A prodrug nanoparticle approach for the oral delivery of a hydrophilic peptide, leucine5-enkephalin, to the brain

Article

pubs.acs.org/molecularpharmaceutics

A Prodrug Nanoparticle Approach for the Oral Delivery of a

Hydrophilic Peptide, Leucine⁵-enkephalin, to the Brain

Aikaterini Lalatsa, Vivian Lee, John P. Malkinson, Mire Zloh, Andreas G. Schatzlein,

̈

and Ijeoma F. Uchegbu*

UCL School of Pharmacy, University of London, 29-39 Brunswick Square, London, WC1N 1AX, U.K.

S

* Supporting Information

ABSTRACT: The oral use of neuropeptides to treat brain

disease is currently not possible because of a combination of

poor oral absorption, short plasma half-lives and the blood−

brain barrier. Here we demonstrate a strategy for neuropeptide

brain delivery via the (a) oral and (b) intravenous routes. The

strategy is exempliﬁed by a palmitic ester prodrug of the model

drug leucine⁵-enkephalin, encapsulated within chitosan

amphiphile nanoparticles. Via the oral route the nano-

particle−prodrug formulation increased the brain drug levels

by 67% and signiﬁcantly increased leucine⁵-enkephalin’s

antinociceptive activity. The nanoparticles facilitate oral

absorption and the prodrug prevents plasma degradation,

enabling brain delivery. Via the intravenous route, the

nanoparticle−prodrug increases the peptide brain levels by 50% and confers antinociceptive activity on leucine⁵-enkephalin.

The nanoparticle−prodrug enables brain delivery by stabilizing the peptide in the plasma although the chitosan amphiphile

particles are not transported across the blood−brain barrier per se, and are excreted in the urine.

KEYWORDS: chitosan amphiphiles, quaternary ammonium palmitoyl glycol chitosan, GCPQ, blood−brain barrier, peptides, oral,

leucine⁵-enkephalin

toxin.¹³However despite knowledge of these technologies,

the oral delivery of peptides to the brain, the main focus of this

INTRODUCTION

The oral use of neuropeptides to treat brain disorders is not

■

work, has not been demonstrated by means of both

possible due to a combination of (a) limited peptide oral

pharmacokinetic and pharmacodynamic data, and in fact

absorption, (b) short peptide plasma half-life and (c) the

lipidized peptides have been reported not to cross the BBB

blood−brain barrier (BBB). A broadly applicable method of

after oral administration.¹⁴

orally delivering neuropeptides to the brain would have a

dramatic impact on the global brain disease morbidity ﬁgures.

Brain diseases span a wide range of conditions including

Currently the BBB limits the passage of most molecules from

the blood into the CNS. Speciﬁcally the BBB is a specialized

psychiatric disorders,²neurodegenerative diseases³and brain

structure where the blood capillaries are characterized by an

tumors.⁴At any time half a billion people are aﬀected by brain

absence of fenestrae, the presence of tight intercellular

junctions, low pinocytotic activity and high levels of eﬄux

transporters at their luminal endothelial surface, all of which

disorders, and the incidence of these conditions is steadily

rising.⁵

successfully limits the penetration of 95% of drugs to the

To deliver hydrophilic neuropeptides to the brain, via the

oral route, would require such drugs to be absorbed from the

brain.¹⁵⁻¹⁹Some peptides and regulatory proteins cross the

BBB by saturable or nonsaturable mechanisms in small

gastrointestinal tract and then be transported across the BBB.

amounts, whereas others cannot cross.^20,21However, the

The oral absorption of most hydrophilic drugs, such as

brain accumulation of many peptides and proteins in

hydrophilic peptides, is hampered by hydrogen bonding with

therapeutically meaningful amounts after systemic adminis-

the aqueous gut lumen contents, which limits diﬀusion across

the gastrointestinal epithelium.⁶Additionally peptides are

tration has not been demonstrated. A number of parenteral

rapidly degraded within the gastrointestinal tract⁷and enjoy

methods of brain delivery have been attempted for hydrophilic

drugs, e.g. the use of ligands for endogenous transporters,²²⁻²⁵

short plasma half-lives (e.g., of 3 min⁸for leucine⁵-enkephalin),

which additionally hampers their transport across the BBB.

Methods to improve the absorption of peptides such as insulin

and calcitonin in the gut have focused on encapsulation⁹⁻¹²

strategies to prevent intestinal degradation or increasing

paracellular gut permeability using the zonula occludens

Received: January 6, 2012

Revised: April 30, 2012

Accepted: May 10, 2012

Published: May 10, 2012

1665

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

a

Scheme 1. The Synthesis of Leucine⁵-enkephalin and TPLENK

a

Reagents and conditions: (a) Fmoc-Phe-OH, HBTU, NMM, DMF, rt, 2 × 25 min; (b) 20% v/v piperidine, DMF, rt, 2 × 10 min; (c) Fmoc-Gly-

OH, HBTU, NMM, DMF, rt, 2 × 25 min; (d) Fmoc-Tyr-OH, HBTU, NMM, DMF, rt, 2 × 25 min, ca. 98% over 7 steps; (e) AcOH/TFE/CH₂Cl₂

2:2:6 v/v/v, rt, 2 h or 5% w/w PhOH, TFA, rt, 2 h; (f) palmitic acid N-hydroxysuccinimide ester, NEt₃, DMF, 25 °C, 24 h.

the inhibition of eﬄux transporters^15,26or temporarily

disrupting the blood−brain barrier.²⁷All of these strategies

have their limitations: the use of endogenous to-brain

transporters tends to be limited by the carrying capacity of

the transporter, eﬄux transporters exist at sites outside the

blood−brain barrier and so are too widespread to be inhibited

routinely, while the temporary disruption of the BBB is

associated with the development of seizures.²⁸As such the

majority of these approaches have not progressed to clinical

products. However the main focus of this work is delivering to

the brain via the oral route, and the oral delivery of

neuropeptides to the brain has seen very little activity.

Our work is concerned with the delivery of peptides across

the blood−brain barrier, and here we present a simple strategy,

amenable to industrial scaleup, for the oral delivery of peptides

to the brain; delivery is achieved by preparing a lipidic prodrug

and encapsulating the prodrug within a nanoparticle. The

strategy is illustrated using the model drug leucine⁵-enkephalin:

an endogenous opioid which binds selectively to δ opioid

receptors,²⁹has a serum half-life of 3 min⁸and is not orally

active.

EXPERIMENTAL METHODS

■

Materials. All chemicals, unless otherwise stated, were

obtained from Sigma-Aldrich (Dorset, U.K.) and used without

further puriﬁcation. Solvents (HPLC grade) and acids were

purchased from Fisher Scientiﬁc (Loughborough, U.K.) and

were used without further puriﬁcation. Fluorenylmethylox-

ycarbonyl (Fmoc)-protected amino acids, O-(1H-benzotria-

zole-1-yl)-N,N,N′,N′-tetramethyluronium hexaﬂuorophosphate

(HBTU) and amino 2-chlorotrityl resin preloaded with leucine

(H-Leu-2-Cl-Trt resin, 0.86 mmol g⁻¹) were obtained from

Novabiochem (Nottingham, U.K.). The leucine⁵-enkephalin

radioimmunoassay (RIA) kit was purchased from Bachem

1666

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

(Merseyside, U.K.). Male CD-1 and Balb/C mice for the in vivo

experiments and male Wistar rats for harvest of biological

samples used in the in vitro experiments were purchased from

Harlan (Oxfordshire, U.K.).

acetonitrile (82:18) mobile phase driven by a Waters HPLC

system (Waters 515 HPLC pump, Waters 717 plus

Autosampler, Waters Ltd., Elstree, U.K.) at a ﬂow rate of 4

mL min⁻¹. The column was heated to 35 °C by a Jones

chromatography column heater model 7971 (Jones Chroma-

tography Ltd., Cardiﬀ, U.K.), and peptides were detected at 280

nm by a Waters 486 variable wavelength UV detector. The

retention time was 7.1 min for leucine⁵-enkephalin. Fractions

(4 mL each) were collected, freeze-dried and analyzed by

electrospray mass spectrometry (Finnigan Mat TSQ7000) as

described below. Fractions 6 and 7 were pooled to obtain

leucine⁵-enkephalin (95% purity), which presented as a ﬂuﬀy

white powder.

Yield = 33 mg (58%). Mp = 139 °C. MS (m/z): calculated

for C₂₈H₃₈N₅O₇= 555.3; found = 556.3 [M + H]⁺. FTIR: ν

(cm⁻¹) = 3288, 3078 (N−H stretch, O−H stretch), 2964,

2939, 2877 (C−H sat. stretch), 1650 (N⁺−H bend), 1516 (N−

H bend), 1442 (C−H sat. bend). ¹H NMR [tBu(Tyr)-LENK],

δ (ppm) 0.93 (m, 6H, CH₃- Leu), 1.39 (s, 9H, CH₃- tBu-Tyr),

1.63 (m, 3H, γ-CH β-CH₂- Leu), 2.10, (m, 1H, β-CH - Leu),

3.07 (m, 1H, β-CH₂- tBu-Tyr), 3.25 (m, 3H, β-CH₂- tBu-Tyr

and Phe), 4.0 (b, 4H, α-CH₂- Gly), 4.27 (b, 2H, α-CH - Leu

and Phe), 4.73 (b, exact number of protons obscured by

solvent, α-CH - Tyr), 7.16 (b, 2H, meta-CH - tBu-Tyr), 7.43

(b, 7H, ortho-CH - tBu-Tyr and ortho-CH - Phe and meta-CH

- Phe and para-CH - Phe).

Synthesis of TPLENK. The synthesis of TPLENK was carried

out as depicted in Scheme 1. Triethylamine (350 μL, 2.5

mmol) was added to a suspension of Fmoc-Tyr-(OH)-Gly-Gly-

Phe-Leu-2-Cl-Trt-Resin (0.181 g, 0.1 mmol) preswelled in

DMF (8 mL), and the resultant suspension was reacted with

the N-hydroxysuccinimide ester of palmitic acid (177 mg, 0.5

mmol) in DMF (5 mL) at 25 °C (constant temperature room

or using a water bath) for 24 h, during which time the

suspension was agitated (120 rpm). The mixture was then

concentrated in vacuo to remove volatile products and the

residue dispersed in DMF (4 mL). The DMF suspension was

ﬁltered, and the residue washed with copious amounts of DMF

(100 mL). The product bound to the resin was treated with

piperidine in DMF (20% v/v, 20 mL) for 20−25 min. After

washing with DMF and ﬁltration, cleavage of the peptide chain

from the resin was performed by treatment with the acetic acid

cleavage mixture described above (1 mL of cleavage mixture for

0.1 g of dried resin) for 2 h at room temperature. The crude

TPLENK was then precipitated with cold puriﬁed water (4 °C,

pH 7.0), left to stand for 12 h at −20 °C, collected by

centrifugation (3 × 2500 rpm at 4 °C for 30 min), with further

washing of the pellet with cold water after the ﬁrst two

centrifugations, and lyophilized.

Peptide puriﬁcation was achieved using semipreparative

reverse-phase HPLC as detailed above. The retention time

for TPLENK was 8.3 min. Fractions (4 mL each) were

collected and freeze-dried and analyzed by electrospray mass

spectrometry (Finnigan Mat TSQ7000) as described below.

Fractions 8 to 9 were pooled to obtain TPLENK (95% purity),

which presented as a white powder after lyophilization.

Yield = 42 mg (58%). Mp =173−174 °C. MS (m/z):

calculated for C₄₄H₆₇N₅O₈= 794.0; found = 792.8 [M − H]⁺.

FTIR: ν (cm⁻¹) = 3300, 3072 (N−H stretch, O−H stretch),

2922, 2860 (C−H sat. stretch), 1723 (CO ester stretch), 1635

(N⁺−H bend), 1515 (N−H bend), 1442 (C−H sat. bend). ¹H

NMR: δ (ppm), 0.86 (m, 9H, CH₃- Leu and palmitoyl), 1.25

(m, 22H, −CH₂−CH₂− CH₂palmitoyl), 1.52 (m, 2H, β-CH₂-

Peptide Synthesis. Leucine⁵-enkephalin and TPLENK

synthesis was carried out manually, using standard Fmoc-based

solid phase methodology (0.5 mmol per batch) as depicted in

Scheme 1.

Synthesis of Leucine⁵-enkephalin. To preswelled H-Leu-2-

Cl-Trt resin (0.581 g, 0.5 mmol) in N-methylmorpholine

[NMM, 4.45% v/v in dimethylformamide (DMF), 10 mL] was

added Fmoc protected amino acid (Fmoc-^L-phenylalanine, 0.48

g, 2.5 mol) and HBTU (0.47 g, 2.5 mmol) dissolved in NMM

(4.45% v/v in DMF, 5 mL). The reaction was left for 25 min.

For each amino acid residue coupled, the above procedure was

performed twice. After coupling each residue the Kaiser test³⁰

was performed to ensure coupling had taken place.

Deprotection of the Fmoc moiety after washing the resin

with DMF (150 mL) was achieved by adding piperidine (20%

v/v in DMF, 10 mL) to the resin beads, which was then

agitated for 10 min (performed twice). The process detailed

above was repeated for each amino acid residue until synthesis

of the peptide was complete. Glycine was double-coupled as

there were two adjacent glycine units in the peptide sequence.

All peptide synthesis steps were performed at room temper-

ature. Once peptide synthesis had been completed, the resin

was washed with copious amounts of DMF (250 mL), followed

by copious amounts of dichloromethane (DCM, 100 mL) and

then by a mixture of DCM, methanol (1: 1, 200 mL). The

resin-bound peptide was dried under vacuum, transferred to a

preweighed glass container, left in a desiccator under vacuum

for 24 h and then stored under nitrogen at −20 °C until

required. The cleavage of leucine⁵-enkephalin from the 2-

chlorotrityl resin, while still preserving the tertiary butyl ether

(tBu) protection on the tyrosyl hydroxyl group, was achieved

by adding an acetic acid cleavage mixture (acetic acid, 2,2,2-

triﬂuoroethanol, DCM, 2:2:6, 1 mL of cleavage mixture for 0.1

g of dried resin) to the peptide bound resin obtained from

above. The reaction was left for 2 h and washed three times

with equal volumes of cleavage mixture. To the ﬁltrate was

added hexane (15 times the total cleavage mixture volume) to

evaporate oﬀ acetic acid as an azeotrope with hexane. To the

resulting reduced volume suspension containing the crude

peptide was added ice cold diethyl ether (40−50 mL) to

precipitate the crude peptide; the latter of which was left for 12

h at −20 °C, with the peptide being collected by centrifugation

(3 × 1500 rpm at 4 °C for 45 min, Hermle Z323K, VWR,

Lutterworth, U.K.) with further washing with frozen diethyl

ether after the ﬁrst two centrifugations. Diethyl ether was

decanted, and the crude peptide was dried with nitrogen,

dissolved in water and lyophilized.

An alternative method of cleaving leucine⁵-enkephalin from

the resin was also employed if preservation of the tBu

protection of the tyrosyl hydroxyl group was not required,

namely, the incubation of the peptide bound resin with Reagent

P [phenol crystals, triﬂuoroacetic acid (TFA), 5:95] for 2 h.

The crude peptide was once again obtained by evaporation of

TFA, precipitation in diethyl ether and centrifugation.

Peptide puriﬁcation was achieved using semipreparative

reverse-phase HPLC (RP-HPLC). Crude peptide (15 mg

mL⁻¹) was chromatographed over a semipreparative Waters

Spherisorb ODS₂C18 column (10 mm × 250 mm, pore size

=10 μm) using an ammonium acetate buﬀer (25 mM),

1667

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

Leu), 1.64 (m, 1H, γ-CH - Leu), 1.97 (−CH₂−CO−

palmitoyl), 2.52 (m, exact number of protons obscured by

solvent peak, β-CH₂, Phe), 2.87, (m, 4H, β-CH₂, Phe and

palmitoyl Tyr), 3.08 (m, 3H, β-CH₂- palmitoyl Tyr), 3.80 (m,

exact number of protons obscured by solvent peak, CH₂-Gly),

4.17 (t, 1H, α-CH - Leu), 4.20 (t, 1H, α-CH - palmitoyl Tyr),

4.58 (t, 1H, α-CH - Phe), 6.65 (m, 2H, ortho CH - palmitoyl

Tyr), 6.93 (m, 2H, meta CH - palmitoyl Tyr), 7.24 (m, 5H,

ortho CH, Phe and meta CH - Phe and para CH - Phe), 8.08

(b, 5H, NH, Leu and Phe and Gly; NH₂- palmitoyl-Tyr).

Mass Spectrometry (MS). Leucine⁵-enkephalin (1 mg) was

dissolved in methanol (1 mL) and infused into a TSQ 7000

mass spectrometer (Thermo Fisher Scientiﬁc, Loughborough,

U.K.) operated in the positive electrospray ionization (EI)

mode at a rate of 1 mL h⁻¹, with a needle voltage of 9.72 × 10⁸

V and a capillary temperature of 250 °C. TPLENK (1 mg) was

dissolved in methanol (1 mL) and infused into a TSQ 7000

mass spectrometer operated in the negative EI mode at a rate of

1 mL h⁻¹, with a cone voltage of 25 V and capillary temperature

of 250 °C.

Nuclear Magnetic Resonance (NMR). ¹H NMR and ¹H−¹H

COSY experiments were performed on all peptides on a Bruker

AMX 400 MHz spectrometer (Bruker Instruments, Coventry,

U.K.). TPLENK was analyzed in deuterated dimethyl sulfoxide

(DMSO, 0.6 mL, concentration 5.8 mM), while tBu(Tyr)-

leucine⁵-enkephalin was analyzed in deuterated water (D₂O, 0.6

mL, 3.6 mM). Analyses were performed at a temperature of

45−50 °C.

Horizontal Attenuated Total Reﬂectance Fourier-Trans-

formed Infrared Spectroscopy (HATR-FTIR). The infrared

absorption spectra for the samples were recorded using a

Perkin-Elmer Spectrum 100 FTIR spectrometer equipped with

a Universal Attenuated Total Reﬂectance accessory and a zinc

selenide crystal (4000−650 cm⁻¹) and Spectrum FTIR

software. A background spectrum was recorded on a clean

zinc selenide window before a sample spectrum was recorded.

HPLC Peptide Analyses. Reverse phase high performance

liquid chromatography (RP-HPLC) analyses were performed

on peptide samples, and chromatography conditions are given

in Table 1.

Polymer Synthesis. Quaternary ammonium palmitoyl

glycol chitosan (GCPQ24, shown in Figure 1) was synthesized

as previously described.³¹

Yield = 124 mg (41%); M_w= 21,130 9,746 g mol⁻¹; M_n=

12,391

3,370 g mol⁻¹; M_w/M_n= 1.79

0.95; mol %

palmitoylation = 15.0 4.94; mol % quaternization = 8.1

2.30 (n = 3).

Conformational Analysis of Peptide. Leucine⁵-enkepha-

lin and TPLENK conformations were built using Maestro v 7.5

(Schrodinger, Camberly, U.K.), and the protonation states were

̈

adjusted using Ligprep (Schrodinger, Camberly, U.K.). All

̈

simulations were carried out using Macromodel,³²the

OPLS2005 force ﬁeld parameters³³and the water was

considered implicitly by a generalized Born solvent accessibility

(GB/SA) continuum solvent model,³⁴with a constant dielectric

function (ε = 1), and an extended nonbonded cutoﬀ (van der

Waals = 8 Å, electrostatic = 20 Å, hydrogen bonding = 4 Å) was

used.³²Two thousand steps of conformational search was

performed on the peptides using the Monte Carlo multiple

minima (MCMM) method.³⁵The energy cutoﬀ was 30 kJ

mol⁻¹above the lowest energy conformation. The root mean

square deviation (rmsd) between the α carbon atoms of two

peptides superpositioned using Maestro was 0.15 Å.

1668

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

Figure 1. Quaternary ammonium palmitoyl glycol chitosan (GCPQ).

The lowest energy structures of both peptides were inserted

into a cubic water box with a 10 Å buﬀer using Maestro System

Builder, with the water molecules interacting through the Single

Point Charge model. Molecular dynamics simulations of fully

solvated systems were carried out by DESMOND (Desmond

Molecular Dynamics System, version 2.0, D. E. Shaw Research,

New York) and the OPLS2005 force ﬁeld. The resulting

trajectories were analyzed using Maestro. Solvated LENK and

TPLENK systems, consisting of 3167 and 5140 atoms

respectively, underwent several short energy minimization

and relaxation runs. After equilibration, data were collected

from 200 ps production runs. The end point structures were

used for superposition of the α carbon atoms for the two

peptides, and the resulting rmsd was 0.81 Å, indicating the

preservation of the backbone conformations of these two

peptides despite the introduction of the palmitoyl group.

Polymer−Peptide Formulations. Preparation of GCPQ−

Peptide Formulations. GCPQ−peptide formulations were

prepared by vortexing a mixture of GCPQ and the peptide in

a suitable aqueous disperse phase followed by probe sonication

of the peptide and GCPQ for 8−15 min on ice with the

instrument set at 50−75% of its maximum output (MSE

Soniprep 150, MSE, London, U.K.). In some cases the peptide

alone was probe sonicated in aqueous media.

TEM was performed by the placement of a drop of the

nanoparticle suspension on the Formvar/carbon coated grid.

Excess sample was blotted oﬀ with Whatman No. 1 ﬁlter paper,

and the samples were negatively stained with uranyl acetate

(1% w/v). Samples were imaged using a FEI CM120 BioTwin

transmission electron microscope (Philips, Eindoven, The

Netherlands). Images were captured using an AMT digital

camera.

Nanoparticle size was measured by PCS on a Malvern

Zetasizer 3000 H_SA(Malvern Instruments, Malvern, U.K.) at 25

°C, a wavelength of 633 nm and a detection angle of 90°. Data

was analyzed by the Contin method of size distribution analysis.

Prior to measurements, polystyrene standards (diameter = 100

nm) were measured; size results were in accordance with the

nominal size of the standard particles. Peptide content was

analyzed by RP-HPLC as described above.

Plasma Stability of Peptides. Fresh blood, obtained from

male Balb/C mice, was collected in evacuated sterile (spray

coated with tripotassium ethylenediaminetetraacetic acid [3.6

mg]), medical grade PET tubes (3 × 75 mm K3E Vacutainer,

BD Biosciences, U.K.) and maintained on ice (4 °C) until

centrifugation. Plasma was obtained as the supernatant after

centrifugation of blood samples at 1600g (4,800 rpm) at 4 °C

for 15 min (Hermle Z323K centrifuge, VWR, Poole, U.K.). The

1669

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

separated mouse plasma was stored at −80 °C until required.

Before each stability experiment, the plasma was quickly thawed

and diluted to 50% (v/v) with isotonic sodium chloride (0.9%

w/v). Diluted plasma (50%, 495 μL) was incubated at 37 °C

(60 rpm) for 30 min prior to the addition of the peptide

formulations (5 μL). Plasma was diluted to slow down the

peptide degradation and allow degradation to be measured. All

peptide formulations were prepared at a 5 mM concentration

(2.78 mg mL⁻¹for leucine⁵-enkephalin or 3.97 mg mL⁻¹for

TPLENK and when required 7.23 mg mL⁻¹or 10.32 mg mL⁻¹

of GCPQ respectively) in isotonic sodium chloride (0.9% w/v)

apart from TPLENK alone, which was prepared in glycerol

(2.25% w/v). Peptide formulations were prepared by probe

sonication. Formulations were ﬁltered (0.8 μm, 13 mm,

Acrodisc syringe ﬁlter, low protein binding, PALL life sciences,

VWR, Poole, U.K.) after probe sonication and analyzed by

HPLC for peptide content. The stability of the peptide

formulations leucine⁵-enkephalin, leucine⁵-enkephalin−GCPQ

[(M_w= 14 kDa, mol % palmitoylation = 17%, mol % quaternary

ammonium groups = 7%) 1:2.6 g g⁻¹], TPLENK and

TPLENK−GCPQ [(M_w= 14 kDa, mol % palmitoylation =

17%, mol % quaternary ammonium groups = 7%) 1:2.6 g g⁻¹]

was determined in triplicate. At various time intervals, aliquots

(50 μL) were removed and methanol (150 μL) was added to

quench enzyme activity. Samples were then immediately placed

in −20 °C for at least 2 h prior to centrifugation at 13,000 rpm

for 15 min (MicroCentaur, MSE, London, U.K.). The

supernatant (∼180 μL) was collected and 40 μL was analyzed

by reverse phase HPLC as described above.

Bioconversion in Plasma, Liver and Brain Homoge-

nates. Fresh blood, obtained from male Wistar rats, was

centrifuged immediately at 3000 rpm and 4 °C for 15 min

(Hermle Z323K centrifuge, VWR, Poole, U.K.). The separated

rat plasma was stored at −80 °C until required. Before each

stability experiment, the plasma was quickly thawed and diluted

with 9 volumes of HBSS [Hanks buﬀered salt solution, pH =

7.4 (sodium chloride = 137 mM, potassium chloride = 5.37

mM, potassium phosphate monobasic = 0.44 mM, glucose =

5.55 mM, sodium phosphate dibasic anhydrous = 0.34 mM,

sodium bicarbonate = 4.17 mM)] to give a 10% v/v plasma

sample.

Rat livers and brains were obtained from male Wistar rats

and the tissues blotted to dryness, weighed, sliced into small

pieces, and homogenized on ice with ice-cold HBBS (1 mL per

g of tissue) using a 3 mL glass homogenizer (Jenkins, Poole,

U.K.). The homogenates were stored at −80 °C until required.

Prior to using these samples for experiments, the homogenates

were quickly thawed and rehomogenized on ice with an equal

volume of ice-cold HBSS. Cell debris and nuclei were removed

by centrifugation for 10 min at 13,000 rpm and 4 °C (MSE,

MicroCentaur, London, U.K.). The supernatants were used for

esterase activity and peptide prodrug stability studies.

The total esterase activity in biological media was assessed at

pH 7.4 and 25 °C using p-nitrophenyl butyrate (PNPB) as a

substrate as previously described.³⁶Brieﬂy the esterase activity

was assessed by monitoring the hydrolysis of PNPB to p-

nitrophenol; the ﬁnal product of this enzymatic reaction was

quantiﬁed spectrophotometrically at λ = 420 nm using a

Shimadzu UV-1650PC UV−vis double-beam spectrophotom-

eter (Shimadzu Ltd., Milton Keynes, U.K.). Esterase activities

were expressed as units per milligram of protein. One unit

represents the amount of enzyme that catalyzes the formation

of 1 μmol of p-nitrophenol per minute in HBSS, pH = 7.4 at 25

°C. A linear enzymatic hydrolysis of PNPB was maintained for

300s between 0.02 and 2 U mL⁻¹.

To determine the amount of esterase B versus esterases A/C

in various biological media, the total esterase activity was

assessed in the presence and absence of paraoxon (1 mM).

Since paraoxon inhibits only the enzymatic hydrolysis of PNPB

by esterase B, the remaining esterase activity can be attributed

to esterase A and/or C.³⁶The total protein concentration in

the biological media was determined using the Bradford protein

assay with bovine serum albumin (BSA) as a standard [λ_max

=

595 nm; BSA calibration graph in the linear range (0.1−1.4 μg

mL⁻¹: y = 0.63931x + 0.516, r²= 0.995)]. The speciﬁc esterase

activities were calculated as follows:

TE

C

SE =

(1)

where SE = speciﬁc esterase activity in U mg⁻¹protein. TE =

total esterase activity U mL⁻¹enzyme and C = concentration of

protein in mg mL⁻¹.

The enzymatic stability of TPLENK was determined in

plasma (diluted to 10% with HBSS), liver homogenates

(diluted to 50% with HBSS) and brain homogenates (diluted

to 50% with HBSS). TPLENK was incubated at a ﬁnal

concentration of 0.75 mM with either the diluted plasma,

diluted liver homogenate or diluted brain homogenate with the

respective biological matrix containing 1% DMSO. The samples

were maintained at 37 °C for at least two half-lives of the

peptide in a temperature-controlled shaking water bath (60

rpm). At various time intervals, aliquots (100 μL) were

removed. The enzyme activity of these samples was

immediately quenched by adding ice-cold, freshly prepared

guanidinium hydrochloride solution (6 M, 60 μL) in acidiﬁed

HBSS containing ortho-phosphoric acid (0.01% v/v) and

HPLC mobile phase (40 μL), followed by the storage of

samples at −80 °C until analyses could be carried out on them.

Samples were analyzed by HPLC as described above. The

apparent half-life (t_1/2) for the disappearance of TPLENK was

calculated by linear regression of log drug concentrations versus

time plots.

In Vitro Stability in Simulated Gastrointestinal Fluids.

Simulated gastric ﬂuid (SGF) without pepsin (pH 1.2) was

prepared as previously described.³⁷Leucine⁵-enkephalin (0.22

mM, 80 μL) or TPLENK (0.93 mM, 80 μL) was suspended in

SGF (1.3 mL) that had been allowed to stand at 37 °C for 15

min. Peptides were incubated in the SGF in the presence and

absence of GCPQ (peptide, GCPQ, 1:5 w/w) at 37 °C and

with shaking at 130 cycles per minute for a maximum of 3 h. At

regular time intervals aliquots (100 μL) were removed and

mixed with Na₂CO₃(0.1 M, 65 μL) to stop digestion and

HPLC mobile phase [acetonitrile:phosphate buﬀer (200 mM),

18:82, pH = 8, 35 μL]. The same amount of peptide was added

to a mixture of SGF with sodium carbonate (Na₂CO₃) (ratio

1:0.65) to serve as controls. Samples were analyzed by HPLC

as described above.

Intestinal washings were obtained from male Wistar rats

(∼250 g) using the method described by Cheng and others.³⁸

The pooled washings from all the intestinal segments were

centrifuged (4,000 rpm, 30 min), and the supernatant was

stored (−20 °C) until used. The protein content (∼0.775 mg

mL⁻¹) was determined using the Bradford assay with bovine

serum albumin (BSA) as a standard. The degradation

experiments were initiated by incubating both peptides (10

mM) with intestinal ﬂuids (46 μg mL⁻¹ﬁnal protein content in

1670

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

1.5 mL) in the absence or presence of GCPQ (peptide,

polymer ratio = 1:5 w/w). The mixtures were shaken at 37 °C

in a water bath (135 rpm). At speciﬁc time points, aliquots (150

μL) of the incubation mixtures were withdrawn and the enzyme

reaction was quenched by adding glacial acetic acid (17.5 M, 10

μL). Samples were analyzed by HPLC as described above.

In Vivo Studies. All experiments were performed under a

UK Home Oﬃce Animal License and in accordance with local

ethics committee rules. Male CD-1 outbred mice were housed

in groups of 5 in plastic cages in controlled laboratory

conditions with ambient temperature and humidity maintained

at ∼22 °C and 60% respectively with a 12 h light and dark cycle

(lights on at 07:00 and oﬀ at 19:00). Food and water were

available ad libitum and the animals acclimatized for 5−7 days

prior to experiments, within the School of Pharmacy Animal

House.

Pharmacokinetic Studies: Intravenous Delivery. Groups of

5 male CD-1 mice (4 weeks old, 18−24 g) were acclimatized to

the testing room for one hour prior to experiments and

subsequently were intravenously administered freshly ﬁltered

(0.8 μm) leucine⁵-enkephalin, leucine⁵-enkephalin−GCPQ

(1:2.3 g g⁻¹) formulations, TPLENK−GCPQ (1:2.3 g g⁻¹)

formulations or sodium chloride (0.9% w/v). At various time

intervals animals were killed and brain and blood sampled.

Plasma was separated as described above, and all tissues were

stored at −80 °C until analyses could be performed on them.

After thawing, brains were boiled for 10 min in 30 volumes of

a mixture of glacial acetic acid (1 M), hydrochloric acid (0.02

M), 2-mercaptoethanol (0.1% v/v). The tissues were

homogenized and the homogenate centrifuged (9,000 rpm,

20 min) at 4 °C to remove any debris. The pellet was discarded

and the supernatant frozen (−20 °C) and lyophilized. The

lyophilized samples were stored at −80 °C for later use.

The lyophilized brain homogenate samples were reconsti-

tuted in triﬂuoroacetic acid (1%, 1 mL) and centrifuged

(13,000 rpm, 20 min) and the supernatant stored. C18 silica

solid phase extraction columns (Waters SEP-PAK, 200 mg C18,

Bachem, U.K.) were equilibrated with buﬀer B (acetonitrile,

water, triﬂuoroacetic acid, water, 69:39:1, 1 mL) and washed

with triﬂuoroacetic acid (1%, 9 mL). The supernatant from the

brain homogenate (300 μL) was loaded onto the equilibrated

column, the column was washed with triﬂuoroacetic acid (1%, 6

mL) and the washings were discarded. The peptides were then

eluted using buﬀer B (3 mL), and the eluate was dried under a

stream of nitrogen at 30 °C using a sample concentrator (Dri-

Block DB-3, Techne Sample Concentrator, VWR, Lutterworth,

U.K.). The samples were then stored at −20 °C until

radioimmunoassay (RIA) based quantiﬁcation could be

performed. Plasma samples (100 μL) were mixed with

triﬂuoroacetic acid (1%, 100 μL) and centrifuged (13,000

rpm for 20 min), and the supernatant was chromatographed

over the solid phase extraction columns as detailed above.

Eluates were dried and the dried residues stored at −20 °C

until analyses could be performed on them.

added (100 μL), and the samples/standards were left for a

further 22 h at 4 °C. After the incubation period, goat-anti-

rabbit IgG (100 μL) was added as a secondary antibody,

normal rabbit serum (100 μL) was also added, and the samples

were left for 90 min at room temperature. Addition of RIA

buﬀer (500 μL) followed, and the standards and samples were

centrifuged (1700 g, 20 min, Eppendorf 5415R, Eppendorf UK,

Cambridge, U.K.). The radioactivity of the precipitate (bound

labeled antigen) and of supernatant (free labeled antigen) were

counted using a Perkin-Elmer (Packard) Cobra II gamma

counter, model E5002 (Perkin-Elmer, Cambridge, U.K.), for 1

min.

Plasma half-life calculations were performed using Origin 6

(Microcal Software Inc., Little Chalfont, U.K.).

Pharmacokinetic Studies: Polymer Biodistribution. The

GCPQ fractions were radiolabeled using a modiﬁcation of the

method previously reported.³⁹To GCPQ (10 mg mL⁻¹, 0.5

mL) dissolved in borate buﬀer (0.1 M, pH = 6.6) was added

trimethylamine (97.5 μL). An aliquot of this GCPQ solution

(300 μL) was then added to dry Bolton−Hunter reagent (10

μCi, 4 μL) and the reaction mixture stirred for 20 min. The

reaction mixture was then puriﬁed using Sephadex G25 spin

columns. The quantity of bound and free [¹²⁵I] iodide in the

preparation was assessed by thin layer chromatography (TLC)

using aluminum backed TLC plates (20 × 20 cm, silica gel 60

F254, layer thickness 200 μm, pore size 60 Å, Merck KGaA,

Darmstadt, Germany) and ethyl acetate:toluene (8:2) as mobile

phase. Column chromatography eﬀectively removed free label

(Figure 2 in the Supporting Information). The speciﬁc

radioactivity (radioactivity per mg of polymer and labeling

eﬃciency) was calculated by labeling the polymer (3 mg) as

detailed above, followed by exhaustive dialysis to removed

unbound label and a measurement of the radioactivity (bound

label) of the polymer (WIZARD 2470 gamma counter,

PerkinElmer, Cambridge, U.K.). Polymer speciﬁc radioactivity

= 471,274 counts per minute (cpm) mg⁻¹of polymer (0.23 μCi

mg⁻¹of polymer).

The labeled polymer (5,000,000 cpm) was diluted with

sodium chloride (0.9% w/v) to 2 mL. Radiolabeled GCPQ

(500,000 cpm, 47 mg kg⁻¹, 5.3 mg mL⁻¹, 200 μL) was injected

into the tail vein of male adult CD-1 mice (20−25 g), and at

various time intervals animals were killed and the blood and

major organs were harvested, weighed and assayed for

radioactivity.

Pharmacokinetic Studies: Oral Delivery. Groups of 5 male

CD-1 mice (4 weeks old, 18−24 g) who had been fasted

overnight were acclimatized to the testing room for one hour

prior to experiments and subsequently were administered by

oral gavage either leucine⁵-enkephalin, leucine⁵-enkephalin−

GCPQ (1:5 g g⁻¹) formulations, TPLENK−GCPQ (1:5 g g⁻¹)

formulations or distilled water. At various time intervals animals

were killed and brain and blood sampled. Plasma was separated

as described above, and all tissues were stored at −80 °C until

analyses could be performed on them. Analyses of the tissues

were carried out as detailed above.

Quantiﬁcation of leucine⁵-enkephalin was carried out using

an RIA kit (S-2118, Bachem, Liverpool, U.K.) according to the

manufacturer’s protocol. The kit is based on the principle of

competitive radioimmunoassay, and it is 100% speciﬁc for

leucine⁵-enkephalin. An aliquot of the reconstituted samples

(reconstituted from the dried residue obtained above) or

standards (100 μL) diluted in RIA buﬀer (Y-1050, Bachem,

Liverpool, U.K.) was incubated with speciﬁc rabbit antiserum

(100 μL) for 22 h at 4 °C. [¹²⁵I]-Tyr-leucine⁵-enkephalin was

Pharmacodynamic Experiments. Antinociception was

assessed in mice using a tail ﬂick warm water bioassay.⁴⁰

Animals for the tail-ﬂick test were acclimatized in the testing

environment for at least 20 h prior to testing (lights oﬀ at 20:00

and on at 8:00, ambient temperature at ∼22 °C and humidity at

∼60%). Groups of male CD-1 mice (4−5 weeks old, 22−28 g)

were intravenously (n = 8) administered or administered by

oral gavage (n = 16) either leucine⁵-enkephalin, TPLENK,

1671

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

GCPQ−TPLENK, GCPQ−leucine⁵-enkephalin formulations

or distilled water (oral gavage only) or sodium chloride

(0.9% w/v, intravenous injection only). Intravenous formula-

tions were ﬁltered (0.8 μm) prior to administration. At various

time intervals the protruding distal part of the tail (∼5 cm) of

conﬁned mice in a Plexiglas restrainer was immersed in

The two aromatic rings are separated by a distance of 7.0 1.3

Å, the hydrophobic aromatic ring is separated from the nitrogen

atom by a distance of 8.2 1.0 Å and the plane of the two

aromatic rings is perpendicular to the phenolic hydroxyl group.

It is known that leucine⁵-enkephalin binds in a (1−4) β-turn

conformation to the δ-opioid receptor,⁴³and the simulated

conformations of both leucine⁵-enkephalin and TPLENK in

aqueous media are shown schematically in Figure 2. There is

circulating warm water, maintained at 55 °C

0.1 °C by a

thermostatically controlled water bath (Grant W14, Cambridge,

U.K.). The response latency times, in centiseconds, for each

mouse to withdraw its tail by a “sharp ﬂick” was recorded using

a digital stopwatch. The ﬁrst sign of a rapid tail ﬂick was taken

as the behavioral end point, which followed in most cases 1−3

slow tail movements. Three separate withdrawal latency

determinations (separated by ≥20 s) were averaged. The

baseline latency was measured for all mice 2 h prior to testing,

and mice not responding within 5 s were excluded from further

testing. The maximum possible response was set at a latency of

10 s, and this maximum latency was assigned a value of a 100%

response. The response times were then converted to

percentage of maximum possible eﬀect (% MPE) by a method

reported previously.⁴⁰

Statistical Analysis. Statistical analyses was performed via a

one-way ANOVA test using Minitab 16 (Minitab Ltd.,

Coventry, U.K.) followed by Tukey’s test.

Figure 2. Simulated lower energy conformations of leucine⁵-

enkephalin and TPLENK. Both compounds are able to adopt the

three-dimensional δ-opioid receptor pharmacophore proposed by

Shenderovich.¹

RESULTS

■

The objective of the current study was to prepare a

nanoparticle−lipidic prodrug formulation, which would enable

the delivery of neuropeptides to the brain via the oral route. We

hypothesized that the nanoparticles would promote absorption

by (a) protecting the peptide from degradation, (b) increasing

the upper gastrointestinal drug residence time due to the

adherence of GCPQ nanoparticles to the gastrointestinal

mucosa³¹and (c) promoting transcellular absorption^31,41,42of

the lipidic prodrug (the lipidic prodrug would enjoy a reduced

level of hydrogen bonding with the aqueous contents of the

gastrointestinal lumen). We also hypothesized that the prodrug

would be stabilized against degradation in the plasma once

absorbed and be delivered to the brain either as the regenerated

parent drug (regenerated by plasma esterases) or as the lipidic

prodrug (with the parent drug being regenerated by brain

esterases). It is envisaged that the lipidic prodrug would be

more favorably transported across the brain endothelial cells

due to its lipidic character.

good agreement between these simulated conformations and

the pharmacophore deﬁned by Shenderovich et al.¹The

presence of the palmitoyl group appears not to interfere with

the three-dimensional arrangement of TPLENK into the δ-

opioid pharmacophore. Selective δ-opioid receptor pharmaco-

phores derive their selectivity by interaction with a hydrophobic

pocket,⁴⁴and we speculate that this interaction may even be

enhanced by the presence of the palmitoyl chain. Hence any

noncleavage of the prodrug in vivo to yield the parent drug is

unlikely to have a negative impact on pharmacological activity,

as the prodrug itself may be intrinsically active.

Simulation studies of leucine⁵-enkephalin in water yield a

compact structure with intramolecular hydrogen bonding⁴⁵and

a hydrophobic association between the tyrosine and leucine

groups.⁴⁶In aqueous media leucine⁵-enkephalin presented as

170 nm nanoparticles, which after ﬁltration (0.22 μm) were 44

0.1 nm in size (polydispersity = 0.44 0.002, Figure 3a). We

know of no other report of leucine⁵-enkephalin intrinsically

forming nanoparticles. TPLENK on the other hand formed

dense nanoﬁbers (Figure 3b) 300−500 nm in length and 20 nm

in diameter. Amphiphilic peptides (bearing hydrophobic non-

amino acid groups at one end) are known to form nanoﬁbers

by a combination of a β-sheet arrangement of the peptide on

the outside of the nanoﬁber and the association of the

hydrophobic non-amino acid moieties in the core of the

nanoﬁber.⁴⁷In aqueous media, on association with the particle

forming chitosan amphiphile, quaternary ammonium palmitoyl

glycol chitosan (GCPQ)⁴⁸(Figure 1), leucine⁵-enkephalin

formed spherical aggregates 40 nm in size (Figure 3c) and

TPLENK formed a mixture of spherical (Figure 3d) and

nanoﬁber aggregates of 200−500 nm in size.

A novel prodrug of leucine⁵-enkephalin was synthesized by

esteriﬁcation of the free phenolic hydroxyl group of the tyrosine

residue (Scheme 1) with palmitic acid to give a palmitoyl

derivative of leucine⁵-enkephalin (tyrosine¹palmitate-leucine⁵-

enkephalin−TPLENK). Although we set out to synthesize a

prodrug of leucine⁵-enkephalin, we opted for derivatization at

the phenolic hydroxyl group of tyrosine, as this was thought to

be less likely to interfere with drug−receptor interactions. It

was thought that this approach would ensure drug activity,

should cleavage of the prodrug, to yield leucine⁵-enkephalin, fail

to occur in a timely manner. The hydroxy moiety was preferred

for esteriﬁcation instead of the free carboxy terminal of

leucine⁵-enkephalin as the latter is crucial for eliciting opioid

activity.

The three-dimensional structure of the δ opioid receptor

pharmacophore has been elucidated using non-peptide δ opioid

receptor agonists.¹The pharmacophore comprises a three-

point structure: an amine moiety (protonated at physiological

pH), a hydrophobic region (aromatic ring) and a phenolic site.

In order to ascertain if the parent drug would be regenerated

from the lipidic ester prodrug in vivo, TPLENK was incubated

with plasma and liver esterases in vitro. The protein contents of

the 50% v/v liver homogenate and the 10% v/v plasma

1672

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

Figure 3. Transmission electron micrographs (TEMs) with negative

staining of peptide nanoparticle formulations. (a) Leucine⁵-enkephalin

nanoparticles (4 mg mL⁻¹) in sodium chloride solution (0.9% w/v),

bar = 100 nm. Particles are 40−200 nm in diameter. (b) TPLENK

nanoﬁbers (2 mg mL⁻¹) in water, bar = 500 nm. Particles are 20 nm in

diameter and 200−400 nm long. (c) GCPQ (6.9 mg mL⁻¹)−leucine⁵-

enkephalin nanoparticles (3 mg mL⁻¹) in sodium chloride (0.9% w/v),

bar = 100 nm. Particles are 40 nm in diameter. (d) GCPQ (6.9 mg

mL⁻¹)−TPLENK nanoparticles (3 mg mL⁻¹) in sodium chloride

(0.9% w/v), bar = 100 nm. Particles are 30−250 nm in diameter.

Figure 4. In vitro peptide prodrug activation (mean SD). (a) The

bioconversion of TPLENK (0.61 mg mL⁻¹) into leucine⁵-enkephalin

in the presence of 10% v/v plasma containing DMSO (1% v/v),

protein content of 10% v/v plasma = 8.8 0.10 mg mL⁻¹, esterase

content of 10% v/v plasma = 734 Units mg⁻¹of protein. TPLENK is

converted to leucine⁵-enkephalin, and leucine⁵-enkephalin is rapidly

degraded by plasma esterases. (b) The bioconversion of TPLENK

(0.61 mg mL⁻¹) into leucine⁵-enkephalin in the presence of 50% v/v

liver homogenate containing DMSO (1% v/v), protein content of 50%

v/v liver homogenate = 24.2 6.75 mg mL⁻¹, esterase content of the

50% v/v liver homogenate = 189 Units mg⁻¹of protein. TPLENK is

homogenate were 24.2 6.75 mg mL⁻¹and 8.8 0.10 mg

mL⁻¹respectively. The total amount of esterases was 189 units

mg⁻¹of protein in the 50% v/v liver homogenate and 734 units

mg⁻¹of protein in the 10% v/v plasma sample, of which only

11.1 units mg⁻¹of protein (6%) was not inhibited by paraoxon

in the liver homogenate and 18.1 units mg⁻¹of protein (3%),

not inhibited by paraoxon, in the plasma. Esterases not

inhibited by paraoxon were type A/C esterases while type B

esterases are inhibited by paraoxon. Type B esterases are

abundant in the plasma and liver homogenate. The degradation

of the ester prodrugs in plasma is reported to involve type B

esterases.⁴⁹

converted to leucine⁵-enkephalin, and leucine⁵-enkephalin is rapidly

5

□

○

degraded by liver esterases. = TPLENK, = leucine -enkephalin.

homogenates of humans, dogs, rats, mice and guinea pigs.³⁶

The lack of in vitro bioconversion of TPLENK to leucine⁵-

enkephalin, observed here with the brain homogenates, could

reﬂect the diﬃculty in preserving esterase B activity in brain

homogenates.

On studying the stability of the drug/formulations in plasma

we found that leucine⁵-enkephalin is rapidly degraded by

plasma peptidases in vitro (Figure 5a). Encapsulation within

GCPQ retards this degradation (Figure 5a). TPLENK

demonstrates superior stability to plasma peptidases when

compared to leucine⁵-enkephalin, with 80% of the prodrug

remaining after 2 h, compared to 0% of the leucine⁵-enkephalin

remaining at the same time point, and encapsulation of

TPLENK within GCPQ nanoparticles further improves the

plasma stability of the prodrug. However, despite the

comparative resistance of TPLENK to degradation shown in

Figure 5a, esterase catalyzed conversion of TPLENK to

leucine⁵-enkephalin is possible (Figure 4). In comparing the

data shown in Figure 5a with that shown in Figure 4a, it is

important to note that the level of TPLENK used for the

plasma stability experiments (Figure 5a) exceeds the level used

for the prodrug activation experiments (Figure 4a) by 6-fold.

We next sought to quantify the stability of the drug/

formulations in gastrointestinal ﬂuids. There was no degrada-

TPLENK was indeed converted to leucine⁵-enkephalin in the

presence of both the plasma and the liver homogenate (Figures

4a and 4b). The half-life of TPLENK was 73 and 44 min in the

plasma sample (10% v/v) and liver homogenate (50% v/v)

respectively as calculated from pseudo-ﬁrst-order rate constants

obtained by linear regression of log drug concentration versus

time plots (y = 6.30 + 80.27^−2.73252T, r²= 0.97 for plasma and y

= 52.00^−0.94672T, r²= 0.98 for liver).

The brain homogenate does not catalyze the bioconversion

of TPLENK to leucine⁵-enkephalin (Figure 1 in the Supporting

Information) as it is rich in esterases A and C (65%) instead of

type B esterases. Esterases A and C are known to be more

abundant in the water-soluble fractions of brain homogenates,

when compared to liver homogenates and plasma³⁶(most type

B esterases are membrane bound in the brain), and it is known

that there are low levels of type B esterases in brain

1673

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

5

□

Figure 5. In vitro peptide stability (mean SD). The stability of various peptide formulations in plasma (50% v/v). = leucine -enkephalin (2.78

−1

5

−1

■

○

●

mg mL ), = GCPQ (7.23 mg mL )−leucine -enkephalin (2.78 mg mL ), = TPLENK (3.97 mg mL ), = GCPQ (10.32 mg mL )−

TPLENK (3.97 mg mL⁻¹). Signiﬁcant diﬀerences * = p < 0.05 versus leucine⁵-enkephalin. (b) The stability of various peptide formulations

5

−1

5

□

■

○

simulated gastric ﬂuid (SGF). = leucine -enkephalin (0.12 mg mL ), = TPLENK (0.70 mg mL-1), = GCPQ (0.6 mg mL )−leucine -

−1

●

enkephalin (0.12 mg mL ) nanoparticles and = GCPQ (3.5 mg mL )−LENK (0.7 mg mL ) nanoparticles. There was no degradation of the

5

−1

□

○

peptides after 3 h. (c) The stability of various peptide formulations in rat intestinal wash. = leucine -enkephalin (8.33 mg mL ), = TPLENK

−1

5

−1

■

●

(12.05 mg mL ), = GCPQ (41.65 mg mL )−leucine -enkephalin (8.33 mg mL ) and = GCPQ (60.25 mg mL )−TPLENK (12.05 mg

mL⁻¹) in the presence of rat intestinal washes. Signiﬁcant diﬀerences: * = p < 0.05 versus leucine⁵-enkephalin. TPLENK is less stable in the presence

of rat intestinal peptidases, when compared to leucine⁵-enkephalin; however, GCPQ nanoparticles protect both TPLENK and leucine⁵-enkephalin

from intestinal degradation.

tion of leucine⁵-enkephalin or TPLENK in simulated gastric

ﬂuid (Figure 5b); however, there was marked degradation of

leucine⁵-enkephalin and especially TPLENK within the rat

intestinal wash (Figure 5c). The stability of TPLENK and

leucine⁵-enkephalin in the intestinal wash was enhanced by

to produce leucine⁵-enkephalin type peptides from plasma

proteins⁵⁰and so leucine⁵-enkephalin is unlikely to be a major

substrate for pepsin. TPLENK was degraded to a much greater

extent than leucine⁵-enkephalin in the intestinal wash, although

TPLENK was not degraded to leucine⁵-enkephalin. It may be

concluded that GCPQ nanoparticles protect the peptides from

degradation by the intestinal enzymes and such protection may

enhance absorption.

Once it had been established that (a) TPLENK could be

converted to leucine⁵-enkephalin by plasma and liver esterases

(although TPLENK could adopt the same conformation as the

δ-opioid receptor pharmacophore and hence may be intrinsi-

cally active) and (b) GCPQ protects TPLENK and leucine⁵-

enkephalin from intestinal and plasma degradation, the next

formulating as GCPQ nanoparticles: with 66.5

0.89% of

TPLENK and 84.5 0.76% of leucine⁵-enkephalin recovered

after 4 h in the presence of GCPQ compared to 17.8 0.66%

and 54.1

1.71% of TPLENK and leucine⁵-enkephalin

recovered respectively in the absence of GCPQ.

Leucine⁵-enkephalin was not degraded by the low pH in the

stomach, and the main source of degradation appears to be the

intestinal enzymes. Pepsin was omitted from the SGF due to its

interference with the peptide assay; however, pepsin is known

1674

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

Table 2. Leucine⁵-enkephalin Pharmacokinetic Parameters Following the Intravenous and Oral Administration of Peptide

Formulations

plasma

brain

a

LENK dose (mg kg⁻¹

admin route

)

T_max

C_max(mean SEM,

AUC_0−t(μg

T_maxC_max(mean SEM, AUC_0−t(μg

no. of MPE

b

formulation

(h)

μg mL⁻¹

)

mL⁻¹h)

0.550

(h)

μg g⁻¹

)

g⁻¹h)

0.296

responders

c

10 intravenous

GCPQ−

0.05

1.66 0.05

1.5

0.085 0.014

4/8

LENK

c

GCPQ−

0.05

1.66 0.69

0.641

1.5

0.141 0.015

0.442

6/8

TPLENK

70 oral

LENK

1

0.293 0.04

0.266 0.05

2.339

2.811

4

0.695 0.088

1.090 0.107

10.043

16.807

1/16

6/16

GCPQ−

LENK

GCPQ−

0.5

0.171 0.03

3.079

4

1.098 0.085

16.508

9/16

TPLENK

a

b

t = 4 h with the intravenous formulation, t = 24 h with the oral formulation. Animals achieving the highest possible latency periods (10 s) when

c

exposed to the thermal stimulus, LENK = leucine⁵-enkephalin. Earliest time point sampled.

step was to study the impact of the nanoparticle−prodrug

approach on leucine⁵-enkephalin brain bioavailability after oral

administration. The radioimmunoassay (RIA) used here is

100% speciﬁc for leucine⁵-enkephalin, as stated by the

manufacturer; hence when analyzing TPLENK samples only

the amount of TPLENK bioconverted in vivo to leucine⁵-

enkephalin was quantiﬁed. Using the RIA kit, there was no

recovery of leucine⁵-enkephalin in biological samples after the

oral administration of water or intravenous administration of

sodium chloride (0.9% w/v) to control animals. Brain and

plasma were analyzed as follows: (a) these two tissues were

relevant to the hypothesis and (b) the drug is centrally active.

On oral administration, the GCPQ−TPLENK formulation

resulted in a 32% increase in leucine⁵-enkephalin plasma

AUC_0−24h(Table 2, Figure 6a). However leucine⁵-enkephalin

plasma C_maxwas lower (Table 2, Figure 6a) with the GCPQ−

TPLENK formulation, when compared to the leucine⁵-

enkephalin formulations, which reﬂects the time taken for in

situ bioconversion of the prodrug. The GCPQ−TPLENK

formulation produced higher plasma levels of leucine⁵-

enkephalin at the later time points (Figure 6a), when compared

to the leucine⁵-enkephalin formulations, an indication that the

prodrug is capable of sustained delivery of the drug.

GCPQ formulations of TPLENK and leucine⁵-enkephalin

itself were compared via the intravenous route and the use of

GCPQ−TPLENK results in a modest (16%) increase in

leucine⁵-enkephalin plasma AUC_0−4h(Table 2, Figure 6c) when

compared to GCPQ−leucine⁵-enkephalin, although the pro-

drug did not alter plasma C_max(Table 2, Figure 6c). The brain

AUC of leucine⁵-enkephalin is increased by 50% however by

the use of GCPQ−TPLENK, when compared to GCPQ−

leucine⁵-enkephalin (Table 2, Figure 6d), and the brain C_max

increased by 66%, despite the fact that a lower dose of leucine⁵-

enkephalin was given with the TPLENK formulations (Figures

6c and 6d). Both particulate formulations resulted in the same

90 min brain T_max(Table 2, Figure 6d). Once again, it must be

stressed that the results of in situ bioconversion are being

measured in the case of the TPLENK formulations and it is

possible that intact TPLENK was present in the brain and

plasma following the intravenous administration of TPLENK

formulations.

The polymer itself is not delivered appreciably across the

blood−brain barrier as no more than 0.05% of the administered

dose is seen in the brain after intravenous dosing (Figure 6e).

Instead the polymer is distributed to other tissues and excreted

via the kidneys and bladder (Figure 6e). The polymer is also

delivered to the skin, intestines and tail, and we speculate that

this could be due to an adherence to the skin capillaries.

Degradation of radiolabeled GCPQ remained low as indicated

by the low thyroid levels (<0.09% of the administered dose)

throughout the whole of the experiment (24 h). The tail levels

are unlikely to be due to an extravascular injection as the tail

levels show an increasing trend from 5 to 10 min before falling.

Interestingly there is no major accumulation in the liver and

spleen and virtually no distribution to the lungs; hence the

polymer evades macrophage uptake by the reticuloendothelial

system, unlike other particulate carriers such as liposomes.⁵²

Crucially the polymer nanoparticles deliver drug across the

blood−brain barrier, without actually signiﬁcantly traversing the

blood−brain barrier.

The most important ﬁnding, however, of this research is that

the GCPQ−nanoparticle formulations signiﬁcantly enhance the

brain delivery of leucine⁵-enkephalin (Figure 5b) when

formulated with the parent drug (leucine⁵-enkephalin) or

prodrug (TPLENK). The leucine⁵-enkephalin brain AUC_0−24h

increased by over 67% with the oral GCPQ formulations

(Table 2, Figure 6b), when compared to the oral administration

of leucine⁵-enkephalin alone. Furthermore brain C_maxwas

increased by 57% when GCPQ nanoparticles were used to

encapsulate the leucine⁵-enkephalin or TPLENK (Table 2,

Figure 6b). However the nanoparticle formulations do not alter

the brain T_max. Although brain levels of leucine⁵-enkephalin

from GCPQ−leucine⁵-enkephalin and GCPQ−TPLENK for-

mulations were indistinguishable, it must be borne in mind that,

in the case of the TPLENK formulation, we are measuring the

result of in situ bioconversion and that the unconverted

prodrug (TPLENK) may also be present in plasma and brain

samples.

Having established that both the prodrug and the nano-

particles increase the brain delivery of leucine⁵-enkephalin, we

now decided to carry out pharmacodynamic evaluations.

Leucine⁵-enkephalin causes central antinociceptive eﬀects in

the brain²⁹by acting on opioid receptors, showing a preference

for δ opioid receptors. These central antinociceptive eﬀects, as

measured by the tail ﬂick bioassay, may be used to demonstrate

the delivery of the peptide to the brain.

Via the oral route 0.02% of the leucine⁵-enkephalin dose was

detected at the C_max; these levels are similar to those detected

on subcutaneous administration of morphine (0.015%),⁵¹

a

more metabolically stable compound, demonstrating the

eﬃcacy of the strategy used.

1675

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

Figure 6. Peptide and polymer pharmacokinetics (mean SD). (a) Mouse plasma levels after the oral administration: leucine⁵-enkephalin (70 mg

−1

5

−1

kg ) ( ), leucine -enkephalin (70 mg kg )−GCPQ (350 mg kg ) nanoparticles ( ) and TPLENK 100 mg kg (= 70 mg kg⁻¹leucine⁵-

□

■

−1

●

enkephalin)−GCPQ (500 mg kg ) nanoparticles ( ) Both GCPQ formulations presented as viscous liquids due to the high concentration of

GCPQ. The concentration of leucine⁵-enkephalin within the dose was 15 mg mL⁻¹, the concentration of GCPQ within the dose was 75 mg mL⁻¹

and the dose volume varied from 85 to 115 μL for the leucine⁵-enkephalin group of animals and 120 to 165 μL for the TPLENK group of animals,

depending on the animal weight. Formulations were prepared as described in the Experimental Methods. Signiﬁcant diﬀerences: * = p < 0.05 versus

5

−1

5

−1

□

leucine -enkephalin. (b) Mouse brain levels after oral administration: leucine -enkephalin (70 mg kg ) ( ), leucine -enkephalin (70 mg kg )−

GCPQ (350 mg kg ) nanoparticles ( ) and TPLENK 100 mg kg⁻¹(= 70 mg kg⁻¹leucine -enkephalin)−GCPQ (500 mg kg ) nanoparticles ( ).

Dose volumes and concentrations are as indicated in panel a. Signiﬁcant diﬀerences: * = p < 0.05 versus leucine⁵-enkephalin. (c) Mouse plasma

levels after the intravenous injection of leucine⁵-enkephalin (10 mg kg⁻¹)−GCPQ (23 mg kg⁻¹) nanoparticles (particle diameter = 172 nm,

−1

5

−1

■

●

polydispersity = 0.40) ( ) and TPLENK [10 mg kg⁻¹(= 7 mg kg⁻¹leucine⁵-enkephalin)]−GCPQ (23 mg kg⁻¹) nanoparticles (particle diameter =

■

5

−1

●

123 nm, polydispersity = 0.27) ( ). The concentration of leucine -enkephalin within the dose was 3 mg mL ; the concentration of GCPQ within

the dose was 6.9 mg mL⁻¹, and the dose volume varied from 80 to 110 μL for the leucine⁵-enkephalin group of animals and the TPLENK group of

animals depending on the animal weight. Formulations were prepared as described in the Experimental Methods and were ﬁltered (0.8 μm) prior to

administration. Signiﬁcant diﬀerences: * = p < 0.05 versus leucine⁵-enkephalin. (d) Mouse brain levels after the intravenous injection of leucine⁵-

enkephalin (10 mg kg )−GCPQ (23 mg kg ) nanoparticles (particle diameter = 172 nm, polydispersity = 0.40) ( ) and TPLENK [10 mg kg⁻¹(=

−1

■

7 mg kg⁻¹leucine -enkephalin)]−GCPQ (23 mg kg ) nanoparticles (particle diameter = 123 nm, polydispersity = 0.27) ( ). Dose volumes and

concentrations are as detailed in panel c. Signiﬁcant diﬀerences: * = p < 0.05 versus leucine⁵-enkephalin. (e) The biodistribution of radiolabeled

GCPQ (47 mg kg⁻¹) nanoparticles on intravenous administration; GCPQ nanoparticles are excreted via the kidneys into the bladder and do not

cross the BBB; U.T. = urinary tract. Nanoparticles also do not distribute to a large extent to the liver and spleen and do not distribute at all to the

heart and lungs.

5

−1

●

On oral administration of 70 mg kg⁻¹leucine⁵-enkephalin,

the analgesia produced by GCPQ−leucine⁵-enkephalin and

GCPQ−TPLENK was signiﬁcantly higher than that produced

by the peptide alone (Figure 7a) and the peptide prodrug−

nanoparticle formulation (GCPQ−TPLENK) produced the

highest and most prolonged level of antinociception (Table 2,

1676

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

▲

Figure 7. Peptide and polymer pharmacodynamics. (a) % antinociception (mean SEM, n = 16) recorded for mice dosed orally with water ( ),

5

−1

leucine⁵-enkephalin (70 mg kg⁻¹, ), leucine -enkephalin (70 mg kg )−GCPQ (350 mg kg ) nanoparticles ( ), TPLENK (100 mg kg )−

□

■

−1

5

●

GCPQ (500 mg kg ) nanoparticles ( ). GCPQ−TPLENK formulations produce signiﬁcantly more analgesia when compared to leucine -

enkephalin alone. The baseline latency for the tail ﬂick response was 2.33 0.70 s. * = signiﬁcant diﬀerences between GCPQ groups and controls (p

< 0.01). At the 360 and 480 min time points there are signiﬁcant diﬀerences between the GCPQ−leucine⁵-enkephalin and GCPQ−TPLENK groups

▲

(p < 0.01). (b) % antinociception (mean SEM, n = 8) recorded for mice dosed intravenously with either sodium chloride (0.9% w/v, ) or

−1

◆

◇

TPLENK (14 mg kg , Δ) in glycerol (2.25% v/v), or orally with either water ( ) or TPLENK (100 mg kg , ) in water. The baseline latency for

the tail ﬂick response was 2.53 0.54 s. Signiﬁcant diﬀerences: * = p < 0.01 versus sodium chloride (0.9% w/v). (c) % antinociception (mean

5

−1

5

□

SEM, n = 8) recorded for mice dosed intravenously with sodium chloride (Δ), leucine -enkephalin (14 mg kg , ), leucine -enkephalin (14 mg

−1

■

●

kg )−GCPQ (32 mg kg ) nanoparticles ( ), TPLENK (20 mg kg )−GCPQ (46 mg kg ) nanoparticles ( ). GCPQ−TPLENK formulations

produce signiﬁcantly more analgesia when compared to leucine⁵-enkephalin. The baseline latency for the tail ﬂick response was 1.84 0.62 s.

Signiﬁcant diﬀerences: * = p < 0.05 versus sodium chloride (0.9% w/v), + = p < 0.001 versus leucine⁵-enkephalin.

Figure 7a). TPLENK is not orally active per se, however (Figure

7b), a possible consequence of its rapid intestinal degradation

(Figure 5c).

GCPQ nanoparticles (Figure 7b), evidence that TPLENK itself

results in enhanced delivery of leucine⁵-enkephalin to the brain,

when compared to leucine⁵-enkephalin alone.

Via the intravenous route and at a dose of 10 mg kg⁻¹

leucine⁵-enkephalin alone did not produce any analgesia

(Figure 7c), while the GCPQ nanoparticle formulations

produced signiﬁcant antinociception and GCPQ−TPLENK

formulations produced the highest level of antinociception

(Table 2, Figure 7c), and this correlated well with the fact that

GCPQ−TPLENK nanoparticles delivered more leucine⁵-

enkephalin to the brain. TPLENK is pharmacologically active

via the intravenous route, even when not encapsulated in

Typical Straub tail eﬀects, characterized by erect tails, were

observed when high levels of analgesia were recorded. The

Straub tail reaction in mice is an S-shaped dorsiﬂexion of the

mouse tail and is seen with high levels of morphine. It is based

on a contraction of the sacrococcygeal dorsalis muscles, which

in the case of morphine is induced by a long-lasting stimulation

of the muscle motor innervation at the level of the lumbosacral

spinal cord.⁵³Thus, this observation, which is a typical index of

opiate receptor activation and involves both central and

1677

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

peripheral components of the nervous system,⁵⁴provides

further evidence for the successful delivery of the peptide to the

site of action in adequate pharmacological amounts.

Our data show that the nanoparticle−peptide prodrug

approach is a suitable strategy for the oral delivery of gut

labile neuropeptides.

antinociceptive activity is detected and there are virtually no

false positive responses.⁵⁶In the current study, the GCPQ

formulations produced signiﬁcant numbers of analgesic

responders achieving the MPE (Table 2).

While the oral dose of TPLENK used in the current study

may appear high (100 mg kg⁻¹), it must be borne in mind that

the human oral dose of opioid receptor agonists such as

morphine is almost 200-fold lower than the mouse oral

dose.⁵⁷⁻⁵⁹The human intravenous dose of morphine (0.06 mg

kg⁻¹) is also 10−20-fold lower than an eﬀective intravenous

mouse dose.^60,61

Previous strategies to achieve brain delivery of the peptide

dalargin via the oral route, using polysorbate coated nano-

particles,^62,63have been hampered by reports that polysorbate

80 is postulated to have a toxic eﬀect at the blood−brain barrier

by increasing the permeability of tight junctions or destruction

of endothelial cells.⁶⁴Additionally other strategies which have

been attempted to achieve the oral delivery of neuropeptides

did not produce prolonged pharmacodynamic eﬀects⁶³or did

not result in the peptide being delivered across the blood−brain

barrier¹⁴and did not present pharmacokinetic evidence of

increased brain delivery.

On intravenous administration, GCPQ nanoparticles protect

leucine⁵-enkephalin and TPLENK from degradation in the

plasma (Figure 5a), and the higher brain levels and

pharmacodynamic activity of leucine⁵-enkephalin obtained on

the intravenous administration of GCPQ−TPLENK nano-

particles when compared to GCPQ−leucine⁵-enkephalin nano-

particles (Figures 6a and 7a) is attributed to the slower

enzymatic destruction of TPLENK within the blood.

DISCUSSION

■

The original hypothesis underpinning this work was that the

oral delivery of peptides to the brain could be achieved by the

encapsulation of a peptide prodrug within nanoparticles. It was

hypothesized that this strategy would promote oral absorption

due to the nanoparticles’ ability to protect the peptide from

degradation within the gut and the ability of GCPQ

nanoparticles to promote the oral absorption of hydrophobic

drugs, in part due to mechanisms linked to GCPQ’s

gastrointestinal mucoadhesion.³¹It was further hypothesized

that the absorbed prodrug would then be transported across the

blood−brain barrier, because of its greater plasma stability and

lipophilicity. GCPQ nanoparticles did indeed increase the oral

bioavailability and pharmacodynamic activity of leucine⁵-

enkephalin and TPLENK (Figures 6 and 7). The increased

oral brain bioavailability is in part derived from the ability of

GCPQ nanoparticles to promote oral absorption by protecting

both TPLENK and leucine⁵-enkephalin from degradation by

intestinal enzymes (Figure 5c) and, in the case of TPLENK, by

a resistance to plasma degradation (Figure 5a) once absorbed.

It is possible that TPLENK is also transported to a greater

extent across the blood−brain barrier due to its lipophilicity.

The main threat to peptide stability arises in the lumen of the

small intestines, where large quantities of endopeptidases and

exopeptidases are secreted by the pancreas.⁵⁵Peptidases also

arise from the mucosal cells of the villi, from the brush border

membrane of the epithelial cells and from the lysosomes which

could degrade any endocytosed peptides.⁵⁵It is conceivable

that GCPQ would form a physical barrier between these

peptidases and their substrates. On surviving the enzymatic

degradation, TPLENK, by virtue of its increased lipophilicity, is

hypothesized to partition into the intestinal epithelial cells due

to a decreased ability to hydrogen bond with the aqueous

environment of the intestinal lumen. Polymeric amphiphiles are

known to promote the oral absorption of hydrophobic

drugs,^31,41,42and GCPQ is known to act in part by

mucoadhesion and thus prolong the drug residence time at

the functional aspects of the gastrointestinal tract.³¹Unlike

underivatized chitosan, GCPQ does not open the tight

intercellular junctions of the gastrointestinal epithelium.³¹

Once absorbed into the blood TPLENK is bioconverted to

leucine⁵-enkephalin by plasma and liver esterases (Figures 4

and 6). However, although the resulting brain levels of leucine⁵-

enkephalin were similar when either GCPQ−TPLENK or

GCPQ−leucine⁵-enkephalin was orally administered (Figure

6b), the pharmacodynamic activity resulting from the oral

administration of GCPQ−TPLENK formulations was longer in

duration (Figure 7a). The similarity in leucine⁵-enkephalin

pharmacokinetics but diﬀerence in duration of pharmacological

activity could be explained by the fact that TPLENK may be

intrinsically active, especially as TPLENK is able to adopt the

three-dimensional conformation of the δ-opioid pharmaco-

phore (Figure 1). However there is currently no direct evidence

of the intrinsic activity of TPLENK.

To summarize, the present study is the ﬁrst to demonstrate

the oral delivery of labile hydrophilic peptides to the CNS via

both pharmacokinetic and pharmacodynamic studies.

ASSOCIATED CONTENT

* Supporting Information

■

S

Brain homogenate (50%) stability of TPLENK and thin layer

chromatography analysis of radiolabeled GCPQ before and

after puriﬁcation. This material is available free of charge via the

Internet at http://pubs.acs.org.

AUTHOR INFORMATION

Corresponding Author

■

*UCL School of Pharmacy, University of London, 29-39,

Brunswick Square, London, WC1N 1AX, U.K. E-mail: ijeoma.

uchegbu@ucl.ac.uk. Tel: +44 (0) 20 7753 5997. Fax: +44 (0)

20 7753 5942.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

This work was supported by the Engineering and Physical

Sciences Research Council (EPSRC), the Wellcome Trust and

the School of Pharmacy, University of London.

REFERENCES

■

(1) Shenderovich, M. D.; Liao, S. B.; Qian, X. H.; Hruby, V. J. A

three-dimensional model of the delta-opioid pharmacophore: Com-

parative molecular modeling of peptide and nonpeptide ligands.

Biopolymers 2000, 53 (7), 565−580.

(2) Jablensky, A. Epidemiology of schizophrenia: the global burden

of disease and disability. Eur. Arch. Psychiatr. Clin. Neurosci. 2000, 250

(6), 274−285.

In the quantal form of the tail ﬂick test, which detects

maximum possible eﬀects (MPE, Table 2), only morphine-like

1678

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

(3) Mayeux, R. Epidemiology of neurodegeneration. Annu. Rev.

Neurosci. 2003, 26, 81−104.

(23) Beduneau, A.; Hindre, F.; Clavreul, A.; Leroux, J. C.; Saulnier,

P.; Benoit, J. P. Brain targeting using novel lipid nanovectors. J.

Controlled Release 2008, 126 (1), 44−49.

(4) Khuntia, D.; Brown, P.; Li, J.; Mehta, M. P. Whole-brain

radiotherapy in the management of brain metastasis. J. Clin. Oncol.

2006, 24 (8), 1295−1304.

(24) Pardridge, W. M.; Boado, R. J. Delivery of pharmaceutical agents

via the human insulin receptor. United States Patent 7,388,079, 2008.

(25) Regina, A.; Demeule, M.; Che, C.; Lavallee, I.; Poirier, J.;

Gabathuler, R.; Beliveau, R.; Castaigne, J. P. Antitumour activity of

ANG1005, a conjugate between paclitaxel and the new brain delivery

vector Angiopep-2. Br. J. Pharmacol. 2008, 155 (2), 185−197.

(26) Bihorel, S.; Camenisch, G.; Lemaire, M.; Scherrmann, J. M.

Modulation of the brain distribution of imatinib and its metabolites in

mice by valspodar, zosuquidar and elacridar. Pharm. Res. 2007, 24 (9),

1720−1728.

(5) World Health Organisation. World Mental Health Day 2006:

Building AwarenessReducing Risks: Suicide and Mental Illness.

2006, http://www.who.int/mediacentre/news/releases/2006/pr53/

en/.

(6) Karls, M. S.; Rush, B. D.; Wilkinson, K. F.; Vidmar, T. J.; Burton,

P. S.; Ruwart, M. J. Desolvation Energya Major Determinant of

Absorption, but Not Clearance, of Peptides in Rats. Pharm. Res. 1991,

8 (12), 1477−1481.

(7) Pauletti, G. M.; Gangwar, S.; Siahaan, T. J.; Aube, J.; Borchardt,

R. T. Improvement of oral peptide bioavailability: Peptidomimetics

and prodrug strategies. Adv. Drug Delivery Rev. 1997, 27 (2−3), 235−

256.

(8) Sokolov, O. Y.; Kurasova, O. B.; Kost, N. V.; Gabaeva, M. V.;

Korneeva, E. V.; Mikheeva, I. G.; Zozulya, A. A. Half-life of leu-

enkephalin in the serum of infants of the ﬁrst year of life on diﬀerent

types of feeding: relationship with temperament. Bull. Exp. Biol. Med.

2004, 137 (4), 342−344.

(9) Elsayed, A.; Al Remawi, M.; Qinna, N. Farouk, A.; Badwan, A.

Formulation and characterization of an oily-based system for oral

delivery of insulin. Eur. J. Pharm. Biopharm. 2009, 73 (2), 269−279.

(10) Paul, W.; Sharma, C. P. Tricalcium phosphate delayed release

formulation for oral delivery of insulin: A proof-of-concept study. J.

Pharm. Sci. 2008, 97 (2), 875−882.

(11) Chalasani, K. B.; Russell-Jones, G. J.; Jain, A. K.; Diwan, P. V.;

Jain, S. K. Eﬀective oral delivery of insulin in animal models using

vitamin B12-coated dextran nanoparticles. J. Controlled Release 2007,

122 (2), 141−150.

(12) Martins, S.; Silva, A. C.; Ferreira, D. C.; Souto, E. B. Improving

Oral Absorption of Samon Calcitonin by Trimyristin Lipid Nano-

particles. J. Biomed. Nanotechnol. 2009, 5 (1), 76−83.

(13) Fasano, A.; Uzzau, S. Modulation of intestinal tight junctions by

Zonula occludens toxin permits enteral administration of insulin and

other macromolecules in an animal model. J. Clin. Invest. 1997, 99 (6),

1158−1164.

(27) Jolliet-Riant, P.; Tillement, J. P. Drug transfer across the blood-

brain barrier and improvement of brain delivery. Fundam. Clin.

Pharmacol. 1999, 13 (1), 16−26.

(28) Marchi, N.; Angelov, L.; Masaryk, T.; Fazio, V.; Granata, T.;

Hernandez, N.; Hallene, K.; Diglaw, T.; Franic, L.; Najm, I.; Janigro,

D. Seizure-promoting eﬀect of blood-brain barrier disruption. Epilepsia

2007, 48 (4), 732−742.

(29) Lord, J. A. H.; Waterﬁeld, A. A.; Hughes, J.; Kosterlitz, H. W.

Endogenous opioid peptidesmultiple agonists and receptors. Nature

1977, 267 (5611), 495−499.

(30) Chan, W. C.; White, P. D. Chapter 3. Basic Procedures. In Fmoc

Solid Phase Peptide Synthesis: a practical approach, 1st ed.; Chan, W. C.,

White, P. D., Eds.; Oxford University Press: Oxford, 2000; p 61.

(31) Siew, A.; Le, H.; Thiovolet, M.; Gellert, P.; Schatzlein, A.;

Uchegbu, I. Enhanced Oral Absorption of Hydrophobic and

Hydrophilic Drugs Using Quaternary Ammonium Palmitoyl Glycol

Chitosan Nanoparticles. Mol. Pharmaceutics 2012, 9 (1), 14−28.

(32) Mohamadi, F.; Richards, N.; Guida, W.; Liskamp, R.; Lipton,

M.; Caulﬁeld, C.; Chang, G.; Hendrickson, T.; Still, W. Macro-

Modelan integrated software system for modeling organic and

bioorganic molecules using molecular mechanics. J. Comput. Chem.

1990, 11, 440−67.

(33) Chang, G.; Guida, W. C.; Still, W. C. An internal coordinate

monte-carlo method for searching conformational space. J. Am. Chem.

Soc. 1989, 111 (12), 4379−4386.

(34) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T.

Semianalytical treatment of solvation for molecular mechanics and

dynamics. J. Am. Chem. Soc. 1990, 112 (16), 6127−6129.

(35) Jorgensen, W. L.; Maxwell, D. S.; TiradoRives, J. Development

and testing of the OPLS all-atom force ﬁeld on conformational

energetics and properties of organic liquids. J. Am. Chem. Soc. 1996,

118 (45), 11225−11236.

(36) Liederer, B. M.; Borchardt, R. T. Stability of oxymethyl-modiﬁed

coumarinic acid cyclic prodrugs of diastereomeric opioid peptides in

biological media from various animal species including human. J.

Pharm. Sci. 2005, 94 (10), 2198−2206.

(37) British Pharmacopoeial Commission, British Pharmacopoeia,

Appendix 1A; TSO: Norwich, U.K., 2007; Vol. IV, p A61.

(38) Cheng, W.; Satyanarayanajois, S.; Lim, L. Y. Aqueous-soluble,

non-reversible lipid conjugate of salmon calcitonin: synthesis,

characterization and in vivo activity. Pharm. Res. 2007, 24 (1), 99−110.

(39) Richardson, S. C. W.; Kolbe, H. J. V.; Duncan, R. Potential of

low molecular mass chitosan as a DNA delivery system: biocompat-

ibility, body distribution and ability to complex and protect DNA. Int.

J. Pharm. 1999, 178 (2), 231−243.

(14) Wang, J.; Hogenkamp, D. J.; Tran, M.; Li, W. Y.; Yoshimura, R.

F.; Johnstone, T. B. C.; Shen, W. C.; Gee, K. W. Reversible lipidization

for the oral delivery of leu-enkephalin. J. Drug Targeting 2006, 14 (3),

127−136.

(15) Deeken, J. F.; Loscher, W. The blood-brain barrier and cancer:

Transporters, treatment, and Trojan horses. Clin. Cancer Res. 2007, 13

(6), 1663−1674.

(16) Pardridge, W. M. The Blood-Brain Barrier: Bottleneck in Brain

Drug Development. J. Am. Soc. Exp. NeuroTher. 2005, 2, 3−14.

(17) Begley, D. J. The blood-brain barrier: Principles for targeting

peptides and drugs to the central nervous system. J. Pharm. Pharmacol.

1996, 48 (2), 136−146.

(18) Begley, D. J. ABC transporters and the blood-brain barrier. Curr.

Pharm. Des. 2004, 10 (12), 1295−1312.

(19) Lalatsa, A.; Schatzlein, A. G.; Uchegbu, I. F. Drug delivery across

̈

the blood brain barrier. In Comprehensive Biotechnology, 2nd ed.; Moo-

Young, M.; Butler, M., Webb, C., Moreira, A., Grodzinski, B., Cui, Z.,

Eds.; Elsevier: Amsterdam, 2011; pp 657−668.

(20) Batrakova, E. V.; Vinogradov, S. V.; Robinson, S. M.; Niehoﬀ,

M. L.; Banks, W. A.; Kabanov, A. V. Polypeptide point modiﬁcations

with fatty acid and amphiphilic block copolymers for enhanced brain

delivery. Bioconjugate Chem. 2005, 16 (4), 793−802.

(21) Rousselle, C.; Clair, P.; Smirnova, M.; Kolesnikov, Y.; Pasternak,

G. W.; Gac-Breton, S.; Rees, A. R.; Scherrmann, J. M.; Temsamani, J.

Improved brain uptake and pharmacological activity of dalargin using a

peptide-vector-mediated strategy. J. Pharmacol. Exp. Ther. 2003, 306

(1), 371−376.

(22) Pardridge, W. M. Drug targeting to the brain. Pharm. Res. 2007,

24 (9), 1733−1744.

(40) Polt, R.; Porreca, F.; Szabo, L. Z.; Bilsky, E. J.; Davis, P.;

Abbruscato, T. J.; Davis, T. P.; Horvath, R.; Yamamura, H. I.; Hruby,

V. J. Glycopeptide enkephalin analogs produce analgesia in mice

evidence for penetration of the blood-brain-barrier. Proc. Natl. Acad.

Sci. U.S.A. 1994, 91 (15), 7114−7118.

(41) Cheng, W. P.; Gray, A. I.; Tetley, L.; Hang, T. L. B.; Schatzlein,

A. G.; Uchegbu, I. F. Polyelectrolyte nanoparticles with high drug

loading enhance the oral uptake of hydrophobic compounds.

Biomacromolecules 2006, 7 (5), 1509−1520.

(42) Uchegbu, I. F.; Lane, M.; Schatzlein, A. G., Nanomedicines from

polymeric amphiphiles. In Handbook of Materials for Nanomedicine;

1679

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Molecular Pharmaceutics

Article

Torchilin, V. P., Amiji, M., Eds.; Pan Stanford Publishing: Singapore,

2010; pp 495−513.

(63) Schroeder, U.; Sommerfeld, P.; Sabel, B. A. Eﬃcacy of oral

dalargin-loaded nanoparticle delivery across the blood-brain barrier.

Peptides 1998, 19 (4), 777−780.

(64) Olivier, J. C.; Fenart, L.; Chauvet, R.; Pariat, C.; Cecchelli, R.;

Couet, W. Indirect evidence that drug brain targeting using

polysorbate 80-coated polybutylcyanoacrylate nanoparticles is related

to toxicity. Pharm. Res. 1999, 16 (12), 1836−1842.

(43) Blomberg, D.; Kreye, P.; Fowler, C.; Brickmann, K.; Kihlberg, J.

Synthesis and biological evaluation of leucine enkephalin turn

mimetics. Org. Biomol. Chem. 2006, 4 (3), 416−423.

(44) Kane, B. E.; Svensson, B.; Ferguson, D. M. Molecular

recognition of opioid receptor ligands. AAPS J. 2006, 8 (1), E126−

E137.

(45) vanderSpoel, D.; Berendsen, H. J. C. Molecular dynamics

simulations of leu-enkephalin in water and DMSO. Biophys. J. 1997, 72

(5), 2032−2041.

(46) Aburi, M.; Smith, P. E. A conformational analysis of leucine

enkephalin as a function of pH. Biopolymers 2002, 64 (4), 177−188.

(47) Jun, H. W.; Paramonov, S. E.; Hartgerink, J. D. Biomimetic self-

assembled nanoﬁbers. Soft Matter 2006, 2 (3), 177−181.

(48) Qu, X. Z.; Khutoryanskiy, V. V.; Stewart, A.; Rahman, S.;

Papahadjopoulos-Sternberg, B.; Dufes, C.; McCarthy, D.; Wilson, C.

G.; Lyons, R.; Carter, K. C.; Schatzlein, A.; Uchegbu, I. F.

Carbohydrate-based micelle clusters which enhance hydrophobic

drug bioavailability by up to 1 order of magnitude. Biomacromolecules

2006, 7 (12), 3452−3459.

(49) Fredholt, K.; Adrian, C.; Just, L.; Hoj Larsen, D.; Weng, S.;

Moss, B.; Juel Friis, G. Chemical and enzymatic stability as well as

transport properties of a Leu-enkephalin analogue and ester prodrugs

thereof. J. Controlled Release 2000, 63 (3), 261−273.

(50) Garreau, I.; Pejoan, C.; Bressollier, P.; Verneuil, B.; Cucumel, K.;

Cupo, A. Puriﬁcation and characterization of enkephalin-related

peptides released by in-vitro peptic digestion of bovine plasma-

proteins. Peptides 1994, 15 (7), 1195−1204.

(51) Zoellner, C.; Mousa, S. A.; Fischer, O.; Rittner, H. L.; Shaqura,

M.; Brack, A.; Shakibaei, M.; Binder, W.; Urban, F.; Stein, C.; Schaefer,

M. Chronic morphine use does not induce peripheral tolerance in a rat

model of inﬂammatory pain. J. Clin. Invest. 2008, 118 (3), 1065−1073.

(52) Gregoriadis, G.; Ryman, B. Fate of protein containing liposomes

injected into rats. Eur. J. Biochem. 1972, 24, 484−491.

(53) Stoiber, W.; Sanger, A. M.; Lembeck, F. Diﬀerent modes of

action of morphine and epibatidine in Straub tail reactions in mice.

Arzneim. Forsch. 2000, 50 (8), 683−7.

(54) Srimal, R. C.; Jaitly, K. D.; Bhargava, K. P. Eﬀect of muscle

relaxants on the straub tail phenomenon. Jpn. J. Pharmacol. 1965, 15,

323.

(55) Woodley, J. F. Enzymatic Barriers for Gi Peptide and Protein

Delivery. Crit. Rev. Ther. Drug Carrier Syst. 1994, 11 (2−3), 61−95.

(56) Franklin, K. B. J.; Abbott, F. V., Techniques for Assessing the

Eﬀects of Drugs on Nociceptive Responses. In Neuromethods 13:

Psychopharmacology; Boulton, A. A., Baker, G. B., Greensham, A. J.,

Eds.; Humana Press Inc.: Clifton, NJ, 1989; pp 145−216.

(57) Field, S.; Newton-Bishop, J. A. Melanoma and vitamin D. Mol.

Oncol. 2011, 5 (2), 197−214.

(58) Jinsmaa, Y.; Miyazaki, A.; Fujita, Y.; Li, T. Y.; Fujisawa, Y.;

Shiotani, K.; Tsuda, Y.; Yokoi, T.; Ambo, A.; Sasaki, Y.; Bryant, S. D.;

Lazarus, L. H.; Okada, Y. Oral bioavailability of a new class of mu-

opioid receptor agonists containing 3,6-bis[Dmt-NH(CH2)(n)]2-

(1H)-pyrazinone with central-mediated analgesia. J. Med. Chem.

2004, 47 (10), 2599−2610.

(59) Stadelmann, W. K.; Reintgen, D. S. Diagnosis in malignant

melanoma. Hematol. Oncol. Clin. North Am. 1998, 12 (4), 767.

(60) Young, S. E.; Martinez, S. R.; Essner, R. The role of surgery in

treatment of stage IV melanoma. J. Surg. Oncol. 2006, 94 (4), 344−

351.

(61) Ott, P. A.; Hamilton, A.; Min, C.; Safarzadeh-Amiri, S.;

Goldberg, L.; Yoon, J.; Yee, H.; Buckley, M.; Christos, P. J.; Wright, J.

J.; Polsky, D.; Osman, I.; Liebes, L.; Pavlick, A. C. A Phase II Trial of

Sorafenib in Metastatic Melanoma with Tissue Correlates. PLoS One

2010, 5 (12), e15588.

(62) Das, D.; Lin, S. S. Double-coated poly (butylcynanoacrylate)

nanoparticulate delivery systems for brain targeting of dalargin via oral

administration. J. Pharm. Sci. 2005, 94 (6), 1343−1353.

1680

dx.doi.org/10.1021/mp300009u | Mol. Pharmaceutics 2012, 9, 1665−1680

Article Doi

A prodrug nanoparticle approach for the oral delivery of a hydrophilic peptide, leucine5-enkephalin, to the brain

DOI: 10.1021/mp300009u

Source and publish data:

Authors:

Article abstract of DOI:10.1021/mp300009u

Full text of DOI:10.1021/mp300009u

Products guided by the article

R&D Labs maybe for 1375483-42-3

Relevant to this article

Hot Product