Shuttle-Cargo fusion molecules of transport peptides and the hD 2/3 receptor antagonist fallypride: A feasible approach to preserve ligand-receptor binding?

Article abstract of DOI:10.1021/jm5004123

To determine if the conjugation of a small receptor ligand to a peptidic carrier to potentially facilitate transport across the blood-brain barrier (BBB) by molecular Trojan horse transcytosis is feasible, we synthesized several transport peptide-fallypride fusion molecules as model systems and determined their binding affinities to the hD₂ receptor. Although they were affected by conjugation, the binding affinities were found to be still in the nanomolar range (between 1.5 and 64.2 nM). In addition, homology modeling of the receptor and docking studies for the most potent compounds were performed, elucidating the binding modes of the fusion molecules and the structure elements contributing to the observed high receptor binding. Furthermore, no interaction between the hybrid compounds and P-gp, the main excretory transporter of the BBB, was found. From these results, it can be inferred that the approach to deliver small neuroreceptor ligands across the BBB by transport peptide carriers is feasible.

Full text of DOI:10.1021/jm5004123

Article
pubs.acs.org/jmc
Shuttle−Cargo Fusion Molecules of Transport Peptides and the hD2/3
Receptor Antagonist Fallypride: A Feasible Approach To Preserve
Ligand−Receptor Binding?
Carmen Wangler,*,†,‡,§ Shafinaz Chowdhury,∥ Georg Hofner,⊥ Petia Djurova,# Enrico O. Purisima,∇
̈
̈
Peter Bartenstein,§ Bjorn Wangler,○ Gert Fricker,# Klaus T. Wanner,⊥ and Ralf Schirrmacher†
̈
̈
†McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University, Montreal H3A 2B4, Canada
‡Biomedical Chemistry, Department of Clinical Radiology and Nuclear Medicine, Medical Faculty Mannheim of Heidelberg
University, Mannheim 68167, Germany
§Department of Nuclear Medicine, University Hospital Munich, Ludwig-Maximilians-University Munich, Munich 81377, Germany
∥Lady Davis Institute, Jewish General Hospital, McGill University, Montreal H3T 1E2, Canada
⊥Pharmaceutical Chemistry, Ludwig-Maximilians-University Munich, Munich 81377, Germany
#Institute of Pharmacy and Molecular Biotechnology, University of Heidelberg, Heidelberg 69120, Germany
∇National Research Council of Canada, Montreal H4P 2R2, Canada
○Molecular Imaging and Radiochemistry, Department of Clinical Radiology and Nuclear Medicine, Medical Faculty Mannheim of
Heidelberg University, Mannheim 68167, Germany
S
* Supporting Information
ABSTRACT: To determine if the conjugation of a small receptor ligand to a
peptidic carrier to potentially facilitate transport across the blood−brain barrier
(BBB) by “molecular Trojan horse” transcytosis is feasible, we synthesized several
transport peptide−fallypride fusion molecules as model systems and determined
their binding affinities to the hD2 receptor. Although they were affected by
conjugation, the binding affinities were found to be still in the nanomolar range
(between 1.5 and 64.2 nM). In addition, homology modeling of the receptor and
docking studies for the most potent compounds were performed, elucidating the
binding modes of the fusion molecules and the structure elements contributing to
the observed high receptor binding. Furthermore, no interaction between the
hybrid compounds and P-gp, the main excretory transporter of the BBB, was
found. From these results, it can be inferred that the approach to deliver small
neuroreceptor ligands across the BBB by transport peptide carriers is feasible.
Following this approach, different substance classes have
been used as shuttle vectors. So far, mostly large bioactive
molecules such as antibodies directed against the transferrin
receptor (TfR) (TfRmAbs, such as OX26) or the insulin
receptor (IRmAbs), other proteins such as melanotransferrin,
receptor associated protein, and CRM197 (a nontoxic mutant
of diphtheria toxin), or liposomes functionalized with peptides
or proteins have been applied.2,3,7−9 However, although
exhibiting a highly specific accumulation in the target, the use
of large molecules as drug carriers is unfavorable with regard to
their drug pharmacokinetics, clearance, immunogenicity, and
bioavailability. Furthermore, a fusion molecule composed of a
large carrier and a small neuroreceptor-binding cargo is
commonly regarded as being unlikely to exhibit a preserved
affinity of the ligand to its intended target.
INTRODUCTION
■
Functional diagnosis and treatment of diseases of the central
nervous system (CNS) are still limited by the blood−brain
barrier (BBB), a combination of endothelial tight junctions and
active drug transporters, preventing the accumulation of
hydrophilic as well as lipophilic drugs in the brain parenchyma
(Figure 1A,B).1 While exercising the very important physio-
logical role of protecting the CNS against potentially toxic
compounds, it limits diagnostic and therapeutic opportunities
in the case of disease. Thus, many efforts have been made
within recent decades to overcome this barrier.2−6 Most
strategies use the so-called molecular Trojan horse approach,
which makes use of drug conjugates combining a shuttle (a
substance that is able to cross the BBB by transcytosis; Figure
1C,D) and a cargo (a diagnostic or therapeutic drug that is
intended to be taken up into the CNS and is not able to cross
the BBB on its own; Figure 1E).
Received: March 16, 2014
Published: April 30, 2014
© 2014 American Chemical Society
4368
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
Figure 1. Schematic depiction of the different mechanisms of BBB function and transcytosis: hydrophilic drugs cannot enter the endothelium (A),
some lipophilic drugs can enter the endothelium but get externalized by active drug transport (B), BBB crossing of receptor-binding substances such
as transferrin and insulin by receptor-mediated transcytosis (C), unspecific BBB crossing of positively charged substances by adsorptive transcytosis
(D), and specific or nonspecific BBB crossing of hydrophilic (blue) or lipophilic (orange) drugs using the molecular Trojan horse approach by
receptor-mediated or adsorptive transcytosis (E).
Therefore, the use of smaller peptides as shuttles across the
BBB has been proposed. Peptides that are, in principle,
applicable for cargo delivery to the brain are neuropeptides that
are transported across the BBB by endogenous transporters
(Figure 1C), such as transferrin receptor binding peptides
(transported via the TfR),10 neurotensin (transported via
neurotensin receptors NTS1, NTS2, and NTS3),11,12 chol-
ecystokinin (transported via cholecystokinin receptors CCKR1
and CCKR2),12,13 neuropeptide Y (transported mostly via
receptors Y1, Y2, Y4, and Y5),12,14 and Substance P
(transported via the neurokinin receptor)15 as well as
AngioPep-2 and other peptide derivatives that are transported
via the low-density lipoprotein related receptor, LRP1.16−20
Furthermore, cell-penetrating peptides such as SynB,21,22
Penetratin,23,24 vasoactive intestinal peptide (VIP),25 Trans-
portan,26 and TAT27−30 derivatives as well as polyarginines31
that have been shown to enter the brain via different
transduction mechanisms32,33 (Figure 1D) are also promising
peptidic carriers for transporting small molecule drugs across
the BBB.
The use of peptides is favorable because they are highly
potent substances that display a low immunogenicity and
molecular weight, resulting in fast pharmacokinetics and
background clearance. Furthermore, they are easily accessible
by chemical synthesis and can be site-specifically derivatized
with a cargo drug. All of this makes peptides promising carriers
for small drug molecules that can have a higher chance of
preserving the biological activity of the drug compared to that
when using larger shuttles.
the cargo drug has to preserve its biological activity upon
conjugation to the shuttle.
Because a small drug molecule is structurally altered
significantly by conjugation to a peptide, it is conceivable that
such a conjugation would most likely result in a fusion molecule
of diminished or even eliminated binding affinity of the ligand
to its target receptor. Similar effects of drastically decreased
binding properties (up to 6 orders of magnitude) upon
conjugation of large constructs have been shown exemplarily on
the biotin−streptavidin system recently.34,35
Thus, we intended to investigate if the conjugation of a small
neuroreceptor-affine drug to a carrier peptide can, in principle,
result in a hybrid compound that is potentially able to transport
the small molecule while at the same time preserving its affinity
to the neuroreceptor. We therefore synthesized a palette of
fusion drugs, each consisting of a peptide carrier and a
derivative of the hD2/3 receptor-binding drug fallypride in order
to test the feasibility of this approach.
Fallypride was used as the model neuroreceptor-binding drug
because is used in its 18F-labeled form in human in vivo hD2/3
receptor imaging with positron emission tomography
(PET),36−38 thus making it a well-studied ligand system. In
addition, for its counterpart, the hD2 receptor system, an in
vitro receptor binding assay is available to study the hD2
receptor affinities of the developed hybrid molecules.
For these fallypride and transport peptide conjugates, we
determined the binding affinities to the hD2 receptor in vitro
and found that although a decrease in hD2 receptor affinity is
observed upon conjugation of fallypride to the peptides, the
remaining binding affinity is still surprisingly high and remains
in the nanomolar range. In order to explain and corroborate
these experimental findings, we conducted homology modeling
of the hD2 receptor, starting from the hD3 receptor as the
template, and performed docking studies of the two most
potent hybrid compounds, elucidating the contribution of
different structure elements to receptor binding. We also
A successful fusion molecule consisting of a peptide and a
small drug thus has to fulfill the following requirements: (i) the
shuttle has to target a transcytosis receptor system with high
affinity or has to be able to pass the BBB by a passive
mechanism, (ii) the ability of the shuttle to cross the BBB has
to be maintained after conjugation of the cargo drug, and (iii)
4369
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
conjugated structural motifs to some of the synthesized hybrid
molecules, allowing for radiolabeling with 68Ga and 18F and
future in vivo PET imaging of hD2/3 receptors. Furthermore,
the interaction between the most promising hybrid compounds
and the main excretory transporter of the BBB, P-glycoprotein,
was studied.
For the first step of this reaction, an exact DMF/H2O ratio of
3:1 and a small solvent volume just sufficient to dissolve the FP-
Tos were found to be critical to obtain the product in high yield
and purity. In general, these conditions represent a valuable
new complement to former approaches using NaSH for the
conversion of tosylates into thiols.40,41
Synthesis of Maleimide-Comprising Peptidic Carriers
2a−22a and Their Reaction with FP-T (1), Yielding
Peptide−Fallypride Fusion Molecules 2−22. Carrier
peptide maleimides 2a−22a were synthesized by standard
Fmoc-based solid-phase peptide synthesis (SPPS) using an
excess of 3.9 equiv of O-(benzotriazol-1-yl)-N,N,N′,N′-
tetramethyluronium hexafluorophosphate (HBTU) and 4.0
equiv of amino acids and N,N-diisopropylethylamine (DIPEA)
in each conjugation step. The products were obtained in yields
of 11−90% after HPLC purification and lyophilization,
depending on the complexity of the synthesized peptide.
1,4,7,10-Tetraazacyclododecane-N,N′,N″,N‴-tetraacetic acid
(DOTA) is a powerful chelator for radiometals such as 68Ga
used for in vivo PET imaging. Thus, some DOTA-derivatized
peptides were also synthesized as carriers. DOTA10−
AngioPep−PEGx−maleimides 9a and 17a−20a and DOTA−
TfR-Ps maleimide 22a, all comprising a lysine ε-amino-
conjugated DOTA moiety, were synthesized via two different
pathways. First, the standard reaction pathway was followed
using Fmoc−Lys(Mtt)−OH as an amino acid derivative during
the SPPS of peptides 9a and 22a followed by Mtt-deprotection
under mildly acidic conditions and subsequent conjugation of
tris-t-Bu-DOTA. However, this strategy gave products 9a and
22a in comparatively low yields of 19 and 11%, respectively.
Thus, another approach was subsequently used, utilizing the
amino acid building block Fmoc−Lys(tris-t-Bu-DOTA)−OH
(23). This chelator-modified amino acid building block was
synthesized from Nα-Fmoc-protected lysine and tris-t-Bu-
DOTA using HBTU as an activating agent. Applying this
building block in standard SPPS, the respective peptides, 9a,
17a−20a, and 22a, could be obtained in significantly improved
but still moderate yields of 20−32% after purification.
RESULTS AND DISCUSSION
■
As peptidic carriers, different peptides with the potential to
enter the brain parenchyma by receptor-mediated transcytosis
(two different transferrin receptor binding peptides as well as
AngioPep-2) or adsorptive transcytosis (TAT, TAT49−57
,
transportan, penetratin, and the nuclear localization sequence
of human NF-κB) were chosen as model peptides. They
strongly vary in length, structure, polarity, and geometry. This
structural diversity was intended to elucidate possible influences
of the peptide architecture on the hD2 receptor affinity of the
fallypride moiety in the resulting fusion drugs. Furthermore, the
influence of linker structures and lengths between the peptide
carrier and the receptor ligand was systematically evaluated.
Synthesis of Thio-fallypride (FP-T, 1). The first step in
the stable conjugation of fallypride to different peptidic carriers
was the synthesis of a fallypride derivative containing a
functional group in the terminal nonpharmacophoric position
of the alkyl chain for peptide coupling. For this purpose, thiol-
functionalization of fallypride was chosen because peptides can
be easily modified with maleimides and subsequently reacted
efficiently and site-specifically with thiol-bearing synthons,
yielding a defined fusion molecule via simple and efficient
click chemistry.
The first attempts to obtain a thiol-functionalized fallypride
molecule were made by implementing protected thiol-bearing
synthons in the fallypride scaffold synthesis, resulting, however,
only in the formation of byproducts. Hence, it was intended
that the thiol functionality would be introduced by reacting
tosyl-fallypride (FP-Tos)39 with S-acetyl-thioethanolamine, 1,2-
ethanedithiol, or 2,2′-dithiobis(ethylamine) dihydrochloride
under mildly basic conditions. Unfortunately, the anticipated
thiol-bearing products could not be isolated.
Obtained peptide maleimides 2a−22a were subsequently
reacted with FP-T (1) in aqueous media at ambient
temperature and neutral pH for 10 min (Scheme 2). Products
2−22 were purified by semipreparative HPLC and obtained in
yields between 36 and 79%.
Influence of Different Structural Elements in the
Shuttle−Cargo Fusion Molecules on the hD2 Receptor
We subsequently developed another method starting from
FP-Tos using NaSH directly as the nucleophile in a DMF/H2O
system. These reaction conditions yielded thiol-derivatized
product FP-T (1) within short reaction times of only 10−20
min in good yields (Scheme 1). As a result of the basic reaction
conditions using NaSH, the dimer of FP-T was initially formed,
which could subsequently be reduced efficiently using tris(2-
carboxyethyl)phosphine hydrochloride (TCEP), giving the
product in good overall yields of 76% after semipreparative
HPLC purification.
Scheme 2. General Reaction Pathway for the Conjugation
Reactions of Peptide Maleimides 2a−22a to FP-T (1),
a
Yielding Peptide−FP-T conjugates 2−22
Scheme 1. Synthesis of FP-T (1) Starting from Tosyl-
a
fallypride
a
Reagents and conditions: (i) NaSH, DMF/H2O 3:1, rt; (ii) TCEP,
a
DMF/H2O, rt.
Reagents and conditions: PB/MeCN, pH 6.7−7.2, rt.
4370
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
Figure 2. Schematic depiction of the synthesized shuttle−cargo fusion molecules of subset I (carrier peptide−FP-T fusion molecules 2−9).
Figure 3. Structures of TfR-Ps−PEGx−FP-T fusion molecules 4 and 10−16 (fusion molecule subset II).
Binding Affinities of the Fallypride Moiety. In order to
determine the influence of peptide architecture as well as
potential linkers or other structure elements such as SiFA
(silicon fluoride acceptor) and DOTA on the binding affinity of
the conjugated fallypride moiety to the hD2 receptor, we
synthesized four different subsets of fusion molecules (I−IV)
by systematically varying different structure elements. Sub-
sequently, we determined the binding affinities of these hybrid
compounds to the hD2 receptor in vitro.
Hybrid Molecule Subset I. This first subset of fusion
molecules (I) was intended to reveal if the complexity of the
carrier peptide exerts a significant influence on the binding
parameters of the conjugated fallypride moiety. Thus, different
peptidic carriers with the potential to cross the BBB by
receptor-mediated processes (TfR-Ps, TfR-Pl, and AngioPep-2)
or transcytosis (TAT, TAT49−57, transportan, penetratin, and
nuclear localization sequence of NF-κB) of different lengths,
amino acid sequence, and complexity were synthesized,
derivatized with maleimides, and finally conjugated to FP-T
(1). The resulting fusion molecules, 2−9, are depicted in Figure
2.
Hybrid Molecule Subset II. The second subset was based on
the relatively short transferrin receptor binding peptide
HAIYPRH (TfR-Ps) that has been shown to be able to
efficiently enter cells by specific transferrin receptor-mediated
transport.10,42,43 This peptide scaffold was derivatized on a solid
support with linkers of different lengths (γ-amino butyric acid,
PEG1, PEG2, PEG3, PEG5, PEG7, and PEG11) in order to
determine if the probable steric hindrance exerted by the
peptide on the binding of the fallypride moiety to the hD2
receptor could be attenuated by using an appropriate linker
structure. The linker-comprising peptides were reacted on the
solid support with maleimido-hexanoic acid, and after cleavage
from the resin and purification they were conjugated to FP-T
(1). The synthesized TfR-Ps−PEGx−FP-T fusion molecules, 4
and 10−16, comprising linkers of different length are depicted
in Figure 3.
Hybrid Molecule Subset III. The third subset of fusion
molecules (III) was based on the AngioPep-2 peptide that
enters the brain by specific receptor-mediated transcytosis
utilizing the low-density lipoprotein related receptor
LRP1.17,44,45 Because it has been shown that a derivatization
in position Lys10 of the peptide is well-tolerated in terms of
4371
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
Figure 4. Structures of AngioPep−PEGx−FP-T fusion molecules 9 and 17−20 (fusion molecule subset III).
Figure 5. Structures of TfR-Ps−SiFA−PEG3−FP-T (21), TfR-Ps−DOTA−PEG3−FP-T (22) and DOTA10−Py515−AngioPep−FP-T (24) (fusion
molecule subset IV).
AngioPep’s transport capacity,16,18 the chelator DOTA was
introduced in this position in all synthesized derivatives. For
this purpose, the amino acid building block Fmoc−Lys(tris-t-
Bu-DOTA)−OH (23) was synthesized and implemented into
routine Fmoc-SPPS, giving improved yields compared to
standard peptide scaffold synthesis, subsequent side-chain
deprotection, and chelator functionalization. AngioPep-2 was
further reacted on a solid support with PEG linkers of different
lengths (PEG3, PEG5, PEG7, and PEG11) and maleimido-
hexanoic acid. After cleavage from the resin and purification,
the peptide derivatives were conjugated to FP-T (1).
Synthesized AngioPep−PEGx−FP-T fusion molecules 9 and
17−20 are shown in Figure 4.
Hybrid Molecule Subset IV. Finally, the fourth group of
fusion molecules comprised either a SiFA46−48 or DOTA
moiety (for radiolabeling with 18F or 68Ga) as well as the near-
infrared-emitting fluorescent dye, Py5.49 These fusion mole-
cules could, in principle, be applicable in in vivo PET imaging
or combined PET/optical imaging of hD2 receptors. From
these conjugates, it should be possible to determine if the
4372
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
Figure 6. Structures PEG3−FP-T (25), TfR-Ps−SiFA−PEG3−cysteine (26), and underivatized TfR-Ps (27), TfR-Pl (28), and TAT49−57 (29).
binding affinity of the fallypride moiety to the hD2 receptor
tolerates complex structure elements essential for radiolabeling
and in vivo imaging such as SiFA, DOTA, and fluorescent dyes
introduced within the carrier peptide part of the molecule.
Therefore, additional different conjugates based on TfR-Ps and
AngioPep-2 were synthesized and derivatized with a SiFA/
DOTA and/or Py5 moiety. SiFA-comprising fusion molecule
TfR-Ps−SiFA−PEG3−FP-T (21) was synthesized using
Fmoc−Lys(Mtt)−OH during the SPPS. The Mtt protecting
group was cleaved under mildly acidic conditions before
reacting with bis-Boc-aminooxy-acetic acid using the standard
coupling protocol on solid support. The peptide was then
modified with the PEG3-linker and maleimido-hexanoic acid,
cleaved from the solid support, and reacted with SiFA-A, an
aldehyde-bearing SiFA building block,48,50 and FP-T (1) after
purification. 22 was synthesized as described earlier. The
DOTA and Py5 near-infrared fluorescent dye comprising fusion
molecule 24 was obtained by reaction of DOTA10−AngioPep−
FP-T (9) with Py5. Obtained fusion molecules 21, 22, and 24
are depicted in Figure 5.
peptide/linker/2,5-dioxo-3-pyrrolidinyl-thioether/lipophilic
SiFA structure of the fusion molecules with the hD2 receptor,
we prepared various negative control compounds that were also
evaluated in vitro with regard to their binding affinities to the
hD2 receptor.
TfR-Ps−SiFA−PEG3−cysteine (26) was obtained by reac-
tion of 21a with cysteine in order to obtain a very structurally
similar analogue to 21 that is, however, not able to specifically
bind to the hD2 receptor because it lacks the fallypride moiety,
but it retains the peptide, PEG linker, SiFA moiety, and 2,5-
dioxo-3-pyrrolidinyl-thioether. Furthermore, underivatized
TfR-Ps (27), TfR-Pl (28), and TAT49−57 (29) as well as a
PEG3−FP-T (25, as a specific binding control) were
synthesized and tested for their in vitro binding affinities to
the hD2 receptor together with fusion molecules 2−22, 24, and
26. The structures of control substances 25−29 are presented
in Figure 6.
Radiolabeling of Fusion Molecules 9, 17−22, and 24
with 18F and 68Ga. In principle, the synthesized fusion
molecules could be interesting candidates for in vivo imaging of
the hD2/3 receptor system with PET. Consequently, some of
the conjugates were designed bearing a SiFA46−48 or DOTA
Synthesis of Control Substances. In order to exclude an
unlikely but theoretically possible nonspecific interaction of the
4373
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
moiety in order to be amenable to direct radiolabeling with 18
F
carried out in triplicate. The binding affinity of fallypride
derivative 1 could not be determined because it showed
considerable dimerization under the conditions of the in vitro
binding experiment that would have falsified the results.
Fallypride (FP), desmethoxy-fallypride (DMFP), raclopride
(RP), haloperidol (HP), chlorpromazine (CP), and (+)-buta-
clamol (BC) were coevaluated as reference compounds. Table
1 summarizes the results of these evaluations.
and 68Ga, respectively. In the following radiolabeling study, we
investigated whether SiFA- and DOTA-comprising fusion
molecules 9, 17−22, and 24 are de facto amenable to
radiolabeling with 18F and 68Ga.
The radiolabeling of the fusion molecule DOTA10−Py515−
AngioPep−FP-T (24) with 68Ga3+ gave only poor radiolabeling
results under standard reaction conditions, as several
unidentified radioactive byproducts were formed that could
not be easily separated by HPLC. Because labeling precursor 24
is structurally identical to 9 (apart from the near-infrared
fluorescent dye Py5 being introduced into the Nε-Lys15 amino
side chain functionality of the peptide), which can be labeled
efficiently using standard conditions as described below, the
formation of byproducts seems to be a result of the sensitivity
of the large delocalized system of π electrons within the
fluorescent dye. Thus, a derivative being amenable to 68Ga
labeling for PET imaging as well as to optical imaging
simultaneously would require the use of a different, less
sensitive near-infrared fluorescent dye.
Table 1. Ki Values of Compounds 2−22 and 24−29 and
b
Internal Standards
Also, when radiolabeling the fusion molecule TfR-Ps−SiFA−
PEG3−FP-T (21) with 18F−, several unidentified radioactive
byproducts were formed, resulting in only poor radiolabeling
results under standard reaction conditions. Similar byproducts
as those observed in the 18F labeling of 21 were also observed
for a structurally similar fusion molecule of TfR-Ps and a SiFA
moiety conjugated via a 2,5-dioxo-3-pyrrolidinyl-thioether
structure (SiFA−S-maleimido−THRPPMWSPVWP).51 Be-
cause neither the presence of peptides comprising various
side chain functionalities nor PEG linkers nor fallypride
interfere with the used 18F-SiFA-labeling conditions,52,53 it
can be inferred from these data that the 2,5-dioxo-3-
pyrrolidinyl-thioether structure element is sensitive to the
slightly basic labeling conditions used, resulting in the
formation of structurally similar, perhaps ring opening,
byproducts. In order to circumvent the formation of these
byproducts, milder 18F-labeling conditions were applied,
resulting in the formation of the desired product, [18F]21,
only. These reaction conditions apply the so-called Munich
method for cartridge-based [18F]fluoride drying54 and a
subsequent in part neutralization of the reaction mixture with
oxalic acid, which has been shown to give excellent radio-
labeling results.48 Under these conditions, the only byproduct is
unreacted [18F]fluoride, which can be easily removed by
cartridge purification, giving final product [18F]21 in radio-
chemical yields of up to 60%, radiochemical purities of ≥98%,
and specific activities of up to 80 GBq/μmol (2160 mCi/μmol)
in an overall synthesis time of only 25 min.
a
Concentrations of up to 1 μM were studied and showed no
displacement of [3H]spiperone. Abbreviations: FP, Fallypride; DMFP,
desmethoxy-fallypride; RP, raclopride; HP, haloperidol; CP, chlorpro-
mazine; and BC, (+)-butaclamol. Yellow, fusion molecule subset I;
orange, fusion molecule subset II; green, fusion molecule subset III;
blue, fusion molecule subset IV; violet, positive and negative controls;
and grey, internal standards.
b
As can be seen from the in vitro data, the binding affinity of
the fallypride moiety to the hD2 receptor is still in the
nanomolar range (between 1.5 and 64.2 nM) for all studied
compounds and thus it is in a comparable range to that of
raclopride (affinity: 1.2 nM) and chlorpromazine (affinity: 6.1
nM), although the affinity is markedly reduced upon
conjugation to complex structural systems such as peptides.
This is remarkable insofar as a more pronounced reduction of
binding affinities was expected, especially in the case of large
peptidic carriers that were hypothesized to considerably hamper
the fallypride moiety from binding to the hD2 receptor by
folding effects, steric hindrance, or entropy effects.
As expected, fusion molecule 26 (comprising all structure
elements of high-affinity binder 21 except for the fallypride
moiety: TfR-binding peptide, PEG linker, lipophilic SiFA
moiety, and 2,5-dioxo-3-pyrrolidinyl-thioether) and unmodified
peptides 27−29 showed no affinity to the hD2 receptor in the
The 68Ga labeling of fusion molecules 9, 17−20, and 22
proceeded efficiently, giving the radiolabeled products in high
radiochemical yields and purities of 96−98% as well as in high
specific activities of 19.2−28.1 GBq/μmol (519−759 mCi/
μmol) using standard labeling conditions.
Determination of hD2 Receptor Binding Affinities of
2−22 and 24−29. Fusion shuttle−carrier fusion molecules 2−
22 and 24 as well as control substances 25−29 were tested in
vitro for their hD2 receptor binding affinity in an established
competition binding assay by displacement of [3H]spiperone
on human D2 receptor-transfected HEK cells.51,55 This is an
established procedure, with the obtained affinities giving a
valuable indication of the biological potency of the newly
developed receptor ligands. All binding experiments were
performed at least thrice separately, with each experiment being
4374
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
Figure 7. (A) Predicted binding mode of 13 to the model hD2 receptor obtained from 6 ns MD simulations. The fallypride part of compound 13 is
depicted in original atom color, the linker part, in green, and the peptidic part (TfR-binding peptide), in yellow. (B) Overlay of the energy-minimized
MD average structures of 25 (in cyan carbon atoms) and 13 (fallypride is depicted in original atom color, linker structure, in green, and peptide, in
yellow) bound to the hD2 receptor. The hD2 receptor is displayed as a molecular surface in gray for improved clarity.
18F-binder or NIR dye resulted in a markedly decreased binding
studied concentration range of up to 1 μM. This conclusively
demonstrates that the observed high hD2 receptor binding
affinities of the fusion molecules are a result of the fallypride
structure element and thus specific binding.
affinity in the case of the introduction of a small lipophilic SiFA
or Py5 moiety (6.0 nM for 21 instead of 1.5 nM for its lead
compound, 13, and 64.2 nM for 24 instead of 8.0 nM in the
case of its lead compound, 9) and even a significant decrease in
affinity in the case of the introduction of a hydrophilic and
sterically demanding DOTA moiety (52.0 nM for 22 instead of
1.5 nM in case of its lead compound, 13). The strong reduction
of affinity observed in the case of DOTA introduction into the
peptide sequence of TfR-Ps (22) interestingly does not seem to
be a result of the position of conjugation (the introduction of
the SiFA moiety in the same position (21) did not result in
such a marked decrease of binding affinity) but might be a
consequence of the higher hydrophilicity/charge of the DOTA
moiety compared to that of the SiFA scaffold.
Molecular Modeling of the hD2 Receptor and Docking
Studies. As a more pronounced reduction in the binding
affinities of fallypride to the human D2 receptor was expected
upon conjugation to large and complex peptidic carriers, we
also performed homology modeling of the receptor and
docking studies of selected compounds (fallypride, 25, and
13; detailed information is available in the Supporting
Information). By this, we intended to determine the influence
of the different structure elements of the fusion molecules on
hD2 receptor binding and to elucidate the reason for the
experimentally observed preserved binding of the hybrid
molecules.
The complexity, length, and individual structure of the used
peptidic carriers (studied within subset I; compounds 2−9)
seems to have no significant influence on the binding affinities
of the conjugated fallypride moiety. Fusion molecules
comprising very short peptides such as 4 exhibit comparable
affinities (10.3 nM) as that of very complex ones such as 7 or 6
(8.1 and 12.3 nM, respectively). This is further corroborated by
the results obtained for compound 9, for which a relatively high
affinity of 8.0 nM was observed (which is the highest observed
affinity within this subgroup) despite the rather long peptide
sequence, the variety in hydrophilic as well as lipophilic amino
acid side chain functionalities, and the additional chelator
structure within the peptide sequence.
Regarding the linker length, whose influence was studied
within the subsets II (compounds 4 and 10−16) and III
(compounds 9 and 17−20) using the transferrin receptor-affine
peptide TfR-Ps and AngioPep as scaffold peptides, significant
effects of linker lengths on binding affinities were observed.
Within TfR-Ps−PEGx−FP-T subgroup II, the best results were
found for 13 (binding affinity of 1.5 nM), whereas shorter or
longer linkers resulted in lower affinities between 10.3 and 53.9
nM, which might be attributable to suboptimal receptor
penetration depths in the case of short linkers and entropy
costs in the case of long linkers.
When docking hybrid molecule 13 to the modeled hD2
receptor, the peptide, in contrast to the PEG linker, does not
seem to contribute to the stabilization of fusion molecule
binding to the hD2 receptor, as it is mostly exposed to solvent.
The predicted binding of 13 to the hD2 receptor is depicted in
Figure 7A,B and shows an overlay of the predicted binding
modes of 25 and 13 for direct comparison.
However, different results were obtained within AngioPep−
PEGx−FP-T series III, where the highest binding affinity of 8.0
nM was found for derivative 9 that does not have any linker
structure at all. The introduction of a linker in this subset
resulted in affinities moderately decreasing from 26.1 to 44.7
nM with increasing linker lengths.
Thus, the linker length resulting in optimal binding
parameters in such fusion molecules has to be determined for
each peptide system independently.
A considerable influence on binding affinities was also
observed for the conjugation of SiFA, DOTA, and Py5 within
the peptide carrier sequence that was studied in subgroup IV,
including fusion molecules 21, 22, and 24. The conjugation of a
Hence, this in silico modeling shows that the fallypride
moiety of the fusion molecules is still able to bind in its binding
pocket of the hD2 receptor and that the linker and peptide do
not seem to negatively influence the ligand binding by not
directly interacting with the binding pocket of the receptor. On
the contrary, it indicates that the PEG linker might contribute
to the observed preserved binding properties of the fallypride
4375
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
Figure 8. Results obtained from the Calcein assay to identify the interaction of the studied substances with P-gp. Intracellular fluorescence in the
absence of test compounds was set as 100%. None of the substances showed any modulation of P-gp in the concentration range of 0.1−10 μM. At
higher concentrations, fallypride as well as 21 showed an interaction with the applied esterase, disturbing the assay at concentrations >10 μM and
resulting in the lower fluorescence values observed, which are not consequence of P-gp modulation by these compounds. Cyclosporine A was used as
a positive control and showed an IC50 value of 0.94 μM under the same conditions.
moiety by creating a spatial separation between the binding
ligand and the peptidic carrier.
known P-gp inhibitor cyclosporine A as an internal positive
control.
These results obtained in silico thus support the three main
experimentally observed effects: (i) the retained binding affinity
of the hybrid molecules to the hD2 receptor, (ii) the increased
affinity of PEG linker-modified hybrid molecules when
optimizing linker lengths, and (iii) the missing influence of
the peptide size and complexity on the fallypride binding
(Table 1).
In this assay, we studied the most promising hybrid
molecules, 9, 13, 21, and 22, as well as their native peptide
lead structures 27, AngioPep-2, and the receptor ligand
fallypride as references and Cyclosporine A as positive control.
Determination of the Toxicity of the Substances. In order
to exclude the possibility that the tested compounds are toxic to
the cells and thus falsifying the Calcein assay, the cytotoxicity of
the compounds was evaluated in an AlamarBlue assay that
identifies cytotoxic substances by their ability to decrease the
metabolic activity and thus the redox potential of viable cells.
The assay revealed no cytotoxic effects for any of the tested
compounds on hCMEC/D3 cells (Figure S1, Supporting
Information).
Determination of the Interaction of Hybrid Molecules and
P-glycoprotein. The interaction of the substances with P-gp
was studied by subjecting them to the Calcein assay, which is a
highly sensitive, specific, and reliable screening method that
allows for the rapid evaluation and classification of drug
candidates according to their P-gp interaction.56,57 Intracellular
accumulation of free Calcein corresponds to the degree of P-gp
interaction of a given test compound. For this assay, the
hCMEC/D3 cell line was used, which is known to stably
express P-gp.58 The P-gp inhibitor, Cyclosporine A, served as
an internal positive control.
Estimation of the Blood−Brain Barrier Permeability of
Hybrid Compounds 9, 13, 21, and 22, Native TfR-
Binding Peptide, AngioPep-2, and Fallypride. After
having shown that the conjugation of a small neuroreceptor-
affine drug to a transport peptide can result in a hybrid
compound that is still able to bind its target receptor, we
intended to estimate the potential of the developed fusion
molecules to enter the brain parenchyma. The BBB is the last
hurdle for the hybrid compounds in the studied molecular
Trojan horse approach using covalent conjugates of receptor-
affine small molecules and peptides as carriers. On the
molecular level, the P-glycoprotein transporter protein (P-gp)
is known to play the major role in the development of drug
resistance that is associated with limited CNS permeation of
drug candidates.56 Because of its broad substrate spectrum and
high expression level at various blood−tissue barriers, P-gp
considerably affects the pharmacokinetics and bioavailability of
various agents.
The Calcein cell assay is a highly sensitive, specific, and
reliable screening method that allows for the rapid evaluation
and classification of drug candidates according to their P-gp
interaction.56 A key detection component in this assay is the
Calcein−acetoxymethylester (Calcein-AM) complex that is
easily internalized in cells and acts as a substrate of both P-
gp and intracellular esterases. In the presence of a P-gp
inhibitor or substrate, the transporter binding sites are
occupied, and as a result, the Calcein-AM complex cannot
interact with P-gp. In this case, Calcein-AM is not transported
out of the cells and is hydrolyzed to the fluorescent product,
Calcein, by the activity of cellular esterases. Depending on the
P-gp affinity of certain test compounds, the amount of Calcein
formation reaches different levels at specific compound
concentrations. The level of the fluorophore formation serves
as an indicator to determine the P-gp interaction profile of the
given test compounds.56,57 In this study, we used the human
brain endothelial capillary cell line hCMEC/D3 that exhibits
stable expression of P-gp.58 In order to verify the assay, we used
None of the studied substances, neither hybrid molecules nor
native peptides nor fallypride, were substrates of P-gp (Figure
8), implying that they are not transported from the abluminal
to the luminal side of the endothelium by P-gp. Although this
assay is not able to reproduce the complex situation in a living
subject, these results indicate that the transport of the
developed fusion molecules over the BBB is not prevented by
the main excretory transporter of the BBB, P-gp.
CONCLUSIONS
■
From the obtained results, it can be inferred that even a small
neuroreceptor ligand such a fallypride binding to the hD2/3
receptor can be conjugated to a relatively large peptidic carrier
molecule while largely retaining its nanomolar binding affinity
to the target receptor (specific binding affinities of up to 1.5 nM
were found in this study).
These promising experimental in vitro results were further
substantiated by in silico molecular modeling studies that
indicated that the fallypride moiety of the fusion molecules is
4376
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
General Synthesis of Carrier Peptide Maleimides 2a−20a.
Peptides were synthesized on solid support by standard Fmoc solid-
phase peptide synthesis as described by Wellings et al.60 using standard
commercially available resins, amino acids, Fmoc−amino−PEG−acids
and Fmoc−Lys(tris-t-Bu-DOTA)−OH (23). The N-terminal mod-
ification with maleimido-hexanoic acid was accomplished using the
standard coupling protocol for amino acids after cleavage of the N-
terminal Fmoc protecting group. The resulting maleimide-modified
peptides were cleaved from the solid support using a mixture of TFA
(trifluoroacetic acid)/TIS (triisopropylsilane)/H2O (95:2.5:2.5) for 90
min, suspended in diethyl ether, and purified by semipreparative
HPLC. The products were isolated as white solids after lyophilization.
Gradients used for HPLC purification and overall synthesis yields are
given below.
still able to bind to its binding pocket of the hD2 receptor.
Despite the significant structural changes exerted by con-
jugation, neither linker nor peptide seem to negatively influence
the ligand binding to its receptor. The linker can even slightly
stabilize the ligand−receptor interaction according to the in
silico modeling.
Furthermore, the most promising hybrid substances were
tested to determine if they possibly interact with P-gp, showing
that these substances are not substrates of this transporter. This
indicates that the transcytosis of the developed fusion
molecules over the BBB is not prevented by its main excretory
transporter, P-gp.
These findings should encourage enlarging the research focus
of molecular Trojan horse drug delivery across the BBB from
the generally used proteins toward transport peptides as
carriers.
2a: gradient, 0−30% MeCN in 6 min; yield, 90%; MALDI-MS (m/
z) for [M + H]+ (calcd): 1912.2 (1912.1). 3a: gradient, 0−30% MeCN
in 6 min; yield, 72%; ESI-MS (m/z) for [M + 2H]2+ (calcd): 738.92
(766.97). 4a: gradient, 15−40% MeCN in 6 min, yield: 76%; ESI-MS
(m/z) for [M + H]+ (calcd): 1085.55 (1085.55), (m/z) for [M +
2H]2+ (calcd): 543.28 (543.27). 5a: gradient, 20−45% MeCN in 6
min; yield, 44%; MALDI-MS (m/z) for [M + H]+ (calcd): 1683.6
(1683.8). 6a: gradient, 30−80% MeCN in 6 min; yield, 11%; MALDI-
MS (m/z) for [M + H]+ (calcd): 3040.6 (3035.8). 7a: gradient, 15−
40% MeCN in 6 min; yield, 28%; ESI-MS (m/z) for [M + 3H]3+
(calcd): 814.13 (832.80), (m/z) for [M + 4H]4+ (calcd): 610.85
(624.85). 8a: gradient, 15−40% MeCN in 6 min; yield, 83%; ESI-MS
(m/z) for [M + H]+ (calcd): 1476.84 (1476.84), (m/z) for [M +
EXPERIMENTAL SECTION
■
General. All commercially available chemicals were of analytical
grade and were used without further purification. Resins for peptide
synthesis, coupling reagents, and Fmoc-protected amino acids were
purchased from NovaBiochem. Calcein and Cyclosporine A were
purchased at the highest purity grade from Sigma-Aldrich. EBM-2
basal medium and Single-Quot Kit were obtained from Lonza, fetal
bovine serum, from Biochrom, and rat-tail collagen type I, from
Hoffmann-LaRoche. SiFA-A, Py5, and tosyl-fallypride were synthe-
sized as described elsewhere.39,50,59 For analytical and semipreparative
HPLC chromatography, an Agilent 1200 system or a Dionex 3000
HPLC system equipped with a Gabi-Star (Raytest) radioactivity
detector was used. The columns used for chromatography were
Chromolith Performance (RP-18e, 100−4.6 mm, Merck, Germany)
and Chromolith (RP-18e, 100−10 mm, Merck, Germany) columns,
operated with flows of 4 and 8 mL/min, respectively. The preparative
HPLC system used was a Knauer Preparative Pump 1800 together
with a Knauer Smartline UV detector 2600 and a Chromolith Prep
(RP-18e, 100−25 mm, Merck, Germany) column, operated with a
flow of 100 mL/min. ESI, MALDI, and NMR spectra were obtained
with a Finnigan MAT95Q, a Bruker Daltonics Microflex, and a Jeol
AS500 spectrometer, respectively.
2H]2+ (calcd): 738.92 (738.92). 9a: gradient, 0−50% MeCN in 6 min;
2+
yield, 31%; ESI-MS (m/z) for [M + H + Na + Kcomplexed
1471.64 (1471.66), (m/z) for [M + 2H + Na + Kcomplexed
]
(calcd):
(calcd):
3+
]
981.10 (981.11). 10a: gradient, 0−30% MeCN in 6 min; yield, 69%;
ESI-MS (m/z) for [M + H]+ (calcd): 1170.60 (1170.60), (m/z) for
[M + 2H]2+ (calcd): 585.80 (585.80). 11a: gradient, 0−30% MeCN in
6 min; yield, 73%; ESI-MS (m/z) for [M + H]+ (calcd): 1230.62
(1230.62), (m/z) for [M + 2H]2+ (calcd): 616.31 (615.81). 12a:
gradient, 0−30% MeCN in 6 min; yield, 61%; ESI-MS (m/z) for [M +
H]+ (calcd): 1288.66 (1288.66), (m/z) for [M + 2H]2+ (calcd):
644.83 (644.83). 13a: gradient, 0−30% MeCN in 6 min; yield, 65%;
ESI-MS (m/z) for [M + H]+ (calcd): 1332.69 (1332.69), (m/z) for
[M + 2H]2+ (calcd): 666.85 (666.85). 14a: gradient, 0−30% MeCN in
6 min; yield, 84%; ESI-MS (m/z) for [M + H]+ (calcd): 1420.74
(1420.74), (m/z) for [M + 2H]2+ (calcd): 710.87 (710.87). 15a:
gradient, 10−40% MeCN in 6 min; yield, 71%; ESI-MS (m/z) for [M
+ H]+ (calcd): 1508.80 (1508.79), (m/z) for [M + 2H]2+ (calcd):
754.90 (754.90). 16a: gradient, 10−40% MeCN in 6 min; yield, 73%;
ESI-MS (m/z) for [M + H]+ (calcd): 1684.90 (1684.90), (m/z) for
[M + 2H]2+ (calcd): 842.95 (842.95). 17a: gradient, 0−40% MeCN in
6 min; yield, 32%; substance was directly reacted with FP-T after
reduction of solvent volume. 18a: gradient, 0−40% MeCN in 6 min;
yield, 31%; substance was directly reacted with FP-T after reduction of
solvent volume. 19a: gradient, 0−40% MeCN in 6 min; yield, 31%;
substance was directly reacted with FP-T after reduction of solvent
volume. 20a: gradient, 0−40% MeCN in 6 min; yield, 22%; substance
was directly reacted with FP-T after reduction of solvent volume.
Synthesis of TfR-Ps−SiFA−PEG3−Maleimide 21a. TfR-Ps (50
μmol) was synthesized on solid support as described above. Fmoc−
Lys(Mtt)−OH was coupled using the standard coupling protocol after
removal of the N-terminal Fmoc protecting group followed by
coupling of Fmoc−PEG3−OH and maleimido-hexanoic acid. For
selective deprotection of the terminal lysine side chain functionality,
the Mtt protecting group was cleaved using TFA (1.75%) and TIS
(1%) in DCM, and the resulting charged ammonium group was
converted into the amine by treatment with DIPEA in DCM (2%, 6 ×
2 mL) before reacting with bis-Boc-aminooxy-acetic acid using the
standard coupling protocol. The peptide was cleaved from the solid
support, deprotected, and purified by semipreparative HPLC using a
gradient of 0−40% MeCN + 0.1% FA over 6 min. The product was
obtained as a white solid after lyophilization (15.6 mg, 10.2 μmol,
20%). ESI-MS (m/z) for [M + H]+ (calcd): 1533.80 (1533.80), (m/z)
for [M + 2H]2+ (calcd): 767.40 (767.40). Subsequently, a solution of
The purity of synthesized compounds 1, 2a−22a, and 2−29 was
≥96% in each case, as determined by analytical HPLC.
The determination of the FP-T−peptide conjugate binding affinities
to the human D2 receptor was performed by competition binding
experiments according to a previously described procedure.51
A detailed description of homology modeling of the hD2 receptor,
model prediction, docking studies, and molecular dynamics
simulations as well as a detailed description of the cytotoxicity and
Calcein assays can be found in the Supporting Information.
Synthesis of 3-(3-((1-Allylpyrrolidin-2-yl)methylcarbamoyl)-
4,5-dimethoxy-phenyl)-propyl-thiol (FP-T, 1). A solution of
sodium hydrosulfide (217 mg, 3.87 mmol) in DMF/H2O (3:1, 1
mL) was added to a solution of tosyl-fallypride (200 mg, 387 μmol) in
DMF (650 μL). The initially dark green color of the solution bleached
within 10 to 15 min, indicating the end of the reaction. After addition
of H2O (1 mL), Tris(2-carboxyethyl)phosphine hydrochloride
(TCEP) (∼150 mg) was added as a solid until gas formation stopped.
The crude product was purified by semipreparative HPLC using a
gradient of 0−40% MeCN + 0.1% FA over 6 min. The product was
obtained as colorless oil after lyophilization (111 mg, 293 μmol, 76%).
1H NMR (acetone-d6): δ 8.56 (bs, 1H), 7.35−7.32 (m, 1H), 7.10−
7.07 (m, 1H), 6.13−6.03 (m, 1H), 5.53 (d, 1H, J2 = 16.5), 5.41 (d, 1H,
J2 = 10.3), 4.06−4.02 (m, 1H), 3.92 (s, 3H), 3.90 (s, 3H), 3.89−3.70
(m, 4H), 3.23−3.16 (m, 1H, J3 = 7.7), 2.72 (t, 2H, J3 = 7.7), 2.68−2.46
(m, 4H), 2.36−2.31 (m, 1H), 2.13−2.07 (m, 2H), 2.01−1.82 (m, 2H),
1.77 (t, 1H, J3 = 7.9). 13C NMR (acetone-d6): δ 166.89, 153.68,
146.88, 138.29, 129.24, 127.48, 124.69, 122.20, 116.73, 66.82, 61.52,
58.07, 56.41, 54.41, 41.53, 36.46, 34.71, 29.11, 24.26, 23.53. ESI-MS
(m/z) for [M + H]+ (calcd): 379.20 (379.20).
4377
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
this aminooxy-functionalized intermediate (9.4 mg, 6.13 μmol) in PB
(0.5 M, pH 4, 150 μL) was added to a solution of SiFA-A (3.26 mg,
12.26 μmol) in MeCN (300 μL). After 10 min, the crude product was
purified by semipreparative HPLC using a gradient of 20−90% MeCN
+ 0.1% FA over 6 min. The product was obtained as a white solid after
lyophilization (5.3 mg, 2.98 μmol, 49%). ESI-MS (m/z) for [M − H]−
(calcd): 1781.92 (1781.94).
hexafluorophosphate (HBTU, 1.0 g, 2.65 mmol) and tris-t-Bu-
DOTA (1.56 g, 2.72 mmol) in DMF (6 mL) was added
diisopropylethylamine (DIPEA, 464 μL, 2.72 mmol). After 2 min
reaction time, the mixture was added to a suspension of Fmoc−lysine
(1.0 g, 2.72 mmol) in DMF (2 mL) and reacted for 30 min. Afterward,
the nonreacted Fmoc−lysine was filtered off, and the solvent was
removed. The crude product was purified by preparative HPLC using
a gradient of 15−70% MeCN + 0.1% TFA over 10 min. The product
was obtained as light yellow solid oil after lyophilization (815 mg, 883
μmol, 33%). 1H NMR (DMSO-d6): δ 8.52 (bs, 1H), 7.91 (d, 2H, J3 =
7.6), 7.75−7.72 (m, 2H), 7.69 (d, 1H, J3 = 8.3), 7.43 (t, 2H, J3 = 7.5),
7.33 (t, 2H, J3 = 7.5), 4.33−4.20 (m, 2H), 3.96−3.90 (m, 2H), 3.61−
2.87 (bm, 18H), 1.74−1.23 (m, 6H), 1.47 (s, 9H), 1.39 (s, 18H). 13C
NMR (DMSO-d6): δ 175.57, 173.86, 158.70, 158.35, 158.01, 156.11,
143.69, 140.65, 127.59, 126.30, 125.21, 120.08, 65.52, 54.56, 54.46,
53.54, 53.21, 53.19, 50.79, 50.69, 50.61, 46.57, 27.59, 23.07. ESI-MS
(m/z) for [M + Na]+ (calcd): 923.55 (923.54).
Synthesis of DOTA10−Py515−AngioPep−FP-T (24). To a
solution of DOTA10−AngioPep−maleimide−FP-T conjugate (9, 3.1
mg, 0.95 μmol) and Py5 (0.7 mg, 1.90 μmol) in DMF (250 μL) was
added DIPEA (0.32 μL, 1.90 μmol), and the mixture was reacted for
10 min. The crude product was purified by semipreparative HPLC
using a gradient of 10−40% MeCN + 0.1% FA over 6 min. The
product was obtained as a dark red solid after lyophilization (1.37 mg,
0.38 μmol, 40%). ESI-MS (m/z) for [M + 2H]3+ (calcd): 1173.91
(1173.90), (m/z) for [M + 3H]4+ (calcd): 880.68 (880.68).
Synthesis of PEG3−FP-T (25). To a commercially available Rink
Amide resin (50 μmol) were subsequently coupled Fmoc−PEG3−OH
and maleimido-hexanoic acid using standard amino acid coupling
conditions. The product was cleaved from the resin as described before
and purified by semipreparative HPLC using a gradient of 10−40%
MeCN + 0.1% FA over 6 min. The product was obtained as colorless
crystallizing oil after lyophilization (19 mg, 41.6 μmol, 83%). ESI-MS
(m/z) for [M + H]+ (calcd): 480.23 (480.23). Subsequently, a
solution of this maleimide intermediate (4.8 mg, 10.5 μmol) in PB
(phosphate buffer, 0.1 M, pH 4.0)/MeCN 1:1 (250 μL) was added to
a solution of FP-T (1, 3.0 mg, 7.9 μmol), and the pH of the solution
was adjusted to 7.0 using PB (0.5 M, pH 7.2, ∼ 50 μL). After 10 min,
the product was purified by semipreparative HPLC using a gradient of
10−50% MeCN + 0.1% FA over 6 min. The product was isolated as
colorless oil after lyophilization (2.72 mg, 3.26 μmol, 41%). ESI-MS
(m/z) for [M + H]+ (calcd): 836.45 (836.44).
Synthesis of TfR-Ps−SiFA−PEG3−Cysteine (26). To a solution
of TfR-Ps−SiFA−PEG3−maleimide (21a, 6.2 mg, 3.48 μmol) in
MeCN/PB (0.1M, pH 5.0) 1:1 (300 μL) was added a solution of
cysteine (1.0 mg, 8.26 μmol) in H2O (100 μL), and the pH of the
mixture was adjusted to 7.0 using PB (0.5 M, pH 7.2, ∼25 μL). After 5
min, the crude product was purified by semipreparative HPLC using a
gradient of 10−65% MeCN + 0.1% FA over 6 min. The product was
obtained as white solid after lyophilization (3.54 mg, 1.86 μmol, 53%).
ESI-MS (m/z) for [M + H]+ (calcd): 1903.96 (1902.96), (m/z) for
[M + 2H]2+ (calcd): 952.48 (951.98).
Synthesis of TfR-Ps−DOTA−PEG3−Maleimide 22a. TfR-Ps (50
μmol) was synthesized on solid support as described above. Fmoc−
Lys(tris-t-Bu-DOTA)−OH (23) was introduced using the standard
coupling protocol after removal of the N-terminal Fmoc protecting
group followed by coupling of Fmoc−PEG3−OH and maleimido-
hexanoic acid. The peptide was cleaved from the solid support,
deprotected, and purified by semipreparative HPLC using a gradient of
0−40% MeCN + 0.1% FA over 6 min. The product was obtained as a
white solid after lyophilization (18.3 mg, 9.91 μmol, 20%). ESI-MS
(m/z) for [M + H + Nasalt + Kcomplexed]+ (calcd): 1909.88 (1909.93),
2+
(m/z) for [M + 2H + Nasalt + Kcomplexed
]
(calcd): 954.94 (954.97).
General Synthesis of Carrier Peptide−FP-T Conjugates 2−
22. To a solution of peptide maleimides 2a−22a (2.5−10 μmol) in PB
(phosphate buffer, 0.1 M, pH 4.0)/MeCN 1:1 (250 μL) was added a
solution of FP-T (1, 1.25 equiv) in MeCN, and the pH of the solution
was adjusted to 6.7−7.2 using PB (0.5 M, pH 7.2, ∼ 50 μL). After 10
min, the reactions were complete, and the crude products were
purified by semipreparative HPLC. The products were isolated as
white solids after lyophilization. Gradients used for HPLC purification
and overall synthesis yields are given below.
2: gradient, 0−40% MeCN in 6 min; yield, 71%; MALDI-MS (m/z)
for [M + H]+ (calcd): 2291 (2289). 3: gradient, 15−40% MeCN in 6
min; yield, 67%; ESI-MS (m/z) for [M + 4H]4+ (calcd): 478.54
(478.53). 4: gradient, 15−50% MeCN in 6 min; yield, 56%; ESI-MS
(m/z) for [M + 2H]2+ (calcd): 732.37 (732.37). 5: gradient, 20−45%
MeCN in 6 min; yield, 62%; ESI-MS (m/z) for [M + 2H]2+ (calcd):
1031.51 (1031.50), (m/z) for [M + 3H]3+ (calcd): 688.01 (688.00). 6:
gradient, 30−80% MeCN in 6 min; yield, 36%; MALDI-MS (m/z) for
[M + H]+ (calcd): 3414.8 (3413.9). 7: gradient, 15−40% MeCN in 6
min; yield, 69%; ESI-MS (m/z) for [M + 4H]4+ (calcd): 733.64
(719.40), (m/z) for [M + 5H]5+ (calcd): 564.52 (575.72). 8: gradient,
15−40% MeCN in 6 min; yield, 66%; ESI-MS (m/z) for [M + 3H]3+
(calcd): 619.02 (619.01). 9: gradient, 10−50% MeCN in 6 min; yield,
46%; ESI-MS (m/z) for [M + 3H]3+ (calcd): 1086.86 (1086.85), (m/
z) for [M + 4H]4+ (calcd): 815.39 (815.39). 10: gradient, 0−40%
MeCN in 6 min; yield, 65%; MALDI-MS (m/z) for [M + H]+ (calcd):
1550.4 (1548.8). 11: gradient, 0−40% MeCN in 6 min; yield, 64%;
MALDI-MS (m/z) for [M + H]+ (calcd): 1610.7 (1608.8). 12:
gradient, 0−40% MeCN in 6 min; yield, 56%; MALDI-MS (m/z) for
[M + H]+ (calcd): 1669.0 (1666.9). 13: gradient, 0−40% MeCN in 6
min; yield, 69%; MALDI-MS (m/z) for [M + H]+ (calcd): 1713.1
(1710.9). 14: gradient, 10−40% MeCN in 6 min; yield, 57%; ESI-MS
(m/z) for [M + 2H]2+ (calcd): 899.97 (899.97), (m/z) for [M +
3H]3+ (calcd): 600.32 (600.31). 15: purification performed using a
Kromasil C8 preparative column, gradient, 0−50% MeCN in 6 min;
yield, 72%; ESI-MS (m/z) for [M + 2H]2+ (calcd): 944.00 (944.00);
(m/z) for [M + 3H]3+ (calcd): 630.00 (629.66). 16: purification
performed using a Kromasil C8 preparative column, gradient, 0−50%
MeCN in 6 min; yield, 64%; ESI-MS (m/z) for [M + 2H]2+ (calcd):
1032.05 (1032.05), (m/z) for [M + 3H]3+ (calcd): 688.70 (688.37).
17: gradient, 0−50% MeOH in 6 min; yield, 40%; MALDI-MS (m/z)
for [M + H]+ (calcd): 3507.8 (3505.7). 18: gradient, 0−50% MeOH
in 6 min; yield, 70%; MALDI-MS (m/z) for [M + H]+ (calcd): 3598.0
(3593.7). 19: gradient, 0−50% MeOH in 6 min; yield, 74%; MALDI-
MS (m/z) for [M + H]+ (calcd): 3682.8 (3681.8). 20: gradient, 0−
50% MeOH in 6 min; yield, 75%; MALDI-MS (m/z) for [M + H]+
(calcd): 3859.3 (3857.9). 21: gradient, 10−65% MeCN in 6 min;
yield, 79%; ESI-MS (m/z) for [M + 2H]2+ (calcd): 1080.66 (1080.57),
(m/z) for [M + 3H]3+ (calcd): 721.05 (720.71). 22: gradient, 10−40%
MeCN in 6 min; yield, 42%; ESI-MS (m/z) for [M + 2H]2+ (calcd):
1113.04 (1113.08), (m/z) for [M + 3H]3+ (calcd): 742.73 (742.39).
Synthesis of Fmoc−Lys(tris-t-Bu-DOTA)−OH (23). To a
solution of N,N,N′,N′-tetramethyl-O-(1H-benzotriazol-1-yl)uronium
Synthesis of Nonmodified Peptides TfR-Ps (27), NTP (28),
and TAT49−57 (29). The peptides were synthesized on solid support,
deprotected, and cleaved using standard protocols as described before.
The products were purified by semipreparative HPLC and isolated as
white solids after lyophilization. Gradients used for HPLC purification,
synthesis yields, and analytical data are given below.
27: gradient, 0−40% MeCN in 6 min; yield, 86%; ESI-MS (m/z)
for [M + H]+ (calcd): 892.47 (892.47). 28: gradient, 0−30% MeCN in
6 min; yield, 73%; ESI-MS (m/z) for [M + 2H]2+ (calcd): 642.39
(642.39). 29: gradient, 0−30% MeCN in 6 min; yield, 71%; ESI-MS
(m/z) for [M + 4H]4+ (calcd): 335.72 (335.72); MALDI-MS (m/z)
for [M + H]+ (calcd): 1340.2 (1339.9).
18F Radiolabeling of 21. Standard radiolabeling conditions: To a
solution of dried [n-Bu4N]18F complex (4.1−6.2 GBq) in DMSO (500
μL), obtained as described elsewhere,61 was added a solution of 21 (25
nmol, 54 μg) in dry DMSO (2.5 μL), and the mixture was reacted for
5 min at ambient temperature without stirring. Subsequently, the
reaction mixture was added to HEPES buffer (0.1 M, pH 4, 9 mL) and
4378
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
loaded on a Waters SepPak C-18 light cartridge, previously
conditioned with ethanol (5 mL) and water (10 mL). The trapped
product was washed with water (5 mL), eluted from the cartridge with
ethanol (200 μL), and diluted with isotonic saline solution (2 mL).
The purified product solution was subjected to analytical radio-HPLC,
which showed an insufficient radiochemical purity of the product of
81−85%. Optimized radiolabeling conditions: Aqueous [18F]fluoride
in target water was passed through a SAX cartridge (Sep-Pak Accell
Plus QMA Carbonate light (46 mg)) followed by 20 mL of air, 5 mL
of anhydrous acetonitrile, and again 20 mL of air. The [18F]fluoride
was eluted with a solution of [K+⊂K2.2.2]OH− in anhydrous
acetonitrile (0.5 mL) containing 110 μmol Kryptofix 2.2.2 and 100
μmol KOH. To this solution were added first a solution of oxalic acid
(1M, 21 μL) and subsequently a solution of 21 (25 nmol, 54 μg) in
anhydrous DMSO (2.5 μL), and the mixture was reacted for 5 min at
ambient temperature without stirring. Subsequently, the reaction
mixture was added to HEPES buffer (0.1 M, pH 2, 9 mL) and loaded
on a Waters SepPak C-18 light cartridge. The trapped product was
washed with water (5 mL), eluted from the cartridge with ethanol
(200 μL), and finally diluted with isotonic saline solution (2 mL). The
purified product solution was subjected to analytical radio-HPLC. The
radiolabeled product, [18F]21, was found to be ≥98% pure and was
obtained in specific activities of up to 80 GBq/μmol and radiochemical
yields of up to 60% in a total synthesis time of 25 min.
amine; DOTA, 1,4,7,10-tetraaza-cyclododecane-N,N′,N″,N‴-
tetraacetic acid; DMF, N,N-dimethylformamide; DMFP,
desmethoxy-fallypride; ESI, electrospray ionization; FP, fall-
ypride; GPCR, G protein-coupled receptor; HBTU, O-
(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium hexafluoro-
phosphate; HP, haloperidol; HPLC, high-performance liquid
chromatography; KRB, Krebs ringer buffer; MALDI, matrix-
assisted laser desorption/ionization; NIR, near-infrared; PEG,
poly(ethylene glycol); PET, positron emission tomography; P-
gp, P-glycoprotein; RP, raclopride; SIE, solvated interaction
energy; SiFA, silicon fluoride acceptor; SPPS, solid-phase
peptide synthesis; TfR, transferrin receptor; VIP, vasoactive
intestinal peptide
REFERENCES
■
(1) Schirrmacher, R.; Bernard-Gauthier, V.; Reader, A.; Soucy, J. P.;
Schirrmacher, E.; Wangler, B.; Wangler, C. Design of brain imaging
̈
̈
agents for positron emission tomography: do large bioconjugates
provide an opportunity for in vivo brain imaging? Future Med. Chem.
2013, 5, 1621−1634.
(2) Pardridge, W. M. Re-engineering biopharmaceuticals for delivery
to brain with molecular trojan horses. Bioconjugate Chem. 2008, 19,
1327−1338.
68Ga Radiolabeling of 9, 17−20, 22, and 24. A solution of the
respective fusion molecule (10−12.5 nmol) in HEPES buffer (0.025
M, pH 4.0, 100 μL) was added to 170−290 MBq of 68Ga3+ in sodium
acetate solution obtained by fractioned elution of a 68Ge/68Ga
generator (IGG100, Eckert & Ziegler, Berlin, Germany) with HCl (0.1
M, 1.2 mL) and subsequent titration to pH 3.5−4.0 by addition of
sodium acetate solution (1.25 M, 90−95 μL). After reaction for 10
min at 99 °C, the reaction mixtures were cooled to ambient
temperature and analyzed by analytical radio-HPLC. The radiolabeled
products, [68Ga]9, [68Ga]17−[68Ga]20, and [68Ga]22 were found to
be 96−98% pure and obtained in specific activities of 19.2−28.1 GBq/
μmol.
(3) Pardridge, W. M. Drug transport across the blood-brain barrier. J.
Cereb. Blood Flow Metab. 2012, 32, 1959−1972.
(4) Pardridge, W. M. Biopharmaceutical drug targeting to the brain. J.
Drug Targeting 2010, 18, 157−167.
(5) Patel, M. M.; Goyal, B. R.; Bhadada, S. V.; Bhatt, J. S.; Amin, A. F.
Getting into the brain: approaches to enhance brain drug delivery.
CNS Drugs 2009, 23, 35−58.
(6) Smith, M. W.; Gumbleton, M. Endocytosis at the blood−brain
barrier: from basic understanding to drug delivery strategies. J. Drug
Targeting 2006, 14, 191−214.
(7) Pardridge, W. M. Blood−brain barrier delivery of protein and
non-viral gene therapeutics with molecular trojan horses. J. Controlled
Release 2007, 122, 345−348.
ASSOCIATED CONTENT
* Supporting Information
■
S
(8) Pardridge, W. M. Drug targeting to the brain. Pharm. Res. 2007,
24, 1733−1744.
Detailed description of the homology modeling of the hD2
receptor; the model prediction, docking studies, molecular
dynamics simulations, and results of these in silico studies; and
a detailed description of the cytotoxicity and Calcein assays.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(9) Gabathuler, R. Approaches to transport therapeutic drugs across
the blood−brain barrier to treat brain diseases. Neurobiol. Dis. 2010,
37, 48−57.
(10) Lee, J. H.; Engler, J. A.; Collawn, J. F.; Moore, B. A. Receptor
mediated uptake of peptides that bind the human transferrin receptor.
Eur. J. Biochem. 2001, 268, 2004−2012.
Accession Codes
(11) Boules, M.; Fredrickson, P.; Richelson, E. Bioactive analogs of
neurotensin: focus on CNS effects. Peptides 2006, 27, 2523−2533.
(12) McGonigle, P. Peptide therapeutics for CNS indications.
Biochem. Pharmacol. 2012, 83, 559−566.
(13) Cano, V.; Merino, B.; Ezquerra, L.; Somoza, B.; Ruiz-Gayo, M.
A cholecystokinin-1 receptor agonist (CCK-8) mediates increased
permeability of brain barriers to leptin. Br. J. Pharmacol. 2008, 154,
1009−1015.
(14) Kastin, A. J.; Akerstrom, V. Nonsaturable entry of neuropeptide
Y into brain. Am. J. Physiol.: Endocrinol. Metab. 1999, 276, E479−E482.
(15) Chappa, A. K.; Audus, K. L.; Lunte, S. M. Characteristics of
substance P transport across the blood−brain barrier. Pharm. Res.
2006, 23, 1201−1208.
(16) Thomas, F. C.; Taskar, K.; Rudraraju, V.; Goda, S.; Thorsheim,
H. R.; Gaasch, J. A.; Mittapalli, R. K.; Palmieri, D.; Steeg, P. S.;
Lockman, P. R.; Smith, Q. R. Uptake of ANG1005, a novel paclitaxel
derivative, through the blood−brain barrier into brain and
experimental brain metastases of breast cancer. Pharm. Res. 2009,
26, 2486−2494.
(17) Bertrand, Y.; Currie, J. C.; Demeule, M.; Regina, A.; Che, C.;
Abulrob, A.; Fatehi, D.; Sartelet, H.; Gabathuler, R.; Castaigne, J. P.;
Stanimirovic, D.; Beliveau, R. Transport characteristics of a novel
PDB ID for protein hD3 receptor in complex with antagonist
eticlopride: 3PBL.
AUTHOR INFORMATION
Corresponding Author
*Phone: +49 (0)621 383 3761; Fax: +49 (0)621 383 1910; E-
mail: Carmen.Waengler@medma.uni-heidelberg.de.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
C.W. thanks the Fonds der Chemischen Industrie for financial
support. R.S. thanks the Natural Sciences and Engineering
Council (NSERC) Canada and the FRSQ funded Quebec Bio-
Imaging Network for financial support.
ABBREVIATIONS USED
■
BBB, blood−brain barrier; BC, (+)-butaclamol; Calcein-AM,
calcein−acetoxymethylester complex; CNS, central nervous
system; CP, chlorpromazine; DIPEA, N,N-diisopropylethyl-
4379
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
peptide platform for CNS therapeutics. J. Cell. Mol. Med. 2010, 14,
2827−2839.
(18) Che, C.; Yang, G. Q.; Thiot, C.; Lacoste, M. C.; Currie, J. C.;
Demeule, M.; Regina, A.; Beliveau, R.; Castaigne, J. P. New angiopep-
modified doxorubicin (ANG1007) and etoposide (ANG1009) chemo-
therapeutics with increased brain penetration. J. Med. Chem. 2010, 53,
2814−2824.
(19) 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, 185−197.
(20) Malcor, J. D.; Payrot, N.; David, M.; Faucon, A.; Abouzid, K.;
Jacquot, G.; Floquet, N.; Debarbieux, F.; Rougon, G.; Martinez, J.;
Khrestchatisky, M.; Vlieghe, P.; Lisowski, V. Chemical optimization of
new ligands of the low-density lipoprotein receptor as potential vectors
for central nervous system targeting. J. Med. Chem. 2012, 55, 2227−
2241.
(21) Tian, X. H.; Wei, F.; Wang, T. X.; Wang, P.; Lin, X. N.; Wang,
J.; Wang, D.; Ren, L. In vitro and in vivo studies on gelatin−siloxane
nanoparticles conjugated with SynB peptide to increase drug delivery
to the brain. Int. J. Nanomed. 2012, 7, 1031−41.
(22) Temsamani, J.; Bonnafous, C.; Rousselle, C.; Fraisse, Y.; Clair,
P.; Granier, L. A.; Rees, A. R.; Kaczorek, M.; Scherrmann, J. M.
Improved brain uptake and pharmacological activity profile of
morphine-6-glucuronide using a peptide vector-mediated strategy. J.
Pharmacol. Exp. Ther. 2005, 313, 712−719.
(23) Mazel, M.; Clair, P.; Rousselle, C.; Vidal, P.; Scherrmann, J. M.;
Mathieu, D.; Temsamani, J. Doxorubicin−peptide conjugates over-
come multidrug resistance. Anticancer Drugs 2001, 12, 107−16.
(24) Rousselle, C.; Clair, P.; Lefauconnier, J. M.; Kaczorek, M.;
Scherrmann, J. M.; Temsamani, J. New advances in the transport of
doxorubicin through the blood−brain barrier by a peptide vector-
mediated strategy. Mol. Pharmacol. 2000, 57, 679−686.
(25) Dogrukol-Ak, D.; Banks, W. A.; Tuncel, N.; Tuncel, M. Passage
of vasoactive intestinal peptide across the blood−brain barrier. Peptides
2003, 24, 437−444.
(26) Lindgren, M. E.; Hallbrink, M. M.; Elmquist, A. M.; Langel, U.
Passage of cell-penetrating peptides across a human epithelial cell layer
in vitro. Biochem. J. 2004, 377, 69−76.
(27) Qin, Y.; Chen, H. L.; Yuan, W. M.; Kuai, R.; Zhang, Q. Y.; Xie,
F. L.; Zhang, L.; Zhang, Z. R.; Liu, J.; He, Q. Liposome formulated
with TAT-modified cholesterol for enhancing the brain delivery. Int. J.
Pharm. 2011, 419, 85−95.
(35) Swift, J. L.; Heuff, R.; Cramb, D. T. A two-photon excitation
fluorescence cross-correlation assay for a model ligand−receptor
binding system using quantum dots. Biophys. J. 2006, 90, 1396−1410.
(36) Ceccarini, J.; Vrieze, E.; Koole, M.; Muylle, T.; Bormans, G.;
Claes, S.; Van Laere, K. Optimized in vivo detection of dopamine
release using F-18-fallypride PET. J. Nucl. Med. 2012, 53, 1565−1572.
(37) Rominger, A.; Cumming, P.; Xiong, G. M.; Koller, G.; Boning,
G.; Wulff, M.; Zwergal, A.; Forster, S.; Reilhac, A.; Munk, O.; Soyka,
M.; Wangler, B.; Bartenstein, P.; la Fougere, C.; Pogarell, O. [F-
18]fallypride PET measurement of striatal and extrastriatal dopamine
D-2/3 receptor availability in recently abstinent alcoholics. Addict. Biol.
2012, 17, 490−503.
(38) Kegeles, L. S.; Slifstein, M.; Xu, X. Y.; Urban, N.; Thompson, J.
L.; Moadel, T.; Harkavy-Friedman, J. M.; Gil, R.; Laruelle, M.; Abi-
Dargham, A. Striatal and extrastriatal dopamine D-2/D-3 receptors in
schizophrenia evaluated with [F-18]fallypride positron emission
tomography. Biol. Psychiatry 2010, 68, 634−641.
(39) Gao, M. Z.; Wang, M.; Mock, B. H.; Glick-Wilson, B. E.; Yoder,
K. K.; Hutchins, G. D.; Zheng, Q. H. An improved synthesis of
dopamine D-2/D-3 receptor radioligands [C-11]fallypride and [F-
18]fallypride. Appl. Radiat. Isot. 2010, 68, 1079−1086.
(40) Majumdar, K. C.; Muhuri, S. Regioselective synthesis of pyrone-
annulated sulfur heterocycles by aryl radical cyclization. Synthesis 2006,
2725−2730.
(41) Majumdar, K. C.; Rafique-ul-Islam. Chemoselective synthesis of
thieno [3,2-c] [1,8]naphthyridin-4(5H-ones by tandem cyclization.
Heteroat. Chem. 2007, 18, 87−92.
(42) Liu, S. H.; Guo, Y. B.; Huang, R. Q.; Li, J. F.; Huang, S. X.;
Kuang, Y. Y.; Han, L.; Jiang, C. Gene and doxorubicin co-delivery
system for targeting therapy of glioma. Biomaterials 2012, 33, 4907−
4916.
(43) Oh, S.; Kim, B. J.; Singh, N. P.; Lai, H.; Sasaki, T. Synthesis and
anti-cancer activity of covalent conjugates of artemisinin and a
transferrin-receptor targeting peptide. Cancer Lett. 2009, 274, 33−39.
(44) Demeule, M.; Currie, J. C.; Bertrand, Y.; Che, C.; Nguyen, T.;
Regina, A.; Gabathuler, R.; Castaigne, J. P.; Beliveau, R. Involvement of
the low-density lipoprotein receptor-related protein in the transcytosis
of the brain delivery vector aAngiopep-2. J. Neurochem. 2008, 106,
1534−1544.
(45) Bertrand, Y.; Currie, J. C.; Poirier, J.; Demeule, M.; Abulrob, A.;
Fatehi, D.; Stanimirovic, D.; Sartelet, H.; Castaigne, J. P.; Beliveau, R.
Influence of glioma tumour microenvironment on the transport of
ANG1005 via low-density lipoprotein receptor-related protein 1. Br. J.
Cancer 2011, 105, 1697−1707.
(28) Qin, Y.; Chen, H. L.; Zhang, Q. Y.; Wang, X. X.; Yuan, W. M.;
Kuai, R.; Tang, J.; Zhang, L.; Zhang, Z. R.; Zhang, Q.; Liu, J.; He, Q.
Liposome formulated with TAT-modified cholesterol for improving
brain delivery and therapeutic efficacy on brain glioma in animals. Int.
J. Pharm. 2011, 420, 304−312.
(29) Rao, K. S.; Reddy, M. K.; Horning, J. L.; Labhasetwar, V. TAT-
conjugated nanoparticles for the CNS delivery of anti-HIV drugs.
Biomaterials 2008, 29, 4429−4438.
(30) Nam, Y. S.; Park, J. Y.; Han, S. H.; Chang, I. S. Intracellular drug
delivery using poly(D,L-lactide-co-glycolide) nanoparticles derivatized
with a peptide from a transcriptional activator protein of HIV-1.
Biotechnol. Lett. 2002, 24, 2093−2098.
(46) Kostikov, A. P.; Chin, J. S.; Orchowski, K.; Schirrmacher, E.;
Niedermoser, S.; Jurkschat, K.; Iovkova-Berends, L.; Wangler, C.;
̈
Wangler, B.; Schirrmacher, R. Synthesis of [F-18]SiFB: a prosthetic
̈
group for direct protein radiolabeling for application in positron
emission tomography. Nat. Protoc. 2012, 7, 1956−1963.
(47) Wangler, B.; Kostikov, A. P.; Niedermoser, S.; Chin, J. S.;
̈
Orchowski, K.; Schirrmacher, E.; Iovkova-Berends, L.; Jurkschat, K.;
Wangler, C.; Schirrmacher, R. Protein labeling with the labeling
̈
precursor [F-18]SiFA-SH for positron emission tomography. Nat.
Protoc. 2012, 7, 1964−1969.
(48) Wangler, C.; Niedermoser, S.; Chin, J. S.; Orchowski, K.;
̈
Schirrmacher, E.; Jurkschat, K.; Iovkova-Berends, L.; Kostikov, A. P.;
(31) Sharma, G.; Modgil, A.; Sun, C. W.; Singh, J. Grafting of cell-
penetrating peptide to receptor-targeted liposomes improves their
transfection efficiency and transport across blood−brain barrier model.
J. Pharm. Sci. 2012, 101, 2468−2478.
(32) Meade, A. J.; Meloni, B. P.; Mastaglia, F. L.; Knuckey, N. W.
The application of cell penetrating peptides for the delivery of
neuroprotective peptides/proteins in experimental cerebral ischemia
studies. J. Exp. Stroke Transl. Med. 2009, 2, 22−40.
(33) Malakotikhah, M.; Teixido, M.; Giralt, E. Toward an optimal
blood−brain barrier shuttle by synthesis and evaluation of peptide
libraries. J. Med. Chem. 2008, 51, 4881−4889.
(34) Swift, J. L.; Cramb, D. T. Nanoparticles as fluorescence labels: Is
size all that matters? Biophys. J. 2008, 95, 865−876.
Schirrmacher, R.; Wangler, B. One-step F-18-labeling of peptides for
positron emission tomography imaging using the SiFA methodology.
Nat. Protoc. 2012, 7, 1946−1955.
̈
(49) Wetzl, B. K.; Yarmoluk, S. M.; Craig, D. B.; Wolfbeis, O. S.
Chameleon labels for staining and quantifying proteins. Angew. Chem.,
Int. Ed. 2004, 43, 5400−5402.
(50) Iovkova, L.; Wangler, B.; Schirrmacher, E.; Schirrmacher, R.;
̈
Quandt, G.; Boening, G.; Schurmann, M.; Jurkschat, K. Para-
functionalized aryl-di-tert-butylfluorosilanes as potential labeling
synthons for F-18 radiopharmaceuticals. Chem.Eur. J. 2009, 15,
2140−2147.
(51) Wangler, C.; Nada, D.; Hofner, G.; Maschauer, S.; Wangler, B.;
̈
̈
Schneider, S.; Schirrmacher, E.; Wanner, K. T.; Schirrmacher, R.;
4380
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381

Journal of Medicinal Chemistry
Article
Prante, O. In vitro and initial in vivo evaluation of 68Ga-labeled
transferrin receptor (TfR) binding peptides as potential carriers for
enhanced drug transport into TfR expressing cells. Mol. Imaging Biol.
2011, 13, 332−341.
(52) Schirrmacher, E.; Wangler, B.; Cypryk, M.; Bradtmoller, G.;
̈
Schafer, M.; Eisenhut, M.; Jurkschat, K.; Schirrmacher, R. Synthesis of
̈
p-(di-tert-butyl[(18)f]fluorosilyl)benzaldehyde ([F-18]SiFA-A) with
high specific activity by isotopic exchange: A convenient labeling
synthon for the F-18-labeling of n-amino-oxy derivatized peptides.
Bioconjugate Chem. 2007, 18, 2085−2089.
(53) Wangler, C.; Waser, B.; Alke, A.; Iovkova, L.; Buchholz, H. G.;
̈
Niedermoser, S.; Jurkschat, K.; Fottner, C.; Bartenstein, P.;
Schirrmacher, R.; Reubi, J. C.; Wester, H. J.; Wangler, B. One-step
̈
(18)F-labeling of carbohydrate-conjugated octreotate-derivatives con-
taining a silicon-fluoride-acceptor (SiFA): in vitro and in vivo
evaluation as tumor imaging agents for positron emission tomography
(PET). Bioconjugate Chem. 2010, 21, 2289−2296.
(54) Wessmann, S. H.; Henriksen, G.; Wester, H. J. Cryptate
mediated nucleophilic F-18-fluorination without azeotropic drying.
Nuklearmedizin 2012, 51, 1−8.
(55) Niessen, K. V.; Hofner, G.; Wanner, K. T. Competitive MS
̈
binding assays for dopamine D-2 receptors employing spiperone as a
native marker. ChemBioChem 2005, 6, 1769−1775.
(56) Bubik, M.; Ott, M.; Mahringer, A.; Fricker, G. Rapid assessment
of p-glycoprotein-drug interactions at the blood-brain barrier. Anal.
Biochem. 2006, 358, 51−58.
(57) Bauer, B.; Miller, D. S.; Fricker, G. Compound profiling for p-
glycoprotein at the blood−brain barrier using a microplate screening
system. Pharm. Res. 2003, 20, 1170−1176.
(58) Weksler, B.; Romero, I. A.; Couraud, P. O. The hCMEC/D3
cell line as a model of the human blood brain barrier. Fluids Barriers
CNS 2013, 10, 16.
(59) Hofelschweiger, B. The Pyrylium Dyes: A New Class of
̈
Biolabels. Synthesis, Spectroscopy, and Application as Labels and in
General Protein Assay, Ph.D. Thesis, University Regensburg,
Regensburg, Germany, 2005.
(60) Wellings, D. A.; Atherton, E. Standard Fmoc protocols. Methods
Enzymol. 1997, 289, 44−67.
(61) Iovkova-Berends, L.; Wangler, C.; Zoller, T.; Hofner, G.;
Wanner, K. T.; Rensch, C.; Bartenstein, P.; Kostikov, A.; Schirrmacher,
R.; Jurkschat, K.; Wangler, B. t-Bu(2)SiF-derivatized D(2)-receptor
ligands: the first SiFA-containing small molecule radiotracers for
target-specific PET-imaging. Molecules 2011, 16, 7458−7479.
4381
dx.doi.org/10.1021/jm5004123 | J. Med. Chem. 2014, 57, 4368−4381