Synthesis of carbamate-type caged derivatives of a novel glutamate transporter blocker

Article abstract of DOI:10.1016/j.bmc.2004.04.011

L-threo-β-Benzyloxyaspartate (L-TBOA) and (2S,3S)-3-{3-[4- (trifluoromethyl)benzoylamino]benzyloxy}aspartate (L-TFB-TBOA) are potent nontransportable blockers for glutamate transporters. We synthesized a carbamate-type coumarin derivative of L-TBOA 3a as

Full text of DOI:10.1016/j.bmc.2004.04.011

Bioorganic & Medicinal Chemistry 12 (2004) 3687–3694
Synthesis of carbamate-type caged derivatives of a novel glutamate
transporter blocker
Kiyo Takaoka,a,b Yoshiro Tatsu,c Noboru Yumoto,c Terumi Nakajimaa and
Keiko Shimamotoa,*
aSuntory Institute for Bioorganic Research, 1-1-1 Wakayamadai, Shimamoto, Osaka 618-8503, Japan
bJapan Society for the Promotion of Science, 1-1-1 Wakayamadai, Shimamoto, Osaka 618-8503, Japan
cNational Institute of Advanced Industrial Science and Technology, Midorigaoka, Ikeda 563-8577, Japan
Received 15 March 2004; revised 13 April 2004; accepted 13 April 2004
Available online 18 May 2004
Abstract—
-TFB-TBOA) are potent nontransportable blockers for glutamate transporters. We synthesized a carbamate-type coumarin
derivative of -TBOA 3a as a caged blocker and compared 3a with the corresponding ester-type analogs 1. The carbamate 3a was
L
-threo-b-Benzyloxyaspartate
(L-TBOA) and (2S,3S)-3-{3-[4-(trifluoromethyl)benzoylamino]benzyloxy}aspartate
(
L
L
less sensitive to photolysis than the ester 1 but was more stable in the aqueous solution. The [6,7-bis(carboxymethoxy)-coumarin-4-
yl]methylcarbonyl (BCMCMC) group exhibited good results both in photoreactivity and stability. Therefore, we examined pho-
tolysis of N-BCMCMC-TBOA 3b and N-BCMCMC-TFB-TBOA 4, which immediately released blockers to show glutamate uptake
inhibition.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
on a millisecond order, thus, methods with a good
kinetic resolution are required. Excellent caged gluta-
mate derivatives were reported for the investigation of
the function of glutamate.5 However, these derivatives
activate both glutamate receptors and EAATs. Since
EAATs together with glutamate receptors regulate the
fast signal transmission at excitatory synapses, caged
blocker of glutamate transporters would be promising
for methods for the selective inactivation of EAATs. In
Caged compounds whose activities are masked by a
photocleavable group are useful tools to elucidate
complex intracellular processes.1–3 The concentration
jumps of an active compound can be generated at the
desired time and position by a photochemical reaction
using caged compounds with UV-irradiation and the
technique has attracted much attention toward over-
coming the limitations in time resolution.
order to inhibit EAATs efficiently, we have developed
TBOA ( -threo-b-benzyloxyaspartate), which is a potent
nontransportable blocker for all subtypes of EAATs.6
As -TBOA did not affect receptors, including iono-
L-
L
L
-Glutamate is a major excitatory neurotransmitter in
the mammalian central nervous system (CNS). Gluta-
mate receptors are believed to be involved in learning,
memory, and pathological phenomena such as ischemia-
related neuronal death. On the other hand, glutamate
transporters, five subtypes of which were cloned as
excitatory amino acid transporters (EAAT1-5), main-
tain extracellular glutamate concentrations at low levels
to limit the receptor activation and to protect neurons
from the excitotoxicity.4 Signal transmission can occur
L
tropic and metabotropic glutamate receptors, its caged
compound, which can block the glutamate uptake at the
desired time and position by UV-irradiation, has
potential as a useful tool for investigation of EAATs. In
the previous paper, we reported the synthesis of its ester-
type caged derivatives 1–2 and showed that they rapidly
provided
L
-TBOA by UV-irradiation.7 However, they
were gradually hydrolyzed at room temperature in the
aqueous or the DMSO solution. As excess amounts of
caged compounds are generally applied to cells or tissue
culture, stability in the aqueous solution before UV-
irradiation is crucial for practical use. Here, we describe
the synthesis of caged TBOA analogs, possessing a
stable caging group even in the aqueous solution.
Keywords: Excitatory amino acid; Glutamate transporter blocker;
Caged compound; Photolysis; Carbamate.
* Corresponding author. Tel.: +81-75-962-6114; fax: +81-75-962-2115;
e-mail: shimamot@sunbor.or.jp
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.04.011

3688
K. Takaoka et al. / Bioorg. Med. Chem. 12 (2004) 3687–3694
Moreover, we synthesized a caged derivative of a novel
TBOA analog (2S,3S)-3-{3-[4-(trifluoromethyl)benzoyl-
obtained at all from the amine 11 and 7a under the same
conditions. The reactivity of the amino group of 11 was
very low probably due to the steric hindrance of two
tert-butyl esters and the benzyloxy group. In contrast,
treatment of 11 with chloroformate 7b prepared from
trichloromethylchloroformate (phosgene dimer) and 6a
provided 12a in 51% yield.11 Deprotection of 12a with
TFA followed by HPLC purification gave N-(7-carb-
oxymethoxycoumarin-4-yl)methoxycarbonyl-TBOA (N-
CMCMC-TBOA: 3a: kmax: 322 nm). We examined the
chemical stability of 3a both in the aqueous solution and
in the DMSO solution. In the aqueous buffer as well as
the DMSO solution, 3a was stable at room temperature
and no decomposed product was observed by HPLC at
least for three months (Fig. 2).
amino]benzyloxy}aspartate (L-TFB-TBOA), which was
about 50- to 1500-fold more potent than
uptake assay.8
L-TBOA in the
2. Results and discussion
As two carboxylic acids and one amino group of
-TBOA are necessary for the inhibitory activity on
L
EAATs, its activity could be masked by introduction of
a photocleavable group to the amino group (Fig. 1). A
carbamate group is generally more stable against
hydrolysis than an ester group.9 We first planned to
synthesize the carbamate-type caged
the [7-(carboxymethoxy)coumarin-4-yl]methoxycar-
bonyl (CMCMC) group, because the [7-(carboxymeth-
oxy)coumarin-4-yl]methyl (CMCM) ester2c of
-TBOA
1a showed superior sensitivity for photolysis to 2-(4,5-
dimethoxy-2-nitrophenyl)ethyl (DMNPE) esters 210 and
good solubility in the aqueous solution (PBS(+),
pH 7.4).7 The protected CMCM-OH 6a was obtained
from the corresponding aldehyde 5a2c with NaBH4
reduction and converted to 7a with 1,1-carbonyldiimi-
dazole (Scheme 1).9 Introduction of the CMCMC group
using 7a to the amino group of the protected glutamate
8 was achieved in high yield. However, the corre-
L-TBOA having
Next, we compared the photosensitivity of 3a with those
of the ester-type compounds 1. All compounds were
decomposed by the irradiation of a UV-handy lamp
(365 nm, max power at 10 cm from the lamp: 2.3 mW/
cm2). Kinetic analyses revealed the half-life of 1a, 1b,
and 3a was 3.7, 3.9, and 19.1 min, respectively (Fig. 3).
Compared to the ester-type compounds, the carbamate-
type caging group was less sensitive for the photolysis by
UV-handy lamp. Therefore, in order to increase the
photosensitivity, an additional carbomethoxy group was
introduced on the coumarin ring of 3a, expecting both
bathochromic and hyperchromic effects.2 N-[6,7-
bis(carboxymethoxy)coumarin-4-yl]methoxycarbonyl-
L
sponding carbamate derivative of L-TBOA 12a was not
Figure 1. Glutamate transporter blockers and their caged derivatives. CMCM: [7-(carboxymethoxy)coumarin-4-yl]methyl, MCM: (7-methoxy-
coumarin-4-yl)methyl, DMNPE: 2-(4,5-dimethoxy-2-nitrophenyl)ethyl, BCMCM: [6,7-bis(carboxymethoxy)coumarin-4-yl]methyl, CMCMC: [7-
(carboxymethoxy)coumarin-4-yl]methoxycarbonyl, BCMCMC: [6,7-bis(carboxymethoxy)coumarin-4-yl]methoxycarbonyl.

K. Takaoka et al. / Bioorg. Med. Chem. 12 (2004) 3687–3694
3689
Scheme 1.
Figure 3. Decomposition of caged compounds (1a, 1b, 3a, and 3b) with
UV-handy lamp irradiation. The concentrations of the remaining
caged compounds during irradiation were measured by HPLC. Values
are presented as the average of three determinations.
Figure 2. Stability of caged compounds (1a, 3a) in PBS(+) (pH 7.4).
The concentration of the caged compounds were quantified by HPLC
and values are presented as the average of two determinations.
TBOA (N-BCMCMC-TBOA: 3b) was synthesized in the
same manner as 3a (Scheme 2). The kmax (340 nm) of 3b
was shifted to longer wavelength to be closer to the
irradiation wavelength (365 nm) and the extinction
coefficient (e) significantly increased at kmax as well as at
365 nm. As expected, 3b was more rapidly decomposed
by UV-irradiation than 3a and the half-life of 3b was
improved to 7.8 min (Fig. 3), which was almost com-
parable to that of 1a. No decomposition product
from 3b was observed in the aqueous solution for
three months (data not shown), indicating the superior
Scheme 2.

3690
K. Takaoka et al. / Bioorg. Med. Chem. 12 (2004) 3687–3694
stability of carbamate-type caging groups to that of the
ester-type (Fig. 2).
irradiated samples. However, although the peak of 4
monitored by HPLC almost completely disappeared
within 30 min,
40% of the expected amount.12 Despite attempts to find
other products, no remarkable peak except for -TFB-
TBOA or BCMCM-OH could be detected either in
HPLC or in Q-TOF-MS. These observations were
consistent with our previous results for the esters 17 and
the recent report that photolysis of 6-bromo-4-(1,2-di-
hydroxyethyl)-7-hydroxycoumarin (Bhc) derivatives of
acetal gave only a half of the expected yield.13 Although
the causes of the differences remain unclear, we con-
L-TFB-TBOA reached a plateau at ca.
L
-TFB-TBOA is a novel blocker, which blocks EAATs
about 50- to 1500-fold more potently than
L
-TBOA
L
with longer duration.8 An excess amount of a caged
compound sometimes interferes in other intracellular
processes. The application dose of caged L-TFB-TBOA
would be expected to decrease and, thus, unexpected
effects could be prevented. Therefore, based on the
results above, we designed N-BCMCMC-TFB-TBOA 4
(Fig. 1). The Boc group of 13 was removed by TFA and
the resulting amino group of 14 was reacted with 7c to
give a protected caged-derivative 15. Removal of tert-
butyl esters by TFA provided the desired 4 as shown in
Scheme 3.
firmed that the blocker activity of the resulting L-TFB-
TBOA in the uptake assay was not impeded by
photolysis byproducts.14
We then examined photolysis of these compounds by
using YAG Laser (SureLite II, Continuum, Santa Clara,
CA, 355 nm, 8 ns pulse, 40 mJ).15 Using a laser pulse,
which has higher energy and much smaller focus than a
UV-handy lamp, the active compound can be generated
at the desired time and position in a moment. Therefore,
the laser pulse in a sample specimen would be more
useful for physiological applications. The chemical
yields of disappearance (Ydis), quantum yields of disap-
pearance (Udis) and e355Udis values (indication of the
efficiency of photolysis: high value reflects high effi-
ciency) with laser pulse irradiation are summarized in
Table 1.7;16 Indication values of the efficiency of pho-
tolysis (e355Udis) of 3b and 4 were larger than that of 1a,
which was not consisted with the results of half-life with
UV-handy lamp irradiation (Fig. 3) probably because
the disparity of the features of light sauces. Laser beam
The stability and the photoreactivity of 4 were then
assessed. It was stable in the aqueous solution for three
months (data not shown). In the case of UV-handy lamp
irradiation experiments, the concentration of both the
decomposed 4 and the produced
directly qualified by HPLC since
strong UV absorption unlike -TBOA. HPLC analysis
L
-TFB-TBOA can be
L
-TFB-TBOA has
L
revealed that 4 was smoothly decomposed while a new
peak, which showed the same retention time as that of
L
-TFB-TBOA (Rt: 12.8 min), increased with UV-irradi-
ation. The half-life of the decomposition of 4 and that of
the production of -TFB-TBOA were 5.5 and 5.8 min,
respectively, and both can be fitted to single exponential
equations. The production of -TFB-TBOA was also
L
L
confirmed by Q-TOF-MS as the molecular ion peak
[m=z (M+H)þ 427] was dominantly observed in the
Scheme 3.
Table 1. Spectroscopic and photolytic characteristics of the caged compounds
Compound
Laser photolysis 1 shot (40 mJ) 355 nm
UV absorption
kmax (nm)a
(e: Mꢀ1 cmꢀ1
)
Ydis
Udis
e355
e355Udis
b
b
c
d
a-CMCM-TBOA 1a
325 (11,381)e
322 (5796)
8
1
2.3
0.8
0.02
0.006
0.03
0.02
2455
738
49
4.4
363
81
N-CMCMC-TBOA 3a
N-BCMCMC-TBOA 3b
N-BCMCMC-TFB-TBOA 4
340 (15,457)
287 (5372) 342 (5047)
24 3.6
10 1.2
12,120
4052
a e: Wavelength of absorbance maximum (kmax) and extinction coefficient in 100 lM solution in PBS(+) solution.
b Ydis: Chemical yield of disappearance (%). Udis: Quantum yield of disappearance (%).
c e355: extinction coefficient at 355 nm (Mꢀ1 cmꢀ1).
d e355
U
dis: Product of the quantum yield of disappearance and extinction coefficient (indication of the efficiency of photolysis).
e emax; e355, Udis, and e355Udis values of 1 reported in Ref. 7 are corrected to these values.

K. Takaoka et al. / Bioorg. Med. Chem. 12 (2004) 3687–3694
3691
is polarized and monochromic while UV-handy lamp
has a spectrum band (330–400 nm). Moreover, the
longer irradiation time of the UV-handy lamp could in
principle offer advantages over laser pulses if the life-
times of multiple excited states of 1 are in a relatively
longer range rather than those of 3b and 4,17 although
remarkable difference of the lifetimes between the tested
compounds was not suggested at this stage. The present
result showed the excellent photosensitivity of
BCMCMC derivatives for flash photolysis using YAG
laser. Together with the photosensitivity and the sta-
bility in the aqueous solution, the ability of the
BCMCMC group as a caging group was evaluated.
We next measured the caging effects on biological
activity by the glutamate uptake inhibition assay using
EAAT2 stably expressed on MDCK cells.7 The activities
of 3a (IC50 > 100 lM) and 3b (IC50 > 100 lM) before the
Figure 5. Inhibition of [14C]Glu uptake in MDCK cells expressing
EAAT2 by N-BCMCMC-TFB-TBOA (4) before and after YAG laser
irradiation (355 nm). Nonirradiated 4 (M) was much less potent than
L
-TFB-TBOA (ꢀ). Irradiated samples of 4 restored the activity
irradiation were less potent than that of
L-TBOA
according to the frequency of laser shots (1, 8 shots). Values are pre-
sented as the mean SEM of at least three determinations.
(IC50 ¼ 1.3 0.12 lM) (Fig. 4). The effects of the car-
bamate-type caging groups were almost the same as that
of the ester (1a: IC50 ¼ 124 lM).7 The activity of
L-TFB-
TBOA (IC50 ¼ 4.1 0.6 nM)18 was effectively masked (4:
IC50 > 1000 nM) by the BCMCMC group (Fig. 5).
Moreover, 4 did not show any inhibition at under
100 nM. Therefore, compared with the effective con-
3. Conclusion
In summary, we have synthesized carbamate-type caged
blockers for glutamate transporters, N-BCMCMC-
TBOA 3b and N-BCMCMC-TFB-TBOA 4. (1) They
were more stable than the ester-type analogs in the
aqueous solution and could be stocked at least for
3 months. (2) The blocker activity of 4 was well masked
before UV-irradiation. (3) They were highly reactive for
the photolysis using a UV-handy lamp or YAG laser to
restore biological activity. (4) As the effective dose of 4
at inhibiting EAATs was lower than that of 3b, the side
effects against other intracellular processes could be
decreased. Therefore, these compounds, especially 4,
would be useful tools for the analysis of signal trans-
duction and for the elucidation of the physiological roles
of transporters.
centration of L-TFB-TBOA, much higher concentration
of 4 could be applied into biological preparations. On
the other hand, blocker activity was obviously restored
by the laser irradiation. The concentration of
L-TFB-
TBOA produced from 4 with 1 shot of the laser pulse
was estimated to be 20 nM from the uptake assay as well
as HPLC. Since it is an averaged value in the whole
sample (100 lL), the localized concentration around the
target point would be higher in the moment of the
irradiation and would be enough to inhibit uptake.
These results indicate that L-TFB-TBOA can be quickly
and effectively generated at the desired time and position
by using the combination of 4 with the laser pulse and
that 4 would be useful for the analysis of the inactivation
of EAATs.
4. Experimental
1H NMR (400 MHz) was measured on JEOL EX-400
spectrometer. Chemical shifts are reported as d values in
ppm relative to CHCl3 (7.26) in CDCl3 or DMSO (2.50)
in DMSO-d6. IR spectra were recorded on Nicolet FT/
IR spectrometer AVATAR 360. Mass spectra were ob-
tained on JEOL JMS-HX110 spectrometer for high-
resolution FAB or Micromass Q-Tof mass spectrometer
for Q-TOF-MS and Micromass MSI mass spectrometer
for ESI-MS. HPLC was carried out with a JASCO UV-
975 (detector), PU-980 (pump) high-pressure liquid
chromatography. For the purification of the caged
compounds, a column of SP-120-5-ODS-AP (Daiso Co.
Ltd. Japan), 250 mm · 20 mm ID was used and peaks
were monitored at only 254 nm to avoid decomposition.
Optical rotations were measured on a JASCO DIP-1000
digital polarimeter. Analytical and preparative TLC
were performed on Merck Silica gel 60 F254 TLC plates
Art. 5715 (layer thickness, 0.25 mm) and 5744 (0.5 mm),
respectively, and monitored by UV light (254 and
Figure 4. Inhibition of [14C]Glu uptake in MDCK cells expressing
EAAT2 by nonirradiation sample of caged compounds (1a, 3a, and 3b)
and L-TBOA.

3692
K. Takaoka et al. / Bioorg. Med. Chem. 12 (2004) 3687–3694
365 nm) or ninhydrin solution. Column chromatogra-
phy was carried out on silica gel (Silica Gel 60, partial
size 63–210 lM, Kanto Chemical).
distilled off. The residue was dissolved in CH2Cl2 (2 mL)
and the resulting solution and i-Pr2NEt (0.4 mL,
2.2 mmol) were added to a solution of compound 11
(85 mg, 0.24 mmol) in CH2Cl2 (2 mL), after which the
mixture was stirred for 10 min. The solvent was distilled
off and the residue was purified on a silica gel-plate
(hexane/EtOAc ¼ 2/1) to obtain 84 mg of compound 12a
4.1. 7-(tert-Butoxycarbonylmethoxy)-4-(hydroxymeth-
oxy)coumarin (6a)
1
(51% yield). H NMR (CDCl3) d 1.43 (s, 9H), 1.47 (s,
To a solution of 7-(tert-butoxy-carbonylmethoxy)cou-
marin-4-carbaldehyde2c (5a: 1.78 g, 5.87 mmol) in
MeOH (60 mL), NaBH4 (346.5 mg, 9.16 mmol) was
added at 0 ꢁC and the mixture was stirred for 3 h at
room temperature. After being treated with 1 N HCl
solution, the mixture was extracted with EtOAc. The
extract was purified by silica gel column chromatogra-
9H), 1.50 (s, 9H), 4.45–4.41 (m, 2H), 4.57 (s, 2H), 4.75–
4.82 (m, 2H), 5.24 (d, 2H, J ¼ 6:4 Hz), 5.67 (d, 1H,
J ¼ 9:6 Hz), 6.33 (s, 1H), 6.78 (d, 1H, J ¼ 2:4 Hz), 6.88
(dd, 1H, J ¼ 2:4, 8.8 Hz), 7.32–7.34 (m, 5H), 7.42 (d,
1H, J ¼ 8:8 Hz).
4.5. (2S,3S)-3-Benzyloxy-N-{[7-(carboxymethoxy)cou-
marin-4-yl-methoxy]carbonyl}aspartate (N-CMCMC-
TBOA: 3a)
phy (hexane/EtOAc ¼ 1/1) to obtain 1.47 g of 6a (100%
KBr
max
yield). IR m
cmꢀ1: 3500, 1755, 1697, 1610, 1394, 1154,
1090. 1H NMR (CDCl3) d 1.48 (s, 9H), 4.55 (s, 2H), 4.84
(s, 2H), 6.46 (s, 1H), 6.75 (d, 1H, J ¼ 2:4 Hz), 6.86 (dd,
1H, J ¼ 2:4, 11.2 Hz), 7.41 (d, 1H, J ¼ 8:8 Hz). MS
(ESI) m=z calcd for C16H18O6(M+H)þ 307, found 307.
Compound 12a (97 mg, 0.12 mmol) was dissolved in
CH2Cl2 (2 mL) and to the solution cooled in an ice bath
TFA (0.5 mL) was added. The mixture was stirred for
1 h at 0 ꢁC and then at room temperature for 4 h. The
solvent was distilled off and the residue was purified by
HPLC (10–30% CH3CN/H2O/0.1%TFA) to obtain
25 mg of 3a (42% yield). 1H NMR (DMSO-d6) d 4.45 (d,
1H, J ¼ 12:0 Hz), 4.50 (d, 1H, J ¼ 3:6 Hz), 4.62 (dd, 1H,
J ¼ 3:6, 10.0 Hz), 4.76 (d, 1H, J ¼ 12:0 Hz), 5.25 (d, 1H,
J ¼ 16:4 Hz), 5.33 (d, 1H, J ¼ 16:4 Hz), 6.4 (s, 1H), 6.95
(dd, 1H, J ¼ 2:4, 9.2 Hz), 6.99 (d, 1H, J ¼ 2:4 Hz), 7.28–
7.60 (m, 5H), 7.61 (d, 1H, J ¼ 9:2 Hz), 7.89 (d, 1H,
4.2. Di-tert-butyl N-{[7-(tert-butoxycarbonylmethoxy)-
coumarin-4-yl-methoxy]carbonyl}glutamate (9)
To a solution of 6a (78.3 mg, 0.31 mmol) in CH2Cl2
(5 mL), 1,1-carbonyldiimidazole (45.6 mg, 0.28 mmol)
was stirred for 3 h at room temperature, and then and
the resulting solution, di-tert-butyl glutamate hydro-
chloride (51.5 mg, 0.17 mmol) and Et3N (28 lL,
0.2 mmol) were stirred at reflux overnight. The solvent
was distilled off and the residue was purified on a silica
gel-plate (hexane/EtOAc ¼ 2/1) to obtain 103 mg of
J ¼ 9:2 Hz).
HRMS
(FAB)
m=z
calcd
for
C24H22NO12(M+H)þ 516.1142, found 516.1142. Fluo-
rescence kex=kem (1 lM): 320 nm/405 nm.
1
compound 9 (100% yield). H NMR (CDCl3) d 1.45 (s,
9H), 1.48 (s, 9H), 1.50 (s, 9H), 1.94–2.38 (m, 4H), 4.26–
4.27 (m, 1H), 4.58 (s, 2H), 5.26 (s, 2H), 5.59 (d, 1H,
J ¼ 7:6 Hz), 6.35 (s, 1H), 6.79 (d, 1H, J ¼ 2:4 Hz), 6.90
(dd, 1H, J ¼ 2:4, 8.8 Hz), 7.43 (d, 1H, J ¼ 8:8 Hz).
4.6. 6,7-Bis(tert-butoxycarbonylmethoxy)-4-(hydroxy-
methoxy)coumarin (6b)
Compound 6b (1.70 g, 64%) was prepared by the same
procedure as 6a, from 6,7-bis(tert-butoxycarbonyl-
methoxy)coumarin-4-carbaldehyde2c (5b: 2.64 g, 6.07
4.3. Di-tert-butyl (2S,3S)-3-benzyloxyaspartate (11)
mmol) and NaBH4 (321.5 mg, 8.50 mmol). IR mKmaBxr cmꢀ1
:
3544, 1739, 1641, 1616, 1370, 1152, 1102. 1H NMR
(CDCl3) d 1.49 (s, 9H), 1.49 (s, 9H), 4.61 (s, 2H), 4.64 (s,
2H), 4.79 (d, 2H, J ¼ 9:6 Hz), 6.46 (s, 1H), 6.71 (s, 1H),
7.03 (s, 1H). MS (ESI) m=z calcd for C22H28O9(M+H)þ
437, found 437.
To a solution of di-tert-butyl (2S,3S)-3-benzyloxy-N-
tert-butoxycarbonylaspartate6a;b (10: 149 mg, 0.33
mmol) in CH2Cl2 (2 mL), TFA (1 mL) was added at 0 ꢁC
and the mixture was stirred for 1 h at 0 ꢁC. Saturated
aqueous NaHCO3 solution was added and the mixture
was extracted with CHCl3. The extract was purified by
silica gel column chromatography (hexane/EtOAc ¼ 1/1)
1
4.7. Di-tert-butyl (2S,3S)-3-benzyloxy-N-{[6,7-bis(tert-
butoxycarbonylmethoxy)coumarin-4-yl-methoxy]carbonyl}-
aspartate (12b)
to obtain 92 mg of 11 (79% yield). H NMR (CDCl3) d
1.42 (s, 9H), 1.52 (s, 9H), 3.76 (d, 1H, J ¼ 2:8 Hz), 4.31
(d, 1H, J ¼ 2:8 Hz), 4.42 (d, 1H, J ¼ 11:2 Hz), 4.80 (d,
1H, J ¼ 11:2 Hz), 7.32–8.33 (m, 5H).
Compound 12b (110.6 mg, 75%) was prepared by the
same procedure as 12a, from 6b (342.7 mg, 0.79 mmol),
trichloromethylchloroformate (0.18 mL, 1.48 mmol), 11
(63 mg, 0.18 mmol), and i-Pr2NEt (0.31 mL, 1.78 mmol).
1H NMR (CDCl3) d 1.36 (s, 9H), 1.37 (s, 9H), 1.40 (s,
9H), 1.42 (s, 9H), 4.34–4.37 (m, 2H), 4.55 (s, 2H), 4.57
(s, 2H), 4.69–4.75 (m, 2H), 5.13 (s, 2H), 5.61 (d, 1H,
J ¼ 10:0 Hz), 6.28 (s, 1H), 6.67 (s, 1H), 6.89 (s, 1H),
7.19–7.26 (m, 5H).
4.4. Di-tert-butyl (2S,3S)-3-benzyloxy-N-{[7-(tert-butoxy-
carbonylmethoxy)coumarin-4-yl-methoxy]carbonyl}-
aspartate (12a)
To a solution of 6a (176 mg, 0.7 mmol) in THF (5 mL),
trichloromethylchloroformate (0.17 mL, 1.4 mmol) was
added prior to reflux for 2 h, and then the solvent was

K. Takaoka et al. / Bioorg. Med. Chem. 12 (2004) 3687–3694
3693
4.8. (2S,3S)-3-Benzyloxy-N-{[6,7-bis(carboxymeth-
oxy)coumarin-4-yl-methoxy]carbonyl}aspartate (N-
BCMCMC-L-TBOA: 3b)
TFA (0.5 mL) was added. The mixture was stirred for
1 h at 0 ꢁC and then at room temperature for 4 h. The
solvent was distilled off and the residue was purified by
HPLC (10–50% CH3CN/H2O/0.1%TFA) to obtain
1
Compound 3b (61.6 mg, 52%) was obtained by the same
procedure as 3a, from compound 12b (164 mg,
42 mg of 4 (49% yield). H NMR (DMSO-d6) d 1.6 (d,
3H, J ¼ 6:4 Hz), 3.8 (s, 3H), 3.9 (s, 3H), 4.2 (d,
1H, J ¼ 11:2 Hz), 4.4 (d, 1H, J ¼ 3:2 Hz), 4.5 (d, 1H,
J ¼ 4:6 Hz), 4.7 (d, 1H, J ¼ 11:2 Hz), 6.4 (q, 1H,
J ¼ 6:4 Hz), 7.04–7.06 (m, 2H), 7.20–7.22 (m, 4H), 7.51
(s, 1H). Q-TOF-MS m=z 777 (M+H)þ, 427. HRMS
1
0.20 mmol) and TFA (2 mL). H NMR (DMSO-d6) d
4.44 (d, 1H, J ¼ 12:0 Hz), 4.49 (d, 1H, J ¼ 3:2 Hz), 4.61
(dd, 1H, J ¼ 3:2, 9.6 Hz), 4.75 (d, 1H, J ¼ 12:0 Hz), 4.78
(s, 2H) 4.86 (s, 2H), 5.21 (d, 1H, J ¼ 16:8 Hz), 5.31 (d,
1H, J ¼ 16:8 Hz), 6.41 (s, 1H), 7.00 (s, 1H), 7.10 (s, 1H),
7.28–7.32 (m, 5H) 7.90 (d, 1H, J ¼ 9:6 Hz). HRMS
(FAB) m=z calcd for C26H24NO15(M+H)þ 590.1146,
(FAB) m/z calcd for C34H28F3N2O16(M+H)þ 777.1391,
25
D
found 777.1395. ½aꢁ +92.7 (c 0.65, DMSO).
found 590.1127. Fluorescence kex=kem (1 lM): 340 nm/
444 nm. ½aꢁ )9.01 (c 0.32, DMSO).
26
D
4.12. Photolysis of caged compounds with UV-handy lamp
A solution (300 lL) of 100 lM of the caged compound
in PBS(+) buffer (phosphate buffered saline including
137 mM NaCl, 2.7 mM KCl, 8.1 mg Na2HPO4, 1.5 mM
KH2PO4, 1 mM CaCl2 and 1 mM MgCl2, pH 7.4) was
placed in quartz cuvettes with a path length of 1 mm in a
dark box and irradiated at room temperature for 5–
120 min. Irradiation condition was as follows: Spectro-
line ENF-260C/J (Spectronics Co., USA), 365 nm, max
power at 10 cm from the lamp: 2.3 mW/cm2. Irradiated
solution was analyzed by HPLC [Column; SP-120-5-
ODS-AP (Daiso Co. Ltd, Japan), 150 mm · 6 mm ID:
Eluent: 30% CH3CN/H2O/0.1%TFA for 1a, 1b, 3a, and
3b or 30–50% CH3CN/H2O/0.1%TFA for 4; Flow rate;
1 mL/min.; Detection; 320 and 254 nm.] As an internal
standard for the quantitative analyses, 10 lM of 30,40-
dimethoxyacetophenone (Rt: 11.1 min) for 1a (Rt:
5.3 min), 1b (Rt: 10.7 min), 3b (Rt: 9.0 min), and 4 (Rt:
25.0 min) or 40 lM of 2-(3,4-dimethoxyphenyl)ethanol
(Rt: 6.7 min) for 3a (Rt: 12.5 min) was included in the
sample solution.
4.9. Di-tert-butyl (2S,3S)-3-{3-[4-(trifluoromethyl)benzoyl]-
aminobenzyloxy}aspartate (14)
To a solution of compound di-tert-butyl (2S,3S)-N-tert-
butoxycarbonyl-3-3-[4-(trifluoromethyl)benzoyl]amino-
benzyloxyaspartate8 (13: 149 mg, 0.23 mmol) in CH2Cl2
(2 mL), TFA (1 mL) was added at 0 ꢁC and the mixture
was stirred for 1 h at 0 ꢁC. Saturated aqueous NaHCO3
solution was added and the mixture was extracted with
CHCl3. The extract was purified by silica gel column
chromatography (hexane/EtOAc ¼ 1/1) to obtain 91 mg
1
of compound 14 (74% yield). H NMR (CDCl3) d 1.44
(s, 9H), 1.52 (s, 9H), 3.80 (d,1H, J ¼ 2:8 Hz), 4.33 (d,
1H, J ¼ 2:8 Hz), 4.43 (d, 1H, J ¼ 11:6 Hz), 4.82 (d, 1H,
J ¼ 11:6 Hz), 7.13 (d, 1H, J ¼ 7:6 Hz), 7.35(t, 1H,
J ¼ 7:6 Hz), 7.50 (s, 1H), 7.77 (d, 2H, J ¼ 8:4 Hz), 7.88
(s, 1H), 8.00 (d, 2H, J ¼ 8:4 Hz).
4.10. Di-tert-butyl (2S,3S)-N-[6,7-bis(tert-butoxycarbon-
ylmethoxy)coumarin-4-yl-methoxy]carbonyl-3-{3-[4-(tri-
fluoromethyl)benzoyl]aminobenzyloxy}aspartate (15)
4.13. Photolysis of caged compounds with YAG-laser
To a solution of 6b (301 mg, 0.69 mmol) in THF (4 mL),
trichloromethylchloroformate (0.17 mL, 1.4 mmol) was
added prior to reflux for 2 h, and then the solvent was
distilled off. The residue was dissolved in CH2Cl2 (2 mL)
and the resulting solution and i-Pr2NEt (0.3 mL,
1.7 mmol) were added to a solution of compound 14
(91 mg, 0.17 mmol) in CH2Cl2 (2 mL), after which the
mixture was stirred for 10 min. The solvent was distilled
off and the residue was purified on a silica gel-plate
(hexane/EtOAc ¼ 2/1) to obtain 111 mg of compound 15
A solution (100 lL) of 100 lM of the caged compound
in PBS(+) was placed in quartz cuvettes with a path
length of 1 mm and irradiated at room temperature. For
the uptake assay 1 lM solution of 4 was used. Irradia-
tion condition was as follows: SureLite II (Continuum,
Santa Clara, CA, USA), 355 nm, 8 ns pulse, 40 mJ.
Irradiated solution was analyzed by HPLC as above.
1
(64% yield). H NMR (CDCl3) d 1.43 (s, 9H), 1.46 (s,
4.14. [14C]Glutamate uptake assay
9H), 1.49 (s, 9H), 1.50 (s, 9H), 4.42–4.47 (m, 2H), 4.61
(s, 2H), 4.63 (s, 2H), 4.78–4.81 (m, 2H), 5.20 (d, 2H,
J ¼ 6:0 Hz), 5.71 (d, 1H, J ¼ 10:4 Hz), 7.10 (d, 1H,
J ¼ 7:6 Hz), 7.36 (t, 1H, J ¼ 7:6 Hz), 7.54 (s, 1H), 7.76
(d, 2H, J ¼ 8:4 Hz), 7.80 (d, 1H, J ¼ 7:6 Hz), 8.02 (d,
2H, J ¼ 8:4 Hz).
MDCK cells stably expressing EAAT2 were seeded onto
96-well plates and cultured for 2 days before the uptake
assay. The subconfluent cells were washed two times
with 300 lL of modified PBS(+) that contained 10 mM
D
-glucose, pH 7.4 and preincubated in 300 lL of the
same buffer at 37 ꢁC for 12 min. After aspiration of the
buffer, cells were incubated with 1 lM
-[14C]glutamate
4.11. (2S,3S)-N-[6,7-Bis(carboxymethoxy)coumarin-4-yl-
methoxy]carbonyl-3-{3-[4-(trifluoromethyl)-benzoyl)amino-
benzyloxy}aspartate (N-BCMCMC-L-TFB-TBOA: 4)
L
in 80 lL of modified PBS(+) in the absence or presence
of test compounds at various concentrations at 37 ꢁC for
12 min. To determine the uptake, cells were washed
three times with ice-cold buffer and radioactivity was
measured by scintillation counting in 100 lL of
Compound 15 (111 mg, 0.11 mmol) was dissolved in
CH2Cl2 (4 mL) and to the solution cooled in an ice bath

3694
K. Takaoka et al. / Bioorg. Med. Chem. 12 (2004) 3687–3694
A.; Corrie, J. E. T. J. Am. Chem. Soc. 1999, 121, 6503–
6504; (c) Matsuzaki, M.; Ellis-Davies, G. C. R.; Nemoto,
T.; Miyashita, Y.; Iino, M.; Kasai, H. Nat. Neurosci. 2001,
4, 1086–1092.
Microscinti 40 (PerkinElmer Life and Analytical Sci-
ences, Inc., Boston, MA, USA). Nonspecific incorpo-
ration was determined in sodium-free solution (140 mM
choline chloride, 5 mM KCl, 1 mM CaCl2, 1 mM MgCl2,
6. (a) Shimamoto, K.; Lebrun, B.; Yasuda-Kamatani, Y.;
Sakaitani, M.; Shigeri, Y.; Yumoto, N.; Nakajima, T.
Mol. Pharmacol. 1998, 53, 195–203; (b) Shimamoto, K.;
Shigeri, Y.; Yasuda-Kamatani, Y.; Lebrun, B.; Yumoto,
N.; Nakajima, T. Bioorg. Med. Chem. Lett. 2000, 10,
2407–2410; (c) Shigeri, Y.; Shimamoto, K.; Yasuda-
Kamatani, Y.; Seal, R. P.; Yumoto, N.; Nakajima, T.;
Amara, S. G. J. Neurochem. 2001, 79, 297–302.
7. Takaoka, K.; Tatsu, Y.; Yumoto, N.; Nakajima, T.;
Shimamoto, K. Bioorg. Med. Chem. Lett. 2003, 13, 965–
970.
20 mM HEPES, and 10 mM D-glucose, pH 7.4). Specific
uptake of [14C]glutamate is given relative to the control.
All values displayed are mean SEM of at least three
determinations. Dose–response curves were generated
by using computer software GraphPad Prism (GraphPad
Software Inc., San Diego, CA, USA).
Acknowledgements
8. Shimamoto, K.; Sakai, R.; Takaoka, K.; Yumoto, N.;
Nakajima, T.; Amara, S. G.; Shigeri, Y. Mol. Pharmacol.
2004, 65, 1008–1015.
9. Rossi, F. M.; Margulis, M.; Tang, C.-M.; Kao, J. P. Y.
J. Bio. Chem. 1997, 272, 32933–32939.
10. (a) Kim, J.-M.; Chang, T.-E.; Park, R. H.; Kim, D. K.;
Han, D. K.; Ahn, K.-D. Chem. Lett. 2000, 712–713; (b)
Walker, J. W.; Reld, G. P.; McCray, J. A.; Trenthan, D.
R. J. Am. Chem. Soc. 1988, 110, 7170–7177.
11. The same reaction using 7b prepared from triphosgen and
6a yielded 12a in only 8%.
12. Even if L-TFB-TBOA was irradiated for 60 min in the
This work was financially supported by Grant-in-Aid
for Creative Scientific Research (#13NP0401) and
Grant-in-Aid for Scientific Research (#13680681) from
the Ministry of Education, Culture, Sports, Science, and
Technology. We wish to thank Prof. Susan. G. Amara
at the University of Pittsburgh for providing the MDCK
cells expressing EAAT2, Dr. Yasushi Shigeri at Na-
tional Institute of Advanced Industrial Science and
Technology for critical reading of this manuscript, and
Dr. Manabu Horikawa and Mr. Tsuyoshi Fujita at
Suntory Institute for Bioorganic Research for mea-
surement of Q-TOF-MS and HR-FABMS.
presence or in the absence of BCMCM-OH, the amount
L-
monitored by HPLC did not change, indicating that
TFB-TBOA does not decompose by UV-irradiation.
13. Lu, M.; Fedoryak, O. D.; Moister, B. R.; Dore, T. M.
Org. Lett. 2003, 5, 2119–2122.
14. The concentration of L-TFB-TBOA in the photolysis
References and notes
samples estimated from the uptake assay was well consis-
tent with that determined from the HPLC analysis. We
also confirmed that neither BCMCM-OH nor CMCM-
OH affected the glutamate uptake.
1. Methods in Enzymology, 291, Caged Compounds; (a)
Marriott, G., Ed.; Academic: San Diego, 1998; (b) Shigeri,
Y.; Tatsu, Y.; Yumoto, N. Pharmacol. Therapeu. 2001, 91,
85–92.
15. The transient absorption during the laser irradiation was
measured using a photomultiplier tube and monochro-
matic light from a 150-W xenon lamp. The photocurrent
was fed into an oscilloscope (TDS340AP, Sony-Tektron-
ics, Tokyo, Japan). The accurate rate of the photolysis of
coumarin-carbamate 3b, 4 could not be obtained in the
same case as the esters 1,7 because these coumarin-
derivatives showed very strong fluorescence, of which the
decay curves were very complex. Nevertheless, the change
in the transient spectroscopy was very fast and was
completed within 1.5 ls. It was reported that the photo-
lytic uncaging and release of the carbamate-type deriva-
tives occur in two steps, consisting of an initial light-
induced cleavage and a subsequent light-independent,
pH-dependent decarboxylation step that proceeds on
the millisecond time scale.9 The latter step would be
2. (a) Hagen, V.; Frings, S.; Bending, J.; Lorenz, D.;
Wiesner, B.; Kaupp, U. B. Angew. Chem., Int. Ed. 2002,
41, 3625–3628; (b) Eckardt, T.; Hagen, V.; Schade, B.;
Schmidt, R.; Schweitzer, C.; Bendig, J. J. Org. Chem.
2002, 67, 703–710; (c) Hagen, V.; Bendig, J.; Frings, S.;
Eckardt, T.; Helm, S.; Reuter, D.; Kaupp, U. B. Angew.
Chem., Int. Ed. 2001, 40, 1045–1048; (d) Hagen, V.;
Bendig, J.; Frings, S.; Wiesner, B.; Schade, B.; Helm, S.;
Lorenz, D.; Kaupp, U. B. J. Photochem. Photobiol. B:
Biol. 1999, 53, 91–102; (e) Sarker, A. M.; Kaneko, Y.;
Neckers, D. C. J. Photochem. Photobiol. A: Chem. 1998,
117, 67–74.
3. (a) Furuta, T.; Torigai, H.; Sugimoto, M.; Iwamura, M.
J. Org. Chem. 1995, 60, 3953–3956; (b) Furuta, T.; Wang,
S. S.-H.; Dantzker, J. L.; Dore, T. M.; Bybee, W. J.;
Callaway, E. M.; Denk, W.; Tsien, R. Y. Proc. Natl. Acad.
Sci. U.S.A. 1999, 96, 1193–1200; (c) Furuta, T.; Hirayama,
Y.; Sugimoto, M.; Iwamura, L. Org. Lett. 2001, 3, 1809–
1812; (d) Fedoryak, O. D.; Dore, T. M. Org. Lett. 2002, 4,
3419–3422.
rate-limiting for release of free amino acid (L-TBOA or L-
TFB-TBOA) if the photocleavage process of the couma-
rin-moiety is fast. To estimate the kinetics of 3b and 4,
electrophysiological measurements would be needed.
16. The chemical yields of the photolysis by repeated irradi-
ation (8 shots) were 5.5 1.5% (3a), 43 1.0% (3b), and
42 1.9% (4), respectively.
4. For review, (a) Danbolt, N. C. Prog. Neurobiol. 2001, 65,
1–105; (b) Amara, S. G.; Fontana, A. C. K. Neurochem.
Int. 2002, 41, 313–318.
5. (a) Wieboldt, R.; Gee, K. R.; Niu, L.; Ramesh, D.;
Carpenter, B. K.; Hess, G. P. Proc. Natl. Acad. Sci. U.S.A.
1994, 91, 8752; (b) Papageorgiou, G.; Ogden, D. C.; Barth,
17. MaCray, J. A.; Trentham, D. R. Annu. Rev. Biophys.
Biophys. Chem. 1989, 18, 239–270.
18. The activity of L-TFB-TBOA was slightly less potent than
the previous result (IC50 ¼ 1.9 0.1 nM).8 The value of
IC50 would vary according to the condition of using cells.