Design, synthesis and brain uptake of lat1-targeted amino acid prodrugs of dopamine

Article abstract of DOI:10.1007/s11095-012-0966-3

Purpose: Drug delivery to the brain is impeded by the blood-brain barrier (BBB). Here, we attempted to enhance the brain uptake of cationic dopamine by utilizing the large amino acid transporter 1 (LAT1) at the BBB by prodrug approach. Methods: Three amino acid prodrugs of dopamine were synthesized and their prodrug properties were examined in vitro. Their LAT1-binding and BBB-permeation were studied using the in situ rat brain perfusion technique. The brain uptake after intravenous administration and the dopamine-releasing ability in the rat striatum after intraperitoneal administration were also determined for the most promising prodrug. Results: All prodrugs underwent adequate cleavage in rat tissue homogenates. The prodrug with phenylalanine derivative as the promoiety had both higher affinity for LAT1 and better brain uptake properties than those with an alkyl amino acid -mimicking promoiety. The phenylalanine prodrug was taken up into the brain after intravenous injection but after intraperitoneal injection the prodrug did not elevate striatal dopamine concentrations above those achieved by corresponding L-dopa treatment. Conclusions: These results indicate that attachment of phenylalanine to a cationic drug via an amide bond from the meta-position of its aromatic ring could be highly applicable in prodrug design for LAT1-mediated CNS-delivery of not only anionic but also cationic polar drugs.

Full text of DOI:10.1007/s11095-012-0966-3

Pharm Res
DOI 10.1007/s11095-012-0966-3
RESEARCH PAPER
Design, Synthesis and Brain Uptake of LAT1-Targeted Amino
Acid Prodrugs of Dopamine
Lauri Peura & Kalle Malmioja & Kristiina Huttunen & Jukka Leppänen & Miia Hämäläinen & Markus M. Forsberg & Jarkko Rautio &
Krista Laine
Received: 23 August 2012 /Accepted: 14 December 2012
#
Springer Science+Business Media New York 2013
ABSTRACT
ABBREVIATIONS
Purpose Drug delivery to the brain is impeded by the
blood-brain barrier (BBB). Here, we attempted to enhance
the brain uptake of cationic dopamine by utilizing the large
amino acid transporter 1 (LAT1) at the BBB by prodrug
approach.
AADC aromatic L-amino acid decarboxylase
AUC
BBB
area under the curve
blood-brain barrier
BBN
CNS
9-borabicyclo[3.3.1]nonane
central nervous system
Methods Three amino acid prodrugs of dopamine were
synthesized and their prodrug properties were examined in
vitro. Their LAT1-binding and BBB-permeation were stud-
ied using the in situ rat brain perfusion technique. The
brain uptake after intravenous administration and the
dopamine-releasing ability in the rat striatum after intraper-
itoneal administration were also determined for the most
promising prodrug.
Results All prodrugs underwent adequate cleavage in rat tissue
homogenates. The prodrug with phenylalanine derivative as the
promoiety had both higher affinity for LAT1 and better brain
uptake properties than those with an alkyl amino acid -
mimicking promoiety. The phenylalanine prodrug was taken
up into the brain after intravenous injection but after intraperi-
toneal injection the prodrug did not elevate striatal dopamine
concentrations above those achieved by corresponding L-dopa
treatment.
COMT catechol-O-methyl transferase
EDC
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
GluT1 glucose transporter 1
HOBt
LAT1
1-hydroxybenzotriazole
large neutral amino acid transporter 1
LLOQ lower limit of quantification
PA brain permeability-surface area
SVTC2 ascorbic acid transporter
TPSA topological polar surface area
INTRODUCTION
The blood-brain barrier (BBB) acts as a major impediment
to hydrophilic drugs as they try to reach the brain making
the central nervous system (CNS) drug development a com-
plicated process (1). The majority of CNS-acting drug can-
didates do not readily enter the brain in pharmacologically
relevant amounts, and the further development of those
candidates has often been discontinued (2). As a conse-
quence, in addition to the discovery of totally novel potential
neurotherapeutic agents, the need for efficient drug delivery
systems to enhance their brain distribution seems worth-
while while the general population is aging and the physi-
cochemical properties of many novel drug molecules tend to
be less favorable for brain uptake (3).
Conclusions These results indicate that attachment of phenylal-
anine to a cationic drug via an amide bond from the meta-position
of its aromatic ring could be highly applicable in prodrug design for
LAT1-mediated CNS-delivery of not only anionic but also cationic
polar drugs.
.
.
KEY WORDS blood-brain barrier brain drug delivery
.
.
dopamine large amino acid transporter 1 prodrugs
One promising strategy for improving the poor brain
uptake of drug molecules is to adopt the prodrug approach
which utilizes specific carrier-mediated transport mecha-
nisms expressed at the BBB (4). These transporters ferry
:
:
:
:
L. Peura (*) K. Malmioja K. Huttunen J. Leppänen
:
:
:
M. Hämäläinen M. M. Forsberg J. Rautio K. Laine
School of Pharmacy, University of Eastern Finland, Kuopio, Finland
e-mail: lauri.peura@uef.fi

Peura et al.
polar pivotal nutrients, such as amino acids, glucose and
vitamins, from blood to the brain across the BBB. In the
carrier-mediated prodrug brain delivery strategy, a sub-
strate of an endogenous transporter is conjugated to an
active drug molecule in a bioreversible manner. This brain
targeting strategy can transfer the active drug molecule to
the site of action by the transporter without modifying the
structure of the active drug and, thus, there is no loss of the
biological activity of the drug. Especially glucose transport-
ers (glucose transporter 1, GluT1, SLC2A1), ascorbic acid
transporters (SVCT2, SLC23A2) and amino acid trans-
porters (large neutral amino acid transporter 1, LAT1,
SLC7A5) have been successfully utilized as drug carriers
to the brain (5–13).
Levodopa (L-dopa), the only clinically used prodrug of
dopamine, is transported into the brain via LAT1-
transporters as a “pseudonutrient” (14). The “pseudonu-
trient” structure of L-dopa is metabolically transformed to
dopamine in the brain while its metabolic transformation in
peripheral tissues is prevented by simultaneous administra-
tion of peripheral inhibitors of L-dopa metabolizing
enzymes, i.e. aromatic L-amino acid decarboxylase (AADC)
and catechol-O-methyl transferase (COMT). The need for
these inhibitors complicates L-dopa treatment. Further-
more, chronic L-dopa treatment is confronted by many
problematic pharmacokinetic obstacles, e.g. extensive me-
tabolism, plasma fluctuations, erratic oral absorption and
variability in the extent of the first-pass effect (15). In ad-
vanced Parkinson’s disease, an extensive loss of nigrostriatal
dopaminergic neurons leads to a significant decrease in
striatal AADC activity (16) that is essential to convert L-
dopa to dopamine. However, L-dopa with AADC- and
COMT-inhibitors still remains the cornerstone of the drug
treatment of Parkinson’s disease.
studies has compared a prodrug with the standard L-dopa
therapy, i.e. L-dopa combined with the peripheral inhibitors
of AADC and COMT.
In this study, we investigated three new CNS-targeted
amino acid prodrugs of dopamine (Fig. 1) which have been
designed as substrates of LAT1. We have previously
reported that the meta-substituted phenylalanine is a poten-
tial promoiety for carrying anionic drugs into the brain via
LAT1 (13). The model drugs used in all previous studies
have been anionic. In the present study, we have studied
three amino acid promoieties, aspartic acid (1), 2-amino-
apidic acid (2) and meta-substituted carboxylic acid ana-
logue of phenylalanine (3) (Fig. 1), as LAT1-targeted pro-
moieties of a cationic dopamine with an aim to enhance its
brain uptake. The model compounds used previously in
LAT1-targeting (i.e., nipecotic acid (11), mercaptopurine
(10), ketoprofen (5,7) and valproic acid (13)) have also been
relatively lipophilic especially after conjugation with the
promoiety. Here, we studied the influence of more hydro-
philic prodrugs on the LAT1-binding and determined their
translocation across the BBB. The prodrugs 1–3 were syn-
thesized, their physicochemical properties were determined
in in vitro assays and their ability to bind to LAT1 and to
cross the BBB were studied by using the in situ rat brain
perfusion technique. Finally, the brain uptake of the selected
prodrug 3 after intravenous administration in rats and its
efficacy to release dopamine in the rat brain after intraper-
itoneal administration were determined.
MATERIALS AND METHODS
Materials
Dopamine is a textbook example of a drug that has been
used for brain targeting by prodrug technology because
there are several pharmacokinetic problems encountered
in its use to treat Parkinson’s disease. First, both the amino
terminal and the catechol hydroxyl groups of dopamine are
rapidly and extensively metabolized after oral dosing (17).
Furthermore, dopamine is largely ionized in the blood-
stream and, thus, it is not able to cross the BBB via passive
diffusion (17). Hence, several prodrugs providing better
structural features for intestinal absorption, protection from
premature metabolism and improved brain distribution
have been investigated. The brain targeted prodrugs consist
mainly of lipophilic analogs of dopamine intended to im-
prove the passive delivery of dopamine especially across
intestinal endothelial cells and the BBB (18–23). However,
quite a few transporter-utilizing prodrugs of dopamine have
also been described previously (24–26), and unfortunately,
the in vivo pharmacokinetic properties have not been studied
with all of these prodrugs, and none of the pharmacokinetic
All materials used in this study were purchased from com-
mercial sources unless otherwise noted. Their origin is spec-
ified in the text. All chemicals were of the highest purity
available. All radio-labeled compounds used in the present
studies were uniformly labeled. Compounds 1–3 were syn-
thesized according to the published procedures. The man-
ufacturers of the instruments are also specified in the text.
Animals
Adult male Wistar rats were obtained from the Laboratory
Animal Centre Kuopio, Finland (originally from Harlan,
the Netherlands). At the beginning of the studies, the rats
were 7–9 weeks old and weighed 200–250 g. Pelleted food
and tap water were supplied ad libitum, and the rats were
housed in artificially light- and temperature-controlled en-
vironment (12 h light/dark cycle, 22 1°C). All procedures
with the animals were performed according to the appro-
priate European Community Guidelines and reviewed by

Dopamine Prodrugs for LAT1
Fig. 1 Chemical structures of
dopamine prodrugs 1–3.
the Animal Ethics Committee at the University of Eastern
Finland, and approved by the local provincial government
(license numbers ESAVI/6317/04.10.03/2011 and ESHL-
2009-06954/Ym-23). The animal welfare 3R principles (re-
placement, refinement and reduction) were followed.
(1.00 g, 2.70 mmol) in DCM (20 ml) was added EDC
(0.84 g, 4.43 mmol) and triethylamine (1.51 ml, 10.81 mmol).
The solution was stirred at room temperature overnight.
The crude product was purified by flash-chromatography
eluting with a gradient of DCM:MeOH mixture. After
combining the appropriate fractions and evaporating the
solvents, the title compound (1.42 g, 2.38 mmol) was
General Synthetic Procedures
1
obtained as a white solid. H NMR (500 MHz, CDCl3)
δ 2.59;2.83 (2H, dd, J=15.76, 4.41 Hz), 2.67 (2H, t, J=
6.94 Hz), 3.34–3.46 (2H, m), 3.75 (3H, s), 4.56 (1H,
dt, J=8.43, 4.14 Hz), 5.13 (2H, s), 5.14 (2H, s), 5.16
(2H, s), 6.67 (1H, dd, J=7.93, 1.83 Hz), 6.77 (1H, s),
6.87 (1H, d, J=8.24 Hz), 7.30–7.38 (11H, m), 7.45
(4H, d, J=7.02 Hz).
Prodrugs 1–3 were synthesized from Cbz-L-aspartic acid 1-
methyl ester (Aldrich, St. Louis, MO, USA), DL-2-
aminoadipic acid (Sigma, St. Louis, MO, USA) and Boc-
3-cyano-L-phenylalanine (Aldrich, St. Louis, MO, USA),
respectively, as shown in Scheme 1. Other reagents were
purchased from commercial suppliers and were used with-
out purification. Reactions were monitored by thin-layer
chromatography using aluminum sheets coated with silica
gel 60 F245 (0.24 mm) with UV and ninhydrin visualization.
Microwave reactions were conducted with Biotage Iniator
(Biotage, Sweden). Purifications by flash chromatography
were performed either on a Combiflash Companion Instru-
ment with RediSep Columns (Teledyne ISCO, Lincoln,
CA, USA) or by normal column chromatography with silica
L-2-Amino-4-((3,4-Dihydroxyphenethyl)Amino)-
4-Oxobutanoic Acid (1)
Compound 4 (1.42 g, 2.38 mmol) in DCM (10 ml) was
added to a solution of MeOH:H2O (1:1) (10 ml). Lith-
ium hydroxide (0.30 g, 7.14 mmol) was added and the
mixture was stirred for 2 h at room temperature. After
completion, the solvents were evaporated and the crude
product was carried to the next step without further
purification.
1
gel 60 (0.063–0.200 mm mesh). H and 13C nuclear mag-
netic resonance (NMR) spectra were recorded on a Bruker
Avance 500 spectrometer (Bruker Biospin, Fallanden, Swit-
zerland) operating at 500.13 and 125.75 MHz, respectively,
using the solvent peak as an internal standard. Furthermore,
the products were characterized by mass spectrometry with
a Finnigan LCQ quadrupole ion trap mass spectrometer
(Finnigan MAT, San Jose, CA) equipped with an electro-
spray ionization source, and the purity was determined by
elemental analysis (C, H, N) with a ThermoQuest CE
Instruments EA 1110-CHNS-O elemental analyzer (CE
Instruments, Milan, Italy). All compounds tested in the
assays were characterized with combustion analysis to be
at least 95% of purity unless otherwise stated.
The crude product from the previous step was dis-
solved in a 4:1 mixture of methanol and 3 M HCl
(20 ml). Palladium activated on charcoal (10% Pd basis)
(0.1 g) was added to the solution and the mixture was
put in a reactor and pressurized with hydrogen (4 bar)
and left to stir for 2 h at room temperature. The
mixture was filtered through a pad of Celite to remove
the catalyst, and the title compound (0.38 g, 1.42 mmol)
was obtained after removing the solvents under reduced
pressure as a yellowish solid. Analytically pure sample
was obtained as a white solid after purification by column
chromatography eluting using gradient of DCM:EtOH:H2O
1
Methyl 2-(((Benzyloxy)Carbonyl)Amino)-4-((3,4-bis
(Benzyloxy)Phenethyl)Amino)-4-Oxobutanoate (4)
mixture. H NMR (500 MHz, MeOD) δ 2.64 (2H, t, J=
7.32), 2.82–2.96 (2H, m), 3.35 (2H, t, J=7.32 Hz), 4.24
(1H, br. s.), 6.53 (1H, dd, J=7.93, 1.53 Hz), 6.66 (1H, d,
J=1.83 Hz ), 6.67–6.70 (1H, m). 13C (125 MHz, MeOH)
33.88, 34.43, 41.02, 56.95, 115.02, 115.46, 119.68, 130.52,
To a solution of Cbz-L-Asp-OMe (0.76 g, 2.70 mmol), HOBt
(0.40 g, 2.97 mmol) and 3,4-dibenzyloxyphenethylamine

Peura et al.
Scheme 1 Synthetic pathways
of the prodrugs of dopamine.
Reaction conditions: (a) 3,4-
dibenzyloxyphenethylamine
hydrochloride, HOBt, EDC,
TEA, DCM; (b) LiOH, MeOH:
H2O; (c) 3 M HCl, Pd, MeOH;
(d) 9-BBN, Toluene, MW (27);
(e) Ethylene diamine, THF; (f)
6 M HCl.
143.37, 144.61, 167.54, 169.45. MS calcd. for C12H17N2O5,
268.27; found 269.18. Anal. Calcd. for C12H16N2O5∙1.6
HCl: C, 44.13; H, 5.43; N, 8.58. Found: C, 43.94; H,
5.37; N, 8.48.
(2H, m), 1.65–1.98 (16H, m), 2.41 (2H, t, J=7.09 Hz),
3.69 (1H, m).
4′-(4-((3,4-Bis(Benzyloxy)Phenethyl)Amino)-4-
Oxobutyl)-5′-Oxospiro[Bicyclo[3.3.1]Nonane-9,2′-
[1,3,2]Oxazaborolidin]-2′-uide (6)
4′-(3-Carboxypropyl)-5′-Oxospiro[Bicyclo[3.3.1]
Nonane-9,2′-[1,3,2]Oxazaborolidin]-2′-uide (5)
To a solution of compound 5 (1.2 g, 4.32 mmol), HOBt
(0.642 g, 4.75 mmol) and 3,4-dibenzyloxyphenethylamine
(1.59 g, 4.23 mmol) in DCM (20 ml) was added EDC
(1.34 g, 8.64 mmol) and triethylamine (2.41 ml,
17.28 mmol). This solution was stirred at room temperature
overnight. The crude product was purified by flash-
chromatography eluting gradiently with hexane:ethyl ace-
tate mixture. After combining the appropriate fractions and
evaporating the solvents, the title compound (2.08 g,
3.49 mmol) was obtained as a white solid. 1H NMR
(500 MHz, MeOD) δ 1.50 (2H, m), 1.65–1.97 (16H, m),
2.33 (2H, t, J=7.09 Hz), 2.73 (2H, t, J=7.09 Hz), 3.39 (2H,
t, J=7.25 Hz), 3.63 (1H, m), 5.10 (2H, s), 5.13 (2H, s), 6.77
(1H, dd, J=8.20, 1.89 Hz), 6.94 (1H, d, J=1.89 Hz ), 6.96
(1H, s), 7.30–7.38 (6H, m), 7.46 (4H, t, J=8.35 Hz).
A mixture of B-methoxy-9-borabicyclo[3.3.1]nonane
1 M in hexanes (3.11 ml, 3.11 mmol) and DL-2-
aminoadipic acid (1.00 g, 6.20 mmol) were placed in
a microwave vial with a stirring bar. Ten ml of toluene
were added and the vial was capped. This mixture was
irradiated in a microwave reactor at 140°C for 20 min.
After the reaction was cooled down the clear solution
was transferred to a flask and the solvent was removed
under reduced pressure, which gave a yellowish solid.
This solid was purified by flash-chromatography using
hexane:ethyl acetate (1:1) as an eluent. After combin-
ing the appropriate fractions, the solvents were evapo-
rated to afford the title compound (1.21 g, 2.18 mmol)
as a white solid. 1H NMR (500 MHz, MeOD) δ 1.50

Dopamine Prodrugs for LAT1
DL-2-Amino-6-((3,4-Dihydroxyphenethyl)Amino)-
6-Oxohexanoic Acid (2)
eluting with hexane:ethyl acetate (1:1). After combining
the appropriate fractions, the solvents were evaporated
to afford the title compound (1.00 g, 3.05 mmol) as a
white solid. 1H NMR (500 MHz, MeOD) δ 1.42–1.87
(12H, m), 3.13 (1H, dd, J=14.66, 8.04 Hz), 3.37 (1H,
dd, J=14.66, 4.89 Hz), 4.01 (1H, t, J=6.46 Hz), 7.59
(1H, d, J=7.57 Hz), 7.86 (1H, t, J=7.57 Hz), 7.94 (1H,
d, J=7.57 Hz), 8.03 (1H, s).
Compound 6 (2.08 g, 3.49 mmol) was dissolved in THF
(20 ml) in a 100 ml Erlenmeyer flask. Ethylene diamine
(1.17 ml, 17.46 mmol) was then added and this solution
was heated with a heat gun until the reaction was complete
(1 to 2 min). The reaction was allowed to cool down and
precipitate. The precipitate was filtered, washed with THF
(3×20 ml) and dried in vacuum.
4′-(3-((3,4-Bis(Benzyloxy)phenethyl)Carbamoyl)Benzyl)-
5′-Oxospiro[Bicyclo[3.3.1]Nonane-9,2′-[1,3,2]
Oxazaborolidin]-2′-uide (9)
The crude product from previous step was dissolved in a
4:1 mixture of methanol and 3 M HCl (20 ml). Palladium
active on charcoal (10% Pd basis) (0.1 g) was added to the
solution and the mixture was placed in a hydrogen chamber
(4 mbar) and left to stir for 2 h at room temperature. The
mixture was filtered through a pad of Celite® (Sigma, St.
Louis, MO, USA) to remove the catalyst.. The compound
was obtained as a solid after removing the solvents under
reduced pressure. To ensure the formation of HCl salt, the
title compound was bubbled with HCl(g) in ACN. Remov-
ing of the solvent gave the title compound as a yellowish
solid (0.31 g, 1.05 mmol). 1H NMR (500 MHz, D2O) δ 1.57
(2H, m), 1.75 (2H, m), 2.21 (2H, t), 2.67 (2H, t, J=6.15 Hz),
3.39 (2H, t, J=6.46 Hz), 3.71 (1H, t, J=5.83 Hz), 6.69 (1H,
d, J=8.20 Hz), 6.77 (1H, s), 6.85 (1H, d, J=7.57 Hz), 13C
NMR (125 MHz, D2O) 21.14, 29.79, 33.83, 35.21, 40.59,
116.32, 116.79, 121.27, 131.16, 142.32, 143.85, 174.40,
176.61. MS calcd. for C14H20N2O5, 296.32; found 297.23.
Anal. Calcd. for C14H20N2O5∙1.5 ACN∙0.5 HCl: C, 54.29;
H, 6.70; N, 13.03. Found: C, 54.20; H, 7.09; N, 12.79.
To a solution of compound 8 (0.8 g, 2.44 mmol), HOBt
(0,33 g, 2.44 mmol) and 3,4-dibenzyloxyphenethylamine
(0.82 g, 2.22 mmol) in DCM (20 ml) were added EDC
(0.69 g, 4.43 mmol) and triethylamine (1.24 ml, 8.86 mmol).
This solution was stirred overnight at room temperature.
The crude product was purified by flash-chromatography,
eluting gradiently with a hexane:ethyl acetate mixture. After
combining the appropriate fractions and evaporating the
solvents, the title compound (0.98 g, 1.52 mmol) was
obtained as a white solid. 1H NMR (500 MHz, CDCl3) δ
1.30–1.80 (12H, m), 2.81 (2H, m), 3.08 (1 H, dd, J=14.98,
9.30 Hz), 3.37 (1H, dd, J=14.66, 4.26 Hz), 3.60 (2H, m),
3.94 (1H, m), 5.07 (2H, s), 5.10 (2H, s), 6.73 (1H, d, J=
7.88 Hz), 6.82 (1H, s), 6.89 (1H, d, J=8.20 Hz), 7.28–7.42
(11H, m), 7.47 (1H, d, J=4.10 Hz), 7.55 (1H, s).
L-2-Amino-3-(3-((3,4-DihydroxyphenethylCarbamoyl))
Phenyl)Propanoic Acid (3)
3-(2-Amino-2-Carboxyethyl)Benzoic Acid (7)
Compound 9 (0.98 g, 1.52 mmol) was dissolved in THF
(20 ml) in a 100 ml Erlenmeyer flask. Ethylene diamine
(0.46 ml, 6.85 mmol) was then added and this solution was
heated with a heat gun until the reaction was complete (1 to
2 min). The reaction was allowed to cool down and precip-
itate. The precipitate was filtered, washed with THF (3×
20 ml) and dried in vacuum.
The white solid from previous step was dissolved in a 4:1
mixture of methanol and 6 M HCl (20 ml). Palladium
activated on charcoal (10% Pd basis) (0.1 g) was added to
the solution and the mixture was put in a hydrogen chamber
(4 mbar) and left to stir for 2 h at room temperature. The
mixture was filtered through a pad of Celite to remove
the catalyst. The title compound (0.19 g, 0.54 mmol)
was obtained after removing the solvents under reduced
pressure as a yellowish solid. 1H NMR (500 MHz, DMSO)
δ 2.64 (2H, t), 3.03 (1H, m), 3.15 (1H, m), 3.44 (2H, m), 3.51
(1H, t, J=5.67 Hz), 6.44 (1H, d, J=7.88 Hz), 6.61 (1H, d, J=
7.88), 6.73 (1H, s), 7.34–7.36 (2H, m), 7.63 (1H, d, J=
7.25 Hz), 7.78 (1H, s). 13C (125 MHz, DMSO) δ34.67,
35.55, 41.27, 55.15, 115.25, 116.28, 119.00, 125.50,
Boc-3-cyano-L-phenylalanine (0.90 g, 3.10 mmol) was dis-
solved in 6 M HCl (20 ml) and refluxed overnight. The
white precipitate was filtered, washed with cold water and
dried under vacuum to afford the title compound (0.65 g,
3.10 mmol) as a white powder. 1H NMR (500 MHz,
DMSO) δ 3.20 (2H, d), 4.18 (1H, t), 7.46 (1H, m), 7.54
(1H, d, J=7.57 Hz), 7.85 (1H, d, J=7.88 Hz), 7.87 (1H, s).
4′-(3-Carboxybenzyl)-5′-Oxospiro[Bicyclo[3.3.1]
Nonane-9,2′-[1,3,2]Oxazaborolidin]-2′-uide (8)
A mixture of B-methoxy-9-borabicyclo[3.3.1]nonane 1 M
in hexanes (3.11 ml, 3.11 mmol) and compound 7 were
placed in a microwave vial with a stirring bar. Ten ml of
toluene were added and the vial was capped. This mixture
was irradiated in a microwave reactor at 140°C for 20 min.
After the reaction was cooled down, the clear solution
was transferred to a flask and the solvent was removed
under reduced pressure, which resulted in a yellowish
solid. This solid was purified by flash-chromatography

Peura et al.
128.11, 130.52, 132.11, 135.03, 137.18, 143.75, 145.40,
152.73, 166.38, 171.96. MS calcd. for C18H20N2O5 344.36;
found 345.13. Anal. Calcd. for C18H20N2O5∙H2O: C, 59.66;
H, 6.12; N, 7.73. Found: C, 59.88; H, 5.99; N, 7.61.
combined. Water was added up to 8 ml to the combined
supernatants after which they were centrifuged (14,000 g at
7°C). The supernatants were loaded into preconditioned
and equilibrated C18 SPE-columns (Supelco Discovery
DSC-18 3 ml, Supelco Park Bellefonte, PA, USA). After
the supernatants had been added, the columns were first
washed with 2 ml of 5% (v/v) MeOH in water and then with
2 ml of water. Finally, the prodrugs were eluted out from the
columns two times with 0.5 ml of 80% (v/v) MeOH in
water, and the solution was evaporated to dryness under a
stream of nitrogen at 40°C. Prior to analysis, the samples
were reconstituted in 150 μl of 0.1 M perchloric acid and
analyzed by HPLC-EC. The prodrugs were shown to re-
main intact during the whole sample preparation procedure
(data not shown). External standards were used for the brain
samples. The calibration curves of the brain methods were
linear over a range of 0.4–2.0 nmol/brain hemisphere for 2
and 0.01–2.0 nmol/brain hemisphere for 3. The lower
limits of quantification (LLOQ) for spiked samples were
0.4 nmol of prodrug/brain hemisphere for 2 and 0.01 nmol
of prodrug/brain hemisphere for 3.
Dopamine levels in rat striatal samples were assayed by
homogenizing the tissue in 0.1 M perchloric acid (1:10, w⁄v)
with MSE Soniprep 150 ultrasonic disintegrator (MSE Scien-
tific Instruments, Crawley, UK). The homogenates were
centrifuged for 15 min at 16,000 g at 4°C. The supernatants
were then filtered by 0.45 μm GHP Acrodisc Syringe Filters
(Pall Gelman Laboratory, Ann Arbor, MI, USA) and diluted
(1:1) with 0.1 M perchloric acid. Dopamine concentrations in
tissue homogenates of both striata were determined by the
HPLC method described above. External standards were
used for the striatal samples. The calibration curve of this
method was linear over a range of 0.05–10.0 pmol/injection.
The LLOQ was 0.01 pmol of dopamine/injection.
Analytical Procedures
The amounts of the prodrugs 1–3 and dopamine in analyt-
ical and biological samples were analyzed by the Agilent
HPLC 1100 Series system (Agilent Technologies Inc., Little
Falls, Wilmington, DE, USA) that consisted of a binary
gradient pump G1312A, a vacuum degasser G1322A, an
autosampler G1313A, a thermostatically controlled column
compartment G1316A and ChemStation software (Rev.
B.04.03; Agilent Technologies Inc., Palo Alto, CA, USA).
The prodrugs and dopamine were detected by an electro-
chemical (EC) detector ESA CouloChem III detector (ESA
Inc., Chelmsford, MA, USA) with an amperometric analyt-
ical cell model 5041 with a glassy carbon-ceramic target and
0.0005″ gasket. The applied potential was +150 mV and
the range 20 nA. Injection volume was 2.5 μl for analyzing
dopamine and 5 μl for the prodrugs. The analytes were
separated on a reversed phase column (Zorbax SB-Aq,
100×2.1 mm, 3.5 μm, AgilentTechnologies Inc., Little
Falls, Wilmington, DE, USA) with a Zorbax-Aq guard
column (AgilentTechnologies Inc., Little Falls, Wilmington,
DE, USA) with isocratic elution. The mobile phase con-
sisted of acetonitrile (20%, v/v) and 150 mM monobasic
sodium phosphate containing 4.8 mM citric acid monohy-
drate, 3 mM 1-octanesulfonic acid sodium salt and 50 μM
disodium EDTA (80%, v/v). The pH of the mobile phase
was adjusted to 5.60 with sodium hydroxide (10 M). The
column temperature was maintained at 30°C with mobile
phase flowing at 0.3 ml/min. The method was shown to be
linear between dopamine concentrations of 0.05–20 pmol/
injection and prodrug concentrations of 0.01–1.0 nmol/
brain hemisphere, respectively.
The prodrug 3 was separated from the thawed plasma
samples simply by protein precipitation. First, the plasma
samples (100 μl) were acidified with 40μl of 5 M hydrochloric
acid and vortexed for 2 min. MeOH (100 μl) was added and
the plasma samples were vortexed for 2 min. After standing
for 10 min, the samples were centrifuged for 5 min (16,000 g
at 7°C). The supernatants were analyzed by the HPLC-EC
method. The prodrug was shown to remain intact during the
whole sample preparation procedure. External standards
were used for the plasma samples. The calibration curves of
the plasma methods were linear over a range of 0.01–
15 nmol/100 μl of plasma. The LLOQ of spiked samples
were 0.01 nmol of prodrug/100 μL of plasma.
Brain and Plasma Sample Preparation Procedures
The prodrugs 2 and 3 were separated from brain tissue (a
whole brain hemisphere) by protein precipitation followed
by liquid-liquid extraction and solid-phase extraction (SPE).
First, the thawed brain hemisphere was homogenized with
2.5 ml of water. To precipitate the proteins, the samples
were then acidified with 200μl of 5 M hydrochloric acid and
vortexed for 5 min. MeOH (1.5 ml) was added and the
homogenates were vortexed for 2 min. After standing for
10 min, the samples were centrifuged for 10 min (14,000 g
at 7°C) after which the supernatants were collected. The
supernatant was evaporated to 3 ml under a stream of
nitrogen at 40°C. The washing of the pellet was repeated
two times with water (1.5 ml) and the supernatants were
Aqueous Solubility and Apparent Partition
Coefficients
The aqueous solubilities of 2 and 3 were determined at
room temperature (21°C) in phosphate buffer (160 mM) at

Dopamine Prodrugs for LAT1
pH 7.4. An excess amount of 2 and 3 was added to 0.5 ml of
buffer, the mixtures were shaken for 2 h, filtered (0.45 μm,
Millipore, Billerica, MA, USA) and analyzed by the HPLC-
EC method described above. The pH of the mixtures was
checked during the shaking, and adjusted if needed.
Apparent partition coefficients (log D) of 1–3 and dopa-
mine were determined in mutually saturated 1-octanol–
phosphate buffer at pH 7.4 at room temperature. 1–3 or
dopamine was first dissolved in the phosphate buffer, com-
bined with 1-octanol (1:10), and the octanol–phosphate
buffer mixtures were shaken for 180 min to reach an equi-
librium distribution. The concentrations of the prodrug or
dopamine in the phosphate phase after the shaking were
analyzed by the HPLC-EC method described above, and
log D values were calculated.
this study, transport mechanisms, LAT1-affinity and brain
uptake of the prodrugs 1–3 were determined by the modified
in situ rat brain perfusion technique (13,29,30). Briefly, the rats
were anesthetized with ketamine (90 mg/kg, i.p., Intervet
International B.V., Boxmeer, Netherlands) and xylazine
(8 mg/kg, i.p., Intervet International B.V., Boxmeer, Nether-
lands), and their right carotid artery system was exposed. The
right external carotid artery was ligated, and the right common
carotid artery was cannulated with heparinized PE-50 cathe-
ters. The blood flow to the rat brain was terminated, and the
right brain hemisphere was then perfused for 30–60 min with
perfusion solution containing prodrug 1–3 or L-dopa. The rat
was decapitated and the brain samples were collected for
further analysis. Validation, description and discussion of the
in situ rat brain perfusion technique used in this study have
been described in more detail in our previous paper (13).
Topological Polar Surface Area
Determination of the LAT1-Affinity for 1–3
Topological polar surface areas (Å2) of 1–3, dopamine and
To determine the ability of the prodrugs or L-dopa to bind to
LAT1 using the in situ rat brain perfusion, 0.2 μCi/ml [14C]-
leucine (i.e. 0.64 μM), an endogenous substrate of LAT1, was
added into the perfusion fluid with 100 μM concentration of
the prodrug 1–3 or L-dopa. The permeability-surface area of
[14C]-leucine was determined after 30 s co-perfusion, and the
LAT1-binding of the LAT1-substrates was evaluated by the
reduction in the permeability-surface area of [14C]-leucine
caused by competitive binding of the substrate to LAT1.
The permeability-surface area of 0.2 μCi/ml [14C]-leucine
after the co-perfusion with a prodrug or L-dopa was compared
with the 100% permeability-surface area of 0.2 μCi/ml [14C]-
leucine without any competitive substrates. The 100%
permeability-surface area product of [14C]-leucine has been
determined to be 19.0 3.6 μl/s/g after 30 s perfusion of
0.2 μCi/ml [14C]-leucine solution (13).
L-dopa were calculated with MOE-2011.10 [Molecular Op-
erating Environment (MOE), 2011.10; Chemical Computing
Group Inc., 1010 Sherbooke St. West, Suite #910, Mon-
treal, QC, Canada, H3A 2R7, 2011] using group contri-
butions to approximate the polar surface area only from
connection table information. The parameterization is that
of Ertl et al. 2000 (28).
In Vitro Stability Studies
To demonstrate the chemical and enzymatic stabilities of 1–3
under physiological conditions, and to demonstrate the bio-
conversion of 1–3 to dopamine, the prodrugs were exposed to
degradation in various media, including phosphate buffer (pH
7.4, 0.16 M, μ=0.5), 75% (v/v in isotonic phosphate buffer pH
7.4) rat plasma, 50% (v/v in isotonic phosphate buffer pH 7.4)
rat liver homogenate and 20% (v/v in isotonic phosphate
buffer pH 7.4) rat brain homogenate. Each prodrug 1–3 was
incubated in a thermostatically controlled water bath at 37°C,
and 100 μl aliquots were withdrawn at regular time intervals.
The withdrawn biological samples were analyzed by protein
precipitation with 200 μl of acetonitrile, and after mixing, the
samples were centrifuged for 10 min (16,000 g at 7°C). The
samples from the phosphate buffer were analyzed immediately
without protein precipitation. The HPLC-EC analysis was
performed to determine the amounts of intact prodrug and
dopamine in the supernatant.
Brain Uptake Studies of 2 and 3
The brain uptake studies of prodrugs 2 and 3 were also
performed by using the in situ rat brain perfusion technique.
To quantify their brain uptake, the right brain hemisphere
was perfused for 60 s with 37°C perfusion medium containing
25–600 μM concentrations of the prodrugs 2 or 3. After the
perfusion, the brain vasculature was washed with cold
prodrug-free perfusion medium (5°C) for 30 s to wash out
the remaining prodrug from the brain vasculature and to
decrease the potential efflux activity of the transporters.
In addition to the uptake studies of 2 and 3, our interest
was to evaluate whether 3 could cross the BBB by a non-
saturable brain transport mode or by carrier-mediated trans-
port via LAT1. In an attempt to delineate the role of the
nonsaturable transport mode, the rat brain capillaries were
first washed for 30 s with cold prodrug-free perfusion medium
In Situ Rat Brain Perfusion Technique
The in situ rat brain perfusion method of Takasato et al. (29)
is a very useful and sensitive method for evaluating the BBB
transport mechanisms and the brain uptake of drugs. In

Peura et al.
(5°C) to decrease the activity of the transporters. To maintain
the low activity of LAT1, also the 60 s perfusion with 100 μM
of the prodrug was performed at 5°C. Finally, the brain
hemisphere was washed with the cold prodrug-free perfusion
medium (5°C) for 30 s to wash out the remaining prodrug
from the brain vasculature and to ensure the reduced activity
of the transporter. The cold-perfusion study of 3 was also
performed with an endogenous competitor, 2 mM L-phenyl-
alanine, to determine the activity of LAT1 at 5°C. A compe-
tition study with 2 mM L-phenylalanine was also conducted
at 5°C to clarify whether 3 was being taken up into the brain
in a carrier-mediated manner.
investigated by treating the rats with 50.71 μmol/kg of 3.
Before the intraperitoneal administration, the rats were
allowed to adapt to the handling and animal room for
5 days. The dopamine levels after the intraperitoneal ad-
ministration of 3 plus carbidopa and entacapone were com-
pared to those after an equivalent intraperitoneal dose L-
dopa (free base) (Sigma-Aldrich) combined with carbidopa
and entacapone. Both L-dopa and prodrug 3 were admin-
istered as suspensions containing a few drops of Tween 80 in
0.9% NaCl solution. Carbidopa (30 mg/kg) and entacapone
(10 mg/kg) were administered intraperitoneally in the same
formulation 30 min before the administration of L-dopa or
3. The rats were sacrificed at 10, 30, 60, and 180 min after
administration of L-dopa or 3, and their dissected striata were
frozen immediately in liquid nitrogen and stored at −70°C
until sample preparation and HPLC-EC analyses.
Pharmacokinetic Studies of 3 in Rats
After Intravenous Bolus Injections
The brain uptake of 3 from blood to the brain and its basic
pharmacokinetic parameters were determined after intrave-
nous bolus injection in rats. The rats were anesthetized with
ketamine (90 mg/kg, i.p.) and xylazine (8 mg/kg, i.p.). Their
right jugular vein was exposed and cannulated with a PE-50
catheter. The rats were treated with buprenorphine
(0.12 mg/kg, i.p.) as an analgesic for 24 h after the cannu-
lation. The cannulated rats were allowed to recover until the
following day. A total of 25.36 μmol/kg of prodrug 3 were
administered into the cannulated jugular vein of the rats as a
0.5 ml bolus injection. The rats were pretreated with
intraperitoneal injection of carbidopa (30 mg/kg) (Orion
Pharma, Espoo, Finland) and entacapone (10 mg/kg) (Orion
Pharma, Espoo, Finland) 30 min before the administration of
3 in order to compare 3 to the standard L-dopa treatment of
Parkinson’s disease. Carbidopa and entacapone were used to
prevent the premature conversion of L-dopa to dopamine in
peripheral tissues. Carbidopa, entacapone and 3 were admin-
istered as suspensions containing a few drops of Tween 80
(Sigma-Aldrich Chemie GmbH, Steinheim, Germany) in
0.9% NaCl solution at pH 7.4. Plasma samples were collected
2, 5, 10, 30, 60 and 80 min after the prodrug administration
and the brain samples (a whole brain hemisphere) were col-
lected at 10, 30, 60 and 180 min after the prodrug adminis-
tration. The collected brain and plasma samples were
immediately frozen, and stored at −70°C. Rat brain and
plasma levels of 3 were analyzed after the sample preparation
by the HPLC-EC method described above. The brain levels
were corrected by the measured brain sample concentration
at the cerebral vascular space component, which was previ-
ously determined to be 0.0149 ml/g (13).
Data Analyses
The figures were created and all statistical analyses were
performed using nonlinear regression by GraphPad Prism
5.0 for Windows (GraphPad Software Inc., La Jolla, USA).
Statistical significance of differences between different
groups were analyzed using one-way ANOVA followed
by a two-tailed Dunnett’s test. The significance is reported
as ***p<0.001 or as *p<0.05 for all tests. The pharmacoki-
netic data analysis of 3 after intravenous administration was
carried out by WinNonlin® Professional 5.3 (Pharsight
Corparation, St. Louis, MO, USA) by using non-
compartmental model for the intravenous bolus injections,
and with 1/Yhat weighting.
RESULTS
In Vitro Studies of 1–3
The aqueous solubilities of 2 and 3 at pH 7.40 were deter-
mined to be higher than 10 mg/ml. The maximum solubil-
ities of 2 and 3, and the aqueous solubility of 1 were not
determined because of the small amount of the prodrugs
available. However, the apparent partition coefficients of 1–3
at pH 7.40 (log D) were determined to be negative, −2.25
0.12, −1.54 0.16 and −1.57 0.16 (mean sd, n=3) respec-
tively, indicating that the prodrugs are highly hydrophilic at pH
7.40. As a comparison, the log D of dopamine at pH 7.40 was
determined to be 0.20 0.11 (mean sd, n=3). In addition,
topological polar surface areas (TPSA) of 1–3, L-dopa and
dopamine were calculated with MOE-2011.10. TPSA of the
prodrugs was calculated to be 132.9 Å2, and that of L-dopa and
dopamine was 103.8 Å2 and 66.5 Å2, respectively (Table I).
The stabilities of 1–3 and their bioconversion to dopa-
mine were studied in aqueous buffer (pH 7.4), in 20% rat
Dopamine Release from 3 in Rat Striatum
After Intraperitoneal Injections
The ability of prodrug 3 to release dopamine in the rat
striatum after intraperitoneal administration was

Dopamine Prodrugs for LAT1
Table I Hydrolysis Rates of 1–3 in Phosphate Buffer (pH 7.4), in Rat Plasma, in Rat Brain and Rat Liver Homogenates (37°C) (T1/2, mean sd; n=3) and
Topological Polar Surface Areas (TPSA) for 1–3, L-dopa and Dopamine
Prodrug
Buffer (pH 7.4) (h)
75% rat plasma (h)
20% rat brain homogenate (h)
50% rat liver homogenate (h)
TPSA (Å2)
1
24.4 3.3
9.9 0.04
19.9 2.6
n.a
4.8 1.0
3.2 0.2
1.5 0.1
n.a
24.1 8.1
30.58a
28.4 5.5
n.a.
4.4 0.9
5.3 1.1
2.7 0.4
n.a.
132.9
132.9
132.9
103.8
66.5
2
3
L-dopa
Dopamine
n.a
n.a
n.a
n.a
a Mean (n=2)
brain homogenate, 50% rat liver homogenate and in 75%
rat plasma at 37°C (Table I). The chemical and enzymatic
degradation of 1–3 followed pseudo-first-order kinetics, and
the half-lives are presented in Table I. The release of dopa-
mine was also verified quantitatively by the HPLC-EC
method. The prodrugs appeared to be reasonably stable in
the aqueous buffer at pH 7.4 having half-lives of approxi-
mately 10 h (2) or longer (1 and 3). The half-lives of the
prodrugs in the rat plasma and in the rat liver homogenates
were notably shorter than those determined in the aqueous
buffer solution (Table I) suggesting that the prodrugs re-
leased the parent drug predominantly enzymatically.
According to the in vitro data, 1–3 were suitable for further
evaluation by the in situ and in vivo methods in rats.
product of [14C]-leucine decreased from 19.0 3.6 μl/s/g to
2.76 0.72 μl/s/g (mean sd, n=3), respectively. This obser-
vation i.e. about 85% inhibition in [14C]-leucine brain uptake
by 3, confirms our previous results which showed that meta-
substituted phenylalanine was a potential promoiety and car-
rier of anionic drugs into the brain (13). In the present study,
the promoiety has been modified also for cationic drugs, and
this modification was shown to be capable of retaining its
LAT1-targeting action. In addition, this study demonstrates
for the first time that hydrophilic prodrugs can also have high
binding affinity for LAT1. Moreover, the inhibition caused by
3 was significantly stronger than that observed after in situ rat
brain perfusion of an equivalent concentration of L-dopa
(Fig. 2) which was 66.26 1.39% (mean sd, n=3). This result
indicates that 3 has better possibilities to compete with the
endogenous LAT1-substrates than L-dopa when being taken
up into the brain by LAT1. This observation encouraged us to
study the brain uptake of 3 in more detail by using the in situ rat
brain perfusion to complement the [14C]-leucine brain uptake
inhibition determination. In contrast, prodrugs 1 and 2 did not
possess as strong affinity for LAT1 as 3, and they were less
potent than L-dopa in their ability to inhibit the brain uptake of
leucine (Fig. 2). 1 was able to decrease the PA product of [14C]-
leucine with 38.80 5.61% and 2 with 34.95 10.54%
Transporter-Mediated Brain Uptake of 1–3 by In Situ
Rat Brain Perfusion
Strong inhibition in the permeability-surface area (PA)
product of [14C]-leucine was apparent when perfusing
0.2 μCi/ml [14C]-leucine (i.e. 0.64 μM) in the presence of
100 μM concentration of the prodrug 3 (Fig. 2) as the PA
(mean
sd, n=3), respectively. This observation was not
consistent with our previous studies where the lysine derivative
of ketoprofen, a structurally-related prodrug to 1 and 2,
achieved almost 80% inhibition of leucine brain uptake at
the equivalent concentration (7) (Fig. 2).
We also studied the total amounts of the prodrugs 2 and
3 in the rat brain after the in situ rat brain perfusion to
confirm that the prodrugs do not only bind to LAT1 but are
also translocated across the BBB, because the entry of a
molecule into the brain does not necessarily occur even
though the prodrug might bind very tightly to the transport-
er (31). As illustrated in Fig. 3, only prodrug 3 possessed the
ability to cross the rat BBB and gain entry into the brain
tissue during the in situ rat brain perfusion. In contrast,
prodrug 2, which showed only low affinity for LAT1
(Fig. 2), was not shown to cross the rat BBB even at as high
as 600 μM concentration in the perfusion fluid (data not
Fig. 2 LAT1-binding data of prodrugs 100 μM 1–3, 100 μM L-dopa and
90 μM lysine-ketoprofen (7). The PA product of 0.2 μCi/ml [14C]-leucine,
19.0 3.6 μl/s/g, was reduced to 11.93 1.10 μl/s/g (38.8% inhibition) in
the presence of 1, to 12.68 μl/s/g (35.0% inhibition) in the presence of 2
and to 2.76 0.72 μl/s/g (85.5% inhibition) in the presence of 3 (means
sd, n=3). As a comparison, the PA product [14C]-leucine decreased to
6.41 0.26 μl/s/g (66.3% inhibition) in the presence of L-dopa and to
4.26 0.19 μl/s/g (79.3% inhibition) in the presence of lysine-ketoprofen
(7) (means
sd, n=3). The asterisk denotes a statistically significant
difference from the respective control (***P<0.001, one-way ANOVA,
followed by Dunnett t-test).

Peura et al.
decreased the brain uptake of 3 from about 0.307 pmol/
mg/min to 0.095 pmol/mg/min, respectively. This corre-
sponded to approximately 69% inhibition and was a clear
evidence of carrier-mediated brain uptake of 3. A similar
reduction of about 61% inhibition in the brain uptake of 3
was shown when co-perfusing the rat brain with 100 μM of
3 and 2 mM L-phenylalanine, and this further confirmed
the involvement of LAT1 in the brain uptake of 3. A
competition study with 2 mM L-phenylalanine was also con-
ducted at 5°C to demonstrate that 3 was being taken up into
the brain by a carrier-mediated manner also at 5°C (Fig. 4). A
significant decrease in the brain uptake of 3 was observed when
3 was co-perfused with 2 mM L-phenylalanine at a reduced
temperature, when compared to the results of the same ex-
periment conducted at 37°C. However, there still remains a
non-saturable, cold-insensitive part in the brain uptake of 3.
Fig. 3 Kinetics of the rat brain uptake of 3. Km and Vmax are 227.9
52.0 μM and 0.99 0.09 pmol/mg/min (mean sd, n=3) respectively.
Brain Uptake and Basic Pharmacokinetic Parameters
of 3 After Intravenous Administration
shown). The poor brain uptake of 2 may indicate the fact
that very high affinity to LAT1 is needed for the brain
uptake of LAT1-targeted prodrugs. Since 1 and 2 lack
affinity for LAT1 and 2 was not taken up into the brain at
detectable amounts, we decided to perform more compre-
hensive in vivo brain uptake studies only for 3.
Roles of reduced temperature and simultaneous LAT1
inhibition in the rat brain uptake of 3 at 100 μM concen-
tration were investigated by using cold-perfusion (5°C) and
2 mM L-phenylalanine as a competitive endogenous LAT1
substrate (Fig. 4) to demonstrate the role of LAT1 in the
brain uptake of 3. A lower perfusion fluid temperature
Encouraged by the in situ results, the brain uptake of 3 in vivo
was determined after intravenous administration in rats. A
total of 25.35 μmol/kg solution of 3 was administered into
the right jugular vein of the rats. The results indicate that
3 was able to cross the BBB in vivo in the presence of
competitive endogenous substrates (Fig. 5). The area under
the curve for 3 during the 180 min after the administration
in the brain (AUCbrain) was 13.39 nmol*min/g and in
plasma (AUCplasma) 1,481 nmol*min/ml (Table II). The
half-life of 3 in plasma during the distribution phase was
as short as 3.10 min indicating that the prodrug 3 is excreted
Fig. 4 Rat brain uptake of 3 after cold perfusion and co-perfusion with 2 mM L-phenylalanine. The uptake of 3 with 100 μM concentration significantly
decreased from 0.307 0.02 pmol/mg/min to 0.095 0.01 pmol/mg/min (mean sd, n=3) (69.1% inhibition) in cold perfusion medium (5°C). A
comparable inhibition (60.8%) was observed during the co-perfusion of 100 μM of 3 and 2 mM of L-phenylalanine. When conducting the study with 2 mM
L-phenylalanine at 5°C the uptake of 3 inhibited 78.6%. There was a statistically significant difference between the column 100 μM 3 (37°C) and the other
columns (*** P<0.001 one-way ANOVA, followed by Dunnett t-test). There was also a significant difference between columns 100μM 3+2 mM Phe
(37°C) and 100μM 3+2 mM Phe (5°C) (* P<0.05 one-way ANOVA, followed by Dunnett t-test).

Dopamine Prodrugs for LAT1
Fig. 5 Mean plasma (a) and
brain (b) concentration-time pro-
files of 3 after intravenous bolus
injection (25.35 μmol/kg) to rats
(mean SD, n=4–5).
rapidly as a hydrophilic compound or distributed rapidly
into the rest of the rat body (Table II).
the therapeutically relevant brain uptake of a drug candi-
date is more of an exception than the rule (2). The CNS-
acting drugs have been traditionally designed to cross the
BBB by simple passive transcellular diffusion. Thus, for
decades only relatively lipophilic and low molecular weight
compounds have been thought to have the ideal physico-
chemical properties to allow the drugs to reach the brain.
However, the novel therapeutically active drug candidates
tend to have higher molecular mass (32) and to possess less
favourable physicochemical characteristics for BBB-
permeation (3) leading to the urgent need to develop better
drug brain delivery strategies.
In recent years, transporter-mediated drug brain delivery
has shown substantial promise as a strategy to enhance
transmembrane permeability of polar compounds, which
are not otherwise able to cross the cell membranes (4).
Prodrugs in which the parent drug molecule has been con-
jugated in a bioreversible manner with an endogenous sub-
strate of a transporter expressed at the BBB have been
Release of Dopamine from 3 in the Rat Striatum
After Intraperitoneal Administration
In addition to the brain uptake studies in vivo, we evaluated
whether 3 was able to elevate the striatal dopamine concen-
trations in the rats. In this study, we compared the dopa-
mine concentrations in striatal samples after intraperitoneal
administration of the prodrug 3 with those induced after
intraperitoneal administration of an equivalent dose of L-
dopa. A dose of 50.71 μmol/kg of 3 or L-dopa suspension
was administered intraperitoneally into the rats, which were
pretreated with AADC- and COMT-inhibitors (i.e., carbi-
dopa (30 mg/kg) and entacapone (10 mg/kg), respectively)
to prevent peripheral metabolism of L-dopa to dopamine.
The results indicate that 3 does not increase the striatal
dopamine concentrations in vivo, whereas the striatal con-
centrations of dopamine were significantly increased 60 min
after the L-dopa administration (Fig. 6).
DISCUSSION
The brain distribution of polar or high molecular weight
drugs is effectively limited by the BBB, leading to a fact that
Table II The Basic Pharmacokinetic Parameters of 3 After Intravenous
Bolus Injection into Rats (mean SD, n=4)
Parameters
Prodrug 3
Dose (μmol/kg)
25.35
Fig. 6 Effect of 3 and L-dopa on rat striatal dopamine concentrations after
equivalent intraperitoneal doses (50.71 μmol/kg). The rats were pre-
treated with carbidopa (30 mg/kg) and entacapone (10 mg/kg) 30 min
before the administration of L-dopa or 3. The asterisks denote a statistically
significant difference from the respective control (***P<0.001, one-way
ANOVA, followed by Dunnett t-test).
AUCbrain (nmol*min/g) (10–180 min)
AUCplasma (nmol*min/ml) (2–180 min)
CL (ml/min/kg)
13.39
1481
13.52 3.92
3.10 0.60
T
1/2 (min)

Peura et al.
reported in many prior studies with the aim being to en-
hance the brain uptake of the parent drug (4). The most
suitable transporters at the BBB for CNS-targeting of drugs
by the prodrug approach seem to be LAT1 and GluT1.
SVCT1, which is mainly expressed at the choroid plexus,
could also be utilized as a drug carrier to the brain.
none of the pharmacokinetic studies has compared a pro-
drug with L-dopa combined with peripheral inhibitors of L-
dopa metabolizing enzymes, e.g. carbidopa and entaca-
pone, the mainstay therapy in treating the symptoms
of Parkinson’s disease. Another published study with a
transporter-mediated dopamine prodrug utilized glutathi-
one transporters at the BBB (24). This dopamine derivative
has been shown to possess high affinity for the glutathione
transporter in a MDCK cell monolayer and to be able
to release dopamine in brain homogenate in vitro but the
in vivo brain uptake capability of that derivative has not
been clarified.
Tyrosine derivatives of e.g. nipecotic acid, phosphonofor-
mate, ketoprofen and valproic acid have been shown
to possess therapeutically significant affinity for LAT1
(8,11,13,33). Most studies of LAT1-targeted prodrugs have
focused only on the measurement of the LAT1-affinity of
the prodrugs, and the actual brain uptake of those com-
pounds has rarely been evaluated. Only the LAT1-targeted
ketoprofen and valproic acid derivatives have been shown to
be taken up into the brain in a LAT1-mediated manner in
the in situ rat brain perfusion in addition to binding with
high affinity to that transporter (7,8,13). Furthermore, pro-
drugs substituted at the meta-position of phenylalanine have
been shown to bind to LAT1 with brain uptake significantly
superior to the corresponding para-substituted prodrugs
(13). Those meta-substituted phenylalanine prodrugs with
an unnatural amino acid promoiety are most probably able
to compete with the endogenous amino acids of the systemic
circulation while utilizing LAT1, because of their high
LAT1-affinity, which is even greater than that of the
endogenous substrates.
In addition to LAT1-targeting, some drugs such as do-
pamine, L-dopa, ketoprofen and 7-chlorokyurenic acid have
been conjugated with a glucose promoiety (5,6,22,23,34) in
order to modify those structures to make them into sub-
strates for GluT1. Those derivatives possess affinity for
GluT1 but there is no clear evidence that these prodrugs
had been taken up from blood into the brain after in vivo
administration. Nipecotic acid, kynurenic acid and diclofe-
namic acid conjugated with ascorbic acid have been shown
to act as substrates for SVCT1 and they are believed to be
taken up into the brain via that transporter (35,36). How-
ever, none of the previous reports has so far investigated the
brain uptake of the SVCT1-targeted prodrugs in vivo.
Though there are several transporters which can be consid-
ered in brain drug targeting, LAT1 is viewed as the most
attractive transporter for carrier-mediated brain drug delivery
approaches (4).
In this study, we have introduced one promising prodrug
strategy, the utilization of LAT1 as a prodrug carrier at the
BBB, which could be capable of enhancing the penetration
of polar cationic drugs into the brain. Dopamine prodrugs
1–3 were designed, synthesized and characterized, their
physicochemical properties were studied in vitro, and their
ability to utilize LAT1 was evaluated with the in situ rat
brain perfusion technique. In addition, the brain uptake of 3
was determined after an intravenous bolus injection in rats,
and the release of dopamine from 3 in the rat striatal tissue
was determined after an intraperitoneal bolus injection.
In the present study, the dopamine prodrugs 1–3 showed
good aqueous solubility at pH 7.40, and their apparent log
D -values at the same pH were negative. In addition, the
topological polar surface areas (TPSA) of the prodrugs were
much higher than the proposed maximum PSA (<90 Å2) for
molecules that can cross the BBB by passive diffusion (37),
indicating that the prodrugs are not very likely penetrated
into the brain by passive diffusion. After evaluation of the in
vitro data, it was assumed that 1–3 permeate through the
biological membranes via transporters or paracellularly at
physiological pH (38). However, it is unlikely that the para-
cellular pathway takes any major role in the brain perme-
ation of polar solutes (1). On the other hand, it is worthwhile
to bear in mind that the hydro- or lipophilicity of a drug
molecule is not the only parameter that could be used as a
surrogate for drug brain penetration (3) even though the log
D -values of the majority of CNS-acting drugs were shown
to be 0–4 (37). It is also worthwhile mentioning that lipo-
philic drugs are usually potential substrates of efflux trans-
porters, extruding neurotherapeutics from the brain back to
the bloodstream (3), and lipophilic derivatives are more
likely to possess high unspecific brain tissue binding and
plasma protein binding (37). Both of these properties, i.e.
being a substrate for an efflux transporter and binding to the
brain tissue and plasma proteins, can decrease the unbound
brain concentration of an active drug compound. In addi-
tion, lipophilic compounds possess a high volume of distri-
bution as they are able to cross the biological membranes in
general, and the specific fraction taken up into the brain
might be relatively low. Thus, the influx transporter utilizing
In addition to the LAT1-targeted amino acid “pseudo-
nutrient” prodrug of dopamine, L-dopa, a series of glycosyl
derivatives of dopamine has been previously investigated in
order to determine the ability of the glucose or galactose
promoiety to enhance the brain uptake of dopamine (23).
The galactosylated derivatives of dopamine have been
shown to bind to GluT1 and to be active in models of
Parkinson’s disease in rats. However, the in vivo pharmaco-
kinetic properties of the GluT1-targeted dopamine deriva-
tives have not been studied with all these prodrugs, and

Dopamine Prodrugs for LAT1
prodrug strategies, in which the parent molecules are rela-
tively hydrophilic and the prodrug specific for certain influx
transporter specific, are very attractive and promising in
drug targeting into the brain.
with the concentration dependent brain uptake of 3 also
further confirmed the proposal that 3 is being taken up into
the brain in a carrier-mediated manner. The role of non-
saturable brain transport mode of 3 was investigated by
using cold-perfusion (5°C) and by competitively inhibiting
LAT1 with co-perfusion of 2 mM L-phenylalanine (Fig. 4).
The lowered temperature of the perfusion fluid decreased
the brain uptake of the prodrug 3 significantly. This signif-
icant decrease in the brain uptake is clear evidence of
carrier-mediated passage of 3 across the BBB. A similar
decrease was shown after the co-perfusion with 2 mM L-
phenylalanine. However, in both experiments the uptake of
3 was not totally inhibited. Thus, we conducted an addi-
tional competitive inhibition study with 2 mM L-phenylala-
nine and 3 at 5°C and found that the brain uptake was
further decreased by about 20%. This observation may
indicate that the activity of LAT1 is not totally inhibited
even in a cold perfusion medium. Alternatively, some part of
3 may be transferred in our study through the BBB by a
non-saturable, cold-insensitive manner, which is the part of
3 that crossed the BBB most probably by passive diffusion.
On the other hand, the physicochemical properties of 3, i.e.
high aqueous solubility (over 10 mg/ml), low log D (−1.57)
and high TPSA (132.9 Å2), do not indicate that 3 is likely to
be taken up by passive diffusion.
To summarize the in vitro and in situ data, it seems that
meta-substituted phenylalanine can be an efficient and po-
tential promoiety for improving the brain uptake of small
molecular weight cationic drugs that are not readily trans-
ported into the brain by passive diffusion. 3 was taken up
into the rat brain in a carrier-mediated manner, rapidly and
at significant amounts via LAT1 whereas the alkyl amino
acid derivatives (1 and 2) did not possess sufficiently high
affinity for LAT1 and, hence, were not taken up into the
brain in appreciable amounts. Thus, it seemed reasonable to
perform more comprehensive in vivo brain uptake studies
with 3.
Although the in vitro and in situ studies indicated that
3 was a promising LAT1-transported prodrug, it did not
evoke desirable pharmacological effects i.e. it could not
elevate the striatal dopamine levels in the in vivo situation
(Fig. 6), even though it was taken up into the rat brain after
intravenous administration (Fig. 5). As a hydrophilic com-
pound, 3 may not reach the brain parenchyma in order to
release dopamine enzymatically. Most probably 3 may be
cleared out from the brain too rapidly to release dopamine
in pharmacologically significant amounts. However, LAT1
and other amino acid transporters are known to be
expressed also in the glia cells and astrocytes (41,42), and
thus the rapid CNS elimination after intravenous adminis-
tration was unexpected because those amino acid transport-
ers would be expected to carry 3 into the cells and trap the
prodrug or the active molecule within the cells. As a
The enzymatic hydrolysis of the prodrugs released dopa-
mine (Table I) indicating that the molecules can, indeed, act
as prodrugs. Enzymatic hydrolysis rates of 1–3 in rat plasma
appeared to be relatively high when compared with those in
the rat liver and brain homogenates. This may be due to the
fact that the plasma hydrolysis studies were conducted in
75% rat plasma whereas the rat liver and brain tissue
homogenates were diluted by isotonic buffer (pH 7.4) to
concentrations of 50% or 20%, respectively. Additionally,
the differences between the hydrolysis rates of the prodrugs
in rat brain and liver homogenates probably correspond
also to the distinct tissue distribution of the prodrug-
hydrolyzing esterases. Amide prodrugs in general may be
too stable for releasing the parent drug but the present
dopamine prodrugs seem to belong to those amide prodrugs
which are capable of releasing the parent drug, such as
lisdextroamphetamine (39), a prodrug of dextroamphet-
amine, lysine-ketoprofen, a prodrug of ketoprofen (7), and
9-[P-(N,N-dipropylsulfamide)] benzoylamino-1,2,3,4-4H-
acridine, a prodrug of tacrine (40).
The inhibition of [14C]-leucine brain uptake in the in situ
rat brain perfusion evoked by 3 was significantly higher than
that of 1 and 2 and the corresponding inhibition achieved
by L-dopa (Fig. 2). The potency of meta-substituted phenyl-
alanine promoiety as a LAT1-targeted drug brain carrier
was in agreement with our earlier findings (13). In this
paper, we have shown that the promoiety could also be
modified as a carboxylic acid derivative for cationic drug
candidates with poor brain penetration. Surprisingly, 1 and
2 did not possess as high affinity for LAT1 as shown by the
lysine-ketoprofen (7), which has the promoiety that structur-
ally resembles the promoieties of 1 and 2. Therefore, the
alkyl promoiety in general does not seem to act as favorable
a LAT1-substrate as its aromatic counterpart, and the alkyl
amino acid promoiety cannot be generalized to be able to
function as a high-affinity LAT1-promoiety for every par-
ent drug. However, the whole prodrug molecule can act
as a high-affinity LAT1 substrate as shown by the lysine-
ketoprofen prodrug.
The results of the in situ uptake studies showed that the
penetration of 3 across the rat BBB was significantly better
than that of 2 (Fig. 3). In fact, 2 did not cross the rat BBB at
concentrations above the lower limit of quantification (i.e.
0.4 nmol/brain hemisphere). The reason why 2, and most
probably also 1, were not taken up into the brain in detect-
able amounts is most probably due to their low affinity for
the transporter. This was the first time when the transloca-
tion of the LAT1 prodrugs across the BBB was shown to be
dependent on their high affinity to LAT1. This observation

Peura et al.
consequence, the time period for 3 to release dopamine at
the site of action may be too short to achieve any favorable
elevating effect on the striatal dopamine levels. In order to
achieve relevant pharmacological actions, the CNS-acting
drug must also gain access to its site of action within the
brain tissue after its passage into brain.
penetrating drugs, and the attachment of phenylalanine to
the parent drug via an amide bond from the meta-position
of its aromatic ring is a feasible promoiety for LAT1-
targeting of cationic drugs. However, further work needs
to be conducted to establish whether this promoiety can be
used as a drug carrier to the brain also in conjunction with
other cationic and poorly brain penetrating parent drugs.
Higher amounts of L-dopa were taken up into the brain
than 3 and this drug was able to elevate the striatal dopa-
mine supply even though its affinity for LAT1 was relatively
weak (Figs. 2 and 6). It is believed that the brain uptake of L-
dopa occurs in both passive and carrier-mediated manners
(43). That may explain why L-dopa was superior in the in
vivo situation when compared to 3. The part of L-dopa being
taken up into the brain by passive diffusion is considered to
increase when it is metabolized into 3-O-met-dopa. This
metabolite binds to LAT1 with much higher affinity than L-
dopa itself, and acts as a competitive inhibitor for L-dopa
transport across the BBB (44). This significantly higher
LAT1-binding affinity of 3-O-met-dopa makes the compet-
ing hypothesis more relevant even though the plasma levels
of 3-O-met-dopa are normally low.
ACKNOWLEDGMENTS AND DISCLOSURES
Lauri Peura, Kalle Malmioja, Jarkko Rautio and Krista
Laine share equal contribution to this work.
The authors would like to express their profound grati-
tude to laboratory technicians Helly Rissanen and Jaana
Leskinen for their skillful assistance. Marko Lehtonen,
M.Sc., is also acknowledged for his skillful help in the
HPLC-EC-analytics, Tiina Kääriäinen, Ph.D., for her skill-
ful help in the animal studies and Ewen MacDonald, Ph.D.,
for refining the English of this paper. We also thank Henna
Härkönen, M.Sc., for the polar surface area calculations.
The work was financially supported by the Graduate School
of Pharmaceutical Research, the Academy of Finland
(# 132637), the Orion-Farmos foundation, the Finnish
Pharmaceutical Society, the Finnish Parkinson Foundation
and the Emil Aaltonen Foundation.
CONCLUSIONS
In the present study, we explored amino acid prodrugs (1–3)
of dopamine and found a new promoiety to carry this
cationic drug molecule into the brain via the LAT1 trans-
porter. Firstly, we showed that these prodrugs possess affin-
ity for LAT1 at the rat BBB during the in situ rat brain
perfusion and they evoked a sustained release of dopamine
in in vitro stability studies, thus fulfilling the specifications
required for tissue-targeted prodrugs. Secondly, only the
meta-substituted phenylalanine promoiety attached to do-
pamine via an amide bond (prodrug 3) exhibited adequate
affinity for LAT1 for carrier-mediated brain uptake. The
aspartic acid (prodrug 1) and adipidic acid (prodrug 2)
prodrugs did not possess as favourable properties as LAT1
substrates as 3 and this was the first time when the translo-
cation of the LAT1 prodrugs across the BBB was shown to
be dependent on their high affinity to LAT1. In addition to
the in situ studies, we showed that prodrug 3 was also taken
up into the rat brain after its intravenous administration in
vivo in rats. Unfortunately, we still could not find enhanced
dopamine concentrations in the rat striata after intraperito-
neal administration of the prodrug 3 when compared with
corresponding equimolar dose of L-dopa. This may be be-
cause 3 is too quickly eliminated from the brain due to its
high polarity. Our in vivo data also highlighted the problem
that a prodrug that shows potential in vitro and in situ, may
not necessarily work well in in vivo studies.
REFERENCES
1. Begley DJ. Delivery of therapeutic agents to the central nervous
system: the problems and the possibilities. Pharmacol Ther.
2004;104(1):29–45.
2. Pardridge WM. Crossing the blood–brain barrier: are we getting it
right? Drug Discov Today. 2001;6(1):1–2.
3. Wager TT, Hou X, Verhoest PR, Villalobos A. Moving beyond
rules: the development of a central nervous system multiparameter
optimization (CNS MPO) approach to enable alignment of drug-
like properties. ACS Chem Neurosci. 2010;1(6):435–49.
4. Pavan B, Dalpiaz A. Prodrugs and endogenous transporters: are
they suitable tools for drug targeting into the central nervous
system? Curr Pharm Des. 2011;17(32):3560–76.
5. Gynther M, Ropponen J, Laine K, Leppänen J, Haapakoski P,
Peura L, et al. Glucose promoiety enables glucose transporter
mediated brain uptake of ketoprofen and indomethacin prodrugs
in rats. J Med Chem. 2009;52(10):3348–53.
6. Bonina F, Puglia C, Rimoli MG, Melisi D, Boatto G, Nieddu M, et
al. Glycosyl derivatives of dopamine and L-dopa as anti-Parkinson
prodrugs: Synthesis, pharmacological activity and in vitro stability
studies. J Drug Target. 2003;11(1):25–36.
7. Gynther M, Jalkanen A, Lehtonen M, Forsberg M, Laine K,
Ropponen J, et al. Brain uptake of ketoprofen-lysine prodrug in
rats. Int J Pharm. 2010;399(1–2):121–8.
8. Gynther M, Laine K, Ropponen J, Leppänen J, Mannila A,
Nevalainen T, et al. Large neutral amino acid transporter enables
brain drug delivery via prodrugs. J Med Chem. 2008;51(4):932–6.
9. Hokari M, Wu H, Schwarcz R, Smith QR. Facilitated brain uptake
of 4-chlorokynurenine and conversion to 7-chlorokynurenic acid.
Neuroreport. 1997;8(1):15–8.
Taken together, these findings imply that LAT1 is a
promising target for brain drug delivery of poorly brain

Dopamine Prodrugs for LAT1
10. Killian DM, Chikhale PJ. A bioreversible prodrug approach
designed to shift mechanism of brain uptake for amino-acid-
containing anticancer agents. J Neurochem. 2001;76(4):966–74.
11. Bonina FP, Arenare L, Palagiano F, Saija A, Nava F, Trombetta
D, et al. Synthesis, stability, and pharmacological evaluation of
nipecotic acid prodrugs. J Pharm Sci. 1999;88(5):561–7.
12. Balakrishnan A, Jain-Vakkalagadda B, Yang C, Pal D, Mitra AK.
Carrier mediated uptake of L-tyrosine and its competitive inhibi-
tion by model tyrosine linked compounds in a rabbit corneal cell
line (SIRC)—strategy for the design of transporter/receptor tar-
geted prodrugs. Int J Pharm. 2002;247(1–2):115–25.
13. Peura L, Malmioja K, Laine K, Leppänen J, Gynther M, Isotalo
A, et al. Large amino acid transporter 1 (LAT1) prodrugs of
valproic acid: new prodrug design ideas for central nervous system
delivery. Mol Pharm. 2011;8(5):1857–66.
14. Gomes P, Soares-Da-Silva P. L-DOPA transport properties in an
immortalised cell line of rat capillary cerebral endothelial cells,
RBE 4. Brain Res. 1999;829(1–2):143–50.
28. Ertl P, Rohde B, Selzer P. Fast calculation of molecular polar
surface area as a sum of fragment-based contributions and its
application to the prediction of drug transport properties. J Med
Chem. 2000;43(20):3714–7.
29. Takasato Y, Rapoport SI, Smith QR. An in situ brain perfusion
technique to study cerebrovascular transport in the rat. Am J
Physiol. 1984;247(3 Pt 2):H484–93.
30. Smith QR, Allen DD. Methods in molecular medicine: the blood-
brain barrier: biology and research protocols. In: Sukriti Nag,
editors. Totowa, NJ: Humana Press Inc.; 2003. p. 209–218.
31. Krupka RM. Expression of substrate specificity in facilitated trans-
port systems. JMB. 1990;117:69–78.
32. Liu X, Smith BJ, Chen C, Callegari E, Becker SL, Chen X, et al.
Use of a physiologically based pharmacokinetic model to study the
time to reach brain equilibrium: an experimental analysis of the
role of blood-brain barrier permeability, plasma protein binding,
and brain tissue binding. J Pharmacol Exp Ther. 2005;313
(3):1254–62.
15. Nutt JG. Pharmacokinetics and pharmacodynamics of levodopa.
Mov Disord. 2008;23(S3):S580–4.
16. Olanow C, Obeso J, Stocchi F. Drug insight: continuous dopami-
nergic stimulation in the treatment of Parkinson’s disease. Nat Clin
Pract Neurol. 2006;2(7):382–92.
17. Chemuturi NV, Donovan MD. Role of organic cation transporters
in dopamine uptake across olfactory and nasal respiratory tissues.
Mol Pharm. 2007;4(6):936–42.
18. Borgman RJ, McPhillips JJ, Stitzel RE, Goodman IJ. Synthesis and
pharmacology of centrally acting dopamine derivatives and analogs
in relation to Parkinson’s disease. J Med Chem. 1973;16(6):630–3.
19. Bodor N, Farag HH, Brewster III ME. Site-specific, sustained
release of drugs to the brain. Science. 1981;214(4527):1370–2.
20. Bodor N, Farag HH. Improved delivery through biological mem-
branes. 11. A redox chemical drug-delivery system and its use for
brain-specific delivery of phenylethylamine. J Med Chem. 1983;26
(3):313–8.
33. Walker I, Nicholls D, Irwin WJ, Freeman S. Drug delivery via
active transport at the blood-brain barrier: affinity of a prodrug of
phosphonoformate for the large amino acid transporter. Int J
Pharm. 1994;104(2):157–67.
34. Ruocco LA, Viggiano D, Viggiano A, Abignente E, Rimoli MG,
Melisi D, et al. Galactosylated dopamine enters into the brain,
blocks the mesocorticolimbic system and modulates activity and
scanning time in Naples high excitability rats. Neuroscience.
2008;152(1):234–44.
35. Dalpiaz A, Pavan B, Scaglianti M, Vitali F, Bortolotti F, Biondi C,
et al. Transporter-mediated effects of diclofenamic acid and its
ascorbyl pro-drug in the in vivo neurotropic activity of ascorbyl
nipecotic acid conjugate. J Pharm Sci. 2004;93(1):78–85.
36. Manfredini S, Pavan B, Vertuani S, Scaglianti M, Compagnone
D, Biondi C, et al. Design, synthesis and activity of ascorbic acid
prodrugs of nipecotic, kynurenic and diclophenamic acids, liable to
increase neurotropic activity. J Med Chem. 2002;45(3):559–62.
37. van de Waterbeemd H, Smith DA, Jones BC. Lipophilicity in PK
design: methyl, ethyl, futile. J Comput Aided Mol Des. 2001;15
(3):273–86.
21. Simpkins JW, Bodor N, Enz A. Direct evidence for brain-specific
release of dopamine from a redox delivery system. J Pharm Sci.
1985;74(10):1033–6.
22. Carelli V, Liberatore F, Scipione L, Impicciatore M, Barocelli E,
Cardellini M, et al. New systems for the specific delivery and
sustained release of dopamine to the brain. J Control Release.
1996;42(3):209–16.
23. Denora N, Laquintana V, Lopedota A, Serra M, Dazzi L, Biggio
G, et al. Novel L-Dopa and dopamine prodrugs containing a 2-
phenyl-imidazopyridine moiety. Pharm Res. 2007;24(7):1309–24.
24. Fernández C, Nieto O, Fontenla JA, Rivas E, De Ceballos ML,
Fernandez-Mayoralas A. Synthesis of glycosyl derivatives as dopa-
mine prodrugs: interaction with glucose carrier GLUT-1. Org
Biomol Chem. 2003;1(5):767–71.
25. Fernández C, Nieto O, Rivas E, Montenegro G, Fontenla JA,
Fernández-Mayoralas A. Synthesis and biological studies of glyco-
syl dopamine derivatives as potential antiparkinsonian agents.
Carbohydr Res. 2000;327(4):353–65.
26. More SS, Vince R. Design, synthesis and biological evaluation of
glutathione peptidomimetics as components of anti-Parkinson pro-
drugs. J Med Chem. 2008;51(15):4581–8.
38. Levin VA. Relationship of octanol/water partition coefficient and
molecular weight to rat brain capillary permeability. J Med Chem.
1980;23(6):682–4.
39. Elia J, Easley C, Kirkpatrick P. Lisdexamfetamine dimesylate. Nat
Rev Drug Discov. 2007;6(5):343–4.
40. Gong T, Huang Y, Zang Z, L L. Synthesis and characterization of
9-[P-(N, N-dipropylsulfamide)] benzoylamino-1,2,3,4–4H-acri-
dine—a potential prodrug for the CNS delivery of tacrine. J Drug
Target. 2004;12(3):177–82.
41. Thurlow RJ, Hill DR, Woodruff GN. Comparison of the uptake of
[ 3H]-gabapentin with the uptake of L-[ 3H]-leucine into rat brain
synaptosomes. Br J Pharmacol. 1996;118(3):449–56.
42. Su T, Lunney E, Campbell G, Oxender DL. Transport of gabapentin,
a γ-amino acid drug, by system L α-amino acid transporters: a com-
parative study in astrocytes, synaptosomes, and CHO cells. J Neuro-
chem. 1995;64(5):2125–31.
43. Nutt JG, Fellman JH. Pharmacokinetics of levodopa. Clin Neuro-
pharmacol. 1984;7(1):35–49.
27. Dent III WH, Erickson WR, Fields SC, Parker MH, Tromiczak EG.
9-BBN: an amino acid protecting group for functionalization of amino
acid side chains in organic solvents. Org Lett. 2002;4(8):1249–51.
44. Gomes P, Soares-da-Silva P. Interaction between L-DOPA and 3-
O-methyl-l-DOPA for transport in immortalised rat capillary ce-
rebral endothelial cells. Neuropharmacology. 1999;38(9):1371–80.