H4
H2
H4
H5
H6
H7
γ-H 1.42, 1.42
δ-H 1.70, 1.70
ε -H 2.63, 2.63
γ-H 1.04
γ-H 5.38
γ-H 1.07
δ-H 3.37, 3.49
H5 7.81
H6 7.38
H7 7.32
H8 7.85
γ-H 1.34, 1.34
H2, H6 6.70, 6.70
H3, H5 6.48, 6.48
H2 7.78
H4 7.51
H5 6.97
D-Trp4
8.57
8.60
4.35
4.02
2.79, 3.03
0.96, 0.96
Hag2
Tyr3
8.12
7.81
4.28
4.54
1.05, 1.05
2.55, 2.69
D-Trp4
8.38
4.31
2.85, 3.06
Lys5
H6 7.04
H7 7.29
Thr6
7.71
7.82
7.37
4.29
4.59
3.97
3.99, 3.99
2.27, 2.27
3.65, 3.65
Lys5
7.92
4.16
1.61, 1.61
γ-H 1.06, 1.06
δ-H 1.40, 1.40
ε -H 2.67, 2.67
γ-H 0.88, 0.88
δ-H 1.63, 1.63
γ-H 0.99
Hag7
Thr(ol)8
Val6
Hag7
Thr8
7.65
7.95
7.41
3.98
4.31
4.83
1.96
1.05, 1.05
4.11
aSignal overlap prohibits unambiguous assignment of aromatic protons.
selection: 4096 points in t2 and 512 points in t1 were ac-
quired with 64 scans per t1 point and a 0.8 s relaxation delay
between scans. The spectral width was 6000 Hz in both di-
mensions. Zero-filling to 4096 points in t2 and forward lin-
ear prediction to 1024 points and zero-filling to 4096 points
in t1 were used. An unshifted sine-bell window function was
applied in both dimensions prior to Fourier transformation.
The two-dimensional TOCSY experiments were per-
formed in phase-sensitive mode using echo/antiecho–TPPI
gradient selection: 4096 points in t2 and 512 points in t1
were acquired with 32 scans per t1 point and a 0.8 s relax-
ation delay between scans. The spectral width was 6000 Hz
in both dimensions. The TOCSY mixing time was 120 ms.
2048 Points in t2 and zero-filling to 1024 points in t1 were
used. A π/4-shifted, sine-squared-bell window function was
applied in both dimensions prior to Fourier transformation.
The two-dimensional NOESY experiments were per-
formed in phase-sensitive mode using States gradient selec-
tion: 4096 points in t2 and 512 points in t1 were acquired
with 64 scans per t1 point and a 0.8 s relaxation delay be-
tween scans. The spectral width was 6000 Hz in both dimen-
sions. The NOESY mixing time was 400 ms. 2048 Points in
t2 and zero-filling to 1024 points in t1 were used. An expo-
nential window function of 3.0 Hz was applied in the t2 di-
Conclusion
The methodology described in this paper is a highly effi-
cient, generic, on-resin route to carbon-based somatostatin
mimics, which could find wide-spread use in peptidomim-
etic research. Our group is currently applying this methodol-
ogy to other peptides, particularly cystine-containing cyclic
peptides.
Experimental
Instrumentation and general procedures
1H and 13C NMR spectra were recorded on a Bruker
DPX300 spectrometer operating at 300.13 MHz (1H) and
75.48 MHz (13C).
The NMR spectra of the peptides were recorded in
DMSO-d6 (referenced to δH 2.49 ppm) at 298 K using a
1
Bruker DRX500 spectrometer at 500.13 MHz for H using a
5 mm inverse-detection H–13C–15N TXI probe with z gradi-
1
ents. Pulse sequences are from the Bruker XWIN-NMR v3.5
library. Chemical shift data were derived from a combination
of the COSY45, COSY, and TOCSY experiments.
Assignments were made using standard peptide NMR as-
signment techniques (24).
The two-dimensional COSY45 experiments were per-
formed in magnitude mode using gradient selection and a
45° pulse: 4096 points in t2 and 512 points in t1 were ac-
quired with 16 scans per t1 point and a 0.8 s relaxation delay
between scans. The spectral width was 6000 Hz in both di-
mensions. 2048 Points in t2 and zero-filling to 1024 points in
t1 were used. An unshifted sine-bell window function was
applied in both dimensions prior to Fourier transformation.
The two-dimensional COSY experiments were performed
in phase-sensitive mode using echo/antiecho–TPPI gradient
mension and
a
π/4-shifted, sine-squared-bell window
function was applied in the t1 dimension prior to Fourier
transformation.
Positive electrospray mass spectral data were obtained on
a VG Platform mass spectrometer using a cone voltage of
50 V.
Analytical HPLC spectra were recorded on a Waters Alli-
ance 2695 Separations Module equipped with a Waters PDA
2996 detector monitoring 190–300 nm. Components were
separated on either an Altima 150 mm × 4.6 mm C18 5 µm
© 2005 NRC Canada

880
Can. J. Chem. Vol. 83, 2005
Table 4. Data for cyclic peptide analogues.
Molecular
formula
Exact
mass
m/z (Pos. ion)
obtained
Retention
%
Yield
%
Purity
No.
time (min)a
980.51
982.53
981.2
983.6
9
C51H68N10O10
C51H70N10O10
C56H70N11O10
C56H72N11O10
C59H71N11O10
C59H73N11O10
30.10
33.37
46.00
43.25
11.42
39.50
46
23
25
4
98.9
98.8
98.6
97.0
99.2
99.6
1b
10
2b
11
3b
1057.54
1059.55
1093.54
1095.55
1058.6
1060.3
1094.5
1096.4
2
20
aPreparative HPLC.
packed column or a Waters 50 mm × 4.6 mm Xterra
C18 2.5 µm packed column. Purification via preparative
HPLC was achieved using a Shimadzu LC-4A with an SPD-
2AS detector (227 nm) and a C18 Altima 250 mm × 22 nm
5 µm packed column.
100.84 (OCHO), 120.07 (C3, C6), 124.74 (C2′, C6′), 126.21
(C2, C7), 127.11 (C4, C5), 127.77 (C1, C8), 129.83 (C4′),
130.30 (C3 , C5 ), 141.31 (C4a, C4b), 143.82 (C1 ), 143.85
′
′
′
(C8a, C9a), 156.52 (CO), 170.94 (COOH). MS (ESI+,
DCM–MeOH) m/z: 350.2 (M + Na+), C19H21NO4Na.
Solvents were purified, dried, and degassed according to
standard procedures. Degassed methanol and dichloromethane
were used in hydrogenation reactions. Deuterated chloro-
form was used as supplied by Merck. Deuterated dimethyl-
sulphoxide (99.96% D, DLM-34) was used as supplied by
Cambridge Isotope Laboratories. Chemical reagents, includ-
ing amino acids, were purchased from Sigma-Aldrich, AusPep,
or GL Biochem and were used without further purification.
Catalysts were used as received from Strem chemicals.
Grubbs catalyst refers to bis(tricyclohexylphosphine)benzyl-
idene ruthenium dichloride. Wilkinson’s catalyst refers to
chlorotris(triphenylphosphine)rhodium(I). All ruthenium-
catalysed metathesis reactions were performed using standard
Schlenk techniques under an atmosphere of nitrogen. In all
rhodium-phosphine hydrogenations, high purity (<10 ppm of
oxygen) hydrogen and argon (supplied by BOC gases) were
used and purified by passage through a series of traps to re-
move water, oxygen, and hydrocarbons.
Peptide synthesis
Linear peptides were synthesized using manual solid-phase
peptide synthesis techniques on Rink amide resin, utilizing
3 equiv. each of 9-fluorenylmethoxycarbonyl (Fmoc) pro-
tected amino acids and the coupling reagent HATU (O-(7-
azabenzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium hexafluoro-
phosphate) and 6 equiv. of NMM (4-methyl morpholine).
Fmoc-L-threoninol p-carboxybenzacetal was coupled directly
to the resin in the synthesis of 5 using 3 equiv. each of the
coupling reagents DIC (diisopropylcarbodiimide) and HOBt
(1-hydroxybenzotriazole), as well as 3 equiv. of the acetal.
Upon completion of the synthesis of the linear Fmoc-
protected octapeptide, a resin aliquot was removed and the
peptide cleaved from the resin with trifluoroacetic acid
(9 mL) containing the scavengers ethanedithiol (0.25 mL),
thioanisole (0.5 mL), and phenol (0.79 g), as well as water
(0.25 mL). The filtrate was concentrated by application of
high pressure air. The peptide was precipitated by the addi-
tion of ice cold diethyl ether and was extracted into a 50%
aqueous solution of acetonitrile and lyophilized to give a
white solid that was characterized by mass spectral analysis.
Synthesis of the linear peptides
Fmoc-L-threoninol p-carboxybenzacetal
Fmoc-threoninol (1.0 g, 3.05 mmol), p-carboxybenzaldehyde
(0.5 g, 3.36 mmol), and a catalytic amount of p-toluene-
sulphonic acid were suspended in dry toluene (70 mL) and
stirred at reflux for 20 h using a Dean–Stark apparatus. The
solvent was removed under reduced pressure to yield a clear
yellow oil, which was extracted into dichloromethane and
washed with water and saturated sodium chloride and dried
over sodium sulphate. Purification of the yellow oil by flash
chromatography (SiO2, dichloromethane – ethyl acetate, 3:1)
yielded the desired acetal (0.71 g, 51%), mp 86–88 °C (lit.
General procedure for ring-closing metathesis
A Schlenk flask was charged with resin-tethered peptide
(50–200 mg) to which dry dichloromethane (20 mL) and de-
gassed 0.4 mol/L lithium chloride in dimethylformamide
(1 mL) were added and the resin allowed to swell for 1 h be-
fore addition of a solution of Grubbs catalyst (20 mol%) in
dry dichloromethane. The suspended resin was stirred at
50 °C for a prescribed period of time (48–96 h) and col-
lected via filtration, washed with dichloromethane, dimethyl-
formamide, and methanol, and finally dried in vacuo. An
aliquot of the resin was removed and the peptide cleaved
from the resin as described previously.
The peptide was purified using preparative reverse-phase
high performance liquid chromatography (RP-HPLC) (20%–
100% aqueous acetonitrile with 0.1% TFA over 2 h,
10 mL/min). Fractions were collected, lyophilized, and ana-
lysed by analytical HPLC, NMR, and mass spectral analysis
(Table 4).
1
value (18) mp 86–88 °C). H NMR (300 MHz, CDCl3) δ:
1.29 (d, J = 6.4 Hz, 3H, CH3), 3.73 (dd, J = 10.1, 1.6 Hz,
1H, NHCHCH(H)), 4.11–4.27 (m, 4H, CHCH3, CH2OCO,
and NHCHCH(H)), 4.51–4.38 (m, 2H, H9 and NHCH), 5.57
(d, J = 9.9 Hz, 1H, NH), 5.65 (s, 1H, OCHO), 7.32 (t, J =
7.4 Hz, 2H, H2, H7), 7.40 (t, J = 7.6 Hz, 2H, H3, H6), 7.62
(d, 2H, J = 7.2 Hz, H2′, H6′), 7.63 (d, J = 8.4 Hz, 2H, H3′,
H5′), 7.77 (d, J = 7.3 Hz, 2H, H1, H8), 8.14 (d, J = 8.3 Hz,
2H, H4, H5), 10.05 (s, 1H, COOH). 13C NMR (75 MHz,
DMSO-d6) δ: 17.78 (CH3), 47.45 (C9), 48.90 (NHCH),
67.12 (NHCHCH2), 72.00 (CH2OCO), 75.61 (CHCH3),
© 2005 NRC Canada

Whelan et al.
881
L.J. Hofland, J.W. Koper, and S.W. Lamberts. Eur. J. Nucl.
Med. 20, 716 (1993); (b) W.H. Bakker, R. Albert, C. Bruns,
W.A.P. Breeman, L.J. Hofland, P. Marbach, J. Pless, D. Pralet,
B. Stolz, J.W. Koper, S.W.J. Lamberts, T.J. Visser, and E.P.
Krenning. Life Sci. 49, 1583 (1991).
General procedure for hydrogenation
A Fischer–Porter tube was charged in a drybox with resin-
tethered peptide (50–200 mg), Wilkinson’s catalyst (1–5 mg),
dry and deoxygenated dichloromethane (9 mL), and dry and
deoxygenated methanol (1 mL). The line and reaction vessel
were degassed with three vacuum–argon cycles each before
the tube was pressurised to 60 psi of hydrogen. The reaction
was stirred gently at room temperature overnight. The resin
was collected via filtration and washed with dichloro-
methane, dimethylformamide, and finally methanol, before
being allowed to dry in vacuo. An aliquot of the resin was
removed and the peptide cleaved from the resin as described
previously.
10. J.S. Lewis, M.R. Lewis, A. Srinivasan, M.A. Schmidt, J.
Wang, and C.J.J. Anderson. J. Med. Chem. 42, 1341 (1999).
11. (a) T. Maina, B. Nock, A. Nikolopoulou, P. Sotiriou, G.
Loudos, D. Maintas, P. Cordopatis, and E. Choitellis. Eur. J.
Nucl. Med. 29, 742 (2002); (b) J. Fichna and A. Janecka.
Bioconjugate Chem. 14, 3 (2003).
12. (a) H. Kolan, J. Li, and M.L. Thakur. Peptide Res. 9, 144
(1996); (b) J.K. Amartey. Nucl. Med. Biol. 20, 539 (1993).
13. A.N. Whelan, J. Elaridi, M. Harte, S.V. Smith, W.R. Jackson,
and A.J. Robinson. Tetrahedron Lett. 45, 9545 (2004).
14. S.T. Nguyen, L.K. Johnson, R.H. Grubbs, and J.W. Ziller. J.
Am. Chem. Soc. 114, 9858 (1992).
15. (a) H. Kessler and N. Schmiedeberg. Org. Lett. 4, 59 (2002);
(b) J.F. Reichwein, C. Versuluis, and R.M.J. Liskamp. J. Org.
Chem. 65, 6187 (2000); (c) T.D. Clark and M.R. Ghadiri. J.
Am. Chem Soc. 117, 12 364 (1995); (d) S.J. Miller and R.H.
Grubbs. J. Am. Chem. Soc. 117, 5855 (1995); (e) E.N.
Prabhakaran, V. Rajesh, S. Dubey, and J. Iqbal. Tetrahedron
Lett. 42, 339 (2001); ( f ) H.E. Blackwell, J.D. Sadowsky, R.J.
Howard, J.N. Samson, J.A. Chao, W.E. Steinmetz, D.J.
O’Leary, and R.H. Grubbs. J. Org. Chem. 66, 5291 (2001).
16. J.L. Stymiest, B.F. Mitchell, S. Wong, and J.C. Vederas. Org.
Lett. 5, 47 (2003).
The peptide was purified using preparative reverse-phase
HPLC (20%–100% aqueous acetonitrile with 0.1% TFA
over 2 h, 10 mL/min). Individual fractions were collected,
lyophilized, and analysed by analytical HPLC, NMR, and
1
mass spectral analysis (Table 4). H NMR data for dicarba
octreotide 1b, unsaturated dicarba octreotide 9, and dicarba
lanreotide 2b are reported in Tables 1–3.
Acknowledgements
The authors would like to acknowledge funding from an
Australian Postgraduate Award (awarded to ANW), as well
as CSIRO and ANSIE grants. Further thanks go to Noel Hart
and Stuart Littler for assistance in purification of the pep-
tides.
17. E. Teoh, E.M. Campi, W.R. Jackson, and A.J. Robinson.
Chem. Commun. 978 (2002).
18. H.P. Hsieh, Y.T. Wu, S.T. Chen, and K.T. Wang. Bioorg. Med.
Chem. 7, 1797 (1999).
References
19. Intolerance of catalysts to basic functionality has been previ-
ously reported. (a) S. Kotha, N. Sreenivasachary, K.
Mohanraja, and S. Durani. Bioorg. Med. Chem. Lett. 11, 1421
(2001); (b) R.H. Grubbs and S. Chang. Tetrahedron, 54, 4413
(1998); (c) T.M. Trnka and R.H. Grubbs. Acc. Chem. Res. 34,
18 (2001); (d) M.S. Sanford, J.A. Love, and R.H. Grubbs. J.
Am. Chem. Soc. 123, 6543 (2003).
20. M.R. Buchmeiser. Chem. Rev. 100, 1565 (2000).
21. R. Santini, M.C. Griffith, and M. Qi. Tetrahedron Lett. 39,
8951 (1998). Incompatible resin–solvent pairing can lead to
poor reaction site availability and diminished reaction rates.
Rink amide resin swells poorly in methanol, but exhibits high
swell in chlorinated solvents.
22. The unit cell of octreotide reveals the flexibility of the peptide
backbone and contains three individual molecules of the pep-
tide. Only one molecule (Molecule II) possesses the rotomer
conformation believed to be associated with biological activity.
This molecule was modelled using Insight II with the Discover
Minimization Module [computer program]. Vol. 2002.
Accelerys, San Diego, Calif. 1996.
1. K. Shin-ya, Y. Masuoka, A. Nagai, K. Furihata, K. Nagai, K.
Suzuki, Y. Hayakawa, and Y. Seto. Tetrahedron Lett. 42, 41
(2001).
2. A. Yurek-George, F. Habens, M. Brimmell, G. Packham, and
A. Ganesan. J. Am. Chem. Soc. 126, 1030 (2004).
3. R. Furumai, A. Matsuyama, N. Kobashi, K.-H. Lee, N.
Nishiyama, H. Nakajima, A. Tanaka, Y. Komatsu, N. Nishino,
M. Yoshida, and S. Horinouchi. Cancer Res. 62, 4916 (2002).
4. P. Grieco, P. Campiglia, I. Gomez-Monterrey, T. Lama, and E.
Novellino. Synlett, 14, 2216 (2003), and refs. cited therein.
5. (a) P. Campiglia, I. Gomez-Monterrey, L. Longobardo, T.
Lama, E. Novellino, and P. Grieco. Tetrahedron Lett. 45, 1453
(2004), and refs. cited therein; (b) X. Zhu and R.R. Schmidt.
Eur. J. Org. Chem. 4069 (2003), and refs. cited therein.
6. (a) R.F. Nutt, R.G. Strachan, D.F. Veber, and F.W.J. Holly. J.
Org. Chem. 45, 3078 (1980); (b) K. Undheim, J. Efskind, and
G. Beate Hoven. Pure Appl. Chem. 75, 279 (2003).
7. W. Bauer, U. Briner, W. Doepfner, R. Haller, R. Huguenin, P.
Marbach, T.J. Petcher, and J. Pless. J. Life Sci. 31, 1133 (1982).
8. D.A. Pearson, J. Lister-James, W.J. McBride, D.M. Wilson,
L.J. Martel, E.R. Civitello, J.E. Taylor, B.R. Moyer, and R.T.
Dean. J. Med. Chem. 39, 1361 (1996).
23. E. Pohl, A. Heine, and G.M. Sheldrick. Acta Crystallogr. Sect.
D, D51, 48 (1995).
24. K. Wüthrich. NMR of proteins and nucleic acids. John Wiley
and Sons, New York. 1986.
9. (a) E.P. Krenning, D.J. Kwekkeboom, W.H. Bakker, W.A.P.
Breeman, P.M. Kooji, H.Y. Oei, M. van Hagen, P.T.E.
Postema, M. de Jong, J.C. Reubi, T.J. Visser, A.E.M. Reijs,
© 2005 NRC Canada