Optimization of a somatostatin mimetic via constrained amino acid and backbone incorporation

Article abstract of DOI:10.1016/S0960-894X(00)00552-7

Constrained analogues 5-7 of the potent and subtype selective somatostatin mimetic 1 were prepared by incorporating conformational constraints into the nine-membered heterocyclic scaffold. Each constrained peptidomimetic showed an altered activity profile relative to lead compound 1, with compound 7 exhibiting a 25-fold and 2-fold binding enhancement against somatostatin receptor subtypes sst4 and sst5, respectively. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00552-7

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2731±2733
Optimization of a Somatostatin Mimetic via Constrained Amino
Acid and Backbone Incorporation
Andrew J. Souers,a Asa Rosenquist,a Emma M. Jarvie,b Mark Ladlow,b
Wasyl Feniukb and Jonathan A. Ellmana,*
aDepartment of Chemistry, University of California, Berkeley, CA 94720, USA
bGlaxo Wellcome Cambridge Chemistry Laboratory, University Chemical Laboratory, Lens®eld Road, Cambridge CB2 1EW, UK
Received 14 August 2000; accepted 26 September 2000
AbstractÐConstrained analogues 5±7 of the potent and subtype selective somatostatin mimetic 1 were prepared by incorporating
conformational constraints into the nine-membered heterocyclic sca€old. Each constrained peptidomimetic showed an altered
activity pro®le relative to lead compound 1, with compound 7 exhibiting a 25-fold and 2-fold binding enhancement against soma-
tostatin receptor subtypes sst4 and sst5, respectively. # 2000 Elsevier Science Ltd. All rights reserved.
Although peptides and proteins mediate numerous bio-
logical processes, their use as therapeutic agents and
biological probes is often accompanied by several
drawbacks, including poor biostability, poor receptor
subtype selectivity, and unfavorable absorption proper-
ties.1 Consequently, much e€ort has been dedicated to
the design and synthesis of selective, high-anity
ligands by the display of appropriate amino acid side
chains upon peptidomimetic sca€olds.2,3 Previously we
reported the ®rst successful design and implementation
of a b-turn mimetic sca€old that displays side chains
corresponding to the i+1, i+2, and i+3 positions of
the b-turn structure, and that is amenable to the rapid
and parallel solid-phase synthesis of multiple derivatives
in high yield.4 We have since demonstrated the ecacy
of this sca€old for producing small molecule ligands to
several distinct receptor targets.4b,5
non-peptidic sst47a and sst57b ligands has only recently
been achieved. To address this problem, we synthesized
and evaluated a collection of somatostatin mimetics
based on our b-turn sca€old resulting in the identi®ca-
tion of the selective and potent sst5 ligand
1
(IC50=87 nM) as shown in Figure 1.4b Herein we report
a study to further enhance potency and selectivity of 1
by introducing conformational constraints into the
¯exible nine-membered heterocyclic sca€old.
The lysine side chain in mimetic 1 (Fig. 1) presented an
opportunistic position for the incorporation of con-
strained amino acids. Since both proline and pipecolic
acid could be used e€ectively in the solid-phase synth-
esis sequence, constrained lysine analogues of these
amino acids (Fig. 2) were selected. Ligand 1 possesses
the unnatural d stereochemistry at the i+2 position,
therefore the ®ve-membered lysine analogues 2 and 3
retained this stereochemistry at the a-carbon, but dif-
fered in the stereochemistry at the 4-position. It was
hoped that this variable would probe the conforma-
tional preference of the side chain while locking the
heterocyclic backbone. The six-membered lysine deriva-
tive 4 was also designed to retain the unnatural d ste-
reochemistry at the a-carbon. In addition, the side chain
would be oriented in a manner cis to the carboxylate in
order to `tie back' the primary amine. The three con-
strained lysine surrogates were synthesized according
to literature methods with only minor modi®cations,8
and were then incorporated into the protected turn
mimetics according to our previously reported solid-
phase synthesis sequence.4 Standard functional group
In one e€ort we targeted the human somatostatin
receptors 1±5.4b The endogenous peptide somatostatin
(SRIF) is a cyclic tetradecapeptide that has attracted
considerable attention as a potential therapeutic agent.
SRIF modulates the secretion of several hormones,
including growth hormone, insulin, and glucagon.6
However, the very short biological half-life and lack of
subtype speci®city of SRIF necessitates that longer act-
ing and more selective small molecule mimetics be
developed, and the identi®cation of potent and selective
*Corresponding author. Tel.: +1-510-642-4488; fax: +1-510-642-
8369; e-mail: jellman@uclink.berkley.edu
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00552-7

2732
A. J. Souers et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2731±2733
manipulation/global deprotection a€orded the desired
mimetics 5, 6, and 7 (Fig. 3), which were then puri®ed
by preparative HPLC and characterized prior to biolo-
gical evaluation.9
Table 1, control compound 1 showed complete inhibi-
tion of SRIF binding to sst5 at 1 mM, while compounds
5 and 6 inhibited binding to sst5 at 50 and 56%,
respectively. Although both second-generation mimetics
exhibited considerably reduced levels of inhibition, it is
interesting to note that the highest inhibitory value for
sst5 was conserved. Bicyclic compound 7 exhibited near-
complete and complete inhibition of sst4 and sst5,
respectively, while showing moderate to low inhibition
for the remaining subtypes. Finally, the backbone-con-
strained heterocycle 10 possessed modest activity for the
various subtypes.
The ¯exible alkanethiol backbone provided an addi-
tional site for introducing conformational constraints.
To assess the ecacy of backbone constraints, we
designed a solution-phase synthesis that employed the
aromatic disul®de 8. This backbone should constrain
the heterocyclic mimetic while simultaneously introdu-
cing an additional hydrophobic contact. Notably, SRIF
and potent cyclic peptide analogues of SRIF require
hydrophobic residues at the entering i residue of their
corresponding b-turn regions.10 Constrained mimetic 10
was synthesized according to Scheme 1 and puri®ed by
preparative HPLC prior to biological evaluation.9
Since compound 7 showed very promising inhibition
data, IC50 values of this compound were determined
and compared to somatostatin for each receptor sub-
type (Table 2). The inhibitory constants of 7 for sst4
and sst5 are both 41 nM. The dual activity and selectiv-
ity for these subtypes over the other receptor subtypes is
quite interesting given that these two receptors come
from distinct groups.6 Furthermore, the IC50 value of
41 nM is a 2-fold enhancement over compound 1 for
sst5, but over 25-fold for sst4. Thus, while the con-
strained amino acid clearly imparts enhanced potency
on mimetic 7 for the two subtypes, it has the e€ect of
lowering the selectivity for sst5 over the non-homo-
logous sst4 when compared to the parent compound 1.
The four individual mimetics were then tested at a con-
centration of 1 mM in a competitive radioligand binding
assay ({125I}-{Tyr11}-SRIF) in membranes from CHO-
K1 cells expressing human sst1±sst5. As displayed in
In summary, we have synthesized and evaluated a small
collection of constrained second-generation mimetics
based on lead SRIF mimetic 1. Incorporation of the
three constrained amino acids and aromatic disul®de
backbone into the ¯exible synthetic sequence provided
compounds with considerably altered biological pro®les
towards the somatostatin receptor subtypes 1±5. Fur-
thermore, this e€ort resulted in the identi®cation of the
bicyclic mimetic 7 as a more potent sst5 ligand than par-
ent compound 1. Further optimization of the constrained
SRIF mimetic will be reported in due course.
Figure 1. sst5 selective turn mimetic.
Figure 2. Constrained amino acids.
Figure 3. Second-generation constrained mimetics.
Scheme 1. Reagents. (a) BH3, 0±25 ꢀC; (b) (i) Dithiobisbenzothiazole, CHCl3, then t-BuSH; (ii) Ms2O, pyridine, CH2Cl2, then 1-napthylmethyl-
amine; (c) Pybop, HOAt, Fmoc-d-Lys-(N-Phth)-OH, i-Pr2EtN, DMF; (d) piperidine/CH2Cl2; (e) EDC, HOBt, a-Br-Trp-(N-Boc)-OH; (f) i-Bu3P,
9:1 dioxane/H2O, then TMG, 55 ꢀC; (g) H2NNH2, EtOH, re¯ux.

A. J. Souers et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2731±2733
2733
Table 1. % Inhibition at 1 mMa
Rosenquist, A.; Fenuik, W.; Ellman, J. A. J. Am. Chem. Soc.
1999, 121, 1817.
Compd
sst1
sst2
sst3
sst4
sst5
5. (a) Souers, A. J.; Virgilio, A. A.; Schurer, S.; Ellman, J. A.;
Kogan, T. P.; West, H. E.; Ankener, W.; Vanderslice, P.
Bioorg. Med. Chem. Lett. 1998, 8, 2297. (b) Haskell-Luevano,
C.; Rosenquist, A.; Souers, A. J.; Khong, K. C.; Ellman, J.;
Cone, R. D. J. Med. Chem. 1999, 42, 4380.
1
5
6
7
10
64
28
13
64
36
33
20
7
74
7
42
16
19
44
13
43
17
12
99
41
99
50
56
89
64
6. Patel, Y. C. J. Endocrinol. Invest. 1997, 20, 348.
7. (a) Ankerson, M.; Crider, M.; Liu, S.; Ho, B.; Andersen, H.
S.; Stidsen, C. J. Am. Chem. Soc. 1998, 120, 1368. (b) Berk, S.
C.; Rohrer, S. P.; Degrado, S. J.; Birzin, E. T.; Mosley, R. T.;
Hutchins, S. M.; Pasternak, A.; Schae€er, J. M.; Underwood,
D. J.; Chapman, K. T. J. Comb. Chem. 1999, 1, 388.
8. (a) For the synthesis of amino acids 2 and 3: Murray, P. J.;
Starkey, I. D.; Davies, J. E. Tetrahedron Lett. 1998, 39, 6721.
(b) For the synthesis of amino acid 4: Souers, A. J.; Ellman, J.
A. J. Org. Chem. 2000, 65, 1222.
aCell membranes (2±30 mg protein) were incubated with 0.03 nM [125I]-
[Tyr11]-SRIF and increasing concentrations of SRIF or SRIF ana-
logues for 90 min at rt. Nonspeci®c binding was de®ned with 1 mM
SRIF. The assay was terminated by rapid ®ltration through Whatman
GF/C glass ®bre ®lters soaked in 0.5% polyethylenimine (PEI), fol-
lowed by 3Â3 mL washes of 50 mM Tris±HCl pH 7.4 containing 5 mg/
mL bovine serum albumin (BSA). Radioactivity in the ®lters was
determined using a Canberra Packard Cobra II Auto-# counter.
9. (5) 1H NMR (MeOH-d4, 500 MHz) d 7.83±7.77 (m, 3H),
7.59 (m, 1), 7.46 (m, 2H), 7.36 (d, J=5.4 Hz, 2H), 7.28 (m,
1H), 7.12 (s, 1H), 7.07±6.92 (m, 2H), 5.78 (d, J=13.7 Hz, 1H),
5.32 (m, 1H), 4.24 (m, 1H), 4.15±3.92 (m, 2H), 3.84 (m, 1H),
3.72±3.51 (m, 2H), 3.41 (m, 1H), 3.33±3.31 (m, 1H), 3.28±3.03
(m, 2H), 2.96±2.72 (m, 2H), 2.50 (m, 1H), 2.37 (m, 1H), 2.18
(m, 1H), 1.89 (m, 2H). HRMS (FAB) calcd for C31H34N4O2S
Table 2. IC50 values of compounds 1, 7, and SRIF (nM)a
Compd
sst1
sst2
sst3
sst4
sst5
SRIF
1
7
0.65
501
407
0.07
1585
275
0.72
3090
1258
1.78
1047
41
0.56
87
41
1
(M+1): 527.2481. Found: 527.2480. (6) H NMR (MeOH-d4,
500 MHz) d 7.92±7.74 (m, 3H), 7.62 (d, J=7.8 Hz, 1H), 7.41±
7.34 (m, 4H), 7.29 (d, J=8.0 Hz, 1H), 7.00 (s, 1H), 7.08±6.92
(m, 2H), 5.73 (d, J=14.8 Hz, 1H), 5.30 (d, J=8.1 Hz, 1H),
4.19±3.38 (m, 4H), 3.43 (dd, J=14.6, 6.2 Hz, 1H), 3.33±3.31
(m, 1H), 3.18±3.01 (m, 3H), 2.72±2.64 (m, 3H), 2.28 (dd,
J=11.6, 5.3 Hz, 1H), 2.17 (dd, J=12.9, 7.3 Hz, 1H), 1.70 (m,
1H), 1.60 (m, 2H). HRMS (FAB) calcd for C31H34N4O2S
aSee conditions for Table 1.
Acknowledgements
1
(M+1): 527.2481. Found: 527.2475. (7) H NMR (MeOH-d4,
This work was generously supported by the National
Institutes of Health (GM-53696). A.J.S. is grateful to
the American Chemical Society, Organic Division, for
their generous support.
500 MHz) d 8.02 (m, 1H), 7.96±7.80 (m, 2H), 7.60 (m, 1H),
7.55±7.40 (m, 4H), 7.25 (m, 1H), 7.11 (s, 1H), 7.03 (m, 1H),
6.96 (m, 1H), 5.73 (d, J=14.7 Hz, 1H), 5.21 (d, J=5.45 Hz,
1H), 4.41 (d, J=10.7 Hz, 1H), 4.06 (m, 2H), 3.95 (m, 1H), 3.47
(m, 1H), 3.40 (dd, J=7.02, 7.49 Hz, 1H), 3.11 (m, 2H), 2.67
(m, 1H), 2.59±2.41 (m, 3H), 1.98 (m, 1H), 1.82 (m, 1H), 1.73±
1.58 (m, 2H), 1.48±1.07 (m, 5H). HRMS (FAB) calcd for
C33H38N4O2S (M+1): 555.2794. Found: 555.2800. (10) 1H
NMR (CDCl3, 500 MHz) d 9.30 (br s, 1H), 7.87 (m, 1H), 7.78
(m, 1H), 7.77±7.63 (m, 2H), 7.59±7.30 (m, 5H), 7.27±7.16 (m,
5H), 7.12±6.89 (m, 2H), 6.00 (m, 1H), 5.18 (m, 1H), 4.87 (m,
1H), 4.66 (m, 1H), 4.06 (m, 1H), 3.97 (m, 2H), 3.70 (m, 1H),
3.63±3.37 (m, 3H), 3.34±3.12 (m, 2H), 1.88±1.65 (m, 3H),
1.65±1.50 (m, 3H). HRMS (FAB) calcd for C35H36N4O2S
(M+1): 577.2637. Found: 577.2641.
References and Notes
1. Sawyer, T. K. In Structure-Based Design: Diseases, Targets,
Techniques and Developments; Veerapandian. P., Ed.; Marcel
Dekker: New York, 1997; p 559.
2. Pauletti, G. M.; Gangwar, S.; Siahaan, T. J.; Aube, J.;
Borchordt, R. T. Adv. Drug Deliver. Rev. 1997, 27, 235.
3. Eguchi, M.; Lee, M. S.; Nakanishi, H.; Stasiak, M.; Lovell,
S.; Kahn, M. J. Am. Chem. Soc. 1999, 121, 12204.
10. Brady, S. F.; Paleveda, W. J.; Arison, B. H.; Saperstein,
R.; Brady, E. J.; Raynor, K.; Reisine, T.; Veber, D. F.; Frei-
dinger, R. M. Tetrahedron 1993, 49, 3449.
4. (a) Virgilio, A. A.; Schurer, S. C.; Ellman, J. A. Tetra-
hedron Lett. 1996, 37, 6961. (b) Souers, A. J.; Virgilio, A. A.;