An imidazole based H-Phe-Phe-NH2 peptidomimetic with anti-allodynic effect in spared nerve injury mice

Article abstract of DOI:10.1016/j.bmcl.2018.06.009

The dipeptide amide H-Phe-Phe-NH₂ (1) that previously was identified as a ligand for the substance P 1–7 (SP_1–7) binding site exerts intriguing results in animal models of neuropathic pain after central but not after peripheral administration. The dipeptide 1 is derived from stepwise modifications of the anti-nociceptive heptapeptide SP_1–7 and the tetrapeptide endomorphin-2 that is also binding to the SP_1–7 site. We herein report a strong anti-allodynic effect of a new H-Phe-Phe-NH₂ peptidomimetic (4) comprising an imidazole ring as a bioisosteric element, in the spare nerve injury (SNI) mice model after peripheral administration. Peptidomimetic 4 was stable in plasma, displayed a fair membrane permeability and a favorable neurotoxic profile. Moreover, the effective dose (ED₅₀) of 4 was superior as compared to gabapentin and morphine that are used in clinic.

Full text of DOI:10.1016/j.bmcl.2018.06.009

Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
An imidazole based H-Phe-Phe-NH peptidomimetic with anti-allodynic
2
effect in spared nerve injury mice
a,e
b,e
a
c
c
Anna Skogh , Anna Lesniak , Christian Sköld , Maria Karlgren , Fabienne Z. Gaugaz ,
c
b
b
a
b
Richard Svensson , Shanti Diwakarla , Anna Jonsson , Rebecca Fransson , Fred Nyberg ,
Mathias Hallberg , Anja Sandström
Department of Medicinal Chemistry, Uppsala University, BMC, Box 574, SE-751 23 Uppsala, Sweden
The Beijer Laboratory, Department of Pharmaceutical Bioscience, Uppsala University, BMC, Box 591, SE-751 24 Uppsala, Sweden
Uppsala Drug Optimization and Pharmaceutical Profiling Platform (UDOPP), Science for Life Laboratory Drug Discovery and Development Platform, Department of
b
a,d,⁎
a
b
c
Pharmacy, Uppsala University, BMC, Box 580, SE-751 23 Uppsala, Sweden
d
The Beijer Laboratory, Department of Medicinal Chemistry, Uppsala University, BMC, Box 574, SE-751 23 Uppsala, Sweden
A R T I C L E I N F O
A B S T R A C T
Keywords:
2
The dipeptide amide H-Phe-Phe-NH (1) that previously was identified as a ligand for the substance P 1–7
Substance P 1–7
Peptidomimetics
Neuropathic pain
SNI mice
(SP1–7) binding site exerts intriguing results in animal models of neuropathic pain after central but not after
peripheral administration. The dipeptide 1 is derived from stepwise modifications of the anti-nociceptive hep-
tapeptide SP1–7 and the tetrapeptide endomorphin-2 that is also binding to the SP1–7 site. We herein report a
strong anti-allodynic effect of a new H-Phe-Phe-NH
bioisosteric element, in the spare nerve injury (SNI) mice model after peripheral administration. Peptidomimetic
was stable in plasma, displayed a fair membrane permeability and a favorable neurotoxic profile. Moreover,
the effective dose (ED50) of 4 was superior as compared to gabapentin and morphine that are used in clinic.
2
peptidomimetic (4) comprising an imidazole ring as a
Phenylalanine bioisostere
4
The neuropeptide substance P (SP, H-Arg-Pro-Lys-Pro-Gln-Gln-Phe-
Phe-Gly-Leu-Met-NH ) that acts through the G-protein coupled neuro-
multiple sclerosis (MS) or spinal cord injury (SCI).13,14 Current first-line
treatment includes tricyclic antidepressants (TCA), serotonin and nor-
adrenaline reuptake inhibitors (SNRI) and anticonvulsants like gaba-
2
kinin-1 receptor, is an important neurotransmitter and neuromodulator
1
15
of pain signaling in the nervous system. As other tachykinins, the C-
pentin and pregabalin, all hampered by CNS side effects. Gabapentin
terminal part accounts for the signal transduction at the receptor,
is commonly used in clinic and is in comparison relatively safe despite
1
6
whereas the N-terminal part of the undecapeptide is important for re-
the high doses needed vide infra to combat neuropathic pain. Since
opioids are addictive and associated with burdensome side effects like
constipation, they are considered only as third line treatment. Morphine
is a representative example of opioids used in clinic and is admini-
2
ceptor subtype selectivity. SP coexists in neurons with classic trans-
mitters and is after release subjected to enzymatic cleavage, which re-
sults in fragments, some with very diverse bioactivities.3–5 The N-
terminal fragment, substance P 1–7 (SP1–7, Fig. 1) has attracted most
attention owing to its biological effects that oppose those elicited by the
parent peptide with regard to nociception as well as opioid tolerance
strated for severe nociceptive pain, which also provides effectiveness in
short-term neuropathic pain relief.1
7,18
In light of this, safe pharma-
cological treatments that more efficiently target other sites in the
neuropathic pain pathways are of interest as future drug candidates.
Several observations suggest that SP1–7 binds to a unique binding
site, probably G-protein dependent and distinct from the neurokinin
6–9
and withdrawal.
Our interest in this heptapeptide was reinforced
owing to its anti-hyperalgesic and anti-allodynic effects in animal
1
0–12
models of neuropathic pain,
and this prompted us to commence a
medicinal chemistry program aimed at converting the heptapeptide
into drug-like peptidomimetics.
receptors or other receptors associated with pain such as opioid re-
ceptors.1
9,20
In line with this, the binding is specific, saturable, and
Neuropathic pain affects 7–10% of the general population and is a
complex clinical condition that can arise as a result of several under-
lying etiologies such as diabetes mellitus, Herpes zoster infection,
reversible. However, it is possible that the heptapeptide elicits its action
by interfering with allosteric sites present on other neuropeptide re-
ceptors. It is notable that the μ-receptor agonist endomorphin-2 (EM-2,
3
⁎
Corresponding author.
E-mail address: anja.sandstrom@ilk.uu.se (A. Sandström).
These authors contributed equally.
e
https://doi.org/10.1016/j.bmcl.2018.06.009
Received 21 May 2018; Accepted 3 June 2018
0960-894X/ © 2018 Elsevier Ltd. All rights reserved.
Please cite this article as: Skogh, A., Bioorganic & Medicinal Chemistry Letters (2018), https://doi.org/10.1016/j.bmcl.2018.06.009

A. Skogh et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Fig. 1. Lead compound 1 derived from SAR studies of SP1–7 and EM-2 along with three rigidified and terminal modified analogues 2, 3 and 4. Compounds evaluated
3
in a displacement binding assay using spinal cord membrane from rats and radioactive [ H]SP1–7 tracer.
Fig. 1), but not EM-1, exhibited high affinity for the SP1–7 binding
site.
Herein, as an alternative to the primary amine in 2 and the carba-
mate in 3, compound 4 was designed and prepared, comprising an C5
phenyl substituted imidazole carboxamide in the N-terminal (Fig. 1).
This phenylalanine isostere match both the side chain phenyl and
contains nitrogen atoms that can act as hydrogen bond donors (HBD),
hydrogen bond acceptors (HBA), and sites for positive charges, de-
pending on tautomeric state and pH. The assessment of in vitro plasma
stability and permeability as well as its in vitro cytotoxic profile in
primary neuronal cell cultures of the new imidazole based H-Phe-Phe-
2
0,21
2
7
In our ongoing medicinal chemistry program, we aimed at devel-
oping pharmacological tools for in depth studies of the SP1–7 system but
also employing SP1–7 as starting point in a design process with the long
term goal to discover new chemical entities as potential new ther-
apeutics useful for treatment of neuropathic pain. Important anchor
points of SP1–7 and EM-2 were identified through a structure activity
relationship (SAR) study.2
2,23
The extensive SAR study revealed that a
C-terminal primary amide is improving the binding affinity and, hence,
resulted in the identification of SP1–7-NH with a K of 0.3 nM (Fig. 1).
N-terminal truncation of SP1–7-NH and EM-2 along with an alanine
scan led to the remarkable discovery of the dipeptide lead compound 1
H-Phe-Phe-NH ) with equipotent affinity (K = 1.5 nM) as the SP1–7
itself (Fig. 1).
2
NH peptidomimetic 4 is presented. Furthermore, the in vivo anti-allo-
dynic potential of the compound in an SNI model of neuropathic pain
after peripheral administration (i.p.) was assessed. Moreover, to con-
front its action with the SP1–7-NH peptide and with gabapentin and
2
morphine used in clinic the activity of the four compounds were com-
pared in the SNI model.
2
i
2
(
2
i
A pharmacophore hypothesis was proposed for 1 where (S,S) con-
figuration of the side-chains, primary amine in the N-terminal and
primary amide in the C-terminal were essential features for binding.24
The in vivo effect of 1 was further evaluated in a diabetic neuropathy
model induced by streptozotocin (STZ). Following central (intrathecal,
i.t.) administration 1 produced potent anti-allodynic and anti-hyper-
The introduction of the ortho-substituted imidazole moiety in 4
complies with a suitable atom-to-atom match to the N-terminal phe-
nylalanine residue in 1 (Fig. 2). In order to investigate if 1 and 4 can
obtain similar spatial arrangement of their potential pharmacophore
groups in low energy conformations, a pharmacophore group alignment
analysis was performed using Phase.28 Conformational analysis of 1 and
2
5
−1
algesic effect at a 0.5–4 pmol dosage range. However, when evaluated
in mice suffering from spared nerve injury (SNI) after peripheral (in-
traperitoneal, i.p.) administration of a dose at 185 nmol/kg, the di-
peptide amide 1 failed to reach any distinct anti-allodynic effect.12 This
observation is probably due to low brain exposure since 1 displayed
poor drug-like properties such as poor plasma stability, high hepatic
metabolism and low permeability and stability over Caco-2 cells.24
With the aim to develop drug-like analogues to 1 acting as the
neuropeptide SP1–7, compound 2 and 3 were designed (Fig. 1).2
Both compounds exhibited retained binding in vitro along with im-
proved metabolic stability and permeability. When evaluated in vivo in
the SNI model, the rigidified compound 2 displayed a strong anti-al-
lodynic effect. The effect, however, was short-lasting with a peak of
effect after 30 min, which may be attributed to low brain exposure due
to high blood-brain barrier (BBB) efflux as indicated in vitro by Caco-2
4 resulted in 69 and 76 conformations, respectively, within 21 kJ mol
of the lowest energy conformation found, which were included in the
pharmacophore group alignment analysis. Requirement of matching all
the mutual potential pharmacophore groups identified by Phase (three
HBD, two HBA, and two aromatic rings), produced six clusters of
pharmacophore group match hypotheses. The match with the lowest
−1
−1
energy conformations (ΔE = 3.6 kJ mol
and 7.8 kJ mol
from the
lowest energy conformation found for 4 and 1, respectively) is pre-
sented in Fig. 2 and shows a very good overlap of the mutual structural
features that can be of importance for target-ligand interactions. Taken
together, the analysis show that compounds 4 and 1 indeed can obtain a
similar spatial arrangement of their potential pharmacophore groups in
low energy conformations. Albeit the rigidified 4 should not be able to
cover all spatial arrangements of the more flexible 1, the analysis
supports that the ortho-substituted imidazole can act as a suitable mimic
of the N-terminal phenylalanine residue in 1.
4,26
1
2
cells measurements. The carbamate 3 on the other hand, displayed no
in vitro efflux over the Caco-2 monolayer and was shown to enter the
CNS according to an in vivo infusion study but failed to exhibit any anti-
allodynic effect in the SNI mice, possibly as a results of fast hydrolysis
of the carbamate functionality in plasma.12,26
The synthesis of 4 was performed adopting a previously described
method27 and is outlined in Scheme 1. The Pd-catalyzed C5 arylation of
1-benzyl-1H-imidazole was conducted under microwave irradiation for
2 3
1 h at 160 °C using bromobenzene, Pd(OAc) , P(2-furyl) , pivalic acid
2

A. Skogh et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
6
was used which enabled ex situ generation of CO from Mo(CO) . The
reaction was run at 120 °C overnight and 7 was obtained in a 75% yield.
Final reductive Pd-catalyzed N-debenzylation provided the target
compound 4 (76%), resulting in an overall yield of 28% over four steps.
First, compound 4 was subjected to in vitro plasma stability and
Caco-2 permeability in both the apical to the basolateral (a – b) and the
basolateral to apical (b – a) direction in order to evaluate important
pharmacokinetic parameters (Table 1). The introduction of an imida-
zole phenylalanine mimetic in the N-terminal improved the plasma
stability and the compound was completely stable for more than
1
80 min in both human and mouse plasma. Moreover, Caco-2 perme-
29
−6
ability revealed a moderate permeability of 0.84 × 10 cm/s (a – b).
In our laboratory’s classification, a Papp value below 0.2 × 10
−
6
cm/s
indicates a low permeability, a Papp value ranging from 0.2 to
−
6
−6
1
.6 × 10
cm/s moderate, and a Papp value above 1.6 × 10
cm/s
30
indicates a high permeability. The a – b Papp value is however affected
by a significant efflux as shown by the ratio of 19, most likely due to the
efflux transporter multidrug resistance protein 1 (protein name: mul-
tidrug resistance protein 1 [MDR1] often referred to as P-glycoprotein
[
P-gp]; gene name: ABCB1).31 It is possible that the high concentration
(
100 µM) used here may, at least partly, saturate the MDR1/P-gp
transporter and thereby result in a higher apical to basolateral (a – b)
permeability than what would have been seen if a lower concentration
was used. According to previous studies (vide supra) an N-terminal basic
moiety is important for binding to the SP1–7 binding site, but also for
the efflux. The introduction of the less basic imidazole scaffold in the N-
terminal did not circumvent the efflux problem.
Fig. 2. Potential pharmacophore group alignment of 1 (green carbons) and 4
(orange carbons). Blue sphere = HBD, red sphere = HBA, orange torus = aro-
matic ring.
Compound 4 did not elicit any significant neurotoxic effect in vitro
in rat primary cortical cell cultures as compared to control. Neither did
4
affect cell viability in the MTT assay that addresses the mitochondrial
function, nor in the LDH assays that is associated with membrane in-
tegrity and measure lactate dehydrogenase release from the cells
(
Fig. 3). Thus, a favorable safety profile was demonstrated.
Compound 4 was evaluated for its in vivo anti-allodynic efficacy in
the SNI model32 of neuropathic pain (Fig. 4). At 14 days after surgery, 4
was administrated i.p. and mechanical threshold was assessed by using
the von Frey test.3
3,34
The anti-allodynic effect of 4 was examined at
12
four different dosages (1.85–1850 nmol/kg)
acidified saline as control.
and compared with
Scheme 1. Synthesis of target compound 4. aReaction conditions: (I) The re-
action was performed under microwave irridation in a sealed vial at 160 °C for
Peripheral (i.p.) administration of 4 produced a time and dose de-
pendent increase of mechanical thresholds in mice with SNI-induced
neuropathic pain (Fig. 4A). Mice injected with a dose of 1850 nmol/kg
and 185 nmol/kg of 4 showed a significant decrease in mechanical
threshold (Fig. 4B). The anti-allodynic effect of 4 was transient and
lasted 30 min whereupon it declined rapidly and reached control values
at 45 min. Gratifyingly, the results suggest that the N-terminal rigidi-
fication displayed in compound 4 is allowed for in vivo efficacy and that
the ortho-phenyl substituted imidazole carboxamide serves as an ef-
fective phenylalanine bioisostere replacement.
1
K
h: 1-Benzyl-1H-imidazole, bromobenzene, Pd(OAc)2, P(2-furyl)3, PivOH and
CO in DMF. (II) NIS, TFA in MeCN, 80 °C for 1 h. (III) The reaction was
2
3
performed in a sealed two-chamber system at 120 °C overnight: Chamber A: Mo
CO) , DBU in DMF. Chamber B: 6, H-Phe-NH %HCl, Pd(PPh , K CO and
DMAP in 1,4-dioxane. (IV) The reaction was performed in an autoclave at 60 °C
for 6 h with H (10 bar): 10% Pd/C in MeOH.
(
6
2
3
)
4
2
3
2
(
PivOH) and potassium carbonate in DMF, securing 5 in 64% yield.
Next, iodination was performed by mixing 5 with N-iodosuccinimide
NIS) and catalytic trifluoroacetic acid (TFA) in acetonitrile (MeCN).
Furthermore, assuming that the observed in vivo effect requires BBB
(or blood-spinal cord barrier or blood-nerve barrier) transport for a
centrally mediated effect, the data (Fig. 4.) suggest that 4 has sufficient
exposure, despite the high efflux measured in vitro and at fairly low
(
The reaction was stirred at 80 °C in 1 h, giving 6 in 78% yield. In the Pd-
catalyzed aminocarbonylation reaction a two-chamber system set-up
Table 1
In vitro plasma stability and permeability study of compound 4.
Compound
Plasma stability
(min)
Caco-2 permeability
a
−6
t
½
P
app (×10
cm/s)
Mouse
> 180
Human
> 180
(a – b)b
0.84 ± 0.11c
Mass balance (%)
65
(b – a)b
Mass balance (%)
67
Efflux ratiod
4
16.0 ± 1.50c
19
a
P
app = apparent permeability coefficient.
b
c
(a – b) = apical to basolateral.
Average of three filters. Results are expressed as mean ± SD.
Efflux ratio (b – a)/(a – b).
d
3

A. Skogh et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Fig. 5. Effect of D-SP1–7 (1850 nmol/kg, i.p.) on the anti-allodynic effect of 4
(185 nmol/kg, i.p.) in SNI mice (n = 6–9). Control mice were injected with
0.1 M acidified saline. Statistical significance was marked as follows:
Fig. 3. Primary cortical cell culture cytotoxic death (A) and viability (B) mea-
sured in the LDH and MTT assay, respectively, after treatment with 4. Doses are
in 10 –10 M. Triton X-100 served as positive control.
*
*
−
9
−5
p < 0.01 (vs. relevant control).
2
pretreated with an antagonist to the SP1–7 binding site, [D-Pro , D-
7
20
Phe ]SP1–7 (D-SP1–7, K
i
= 1.8 nM) (Fig. 5). This antagonist have
previously been documented to reverse several in vivo effects displayed
8,35
by the SP1–7
.
Here, we could for the first time reverse the analgesic
effect produced by a small peptidomimetic compound with D-SP1–7
.
Mice receiving 4 that were pretreated with D-SP1–7 showed lower peak
mechanical threshold values than mice injected with injected with 4
alone. D-SP1–7 alone had no effect on mechanical thresholds in SNI
mice. This observation implies that 4 and the full-length D-SP1–7 an-
tagonist interacts with the same molecular target.
As mentioned before, several interesting effects on neuropathic pain
have been observed for the full-length SP1–7 and its amidated analogue
2
SP1–7-NH . More specifically, the peptides have displayed anti-hyper-
algesic effect in diabetic mice following i.t. administration as well as
anti-allodynic effect in rats suffering from the spinal cord injury (SCI)
and in SNI mice following i.p. administration, even though the plasma
Fig. 4. Time course (A) and total effect (B) of i.p. administration of 4 on me-
chanical von Frey threshold SNI mice (n = 15–18). Doses are given in nmol/kg.
Control mice were injected with 0.1 M acidified saline. Statistical significance
1
0–12,38,39
half-life was measured to 4.4 and 6.4 min, respectively.
Throughout the studies, the observed effect for the amidated analogue
is higher than for the endogenous SP1–7 which corresponds well with
binding data (Fig. 1). Here, the potent anti-allodynic effect evoked by 4
was marked as follows: * p < 0.05; p < 0.01;
**
***
p < 0.001 (vs. control).
doses. Thus, the in vitro Caco-2 assay may be a poor predictor of brain
exposure in these mice, i.e. 4 is an apparent substrate for MDR1/P-gp,
but has sufficient passive permeation and/or aided by active uptake
was compared to corresponding treatment with the full-length SP1–7
NH peptide and gratifyingly the small peptidomimetic compound 4
displayed equipotent effect as the full-length SP1–7-NH (Fig. 6A). In-
-
2
2
3
1
transporters in vivo.
terestingly, this observation indicates that the prolonged plasma sta-
bility (180 min vs. 6.4 min) did neither result in a longer-lasting or
increased in vivo anti-allodynic effect, which is in agreement with
However, it cannot be excluded that a peripheral target for full-
length SP1–7 and corresponding peptidomimetics, such as 4, is involved
in signal transduction of the anti-allodynic effect. Observations
strengthening this hypothesis were shown in a previous study in-
dicating that SP1–7 modulates neurogenic inflammatory process in
peripheral tissue by inhibition of the vascular inflammatory response. It
has been speculated that the heptapeptide interacts with a post-term-
4
0
2
previous observation from N-methylated SP1–7-NH peptides.
Finally, compound 4 was compared to gabapentin and morphine
and dose-response curves are shown in Fig. 7A. When compared with
3
5
inal binding site located on endothelial cells. In contrast, observations
indicating that the SP1–7 analogues exert their effect through a centrally
located target is that specific binding sites have been observed in the
1
9,20
CNS.
In line with this, as mentioned before, when 1 is delivered i.t. at a
dose in pico molar range it exerts anti-allodynic and anti-hyperalgesic
2
5
effects in diabetic mice. The effectiveness of low dose administration
indicated a centrally located binding site. Additionally, it has been
shown that peripheral nerve injury can cause an increase in BBB, as
well as blood-spinal cord barrier and blood-nerve barrier perme-
3
6,37
ability.
Altogether, whether the SP1–7 peptidomimetic, such as
compound 4 described herein, exerts its effect through a central or
peripheral target, or both, needs to be further elucidated.
To further evaluate whether the small peptidomimetic compound 4
really acts and bind as the mother peptide SP1–7, the SNI mice were
Fig. 6. Time course (A) and total effect (B) of i.p. administration of 185 nmol/
kg of SP1–7-NH
2
and 4 on mechanical von Frey threshold in SNI mice
(n = 13–18). No statistical significance between the groups.
4

A. Skogh et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
experiments. Design-wise, compound 4 proves that an ortho-phenyl
substituted imidazole carboxamide serves as a bioisosteric replacement
of the N-terminal phenylalanine in 1 based on the strong in vivo effect.
Altogether, these encouraging data suggest that 4 (called AS097)
might serve as a promising starting point for further development into
improved analgesics aimed for the management of neuropathic pain.
Acknowledgement
We gratefully acknowledge Kjell and Märta Beijer Foundation, the
Berzelii Technology Centre for Neurodiagnostics, the Swedish Medical
Research Council (Grant no. 9459) for financial support.
Fig. 7. Dose-response curves of 4 (1.85–1850 nmol/kg, i.p.), gabapentin
(
18.5–585 µmol/kg, i.p.) and morphine (0.185–12 µmol/kg, i.p.) (A) and the
A. Supplementary data
control subtracted total anti-allodynic effect (AUC) (B) on mechanical von Frey
threshold in SNI mice (n = 10–18). Statistical significance was marked as fol-
lows: p < 0.001 (vs. gabapentin), # p < 0.05 (vs. morphine).
*
*
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.bmcl.2018.06.009.
Table 2
References
ED50 values of 4, gabapentin and morphine.a
1.
2.
3.
4.
5.
6.
Otsuka M, Yoshioka K. Physiol Rev. 1993;73:229–308.
Nakanishi S, Nakajima Y, Yokota Y. Regul Pept. 1993;46:37–42.
Hallberg M. Med Res Rev. 2015;35:464–519.
Nyberg F, Hallberg M. Curr Drug Targets. 2007;8:147–154.
Hallberg M, Nyberg F. Curr Protein Pept Sci. 2003;4:31–44.
Sakurada C, Watanabe C, Sakurada S, Tan-No K, Sakurada T. J Pharm Sci.
Compound
ED50 (µmol/kg)
4
0.21**#
336
6.5
Gabapentin
Morphine
1
999;88:1127–1132.
a
Statistical significance was marked as follows: **
7. Sakurada C, Watanabe C, Sakurada T. Methods Find Exp Clin Pharmacol.
2004;26:171–176.
Kreeger JS, Larson AA. J Pharmacol Exp Ther. 1996;279:662–667.
p < 0.001 (vs. gabapentin), # p < 0.05 (vs. mor-
phine).
8
9
.
.
Zhou Q, Carlsson A, Hallberg M, Nyberg F. Peptides. 2011;32:1661–1665.
1
0. Carlsson A, Ohsawa M, Hallberg M, Nyberg F, Kamei J. Eur J Pharmacol.
2
010;626:250–255.
11. Carlsson-Jonsson A, Gao T, Hao J-X, et al. Eur J Pharmacol. 2014;738:319–325.
12. Jonsson A, Fransson R, Haramaki Y, et al. Neuroscience. 2015;298:112–119.
13. Van Hecke O, Austin SK, Khan RA, Smith BH, Torrance N. Pain. 2014;155:654–662.
14. Costigan M, Scholz J, Woolf CJ. Annu Rev Neurosci. 2009;32:1–32.
gabapentin, 4 showed effective pain relief at a 1600-times lower ef-
fective dose. This observation is in line with clinical data showing that
very high doses of gabapentin ranging from 1800 to 3600 mg/day are
15. Finnerup NB, Attal N, Haroutounian S, et al. Lancet Neurol. 2015;14:162–173.
16. Kremer M, Salvat E, Muller A, Yalcin I, Barrot M. Neuroscience. 2016;3:183–206.
required to achieve adequate pain suppression.41 Furthermore, a nearly
1
7. Gilron I, Bailey JM, Tu D, Holden RR, Weaver DF, Houlden RL. N Engl J Med.
30-times superiority in effectiveness was also observed over morphine.
2
005;352:1324–1334.
Additionally, the total anti-allodynic effect confirmed that it was a
significant difference between the groups (Fig. 7B), since mice injected
with 0.185 µmol/kg of 4 showed significantly higher AUC values than
mice treated with 18.5 µmol/kg gabapentin or 0.185 µmol/kg of mor-
phine.
ED50 values for these compounds are displayed in Table 2. The in-
jection of 4 evoked a half-maximum peak anti-allodynic effect at a dose
of 0.21 µmol/kg. Thus, 4 was more potent than both gabapentin with an
ED50 value of 336 µmol/kg and morphine with an ED50 value of
18. McNicol ED, Midbari A, Eisenberg E. Opioids for Neuropathic Pain. Rev: Cochrane
Database Syst; 2013.
1
9. Igwe OJ, Kim DC, Seybold VS, Larson AA. J Neurosci. 1990;10:3653–3663.
2
0. Botros M, Hallberg M, Johansson T, et al. Peptides. 2006;27:753–759.
21. Botros M, Johansson T, Zhou Q, et al. Peptides. 2008;29:1820–1824.
2
2. Fransson R, Botros M, Nyberg F, Lindeberg G, Sandström A, Hallberg M.
Neuropeptides. 2008;42:31–37.
2
3. Fransson R, Botros M, Sköld C, et al. J Med Chem. 2010;53:2383–2389.
24. Fransson R, Sköld C, Kratz JM, et al. J Med Chem. 2013;56:4953–4965.
2
2
2
5. Ohsawa M, Carlsson A, Asato M, et al. Mol Pain. 2011;7:85.
6. Fransson R, Nordvall G, Bylund J, et al. ACS Med Chem Lett. 2014;5:1272–1277.
7. Skogh A, Fransson R, Sköld C, Larhed M, Sandström A. J Org Chem.
2013;78:12251–12256.
6
5
3
.5 µmol/kg. However, when increasing the dose of gabapentin to
85 µmol/kg a strong anti-allodynic effect was observed with a peak at
28. Phase, version 4.5; Schrödinger, LLC: New York, NY, 2015.
29. Hubatsch I, Ragnarsson EGE, Artursson P. Nat Protoc. 2007;2:2111–2119.
0 min, lasting for 60 min and reaching control values after 180 min.40
3
0. Bergström CAS, Strafford M, Lazorova L, Avdeef A, Luthman K, Artursson P. J Med
Chem. 2003;46:558–570.
Morphine, displayed anti-allodynic effect when mice were injected with
2 µmol/kg. The effect was short-lasting, with a peak at 45 min (see
1
31. Giacomini KM, Huang S-M, Tweedie DJ, et al. Nat Rev Drug Discov. 2010;9:215–236.
3
3
2. Bourquin AF, Süveges M, Pertin M, et al. Pain. 2006;122 14.e1-14.e14.
3. Chaplan SR, Bach FW, Pogrel JW, Chung JM, Yaksh TL. J Neurosci Methods.
Supplementary data).
In conclusion, evaluation of a conformationally constrained imida-
1
994;53:55–63.
2
zole based H-Phe-Phe-NH analogue 4 revealed improved plasma sta-
34. Dixon W. J Am Stat Assoc. 1965;60:967–978.
35. Wiktelius D, Khalil Z, Nyberg F. Peptides. 2006;27:1490–1497.
36. Beggs S, Liu XJ, Kwan C, Salter MW. Mol Pain. 2010;6:74.
37. Lim TKY, Shi XQ, Martin HC, et al. Pain. 2014;155:954–967.
bility, a good cell permeability but also a relatively high efflux ratio in
the in vitro experiments. More importantly, compound 4 exhibits a
strong anti-allodynic effect in the SNI mice model after peripheral ad-
ministration. In fact, 4 exhibits superior effects as compared to both
gabapentin and morphine that are used in clinic. Moreover, 4 displayed
a favorable neurotoxic profile as deduced from cortical cell culture
38. Ohsawa M, Carlsson A, Asato M, et al. Peptides. 2011;32:93–98.
39. Skogh A, Lesniak A, Gaugaz FZ, et al. Eur J Pharm Sci. 2017;106:345–351.
40. Skogh A, Lesniak A, Gaugaz FZ, et al. Eur J Pharm Sci. 2017;109:533–540.
41. Backonja M, Glanzman RL. Clin Ther. 2003;25:81–104.
5