L-Type Amino Acid Transporter 1 Enables the Efficient Brain Delivery of Small-Sized Prodrug across the Blood-Brain Barrier and into Human and Mouse Brain Parenchymal Cells

Article abstract of DOI:10.1021/acschemneuro.0c00564

Membrane transporters have long been utilized to improve the oral, hepatic, and renal (re)absorption. In the brain, however, the transporter-mediated drug delivery has not yet been fully achieved due to the complexity of the blood-brain barrier (BBB). Because L-type amino acid transporter 1 (LAT1) is a good candidate to improve the brain delivery, we developed here four novel LAT1-utilizing prodrugs of four nonsteroidal anti-inflammatory drugs. As a result, all the prodrugs were able to cross the BBB and localize into the brain cells. The brain uptake of salicylic acid (SA) was improved five times, not only across the mouse BBB but also into the cultured mouse and human brain cells. The naproxen prodrug was also transported efficiently into the mouse brain achieving less peripheral exposure, but the brain release of naproxen from the prodrug was not improved. Contrarily, the high plasma protein binding of the flurbiprofen prodrug and the premature bioconversion of the ibuprofen prodrug in the mouse blood hindered the efficient brain delivery. Thus, the structure of the parent drug affects the successful brain delivery of the LAT1-utilizing prodrugs, and the small-sized LAT1-utilizing prodrug of SA constituted a successful model to specifically deliver its parent drug across the mouse BBB and into the cultured mouse and human brain cells.

Full text of DOI:10.1021/acschemneuro.0c00564

pubs.acs.org/chemneuro
Research Article
L‑Type Amino Acid Transporter 1 Enables the Efficient Brain
Delivery of Small-Sized Prodrug across the Blood−Brain Barrier and
into Human and Mouse Brain Parenchymal Cells
̈
̈
Ahmed B. Montaser,* Juulia Jarvinen, Susanne Loffler, Johanna Huttunen, Seppo Auriola,
Marko Lehtonen, Aaro Jalkanen, and Kristiina M. Huttunen
Cite This: https://dx.doi.org/10.1021/acschemneuro.0c00564
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Membrane transporters have long been utilized to
improve the oral, hepatic, and renal (re)absorption. In the brain,
however, the transporter-mediated drug delivery has not yet been
fully achieved due to the complexity of the blood−brain barrier
(BBB). Because L-type amino acid transporter 1 (LAT1) is a good
candidate to improve the brain delivery, we developed here four
novel LAT1-utilizing prodrugs of four nonsteroidal anti-inflamma-
tory drugs. As a result, all the prodrugs were able to cross the BBB
and localize into the brain cells. The brain uptake of salicylic acid
(SA) was improved five times, not only across the mouse BBB but
also into the cultured mouse and human brain cells. The naproxen
prodrug was also transported efficiently into the mouse brain
achieving less peripheral exposure, but the brain release of
naproxen from the prodrug was not improved. Contrarily, the high plasma protein binding of the flurbiprofen prodrug and the
premature bioconversion of the ibuprofen prodrug in the mouse blood hindered the efficient brain delivery. Thus, the structure of
the parent drug affects the successful brain delivery of the LAT1-utilizing prodrugs, and the small-sized LAT1-utilizing prodrug of SA
constituted a successful model to specifically deliver its parent drug across the mouse BBB and into the cultured mouse and human
brain cells.
KEYWORDS: L-type amino acid transporter 1 (LAT1), LAT1-utilizing prodrugs, intracellular brain drug delivery,
transporter-mediated brain delivery, BBB, pharmacokinetics
acid transporter 1 (LAT1) to reach their target sites in the brain.
Therefore, LAT1 utilization is generally considered very
promising regarding the effective brain delivery. Among over
20 carriers expressed on the BBB, LAT1 showed better
characteristics over the other amino acid, glucose, organic
cation/anion, choline, or monocarboxylic acid transporters.8
Despite the low transport capacity of LAT1 compared to that of
glucose transporter(s),9 LAT1 is highly and selectively ex-
pressed in both luminal and abluminal sides of the BBB.10,11 In
addition, LAT1 is also expressed in mouse brain parenchymal
cells such as astrocytes and microglia.12 Moreover, LAT1
utilization does not interrupt the brain amino acid homeo-
stasis,8,13,14 and its expression and function are not altered by the
1. INTRODUCTION
Membrane transporters are integral transmembrane proteins
expressed ubiquitously on the cellular or blood−organ interfaces
in order to control the permeability of solutes and nutrients.1 In
the blood−brain barrier (BBB), for instance, the transporters
have been estimated to constitute 10−15% of all proteins in the
neurovascular unit.2 Hence, transporters have been utilized as
drug targets to develop outstanding therapies. For example,
inhibition of glucose and sodium transporters led to the
development of clinically approved antidiabetic and diuretic
drugs, respectively, such as gliflozins and thiazides.3 Moreover,
influx transporters regulate the drug dispositions, and thus, they
are considered potential drug carriers. For example, the drug
delivery of some oral antiviral drugs, such as valacyclovir and
valganciclovir, has been improved via utilizing intestinal
transporters.4,5
Received: August 28, 2020
Accepted: November 6, 2020
Transporters expressed at the BBB have been investigated as
effective drug carriers into the brain. It has been found, for
instance, that the dopamine prodrug (levodopa), baclofen,
melphalan, gabapentin,6 and pregabalin7 utilize L-type amino
https://dx.doi.org/10.1021/acschemneuro.0c00564
© XXXX American Chemical Society
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
A

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
a
Scheme 1. Synthetic Steps for the Preparation of Prodrugs 1−4
a
Reagents and conditions: (a) SOCl2, CH2Cl2, reflux, overnight; (b) BOC-L-3-aminophenylalanine, NaOH, CH2Cl2, RT, overnight, 44−88%; (c)
TFA, CH2Cl2, RT, overnight, 51−74%.
Table 1. Physiochemical Properties and Protein Binding of the Studied Compounds
unbound drug fraction (fu, %) (n = 3, mean SD)
compound
molecular weight (Mw) (g/mol)
cLogDpH 7.4
mouse serum
mouse liver S9
mouse brain S9
prodrug 1
FLB
prodrug 2
SA
prodrug 3
IBU
prodrug 4
NAP
406.46
244.27
300.31
138.12
368.48
206.29
392.46
230.26
2.22
1.41
−0.43
−1.47
2.12
1.71
1.26
0.25
1.07 0.31
0.97 0.16
81.69 7.94
0.71 0.03
17.80 4.80
0.09 0.05
13.63 1.03
0.12 0.01
0.84 0.32
11.09 1.19
60.59 15.59
46.17 9.21
15.07 4.22
18.74 2.89
1.97 0.20
1.17 0.30
38.00 1.83
50.62 16.63
51.44 3.70
5.75 0.03
34.09 6.01
5.86 2.17
64.65 2.75
41.06 8.61
inflammatory insult.15 Therefore, LAT1 is believed to transport
its substrates across the BBB and the cellular barriers of the brain
parenchyma effectively and harmlessly.
findings, we developed four novel LAT1-utilizing prodrugs of
four nonsteroidal anti-inflammatory drugs (NSAIDs) by
conjugation to the same amino acid moiety. We investigated
the brain and liver distribution in vivo and the intracellular
uptake of these prodrugs into cultured human and mouse brain
parenchymal cells with the aim to improve the brain delivery and
the intracellular localization of the anti-inflammatory drugs.
Neuroinflammation is a hallmark of the neurodegenerative
diseases.26 Thus, our main goal is to deliver the anti-
inflammatory drugs specifically into the brain and further target
them into the activated astrocytes and microglia, which are
releasing most of the inflammatory mediators27,28 in the brain.
Consequently, neuroinflammation can be alleviated, and hence,
better therapies can be developed for neurodegenerative
diseases.
LAT1 is a nonglycosylated membrane protein and commonly
referred and shortened from a heterodimeric complex of LAT1
light chain and 4F2hc (CD98) heavy chain.16 The transport
activity of LAT1 is dependent on both heterodimeric subunits
but independent of the surrounding media such as pH or
sodium. LAT1 exchanges large and neutral amino acids such as
L-leucine (L-Leu), L-histidine, and L-phenylalanine (L-Phe) as
well as thyroid hormones across the plasma membranes. Since
LAT1 has been first cloned16 several studies have investigated
the structural features for optimal LAT1 utilization.17−19 Hence,
a prodrug approach has emerged that suggests the conjugation
of the parent drugs to natural LAT1 substrates such as L-Phe or
L-tryptophan.19−25 As a consequence, LAT1-utilizing prodrugs
of ketoprofen, ferulic acid, dopamine, and valproic acid were
developed based on the published quantitative structure−
activity relationship (QSAR) model.19 The in vitro cellular
uptake, in situ rat brain perfusion, and pharmacokinetics studies
of the previously mentioned prodrugs have paved the way to
identify the best amino acid moiety and conjugation site for
optimal LAT1 utilization. However, it is still unclear whether the
structures of the parent drug play a role in the brain uptake and
the intracellular localization in brain cells following the LAT1
prodrug approach. Therefore, and in light of these latest
2. RESULTS
2.1. Drug Design. Our research group has previously
reported five LAT1-utilizing prodrugs of ketoprofen having
structurally different promoieties in their prodrug structure.20
According to these results and in accordance to our previously
published 3D-QSAR model,19 the phenylalanine attached at the
meta position of the aromatic ring with an amide bond to the
parent drug has the highest ability to utilize LAT1 for cellular
and brain uptake after intraperitoneal administration to mice.
Therefore, in the present study, this L-Phe-promoiety was
B
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Figure 1. Membrane transport in SV40 cells is demonstrated by (1) the protein expression levels of LAT1, 4F2hc, GLUT1, Na+K+ ATPase, ASCT1,
□
■
OATP1A2, OATP2B1, OATP1A2, OATP2B1, OATP1B3, MRP1, and MRP4 in the crude membrane ( ) and plasma membrane ( ) fractions. The
results are presented as mean SD (n = 3).
derivatized by a similar manner with four different anti-
inflammatory drugs, namely, flurbiprofen (FLB), salicylic acid
(SA), ibuprofen (IBU), and naproxen (NAP). The amide bond
was selected to produce prodrugs that are stable enough in the
mouse first-pass metabolism enabling the highest possible brain
drug delivery without premature bioconversion. We have
previously reported that this is impossible to achieve in mice
with, e.g., ester prodrugs.23,29,30 Other alternatives would have
been to produce prodrugs with carbamate or carbonate bonds;
however, due to the acid group of the selected parent drugs, only
amide or ester prodrugs are possible to be designed. The
prodrugs 1−4 were synthesized as previously described21 with
overall good yields after two steps (Scheme 1). Briefly, FLB, SA,
IBU, or NAP was converted to corresponding acid chlorides by
refluxing the acids with thionyl chloride. The acid chlorides were
then coupled with tert-butyloxycarbonyl (BOC)-protected 3-
aminophenylalanine, prepared as previously described,21 in the
presence of sodium hydroxide. The intermediates were finally
deprotected by trifluoroacetic acid to give desired prodrugs 1−4.
2.2. Pharmaceutical Properties of the Studied Com-
pounds. The physiochemical properties of the prodrugs and
their nonspecific protein binding were studied and compared to
those of their parent drugs (Table 1). The calculated
distribution coefficients (cLogDpH 7.4) of all the prodrugs (2 <
4 < 3 < 1) were higher than those of their corresponding parent
drugs. Furthermore, cLogD values correlated significantly with
the unbound fractions of the prodrugs (Y = −18.07 4.2 cLogD
lipophilic prodrug (prodrug 2, cLogD = −0.43) showed the
highest percentage of unbound portions in mouse plasma (81.69
7.94), brain S9 (50.62 16.63), and liver S9 (60.59 15.59)
fractions. In contrast, the highest lipophilic prodrug (prodrug 1,
cLogD = 2.22) showed the lowest percentage of unbound
portions in mouse plasma (1.07 0.31), brain S9 (1.17 0.30),
and liver S9 (0.84 0.32) fractions.
2.3. Chemical and Enzymatic Stability of the Studied
Prodrugs. The stability and bioconversion of the prodrugs were
followed over a period of 6 h in mouse and human plasma as well
as in liver and brain S9 subcellular fractions. The amounts of the
prodrugs and the released parent drugs were quantified after
incubation in different conditions at several time points by liquid
chromatography−tandem mass spectrometric methods (LC−
MS/MS) as described in Liquid Chromatography−Tandem
Mass Spectrometry Analysis section. Prodrugs 2−4 were
completely stable in all the conditions (supplementary data,
Table S1). However, prodrug 1 was completely stable only in
mouse and human plasma but exhibited about 30% degradation
in mouse and human S9 liver fractions as well as in mouse brain
after 6 h incubation. Because the degradation percentage was
almost the same among all media, it is likely that the degradation
was chemical and not enzymatic.
2.4. Membrane Transporters in the Immortalized
Human Microglia. A functional LAT1 transporter was
identified earlier in the primary mouse astrocytes and BV2
cells.12 Here, we evaluated the expression of LAT1/4F2hc
transporter as well as some common influx and efflux
transporters such as ASCT1, ASCT2, OATP1A2, OATP2B1,
+ 39.20
(supplementary data, Figure S1). For instance, the least
6.9, R2 = 0.9059) in brain homogenates
C
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
OATP1B3, MRP1, and MRP4 in the human immortalized
microglia (SV40) using the LC−MS/MS method described in
the Membrane Transporters in the Immortalized Human
Microglia section. The amounts of LAT1 light subunit (1.37
0.19 fmol/μg protein) and 4F2hc heavy subunit (1.19 0.06
fmol/μg protein) in the crude membrane fraction of SV40 cells
were almost equal (Figure 1A). In the plasma membrane
fraction, however, the amount of LAT1 (1.52 0.3 fmol/μg
protein) was slightly higher than 4F2hc (0.73 0.34 fmol/μg
protein). Other influx transporters such as ASCT1 and efflux
transporter such as MRP1 were markedly expressed (Figure
1A), whereas the expressions of ASCT2, OATP1A2, OATP2B1,
OATP1B3, and MRP4 were low or under the detection limits.
Additionally, both membrane transporters GLUT1 and Na+/
K+ATPase, which were used as markers of consistent cellular
fractionation,31 were detected in good yield and low variation
between the replicates.
Moreover, we evaluated the function of LAT1 in SV40 cells
via a concentration-dependent uptake study using [14C]-L-Leu
as a probe substrate (Figure 1 B−D) and in the presence and
absence of a selective LAT1 inhibitor (KMH-233).24 The LAT1
inhibitor was able to inhibit the uptake of [14C]-L-Leu
significantly (92 0.72% inhibition) (Figure 1B). In addition,
the uptake of [14C]-L-Leu was linear over the range 0.76−20 μM
and was saturated at higher concentrations (>100 μM) with a
(supplementary data, Figures S2, S3, and S4). Additionally,
the Eadie-Hofstee plot revealed that the prodrugs were able to
utilize a secondary transport system at the higher concentrations
(supplementary data, Figures S5, S6, and S7). This was
particularly seen for prodrugs 1, 3, and 4 in SV40 cells as
compared to the other cell types (Table 3). However, prodrug 2
showed a consistent uptake pattern in all the cell types.
Moreover, prodrugs 3 and 4 showed significantly lower
transport capacity as compared to those of the other prodrugs
in astrocytes (Vmax = 2.6 and 0.4 pmol/(min × mg protein),
respectively) and in BV2 cells (Vmax = 9.6 and 3 pmol/(min ×
mg protein), respectively).
Furthermore, all the prodrugs were able to utilize LAT1 in
higher affinity (Km = 1.7−20.8 μM) values than those of L-Leu
(Km = 26.7−85.8 μM) in all the cell types (Table 4).
2.6. Pharmacokinetic Study in Mice. As the prodrugs
showed promising in vitro cellular uptake, this encouraged us to
examine their in vivo pharmacokinetics in mice. After
administrating a bolus intraperitoneal (i.p.) injection of 25
μmol/kg to the mice, we followed the drug’s concentrations in
the plasma, brain, and liver over a period of 5 h. All prodrugs and
their parent drugs were rapidly absorbed to the systemic
circulation. They reached the peak concentration in plasma
(Cmax, plasma = 55−205 nmol/mL) already 5−10 min after drug
administration (Table 4). Importantly, all prodrugs rapidly
penetrated the brain (within 2−30 min after drug admin-
istration) with peak concentrations varying between 0.2 and 1.2
nmol/g brain (Figure 3 and Table 4). At each time point, the
drug concentration ratio Cbrain/Cplasma was higher for prodrugs 1,
2, and 4 than that of their parent drugs (Figure 4) indicating
improved plasma-to-brain distribution. Except for prodrug 3,
which showed slow brain distribution, and its Cbrain/Cplasma ratios
remained lower than those of the parent drug IBU for up to 120
min.
V
max value of 530 pmol/(min × mg protein) and Km value of 26.7
(Figure 1C). Moreover, the Eadie-Hofstee plot did not show any
other transport system for [14C]-L-Leu in SV40 cells (Figure
1D).
2.5. Lat1 Utilization in the Cultured Brain Cells. The
cellular kinetics of the parent and their prodrugs were evaluated
against the functional LAT1 transporter in the primary mouse
astrocytes, BV2 and SV40 cells. The binding affinity and
transport capacity were evaluated by the ligand-competition
uptake and total cellular uptake studies, respectively. As
expected, the parent drugs were not able to inhibit the uptake
of [14C]-L-Leu in any of the studied cell types. In contrast, all the
prodrugs inhibited the uptake of [14C]-L-Leu with IC50 values in
the low micromolar range (0.4−54 μM; Table 2). While
Moreover, to better understand the distribution kinetics
between plasma and different organs, we calculated the
distribution coefficients for brain and liver (Kp and Kp,u values)
in respect to the total or the unbound concentrations in plasma,
respectively (Table 4). In the case of total brain exposure and
regardless of the plasma protein binding, the AUC ratios
between the brain and plasma varied between 0.003 and 0.03
among prodrugs (Table 4). Notably, prodrugs 1 and 4 showed
higher Kp values (0.02 and 0.03) than those of their
corresponding parent drugs (0.006) indicating higher total
brain exposure. In contrast, prodrug 2 showed a slightly lower Kp
value as compared to that of SA, while prodrug 3 showed a
considerably lower Kp value as compared to that of IBU.
However, because the Kp value overlooks the differences in
plasma protein binding, it is crucial also to calculate the Kp,u
values. Prodrug 1 had a substantially higher Kp,u_brain value (2.7
fold) than that of its parent drug FLB, indicating improved BBB
penetration, because a higher fraction of the unbound prodrug 1
in plasma had distributed into the brain. Contrarily, prodrugs 2,
3, and 4 had lower Kp,u_brain values than those of their parent
drugs.
Table 2. IC50 values of L-Leucine inhibition by the prodrugs
IC50 (μM) of L-Leucine uptake
compound
astrocyte
BV2
SV40 cell
prodrug 1
prodrug 2
prodrug 3
prodrug 4
2.88 1.46
21.30 1.24
17.66 1.17
7.66 1.13
6.00 1.47
18.88 1.62
13.02 1.21
10.16 1.14
0.43 1.45
21.10 1.33
54.17 1.33
16.88 1.88
prodrug 1 was the most potent to utilize LAT1, the other
prodrugs showed similar and consistent potencies in all the cell
types. Moreover, the selective LAT1 inhibitor (KMH-233)
significantly reduced the uptake of prodrug 2 (25, 50, and 100
μM) and prodrug 1 (50 and 100 μM) in all the cell types (Figure
2). In contrast, the uptake of prodrugs 3 and 4 was not affected
by the LAT1 inhibitor, and that was consistent in all the cell
types.
2.5.1. LAT1 Transport Capacity. The total cellular uptake is
another kinetic parameter that directly explains the transport
capacity in the cultured cells. After 30 min incubation with the
cells, the uptake of the studied compounds was measured by
LC−MS/MS. Interestingly, the prodrugs but not the parent
drugs were transported efficiently into all the cell types
However, the improved brain delivery of the prodrug across
the BBB does not merely guarantee improved efficacy, because
the prodrug still needs to release the parent drug in the brain.
Importantly, prodrug 2 released its parent drug SA specifically
and considerably in the mouse brain, as indicated by a 5-fold
higher Kp,u_brain value than that of SA itself.
Similarly, prodrug 4 was able to release NAP in mouse brain
but to a lesser extent than prodrug 2, as indicated by nearly half
D
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
■
□
Figure 2. Cellular uptake of the prodrugs in the presence ( ) and absence ( ) of the selective LAT1 inhibitor (KMH-233) in mouse astrocytes (A),
BV2 cells (B), and SV40 cells (C). The results are expressed as mean SD (n = 3) and analyzed by one-way ANOVA followed by Bonferroni’s multiple
comparison test; asterisks denote statistical significance *** P < 0.0001.
Kp,u_brain value as compared to that of NAP itself. In contrast,
brain. Interestingly, the parent drugs were not detected in the
prodrugs 1 and 3 did not release their parent drugs in the mouse
mouse liver after the prodrugs’ administration over the period of
E
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Table 3. Kinetics of prodrugs cellular uptake
V
max (pmol/(min × mg protein))
Km (μM)
transport system
astrocyte
BV2
SV40
astrocyte
BV2
SV40
prodrug 1
prodrug 2
prodrug 3
prodrug 4
[14C]-L-Leu
LAT1
other(s)
LAT1
other(s)
LAT1
other(s)
LAT1
42.3 13.0
432.4 100.0
44.5 5.0
134.4 16.0
2.6 0.0
42.6 6.0
0.4 0.0
7.3 0.0
26.3 6.0
234.2 0.0
3133.0 1.0
39.4 0.0
3.2 3.0
262.5 23.0
4.5 1.0
39.3 14.0
1.7 0.0
2.0 1.0
4.8 1.0
306.7 42.0
2.2 1.0
a
NI
NI
27.3 2.0
108.3 5.0
9.6 1.0
3.5 0.0
47.2 5.0
2.9
134.8 5.0
19.8 2.0
75.2 0.0
NI
3.4 1.0
NI
55.1 9.0
232.5 8.0
1.7 1.0
117.5 29.0
3.0 1.0
NI
13.0 3.0
NI
20.8 7.0
NI
4.3 2.0
NI
other(s)
LAT1
209.0 38.0
65.9 14.0
b
b
b
b
2920.0 400.0
840.0 100.0
530.0 40.0
85.8 19.0
26.7 6.0
a
b
NI = not identified due to insufficient data. Published elsewhere.12
5 h. At the same time, prodrugs 1, 2, and 4 released small
amounts of their corresponding parent drugs (3, 0.7, and 4%,
respectively) in the mouse plasma, while prodrug 3 already
released 26% of its parent drug IBU in the mouse plasma.
3.1. Prodrug’s Synthesis, Physiochemical Properties,
and Protein Binding. We selected structurally different
NSAIDs such as FLB, SA, IBU, and NAP due to the differences
in their lipophilicities (cLogDpH 7.4 = −1.47−2.2, Table 1) and
molecular weights (138−244 Da, Table 1). The LAT1-utilizing
prodrugs were then synthesized by conjugating L-Phe at the
meta position to the carboxylic group of the parent drug via an
amide bond, as described above and previously reported.20 One
known factor that limits the brain exposure to CNS drugs is the
high plasma protein binding that decreases the unbound fraction
of the drugs in plasma. NSAIDs are known to be extensively
bound to plasma proteins as reported earlier37 and also shown
here (Table 1), which limits their brain uptake. The prodrugs
were less bound to the mouse plasma proteins than their parent
drugs (Table 1). Because the ionized acids were shown to be
highly bound to plasma proteins,38 conjugating the acidic parent
drugs with the neutral phenylalanine could explain the decrease
in the plasma protein binding between the prodrugs and their
parent drugs.
3.2. Membrane Transporters in the Immortalized
Human Microglia. To investigate whether the drug delivery
via LAT1 utilization can also be applied to human brain cells, we
studied the expression and function of LAT1 in SV40 cells. We
found that both LAT1/4F2hc subunits were expressed in the
crude and plasma membranes of SV40 cells. However, the LAT1
expression levels were almost 4-fold higher in primary mouse
astrocytes and BV2 12 than in the human SV40 cells, which was
in agreement with the human-to-mouse LAT1 ratio in the brain
microvessels.39 The transporters detected in the crude
membrane fraction were enriched in the plasma membrane
fraction except for 4F2hc protein. This indicates that these
transporters, except for 4F2hc protein, were mainly expressed in
the plasma membranes of the SV40 cells. Moreover, SV40 cells
expressed ASCT1, a transporter for small neutral amino acid,
which supposedly will not affect the function of LAT1 that
transports large neutral amino acids. Besides, the expression of
the three quantified OATPs and ASCT2 were very low or under
the detection limits. Thus, SV40 cells expressed a functional
LAT1 transporter as indicated by the selective transport of the
natural LAT1 substrate (L-Leu) across the plasma membrane
(Figure 1B−D), and this transport was inhibited by the selective
LAT1 inhibitor.24 Nevertheless, SV40 cells also expressed the
multidrug resistance-associated protein 1 (MRP1), which may
interact with NSAIDs as previously proposed in the literature.40
3.3. Uptake Kinetics in the Human and Mouse Brain
Cells. The target enzymes (cyclooxygenases) of NSAIDs are
located intracellularly in the brain parenchymal cells,41 as it is the
case for many other CNS drug targets. Thus, it is essential to
3. DISCUSSION
Neuroinflammation and the brain delivery of therapeutical
agents are important issues to consider in order to combat the
neurodegenerative diseases. Epidemiological studies have
shown contradictory results about using the NSAIDs in
preventing some neurodegenerative diseases, and the clinical
trials have failed to reach their efficacy end points.32,33 These
failures may be due to the difficulties in selecting the target
population, dosing, and duration of NSAID exposure.
Furthermore, the significant peripheral side effects of NSAIDs
may restrict their long-term use. Nonetheless, there is a strong
therapeutic rationale for the use of NSAIDs in restoring the
dysregulation of glial cells characteristic of the neurodegener-
ative processes. For instance, they can modulate several
inflammatory mediators such as cyclooxygenases, γ-secretases,
and NF-kB.34 Immunomodulation such as the inhibition of the
voltage-gated K+ channel Kv1.3 can also reduce the microglia
activation and the associated inflammatory responses.35 Thus,
the improved delivery of NSAIDs or immunomodulatory drugs
across the BBB and specifically into the activated microglia and
astrocytes could also improve their pharmacological outcomes
in neurodegenerative diseases.
Transporter-mediated brain delivery via a prodrug approach
constitutes a promising way to deliver drugs into the brain.36
The improved brain delivery of the LAT1-utilizing prodrugs is
evident and reported in several studies.19−25 However, the
released amount of the parent drugs from the amide prodrugs in
the mouse or rat brain has been too low. However, the balance
between efficient brain uptake and sufficient release of the amide
and ester prodrugs vs their premature bioconversion in plasma is
very delicate. Moreover, the species differences in prodrug
activating enzymes among rodents and humans makes the
translation challenging. Therefore, we, for example, study amide
and ester prodrugs in parallel, as it is highly likely that ester
prodrugs can be suitable prodrugs in human situation. In the
present study, we examined whether the structural differences of
four NSAIDs contribute to the brain delivery of the LAT1-
utilizing amide prodrugs and the release of the parent drugs into
the mouse brain. In addition, the findings from this study can
also be applied to the other neuroprotective and immunomo-
dulatory drugs.
F
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
G
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Figure 3. Total tissue concentrations of flurbiprofen (A), prodrug 1 (B), salicylic acid (C), prodrug 2 (D), ibuprofen (E), prodrug 3 (F), naproxen (G),
■
○
and prodrug 4 (H) in the mouse brain as nmol/g tissue ( ) and in the mouse plasma as nmol/mL ( ) and the concentration of parent drugs released
▲
from the prodrugs in the brain as nmol/g tissue ( ). The results are expressed as mean SD (n = 3).
deliver the CNS drugs not only across the BBB but also into their
target sites in the brain. Therefore, we examined here the cellular
kinetics of the prodrugs in the primary mouse astrocytes and
BV2 cells as well as in the human immortalized microglia cells
(SV40). The competitive ligand-binding assays using known
LAT1 substrate or inhibitor give information about the LAT1
utilization, while the total cellular uptake shows the transport’s
capacity and detect if there are other secondary transporters.
3.3.1. LAT1 Utilization in the Cultured Brain Cells. All the
prodrugs were able to compete with [14C]-L-Leu for LAT1
H
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
○
●
Figure 4. Brain-to-plasma concentration ratios of prodrugs ( ) and their parents ( ) in mice. The results are expressed as mean SD (n = 3).
utilization. The prodrugs showed high potency to utilize LAT1
in all the cell types with IC50 values in the low micromolar range
(3−54 μM) (Table 3). This finding confirms not only that the
prodrugs can utilize LAT1 even in the presence of a natural
amino acid as previously reported12,20 but also that this
approach can be applied to noncancer human brain cells
(SV40). Additionally, the Michalis-Menten affinity constants
(Km) further confirmed the high LAT1 utilization of the
prodrugs when compared with that of L-Leu (Table 4).
prodrugs are transported into all the cultured cell types with high
affinity and high transport capacity. However, prodrugs 3 and 4
showed higher affinity for LAT1 but low transport capacity. This
indicates that the parent drug moiety may affect the transport
capacity but not the affinity of LAT1 utilization.
3.4. Pharmacokinetic Study in Mice. While the cellular
uptake of the prodrugs into the brain cells was promising, it is
essential to determine how the prodrugs are distributing
centrally and peripherally in mice. After an i.p. dose of 25
μmol/kg, the concentrations of the prodrugs and their parent
drugs were followed in the brain, liver, and plasma over a period
of 5 h. The peak concentrations in plasma (Cmax) varied
considerably between the parent drugs, ranging from 55 to 205
nmol/mL. This indicates differences in the diffusion properties
of the parent drugs from the intraperitoneal cavity into the
systemic circulation. In contrast, the prodrugs showed small
differences in their plasma Cmax values (109−176 nmol/mL)
(Table 4). This suggests that the model of LAT1 utilization
showed somehow consistent diffusion properties regardless of
the different parent drug moiety. Nevertheless, the distribution
of the prodrugs in mice and the in vivo release of the parent
drugs varied markedly.
The prodrugs, as amide conjugates of their parent drugs, are
expected to be stable in vivo. This was particularly true in the
mouse liver, as no parent drugs were detected or the release was
just under the detection limits, which would also be negligible in
that case. However, the amide prodrugs can release their parent
drugs due to the enzymatic actions of carboxylesterases,
peptidases, or proteases.42 In the mouse plasma, prodrugs 1, 2,
and 4 released their parent drugs but in very small amounts
(0.7−4%), while prodrug 3 was particularly unstable with 26%
delayed release of its parent drug IBU. Additionally, the peak
concentration of the released parent drug was detected at the 2 h
time point in the cases of prodrugs 3 and 4, at the 5 h time point
3.3.2. LAT1 Transport Capacity. The transport capacity or
the ability of the drugs to accumulate into the cells is another
parameter to determine the optimal drug uptake. The prodrugs
were transported efficiently into all the cell types after the 30 min
incubation. However, the NSAID parent drugs were not
detected in all the cell types, which is probably because of the
efflux transporters such as MRP1 that is expressed in SV40 cells
(Figure 1). Most prodrugs showed a secondary transport system
at only the higher concentrations (>50 μM). This secondary
transport system is characterized by being low-affinity and high-
capacity transporter(s) as it is only activated when LAT1 is
saturated at high concentrations (supplementary data, Figures
S5, S6, and S7). The prodrugs showed different transport
capacities via LAT1; prodrugs 1 and 2 were transported in
higher quantities than prodrugs 3 and 4. However, the
exceptional high uptake of prodrug 3 in SV40 cells (Vmax ≈ 75
pmol/(min × mg protein)) might be due to another secondary
transporter that is not expressed in mouse astrocytes and BV2
cells. Thus, prodrugs 3 and 4 were not efficiently transported
from the LAT1 recognizing cavity, and they were better binders
rather than substrates of LAT1. This also explains why the LAT1
inhibitor was able to block the uptake of prodrugs 1 and 2 but
not prodrugs 3 and 4 in all the cell types.
Taking the LAT1 utilization and transport capacity together,
prodrugs 1 and 2 showed promising uptake properties. Both
I
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
The intermediates (0.49−2.86 mmol) were dissolved in anhydrous
CH2Cl2 (ca. 10 mL) and reacted with trifluoroacetic acid (1−2 mL) by
stirring the reaction mixture at RT overnight. The solvents were
removed, and the residue was redissolved in THF and stirred with 1 M
HCl (2−4 mL) at RT for 30 min. The mixture was evaporated, and the
residue was purified by flash column chromatography eluting with 1−
99% MeOH/CH2Cl2 to yield the prodrugs 1−4 as off-white solids (51−
74%).
4.2.1. (2S)-2-Amino-3-(3-(2-(2-fluoro-[1,1′-biphenyl]-4-yl)
propanamido)phenyl)propanoic Acid 1. 1H NMR (500 MHz,
(CD3)2SO): δ ppm 10.12 (d, J = 13.25 Hz, 1H), 7.59−7.50 (m, 3H),
7.50−7.44 (m, 4H), 7.40−7.33 (m, 2H), 7.31 (s, 1H), 7.19 (t, J = 7.7
Hz, 1H), 6.95 (d, J = 7.5 Hz, 1H), 3.98 (q, J = 6.8 Hz, 1H), 3.47−3.42
(m, 1H), 3.15−3.08 (m, 1H), 2.88−2.79 (m, 1 H), 1.44 (d, J = 6.9 Hz,
3H). 13C NMR (125 MHz, (CD3)2SO): δ ppm 171.60, 169.59, 159.79
(J = 245.8 Hz), 143.87 (J = 8.2 Hz), 139.20, 138.03 (J = 7.2 Hz),
134.93, 130.60 (J = 3.5 Hz), 128.69 (J = 2.5 Hz, 2C), 128.58 (2C),
127.73, 126.58 (J = 13.4 Hz), 124.28, 123.85 (J = 3.0 Hz), 120.10,
117.48, 115.09, 114.82, 55.36, 45.23, 37.03, 18.34. MS (ESI−) for
C24H22FN2O3 (M − H)−: calcd 405.45, found 405.10. Anal. calcd for
(C24H22FN2O3 × 1.0 CH2Cl2): C, 61.11; H, 4.92; N, 5.70; found = C,
61.12; H, 5.15; N, 5.85.
4.2.2. (S)-2-Amino-3-(3-(2-hydroxybenzamido)phenyl)propanoic
Acid 2. 1H NMR (500 MHz, (CD3)2SO): δ ppm 10.78 (s, 1 H), 8.00 (d,
J = 8.1 Hz, 1H), 7.67 (d, J = 8.2 Hz, 1H), 7.52 (s, 1H), 7.38 (t, J = 8.2
Hz, 1H), 7.26 (t, J = 7.8 Hz, 1H), 7.04 (d, J = 7.5 Hz, 1H), 6.98 (d, J =
8.2 Hz, 1H), 6.91 (t, J = 7.6 Hz, 1H), 3.49−3.67 (m, 1H), 3.21−3.19
(m, 1H), 2.92−2.88 (m, 1 H). 13C NMR (125 MHz, (CD3)2SO): δ
ppm 172.98, 166.16, 158.78, 138.17, 138.03, 133.39, 129.10, 128.51,
125.04, 121.62, 119.07, 118.62, 117.29, 117.27, 55.44, 36.95. MS
(ESI−) for C16H15N2O4 (M − H)−: calcd 299.30, found 299.15. Anal.
calcd for (C16H16N2O4 × 1.09 EtOH): C, 62.29; H, 6.48; N, 7.99;
found = C, 62.73; H, 6.15; N, 7.55.
in the case of prodrug 1, and at the 10 min time point in the case
of prodrug 2 (see Tmax_plasma, Table 4). Most importantly,
prodrug 2 released a significant amount of its parent drug SA in
the mouse brain, which was estimated as 5 times higher than the
amount that reached the brain after the SA dosing itself.
Similarly, prodrug 4 released NAP in the mouse brain in an
amount estimated as half the amount that reached the brain after
the NAP dosing itself.
The brain-to-plasma partition coefficients such as Kp,brain and
Kp,u_brain explained the brain delivery of the prodrugs as
compared to that of their parent drugs. Prodrug 2 improved
the brain exposure of its parent drug SA as demonstrated by the
5 times higher Kp,u_brain value than that of SA itself. Additionally,
the total brain exposure of prodrug 4 was supposedly improved
4.5-fold over its parent drug as demonstrated by its Kp,brain value.
However, the distribution of prodrug 4 into the mouse brain was
limited due to the unspecific protein binding (Table 1). The
Kp,u_brain of NAP was higher than the combined values of
prodrug 4 and the released NAP in the mouse brain (Figure 3
and Table 4). Nevertheless, both prodrugs 2 and 4 released their
parent drugs specifically in the mouse brain and not in the
periphery (plasma or liver). Hence, prodrugs 2 and 4 delivered
not only their parent drugs into the brain but also minimized
their peripheral exposure, as previously reported.20 On the other
hand, prodrug 1 was not able to improve the brain delivery of its
parent drug FLB despite its high LAT1 affinity and transport
capacity. This could be mainly explained by the high nonspecific
protein binding in the mouse plasma and liver (Table 1). Thus, a
small fraction is only free in the plasma to diffuse into the brain.
Similarly, the unexpected instability of prodrug 3 in the mouse
plasma makes it difficult to speculate its brain delivery (Table 4).
4.2.3. (2S)-2-Amino-3-(3-(2-(4-isobutylphenyl)propanamido)-
1
phenyl)propanoic Acid 3. H NMR (500 MHz, (CD3)2SO): δ ppm
10.14 (d, J = 8.0 Hz, 1H), 7.55−7.46 (m, 2H), 7.29 (d, J = 7.1 Hz, 2H),
7.18 (t, J = 7.7 Hz, 1H), 7.09 (d, J = 7.1 Hz, 2H), 6.93 (d, J = 7.4 Hz,
1H), 3.87−3.80 (m, 1H), 3.51−3.45 (m, 1H), 3.14−3.06 (m, 1H),
2.86−2.79 (m, 1 H), 2.39 (d, J = 7.0 Hz, 2H), 1.79 (hept, J = 6.7 Hz,
1H), 1.38 (d, J = 6.9 Hz, 3H), 0.84 (d, J = 6.5 Hz, 6H). 13C NMR (125
MHz, (CD3)2SO): δ ppm 172.36, 169.75, 139.40, 139.37, 139.21,
137.79, 128.85 (2C), 128.58, 127.01 (2C), 124.08, 119.99, 117.46,
55.25, 45.40, 44.22, 36.96, 29.59, 22.17 (2C), 18.62. MS (ESI−) for
C22H27N2O3 (M − H)−: calcd 367.47, found 367.20. Anal. calcd for
(C22H28N2O3 × 1.4 CH2Cl2 × 0.4 MeOH): C, 57.15; H, 6.25; N, 5.60;
found = C, 56.92; H, 6.52; N, 5.72.
4.2.4. (2S)-2-Amino-3-(3-(2-(6-methoxynaphthalen-2-yl)-
propanamido)phenyl)propanoic Acid 4. 1H NMR (500 MHz,
(CD3)2SO): δ ppm 10.20 (s, 1H), 7.83−7.75 (m, 3H), 7.56−7.48
(m, 3H), 7.26 (s, 1H), 7.20−7.13 (m, 2H), 6.92 (d, J = 7.7 Hz, 1H),
4.00 (q, J = 7.0 Hz, 1H), 3.85 (s, 3H), 3.45−3.40 (m, 1H), 3.14−3.07
(m, 1H), 2.83−2.77 (m, 1H), 1.48 (d, J = 6.9 Hz, 3H). 13C NMR (125
MHz, (CD3)2SO): δ ppm 172.29, 169.57, 157.03, 139.33, 137.98,
137.07, 133.19, 129.11, 128.57, 128.35, 126.71, 126.39, 125.39, 124.11,
120.05, 118.62, 117.47, 105.68, 55.39, 55.13, 45.73, 37.04, 18.63. MS
(ESI−) for C23H23N2O4 (M − H)−: calcd 391.45, found 391.20. Anal.
calcd for (C23H23N2O4 × 1.1 CH2Cl2): C, 59.58; H, 5.21; N, 5.77;
found = C, 59.36; H, 5.57; N, 5.69.
4. EXPERIMENTAL SECTION
All reagents and materials used in the analytical analysis were
commercial of high purity analytical grade or ultrapure HPLC grade
purchased from Sigma (St. Louis, MO, USA), Acros Organics
(Waltham, MA, USA), or Merck (Darmstadt, Germany). Water was
purified using a Milli-Q Gradient system (Millipore, Milford, MA,
USA).
4.1. General Synthetic Procedures. The reactions were
monitored by thin-layer chromatography using aluminum sheets
coated with silica gel 60 F245 (0.24 mm) with suitable visualization.
Purifications by flash chromatography were performed on silica gel 60
1
(0.063−0.200 mm mesh). H and 13C nuclear magnetic resonance
(NMR) spectra were recorded on a Bruker Avance 500 spectrometer
̈
(Bruker Biospin, Fallanden, Switzerland) operating at 500.13 and
125.75 MHz, respectively, using tetramethylsilane as an internal
standard. ESI-MS spectra were recorded by a Finnigan LCQ
quadrupole ion trap mass spectrometer (Finnigan MAT, San Jose,
CA, USA) equipped with an electrospray ionization source. Over 95%
purities were confirmed for the final products by elemental analysis (C,
H, N) with a PerkinElmer 2400 Series II CHNS/O organic elemental
analyzer (PerkinElmer Inc., Waltham, MA, USA).
4.2. General Procedure for Preparing Prodrugs 1−4. Prodrugs
1−4 were prepared according to the literature procedure.21 SA, FLB,
IBU, or NAP (0.71−2.17 mmol) in anhydrous CH2Cl2 (10−20 mL)
was reacted with SOCl2 (1.07−4.34 mmol) in a microwave reactor
(Biotage Initiator, Biotage AB, Uppsala, Sweden) at 100 °C for 60 min
under Ar atm. The reaction mixture was evaporated, and the residue
was redissolved in CH2Cl2 (10−20 mL) and reacted with t-Boc-3-
amino-L-phenylalanine (0.71−2.17 mmol) in the presence of powdered
NaOH (1.43−4.34 mmol) at room temperature (RT) under Ar atm
overnight.21 The solvent was removed, and the residue was purified by
flash column chromatography eluting with 1−30% MeOH/CH2Cl2 to
yield the prodrug intermediates (44−88%).
4.3. Physiochemical Properties and Nonspecific Protein
Binding. The unbound fractions of the studied compounds were
determined in mouse serum as well as in the S9 subcellular fractions of
the mouse liver and brain by using Rapid Equilibrium Dialysis (RED)
plates (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Briefly, the
studied compounds (10 μM) were spiked to 100 μL of mouse S9
subcellular fraction or serum and added to the reaction chamber. A total
of 350 μL of HBSS buffer was added to the buffer chamber of the RED
plate. The dialysis plate was incubated at 37 °C for 4 h while shaking. A
total of 50 μL of samples was taken from the reaction and buffer
chambers, and equal amounts of buffer or blank homogenate were
added, respectively, to yield identical matrices. The samples were
J
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
treated with 100 μL of ice-cold ACN (including the selected internal
standards) to precipitate the proteins, and the supernatants were
collected for LC−MS analysis after centrifugation at 12 000 × g for 10
min. The unbound drug fraction (fu,tissue) was calculated and corrected
with the homogenate dilution factor (D), as described by Kalvass et al.43
Additionally, cLogDpH7.4 values of the studied compounds were
predicted using MarvinSketch, ChemAxon software.
4.4. Chemical and Enzymatic Stabilities of the Prodrugs. The
control mouse brain and liver were collected freshly and homogenized
with Omni Bead Ruptor 24 Elite homogenizer (Omni International,
Kennesaw, GA, USA) in (1:4 w/v) 50 mM Tris-buffered saline (TBS)
(pH 7.4). The brain and liver homogenates were centrifuged at 9 000 ×
g for 20 min at 4 °C to prepare the S9 subcellular fractions. The mouse
blood was aseptically collected from control animals, while the human
plasma was supplied from the Finnish Red Cross as freshly frozen
(Helsinki, Finland). Then, the mouse plasma was prepared by
centrifugation at 1200 × g for 10 min at 4 °C. The protein
concentrations were measured for each homogenate by Bio-Rad
Protein Assay, based on the Bradford dye-binding method (EnVision,
PerkinElmer, Inc., Waltham, MA, USA).
The bioconversion and stability of the prodrugs were determined in
mouse brain, S9 subcellular fractions of mouse or human liver (Sigma-
Aldrich, St. Louis, MO, USA), and mouse or human plasma. A total
amount of 100 μM prodrugs in 2% DMSO were incubated with 1 mg/
mL protein of the above-mentioned homogenates at 37 °C. Aliquots of
100 μL were taken from the incubation mixture at appropriate intervals.
Then, the reaction was stopped, and the proteins were precipitated by
adding 100 μL of ice-cold acetonitrile (ACN). The samples were then
centrifuged at 12 000 × g for 5 min at RT, and the supernatants were
collected and analyzed by LC−MS/MS methods described in the
Liquid Chromatography−Tandem Mass Spectrometry Analysis
section. In the blank reactions (chemical stability), the biological
material was replaced with the same volume of buffer. Because pseudo-
first-order half-lives (t1/2) of bioconversion were not calculated due to
the relatively high stability of the prodrugs in vitro, the degradation was
reported herein as percentages (%), calculated as the difference in drug
concentrations between the last (6 or 24 h) and the first (0 h) time
points.
4.5. Cultured Cells. Primary astrocytes were isolated from cortices
of two-day-old C57BL/6 wild type mice as previously described.12,44,45
The cells were cultured in DMEM (+ 4500 mg/L glucose, + L-
glutamine, − pyruvate; Gibco-BRL) supplemented with 10% heat-
inactivated fetal bovine serum (GIBCO, MD, USA) and 100 U/mL
penicillin and streptomycin (EuroClone, Milan, Italy).
Immortalized BV2 cells were a generous gift from Prof. Mikko
Hiltunen, while SV40 cells were originated from Tardieu lab46 and were
a kind gift from Dr. Dora Brites and Prof. Tarja Malm. Both cell lines
were cultured in RPMI-1640 medium containing L-glutamine (2 mM),
heat-inactivated fetal bovine serum (10%), penicillin (50 U/mL), and
streptomycin (50 μg/mL). The experiments were done using cell
passaging numbers between 9 and 13.
4.6. Membrane Transporters in the Immortalized Human
Microglia. The LAT1/4F2hc transporter was quantified in the crude
and plasma membrane fractions of SV40 cells using a triple quadrupole
mass spectrometer (MSD 6495, Agilent Technologies, Santa Clara, CA,
USA) following the targeted proteomic approach. Additionally, other
influx and efflux transporters such as GLUT1, Na+/K+ ATPase, ASCT1,
ASCT2, OATP1A2, OATP2B1, OATP1B3, MRP1, and MRP4 were
quantified from the same samples. Briefly, the cellular fractions were
enzymatically digested into peptides, and a marker peptide for each
protein was selected based on the protocol described by ref 47 and
validated by ref 48. Thereafter, the proteins were quantified using the
selected/multiple reaction monitoring (SRM/MRM) mode of at least
three transitions derived from the isotopically labeled and quantified
peptides (JPT Peptide Technologies GmbH, Berlin, Germany). The
proteins were then quantified based on the ratio between the
endogenous digested peptides and the heavy-labeled peptides
(supplementary data, Table S2).
using a cell scraper and centrifuged at 250 × g at 4 °C for 5 min. The
crude and plasma membrane fractions were isolated from the cells using
the membrane protein extraction kit (BioVision Incorporated, Milpitas,
CA, USA) by following the manufacturer’s protocol. The protein
concentration was measured by Bio-Rad Protein Assay, based on the
Bradford dye-binding method (EnVision, PerkinElmer, Inc., Waltham,
MA, USA).
The membrane fractions were denatured, reduced, and S-
carboxymethylated before the digestion with LysC (Sigma-Aldrich,
St. Louis, MO, USA) and TPCK-trypsin (Promega Biotech AB, Nacka,
Sweden). Briefly, a total amount of 50 μg of protein was mixed with
denaturing solution containing 7 M Guanidine hydrochloride, 0.5 M
Tris-HCl and 10 mM EDTA-Na. Thereafter, the proteins were reduced
by dithiothreitol (1:50, w/w) and S-carboxymethylated by iodoaceta-
mide (1:20, w/w) (Sigma-Aldrich, St. Louis, MO, USA). The alkylated
proteins were precipitated by methanol/chloroform/water (4:1:3) and
centrifuged at 18 000 × g for 5 min at 4 °C. The pellet was then
resuspended in 6 M urea and mixed for 10 min at room temperature.
The samples were then diluted with 0.1 M Tris−HCl to a final
concentration of 1.2 M urea and dissolved completely by intermittent
sonication (Branson 3510, Danbury, CT, USA). The dissolved proteins
were first digested with LysC (1/100, w/w) (Sigma-Aldrich, St. Louis,
MO, USA) and 0.05% ProteaseMax (Promega Biotech AB, Nacka,
Sweden) for 3 h at room temperature. Then, the samples were spiked
with 10 μL (30 fmol) of the labeled peptides for absolute quantification
(JPT Peptide Technologies GmbH, Berlin, Germany) (supplementary
data, Table S2). The samples were further incubated with (1/100, w/
w) TPCK-Trypsin (Promega Biotech AB, Nacka, Sweden) for 18 h at
37 °C. The tryptic digestion was then quenched by adding 40 μL of 5%
formic acid. The samples were then centrifuged at 18 000 × g for 5 min
at 4 °C, and the supernatant was transferred to HPLC vial for the
analysis.
4.6.2. LC−MS/MS−Selective Reaction Monitoring (SRM) Targeted
Protein Quantification. The digested peptides were analyzed using a
UPLC system coupled with a triple quadrupole mass spectrometer with
a heated electrospray ionization source in the positive mode (UPLC
1290 and MSD 6495, Agilent Technologies, Santa Clara, CA, USA). A
total amount of 20 μL of the digested peptides (10 μg) was separated
using AdvanceBio Peptide Map 2.1 × 250 mm, 2.7 μm column (Agilent
Technologies, Santa Clara, CA, USA) and LC eluents of 0.1% formic
acid in water (A) and acetonitrile (B). The peptides were eluted
following a constant flow rate of 0.3 mL/min and a gradient of 2−7% B
for 2 min, followed by 7−30% B for 48 min, 30−45% B for 3 min, and
45−80% B for 2.5 min before re-equilibrating the column again for 4.5
min. Data was acquired using Agilent MassHunter Workstation
Acquisition and processed using Skyline software 20.1.
4.7. LAT1 Utilization in the Cultured Cells. The uptake of the
prodrugs (25, 50, 100 μM) and [14C]-L-Leu (0.76 uM) were quantified
in astrocytes, BV2, and SV40 cells in the presence and absence of the
selective LAT1 inhibitor (KMH-233).24 First, the LAT1 inhibitor (100
μM) or HBSS were preincubated with the cells for 10 min. Thereafter,
the preincubation mixture was removed, and the prodrugs or [14C]-L-
Leu were added in the presence of the LAT1 inhibitor or HBSS for 30
min. The incubation mixture was then aspirated, and the cells were
washed and lysed for 1 h with 0.1 M NaOH (250 μL) and stored at −20
°C prior to the analysis with the LC−MS/MS method as described in
the Liquid Chromatography−Tandem Mass Spectrometry Analysis
section.
The ability of the prodrugs to inhibit uptake [14C]-L-Leu was studied
in the primary astrocytes, BV2, and SV40 cells. Cells were incubated for
5 min with 0.1−1000 μM prodrugs and 0.76 μM [14C]-L-Leu
(PerkinElmer, Waltham, MA, USA) in HBSS. After washing the cells
with HBSS, the cells were lysed with 0.1 M NaOH (250 μL), and the
lysates were mixed with 1 mL of emulsifier cocktail (PerkinElmer,
Waltham, MA, USA). The radioactivity of the cell lysates was measured
by a liquid scintillation counter (MicroBeta2, PerkinElmer, MA, USA).
The concentrations at which 50% of the uptake is inhibited (IC50) were
determined for each compound using GraphPad Prism 5.
4.6.1. Extraction and Digestion of Membrane Proteins. The SV40
cells were obtained from ten confluent culture plates per one replicate
4.8. LAT1 Transport Capacity in the Cultured Cells. The cells
were seeded at a density of 1 × 105 cells/well onto 24-well plates 1 day
K
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
before the uptake experiments. The culture media was aspirated, and
the cells were washed with prewarmed HBSS solution (Hank’s balanced
salt solution) containing 125 mM choline chloride, 4.8 mM KCl, 1.2
mM MgSO4, 1.2 mM KH2PO4, 1.3 mM CaCl2, 5.6 mM glucose, and 25
mM HEPES (pH 7.4 adjusted with 1 M NaOH). A total of 500 μL of
prewarmed HBSS was preincubated with the cells at 37 °C for 10 min
before adding the studied compounds. Different concentrations of the
studied compounds (1−200 μM) were diluted in HBSS (250 μL) and
incubated with the cells for 30 min in RT. The reaction was stopped by
removing the incubation medium and adding 500 μL of ice-cold HBSS,
and the cells were washed twice with HBSS on ice. Thereafter, the cells
were lysed for 1 h with 0.1 M NaOH (250 μL) and stored at −20 °C
prior to the analysis with the LC−MS/MS method as described in the
Liquid Chromatography−Tandem Mass Spectrometry Analysis
section.
18 000 × g for 10 min at 4 °C. The supernatants were further diluted
with 50% ACN containing labetalol and diclofenac. Standard samples
were prepared the exact same way by spiking 5 μL of at least eight
different concentrations of the analytes into the matrix blank.
4.10.2. LC−MS/MS Quantification of the Studied Compounds.
The studied compounds were separated using a reversed-phase column
(Zobrax SB-C18, 50 mm × 2.1 mm, 1.8 μm; Agilent Technologies,
Santa Clara, CA, USA) and LC eluents of 0.1% formic acid in water (A)
and acetonitrile (B) for the prodrugs and SA and using 35% methanol in
water (A) and methanol (B) for FLB, NAP, and IBU. Labetalol and
diclofenac were used as internal standards for the prodrugs and parent
drugs, respectively. The injection volume was 5 μL for all the samples,
and the column temperature was 40 °C for all the prodrugs and SA
samples and 60 °C for FLB, NAP, and IBU samples. The prodrugs and
SA were then eluted following a constant flow rate of 0.3 mL/min and a
gradient of 5% B for 2 min, followed by 5−95% B for 8 min before re-
equilibrating the column again for 3 min. FLB and NAP were eluted
following a constant flow rate of 0.3 mL/min and a gradient of 60% B
for 1.5 min, followed by 60−95% B for 1.5 min and 95% B of the column
washing phase for 3 min before re-equilibrating the column again for 3
min. IBU was eluted, however, using a different constant flow rate of 0.4
mL/min and a gradient of 15% B for 1 min, followed by 15−90% B for 8
min and 90% B of column washing phase for 3 min before re-
equilibrating the column again for 3 min. The prodrugs and labetalol
were analyzed using a UPLC system coupled with a triple quadrupole
mass spectrometer with an electrospray ionization source in the positive
mode (UPLC 1290 and MSD 6410, Agilent Technologies, Santa Clara,
CA, USA) following the transitions listed in Table 5. Salicylic acid and
4.9. Pharmacokinetic Studies in Mice. Eight-week-old male
healthy mice (C57BL/6JOlaHsd) weighing 30 5 g were purchased
from Envigo, Netherlands. Animals were housed in well-ventilated
stainless-steel cages with ad-libitum consumption of tap water and food
pellets (Teklad 2016, Envigo), temperature of 22
2 °C, relative
humidity of 55 15%, and 12/12 h light−dark cycle. The procedures
were conducted under a license (ESAVI-2015-003347) approved by
the Finnish National Animal Experiment Board and in accordance with
the European Community Guidelines and Guide for the Care and Use
of Laboratory Animals. All effort was taken to minimize the number of
animals used and their suffering. Stock solutions (80 mM) of the
studied compounds were prepared in DMSO 1 day before the study and
were dosed to animals as a solution in normal saline via intraperitoneal
(i.p.) bolus administration (25 μmol/kg), while the DMSO final
concentration was 3%. The mice were anaesthetized using a mixture of
ketamine (140 mg/kg) and xylazine (8 mg/kg) (Intervet International,
Netherlands) prior to the transcardial perfusion of ice-cold
physiological saline for 1 min. Blood (for preparation of plasma),
liver, and brain samples were collected at 5, 10, 30, 60, 120, and 300 min
after injection. Brain and liver samples were immediately snap-frozen in
liquid nitrogen and stored at −80 °C. Blood samples were immediately
vortexed with 1 μL of 1000 IU heparin and centrifuged at 1500 × g for
10 min at 4 °C. Finally, the supernatants (plasma) were stored at −80
°C.
Table 5. Ionization Parameters of the Studied Compounds
Required for the LC−MS/MS Methods
collision
energy
(V)
precursor product fragmentor
compound
ion
ion
voltage (V)
polarity
flurbiprofen
prodrug 1
naproxen
243
407
229
393
205
369
137
301
294
329
199
361
170
347, 185
161
323, 161
93
255, 181
250
380
170
380
150
380
140
100
100
50
10
13
10
15
5
13
13
11
3
negative
positive
negative
positive
negative
positive
negative
positive
negative
positive
prodrug 4
ibuprofen
prodrug 3
salicylic acid
prodrug 2
4.10. Liquid Chromatography−Tandem Mass Spectrometry
Analysis. 4.10.1. Sample Preparation. The cell lysates were diluted
(1:4, v/v) and acidified with 0.1% formic acid in ACN and centrifuged
at 18000 × g for 10 min at 4 °C. For the uptake studies, the drug
concentrations were calculated from a calibration curve (1−2500 nM)
prepared the same way as the samples by spiking known standard
concentrations to control cell lysates. The protein concentrations on
each plate were determined as the mean of three samples by Bio-Rad
Protein Assay, based on the Bradford dye-binding method (EnVision,
PerkinElmer, Inc., Waltham, MA, USA). The uptake results were then
normalized to the protein content and presented as pmol/(min × mg
protein). Michaelis−Menten kinetic parameters (Vmax and Km) were
calculated using GraphPad Prism 5, where Km is the concentration that
gives half of the maximum uptake concentration (Vmax).
The frozen mouse plasma was first melted in wet ice. A total of 25 μL
was mixed carefully with 75 μL of ACN acidified with 0.1% formic acid
and incubated at 4 °C for 1 h prior to centrifugation at 18 000 × g and 4
°C for 10 min. The supernatant was further diluted with 0.1% formic
acid in 50% ACN (ACN/H2O (1:1)) containing 200 nM labetalol and
diclofenac as internal standards. Standard samples were prepared the
same way by spiking 5 μL of at least eight different concentrations of the
analytes into the matrix blank.
The frozen mouse brain and liver were first weighed while frozen and
transferred to 2 mL of Bead Ruptor bead beating tubes prefilled with 1.4
mm ceramic beads (Omni International, Kennesaw, GA, USA). Milli-Q
deionized water (Millipore, Milford, MA, USA) was added (1:4, w/v),
and the samples were homogenized using Omni Bead Ruptor 24 Elite
homogenizer coupled with the Omni BR Cryo cooling unit. Samples
handling and homogenization were conducted at a cold temperature
(≈4 °C). A total of 50 μL of tissue homogenates was then mixed
carefully with 150 μL of 0.1% formic acid in ACN and centrifuged at
a
diclofenac
labetalol
a
294, 162
70
10
a
Internal standards.
diclofenac, however, were analyzed in the negative ionization mode. On
the other hand, FLB, NAP, and IBU were analyzed using a triple
quadrupole instrument with iFunnel technology and an electrospray
ionization source in the negative mode (UPLC 1290 and MSD 6495,
Agilent Technologies, Santa Clara, CA, USA) following the transitions
listed in Table 5. The method optimizations and instrumentation are
explained further in appendix 1 in the Supplementary Information.
4.10.3. Method Validation. The LC−MS methods were validated
according to the U.S. Food and Drug Administration guidance for
method validation.49 The validation parameters of the assays were
followed such as calibration, quality control, selectivity, sensitivity,
accuracy, precision, recovery, stability, and dilution effects (Supple-
mentary Information, Appendix 1). The lower limits of quantifications
of FLB, IBU, and NAP, were 0.77, 0.65, and 0.72 ng/mL, respectively,
which is, to our knowledge, the most sensitive, easy, and quick LC−MS
method for their quantifications in biological samples.
4.11. Pharmacokinetics Modeling, Calculations, and Stat-
istical Analysis. The unbound fractions of the studied compounds in
homogenates (fu,homogenate) were calculated by equilibrium dialysis
following the instructions from the manufacturer (Thermo Scientific,
Single-Use RED Plates) and the formula
L
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
Concentration in buffer chamber
pubs.acs.org/chemneuro
Research Article
slowly in the mouse plasma, and thus, it did not improve the
brain delivery of its parent drug. Lastly, the naproxen prodrug
was transported and released its parent drug into the mouse
brain. Although the released amount of naproxen from its LAT1-
utilizing prodrug was lower than what was achieved from
naproxen itself, the naproxen prodrug released the naproxen
specifically in the mouse brain. Hence, minimal peripheral
exposure and less adverse effects can be achieved. Therefore,
when designing LAT1-utilizing brain-targeted prodrugs, the
properties of the whole compound, including molecular weight,
lipophilicity, and nonspecific binding, should be considered.
(fu,homogenate ) =
× 100%
Concentration in reaction chamber
The effect of tissue dilution of the homogenates (D) was
compensated to calculate the unbound tissue fraction (fu,tissue ) as
described by Kalvass et al.43 using the following formula:
fu,homogenate
(fu,tissue ) =
× 100%
D − (D − 1) × fu,homogenate
The brain, liver, and plasma drug concentrations were analyzed using
an add-in program in Microsoft Excel (PKSolver).50 After we followed
the noncompartmental analysis mode, several PK parameters were
generated. The area under the curve (AUC) was calculated based on
the linear trapezoidal method, while the terminal elimination slope was
calculated based on at least the last three data points. Other PK
parameters were also obtained, such as Tmax, Cmax, half-life (t1/2), and
the apparent volume of distribution (Vz/F).
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acschemneuro.0c00564.
■
sı
The brain uptake partition coefficient parameters were calculated as
Correlation between the lipophilicity and unbound
fractions of the studied compounds in the mouse tissues;
stability of the studied compounds; cellular uptake of the
studied compounds in mouse primary astrocytes, BV2,
and SV40 cells; LC−MS−SRM/MRM transitions of
absolute protein quantification for the membrane trans-
porters and examples of the signal peaks; and validation of
the quantitative LC−MS/MS methods for the studied
compounds (PDF)
described by Hammarlund-Udenaes et al.51 The values of Kp,brain and
K
p,u_brain were calculated from the following formulas:
AUCbrain,total
Kp,brain
=
AUCplasma,total
AUCbrain,total
Kp,u_brain
=
AUCplasma,unbound
Whereas the AUCtissue,unbound values were calculated from the formula
AUCtissue,unbound = AUCtissue_total × fu,tissue
AUTHOR INFORMATION
Corresponding Author
■
The cellular kinetic parameters (Vmax, Km, and IC50) were obtained
by performing nonlinear regression analysis. The statistical difference
between groups were tested by performing one-way ANOVA followed
by Bonferroni’s multiple comparison test using GraphPad Prism 5,
while probability values of <0.05 are considered statistically significant.
Ahmed B. Montaser − School of Pharmacy, Faculty of Health
Sciences, University of Eastern Finland, FI-70211 Kuopio,
Finland; orcid.org/0000-0003-4511-467X;
Phone: +358417537797; Email: ahmed.montaser@uef.fi
Authors
5. CONCLUSION
̈
Juulia Jarvinen − School of Pharmacy, Faculty of Health
The transporter-mediated brain delivery via LAT1 utilization is a
promising approach to deliver small-sized drugs not only across
the blood−brain barrier but also into the brain parenchymal cells
(astrocytes and microglia). Because cyclooxygenases are located
intracellularly, delivering the anti-inflammatory drugs into the
activated astrocytes and microglia can better help with
alleviating the neuroinflammation. This can be achieved by
developing LAT1-utilizing prodrugs that can release the active
parent drugs specifically into the target sites. Furthermore, in the
present study, we identified a functional LAT1 transporter in the
human immortalized microglia. Hence, this approach can be
further translated into a human approach. As an example, we
showed the improved localization of the LAT1-utilizing
prodrugs into the mouse and human brain parenchymal cells.
Additionally, the LAT1-utilizing prodrugs showed promising
pharmacokinetics in mice. The salicylic acid prodrug was able to
cross the BBB at a rate 5 times higher than that of salicylic acid
itself, and it releases its parent drug specifically in the mouse
brain. This targeted delivery achieved both enhanced local-
ization into brain cells and minimal peripheral exposure.
The other structurally similar LAT1-utilizing prodrugs of
flurbiprofen, ibuprofen, and naproxen also showed improved
localization into the mouse and human brain parenchymal cells.
However, their pharmacokinetics in mice were affected by other
factors such as molecular weight, protein binding, and
bioconversion. The flurbiprofen prodrug was not able to release
its parent drug in any tissue due to the high plasma and liver
protein binding. The ibuprofen prodrug released its parent drug
Sciences, University of Eastern Finland, FI-70211 Kuopio,
Finland
̈
Susanne Loffler − School of Pharmacy, Faculty of Health
Sciences, University of Eastern Finland, FI-70211 Kuopio,
Finland
Johanna Huttunen − School of Pharmacy, Faculty of Health
Sciences, University of Eastern Finland, FI-70211 Kuopio,
Finland
Seppo Auriola − School of Pharmacy, Faculty of Health
Sciences, University of Eastern Finland, FI-70211 Kuopio,
Finland
Marko Lehtonen − School of Pharmacy, Faculty of Health
Sciences, University of Eastern Finland, FI-70211 Kuopio,
Finland
Aaro Jalkanen − School of Pharmacy, Faculty of Health
Sciences, University of Eastern Finland, FI-70211 Kuopio,
Finland
Kristiina M. Huttunen − School of Pharmacy, Faculty of Health
Sciences, University of Eastern Finland, FI-70211 Kuopio,
Finland; orcid.org/0000-0002-1175-8517
Complete contact information is available at:
https://pubs.acs.org/10.1021/acschemneuro.0c00564
Author Contributions
The manuscript was written through the contributions of all
authors. A.B.M., M.L., A.J., and K.M.H. participated in research
design. A.B.M., J.J., S.L., J.H., and A.J. conducted experiments.
M
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
(10) Duelli, R., Enerson, B. E., Gerhart, D. Z., and Drewes, L. R.
(2000) Expression of large amino acid transporter LAT1 in rat brain
endothelium. J. Cereb. Blood Flow Metab. 20, 1557−1562.
(11) Scalise, M., Galluccio, M., Console, L., Pochini, L., and Indiveri,
C. (2018) The Human SLC7A5 (LAT1): The Intriguing Histidine/
Large Neutral Amino Acid Transporter and Its Relevance to Human
Health. Front. Chem. 6, 243.
(12) Huttunen, J., Peltokangas, S., Gynther, M., Natunen, T.,
Hiltunen, M., Auriola, S., Ruponen, M., Vellonen, K., and Huttunen,
K. M. (2019) L-Type Amino Acid Transporter 1 (LAT1/Lat1)-
Utilizing Prodrugs Can Improve the Delivery of Drugs into Neurons,
Astrocytes and Microglia. Sci. Rep. 9, 12860.
(13) Puris, E., Gynther, M., de Lange, E. C.M., Auriola, S.,
Hammarlund-Udenaes, M., Huttunen, K. M., and Loryan, I. (2019)
Mechanistic Study on the Use of the l-Type Amino Acid Transporter 1
for Brain Intracellular Delivery of Ketoprofen via Prodrug: A Novel
Approach Supporting the Development of Prodrugs for Intracellular
Targets. Mol. Pharmaceutics 16, 3261−3274.
(14) Markowicz-Piasecka, M., Huttunen, J., Montaser, A., and
Huttunen, K. M. (2020) Hemocompatible LAT1-inhibitor can induce
apoptosis in cancer cells without affecting brain amino acid homeo-
stasis. Apoptosis 25, 426−440.
(15) Gynther, M., Puris, E., Peltokangas, S., Auriola, S., Kanninen, K.
M., Koistinaho, J., Huttunen, K. M., Ruponen, M., and Vellonen, K.
(2019) Alzheimer’s Disease Phenotype or Inflammatory Insult Does
Not Alter Function of L-Type Amino Acid Transporter 1 in Mouse
Blood-Brain Barrier and Primary Astrocytes. Pharm. Res. 36, 17.
(16) Kanai, Y., Segawa, H., Miyamoto, K., Uchino, H., Takeda, E., and
Endou, H. (1998) Expression cloning and characterization of a
transporter for large neutral amino acids activated by the heavy chain of
4F2 antigen (CD98). J. Biol. Chem. 273, 23629−23632.
(17) Uchino, H., Kanai, Y., Kim, D. K., Wempe, M. F., Chairoungdua,
A., Morimoto, E., Anders, M. W., and Endou, H. (2002) Transport of
amino acid-related compounds mediated by L-type amino acid
transporter 1 (LAT1): insights into the mechanisms of substrate
recognition. Mol. Pharmacol. 61, 729−737.
(18) Geier, E. G., Schlessinger, A., Fan, H., Gable, J. E., Irwin, J. J., Sali,
A., and Giacomini, K. M. (2013) Structure-based ligand discovery for
the Large-neutral Amino Acid Transporter 1, LAT-1. Proc. Natl. Acad.
Sci. U. S. A. 110, 5480−5485.
A.B.M., S.A., M.L., A.J., and K.M.H. contributed to new reagents
or analytical tools. A.B.M. wrote the manuscript. S.A., M.L., A.J.
and K.M.H. revised the manuscript. All authors have given
approval to the final version of the manuscript.
Funding
The study was financially supported by the Academy of Finland
[grant numbers 294227, 294229, 307057, 311939], Sigrid
Juselius Foundation [2015−2020], and the doctoral program of
the University of Eastern Finland.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors would like to thank Prof. Mikko Hiltunen for
providing the BV2 cells, Prof. Tarja Malm for providing the
SV40 cells, Dr. Kati-Sisko Vellonen for providing the primary
astrocytes, Mrs. Tiina Koivunen for the technical assistance with
prodrug syntheses and nonspecific protein-binding studies, Mrs.
Sari Ukkonen for the technical assistance with the cellular uptake
studies, and Ms. Agathe Hugele for performing the stability
studies in the cellular homogenates. The Mass Spectrometry
Laboratory is supported by Biocentre Finland and Biocentre of
Kuopio.
ABBREVIATIONS
■
LAT1
BBB
L-type amino acid transporter 1
blood−brain barrier
liquid chromatography−tandem mass spectrom-
LC-MS
etry
UPLC
ultra performance liquid chromatography
SRM/MRM selected reaction monitoring/multiple reaction
monitoring
ACN
DMSO
acetonitrile
dimethyl sulfoxide
REFERENCES
■
(1) Lodish, H. F., Berk, A., Kaiser, C., Krieger, M., Bretscher, A.,
Ploegh, H. L., Amon, A., and Martin, K. C. (2016) Transmembrane
Transport of Ions and small Molecules. In Molecular Cell Biology, pp
474−510, W.H. Freeman-Macmillan Learning, New York.
(2) Enerson, B. E., and Drewes, L. R. (2006) The Rat BloodBrain
Barrier Transcriptome. J. Cereb. Blood Flow Metab. 26, 959−973.
(3) Cesar-Razquin, A., Snijder, B., Frappier-Brinton, T., Isserlin, R.,
Gyimesi, G., Bai, X., Reithmeier, R. A., Hepworth, D., Hediger, M. A.,
Edwards, A. M., and Superti-Furga, G. (2015) A Call for Systematic
Research on Solute Carriers. Cell 162, 478−487.
(19) Ylikangas, H., Malmioja, K., Peura, L., Gynther, M., Nwachukwu,
E. O., Leppanen, J., Laine, K., Rautio, J., Lahtela-Kakkonen, M.,
Huttunen, K. M., and Poso, A. (2014) Quantitative insight into the
design of compounds recognized by the L-type amino acid transporter 1
(LAT1). ChemMedChem 9, 2699−2707.
(20) Puris, E., Gynther, M., Huttunen, J., Petsalo, A., and Huttunen, K.
M. (2017) L-type amino acid transporter 1 utilizing prodrugs: How to
achieve effective brain delivery and low systemic exposure of drugs. J.
Controlled Release 261, 93−104.
(21) Peura, L., Malmioja, K., Laine, K., Leppanen, J., Gynther, M.,
Isotalo, A., and Rautio, J. (2011) Large amino acid transporter 1
(LAT1) prodrugs of valproic acid: new prodrug design ideas for central
nervous system delivery. Mol. Pharmaceutics 8, 1857−1866.
(4) De Vrueh, R. L. A., Smith, P. L., and Lee, C. P. (1998) Transport of
L-Valine-Acyclovir via the Oligopeptide Transporter in the Human
Intestinal Cell Line, Caco-2. J. Pharmacol. Exp. Ther. 286, 1166−1170.
(5) Ganapathy, M. E., Huang, W., Wang, H., Ganapathy, V., and
Leibach, F. H. (1998) Valacyclovir: A substrate for the intestinal and
renal peptide transporters PEPT1 and PEPT2. Biochem. Biophys. Res.
Commun. 246, 470−475.
(6) Vale, N. (2016) Amino Acids and Peptides in Medicine: Old or
New Drugs?. In Biomedical Chemistry: Current Trends and Developments,
pp 193, De Gruyter.
(7) Takahashi, Y., Nishimura, T., Higuchi, K., Noguchi, S., Tega, Y.,
Kurosawa, T., Deguchi, Y., and Tomi, M. (2018) Transport of
Pregabalin Via L-Type Amino Acid Transporter 1 (SLC7A5) in
Human Brain Capillary Endothelial Cell Line. Pharm. Res. 35, 246.
(8) Smith, Q. R. (2005) Carrier-mediated transport to enhance drug
delivery to brain. Int. Congr. Ser. 1277, 63−74.
(9) Wang, B., Hu, L., and Siahaan, T. J. (2016) Pathways for drug
delivery to the central nervous system. In Drug Delivery: Principles and
Applications, 2nd ed., pp 360−370, John Wiley & Sons, Inc..
̈
̈
̈
̈
(22) Peura, L., Malmioja, K., Huttunen, K., Leppanen, J., Hamalainen,
M., Forsberg, M. M., Rautio, J., and Laine, K. (2013) Design, Synthesis
and Brain Uptake of LAT1-Targeted Amino Acid Prodrugs of
Dopamine. Pharm. Res. 30, 2523−2537.
(23) Puris, E., Gynther, M., Huttunen, J., Auriola, S., and Huttunen, K.
M. (2019) L-type amino acid transporter 1 utilizing prodrugs of ferulic
acid revealed structural features supporting the design of prodrugs for
brain delivery. Eur. J. Pharm. Sci. 129, 99−109.
(24) Huttunen, K. M., Gynther, M., Huttunen, J., Puris, E., Spicer, J.
A., and Denny, W. A. (2016) A Selective and Slowly Reversible
Inhibitor of l-Type Amino Acid Transporter 1 (LAT1) Potentiates
Antiproliferative Drug Efficacy in Cancer Cells. J. Med. Chem. 59,
5740−5751.
̈
(25) Gynther, M., Laine, K., Ropponen, J., Leppanen, J., Mannila, A.,
̈
Nevalainen, T., Savolainen, J., Jarvinen, T., and Rautio, J. (2008) Large
N
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Neutral Amino Acid Transporter Enables Brain Drug Delivery via
Prodrugs. J. Med. Chem. 51, 932−936.
(26) Guzman-Martinez, L., Maccioni, R. B., Andrade, V., Navarrete, L.
P., Pastor, M. G., and Ramos-Escobar, N. (2019) Neuroinflammation
as a Common Feature of Neurodegenerative Disorders. Front.
Pharmacol. 10, 1008.
(27) Liddelow, S. A., and Barres, B. A. (2017) Reactive Astrocytes:
Production, Function, and Therapeutic Potential. Immunity 46, 957−
967.
(28) Streit, W. J., Mrak, R. E., and Griffin, W. S. (2004) Microglia and
neuroinflammation: a pathological perspective. J. Neuroinflammation 1,
14.
(45) Jankowsky, J. L., Fadale, D. J., Anderson, J., Xu, G. M., Gonzales,
V., Jenkins, N. A., Copeland, N. G., Lee, M. K., Younkin, L. H., Wagner,
S. L., Younkin, S. G., and Borchelt, D. R. (2004) Mutant presenilins
specifically elevate the levels of the 42 residue beta-amyloid peptide in
vivo: evidence for augmentation of a 42-specific gamma secretase. Hum.
Mol. Genet. 13, 159−170.
́
(46) Janabi, N., Peudenier, S., Heron, B., Ng, K. H., and Tardieu, M.
(1995) Establishment of human microglial cell lines after transfection
of primary cultures of embryonic microglial cells with the SV40 large T
antigen. Neurosci. Lett. 195, 105−108.
(47) Uchida, Y., Tachikawa, M., Obuchi, W., Hoshi, Y., Tomioka, Y.,
Ohtsuki, S., and Terasaki, T. (2013) A study protocol for quantitative
targeted absolute proteomics (QTAP) by LC-MS/MS: application for
inter-strain differences in protein expression levels of transporters,
receptors, claudin-5, and marker proteins at the blood−brain barrier in
ddY, FVB, and C57BL/6J mice. Fluids Barriers CNS 10, 21.
(48) Gynther, M., Proietti Silvestri, I., Hansen, J. C., Hansen, K. B.,
Malm, T., Ishchenko, Y., Larsen, Y., Han, L., Kayser, S., Auriola, S.,
Petsalo, A., Nielsen, B., Pickering, D. S., and Bunch, L. (2017)
Augmentation of Anticancer Drug Efficacy in Murine Hepatocellular
Carcinoma Cells by a Peripherally Acting Competitive N-Methyl-d-
aspartate (NMDA) Receptor Antagonist. J. Med. Chem. 60, 9885−
9904.
(29) Huttunen, K. M. (2018) Identification of human, rat and mouse
hydrolyzing enzymes bioconverting amino acid ester prodrug of
ketoprofen. Bioorg. Chem. 81, 494−503.
́
̈
̈
̈
(30) Gynther, M., Peura, L., Vernerova, M., Leppanen, J., Karkkainen,
J., Lehtonen, M., Rautio, J., and Huttunen, K. M. (2016) Amino Acid
Promoieties Alter Valproic Acid Pharmacokinetics and Enable
Extended Brain Exposure. Neurochem. Res. 41, 2797−2809.
(31) Kubo, Y., Ohtsuki, S., Uchida, Y., and Terasaki, T. (2015)
Quantitative Determination of Luminal and Abluminal Membrane
Distributions of Transporters in Porcine Brain Capillaries by Plasma
Membrane Fractionation and Quantitative Targeted Proteomics. J.
Pharm. Sci. 104, 3060−3068.
(49) U.S. Food and Drug Administration (2018) Bioanalytical Method
Validation Guidance for Industry, https://www.fda.gov/downloads/
drugs/guidances/ucm070107.pdf (accessed July 2019).
(50) Zhang, Y., Huo, M., Zhou, J., and Xie, S. (2010) PKSolver: An
add-in program for pharmacokinetic and pharmacodynamic data
analysis in Microsoft Excel. Comp. Meth. Prog. Bio. 99, 306−314.
(51) Hammarlund-Udenaes, M., Friden, M., Syvanen, S., and Gupta,
A. (2008) On the rate and extent of drug delivery to the brain. Pharm.
Res. 25, 1737−1750.
(32) Ozben, T., and Ozben, S. (2019) Neuro-inflammation and anti-
inflammatory treatment options for Alzheimer’s disease. Clin. Biochem.
72, 87−89.
(33) T. Heneka, M., P. Kummer, M., Weggen, S., Bulic, B., Multhaup,
G., Munter, L., Hull, M., Pflanzner, T., and U. Pietrzik, C. (2011)
Molecular mechanisms and therapeutic application of NSAIDs and
derived compounds in Alzheimer’s disease. Curr. Alzheimer Res. 8, 115−
131.
(34) Lleo, A., Galea, E., and Sastre, M. (2007) Molecular targets of
non-steroidal anti-inflammatory drugs in neurodegenerative diseases.
Cell. Mol. Life Sci. 64, 1403−1418.
(35) Nguyen, H. M., di Lucente, J., Chen, Y., Cui, Y., Ibrahim, R. H.,
Pennington, M. W., Jin, L., Maezawa, I., and Wulff, H. (2020)
Biophysical basis for Kv1.3 regulation of membrane potential changes
induced by P2 × 4-mediated calcium entry in microglia. Glia 68, 2377−
2394.
(36) Rautio, J., Laine, K., Gynther, M., and Savolainen, J. (2008)
Prodrug approaches for CNS delivery. AAPS J. 10, 92−102.
(37) Parepally, J. M., Mandula, H., and Smith, Q. R. (2006) Brain
uptake of nonsteroidal anti-inflammatory drugs: ibuprofen, flurbipro-
fen, and indomethacin. Pharm. Res. 23, 873−881.
(38) Kamble, S., Loadman, P., Abraham, M. H., and Liu, X. (2018)
Structural properties governing drug-plasma protein binding deter-
mined by high-performance liquid chromatography method. J. Pharm.
Biomed. Anal. 149, 16−21.
(39) Uchida, Y., Ohtsuki, S., Katsukura, Y., Ikeda, C., Suzuki, T.,
Kamiie, J., and Terasaki, T. (2011) Quantitative targeted absolute
proteomics of human blood−brain barrier transporters and receptors. J.
Neurochem. 117, 333−345.
(40) Stefan, S. M., and Wiese, M. (2019) Small-molecule inhibitors of
multidrug resistance-associated protein 1 and related processes: A
historic approach and recent advances. Med. Res. Rev. 39, 176−264.
(41) Chandrasekharan, N. V., and Simmons, D. L. (2004) The
cyclooxygenases. Genome Biol. 5, 241.
(42) Rautio, J., Kumpulainen, H., Heimbach, T., Oliyai, R., Oh, D.,
̈
Jarvinen, T., and Savolainen, J. (2008) Prodrugs: design and clinical
applications. Nat. Rev. Drug Discovery 7, 255−270.
(43) Kalvass, J. C., and Maurer, T. S. (2002) Influence of nonspecific
brain and plasma binding on CNS exposure: implications for rational
drug discovery. Biopharm. Drug Dispos. 23, 327−338.
̈
(44) Pihlaja, R., Koistinaho, J., Malm, T., Sikkila, H., Vainio, S., and
Koistinaho, M. (2008) Transplanted astrocytes internalize deposited
beta-amyloid peptides in a transgenic mouse model of Alzheimer’s
disease. Glia 56, 154−163.
O
https://dx.doi.org/10.1021/acschemneuro.0c00564
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX