3-O-Methyl-D-glucose mutarotation and proton exchange rates assessed by 13C, 1H NMR and by chemical exchange saturation transfer and spin lock measurements

Abstract

3-O-Methyl-d-glucose (3OMG) was recently suggested as an agent to image tumors using chemical exchange saturation transfer (CEST) MRI. To characterize the properties of 3OMG in solution, the anomeric equilibrium and the mutarotation rates of 3OMG were studied by ¹H and ¹³C NMR. This information is essential in designing the in vivo CEST experiments. At room temperature, the ratio of α and β 3OMG anomers at equilibrium was 1:1.4, and the time to reach 95% equilibrium was 6 h. The chemical exchange rates between the hydroxyl protons of 3OMG and water, measured by CEST and spin lock at pH 6.14 and a temperature of 4 °C, were in the range of 360–670 s⁻¹.

Full text

1 23
Journal of Biomolecular NMR
ISSN 0925-2738
Volume 72
Combined 1-2
J Biomol NMR (2018) 72:93-103
DOI 10.1007/s10858-018-0209-y
3-O-Methyl-d-glucose mutarotation and
proton exchange rates assessed by 13C, 1H
NMR and by chemical exchange saturation
transfer and spin lock measurements
Michal Rivlin & Gil Navon

1 23
Your article is protected by copyright and
all rights are held exclusively by Springer
Nature B.V.. This e-offprint is for personal
use only and shall not be self-archived
in electronic repositories. If you wish to
self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

Vol.:(0123456789)
1 3
Journal of Biomolecular NMR (2018) 72:93–103
https://doi.org/10.1007/s10858-018-0209-y
ARTICLE
3-O-Methyl-d-glucose mutarotation andproton exchange rates
assessed by13C, 1H NMR andbychemical exchange saturation transfer
andspin lock measurements
MichalRivlin1· GilNavon1
Received: 24 June 2018 / Accepted: 5 September 2018 / Published online: 10 September 2018
© Springer Nature B.V. 2018
Abstract
3-O-Methyl-d-glucose (3OMG) was recently suggested as an agent to image tumors using chemical exchange saturation
transfer (CEST) MRI. To characterize the properties of 3OMG in solution, the anomeric equilibrium and the mutarotation
rates of 3OMG were studied by 1H and 13C NMR. This information is essential in designing the invivo CEST experiments.
At room temperature, the ratio of α and β 3OMG anomers at equilibrium was 1:1.4, and the time to reach 95% equilibrium
was 6h. The chemical exchange rates between the hydroxyl protons of 3OMG and water, measured by CEST and spin lock
at pH 6.14 and a temperature of 4°C, were in the range of 360–670s−1.
Keywords 3-O-Methyl-d-glucose (3OMG)· Mutarotation kinetics· Anomeric equilibrium· NMR spectroscopy· Chemical
exchange saturation transfer (CEST)· Spin lock (SL)
Abbreviations
α0 The initial concentration of alpha 3OMG
α Alpha anomer of 3OMG
β Beta anomer of 3OMG
BM Bloch–McConnell equations
c The time elapsed before the measurements had
started
CESL Chemical exchange spin lock
CEST Chemical exchange saturation transfer
CW Continuous wave
fb
The fraction of the labile proton
HSQC Heteronuclear single quantum correlation
K Equilibration constant
k1 The forward mutarotation rate constant
k2 The reverse mutarotation rate constant
k k1 + k2
kb The exchange rate from the labile proton to
water
M0 The water resonance intensity at a very far irra-
diation offset
MTRasym Magnetization transfer asymmetry ratio
NMR Nuclear magnetic resonance
NOE Nuclear overhauser effect
3OMG 3-O-Methyl-d-glucose
QUESP Quantification of exchange rate using varying
saturation power
R1a
The water longitudinal relaxation rate
R1ρ NMR spin relaxation in the rotating frame
SL Spin lock
T1 The longitudinal relaxation time
T0
2
The transverse relaxation time of water in the
absence of chemical exchange
ω1 CW irradiation amplitude
Δ𝜔
The frequency difference from the water peak
Zlab
The normalized Z magnetization after satura-
tion at the CEST resonance
Zref
The normalized Z magnetization after satura-
tion without CEST effect (estimated from the
opposite frequency)
Introduction
Interest is growing in chemical exchange saturation trans-
fer (CEST) MRI techniques as a means of producing new
contrast in magnetic resonance images with potential
usefulness in the clinic (Jones etal. 2018). CEST effect
obtained by functional exchangeable hydroxyl protons
groups of glucose and its analogues has been shown to be
* Gil Navon
navon@post.tau.ac.il
1 School ofChemistry, Tel Aviv University, TelAviv, Israel
Author's personal copy

94 Journal of Biomolecular NMR (2018) 72:93–103
1 3
a promising modality for the noninvasive diagnosis of can-
cer (Chan etal. 2012; Walker-Samuel etal. 2013; Rivlin
etal. 2013, 2014; Xu etal. 2015; Rivlin and Navon 2016,
2018; Wang etal. 2016; Longo etal. 2017). Also promis-
ing is the use of the non-metabolized (Hwang etal. 1992;
Xu and Krnjevic̀ 2001; Csaky and Glenn 1957; Jay etal.
1990) glucose analogue 3-O-methyl-d-glucose (3OMG) as
a CEST and CESL MRI agent for the detection of tumors
and the effects of stroke (Rivlin etal. 2014; Rivlin and
Navon 2018; Jin etal. 2018; Zu etal. 2018). 3OMG is
generally considered nontoxic, but detailed studies of its
toxicity are lacking. The present invitro work examines
the equilibria and kinetics of 3OMG required to assure
proper and effective use of this innovative product. Like
other cyclic forms of glucose (Los etal. 1956; Yamabe and
Ishikawa 1999) and its analogues, in water, 3OMG under-
goes a mutarotation reaction (Fig.1a) until equilibrium
between its α and β anomeric forms is reached (1H NMR
spectrum of 3OMG in D2O at 4°C is shown in Fig.1b).
The knowledge of the rate of mutarotation and the value
of the anomeric ratio at different temperatures and pH val-
ues is essential for the design of the MRI experiments.
Determination of the proton exchange rates can help in the
estimation of the tissue pH.
Fig. 1 a 3OMG mutarotation equilibrium, characterized by forward (k1) and reverse (k2) mutarotation rate constants. b 1H NMR spectrum of
0.5M 3OMG in D2O at T = 4°C, B0 = 11.7T
Author's personal copy

95Journal of Biomolecular NMR (2018) 72:93–103
1 3
Mutarotation kinetics
Because the mutarotation reaction is very slow, and the ratio
of the α and β anomeric forms may affect the results of the
CEST measurements in invivo experiments, it is essential
to determine the precise rate of mutarotation as a function
of temperature and pH.
3OMG crystalizes in the α anomer form and the mutaro-
tation reaction takes place upon its dissolution, reaching an
equilibrium between the two anomers. If α and β are the
concentrations of the α and β-anomers at time t, respectively,
then
where
α0
is the initial concentration of α-3OMG and
c
is the
time elapsed from the dissolution until the beginning of the
measurements.
The equilibration constant is defined as K = βeq/αeq.
In the present work 1H and 13C NMR spectroscopy were
used to follow the mutarotation process in freshly prepared
3OMG solutions.
Experimental
Chemicals
α-3OMG in powdered form and D2O (99.9% isotopically
enriched) were purchased from Sigma-Aldrich Chemical
(Rehovot, Israel).
Determination of3OMG mutarotation reaction
Mutarotation was initiated by dissolving samples of 3OMG
in D2O solution. Kinetic measurements were begun as
quickly as possible with a Bruker DRX 500MHz spectrom-
eter, set to the desired temperature.
1H NMR spectra were recorded every few minutes and
α and β anomeric peaks were determined by their intensi-
ties and integrals. The curve plots and the kinetics calcula-
tions were made using Matlab software (Matlab R2017a, the
MathWorks, Inc). At steady state, α and β peaks’ intensities
(1)
α=A+Be−k(t+c)
(2)
β=B(1−e−k(t+c))
(3)
A
=
α
0
k
2
k1+k2
(4)
B
=
α
0
k
1
k1+k2
(5)
k=k1+k2
were normalized according to the ratio of their integrals.
To select relaxation delay (d1), longitudinal relaxation
times, T1, were measured using the ‘inversion-recovery’
pulse sequence. The longest T1 for the high temperature
(T = 42°C) was 13.8, 3.6 and 2.2s for the HDO, α and β
anomeric peaks, respectively (for deuterated 3OMG solu-
tion) and thus, the relaxation delay was selected as 3–5T1.
Acquisition parameters were as follows: spectral width
3kHz, data size 32K, pulse width 3µs, relaxation delay
30–60s (depending on the temperature setting), acquisition
time 5.5s.
13C NMR spectra were recorded at 125.76MHz using
a Bruker DRX 500MHz spectrometer. Each sample was
scanned overnight with 1H decoupling. Some of the 13C
NMR measurements were performed without 1H decoupling.
Acquisition parameters were as follows: spectral width 7.5
kHz, data size 16K, pulse width 7.7µs, relaxation delay
12s, acquisition time 2.2s. The longest relaxation time T1
was determined for T = 45 °C as 1.5 and 1.7s for the α
and β anomeric peaks, respectively (for deuterated 3OMG
solution).
The conventional heteronuclear single quantum correla-
tion (HSQC) spectra were recorded with 256 increments,
32 scans, a recovery delay of 9s and an acquisition time
of 0.6s.
All data were recorded and analyzed with TopSpin 3.1
software (Bruker).
CEST measurements
CEST NMR experiments were performed on 20mM 3OMG
solutions (containing 5mM PBS and 5% D2O for lock) at pH
values of 6.14–7.47, and measured at 4°C and 37°C using
a Bruker DRX 500MHz spectrometer. The water intensity
was measured after long pre-saturation pulse (3–5s) at a
series of offset frequencies in the range of ± 5ppm. B1 power
was varied in the range of 0.5–5.9µT. The CEST effect is
described here in terms of a magnetization transfer asym-
metry ratio (MTRasym), a parameter that is defined as:
where
Δ𝜔
is the difference between the irradiation frequency
and the water resonance,
M(Δ
𝜔
)
and
M(−Δ
𝜔
)
are the water
intensities after a long pre-saturation pulse at the offset
frequency and the offset frequency on the opposite side of
the water signal, respectively. M0 is the water resonance
intensity at a very far irradiation offset (at which the water
resonance intensity is unaffected by functional groups).
To define the CEST signal by the MTR asymmetry of the
normalized Z magnetization, two Z values were used: the
normalized Z magnetization after saturation at the CEST
(6)
MTR
asym(Δ𝜔)=Zref
(
B1
)
−Zlab
(
B1
)
=
M(−Δ
𝜔
)−M(Δ
𝜔
)
M
0
Author's personal copy

96 Journal of Biomolecular NMR (2018) 72:93–103
1 3
resonance
Zlab =Z(Δ
𝜔
),
and the reference scan which rep-
resents the saturation without CEST effect that can be esti-
mated from the opposite frequency
Zref =Z(−Δ
𝜔
).
CEST quantification
The exchange rates between the 3OMG hydroxyl protons and
water were estimated using the dependence of the water Z
magnetization on the CW irradiation amplitude ω1, identified
as quantification of exchange rate using varying saturation
power (QUESP) (McMahon etal. 2006) equation. The inverse
Z-spectrum analysis (Zaiss etal. 2014, 2018; Wu etal. 2015)
that allows for the correction of spillover effects was used:
where
fb
is the fraction of the labile proton,
kb
is its exchange
rate,
Δ𝜔
is the frequency difference from the water peak, and
R1a
is the water longitudinal relaxation rate. The term
(
ω1
Δ
𝜔
)2
that was neglected in a previous publication (Zaiss etal.
2018) was included in the present analysis because of the
proximity of the hydroxyl peaks to the water peak. Nonlinear
fit was used rather than the linear omega plot (Wu etal.
2015; Meissner etal. 2015), since in the latter, excessive
weight is given to low values of ω1.
Spin lock measurements
For on-resonance
R1ρ
(1
/
T
1ρ)
dispersion experiments, nuclei
were excited by a 90° pulse and then spin locked by a long CW
pulse and variable B1 fields.
Spin lock quantification
For the calculation of the exchange rate from R1ρ dispersion
measurements, data were fitted to Eq.(8) given by Trott and
Palmer (2002):
where
T0
2
is the transverse relaxation time of water in the
absence of chemical exchange, and the definitions of
fb
,
kb
and
Δ𝜔
are the same as above. In the fitting,
1
∕T
0
2,
fb
,
kb
and
Δ𝜔
were kept as free parameters.
(7)
MTR
Rex
=1
Zref
−1
Zlab
=1
R1a
fbkb
𝜔
2
1
𝜔2
1
+k2
b
[1+
(
ω1
Δ𝜔
)
2
]
(8)
R
1ρ=
1
T0
2
+fbΔ𝜔2kb∕(Δ𝜔2+ω
2
1+k2
b
)
Results anddiscussion
Assignment
The assignment of the aliphatic and the hydroxyl protons
of 3OMG in aqueous solution was based on the known
assignment of α-3OMG in DMSO (http://www.sdbs.
db.aist.go.jp/sdbs/cgi-bin/direc t_frame _disp.cgi?sdbsn
o=13453 ; Onche etal. 2013). 1H NMR spectra at 18°C
before and after reaching anomeric equilibrium are pre-
sented in Fig.2a, b, respectively. Both α-H1 and β-H1
resonances are doublets with JH−H of 3.70 and 7.56Hz,
respectively. While the 13C NMR is not expected to differ
greatly between the two solvents, this is different for 1H
spectra. Hence the 13C–1H HSQC experiment was used for
the assignments of the aliphatic 3OMG protons (Fig.3).
Since the spectral peaks of the hydroxyl protons are well
separated at low pH and low temperature (Fig.4), their
identification was based on the known assignment in
DMSO (http://www.sdbs.db.aist.go.jp/sdbs/cgi-bin/direc
t_frame _disp.cgi?sdbsn o=13453 ) as well as in analogy
to the known assignment of glucose (Onche etal. 2013).
Mutarotation
The hydroxyl protons of the 3OMG are observable only
at low pH and low temperature and therefore are not suit-
able for the study of the mutarotation rates. Since the large
water signal interferes with following the aliphatic pro-
tons, the experiments were run in D2O solutions. It should
be noted that the mutarotation is presumed to be somewhat
slower in D2O than in H2O due to a combination of sol-
vent and kinetics isotope effects (Isbell and Pigman 1969;
Drake and Brown 1977). While at low temperatures both
1α and 1β peaks are well separated from the residual HDO
peak, at temperatures higher than 25°C the HDO peak
moves closer to the 1β peak. Thus, at these temperatures
the signal of the 2, 3β was used to follow the buildup of
the β anomer.
Time course of the mutarotation process was monitored
by running a set of 1H NMR single pulse experiments at
several different temperatures and pH values. For that pur-
pose, fresh solutions of 0.5M 3OMG were prepared in
D2O and immediately transferred to a 5mm NMR tube;
NMR measurements were started when the temperature
stabilized. Spectra were collected every few minutes for
several hours until equilibrium was reached. We simul-
taneously monitored the decrease of the α anomer signal
(peak at 5.26ppm) and the increase of the β anomer sig-
nals (peaks at 4.70ppm and 3.34ppm) that characterized
the mutarotation reaction. The rate of mutarotation for the
Author's personal copy

97Journal of Biomolecular NMR (2018) 72:93–103
1 3
α anomer was determined by measuring the kinetics of the
α-H1 peak. The kinetics of the β anomer was determined
by measuring the β-H1 peak at the low temperatures, w
hen it was well separated from the HDO peak, and the
β-H2, 3 peak at temperatures above 25°C.
The kinetics constants of mutarotation were observed by
optimizing the fit between Eqs.(2) and (3), and the data.
Examples are given in Fig.5.
The dependence of the rate constants on temperature
(Fig.6) yielded an activation energy of 35.3 (kJ/mol). The
inset shows the time required to reach 95% equilibrium
between the α and β anomer, in the temperature range of
10–45°C.
Figure7 illustrates the dependence of the reaction rates
on the pH of the solutions of 0.5M 3OMG in D2O at 37°C,
in the pH range of 4.53–8.15. k was determined by fitting
the signal intensities to Eqs.(2) and (3). As expected in a
base-catalyzed exchange process, k increases with pH. The
reaction rates for both the temporal plots of the α and β ano-
mers were very similar. This was expected since they are the
Fig. 2 1H NMR spectra of 0.5 M 3OMG in D2O solution and its
anomeric configurations (α- and β-forms) at 18 °C. Spectra were
recorded with B0 = 11.7 T on a a fresh sample, b and 22 h after
3OMG dissolution. The net spectra of the α and β anomers are pre-
sented in c, d, respectively. They are obtained by subtracting spec-
trum b from spectrum a and the reverse, respectively
Fig. 3 2D 13C–1H HSQC spectrum (absolute value) of 0.5M 3OMG in D2O solution before (a) and after (b) anomeric equilibrium was reached,
recorded at 7°C (a) and 25°C (b). The anomeric 13C–1H cross peaks are indicated
Author's personal copy

98 Journal of Biomolecular NMR (2018) 72:93–103
1 3
Fig. 4 1H NMR spectra of the hydroxyl protons of 0.5M 3OMG in
D2O solution at 4°C. Spectra were recorded with B0 = 11.7T (a) sev-
eral days following the sample preparation, and on a b fresh sample.
The net spectra of the α and β hydroxyl peaks are presented in c, d,
respectively. They are obtained by subtracting spectrum b from spec-
trum a and the reverse, respectively
Fig. 5 Kinetics of α and β anomers as determined by 1H NMR meas-
urements during the mutarotation of 0.5 M 3OMG in D2O at 13 °C
(a, b), 25°C (c, d) and 37°C (e, f). The kinetics of α anomer was fol-
lowed by the α-H1 peak and that of the β anomer by the β-H2, 3 peak.
Solid lines represent the fit adjustment to Eqs.(2) and (3) for α and β
anomers according to their evolution with time, respectively
Author's personal copy

99Journal of Biomolecular NMR (2018) 72:93–103
1 3
sums of the forward and reverse conversion rates: k = k1 + k2.
The reported results are the averaged of the two. The rate of
mutarotation for 3OMG at 25°C, k = 2.75E−04s−1, can be
compared to the value of k = 4.13 × E−04s−1 reported for
d-glucose (Dona etal. 2009).
13C NMR spectroscopy, which shows the unique features
of the α and β anomers, can be used to identify the two ano-
meric forms and the contribution of each at equilibrium. As
described earlier, signals at 98.8 and 88.2ppm are assigned
to β anomeric carbons of C1 and C3 of 3OMG, respectively,
and signals at 95.0 and 85.5ppm are assigned to α anomeric
carbons of C1 and C3 of 3OMG, respectively. Based on
13C NMR studies, (β/α)eq ratio of equilibrium population of
3OMG anomers is slightly temperature dependent, with an
average (β/α)eq ratio of 1.39 ± 0.01 in the temperature range
of 4–50°C (calculation based on the average of two sets
of intensities of the C1 and C3 of 3OMG). To verify that
there was no contribution from NOE, some of the 13C NMR
experiments were recorded without proton decoupling and
the intensity ratio of α to β anomeric forms was compared
with the 1H decoupled spectrum giving identical results.
The anomeric ratio was also evaluated from the 1H NMR
kinetic studies. The results at 12 pH values in the range of
4.53–8.15, showed no pH dependence giving an average
(α/β)eq ratio of 1.39 ± 0.09at 37°C. Hence, at equilibrium
the β anomer is favored (58%) compared to the α anomer
(42%), as illustrated in Fig.8.
CEST NMR studies
Earlier studies (Rivlin etal. 2014) have shown the 3OMG
MTRasym at 1.2ppm, 25°C, and pH range of 6.34–8.02,
exhibits maximum MTRasym effect at pH = 7.1. A significant
MTRasym effect is also detected at the same examined OH
group (1.2ppm) at 37°C (Fig.9).
Analyzing the Z spectrum at physiological condition,
where the hydroxyl protons are not resolved, requires the use
of the Bloch–McConnell (BM) equations (Zaiss and Bachert
Fig. 6 The dependence of
reaction rate on temperature dis-
played as Arrhenius plot (main
plot) and the time required to
reach 95% equilibrium (inset) in
fresh solutions of 0.5M 3OMG
in D2O at different tempera-
tures. Data were obtained from
1H NMR measurements of α
and β signals evolution during
mutarotation reaction
Fig. 7 The dependence of reaction rate on pH in fresh solutions of
0.5 M 3OMG in D2O at 37 °C [reaction rates are presented as an
average of α and β anomers and were obtained from their fit adjust-
ment to Eqs. (2) and (3), based on 1H NMR mutarotation measure-
ments]
Author's personal copy

100 Journal of Biomolecular NMR (2018) 72:93–103
1 3
2013; Zaiss etal. 2015), which is beyond the scope of this
work. Our calculations were based on the analytical expres-
sions for the inverse Z-spectrum analysis (Zaiss etal. 2014,
2018; Wu etal. 2015). As mentioned in the “Experimental”
section, this analysis (based on QUESP formula) enables the
elimination of spillover effects. To simplify its use, the case
of slow exchange was examined (all hydroxyls’ protons are
well observed). Hence, a series of multi B1 experiments with
a solution of 20 mM 3OMG (5 mM PBS, 5% D2O,
pH = 6.14, B0 = 11.7T) at 4°C were performed with satura-
tion pulse duration of 5s. Applying the inverse asymmetry
method (Fig.10) led to exchange rates of the hydroxyl pro-
tons, kb, of: 414, 390, 670 and 581s−1 for the hydroxyl peaks
at 0.75, 1.3, 2.0 and 2.7ppm, respectively. The same values
were obtained for the negative intercepts on the X axis in the
linear (“omega”) plots of
[
1+
(
ω1
Δ𝜔
)
2
]/(
1
Zref
−1
Zlab
)
versus
(
1
ω
1)2
,
which is independent of the values of the fractions,
but as is mentioned in the “Experimental”section above, in
the omega plots excessive weight is given to low values of
ω1. The fractions obtained from the fitting (Fig.10) (as well
as from the intercept on the Y axis in the omega plots) were
3.68E−04, 1.96E−04, 0.46E−04 and 0.54E−04 for the
hydroxyl peaks observed at 0.75, 1.3, 2.0 and 2.7ppm from
water signal, respectively. These value deviate from the
actual hydroxyl concentrations of 3OMG solution, i.e.
1.8E−04, 3.6E−04, 0.75E−04 and 1.05E−04. Apparently
such deviation is inherent in the analysis, as it occurred also
in the analysis of the simpler system of the paramagnetic
species with single resonance far from the water peak (Zaiss
etal. 2018).
Quantification of the proton exchange rates can also be
made by using the on-resonance spin lock (SL) (Jin etal.
2011) method. However, it should provide in principal a
weighted average exchange rate for all the exchangeable
protons. Figure11 shows the on-resonance R1ρ dispersions
for the same 3OMG solution that was used for the CEST
measurements. The R1ρ curve pattern is in agreement with
previously published results by Jin etal. (2018). The result
of the fitted total fraction is in agreement with 3OMG con-
centration (fb = 7.2E−04, R2 = 0.9976). The fitted
T0
2
was
1.67s, in agreement with the independently measured value.
The result of the exchange rate, kb, was 362s−1, which is in
the same range to those found with CEST experiments but
is slightly lower than the weighted average of the exchange
rates of the hydroxyl protons obtained in the CEST experi-
ments namely:
kav
=
∑
f
i
k
i
∕
∑
f
i
= 453s−1. However, the
fitted value of the frequency shift from the water peak,
Δ
𝜔
,
was 0.78ppm rather than the expected weighted average
Fig. 8 The dependence of anomeric composition (β/α) on tempera-
ture in a sample of 0.5M 3OMG in D2O. Filled blue circle based on
13C NMR peaks’ intensities at equilibrium (each point represents an
average of two sets of intensities for 13C-3 and 13C-1); red color times
symbol based on 1H NMR kinetics measurements
Fig. 9 MTRasym plots of 20 mM 3OMG solution at 37 °C (con-
taining 5 mM PBS and 5% D2O) with a pH values from 6.14 to
7.49 measured at a frequency offset of 1.2 ppm from water signal
(B1 = 2.4 µT), and b pH value of 7.1at frequency offsets of 1.2, 2.1
and 2.8 from water signal and at different B1 values
Author's personal copy

101Journal of Biomolecular NMR (2018) 72:93–103
1 3
shift. The SL results were also tried to fit to the following
expression, which takes into account the independent contri-
butions of the individual hydroxyl protons, using the known
values of fi and Δωi:
(9)
R
1ρ=
1
T0
2
+
∑
i
fiΔ𝜔2
iki∕(Δ𝜔2
i+ω
2
1+k2
i
)
However, the fitting to the experimental results was not
good and the resulted fitted parameters were unrealistic.
Thus we can conclude that the SL method gives a correct
rough estimate of the average exchange rates of the mul-
tiple exchange sites but cannot give information about the
individual exchange rates.
Fig. 10 The QUESP fitting (dots) to the inverse Z-spectrum analysis
(Eq.7, blue solid lines) for the hydroxyl peaks observed at a 0.75, b
1.3, c 2.0 and d 2.7ppm from the water peak, as obtained from exper-
imental CEST data of 20mM 3OMG solution (containing 5mM PBS
and 5% D2O, at pH = 6.14, T = 4°C, B0 = 11.7T)
Fig. 11 On-resonance R1ρ
dispersion versus spin lock
amplitude for 20mM 3OMG
solution (containing 5mM
PBS and 5% D2O, at pH 6.14,
T = 4°C, B0 = 11.7T)
Author's personal copy

102 Journal of Biomolecular NMR (2018) 72:93–103
1 3
Conclusions
Since 3OMG was proposed for the imaging of tumors
using CEST–MRI, it was important to establish the ano-
meric ratio at equilibrium and the time needed to reach
this equilibrium. Indeed, this time was found to be very
long, and at low temperatures can last for several days.
This should be taken into consideration when perform-
ing MRI experiments. For instance, doing the experiments
using fresh 3OMG samples can give considerable errone-
ous and irreproducible results. The determination of the
anomeric ratio at equilibrium is essential for future cal-
culation by BM of the proton exchange rates, which can
help determine the pH in tumors. The determination of the
exchange rates of the hydroxyl protons by CEST and by SL
gives order of magnitude values but points to difficulties
in treating systems of multiple exchange sites.
Acknowledgements The study was supported by the Israel Science
Foundation (ISF) and the European Union’s Horizon 2020 research and
innovation program under Grant Agreement No. 667510.
Compliance with ethical standards
Conflict of interest The authors declare no conflict of interest.
References
Chan KW, McMahon MT, Kato Y, Liu G, Bulte JW, Bhujwalla ZM,
Artemov D, van Zijl PC (2012) Natural d-glucose as a biode-
gradable MRI contrast agent for detecting cancer. Magn Reson
Med 68:1764–1773
Csaky TZ, Glenn JE (1957) Urinary recovery of 3-methylglucose
administered to rats. Am J Physiol 188:159–162
Dona A, Pagès G, Gilbert G, Gaborieau R, Kuchel MP (2009) Kinet-
ics of invitro digestion of starches monitored by time-resolved
1H nuclear magnetic resonance. Biomacromolecules 10:638–644
Drake EN, Brown CE (1977) Application of nmr to biochemical
kinetics. A laboratory experiment in physical biochemistry. J
Chem Edu 54:124–127
Hwang YY, Kim SG, Evelhoch JL, Ackerman JJ (1992) Nongly-
colytic acidification of murine radiation-induced fibrosarcoma
1 tumor via 3-O-methyl-d-glucose monitored by 1H, 2H, 13C,
and 31P nuclear magnetic resonance spectroscopy. Cancer Res
52:1259–1266
Isbell HS, Pigman W (1969) Mutarotation of sugars in solution. II.
Catalytic processes, isotope effects, reaction mechanisms, and
biochemical aspects. Adv Carbohydr Chem Biochem 24:13–65
Jay TM, Dienel GA, Cruz NF, Mori K, Nelson T, Sokoloff L (1990)
Metabolic stability of 3-O-methyl-d-glucose in brain and other
tissues. J Neurochem 55:989–1000
Jin T, Autio J, Obata T, Kim S-G (2011) Spin-locking vs. chemical
exchange saturation transfer MRI for investigating chemical
exchange process between water and labile metabolite protons.
Magn Reson Med 65:1448–1460
Jin T, Mehrens H, Wang P, Kim SG (2018) Chemical exchange-sen-
sitive spin-lock MRI of glucose analog 3-O-methyl-d-glucose
in normal and ischemic brain. J Cereb Blood Flow Metab
38:869–880
Jones KM, Pollard AC, Pagel MD (2018) Clinical applications of
chemical exchange saturation transfer (CEST) MRI. J Magn
Reson Imaging 47:11–27
Longo DL, Moustaghfir FZ, Zerbo A, Consolino L, Anemone A,
Bracesco M, Aime S (2017) EXCI-CEST: exploiting pharma-
ceutical excipients as MRI-CEST contrast agents for tumor
imaging. Int J Pharm 525:275–281
Los JM, Simpson LB, Wiesner K (1956) The kinetics of mutarota-
tion of d-glucose with consideration of an intermediate free-
aldehyde form. J Am Chem Soc 78:1564–1568
McMahon MT, Gilad AA, Zhou J, Sun PZ, Bulte JW, van Zijl PC
(2006) Quantifying exchange rates in chemical exchange satu-
ration transfer agents using the saturation time and saturation
power dependencies of the magnetization transfer effect on the
magnetic resonance imaging signal (QUEST and QUESP): Ph
calibration for poly-l-lysine and a starburst dendrimer. Magn
Reson Med 55:836–847
Meissner JE, Goerke S, Rerich E, Klika KD, Radbruch A, Ladd ME,
Bachert P, Zaiss M (2015) Quantitative pulsed CEST-MRI using
omega-plots. NMR Biomed 28:1196–1208
Onche EU, Tukura BW, Bako SS (2013) Nuclear magnetic resonance
(NMR) analysis of d-(+)-glucose: a guide to spectrometric
structural elucidation of sugars. IOSR J Appl Chem 6:45–51
Rivlin M, Navon G (2016) Glucosamine and N-acetyl glucosamine
as new CEST MRI agents for molecular imaging of tumors. Sci
Rep 6:32648
Rivlin M, Navon G (2018) CEST MRI of 3-O-methyl-d-glucose on
different breast cancer models. Magn Reson Med 79:1061–1069
Rivlin M, Horev J, Tsarfaty I, Navon G (2013) Molecular imaging
of tumors and metastases using chemical exchange saturation
transfer (CEST) MRI. Sci Rep 3:3045
Rivlin M, Tsarfaty I, Navon G (2014) Functional molecular imag-
ing of tumors by chemical exchange saturation transfer MRI of
3-O-methyl-d-glucose. Magn Reson Med 72:1375–1380
Trott O, Palmer AG (2002) R1ρ relaxation outside of the fast-
exchange limit. J Magn Reson 154:157–160
Walker-Samuel S, Ramasawmy R, Torrealdea F, Rega M, Rajku-
mar V, Johnson SP, Richardson S, Goncalves M, Parkes HG,
Arstad E, Thomas DL, Pedley RB, Lythgoe MF, Golay X (2013)
Invivo imaging of glucose uptake and metabolism in tumors.
Nat Med 19:1067–1072
Wang J, Weygand J, Hwang KP, Mohamed AS, Ding Y, Fuller CD,
Lai SY, Frank SJ, Zhou J (2016) Magnetic resonance imaging of
glucose uptake and metabolism in patients with head and neck
cancer. Sci Rep 6:30618
Wu R, Xiao G, Zhou IY, Ran C, Sun PZ (2015) Quantitative chemi-
cal exchange saturation transfer (qCEST) MRI—omega plot
analysis of RF spillover-corrected inverse CEST ratio asym-
metry for simultaneous determination of labile proton ratio and
exchange rate. NMR Biomed 28:376–383
Xu YZ, Krnjevic̀ K (2001) Unlike 2-deoxy-d-glucose, 3-O-methyl-
d-glucose does not induce long-term potentiation in rat hip-
pocampal slices. Brain Res 895:250–252
Xu X, Yadav NN, Knutsson L, Hua J, Kalyani R, Hall E, Laterra
J, Blakeley J, Strowd R, Pomper M, Barker P, Chan K, Liu
G, McMahon MT, Stevens RD, van Zijl PC (2015) Dynamic
glucose-enhanced (DGE) MRI: translation to human scanning
and first results in glioma patients. Tomography 1:105–114
Yamabe S, Ishikawa T (1999) Theoretical study of mutarotation of
glucose. J Org Chem 64:4519–4524
Zaiss M, Bachert P (2013) Chemical exchange saturation transfer
(CEST) and MR Z-spectroscopy invivo: a review of theoretical
approaches and methods. Phys Med Biol 58:221–269
Author's personal copy

103Journal of Biomolecular NMR (2018) 72:93–103
1 3
Zaiss M, Junzhong X, Steffen G, Imad SK, Robert JS, John CG, Dan-
iel FG, Peter B (2014) Inverse Z-spectrum analysis for spillo-
ver-, MT-, and T1-corrected steady-state pulsed CEST-MRI—
application to pH-weighted MRI of acute stroke. NMR Biomed
27:240–252
Zaiss M, Zu Z, Zhongliang XJ, Schuenke P, Gochberg DF, Gore JC,
Ladd ME, Bachert P (2015) A combined analytical solution for
chemical exchange saturation transfer and semi-solid magnetiza-
tion transfer. NMR Biomed 28:217–230
Zaiss M, Angelovski G, Demetriou E, McMahon MT, Golay X, Schef-
fler K (2018) QUESP and QUEST revisited—fast and accurate
quantitative CEST experiments. Magn Reson Med 79:1708–1721
Zu Z, Jiang X, Xu J, Gore JC (2018) Spin-lock imaging of 3-O-methyl-
d glucose (3OMG) in brain tumors. Magn Reson Med
80:1110–1117
Author's personal copy