Impact of the Proline Residue on Ligand Binding of Neurotensin Receptor2 (NTS2)-Selective Peptide-Peptoid Hybrids

Article abstract of DOI:10.1002/cmdc.201300054

To investigate the binding mode and structure-activity relationships (SARs) of selective neurotensin receptor2 (NTS2) ligands, novel peptide-peptoid hybrids that simulate the function of the endogenous ligand were developed. Starting from our recently described NTS2 ligands of type 1, structural variants of type 2 and the metabolically stable analogues 3a,b were developed. Replacement of the proline unit by a collection of structural surrogates and evaluation of the respective molecular probes for NTS2 affinity and selectivity indicated similar SARs as described for NT(8-13) derivatives bound to the subtype NTS1. Peptide-peptoid hybrids 2d, 3a,b showed substantial NTS2 binding affinity (K_i=8.1-16nM) and 2400-8600-fold selectivity over NTS1. The thiazolidine derivative 3b showed metabolic stability over 32h in a serum degradation assay. In an inositol phosphate accumulation assay, the neurotensin mimetics 3a and 3b displayed an inhibition of constitutive activity exceeding 1.7-2.0times the activity of NT(8-13). The fluorinated derivative 3a could afford attractive opportunities to detect NTS2 by ¹⁹F magnetic resonance imaging.

Full text of DOI:10.1002/cmdc.201300054

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201300054
Impact of the Proline Residue on Ligand Binding of
Neurotensin Receptor 2 (NTS2)-Selective Peptide–Peptoid
Hybrids
Cornelia Held,[a] Harald Hꢀbner,[a] Ralf Kling,[a] Yvonne A. Nagel,[b] Helma Wennemers,[b] and
Peter Gmeiner*[a]
To investigate the binding mode and structure–activity rela-
tionships (SARs) of selective neurotensin receptor 2 (NTS2) li-
gands, novel peptide–peptoid hybrids that simulate the func-
tion of the endogenous ligand were developed. Starting from
our recently described NTS2 ligands of type 1, structural var-
iants of type 2 and the metabolically stable analogues 3a,b
were developed. Replacement of the proline unit by a collec-
tion of structural surrogates and evaluation of the respective
molecular probes for NTS2 affinity and selectivity indicated
similar SARs as described for NT(8–13) derivatives bound to
the subtype NTS1. Peptide–peptoid hybrids 2d, 3a,b showed
substantial NTS2 binding affinity (Ki =8.1–16 nm) and 2400–
8600-fold selectivity over NTS1. The thiazolidine derivative 3b
showed metabolic stability over 32 h in a serum degradation
assay. In an inositol phosphate accumulation assay, the neuro-
tensin mimetics 3a and 3b displayed an inhibition of constitu-
tive activity exceeding 1.7–2.0 times the activity of NT(8–13).
The fluorinated derivative 3a could afford attractive opportuni-
ties to detect NTS2 by 19F magnetic resonance imaging.
Introduction
NT(8–13) (H-Arg-Arg-Pro-Tyr-Ile-Leu-OH) is the fully active C-
terminal hexapeptide of the neuropeptide neurotensin,[1–4]
which exerts its function through activation of the G-protein-
coupled receptors (GPCRs) NTS1[5,6] and NTS2[7,8] or through
subtype 3 (NTS3), which belongs to the Vps10p family of sort-
ing receptors.[9] Grisshammer and co-workers recently deter-
mined the crystal structure of neurotensin receptor 1 (NTS1)
bound to the activating peptide NT(8–13).[10] This excellent
work provides valuable insight into the shape and electrostatic
properties of the binding pocket and the bioactive conforma-
tion of the endogenous ligand. Moreover, the crystal structure
will serve as a guideline for the rational design of subtype-se-
lective NT(8–13) analogues. Neurotensin receptor 2 (NTS2),
which is structurally related to NTS1, has attracted growing in-
terest because its physiological effects range from hypother-
mia to antipsychotic properties.[11] Promoting m-opioid-inde-
pendent antinociception,[12] NTS2 is mainly involved in the
modulation of tonic pain sensitivity.[13] Thus, NTS2-selective li-
gands could serve as valuable therapeutic agents.
to the highly potent and NTS2-selective peptide–peptoid hy-
brids of type 1 (Figure 1).[14] Except for an N-2-(4-hydroxyphe-
Figure 1. Structural modification leading to target compounds of type 2.
In the course of our recent studies, identification of screen-
ing hits and the chemical synthesis of structural variants led us
nyl)ethylglycine peptoid moiety[15] (N-homotyrosine, Nhtyr) in
place of the tyrosine unit of the parent peptide, the sequence
of 1a is identical to that of the hexapeptide NT(8–13). Further
replacements of the basic amino acids with N-methylarginine
and lysine resulted in the discovery of the metabolically stable
derivative 1b.[16]
SAR studies involving alanine,[17] d-amino acid,[18,19] and
homo-b-amino acid scans[20] on NT(8–13), as well as the intro-
duction of conformational constraints,[21] indicated a crucial
role for the proline residue of neurotensin and NT(8–13) in
[a] Dr. C. Held, Dr. H. Hꢀbner, R. Kling, Prof. Dr. P. Gmeiner
Department of Chemistry and Pharmacy
Emil Fischer Center, Friedrich Alexander University
Schuhstraße 19, 91052 Erlangen (Germany)
E-mail: peter.gmeiner@medchem.uni-erlangen.de
[b] Dr. Y. A. Nagel, Prof. Dr. H. Wennemers
ETH Zꢀrich, Laboratory for Organic Chemistry
Wolfgang Pauli Strasse 10, 8093, Zꢀrich (Switzerland)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201300054.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 772 – 778 772

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
NTS1 binding. This was corroborated by the crystal structure of
the GPCR–peptide complex, which shows a tight hydrophobic
binding site for the pyrrolidine ring of the proline residue
(Figure 2).
Scheme 1. Reagents and conditions: a) 1. Fmoc-AA-OH, PyBOP, DIPEA, HOBt,
DMF, m-wave, except for proline surrogates: Fmoc-Xaa-OH, HATU, DIPEA, m-
wave, 2. piperidine/DMF, m-wave; b) TFA/phenol/H2O/TIS, RT, 2 h.
The ring pucker of proline can be strongly influenced by
substituents at position Cg.[24–28] This conformational control,
which is caused by stereoelectronic,[25,26] steric,[27] and/or hy-
drogen bonding effects,[28] can be used as a SAR investigation
tool. For example, with electron-withdrawing groups at Cg
(e.g., fluorine as the most electronegative element), the sub-
stituents of the CdÀCg bond are not arranged “anti-periplanar”
(as expected based on steric considerations) but “gauche”.
Thus, an electron-withdrawing substituent at a 4R-configured
center stabilizes a Cg exo pucker, and a 4S-configured substitu-
ent induces a Cg endo pucker[25,26] (Figure 3).
Figure 2. Close view of the crystal structure of the NTS1–NT(8–13) complex:
NTS1 is represented by green ribbons, the ligand NT(8–13) is shown in blue
sticks. The proline residue of NT(8–13) and the residues of the receptor’s hy-
drophobic binding site are highlighted as blue and grey spheres, respective-
ly. The image was prepared with PyMOL.
To learn more about the binding mode and the bioactive
conformation of NTS2-selective NT(8–13) analogues, we envi-
sioned the synthesis of target compounds of type 2 incorpo-
rating a collection of proline surrogates and to study these
molecular probes for NTS2 affinity and selectivity over NTS1.
Using alternative cyclic systems and substituents at position 4
of the pyrrolidine ring that can control ring pucker, we sought
to investigate the impact of shape and conformational proper-
ties, respectively, on receptor–ligand recognition.
Figure 3. Stereo projection of 4-substituted prolines.
These properties have also been described for azido
groups.[26,29] Using SPPS, the chiral building blocks Fmoc-pro-
tected (2S,4S)-4-azidoproline, (2S,4R)-4-azidoproline,[26] (2S,4R)-
4-fluoroproline, and (2S,4R)-4-fluoroproline[25,30] were incorpo-
rated to afford the peptide–peptoid hybrids 2e–h.
To investigate the effect of charge and hydrogen bonding,
amino- and acetamido-substituted analogues were also syn-
thesized.[28] The peptide–peptoid hybrids containing (2S,4S)-4-
aminoproline, (2S,4R)-4-aminoproline, (2S,4S)-4-acetamidopro-
line and (2S,4R)-4-acetamidoproline were prepared from the
fully protected solid-phase-bound (2S,4S)-4-azidoproline and
(2S,4R)-4-azidoproline derivatives 4a,b, respectively. Using tri-
methylphosphine, intermediates 4a,b were reduced to the pri-
mary amines 4c,d, which could be N-deprotected and cleaved
from the resin to obtain aminoprolines 2i,j.[31,32] N-Acetylation
of 4c,d by acidic anhydride and subsequent N-deprotection
and cleavage resulted in the formation of acetamidoproline
derivatives 2k,l (Scheme 2).
To examine the effect of fluorine substituents in greater
detail, difluoroproline[33] was used to synthesize peptide–pep-
toid 2m (Scheme 1). Representatively, (2S,4R)-4-fluoroproline
and thiazolidine-2-carboxylic acid were converted into the pep-
tide–peptoids 3a,b, which contain the metabolically stable N-
terminal motif N-methylarginine-lysine instead of Arg-Arg (see
Supporting Information).
Results and Discussion
Chemistry
Several proline surrogates were incorporated into NT(8–13)
and investigated for NTS1 receptor binding.[22,23] A significant
decrease in affinity was observed with the use of 3,4-dehydro-
proline, azetidine-2-carboxylic acid, and pipecolic acid. Thiazoli-
dine-2-carboxylic acid proved to be a potent bioisostere, lead-
ing to a slight increase in NTS1 binding affinity. To determine if
the pharmacophore of the peptide–peptoid hybrids 1 causes
analogous SARs at NTS2, the commercially available proline
surrogates mentioned above were used to synthesize target
compounds 2a–d. Solid-phase peptide synthesis (SPPS) was
started from Fmoc-leucinyl-loaded Wang resin. The peptide–
peptoid hybrids 2a–d were prepared by microwave irradiation
to accelerate Fmoc deprotection, PyBOP-promoted acylation,
and HATU-induced proline surrogate incorporation (Scheme 1).
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 772 – 778 773

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
uration, is favored. Introduction of a primary amino group at
position 4, facilitating formation of a cationic center upon pro-
tonation, led to a drastic loss of NTS2 affinity for 2i and 2j. In-
terestingly, the geminally substituted difluoride 2m (Ki =
30 nm) showed higher NTS2 binding affinity than the 4R fluoro
analogue 2 f, indicating that the exo ring pucker might be
more favorable; however, a fluorine substituent will be bound
better by the receptor if located at the 4S position of the
ligand. To take advantage of the metabolic stability that results
from a formal exchange of the arginine residues at the N ter-
minus by the sequence N-Me-Arg-Lys, the peptide–peptoid hy-
brids 3a and 3b were included into the collection of test com-
pounds. Interestingly, the NTS2 ligands showed high binding
affinity (8.1 nm for 3a, 16 nm for 3b) and selectivity over NTS1
(>2000). The thiazolidine derivative 3b was subjected to
a serum degradation assay that indicated metabolic stability
over 32 h (see Supporting Information).
To investigate the intrinsic activity of the peptide–peptoid
hybrids 2d, 2 f, 2m, 3a, and 3b in comparison with the lead
compound 1a and NT(8–13),[34] we used a previously estab-
lished inositol phosphate (IP) accumulation assay (Table 2).[16,35]
Gaq-promoted modulation of IP3 production in HEK293 cells
transiently expressing NTS2 was recorded. Constitutive activity
could be attenuated by the reference agents neurotensin,
NT(8–13), and 1a. Remarkably, all peptide–peptoid hybrids 2d,
2 f, 2m, 3a, and 3b displayed strong intrinsic activity, substan-
tially exceeding the maximal efficacy of the endogenous
ligand neurotensin. The strongest effect was exerted by the
thiazolidine derivative 3b, showing a maximum effect that is
twofold stronger than that of NT(8–13).
Scheme 2. Reagents and conditions: a) H2O, Me3P, THF, RT, 2 h; b) DIPEA,
Ac2O, CH2Cl2, RT, 30 min; c) 1. piperidine/DMF, m-wave, 2. TFA/phenol/H2O/
TIS, RT, 4 h.
Biological investigations
Radioligand binding studies were conducted to evaluate the
peptide–peptoids 2a–m and 3a,b for NTS1 and NTS2 affinity
(Table 1). Binding data were determined using the radioligand
[3H]neurotensin and stably transfected Chinese hamster ovary
(CHO) cells expressing human NTS1. [3H]NT(8–13) was used for
binding assays investigating human NTS2, which was transient-
ly transfected in human embryonic kidney (HEK293) cells.
The binding data of dehydropyrrolidine and piperidine deriv-
atives 2a and 2c show that slight modifications to the geome-
try of the proline-mimicking ring system made by introduction
of a double bond or an additional methylene unit lead to
slightly decreased NTS2 affinity (Ki =23 and 24 nm) relative to
the lead structure 1a. This was also observed for the respec-
tive NT(8–13) derivatives at NTS1.[22] Employing azetidine-2-car-
boxylic acid instead of proline resulted in a 10-fold decrease in
NTS2 binding affinity for 2b which is similar to the 50-fold loss
of affinity in NTS1 binding described for the respective NT(8–
13) analogue.[22] Formal exchange of the pyrrolidine moiety of
1a by a thiazolidine unit resulted in preservation of NTS2 bind-
ing and an increase in selectivity; the thia-analogue 2d
showed Ki values of 10 and 66000 nm for NTS2 and NTS1, re-
spectively.
Conclusions
Using the NTS2 ligands of type 1 as lead compounds, structur-
al variants of type 2 were synthesized. Replacement of the pro-
line unit with a collection of structural surrogates and evalua-
tion of the respective molecular probes for NTS2 affinity and
selectivity indicated a related binding mode and similar SARs
as described for NT(8–13) derivatives bound to the subtype
NTS1. The peptide–peptoid hybrids of type 2e–m suggest that
NTS2 has a preference for an exo-puckered conformation of
the pyrrolidine ring. The thiazolidine-2-carboxlylic acid deriva-
tive 2d and the N-terminally modified (2S,4R)-4-fluoroproline
NT(8–13) analogue 3a show excellent receptor binding affinity
and subtype selectivity. These SAR investigations will be of par-
ticular interest in drug discovery because selective NTS2 li-
gands could serve as useful molecular probes and analgesic
agents without the side effects of opioid-mediated antinoci-
ception. Furthermore, the fluorinated derivative 3a might pro-
vide attractive opportunities to develop NTS2-selective 19F
magnetic resonance imaging agents.
Stereospecific substitution of the proline Cg position of 1a
for fluoro, azido, and acetamido groups resulted in respective
Ki values of Ki =74, 83, and 52 nm for the 4R isomers 2 f, 2h,
and 2l. In contrast, the affinities of the 4S isomers 2e, 2g, and
2k were two- to fivefold lower. This indicates that the adop-
tion of an exo ring pucker, which is stabilized by the 4R config-
Experimental Section
Chemistry: Reagents and dry solvents were obtained from com-
mercial sources and were used as received. Reactions were con-
ducted under dry N2. Evaporations of product solutions were car-
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 772 – 778 774

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
ried out in vacuo with a rotatory
evaporator. Column chromatogra-
phy was performed using 60 mm
silica gel. For thin-layer chroma-
tography (TLC) silica gel 60 mm
plates were used (UV, I2, or ninhy-
drin detection). Melting tempera-
tures are uncorrected. IR spectra
were registered from a thin film
on a NaCl crystal. NMR data were
acquired with 360 or 600 MHz
spectrometers, if not stated other-
wise in CDCl3; 13C NMR spectra
were recorded at 90 MHz if not
stated otherwise in CDCl3. Chemi-
cal shifts (d) are given in ppm rel-
ative to tetramethylsilane (TMS),
with TMS=0. MS data were ac-
quired using electronic ionization
(EI), atmospheric pressure chemi-
cal ionization (APCI), or electro-
spray ionization (ESI) techniques.
HPLC analysis revealed a purity of
>95% for all SAR compounds.
Table 1. Human NTS1 and NTS2 receptor binding data for the peptide–peptoids 1a, 2a–m, and 3a,b in com-
parison with the reference agent NT(8–13).
Compd
Sequence
Ki [nm][a]
SR[f]
NTS1[b]
NTS2[c]
NT(8–13)
0.24Æ0.024
1.2Æ0.17[d]
0.20
1a
31000Æ7000
60000Æ8700
59000Æ15000
8.0Æ0.94
3900
2600
650
2a
2b
23Æ5.4
91Æ12
2c
23000Æ7800
24Æ5.6
10Æ1.7
930
2d
66000Æ21000
6600
2e
2 f
2g
2h
2i
25000Æ5500
44000Æ9000
59000Æ4900[e]
57000Æ13000
340Æ100
74Æ10
67
590
310
690
55
Synthesis of peptide–peptoid
hybrids: Peptide synthesis was
performed according to standard
protocols as described below. The
synthesis was performed by start-
ing from commercially available
Fmoc-Leu Wang resin. a-Amino
acids were incorporated as their
commercially available derivatives:
Fmoc-Ile-OH, Fmoc-Lys(Boc)-OH,
190Æ0[e]
83Æ19
Fmoc-Arg(Pbf)-OH,
Fmoc-N-Me-
Arg(Mtr)-OH, Fmoc-(2S)-thiazoli-
dine-2-carboxylic acid, Fmoc-(2S)-
azetidine-2-carboxylic acid, Fmoc-
(2S)-3,4-dehydroproline,
Fmoc-
(2S)-pipecolic acid. Elongation of
the peptide chain was done by re-
petitive cycles of Fmoc deprotec-
tion with the help of microwave
irradiation and subsequent cou-
pling of the amino acid derivative.
After the final acylation step, the
N-terminal Fmoc residue was de-
protected, the resin was rinsed
with CH2Cl2 (10ꢂ) and dried in va-
cuo. Cleavage from the resin was
performed with a mixture of tri-
fluoroacetic acid (TFA)/phenol/
>100000
1800Æ150
880Æ62
2j
59000Æ8700
20000Æ3500
67
2k
200Æ20
100
H2O/triisopropylsilane
(TIS)
88:4:6:2 for 4 h, followed by filtra-
tion of the resin. After evapora-
tion of the solvent in vacuo and
precipitation in tert-butylmethyl
ether, the crude peptides were
purified by preparative RP-HPLC:
Agilent 1100 preparative series,
column Zorbax Eclipse XDB-C8,
21.2 mmꢂ150 mm, 5 mm particles
[C8], employing solvent systems,
2l
55000Æ23000
67000Æ17000
52Æ11
30Æ11
1100
2200
2m
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 772 – 778 775

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
DMF. Fmoc-Pro-OH (5 equiv) was
routinely coupled 2ꢂ with HATU
(5 equiv) and DIPEA (10 equiv) in
DMF.
Table 1. (Continued)
Compd
3a
Sequence
Ki [nm][a]
SR[f]
8600
2400
NTS1[b]
NTS2[c]
H-Arg-Arg-3,4-dehydroproline-
Nhtyr-Ile-Leu-OH (2a): Synthe-
sized according to EM. Purification
[C18]: CH3CN (A) and 0.1% HCO2H
in H2O (B) applying a linear gradi-
ent starting from 3% A in 97% B
to 32% A in 68% B in 10.9 min;
70000Æ15000
39000Æ21000
8.1Æ2.3
3b
16Æ5.1
FR: 32.0 mLminÀ1
; tR: 8.6 min;
[a] Values are the means ÆSEM of 3–9 individual experiments, each done in triplicate. [b] Determined with
[3H]neurotensin and membranes from CHO cells stably expressing human NTS1. [c] Determined with [3H]NT(8–
13) and homogenates from HEK293 cells transiently expressing human NTS2. [d] KD value. [e] Values are the
means ÆSD of two individual experiments, each done in triplicate. [f] Selectivity ratio: Ki(NTS1)/Ki(NTS2).
purity: S1B 97% (tR: 12.0 min);
S2A >97% (tR: 14.8 min); [M+
H]+: calcd 829.5, found 829.6.
H-Arg-Arg-azetidine-2-carboxylic
acid-Nhtyr-Ile-Leu-OH (2b): Syn-
thesized according to EM. Purifica-
Table 2. Determination of the functional activity of selected test com-
pounds, reference agent NT(8–13), and neurotensin obtained from an IP
accumulation assay in NTS2-expressing cells.
tion [C8]: CH3CN (A) and 0.1% HCO2H in H2O (B) applying a linear
gradient starting from 3% A in 97% B to 37% A in 63% B in
18.0 min; FR: 10.0 mLminÀ1; tR: 10.1 min; purity: S1A >99% (tR:
14.8 min); S2A >99% (tR: 15.6 min); [M+H]+: calcd 817.5, found
817.5.
Compd
IP accumulation[a]
1a
2d
2 f
2m
3a
1.5Æ0.21
2.0Æ0.11
1.4Æ0.18
1.6Æ0.14
1.7Æ0.13
2.0Æ0.11
1.0Æ0.21
H-Arg-Arg-pipecolic acid-Nhtyr-Ile-Leu-OH (2c): Synthesized ac-
cording to EM. Purification [C18]: CH3CN (A) and 0.1% HCO2H in
H2O (B) applying a linear gradient starting from 3% A in 97% B to
29.6% A in 70.4% B 10.0 min; FR: 32.0 mLminÀ1; tR: 9.0 min; purity:
S1B 99% (tR: 12.7 min); S2A >99% (tR: 15.3 min); [M+H]+: calcd
845.5, found 845.7.
3b
NT(8–13)
[a] Fold effect relative to NT(8–13): ligand efficacy at 1 mm [NT(8–13)] and
10 mm [test compounds] measured in an IP accumulation assay; values
are the means ÆSEM of five individual experiments, each done in tripli-
cate.
H-Arg-Arg-thiazolidine-2-carboxylic acid-Nhtyr-Ile-Leu-OH (2d):
Synthesized according to EM. Purification [C8]: CH3CN (A) and 0.1%
HCO2H in H2O (B) applying a linear gradient starting from 3% A in
97% B to 37% A in 63% B in 18.0 min; FR: 10.0 mLminÀ1; tR:
10.4 min; purity: S1A >99% (tR: 15.5 min); S2A >99% (tR:
16.0 min); [M+H]+: calcd 849.5, found 849.6.
linear gradient and flow rate [FR] as specified below or VP 250/32
Nucleodur C18 HTec, 5 mm particles [C18] employing solvent sys-
tems, linear gradient, and flow rate [FR] as specified below.
H-Arg-Arg-(2S,4S)-4-fluoroproline-Nhtyr-Ile-Leu-OH (2e): Synthe-
sized according to EM. Purification [C8]: CH3OH (A) and 0.1%
HCO2H in H2O (B) applying a linear gradient starting from 10% A in
90% B to 50% A in 50% B in 18.0 min; FR: 10.0 mLminÀ1; tR:
9.5 min; purity: S1A >99% (tR: 14.4 min); S2A >99% (tR:
16.3 min); [M+H]+: calcd 849.5, found 849.7.
After separation, the peptides were lyophilized, and peptide purity
and identity were assessed by analytical HPLC (Agilent 1100 analyt-
ical series equipped with QuatPump and VWD detector; column
Zorbax Eclipse XDB-C8 analytical column, 4.6 mmꢂ150 mm, 5 mm,
FR: 0.5 mLminÀ1) coupled to a Bruker Esquire 2000 mass detector
equipped with an ESI-trap. System 1 (S1): x–y% CH3OH in H2O+
0.1% HCO2H in 18 min (S1A: 10–55% in 18 min, 55–95% in 2 min,
95–95% in 2 min; S1B: 10–65%; S1C: 10–40%). System 2 (S2): x–
y% CH3CN in H2O+0.1% HCO2H in 26 min (S2A: 3–40%).
H-Arg-Arg-(2S,4R)-4-fluoroproline-Nhtyr-Ile-Leu-OH (2 f): Synthe-
sized according to EM. Purification [C8]: CH3OH (A) and 0.1%
HCO2H in H2O (B) applying a linear gradient starting from 10% A in
90% B to 50% A in 50% B in 18.0 min; FR: 10.0 mLminÀ1; tR:
9.0 min; purity: S1A >99% (tR: 13.8 min); S2A >99% (tR:
14.6 min); [M+H]+: calcd 849.5, found 849.6.
Elongation method (EM): Microwave-assisted (Discover microwave
oven, CEM Corp.) peptide synthesis was carried out in silanized
glass tubes loosely sealed with a silicon septum. The development
of overpressure was avoided by using DMF as the solvent along
with intermittent cooling. Fmoc deprotection: the resin was treat-
ed 1ꢂ with 25% piperidine in DMF (microwave irradiation: 7ꢂ5 s,
100 W), followed by washings with DMF (5ꢂ). Peptide coupling
H-Arg-Arg-(2S,4S)-4-azidoproline-Nhtyr-Ile-Leu-OH (2g): Synthe-
sized according to EM. Purification [C8]: CH3OH (A) and 0.1%
HCO2H in H2O (B) applying a linear gradient starting from 10% A in
90% B to 50% A in 50% B in 18.0 min; FR: 10.0 mLminÀ1; tR:
9.6 min; purity: S1A >99% (tR: 13.7 min); S2A 96% (tR: 14.9 min);
[M+H]+: calcd 872.5, found 872.6.
was done by using 5 equiv each of Fmoc-AA/PyBOP/DIPEA and H-Arg-Arg-(2S,4R)-4-azidoproline-Nhtyr-Ile-Leu-OH (2h): Synthe-
7.5 equiv 1-hydroxybenzotriazole (HOBt) dissolved in a minimum sized according to EM. Purification [C8]: CH3OH (A) and 0.1%
amount of DMF (irradiation: 20ꢂ10 s, 50 W). Between each irradia- HCO2H in H2O (B) applying a linear gradient starting from 10% A in
tion step, cooling of the reaction mixture to À108C was achieved 90% B to 50% A in 50% B in 18.0 min; FR: 10.0 mLminÀ1; tR:
by sufficient agitation in an EtOH–ice bath. The peptoid (3 equiv) 9.6 min; purity: S1A 95% (tR: 13.6 min); S2A 95% (tR: 14.8 min);
was coupled with 3 equiv PyBOP/DIPEA and 4.5 equiv HOBt in [M+H]+: calcd 872.5, found 872.6.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 772 – 778 776

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
H-Arg-Arg-(2S,4S)-4-aminoproline-Nhtyr-Ile-Leu-OH (2i): Peptide
2g was synthesized according to EM. After completed peptide
coupling, the resin-bound peptide–peptoid [H-Arg(Pbf)-Arg(Pbf)-
(2S,4S)-4-azidoproline-Nhtyr(tBu)-Ile-Leu Wang resin (4a)] was
washed with CH2Cl2 (5ꢂ2 mL), dried, and swollen with THF (2 mL)
for 30 min. After solvent removal, a mixture of H2O (130 mL), Me3P
(1m in THF, 30 equiv), and THF (2 mL) was added, and the reaction
vessel was shaken for 2 h at RT (reaction monitored by Kaiser test).
After complete reaction, the resin was filtered and washed with
THF (5ꢂ2 mL), DMF (5ꢂ2 mL), and CH2Cl2 (3ꢂ2 mL) and dried af-
terward. Then, the resin 4c was divided into two parts: one part
was processed to 2k, the other part was swollen in DMF for
30 min to deprotect the Fmoc group via standard protocol and
cleave the peptide from the resin. Purification [C8]: CH3CN (A) and
0.1% HCO2H in H2O (B) applying a linear gradient starting from
Rodgau, Germany) at a concentration of 0.50 nm. Specific binding
of the radioligand was determined at KD values of 0.37–0.96 nm
and a Bmax of 6170–9300 fmol(mg protein)À1. Nonspecific binding
was determined in the presence of 10 mm neurotensin. NTS2 bind-
ing assays were carried out by the calcium phosphate method,
using homogenates of membranes from HEK293 cells, which were
transiently transfected with the pcDNA3.1 vector containing the
human NTS2 gene (Missouri S&T cDNA Resource Center (UMR),
Rolla, MO, USA).[37] Membranes were incubated at a final concentra-
tion of 6–20 mg per well together with 0.50 nm [3H]NT(8–13) (spe-
cific activity: 136 CimmolÀ1; custom synthesis of [leucine-3H]NT(8–
13) by GE Healthcare, Freiburg, Germany) at KD values in the range
of 0.67–2.02 nm and a Bmax value of 310–930 fmol(mg protein)À1
.
Nonspecific binding was determined in the presence of 10 mm
NT(8–13), and the protein concentration was generally determined
by the method of Lowry using bovine serum albumin as stan-
dard.[38]
3% A in 97% B to 35% A in 65% B in 18.0 min; FR: 10.0 mLminÀ1
;
tR: 7.8 min; purity: S1C 99% (tR: 10.4 min); S2A 99% (tR: 12.0 min);
[M+H]+: calcd 846.5, found 846.5.
H-Arg-Arg-(2S,4R)-4-aminoproline-Nhtyr-Ile-Leu-OH (2j): Synthe-
sized analogously to 2i using (2S,4R)-4-azidoproline instead of
(2S,4S)-4-azidoproline. One part of the resin-bound intermediate
[Fmoc-Arg(Pbf)-Arg(Pbf)-(2S,4R)-4-aminoproline-Nhtyr(tBu)-Ile-Leu
Wang resin (4d)] was used for the synthesis of 2l, the other part
was N-terminally cleaved from the Fmoc protecting group. After-
ward, resin cleavage was performed. Purification [C8]: CH3CN (A)
and 0.1% HCO2H in H2O (B) applying a linear gradient starting
from 3% A in 97% B to 35% A in 65% B in 18.0 min; FR:
10.0 mLminÀ1; tR: 8.1 min; purity: S1C 99% (tR: 12.3 min); S2A 99%
(tR: 12.2 min); [M+H]+: calcd 846.5, found 846.6.
Data analysis: Data analysis of the competition curves from the
radioligand binding experiments was accomplished by nonlinear
regression analysis using the algorithms in Prism 5.0 (GraphPad
Software, San Diego, CA, USA). EC50 values derived from the result-
ing dose–response curves were transformed into the correspond-
ing Ki values using the equation of Cheng and Prusoff.[39]
Determination of IP accumulation: The functional test on NTS2 re-
ceptor-mediated IP accumulation was performed according to
published procedures.[14,35a] Briefly, HEK293 cells were transiently
transfected and transferred into 24-well plates at a density of 1ꢂ
105 cells per well at a volume of 270 mL. After the addition of 30 mL
H-Arg-Arg-(2S,4S)-4-acetamidoproline-Nhtyr-Ile-Leu-OH (2k): The
resin-bound intermediate from the synthesis of 2i [Fmoc-Arg(Pbf)-
Arg(Pbf)-(2S,4S)-4-aminoproline-Nhtyr(tBu)-Ile-Leu Wang resin (4c)]
was swollen in CH2Cl2 for 30 min. Afterward, the solvent was re-
moved and a mixture of DIPEA (30 equiv) and acetic anhydride
(30 equiv) in CH2Cl2 (2 mL) was added, and the reaction vessel was
shaken for 30 min at RT (reaction monitored by Kaiser test). After
solvent removal the N-terminal Fmoc protecting group was
cleaved before the resin cleavage was performed. Purification [C8]:
CH3CN (A) and 0.1% HCO2H in H2O (B) applying a linear gradient
starting from 3% A in 97% B to 35% A in 65% B in 18.0 min; FR:
10.0 mLminÀ1; tR: 9.6 min; purity: S1A >99% (tR: 13.1 min); S2A
>99% (tR: 14.5 min); [M+H]+: calcd 888.5, found 888.6.
myo-[3H]inositol (specific activity: 22.5 CimmolÀ1
, PerkinElmer,
Rodgau, Germany), cells were incubated for 15 h. The medium was
then aspirated, cells were washed with medium supplemented
with 10 mm LiCl, and test compounds were added for 60 min at
378C. To stop incubation, cells were lysed by adding 150 mL ice-
cold 0.1m NaOH. The cell extract was diluted in buffer (5 mm
sodium tetraborate, 0.5 mm Na·EDTA) and was separated by anion-
exchange chromatography using an AG1-X8 resin (BioRad, Munich,
Germany). After washing with water and elution buffer I (5 mm
sodium tetraborate, 60 mm sodium formate) and again with water,
total IP was eluted with 2.5 mL elution buffer II (1.0m ammonium
formate) and directly collected into scintillation counting vials. Ra-
dioactivity was measured by scintillation counting after adding
2.5 mL of Emulsifier-Safe (PerkinElmer, Rodgau, Germany). Data
were analyzed by normalizing the dpm values; this was done by
setting the data for non-stimulated receptor (buffer) equal to zero
and the effect for NT(8–13) equal to one.
H-Arg-Arg-(2S,4R)-4-acetamidoproline-Nhtyr-Ile-Leu-OH (2l): Syn-
thesized analogously to 2k by using the intermediate 4d instead
of 4c. Purification [C8]: CH3CN (A) and 0.1% HCO2H in H2O (B) ap-
plying a linear gradient starting from 3% A in 97% B to 35% A in
65% B in 18.0 min; FR: 10.0 mLminÀ1; tR: 10.3 min; purity: S1A
99% (tR: 13.7 min); S2A 99% (tR: 14.8 min); [M+H]+: calcd 888.5,
found 888.6.
Glossary: NTS1, neurotensin receptor 1; NTS2, neurotensin recep-
tor 2; NTS3, neurotensin receptor 3; NT, neurotensin; GPCR, G-pro-
tein-coupled receptor; AA, amino acid; Nhtyr, N-homotyrosine;
CHO, Chinese hamster ovary; HEK, human embryonic kidney;
SPPS: solid-phase peptide synthesis; Fmoc, 9-fluorenylmethyloxy-
carbonyl; HATU, 2-(1H-7-azabenzotriazol-1-yl)-1,1,3,3-tetramethy-
luronium hexafluorophosphate; PyBOP, (benzotriazol-1-yloxy)tripyr-
rolidinophosphonium hexafluorophosphate; Pbf, 2,2,4,6,7-pentam-
ethyldihydrobenzofuran-5-sulfonyl; Boc, tert-butyloxycarbonyl; Mtr;
4-Methoxy-2,3,6-trimethylbenzenesulfonyl; HOBt, 1-hydroxybenzo-
triazole; DMF, N,N-dimethylformamide; DIPEA, diisopropylethyla-
mine; TFA, trifluoroacetic acid; TIS, triisopropylsilane; MALDI-TOF-
MS, Matrix-assisted laser desorption/ionization time-of-flight mass
spectrometry; IS, internal standard; IP inositol phosphate.
H-Arg-Arg-4,4-difluoroproline-Nhtyr-Ile-Leu-OH (2m): Synthe-
sized according to EM. Purification [C8]: CH3OH (A) and 0.1%
HCO2H in H2O (B) applying a linear gradient starting from 10% A in
90% B to 50% A in 50% B in 18.0 min; FR: 10.0 mLminÀ1; tR:
11.0 min; purity: S1A >99% (tR: 14.7 min); S2A >99% (tR:
16.3 min); [M+H]+: calcd 867.5, found 867.6.
Receptor binding experiments: Receptor binding data were deter-
mined according to protocols as described previously.[14,36] In
detail, NTS1 binding was measured using homogenates of mem-
branes from CHO cells stably expressing human NTS1 at a final
concentration of 1–2 mg per well, and the radioligand
[3H]neurotensin (specific activity: 116 CimmolÀ1
;
PerkinElmer,
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 772 – 778 777

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[20] J. Einsiedel, H. Hꢀbner, M. Hervet, S. Hꢃrterich, S. Koschatzky, P. Gmeiner,
Bioorg. Med. Chem. Lett. 2008, 18, 2013–2018.
Acknowledgements
[21] H. Bittermann, J. Einsiedel, H. Hꢀbner, P. Gmeiner, J. Med. Chem. 2004,
47, 5587–5590.
[22] D. L. Heyl, A. M. Sefler, J. X. He, T. K. Sawyer, D. J. Wustrow, H. C. Akunne,
M. D. Davis, T. Pugsley, T. G. Heffner, A. E. Corbin, W. Cody, Int. J. Pept.
Protein Res. 1994, 44, 233–238.
We thank J. Meltretter and M. Pischetsrieder (Chair of Food
Chemistry, Friedrich Alexander University) for their help with the
serum degradation assay.
[23] a) F. Cavelier, B. Vivet, J. Martinez, A. Aubry, C. Didierjean, A. Vicherat, M.
Marraud, J. Am. Chem. Soc. 2002, 124, 2917–2923; b) A. Flores-Ortega,
J. Casanovas, D. Zanuy, R. Nussinov, C. Alemꢄn, J. Phys. Chem. B 2007,
111, 5475–5482; c) A. Choudhary, K. H. Pua, R. T. Raines, Amino Acids
2011, 41, 181–186.
Keywords: GPCRs · neurotensin · NTS2 · peptide–peptoid
hybrids · proline analogues · structure–activity relationships
[24] A. K. Pandey, D. Naduthambi, K. M. Thomas, N. J. Zondlo, J. Am. Chem.
Soc. 2013, DOI: 10.1021/ja3109664.
[1] A. M. Sefler, J. X. He, T. K. Sawyer, K. E. Holub, D. O. Omecinsky, M. D.
Reily, V. Thanabal, H. C. Akunne, W. L. Cody, J. Med. Chem. 1995, 38,
249–257.
[2] F. Hong, B. Cusack, A. Fauq, E. Richelson, Curr. Med. Chem. 1997, 4,
412–434, and references therein.
[3] a) J. T. Lundquist IV, T. A. Dix, Bioorg. Med. Chem. Lett. 1999, 9, 2579–
2582; b) J. T. Lundquist IV, T. A. Dix, J. Med. Chem. 1999, 42, 4914–4918.
[4] S. Achilefu, A. Srinivasan, M. A. Schmidt, H. N. Jimenez, J. E. Bugaj, J. L.
Erion, J. Med. Chem. 2003, 46, 3403–3411.
[5] K. Tanaka, M. Masu, S. Nakanishi, Neuron 1990, 4, 847–854.
[6] N. Vita, P. Laurent, S. Lefort, P. Chalon, X. Dumont, M. Kaghad, D. Gully,
G. Le Fur, P. Ferrara, D. Caput, FEBS Lett. 1993, 317, 139–142.
[7] P. Chalon, N. Vita, M. Kaghad, M. Guillemot, J. Bonnin, B. Delpech, G.
Le Fur, P. Ferrara, D. Caput, FEBS Lett. 1996, 386, 91–94.
[8] J. Mazella, J. M. Botto, E. Guillemare, T. Coppola, P. Sarret, J. P. Vincent, J.
Neurosci. 1996, 16, 5613–5620.
[25] a) E. S. Eberhardt, N. Panasik, Jr., R. T. Raines, J. Am. Chem. Soc. 1996,
118, 12261–12266; b) D. O’Hagan, C. Bilton, J. A. K. Howard, L. Knight,
D. J. Tozer, J. Chem. Soc. Perkin Trans. 2 2000, 605–607; c) M. L. DeRider,
S. J. Wilkens, M. J. Waddell, L. E. Bretscher, F. Weinhold, R. T. Raines, J. L.
Markley, J. Am. Chem. Soc. 2002, 124, 2497–2505; d) C. R. S. Briggs, D.
O’Hagan, J. A. K. Howard, D. S. Yufit, J. Fluorine Chem. 2003, 119, 9–13;
e) J. A. Hodges, R. T. Raines, J. Am. Chem. Soc. 2003, 125, 9262–9263;
f) U. Kusebauch, S. A. Cadamuro, H.-J. Musiol, M. O. Lenz, J. Wachtveitl,
L. Moroder, C. Renner, Angew. Chem. 2006, 118, 7170–7173; Angew.
Chem. Int. Ed. 2006, 45, 7015–7018.
[26] L.-S. Sonntag, S. Schweizer, C. Ochsenfeld, H. Wennemers, J. Am. Chem.
Soc. 2006, 128, 14697–14703.
[27] R. S. Erdmann, H. Wennemers, J. Am. Chem. Soc. 2012, 134, 17117–
17124.
[28] M. Kuemin, Y. A. Nagel, S. Schweizer, F. W. Monnard, C. Ochsenfeld, H.
Wennemers, Angew. Chem. 2010, 122, 6468–6471; Angew. Chem. Int. Ed.
2010, 49, 6324–6327.
[9] J. Mazella, N. Zsꢀrger, V. Navarro, J. I. Chabry, M. Kaghad, D. Caput, P.
Ferrara, N. Vita, D. Gully, J.-P. Maffrand, J.-P. Vincent, J. Biol. Chem. 1998,
273, 26273–26276.
[29] Y. A. Nagel, M. Kuemin, H. Wennemers, Chimia 2011, 65, 264–267.
[30] M. Doia, Y. Nishib, N. Kiritoshia, T. Iwataa, M. Nagoa, H. Nakanob, S.
Uchiyamab, T. Nakazawac, T. Wakamiyad, Y. Kobayashi, Tetrahedron
2002, 58, 8453–8459.
[31] a) M. Meldal, M. A. Juliano, A. M. Jansson, Tetrahedron Lett. 1997, 38,
2531–2534; b) J. T. Lundquist IV, J. C. Pelletier, Org Lett. 2001, 3, 781–
783.
[10] N. Noinaj, Y. Shibata, B. Kloss, F. Xu, P. Shah, C. G. Tate, R. Grisshammer,
Nature 2012, 490, 508–513.
[11] M. Boules, Y. Liang, S. Briody, R. Miura, I. Fauq, A. Oliveros, M. Wilson, S.
Khaniyev, K. Williams, Z. Li, Y. Qi, M. Katovich, E. Richelson, Brain Res.
2010, 1308, 35–46.
[12] E. Richelson, D. J. McCormick, Y.-P. Pang, K. S. Phillips, Patent Pub. No.:
US 2009/0062212 A1, 2009 [CAN150:275882].
[13] G. Roussy, M.-A. Dansereau, S. Baudisson, F. Ezzoubaa, K. Belleville, N.
Beaudet, J. Martinez, E. Richelson, P. Sarret, Mol. Pain 2009, 5, 38.
[14] J. Einsiedel, C. Held, M. Hervet, M. Plomer, N. Tschammer, H. Hꢀbner, P.
Gmeiner, J. Med. Chem. 2011, 54, 2915–2923.
[15] a) R. J. Simon, R. S. Kania, R. N. Zuckermann, V. D. Huebner, D. A. Jewell,
S. Banville, S. Ng, L. Wang, S. Rosenberg, C. K. Marlowe, Proc. Natl. Acad.
Sci. USA 1992, 89, 9367–9371; b) R. N. Zuckermann, E. J. Martin, D. C.
Spellmeyer, G. B. Stauber, K. R. Shoemaker, J. M. Kerr, G. M. Figliozzi,
D. A. Goff, M. A. Siani, R. J. Simon, S. C. Banville, E. G. Brown, L. Wang,
L. S. Richter, W. H. Moos, J. Med. Chem. 1994, 37, 2678–2685.
[16] C. Held, M. Plomer, H. Huebner, J. Meltretter, M. Pischetsrieder, P.
Gmeiner, ChemMedChem 2013, 8, 75–81.
[17] J. A. Henry, D. C. Honvell, K. G. Meecharn, D. C. Rees, Bioorg. Med. Chem.
Lett. 1993, 3, 949–952.
[18] Y.-P- Pang, B. Cusack, K. Groshan, E. Richelson, J. Biol. Chem. 1996, 271,
15060–15068.
[32] R. S. Erdmann, H. Wennemers, Angew. Chem. 2011, 123, 6967–6970;
Angew. Chem. Int. Ed. 2011, 50, 6835–6838.
[33] A. Kamal, Rajender, D. R. Reddy, M. K. Reddy, G. Balakishan, T. B. Shaik,
M. Chourasia, G. N. Sastry, Bioorg. Med. Chem. 2009, 17, 1557–1572.
[34] J. Mazella, J.-P. Vincent, Peptides 2006, 27, 2469–2475.
[35] a) M. J. S. Chee, K. Mçrl, D. Lindner, N. Merten, G. W. Zamponi, P. E.
Light, A. G. Beck-Sickinger, W. F. Colmers, J. Biol. Chem. 2008, 283,
33337–33346; b) M. J. Berridge, Biochem. J. 1983, 212, 849–858.
[36] H. Hꢀbner, C. Haubmann, W. Utz, P. Gmeiner, J. Med. Chem. 2000, 43,
756–762.
[37] M. Jordan, A. Schallhorn, F. M. Wurm, Nucleic Acids Res. 1996, 24, 596–
601.
[38] O. H. Lowry, N. J. Rosebrough, A. L. Farr, R. J. Randall, J. Biol. Chem.
1951, 193, 265–275.
[39] Y. Cheng, W. H. Prusoff, Biochem. Pharmacol. 1973, 22, 3099–3108.
[19] J. A. Gilbert, D. J. McCormick, M. A. Pfennig, K. S. Kanba, L. J. Enloe, A.
Moore, E. Richelson, Biochem. Pharmacol. 1989, 38, 3377–3382.
Received: February 4, 2013
Published online on March 26, 2013
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2013, 8, 772 – 778 778