Design, synthesis, and radiosensitizing activities of sugar-hybrid hypoxic cell radiosensitizers

Article abstract of DOI:10.1016/j.bmc.2007.10.035

We have designed sugar-hybrid TX-1877 derivatives conjugated with sugar moieties including β-glucose (β-Glc), β-galactose (β-Gal), α-mannose (α-Man) and N-acetyl-β-galactosamine (β-GalNAc). Compound 1 (TX-1877) was glycosylated with appropriate peracetylated sugars using BF₃-OEt₂ to give acetylated sugar-hybrids, 5 (TX-2244), 6 (TX-2245), 7 (TX-2246), and 10 (TX-2243). Removal of the acetyl groups afforded the sugar-hybrids having free hydroxyl groups, 11 (TX-2141), 12 (TX-2218), 13 (TX-2217) and 14 (TX-2068). We evaluated their radiosensitizing activities by an in vitro radiosensitization assay. All free hydroxyl hybrids have lower enhancement ratio (ER) values (ER ≤ 1.43) and lower n-octanol/water partition coefficient (P_oct) values (P_oct < 1.00 × 10^-2) than does 1 (TX-1877, ER = 1.75, P_oct: 5.60 × 10^-2). All acetylated hybrids have similar P_oct values (3.55 × 10^-2-1.05 × 10^-1) to 1 (TX-1877) and have improved ER values (ER ≥ 1.47) compared to the hybrids having free hydroxyl groups. Among these, 5 (TX-2244) is the most active radiosensitizer (ER = 2.30). We found a good correlation (r = 0.866) between the magnitude of P_oct (logP_oct) and the ER value of 5 (TX-2244), 6 (TX-2245), 7 (TX-2246), 10 (TX-2243) and 1 (TX-1877), suggesting that increasing the hydrophobicity is reflected in increased in vitro radiosensitizing activity. In the present study, we have succeeded in producing sugar-hybrid hypoxic cell radiosensitizers that have an increased radiosensitizing activity that does not depend on increased hydrophobicity.

Full text of DOI:10.1016/j.bmc.2007.10.035

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 675–682
Design, synthesis, and radiosensitizing activities
of sugar-hybrid hypoxic cell radiosensitizers
Takashi Nakae,a Yoshihiro Uto,a,* Motoko Tanaka,a Haruna Shibata,a Eiji Nakata,a
Masahide Tominaga,b Hiroshi Maezawa,b Toshihiro Hashimoto,c Kenneth L. Kirk,d
Hideko Nagasawae and Hitoshi Horia,*
aDepartment of Life System, Institute of Technology and Science, Graduate School, The University of Tokushima,
Minamijosanjimacho-2, Tokushima 770-8506, Japan
bFaculty of Medicine, The University of Tokushima, Tokushima 770-8509, Japan
cFaculty of Pharmaceutical Sciences, Tokushima Bunri University, Yamashiro-cho, Tokushima 770-8514, Japan
dLaboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health,
DHHS, Bethesda, MD 20892, USA
eLaboratory of Pharmaceutical Chemistry, Gifu Pharmaceutical University, Mitahorahigashi-5, Gifu 502-8585, Japan
Received 21 September 2007; revised 9 October 2007; accepted 12 October 2007
Available online 17 October 2007
Abstract—We have designed sugar-hybrid TX-1877 derivatives conjugated with sugar moieties including b-glucose (b-Glc),
b-galactose (b-Gal), a-mannose (a-Man) and N-acetyl-b-galactosamine (b-GalNAc). Compound 1 (TX-1877) was glycosylated with
appropriate peracetylated sugars using BF3-OEt2 to give acetylated sugar-hybrids, 5 (TX-2244), 6 (TX-2245), 7 (TX-2246), and 10
(TX-2243). Removal of the acetyl groups afforded the sugar-hybrids having free hydroxyl groups, 11 (TX-2141), 12 (TX-2218), 13
(TX-2217) and 14 (TX-2068). We evaluated their radiosensitizing activities by an in vitro radiosensitization assay. All free hydroxyl
hybrids have lower enhancement ratio (ER) values (ER 6 1.43) and lower n-octanol/water partition coefficient (Poct) values
(Poct < 1.00 · 10ꢀ2) than does 1 (TX-1877, ER = 1.75, Poct: 5.60 · 10ꢀ2). All acetylated hybrids have similar Poct values
(3.55 · 10ꢀ2–1.05 · 10ꢀ1) to 1 (TX-1877) and have improved ER values (ER P 1.47) compared to the hybrids having free hydroxyl
groups. Among these, 5 (TX-2244) is the most active radiosensitizer (ER = 2.30). We found a good correlation (r = 0.866) between
the magnitude of Poct (logPoct) and the ER value of 5 (TX-2244), 6 (TX-2245), 7 (TX-2246), 10 (TX-2243) and 1 (TX-1877), sug-
gesting that increasing the hydrophobicity is reflected in increased in vitro radiosensitizing activity. In the present study, we have
succeeded in producing sugar-hybrid hypoxic cell radiosensitizers that have an increased radiosensitizing activity that does not
depend on increased hydrophobicity.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
Representative hypoxic cell radiosensitizers are listed in
Figure 1. Etanidazole and misonidazole are well-known
hypoxic cell radiosensitizers but have had limited thera-
Hypoxic cells that exist in many human solid tumors are
characterized by radioresistance.1 The resistance to radi-
ation treatment caused by tumor hypoxia is a major
complicating factor in cancer therapy.2–4 Thus, the
development of compounds that can sensitize hypoxic
tumor cells to radiation has been an important goal
for medicinal chemists.5
Keywords: Hypoxic cell radiosensitizers; Sugar-hybrids; Electronic
states; Hydrophobicity.
*
Corresponding authors. Tel./fax: +81 88 656 7522 (Y.U.); tel.: +81 88
656 7514; fax: +81 88 656 9164 (H.H.); e-mail addresses:
uto@bio.tokushima-u.ac.jp; hori@bio.tokushima-u.ac.jp
Figure 1. Representative hypoxic cell radiosensitizers.
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.10.035

676
T. Nakae et al. / Bioorg. Med. Chem. 16 (2008) 675–682
peutic impact on radiotherapy due to clinical problems
that include dose-limiting side effects such as neurotox-
icity.6 Nimorazole, used locally in Denmark for radio-
therapy of head and neck cancer, is the only hypoxic
cell radiosensitizer currently in clinical use.7
2. Results
As the substrate to be linked to sugars, we chose TX-
1877 (Scheme 1), previously reported by us to be more
potent radiosensitizer than etanidazole15 and have many
additional biological activities such as antimetastatic15
and immunopotentiation activity.15–18 TX-1877 was
conjugated with several sugar moieties, including b-glu-
cose (b-Glc), b-galactose (b-Gal), a-mannose (a-Man),
N-acetyl-b-galactosamine (b-GalNAc), tetra-O-acetyl
b-Glc (b-Glc(OAc)4), tetra-O-acetyl b-Gal (b-Ga-
l(OAc)4), tetra-O-acetyl a-Man (a-Man(OAc)4), and
tri-O-acetyl b-GalNAc (b-GalNAc(OAc)3), to provide
the sugar-conjugated molecules shown in Figure 2.
Our research has been directed toward exploring new
strategies for the development of more effective radio-
sensitizers that will be useful in the clinic.
The sugar moiety has been used effectively in drug
design to improve water solubility and for molecular
recognition. An additional factor relevant to this study
is the significant increases of glucose uptake into tumor
tissue that have been ascribed to enhanced tumor glycol-
ysis. Thus, one possible strategy to improve tumor
affinity of drugs is to link a sugar to a drug or drug phar-
macophoric moiety. Sugar-hybrid molecules designed to
exploit this strategy have been prepared by other
groups. Examples include glyco-conjugated porphy-
rins,8,9 sugar-pendant diamines,10 and other drug
candidates (e.g., cytosine,11 quinoxaline,12 glycidol,13
and b-carboline14). The above considerations led us to
initiate research into the design of sugar-hybrid hypoxic
cell radiosensitizers.
2.1. Molecular orbital calculation
We carried out molecular orbital calculations of the su-
gar-conjugated molecules in order to evaluate the utility
of our sugar-hybrid drug design as a novel hypoxic cell
radiosensitizer (Fig. 3). In all compounds the LUMO is
characterized by a localized increased electron density
on the 2-nitroimidazole ring as same as 1.15 In contrast
to the LUMO, the HOMO is dependent on the structure
of the sugar. Thus, the HOMO for 11, 5, 12, 6, 13, and 7
are localized on the N-methylamide moiety, while for 14
and 10 increased electron densities are localized on the
2-acetamide moiety of the sugar. Energy values of
LUMO (ELUMO) and HOMO (EHOMO) for the series
are given in Table 1. The values ranged from ꢀ1.2 to
ꢀ0.9 eV and were similar to 1.
In this paper, we describe the design and synthesis of su-
gar-hybrid hypoxic cell radiosensitizers. We then discuss
the correlations of the in vitro radiosensitizing activities
and the physicochemical characteristics.
2.2. Chemistry
Reaction of 2-nitroimidazole with methyl bromoacetate
in the presence of potassium carbonate in acetonitrile,
followed by reaction with N-methylethanolamine (one
pot), gave 1 in 92% yield (Scheme 1). Consistent with
1
our previous report,15 the H NMR spectrum of 1 at
25 ꢁC showed proton signals corresponding to the pres-
ence of two rotamers due to a C-N rotational barrier
Scheme 1. One-pot synthesis of TX-1877.
Figure 2. Structure of sugar-hybrid hypoxic cell radiosensitizers.

T. Nakae et al. / Bioorg. Med. Chem. 16 (2008) 675–682
677
Figure 3. Distribution of the electron density (white and blue lobe) of LUMO and HOMO for sugar-hybrid hypoxic cell radiosensitizers.
Table 1. Energy of LUMO (ELUMO) and HOMO (EHOMO) for sugar-
hybrid hypoxic cell radiosensitizers and 1
(Scheme 2). The 1H NMR spectra of all compounds
also showed proton signals corresponding to rotamers,
as described above.
Compound
ELUMO (eV)
EHOMO (eV)
1
5
ꢀ1.1095
ꢀ0.9610
ꢀ1.0060
ꢀ0.9731
ꢀ1.1428
ꢀ0.9912
ꢀ0.9670
ꢀ1.1479
ꢀ1.0418
ꢀ10.1018
ꢀ9.9749
ꢀ9.9804
ꢀ9.7424
ꢀ9.9510
ꢀ9.9044
ꢀ9.8643
ꢀ10.2101
ꢀ9.8371
2.3. Hydrophobic parameters
6
Two hydrophobic parameters-the calculated partition
coefficients (calcdP) and n-octanol/water partition coef-
ficients (Poct)-were determined for each compound in the
series. As shown in Table 3, the calcdP values of 11 or
7
10
11
12
13
14
12 or 13, and 14, 5 or 6 or 7, and 10 were 4.0 · 10ꢀ4
,
1.3 · 10ꢀ4, 5.0 · 10ꢀ2, and 5.1 · 10ꢀ3, respectively. The
Poct values of 5, 6, 7, and 10 were 1.05 · 10ꢀ1
,
3.55 · 10ꢀ2, 6.97 · 10ꢀ2, and 5.10 · 10ꢀ2, respectively.
The Poct values of 11, 12, 13, and 14 were undetectable
(<1.00 · 10ꢀ2).
(see Experimental section). Compound 1 was glycosyl-
ated with various peracetylated sugars in the presence
of BF3-OEt2 in nitromethane (Table 2). The reaction
of 2, 3, and 4 gave 5, 6, and 7 in poor yield (7–15%)
along with aglycon acetate 8 (12–18% yield). Forma-
tion of aglycon acetate by using peracetylated sugars
as glycosyl donor has been studied by some research-
ers.19–22 In contrast, glycosylation of 9 with 1 under
the same conditions produced 10 in 37% without for-
mation of 8. The reason for this difference may be that
the N-acetyl group of 9 can stabilize an N,O-oxazoline
intermediate. Removal of the O-acetyl group from 5, 6,
7, and 10 was carried out with sodium methoxide in
methanol to give 11, 12, 13, and 14 in 45–99% yield
2.4. Radiosensitizing activity
In vitro radiosensitizing activities of the sugar-hybrid
hypoxic cell radiosensitizers were measured at a dose
of 1 mM in EMT6/KU cells under hypoxic conditions.
The enhancement ratios (ER) are shown in Table 3. The
order of radiosensitizing activity was 5 (b-Glc(OAc)4) ꢁ 7
(a-Man(OAc)4) > 1 > 6 (b-Gal(OAc)4) > 10 (b-Gal-
NAc(OAc)3) > 14 (b-GalNAc)
(b-Gal) > 11 (b-Glc). All sugar-hybrids having free
ꢂ 13 (a-Man) ꢂ 12

678
T. Nakae et al. / Bioorg. Med. Chem. 16 (2008) 675–682
Table 2. Glycosylation of 1 with various peracetylated sugars, 2–4 and 9
Run
Sugar
BF3-OEt2
Glycoside (%)
8 (%)
1
2
3
4
2
3
4
9
a-Glc(OAc)5
a-Gal(OAc)5
a-Man(OAc)5
b-GalNAc(OAc)4
2.0 equiv
2.0 equiv
2.0 equiv
2.0 equiv
5
6
(TX-2244)
(TX-2245)
(TX-2246)
(TX-2243)
8
7
12
18
17
—
7
10
15
37
Scheme 2. Deprotection of 5–7 and 10.
Table 3. Hydrophobic parameter (calcdP and Poct) and radiosensitiz-
ing activities of sugar-hybrid hypoxic cell radiosensitizers and 1
were smaller than that of 1 (ER = 1.75). On the other
hand, the acetylated sugar-hybrids, 5 (ER = 2.30), 6
(ER = 1.63), 7 (ER = 1.88), and 10 (ER = 1.47), have
Compound
calcdP
Poct
5.60 · 10ꢀ2
ERa
similar hydrophobicities (Poct: 3.55 · 10ꢀ2–1.05 · 10ꢀ1
)
1
4.5 · 10ꢀ2
5.0 · 10ꢀ2
5.0 · 10ꢀ2
5.0 · 10ꢀ2
5.1 · 10ꢀ3
4.0 · 10ꢀ4
4.0 · 10ꢀ4
4.0 · 10ꢀ4
1.3 · 10ꢀ4
1.75b
2.30
1.63
1.88
1.47
1.33
1.40
1.41
1.43
1.05 · 10ꢀ1
3.55 · 10ꢀ2
6.97 · 10ꢀ2
5.10 · 10ꢀ2
<1.00 · 10ꢀ2
<1.00 · 10ꢀ2
<1.00 · 10ꢀ2
<1.00 · 10ꢀ2
to 1 and have improved ER values compared to the cor-
responding hydroxyl-free hybrids. Among these, 5 was
the most active radiosensitizer in the series.
5
6
7
10
11
12
13
14
3. Discussion
Hypoxic cells present in tumors are a major complicat-
ing factor in cancer therapy, and are an important target
for anticancer drug design. Previous work in drug candi-
date development has explored the introduction of a su-
gar moiety as a device to improve tumor affinity.8–14 The
focus of the present work is on the design, synthesis, and
evaluation of a series of sugar-hybrid radiosensitizers as
an approach to preparing more effective hypoxic cell
radiosensitizers.
a Based on ER value of etanidazole (ER = 1.72).26
b Data from Ref. 15.
hydroxyl groups including 11 (ER = 1.33), 12 (ER =
1.40), 13 (ER = 1.41), and 14 (ER = 1.43) are more
hydrophilic radiosensitizers (Poct < 1.00 · 10ꢀ2) than 1
(Poct = 5.60 · 10ꢀ2). The ER values of these hybrids

T. Nakae et al. / Bioorg. Med. Chem. 16 (2008) 675–682
679
The effectiveness of the sugar-hybrid radiosensitizers
4. Experimental
4.1. General procedures
was evaluated by an in vitro radiosensitizing assay under
hypoxic condition. As shown in Table 3, 11 (TX-2141),
12 (TX-2218), 13 (TX-2217), and 14 (TX-2068) were less
potent radiosensitizers having ER less than 1.6 (the min-
imum effective ER of radiosensitizers in the in vitro as-
say) compared to the ER of 1 (1.75). These hybrid
radiosensitizers were more hydrophilic (Poct < 1.00 ·
10ꢀ2) than 1 (TX-1877, Poct = 5.60 · 10ꢀ2). We found
a good correlation (r = 0.866) between the magnitude
of Poct (logPoct) and the ER value of 5 (TX-2244), 6
(TX-2245), 7 (TX-2246), 10 (TX-2243), and 1 (TX-
1877), suggesting that increasing the hydrophobicity is
reflected in increased radiosensitizing activity. This
result suggests that the hydrophobicity is one of the
major factors affecting the in vitro radiosensitizing
activity.
All reactions were carried out under a nitrogen atmo-
sphere and monitored by TLC using UV light and an
ethanol solution of ammonium molybdate as a develop-
ing agent. Column chromatography was performed on
Kanto Chemical silica gel 60 N (spherical, neutral, 40–
1
50 or 230–400 mesh). H NMR spectra were recorded
in CDCl3, CD3OD or CD3CN on a JOEL JNMEX400
spectrometer (400 MHz) with tetramethylsilane as the
internal standard. Chemical shifts are reported in ppm
and coupling constants are reported in Hz. HRMS were
measured on a JOEL JMS-SX 102A mass spectrometer
using a FAB method. IR spectra were measured on a
Perkin Elmer spectrum RX 1. All melting points were
determined with a Yanagimoto micro melting point
apparatus (MP-S3 model) and are uncorrected. Calcu-
lated partition coefficients (calcdP) were determined by
the PrologP modules in the Pallas 3.0 program of Com-
puDrug Chemistry Ltd. (South San Francisco, Califor-
nia). A Hitachi U-2000 spectrometer was used for
measurement of Poct. All chemicals and solvents were
purchased from Wako Pure Chemical Industries, Ltd.
(Osaka, Japan), Kanto Chemical Co., Inc. (Tokyo,
Japan), Tokyo Chemical Industry Co., Ltd. (Tokyo,
Japan), and Sigma Aldrich Japan (Tokyo, Japan).
Electron affinity (EA = ꢀ ELUMO) is one of the indica-
tors of radiosensitizing activity.23 Previously, Kasai
et al. demonstrated that a decreased EA value or a
shift of the localization of the LUMO electron-density
to the amide moiety resulted in decreased radiosensitiz-
ing activity.15 As shown in Figure 3 and Table 1, no
significant difference in ELUMO value and LUMO local-
ization of the sugar-hybrids was observed from MO
calculations, indicating that the electronic states of su-
gar-hybrid hypoxic cell radiosensitizers do not affect
their radiosensitizing activities. Although it is unclear
why differences in the radiosensitizing activity were ob-
served depending on the nature of the sugar, the most
active radiosensitizer is 5 (TX-2244) having an acety-
lated glucose moiety. Since the acetyl group is a hydro-
lytically labile functional group, it is possible that
intracellular glucose metabolism could contribute to
the radiosensitizing activity of this derivative.24,25 To
reduce toxicity, it is necessary for a drug candidate
to have a Poct value no greater than that of misonidaz-
ole.26 In this regard, 5 (TX-2244) not only has lower
hydrophobicity (Poct = 1.05 · 10ꢀ1) but also higher
radiosensitizing activity (ER = 2.30 at 1 mM) than
4.2. Synthesis
4.2.1. One-pot synthesis of 1 (TX-1877). Methyl bromo-
acetate (1.1 mL, 11.3 mmol, 1.1 equiv) was added to a
mixture of 2-nitroimidazole (1.13 g, 10.0 mmol) and
potassium carbonate (845 mg, 6.00 mmol, 0.6 equiv) in
CH3CN (40 mL), and the mixture was refluxed for
20 min. After cooling to room temperature, N-methy-
lethanolamine (2.5 mL, 31.1 mmol, 3.1 equiv) was added
and the mixture was refluxed for 17 h. After cooling to
room temperature, the precipitate was filtered and
washed with CH3CN. The filtrate was evaporated under
reduced pressure and the residue was purified by silica
gel column chromatography (CH2Cl2/MeOH, 20:1) to
misonidazole
(Poct = 4.22 · 10ꢀ1
,
ER = 1.72
at
1 mM).27 We thus have achieved the design of a su-
gar-hybrid hypoxic cell radiosensitizer that has in-
creased radiosensitizing activity along with controlled
hydrophobicity.
1
give TX-1877 (2.09 g, 92%) as a light yellow solid. H
NMR (400 MHz, CD3OD): d 7.40 (d, J = 0.98 Hz,
0.45H), 7.37 (d, J = 0.98 Hz, 0.55H), 7.16 (d,
J = 0.98 Hz, 0.55H), 7.15 (d, J = 1.2 Hz, 0.45H), 5.54
(s, 1.1H), 5.44 (s, 0.9H), 3.79 (t, J = 5.1 Hz, 1.1H),
3.67 (t, J = 5.6 Hz, 0.9H), 3.55 (t, J = 5.1 Hz, 1.1H),
3.51 (t, J = 5.6 Hz, 0.9H), 3.21 (s, 1.35H), 2.98 (s,
1.65H). Other experimental data were consistent with
previous work.15
In summary, we have presented herein the design and
synthesis of sugar-hybrid hypoxic cell radiosensitizers.
The synthesis was based on glycosylation of 1 (TX-
1877), a radiosensitizer previously prepared in our
laboratory, with appropriate peracetylated sugars in
the presence of BF3-OEt2. Removal of the O-acetyl
group afforded sugar-hybrids containing the free hy-
droxyl groups. In vitro radiosensitizing activity of
the compounds in the series was related to their
hydrophobicities. Among these, 5 (TX-2244) is the
most active radiosensitizer having both higher radio-
sensitizing activity and lower hydrophobicity com-
pared to misonidazole. We thus succeeded in the
design of a hybrid molecule which has an increased
radiosensitizing activity that does not depend on in-
creased hydrophobicity.
4.2.2. General procedure for glycosylation. BF3-OEt2 (2.0
equiv) was added dropwise to a mixture of TX-1877 and
peracetylated sugar (1.0 equiv) in CH3NO2 at 0 ꢁC.
After stirring for 30 min, the mixture was warmed to
room temperature and then stirred for 3 days. The mix-
ture was diluted with CH2Cl2 and saturated NaHCO3
solution. The organic layer was washed with saturated
NaCl solution and dried over anhydrous Na2SO4. The
solvent was evaporated under reduced pressure and
the residue was purified by silica gel column chromatog-

680
T. Nakae et al. / Bioorg. Med. Chem. 16 (2008) 675–682
raphy (EtOAc/MeOH or CH2Cl2/MeOH as the eluent)
to give the product and/or 8.
2.13H), 2.01 (s, 0.87H), 1.96 (s, 0.87H), 1.90 (s,
2.13H); FAB-HRMS (m/z): [M+H]+ calcd for
C22H32N5O12, 558.2048; found, 558.2075.
4.2.2.1. [2-(2-Nitro-1H-imidazol-1-yl)-N-methyl-acet-
amido]ethyl 2,3,4,6-tetra-O-acetyl-b-D-glucopyranoside
(5, TX-2244). Light yellow solid (8%); mp 67–69 ꢁC H
4.2.3. General procedure for deprotection. The mixture of
peracetylated glycoside and a small amount of sodium
methoxide in MeOH was stirred at room temperature.
After consumption of starting material, Dowex resin
(50W · 4, H+ form) was added and the precipitate was
filtered. The filtrate was evaporated under reduced pres-
sure. The residue was purified by silica gel column chro-
matography (CH2Cl2/MeOH, 4:1) to give the product.
1
NMR (CDCl3, 400 MHz): d 7.44 (d, J = 0.98 Hz,
0.67H), 7.19 (d, J = 0.98 Hz, 0.33H), 7.18 (d,
J = 0.98 Hz, 0.67H), 7.13 (d, J = 0.98 Hz, 0.33H),
5.40–4.98 (m, 5H), 4.52 (d, J = 8.1 Hz, 0.67H), 4.50 (d,
J = 7.8 Hz, 0.33H), 4.27 (dd, J = 4.4, 12.4 Hz, 1H),
4.19 (dd, J = 2.4, 4.4 Hz, 0.67H), 4.16 (dd, J = 2.4,
4.4 Hz, 0.33H), 4.11–4.07 (m, 0.67H), 3.98–3.93 (m,
0.33H), 3.76–3.62 (m, 3H), 3.56–3.50 (m, 1H), 3.16 (s,
2H), 2.96 (s, 1H), 2.09 (s, 2H), 2.08 (s, 1H), 2.06 (s,
1H), 2.04 (2s, 3H), 2.03 (s, 2H), 2.01 (s, 3H); FAB-
HRMS (m/z): [M+H]+ calcd for C22H31N4O13,
559.1888; found, 559.1909.
4.2.3.1. [2-(2-Nitro-1H-imidazol-1-yl)-N-methyl-acet-
amido]ethyl b-D-glucopyranoside (11, TX-2141). White
solid (64%); mp 89–91 ꢁC; 1H NMR (CD3CN, 400
MHz):
d
7.25(d, J = 1.2 Hz, 0.33H), 7.21 (d,
J = 1.2 Hz, 0.67H), 7.11 (d, J = 1.2 Hz, 0.33H), 7.10
(d, J = 0.98 Hz, 0.67H), 5.46 (d, J = 16.6 Hz, 0.67H),
5.40 (d, J = 16.6 Hz, 0.67H), 5.26 (s, 0.66H), 4.30 (d,
J = 7.6 Hz, 0.67H), 4.24 (d, J = 7.8 Hz, 0.33H), 3.95–
3.86 (m, 2H), 3.71–3.64 (m, 2H), 3.52–3.43 (m, 4.33H),
3.31–3.18 (m, 5.34H), 3.12 (s, 1H), 3.07–2.97 (m,
0.33H), 2.92 (s, 2H); FAB-HRMS: m/z [M + H]+ calcd
for C14H23N4O9, 391.1465; found, 391.1473; FT-IR
(KBr): 3384, 2892, 1654, 1543, 1493, 1416, 1369, 1299,
4.2.2.2. [2-(2-Nitro-1H-imidazol-1-yl)-N-methyl-acet-
amido]ethyl 2,3,4,6-tetra-O-acetyl-b-D-galactopyranoside
1
(6, TX-2245). Light yellow solid (7%); mp 74-76 ꢁC; H
NMR (CDCl3, 400 MHz): d 7.48 (d, J = 0.98 Hz,
0.67H), 7.19 (d, J = 1.2 Hz, 0.33H), 7.17 (d, J = 1.2
Hz, 0.67H), 7.12 (d, J = 0.98 Hz, 0.33H), 5.44–5.40 (m,
2H), 5.30–5.18 (m, 1H), 5.13 (d, J = 10.0 Hz, 0.67H),
5.09 (d, J = 10.0 Hz, 0.33H), 5.06–5.01 (m, 1H), 4.49
(d, J = 7.8 Hz, 0.67H), 4.48 (d, J = 7.8 Hz, 0.33H),
4.22 (dd, J = 6.6, 8.5 Hz, 0.67H), 4.19 (dd, J = 6.6,
8.5 Hz, 0.33H), 4.15–4.10 (m, 2H), 4.00–3.90 (m, 1H),
3.78–3.63 (m, 2H), 3.56–3.47 (m, 1H), 3.17 (s, 1H),
2.97 (s, 2H), 2.19 (s, 2H), 2.17 (s, 1H), 2.07 (s, 1H),
2.06 (s, 2H), 2.05 (s, 1H), 2.02 (s, 2H), 2.01 (s, 2H),
1.99 (s, 1H); FAB-HRMS (m/z): [M+H]+ calcd for
C22H31N4O13, 559.1888; found, 559.1907.
1162, 1078, 1034 cmꢀ1
.
4.2.3.2. [2-(2-Nitro-1H-imidazol-1-yl)-N-methyl-acet-
amido]ethyl b-D-galactopyranoside (12, TX-2218). White
solid (45%); mp 91–94 ꢁC; 1H MNR (CD3OD,
400 MHz): d 7.52 (d, J = 0.98 Hz, 0.67H), 7.42 (d,
J = 0.98 Hz, 0.33H), 7.16 (d, J = 0.98 Hz, 0.33H), 7.13
(d, J = 0.98 Hz, 0.67H), 5.68 (d, J = 16.6 Hz, 0.67H),
5.58 (d, J = 16.6 Hz, 0.67H), 5.46 (d, J = 16.6 Hz,
0.33H), 5.42 (d, J = 16.6 Hz, 0.33H), 4.31 (d,
J = 7.8 Hz, 0.67H), 4.22 (d, J = 7.1 Hz, 0.33H), 4.06–
3.93 (m, 1.67H), 3.82 (d, J = 2.2 Hz, 1H), 3.79–3.63
(m, 4H), 3.59–3.44 (m, 3.33H), 3.23 (s, 0.99H), 2.98 (s,
2.01H); FAB-HRMS: m/z [M+H]+ calcd for
C14H23N4O9, 391.1465; found, 391.1492; FT-IR (KBr):
3400, 2935, 1651, 1548, 1494, 1370, 1299, 1161,
4.2.2.3. [2-(2-Nitro-1H-imidazol-1-yl)-N-methyl-acet-
amido]ethyl 2,3,4,6-tetra-O-acetyl-a-D-mannopyranoside
(7, TX-2246). Light yellow solid (15%); mp 60-62 ꢁC; 1H
NMR (CDCl3, 400 MHz): d 7.46 (d, J = 0.98 Hz,
0.43H), 7.19 (2d, J = 1.2 Hz, 1H), 7.17 (d, J = 0.98 Hz,
0.57H), 5.36–5.28 (m, 3H), 5.26–5.18 (m, 2H), 4.89 (d,
J = 1.7 Hz, 0.43H), 4.82 (d, J = 1.7 Hz, 0.57H), 4.30
(dd, J = 5.4, 12.4 Hz, 0.43H), 4.25 (dd, J = 5.7,
12.3 Hz, 0.57H), 4.16–4.10 (m, 1H), 4.02–3.97 (m, 1H),
3.93–3.86 (m, 1H), 3.74–3.68 (m, 1H), 3.67–3.44 (m,
2H), 3.27 (s, 1.71H), 3.04 (s, 1.29H), 2.18 (s, 1.29H),
2.17 (s, 1.71H), 2.07 (s, 3H), 2.06 (s, 1.71H), 2.05 (s,
1.29H), 2.01 (s, 3H); FAB-HRMS (m/z): [M+H]+ calcd
for C22H31N4O13, 559.1888; found, 559.1901.
1075 cmꢀ1
.
4.2.3.3. [2-(2-Nitro-1H-imidazol-1-yl)-N-methyl-acet-
amido]ethyl a-D-mannopyranoside (13, TX-2217). White
solid (99%); mp 187–189 ꢁC; 1H NMR (CD3OD,
400 MHz): d 7.44 (d, J = 1.2 Hz, 0.5H), 7.40 (d,
J = 1.2 Hz, 0.5H), 7.17 (d, J = 1.2 Hz, 1H), 5.56–5.45
(m, 2H), 4.83 (d, J = 1.7 Hz, 0.5H), 4.76 (d,
J = 1.7 Hz, 0.5H), 4.02–3.98 (m, 0.5H), 3.93–3.86 (m.
2.5H), 3.85–3.47 (m, 7H), 3.23 (s, 1.5H), 3.00 (s,
1.5H); FAB-HRMS: m/z [M+H]+ calcd for
C14H23N4O9, 391.1465; found, 391.1444; FT-IR (KBr):
3383, 2938, 1649, 1544, 1495, 1414, 1371, 1299, 1226,
4.2.2.4. [2-(2-Nitro-1H-imidazol-1-yl)-N-methyl-acet-
amido]ethyl 2-acetamido-2-deoxy-3,4,6-tri-O-acetyl-b-D-
galactopyranoside (10, TX-2243). Light yellow solid
1
(37%); mp 90-92 ꢁC; H NMR (CDCl3, 400 MHz): d
7.56 (d, J = 0.98 Hz, 0.71H), 7.20 (d, J = 0.98 Hz,
0.29H), 7.17 (d, J = 0.98 Hz, 0.71H), 7.15 (d,
J = 0.98 Hz, 0.29H), 5.84 (d, J = 9.0 Hz, 1H), 5.40–
5.04 (m, 4H), 4.62 (d, J = 8.3 Hz, 0.71H), 4.49 (d,
J = 8.5 Hz, 0.29H), 4.27–4.10 (m, 4H), 4.02–3.83 (m,
1.71H), 3.77–3.64 (m, 1.71H), 3.52–3.47 (m, 0.29H),
3.39–3.32 (m, 0.29H), 3.17 (s, 0.87H), 2.95 (s, 2.13H),
2.17 (2s, 3H), 2.06 (s, 2.13H), 2.04 (s, 0.87H), 2.02 (s,
1137, 1063, 975, 845, 809, 682, 656, 637, 609 cmꢀ1
.
4.2.3.4. [2-(2-Nitro-1H-imidazol-1-yl)-N-methyl-acet-
amido]ethyl 2-acetamido-2-deoxy-b-D-galactopyranoside
1
(14, TX-2068). White solid (79%); mp 121–123 ꢁC; H
NMR (CD3OD, 400 MHz): d 7.72 (d, J = 1.2 Hz,
0.78H), 7.43 (d, J = 1.2 Hz, 0.22H), 7.17 (d,
J = 1.2 Hz, 0.22H), 7.15 (d, J = 1.2 Hz, 0.78H), 5.57

T. Nakae et al. / Bioorg. Med. Chem. 16 (2008) 675–682
681
(d, J = 16.6 Hz, 1H), 5.44 (d, J = 16.6 Hz, 1H), 4.34 (d,
J = 8.3 Hz, 0.78H), 4.33 (d, J = 8.3 Hz, 0.22H), 4.11–
4.05 (m, 1.67H), 4.00–3.95 (m. 0.33H), 3.83 (d,
J = 3.2 Hz, 1H), 3.80–3.61 (m, 4.67H), 3.58–3.48 (m,
2.33H), 3.19 (s, 0.67H), 2.96 (s, 2.33H), 1.99 (s,
0.67H), 1.98 (s, 2.33H); FAB-HRMS: m/z [M + Na]+
calcd for C16H25N5O9Na, 454.1544; found, 454.1571;
FT-IR (KBr): 3397, 2945, 2894, 1650, 1600, 1549,
1508, 1495, 1413, 1371, 1300, 1227, 1161, 1146, 1123,
Japan. The authors thank Dr. Hiromu Nishitani, Mr.
Yasuo Nishimoto, Mr. Taro Kishi, and the staff of
Department of Radiology, School of Medicine, The
University of Tokushima, for generous help with the
radiological experiments. The authors also thank Ms.
Maki Nakamura, Ms. Kayoko Yamashita, and the staff
of our Faculty for measurement of NMR, and Ms. K.
Kirk acknowledges support from the NIDDK intramu-
ral research program.
1082, 1058, 845, 789, 750, 650, 613 cmꢀ1
.
4.3. Molecular orbital calculation
References and notes
1. Ewing, D.; Jones, S. R. Arch. Biochem. Biophys. 1987, 254,
53.
2. Nordsmark, M.; Bentzen, S. M.; Rudat, V.; Brizel, D.;
Lartigau, E.; Stadler, P.; Becker, A.; Adam, M.; Molls,
M.; Dunst, J.; Terris, D. J.; Overgaard, J. Radiother.
Oncol. 2005, 77, 18.
The semi-empirical molecular orbital calculations were
carried out using the program MOPAC 2000 (Fujitsu
WinMOPAC ver. 3.0. Fujitsu Ltd., Tokyo, Japan) with
PM 3 Hamiltonian.
4.4. Measurement of n-octanol/water partition coefficient
(Poct
3. Adams, G. E.; Hasan, N. M.; Joiner, M. C. Radiother.
Oncol. 1997, 44, 101.
)
4. Hall, E. J.; Giaccia, A. J. In Radiobiology for the
Radiologist, 6th ed.; Lippincott Williams & Wilkins:
Philadelphia, 2006; pp 85–105.
5. Nagasawa, H.; Uto, Y.; Kirk, K. L.; Hori, H. Biol. Pharm.
Bull. 2006, 29, 2335.
6. Paulson, O. B.; Melggard, B.; Hansen, H. S.; Kamienie-
cka, Z.; Køhler, O.; Hansen, J. M.; Pedersen, A. G.; Tang,
X.; Trojaborg, W. Acta Neurol. Scand. Suppl. 1984, 100,
133.
7. Overgaard, J.; Eriksen, J. G.; Nordsmark, M.; Alsner, J.;
Horsman, M. R. Lancet. Oncol. 2005, 6, 757.
8. Pandey, S. K.; Zheng, X.; Morgan, J.; Missert, J. R.; Liu,
T. H.; Shibata, M.; Bellnier, D. A.; Oseroff, A. R.;
Henderson, B. W.; Dougherty, T. J.; Pandey, R. K. Mol.
Pharm. 2007, 4, 448.
The measurement of partition coefficients (Poct) was
referred to Shake Flask Method. Stock solutions of each
compound were prepared by dissolving 1 mg of each
compound in 0.1 M phosphate buffer (pH 7.40) and
the absorbance at 325 nm (ABSpre) was measured. The
stock solution (3 mL) and phosphate buffer-saturated
n-octanol (20–300 mL) were combined and the mixture
was vigorously stirred at room temperature for 2 h.
Following separation of the two phases, the aqueous
layer was collected and the absorbance was measured
at 325 nm (ABSpost). The values of Poct were calculated
as shown in Eq. 1.
9. Sylvain, I.; Zerrouki, R.; Granet, R.; Huang, Y. M.;
Lagorce, J. F.; Guilloton, M.; Blais, J. C.; Krausz, P.
Bioorg. Med. Chem. 2002, 10, 57.
ABSpre ꢀ ABSpost stock solution ðmlÞ
Poct
¼
ꢃ
ð1Þ
ABSpost
n ꢀ octanol ðmlÞ
10. Mikata, Y.; Shinohara, Y.; Yoneda, K.; Nakamura, Y.;
Esaki, K.; Tanahashi, M.; Brudzin˜ska, I.; Hirohara, S.;
Yokoyama, M.; Mogami, K.; Tanase, T.; Kitayama, T.;
Takashiba, K.; Nabeshima, K.; Takagi, R.; Takatani, M.;
Okamoto, T.; Kinoshita, I.; Doe, M.; Hamazawa, A.;
Morita, M.; Nishida, F.; Sakakibara, T.; Orvig, C.; Yano,
S. J. Org. Chem. 2001, 66, 3783.
11. Idutsu, Y.; Sasaki, A.; Matsumura, S.; Toshima, K.
Bioorg. Med. Chem. Lett. 2005, 15, 4332.
12. Toshima, K.; Ozawa, T.; Kimura, T.; Matsumura, S.
Bioorg. Med. Chem. Lett. 2004, 14, 2777.
13. Toshima, K.; Okuno, Y.; Matsumura, S. Bioorg. Med.
Chem. Lett. 2003, 13, 3281.
14. Toshima, K.; Okuno, Y.; Nakajima, Y.; Matsumura, S.
Bioorg. Med. Chem. Lett. 2002, 12, 671.
15. Kasai, S.; Nagasawa, H.; Yamashita, M.; Masui, M.;
Kuwasaka, H.; Oshodani, T.; Uto, Y.; Inomata, T.; Oka,
S.; Inayama, S.; Hori, H. Bioorg. Med. Chem. 2001, 9, 453.
16. Kasai, S.; Nagasawa, H.; Kuwasaka, H.; Oshodani, T.;
Nishioka, A.; Ogawa, Y.; Yoshida, S.; Inayama, S.;
Inomata, T.; Hori, H. Int. J. Radiat. Oncol. Biol. Phys.
1998, 42, 799.
4.5. In vitro radiosensitizing assay
In vitro radiosensitization was measured in EMT6/KU
single cells under hypoxic conditions. The hypoxic cell
radiosensitizer was added to an EMT6/KU cell
(2 · 106 cells/mL) suspension at a dose of 1 mM in test
tubes and treated with 95% N2–5% CO2 gas for 1 hr.
The cells were irradiated with a linear accelerator
(MEVATRON KD2 PRIMUS) by 8–36 Gy at an X-
ray dose rate of 2.0 Gy/min. After irradiation, colony
formation assays were performed. Enhancement ratios
(ERs) were determined from the ratio of radiation doses
required to reduce the surviving fraction of EMT6/KU
cells to 1%. Usually, each ER value of radiosensitizer
was obtained from survival curves consisting of four
or five points per curve and converted based on the
ER value of etanidazole (ER = 1.72).26
17. Abou-Bedair, F. A.; Hori, H.; Nagasawa, H.; Uto, Y.;
Abu-Zeid, M.; Inayama, S. Biol. Pharm. Bull. 2002, 25,
591.
Acknowledgments
18. Oshikawa, T.; Okamoto, M.; Ahmed, S. U.; Furuichi, S.;
Tano, T.; Sasai, A.; Kan, S.; Kasai, S.; Uto, Y.;
Nagasawa, H.; Hori, H.; Sato, M. Int. J. Cancer 2005,
116, 571.
This work was partially supported by Grant-in-Aid for
Young Scientist (B) (No. 18790892) from the Ministry
of Education, Culture, Sports, Science and Technology,

682
T. Nakae et al. / Bioorg. Med. Chem. 16 (2008) 675–682
19. Murakami, T.; Hirono, R.; Sato, Y.; Furusawa, K.
Carbohydr. Res. 2007, 342, 1009.
20. Kong, F. Carbohydr. Res. 2007, 342, 345.
21. Aich, U.; Loganathan, D. Carbohydr. Res. 2006, 341, 19.
22. Gouin, S. G.; Pilgrim, W.; Porter, R. K.; Murphy, P. V.
Carbohydr. Res. 2005, 340, 1547.
24. Kim, J. W.; Dang, C. V. Cancer Res. 2006, 66, 8927.
25. Pavlovic, M.; Wroblewski, K.; Manevich, Y.; Kim, S.;
Biaglow, J. E. Br. J. Cancer Suppl. 1996, 74, S222.
26. Brown, J. M.; Workman, P. Radiat. Res. 1980, 82,
171.
27. Nagasawa, H.; Bando, M.; Hori, H.; Satoh, T.; Tada, T.;
Onoyama, Y.; Inayama, S. Int. J. Radiat. Oncol. Biol.
Phys. 1992, 22, 561.
23. Adams, G. E.; Flockhart, I. R.; Smithen, C. E.; Stratfold,
I. J.; Wardman, P.; Watts, M. E. Radiat. Res. 1976, 67, 9.