Novel sst5-selective somatostatin dicarba-analogues: Synthesis and conformation-affinity relationships

Article abstract of DOI:10.1021/jm070886i

We describe synthesis, conformational studies, and binding to the five somatostatin receptors (sst_1-5) of a few analogues of the cyclic octapeptide octreotide (1), where the disulfide bridge was replaced by a dicarba group. These analogues were

Full text of DOI:10.1021/jm070886i

512
J. Med. Chem. 2008, 51, 512–520
Novel sst5-Selective Somatostatin Dicarba-Analogues: Synthesis and Conformation-Affinity
Relationships
Debora D’Addona,† Alfonso Carotenuto,*,‡ Ettore Novellino,‡ Véronique Piccand,§ Jean Claude Reubi,§ Alessandra Di Cianni,†
Francesca Gori,4 Anna Maria Papini,† and Mauro Ginanneschi*,†
Laboratory of Peptides and Proteins, Chemistry and Biology, Department of Organic Chemistry, UniVersity of Firenze, Via
Lastruccia 13 I-50019, Sesto Fiorentino, Italy, Department of Pharmaceutical Chemistry and Toxicology, UniVersity of Napoli, Via
Domenico Montesano 49, Italy, DiVision of Cell Biology and Experimental Cancer Research, Institute of Pathology, UniVersity of Berne,
Murtenstrasse 31 CH-3010 Berne, Switzerland, and Laboratory of Peptides and Proteins, Chemistry and Biology, Department of
Pharmaceutical Sciences, UniVersity of Firenze, Via U. Schiff 6 I-50019, Sesto Fiorentino, Italy
ReceiVed July 24, 2007
We describe synthesis, conformational studies, and binding to the five somatostatin receptors (sst1–5) of a
few analogues of the cyclic octapeptide octreotide (1), where the disulfide bridge was replaced by a dicarba
group. These analogues were prepared by on-resin RCM of linear hepta-peptides containing two allylglycine
residues; first- and second-generation Grubbs catalyst efficiencies were compared. The CdC bridge was
hydrogenated via two different methods. Binding experiments showed that two analogues had good affinity
and high selectivity for the sst5 receptor. Three-dimensional structures of the active analogues were determined
by 1H NMR spectroscopy. Conformation-affinity relationships confirmed the importance of D-Phe2 orientation
for sst2 affinity. Moreover, helical propensities well correlates with the peptide sst5 affinity. The presence of
the bulky aromatic side chain of Tyr(Bzl)10 favored the formation of a 310-helix and enhanced the sst5
selectivity suppressing the sst2 affinity. Finally, a new pharmacophore model for the sst5 was developed.
Introduction
clarified,4 being complicated by transmembrane domain interac-
tions. It is known that some of these receptors, mainly the sst2
and sst5 subtypes, mediate the antiproliferative effects of SRIF
on cellular growth of many sort of tumors. The clinical use of
native SRIF has been precluded by its short half-life in vivo
(1–2 min) but the broad spectrum of physiological activity of
this hormone has prompted many researchers to investigate the
structural requirements necessary to elicit the biological activity.
As a consequence, in the past decades, hundreds of peptide and
nonpeptide analogues were synthesized and tested for their
structure–activity relationships.5 Among them, octreotide (1)
(Figure 1; Sandostatin, SMS 201-995), a cyclic octapeptide SRIF
agonist, was synthesized by Sandoz researchers.6 It showed high
affinity and specificity toward sst2 receptor subtype.4 However,
the administration of 1 itself as a cell growth inhibitor was
successful only in a few cancer types as acromegaly and
metastatic carcinoid diseases.7 It is used in clinical protocols
mainly as a carrier of radionuclides for diagnostic or therapeutic
purposes.4,8,9
Peptide 1 incorporates D-Trp in place of Trp8 of SRIF and
exhibits the so-called pharmacophore sequence, Phe7-D-Trp8-
Lys9-Thr10 (Note: numbering of the residues follows that of
native SRIF). The disulfide tether of 1 (Figure 1) stabilizes a
ꢀ-turn structure spanning the D-Trp8-Lys9 residues, supposed
to be mandatory for the binding to the ssts.10 On the other hand,
it is well-known that the disulfide bridge is prone to be attacked
by endogenous reducing enzymes or by nucleophilic and basic
agents.11 Furthermore, the labeling of the octreotide derivatives
with isotopes such as 99mTc and 188Re needs a step in reducing
medium that can afford ring cleavage, especially at the level of
the disulfide linkage.9 The search for more robust octreotide
derivatives, carrying a stable bridging tether, prompted us to
consider the dicarba-analogues as a convenient target. In the
past, Veber and co-workers prepared nonreducible derivatives
of SRIF by using aminosuberic or diaminosuberic acid as
building blocks in the peptide chain synthesis.12,13 Similarly,
The highly potent cyclic peptide somatostatin (SRIFa;
H-Ala1-Gly2-c[Cys3-Lys4-Asn5-Phe6-Phe7-Trp8-Lys9-Thr10-Phe11-
Thr12-Ser13-Cys14]-OH, SRIF-14) was isolated from mammalian
hypothalamus1 and first synthesized by J. Rivier.2 Later on, it
was found that native SRIF occurs in two biologically active
forms, SRIF-14 and a 28-residue peptide, SRIF-28.3 The natural
hormone is widely distributed in the endocrine and central
nervous systems and it has many modulating actions in the body,
such as growth hormone (GH) inhibition, glucacon, insulin, and
gastrin release suppression. A family of five G-protein-coupled
somatostatin receptors (sst1–5) located in the cell membrane
mediates the biological effects of SRIF.4 The individual
signaling pathway of each SRIF receptor is still not completely
* To whom correspondence to be addressed. Phone: +39-081-
678626(A.C.); +39-055-4573525 (M.G.). Fax: +39-081-678644 (A.C.);
+39 055 4573531 (M.G.). E-mail: alfonso.carotenuto@unina.it (A.C.);
mauro.ginanneschi@unifi.it (M.G.).
†
Department of Organic Chemistry, University of Firenze.
University of Napoli.
University of Berne.
Department of Pharmaceutical Sciences, University of Firenze.
‡
§
4
aAbbreviations: Bzl, benzyl; dhDsa-C, dehydrodiaminosuberic acid
C-terminus; dhDsa-N, dehydrodiaminosuberic acid N-terminus; Dsa-C,
diaminosuberic acid C-terminus; Dsa-N, diaminosuberic acid N-terminus;
DMF, N,N-dimethylformamide; DMSO-d6, hexadeuteriodimethylsulfoxide;
DOTA, tetraazacyclododecanetetraacetic acid; DQF-COSY, double quantum
filtered-correlation spectroscopy; DTPA, diethylene-tetraaminepentaacetic
acid; EDT, 1,2-ethanedithiol; ESI-MS, electrospray ionization mass spec-
trometry; Fmoc, 9-fluorenylmethoxycarbonyl; GH, growth hormone; Hag,
L-2-allyl-Gly; HATU, hexafluorophosphate salt of the O-(7-azabenzotriazol-
yl)-tetramethyl uranium cation (this acronym no longer corresponds to the
true structure); MD, molecular dynamics; NMM, N-methyl morpholine;
NOE, nuclear Overhauser effect; NOESY, nuclear Overhauser enhanced
spectroscopy; PE COSY, primitive exclusive correlated spectroscopy; RCM,
ring-closing metathesis; RMSD, root mean square deviation; RP-HPLC,
reverse phase high performance liquid chromatography; SPE, solid phase
extraction; SPPS, solid phase peptide synthesis; SRIF, somatostatin; sst,
somatostatin receptor; TFA, trifluoroacetic acid; TOCSY, total correlated
spectroscopy; TSP, 3-trimethylsilanylpropionic acid.
10.1021/jm070886i CCC: $40.75
2008 American Chemical Society
Published on Web 01/23/2008

sst5-SelectiVe Somatostatin Dicarba-Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 3 513
Figure 1. Structures of 1 and of dicarba SRIF mimetic (2).
Table 1. Peptide Sequences; General Formula: D-Phe2-c[Xaa3-Phe7-
D-Trp8-Lys9-Yaa10-Zaa14]-Thr(ol)15-OH
peptide
Xaa
Cys
Yaa
Thr
Zaa
Cys
1
2
dhDsa-N
dhDsa-N
dhDsa-N
Dsa-N
Thr
Tyr(Bzl)
Phe
Thr
Tyr(Bzl)
dhDsa-C
dhDsa-C
dhDsa-C
Dsa-Cd
11
12
13
14
Figure 2. First-generation (3) and second-generation Grubbs catalyst
(4).
Dsa-N
Dsa-C
modifying the synthetic pathway and comparing the efficiency
of the first and second generation Grubbs catalysts (namely,
structures 3 and 4, respectively; Figure 2).
The ethylene group of the analogue 11 was hydrogenated,
affording the saturated derivatives 14 (Table 1). All the
synthesized analogues were tested for the binding to the sst1–5
receptor subtypes. Herewith, we also suggest correlations
between the binding data and the results of the conformational
analysis carried out on these compounds via NMR and MD
studies.
Gilon’s group has recently prepared constrained backbone-cyclic
analogues with high conformational stability by cyclization
through synthetic CR bifunctional aminoacids.14,15 The discovery
of the olefin metathesis reaction disclosed the way for synthesis
and cleavage of carbon-carbon bonds under mild conditions.16
In particular, the use of ring-closing metathesis (RCM) reactions
catalyzed by ruthenium complexes has become a popular method
for the formation of CdC bridged structures in organic
syntheses.17 Application of this method to amino acids bearing
unsaturated side chains (i.e., allylglycine) and located in strategic
positions of the peptide motif allows the preparation of carba-
cyclic peptides by the SPPS method.18–21 Moreover, it is known
that constrained carba-tethers in cyclic peptides can stabilize
helices or ꢀ-turns.22,23 However, the effect of the change of the
dihedral angle of disulphide bridge on receptors affinity of these
synthetic peptides is not easily predictable.18
We first applied RCM to the on-resin cyclization of D-Phe2-
Hag3-Phe7-D-Trp8-Lys9-Thr10-Hag14-Thr(ol)15, obtaining the
analogue 2 (Figure 1, and Table 1). This compound is stable in
the enzymatic pool of the human serum for more than 30 h.
NMR analysis in aqueous solution followed by restrained MD
simulations showed that the peptide has a conformational
behavior very similar to that of the parent peptide 1.24 In the
present work we have resynthesized compound 2 with an
improved synthetic strategy, and we have also prepared the
corresponding reduced analogue 13, containing the more flexible
CH2-CH2 tether (Table 1).
Almost at the same time, another group published the
synthesis of dicarba-analogues of 1, lanreotide, and vapreotide
and their hydrogenation products.25 However, no data on the
conformations of these compounds or on their ability to bind
with sst receptors were reported by these authors.
Apart from peptides 2 and 13, in the present paper new
analogues have been developed. Recent studies on the struc-
ture–activity relationships of sst ligands, which have called
attention to the role of lipophilic residues (Phe6,7,11) of the native
hormone26 prompted us to prepare new synthetic dicarba-
analogues by substituting Thr10 residue with Tyr(Bzl) (Bzl )
benzyl) and Phe (compounds 11 and 12, respectively, Table
1). To aim this, we exploited a new solid phase support,
Results and Discussion
Peptide Synthesis and Purification. In our previous work,24
the synthesis of the linear precursor of the analogue 2 started
by anchoring the building block Fmoc-threoninol p-carboxy-
benzaldehyde acetal linker to the Rink amide resin.27The
synthesis of this substance was complicated by the presence of
an unidentified impurity, not detected by the preceding authors
and difficult to remove. Moreover, for reducing the high
substitution level of the Rink amide resin (1 mmol/g) to reach
the pseudodilution conditions necessary to the cyclization step,
we end-capped the free NH2 groups with a (CH3CO)2O/
CH3COOH mixture. The following washing of the residual
acetic acid, which is harmful for catalyst 3, was a really time-
consuming step. In the method here reported we used the H-L-
Thr(t-Bu)-ol-2-chlorotrityl resin which already contains the
[Thr(ol)15] of the sequence (Scheme 1).
After the coupling of the Fmoc-Hag3, the resin loading (0.5
mmol/g) already met the requirements of the pseudodilution
effect, minimizing the risk of the formation of intermolecular
bonds. However, the RCM with catalyst 3 of the linear
octapeptide precursors linked to the Rink amide resin as well
as to the trityl resin gave unsatisfactory yield of the cyclic
compounds 2 and 11 and only traces of 12 (Supporting
Information). The factors that affect the success of the RCM
on peptide cyclization are not yet fully understood. Probably
inappropriate orientation of the alkenyl side chains may hamper
the RCM. In general, the solid-phase RCM was carried out on
peptides containing proline or N-alkyl residues in the sequence28
and only recently dicarba-analogues of oxytocin were prepared
by RCM on solid phase of the parent linear peptides.20,21

514 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 3
Scheme 1. Synthesis of Linear Peptidesa
D’Addona et al.
Scheme 2. Synthesis of Cyclic Peptides (2, 11, 12)a
a Reagents and conditions: (a) (i) Fmoc-Hag, HATU/NMM, 50 min, rt;
(ii) 20% piperidine in DMF (2 × 15 min); (iii) coupling with the following
amino acids: Fmoc-L-Thr(t-Bu) (5), Fmoc-L-Tyr(Bzl) (6), Fmoc-L-Phe (7),
Fmoc-L-Lys(Boc), Fmoc-D-Trp(Boc), Fmoc-L-Phe; Fmoc-Hag.
Consequently, we planned a more convenient synthetic approach
to these cyclopeptides. The elongation of the peptide sequence
was stopped after the coupling of Hag3 residue (Scheme 1),
with the aim of removing any possible interference of the
aromatic ring of D-Phe on the correct orientation of the
allylglycine side chains. The linear hepta-peptides 5–7 were then
converted by RCM with catalyst 3 to the corresponding cyclic
analogues 8–10 (Scheme 2).
The D-Phe2 terminal residue was added only after the
cyclization step (Scheme 2), affording the analogues 2 and 11
in an amount suitable for the next physicochemical and
biological studies. In the case of 12, due to the still persisting
poor linear-to-cyclic conversion, we were able to isolate the
cyclic peptide only for characterization purposes. A new
generation Grubbs catalysts, like 1,3-dimesitylimidazol-2-
ylidene 4 (Figure 2), was considered. The second generation
showed higher stability than the phosphine analogue 3, less
sensibility to nucleophilic groups present in the molecule, and
was more active in most metathesis reactions.29,30 Comparison
between the efficiency of catalysts 3 and 4 on the RCM reaction
of the acyclic heptapeptides 5-7 is reported in Table 2. Catalyst
4 increased the yield by 10% and shortened the reaction time
for all peptide sequences. It is worth noting that the presence
of Phe10 negatively affected the RCM no matter what kind of
catalyst was used. The Phe residue close to Hag14 hampered
the RCM more than the bulky aromatic Tyr(Bzl)10. This suggests
that RCM is not affected by the size of the residue in position
10 but probably by the influence it has on the proper orientation
of the allyl side chains of Hag3 and Hag14. In both reaction
conditions (catalysts 3 and 4), no significant dimerization or
oligomerization byproduct was observed by HPLC-ESI inspection.
The cleavage of the cyclopeptide 11 from the resin required
a more diluted TFA solution than for 2 and 12 to prevent the
contemporary hydrolysis of the Bzl group. The isolation of
peptides obtained on-resin by RCM is often complicated by the
tendency of the organometallic molecules to bind the solid
support. Several washing methods have been suggested to
eliminate Ru contaminants.21,31–33 However, not one of them
gave satisfactory results in our hands. Notably, we found that
repeated treatments of the resin with fresh piperidine solutions
before the coupling with Fmoc-D-Phe, centrifugation of the
aqueous suspension after the cleavage and separation by SPE,
if necessary, afforded compounds that were then easily purified
a Reagents and conditions: (b) catalyst 3, 40 °C, 48 h (8, 9) or 52 h
(10); catalyst 4, 40 °C, 24 h (8, 9) or 48 h (10); (c) (i) 20% piperidine in
DMF, Fmoc-D-Phe/HATU/NMM 50 min, rt; (ii) 20% piperidine in DMF;
cleavage of 2 and 12 [TFA/H2O/EDT/phenol (94:2:2:2)]; 11 [TFA/H2O/
EDT/phenol (70:26:2:2)].
Table 2. Cyclization of the Linear Hepta-Peptides 5, 6, and 7 with
catalyst 3 or 4
catalyst
dicarba-analogues (hours of reflux) % yielda cyclic/linear ratiob
2
2
11
11
12
12
3 (48)
4 (24)
3 (52)
4 (48)
3 (48)
4 (24)
25
35
10
20
3
85:15
90:10
50:50
80:20
30:70
50:50
13
a Overall yield of isolated pure compounds, calculated on the basis of
an average peptide loading of 0.5 mmol/g of resin. b Calculated on the basis
of peak area in analytical RP-HPLC.
from the residual catalyst contaminants by semipreparative
HPLC. Compounds 2 and 11 obtained by RCM with catalyst 4
were directly purified by RP-HPLC due to the low level of
contaminants. Finally, analytical RP-HPLC and ESI-MS analysis
of the crude compounds 2, 11, and 12 showed two chromato-
graphic peaks with the same MW, probably corresponding to
geometric isomers (ratio ≈ 90:10). In any case, binding and
conformational studies were performed only on the most
abundant isomer. The olefinic bridge of the analogues 2 and 11
was hydrogenated to study the effect of the higher flexibility
of the -CH2-CH2- group on the conformational behavior.
Low yield of peptide 12 prevented us from preparing its
hydrogenated derivative. Hydrogenation of 2 and 11 was
accomplished using two different methods because of the
different substitution in position 10. The Pd/C catalyst, which
is the most used tool to reduce olefinic bridges obtained by RCM

sst5-SelectiVe Somatostatin Dicarba-Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 3 515
Scheme 3. Synthesis of Saturated Analogues 13 and 14a
of a small molecule ligand selective toward sst5.38 Its affinity
toward sst5 (IC50 ) 87 nM) was lower compared to our selective
compounds. Finally, two highly potent sst4/sst5 peptidomimetics
have been recently described by Feytens et al.39 Their structure
is based on a constrained tryptophan scaffold, and the most
potent analogue has an IC50 value of 1.2 nM for the sst5. These
compounds showed also high affinity toward sst4 being only 3-
and 9-fold more selective for sst5 as compared to sst4.
Hence, peptides 11 and 14 are among the most potent and
sst5-selective compounds reported to date. The potential thera-
peutic utility of selective agonists for sst5, which is expressed
in the lymphoid system, in pancreatic beta cells, and in
corticotroph adenoma cells has been reported.4
NMR Analysis. NMR analysis of the analogues 2, 11, 13,
and 14, was performed using 1D and 2D proton homonuclear
experiments. NMR analysis of compound 12 was prevented by
the low yield of this peptide. NMR experiments were recorded
on a Varian Inova-Unity 700 MHz spectrometer in water/
DMSO-d6 8:2 solution (277.1 K) at 2 mM concentration. Solvent
mixture was chosen since it allowed a better dispersion of the
amide proton signals of some of the studied peptides compared
to the water solution.
a Reagents and conditions: (d) 20% Pd(OH)2/C, H2, 24 h, 30 °C; (e) (i)
MeOH/DCM, Wilkinson’s catalyst 2.5%, H2 (60 psi); (ii) 20% piperidine
in DMF, Fmoc-D-Phe/HATU/NMM, 50 min, rt; (iii) 20% piperidine in
DMF, cleavage.
on peptides, did not work well for the reduction of 2. Therefore,
the saturated analogue 13 was obtained by using H2 fluxing in
the presence of Pd(OH)2/C. However, heating at 30 °C of the
reaction mixture was needed to obtain compound 13 in good
yields (Scheme 3).
The analogue 14, obtained from the parent 11 (Table 1),
was achieved by using Wilkinson’s catalyst,34,35 to prevent
the cleavage of the Bzl group by the palladium catalyst
(Scheme 3).
Binding Affinity to sst1–5 Receptors. All compounds were
tested for their ability to bind to the five human sst1–5 receptor
subtypes in complete displacement experiments using the
universal SRIF radioligand [125I]-[Leu8,D-Trp22,Tyr25]-SRIF-
28. SRIF-28 was run in parallel as control. IC50 values were
calculated after quantification of the data using a computer-
assisted image processing system. The data are shown in Table
3. Binding data (Table 3) indicate that 2 shows nM binding
affinities toward the sst2, sst4, and sst5.
NMR Analysis of 13. Compound 13 is a carbocyclic
analogue of 1, with the disulfide group replaced by a CH2-CH2
bridge (Scheme 3). A qualitative analysis of short- and medium-
3
range NOEs, JNH-HR coupling constants, and temperature
coefficients for exchanging NH was used to characterize the
secondary structure of 13 (Supporting Information). Spectra
analysis supported the presence of a ꢀ-turn about residues 3–6.
Interestingly, the upfield shift observed for Hγs of Lys9 (δ )
0.51, and δ ) 0.30 ppm) has been used for decades as diagnostic
for biological activity.40,41
Surprisingly, hydroxyl proton signal of Thr10, which usually
exchanges too quickly to be observed in water solution, was
detected in the NMR spectra of 13. This feature indicates that
it can be engaged in a H-bond. NMR-derived constraints
obtained for 13 were used as the input data for a simulated
annealing structure calculation, as implemented within the
standard protocol of the DYANA program.42 NOEs were
translated into interprotonic distances and used as upper limit
constraints in subsequent annealing procedures to produce 200
conformations from which 50 structures were chosen, whose
interprotonic distances best fitted NOE-derived distances and
then refined through successive steps of restrained and unre-
strained energy minimization calculations using the program
Discover (Biosym, San Diego). A family of 20 structures,
satisfying the NMR-derived constraints (violations smaller than
0.40 Å), was chosen for further analysis. The PROMOTIF
program was used to extract details of the location and types
of structural secondary motifs.43 Backbone arrangement of 13
was well-defined, possessing an average root mean square
deviation (rmsd) of the heavy atoms equal to 0.32 Å (Figure
3a).
The analysis of the secondary structure showed the existence
of a distorted type II′ ꢀ-turn conformation about residues Phe7-
D-Trp8-Lys9-Thr10. The turn structure is stabilized by hydrogen
bonds between Phe7 CO and Thr10 NH and between Phe7 CO
and Thr10 OH. The turn is flanked by two short, extended
regions defining an antiparallel ꢀ-sheet. Also, the side chains
have defined orientations (rmsd of all the heavy atoms equal to
0.88 Å). These orientations allowed a close spatial proximity
between D-Phe2-Dsa-N3, Phe7-Thr10, and D-Trp8-Lys9 side
chains. Finally, the sequential Thr10 NH-Dsa-C14NH and Dsa-
C14NH-Thr(ol)15 NH NOEs, indicative of folded structures, are
Compared to the parent peptide 1,5 the affinities toward sst2
and sst5 receptors were reduced (about 20-fold for sst2 and about
5-fold for sst5) and the affinity toward sst3 receptor was
abolished. Peptide 13 shows similar binding affinity profiles
toward the ssts compared to that of 2. The affinity toward sst5
significantly decreased (203 vs 28 nM, Table 3). Compound
12 shows only low affinity toward sst4/5. Therefore, we did not
undertake the scale-up of the synthesis to provide the starting
material for the double bond hydrogenation.
Most interestingly, binding data indicate that 11 is a ligand
less potent than SRIF (IC50 ) 29 nM) but by far more
selective (at least 30-fold) for the sst5 subtype (Table 3).
Peptide 14 shows an enhanced affinity toward the sst5
compared to 11 (IC50 ) 12 nM), while its selectivity appears
only slightly reduced (Table 3).
Not many selective sst5 ligands have been found until now.
Gilon et al. developed a backbone-cyclic sst (PTR 3046) with
high selectivity toward sst5.36 The IC50 value for the binding of
PTR 3046 to sst5 was 67 nM, more than double compared to
our compounds 11 and 14. Some sst5-selective nonpeptide
peptidomimetics have been also reported. One of them, named
L-817,818 with some selectivity to sst5 has been obtained at
the Merck laboratories through combinatorial chemistry ap-
proach.37 L-817,818 showed sub-nM affinity toward sst5 (IC50
) 0.4 nM), but it was only 8-fold more selective for sst5 as
compared to sst1. The synthesis and screening of a focused
library of ꢀ-turn mimetics based upon the crucial Trp-Lys motif,
found in the turn region of SRIF, resulted in the identification

516 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 3
Table 3. Receptor Affinities of the Somatostatin Analogues
D’Addona et al.
a
IC50 (nM)
No.
sst1
sst2
sst3
sst4
sst5
SRIF-28
2
13
12
11
14
2.3 ( 0.4 (7)
>1000 (2)
3.0 ( 0.2 (7)
44 ( 1 (2)
49 ( 12 (2)
>1000 (2)
>1000 (2)
393 ( 20 (2)
3.6 ( 0.5 (7)
>1000 (2)
1.6 ( 0.3 (7)
412 ( 68 (2)
214 ( 9 (2)
141 ( 27 (2)
>1000 (2)
2.1 ( 0.2 (7)
28 ( 2 (2)
>1000 (2)
>1000 (2)
203 ( 106 (2)
206 ( 97 (2)
29 ( 1 (2)
961 ( 294 (2)
>1000 (2)
593 ( 39 (2)
>1000 (2)
892 ( 245 (2)
277 ( 123 (2)
846 ( 400 (2)
12 ( 2 (2)
a The number of independent repetitions to obtain the mean values ( SEM are indicated between brackets. SRIF-28 is used as internal control.
Figure 3. Stereoview of the 10 lowest-energy conformers of 13 (a), 2 (b), 14 (c), and 11 (d). Structures were superimposed using the backbone
heavy atoms of residues 3–14. Heavy atoms are shown with different colors (carbon, green; nitrogen, blue; oxygen, red). Hydrogen atoms are not
shown for clarity. C-terminal helical structure of compounds 14 and 11 was evidenced as a ribbon.
consistently violated by the ꢀ-sheet conformations (violations
of about 0.30 Å). Melacini et al. found similar inconsistencies
studying 1 in DMSO solution.44 The authors hypothesized the
existence of a conformational equilibrium between the ꢀ-sheet
structural cluster and a second conformational ensemble involv-
ing folded structures for the C-terminal residues where the
inconsistencies were observed. Therefore, a similar equilibrium
should hold also for 13. Superposition of the structure of 13
with that of 144 (Protein Data Bank entry 1SOC) showed very
similar structures in the active region (Figure 4). In contrast,
the structure of the N-terminal D-Phe2 and C-terminal Thr(ol)15
residues are significantly different in the two peptides.
The distances between the Cγ atoms of the putative phar-
macophoric side chains are reported in Table 4 together with
those found by Melacini et al.45 It can be observed that the
distances found for 13 are in good accordance with the sst2/3/5
pharmacophore.
Figure 4. Superposition of the most representative structure of 13
(blue) with the structure of 1 (PDB entry 1SOC, red). Structures were
superimposed using the backbone heavy atoms of residues 3–14.
NMR Analysis of 2. Compound 2 is the chemical precursor
of 13, with an unsaturated CHdCH bridge replacing the
disulfide group of 1 (Scheme 2). In a recent paper, we have
reported the NMR analysis of 2.24 The study showed that this
peptide had a conformational behavior, in aqueous solution, very
similar to the parent 1. We reanalyzed the peptide in DMSO/
Hydrogen atoms are not shown for clarity.
H2O solution at 277.1 K to compare its conformational behavior
with the other analogues. A coupling constant value (3JCH)CH
) 8.1 Hz) between the two olefinic protons of the bridge and
a relatively strong NOE between the same protons established

sst5-SelectiVe Somatostatin Dicarba-Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 3 517
Lys9 side chains; moreover, the tyrosyl group of the residue 10
points toward the Lys9 side chain. In contrast, the Phe7 side
chain showed almost free rotation about ꢁ1 angle. Also, the Bzl
group of residue 10 was highly flexible.
Table 4. Cγ-Cγ Distances (Å) between Putative Pharmacophoric
Residues
a
cmpd
13
14
11
sst2/3/5
5–11
11–15
12–15
Ar2-Ar7
9.9 ( 0.5b
12.3 ( 0.5
13.4 ( 0.5
8.8 ( 0.5
12.2 ( 0.5
13.0 ( 0.6
12.8 ( 0.5
7.1 ( 0.5
10.1 ( 0.6
4.7 ( 0.4
6.6 ( 0.5
4.8 ( 0.5
9.5 ( 0.3
13.3 ( 0.2
14.3 ( 0.2
12.6 ( 0.6
7.3 ( 0.3
11.1 ( 0.3
4.9 ( 0.2
7.3 ( 0.6
5.6 ( 0.6
Ar2-Ar8
NMR Analysis of 11. Compound 11 is the chemical precursor
of 14, with an unsaturated CHdCH bridge replacing the
disulfide group of 1 (Scheme 2). The coupling constant value
(3JCHdCH ) 15.1 Hz) between the two olefinic protons of the
bridge and NOE contacts between the same olefinic Hꢀs of
residue 14 (3) established the E-geometry about the double bond.
No signal attributable to the Z-isomer could be detected in the
spectra, indicating a unique configurational structure. Compound
11 shows spectral features resembling those found for 14, with
the exception of those of dhDsa-N3 and dhDsa-C14 residues,
which obviously feel the effect of the double bond in the side
chains. The 3D structures of 11 were calculated as described
above. The quality of the structure is reflected by the small rmsd
(backbone heavy atoms rmsd ) 0.40 Å), which can also be
visually depicted from Figure 3d, showing the bundle of 10
conformers representing the 3D structure. A ꢀ-turn about
residues D-Trp8-Lys9, followed by a short 310-helix along
residues Tyr(Bzl)10-Dsa-C14-Thr(ol)15 were the main backbone
features. An overall enhancement of the backbone conforma-
tional rigidity is observed for 11 compared to 14 (backbone
rmsd ) 0.40 Å vs 0.48 Å). Reduced entities of the NOE
violations for the proton couples HR-9/NH-10, HR-10/NH-14,
and HR-14/NH-15 (about 0.3 Å), and lower values of the
temperature coefficients observed for NH-14 and NH-15 (Sup-
porting Information), compared to the corresponding 14, indicate
that the conformational equilibrium between the extended and
the helical structures is shifted toward the helical conformation
in 11. Finally, the side-chain orientations of 11 were also similar
to those shown by 14 apart from the Tyr(Bzl)10 side chain, which
is more flexible in 11, showing an equilibrium between trans
and gauche- orientations (Figure 3d).
Ar2-Lys9
Ar2-Ar10
Ar7-Ar8
7.2 ( 0.3
9.4 ( 0.2
4.8 ( 0.2
7–9
9–11
5
Ar7-Lys9
Ar8-Lys9
Ar8-Ar10
Lys9-Ar10
a Pharmacophore for the sst2-, sst3-, and sst5-selective SRIF analogues.45
b Average distance and standard deviation calculated from the ensemble of
10 structures.
the Z-geometry about the double bond. No signal attributable
to the E-isomer could be detected in the spectra, indicating a
unique configurational structure.
Compound 2 shows spectral features (Supporting Information)
similar to those found in 13 but with some higher tendency to
the helical structure. For instance, its NOESY spectra shows,
at the same time, both strong HR-9/NH-10, HR-10/NH-14, and
HR-14/NH-15 NOE contacts, consistent with a ꢀ-sheet structural
cluster and dRN(i, i + 2) diagnostic connectivities between HR-
8/NH-10, HR-9/NH-14, and HR-10/NH-15, characteristic of
folded structures for the C-terminal residues. These data can
be explained only hypothesizing an equilibrium between
extended and folded conformational states, as already proposed
for 1.44 Notably, calculated structures (Figure 3b) are character-
ized by an antiparallel ꢀ-pleated sheet with a type II′ ꢀ-turn at
residues D-Trp8/Lys9, but several violations (NH-10/NH-14, HR-
10/NH-15, and NH-14/NH-15, Hꢀ-14/NH-15) were observed
consistently with the hypothesized ꢀ-sheet/helical equilibrium.
NMR Analysis of 14. In compound 14, as in 13, the disulfide
group of 1 is replaced by a CH2-CH2 bridge (Scheme 3).
Compound 14 differs from 13 since Thr10 residue is replaced
by the Tyr(Bzl) residue. Similar substitution into a cyclo-
hexapeptide SRIF mimic led to the discovery of a potent sst1–5
ligand.26 NMR data (Supporting Information) clearly indicated
the presence of a nascent helix or consecutive ꢀ-turn structures
along residues 8–15. NMR-derived constraints obtained for 14
were used as the input data for a simulated annealing structure
calculation, as described above.
Conformation-Affinity Relationships and Pharmacoph-
ore Model for sst5-Selective Analogues. On the basis of the
discussed results, we can draw some conformation-affinity
relationships concerning the binding to the sst receptors. As most
of the bioactive analogues of SRIF are reported so far,46 the
structures of the peptidomimetics presented here have a ꢀ-turn
of type II′ (Figures 3a-d) about residues D-Trp8 and Lys9. The
side chain of D-Trp8 is in the trans-conformer, and the side chain
of Lys9 is in the gauche- conformer, bringing the two side
chains adjacent to each other in close proximity. In compounds
2 and 13, which retain affinity toward sst2, the turn motif is
part of a ꢀ-hairpin structure. This conformation is very similar
to that observed for 1 (Figure 4). A detailed comparison of the
structures of 13 (and 2) with 1 shows a different spatial
orientation of the D-Phe2 aromatic cycle. Because, according
to Grace et al.,47 this moiety belongs to the consensus structural
motif of sst2-selective analogues, the different orientation
observed for D-Phe2 aromatic cycle could explain the reduced
affinities of 2 and 13 toward sst2 compared to the parent 1. On
this point, a direct involvement of the disulfide bridge in the
interaction with the sst2 has also been demonstrated.45,48 In fact,
a disulfide can replace an aromatic moiety in key sst–ligand
interactions.48
The quality of the structure is reflected by the small rmsd
(backbone heavy atoms rmsd ) 0.48 Å), which can also be
visually depicted from Figure 3c showing the bundle of 10
conformers representing the 3D structure.
From the backbone torsion angles, a distorted type II′ ꢀ-turn
about residues D-Trp8-Lys9 followed by a short 310-helix along
residues Tyr(Bzl)10-Dsa-C14-Thr(ol)15 can be identified (Figure
3c). The helical structure is stabilized by H-bonds between
D-Trp8 CO and Dsa-C14 NH and between Lys9 CO and Thr(ol)15
NH. Helical structure is in conformational equilibrium with more
extended conformations. In fact, violations of the NOE-derived
distances HR-9/NH-10, HR-10/NH-14, and HR-14/NH-15 (about
0.5 Å), together with relatively high temperature coefficients
observed for NH-10, NH-14, and NH-15 (-∆δ/∆T > 3.0 ppb/
K, Supporting Information) indicate that the helical structure is
quite flexible. Finally, side chains of D-Phe2, Dsa-N3, D-Trp8,
Lys9, Tyr(Bzl)10, and Dsa-C14 showed well-defined ꢁ1 values
(i.e., trans, gauche+, trans, gauche-, gauche-, and gauche-
orientations, respectively). ꢁ1 values obtained by MD calculation
As for 1, a dynamical equilibrium between ꢀ-sheet and helical
conformations was observed at the C-terminal residues of
analogues 2 and 13. The increased helical character observed
in compound 2 compared to 13 parallels its increased affinity
toward sst5 receptor. Actually, Grace et al. have reported that 1
analogues should prefer the helical conformation to fit the sst5
3
are in accordance with the experimental JHRHꢀ coupling
constants (Supporting Information). These orientations allowed
a close spatial proximity between D-Phe2/Dsa-N3, and D-Trp8/

518 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 3
D’Addona et al.
pharmacophore.47 Accordingly, analogues 11 and 14, which
show high affinity and selectivity toward sst5, feature a stable
helical structure at the C-terminus. In these analogues, the ꢀ-turn
motif is followed by a short 310-helix along residues Tyr(Bzl)10-
Dsa-C14-Thr(ol)15. The side chain of Tyr(Bzl)10, which appar-
ently stabilizes the helical structure, is in the gauche- conformer
and is located in close proximity to the D-Trp8-Lys9 pair (Figure
3c,d). If this structure corresponds to the bound conformation
of the analogue 14 (and 11), then the Tyr(Bzl) bulky side chain
must accommodate in a suitable hydrophobic pocket within the
sst5 binding site. To explain sst5 selectivity of analogues 11 and
14, it can be hypothesized that this side chain is hardly
accommodated into the sst1–4 binding sites. Nonetheless, the
Tyr(Bzl) residue replaced the Thr10 in other SRIF analogues
showing high-affinity binding toward all the sst receptors.26
Hence, the selectivity of the new analogues 11 and 14 toward
sst5 should depend on the conformationally restricted helical
structure that they adopt. Finally, considering our sst5-selective
analogues, the increased Tyr(Bzl) side chain, flexibly observed
in 11 compared to 14, can account for the reduced affinity
toward the sst5 observed in the first one.
Based on the results reported above, we propose a pharma-
cophore model for sst5-selective analogues. The model involves
the classical four side chains of the sst2/3/5 pharmacophore,45
namely, those of residues D-Phe2, Phe7, D-Trp8, and Lys9, plus
the Tyr(Bzl)10 side chain, which enhanced sst5 affinity for both
compounds 11 and 14. The distances between the Cγ atoms of
these side chains are reported in Table 4. Because 14 is the
most potent sst5 ligand among the analyzed peptides, the relevant
distances reported in Table 4 can be proposed as sst5 pharma-
cophoric distances. In the same table we also report the Cγ-Cγ
distances found by Melacini et al. for the sst2/3/5-selective SRIF
analogues.45 It can be observed that the distances found for 14
(and 11) agree with the sst2/3/5 pharmacophore. Recently, the
groups of J. Rivier and J. C. Reubi proposed pharmacophore
models for sst1-,49 sst2-,47 sst3-,50 and sst4-selective51 analogues,
by examining the NMR behavior of a large number of cyclic
peptides containing the disulfide bridge. To the best of our
knowledge, a model of the sst5 pharmacophore has never been
proposed so far.
tredwitz, Germany). HATU was obtained from Advanced Biotech
Italia (Milan, Italy). Peptide grade DMF was from Scharlau
(Barcellona, Spain). All the other solvents and reagents used for
SPPS were of analytical quality and used without further purifica-
tion. Analytical RP-HPLCs were performed on a Beckmann System
Gold instrument (model 125S) equipped with a System Gold 166
detector on a Phenomenex Jupiter C18 column (5 µm, 250 × 4.6
mm) using a flow rate of 1 mL/min, with the following solvent
system: 0.1% TFA in H2O (A), 0.1% TFA in MeCN (B)).
Semipreparative RP-HPLC analyses were performed on the same
instrument using a flow rate of 4 mL/min with the same solvent
system on a Phenomenex Jupiter C18 column (10 µm, 250 × 10
mm). Mass spectra were registered on an ESI LCQ advantage mass
spectrometer (Thermo-Finnigan). LC-ESI-MS analyses were per-
formed on a Phenomenex Jupiter C18 column (5 µm, 150 × 2.0
mm) using a flow rate of 200 µL/min on a ThermoFinnigan
Surveyor HPLC system coupled to ESI-MS, using the solvent
system: H2O (A), MeCN (B), 1% TFA in H2O (C). Routine NMR
spectra were acquired on a Varian Inova 700 apparatus. DMSO-d6
was obtained from Aldrich (Milwaukee, U.S.A.). SPPS was
performed in Teflon reactor on a manual synthesizer PLS 4 × 4
(AdvancedChemTech).
Synthesis of Linear Peptides 5–7 (Scheme 1). Peptides were
synthesized in a Teflon reactor fitted with a polystyrene porous
frit. Peptides were prepared using the general Fmoc-SPPS strategy
on preswelled H-L-Thr(t-Bu)-ol-2-chlorotrityl resin. Couplings were
performed by adding 2 equiv of protected amino acid activated by
HATU and 4 equiv NMM in DMF, stirring for 45 min for each
coupling (Scheme 1) and monitoring by the qualitative ninhydrin
(Kaiser) test.52 At the end of the linear peptides synthesis, a
microscale cleavage was performed. Compound 5 and 7 were
treated with TFA/H2O/EDT/phenol (94:2:2:2, 3 h), while compound
6 was treated with TFA/H2O/EDT/phenol (70:26:2:2, 2.30 h). The
resin was filtered off, the solutions were concentrated, the peptide
was precipitated from Et2O, centrifuged dissolved in water, and
lyophilized. RP-HPLC analysis of the crude products revealed the
presence of the linear peptides in approximately 95% purity, without
traces of isomers due to amino acid racemization.
Cyclization and Purification Methods (Scheme 2). The resin
aliquots containing the peptides 5–7 were swollen for 2 h in
anhydrous DCM. The vessels were heated to 45 °C and a DCM
solution of catalyst 3 (0.5 mol equiv. calculated on the basis of 0.5
mmol/g of peptide) was added.11,18,53 The suspension was then
stirred for 48 h at 45 °C in the case of compounds 5 and 6 and
52 h for compound 7. The cyclization was also performed with
catalyst 6 in the same conditions, stirring for 24 h for compounds
5 and 6, and 48 h for compound 7. The resin aliquots were washed
with DCM, DMF, and MeOH, then swelled for 45 min at room
temperature in DMF. Fmoc-Hag was deprotected (2.5 mL of 20%
piperidine in DMF for 5 min, 4 time repeated) and coupled with
Fmoc-D-Phe, as described above, affording the on-resin peptides
8–10, which were deprotected and cleaved, as previously described.
The solutions were concentrated and treated with Et2O giving crude
dicarba-analogues 2, 11, and 12, which were suspended in water,
and the precipitate was centrifuged off. The aqueous solutions of
the peptides 2 and 11, obtained by RCM with 3, were prepurified
by SPE, eluting with an increased percentage of MeOH in H2O
(from 0 to 50%). The fractions containing 0% MeOH (11) and 20%
MeOH (2) were collected, and the solvent was removed under
reduced pressure and lyophilized, giving colorless powders. The
small amount of 12 obtained by RCM with 3 was not subjected to
SPE and was directly purified by RP-HPLC. Differently, among
the peptides achieved with the use of catalyst 4, only 12 needed
prepurification by SPE, collecting the fraction eluted with 20%
MeOH. Compounds 2, 11, and 12 from both methods were then
purified by semipreparative RP-HPLC, and the most abundant
chromatographic peaks were collected (Table 1). For all the
products, HPLC purity was >98%; ESI-MS (2, [M + H]+) calcd,
981.10; found, 981.45; (11, [M + H]+) calcd, 1133.34; found,
1133.35; (12, [M + H]+) calcd, 1027.22; found, 1027.38 (full
experimental data are reported in the Supporting Information).
Conclusions
We have prepared five biostable SRIF analogues with a
dicarba moiety replacing the disulfide bridge of the parent 1.
These analogues were prepared by on-resin RCM of allylgly-
cine-containing linear hepta-peptides throughout a new and more
efficient synthetic route. First- and second-generation Grubbs
catalyst efficiencies were compared. All the analogues were
tested for their affinity toward the sst1–5 receptor subtypes. Two
of them showed moderately high affinity and selectivity for the
sst5 receptor. 3D structures of the active analogues have been
1
determined by H NMR spectroscopy. Conformation-affinity
relationships indicate that helical propensity well correlates with
the peptide-sst5 affinity. A new pharmacophore model for sst5
was developed. The proposed pharmacophore model will play
an important role in designing highly selective peptide as well
as nonpeptide ligands at the sst5 receptor.
Experimental Section
General Procedures. Fmoc-protected amino acids were pur-
chased from Calbiochem-Novabiochem (Laufelfingen, Switzerland).
First- and second-generation Grubbs catalyst and Wilkinson’s
catalyst were obtained from Aldrich. Fmoc-Hag and Fmoc-O-Bzl-
L-tyrosine and Pd(OH)2/C were purchased from Fluka. H-L-Thr(t-
Bu)-ol-2-chlorotrityl resin was purchased from Iris Biotech (Mark-

sst5-SelectiVe Somatostatin Dicarba-Analogues
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 3 519
Reduction Methods. Method A (Scheme 3). A total of 10%
w/w of 20% Pd(OH)2/C was added to the pure peptide 2 (Z-isomer)
and anhydrous MeOH were added to the mixture. The reaction
vessel was purged with N2 and then H2 was flushed. The suspension
was stirred at 30 °C for 24 h, filtered on celite, the solvent was
evaporated under reduced pressure and the crude product was
lyophilized. Analytical RP-HPLC still revealed the presence of the
unsaturated 2 (20% by HPLC). The product was purified by
semipreparative RP-HPLC obtaining 13 (35% yield, HPLC purity
>98%). ESI-MS [M + H]+ calcd, 983.53; found, 983.44.
ing system. Tissue standards containing known amounts of isotopes,
cross-calibrated to tissue-equivalent ligand concentrations, were
used for quantification.63
Acknowledgment. We thank the Italian Ministry of the
University and of the Research (MIUR) for the financial support
(PRIN 2005). We also acknowledge the Ente Cassa di Risparmio
of Florence for funding and for providing a scholarship. We
also thank Prof. P. Frediani and Dr. L. Rosi for giving disposal
of the apparatus for Wilkinson’s reduction.
Method B (Scheme 3). To 9, suspended in a solution of 10%
anhydrous MeOH in DCM, Wilkinson’s catalyst (2.5 mol %) was
added.25 The reaction vessel was pressured with H2 (60 psi) and
left to stand under stirring at 30 °C for 30 h. The resin was
repeatedly washed with DCM and DMF, then D-Phe was coupled
to the resin-tethered reduced cyclopeptide 9. The saturated analogue
14 was cleaved from the resin as previously described and purified
by semipreparative RP-HPLC, affording pure 14 (13% yield, HPLC
purity >98%). ESI-MS [M + H]+ calcd, 1135.35; found, 1135.50.
NMR Spectroscopy. The samples for NMR spectroscopy were
prepared by dissolving the appropriate amount of peptides in 0.40
mL of 1H2O (pH 5.0) and 0.10 mL of DMSO-d6, obtaining 2 mM
solutions. TSP was used as internal chemical shift standard. The
water signal was suppressed by the hard pulse WATERGATE
scheme. NMR experiments were recorded on a Varian Inova-Unity
Supporting Information Available: Analytical data of the
synthesized peptides and NMR data of the analyzed peptides. This
material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Burgus, R.; Brazeau, P.; Vale, W. W. Isolation and Determination of
the Primary Structure of Somatostatin (a Somatotropin Release
Inhibiting Factor) of Bovin Hypothalamic Origin. Advances in Human
Growth Hormone Research; U.S. Government Printing Office: Wash-
ington, D.C., 1973; 44-158, Publ. No. (NIH) 74-612.
(2) Rivier, J. Somatostatin: Total Solid Phase Synthesis. J. Am. Chem.
Soc. 1974, 96, 2986–2992.
(3) Pradayrol, L.; Joernvall, H.; Mutt, V.; Ribet, A. N-Terminally Extended
Somatostatin: The Primary Structure of Somatostatin-28. FEBS Lett.
1980, 109, 55–58.
1
700 MHz at 277.1 K. Complete H NMR chemical shift assign-
ments were effectively achieved for all the analyzed peptides
(Supporting Information) according to the Wüthrich procedure54
via the usual systematic application of DQF-COSY,55,56 TOCSY,57
and NOESY58 experiments recorded in the phase-sensitive mode
using the method from States et al.59
(4) Weckbecker, G.; Lewis, I.; Albert, R.; Schmid, H. A.; Hoyer, D.;
Bruns, C. Opportunities in Somatostatin Research: Biological, Chemi-
cal and Therapeutical Aspects. Nat. ReV. Drug DiscoVery 2003, 2,
999–1018.
(5) Janecka, A.; Zubzycka, M.; Janecki, T. Somatostatin Analogs. J. Pept.
Res. 2001, 58, 91–107.
Typical data block sizes were 2048 addresses in t2 and 512
equidistant t1 values. Before Fourier transformation, the time domain
data matrices were multiplied by shifted sin2 functions in both
dimensions. A mixing time of 70 ms was used for the TOCSY
experiments. NOESY experiments were run, with mixing times in
the range of 150–300 ms. The qualitative and quantitative analyses
(6) Bauer, W.; Briner, U.; Dopfner, W.; Haller, R.; Huguenin, R.; Marbach,
P.; Petcher, T. J.; Pless, J. SMS201–995: A Very Potent and Selective
Analogue of Somatostatin with Prolonged Action. Life Sci. 1982, 31,
1133–1140.
(7) Lamberts, S. W.; Van der Lely, A. J.; Hofland, L. J. New Somatostatin
Analogs: Will They Fulfill Old Promises? Eur. J. Endocrinol. 2002,
146, 701–705.
(8) Deshmukh, M. V.; Voll, G.; Kühlewein, A.; Schmitt, J.; Kessler, H.;
Gemmecker, G. NMR Studies Reveal Structural Differences Between
the Gallium and Yttrium Complexes of DOTA-D-Phe1-Tyr3-Oct-
reotide. J. Med. Chem. 2005, 48, 1506–1514.
(9) Fichna, J.; Janecka, A. Synthesis of Target-Specific Radiolabeled
Peptides for Diagnostic Imaging. Bioconjug. Chem. 2003, 14, 3–17.
(10) Van Binst, G.; Tourwe′, D. Backbone Modifications in Somatostatin
Analogues: Relation Between Conformation and Activity. Pept. Res.
1992, 5, 8–13.
(11) Willliams, R. M.; Liu, J. Asymmetric Synthesis of Differentially
Protected 2,7-Diaminosuberic Acid, a Ring-Closure Metathesis Ap-
proach. J. Org. Chem. 1998, 63, 2130–2132.
(12) Veber, D. F.; Strachan, R. G.; Bergstarand, S. J.; Holly, F. W.;
Homnick, C. F.; Hirschmann, R. Nonreducible Cyclic Analogues of
Somatostatin. J. Am. Chem. Soc. 1976, 98, 2367–2369.
(13) Nutt, R. F.; Veber, D. F.; Saperstein, R. Synthesis of Nonreducible
Bicyclic Analogues of Somatostatin. J. Am. Chem. Soc. 1980, 102,
6539–6545.
(14) Gazal, S.; Gelerman, G.; Ziv, O.; Karpov, O.; Litman, P.; Bracha,
M.; Afargan, M.; Gilon, C. Human Somatostatin Receptor Specificity
of Backbone-Cyclic Analogues Containing Novel Sulfur Building
Units. J. Med. Chem. 2002, 45, 1665–1671.
(15) Hornik, V.; Seri-Levy, A.; Gellerman, G.; Gilon, C. Conformationally
constrained Backbone Cyclized Somatostatin Analogs. PCT WO
9804583 A1, 1998.
(16) Grubbs, R. H. Olefin Metathesis. Tetrahedron 2004, 60, 7117–7140.
(17) Connon, S. J.; Blechert, S. Recent Advances in Alkene Metathesis.
Top. Organomet. Chem. 2004, 11, 93–124.
(18) Miller, S. J.; Blackwell, H. E.; Grubbs, R. H. Application of Ring-
Closing Metathesis to the Synthesis of Rigidified Amino Acids and
Peptides. J. Am. Chem. Soc. 1996, 118, 9606–9614.
(19) Reichwein, J. F.; Versluis, C.; Liskamp, R. B. Synthesis of Cyclic Peptides
by Ring Closing Metathesis. J. Org. Chem. 2000, 65, 6187–6195.
(20) Stymiest, J. L.; Mitchell, B. F; Wong, S.; Vederas, J. C. Synthesis of
Oxytocin Analogues with Replacement of Sulfur by Carbon Gives
Potent Antagonists with Increased Stability. J. Org. Chem. 2005, 70,
7799–7809.
(21) Schmiedeberg, N.; Kessler, H. Reversible Backbone Protection Enables
Combinatorial Solid-Phase Ring-Closing Metathesis Reaction (RCM)
in Peptides. Org. Lett. 2002, 4, 59–62.
of DQF-COSY, TOCSY, and NOESY spectra were obtained with
60 3
the support of the XEASY software package.
J
coupling
HN-HR
1
constants were obtained from 1D H NMR and 2D DQF-COSY
spectra. 3JHR-Hꢀ coupling constants were obtained from 1D 1H NMR
and 2D PE-COSY spectra, with the last performed with a ꢀ flip
angle of 35°.61
Structural Determinations and Computational Modeling. The
NOE-based distance restraints were obtained from NOESY spectra
collected with a mixing time of 200 ms. The NOE cross peaks
were integrated with the XEASY program and were converted into
upper distance bounds using the CALIBA program incorporated
into the program package DYANA.42 50/200 structures were chosen
whose interprotonic distances best fitted NOE-derived distances and
then were refined through successive steps of restrained and
unrestrained energy minimization calculations using the Discover
algorithm (Accelrys, San Diego, CA) and the consistent valence
force field (CVFF).62
The minimization lowered the total energy of the structures; no
residue was found in the disallowed region of the Ramachandran
plot. The final structures were analyzed using the InsightII program
(Accelrys, San Diego, CA). Graphical representation was carried
out with the InsightII program (Accelrys, San Diego, CA).
Determination of Somatostatin Receptor Affinity Profiles.
CHO-K1 and CCL39 cells stably expressing human sst1-sst5
receptors were grown, as described previously.63 Cell membrane
pellets were prepared and receptor autoradiography was done on
20 µm thick pellet sections (mounted on microscope slides), as
described in detail previously.63 For each of the tested compounds,
complete displacement experiments were done with the universal
SRIF radioligand [125I]-[Leu8,D-Trp22,Tyr25]-SRIF-28 using in-
creasing concentrations of the unlabeled compounds ranging from
0.1 to 1000 nmol/L. SRIF-28 was run in parallel as control using
the same increasing concentrations. IC50 values were calculated after
quantification of the data using a computer-assisted image process-

520 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 3
D’Addona et al.
(22) Blackwell, H. E.; Sadowsky, J. D.; Howard, R. J.; Sampson, J. N.;
Chao, J. A.; Steinmetz, W. E.; O’Leary, D. J.; Grubbs, R. H. Ring-
Closing Metathesis of Olefinic Peptides: Design Synthesis, and
Structural Characterization of Macrocyclic Helical Peptides. J. Org.
Chem. 2001, 66, 5291–5302.
(23) Kaptein, B.; Broxterman, Q. B.; Schoemaker, H. E.; Ruties, F. P. J.;
Veerman, J. J. N.; Kamphuis, J.; Peggion, C.; Formaggio, F.; Toniolo,
C. Enantiopure CR-Tetrasubstituted R-Amino Acids. Chemoenzymatic
Synthesis and Application to Turn-Forming Peptides. Tetrahedron
2001, 57, 6567–6577.
(24) Carotenuto, A.; D’Addona, D.; Rivalta, E.; Chelli, M.; Papini, A. M.;
Rovero, P.; Ginanneschi, M. Synthesis of a Dicarba-Analogue of
Octreotide Keeping the Type II′ ꢀ-Turn of the Pharmacophore in Water
Solution. Lett. Org. Chem. 2005, 2, 274–279.
(25) Wheland, A. N.; Elaridi, J.; Mulder, R. J.; Smith, S. V.; Robinson, A.
J; Jackson, W. R. Metal-Catalyzed Tandem Metathesis-Hydrogenation
Reactions for the Synthesis of Cyclic Peptides. Can. J. Chem. 2005,
83, 875–881.
(26) Lewis, I.; Bauer, W.; Albert, R.; Chandramouli, N.; Pless, J.;
Weckbecker, G.; Bruns, C. A Novel Somatostatin Mimic with Broad
Somatotropine Release Inhibitory Factor Receptor Binding and
Superior Therapeutic Potential. J. Med. Chem. 2003, 46, 2334–2344.
(27) Hsieh, H. P.; Wu, Y. T; Chen, S. T; Wang, K. T. Direct Solid-Phase
synthesis of Octreotide Conjugates: Precursors for Use as Tumor-
Targeted Radiopharmaceuticals. Bioorg. Med. Chem. 1999, 7, 1797–
1803.
(28) Harris, P. W. R.; Brimble, M. A.; Gluckman, P. D. Synthesis of Cyclic
Proline-Containing Peptides via Ring-Closing Metathesis. Org. Lett.
2003, 5, 1847–1850.
(29) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Synthesis and Activity
of a New Generation of Ruthenium-Based Olefin Metathesis Catalysts
Coordinated with 1,3-Dimesityl-4,5-dihydroimidazol-2-ylidene Ligands.
Org. Lett. 1999, 1, 953–956.
(30) Morgan, J. P.; Grubbs, R. H. In Situ Preparation of a Highly Active
N-Heterocyclic Carbene-Coordinated Olefin Metathesis Catalyst. Org.
Lett. 2000, 2, 3153–5155.
(31) Stimyest, J. L.; Mitchell, B. F.; Wong, S.; Vederas, J. C. Synthesis of
Biologically Active Dicarba Analogues of the Peptide Hormone
Oxytocin Using Ring-Closing Metathesis. Org. Lett. 2003, 5, 47–49.
(32) Heather, D. M.; Grubbs, R. H. Purification Technique for the Removal
of Ruthenium from Olefin Metathesis Reaction Products. Tetrahedron
Lett. 1999, 40, 4137–4141.
(33) Paquette, L. A.; Schloss, J. D.; Efremov, I.; Fabris, F.; Gallou, F.;
Méndez-Andino, J.; Yang, J. A Convenient Method for Removing all
Highly-Colored Byproducts Generated During Olefin Metathesis
Reaction. Org. Lett. 2000, 2, 1259–1261.
(34) Jourdant, A.; Gonzalez-Zamora, E.; Zhu, J. Wilkinson’s Catalyst
Catalyzed Selective Hydrogenation of Olefin in Presence of Aromatic
Nitro Function: A Remarkable Solvent Effect. J. Org. Chem. 2002,
67, 3163–3164.
(35) Burgess, K.; Van der Donk, W. A. In Encyclopedia of Reagent for
Organic Synthesis; Paquette L. A., Ed.; Wiley: New York, 1995; Vol
2, pp 1253–1261.
(36) Gilon, C.; Huenges, M.; Mathä, B.; Gellerman, G.; Hornik, V.;
Afargan, M.; Amitay, O.; Ziv, O.; Feller, E.; Gamliel, A.; Shohat, D.;
Wanger, M.; Arad, O.; Kessler, H. A Backbone-Cyclic, Receptor
5-Selective Somatostatin Analogue: Synthesis, Bioactivity, and Nuclear
Magnetic Resonance Conformational Analysis. J. Med. Chem. 1998,
41, 919–929.
(37) Rohrer, S. P.; Birzin, E. T.; Mosley, R. T.; Berk, S. C.; Hutchins,
S. M.; Shen, D. M.; Xiong, Y. S.; Hayes, E. C.; Parmar, R. M.; Foor,
F.; Mitra, S. W.; Degrado, S. J.; Shu, M.; Klopp, J. M.; Cai, S. J.;
Blake, A.; Chan, W. W. S.; Pasternak, A.; Yang, L. H.; Patchett, A. A.;
Smith, R. G.; Chapman, K. T.; Schaeffer, J. M. Rapid Identification
of Subtype-Selective Agonists of the Somatostatin Receptor Through
Combinatorial Chemistry. Science 1998, 282, 737–740.
(38) Souers, A. J.; Virgilio, A. A.; Rosenquist, A.; Fenuik, W.; Ellman,
J. A. Identification of a Potent Heterocyclic Ligand to Somatostatin
Receptor Subtype 5 by the Synthesis and Screening of Beta-Turn
Mimetic Libraries. J. Am. Chem. Soc. 1999, 121, 1817–1825.
(39) Feytens, D.; Cescato, R.; Reubi, J. C.; Tourwe, D. New sst4/5-Selective
Somatostatin Peptidomimetics Based on a Constrained Tryptophan
Scaffold. J. Med. Chem. 2007, 50, 3397–3401.
(41) Wynants, C.; Van Blist, G.; Loosli, H. R. SMS 201–995, an
Octapeptide Somatostatin Analogue. Int. J. Pept. Prot. Res. 1985, 25,
615–621.
(42) Güntert, P.; Mumenthaler, C.; Wüthrich, K. Torsion Angle Dynamics
for NMR Structure Calculation with the New Program DYANA. J.
Mol. Biol. 1997, 273, 283–298.
(43) Hutchinson, E. G.; Thornton, J. M. PROMOTIFsA Program to
Identify and Analyze Structural Motifs in Proteins. Protein Sci. 1996,
5, 212–220.
(44) Melacini, G.; Zhu, Q.; Goodman, M. Multiconformational NMR
Analysis of Sandostatin (Octreotide): Equilibrium between ꢀ-Sheet
and Partially Helical Structures. Biochemistry 1997, 36, 1233–1241.
(45) Melacini, G.; Zhu, Q.; Osapay, G.; Goodman, M. A. Refined Model
for the Somatostatin Pharmacophore: Conformational Analysis of
Lanthionine-Sandostatin Analogs. J. Med. Chem. 1997, 40, 2252–2258.
(46) Tyndall, J. D. A.; Pfeiffer, B.; Abbenante, G.; Fairlie, D. P. Over One
Hundred Peptide-Activated G Protein-Coupled Receptors Recognize
Ligands with Turn Structure. Chem. ReV. 2005, 105, 793–826.
(47) Grace, C. R. R.; Erchegyi, J.; Koerber, S. C.; Reubi, J. C.; Rivier,
J. E.; Riek, R. Novel sst2-Selective Somatostatin Agonists. Three-
Dimensional Consensus Structure by NMR. J. Med. Chem. 2006, 49,
4487–4496.
(48) Brady, S. F.; Paleveda, W. J.; Arison, B. H.; Saperstein, R.; Brady,
E. J.; Raynor, K.; Reisine, T.; Veber, D. F.; Freidinger, R. M.
Approaches to Peptidomimetics which Serve as Surrogates for the cis
Amide Bond: Novel Disulfide-Constrained Bicyclic Hexapeptide
Analogs of Somatostatin. Tetrahedron 1993, 49, 3449–3466.
(49) Grace, C. R. R.; Durrer, L.; Koerber, S. C.; Erchegyi, J.; Reubi, J. C.;
Rivier, J. E.; Riek, R. Somatostatin Receptor 1 Selective Analogues.
4. Three-Dimensional Consensus Structure by NMR. J. Med. Chem.
2005, 48, 523–523.
(50) Gairì, M.; Saiz, P.; Madurga, S.; Roig, X.; Erchegyi, J.; Koerber, S. C.;
Reubi, J. C.; Rivier, J. E.; Giralt, E. Conformational Analysis of a
Potent SSTR3-Selective Somatostatin Analogue by NMR in Water
Solution. J. Pept. Sci. 2006, 12, 82–91.
(51) Grace, C. R. R.; Koerber, S. C.; Erchegyi, J.; Reubi, J. C.; Rivier, J.;
Riek, R. Novel sst4-Selective Somatostatin (SRIF) Agonists. 4. Three-
Dimensional Consensus Structure by NMR. J. Med. Chem. 2003, 46,
5606–5618.
(52) Stewart, J. M.; Young, J. D. In Solid Phase Peptide Synthesis, 2nd
ed.; Pierce Chemical Company: Rockford, IL, 1984.
(53) Gao, Y.; Lane-Bell, P.; Vederas, J. C. Stereoselective Synthesis of
meso-2,6-Diaminopimelic Acid and Its Selectively Protected Deriva-
tives. J. Org. Chem. 1998, 63, 2133–2143.
(54) Wüthrich, K. NMR of Proteins and Nucleic Acids; John Wiley & Sons,
Inc.: New York, 1986.
(55) Piantini, U.; Sørensen, O. W.; Ernst, R. R. Multiple Quantum Filters
for Elucidating NMR Coupling Network. J. Am. Chem. Soc. 1982,
104, 6800–6801.
(56) Marion, D.; Wüthrich, K. Application of Phase Sensitive Two-
Dimensional Correlated Spectroscopy (COSY) for Measurements of
1H-1H Spin-Spin Coupling Constants in Proteins. Biochem. Biophys.
Res. Commun. 1983, 113, 967–974.
(57) Braunschweiler, L.; Ernst, R. R. Coherence Transfer by Isotropic
Mixing: Application to Proton Correlation Spectroscopy. J. Magn.
Reson. 1983, 53, 521–528.
(58) Jenner, J.; Meier, B. H.; Bachman, P.; Ernst, R. R. Investigation of
Exchange Processes by Two-Dimensional NMR Spectroscopy.
J. Chem. Phys. 1979, 71, 4546–4553.
(59) States, D. J.; Haberkorn, R. A.; Ruben, D. J. A Two-Dimensional
Nuclear Overhauser Experiment with Pure Absorption Phase Four
Quadrants. J. Magn. Reson. 1982, 48, 286–292.
(60) Bartels, C.; Xia, T.; Billeter, M.; Güntert, P.; Wüthrich, K. The Program
XEASY for Computer-Supported NMR Spectral Analysis of Biological
Macromolecules. J. Biomol. NMR 1995, 6, 1–10.
(61) Mueller, L. P. E. COSY, a Simple Alternative to E. COSY. J. Magn.
Reson. 1987, 72, 191–196.
(62) Maple, J.; Dinur, U.; Hagler, A. T. Derivation of Force Fields for
Molecular Mechanics and Dynamics from Ab Initio Energy Surface.
Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 5350–5354.
(63) Reubi, J. C.; Schar, J. C.; Waser, B.; Wenger, S.; Heppeler, A.; Schmitt,
J. S.; Macke, H. R. Affinity Profiles for Human Somatostatin Receptor
Subtypes SST1-SST5 of Somatostatin Radiotracers Selected for
Scintigraphic and Radiotherapeutic Use. Eur. J. Nucl. Med. 2000, 27,
273–282.
(40) Arison, B. H.; Hirschmann, R.; Veber, D. F. Inferences About the
Conformation of Somatostatin at a Biologic Receptor Based on NMR
Studies. Bioorg. Chem. 1978, 7, 447–451.
JM070886I