Syntheses of 2-NBDG analogues for monitoring stereoselective uptake of D-glucose

Article abstract of DOI:10.1016/j.bmcl.2011.04.148

2-NBDG is a widely used fluorescent tracer for monitoring d-glucose uptake into single living cells. However, 2-NBDG alone is not sufficient for monitoring the net stereoselective uptake of d-glucose, unless its possible non-stereoselective uptake is properly evaluated. l-Glucose derivatives, which emit fluorescence distinct from that of 2-NBDG, should provide valuable information on the stereoselective uptake, when used with 2-NBDG in combination. In the present study, we synthesized Texas Red (sulforhodamine 101 acid)-coupled and [2-(benz-2-oxa-1,3-diazol-4-yl)amino]-coupled 2-deoxy-d-glucose, referred to as [2-TRG] and [2-BDG], respectively. These derivatives showed emission wavelength longer and shorter than that of 2-NBDG, respectively. 2-TRLG, an antipode of 2-TRG, proved to be an effective tracer for evaluating the extent of non-stereoselective uptake of 2-NBDG when used simultaneously with 2-NBDG. On the other hand, 2-BDG exhibited very weak fluorescence, but the application of a novel cross coupling in the presence of a benzoxadiazole group may be useful for the future development of effective glucose tracers.

Full text of DOI:10.1016/j.bmcl.2011.04.148

Bioorganic & Medicinal Chemistry Letters 21 (2011) 4088–4096
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Syntheses of 2-NBDG analogues for monitoring stereoselective uptake
of D-glucose
Toshihiro Yamamoto a, , Shin-ichi Tanaka a, Sechiko Suga b,c, Seiji Watanabe d, Katsuhiro Nagatomo b,
⇑
Ayako Sasaki b, Yuji Nishiuchi a,e, Tadashi Teshima a,e, Katsuya Yamada b,
⇑
a Peptide Institute, Inc., Saito Research Center, Ibaraki, Osaka 567-0085, Japan
b Department of Physiology, Hirosaki University Graduate School of Medicine, Hirosaki, Aomori 036-8562, Japan
c Hirosaki University of Health and Welfare, Hirosaki, Aomori 036-8102, Japan
d Department of Anatomical Science, Hirosaki University Graduate School of Medicine, Hirosaki, Aomori 036-8562, Japan
e Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
2-NBDG is a widely used fluorescent tracer for monitoring
ever, 2-NBDG alone is not sufficient for monitoring the net stereoselective uptake of
possible non-stereoselective uptake is properly evaluated. -Glucose derivatives, which emit fluorescence
distinct from that of 2-NBDG, should provide valuable information on the stereoselective uptake, when
used with 2-NBDG in combination. In the present study, we synthesized Texas Red (sulforhodamine
D
-glucose uptake into single living cells. How-
Received 3 March 2011
Revised 11 April 2011
Accepted 13 April 2011
Available online 10 May 2011
D
-glucose, unless its
L
101 acid)-coupled and [2-(benz-2-oxa-1,3-diazol-4-yl)amino]-coupled 2-deoxy-D-glucose, referred to
Keywords:
2-NBDG
2-NBDLG
2-TRLG
as [2-TRG] and [2-BDG], respectively. These derivatives showed emission wavelength longer and shorter
than that of 2-NBDG, respectively. 2-TRLG, an antipode of 2-TRG, proved to be an effective tracer for eval-
uating the extent of non-stereoselective uptake of 2-NBDG when used simultaneously with 2-NBDG. On
the other hand, 2-BDG exhibited very weak fluorescence, but the application of a novel cross coupling in
the presence of a benzoxadiazole group may be useful for the future development of effective glucose
tracers.
2-BDG
Glucose uptake
Buchwald–Hartwig reaction
Ó 2011 Elsevier Ltd. All rights reserved.
Measurements of the cellular uptake of
D
-glucose have
and organs, leading to new findings such as metabolic
communication between astrocytes in the brain.8–11 On the other
hand, there was an increasing need for control substrates against
2-NBDG (1) to properly evaluate an occurrence of non-stereoselec-
tive uptake and membrane adsorption of 2-NBDG (1) in addition
to its time-dependent degradation and extinction at the individual
experimental condition used.5
contributed to broad spectra of research fields from functional brain
mapping to cancer diagnosis.1,2 To evaluate the glucose uptake
activity, a variety of radio-labeled analogues of
D
-glucose, such as
[
14C] 2-deoxy-
D
-glucose (2-DG)3 and [18F] fluoro-2-deoxy-
D
-glucose
(2-FDG),4 have been effectively used as tracers. However, a draw-
back of such tracers is their poor spatial and temporal resolution,
and they can not monitor the glucose uptake in single, living cells
in real time. 2-[N-(7-Nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2-
To approach these issues, we synthesized transporter-
unrecognizable (
substrates for 2-NBDG (1),12,13 because it is known that
-glucose binds to GLUTs.14 Of these, 2-NBDLG (2) (Fig. 1), an enan-
L
-isomer) fluorescent tracers of glucose as control
deoxy-
D
-glucose [2-NBDG (1)] (Fig. 1), a fluorescent analogue of
D- but not
D
-glucose, provides more sensitive measurement of the glucose
L
uptake.5 2-NBDG (1), which was originally designed to monitor
the viability ofEscherichia coli cells,6 is taken up through mammalian
glucose transporters (GLUTs) in a time, concentration and tempera-
ture-dependent manner with Km values comparable to those
reported for radio-labeled glucose tracers.7 In the last decade,
2-NBDG (1) has been successfully applied to a wide variety of cells
tiomer of 2-NBDG (1), is an ideal control substrate providing us
with the valuable information on the net stereoselective uptake of
D-glucose. However, the emission spectrum of 2-NBDLG (2), which
is identical to that of 2-NBDG (1), makes it impossible to use
2-NBDLG (2) simultaneously with 2-NBDG (1). This is critical when
examining if the stereoselective uptake of delicate cells such as
neurons is preserved or not, because such cells often show
gradually changing membrane states between healthy and dead
especially during in vitro experiments.15 Membrane-impermeable
large molecules such as dextran-Texas Red may not properly
monitor the stereoselectivity.16 Indeed, it is known that severe
metabolic stress induces an altered glucose transport prior to total
⇑
Corresponding authors. Tel.: +81 72 643 4411; fax: +81 72 643 4422 (T.Y. for
synthesis); tel.: +81 172 39 5008; fax: +81 172 39 5009 (K.Y. for biological concept
and optical properties).
E-mail addresses: yamamoto@peptide.co.jp (T. Yamamoto), kyamada@cc.hiro-
saki-u.ac.jp (K. Yamada).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.04.148

T. Yamamoto et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4088–4096
4089
HO
HO
HO
HO
O
O
OH
HO
HO
O
O
OH
HO
HO
OH
SO2
OH
HO
OH
SO2
HN
HO
HN
HN
OH
HN
N
O3S
N
O
N
O
N
N
O2N
SO3
NO2
N
O
O
2-NBDLG (2)
2-NBDG (1)
N
N
2-TRG (3a) (para-isomer)
2-TRG (3b) (ortho-isomer)
OH
OH
HO
HO
HO
O
O
OH
OH
O
OH
OH
HO
HO
HN
HN
OH
HN
SO2
SO2
N
O3S
N
O
N
N
O
SO3
2-BDG (5)
O
N
N
2-TRLG (4a) (para-isomer)
2-TRLG (4b) (ortho-isomer)
Figure 1. Structures of 2-NBDG (1) and its analogues.
disruption of the membrane integrity, possibly due to a rapid
change in both protein expression and protein/lipid polarity in
the plasma membrane.15 Instead,
L-glucose derivatives, which have
20 M 2-TRLG
wavelength distinct from that of 2-NBDG (1), might serve as reli-
able molecules informing the stereoselectivity status at the cost
of their physicochemical properties that may not be identical to
those of 2-NBDG (1).
100
90
80
70
60
50
40
30
20
10
0
2-Amino-2-deoxy-
either para (3a) or ortho (3b) isomer of sulforhodamine 101 acid
(Texas Red), was prepared from -glucosamine hydrochloride with
D-glucose derivative 2-TRG (3), which bears
D
a mixture of para- and ortho-101-rhodamine sulfonyl chloride
(Texas Red sulfonyl chloride) (Fig. 1).17 Both para (3a) and ortho
(3b) isomers emit strong red fluorescence with identical emission
spectra (data not shown). Using L-glucosamine hydrochloride as a
500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800
Wavelength (nm)
Figure 2A. Normalized emission spectrum of 2-TRLG (4) (in HEPES buffer, pH 7.35).
Argon laser (488 nm) was used for excitation.
L-glucosamine hydrochloride
Sulforhodamine
48 %
101 acid chloride
(ortho/para mixture)
100 M 2-NBDG
OH
OH
O
O
OH
OH
100
90
80
70
60
50
40
30
20
10
OH
HO
OH
HO
HN
HN
SO2
SO2
N
O3S
N
+
SO3
O
O
N
N
0
500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800
2-TRLG (4b)
2-TRLG (4a)
ortho
Wavelength (nm)
para
Scheme 1. Synthesis of 2-TRLG (4) as a mixture of para (4a) and ortho (4b) isomers.
Figure 2B. Similar to 2A, but emission spectrum of 2-NBDG (1).

4090
T. Yamamoto et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4088–4096
Fig. 2C) may well reflect the extent of 2-TRLG (4) entry, in other
words, an occurrence of non-stereoselective uptake.
Actually, when 20 M of 2-TRLG (4) was applied simulta-
neously with 100 M of 2-NBDG (1) to a neuron freshly dissociated
100 M 2- NBDG + 20 M 2- TRLG
100
90
80
70
60
50
40
30
20
10
0
l
l
from an adult mouse,18 an increase in fluorescence in the green
channel was detected in the cytosolic area excluding the nucleus
(Fig. 3A–C). In contrast, only a minimum increase in the red chan-
nel was detected, which was hardly seen in this panel, suggesting
that 2-NBDG (1), but not 2-TRLG (4), was taken up into the cell
(Fig. 3C).19 According to the extent of impaired stereoselectivity,
2-TRLG (4) was taken up into cells in addition to 2-NBDG (1) to a
variable degree (see Supplementary figure).
500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800
(nm)
To compare stereoselectivity of 2-TRLG (4) with that of 2-
NBDLG (2), 20
lM 2-TRLG (4) was applied simultaneously with
Green Channel
Red Channel
(580-740 nm)
100 M of 2-NBDLG (2) to neurons in various conditions. A neu-
l
(500-580 nm)
ron exemplified in Fig. 3D–E showed no discernable increase in
fluorescence both in the red and the green channels for the first
application, suggesting that the stereoselectivity was mostly pre-
served at that time. However, the second application of the mix-
ture after damaging the same neuron by a light touch with a
small glass pipette, resulted in a rapid increase both in the red
580 nm
Figure 2C. Similar to 2A and 2B, but emission spectrum of 100
and 20 of 2-TRLG (4) mixture (solid line). Spectrum of 2-NBDG (1) is
superimposed (broken line) for comparison.
lM of 2-NBDG (1)
lM
starting material in place of
D-isomer, we further synthesized
and green channels, indicating
(Fig. 3F and G).
a loss of stereoselectivity
2-TRLG (4), an antipode of 2-TRG (3), as a mixture of para (4a) and
ortho (4b) isomers (Fig. 1 and Scheme 1). 2-TRLG (4) is characterized
by its intermediate size (molecular weight <1000), and the emission
maximum (600–610 nm, Fig. 2A) is well apart from that of 2-NBDG
(1) (550 nm, Fig. 2B), making double peaked spectrum when mixed
with 2-NBDG (1) (Fig. 2C). For detection, two photomultipliers were
simultaneously used. Fluorescence at 500–580 nm (restricted by a
prism and slit in Leica SP5) was introduced to the first one (green
channel) for detecting 2-NBDG (1) entry, while fluorescence at
580–740 nm was assigned to the second one (red channel) for
detecting 2-TRLG (4) entry (Fig. 2C). If the relative fluorescence
intensity in the red channel to that in the green channel surpasses
the value expected solely by 2-NBDG (1) entry (red area in Fig. 2C),
the difference in the fluorescence intensity (hatched area in
Apart from such extreme cases, cells in actual biological exper-
iments often show various intermediate states between fully pre-
served stereoselectivity and total loss of membrane integrity,
indicating a need for some quantification. In separate experiments
using solution containing fluorescent glucose tracers (Fig. 4), it was
confirmed that the ratio of fluorescence intensity detected in the
red channel to that in the green channel (referred to as the red/
green ratio) was constant for 100 lM of 2-NBDG (1) solution, when
photomultiplier sensitivity and the intensity of excitation light
were adequately selected (a and b in Fig. 4A; green and red lines
in Fig. 4B, in the shaded area). In such ranges, a linear relationship
was also confirmed for the mixture of 100
lM of 2-NBDG (1) and
20 M of 2-TRLG (4) (c and d in Fig. 4A; dotted green and red lines
l
Green Channel
Red Channel
DIC Image
Overlay
Figure 3A. Confocal microscopic images before applying tracers to a neuron acutely dissociated from substantia nigra pars reticulata of an adult mouse. The intensity of
excitation and sensitivity of detector were adjusted so that an intrinsic fluorescence in the green channel was slightly discernable. DIC: differential interference contrast.

T. Yamamoto et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4088–4096
4091
Green Channel
Red Channel
DIC Image
Overlay
Figure 3B. Similar to 3A, but during simultaneous application of 100 lM 2-NBDG (1) and 20 lM 2-TRLG (4) for 1 min at 37 °C (extracellular glucose concentration was
5 mM). The shadow of the neuron is larger and clearer in the red than in the green channel, suggesting an exclusion of 2-TRLG (4) from the cytosol.
in Fig. 4B), indicating that the red/green ratio can be used as a mea-
sure for 2-TRLG (4) entry into the cytosol.
The red/green ratio should reflect the extent of 2-TRLG (4) entry
proportionally in such ranges from the minimum value seen in sole
2-NBDG (1) entry (red versus green line in Fig. 4B) up to the max-
line in Fig. 4B). In contrast, only an ignorable increase in the fluo-
rescence intensity was detected in the green channel by 2-TRLG
(4) entry (green line and green dotted line in Fig. 4B). This is due
to the sharp cut-off at the shorter end in the emission spectrum
of 2-TRLG (4) (Fig. 2A). These data suggest that the extent of 2-
TRLG (4) entry can be evaluated by calculating the red/green ratio
for the region of interest in the cytosol in reference to the value of
imum value corresponding to the external solution (100
lM of 2-
NBDG (1) and 20 M of 2-TRLG (4); red dotted versus green dotted
l
Green Channel
Red Channel
DIC Image
Overlay
Figure 3C. Similar to 3A and B, but after rapid removal of fluorescent tracers. An enhanced image is presented in the inset to show cytosolic incorporation of 2-NBDG (1)
excluding the nucleus.

4092
T. Yamamoto et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4088–4096
Green Channel
Red Channel
DIC Image
Overlay
Figure 3D. Similar to 3A–C, but 100 lM of 2-NBDLG (2), the antipode of 2-NBDG (1), was applied simultaneously with 20 lM 2-TRLG (4). These subsets of figures represent
images before application of fluorescent tracers.
Green Channel
Red Channel
DIC Image
Overlay
Figure 3E. Similar to 3D, but images after removal of tracers. L-Form glucose derivatives produced only a minimum increase in the fluorescence to the neuron of interest,
contrasted to considerable increase seen in small debris (arrows) possibly due to membrane damage.

T. Yamamoto et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4088–4096
4093
Green Channel
Red Channel
DIC Image
Overlay
Figure 3F. The same neuron as in 3D–E, but after a slight touch to the dendrite with a small glass pipette approaching from the left side. These subsets of figures represent
images just before the second application of tracers.
Green Channel
Red Channel
DIC Image
Overlay
Figure 3G. Similar to 3F, but images after the second application of tracers. In contrast to 3D-3E, an immediate increase in the fluorescence was apparent both in the green
and the red channels, suggesting a massive entry of 2-TRLG (4) as well as 2-NBDLG (2). This makes yellow color in the overlay, contrasting to Fig. 3E.

4094
T. Yamamoto et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4088–4096
(a) 2-NBDG fluorescence detected in the GREEN channel
(b) 2-NBDG fluorescence detected in the RED channel
200
180
160
140
120
100
80
200
180
160
140
120
100
80
Laser 60%
Laser 70%
Laser 80%
Laser 90%
Laser 100%
Laser 60%
Laser 70%
Laser 80%
Laser 90%
Laser 100%
60
60
40
40
20
20
0
0
300
500
700
900
1100
1300
300
500
700
900
1100
1300
PMT sensitivity (Volt)
PMT sensitivity (Volt)
(c) 2-NBDG+2-TRLG fluorescence in the GREEN channel
(d) 2-NBDG+2-TRLG fluorescence in the RED channel
200
180
160
140
120
100
80
200
180
160
Laser 60%
Laser 60%
Laser 70%
Laser 80%
Laser 90%
Laser 100%
140
120
100
80
Laser 70%
Laser 80%
Laser 90%
Laser 100%
60
60
40
40
20
20
0
0
300
500
700
900
1100
1300
300
500
700
900
1100
1300
PMT sensitivity (Volt)
PMT sensitivity (Volt)
Figure 4A. Fluorescence intensities detected in the green (500–580 nm) and red (580–740 nm) channel are plotted for 2-NBDG (1) (100
lM) alone (a and b) and for a mixture
of 100 M 2-NBDG (1) and 20 M 2-TRLG (4) (c and d) at various sensitivity of photomultipliers (PMT) and laser intensities (488 nm argon laser of SP5). Values are expressed
l
l
as mean of two separate experiments in arbitrary unit.
Pd2(dba)3 (10 mol% to 7)
Phosphine Ligand 8
(20 mol% to 7)
Ph
O
Ph
O
O
extracellular solution solely containing 2-NBDG (1) and that com-
prised of 2-NBDG (1)/2-TRLG (4) mixture.
O
O
O
HO
HO
H2N
OMe
HN
To date, 2-NBDG (1) and its regioisomer 6-NBDG20 are the only
NaOtBu (2.5 eq. to 6)
Toluene
120 oC, 3h
OMe
6
fluorescent D-glucose tracers that are authorized to enter through
N
O
+
GLUTs of normal cells by many groups,5,7,8,21,22 although 6-NBDG
can not be phosphorylated nor accumulated in the cell.15 Actually,
fluorophores attached to glucose with emission wavelength longer
than NBDGs tend to have larger molecular mass, being extremely
difficult to pass through GLUTs.21,22 As such, even 2-TRG (3), which
has a relatively small molecular mass (=767.87), was not taken up
Cl
N
Ph
9 (56%)
N
Ph
N
N
O
N
7 (1.7 eq.)
Phosphine Ligand 8
by intact neurons (data not shown). However, it is of noted that
glucose derivatives bearing a large chromophore often compete
D-
HO
O
HO
1 M HCl
60 o
HO
80% AcOH aq.
40 oC
with D-glucose to some extent, suggesting that they still have a
HN
possibility to interact with GLUTs.21,22 Collectively, 2-TRLG (4)
OMe
C
may be utilized as a marker for monitoring the loss of stereoselec-
N
tivity much more reliably compared to GLUT-impermeable
cose derivatives or large dead-cell markers.
D-glu-
O
N
10
In this context, fluorescent
emission wavelength shorter than that of 2-NBDG (1), blue
emitting molecules, might have a possibility to pass through
GLUTs. In that case, a combined use of such
D-glucose analogues having
HO
O
H+
HO
HO
2-BDG (5)
H2N
OMe
D-form blue mole-
cules with 2-NBDLG (2), or use of -form blue molecules with
2-NBDG (1), could provide excellent information on the mainte-
nance of stereoselectivity. As an initial attempt, we have designed
L
NH
O
N
[2-(benz-2-oxa-1,3-diazol-4-yl)amino]-2-deoxy-
(5)] (Fig. 1), a -glucose analogue having structural similarity
with 2-NBDG (1). 2-NBDG (1) was readily synthesized from
D-glucose [2-BDG
Scheme 2. Synthetic study of 2-BDG (5).
D
the nucleophilic attack and facilitates the substitution reaction.
D
-
In contrast, the absence of a nitro group on the 2-BDG (5) makes
glucosamine hydrochloride with NBD-F in one-step procedure
using a nucleophilic aromatic substitution, because the nitro
group on the NBD moiety activates an aromatic ring towards
it difficult to introduce the BD group at the C2-position of
D-
glucosamine. Therefore, we selected palladium-catalyzed cross

T. Yamamoto et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4088–4096
4095
to give triol 10. However, the final deprotection of the methyl gly-
coside 10 did not proceed under the acidic hydrolysis in 1 M HCl
at 60 °C and prolonged hydrolysis under harsh conditions led to
decomposition of the BD group. Presumably the protonation of
nitrogen atoms prevented the attack of a proton at the glycosidic
oxygen atom (Scheme 2).
(a) Fluorescence intensity of 2-NBDG alone in the green and red
channel and of 2-NBDG / 2-TRLG mixture in these channels measured
at 70% laser excitation
80
NB GREEN
NB RED
NB+TR GREEN
70
60
NB+TR RED
To overcome this problem, we next used more acid-labile
TMSEt (2-trimethylsilylethyl-) group31,32 as an alternative anomer-
ic protecting group (Scheme 3). TMSEt glycosides are widely used
for the various glycosylation because its deblocking reactions are
mild enough not to decompose the BD group. TMSEt-glycosides
are compatible with most reaction conditions used in the carbohy-
drate synthesis, and importantly can be hydrolyzed much faster
50
40
30
20
10
0
than the other alkyl glycosides.33
D-Glucosamine derivative 11
400
500
600
700
800
900
1000
1100
with the Troc group at C2-position was transformed into TMSEt
glycoside 12 by 3 steps. Deacetylation of triacetate 12 followed
by 4,6-O-benzylidation gave the acetal 13. Deprotection of the Troc
group of the compound 13 was carried out to give amine 14. Reac-
tion of the compound 14 with aryl chloride 7 under the same con-
ditions as those employed for the previous coupling as shown in
Scheme 2, provided the desired product 1534 in 32% yield. The yield
slightly decreased in comparison to that obtained in the previous
coupling presumably due to the steric hindrance of the TMSEt
group. Finally the deprotection of the benzylidene group with
80% AcOH aq. and the deblocking of the TMSEt group with 70%
TFA in AcOH furnished 2-BDG (5)35 in 40% yield after the purifica-
tion with reversed-phase chromatography. Its structure was con-
firmed by 1H NMR and MS spectra. Although the fluorescent
intensity of 2-BDG (5) was very weak (not shown) and so no fur-
ther biological evaluation was done, the novel synthetic strategy
introduced here would allow the future development of effective
tracers.
PMT Sensitivity (Volt)
(b) Relationships in a) in semi-logarithmic scale
100
NB GREEN
NB RED
NB+TR GREEN
NB+TR RED
10
1
400 500 600 700 800 900 1000 1100
PMT Sensitivity (Volt)
In summary, we introduced in the present study a combined
Figure 4B. (a) Similar to Fig. 4A, but fluorescent intensities in the green and red
channel of 2-NBDG (1) alone (solid green and red line, respectively) and of 2-NBDG
(1)/2-TRLG (4) mixture (dotted green and red lines, respectively) are superimposed
at 70% laser excitation. (b) Similar to (a), but expressed in semi-logarithmic scale.
Linear relationships are seen in the shaded area.
use of a newly synthesized GLUT-unrecognizable
logue 2-TRLG (4) simultaneously with 2-NBDG (1) for monitoring
stereoselective uptake of -glucose. 2-TRLG (4) is a promising mol-
L-glucose ana-
D
ecule for visualizing disruption of the stereoselectivity, and in par-
ticular, for quantifying partially preserved stereoselectivity that we
met frequently in biological experiments. Contrasted to 2-TRLG (4),
coupling reaction (Buchwald–Hartwig N-arylation reaction)23–28
to construct a C–N bond instead of the nucleophilic aromatic sub-
stitution. The protected 2-amino-glycoside 6 which contains a
L
-glucose analogues bearing compact chromophore as in 2-BDG (5)
have a possibility to pass through GLUTs. The synthesis of another
candidate compound, which is smaller than 2-NBDG (1), is cur-
rently underway in our laboratory.
free 3-OH was prepared from D-glucosamine hydrochloride, while
4-chlorobenz-2-oxa-1,3-diazole (7) could be used as an aryl ha-
lide. Then we investigated the key reaction involved in the forma-
tion of the C–N bond. Reaction of the amine 6 with aryl chloride 7
in the presence of Pd2(dba)3, pyrazole-based phosphine ligand
8,29,30 and sodium t-butoxide successfully produced the product
9 in a moderate yield (Scheme 2). Subsequently, under the mild
acidic conditions the 4,6-O-benzylidene group was easily cleaved
Acknowledgements
This research was supported by Science and Technology Incuba-
tion Program in Advanced Regions from JST (K.Y.), Grant-in-Aid for
1) Ac2O, pyridine, rt
2) HBr/AcOH, CH2Cl2, rt
HO
HO
HO
AcO
Ph
O
1) NaOMe, MeOH
O
O
O
AcO
O
OTMSEt
OTMSEt
AcO
HO
2) PhCH(OMe)
DMF, rt
2, TsOH
3) HO(CH2)2TMS, AgOTf
2,4,6-collidine, CH2Cl2, rt
OH
TrocHN
11
TrocHN
12
RHN
(75% in 2 steps)
R = Troc
R = H
13
14
(71% in 3 steps)
Zn / AcOH
(99% )
Ph
O
O
1) 80% AcOH aq., 40 oC
2) 70% TFA in AcOH, rt
O
7, Pd2(dba)3,
Phosphine Ligand 8
OTMSEt
HO
HN
14
2-BDG (5)
NaOtBu, Toluene
120 oC
(40% in 2 steps )
N
O
(32% )
N
15
Scheme 3. Synthesis of 2-BDG (5).

4096
T. Yamamoto et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4088–4096
found 768.2; Data of low-polar fraction: 1H-NMR (400 MHz, CD3OD): d = 1.85
and 2.00 [m, 8H, H200, H600, H1200and H1600
/b)], 2.60–3.70 [m, 21H, H3, H4,
H5, H6a, H6b, H100, H300, H500, H700, H1100, H1300, H1500 and H1700
/b)], 4.42 [d,
)],6.54 and 6.56 [br
/b)], 7.23 [d, 0.25H, J = 8.0 Hz, H60 of Ph (b)], 7.27 [d,
)], 7.98 [dd, 0.25H, J = 1.8 and 8.0 Hz, H50 of Ph
Scientific Research on Priority Areas (20019003, 20056001, K.Y.),
Research Funds from Research Foundation for Opto-Science and
Technology (K.Y.), and Grant for Hirosaki University Institutional
Research (K.Y.). We are grateful to Dr. Hideaki Matsuoka of Tokyo
University of Agriculture and Technology, Dr. Naoyoshi Chino, Dr.
Kumiko Yoshizawa-Kumagaye and Mr. Takehiro Ishizu of Peptide
Institute, Inc., for their helpful discussions, and Mr. Masahito Kog-
awa and Ms. Rumiko Narita of Hirosaki University for technical
assistance.
(
a
(a
0.25H, J = 8.3 Hz, H1 (b)], 4.90 [d, 0.75H, J = 3.4 Hz, H1 (
a
s ꢀ 2, 2H, H800 and H1000
(a
0.75H, J = 8.0 Hz, H60 of Ph (
a
(b)], 8.02 [dd, 0.75H, J = 1.8 and 8.0 Hz, H50 of Ph (
a
)], 8.54 [d, 0.25H, J = 1.8 Hz,
H30 of Ph (b)], 8.58 [d, 0.75H, J = 1.8 Hz, H30 of Ph (
a)]; MS (ESI) m/z calcd for
C37H42N3O11S2 [M+H]+ 768.2, found 768.2.
18. Dissociation procedure was performed under urethane anesthesia (1.4 g/kg,
i.p.) and neurons were recorded as described previously at 37 °C.
Carbenoxolone (100 lM, SIGMA C4790) was added to the recording solution
to block hemichannels. For details, see: Yamada, K. et al. Science 2001, 292,
1543. C57BL/6J mice were used in accordance with a protocol approved by the
Hirosaki University Institutional Committee for Animal Experimentation.
19. In principle, 2-NBDG (1) entry should produce an increase in the fluorescence
intensity not only in the green but also in the red channel due to the red
component of 2-NBDG (1) spectrum (Fig. 2C, red area). In the experiment in
Fig. 3, however, the detector sensitivity in the red channel was adjusted so that
no such small increase in red fluorescence is discernable, but an abnormal
entry of 2-TRLG (4) could easily be noticed.
20. Speizer, L.; Haugland, R.; Kutchai, H. Biochim. Biophys. Acta 1995, 815, 75.
21. Cheng, Z.; Xiong, Z.; Gheysens, O.; Keren, S.; Chen, X.; Gambhir, S. S.
Bioconjugate Chem. 2006, 17, 662.
22. Kovar, J. L.; Volcheck, W.; Sevick-Muraca, E.; Simpson, M. A.; Olive, D. M. Anal.
Biochem. 2009, 384, 254.
23. Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem. Soc. 1996, 118, 7215.
24. Wolfe, J. P.; Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem. Res. 1998, 31,
805.
25. Muci, A. R.; Buchwald, S. L. In Topics in Current Chemistry; Miyaura, N., Ed.;
Springer: Berlin, 2001; Vol. 219, pp 131–209.
26. Hartwig, J. F. In Modern Arene Chemistry; Astruc, C., Ed.; Wiley-VCH: Weinheim,
2002; pp 107–168.
27. Huang, X.; Anderson, K. W.; Zim, D.; Jiang, L.; Klapars, A.; Buchwald, S. L. J. Am.
Chem. Soc. 2003, 125, 6653.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.04.148.
References and notes
1. Sokoloff, L. Dev. Neurosci. 1993, 15, 194.
2. Gambhir, S. S. Nat. Rev. Cancer 2002, 2, 683.
3. Sokoloff, L.; Reivich, M.; Kennedy, C.; Des Rosiers, M. H.; Patlak, C. S.; Pettigrew,
K. D.; Sakurada, O.; Shinohara, M. Journal. Neurochem. 1977, 28, 897.
4. Schmidt, K. C.; Lucignani, G.; Sokoloff, L. Journal. Nucl. Med. 1996, 37, 394.
5. Yamada, K.; Saito, M.; Matsuoka, H.; Inagaki, N. Nat. Protoc. 2007, 2, 753.
6. Yoshioka, K.; Takahashi, H.; Homma, T.; Saito, M.; Oh, K.-B.; Nemoto, Y.;
Matsuoka, H. Biochim. Biophys. Acta 1996, 1289, 5.
7. Yamada, K.; Nakata, M.; Horimoto, N.; Saito, M.; Matsuoka, H.; Inagaki, N. J. Biol.
Chem. 2000, 275, 22278.
8. Bernardinelli, Y.; Magistretti, P. J.; Chatton, J.-Y. Proc. Natl. Acad. Sci. U.S.A. 2004,
101, 14937.
9. Blomstrand, F.; Giaume, C. J. Neurosci. Res. 2006, 83, 996.
10. Rouach, N.; Koulakoff, A.; Abudara, V.; Willeecke, K.; Giaume, C. Science 2008,
322, 1551.
11. Gandhi, G. K.; Cruz, N. F.; Ball, K. K.; Theus, S. A.; Dienel, G. A. J. Neurochem.
2009, 110, 857.
28. Strieter, E. R.; Blackmond, D. G.; Buchwald, S. L. J. Am. Chem. Soc. 2003, 125,
13978.
29. Singer, R. A.; Caron, S.; McDermott, R. E.; Arpin, P.; Do, N. M. Synthesis 2003,
1727.
30. Singer, R. A.; Dore, M.; Sieser, J. E.; Berliner, M. A. Tetrahedron Lett. 2006, 47,
3727.
12. Yamamoto, T.; Nishiuchi, Y.; Teshima, T.; Matsuoka, H.; Yamada, K. Tetrahedron
Lett. 2008, 49, 6876.
31. Jansson, K.; Frejd, T.; Kihlberg, J.; Magnusson, G. Tetrahedron Lett. 1988, 29, 361.
32. Schlessinger, R. H.; Poss, M. A.; Richardson, S. J. Am. Chem. Soc. 1986, 108, 3112.
33. Jansson, K.; Noori, G.; Magnusson, G. J. Org. Chem. 1990, 55, 3181.
34. Data of compound 15: 1H NMR (400 MHz, CDCl3): d = 0.00 (s, 9H, –CH2–CH2–
TMS), 0.87–0.92 (m, 2H, –CH2–CH2–TMS), 2.82 (br s, 1H, OH), 3.56–3.62 (m, 2H,
–CH2–CH2–TMS, 5-H), 3.66–4.76 (m, 2H, 4-H, 2-H), 3.90–4.05 (m, 2H, –CH2–
CH2–TMS, 6-H), 4.48 (dd, 1H, J = 5.03, 10.5 Hz, 60-H), 4.55 (d, 1H, J = 7.78 Hz, 1-
H), 4.95 (d, 1H, J = 7.78 Hz, NH), 5.66 (s, 1H, Ph-CH), 6.66 (d, 1H, J = 7.32 Hz,
Aromatic), 7.15 (d, 1H, J = 8.69 Hz, Aromatic), 7.30–7.34 (m, 1H, Aromatic),
7.45–7.49 (m, 3H, Aromatic), 7.57–7.60 (m, 2H, Aromatic); 13C NMR (100 MHz,
CDCl3): d = ꢁ1.57, 18.11, 60.52, 65.97, 68.09, 68.62, 72.55, 81.10, 101.87,
102.70, 103.54, 104,09, 126.43, 128.42, 129.39, 136.23, 136.98, 145.04, 149.81.;
MS (ESI) m/z calcd for C24H31N3O6Si [M + H]+ 486.2, found 486.1.
13. Yamada, K.; Matsuoka, H.; Teshima, T.; Yamamoto, T. PCT/JP2009/064053 (WO
2010/016587 A1).
14. LeFevre, P. G. Pharmacol. Rev. 1961, 13, 39.
15. Haddad, G. G.; Jiang, C. Prog. Neurobiol. 1993, 40, 277.
16. Etxeberria, E.; Gonzalez, P.; Tomlinson, P.; Pozueta-Romero, J. J. Exp. Botany
2005, 56, 1905.
17. Since Texas Red sulfonyl chloride consists of a para and an ortho isomer, the
conjugated product is ultimately an isomeric mixture of 2-TRG (3a) (para-
isomer) and 2-TRG (3b) (ortho-isomer). 2-TRG (3) was purified by reverse
phase HPLC on C18 column and eluted as two isomer peaks at 18.9 min (high
polar fraction) and 20.1 min (less polar fraction). Data of high-polar fraction:
1H NMR (400 MHz, CD3OD): d = 1.86 and 2.01 [m, 8H, H200, H600, H1200and H1600
(a
/b) ], 2.60-3.72 [m, 21H, H3, H4, H5, H6a, H6b, H100, H300, H500, H700, H1100,
35. Data of 2-BDG (5): 1H NMR (400 MHz, D2O): (
a isomer) d = 3.40–3.84 (m, 6H,
H1300, H1500 and H1700
(a/b) ], 4.35 [d, 0.3H, J = 8.5 Hz, H1 (b)], 6.55 and 6.73
2-H, 3-H, 4-H, 5-H, 6-H, 60-H), 5.30 (d, 1H, J = 2.75 Hz, 1-H), 6.34–6.37 (m, 1H,
Aromatic), 6.92–6.98 (m, 1H, Aromatic), 7.22–7.26 (m, 1H, Aromatic): (b
isomer) d = 3.40–3.84 (m, 6H, 2-H, 3-H, 4-H, 5-H, 6-H, 60-H), 4.65–4.73 (m, 1H,
1-H), 6.34–6.37 (m, 1H, Aromatic), 6.92–6.98 (m, 1H, Aromatic), 7.22–7.26 (m,
1H, Aromatic), MS (ESI) m/z calcd for C12H15N3O6 [M+H]+ 298.1, found, 298.1.
[s ꢀ 2, 0.7H ꢀ 2, H800 and H1000
(a
)], 6.68 and 6.84 [s ꢀ 2, 0.3H ꢀ 2, H800 and
H1000 )], 719 [d, 0.3H, J = 8.0 Hz, H60 of Ph (b)], 7.24 [d, 0.7H, J = 8.0 Hz, H60 of
(a
Ph (
and 8.0 Hz, H50 of Ph (
J = 1.7 Hz, H30 of Ph (
a
)], 7.98 [dd, 0.3H, J = 1.7 and 8.0 Hz, H50 of Ph (b)], 8.02 [dd, 0.7H, J = 1.7
)], 8.49 [d, 0.3H, J = 1.7 Hz, H30 of Ph (b)], 8.57 [d, 0.7H,
)]; MS (ESI) m/z calcd for C37H42N3O11S2 [M+H]+ 768.2,
a
a