Novel D-xylose derivatives stimulate muscle glucose uptake by activating AMP-activated protein kinase α

Article abstract of DOI:10.1021/jm8008713

Type 2 diabetes mellitus has reached epidemic proportions; therefore, the search for novel antihyperglycemic drugs is intense. We have discovered that D-xylose increases the rate of glucose transport in a non-insulin-dependent manner in rat and human myot

Full text of DOI:10.1021/jm8008713

8096
J. Med. Chem. 2008, 51, 8096–8108
Novel D-Xylose Derivatives Stimulate Muscle Glucose Uptake by Activating AMP-Activated
Protein Kinase r
Arie Gruzman,† Ofer Shamni,† Moriya Ben Yakir,† Daphna Sandovski,† Anna Elgart,‡ Evgenia Alpert,† Guy Cohen,†
Amnon Hoffman,‡ Yehoshua Katzhendler,§ Erol Cerasi,| and Shlomo Sasson*,†
Departments of Pharmacology, Medicinal Chemistry, and Pharmaceutics, School of Pharmacy, Faculty of Medicine,
The Hebrew UniVersity, 91120 Jerusalem, Israel, and Endocrinology and Metabolism SerVice, Department of Internal Medicine,
Hadassah-Hebrew UniVersity Medical Center, 91120 Jerusalem, Israel
ReceiVed July 15, 2008
Type 2 diabetes mellitus has reached epidemic proportions; therefore, the search for novel antihyperglycemic
drugs is intense. We have discovered that D-xylose increases the rate of glucose transport in a non-insulin-
dependent manner in rat and human myotubes in vitro. Due to the unfavorable pharmacokinetic properties
of D-xylose we aimed at synthesizing active derivatives with improved parameters. Quantitative structure-activity
relationship analysis identified critical hydroxyl groups in D-xylose. These data were used to synthesize
various hydrophobic derivatives of D-xylose of which compound 19 the was most potent compound in
stimulating the rate of hexose transport by increasing the abundance of glucose transporter-4 in the plasma
membrane of myotubes. This effect resulted from the activation of AMP-activated protein kinase without
recruiting the insulin transduction mechanism. These results show that lipophilic D-xylose derivatives may
serve as prototype molecules for the development of novel antihyperglycemic drugs for the treatment of
diabetes.
Introduction
(GLUT4) translocation to the plasma membrane of insulin-
sensitive cells and thus increases glucose influx in a non-insulin-
dependent manner.5 In view of the reduced insulin secretory
capacity and peripheral insulin resistance of type 2 diabetic
patients, this mechanism is extremely attractive.
The World Health Organization has defined type 2 diabetes
a worldwide epidemic and predicted that the number of people
diagnosed with the disease would reach over 380 million in
2025. A major pathogenic defect in the disease is insulin
resistance, which is characterized by the impeded capacity of
peripheral tissues to utilize glucose effectively in face of
hyperinsulinemia.1 Hyperglycemia contributes to this phenom-
enon by downregulating the rate of glucose transport and
utilization in peripheral tissues, most notably in skeletal
muscle,2,3 which is the main consumer of glucose. Modern
antidiabetic drug therapy aims at a strict regulation of glucose
homeostasis to prevent late complications of diabetes. However,
monotherapy and combination therapy with oral agents often
fail to achieve near-normoglycemia in diabetic patients, hence
the frequent need for insulin treatment.4 Therefore, the search
for novel antidiabetic drugs is intense. Recent work on the
molecular mechanisms mediating insulin- and non-insulin-
dependent augmentation of glucose transport in insulin-sensitive
tissues has identified new potential targets for antidiabetic
drugs.4 Among these, the enzyme 5′-AMP-activated protein
kinase (AMPK)a emerges as a unique target for drug develop-
ment, since in its active form it induces glucose transporter-4
We have discovered, and extensively investigated, a new
regulatory pathway of carbohydrate metabolism, the phenom-
enon of glucose-induced downregulation of glucose transport,
which is operative in a variety of cell types, most notably
myotubes and skeletal muscle.2,4,6,7 In a search for various
carbohydrates that mimic this effect of glucose, we screened
various pentoses and made a unique discovery: high levels of
D-xylose upregulated the glucose transport system in myotubes.
We used D-xylose as a prototype molecule for the synthesis of
active derivatives with adequate pharmacodynamic and phar-
macokinetic parameters. We now report the results of this effort
and describe the cellular and molecular mechanisms of action
of D-xylose and such active hydrophobic derivatives.
Results and Discussion
D-Xylose Augments the Rate of Hexose Transport in
Myotubes in a Dose- and Time-Dependent Manner. In a
search for carbohydrates that regulate glucose transport in
skeletal muscles, we tested the pentoses D-xylose, D-arabinose,
D-lyxose, and D-ribose in cultured L6 myotubes. The myotubes
were preconditioned for 24 h at 23.0 mM D-glucose, then
washed, and incubated for up to 10 h in medium containing
the same D-glucose concentration and 20.0 mM pentose. The
cultures were thoroughly washed with PBS at room temperature
at the indicated times to remove all extracellular carbohydrates,
and the rate of dGlc uptake was measured. Figure 1A shows
that the rate of hexose transport increased markedly within 6 h
of exposure to D-xylose and D-arabinose (1.77 ( 0.03- and 1.38
( 0.02-fold, respectively, over the basal rate) and then gradually
declined and approached the basal uptake level at 10 h. D-Lyxose
and D-ribose had no significant effect on the hexose transport
capacity of the myotubes, nor did 20.0 mM nonmetabolizable
* To whom correspondence should be addressed. Phone: +972-2-675-
8798. Fax: +972-2-675-8741. E-mail: ShlomoSasson@huji.ac.il.
†
Department of Pharmacology.
Department of Pharmaceutics.
Department of Medicinal Chemistry.
Endocrinology and Metabolism Service.
‡
§
|
a Abbreviations: AICAR, 5-aminoimidazole-4-carboxamide-1-ꢀ-ribo-
furanosyl 5-ꢀ-monophosphate; Akt/PKB, protein kinase B; AMPK, AMP-
activated protein kinase; RMEM, R-minimal essential medium; BSA, bovine
serum albumin; dGlc, 2-deoxy-D-glucose; DMF, dimethylformamide;
DMSO, dimethyl sulfoxide; DNP, 2,4-dinitrophenol; DTT, dithiothreitol;
FCS, fetal calf serum; GLUT, glucose transporter; MeGlc, 3-O-methyl-D-
glucose; NFDM, nonfat dry milk; OPD, O-phenylenediamine; PBS,
phosphate-buffered saline; PEG, polyethylene glycol; PPTS, pyridinium
p-toluenesulfonic acid; QSAR, quantitative structure-activity relationship;
SDS, sodium dodecyl sulfate; TCA, trichloroacetic acid; THF, tetrahydrofuran.
10.1021/jm8008713 CCC: $40.75
2008 American Chemical Society
Published on Web 12/02/2008

D-Xylose DeriVatiVes Stimulate Muscle Glucose Uptake
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 8097
effects of D-xylose were observed at 5.8 and 13.7 mM
(Supporting Information Figure 1). Therefore, in further experi-
ments we studied the maximal effect of D-xylose at 20 mM
following 7 h of incubation.
To determine whether D-xylose increased the rate of hexose
transport via glucose transporters, or also stimulated intracellular
phosphorylation by hexokinase, we measured the rate of
transport of the glucose analogue [3H]MeGlc, which, unlike
dGlc, is not a substrate for hexokinase.8 Figure 1B shows that
D-xylose induced comparable stimulations of dGlc and MeGlc
accumulation in myotubes (1.51 ( 0.12- and 1.69 ( 0.15-fold
above the respective control values), suggesting that D-xylose
increased the plasma membrane content and/or activity of
glucose transporters in L6 myotubes. To test this, we used L6
myotubes stably expressing GLUT1myc or GLUT4myc. The
cell surface expression of these transporters was estimated by
immunodetection of the myc epitope in an extracellular loop
of the transporter.9 D-Xylose increased significantly the abun-
dance of GLUT4myc, but not GLUT1myc, in the plasma
membrane of L6 myotubes (Figure 1C). The figure also shows
that the plasma membrane abundance of GLUT4myc and
GLUT1myc in L6 myotubes exposed to 23.0 mM D-glucose
for 24 h was reduced nearly by 50% compared to cultures
maintained at 5.5 mM D-glucose. These findings confirm
previous data on glucose-induced downregulation of glucose
transporters in skeletal muscle2,7,10 and reconfirm the suitability
of the myc-tagged L6 cell lines for investigating the muscle
glucose transport system.
Although it has been shown that pentoses enter cells,11 little
is known about the pentose transport kinetics or pentose
transporters in skeletal muscle. We therefore determined the
kinetics of D-xylose transport in L6 myotubes. Figure 2A shows
a saturable D-xylose influx into L6 myotubes, which reaches
maximum within 30 s. Moreover, Figure 2B shows that D-xylose
does not utilize glucose transporters to enter myotubes: when
it was added directly to the [3H]dGlc uptake mixture, even at a
concentration as high as 20 mM, this pentose failed to alter the
rate of [3H]dGlc uptake. As expected, when unlabeled D-glucose
and dGlc, which utilize glucose transporters to enter cells, were
added to the uptake mixture, they reduced effectively and dose-
dependently the accumulation of [3H]dGlc in the myotubes
(Figure 2B). These data suggest the presence of a highly specific
pentose transport system in skeletal muscles. Such a specific
transporter of D-xylose explains well the ability of the pentose
to exert significant biological effects in the presence of high
concentrations of D-glucose.
Figure 1. D-Xylose augments the rate of hexose transport in L6
myotubes in a dose- and time-dependent manner. (A) Time-course
analysis. Myotube cultures were washed and received similar fresh
medium supplemented with 20.0 mM D-xylose (9), D-arabinose (0),
D-lyxose (b), or D-ribose (∆). Control myotubes received fresh medium
without (O) or with 20.0 mM L-glucose (2). Myotubes incubated with
20.0 mM D-xylose for 6 h were washed and received fresh medium
supplemented with 23.0 mM D-glucose ([). The myotubes were then
washed at the indicated times and taken for the standard [3H]dGlc uptake
assay, as described in the Experimental Section. (B) Effect of D-xylose
on the rate of MeGlc transport. L6 myotubes were pretreated as
described above, washed and incubated in fresh medium supplemented
with 20.0 mM D-xylose for 6 h, and taken for the [3H]MeGlc or
[3H]dGlc uptake assays. The respective rates of uptake in myotubes
that were not exposed to D-xylose (1.27 ( 0.08 nmol of dGlc (mg of
protein)-1 min-1 and 1.02 ( 0.01 pmol of MeGlc (mg of protein)-1
min-1, respectively) were taken as the 100% values. (C) L6 myotubes
expressing GLUT1myc and GLUT4myc were preconditioned and
treated with 20.0 mM D-xylose for 6 h, as described above. At the end
of incubation the cultures were taken for immunodetection of surface
GLUT1myc or GLUT4myc. *, p < 0.05, in comparison with the
respective control values following incubation at 23.0 mM glucose only.
D-Xylose was not added alone to myotube cultures in glucose-
free medium because this condition mimics the effect of glucose
starvation. Although D-xylose enters the cells, its intracellular
metabolism is different from that of D-glucose and is unable to
sustain the energy balance required for normal cell function.
The findings the D-xylose and D-glucose utilize independent
transport systems (Figure 2B) and that D-xylose augments the
rate of hexose transport in myotubes already exposed to high
D-glucose concentration (Figure 1A and 1B) indicate an
important regulatory role of this pentose on the hexose transport
system in hyperglycemic-like conditions.
Thus, we show for the first time that D-xylose is a potent
stimulator of glucose uptake in myotubes by increasing the
plasma membrane abundance of GLUT4 in a non-insulin-
dependent manner. It is, however, evident that despite these
remarkable qualities, D-xylose presents unfavorable pharmaco-
kinetic and pharmacodynamic characteristics which excludes
its consideration as a potential drug for the treatment of
glucose isomer L-glucose (Figure 1A). Importantly, the effects
of D-xylose and D-arabinose were observed in the absence of
insulin. A thorough washout of D-xylose from cultures at 6 h
resulted in a rapid decline of the uptake rate, reaching the basal
level within 90 min. Half-maximal and maximal stimulatory

8098 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Gruzman et al.
Scheme 1. Synthesis of 7a
a Reagents and conditions: (a) 2-methoxypropene, dry DMF, 5 M HCl
in methanol (catalytic amount), 4 °C, 1.5 h; (b) methyl iodide, Ag2O, DMF,
room temperature, 24 h; (c) acetic acid, 2 M H2SO4 in water, 100 °C, 24 h.
for 1. The hydroxyl group in position 2 was then interacted
with methyl iodide in the presence of Ag2O to produced 1,2-
O-dimethyl-3,5-O-dibenzyl-R-D-xylofuranoside. Compound 2
was obtained by the removal of the benzyl moieties by
hydrogenation and demethylation of the glycosylic position by
acid-catalyzed hydrolysis. The synthesis of 3 was performed
according to Levine and Raymond16 by interacting 1,2-O-
isopropylidene-R-D-xylofuranose in dry pyridine with benzoyl
chloride, which reacts with the primary alcohol in the fifth
position, to produce 5-O-benzoyl-1,2-O-isopropylidene-R-D-
xylofuranose. The hydroxyl group in the third position remained
free to interact with methyl iodide to produce 5-O-benzoyl-3-
O-methyl-1,2-O-isopropylidene-R-D-xylofuranose. Following the
removal of the benzoyl moiety with barium oxide and the
isopropylidene ring by acid hydrolysis, 3 was obtained. The
principles of the procedure of Levine and Raymond16 were also
used for a selective methylation of the fifth position in D-xylose
to produce 4. Briefly, 1,2-O-isopropylidene-R-D-xylofuranose
was interacted with p-toluenesulfonyl chloride in dry pyridine
to obtain 5-O-tosyl-1,2-O-isopropylidene-R-D-xylofuranose. Treat-
ment with sodium methoxide exchanged the tosyl moiety with
a methyl group, and acid hydrolysis removed the isopropylidene
ring to obtain 4.
Scheme 1 depicts the novel synthesis of 4-O-R,ꢀ-methyl-D-
xylopyranose (7). Briefly, 1-O-R-methyl-2,3-isopropylidene-D-
xylopyranoside (5) was synthesized from 1 according to the
synthetic strategy described by Takeo et al.17 The hydroxyl
group in the fourth position of 5 was methylated as described
above for 3. The resulting compound, 1-O-R-methyl-4-O-
methyl-2,3-isopropylidene-D-xylopyranoside (6), was subjected
to an acid-catalyzed hydrolysis to remove the protection groups
and obtained 7.
Compounds 2, 4, and 7, used at concentrations as high as 20
mM for up to 48 h incubations, had no marked effect on hexose
transport in L6 myotubes (Supporting Information Table 2).
Figure 3 shows that 1 and 3 (5 mM each) were more potent
than D-xylose (20 mM) in increasing the rate of hexose uptake
(1.98 ( 0.08- and 2.10 ( 0.16-fold, respectively, in comparison
with the 1.49 ( 0.07-fold increase of D-xylose). Moreover, the
effects of D-xylose, 1, and 3 were additive to the hexose
transport-enhancing effect of insulin.
Figure 2. D-xylose uptake in L6 myotubes. The myotubes were
preconditioned at 2.0 mM glucose for 24 h to increase the rate of hexose
transport to the maximal level. (A) The cultures were washed at the
indicated times and taken for the [14C]xylose uptake assay, as described
in the Experimental Section. (B) Competition assay: similar myotube
cultures were washed and taken for the [3H]dGlc uptake assay.
However, the standard uptake mixture was supplemented with the
indicated concentrations of nonradioactive D-glucose (0), dGlc (2),
or D-xylose (b). The basal rate of dGlc uptake in the standard uptake
mixture (1.47 ( 0.09 nmol (mg of protein)-1 min-1) was taken as 100%.
hyperglycemia: its rate of influx is low, its effective extracellular
concentration is very high (>10 mM), its biological half-life is
very short (30-60 min12), and its hexose transport-augmenting
effect declines shortly after being washed out. Therefore, we
used D-xylose as a prototype molecule for synthesis of long-
acting and more potent derivatives.
QSAR of D-Xylose. A structure-activity relationship analysis
of D-xylose was imperative to plan the synthesis of potent
derivatives. Methyl ether derivatives of xenobiotics are intra-
cellularly stable and not easily cleaved. Therefore, a selective
methylation of each hydroxyl group in the D-xylose molecule
was performed, and the capacity of the methylated derivatives
to modulate the hexose transport of L6 myotubes was tested.
The procedure for the synthesis of 1-O-R-methyl-D-glucopy-
ranoside13 was used to obtain 1-O-R-methyl-D-xylopyranoside
(1) by interacting D-xylose, instead of D-glucose, with methanol
in the presence of Dowex 50. 1,2-O-Isopropylidene-R-D-
xylofuranose, which was the starting material for the synthesis
of 2-O-R,ꢀ-methyl-D-xylopyranose (2), 3-O-R,ꢀ-methyl-D-xy-
lopyranose (3), and 5-O-R,ꢀ-methyl-D-xylofuranose (4), was
synthesized according to Moravcova et al.14 This one-step
reaction between D-xylose and acetone results in an isopropy-
lidene addition to the glycosylic and second hydroxyl positions
in D-xylose, leaving the third and fifth hydroxyl groups available
for chemical manipulations. The synthesis of 2 was carried
according to Bowering15 by interacting 1,2-O-isopropylidene-
R-D-xylofuranose with benzyl bromide in the presence of
potassium hydroxide. The resulting 3,5-O-dibenzyl-1,2-O-
isopropylidene-R-D-xylofuranose was subjected to a HCl(g)-
induced hydrolysis to remove the isopropylidene ring, followed
by methylation of the glycosylic position, as described above
In general, the QSAR analyses indicated that positions 2, 4,
and 5 in D-xylose are critical for its biological activity and that
methylation of positions 1 and 3 not only conserved activity
but rendered the derivatives more efficacious and potent than
D-xylose.
The unexpected finding that 1-O- and 3-O-methylation
augments considerably the hexose transport capacity of the
resulting molecules in comparison with D-xylose could imply
that longer alkyl substitutions at these two positions further
improve their potency and efficacy. However, 1-O-R,ꢀ-ethyl-
D-xylopyranoside (8)18 failed to modulate hexose transport in
myotubes, indicating that elongation of the alkyl group at the

D-Xylose DeriVatiVes Stimulate Muscle Glucose Uptake
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 8099
efficient enzymatic degradation either in the culture medium
or intracellularly, whereas the cell membrane permeability of
others might be impeded.
In contrast to these inactive derivatives, dithioether addition
to the glycosylic position combined with acetal substitutions at
positions 2,4 and/or 3,5 produce potent and long-acting deriva-
tives. It has been shown that thioether bonds in xenobiotics are
oxidized by the cytochrome P450 system to release the core
compound and sulfoxide.21 In addition, Nomura et al. showed
that cyclic acetal substitutions on D-ribose moieties were
effectively reduced under physiological conditions, releasing the
intact pentose.22 The compound 2,4:3,5-dibenzylidene-D-xylose
diethyl dithioacetal (19) was synthesized according to Curtis
and Jones23 (Scheme 2). Briefly, D-xylose was converted to
D-xylose diethyl dithioacetal (20) by reaction with ethanediol
in the presence of concentrated HCl. Benzaldehyde was
interacted with 20 in the presence of HCl(g), and the resulting
19 was then crystallized from ethanol. 2,4-Benzylidene-D-xylose
diethyl dithioacetal (22) was prepared by a selective acid
hydrolysis of the 3,5-acetal ring in 19 as described24 (Scheme
2). Briefly, 19 was dissolved in an ice-cold mixture of
chloroform:methanol:water (42:56:2) and a catalytic volume of
concentrated HCl. Following stirring at 4 °C for 3 days, three
products were obtained: 20, 21, and 3,5-benzylidene-D-xylose
diethyl dithioacetal. Compound 21 was purified by PLC (using
diethyl ether as the eluent), extracted with chloroform, and
crystallized with petroleum ether:diethyl ether (40:60).
Figure 3. Effects of D-xylose, 1, 3, 16, and 17 on hexose uptake in
L6 myotubes. Myotube cultures were washed and received fresh RMEM
containing 23.0 mM D-glucose and 20.0 mM D-xylose or 5.0 mM 1, 3,
16, or 17 and incubated for 7 h. Insulin effect was measured as described
in the Experimental Section. The basal rate of dGlc uptake (1.32 (
0.05 nmol (mg of protein)-1 min-1) was taken as 100%. *, p < 0.05 in
comparison with the respective 23.0 mM glucose control.
glycosylic position was disadvantageous. Ethyl (16), propyl (17),
and n-butyl (18) substitutions at position 3 in D-xylose were
performed as described in Scheme 2: 1,2-O-R-isopropylidene-
D-xylofuranose was interacted with trityl chloride to produce
5-O-trityl-1,2-O-R-isopropylidene-D-xylofuranose (9).19 Fol-
lowing its alkylation with ethyl iodide, propyl iodide, or n-butyl
iodide in the presence of Ag2O and removal of the trityl and
isopropylidene protective groups by acid catalysis, the corre-
sponding 3-O-alkyl-D-xylopyranoses (16, 17, and 18) were
obtained.
Figure 3 shows that 16 and 17 increased the rate of hexose
transport in myotubes 2.79 ( 0.11- and 3.89 ( 0.27-fold above
the basal uptake level, whereas 18 was inactive (Supporting
Information Table 2). Similar to D-xylose, 1, and 3, the effects
of 16 and 17 were additive to the effect of insulin. These data
indicate that a 3C alkyl elongation at position 3 of D-xylose
was optimal. Nevertheless, although the efficacy of 16 and 17
was improved in comparison with 1 and 3, their potency
remained modest (2.0 and 5.0 mM for half-maximal and
maximal effects, respectively).
Since the addition of an alkyl group in position 3 of D-xylose
improved both the efficacy and potency of the molecule, we
also synthesized 3-O-alkyl derivatives of 21. Scheme 2 also
shows the synthesis of the 3-O-methyl- and 3-O-ethyl derivatives
(24 and 25). Dithioacetal derivatives at the glycosylic position
of 3 and 16 were prepared by coupling with ethanethiol
employing an acidic catalysis to produce the corresponding 22
and 23. Selective 2,4-benzaldehyde acetal ring addition to these
compounds, using benzaldehyde dimethyl acetal under dry
conditions and gentle acid catalysis, produced 24 and 25. The
chirality of carbon atoms in the D-xylose backbone in 19, 21,
and 24 is maintained. In 19 two additional chiral centers were
introduced in the 2,4- and 3,5-benzylidene rings. According to
Rao and Grindley25 the conformations of the benzylidene carbon
atom in 19 are 2,4(R) and 3,5(S) while in 21 it is 2,4(R). The
synthetic procedures of 24 and 25 produced a mixture of S/R
stereoisomers in position 2,4 (46:54 and 39:61, respectively).
Compounds 19, 21, and 24 increased the rate of hexose
transport in L6 myotubes in a time-dependent manner; 21 and
24 peaked at 7 h while 19 peaked at 18 h, after which a gradual
decline was observed (Figure 4A). The concentrations used in
this experiment were based on the dose-response analysis
(Figure 4B), in which maximal stimulatory effects of 19, 21,
and 24 were observed at 5, 150, and 50 µM, respectively.
The methyl ether substitution at position 3 in 24 augmented
its potency 3-fold compared to that of 21. Interestingly, an ethyl
ether substitution at this position rendered 25 inactive (Sup-
porting Information Table 2). Of major interest is the finding
that the potency of 19 was 30- and 10-fold higher than those of
21 and 24, respectively. The high lipophilicity of 19 (calcd log
P ) 4.99)26 correlates well with this enhanced potency, in
contrast to the significantly lower log P values of 21, 24, and
obviously D-xylose (2.40, 2.80, and -0.83, respectively). It is
important to note that we used 10% rather than the usual 2%
(v/v) FCS in the medium culture of mature myotubes to improve
the solubility of 19. We measured the protein binding capacity
of 21 in this high-serum culture medium, as described in the
Effects of Lipophilic Derivatives of D-Xylose. A common
method to improve the pharmacokinetic parameters of hydro-
philic molecules, such as D-xylose, 16, or 17, is to synthesize
active lipophilic derivatives that readily diffuse into cells through
the lipid phase of the plasma membrane and act intracellularly,
thus yielding long-acting and more potent and efficacious
derivatives than the parent compound.20 Often, such derivatives
are prodrugs, which are activated in vivo by spontaneous or
enzymatic dissociation of the lipophilic substituents. On the basis
of the QSAR data we planned D-xylose glycoside derivatives
and derivatives in which positions 2, 4, and 5 were maintained
intact or bound to dissociable groups. In line with this strategy,
we synthesized and screened various glycoside-, ester-, phos-
phate ester-, ketal-, and ether bond-based derivatives of these
positions in D-xylose. The synthetic procedures, effects in the
hexose uptake assay, and cytotoxicity of these compounds are
presented in the section on inactive D-xylose derivative in the
Supporting Information. Most of these compounds had no effect
on the hexose transport capacity in the myotubes. We hypoth-
esize that some of these derivatives are subjected to rapid and

8100 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Gruzman et al.
Scheme 2. Synthesis of 3-Alkyl and Benzylidene Derivatives of D-Xylosea
a Reagents and conditions: (a) respective alkyl iodide, Ag2O, DMF, room temperature, 24 h; (b) TFA, chloroform; (c) 0.1% H2SO4, water, 80 °C, 2 h;
(d) ethanethiol, cold HCl, overnight, room temperature; (e) benzaldehyde dimethyl acetal, molecular sieve, 3 Å, dry DMF, dry HCl in dioxane, room
temperature, overnight; (g) benzaldehyde, dry HCl gas, 20 min, ice bath; (f) chloroform/methanol/water (15:20:0.75), HCl, 4 °C, 3 days.
Experimental Section, and found it to be 98.0 ( 0.3%. Thus,
when the total (added) concentration of 19 in the culture medium
was 5 µM, its calculated free (non-protein-bound) concentration
was nearly 100 nM. Of importance are the findings that the
viability of myotubes was not compromised on the presence of
19, 21, 24, or D-xylose under the incubation conditions described
above (Supporting Information Figure 2).
Another aspect of the higher lipophilicity of 19 in comparison
to 21 and 24 is depicted in Figure 4C: L6 myotubes were treated
with optimal concentrations of these derivatives and of D-xylose.
At the times of respective peak effects, the cultures were washed
thoroughly, received fresh medium with 23.0 mM D-glucose,
and the rate of hexose transport was determined thereafter. These
data were used to calculate the half-life of the effect of the test
compounds after washing. A positive correlation between log
P values and half-life values was revealed (inset to Figure 4C).
The emerging rank order, D-xylose < 21 < 24 < 19, supports
the view that the extended duration of action of 19 results from
its higher lipid solubility in cells. Worth mentioning is the
observation that 19, 21, and 24 were active in the absence of
insulin and that their effects were additive to that of the hormone
when incubated together (Figure 4D), suggesting that the
mechanisms of action of these synthetic compounds and insulin
are distinct.
greatly. While insulin acts within minutes and its effect
disappears shortly after its removal, the onset and off rates of
the effects of D-xylose, and especially of its lipophilic deriva-
tives, are long and positively correlated with the lipophilicity
of the compounds. Nevertheless, the end point effects of insulin,
D-xylose, and its lipophilic derivatives on the muscle-specific
glucose transporter GLUT4 are similar: Panels A and B of
Figure 5 show that insulin, D-xylose, and the derivatives
increased the abundance of GLUT4myc, but not of GLUT1myc,
in the plasma membrane of L6 myotubes. However, the effect
of insulin on GLUT4myc abundance in the plasma membrane
of myotubes was fast (30 min) whereas several hours were
required for D-xylose, 19, 21, and 24 to exert their maximal
effects (7 h of incubation for all except 18 h for 19). These
treatments did not alter the total cell content of GLUT1 or
GLUT4, as was determined in whole call lysates of myc epitope-
tagged L6 myotubes (Figure 5) or wild-type L6 myotubes
(Supporting Information Figure 3).
The experimental conditions of the standard hexose uptake
and the colorimetric determination of surface GLUT1myc and
GLUT4myc differ greatly: the former is short and carried at
room temperature, while the latter is longer and entails extensive
washes in a buffer containing goat serum at 4 °C. Therefore, to
correlate plasma membrane GLUT1myc and GLUT4myc data
to hexose transport levels, we also determined the rate of
[3H]dGlcuptakeinwild-typeandinGLUT4myc-andGLUT1myc-
tagged L6 myotubes under the experimental conditions em-
ployed in the colorimetric assay (except for the fixation stage;
the uptake period at 4 °C was 15 min). The relative effects of
low and high glucose concentrations, insulin, D-xylose, and 19,
21, and 24 were similar to those observed under conditions of
the standard uptake assay (Supporting Information Figure 4).
Thus, the changes found in plasma membrane abundance of
To be of relevance for diabetes treatment, these compounds
should be effective also in nontransformed human muscle cells.
This was indeed the case (Figure 4E): primary cultures of human
myotubes exposed to D-xylose, 19, 21, or 24 increased signifi-
cantly the rate of hexose transport (1.43 ( 0.15-, 1.59 ( 0.18-,
1.49 ( 0.12-, and 1.40 ( 0.17-fold, respectively, over the
control level).
Mechanism of Action of D-Xylose, 19, 21, and 24. The
temporal characteristics of the hexose transport-augmenting
effects of insulin and of D-xylose and its derivatives differ

D-Xylose DeriVatiVes Stimulate Muscle Glucose Uptake
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 8101
Figure 4. Compounds 19, 21, and 24 increase the rate of hexose uptake in L6 myotubes and human skeletal muscle myotubes. (A) Time-course
analysis. Myotube cultures were washed and received fresh medium supplemented with 10% (v/v) FCS, 23.0 mM D-glucose, and 5 µM 19 (b), 150
µM 21 (9) or 50 µM 24 (O). Control myotubes received the vehicle (V) only (0). After incubation for the indicated time periods the cultures were
taken for the dGlc uptake assay. The basal rate of uptake at zero time (1.17 ( 0.09 nmol of dGlc (mg of protein)-1 min-1) was taken as 100%. (B)
Dose-response analysis. Myotube cultures were incubated with medium supplemented with 10% (v/v) FCS, 23.0 mM D-glucose, and increasing
concentrations of 19 (b, 7 h), 21 (9, 18 h), or 24 (O, 7 h). Control myotubes received the vehicle only (0) and were incubated for 18 h. The
cultures were then washed and taken for the standard [3H]dGlc uptake assay. The basal rate of dGlc uptake at zero time (1.58 ( 0.21 nmol (mg
of protein)-1 min-1) was taken as 100%. (C) Myotube cultures were exposed to D-xylose, 19, 21, and 24 as described for (A). The cultures were
then washed at the respective peak times depicted in (B) and received fresh medium supplemented with 23.0 mM D-glucose. The rate of dGlc
uptake was assayed at the indicated times. Inset: Correlation between log P values of D-xylose, 19, 21, and 24 and their half-maximal hexose uptake
augmenting capacity. (D) Combined effects of 19, 21, and 24 and of insulin. The cultures were treated as described above and incubated for 24 h.
The basal rate of dGlc uptake at zero time (1.13 ( 0.06 nmol (mg of protein)-1 min-1) was taken as 100%. (E) Human skeletal muscle myotubes
were prepared as described in the Experimental Section. The cultures were treated with 19, 21, and 24 as described above for (C) and taken for the
standard [3H]dGlc uptake assay. *, p < 0.05, in comparison with the respective 23.0 mM glucose control.
GLUT4myc reflected the changes in the hexose transport
capacity of myotubes.
alin-7-yl)phenyl)methyl)-4-piperidinyl)-2H-benzimidazol-2-
one trifluoroacetate salt hydrate).30 Panels A and B of Figure 6
show the expected abolishment of insulin-stimulated hexose
uptake in L6 myotubes by inhibiting PI3-kinase and Akt/PKB.
In contrast, the hexose transport-stimulatory effects of D-xylose,
19, 21, and 24 remained intact in the presence of these inhibitors,
indicating that neither PI3-kinase nor Akt/PKB participates in
this action. Further support for this conclusion came from the
determination of the extent of Ser473 and Thr308 phosphorylation
in Akt/PKB: Figure 6C depicts the predictable insulin-induced
phosphorylation of both moieties in Akt/PKB.27 D-xylose, 19,
21, and 24 had no comparable phosphorylating capacity, except
Since D-xylose and its derivatives increased the plasma
membrane GLUT4 content like insulin, it was questioned
whether they activated the insulin signaling pathways, including
PI3-kinase and its downstream effector protein kinase B (Akt/
PKB).27 Inhibition of PI3-kinase or Akt/PKB silences the insulin
transduction pathway;28 we therefore postulated that if D-xylose
derivatives act through this pathway, their effects on glucose
transport would be abolished by wortmannin that inhibits PI3-
kinase29 and by the commercially available inhibitor of Akt/
PKB (1,3-dihydro-1-(1-((4-(6-phenyl-1H-imidazo[4,5-g]quinox-

8102 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Gruzman et al.
Figure 5. D-Xylose, 19, 21, and 24 increase the abundance of GLUT4, but not GLUT1, in the plasma membrane of L6 myotubes. Myotubes
expressing GLUT1myc and GLUT4myc were preconditioned with the indicated glucose levels and then received the vehicle (V) or 200 nM insulin,
20 mM D-xylose, 5 µM 19, 150 µM 21, or 50 µM 24, as described for Figure 4B. At the end of incubation, the cultures were taken for immunodetection
of surface GLUT1myc (A) or GLUT4myc. (B), as described in the Experimental Section. The Western blots depict the total cell content of the
corresponding glucose transporters in whole cell lysates. *, p < 0.05, in comparison with the control (V).
Figure 6. D-Xylose and 19, 21, and 24 augment the rate of hexose transport in L6 myotubes in a PI3K- and Akt/PKB-independent manner.
Myotube cultures were washed and received fresh medium containing the same D-glucose concentration and 20 mM D-xylose, 5 µM 19, 150 µM
21, 50 µM 24, or the vehicle (V) in the absence or presence of 100 nM wortmannin (A) or 100 nM Akt/PKB inhibitor (B), as described under
Figure 4B. Insulin effect was measured as described in the Experimental Section. Wortmannin or Akt/PKB inhibitor was added 30 min before the
addition of test compounds or insulin. The basal rate of dGlc uptake (1.02 ( 0.09 nmol (mg of protein)-1 min-1) in myotubes exposed to 23 mM
D-glucose (V) was taken as 100%. *, p < 0.05, in comparison with the respective controls. (C) Whole cell lysates were prepared from myotube
cultures that were treated with D-xylose, 19, 21, and 24 as described above. Western blot analyses were performed with antibodies against PKB,
pThr308-PKB, and pSer473-PKB, as described in the Experimental Section and Supporting Information Table 3.
for a very weak phosphorylation of Thr308. We doubt that the
latter plays a role in mediating the effect of D-xylose derivatives
since, as suggested by Bouzakri et al.,31 the phosphorylation of
Ser473, but not of Thr308, is imperative for insulin-induced
stimulation of glucose uptake in skeletal muscles. Therefore,
our findings strongly imply that D-xylose, 19, 21, and 24 do
not recruit Akt/PKB for transducing their hexose transport-
stimulatory effects.
The translocation of GLUT4 to the plasma membrane of
skeletal muscle is also mediated by non-insulin-dependent
stimuli, such as muscle contraction and reduced energy charge
that activate 5-AMP-activated kinase (AMPKR)5. Figure 7A

D-Xylose DeriVatiVes Stimulate Muscle Glucose Uptake
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 8103
Figure 7. D-Xylose, 19, 21, and 24 activate AMPKR and AS160. (A) Myotube cultures were treated with D-xylose (D-xyl), 19, 21, 24, insulin
(Ins), or the vehicle (V) as described in the legend to Figure 4A. AICAR (4 mM) and D-sorbitol (S, 0.25 M) were present for 1 h and 30 min,
respectively. Whole cell lysates were prepared, and Western blot analyses were performed with antibodies against AMPKR and pTyr172-AMPKR.
(B) Time-dependent phosphorylation of Tyr172-AMPKR. The experiment was performed as described in (A). The incubation was terminated at the
indicated times, and the myotubes were lysed and taken for Western blot analyses of AMPKR and pTyr172-AMPKR. (C) Human myotubes were
treated as described above for (A) and taken for Western blot analysis of AMPKR and pTyr172-AMPKR. (D) The cultures were preincubated for
1 h with 5 µM Compound C prior to the addition of D-xylose (D-xyl), 19, 21, or 24. The basal rate of dGlc uptake (1.45 ( 0.09 nmol (mg of
protein)-1 min-1) in myotubes exposed to the vehicle was taken as 100%. (E) Whole cell content of AS160 and pThr642-AS160 was determined by
Western blot analysis in samples that were prepared as described above for (A). Representative blot and a summary of n ) 3. *, p < 0.05, in
comparison with the respective controls.
shows that the classical AMPKR activators AICAR32 and a
hyperosmolar shock (0.25 M D-sorbitol),33 but not insulin,
induced significant Tyr172 phosphorylation in AMPKR in L6
myotubes. D-Xylose, 19, 21, and 24 significantly induced Tyr172
phosphorylation (3.23 ( 0.57-, 4.89 ( 0.77-, 5.25 ( 0.49-, and
3.98 ( 0.38-fold above control, respectively), whereas total cell
content of AMPKR was not altered in their presence. These
effects of D-xylose and its derivatives on AMPKR phosphory-
lation were time-dependent (Figure 7B): Peak effects were
obtained at 1-2 h for D-xylose, 21, and 24 and at 12 h for 19,
demonstrating that the activation of AMPKR precedes the
maximal stimulation of hexose transport. Again, a similar
positive correlation between the onset of the stimulatory effects
and log P values of these compounds is apparent (Figure 4C).
D-Xylose-, 19-, 21-, and 24-induced phosphorylation of Tyr172
in AMPKR was also evident in primary cultures of human
skeletal muscles (Figure 7C).
We also used Compound C, a selective inhibitor of
AMPKR,34 to evaluate directly the role of AMPKR in mediating
the action of D-xylose, 19, 21, and 24. Compound C (5 µM)
significantly impeded the ability of D-xylose and its derivatives
to augment the rate of hexose transport in L6 myotubes (Figure

8104 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Gruzman et al.
Table 1. ATP Content Is Not Altered in L6 Myotubes Following
Treatment with D-Xylose, 19, 21, or 24a
D-xylose, D-ribose, L-glucose, 1-O-methyl-ꢀ-D-xylopyranoside,
AICAR (5-aminoimidazole-4-carboxamide-1-ꢀ-ribofuranosyl 5-ꢀ-
monophosphate), Akt/PKB (Akt1/2 kinase) inhibitor, BSA (bovine
serum albumin, fraction V), chloroform-d, D2O, dimethyl sulfoxide-
d6, Igepal CA-630, O-phenylenediamine (OPD), and wortmannin
were purchased from Sigma Chemical Co (St. Louis, MO).
Anhydrous acetonitrile and dry DMF were also from the same
supplier and used as received. D-Sorbitol was from BDH (Poole,
U.K.). Compound C was from Calbiochem (Darmstadt, Germany).
American Radiolabeled Chemicals (St. Louis, MO) supplied 2-[1,2-
3H(N)]deoxy-D-glucose (2.22 TBq/mmol), 3-O-[methyl-3H]-D-
glucose (2.22 TBq/mmol), [14C-(U)]sucrose (22.2 GBq/mmol), and
[1-14C]-D-xylose (2.03 GBq/mmol). The following antibodies were
used: anti-AMPKR and anti-pThr172-AMPKR from Cell Signaling
Technology (Beverly, MA); anti Akt/PKBR (PH domain) and anti
pSer473-Akt/PKBR from Upstate Biotechnology (Lake Placid, NY);
anti pThr308-Akt1/PKB and anti-C-Myc (A-14) from Santa Cruz
condition
ATP (mmol/mg of protein)
5.0 mM D-glucose (L)
23.0 mM D-glucose (H)
H + insulin (100 nM)
H + DMSO (1.0%)
H + D-xylose (20 mM)
H + 19 (5 µM)
1.84 ( 0.09
1.94 ( 0.08
1.66 (0.10
1.90 ( 0.06
1.60 ( 0.03
1.99 ( 0.12
1.73 ( 0.08
1.69 ( 0.05
0.72 ( 0.08b
H + 21 (150 µM)
H + 24 (50 µM)
H + DNP (5 mM, 10 min)
a L6 myotubes were preconditioned with the indicated glucose concentra-
tion and treated with the various compounds as described in the legend to
Figure 5. At the end of incubation the myotubes were lysed, and the ATP
content was determined as described in the Experimental Section. b p <
0.05, in comparison with the 23.0 mM D-glucose (H) incubation. L, low
(5.0 mM D-glucose); H, high (23.0 mM D-glucose).
Biotechnology (Santa Cruz, CA); and anti AS160 and anti pT642
-
7D): 68.9 ( 10.7%, 48.8 ( 3.4%, 60.0 ( 8.1%, and 69.2 (
8.6% reduction, respectively (p < 0.05), in comparison with
the respective control cells. Table 1 shows that D-xylose, 19,
21, and 24 activate AMPKR in an adenosine-independent
manner because total cell ATP content in L6 myotubes was
not altered significantly in their presence. The uncoupler DNP
(5 mM, 10 min), which reduced significantly the ATP content,
served as a positive control.
The protein AS160 mediates insulin-stimulated GLUT4
translocation in myotubes following the activation of Akt/PKB.
Equally important, AS160 is also activated under conditions of
AMPKR stimulation.35 We therefore determined the extent of
Thr642 phosphorylation in AS160, a marker of its activation, in
L6 myotubes treated with insulin and with D-xylose, 21, 22,
and 25. Figure 7E shows that, in concert with the stimulation
of the rate of hexose transport and AMPKR phosphorylation,
these treatments elicited significant Thr642 phosphorylation in
AS160.
In summary, this study shows that high levels of D-xylose
increase the rate of hexose transport in L6 myotubes by
increasing the plasma membrane content of GLUT4 in an
AMPKR-dependent manner. Interestingly, AMPKR is consid-
ered a key target for the development of new pharmaceuticals
to treat obesity, type 2 diabetes, and the metabolic syndrome.5
For instance, recent studies have linked the blood glucose
lowering effects of the antihyperglycemic drug metformin (1,1-
dimethylbiguanide) to the activation of AMPKR.36,37 Lipophilic
derivatives of D-xylose were synthesized to overcome the poor
pharmacokinetic and pharmacodynamic parameters of D-xylose.
The highly lipophilic 19, 21, and 24 exerted significant hexose
transport-stimulatory effects at low concentrations in rat and
human cultured myotubes in an AMPKR-dependent manner. It
is assumed that these compounds act as prodrugs and release
the active parent compound upon intracellular oxidation of the
thioether group and the reduction of the acetal substitutions.
These findings emphasize the great potential of this and
similar compounds in the development of novel potent and long-
acting antihyperglycemic compounds. Thus, future syntheses
and criteria for development of such D-xylose derivatives need
to be based on selection of suitable dissociable lipophilic
moieties to be coupled to the core molecule, the lipophilicity
and membrane permeability of the derivatives, the intracellular
release of the parent molecule, and its potential to activate
AMPKR to increases the cell surface abundance of GLUT4 in
skeletal muscles.
AS160 from Acris antibodies (Hiddenhausen, Germany). Horserad-
ish peroxidase-conjugated anti-rabbit IgG was from Jackson
ImmunoResearch (West Grove, PA). R-Minimal essential medium
(RMEM), EZ-ECL chemoluminescence detection kit for HRP, goat
serum, fetal calf serum (FCS), L-glutamine, and antibiotics were
from Biological Industries (Beth-Haemek, Israel). Preparative silica
gel glass plates were from Yoel Naim Ldt. (Rehovot, Israel). Silica
gel 60 F254 TLC plates were purchased from Merck (Darmstadt,
Germany) Organic solvents (HPLC grade) were purchased from
Frutarom Ltd. (Haifa, Israel) and Mallinckrodt Baker B.V. (De-
venter, Holland). The boiling range of petroleum ether was 35-60
°C.
Chemical Procedures. Dry pyridine was prepared via reflux with
potassium hydroxide for 5 h followed by distillation. Dry methanol
was prepared by reflux (8 h) with magnesium and small amount of
iodine crystals followed by distillation. Dry dichloromethane was
prepared via reflux with P2O5 for 12 h followed by distillation.
Melting points were determined using a Melting Point SMP
apparatus (Stuart Scientific, Stone, Staffordshire, U.K.). Elemental
analysis was performed in the Microanalysis Laboratory of the
Institute of Chemistry, Faculty of Life Sciences, The Hebrew
University (Jerusalem, Israel). Proton and carbon NMR spectra of
compounds were obtained with a Varian VXR-300 (300 MHz)
spectrometer equipped with a 5 mm probe, and data were processed
using VNMR software. Chloroform-d, D2O, and dimethyl sulfoxide-
d6 were used as solvents, using tetramethylsilane (δ ) 0.00 ppm)
as an internal standard. The splitting pattern abbreviations are as
follows: s, singlet; d, doublet; t, triplet; q, quartet; m, unresolved
multiplet due to the field strength of the instrument; ds, doublet of
singlet; a, axial; e, equatorial position. Electrospray ionization mass
spectrometry was measured on a Thermo Quest Finnigan LCQ-
Duo (Fisher Scientific, Waltham, MD) in the positive ion mode.
In most cases, elution was performed in a 20:79:1 water:methanol:
acetic acid mixture at a flow rate of 15 µL/min. Data were processed
using ThermoQuest Finnigan’s Xcalibur Biomass Calculation and
Deconvolution software. For purification we employed flash column
chromatography on Merck silica gel 60 (particle size 230-400
mesh) using dichloromethane and methanol, at various ratios, as
eluent. Purification on HPLC (Gilson, Middleton, WI) was per-
formed on C18 preparative and semipreparative columns (Beckman
Ultrasphere ODS column, 250 mm × 10 mm i.d., 5 µM, and
Beckman Ultrasphere ODS column, 250 mm × 4.6 mm i.d., 5 µM)
using acetonitrile and doubly distilled water in different ratios as
the eluent solvent. For analytical TLC silica gel 60 F254 precoated
plates (Merck, Darmstadt, Germany) were used, and compounds
were visualized with UV light or I2 vapor.
1-O-Methyl-2,3-O-isopropylidene-r-D-xylopyranoside (5). This
compound was synthesized according to the general synthetic
procedure described by Takeo et al.17 with minor modifications:
2-Methoxypropene (6.71 mL) was added dropwise during 15 min
to an ice-cold stirred suspension of 1 (5 g) in dry DMF (16 mL)
and 5 M HCl in methanol (0.1 mL). The reaction mixture cleared
within 10 min and was further stirred for 1 h at room temperature.
Experimental Section
Materials. D-Arabinone, 2,4-dinitrophenol (DNP), 2-deoxy-D-
glucose, D-glucose, D-lyxose, 3-O-methyl-D-glucose, 3-O-methyl-

D-Xylose DeriVatiVes Stimulate Muscle Glucose Uptake
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 8105
Then, 150 mL of chloroform was added, and the solution was
extracted three times with 100 mL of water. The resulting organic
layer was dried with anhydrous sodium sulfate, filtered, and
evaporated to obtain a light brown syrup that was then purified by
silica gel chromatography (hexane:ethyl acetate 60:40). After
evaporating the organic solvents colorless syrup was obtained. Anal.
Calcd for C8H16O5: C, 52.93; H, 7.90. Found: C, 51.08; H, 7.37.
1H NMR (CDCl3) δ, ppm, 1.54 (ds, 6H, [CH3-C-CH3]), 3.30 (s,
3H, [O-C-H3]), 3.37 (t, 1H, [C-H2]), 3.41 (d, 1H, [C-aH5]), 3.47
(m, 1H, [C-H3]), 3.83 (m, 1H, [C-H4]), 4.01 (d, 1H, [C-eH5]),
4.84 (d, 1H, [C-H1]).
respectively, instead of iodoethane. Analytical data: 11, yield )
81%; mp 128-132 °C. Anal. Calcd for C30H34O5: C, 75.81; H,
7.21. Found: C, 75.92; H, 7.22. ESMS [MW + (Na+)]: m/e 497.07,
calcd m/e 497.24. 1H NMR (CDCl3) δ, ppm, 1.05 (t, 3H, [O-CH2-
CH2-CH3]), 1.28 (q, 2H, [O-CH2-CH2-CH3]), 1.51-1.31 (ds, 6H,
[CH3-C-CH3]), 3.88 (q, 2H, [O-CH2-CH2-CH3]), 4.21 (q, 1H,
[C-H3]), 4.36 (d, 2H, [C-H25]), 4.57 (q, H, [C-H2]), 4.72 (q, 1H,
[C-H4]), 5.91 (d, 1H, [C-H1]), 7.10-7.29 (m, 15H, [trityl]). 12,
yield ) 94%. Anal. Calcd for C31H36O5: C, 75.69; H, 7.13. Found:
C, 76.20; H, 7.43. ESMS [MW + (Na+)]: m/e 511.31, calcd m/e
511.26. 1H NMR (CDCl3) δ, ppm, 1.05 (t, 3H, [O-CH2-CH2-CH2-
CH3]), 1.22-1.28 (m, 4H, [O-CH2-CH2-CH2-CH3]), 1.51-1.31 (ds,
6H, [CH3-C-CH3]), 3.88 (q, 2H, [O-CH2-CH2-CH2-CH3]), 4.21 (q,
1H, [C-H3]), 4.36 (d, 2H, [C-H25]), 4.57 (q, 1H, [C-H2]), 4.72 (q,
1H, [C-H4]), 5.91 (d, 1H, [C-H1]), 7.10-7.29 (m, 15 H, [trityl]).
3-O-Ethyl-1,2-isopropylidene-r-xylopyranoside (13), 3-O-Pro-
pyl-1,2-isopropylidene-r-xylopyranoside (14), and 3-O-Butyl-
1,2-isopropylidene-r-xylopyranoside (15). These compounds were
synthesized by de-O-tritylization19 of 10, 11, and 12, respectively.
Analytical data: 13, yield ) 21.3%. ESMS [MW + (Na+)]: m/e
1,4-O-Dimethyl-2,3-isopropylidene-r-xylopyranoside (6). To
a solution of 5 g of 5 in 50 mL of DMF and 4.77 mL of methyl
iodide was added gradually 16.6 g of silver oxide (I) during 2 h
and then left in the dark at room temperature for 48 h. Then 200
mL of chloroform was added, and the suspension was filtered. The
filtrate was washed with 50 mL of chloroform and extracted three
times with 100 mL of water. Following drying with anhydrous
sodium sulfate, filtering, and evaporating, light brown syrup was
obtained. It was purified by silica gel chromatography (ethyl acetate:
petroleum ether 20:80) and monitored by TLC (ethanol:diethyl ether
30:70). Colorless syrup was received with a yield of 34.7%.
Crystallization attempts were unsuccessful, and the compound was
further used as syrup. Anal. Calcd for C10H18O5: C, 55.03; H, 8.31.
Found: C, 55.19; H, 7.97. ESMS [MW + (Na+)] (C10H18O5): m/e
241.19, calcd m/e 241.12. 1H NMR (CDCl3) δ, ppm, 1.53 (ds, 6H,
[CH3-C-CH3]), 3.34 (t, 1H, [C-H2]), 3.42 (s, 3H, [O-CH3]), 3,49
(s, 3H, [O-CH3]), 3.51 (d, 1H, [C-aH5]), 3.54 (m, 1H, [C-H3]),
3.83 (m, 1H, [C-H4]), 4.1 (d, 1H, [C-eH5]), 4.9 (d, 1H, [C-H1]).
4-O-Methyl-r,ꢀ-xylopyranose (7). Compound 6 (1.7 g) was
dissolved in 50 mL of hot glacial acetic acid, diluted with 10 mL
of boiling 2 M sulfuric acid, and left for 2 h on a steam bath. The
mixture then received 10 mL of boiling 2 M sulfuric acid and was
similarly heated for an additional 24 h. After the mixture was cooled
to room temperature it was mixed with 100 mL of water and
neutralized with sodium carbonate. After filtration and lyophilization
the white crude mass was extracted with methanol and filtered, and
the methanol was evaporated. A remaining yellow syrup was further
purified by silica gel short column chromatography (methanol),
while monitoring on TLC (methanol:water 90:10). Attempts to
crystallize the syrup were unsuccessful. Yield ) 77.1%. Anal. Calcd
for C6H12O5: C, 43.90; H, 7.37. Found: C, 43.39; H, 7.21. 1H NMR
(D2O) δ, ppm, 3.31 (t, 1H, [C-H2]), 3.47 (s, 3H, [O-CH3]), 3.51
(d, 1H, [C-aH5]), 3.57 (m, 1H, [C-H3]), 3.88 (m, 1H, [C-H4]),
4.12 (d, 1H, [C-eH5]), (R) 5.29, (ꢀ) 4.71 (d, 1H, [C-H]). 13C NMR
(D2O) δ, ppm, 52.9 (C4-O-CH3), 64.9 (C5), 70.1 (C3), 76.1 (C4),
77.5 (C2), 93.4 (C1).
1
241.16, calcd m/e 241.12. H NMR (CDCl3) δ, ppm, 1.05 (t, 3H,
[O-CH2-CH3]), 1.51-1.31 (ds, 6H, [CH3-C-CH3]), 3.28 (t, 2H,
[O-CH2-CH3]), 3.64 (d, 2H, [C-H25]), 4.21 (q, 1H, [C-H3]), 4.47
(q, 1H, [C-H2]), 4.62 (q, 1H, [C-H4]), 5.91 (d, 1H, [C-H1]). 14,
yield ) 74.2%. ESMS [MW + (Na+)]: m/e 254.96, calcd m/e
255.13. 1H NMR (CDCl3) δ, ppm, 1.05 (t, 3H, [O-CH2-CH2-CH3]),
1.27 (m, 2H, [O-CH2-CH2-CH3]), 1.51-1.31 (ds, 6H, [CH3-C-
CH3]), 3.28 (t, 2H, [O-CH2-CH2-CH3]), 3.64 (d, 2H, [C-H25]), 4.21
(q, 1H, [C-H3]), 4.47 (q, 1H, [C-H2]), 4.62 (q, 1H, [C-H4]), 5.91
(d, 1H, [C-H1]). 15, yield ) 34.6%. ESMS [MW + (Na+)]: m/e
1
268.87, calcd m/e 269.15. H NMR (CDCl3) δ, ppm, 1.05 (t, 3H,
[O-CH2-CH2-CH2-CH3]), 1.23-1.28 (m, 4H, [O-CH2-CH2-CH2-
CH3]), 1.51-1.31 (ds, 6H, [CH3-C-CH3]), 3.28 (t, 2H, [O-CH2-
CH2-CH2-CH3]), 3.64 (d, 2H, [C-H25]), 4.21 (q, 1H, [C-H3]), 4.47
(q, 1H, [C-H2]), 4.62 (q, 1H, [C-H4]), 5.91 (d, 1H, [C-H1]).
3-O-Ethyl-r,ꢀ-xylopyranose (16), 3-O-Propyl-r,ꢀ-xylopyra-
nose (17), and 3-O-Butyl-r,ꢀ-xylopyranose (18). These com-
pounds were synthesized from 13, 14, and 15, respectively, as
described for 6 using ethyl iodide, propyl iodide, and butyl iodide,
respectively. The syrups that were obtained were purified by
preparative HPLC, wavelength 220 nm. Analytical data: 16, yield
) 23%; mp 80-84 °C. Anal. Calcd for C7H14O5: C, 47.18; H, 7.92.
1
Found: C, 46.99; H, 7.77. H NMR (D2O) δ, ppm, 1.04 (t, 3H,
[O-CH2-CH3]), 3.25 (d, 2 H, [C-eH5]), 3.29 (t, 2H, [O-CH2-CH3]),
3.41 (q, 1H, [C-H3]), 3.56 (d, 2H, [C-aH5]), 3.71 (m, 1H, [C-H4]),
3.81 (m, 1H, [C-H2]), (R) 5.71, (ꢀ) 4.65 (d, 1H, [C-H]). 13C NMR
(D2O) δ, ppm, 14.84 (CH2CH3), 61.17 (C5), 63.99 (CH2CH3), 68.85
(C4), 71.04 (C3), 73.72 (C2), 92.07 (HRC), 98.15 (HꢀC). 17, yield
) 29%; mp 78-82 °C. Anal. Calcd for C8H16O5: C, 49.19; H, 8.72.
5-O-Trityl-3-O-ethyl-1,2-isopropylidene-r-xylopyranoside
(10). To an ice-cold solution of 8.3 g of 5-O-trityl-1,2-isopropy-
lidene-R-xylopyranoside (9)19 and 2.5 mL of iodoethane in DMF
(50 mL) was added 0.45 g of sodium hydride in three portions.
After stirring for 15 min on ice the reaction was left stirring at
room temperature for an additional 2 h and then quenched with 30
mL of water and 100 mL of dichloromethane. After separation of
layers, the organic phase was washed three times with 100 mL of
water, and the common organic phase was dried by anhydrous
sodium sulfate and filtered, and the dichloromethane was evapo-
rated. A yellow crystalline mass was obtained and then dissolved
in chloroform and purified by silica gel chromatography. The
compound was crystallized from ethanol:diethyl ether (5:95) at 4
°C. Yield ) 91%; mp 149-154 °C. Anal. Calcd for C29H32O5: C,
75.21; H, 7.32. Found: C, 75.23; H, 7.30. ESMS [MW + (Na+)]:
1
Found: C, 49.31; H, 8.31. H NMR (D2O) δ, ppm, 1.04 (t, 3H,
[O-CH2-CH3]), 1.39 (m, 2H, [O-CH2-CH2-CH3]), 3.25 (d, 2H,
[C-eH5]), 3.29 (t, 2H, [O-CH2-CH2-CH3]), 3.41 (q, 1H, [C-H3]),
3.56 (d, 2H, [C-aH5]), 3.71 (m, 1 H, [C-H4]), 3.81 (m, 1H, [C-H2]),
(R) 5.71, (ꢀ) 4.65 (d, 1 H, [C-H]). 13C NMR (D2O) δ, ppm, 14.84
(CH2CH2CH3), 61.17 (C5), 63.99 (CH2CH2CH3), 64.12
(CH2CH2CH3), 68.85 (C4), 71.04 (C3), 73.72 (C2), 92.07 (HRC),
98.15 (HꢀC). 18, yield ) 31.0%. Anal. Calcd for C9H18O5: C, 51.02;
H, 7.92. Found: C, 52.11; H, 8.20. 1H NMR (D2O) δ, ppm, 1.04 (t,
3H, [O-CH2-CH2-CH2-CH3]), 1.24-1.39 (m, 4H, [O-CH2-CH2-CH2-
CH3]), 3.25 (d, 2 H, [C-eH5]), 3.29 (t, 2 H, [O-CH2-CH2-CH2-
CH3]), 3.41 (q, 1H, [C-H3]), 3.56 (d, 2H, [C-aH5]), 3.71 (m, 1H,
[C-H4]), 3.81 (m, 1H, [C-H2]), (R) 5.67, (ꢀ) 4.43 (d, 1H, [C-H]).
13C NMR (D2O) δ, ppm, 14.84 (CH2CH2CH2CH3), 61.17 (C5),
63.99 (CH2CH2CH2CH3), 64.12 (O-CH2CH2CH2CH3), 67.80 (O-
CH2CH2CH2CH3), 68.85 (C4), 71.04 (C3), 73.72 (C2), 92.07 (HRC),
98.15 (HꢀC).
1
m/e 483.19, calcd m/e 483.22. H NMR (CDCl3) δ, ppm, 1.04 (t,
3H, [O-CH2-CH3]), 1.51-1.31 (ds, 6H, [CH3-C-CH3]), 3.86 (q, 2H,
[O-CH2-CH3]), 4.11 (q, 1H, [C-H3]), 4.35 (d, 2 H, [C-H25]), 4.55
(q, 1H, [C-H2]), 4.72 (q, 1H, [C-H4]), 5.91 (d, 1H, [C-H1]),
7.10-7.29 (m, 15 H, [trityl]).
3-O-Methyl-D-xylose Diethyl Dithioacetal (22). Cold ethaneth-
iol (15 mL) was added to an ice-cold solution of 17.4 g of 3 in 15
mL of 6 M hydrochloric acid. The reaction was left stirring for
12 h and then treated with 2 M sodium carbonate. When the pH
became alkaline, 100 mL of acetone and 2.0 g of active carbon
5-O-Trityl-3-O-propyl-1,2-isopropylidene-r-xylopyranoside (11)
and 5-O-Trityl-3-O-butyl-1,2-isopropylidene-r-xylopyranoside
(12). These compounds were synthesized as described in the
synthetic procedure of 10 using iodopropane or iodobutane,

8106 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Gruzman et al.
were added to the reaction, and the mixture was filtered. The filtrate
was washed with acetone, and the remaining ethanethiol, water,
and acetone were distilled out. A syrup was obtained and purified
by column silica gel chromatography (ethyl acetate:petrolium ether
30:70). Yield ) 76%. ESMS [MW + (Na+)]: m/e 293.11, calcd
m/e 293.10. 1H NMR (CDCl3) δ, ppm, 1.93 (t, 6H, [S-CH2-CH3]),
2.67 (m, 4H [S-CH2-CH3]), 3.56 (s, 3H [O-CH3) 3.72 (q, 1H,
[C-H3]), 3.77 (m, 2H, [C-H5]), 3.95 (d, 1H, [C-H1]), 4.03 (m, 1H,
[C-H4]) 4.08 (q, 1H, [C-H2]).
to Konrad et al.41 Briefly, PBS-washed myotubes were incubated
with PBS supplemented with 10 µM 3-O-methyl-D-glucose and 5.0
µCi/mL [3H]-3-O-methyl-D-glucose and 0.5 µCi/mL [14C]sucrose
for 90 s at room temperature. The reaction was stopped by three
washes with ice-cold PBS containing 20 mM D-glucose, 10 µM
cytochalasin B, and 2 mM HgCl2. Following all uptake assays, the
myotubes were then lysed with 0.1% (w/v) SDS in water and taken
3
for liquid scintillation counting of H and/or 14C. The counts of
extracellular [3H]-3-O-methyl-D-glucose were calculated from the
counts of the extracellular marker [14C]sucrose and deducted from
the total counts of tritium.
3-O-Ethyl-D-xylose Diethyl Dithioacetal (23). This compound
was synthesized as 22 but using 16 as starting material. Yield )
1
39.8%. ESMS [MW + (Na+)]: m/e 307.78, calcd m/e 307.11. H
MTT Cell Viability Assay. This assay measures the reduction
of a tetrazolium component in MTT [3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyl tetrasodium bromide] into an insoluble formazan
product by the mitochondria of viable cells.42 Myotubes were
incubated with MTT (2 mg/mL in RMEM) for 30 min at 37 °C.
The medium was then aspirated, DMSO was added to solubilize
the colored crystals, and absorbance at 570 nm was measured in
an ELISA reader. The amount of color produced is directly
proportional to the number of viable cells.
Colorimetric Determination of Surface GLUT1myc and
GLUT4myc in L6 Myotubes. Colorimetric detection of surface
GLUT4myc or GLUT1myc was performed on the respective myc-
tagged L6 myotube cultures as previously described.39 Briefly,
cultured myotubes were incubated with rabbit anti-C-Myc antibody
(1:200 dilution), washed and fixed with 3% formaldehyde, and
further interacted with goat HRP-conjugated anti-rabbit IgG (1:
2000 dilution). Following washes, a solution of OPD was added,
and the culture plates were taken for absorbance measurement at
492 nM to estimate the relative abundance of GLUT1myc or
GLUT4myc on the plasma membrane of myotubes.
NMR (CDCl3) δ, ppm, 0.96 (t, 3H, [O-CH2-CH3), 1.93 (t, 6H,
[S-CH2-CH3]), 2.67 (m, 4H [S-CH2-CH3]), 3.53 (q, 2H [O-CH2-
CH3), 3.72 (q, 1H, [C-H3]), 3.77 (m, 2H, [C-H5]), 3.95 (d, 1H,
[C-H1]), 4.03 (m, 1H, [C-H4]), 4.08 (q, 1H, [C-H2]).
2,4-Benzylidene-3-O-methyl-D-xylose Diethyl Dithioacetal
(24). Compound 22 (1.05 g) was dried by three washes with toluene
and evaporation. It was then dissolved in 10 mL of dry DMF to
which 1.4 mL of benzaldehyde dimethyl acetal, 0.5 g of molecular
sieve 3 Å, and 100 mL of dry hydrochloric acid 4 M (in dioxane)
were added. The reaction mix was left stirring for 12 h followed
by neutralization with 15% (w/v) sodium bicarbonate. The mixture
was then filtered, and the filtrate was washed with chloroform and
evaporated. Yellow syrup that was obtained was further purified
by HPLC using a stepwise gradient of ethyl acetate (10-50%) in
petroleum ether. The compound was crystallized from diethyl ether,
and 0.29 g of white crystals was obtained. Yield ) 11.4%; mp )
47-54 °C. Anal. Calcd for C17H26O4S2: C, 56.95; H, 7.31; S, 17.89.
Found: C, 56.94; H, 7.18; S, 17.56. ESMS [MW + (K+)]: m/e
1
397.0, calcd m/e 397.05. H NMR (CDCl3) δ, ppm, 1.23 (t, 6H,
[S-CH2-CH3]), 2.67 (m, 4H [S-CH2-]), 3.62 (s, 3H [O-CH3), 3.72
(q, 1H, [C-H3]), 3.81 (m, 2H, [C-H5]), 3.99 (d, 1H, [C-H1]), 4.23
(m, 1H, [C-H4]), 4.66 (q, 1H, [C-H2]), 5.84 (R), 5.98 (S), S/R-
ratio 46%/54% (s, 1H, [O-CH-O]), 7.35-7.47 (m, 5H, [Bz]).
2,4-Benzylidene-3-O-ethyl-D-xylose Diethyl Dithioacetal (25).
This compound was synthesized as 24 but using 23 as starting
material. Yield ) 19%; mp ) 49-51 °C. Anal. Calcd for
C18H28O4S2: C, 58.03; H, 7.58; S, 17.21. Found: C, 58.14; H, 7.08;
S, 17.36. ESMS [MW + (Na+)]: m/e 395.17, calcd m/e 395.14. 1H
NMR (CDCl3) δ, ppm, 0.97 (t, 2H, [O-CH2-CH3]), 1.23 (t, 6H,
[S-CH2-CH3]), 2.67 (m, 4H, [S-CH2-]), 3.52 (q, 2H, [O-CH2-CH3
]), 3.72 (q, 1H, [C-H3]), 3.81 (m, 2H, [C-H5]), 3.99 (d, 1H, [C-H1]),
4.23 (m, 1H, [C-H4]), 4.66 (q, 1H, [C-H2]), 5.86 (R), 6.06 (S),
S/R-ratio 39%/61% (s, 1H, [O-CH-O]), 7.37-7.49 (m, 5H, [Bz]).
Cell Cultures. L6 skeletal myocytes were grown and let to
differentiate to multinuclear myotubes (85-95% efficiency) in
RMEM supplemented with 2% (v/v) FCS, as described.38 L6
myoblasts expressing c-myc epitope-tagged GLUT1 (GLUT1myc)
and GLUT4 (GLUT4myc) were the courtesy of Dr. A. Klip (The
Hospital for Sick Children, Toronto, Canada). These cells were
grown and let to differentiate into multinucleated myotubes as
described.39 A clone of primary cultures of human skeletal muscle
myoblasts isolated from Rectus abdominis M. of a 38-year-old
Caucasian female donor was purchased from PromoCell (clone
SkMc-c12530; Heidelberg, Germany). Myoblast cultures were
prepared and let to differentiate into multinuclear myotubes using
the manufacturer’s protocols, media, and reagents.
Preparation of Cell Lysates and Western Blot Analyses.
Whole myotube lysates were prepared as previously described.43
The lysis buffer was 50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 1
mM EGTA, 1 mM Na3VO4, 150 mM NaCl, 50 mM NaF, 10 sodium
ꢀ-glycerophosphate, 5 mM sodium pyrophosphate, and 1 mM
PMSF, supplemented with 0.1% (v/v) NP-40, 0.1% (v/v) 2-ꢀ-
mercaptoethanol, and protease inhibitor cocktail (1:100 dilution).
Protein content in the supernatant was determined according to the
method of Bradford,44 using BSA standard dissolved in the same
buffer. Aliquots (5-60 µg of protein) were mixed with sample
buffer [62.5 mM Tris-HCl, pH 6.8, 2% (w/v) SDS, 10% (v/v)
glycerol, 50 mM DTT, and 0.01% (w/v) bromophenol blue], heated
at 95 °C for 5 min, except for samples used for GLUT determina-
tions that were diluted in a urea denaturation buffer [10 mM Tris-
HCl, pH 6.8, 8 M urea, 2% (w/v) SDS, 10 mM, and 0.01% (w/v)
bromophenol blue], and warmed at 37 °C for 15 min. Proteins were
separated on 6-12% SDS-PAGE, and Western blot analyses were
performed using antibodies according to the antibody supplier’s or
previous established protocols. Supporting Information Table 3
gives the main technical details of all buffers used, washing
procedures, and antibody dilutions. Following the transfer of
proteins from the gel to nitrocellulose membranes and incubation
with the second antibody (horseradish peroxidase-conjugated anti-
rabbit IgG), the membranes were washed and taken for chemolu-
minescence detection using the EZ-ECL kit, according to the
manufacturer’s protocol. Band density measurement was performed
after scanning of films using Scion Image software.
ATP Determination. ATP content in lysates of L6 myotubes
was measured with the ENLITEN assay kit (Promega, Madison
WI). Myotubes in six-well plates were washed three times with
cold 0.15 M Tris-HCl, pH 7.75. The wells were then treated with
200 µL of 0.5% (w/v) ice-cold TCA, followed by the addition of
400 µL of this Tris-HCl buffer. The mixture was centrifuged in
Eppendorf tubes (12000 rpm, 30 min at 4 °C), and the supernatant
fractions were taken for ATP determination according to the
manufacturer’s instruction. Briefly, aliquots were mixed with
recombinant luciferase and D-luciferin. In the presence of oxygen
and ATP the enzyme catalyzes the oxidation of D-luciferase to
oxyluciferin coupled to hydrolysis of ATP to ADP and pyrophos-
phate and light emission at 560 nm. The assay was performed in
Hexose Uptake Assay. The rate of [3H]dGlc uptake in myotubes
in the absence or presence of insulin was measured as described.40
Briefly, L6 myotube cultures were usually preincubated in RMEM
supplemented with 2% (v/v) FCS and 23.0 mM D-glucose for 24 h,
after which they were treated as described. Insulin effect was
measured following its addition (200 nM) to cultures during the
last 30 min of treatment. The cultures were then rinsed three times
with PBS at room temperature and incubated with PBS, pH 7.4,
containing 0.1 mM dGlc and 1.3 µCi/mL [3H]dGlc for 5 min at
room temperature. The uptake was stopped by three rapid washes
with ice-cold PBS. The PBS solution for the D-xylose uptake assay
contained 0.1 mM D-xylose and 1.3 µCi/mL [14C]xylose. The rate
of [3H]-3-O-methyl-D-glucose transport was determined according

D-Xylose DeriVatiVes Stimulate Muscle Glucose Uptake
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24 8107
(6) Koivisto, U. M.; Martinez-Valdez, H.; Bilan, P. J.; Burdett, E.; Ramlal,
T.; Klip, A. Differential regulation of the GLUT-1 and GLUT-4
glucose transport systems by glucose and insulin in L6 muscle cells
in culture. J. Biol. Chem. 1991, 266, 2615–2621.
(7) Sasson, S.; Cerasi, E. Substrate regulation of the glucose transport
system in rat skeletal muscle. Characterization and kinetic analysis in
isolated soleus muscle and skeletal muscle cells in culture. J. Biol.
Chem. 1986, 261, 16827–16833.
low-background 96-multiwell plates for fluorescence measurement
(Nunc, Roskilde, Denmark) in a Mithras LB-940 luminometer
(Berthold Technologies, Bad Wilbad, Germany) at room temper-
ature. The ATP standard curve was constructed according to the
manufacturer’s protocol, using ATP supplied in the kit.
Determination of Protein Binding of 19. Culture medium
containing 10% (v/v) FCS was spiked with 4.6-23.0 µM 19 and
incubated at 37 °C for 60 min to determine Ct (bound and unbound
concentrations). Aliquots were centrifuged in 10 kDa cutoff
Microcon tubes (4400g for 60 min; Millipore, Bedford, MA). The
protein-free filtrates were collected and analyzed by LC-MS to
determine the concentration of free (Cf) 19 in the samples. The
protein binding capacity (B) of 19 was calculated by the equation
B (%) ) [(Ct - Cf)/Ct] × 100.
(8) Campbell, P. N. Metabolic studies with 3-methyl glucose. II. In vitro
studies. Biochem. J. 1952, 52, 444–447.
(9) Li, D.; Randhawa, V. K.; Patel, N.; Hayashi, M.; Klip, A. Hyperos-
molarity reduces GLUT4 endocytosis and increases its exocytosis from
a VAMP2-independent pool in l6 muscle cells. J. Biol. Chem. 2001,
276, 22883–22891.
(10) Sasson, S.; Kaiser, N.; Dan-Goor, M.; Oron, R.; Koren, S.; Wertheimer,
E.; Unluhizarci, K.; Cerasi, E. Substrate autoregulation of glucose
transport: Hexose 6-phosphate mediates the cellular distribution of
glucose transporters. Diabetologia 1997, 40, 30–39.
(11) Kiyosawa, K. Volumetric properties of polyols (ethylene glycol,
glycerol, meso-erythritol, xylitol and mannitol) in relation to their
membrane permeability: group additivity and estimation of the
maximum radius of their molecules. Biochim. Biophys. Acta 1991,
1064, 251–255.
(12) Huguenin, P.; Cochet, B.; Balant, L.; Loizeau, E. D-Xylose absorption
test. A pharmacokinetic and statistical study. Schweiz. Med. Wochen-
schr. 1978, 108, 206–214.
(13) Bollenback, G. Methyl-alpha-D-glucopyranoside. In Methods in Car-
bohydrate Chemistry; Whistler, R., Wilfrom, M., BeMiller, J., Eds.;
Academic Press: New York, 1963; Vol. 2, pp 326-327.
(14) Moravcova, J.; Capkova, J.; Stanek, J. One-pot synthesis of 1,2-O-
isopropylidene-alpha-D-xylopyranose. Carbohydr. Res. 1994, 263, 61–
66.
(15) Bowering, W. An alternative synthesis of 2-O-methyl-D-xylose. Can.
J. Chem. 1958, 36, 283–284.
(16) Levine, P.; Raymond, A. 3-Methyl-xylose and 5-methyl-xylose. J. Biol.
Chem. 1933, 102, 317–324.
The sample for mass spectrometric detection was prepared as
follows: Acetonitrile (150 µL) was mixed with 150 µL of filtrates.
Testosterone (0.1 µg), which served as an internal standard, was
also added. After extraction with 3 mL of n-hexane for 1 min and
centrifugation (1000g for 7 min) the upper organic layer was
separated and evaporated to dryness. The residue was dissolved in
100 µL of acetonitrile, and 20 µL was injected into the LC-MS
system. The quantity of 19 was measured using the following
system: HPLC-MS Waters Millennium instrument equipped with
Waters Micromass ZQ detector, 600 Controller gradient pump, and
717 autosampler (Waters Co., Milford, MA). Settings: nitrogen
flow, 500 L/h; source temperature, 400 °C; cone voltage, 35 V;
column, XTerra RP-18 (3.5 µm), 2.1 × 100 mm, heated to 35 °C.
The mobile phase (0.3 mL/min) was water and acetonitrile (1:1)
containing 0.1% (v/v) formic acid. 19 was monitored at SIR m/z
432 and retained for 24 min. Testosterone was monitored at SIR
m/z 289 and retained for 3 min.
Statistical Analysis. Statistical significance (p < 0.05) was
calculated among experimental groups using the Mann-Whitney
test. Results are given as mean ( SEM (n ) 3).
(17) Takeo, K.; Murata, Y.; Kitamura, S. A facile synthesis of 4-O-allyl-
D-xylopyranoses and it’s in the preparation of xylo-oligosaccharides.
Carbohydr. Res. 1992, 224, 311–318.
(18) de Bruyne, C. K.; Loontiens, F. G. Synthesis of alkyl beta-xylopy-
ranoside in the presence of mercuric salts. Nature 1966, 209, 396–
397.
(19) Ly, Y.; Just, G. Study on the diastereoselective synthesis of dithymidine
phosphorothionates through a D-xylose derived chiral auxiliary and
development of novel catalyst. Tetrahedron Lett. 2000, 41, 9223–
9227.
(20) Silverman, R. B. The organic chemistry of drug design and drug action;
Academic Press: San Diego, 1992; pp 352-396.
(21) Silverman, R. B. The organic chemistry of drug design and drug action;
Academic Press: San Diego, 1992; pp 289-319.
(22) Nomura, M.; Shuto, S.; Matsuda, A. Synthesis of the cyclic and acyclic
acetal derivatives of 1-(3-C-ethynyl-beta-D-ribo-pentofuranosyl)cy-
tosine, a potent antitumor nucleoside. Design of prodrugs to be
selectively activated in tumor tissues via the bio-reduction-hydrolysis
mechanism. Bioorg. Med. Chem. 2003, 11, 2453–2461.
(23) Curtis, J.; Jones, K. The synthesis of 3-O-beta-D-xylopyranosyl-D-
xylose. Can. J. Chem. 1960, 38, 1305–1315.
Acknowledgment. We are thankful to Mrs. R. Oron and
Ms. E. Oliver for excellent technical assistance. S.S., A.H., and
Y.K. are members of the David R. Bloom Center for Pharmacy
at the Hebrew University of Jerusalem. A.G., D.S., M.B.Y.,
E.A., and G.C. received fellowships from the Diabetes Research
Center of The Hebrew University. This study was supported
by grants from the Applied Research Program A of The Hebrew
University, The Deutch Foundation for Applied Sciences, Alex
Grass Center for Drug Design and Synthesis of Novel Thera-
peutics, David R. Bloom Center for Pharmacy at The Hebrew
University School of Pharmacy, the Nophar Program of the
Israel Ministry of Commerce and Trade, and the Israel Science
Foundation of The Israel Academy of Sciences and Humanities.
(24) Ferrier, R.; Hatton, L. The acid-catalyzed condensation of D-xylose
with benzaldehyde in the presence of alcohols. Two diastereoisomeric
1,2:3,5-di-O-benzylidene-alpha-D-xylofuranoses. Carbohydr. Res.
1967, 5, 132–139.
(25) Rao, S. P.; Grindley, T. B. The reduction of benzylidene derivatives
of pentose diethyl dithioacetals with lithium aluminium hydrine-
aluminium trichloride. Carbohydr. Res. 1991, 218, 83–93.
(26) Eros, D.; Kovesdi, I.; Orfi, L.; Takacs-Novak, K.; Acsady, G.; Keri,
G. Reliability of log P predictions based on calculated molecular
descriptors: A critical review. Curr. Med. Chem. 2002, 9, 1819–1829.
(27) Zaid, H.; Antonescu, C. N.; Randhawa, V. K.; Klip, A. Insulin action
on glucose transporters through molecular switches, tracks and tethers.
Biochem. J. 2008, 413, 201–215.
(28) Khan, A. H.; Pessin, J. E. Insulin regulation of glucose uptake: a
complex interplay of intracellular signalling pathways. Diabetologia
2002, 45, 1475–1483.
(29) Kanai, F.; Ito, K.; Todaka, M.; Hayashi, H.; Kamohara, S.; Ishii, K.;
Okada, T.; Hazeki, O.; Ui, M.; Ebina, Y. Insulin-stimulated GLUT4
translocation is relevant to the phosphorylation of IRS-1 and the
activity of PI3-kinase. Biochem. Biophys. Res. Commun. 1993, 195,
762–768.
Supporting Information Available: Experimental details of the
purity of compounds, data of in vitro assay, chemistry, biological
effects, and cytotoxicity of nonactive derivatives, technical details
of Western blot analyses, and tracing profiles of lead compounds.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Lann, D.; LeRoith, D. Insulin resistance as the underlying cause for
the metabolic syndrome. Med. Clin. North Am. 2007, 91, 1063–1077.
(2) Alpert, E.; Totary, H.; Sasson, S. Cellular mediators of glucose-induced
autoregulation of hexose transport. In Frontiers in animal diabetes
research: Muscel metabolism, Zierath, J. R., Walberg-Henriksson, H.,
Eds.; Taylor & Francis Books, London, 2002; pp 155-167.
(3) Zierath, J. R.; Kawano, Y. The effect of hyperglycaemia on glucose
disposal and insulin signal transduction in skeletal muscle. Best Pract.
Res., Clin. Endocrinol. Metab. 2003, 17, 385–398.
(4) Ben Yakir, M.; Gruzman, A.; Alpert, E.; Sasson, S. Glucose transport
regulators. Curr. Med. Chem.: Immunol., Endocr. Metab. Agents 2005,
5, 519–527.
(5) Misra, P. AMP activated protein kinase: a next generation target for
total metabolic control. Expert Opin. Ther. Targets 2008, 12, 91–
100.
(30) Barnett, S. F.; Defeo-Jones, D.; Fu, S.; Hancock, P. J.; Haskell, K. M.;
Jones, R. E.; Kahana, J. A.; Kral, A. M.; Leander, K.; Lee, L. L.;
Malinowski, J.; McAvoy, E. M.; Nahas, D. D.; Robinson, R. G.; Huber,
H. E. Identification and characterization of pleckstrin-homology-

8108 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 24
Gruzman et al.
domain-dependent and isoenzyme-specific Akt inhibitors. Biochem.
J. 2005, 385, 399–408.
Efendic, S.; Moller, D. E.; Thorell, A.; Goodyear, L. J. Metformin
increases AMP-activated protein kinase activity in skeletal muscle of
subjects with type 2 diabetes. Diabetes 2002, 51, 2074–2081.
(38) Bashan, N.; Burdett, E.; Guma, A.; Sargeant, R.; Tumiati, L.; Liu, Z.;
Klip, A. Mechanisms of adaptation of glucose transporters to changes
in the oxidative chain of muscle and fat cells. Am. J. Physiol. 1993,
264, C430–C440.
(39) Wang, Q.; Khayat, Z.; Kishi, K.; Ebina, Y.; Klip, A. GLUT4
translocation by insulin in intact muscle cells: Detection by a fast and
quantitative assay. FEBS Lett. 1998, 427, 193–197.
(40) Alpert, E.; Gruzman, A.; Studler-Lardi, B.; Blejter, R.; Reich, R.;
Sasson, S. COX-2 inhibitors augment the rate of hexose transport in
L6 skeletal myotubes. Diabetologia 2006, 49, 562–570.
(41) Konrad, D.; Rudich, A.; Bilan, P. J.; Patel, N.; Richardson, C.; Witters,
L. A.; Klip, A. Troglitazone causes acute mitochondrial membrane
depolarisation and an AMPK-mediated increase in glucose phospho-
rylation in muscle cells. Diabetologia 2005, 48, 954–966.
(42) Mutoh, T.; Kumano, T.; Nakagawa, H.; Kuriyama, M. Involvement
of tyrosine phosphorylation in HMG-CoA reductase inhibitor-induced
cell death in L6 myoblasts. FEBS Lett. 1999, 444, 85–89.
(43) Alpert, E.; Gruzman, A.; Totary, H.; Kaiser, N.; Reich, R.; Sasson, S.
A natural protective mechanism against hyperglycaemia in vascular
endothelial and smooth-muscle cells: Role of glucose and 12-
hydroxyeicosatetraenoic acid. Biochem. J. 2002, 362, 413–422.
(44) Bradford, M. M. A rapid and a sensitive method for the quantitation
of microgram quantites of protein utilizing the principle of protein-
dye binding. Anal. Biochem. 1972, 72, 248–254.
(31) Bouzakri, K.; Zachrisson, A.; Al-Khalili, L.; Zhang, B. B.; Koistinen,
H. A.; Krook, A.; Zierath, J. R. siRNA-based gene silencing reveals
specialized roles of IRS-1/Akt2 and IRS-2/Akt1 in glucose and lipid
metabolism in human skeletal muscle. Cell Metab. 2006, 4, 89–96.
(32) Kramer, H. F.; Witczak, C. A.; Fujii, N.; Jessen, N.; Taylor, E. B.;
Arnolds, D. E.; Sakamoto, K.; Hirshman, M. F.; Goodyear, L. J.
Distinct signals regulate AS160 phosphorylation in response to insulin,
AICAR, and contraction in mouse skeletal muscle. Diabetes 2006,
55, 2067–2076.
(33) Fujii, N.; Hirshman, M. F.; Kane, E. M.; Ho, R. C.; Peter, L. E.; Seifert,
M. M.; Goodyear, L. J. AMP-activated protein kinase alpha2 activity
is not essential for contraction- and hyperosmolarity-induced glucose
transport in skeletal muscle. J. Biol. Chem. 2005, 280, 39033–39041.
(34) Isakovic, A.; Harhaji, L.; Stevanovic, D.; Markovic, Z.; Sumarac-
Dumanovic, M.; Starcevic, V.; Micic, D.; Trajkovic, V. Dual anti-
glioma action of metformin: cell cycle arrest and mitochondria-
dependent apoptosis. Cell. Mol. Life Sci. 2007, 64, 1290–1302.
(35) Bruss, M. D.; Arias, E. B.; Lienhard, G. E.; Cartee, G. D. Increased
phosphorylation of Akt substrate of 160 kDa (AS160) in rat skeletal
muscle in response to insulin or contractile activity. Diabetes 2005,
54, 41–50.
(36) Hawley, S. A.; Gadalla, A. E.; Olsen, G. S.; Hardie, D. G. The
antidiabetic drug metformin activates the AMP-activated protein kinase
cascade via an adenine nucleotide-independent mechanism. Diabetes
2002, 51, 2420–2425.
(37) Musi, N.; Hirshman, M. F.; Nygren, J.; Svanfeldt, M.; Bavenholm,
P.; Rooyackers, O.; Zhou, G.; Williamson, J. M.; Ljunqvist, O.;
JM8008713