Chemical conjugation of the neuropeptide kyotorphin and ibuprofen enhances brain targeting and analgesia

Article abstract of DOI:10.1021/mp2003016

The pharmaceutical potential of natural analgesic peptides is mainly hampered by their inability to cross the blood-brain barrier, BBB. Increasing peptide-cell membrane affinity through drug design is a promising strategy to overcome this limitation. To address this challenge, we grafted ibuprofen (IBP), a nonsteroidal anti-inflammatory drug, to kyotorphin (l-Tyr-l-Arg, KTP), an analgesic neuropeptide unable to cross BBB. Two new KTP derivatives, IBP-KTP (IbKTP-OH) and IBP-KTP-amide (IbKTP-NH₂), were synthesized and characterized for membrane interaction, analgesic activity and mechanism of action. Ibuprofen enhanced peptide-membrane interaction, endowing a specificity for anionic fluid bilayers. A direct correlation between anionic lipid affinity and analgesic effect was established, IbKTP-NH₂ being the most potent analgesic (from 25 μmol·kg^-1). In vitro, IbKTP-NH ₂ caused the biggest shift in the membrane surface charge of BBB endothelial cells, as quantified using zeta-potential dynamic light scattering. Our results suggest that IbKTP-NH₂ crosses the BBB and acts by activating both opioid dependent and independent pathways.

Full text of DOI:10.1021/mp2003016

ARTICLE
pubs.acs.org/molecularpharmaceutics
Chemical Conjugation of the Neuropeptide Kyotorphin and Ibuprofen
Enhances Brain Targeting and Analgesia
Marta M. B. Ribeiro,† Antꢀonia R. T. Pinto,‡,§ Marco M. Domingues,† Isa Serrano,† Montserrat Heras,||
Eduard R. Bardaji,|| Isaura Tavares,‡,§ and Miguel A. Castanho*,†
†Instituto de Medicina Molecular, Faculdade de Medicina de Lisboa, Av. Professor Egas Moniz, 1649-028 Lisboa, Portugal
‡Instituto de Histologia e Embriologia, Faculdade de Medicina do Porto, Alameda Professor Hern^ani Monteiro, 4200-319 Porto,
Portugal
§Instituto de Biologia Celular e Molecular, Rua Campo Alegre, 4150-180 Porto, Portugal
Laboratori d'Innovaciꢀo en Processos i Productes de Síntesi Orgꢁanica (LIPPSO), Departament de Química, Universitat de Girona,
Campus Montilivi, 17071 Girona, Spain
S Supporting Information
b
ABSTRACT: The pharmaceutical potential of natural analgesic
peptides is mainly hampered by their inability to cross the
bloodꢀbrain barrier, BBB. Increasing peptideꢀcell membrane
affinity through drug design is a promising strategy to overcome
this limitation. To address this challenge, we grafted ibuprofen
(IBP), a nonsteroidal anti-inflammatory drug, to kyotorphin
(L-Tyr-L-Arg, KTP), an analgesic neuropeptide unable to cross
BBB. Two new KTP derivatives, IBP-KTP (IbKTP-OH) and
IBP-KTP-amide (IbKTP-NH2), were synthesized and charac-
terized for membrane interaction, analgesic activity and mechanism of action. Ibuprofen enhanced peptideꢀmembrane interaction,
endowing a specificity for anionic fluid bilayers. A direct correlation between anionic lipid affinity and analgesic effect was
established, IbKTP-NH2 being the most potent analgesic (from 25 μmol kgꢀ1). In vitro, IbKTP-NH2 caused the biggest shift in the
3
membrane surface charge of BBB endothelial cells, as quantified using zeta-potential dynamic light scattering. Our results suggest
that IbKTP-NH2 crosses the BBB and acts by activating both opioid dependent and independent pathways.
KEYWORDS: kyotorphin, ibuprofen, analgesia, bloodꢀbrain barrier, peptideꢀmembrane interaction, lipophilicity
KTP into lipophilic analogues.6,7 Accordingly, a direct correla-
1. INTRODUCTION
tion between improved interaction with lipid bilayers and
Kyotorphin (KTP) is an endogenous dipeptide (L-Tyr-L-Arg)
analgesic activity for KTP derivatives has been reported.8
first isolated from bovine brain1 and subsequently found in the
Recently, we discovered and characterized a new kyotorphin
brains of other mammals, including humans.2 It acts as a neuro-
derivative, KTP-amide.9 This new peptide, KTP-NH2, is analgesic
transmitter/neuromodulator in nociceptive responses in the central
upon systemic administration to animal models, which, combined
nervous system, having an analgesic potency approximately 4.2-fold
with an inhibition of nociceptive responses of spinal neurons,
higher than endogenous opioid peptides.3 Other activities have been
suggests a centraleffect of the drug.9 Although KTP-NH2 analgesia
proposed such as inhibiting cell proliferation and antihibernating
is inhibited by naloxone administration (either ip or i.t.), direct
regulation.4 The mechanism of action of this multifunctional peptide
binding of KTP-NH2 to opioid receptors is nearly absent, similarly
remains, however, unknown. There is evidence suggesting that
to the original, nonamidated, dipeptide. Also, amidation of KTP
KTP does not bind to opioid receptors, raising the possibility
appeared to enhance peptide partition to biomembranes. Never-
of a specific receptor for KTP, not yet identified. Whatever the
theless, the peptide’s reduced size and susceptibility to enzyme
mechanism, a release of Met-encephalin is acknowledged by all
degradation (resulting in a high oral dose required to induce
the authors.5
analgesia9) prompted us to design an improved derivative.
In the clinics, a combination of more than one analgesic is
frequently used with success. As a matter of fact, the complexity
Despite its interesting properties, KTP is only active when
injected directly into the brain. When systemically administered,
it only shows brief activity and at a high dose of 200 mg kgꢀ1 6
.
3
This limited capacity to cross the BBB, mainly related to both
Received: June 16, 2011
Accepted: August 10, 2011
insufficient lipophilicity and susceptibility to various lytic en-
zymes, confines KTP pharmacological applications. A logical
strategy to target KTP to the brain is the chemical modification of
Revised:
July 28, 2011
Published: August 10, 2011
r
2011 American Chemical Society
1929
dx.doi.org/10.1021/mp2003016 Mol. Pharmaceutics 2011, 8, 1929–1940
|

Molecular Pharmaceutics
ARTICLE
of tissue injury, inflammation, signal transduction and pain path-
ways demands a multitarget strategy, attacking multiple pain
sources and mechanisms altogether.10 Herein, we propose an
innovative covalent combination of kyotorphin (both amidated
and nonamidated) with ibuprofen, a lipophilic nonsteroidal anti-
inflammatory drug (NSAID) that is toxicologically safe, hereafter
referred to as IbKTP-OH and IbKTP-NH2 (Figure 1A, structures
5 and 7). This construct results from both chemical reasoning
(IBP renders KTP more lipophilic) and medical reasoning
(a combination of painkillers of different classes). Ibuprofen is
an anti-inflammatory drug (nonselectively inhibits cyclooxygenase
1 and 2) with analgesic properties: it enhances the analgesic action
of anandamide in a way sensitive to cannabinoid receptors (both
CB1 and CB2).11 Only limited information exists on the brain
distribution of ibuprofen:while someauthorsdefendthatit can get
into the brain,12 others reported a nearly absent permeability of
BBB to IBP.13 IBP has received much attention lately because it
may be a neuroprotectiveagent,12 which isbelieved to berelatedto
its nitric oxide radical scavenging activities in neuronal cells.
Accordingly, IBP was shown to be the only NSAID that, when
taken regularly, reduced by 40% the likelihood of developing
Parkinson’s disease in humans.14 These new reports gathered
further evidence of BBB crossing by IBP and increase the interest
on studying IBP containing compounds.
In the present work, we first used the intrinsic fluorescence of
tyrosine to study the lipophilicity of IbKTP derivatives in vitro in
lipid vesicles. The influences of membrane charge, lipid phase and
sterol presence were also studied. We then moved toward an in vivo
proof of concept of the biophysical rationale, through in vivo antino-
ciception behavior analysis. Our data presented here show that an
unusual chemical grafting of an analgesic peptide with ibuprofen, a
nonsteroidal anti-inflammatory drug, resulted in a highly analgesic
peptide. Moreover, our results suggest that a preference for anionic
membranes correlates with BBB crossing. Indeed, we showed that
IbKTP-NH2 interacts with BBB endothelial cells by binding to their
membranes because it partially neutralizes the membrane negative
charge. Moreover, our results indicate a dual action of IbKTP-NH2:
opioid dependent and independent.
quencher concentration [Q], and I0 is the fluorescence intensity
in the absence of quencher). See Supporting Information for
details on peptides’ biophysical characterization.
2.3. Lipid Preparation. Large unilamellar vesicles (LUVs)
with 100 nm diameter were prepared by the extrusion method
described elsewhere16 and used as model of biological mem-
branes. Several systems were analyzed: POPC (1-palmitoyl-2-
oleoyl-sn-glycero-3-phosphocholine), POPC:POPG {POPC:[1-
palmitoyl-2-oleoyl-sn-glycero-3-(phospho-rac-(1-glycerol))]}
(80%:20%, 50%:50%), POPG, POPC:cholesterol (82%:18%,
75%:25%, 67%:33%).
2.4. Membrane Partition Studies. Membrane partition stud-
ies were performed by successive additions of a 15 mM lipid
suspensiontoanIbKTP-NH2 or IbKTP-OHsolution, with10min
incubation time in between additions and before each measure-
ment. The partition coefficients, Kp, were calculated from the fit
of the experimental data with eq 2,17
IW þ KpγL½LꢁIL
I ¼
ð2Þ
1 þ KpγL½Lꢁ
where IW and IL are the limit fluorescence intensities in aqueous
solution and in lipid, respectively, γL is the molar volume of the
lipid18 and [L] is the lipid concentration. The fraction of peptide
interacting with membranes (XL) and the fraction of light
emitted by the peptide incorporated in the vesicles (fL) were
determined according to ref 18.
2.5. Localization in Lipidic Membranes. The membrane in-
depth location of the Tyr residue in IbKTP derivatives was
studied by differential quenching methodologies. Titration of
peptides in the presence of LUVs (3 mM) was carried with
successive additions of 5-doxylstearic acid (5-NS) and 16-doxyl-
stearic acid (16-NS) (ethanolic solutions) up to 1.2 mM. The
assays consisted in measuring fluorescence emission intensity
(λexc = 282 nm, λem = 306 nm). Data was analyzed through the
SternꢀVolmer equation (eq 1). If a positive deviation was
observed, data was interpreted as a quenching sphere-of-action,19
which states the existence of a sphere of volume V centered at the
fluorophore, where the quenching occurs with a determined
efficiency γ and analyzed using eq 3:
2. EXPERIMENTAL SECTION
I0
¼ ð1 þ KSV½Q ꢁÞ eV½Q ꢁN γ
ð3Þ
2.1. General Synthetic Procedures. The peptides were syn-
thesized by a standard solution peptide synthesis using benzotria-
zol-1-yloxytris(dimethylamino)phosphonium hexafluorophosphate
(BOP) and N-hydroxybenzotriazole (HOBt) as coupling re-
agents and N-methylmorpholine (NMM) as a base. See Support-
ing Information for details of ibuprofenꢀkyotorphin derivative
synthesis.
2.2. Characterization in Aqueous Solution. IbKTP-NH2
(153 μM), IbKTP-OH (263 μM) and Tyr (143 μM) solutions
were prepared in 10 mM 4-(2-hydroxyethyl)-1-piperazineethane-
sulfonic acid (HEPES) buffer, pH 7.4, 150 mM NaCl, and all assays
were carried out at room temperature. Spectral characterization of
KTP peptides and Tyr in aqueous solution was performed at an
excitation wavelength of 275 nm, except for the acrylamide quench-
ing experiments with excitation at 284 nm. Fluorescence quenching
data were analyzed using the SternꢀVolmer equation15 (eq 1):
A
I
Regarding the positive deviation to the SternꢀVolmer rela-
tionship, the Lehrer equation,15 eq 4, was applied:
I0
I
1 þ KSV½Q ꢁ
¼
ð4Þ
ð1 þ KSV½Q ꢁð1 ꢀ fBÞ þ fBÞ
An improved version of the method described in ref 20 was
used to obtain the in-depth distribution in the bilayer of the Tyr
residues of IbKTP-NH2 and IbKTP-OH. Quencher concentra-
tion shown is relative to the effective concentration in the
vesicles.
2.6. Dipole Potential Membrane Assays. Changes in the
membrane dipole potential magnitude were monitored by means
of fluorescence excitation spectral shifts of the dye (4-[2-[6-
(dioctylamino)-2-naphthalenyl]ethenyl]-1-(3-sulfopropyl) (di-8-
ANEPPS).21 Small volumes of di-8-ANEPPS dissolved in
ethanol were added to LUV and incubated overnight. Final
concentrations used were 200 μM for lipids and 5 μM for
di-8-ANEPPS.
I0
¼ 1 þ KSV½Qꢁ
ð1Þ
I
(KSV is the SternꢀVolmer constant, which depends on the
kinetic bimolecular rate constant; I is the fluorescence intensity at
1930
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
Figure 1. Synthesis of IbKTP derivatives: IbKTP-OH (structure 5) and IbKTP-NH2 (structure 7). (A) Reagents and conditions: (a) H-Tyr(tBu)-
OMe HCl (1 equiv), BOP (1 equiv), NMM (3 equiv), DMF, rt, 6 h; (b) LiOH (2.5 equiv) THF/MeOH/H2O (1:2:2), rt, overnight; (c) H-Arg-
3
OMe 2HCl (1 equiv), BOP (1 equiv), HOBt (6 equiv), NMM (3 equiv), DMF, ꢀ15 °C to rt, 21 h; (d) TFA/DMF (1:1), rt, 2 h; (e) HCl (1M),
3
lyophilization, 3 times; (f) H-Arg-NH2 2HCl (1 equiv), BOP (1 equiv), HOBt (6 equiv), NMM (3 equiv), DMF, ꢀ15 °C to rt, 21 h. (B) IbKTP
3
derivative optimization study.
1931
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
2.7. Brain Capillary Endothelial Cell (BCEC) Preparation.
Confluent BCEC were 3 times washed with phosphate buffer
saline (PBS) buffer followed by trypsinization. Cells were
counted and diluted with PBS for final concentration of 1 ꢂ
105 cells/mL and measured for zeta-potential. See Supporting
Information for details on BCEC preparation.
differences for each dose and time point compared with the KTP-
treated controls was analyzed with the nonparametric two-tailed
MannꢀWhitney test. All statistical analyses were calculated with
Prism software (GraphPad Software, version 5).
3. RESULTS
2.8. Human Umbilical Vein Endothelial Cell (HUVEC) Pre-
paration. HUVEC were kindly provided by Dra. Susana C.
Santos (Portugal) and cultured in complete endothelial medium.
Upon reaching confluence, cells were passed onto other gelatin
coated flasks and used at the sixth passage. For zeta-potential
measurements, HUVEC were 3 times washed with PBS buffer
followed by trypsinization. Finally, cells were counted and
diluted with PBS for final concentration of 1 ꢂ 105 cells/mL.
2.9. Zeta-Potential Measurements. Measurements were
conducted on a dynamic light scattering and zeta-potential
equipment, equipped with a HeꢀNe laser (λ = 632.8 nm).
The zeta-potential of the samples was determined at 25 °C using
disposable zeta cells with platinum gold-coated electrodes as
described in ref 22. 1-(4-(Trimethylamino)phenyl)-6-phenyl-
1,3,5-hexatriene (TMA-DPH) was added for a final concentra-
tion of 54 μM and incubated for 30 min previous to measure-
ments. IbKTP derivatives were added to the cell sample both at
200 μM and equilibrated for 15 min prior to the measurement.
Each group value is an average of at least two independent
measurements. The electrophoretic mobility obtained was used
for the zeta-potential calculation through the Smoluchowski
equation,23
3.1. IbKTP-NH2 and IbKTP-OH Synthesis. Nonamidated and
amidated ibuprofen-kyotorphin derivatives, IbKTP-OH (5) and
IbKTP-NH2 (7), have been prepared following standard solu-
tion peptide synthesis as illustrated in Figure 1A. Coupling
between optically pure (S)-ibuprofen and protected L-tyrosine
was mediated by BOP, 1 equiv in the presence of NMM
(3 equiv)27 and afforded compound 1 with good yield. Subse-
quent saponification of methyl ester 1 with LiOH28 provided the
common intermediate 2 also with good yield. Initially, com-
pound 2 and L-arginine methyl ester were coupled using the same
reaction conditions. Disappointingly, dipeptide 3 was obtained as
a diastereoisomeric mixture (Figure 1B, entry 1). A similar result
was obtained for compound 6 when compound 2 was coupled
1
with L-arginine amide (Figure 1B, entry 2). H NMR (nuclear
magnetic resonance) spectra of compounds 3 and 6 recorded in
DMSO-d6 showed split signals. Compared to their precursors,
the 1H NMR spectrum of dipeptide 3 showed almost all
protons split as two sets of signals (Figure S1 in the Supporting
Information). In order to prove racemization or a possible
conformational equilibrium, an NMR study was undertaken.
1
The H NMR of compound 3 did not exhibit line broadening
from 25 to 85 °C in DMSO-d6, which would be expected if a slow
conformational equilibrium occurred (Figure S2 in the Support-
ing Information). An optimization study was conducted for the
synthesis of compound 3 in order to reduce the racemization
degree. The different variables (coupling reagent equivalents,
temperature and reaction time) were systematically changed
(Figure 1B). No reaction took place when dicyclohexylcarbodii-
mide (DCC) was employed as a coupling reagent in the presence
of the commonly used racemization suppressant agent HOBt29
(Figure 1B, entry 3). However, racemization could be signifi-
cantly diminished when BOP was used as coupling reagent and a
large amount of HOBt30 was added to the reaction mixture
(Figure 1B, entries 4ꢀ7) at low temperature. An increase of the
amount of HOBt led to a decrease of racemization until an
acceptable diastereoisomeric ratio of 88:12 (Figure 1B, entry 7).
Dipeptide 6 was also obtained in a similar racemization degree
using the above coupling reaction conditions. In all cases,
diastereoisomeric ratio was determined by NMR integration of
the aromatic protons signals at 6.7 and 6.85 ppm respectively
(Figures S3 and S4 in the Supporting Information). The IbKTP-
OH derivative 5 was obtained after methyl ester saponification of
dipeptide 3 followed by treatment of the resulting free acid 4 with
trifluoroacetic acid (TFA) to remove tert-butyl protecting group.
Dipeptide 6 was converted into IbKTP-NH2 derivative 7 by tert-
butyl removal with TFA (Figure 1A). Final compounds 5 and 7
were dissolved in 1 M HCl and lyophilized. This last step was
repeated three times in order to obtain derivatives 5 and 7 as
hydrochloride salts instead of a trifluoroacetic salt for subsequent
administration to model animals.
4πημ
ζ ¼
ð5Þ
ε
where μ represents the electrophoretic mobility, η the viscosity
of the solvent and ε its dielectric constant.
2.10. Animals. The experiments were performed according to
the ethical Guidelines of the International Association for the
Study of Pain24 and the European Community Council Directive
(86/609/EEC). Adult male Wistar rats (Charles River, Spain)
weighing 275ꢀ300 g at the beginning of experiments were used.
All peptides were dissolved in saline containing 10% of dimethyl
sulfoxide (DMSO). Intrathecal catheters were implanted as
described before.9 On the experiment day, naloxone was i.t.
injected (10 μL; 5 mg mLꢀ1) followed, 10 min later, by the ip
3
administration of either of the peptides.
2.11. Acute Pain Experiments. The effect of the peptides at
25 μmol, 50 μmol and 96 μmol kgꢀ1 bw was evaluated by the
3
tail-flick and hot-plate tests as described elsewhere.9 See Support-
ing Informationfor detailsonacute pain experimentmethodology.
2.12. Inflammatory Pain Experiments. The formalin test
was performed as described previously.9,25 Data is presented as
the number of jerks/flinches (named here as paw-jerks) and focused-
pain related activity of the injected hind paw in 12 successive periods
of 5 min each.
2.13. Immunostaining for Fos. Two hours after formalin
injection, the time point of maximal expression of the proto-
oncogene c-fos,26 animals were sacrificed for the analysis of
c-fos activation as previously described.9 Mean numbers of Fos-
immunoreactive neurons per section were calculated. See Sup-
porting Information for details on Fos immunostaining.
2.14. Statistical Analyses. Data are represented as the groups'
mean + SEM (standard error of the mean). The significance of
differences in each group was analyzed with Friedman’s test
followed by Dunn’s multiple comparison test. The significance of
3.2. Characterization in Aqueous Solution. KTP has one
Tyr residue, which allows structural characterization by fluores-
cence spectroscopy. In aqueous solution and under physiological
pH and ionic strength conditions, absorption and emission
spectra profiles of both derivatives, IbKTP-NH2 (153 μM) and
1932
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
IbKTP-OH (263 μM), are similar to those of free tyrosine
(Figure 2A). With an absorption maximum at 273 nm and an
emission maximum at 302 nm, a small deviation on the maximal
absorption wavelength in comparison to Tyr is observed which is
favored the selective interaction toward anionic vesicles. The
strong influence of the negative charge of lipids is also noticed on
the fraction of peptide partitioned to the membrane at a certain
fixed lipid concentration: XL18 (Figure 2G). While for KTP-NH2
XL values at 3 mM lipid are between 0.79 and 0.87 irrespective of
POPG proportion in the vesicles (Table S1 in the Supporting
Information), in the case of IbKTP-NH2, they increase markedly
with anionic charge in the membranes.
probably related to the additional effect of IBP spectrum (λabs,max
264 nm and λem,max = 298 nm).
=
No red-edge excitation-shift effects were detected. The fluor-
escence quantum yield (ϕ) in aqueous solution at 23 °C is 0.054
and 0.069 for IbKTP-NH2 and IbKTP-OH, respectively, which is
lower than that of free Tyr (ϕ = 0.14031). The fluorescence of an
aromatic amino acid side chain can be quenched by the peptide
backbone as a consequence of a charge transfer between the
excited chromophore (phenol ring), actingasa donor, and electro-
philic units in the amino acid backbone, acting as an acceptor.31
Ross et al.32 also observed that if the phenol side chain is shielded
from solvent and the local environment contains no proton
acceptors, then many intra- and intermolecular interactions
result in a reduction of the quantum yield. Moreover, the fact
that IbKTP-NH2 reveals the lowest ϕ is in agreement with studies
that showed amide group as a quencher of tyrosine fluorescence.33
These results are in conformity with results observed for other
KTP derivatives.7,8 The study of intermolecular quenching can
be used to evaluate the accessibility of fluorophore to quenchers
whose location is known.19 Linear SternꢀVolmer plots for
the fluorescence quenching of the peptide by the hydrophilic
molecule acrylamide, for instance, are evidence that Tyr residues
are equally exposed to the surrounding aqueous medium (Figure
S5A in the Supporting Information), which rules out the
possibility of compact aggregates. Fitting the experimental data
to SternꢀVolmer equation (eq 1), KSV was calculated to be
9.9 and 14.8 Mꢀ1 for IbKTP-NH2 and IbKTP-OH, respectively.
The slightly lower value observed for IbKTP-NH2 may be related
to the addition of the amide group to the Arg residue, resulting in
a conformational alteration that restrains the accessibility of
acrylamide to the Tyr residue. To further evaluate possible
changes in the supramolecular structure of KTP derivatives,
namely, aggregation, the fluorescence quantum yield dependence
on peptide concentration was studied. No significant deviations
from linearity were observed in a fluorescence intensity vs con-
centration plot, i.e., there is no dependence of quantum yield on
concentration (Figure S5B in the Supporting Information).
It may then be concluded that the dominant forms of the deri-
vatized dipeptides are monomers or small loose oligomers.
3.3. Extent of Partition into Fluid Lipid Bilayers. We started
by evaluating the partition of both peptides toward liquid-crystal
(fluid) bilayers, constituted by zwitterionic POPC and anionic
POPG in different proportions, to model human cells’ mem-
branes and appraise the effect of anionic lipids. Titration of aqueous
suspensions of IbKTP-NH2 with LUVs led to an increase in the
fluorescence quantum yield, indicating that the peptide is inter-
acting with the membranes (Figure 2B). This change in fluores-
cence intensity is related to the partition constant, Kp, by eq 2.
When titrated with POPC, Kp is (0.19 ( 0.06) ꢂ 103. There is a
clear effect of charge on the partition of peptide toward the
membrane: Kp increases exponentially with the molar fraction
content of the negatively charged lipid POPG in the POPC
bilayer (Figure 2D,G), reaching a maximum when the membrane
is 100% POPG [(8.1 ( 0.8) ꢂ 103]. In the case in which the IBP
moiety is absent (KTP-NH2), it is difficult to define a specific
pattern of interaction dependent on the anionic lipidic composi-
tion, with Kp varying in a small range of values (Table S1 in the
Supporting Information). Interestingly, the grafting of ibuprofen
The original endogenous nonamidated kyotorphin molecule,
KTP-OH, partition at physiologic conditions has been described
as reduced, with preference for cholesterol (chol) containing
membranes.7 When derivatized with IBP, an increase in the
interaction of the dipeptide with the membrane was observed,
accompanied by a change of preference toward fluid phase lipids
(Figure 2C). The highest Kp reaches (1.8 ( 0.5) ꢂ 103 for an
equimolar mixture of POPC and POPG (Figure 2C,G). Thus,
both hydrophobic and electrostatic interactions seem to con-
tribute for IbKTP-OH partition toward the membrane.
3.4. Extend of Partition into Cholesterol-Containing
Bilayers. POPC vesicles containing 18% and 33% molar choles-
terol were used to study the partition of peptides toward liquid-
ordered membranes.34 For IbKTP-NH2, Kp decreased with
an increase in the cholesterol content in the membranes
(Figure 2B,G). At variance, improved interaction was observed
for IbKTP-OH, when compared to POPC vesicles (Figure 2C,G).
These IbKTP-OH results are in agreement with similar data for
KTP-OH.7
3.5. Localization of Peptides in Lipid Bilayers. In-depth
location of Tyr residue in kyotorphin peptides was carried out by
means of differential fluorescence quenching using two lipophilic
doxyl stearic acids (5-NS and 16-NS). The closer the Tyr residues
are to the quencher group, the more efficient is the quenching.
Considering their distributions, 5-NS probes the bilayer interface
whereas 16-NS probes its core. By combining the results from
both quenchers, one can estimate the membrane depth distribu-
tion of the fluorophore.
5-NS was the most efficient quencher of the fluorescence for
both peptides, indicating a preferred location near the lipid
bilayer interface (Figure 2E,F). When titrated with 5-NS, IbKTP-
NH2 SternꢀVolmer plots show a positive deviation from linearity,
which may be ascribed to simultaneous static and dynamic
quenching effects.19 A negative deviation from linearity in the
SternꢀVolmer plot was observed for IbKTP-OH, indicating the
presence of heterogeneous subpopulations of the fluorophore
relative to the accessibility to the quenchers (Figure 2E). Equa-
tion 3 and eq 4 were applied to data to obtain SternꢀVolmer
constants, KSV, and the fraction of fluorophore accessible to the
quencher, fB. The fB values can be compared to the fraction of
fluorescence intensity originating from the peptide in the lipid,
fL (Figure 2G and Table S1 in the Supporting Information18).
Brownian dynamics simulation of the quenchers in membrane20
allowed the in-depth distribution of Tyr residues to be obtained
taking into account the possibility of static quenching by using a
sphere-of-action methodology.19 All peptides showed a super-
ficial localization (Figure 2F).
3.6. Effect of IbKTP Derivatives on Brain Endothelial Cell
Membrane. In order to underpin the effect of each IbKTP
derivative on the membrane of cells constituting the BBB, we
measured membrane zeta-potential, ζ, of primary bovine cultures
of BCEC in the presence of each IbKTP derivative (Figure 3).
Although endothelial cell membranes are generally known as
displaying a negative charge,35 a direct quantification of their
1933
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
Figure 2. Biophysical characterization of IbKTP derivatives in aqueous solution and in interaction with membranes. (A) Normalized absorption and
emission (λexc = 273 nm) spectra of IbKTP-NH2 (Dotted line) and IbKTP-OH (Dashed line) in aqueous solution. (B-D) Partition of IbKTP derivatives
with vesicles. Titration of IbKTP-NH2 (B) and IbKTP-OH (C) with concentrated LUV suspension. Lines represent fittings of eq 217 to the data. POPG
(2); POPC:POPG (50%) (4); POPC:POPG (20%) (O); POPC (b); POPC:Chol (18%) (0); POPC:Chol (33%) (9). (D) Kp (logarithmic scale) of
IbKTP-NH2 dependence on POPC:POPG molar ratio. (E, F) IbKTP derivatives localization in the membrane. (E) SternꢀVolmer plot of the
fluorescence quenching of IbKTP-NH2 (b; O) and IbKTP-OH (2; 4) by 5-NS (b; 2) and 16-NS (O; 4). Solid lines represent fittings of eqs 3 and 4 to
the experimental data. (F) In-depth distribution of Tyr residues of IbKTP-NH2 and IbKTP-OH in lipid bilayers is obtained from the data in (E) upon
application of a differential quenching method.20 The fluorophore distribution in the membrane is given as probability density function relative to the
distance to the bilayer center. (G) Parameters obtained for the partition and quenching of the fluorescence of IbKTP-NH2 and IbKTP-OH by the
lipophilic probes 5-NS and 16-NS in membranes using different lipidic mixtures. Assays performed at room temperature in a 10 mM HEPES buffer, pH
7.4, containing 150 mM NaCl. Quenching experiments performed at a total lipidic concentration of 3 mM. Kp was determined by fitting eq 2 to the data.
1934
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
Figure 3. Zeta-potential of primary endothelial cells in the presence of IbKTP derivatives. BCEC (A) and HUVEC (B) at 1 ꢂ 105 cells/mL
(PBS buffer) were incubated with TMA-DPH (54 μM), IbKTP-NH2 and IbKTP-OH (both at 200 μM) at 25 °C, and zeta potential was measured. Data
shown as means ( SEM; each group value is an average of at least two independent measurements. **P < 0.01; ***P < 0.001 versus BCEC or HUVEC
control; one-way ANOVA, Bonferroni’s multiple comparison test.
membrane charge has never been performed to the best of
our knowledge. BCEC membrane surface charge through zeta-
potential was revealed to be highly negative, ζ= ꢀ15.28 ( 0.582 mV.
TMA-DPH, a cationic probe of membrane dynamics in living
cells with a cationic charge (+1), was used as a positive control
because it intercalates at the level of hydrophilic head groups of
membrane phospholipids and neutralizes the anionic charges of
lipids of the outer layer of the cytoplasmatic membrane. TMA-
DPH, as expected, caused an increase in the zeta value toward
neutralization (ζ = ꢀ12.13 ( 0.490 mV) (Figure 3A). The effect
of IbKTP-NH2 and IbKTP-OH on BCEC membrane surface
charge was also measured. From these, only IbKTP-NH2 sig-
nificantly led to a shift in the zeta-potential of the BCEC cell
membrane, toward neutralization (ζ = ꢀ12.85 ( 0.492 mV,
Figure 3B). This selectivity to BBB cell membranes, observed for
IbKTP-NH2 but not for IbKTP-OH, may be the basis of different
BBB-crossing potential between the two peptides. To further
evaluate if this interaction profile applies to all endothelial cells,
we took HUVEC and performed the same experiments. The
zeta-potential of HUVEC (ζ = ꢀ12.89 ( 0.564 mV) is sig-
nificantly more positive than that of BCEC (Figure 3B). Unlike
the zeta-potential of BCEC, the zeta-potential of HUVEC is
affected by both IbKTP-NH2 (ζ = ꢀ10.2 ( 0.571 mV) and
IbKTP-OH (ζ = ꢀ11.3 ( 0.482 mV) (Figure 3B). These results
correlate well with the results obtained with membrane vesicles,
which revealed IbKTP-NH2 to be very lipophilic, mainly when
anionic membranes are involved. Zeta-potential results show that
grafting KTP with IBP (IbKTP-OH) grants increased interaction
with anionic surface endothelial cells, and additional amidation
(IbKTP-NH2) allows a certain degree of selectivity to BCECs.
3.7. IbKTP-NH2 and IbKTP-OH Analgesia in Acute Pain
Models. The peptides’ analgesic effect was evaluated following ip
administration to male Wistar rats. Acute pain models using thermal
stimuli were applied: tail flick and hot plate tests.36 The doses of
both peptides tested were based on previous results for KTP-NH2.9
Briefly, we started with the intermediate dose of KTP-NH2 that was
efficient in all pain models tested (32.3 mg kgꢀ1 = 96 μmol kgꢀ1).
The covalent addition of ibuprofen grants KTP-OH analgesic
properties, following systemic administration. IbKTP-OH pro-
voked a long-lasting analgesia at a dose of 96 μmol kgꢀ1 bw as
3
revealed by the hot plate test (Figure 4D). However, the tail flick
method did not reveal antinociception for IbKTP-OH
(Figure 4C). Mechanistically, tail flick is considered a spinal
reflex, while reactions to hot plate are related to supraspinal
modulation36 and are thus more reliable. As in the case of IbKTP-
NH2, IbKTP-OH activity was revealed to be higher than that of
the control (IBP + KTP-OH), confirming a specific action for the
“tandem chimeric” molecule.
Hydrophobization of both KTP-NH2 and KTP-OH resulted
in analgesic action following systemic administration, which is
compatible with BBB crossing. Comparatively, IbKTP-NH2 is
the most active compound, about 3.5 times more effective than
IbKTP-OH. At this point, and considering the 3R policy on
animal experimentation, we selected the most promising peptide,
IbKTP-NH2, to continue the studies with model animals.
3.8. Effect of IbKTP-NH2 in a Model of Inflammatory Pain.
In order to increase IbKTP-NH2 pharmacological relevance, we
proceeded to investigate its effect in a model of sustained pain,
the formalin test, which encompasses two pain phases: acute
(t < 10 min) and inflammatory (t > 10 min),25 thus being a model
with translational value.36 Moreover, it consists of a chemical
stimulus, thus broadening the application range of the drug under
study. Based on the time course of the effects of the drug at the
acute pain tests, IbKTP-NH2 was ip administered 10 min before
formalin at the dose of 5 μmol kgꢀ1 bw. Unlike KTP-OH (used
3
as a control due to structural similarities and inability to cross the
BBB9), IbKTP-NH2 reverted the typical nociception pattern at
the chronic pain phase regarding focused pain-related activity
and paw-jerks (Figure 5A,B). On the other side, animals injected
with KTP-OH showed a pattern of behavioral responses similar
to that of animals injected only with formalin.37 Fos is a neuronal
marker of nociceptive activation of spinal neurons.26 The noci-
ceptive activation of spinal dorsal horn neurons, laminae IꢀVI,
induced by formalin administration was assessed by immunode-
tection of the Fos protein.
3
3
For IbKTP-NH2, 50 μmol kgꢀ1 was used so that the test cutoff
3
Animals injected with IbKTP-NH2 showed a decrease in the
number of Fos-immunoreactive neurons at the deep dorsal horn
layers (Figure 5C). This constitutes evidence on the central
analgesic consequences of this new peptide.
3.9. IbKTP-NH2 Mechanism of Action. Whether kyotorphin
is an opioid molecule or not has been a controversial matter of
debate since its discovery. A specific KTP receptor has not been
identified, and the activity of KTP-OH and KTP-NH2 is completely
limits were not reached. Not only did IbKTP-NH2 show a remark-
able analgesic efficacy—from 25 μmol kgꢀ1 bw (P = 0.0032, tail
3
flick; P = 0.0005, hot plate, Friedman test)—but its potency was also
strikingly higher than that of the combined injection of KTP-NH2
and ibuprofen (Figure 4A,B). In fact, linking IBP to KTP-NH2
increased by 2-fold the analgesic efficacy in relation to the
control IBP + KTP-NH2 mixture. The analgesic effects were dose-
dependent, starting at 15 min and lasting at least for 45 min.
1935
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
Figure 4. Analgesic action of IbKTP derivatives. Inhibition of acute nociceptive responses in the tail flick (A, C) and hot plate (B, D) tests following
systemic administration of IbKTP derivatives. Results expressed as the variation of the behavioral responses at each evaluation time point in relation to
baseline values (time 0) obtained immediately before injection. Doses represented for kg bw. In all experiments, n g 6. *P < 0.05; **P < 0.01; ***P < 0.001
versus basal response, Friedman test, and #P < 0.05, ##P < 0.01, ###P < 0.001 versus IBP + KTP treated controls, MannꢀWhitney test. Mean + SEM for all
groups.
abolished by naloxone administration, which is an unspecific
opioid receptor antagonist.9 However, direct binding between
each peptide and opioid receptors is rather weak.9,38
may change the membrane dipole potential, which can be moni-
tored by spectral shifts of the fluorescence dye di-8-ANEPPS.40
Fluorescence intensity difference spectra of di-8-ANEPPS when
IbKTP-NH2 is used are in agreement with partition data for this
molecule (Figure 2B,D): the presented spectral red shifts in-
dicate decrease in the dipole potential, and this perturbation
increases with the anionic charge in the membrane (Figure 7A).
As to morphine, no significant spectral shifts were observed
except for cholesterol-containing bilayers (Figure 7B). The
highest shift was recovered for POPC:chol (18%), in agree-
ment with the SargentꢀSchwyzer hypothesis41 since opioid
receptors are located in cholesterol-rich domains of mem-
branes.39 Moreover, morphine caused a blue shift effect as a
result of an increase in the membrane dipole potential. This
potential is originated from the alignment of dipolar residues of
lipids (polar head groups and glycerol-ester regions) and
oriented water molecules hydrating the outer surface of the
membrane.21
The effect of central naloxone administration (i.t., 10 μL,
5 mg mLꢀ1) over the analgesic efficacy of IbKTP-NH2
3
(50 μmol kgꢀ1) was evaluated to assess the involvement of
3
opioid pathways in its analgesic mode of action. Naloxone
injection significantly decreased IbKTP-NH2 analgesia in the
hot plate test. However, unlike KTP-NH2,9 the IbKTP-NH2
antinociceptive effect was not completely inhibited by naloxone
(Figure 6). This indicates that for this derivative, in addition to an
opioid-dependent mode of action, there seems to be an opioid-
independent mechanism. The combination of both results in a
peptide with enhanced analgesic activity. It is worth stressing that
we observed increased partition of IbKTP-NH2 to lipid mem-
branes as a result of IBP grafting to KTP-NH2 with a preference
for liquid-crystal anionic membranes. Morphine, on the other
hand, is known to prefer rigid raftlike domains in the membrane,
where opioid receptors tend to concentrate.39
The distinctive preference of IbKTP-NH2 and morphine for
POPC:POPG and POPC:Chol, respectively, points to different
molecular targets and may explain why naloxone does not
completely abolish the analgesic effect of IbKTP-NH2.
To further shed light on the differences on the molecular basis of
action between opioids and IbKTP-NH2, we compared the mem-
branotropic effect of morphine and IbKTP-NH2. Since morphine
does not display environment-sensitive intrinsic fluorescent char-
acteristics, we used di-8-ANEPPS (4-[2-[6-(dioctylamino)-2-naphtha-
lenyl]ethenyl]-1-(3-sulfopropyl)pyridinium) as a polarity-sensitive
probe for assessing IbKTP-NH2 and morphine interaction with
the membrane. Binding and insertion of molecules in the membrane
4. DISCUSSION
Derivatizing peptides in order to improve their properties is a
common approach, whether to avoid enzyme degradation, enable
1936
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
Figure 6. Opioid related mechanism of action of IbKTP-NH2. A partial
reversal of IbKTP-NH2 induced analgesia occurred after i.t. administra-
tion of naloxone (NLX, 5 mg kgꢀ1) in the hot plate test. Results
3
expressed as the variation of the behavioral responses at each evaluation
time point in relation to baseline values (time 0) obtained immediately
before injection. In all experiments n g 6. *P < 0.05; **P < 0.01; ***P <
0.001 versus basal response, Friedman test, and ##P < 0.01, ###P < 0.001
versus NLX + IbKTP-NH2 treated animals, MannꢀWhitney test. Mean +
SEM for all groups.
fluorescence intensity with peptide concentration confirmed that
IbKTP-NH2 and IbKTP-OH do not form compact aggregates in
solution, a fundamental feature when aiming a pharmacological
application.
Overall, derivatization with ibuprofen largely increased parti-
tion and insertion in lipid membranes, with a clear affinity of the
peptides for anionic fluid phase membranes. While the original
peptide, KTP, preferred cholesterol rich membranes, IbKTP-OH
displayed the highest Kp for POPC:POPG (50%:50%). In
contrast, IbKTP-NH2ꢀmembraneinteraction increased exponen-
tiallywiththemolarfraction content ofthe negativelycharged lipid
POPG. Thus, electrostatic attraction plays a fundamental role in
the binding of IbKTP-NH2 to model membranes, in a similar way
as observed for encephalin peptides.43
The zeta-potential measurements of BCEC and HUVEC in
the presence of IbKTP-NH2 and IbKTP-OH revealed a pre-
ference of IbKTP-NH2 for BBB cells. Compared to HUVEC,
BCEC display a more negative membrane charge, which may
explain the preference of more cationic peptides to interact with
and cross the BBB.44 Therefore, we observed a preference of
BCEC for IbKTP-NH2 [positively charged (+1) at variance with
IbKTP-OH, with a net neutral charge at plasma pH], which
showed to be the more analgesic compound. Thus, there is an
agreement between the model membrane results and the cells
results. Although human endothelial cells are mainly constituted by
phosphatidylcholine with smaller quantities of anionic phosphati-
dylserine (PS), Yeung and collaborators45 observed that the
negative charge associated with the presence of PS directed proteins
with moderately positive charge to the endocytic pathway, one of
the major mechanisms of absorption of peptides at the BBB level,
and this is in accordance with the results reported here.
Figure 5. Analgesic profile of IbKTP-NH2 in formalin inflammatory
pain model. Time spent in focused pain-related activity (A) and number
of paw-jerks (B) in 5 min time periods. The analgesic effects of an ip
injection of IbKTP-NH2 10 min before formalin injection were detected
in the inflammatory phase (t > 10 min). Doses represented for kg bw. In
all experiments n g 6. *P < 0.05; **P < 0.01 versus KTP-OH treated
controls, MannꢀWhitney test. Mean + SEM for all groups. (C) In
formalin-injected rats, nociceptive activation of spinal neurons induced
by formalin was inhibited by IbKTP-NH2, as indicated by the lower
number of Fos-immunoreactive neurons at laminae IIIꢀV at the spinal
dorsal horn. *P < 0.05, unpaired t test. Mean ( SEM for all groups.
oral availability or increase biodistribution, for instance. In the
specific case of brain-targeted molecules, being able to cross the
bloodꢀbrainbarrieristhemostcommonquest. Inthisstudy, wetried
to enhance kyotorphin analgesic properties by linking it to ibuprofen:
a well-known lipophilic anti-inflammatory and analgesic drug whose
ability to cross the BBB is still unclear. The influence of ibuprofen
addition on the dipeptide activity was assessed at different levels: in
vitro with membrane model systems and primary brain and umbilical
cells, and in vivo with model animals. The absence of red-edge
excitation shift, linear SternꢀVolmer plots and linear dependence of
Concomitantly with biophysical results, which revealed IbKTP-
NH2 as the peptide with highest affinity to model and BCEC
membranes, in vivo antinociception evaluation also revealed that
this is the most effective analgesic compound. For this reason, only
IbKTP-NH2 was further studied. Insertion into the lipid matrix
1937
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
molecule, KTP, shows no analgesic effect at all when systemically
administered, while IbKTP-OH is analgesic. The key difference
between them is lipophilicity. This evidence strengthens that
hypothesis 1 above may be operative. Mechanistically, ibuprofen
seems to function as a carrier, driving KTP through the blood
ꢀbrain barrier to the target site, the brain.
Reduction in the nociceptive activation of spinal neurons as a
result of IbKTP-NH2 administration provided evidence of
central consequences of the analgesic action of the new drug.
Curiously, the effects were more pronounced in the inhibition of
the deep dorsal horn, which harbors mainly wide dynamic range
neurons and is the site of termination of Aβ fibers.47 Descending
modulation from the main pain control center in the brain (the
periaqueductal gray PAG) also targets mainly deep dorsal horn
neurons, which might be an additional indication of central
actions of IbKTP-NH2. On the other hand, even with naloxone,
the IbKTP-NH2 analgesic effect was still significant, although
reduced. Possibly, IbKTP-NH2 analgesic action may be through
two distinct mechanisms: opioid-dependent (prone to be in-
hibited by naloxone) and opioid-independent. Moreover, a
combination of a central and a peripheral effect is likely to occur.
This multitarget mechanism of action justifies IbKTP-NH2
higher analgesic potential.
Unlike morphine, these peptides do not show a preference for
cholesterol-rich domains. Furthermore, IbKTP-NH2 and mor-
phine affect differently the membrane dipole potential. Opioid
pathways are involved to some extent, but, in light of these data,
IbKTP-NH2 and morphine do not share the same mechanism of
action. Taking together, these new data suggest that, by avoiding
the typical opioid-like mechanism of action, the narcotic side
effects associated with them might as well be reduced.
’ ASSOCIATED CONTENT
S
Supporting Information. Localization of peptides in lipid
b
bilayers, additional figures and tables: 1H NMR spectra of
compounds 2, 3 and 5; biophysical characterization of IbKTP
derivatives in aqueous solution and parameters obtained for the
partition and quenching of the fluorescence of KTP-NH2. Addi-
tional experimental details: detailed synthesis of IbKTP deriva-
tives, aqueous characterization of compounds, BCEC preparation,
animals, acute and inflammatory pain experiments and immunos-
taining for Fos. This material is available free of charge via the
Internet at http://pubs.acs.org.
Figure 7. Polarity-sensitive membrane-dependent differential excita-
tion fluorescence spectra of di-8-ANEPPS-labeled vesicles in the pre-
sence of IbKTP-NH2 (A) and morphine (B). Lipid concentration and
di-8-ANEPPS were kept constant at 200 μM and 5 μM, respectively. All
the assays were performed in 10 mM HEPES buffer pH 7.4, containing
NaCl 150 mM. Fluorescence difference spectra were obtained by
subtracting the excitation spectrum of di-8-ANEPPS-labeled LUVs in
the absence and presence of IbKTP-NH2 or morphine. Before subtrac-
tion, the spectra were normalized to the integrated areas so that the
difference spectra reflect only spectral shifts.42.
’ AUTHOR INFORMATION
appears not to be fundamental for either IbKTP derivative action,
contributing to the idea of a membrane as a concentrator of
peptide at its surface for receptor binding.41
Therefore, lipophilicity correlates to analgesic activity, hypothe-
tically due to (1) a more efficient BBB-translocation process, and/
or (2) a membrane-induced optimization of conformation for
receptor docking, and/or (3) a simple concentration effect in the
vicinity of target receptors.
Two new analgesic compounds were characterized in this
study: IbKTP-NH2 and IbKTP-OH. Synergistic analgesic effects
as a result of a concomitant administration of NSAIDs, namely,
IBP, and opioids have been reported before.46 Here, we pre-
sented an improved solution: grafting IBP to KTP-NH2 or KTP-
OH resulted in a significantly higher analgesia, when compared
with the controls (IBP + KTP-NH2 or IBP + KTP-OH). It is worth
stressing that, in the specific case of IbKTP-OH, the underivatized
Corresponding Author
*Faculdade de Medicina de Lisboa, Instituto de Medicina Mole-
cular, Av. Professor Egas Moniz, 1649-028 Lisboa, Portugal. E-mail:
macastanho@fm.ul.pt. Tel: (+351) 217 985 136. Fax: (+351)
217 999 477.
’ ACKNOWLEDGMENT
Fundac-~ao para a Ci^encia e Tecnologia (Portugal) is acknowl-
edged for funding (SFRH/BD/42158/2007 fellowship to M.M.
B.R., SFRH/BD/41750/2007 fellowship to M.D., SFRH/BPD/
37998/2007 fellowship to I.S. and project PTDC/SAU-FCF/
69493/2006). Marie Curie Industry-Academia Partnerships
and Pathways (European Commission) is also acknowledged
for funding (FP7-PEOPLE-2007-3-1-IAPP, Project 230654).
1938
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
Dra. SusanaC. Santos (Instituto de Medicina Molecular, Portugal)
is acknowledged for kindly providing HUVEC.
responses after systemic administration of amidated kyotorphin. Br.
J. Pharmacol.
(10) Woodcock, J.; Witter, J.; Dionne, R. A. Stimulating the devel-
opment of mechanism-based, individualized pain therapies. Nat. Rev.
Drug Discovery 2007, 6, 703–10.
’ ABBREVIATIONS USED
(11) Di Marzo, V.; Bisogno, T.; De Petrocellis, L. Endocannabinoids
and related compounds: walking back and forth between plant natural
products and animal physiology. Chem. Biol. 2007, 14, 741–56.
(12) Prins, L. H.; du Preez, J. L.; van Dyk, S.; Malan, S. F. Polycyclic
cage structures as carrier molecules for neuroprotective non-steroidal
anti-inflammatory drugs. Eur. J. Med. Chem. 2009, 44, 2577–82.
(13) Mannila, A.; Rautio, J.; Lehtonen, M.; Jarvinen, T.; Savolainen,
J. Inefficient central nervous system delivery limits the use of ibuprofen
in neurodegenerative diseases. Eur. J. Pharm. Sci. 2005, 24, 101–5.
(14) Gao, X.; Chen, H.; Schwarzschild, M. A.; Ascherio, A. Use of
ibuprofen and risk of Parkinson disease. Neurology 2011, 76, 863–9.
(15) Santos, N. C.; Castanho, M. Fluorescence spectroscopy meth-
odologies on the study of proteins and peptides. On the 150th
anniversary of protein fluorescence. Trends Appl. Spectrosc. 2002, 4,
113–125.
KTP, kyotorphin (Tyr-Arg); KTP-NH2, kyotorphin amide; IbKTP-
NH2, kyotorphin amide linked to ibuprofen; IbKTP-OH, kyotor-
phin linked to ibuprofen; IBP, ibuprofen; BBB, bloodꢀbrain barrier;
BOP, benzotriazol-1-yloxytris(dimethylamino)phosphonium
hexafluorophosphate; NMM, N-methylmorpholine; DMSO, di-
methyl sulfoxide; NMR, nuclear magnetic resonance; DCC, di-
cyclohexylcarbodiimide; HOBt, N-hydroxybenzotriazole; TFA,
trifluoroacetic acid; ϕ, fluorescence quantum yield; KSV, Sternꢀ
Volmer constant; POPC, 1-palmitoyl-2-oleoyl-sn-glycero-3-phos-
phocholine; POPG, 1-palmitoyl-2-oleoyl-sn-glycero-3-(phospho-
rac-(1-glycerol)); Chol, cholesterol; Kp, partition constant; XL,
fraction of peptide partitioned to the membrane; 5-NS,
5-doxylstearic acid; 16-NS, 16-doxylstearic acid; fB, fraction of
fluorophore accessible to the quencher; fL, fraction of fluores-
cence intensity originating from the peptide in the lipid; ip,
intraperitoneal; bw, body weight; di-8-ANEPPS, 4-[2-[6-(dioctyl-
amino)-2-naphthalenyl]ethenyl]-1-(3-sulfopropyl)pyridinium;
PAG, periaqueductal gray; NSAID, nonsteroidal anti-inflamma-
tory drug;TMA-DPH, trimethylammonium-diphenylhexatriene;
ζ, zeta-potential; HPLC, high performance liquid chromatogra-
phy;HRMS, high resolution mass spectrometry;DMF, N,N-di-
methylformamide;dr, diastereoisomeric ratio;ESI, electrospray
ionization; rt, room temperature; TLC:, thin layer chromato-
graphy; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfo-
nic acid; i.t., intrathecal; sc, subcutaneous; PBS, phosphate
buffer saline; IW, limit fluorescence intensity in aqueous
medium; IL, limit fluorescence intensity in lipid; γL, molar
volume of the lipid
(16) Mayer, L. D.; Hope, M. J.; Cullis, P. R. Vesicles of variable sizes
produced by a rapid extrusion procedure. Biochim. Biophys. Acta 1986,
858, 161–8.
(17) Santos, N. C.; Prieto, M.; Castanho, M. A. Quantifying molecular
partition into model systems of biomembranes: an emphasis on optical
spectroscopic methods. Biochim. Biophys. Acta 2003, 1612, 123–35.
(18) Henriques, S. T.; Castanho, M. A. Environmental factors that
enhance the action of the cell penetrating peptide pep-1 A spectroscopic
study using lipidic vesicles. Biochim. Biophys. Acta 2005, 1669, 75–86.
(19) Castanho, M. A.; Prieto, M. J. Fluorescence quenching data
interpretation in biological systems. The use of microscopic models for
data analysis and interpretation of complex systems. Biochim. Biophys.
Acta 1998, 1373, 1–16.
(20) Fernandes, M. X.; Garcia de la Torre, J.; Castanho, M. A. Joint
determination by Brownian dynamics and fluorescence quenching of the
in-depth location profile of biomolecules in membranes. Anal. Biochem.
2002, 307, 1–12.
(21) Matos, P. M.; Goncalves, S.; Santos, N. C. Interaction of peptides
with biomembranes assessed by potential-sensitive fluorescent probes.
J. Pept. Sci. 2008, 14, 407–15.
(22) Domingues, M. M.; Castanho, M. A.; Santos, N. C. rBPI(21)
promotes lipopolysaccharide aggregation and exerts its antimicrobial effects
by (hemi)fusion of PG-containing membranes. PLoS One 2009, 4, e8385.
(23) Domingues, M. M.; Santiago, P. S.; Castanho, M. A.; Santos,
N. C. What can light scattering spectroscopy do for membrane-active
peptide studies? J. Pept. Sci. 2008, 14, 394–400.
(24) Zimmermann, M. Ethical guidelines for investigations of ex-
perimental pain in conscious animals. Pain 1983, 16, 109–10.
(25) Tjolsen, A.; Berge, O. G.; Hunskaar, S.; Rosland, J. H.; Hole, K.
The formalin test: an evaluation of the method. Pain 1992, 51, 5–17.
(26) Coggeshall, R. E. Fos, nociception and the dorsal horn. Prog.
Neurobiol. 2005, 77, 299–352.
’ REFERENCES
(1) Takagi, H.; Shiomi, H.; Ueda, H.; Amano, H. A novel analgesic
dipeptide from bovine brain is a possible Met-enkephalin releaser.
Nature 1979, 282, 410–2.
(2) Nishimura, K.; Kaya, K.; Hazato, T.; Ueda, H.; Satoh, M.; Takagi,
H. [Kyotorphin like substance in human cerebrospinal fluid of patients
with persistent pain]. Masui 1991, 40, 1686–90.
(3) Shiomi, H.; Ueda, H.; Takagi, H. Isolation and identification of
an analgesic opioid dipeptide kyotorphin (Tyr-Arg) from bovine brain.
Neuropharmacology 1981, 20, 633–8.
(4) Ignat’ev, D. A.; Vorob’ev, V. V.; Ziganshin, R. Effects of a number
of short peptides isolated from the brain of the hibernating ground
squirrel on the EEG and behavior in rats. Neurosci. Behav. Physiol. 1998,
28, 158–66.
(27) Amblard, M.; Rodriguez, M.; Lignon, M. F.; Galas, M. C.;
Bernad, N.; Artis-Noel, A. M.; Hauad, L.; Laur, J.; Califano, J. C.; Aumelas,
A.; et al. Synthesis and biological evaluation of cholecystokinin analogs in
which the Asp-Phe-NH2 moiety has been replaced by a 3-amino-
7-phenylheptanoic acid or a 3-amino-6-(phenyloxy)hexanoic acid. J. Med.
Chem. 1993, 36, 3021–8.
(28) Nazare, M.; Waldmann, H. Enantiospecific synthesis of the
(9S,18R)-diastereomer of the leukocyte adhesion inhibitor cyclamenol
A. Chem.—Eur. J. 2001, 7, 3363–76.
(29) Konig, W.; Geiger, R. A New Method for Synthesis of Peptides -
Activation of Carboxyl Group with Dicyclohexylcarbodiimide Using
1-Hydroxybenzotriazoles as Additives. Chem. Ber. 1970, 103, 788.
(30) Steinauer, R.; Chen, F. M. F.; Benoiton, N. L. Studies on Race-
mization Associated with the Use of Benzotriazol-1-yl-tris(dimethyl-
amino)phosphonium hexafluorophosphate (Bop). Int. J. Pept. Protein Res.
1989, 34, 295–298.
(5) Godlevsky, L. S.; Shandra, A. A.; Mikhaleva, I. I.; Vastyanov,
R. S.; Mazarati, A. M. Seizure-protecting effects of kyotorphin and
related peptides in an animal model of epilepsy. Brain Res. Bull. 1995,
37, 223–6.
(6) Chen, P.; Bodor, N.; Wu, W. M.; Prokai, L. Strategies to target
kyotorphin analogues to the brain. J. Med. Chem. 1998, 41, 3773–81.
(7) Lopes, S. C.; Soares, C. M.; Baptista, A. M.; Goormaghtigh, E.;
Cabral, B. J.; Castanho, M. A. Conformational and orientational
guidance of the analgesic dipeptide kyotorphin induced by lipidic
membranes: putative correlation toward receptor docking. J. Phys. Chem.
B 2006, 110, 3385–94.
(8) Lopes, S. C.; Fedorov, A.; Castanho, M. A. Chiral recognition of
D-kyotorphin by lipidic membranes: relevance toward improved an-
algesic efficiency. ChemMedChem 2006, 1, 723–8.
(9) Ribeiro, M.; Pinto, A.; Pinto, M.; Heras, M.; Martins, I.; Correia,
A.; Bardaji, E.; Tavares, I.; Castanho, M. Inhibition of nociceptive
1939
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940

Molecular Pharmaceutics
ARTICLE
(31) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.;
Kluwer Academic/Plenum Publishers: New York, 1999.
(32) Ross, J. B. A.; Laws, W. R.; Rousslang, K. W.; Wyssbrod, H. R.
Tyrosine fluorescence and phosphorescence from proteins and poly-
peptides. In Topics in Fluorescence Spectroscopy, Vol. 3: Biochemical
Applications; Lakowicz, J. R., Ed.; Plenum Press: New York, 1992; pp
1ꢀ63.
(33) Mrozek, J.; Rzeska, A.; Guzow, K.; Karolczak, J.; Wiczk, W.
Influence of alkyl group on amide nitrogen atom on fluorescence
quenching of tyrosine amide and N-acetyltyrosine amide. Biophys. Chem.
2004, 111, 105–13.
(34) Reyes Mateo, C.; Ulises Acuna, A.; Brochon, J. C. Liquid-
crystalline phases of cholesterol/lipid bilayers as revealed by the
fluorescence of trans-parinaric acid. Biophys. J. 1995, 68, 978–87.
(35) Goldenberg, N. M.; Steinberg, B. E. Surface charge: a key determi-
nant of protein localization and function. Cancer Res. 2010, 70, 1277–80.
(36) Bars, D. l.; Gozariu, M.; Cadden, S. W. Animal Models of
Nociception. Pharmacol. Rev. 2001, 53, 597–652.
(37) Dubuisson, D.; Dennis, S. G. The formalin test: a quantitative
study of the analgesic effects of morphine, meperidine, and brain stem
stimulation in rats and cats. Pain 1977, 4, 161–74.
(38) Rackham, A.; Wood, P. L.; Hudgin, R. L. Kyotorphin (tyrosine-
arginine): further evidence for indirect opiate receptor activation. Life
Sci. 1982, 30, 1337–42.
(39) Huidobro-Toro, J. P.; Canessa, M.; Fischer, S. Interaction of
morphine with cholesterol monolayers. Biochim. Biophys. Acta 1976,
436, 237–41.
(40) Cladera, J.; Martin, I.; O’Shea, P. The fusion domain of HIV
gp41 interacts specifically with heparan sulfate on the T-lymphocyte cell
surface. EMBO J. 2001, 20, 19–26.
(41) Castanho, M. A.; Fernandes, M. X. Lipid membrane-induced
optimization for ligand-receptor docking: recent tools and insights for
the “membrane catalysis” model. Eur. Biophys. J. 2006, 35, 92–103.
(42) Veiga, S.; Yuan, Y.; Li, X.; Santos, N. C.; Liu, G.; Castanho,
M. A. Why are HIV-1 fusion inhibitors not effective against SARS-CoV?
Biophysical evaluation of molecular interactions. Biochim. Biophys. Acta
2006, 1760, 55–61.
(43) Romanowski, M.; Zhu, X. Y.; Kim, K.; Hruby, V. J.; O’Brien,
D. F. Interaction of enkephalin peptides with anionic model membranes.
Biochim. Biophys. Acta, Biomembr. 2002, 1558, 45–53.
(44) Tamai, I.; Sai, Y.; Kobayashi, H.; Kamata, M.; Wakamiya, T.;
Tsuji, A. Structure-internalization relationship for adsorptive-mediated
endocytosis of basic peptides at the blood-brain barrier. J. Pharmacol.
Exp. Ther. 1997, 280, 410–5.
(45) Yeung, T.; Gilbert, G. E.; Shi, J.; Silvius, J.; Kapus, A.; Grinstein,
S. Membrane phosphatidylserine regulates surface charge and protein
localization. Science 2008, 319, 210–3.
(46) Kolesnikov, Y.; Soritsa, D. Analgesic synergy between topical
opioids and topical non-steroidal anti-inflammatory drugs in the mouse
model of thermal pain. Eur. J. Pharmacol. 2008, 579, 126–33.
(47) Morris, R.; Cheunsuang, O.; Stewart, A.; Maxwell, D. Spinal
dorsal horn neurone targets for nociceptive primary afferents: do single
neurone morphological characteristics suggest how nociceptive infor-
mation is processed at the spinal level. Brain Res. Brain Res. Rev. 2004,
46, 173–90.
1940
dx.doi.org/10.1021/mp2003016 |Mol. Pharmaceutics 2011, 8, 1929–1940