Peptide backbone modifications on the C-terminal hexapeptide of neurotensin

Article abstract of DOI:10.1016/j.bmcl.2008.01.110

To compare backbone-induced susceptibilities with affinity changes that are caused by side-chain modifications in the respective positions, structure activity relationship studies on a series of NT(8-13) analogues were performed providing valuable insight

Full text of DOI:10.1016/j.bmcl.2008.01.110

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2013–2018
Peptide backbone modifications on the C-terminal
hexapeptide of neurotensin
Jurgen Einsiedel, Harald Hubner, Maud Hervet, Steffen Ha¨rterich,
¨
¨
Susanne Koschatzky and Peter Gmeiner*
Department of Chemistry and Pharmacy, Emil Fischer Center, Friedrich Alexander University,
Schuhstraße 19, D-91052 Erlangen, Germany
Received 11 December 2007; revised 28 January 2008; accepted 29 January 2008
Available online 2 February 2008
Abstract—To compare backbone-induced susceptibilities with affinity changes that are caused by side-chain modifications in the
respective positions, structure activity relationship studies on a series of NT(8-13) analogues were performed providing valuable
insights into the major requirement for neurotensin receptor recognition and activation. The data led us to highly potent NTR1
ligands and the generation of a pharmacophore model that will be helpful for the discovery of therapeutically relevant non-peptidic
NTR1 agonists.
ꢀ 2008 Elsevier Ltd. All rights reserved.
The neuropeptide neurotensin (NT, pGlu-Leu-Tyr-Glu-
Asn-Lys-Pro-Arg-Arg-Pro-Tyr-Ile-Leu) acts as a neuro-
modulator and is associated with various physiological
effects. Besides its location in the periphery, NT is found
in the central nervous system especially regulating dopa-
minergic transmission of the mesocorticolimbic path-
ways.1 Based on behavioural observations, neurotensin
produces preclinical effects similar to antipsychotics,
which is likely due to a modulation of dopaminergic
activity by stimulation of the G-protein coupled receptor
NTR1 being co-localized with the dopamine receptor
subtype D2.2,3 SAR studies including a gradual trunca-
tion of the peptide sequence demonstrated that the C-ter-
minal hexapeptide NT(8-13) is sufficient for NTR1
binding and ligand efficacy.4 Probing the contributions
of the ligand’s side chains employing Ala- and D-amino
acid species exchange led to the conclusion that all resi-
dues appear to be important for the receptor binding pro-
cess.5–10 Structural properties mediated by Pro10 and
Tyr11 within the core region of NT(8-13) turned out
highly crucial for receptor recognition.11 Very recently,
M. Baldus and co-workers were able to gain experimental
data on the bioactive conformation of NT(8-13) bound
to NTR1 when 2D solid-state NMR spectroscopy was
utilized.12 According to these results and our investiga-
tions of conformational constraints on NTR1 binding,13
a linear rearrangement of the NT(8-13) backbone can be
concluded to represent the bioactive conformation. As a
complement to previously described SAR studies,14–17 we
herein describe NT(8-13) backbone modifications when
we compare the susceptibilities on receptor recognition
with affinity changes that are caused by side-chain mod-
ifications in the respective positions. Besides systematic
homo-b-amino acid exchange and a peptoid scan, our
backbone manipulations involve the insertion of lactam
bridged scaffolds and the substitution of both amino
and the carboxyl functions at the N- and C-terminal ends,
respectively.
Peptide analogues composed exclusively of b-residues
(b-peptides) or of a combination of b- and a-amino acid
residues have been designed to adopt a variety of con-
formations that resemble protein secondary structures.18
Such foldamer classes can serve as rich sources of inhib-
itors of protein-protein interactions and show remark-
able resistance to degradation by proteases,19 making
such oligomers attractive in medicinal chemistry.
Employing commercially available chiral building
blocks, our initial investigations were directed to the
synthesis of the b-homologues 1a–c and 1f, g when
straightforward solid phase supported preparations
employing Fmoc strategy and HATU as the coupling
reagent gave rise to analytically pure oligomers. Proline
Keywords: Neurotensin; GPCR; Backbone modifications; Peptide
analogues; Ligand binding; Pharmacophore model; NT(8-13).
*
Corresponding author. E-mail: gmeiner@pharmazie.uni-erlangen.de
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.01.110

2014
J. Einsiedel et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2013–2018
β−homo-NT(8-13)/Ala-Insertion
HN
NH2
β-hArg - Arg - Pro - Tyr - Ile - Leu - OH (1a)
Arg - β-hArg - Pro - Tyr - Ile - Leu - OH (1b)
Arg - Arg - β-hPro - Tyr - Ile - Leu - OH (1c)
Arg - Arg - β-isoPro - Tyr - Ile - Leu - OH (1d)
Arg - Arg - Pro - β-hTyr - Ile - Leu - OH (1e)
Arg - Arg - Pro - Tyr - β-hIle - Leu - OH (1f)
Arg - Arg - Pro - Tyr - Ile - β-hLeu - OH (1g)
Arg - Arg - Ala - Pro - Tyr - Ile - Leu - OH (1h)
Arg - Arg - Pro - Tyr - Ala - Ile - Leu - OH (1i)
NH
O
O
O
H
N
H
N
H
N
8
9
10
O
11
12
13
H2N
N
N
H
OH
O
O
terminal modifications
NH
OH
Syn1 - Arg - Pro - Tyr - Ile - Leu - X
Syn1 = Me3N+CH2CO; X = OH (2a)
Syn1 = HO2C(CH2)2CO; X = OH (2b)
HN
NH2
Syn1 = Asp; X = OH (2c)
Syn1 = Arg; X = NH2 (2d)
Syn1 = Arg; X = Ala -OH (2e)
NT(8-13): Arg - Arg - Pro - Tyr - Ile - Leu - OH
peptoids
Lys - Lys - Pro - Tyr - Ile - Leu - OH
NLys - Lys - Pro - Tyr - Ile - Leu - OH (4a)
Lys - NLys - Pro - Tyr - Ile - Leu - OH (4b)
Lys - Lys - Pro - NTyr - Ile - Leu - OH (4c)
Lys - Lys - Pro - Ser(OCH2PhOH)-Ile - Leu - OH (4d)
Lys - Lys - Pro - Tyr - NIle(1) - Leu - OH (4e)
Lys - Lys - Pro - Tyr - NIle(2) - Leu - OH (4f)
Lys - Lys - Pro - Tyr - Ile - NLeu - OH (4g)
lactam-bridged mimetics:
Arg - Arg - Scaffold - Ile - Leu-OH
Scaffold:
Scaffold:
N
N
O
N
N
O
- β-hPro - :
- β-isoPro - :
OH
O
(3a)
(3b)
O
O
O
OH
O
H
N
N
Scaffold:
Scaffold:
N
O
N
N
- NLys - :
N
N
O
NH2
( )4
O
O
OH
O
O
N
(3c)
(3d)
OH
Scheme 1. Backbone modifications leading to a variety of NT(8-13) analogues.
was replaced by its congeners (S)-2-carboxymethylpyrr-
olidine (b-homoproline) and (R)-3-carboxypyrrolidine
(b-isoproline)20 resulting in formation of the peptide
analogues 1c and 1d, respectively (Scheme 1).
To circumvent the perceived major limitations associ-
ated with peptides as therapeutic agents, peptoids or
peptide–peptoid combinations have been frequently
used.25 The peptoid design strategy also allows to ex-
plore the significance of backbone NH functions and
to increase the number of energetically relevant confor-
mations for ligand binding.26 A solid phase supported
sub-monomer approach was chosen for the synthesis
of the test compounds 4a, c, e/f, g containing peptoid
subunits to simulate the amino acid residues of the ref-
erence NTR1 agonist [Lys8,Lsy9]NT(8-13).27,28 Since
we employed 2-butylamine as a racemic building block,
the NIle derivative was obtained as a mixture of diaste-
reomers (4e/f) that had to be separated by HPLC. To
circumvent an acetal-type cleavage of the hydroxyben-
zylamine substructure of NTyr, we applied a protocol
employing 4-allyloxybenzylamine29 and Fmoc-Lys(Al-
loc)-OH for the generation of oligomer 4c. When we uti-
lized 2-chlorotrityl resin, this strategy facilitated a
smooth, palladium catalyzed side-chain deprotection
and subsequent HFIP promoted cleavage. To further
probe the tyrosine binding portion of the receptor, the
4-HO-benzyl function was displaced not only to the
amide nitrogen but also onto the primary alcohol func-
For the synthesis of the b-homotyrosine derived peptide
analogue 1e, we relied on our previously reported
homologization protocol giving access to enantiopure
N,N-dibenzyl protected b-homotyrosine21 that was
esterified, debenzylated and N-protected by a Boc unit.
After alkylation of the aromatic HO-group with 2,6-dic-
hlorobenzylbromide and saponification,22 the chiral
building block was subjected to Boc-SPPS.23 During
the course of this work, our attention was drawn to-
wards the microwave assisted Fmoc deprotection and
peptide coupling principle employing PYBOP as the
coupling reagent. This fast method is highly useful to
achieve excellent purities avoiding the application of
cost-intensive HATU.24 Besides homologization by for-
mally pasting a CH2 group into the backbone, an Ala
moiety as a three-atom spacer was inserted into the pep-
tide bonds connecting the crucial Pro-Tyr unit with the
basic N-terminal and the lipophilic C-terminal frag-
ments to give the heptapeptides 1h and 1i, respectively.

J. Einsiedel et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2013–2018
2015
tion of [Ser11]NT(8-13). In detail, Boc-protected serine
was O-alkylated by 4-allyloxybenzylbromide30 using
NaH as a base. After Boc-detachment promoted by sul-
furic acid in dioxane31 and subsequent Fmoc-protection,
the resulting unnatural amino acid was subjected to so-
lid phase supported peptide synthesis. We took advan-
tage of Alloc protected lysine and 2-chlorotrityl resin
to give peptide surrogate 4d. As the sub-monomer ap-
proach did not prove successful for the synthesis of H-
Lys-NLys-Pro-Tyr-Ile-Leu-OH (4b), we incorporated
NLys as a Fmoc-protected monomer.32
120
100
80
60
40
20
0
Ala
D-aa
4c
4d
4g
1d
4f
4e
1f
To compare backbone-induced susceptibilities on recep-
tor recognition with affinity changes that are caused by
side-chain modifications in the respective positions,
NT(8-13) related hexapeptides obtained upon sequential
alanine and D-amino acid replacements were synthesized
as reference agents according to Fmoc SPPS proto-
cols.33,34 Using PYBOP or HATU as activating re-
agents, spirocyclic and fused Pro-Tyr surrogates could
be introduced into the sequence giving access to the lac-
tam bridged NT(8-13) mimics 3a–d.35,36 For both oligo-
mers 3c and 3d, diastereomers were separated by HPLC
and further investigated in isomerically pure form. To
formally replace the C-terminal carboxy function by a
carboxamide isostere, Rink amide resin was used to give
rise to the test compound H-Arg-Arg-Pro-Tyr-Ile-Leu-
NH2 (2d)8 after sequential coupling of the respective
Fmoc-protected amino acids and TFA cleavage. To fur-
ther explore the C-terminus-binding portion of the neu-
rotensin receptor, the heptapeptide NT(8-13)-Ala-OH
(2e) was synthesized. Finally, the N-terminal moiety
was modified to evaluate if a cationic element is neces-
sary for receptor recognition. Thus, carboxymethyl(tri-
methyl)ammonium (betaine), succinic acid and aspartic
acid were attached to resin-bound NT(9-13) to afford
oligomers 2a–c.
1c
1i
1e
1b
1h
4b
1a
4a
1g
8
9
10 11 12 13
Arg - Arg - Pro - Tyr - Ile - Leu
(Lys - Lys - Pro - Tyr - Ile - Leu)
Figure 1. Impact of peptide backbone modifications on the binding
properties of the C-terminal hexapeptide of neurotensin. Susceptibility
is derived from bound radioactivity determined in a heterologous
binding experiment using porcine NTR1 receptors, [3H]neurotensin
and the test compounds at 300 nM and is expressed as the ratio of
specific radioactivity of test compound over that of reference. Specific
radioactivity was calculated using the equation: ((radioactivity À UB)/
(TB À UB)) · 100%. The b-homo derivatives 1a–i (full circles (•)), the
hexapeptides of the Ala scan (open hash (e)) and the hexapeptides of
the D-amino acid scan (open circles (ꢀ)) are compared to the effect of
NT(8-13). The peptoid derivatives 4a–g (full squares (j)) are
compared to [Lys[8],Lys[9]]NT(8-13).
values when the 8-b-homo-NT(8-13) showed a Ki of
0.13 nM indicating an even higher affinity than the refer-
ence NT(8-13) (Ki = 0.23 nM). For the 9-b-homo-NT(8-
13) (1b), a Ki value of 2.3 nM was observed.
Our initial biological investigations were directed to an
exchange of the amino acid residues 8-13 by b-amino
acid homologues 1a–c and 1e–g. Since we anticipated
the Pro-Tyr fragment to be especially crucial, the
b-isoproline derived congener 1d and the peptide homo-
logues 1h, i that display an insertion of Ala departing the
Pro-Tyr moiety from the N- and C-terminal sequence,
respectively, were also investigated. In detail, suscepti-
bilities on NT(8-13) backbone modifications were com-
pared to affinity changes that are caused by side-chain
modifications when employing Ala- and D-amino acid
species exchange in the positions 8-13. Susceptibility
data were derived from specific radioactivity ratios of
test compounds and reference agents. An attenuation
of affinity compared to NT(8-13) is displayed in Figure
1. The data clearly indicate the low susceptibility for
both backbone and side-chain modifications in position
8 when the amount of bound radioligands after compe-
tition with the test compounds in 300 nM concentration
was comparably low (data in detail, see supporting
information). An exchange of arginine in position 9 by
homo-arginine led to an approximately five-fold higher
amount of bound radioligand indicating a significantly
lower but still substantial ligand affinity. These screening
data could be corroborated by the measurement of Ki
As expected, b-amino acid exchange in position 10 was
more crucial, when we lost a factor of approximately
30 for the ability of 1c to displace [3H]neurotensin. An
alternative evaluation of the b-proline derivative 1d
bearing a carboxylate function in position 3 of the pyr-
rolidine ring induced an even stronger loss of affinity,
which was indicated by a factor of about 70. On the
other hand, investigation of the [b-Tyr]NT(8-13) (1e)
showed a minor susceptibility when repeatedly per-
formed heterologous displacement experiments at eight
different concentrations (Table 1) showed an average
Ki value in the single digit nanomolar range
(Ki = 8.1 nM). Since we inferred from these experiments
a certain flexibility of the distance between the Pro-Tyr
fragment and both the N- and the C-terminal portions,
alanine was inserted as a three-atom spacer instead of
the one-carbon homologation performed before. In fact,
in vitro displacement studies corroborated our hypothe-
sis displaying Ki values of 9.6 nM and 6.8 nM for the
heptapeptides 1h and 1i, respectively. On the other
hand, our heterologous competition experiments clearly
indicate that chemical manipulation of Ile12 led to a

2016
J. Einsiedel et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2013–2018
of the 5-membered lactam by a sterically more demand-
ing surrogate either leads to repulsive interactions or to
an increase of the proline w-angle resulting in a back-
bone structure that is significantly different to the bioac-
tive conformation.37 As a consequence of all data
discussed above, we concluded that both N- and C-ter-
minal groups might be directed to the surface of the
receptor whereas the central portion obviously interacts
with a tight binding site crevice. To learn if the C-termi-
nal carboxyl function and N-terminal amino group
specifically interact with the neurotensin receptor by
an ion-ion interaction or if their hydrophilicity is
beneficial to ligand binding, we exchanged the carboxyl
function by a carboxamide (oligomer 2d) and the argi-
nine residue in position 8 by the anionic residues aspar-
tic acid and succinic acid. Additionally, betaine, a
permanently positively charged quaternary ammonium
salt, was coupled instead of arginine 8. Actually, we
were surprised to see that NT(8-13)-NH2 (2d) showed
an affinity that even exceeded the binding properties of
NT(8-13) (Ki = 0.18 nM). This is in disagreement with
previous studies suggesting the C-terminal carboxy func-
tion to be of primary importance for NTR1 binding and
activation.38 Interestingly, the acidic terminal OH func-
tion can be displaced by an Ala residue without signifi-
cant reduction of affinity for the heptapeptide (2e). On
the other hand, exchange of the basic arginine residue
in position 8 by aspartate or succinate led to a strong de-
crease of NTR1 recognition indicating that not only the
hydrophilicity but also the basicity and, thus, the forma-
tion of a cationic species is crucial for specific binding.
This could be confirmed by investigation of the quarter-
nary ammonium salt 2a revealing a Ki value in the
sub-nanomolar range (Ki = 0.65 nM).
Table 1. Receptor binding data for the oligomeres 1a–i, 2a–e, 3a–d
and 4a–d in comparison to the reference hexapeptides NT(8-13) and
[Lys8Lys9]NT(8-13) employing porcine NTR1 receptors
Oligomer Kia (nM)
Oligomer
Kia (nM)
1a
1b
1c
1d
1e
1f
0.13 0.0050 3a
7400 1200
18,000 5000
1400 50
2.3 0.95
28 5.5
3b
3cb
3cc
3dd
3de
4a
260 35
8.1 0.72
100 26
3.2 0.75
9.6 4.4
6.8 2.3
0.65 0.19
10,000
0
8600 1,500
2400 200
0.45 0.14
5.7 2.0
1g
1h
1i
4b
4c
11,000 0
530 80
2a
2b
2c
2d
2e
4d
2300 0
150 25
0.18 0.046
0.27 0.056
NTf
1.2 0.21
2.6 0.55
NT(8-13) 0.23 0.042
[Lys8Lys9]NT(8-13)
a Ki values in nM SEM are based on the means of 2–7 experiments
each done in triplicate.
b Ki value of diastereomer 1 of 3c.
c Ki value of diastereomer 2 of 3c.
d Ki value of diastereomer 1 of 3d.
e Ki value of diastereomer 2 of 3d.
f Kd value from 17 homologous competition experiments.
substantial loss of affinity for both side-chain and back-
bone modifications. Displacement of the leucine residue
in position 13 resulted in a surprising biological beha-
viour of the b-amino acid homologue 1g since it showed
comparable affinity to the reference peptide NT(8-13)
(Ki = 3.2 nM). This is different to the observation from
our Ala- and D-amino acid scan that indicated an
approximately 100-fold loss of affinity.
To evaluate the effect of our structural modifications on
Gq transactivation, the most promising NT(8-13) ana-
logues exhibiting subnanomolar affinities were tested
for their ability to increase the intracellular calcium con-
centration. Comparison of calcium release at concentra-
tions of 100 nM with the ligand efficacy of the standard
agents NT, NT(8-13) and [Lys8,Lsy9]NT(8-13) indicated
full agonist properties and receptor response for the 8-b-
homo-NT(8-13) 1a, the quarternary ammonium salt 2a,
NT(8-13)-NH2 (2d), the C-terminally modified hepta-
peptide 2e and N-Lys derived oligomer 4a (Table 2).
For the peptoid scan, [Lys8,Lsy9]NT(8-13) was em-
ployed as a reference. The blue curve depicted in Figure 1
indicates susceptibilities that are comparable to the re-
sults of the alanine and D-amino acid exchange for the
respective residue positions. Interestingly, the substan-
tial loss of affinity for the NTyr-derivative (4c) could
be partially compensated by an elongation of the side-
chain leading to a higher affinity of the oligomer 4d.
As expected, modifications in position 12 were again
highly crucial when both diastereomers 4e and 4f
showed very poor ability to displace [3H]neurotensin.
The strong susceptibility of the peptoid analogue in po-
sition 13 compared to the highly potent b-amino acid
analogue 1g led us to conclude that a NH-backbone po-
sition close to the C-terminal end might be of primary
importance for the receptor recognition process. As a
complement to our previous investigations of lactam
bridged NT(8-13) mimetics conformationally constrain-
ing the backbone w-angle in position 11, the spirocyclic
NT(8-13) analogues 3a–c and their fused congener 3d
were also investigated for their ability to compete with
neurotensin at the NTR1 target receptor. Although
our previously reported spirocyclic c-lactam has sub-
stantial NTR1 affinity,13 the enlarged congeners 3a–d
incorporating 6-, 7- and 9-membered rings showed only
poor receptor binding with Ki values in the micromolar
range (Ki = 1.4–18 lM). This indicates that the exchange
Using data from previous efforts9 and the above de-
scribed SAR analysis, one major aim of this study is
to create a pharmacophore model that guides us to dis-
cover non-peptidic neurotensin receptor agonists. The
conceptional drawing depicted in Figure 2 displays the
Table 2. Ligand potency in the funtional calcium assay
Oligomer Rel. potencya Oligomer
Rel. potencya
NT 100% 16%
NT(8-13) 382% 24%
2d
2e
387% 14%
220% 25%
1a
2a
293% 25%
101% 35%
[Lys8Lys9]NT(8-13) 184% 19%
4a
309% 33%
Relative potency of derivatives showing subnanomolar affinities as
compared to the native ligand NT.
a Values SEM are based on the means of 6–12 experiments.

J. Einsiedel et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2013–2018
2017
major requirements that we suggest to be essential for
NTR1 recognition and activation. Two positive charges
attached to the Ca in position 9 within distances of 3–6
Acknowledgments
The authors thank A. Pinsker, S. Hemmers, I. Torres-
Berger and M. Kettler for skillful technical assistance
and Dr. H. Bittermann for helpful discussions. This
work was supported by the Deutsche Forschungsgeme-
inschaft (DFG Gm 13/7).
˚
and 5 A, respectively, are suggested to be required since
substitution of Arg8 by betaine (2a) retained ligand
affinity whereas the introduction of a negatively charged
carboxylate (2b, 2c) had an adverse effect on ligand
binding. The N-terminal portion is connected by a var-
iable linker to the functional unit of Pro10 and Tyr11
since the introduction of b-homo-Arg (1a), D-Arg or a
peptoid (4a) had no effect on ligand binding. Obviously,
the linker between the cationic centre and the backbone
is not fully extended at the native receptor-ligand com-
plex. Proline works as a unique ‘kink generating ele-
ment’ specifically redirecting Tyr11 towards an
aromatic pocket and an H-bond acceptor provided by
NTR1 residues that has to be addressed by Tyr11. This
explains the loss of affinity observed for our lactam-
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2008.01.110.
References and notes
1. Binder, E. B.; Kinkead, B.; Owens, M. J.; Nemeroff, C. B.
Pharmacol. Rev. 2001, 53, 453.
˚
bridged analogues (3a–d). Separated by a 1–4 A spacer
element, two bulky, hydrophobic moieties, which are
provided by Ile12 and Leu13 within the natural ligand,
are necessary. Additionally, H-bonding backbone-back-
bone interactions are crucial at this portion of the ligand
receptor complex, which was indicated by the above
mentioned SAR analysis. A polar, H-bonding C-termi-
nal portion and, not necessarily, a negatively charged
carboxylate is needed for high affinity ligand binding.
2. Tyler-McMahon, B.; Boules, M.; Richelson, E. Regul.
Peptides 2000, 93, 125.
3. Shilling, P. D.; Richelson, E.; Feifel, D. Behav. Brain Res.
2003, 143, 7.
4. St-Pierre, S.; Lalonde, J.-M.; Gendreau, M.; Quirion, R.;
Regoli, D.; Rioux, F. J. Med. Chem. 1981, 24, 370.
5. Hong, F.; Cusack, B.; Fauq, A.; Richelson, E. Curr. Med.
Chem. 1997, 4, 412, and references cited therein.
6. Rivier, J. E.; Lazarus, L. H.; Perrin, M. H.; Brown, M. R.
J. Med. Chem. 1977, 20, 1409.
In conclusion, structure activity relationship studies on a
series of NT(8-13) analogues provided valuable insights
into the major requirement for neurotensin receptor rec-
ognition and activation. The data facilitated the genera-
tion of a pharmacophore model. In combination with
our homology-based protein modelling and site directed
mutagenesis studies, this approach will guide us to non-
peptidic neurotensin receptor agonists that are suggested
to be of special interest for the treatment of
schizophrenia.
7. Cusack, B.; McCormick, D. J.; Pang, Y.-P.; Souder, T.;
Garcia, R.; Fauq, A.; Richelson, E. J. Biol. Chem. 1995,
270, 18359.
8. Henry, J. A.; Horwell, D. C.; Meecham, K. G.; Rees, D.
C. Bioorg. Med. Chem. Lett. 1993, 3, 949.
9. Pang, Y.-P.; Cusack, B.; Groshan, K.; Richelson, E.
J. Biol. Chem. 1996, 271, 15060.
`
10. Barroso, S.; Richard, F.; Nicolas-Etheve, D.; Reversat,
J.-L.; Bernassau, J.-M.; Kitabgi, P.; Labbe-Julie, C. J.
´
´
Biol. Chem. 2000, 275, 328.
11. Heyl, D. L.; Sefler, A. M.; He, J. X.; Sawyer, T. K.;
Wustrow, D. J.; Akunne, H. C.; Davis, M. D.; Pugsley, T.
A.; Heffner, T. G.; Corbin, A. E.; Cody, W. L. Int.
J. Peptide Protein Res. 1994, 44, 233.
12. Luca, S.; White, J. F.; Sohal, A. K.; Filippov, D. V.; van
Boom, J. H.; Grisshammer, R.; Baldus, M. Proc. Nat.
Acad. Soc. 2003, 100, 10706.
13. Bittermann, H.; Einsiedel, J.; Hubner, H.; Gmeiner, P.
¨
J. Med. Chem. 2004, 47, 5587.
14. Sefler, A. M.; He, J. X.; Sawyer, T. K.; Holub, K. E.;
Omecinsky, D. O.; Reily, M. D.; Thanabal, V.; Akunne,
H. C.; Cody, W. L. J. Med. Chem. 1995, 38, 249.
15. (a) Lundquist, J. T., IV; Dix, T. A. Bioorg. Med. Chem.
Lett. 1999, 9, 2579; (b) Lundquist, J. T., IV; Dix, T. A.
J. Med. Chem. 1999, 42, 4914.
16. (a) Nock, B. A.; Nikolopoulou, A.; Reubi, J.-C.; Maes, V.;
Conrath, P.; Tourwe, D.; Maina, T. J. Med. Chem. 2006,
49, 4767; (b) Maes, V.; Hultsch, C.; Kohl, S.; Bergmann,
R.; Hanke, T.; Tourwe, D. J. Pept. Sci. 2006, 12(8), 505.
17. (a) Achilefu, S.; Srinivasan, A.; Schmidt, M. A.; Jimenez,
H. N.; Bugaj, J. E.; Erion, J. L. J. Med. Chem. 2003, 46,
3403; (b) Hong, F.; Zaidi, J.; Cusack, B.; Richelson, E.
Bioorg. Med. Chem. 2002, 10, 3849.
18. For recent examples, see: (a) Hayen, A.; Schmitt, M. A.;
Ngassa, F. N.; Thomasson, K. A.; Gellman, S. H. Angew.
Chem. Int. Ed. 2004, 43, 505; (b) De Pol, S.; Zorn, C.;
Klein, C. D.; Zerbe, O.; Reiser, O. Angew. Chem. Int. Ed.
2004, 43, 511.
Figure 2. Pharmacophore model for NT(8-13) agonists. Functional
units are depicted as gray spheres. Straight black lines indicate fixed
distances and zigzag-lines indicate variable distances. Available space
is given for rough guidance on the left hand side.

2018
J. Einsiedel et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2013–2018
19. Frackenpohl, J.; Arvidsson, P. I.; Schreiber, J. V.;
Seebach, D. Chem. Bio. Chem. 2001, 2, 445.
29. Leuck, M.; Kunz, H. Carbohydr. Res. 1998, 312, 33.
30. Nakayama, T.; Nomura, M.; Haga, K.; Ueda, M. Bull.
Chem. Soc. Jpn. 1998, 71, 2979.
31. Trivedi, H. S.; Anson, M.; Steel, P. G.; Worley, J. Synlett
2001, 12, 1932.
32. Synthesis of the monomer is described in Kruijtzer, J. A.
W.; Liskamp, R. M. J. Tetrahedron Lett. 1995, 36,
6969.
20. (R)-3-Carboxypyrrolidine was synthesized from (R)-aspar-
tic acid according to: (a) Gmeiner, P.; Orecher, F.;
Thomas, C.; Weber, K. Tetrahedron Lett. 1995, 381; (b)
Thomas, C.; Orecher, F.; Gmeiner, P. Synthesis 1998,
1491; Subsequent Fmoc-protection was done according to
(c) Kim, Y. J.; Kaiser, D. A.; Pollard, T. D.; Ichikawa, Y.
Bioorg. Med. Chem. Lett. 2000, 10, 2417.
21. Weber, K.; Kuklinski, S.; Gmeiner, P. Org. Lett. 2000, 2,
647.
22. Stachowiak, K.; Khosla, M. C.; Plucinska, K.; Khairallah,
P. A.; Bumpus, F. M. J. Med. Chem. 1979, 22, 1128.
23. Boc-Leu-PAM resin was used (see, supporting
information).
33. For [D-Arg8]NT(8-13) and [D-Tyr11]NT(8-13), see: Ref. 9.
34. For [D-Arg9]NT(8-13) and [D-Leu13]NT(8-13), see: Gil-
bert, J. A.; McCormick, D. J.; Pfenning, M. A.; Kanba, K.
S.; Enloe, L. J.; Moore, A.; Richelson, E. Biochem.
Pharmacol. 1989, 38, 3377.
35. Bittermann, H.; Boeckler, F.; Einsiedel, J.; Gmeiner, P.
Chem. Eur. J. 2006, 24, 6315.
24. Prante, O.; Einsiedel, J.; Haubner, R.; Wester, P.; Gme-
iner, H.-J.; Kuwert, T.; Maschauer, S. Bioconjugate Chem.
2007, 18, 254.
25. For a recent example, see Kruijtzer, J. A. W.; Nijenhuis,
W. A. J.; Wanders, N.; Gispen, W. H.; Liskamp, R. M. J.;
Adan, R. A. H. J. Med. Chem. 2005, 48, 4224.
26. Kessler, H. Angew. Chem. Int. Ed. 1993, 32, 543.
27. Elongation with BrCH2CO2H/BrCH2COBr and nucleo-
philic substitution with the respective amine derivatives
was performed according to: Schmitt, J.; Bernd, M.;
Kutscher, B.; Kessler, H. Bioorg. Med. Chem. Lett. 1998,
8, 385, For 4a, b, f, g Wang resin was employed. For the
synthesis of 4g, 2-chlorotrityl resin was utilized to suppress
diketopiperazine formation. For the coupling of the
Fmoc-a-amino acids, HATU was applied.
36. Einsiedel, J.; Lanig, H.; Waibel, R.; Gmeiner, P. J. Org.
Chem. 2007, 72, 9102.
37. Our NTR1 homology model, containing NT(8-13) as a
docked ligand, indicated that steric hindrance in the region
of Ile329 could be a reason for the loss of affinity observed
for the bulky bridged NT(8-13) analogues. However, the
human NTR1 triple mutant YIS328-330AAA, constructed
by an overlap PCR methodology and expressed in HEK-
293 cells, did not regain affinity for the sterically most
demanding peptide mimetic 3d (Ki > 1000 nM). Thus, the
increase of the proline w-angle, which is a consequence of
the lactam-ring enlargement obviously leads to an inactive
backbone conformation.
38. NT(8-13) and NT(8-13)-NH2 affinities have also been
investigated in Ref. 9 (Ki values: 0.018 and 10 nM,
respectively) when NT receptors in neonatal mouse whole
brain preparations have been used, and in Ref.10.
28. Zuckermann, R. N.; Kerr, J. M.; Kent, S. B. H.; Moos, W.
H. J. Am. Chem. Soc. 1992, 114, 10646.