Dephosphorylation and biodistribution of 1-13C-phospholactate in vivo

Article abstract of DOI:10.1002/jlcr.3207

Here, we present a new approach for the delivery of a metabolic contrast agent for in vivo molecular imaging. The use of a phosphate-protecting group that facilitates parahydrogen-induced polarization of 1-¹³C- phospholactate potentially enables the in vivo administration of a hydrogenated hyperpolarized adduct. When injected, nonhyperpolarized 1-¹³C- phospholactate is retained in the vasculature during its metabolic conversion to 1-¹³C-lactate by blood phosphatases as demonstrated here using a mucin 1 mouse model of breast cancer and ex vivo high-resolution ¹³C NMR. This multisecond process is a suitable mechanism for the delivery of relatively short-lived ¹³C and potentially ¹⁵N hyperpolarized contrast agents using -OH phosphorylated small molecules, which is demonstrated here for the first time as an example of 1-¹³C- phospholactate. Through this approach, dl-1-¹³C-lactate is taken up by tissues and organs including the liver, kidneys, brain, heart, and tumors according to a timescale amenable to hyperpolarized magnetic resonance imaging. The use of a phosphate-protecting group that facilitates parahydrogen-induced polarization of 1-¹³C-phospholactate potentially enables the in vivo administration of a hydrogenated hyperpolarized adduct. When injected, nonhyperpolarized 1-¹³C-phospholactate is retained in the vasculature during its metabolic conversion to dl-1-¹³C-lactate by blood phosphatases, which in turn taken up by tissues and organs according to a timescale amenable to hyperpolarized magnetic resonance imaging (MRI). This represents a new approach for in delivery of a metabolic hyperpolarized MRI contrast agent for molecular imaging. Copyright

Full text of DOI:10.1002/jlcr.3207

Research Article
Received 15 March 2014,
Revised 07 April 2014,
Accepted 20 April 2014
Published online 3 July 2014 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.3207
Dephosphorylation and biodistribution of
†
13
1- C-phospholactate in vivo
Roman V. Shchepin, Wellington Pham,a,b and Eduard Y. Chekmenev
a
*
a,b**
Here, we present a new approach for the delivery of a metabolic contrast agent for in vivo molecular imaging. The use of a
phosphate-protecting group that facilitates parahydrogen-induced polarization of 1-13C-phospholactate potentially enables
the in vivo administration of a hydrogenated hyperpolarized adduct. When injected, nonhyperpolarized 1-13C-phospholactate
is retained in the vasculature during its metabolic conversion to 1-13C-lactate by blood phosphatases as demonstrated here
using a mucin 1 mouse model of breast cancer and ex vivo high-resolution 13C NMR. This multisecond process is a suitable
mechanism for the delivery of relatively short-lived 13C and potentially 15N hyperpolarized contrast agents using –OH
phosphorylated small molecules, which is demonstrated here for the first time as an example of 1-13C-phospholactate. Through
this approach, DL-1-13C-lactate is taken up by tissues and organs including the liver, kidneys, brain, heart, and tumors according
to a timescale amenable to hyperpolarized magnetic resonance imaging.
Keywords: NMR; hyperpolarization; contrast agent; phospholactate; parahydrogen
13C-succinate20–22 and tetrafluoropropyl 13C-propionate23,24 reporting
on cancer metabolism and plaque lipid content, respectively, in
Introduction
preclinical studies in mice. Nevertheless, hyperpolarized 13C-succinate
is similar to hyperpolarized 13C-fumarate in that it is a negative
contrast agent of cancer, meaning that their uptake and metabolism
decrease with cancer progression, which potentially limits the
biomedical utility of such contrast agents. Hyperpolarized 1-13C-
pyruvate, on the other hand, is a positive contrast agent because it
reports on the elevated rate of glycolysis, which increases with tumor
grade and cancer progression10. Unfortunately, 1-13C-pyruvate cannot
be polarized with PHIP25.
Hyperpolarized magnetic resonance imaging is
a rapidly
emerging field of molecular imaging intended to explore
metabolically impaired diseases such as cancer. NMR
hyperpolarization increases nuclear spin alignment by several
orders of magnitude1, an outcome that can be exploited by
preparation of exogenous metabolic contrast agents, which
can interrogate the metabolism in living organisms. In less than
a decade, hyperpolarized magnetic resonance has progressed
from the proof-of-principle studies centered on the preparation
of hyperpolarized contrast agents2,3 to the first clinical trial4.
The key feature of these contrast agents is their relatively short
lifetime of approximately 1–3 min5–7. During this time, the
contrast agent is taken up by the tissue of interest, for example,
tumor, converted to the metabolite(s) of interest and is then
However, a recently introduced concept of –OH protection of the
C=C–OH motif during PHIP process25,26 significantly expanded the
reach of molecular contrast agents that can be hyperpolarized by
PHIP. The use of phosphate-protecting group25, in particular, is an
attractive approach because blood phosphatases can potentially
remove the phosphate-protecting group in vivo and therefore
activate the hyperpolarized contrast agent in vivo (Figure 1a)25.
Combined with the biological compatibility of nontoxic inorganic
detected spectroscopically and spatially via NMR8,9
detected metabolic signatures serve as biomarkers of cancer10,
disease aggressiveness, and treatment response4,5,11
Several hyperpolarization technologies are available. While
dissolution dynamic nuclear polarization (DNP) leads the field
of hyperpolarized in vivo imaging, parahydrogen-induced
. The
.
aVanderbilt University Institute of Imaging Science (VUIIS)Department of
polarization (PHIP)12 is an ultrafast technology capable of RadiologyNashville, TN, 37232, USA
producing a hyperpolarized contrast agent within a minute13,14
However, PHIP requires molecular hydrogenation with
parahydrogen gas, which governs the source of spin order15,16
.
bDepartment of Biomedical Engineering and Biochemistry, Vanderbilt-Ingram
Cancer Center, Vanderbilt University, Nashville, TN, USA
.
*Correspondence to: Roman Shchepin, Vanderbilt University Institute of Imaging
Science, 1161 21st Ave South AA-1105, Nashville, TN 37232, USA.
E-mail: roman.shchepin@vanderbilt.edu
The subsecond chemical reaction is capable of preserving the
parahydrogen spin order long enough for subsequent
polarization transfer to the long-lived 13C and 15N spin sites of
the hyperpolarized contrast agent. However, the requirement **Correspondence to: Eduard Y. Chekmenev, Department of Biomedical
Engineering and Biochemistry, Vanderbilt-Ingram Cancer Center, Vanderbilt
University, Nashville, TN, USA
E-mail: eduard.chekmenev@vanderbilt.edu
of having a custom isotopically labeled molecular precursor with
C=C or C≡C bond typically adjacent to the 13C17,18 or 15N19 site
severely restricts the availability of biomolecules of interest capable
†Additional supporting information may be found in the online version of this
of reporting altered metabolism in living organisms. Nevertheless,
the first generation of in vivo metabolic contrast agents featured article at the publisher’s web-site.
J. Label Compd. Radiopharm 2014, 57 517–524
Copyright © 2014 John Wiley & Sons, Ltd.

R. V. Shchepin et al.
High-resolution 13C{1H} NMR of samples derived from the
serum, kidneys, livers, tumors, brains, and hearts of tumor-
bearing mice was used to determine the fate of IV injected PLAC.
1-13C-phospholactate is of significant interest because it has
already been successfully hyperpolarized25 to ≥15% of 13C nuclear
spin polarization (corresponding to ≥60,000-fold signal
enhancement at 3.0T) with in vitro T1 of 58 s36, and 1-13C-lactate
(LAC), the daughter molecule after PLAC in vivo
dephosphorylation, can be used for molecular imaging of elevated
glycolysis in cancer37,38. Injections of hyperpolarized LAC have
recently been reported as useful metabolic contrast agent to
noninvasively study metabolism of tumor and heart37,39–41, as well
as report of the treatment efficacy40. Specifically, the metabolic
conversion of LAC into 1-13C-pyruvate, 1-13C-alanine, and
13C-bicarbonate is useful for molecular imaging in vivo.
Experimental
General
All solvents were purchased from common vendors and were used as
received. HPLC grade water was used in the synthesis and buffer solution
preparation. High-resolution 13C NMR spectra were recorded using
Bruker 600-MHz NMR spectrometer equipped with cryogenic radio
frequency probe. All 13C NMR spectra were referenced externally to
dimethyl sulfoxide (39.5 ppm).
Figure 1. The concept of preparation and in vivo use of procontrast agent using
phosphate protection group: (a) overall approach (b) demonstration using DL-1-13C-
phospholactate (PLAC) followed by its in vivo activation to become 1-13C-lactate (LAC)
and subsequent uptake and in vivo metabolism. (c) Synthetic scheme for preparation
of PLAC from 1-13C-pyruvic acid. This figure is available in colour online at
wileyonlinelibrary.com/journal/jlcr.
Synthesis of 1-13C-phospholactate
13C-phosphoenolpyruvate was prepared according to previously
published procedures42,43
.
Briefly, 1-13C-pyruvic acid (1.00 g,
11.2 mmol) was dissolved in CCl4 (20 mL) at room temperature.
Bromine (0.58 mL, 11.2 mmol) and one drop of 37% HCl were added.
Upon color disappearance after approximately 12 h, the solvent was
phosphates as by-products of the dephosphorylation reaction, this
concept can lead to the next generation of in vivo PHIP
hyperpolarized contrast agents, including 1-13C-phospholactate25 evaporated, and the residue was redissolved in CCl4 (10 mL) and
evaporated again. Crude 13C-bromopyruvic acid was used without
further purification. Trimethyl phosphite, P(OMe)3 (1.58 mL,
(PLAC) demonstrated previously as well as the potential feasibility
of 15N-phosphocholine27 and others.
11.2 mmol), was dissolved in THF (5 mL) and cooled to 0°C, and the
The underlying hypothesis of this work is that the phosphate-
solution of 13C-bromopyruvic acid was added dropwise over a period
protecting group of small molecules can be removed by blood
of 10 min. The reaction mixture was stirred for 12 h. Then, the solvent
phosphatases on the timescale compatible with the
was evaporated in vacuo, and the mixture was redissolved in water
hyperpolarization lifetime of hyperpolarized contrast agents. We
(20 mL) and let to stay at room temperature for 12 h before pH was
are interested specifically in testing this notion in breast cancer as
part of our overall approach to using hyperpolarized imaging to
assess tumor metabolic state in order to monitor treatment
adjusted to 2.8 with KOH (3 M) solution. The solution was filtered
and evaporated under reduced pressure. The suspension was
recrystallized twice from water (5 mL)–ethanol (60 mL) to provide a
response in a preclinical mouse model. One of the tumor antigens product (1.30 g), which contained 87% 1-13C-phosphoenolpyruvate
and 13% PLAC with combined yield of 69% with respect to starting
tested most frequently in clinical trials is mucin 1 (MUC1)28,29
.
material 1-13C-pyruvic acid.
MUC1 is a large transmembrane glycoprotein, which becomes
underglycosyated, loses its polarity, and is overexpressed in more
than 90% of human breast cancer30–33. In this mouse model, we
induced MUC1-transfected murine breast cancer cells C57MG, and
then, the cells were used to develop a xenograft tumor on MUC1
transgenic mice34. These mice were used to study the metabolic fate
of PLAC after its intravenous (IV) administration. Study in the past
from our group demonstrated that this animal model is a tumor-
tolerant model suitable for breast cancer research34. It has been
demonstrated that MUC1 modulates cancer cell metabolism
and facilitates cancer cell growth. Notably, MUC1
overexpression enhances glycolytic activity in epithelial-based
cancer35. The latter can be potentially imaged by hyper-
polarized contrast agents involved in glycolysis, and this was
the rational for using this model of cancer described here in
feasibility studies using PLAC (Figure 1c). In addition, MUC1 acts
as a modulator of hypoxic response in cancer cells by regulating
the expression of hypoxia-inducible factor.
13
The C-phosphoenolpyruvate42,43 (0.20 g, 1.2 mmol) was dissolved in
water (50 mL) and Pd/C (10% wet concentration, 0.12 g). After
hydrogenation at 1 atm for 3 h using hydrogen bag, the liquid was
filtered and concentrated in vacuo to provide a quantitative yield of
PLAC. Note that measured weight of 0.22 g was probably because of
the presence of some water; 0.202 g was used for yield calculations of
100%. We note that all 13C isotopic material of PLAC was prepared and
used as a mixture of D-form and L-form.
The 1-13C-phospholactate prepared as described above was
suspended in 1× PBS buffer (3 mL); then, the pH was adjusted to 7.4 by
dropwise addition of NaOH solution (1 N) with an aid of a pH meter. Final
volume was brought to 5 mL by addition of water.
Animal model of cancer
Mucin 1-transfected murine breast cancer cell lines C57MG were cultured
and maintained in Dulbecco Modified Eagle Medium (Mediatech,
Manassas, VA) in the presence of 10% fetal calf serum (Invitrogen,
Carlsbad, CA), penicillin–streptomycin antibiotics (Mediatech), and
www.jlcr.org
Copyright © 2014 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2014, 57 517–524

R. V. Shchepin et al.
10 μg/mL insulin (Sigma-Aldrich, St Louis, MO) at 37°C and 5% CO2
incubator. A colony of MUC1 transgenic mice was maintained by crossing
MUC1 transgenic mice obtained from Sandra Gendler (Mayo Clinic) with
wild-type C57BL/6 strain (Jackson Laboratory, Bar Harbor, ME). The mice were
genotyped by a standard polymerase chain reaction using DNA isolated from
tail tips with the following primers: forward, 5′-CTTGCCAGCCATAGCACCAAG-
3′; reverse, 5′-CTCCACGTCGTGGACATTGATG-3′. After polymerase chain
reaction amplification, the DNA product of each reaction was analyzed by size
fractionation through a 1% agarose gel. The size of DNA product form MUC1-
positive mice corresponded with a 500-bp fragment. MUC1 transgenic mice
were maintained as hemizygous animals. All animal experiments were carried
out in accordance with the guidelines provided by Vanderbilt University’s
Institutional Animal Care and Use Committee.
a)
Serum
PLAC
Time after injection (minutes)
control 0.6 min
1 min
PLAC PLAC
LAC LAC
3 min
5 min
LAC
LAC
LAC
PLAC
Mouse model and biodistribution studies
One million MUC1-transfected C57MG cells in 100-μL serum-free DMEM
medium were injected subcutaneously into the mammary fat pad as
described previously44. Approximately, 2 weeks after the inoculation, small
palpable tumors of approximately 0.4cm were confirmed at the injections
site, and this initiated the biodistribution and metabolite study. The tail
veins of awake or isoflurane-anesthetized tumor-bearing mice were
injected with PLAC compound (5mg, 100 μL) buffered to physiological pH.
Experiments were conducted to study temporal dependence of PLAC
dephosphorylation to LAC in the serum and liver (Figure 2) of tumor-
bearing mice (n = 5), while the control animals were injected with saline.
To compare the effect of anesthesia interference with metabolism, all
animals in this group were injected intravenously with the compounds
while still awake. NMR samples, such as the serum and liver, were
prepared and normalized across the cohort at 0.6, 1, 3, and 5 min after
injection for PLAC-treated and at 0.6 min for control animals. While
sample volume and the number of 13C spectroscopic acquisitions varied,
C1 13C resonances of glucose were used as internal concentration
reference to quantitate PLAC concentration in serum.
183 182
13C Chemical Shift (ppm)
b)
Liver
C2,C3,C4,C5-glc
C6-glc
C1-glc
C1-glygogen
C2-LAC
5 min
3 min
1 min
0.6 min
The second cohort of tumor-bearing mice was subjected to
biodistribution experiments (n = 6) of a more rigorous nature, and this
cohort utilized isoflurane as described in the preceding texts. The animals
were divided into three groups for IV injection under anesthesia using
isoflurane: (i) with buffered solution of PLAC (5 mg per animal, n = 2);
(ii) others were injected with buffered LAC (5 mg per animal, n = 2); and
(iii) control mice were injected with the same volume of buffer solution
(n = 2). Eighty seconds after injection, the animals were euthanized, and
serum and peripheral organ tissues were harvested, normalized, and
prepared for NMR analysis. The normalization procedure included
scaling NMR spectral intensities to adjust for the weight of the
collected tissues, which varied, (Figures S1–S6). NMR spectra of
serum were not normalized, Figure S1, because PLAC or LAC peaks
had very high intensity with respect to metabolites found in serum
of control animals. The spectra presented in Figures S1–S6 were used
for analysis of any extra metabolites and for analysis of partial 13C
enrichment of LAC-C1 carbon.
0.6 min control
100
95
90
85
80
75
70
65
60
13C Chemical Shift (ppm)
Figure 2. Time course of 1-13C-phospholactate (PLAC) metabolism in (a) serum
and (b) liver after tail vein injection of 5 mg of sodium PLAC in 150 mL of PBS buffer
at pH = 7.4. Individual tumor-bearing mouse was used for every data point
corresponding to 13C{1H} high-resolution spectrum recorded using serum sample
of animal or (b) the entire liver sample of a mouse. Note the appearance of extra
resonances in panel (b) because of 13C enrichment from 1-13C-lactate (LAC)
entering the liver after dephosphorylation in blood. Animals used for this substudy
were not anesthetized. No PLAC was detected in the liver. This figure is available in
colour online at wileyonlinelibrary.com/journal/jlcr
Results
For all animals in both cohorts, serum was collected and normalized
based on the weight of the animals after centrifugation of the blood
sample. For the organs, the fresh tissues were weighed and normalized
across the animals after which the samples were homogenized and
centrifuged. The supernatant fractions were collected in NMR tubes for
analysis. Notably, all ex vivo steps were performed under temperature
conditions of 4°C to reduce tissue decomposition. To facilitate NMR
experiments, 0.15 mL of 99.8% deuterium oxide was added to every
NMR sample to enable deuterium lock. 13C spectra were also
collected using 256–512 scans for serum, 3072 scans for tumor, and
512–1024 scans for the remainder of the tissues with a repetition
time of approximately 12 s. It should be noted that the same number
of scans was used for the spectra corresponding to the same type of
tissue/organ in Figures S1–S6 with an exception for serum, where
one spectrum was acquired with 650 scans versus 512 scans. For
example, all 13C{1H} spectra acquired from liver samples utilized
1024 signal averages.
Dephosphorylation
As described in the Materials and Methods, 5 mg of PLAC was
injected into the tail vein of a tumor-bearing mouse, and
biodistribution of 13C label was monitored by ex vivo 13C high-
resolution NMR spectroscopy. The samples of serum collected
at different time points after IV injection of PLAC (Figure 2a)
revealed rapid conversion of the injected 13C substrate into
LAC. A likely explanation for this dephosphorylation catabolic
step (Figure 1) is action of blood phosphatases. Remarkably,
most PLAC is converted to LAC in the vasculature within 1 min
(Figure 2a), despite supraphysiological level of ~270 mM of
injected PLAC resulting in an estimated 27-mM PLAC
concentration in blood vasculature after dilution of the
administered 5-mg dose. The 13C spectral intensities of PLAC
J. Label Compd. Radiopharm 2014, 57 517–524
Copyright © 2014 John Wiley & Sons, Ltd.
www.jlcr.org

R. V. Shchepin et al.
(Figure 2) and C1 resonances of serum glucose were used to
quantitate PLAC concentration in serum sample. Serum glucose
concentration (5.5 mM), 1.1% 13C natural abundance, and
relative intensities of the previously mentioned PLAC and
glucose (internal concentration reference) resonances that yield
1.8, 1.3, 0.76, and 0.54 mM for time points corresponding to
spectra acquired serum samples derived from mice after 0.6-,
1-, 3-, and 5-min PLAC injection in the tail vein, Figure 2a inset.
The PLAC concentration values have 20% error bar of their
value because of physiological variations in blood glucose as
well as the blood volume of individual animals. Moreover, it is
likely that the dephosphorylation is not likely immediately
quenched after animal termination, and some PLAC could be
dephosphorylating in the process of serum collection. The latter
is a clear limitation of this approach. Nevertheless, it provides a
good estimate of PLAC in vivo deprotection. The exponential model
of PLAC concentration decay (red curve on inset of Figure 2a)
provides an estimate of PLAC concentration decay with time
constant of 11 2 s, which is less than T1 of 13C1 of in vitro PLAC
(36 s) and in vivo LAC (>40 s)39.
Uptake and liver metabolism
Tumors have an elevated rate of glycolysis with the end product
of lactate being released from cancer cells. As a result, the livers
of these animals are metabolically predisposed to rapid uptake
and conversion of lactate into glucose and glycogen.
When NMR samples derived from the livers of the same
animals were analyzed, extra 13C NMR resonances were detected
compared with samples derived from the control experiments
without PLAC (Figure 2b). These extra NMR resonances
correspond to extra species being enriched using PLAC as a
source of 13C spin label. This is not surprising, as LAC entering
the gluconeogenesis pathway enriches some carbons and
eventually the third and fourth positions in glucose (Figure 3).
It should be noted that quantitative assessments and
assignments are rather challenging because of the dynamic
nature of the process, partial enrichment, multiple metabolic
intermediates, lack of internal reference in the same animal,
Figure 3. 13C enrichment of gluconeogenesis pathway in the liver from 1-13C-
lactate (LAC). Note that only the third and fourth 13C positions of glucose can be
13C enriched as well as corresponding relevant intermediates. This figure is
available in colour online at wileyonlinelibrary.com/journal/jlcr
and T1 relaxation effects. While metabolic conversion of PLAC 71.6 ppm. We also note the additional splittings for glucose C1
and LAC was not the main goal of this study, we report the resonances (Figure 2b). The resonance at 100.2 ppm corresponds
chemical shift of detected metabolic intermediates.
to enrichment of C1 of glycogen. The latter is consistent with
The extra 13C resonances appear because of 13C enrichment extra splittings detected for glycogen C1. We note that the extra
from LAC, while extra splittings of some NMR peaks in the resonance peak for glycogen C1 is likely because of additional
spectra are attributed to homonuclear 13C–13C spin–spin 13C–13C spin–spin interactions with other enriched carbons
couplings, which cannot be removed by proton decoupling. (most likely 13C1–13C4 because glucoses are linked via 1–4
For example, when no 13C labeling is used, 1.1% 13C natural glycosidic bonds and 13C enrichment is expected for C4 carbon
abundance results in a 2.2% probability that at least one of the from C1 of LAC entering liver) because direct enrichment of C1
two neighboring carbons of 13C2 is indeed 13C (1.1% probability of glucose and glycogen is unlikely from 13C source of LAC
in position C1 and 1.1% probability in position C3). This is (Figure 3). The other two extra resonances at 77.2 and
negligible with respect to the level of detected signal-to-noise 71.6 ppm found in a highly populated C2–C6 region are likely
ratio (SNR) and cannot be readily seen. When a significant C3 and C4 carbons of glucose, respectively, based on earlier
quantity of 13C label enters gluconeogenesis, this probability of assignments by Kalderon and coworkers in U-13C-glucose
13C–13C pair significantly increases well above 2.2% and results in vivo studies45. The observation of extra resonances in 13C
in the appearance of extra splittings of resonances. Note that spectra in liver samples proves that 13C-labeled compound
the new resonances appear because of 13C enrichment, while indeed enters the most likely pathway; most of these extra
extra resonance splittings are the result of 13C–13C spin–spin resonances are detected only at the 1-min mark, which indicates
couplings because of increased probability of adjacent 13C–13C that 13C spin label enters as bolus, and once processed, the
pair. We note the complex nature of these splitting/resonance 13C-glucose end product leaves the liver. The timing of this event
patterns because of the potentially complex network of spin– correlates very well with the values reported previously by Exton
spin couplings and multiple intermediates. The extra series/ and Park in their pioneering studies of gluconeogenesis using
multiplets of resonances are detected at 100.2, 77.2, and radioactive tracers46.
www.jlcr.org
Copyright © 2014 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2014, 57 517–524

R. V. Shchepin et al.
Encouraged by the studies with a pilot cohort of animals, a cleavage of the protecting group, and tumor-bearing mice were
series of samples were derived from the liver, kidneys, heart, injected with PLAC that contained both D-form and L-form. While
tumor, serum, and brain of anesthetized tumor-bearing mice it was difficult to differentiate the rates of utilization of D-form
after 80 s of metabolism following the injection of 5 mg of either and L-form using the injection of a mixture of D-form and L-form
PLAC, LAC, or buffer (Figures S1–S6). The assignment of relevant of PLAC because their chemical shifts could not be distinguished,
13C resonances used is provided in Figures S1–S6. No PLAC was we note that both forms were cleaved to LAC in the vasculature
detected anywhere except in the samples derived from serum (Figure 2a) and processed (Figure 2a) and other data not shown.
in all anesthetized and nonanesthetized animals injected with This is an important observation from the perspective of the
PLAC. This important observation suggests the potentially useful future use of PLAC as a metabolic contrast agent. While both
in vivo applications of PLAC in hyperpolarized-based imaging forms can be metabolized, it could be speculated that different
(Figure 4). Based on the data obtained, we think that PLAC is forms have different enzymatic rates of conversion and further
converted into LAC in the vasculature after its IV injection. The metabolism.
LAC is then taken up by tissues and organs and is then further
catabolized to 13C-glucose and other metabolites (Figures S1–
Metabolism in tumors
S6). Next, we wanted to test the specificity of in vivo-induced
The analysis of 13C spectra acquired from samples of liver,
kidneys, tumors, serum, brain and heart (Figures S1–S6) revealed
no extra 13C resonances except for liver (as discussed previously),
and tumors, thus suggesting that metabolic events are limited to
13C-lactate enrichment in position C1 in most organs except for
tumors and liver on the timescale tested. Extra resonances were
detected in samples derived from animals that received both
PLAC and LAC injection (Figure S3); however, these extra
resonances had a very low SNR and were likely suppressed in
the noise in some instances. Therefore, it is only possible to state
qualitatively that taken up LAC undergoes some metabolic
conversion. Two potential explanations may be offered for why
SNR was low: (i) lack or low level of metabolism and (ii) the
metabolic snapshot at that particular time missed the metabolic
event or/and the maximum of LAC bolus entering tumors. We
consider the second explanation to be more likely because of
similar bolus signatures observed in samples derived from livers
(Figure 2b). Nevertheless, these are challenging experiments and
provide only a single time point from one animal. Future
imaging experiments with hyperpolarized PLAC should yield
substantially more insight into the dynamics of the contrast
agent with sufficient spatial and temporal resolution, thus
providing further data with which to refine the initial simplistic
model developed here (Figure 4).
13C enrichment of lactate
The injected PLAC and LAC result in the increase of 13C
enrichment of C1 carbon of lactate in blood and other tissues
and organs. Unlike other metabolic agents, such as pyruvate
and succinate, lactate in vivo concentration is substantial and
can reach several millimolar levels. Therefore, it is possible to
achieve only partial enrichment of C1-lactate. Nevertheless, it
can be used as a quantitative metrics of metabolic penetration.
Figure 5 summarizes 13C enrichment of C1-lactate in samples
derived from serum, liver, brain, tumor, kidneys, and brain after
injection of 5 mg of LAC, PLAC, and blank buffer (control).
Theoretically, all three carbons of lactate should result in the
same NMR peak integral intensities under conditions of their
equal enrichment and scan repetition time much longer than
spin lattice relaxation time T1 of any of the three carbons. This
is not practical because it would require much longer acquisition
time, orders of magnitude more than 12-s repetition time used
here. 13C T1 relaxation of three lactate carbons is different and
results in different integral intensities of these carbons with
respect to each other. As a result of this limitation, a surrogate
metrics of Esurr = 2•IC1/(IC2 + IC3) was used, where IC1, IC2, and IC3
Figure 4. Metabolism of intravenously injected 1-13C-phospholactate (PLAC). The
steps of enzymatic conversion to 1-13C-lactate (LAC) in vasculature and subsequent
update by organs and further metabolism are shown. This figure is available in
colour online at wileyonlinelibrary.com/journal/jlcr
J. Label Compd. Radiopharm 2014, 57 517–524
Copyright © 2014 John Wiley & Sons, Ltd.
www.jlcr.org

R. V. Shchepin et al.
a) Liver
d) Serum
6.0
heart, kidneys, and serum. The elevation in Esurr in the heart
was the most striking, up to 1.5 for LAC and >7 for PLAC (Figure
5f). It is somewhat surprising that Esurr of PLAC was much higher
than that of LAC injection because PLAC needs to be converted
to LAC first. There are two potential explanations. First, the peak
of LAC in the heart happens much earlier than that in case of
PLAC injection and was consequently missed at 80 s after
injection. Second, because it takes time for PLAC to be released
to LAC form, it circulates longer in vasculature making the tracer
available longer for its uptake by heart. There was no significant
differentiation between PLAC and LAC in Esurr values of in other
sites (Figure 5). The smallest changes in Esurr with respect to
controls were detected in brain and tumors. Nevertheless, the
elevation Esurr is well above the control values and serves as a
proof that injected PLAC and LAC reach those sites by 80 s
following their IV administration in tumor-bearing mice.
Relatively low value in Esurr in LAC and PLAC injections could
also be associated with the fact that LAC in the liver is rapidly
processed via gluconeogenesis.
3
4.5
3.0
1.5
0
2
1
0
Control LAC
PLAC
Control LAC
PLAC
b) Kidneys
8
e) Brain
2.0
1.5
1.0
0.5
0
6
4
2
0
Discussion
13C hyperpolarization of 1-13C-phospholactate: %
Control LAC
PLAC
Control LAC
PLAC
c) Tumor
1.6
f) Heart
polarization and T1
8
6
4
2
0
We present here the initial in vivo evaluation of PLAC for its
potential use as a metabolic contrast agent. PLAC can be
hyperpolarized to 1% using PHIP25. Despite significant sensitivity
gains that can be obtained with 1% hyperpolarized PLAC, the
studies involving hyperpolarized DNP hyperpolarized LAC
required much higher sensitivity gains because the exchange
of hyperpolarized LAC pool into metabolites, for example,
pyruvate, alanine, and bicarbonate, is significantly smaller than
that of pyruvate37,40,47. While the metabolic imaging sensitivity
gains can be improved via indirect proton imaging of
1.2
0.8
0.4
0
Control LAC
PLAC
Control LAC
PLAC
Figure 5. Fractional 13C enrichment (vertical axis) of C1 carbon of LAC measured
using a surrogate metrics of 2•IC1/(IC2 + IC3), where IC1, IC2, and IC3 are the integrated
peak intensities of corresponding carbons. Analysis of 13C{1H} high-resolution NMR
spectra by measuring 13C enrichment of C1 site of lactate is provided in six
compartments (a–e) for three groups of animals injected (as labeled in figure displays)
with PLAC, LAC, or blank buffer (control). The values of 2•IC1/(IC2 + IC3) for each animal
sample are shown in corresponding two animals per group, except for kidneys where
only one control sample was available. Note the absence of 13C enrichment in results
in 2•IC1/(IC2 + IC3) < 1 as a result of different spin lattice relaxation time T1 of the three
sites resulting in lower than anticipated intensity of C1 peak as a result of relatively
short repetition time of ~12 s. Animals used for this substudy were anesthetized. This
figure is available in colour online at wileyonlinelibrary.com/journal/jlcr
hyperpolarized contrast agent48,
a more straightforward
approach is the use of significantly higher polarized LAC or PLAC.
For example, approximately 20% polarized LAC was prepared by
DNP technique, which allowed for metabolic imaging37,40,47. A
comparable % polarization for hyperpolarized PLAC is likely
needed to ensure the success of in vivo studies using PHIP
polarizing technology. The percentage of hyperpolarization can
potentially be improved significantly using deuterated molecular
PHIP precursor similarly to the succinate developed
previously20,21. For example, % hyperpolarization was less
than 2%20 for nondeuterated 1-13C-succinate despite addition
are the integrated peak intensities of corresponding carbons. of two parahydrogen pairs, while 20%21 1-13C-succinate
Note that absence of 13C enrichment results in 2•IC1/(IC2 + IC3) < 1 polarization was successfully demonstrated for a fully deuterated
because of different spin lattice relaxation time T1 of the three molecular precursor using one parahydrogen pair. Most
sites resulting in lower than anticipated intensity of C1 peak successfully PHIP molecule of 2-hydroxyethyl propionate-1-13C-
because of relatively short repetition time of ~12 s. All animals d2,3,3, where neighboring to 1-13C protons were fully deuterated,
used for this second cohort were anesthetized. The error bars was polarized to 20–50% by several groups. We point that
(Figure 5) were calculated as 2•Esurr/SNRC1, where SNRC1 is SNR nondeuterated version of this molecule yields much less than
of C1 carbon. In cases, when Esurr > 2, the error of measurement 1% polarization based on our experience. Therefore, we estimate
was no longer limited by the intensity of SNR of C1 carbon and that fully deuterated % 13С hyperpolarization of PLAC can be
was estimated (not shown in Figure 5) to be 10% of Esurr value. improved by 10- to50-fold based on the reported % polarization
Esurr values for all controls (Figures 5a–5f) were in the range of values for 1-13C-succinate and 2-hydroxyethyl propionate-1-13C.36
0.4 to 0.7 reflecting the variability because of low SNR. Figure 5
Moreover, deuteration of the precursor for PHIP
provides a snapshot of Esurr after 80 s following the IV injection. hyperpolarization was shown to significantly improve the
Esurr values were elevated above the control following the lifetime of 13C hyperpolarization in succinate from T1 of 6 s20 to
injection of LAC and PLAC in nearly all animals in almost all sites T1 of 28 s21 in water. It can be speculated that deuteration of
tested. The most significant differences were observed in the phosphoenolpyruvate (PEP) precursor can also increase previously
www.jlcr.org
Copyright © 2014 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2014, 57 517–524

R. V. Shchepin et al.
reported 13C T1 of 36s measured at 0.0475 T and 13C T1 of 18s administration of hyperpolarized LAC. Moreover, the 13C
measured at 11.7 T25. It should be noted that the T1 gains are enrichment of the C1 carbon of lactate in the liver, heart, tumors,
expected to be longer with deuteration at low field compared with brain, and kidneys suggests that PLAC might also be a useful
those at higher magnetic fields, where large 13C chemical shift contrast agent in its hyperpolarized form because the 13C label
anisotropy rather than dipole–dipole interactions with protons derived from PLAC reaches these sites within 80 s (Figure 5).
is dominant relaxation mechanism. For example, 13C T1 of Therefore, hyperpolarized PLAC has the potential to become
2-hydroxyethyl propionate-1-13C-d2,3,3 was 106 s49 at 0.0475 T a suitable contrast agent from the perspective of its lifetime
compared with 50 s14 at 4.7 T. Relevant to LAC, deuteration and delivery.
had a negligible effect on 1-13C T1 of ~44 s measured at 9.4 T
because these measurements are conducted at high magnetic
field, where 13C chemical shift anisotropy dominates the
In vivo delivery mechanism, limitations, and prospects
relaxation mechanism50
.
Most of the studies presented here were conducted as snapshots
at 80 s following the administration of PLAC or LAC. Notably,
injections of PLAC and LAC produced a similar enrichment of
the C1-lactate carbon in blood, vasculature, brain, tumors, liver,
and kidneys with the exception of the heart, where partial
enrichment of C1-lactate measured as Esurr after injection of
PLAC was significantly greater than that of LAC. The static nature
of the experiments performed demonstrates a clear limitation of
the nonhyperpolarized agent, such that future imaging studies
conducted with hyperpolarized PLAC should provide
significantly more information about the spatial biodistribution
of polarized labels and, more importantly, multiple time
points reporting on kinetics in the same animal. This can be
performed using magnetic resonance spectroscopic imaging8
sequences as well as FIESTA51 imaging sequences. While it is
difficult to speculate on whether hyperpolarized PLAC will
become more or less useful than hyperpolarized LAC, we
point out that their delivery mechanisms are somewhat
different. Further, the results from partial enrichment in
the heart (Figure 5) enable us to speculate that a PLAC
delivery mechanism could offer more advantages than
nonphosphorylated methods. At the same time, the rationale
for the administration of a procontrast agent is not new, as it
has been used previously for succinate and pyruvate using
PHIP hyperpolarized diethyl succinate22 and dissolution DNP
polarized ethyl pyruvate52. However, the latter agents were
developed with the goal of delivering the protected form of
a contrast agent into the site of interest, such as a tumor
and brain, respectively. Here, the phosphorylation of the
potential contrast agent prevents its uptake by the organs
and tissues studied. Instead, the pro-agent must first be
activated (dephosphorylated) in blood vasculature before it
can be metabolized. From that perspective, this certainly
represents a new concept for the in vivo delivery of contrast
agents, particularly hyperpolarized imaging agents, which can
likely be extended to other PHIP hyperpolarized contrast
agents in the future.
1-13C-lactate 13C T1 and in vivo metabolism
1-13C-lactate is the immediate product of dephosphorylation,
which enters organs. Therefore, in vivo 13C T1 of lactate is a
determining factor for detection of in vivo metabolism. The
reported in vivo values of 13C T1 for DNP hyperpolarized LAC
range from 39 s at 3 T40 to 44 s at 339 and 9.4 T37, which is
sufficiently long for detection of in vivo metabolism of cancer39
and in the heart40.
Although in vivo administration of hyperpolarized LAC leads to
production of hyperpolarized pyruvate, alanine, and bicarbonate
39,40, which is expected here on the timescale of 1–3 min, the
relative concentration of these products is significantly smaller
than that of in vivo LAC. As a result of relatively low SNR of
ex vivo 13C{1H} spectra, these metabolites could not be detected
in the study presented here.
13C chemical shift dispersion
However, the chemical shift difference of 1-13C carbon of lactate
and phospholactate is less than 1 ppm, which is rather small and
can be difficult to resolve using MR spectroscopic imaging
approaches in vivo8. Moreover, subsecond fast imaging
employing steady state acquisition (FIESTA), demonstrated
successfully with hyperpolarized 1-13C-succinate imaging, would
lack chemical shift specificity20. A priori knowledge of the
biodistribution of IV injected PLAC, which is studied here, is
therefore important.
Biodistribution
The nonhyperpolarized form of this agent was used to study
the hypothesis that phosphorylated agents can be dephos-
phorylated in the vasculature to convert a pro-agent into an active
form of a metabolic agent that can be taken up and processed by
organs and tissues. Here, the hypothesis25 is indeed successfully
validated as exemplified by PLAC. The inset of Figure 2a
demonstrated PLAC dephosphorylation monitored as the decay of
PLAC concentration in serum samples. The in vivo concentration at
the time point at 0s using 5 mg dose of injected PLAC was calculated
Conclusion
as the homogeneous distribution of 5 mg of injected PLAC over an A simplistic model of the proposed delivery of PLAC and
estimated 0.5 mL of blood pool allowing for 20% physiological metabolic utilization in vivo is developed (Figure 4), using
variation in the estimated blood pool. Moreover, the technique the results available. This model will likely to be refined by
offers the potential for application to other hyperpolarized contrast future imaging studies using hyperpolarized PLAC.
agents including 15N-phosphocholine. The latter can be polarized Nevertheless, the results presented here make it clear that
by dissolution DNP and potentially by PHIP19 using a recently hyperpolarized PLAC is a promising new contrast agent that
developed synthetic approach for preparation of corresponding warrants future preclinical validation in animals. While the
molecular precursors for PHIP hyperpolarization27.
When intravenously administered, PLAC is converted to LAC results presented here suggest that PLAC may also be useful
within the vasculature on timescale suitable for the for cardiac imaging.
past work has been largely focused on cancer imaging, the
a
J. Label Compd. Radiopharm 2014, 57 517–524
Copyright © 2014 John Wiley & Sons, Ltd.
www.jlcr.org

R. V. Shchepin et al.
[23] E. Y. Chekmenev, S. K. Chow, D. Tofan, D. P. Weitekamp, B. D. Ross,
P. Bhattacharya, J. Phys. Chem. B 2008, 112, 6285.
[24] P. Bhattacharya, E. Y. Chekmenev, W. F. Reynolds, S. Wagner,
N. Zacharias, H. R. Chan, R. Bünger, B. D. Ross, NMR Biomed.
2011, 24, 1023.
[25] R. V. Shchepin, A. M. Coffey, K. W. Waddell, E. Y. Chekmenev, J. Am.
Chem. Soc. 2012, 134, 3957–3960.
[26] T. Trantzschel, J. Bernarding, M. Plaumann, D. Lego, T. Gutmann,
T. Ratajczyk, S. Dillenberger, G. Buntkowsky, J. Bargon, U.
Bommerich, Phys. Chem. Chem. Phys. 2012, 14, 5601.
[27] R. V. Shchepin, E. Y. Chekmenev, J. Labelled Comp. Radiopharm.
2013, 56, 655.
Acknowledgements
We thank Dr. Panayiotis Nikolaou for preparation of graphics
elements in Figure 4. We thank the following funding sources:
NIH 5R00 CA134749-03 (E.Y.C.), 3R00CA134749-02S1 (E.Y.C.), DoD
CDMRP Breast Cancer Program Era of Hope Award W81XWH-12-
1-0159/BC112431 (E.Y.C.), and NIH/NCI R01CA160700 (W.P.). We
would also like to acknowledge the NIH grant award number
S10 RR019022 for supplying funding to purchase high-resolution
600-MHz NMR spectrometer.
[28] J. S. Goydos, E. Elder, T. L. Whiteside, O. J. Finn, M. T. Lotze, J. Surg.
Res. 1996, 63, 298.
Conflict of Interest
[29] M. A. Reddish, G. D. MacLean, R. R. Koganty, J. Kan-Mitchell, V. Jones,
M. S. Mitchell, B. M. Longenecker, Int. J. Cancer 1998, 76, 817.
[30] D. F. Hayes, R. Mesatejada, L. D. Papsidero, G. A. Croghan, A. H. Korzun,
L. Norton, W. Wood, J. A. Strauchen, M. Grimes, R. B. Weiss, H. J. Ree, A.
D. Thor, F. C. Koerner, M. A. Rice, M. Barcos, D. W. Kufe, J. Clin. Oncol.
1991, 9, 1113.
[31] L. Perey, D. F. Hayes, P. Maimonis, M. Abe, C. Ohara, D. W. Kufe,
Cancer Res. 1992, 52, 2563.
[32] D. Avichezer, J. Taylor-Papadimitriou, R. Arnon, Cancer Biochem.
Biophys. 1998, 16, 113.
[33] M. Nacht, A. T. Ferguson, W. Zhang, J. M. Petroziello, B. P. Cook, Y. H.
Gao, S. Maguire, D. Riley, G. Coppola, G. M. Landes, S. L. Madden, S.
Sukumar, Cancer Res. 1999, 59, 5464.
[34] S. Kobukai, G.-J. Kremers, J. G. Cobb, R. Baheza, J. Xie, A. Kuley, M.
Zhu, W. Pham, Transl. Oncol. 2011, 4, 1.
[35] S. Czernecki, J.-M. Valery, Synthesis 1991, 1991, 239.
[36] R. V. Shchepin, A. M. Coffey, K. W. Waddell, E. Y. Chekmenev, Anal.
Chem. 2014, DOI: 10.1021/ac500952z
[37] B. W. C. Kennedy, M. I. Kettunen, D.-E. Hu, K. M. Brindle, J. Am. Chem.
Soc. 2012, 134, 4969.
[38] A. Kaech, M. Hofer, D. Rentsch, C. Schnider, T. Egli, Biodegradation
2005, 16, 461.
[39] A. P. Chen, J. Kurhanewicz, R. Bok, D. Xua, D. Joun, V. Zhang, S. J.
Nelson, R. E. Hurd, D. B. Vigneron, Magn. Reson. Imaging 2008, 26,
721.
[40] S. Bluml, A. Moreno-Torres, F. Shic, C. H. Nguy, B. D. Ross, NMR
Biomed. 2002, 15, 1.
[41] R. Huang, G. Chen, M. Sun, C. Gao, Carbohydr. Res. 2006, 341, 2777.
[42] B. L. Hirschbein, F. P. Mazenod, G. M. Whitesides, J. Org. Chem. 1982,
47, 3765.
[43] G. D. Dotson, R. K. Dua, J. C. Clemens, E. W. Wooten, R. W. Woodard,
J. Biol. Chem. 1995, 270, 13698.
[44] W. Pham, Y. D. Choi, R. Weissleder, C. H. Tung, Bioconjugate Chem.
2004, 15, 1403.
[45] B. Kalderon, S. H. Korman, A. Gutman, A. Lapidot, Proc. Natl. Acad. Sci.
U. S. A. 1989, 86, 4690.
[46] J. H. Exton, C. R. Park, J. Biol. Chem. 1967, 242, 2622.
[47] R. E. Hurd, D. Spielman, S. Josan, Y.-F. Yen, A. Pfefferbaum, D. Mayer,
Magn. Reson. Med. 2012, doi: 10.1002/mrm.24544.
[48] M. Mishkovsky, T. Cheng, A. Comment, R. Gruetter, Magn. Reson.
Med. 2012, 68, 349.
The authors did not report any conflict of interest.
References
[1] A. Abragam, M. Goldman, Rep. Prog. Phys. 1978, 41, 395.
[2] J. H. Ardenkjaer-Larsen, B. Fridlund, A. Gram, G. Hansson, L. Hansson,
M. H. Lerche, R. Servin, M. Thaning, K. Golman, Proc. Natl. Acad. Sci.
U. S. A. 2003, 100, 10158.
[3] M. Goldman, H. Johannesson, C. R. Physique 2005, 6, 575.
[4] J. Kurhanewicz, D. Vigneron, K. Brindle, E. Chekmenev, A. Comment,
C. Cunningham, R. DeBerardinis, G. Green, M. Leach, S. Rajan, R. Rizi,
B. Ross, W. S. Warren, C. Malloy, Neoplasia 2011, 13, 81.
[5] S. E. Day, M. I. Kettunen, F. A. Gallagher, D. E. Hu, M. Lerche, J. Wolber,
K. Golman, J. H. Ardenkjaer-Larsen, K. M. Brindle, Nat. Med. 2007, 13, 1382.
[6] F. A. Gallagher, M. I. Kettunen, S. E. Day, M. Lerche, K. M. Brindle,
Magn. Reson. Med. 2008, 60, 253.
[7] C. Cudalbu, A. Comment, F. Kurdzesau, R. B. van Heeswijk, K.
Uffmann, S. Jannin, V. Denisov, D. Kirik, R. Gruetter, Phys. Chem.
Chem. Phys. 2010, 12, 5818.
[8] C. H. Cunningham, A. P. Chen, M. J. Albers, J. Kurhanewicz, R. E. Hurd,
Y. F. Yen, J. M. Pauly, S. J. Nelson, D. B. Vigneron, J. Magn. Reson.
2007, 187, 357.
[9] Y. F. Yen, S. J. Kohler, A. P. Chen, J. Tropp, R. Bok, J. Wolber, M. J. Albers, K.
A. Gram, M. L. Zierhut, I. Park, V. Zhang, S. Hu, S. J. Nelson, D. B. Vigneron,
J. Kurhanewicz, H. Dirven, R. E. Hurd, Magn. Reson. Med. 2009, 62, 1.
[10] M. J. Albers, R. Bok, A. P. Chen, C. H. Cunningham, M. L. Zierhut,
V. Y. Zhang, S. J. Kohler, J. Tropp, R. E. Hurd, Y.-F. Yen, S. J. Nelson, D. B.
Vigneron, J. Kurhanewicz, Cancer Res. 2008, 68, 8607.
[11] F. A. Gallagher, M. I. Kettunen, D. E. Hu, P. R. Jensen, R. in’t Zandt,
M. Karlsson, A. Gisselsson, S. K. Nelson, T. H. Witney, S. E. Bohndiek,
G. Hansson, T. Peitersen, M. H. Lerche, K. M. Brindle, Proc. Natl. Acad. Sci.
U. S. A. 2009, 106, 19801.
[12] T. C. Eisenschmid, R. U. Kirss, P. P. Deutsch, S. I. Hommeltoft,
R. Eisenberg, J. Bargon, R. G. Lawler, A. L. Balch, J. Am. Chem. Soc.
1987, 109, 8089.
[13] J.-B. Hövener, E. Chekmenev, K. Harris, W. Perman, L. Robertson, B. Ross,
P. Bhattacharya, Magn. Reson. Mat. Phys. Biol. Med. 2009, 22, 111.
[14] J.-B. Hövener, E. Chekmenev, K. Harris, W. Perman, T. Tran, B. Ross,
P. Bhattacharya, Magn. Reson. Mat. Phys. Biol. Med. 2009, 22, 123.
[15] C. R. Bowers, D. P. Weitekamp, Phys. Rev. Lett. 1986, 57, 2645.
[16] C. R. Bowers, D. P. Weitekamp, J. Am. Chem. Soc. 1987, 109, 5541.
[17] F. Reineri, A. Viale, G. Giovenzana, D. Santelia, W. Dastru, R. Gobetto,
S. Aime, J. Am. Chem. Soc. 2008, 130, 15047.
[18] D. Rivenzon-Segal, S. Boldin-Adamsky, D. Seger, R. Seger, H. Degani,
Int. J. Cancer 2003, 107, 177.
[19] P. Bhattacharya, S. R. Wagner, H. R. Chan, E. Y. Chekmenev,
W. H. Perman, B. D. Ross, in 17th ISMRM Conference, April 18-24,
Honolulu, Hawaii, 2009.
[49] K. W. Waddell, A. M. Coffey, E. Y. Chekmenev, J. Am. Chem. Soc. 2011,
133, 97.
[50] G. C. Levy, U. Edlund, J. Am. Chem. Soc. 1975, 97, 5031.
[51] P. Bhattacharya, K. Harris, A. P. Lin, M. Mansson, V. A. Norton, W. H.
Perman, D. P. Weitekamp, B. D. Ross, Magn. Reson. Mat. Phys. Biol.
Med. 2005, 18, 245.
[52] R. E. Hurd, Y.-F. Yen, D. Mayer, A. Chen, D. Wilson, S. Kohler, R. Bok, D.
Vigneron, J. Kurhanewicz, J. Tropp, D. Spielman, A. Pfefferbaum,
Magn. Reson. Med. 2010, 63, 1137.
[20] P. Bhattacharya, E. Y. Chekmenev, W. H. Perman, K. C. Harris, A. P. Lin,
V. A. Norton, C. T. Tan, B. D. Ross, D. P. Weitekamp, J. Magn. Reson.
2007, 186, 150.
[21] E. Y. Chekmenev, J. Hovener, V. A. Norton, K. Harris, L. S. Batchelder,
P. Bhattacharya, B. D. Ross, D. P. Weitekamp, J. Am. Chem. Soc. 2008,
130, 4212.
Supporting Information
Additional supporting information may be found in the online
[22] N. M. Zacharias, H. R. Chan, N. Sailasuta, B. D. Ross, P. Bhattacharya,
J. Am. Chem. Soc. 2012, 134, 934.
version of this article at the publisher’s web-site.
www.jlcr.org
Copyright © 2014 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2014, 57 517–524