• hyperpolarized carbon-13;
  • dynamic nuclear polarization;
  • imaging;
  • metabolism;
  • tumor;
  • pyruvate


  1. Top of page
  2. Abstract
  8. Acknowledgements

Dynamic nuclear polarization is an emerging technique for increasing the sensitivity of magnetic resonance imaging and spectroscopy, particularly for low-γ nuclei. The technique has been applied recently to a number of 13C-labeled cell metabolites in biological systems: the increase in signal-to-noise allows the spatial distribution of an injected molecule to be imaged as well as its metabolic product or products. This review highlights the most significant molecules investigated to date in preclinical cancer models, either in terms of their demonstrated metabolism in vivo or the biological processes that they can probe. In particular, label exchange between hyperpolarized 13C-labeled pyruvate and lactate, catalyzed by lactate dehydrogenase, has been shown to have a number of potential applications. Finally, techniques to image these molecules are also discussed as well as methods that may extend the lifetime of the hyperpolarized signal. Hyperpolarized magnetic resonance imaging and magnetic resonance spectroscopic imaging have shown great promise for the imaging of cancer in preclinical work, both for diagnosis and for monitoring therapy response. If the challenges in translating this technique to human imaging can be overcome, then it has the potential to significantly alter the management of cancer patients. Magn Reson Med, 2011. © 2011 Wiley-Liss, Inc.

The application of magnetic resonance in medicine has been dominated by imaging of tissue anatomy. However, one of the great strengths of magnetic resonance (MR) is spectroscopy, which allows imaging of tissue biochemistry; this was recognized in the earliest applications of MR to intact biological systems (1). Although magnetic resonance imaging (MRI) has become a routine clinical tool, the use of magnetic resonance spectroscopy (MRS) in patients has lagged far behind; this has been primarily due to a lack of sensitivity, which leads to long measurement times and poor image resolution (2–5). In spite of this, 1H-MRS measurements of cellular metabolites in a variety of tumor types have been shown to provide a sensitive means to diagnose disease and detect response to treatment (2–4). However, these measurements generally give a static picture of cellular metabolism.

In contrast, 13C-MRS measurements of cellular metabolism in systems incubated with 13C-labeled cell substrates give a dynamic, and therefore potentially more useful, measurement of tissue metabolism (6–8). For example, dynamic measurements of 13C-labeled glucose incorporation into muscle glycogen have been used to dissect the relative importance of glucose transport and hexokinase and glycogen synthase activity in controlling muscle glycogen synthesis (9). However, the problem of low sensitivity is even more acute in the case of 13C-MRS. Dynamic nuclear polarization (DNP or hyperpolarization) of 13C-labeled cell substrates enhances their sensitivity to detection by >104-fold (10, 11). Subsequent spectroscopic imaging of their metabolism following intravenous injection offers a solution to the problem of low sensitivity and has the potential to make dynamic metabolic imaging using MRS a routine clinical application.

There have been a number of recent review articles and book chapters on this subject (12–17), and therefore, the purpose of this brief review is to: highlight those DNP substrates that have shown the greatest promise for oncological applications in vivo; summarize the biochemical mechanisms responsible for label transfer from pyruvate to other metabolites in tumors; and finally provide an overview of the main challenges in label detection and imaging and how these may be addressed.


  1. Top of page
  2. Abstract
  8. Acknowledgements

Although there are many potential candidate molecules for DNP, very few fulfill the numerous criteria required for successful detection of their metabolism in vivo. These include:

  • 1
    High solubility and the ability to form a glass in the solid state.
  • 2
    Long spin-lattice (T1) relaxation time in the liquid state.
  • 3
    Rapid plasma membrane transport; this can be avoided if the reaction involving the substrate takes place in the extracellular space.
  • 4
    Rapid metabolism (within approximately five times the T1; usually 1–3 min) of the nontoxic DNP substrate to a nontoxic product with a significant chemical shift difference between the two.

These criteria are likely to limit useful substrates to those used catabolically (to generate energy) rather than those used in anabolism (to make cell structures) because in general, catabolism is faster than anabolism. For example, pyruvate, which has been widely used for hyperpolarized MRI in vivo (18–22), is a catabolic substrate that plays a central role in several metabolic pathways such as glycolysis and the citric acid cycle. [1-13C]pyruvate fulfills the above criteria: in the acid form, it is a high-concentration (14 M) liquid that forms a glass in the solid state without the addition of a glassing agent; the C1 carbon has a relatively long T1 in vivo (∼30 s), and the molecule is very rapidly transported into cells by the monocarboxylate transporter (23) and the subsequent metabolism is very fast. Contrast this with an anabolic substrate such as choline, which would be very interesting to study in cancer because there is often upregulation of choline metabolism in tumors. Although 15N-labeled choline has a very long T1 (∼4 min) and has been successfully hyperpolarized [the polarization being detected either directly (24) or indirectly via J-coupled protons (25)], there is as yet no evidence that its cellular uptake and subsequent phosphorylation to phosphocholine is sufficiently rapid to allow detection within the lifetime of the polarization. Some of the promising DNP substrates for oncological imaging are described below and summarized in Table 1; several of these demonstrate significant metabolism in vivo and enable fundamental biological processes to be probed.

Table 1. Molecules That Have Been Polarized Using DNP as Well as the Published Polarization Levels at 3.35 T, Method for Polarization, and the Biological Process(es) That Can Be Probed with Each Substrate
MoleculeReported polarizationMethod for polarizationWhat it can probeReferences
  1. In many cases, higher unpublished levels of polarization have been achieved with these molecules, and in some cases a gadolinium chelate has also been added to improve polarization (47). Results obtained at higher polarization field show further increase in polarization levels (33, 43). Abbreviations: ALT, alanine transaminase; BCAT, branched-chain amino acid transferase; GLUT5, glucose transporter; LDH, lactate dehydrogenase; PDH, pyruvate dehydrogenase; PPP, pentose phosphate pathway.

[1-13C]PyruvateUp to 40% at 3.35 T (up to 64% at 4.64 T)Neat acidLDH18–22,26–33
Treatment response
Intracellular pH
[1-13C]Ethyl pyruvate28–35%EthanolLDH34
Brain metabolism
13C-Bicarbonate15%CsH13CO3 or NaH13CO3pH36,37
Carbonic anhydrase activity
[2-13C]Fructose12%Aqueous solutionGLUT542
[1-13C]Ketoisocaproate32% (∼2-fold higher at 4.64 T)Neat acidBCAT43
[1-13C]Glutamate28%Tris baseALT45


To date, [1-13C]pyruvate has been the most commonly used metabolite for DNP. Intravenous injection of [1-13C]pyruvate can result in the appearance of [1-13C]lactate, [1-13C]alanine, and 13CO2 resonances, depending on the tissue probed and its metabolic state. Lactate labeling results from the reaction catalyzed by the enzyme lactate dehydrogenase (LDH) and alanine labeling from a transamination reaction catalyzed by alanine transaminase. 13C-labeled carbon dioxide is formed by the irreversible decarboxylation of [1-13C]pyruvate in the reaction catalyzed by the mitochondrial enzyme, pyruvate dehydrogenase (20). This 13CO2 rapidly equilibrates with [13C]bicarbonate in the reaction catalyzed by carbonic anhydrase, and it is the signal from the latter, which is usually observed at physiological pHs. For further discussion on pyruvate kinetics, see the section “Net Flux or Exchange? Understanding Hyperpolarized Substrate Kinetics.”

In tumors, lactate labeling is increased and this can be used to help distinguish tumor from normal tissue (26, 27). There is also evidence that pyruvate–lactate exchange can be used to identify tumor grade in a transgenic adenocarcinoma of mouse prostate (TRAMP) (28), with hyperpolarized lactate levels showing a strong correlation with the histologic grade of the excised tumors. The alanine and bicarbonate labeling varies more between tumor models, possibly due to variations in relative alanine transaminase and pyruvate dehydrogenase activity (28,29,48). In particular, alterations in labeled alanine levels have been suggested as a useful marker in rat hepatocellular carcinoma (29).

Assessment of tumor response has conventionally been made from measurements of changes in tumor size (49). However, this is not always appropriate, particularly if early response is to be detected or if the drug used does not result in tumor shrinkage, such as an antiangiogenic drug. There is increasing evidence for an early reduction in pyruvate–lactate exchange in a range of cancer models following treatment with cytotoxic chemotherapy (19, 38), targeted drugs (30,31,39), and radiotherapy (32). Changes in the metabolism of pyruvate following therapy in prostate (50) and brain tumors (32) have also been observed (Fig. 1), where imaging of 18F-fluorodeoxyglucose uptake with positron emission tomography can be difficult because of low uptake in prostate and background uptake in the brain (51) and, therefore, pyruvate could offer an alternative way to assess therapy in these tumors.

thumbnail image

Figure 1. Demonstration of the imaging of decreased lactate labeling in glioma following exposure to 15-Gy radiation. Representative images of hyperpolarized 13C pyruvate (c) and lactate (d) in a C6 glioma-bearing animal before (top) and 96 h after radiotherapy (bottom). A CSI data set is shown in (a). The chemical shift images were superimposed on the grayscale T1-weighted proton images (b) for anatomical reference. The lactate signals, in the false color images, were normalized to the maximum pyruvate signal in each dataset. From Ref.32. Data courtesy of Dr. Sam Day.

Download figure to PowerPoint

Hyperpolarized pyruvate has also been demonstrated as a valuable tool to detect response to the PI3K inhibitor LY294002 in both glioblastoma and breast cancer mouse models; reductions in the hyperpolarized lactate signal correlating with reductions in LDH activity as a consequence of PI3K pathway inhibition (31). Treatment of a prostate cancer model with imatinib, a tyrosine kinase inhibitor, also resulted in reduced hyperpolarized lactate signal due to lower expression of LDH (30), providing further evidence that hyperpolarized pyruvate provides a means to image pharmacodynamic activity.

[1-13C]Ethyl Pyruvate

Brain tumors usually have a compromised blood–brain barrier and demonstrate significant levels of [1-13C]lactate signal following [1-13C]pyruvate injection (27, 32). However, where the blood–brain barrier is still intact, such as in the normal brain, there is relatively slow uptake of pyruvate. Recently, hyperpolarized [1-13C]ethyl pyruvate, a lipophilic analog of pyruvate that can be hydrolyzed to pyruvate by esterases, was demonstrated to have rapid and preferential uptake into the brain, with subsequent formation of both labeled pyruvate and lactate (34). Such an approach represents an interesting way of bypassing a rate-limiting membrane transporter.


One potential limitation of [1-13C]pyruvate for future human studies is that the pyruvate concentration used is usually supraphysiological. In contrast, the concentration of lactate in the blood following the injection of hyperpolarized [1-13C]lactate is similar to that seen in an exercising animal, which could offer an important advantage for the introduction of the hyperpolarized 13C label. However, only relatively low levels of 13C-labeled pyruvate, alanine, and CO2 have been detected following intravenous [1-13C]lactate injection (35), reflecting LDH-catalyzed exchange of label into a relatively small endogenous pyruvate pool (see the section “Net Flux or Exchange? Understanding Hyperpolarized Substrate Kinetics”).


Imaging of necrosis is an important clinical challenge, and there are currently no techniques for directly and specifically imaging necrosis, either preclinically or clinically. Diffusion-weighted imaging detects the end result of necrosis, which is loss of cellularity (52), but is unable to detect low levels of diffuse necrosis or early necrosis (53). Fumarate is an example of a substrate whose intracellular metabolism is rapid but cellular uptake is slow. This property has been exploited in hyperpolarized studies to detect cellular necrosis (40) (Fig. 2). Viable cells demonstrate slow uptake, and, consequently, there is little detectable fumarate metabolism within the lifetime of the polarization. In contrast, rapid hydration to malate is seen in necrotic cells where the plasma membrane permeability barrier is compromised. The enzyme catalyzing this reaction—fumarase—is found widely within cells, has high activity, and only requires water as a cosubstrate; all of these features make it an ideal reporter of necrosis. Hyperpolarized fumarate has been shown to be hydrated to malate both in cancer cells in vitro and xenograft tumors in vivo following chemotherapy (38–40); the levels of malate produced correlating with the levels of necrosis in vitro and in vivo (38–40). In a murine lymphoma model, the increased malate signal observed following treatment with a vascular disrupting agent preceded the posttherapy changes seen with diffusion-weighted imaging, indicating that it could be used as a very early marker of necrosis (39).

thumbnail image

Figure 2. Detection of drug-induced necrosis in a lymphoma tumor with hyperpolarized fumarate. a: Correlation between the initial rate of malate production and the degree of necrosis in murine lymphoma cell suspensions in vitro. b: Representative spectrum from an etoposide-treated mouse with a subcutaneous implanted lymphoma tumor. The signal from hyperpolarized fumarate is seen at 175.4 ppm, the signal from [1-13C]malate at ∼181.8 ppm, and the signal from [4-13C]malate at ∼180.6 ppm. c: Total normalized hyperpolarized 13C malate signal over time for untreated (○) and etoposide-treated (▪) animals. The superimposed curves represent the average fits for both the fumarate and malate resonances and show an increase in malate production following treatment. From Ref.40.

Download figure to PowerPoint


Many disease processes are associated with alterations in extracellular pH—for example, most tumors have an acidic extracellular environment (54), which has been shown to mediate tumor cell migration and invasion (55). Furthermore, many drugs are either weak acids or bases and therefore tumor extracellular pH can influence their uptake (56). Despite the importance of the extracellular pH in disease in general and tumor biology and drug sensitivity in particular, there is currently no practical method for imaging extracellular pH distribution in humans. The ratio of the amplitudes of the hyperpolarized [13C]bicarbonate and 13CO2 resonance intensities (the latter is produced rapidly as a consequence of carbonic anhydrase activity) can be used to calculate pH using the Henderson–Hasselbalch equation (36, 37). This technique has allowed the spatial distribution of extracellular pH to be imaged within a tumor-bearing animal (36). Bicarbonate is already safely injected into humans (57) and therefore potentially offers a noninvasive method for imaging extracellular pH in humans and consequently a generic method for detecting the presence of disease and the response of this disease to treatment.


Imaging glucose metabolism in real time would be very attractive as a complementary technique to 18F-fluorodeoxyglucose positron emission tomography, which measures accumulation of the glucose analog over periods of minutes to hours. Glucose has been polarized with DNP (58), but the very short 13C T1s (<1 s) have prevented it from being used in vivo (42). In contrast, the T1 of the 13C at the C2 position of fructose is much longer, ∼16 s. Fructose exists as an isomeric mixture of five- and six-membered rings and is largely transported into the cell by the glucose transporter, GLUT5, which is highly expressed in some cancers (59). Differences in uptake of hyperpolarized [2-13C]fructose have been demonstrated between tumor and surrounding tissue in vivo (42). Although the small chemical shift difference between the fructose isomers and their fructose-6-phosphate counterparts makes them difficult to differentiate in vivo, differences in the ratios of the isomers may help to differentiate tumor from normal tissue as well as probing GLUT5 distribution and hexokinase activity. It may also be a measure of activity within the pentose phosphate pathway, which in turn may be a marker of oxidative stress (60).


Ketoisocaproate is converted to leucine, with the associated conversion of glutamate to α-ketoglutarate, by the enzyme branched-chain amino acid transferase. Branched-chain amino acid transferase enzyme activity has been shown to be upregulated in some tumors (61). The production of [1-13C]leucine has been demonstrated in murine lymphoma in vivo following the injection of hyperpolarized [1-13C]ketoisocaproate (43). An association between the branched-chain amino acid transferase activity and the appearance of leucine was observed, so it is possible that this could be used as a method to probe branched-chain amino acid transferase activity as well as the concentration of glutamate within the cell. However, it is not clear to what extent the rate of leucine production is influenced by other factors such as the rate of ketoisocaproate transport into the cell.


Urea was one of the first molecules to be polarized (10, 11) but does not demonstrate significant metabolism in vivo. This lack of metabolism means that the molecule can be used as a perfusion marker when coinjected with other substrates, such as pyruvate, either sequentially or simultaneously (37, 46). An independent marker of perfusion will be important for modeling the metabolism of hyperpolarized substrates in vivo, in both preclinical models and in the clinic.


Elevated glutamine metabolism is postulated to be a marker of tumor growth and cell proliferation. For example, altered glutamine metabolism might be used to help identify hepatocellular carcinoma within a cirrhotic liver (62) or identify malignant from benign disease when repeated biopsies are not possible. Glutamine is already safely administered to patients (57). The real-time conversion of hyperpolarized [5-13C]glutamine to [5-13C]glutamate, catalyzed by intramitochondrial glutaminase, has been demonstrated in hepatocellular carcinoma cells in vitro (44). C5-labeled glutamine was used instead of the C1-labeled compound as there is a larger chemical shift between [5-13C]glutamine and [5-13C]glutamate compared with the C1-labeled compounds, although [1-13C]glutamine has a longer T1 than [5-13C]glutamine. However, imaging of hyperpolarized glutamine in vivo has so far been limited by the modest polarizations that have been achieved to date, and it is not yet clear whether glutamine metabolism is sufficiently fast to allow its imaging in vivo.

Transfer of Polarization Between Molecules

Finally, although not all molecules may be polarized to a high degree or at high concentration, an interesting alternative to overcome this problem is to transfer polarization from a molecule which polarizes to a high degree, to another which polarizes less well but may be of greater biological interest. A recent proof-of-concept paper has shown that [1,1-13C]acetic anhydride can be polarized and that the hyperpolarized 13C can be chemically transferred to other molecules because of the preferential reaction of acetic anhydride with amine nucleophiles; this was demonstrated with a number of hyperpolarized [1-13C]N-acetylated amino acids (63). Although the polarizations obtained were quite modest (∼1000-fold compared to the thermal signal), future methods for creating secondary polarization may greatly increase the number of molecules that can be polarized with DNP.

Simultaneous Polarization of Multiple Substrates

One of the possible strengths of DNP is that more than one substrate could be copolarized and coinjected to probe several cellular processes simultaneously (37, 38). Given the heterogeneous nature of tumors and the genetic variability of patients, such multiparametric imaging could be of increasing importance for the future of personalized medicine. Wilson et al. successfully copolarized four metabolites (pyruvate, fumarate, urea, and bicarbonate) and used them in vitro and in vivo (37). The challenge for this approach is that the labeled molecules must have sufficiently different chemical shifts to enable them to be distinguished; this is particularly a problem for similarly labeled chemical groups (often carbonyls) and the separation of these peaks in vivo could be problematic. For example, although copolarization of fumarate and pyruvate was used successfully to assess therapy response in breast cancer cells in vitro at 9.4 T, the same approach was hampered in vivo because of the large lactate signal obscuring the smaller malate signals (38).


  1. Top of page
  2. Abstract
  8. Acknowledgements

As described in the previous section, hyperpolarized [1-13C]pyruvate has been the most widely studied substrate to date. However, for widespread use in medicine, standardization of the analysis of [1-13C]pyruvate labeling kinetics and that of other substrates will be required. The appearance of the hyperpolarized 13C label in lactate and alanine is due to the reactions catalyzed by LDH and alanine transaminase, respectively. The question arises as to whether this is net flux, i.e., net chemical conversion of pyruvate into lactate and alanine.

The reaction catalyzed by LDH is readily reversible, and the enzyme is commonly assumed to catalyze a reaction that is near-to-equilibrium in the cell. Indeed, measurements of the extracellular lactate/pyruvate ratio have been used to calculate the cytosolic NADH/NAD+ ratio (the enzyme is located exclusively in the cytosol), a calculation that assumes the enzyme-catalyzed reaction is near-to-equilibrium in the cell (64, 65). The equilibrium constant for the reaction is as follows (65):

  • equation image

Tumors typically contain lactate at a concentration of a few millimolar; thus, when hyperpolarized [1-13C]pyruvate enters the cell, the LDH-catalyzed reaction will come to near-chemical equilibrium with net conversion of a very small fraction of the pyruvate into lactate, following which there will be exchange of hyperpolarized 13C label between the pyruvate and lactate pools at near-chemical equilibrium. This has been demonstrated experimentally in isolated tumor cells, where adding lactate to the cell suspension increased the rate of label flux between the added hyperpolarized [1-13C]pyruvate and lactate (19) (Fig. 3). This observation is incompatible with net flux (where product inhibition by lactate would be expected) but is compatible with an exchange reaction (where the added lactate increases the near-equilibrium concentration of NADH, which in turn stimulates the rate of isotope exchange between pyruvate and lactate). The importance of exchange has also been demonstrated in tumors in vivo using magnetization transfer measurements. In lymphoma-bearing mice injected with hyperpolarized [1-13C]pyruvate, saturation or inversion of the [1-13C]lactate resonance resulted in an increased rate of decay of the pyruvate polarization, which clearly demonstrated flux of hyperpolarized 13C label from lactate to pyruvate and hence the presence of isotope exchange (66). Using the pyruvate and lactate concentrations measured in tumor extracts, the isotope flux was shown to be comparable in both directions, consistent with LDH catalyzing a near-equilibrium reaction in the cell. Furthermore, in a series of inversion transfer measurements, in which the inverted polarization was repeatedly returned to the z axis, showed that the isotope flux changed little during the time over which the hyperpolarized label was observable. This suggested that the isotope exchange velocity was constant over this time period and was consistent with the system being in a near-chemical equilibrium steady state. These magnetization transfer measurements could be combined with fast imaging techniques to generate maps of the rate of label flux between pyruvate and lactate (66) (Fig. 4).

thumbnail image

Figure 3. Evidence for the contribution of exchange to lactate labeling. An increase in the amount of unlabeled added lactate increased the level of lactate labeling following injection of hyperpolarized [1-13C]pyruvate into a tumor cell suspension. From Ref.19.

Download figure to PowerPoint

thumbnail image

Figure 4. Echo-planar 13C images obtained before and after selective inversion of the hyperpolarized [1-13C]lactate resonance (a) or [1-13C] pyruvate resonances in a subcutaneous murine lymphoma (b). The corresponding signal intensity changes in regions of interest that covered the tumor are shown in (c) and (d), respectively. From Ref.66.

Download figure to PowerPoint

The observed isotope exchange velocity will depend on the delivery of hyperpolarized [1-13C]pyruvate to the tumor, the rate of pyruvate transport across the cell membrane, and the kinetics of LDH. The latter will depend on the concentration of the enzyme and substrate concentrations at steady state (NAD+, NADH, pyruvate, and lactate) as well as the intracellular pH. In human breast cancer cells, the similarity of the apparent Km of pyruvate for label flux with the Km of the monocarboxylate transporter and the fact that an inhibitor of the transporter slowed the appearance of label in lactate were cited as evidence that the transporter is “rate limiting” (67). However, the relative importance of membrane transport and LDH activity for the kinetics of the observed label flux is likely to be different in different tumor types. Moreover, elucidation of the relative importance of these steps may be better determined by selective modulation of their activities and determination of the effect on the observed exchange velocity (68). The T2 relaxation times for both [1-13C]pyruvate and lactate show apparent multiexponential behavior, with a very short T2 component of ∼100 ms (66). The relative fractions of these components are consistent with them being from intracellular and extracellular pools, although this has not been demonstrated and care is required when interpreting data from such measurements (29). If this is the case, then by selecting signal from these different components, it may be possible to determine the relative importance of membrane transport. A corollary is that any pulse sequence that involves a delay in signal acquisition may bias signal detection toward either the intracellular or extracellular pools of these molecules (69).

Although it is possible to fit the exchange data to the Michaelis–Menten equation, which describes net chemical flux at steady state (50,67,70), it is important to recognize that the estimated kinetic constants derived from the isotope exchange data are only apparent constants and that their values need not correspond to those determined from steady-state kinetic measurements of net flux. For example, Michaelis–Menten kinetics cannot predict an increase in label flux between pyruvate and lactate when the lactate concentration is increased. In a steady-state kinetic experiment, the pyruvate is added to a reaction mixture containing NADH and the initial rate of net conversion of pyruvate to lactate is measured over a time period where there is only a very small change in NADH concentration. Contrast this to what happens in an experiment with hyperpolarized [1-13C]pyruvate, all of the LDH substrates are present and the addition of pyruvate will result in a rapid adjustment of the chemical equilibrium before isotope exchange between the pyruvate and lactate pools. Increasing the pyruvate concentration might be expected to increase the exchange velocity; however, this effect will be offset to some extent by a decrease in the equilibrium concentration of NADH, which will tend to decrease the exchange velocity (71, 72).

The fact that isotope exchange is being measured, rather than net chemical flux, means that the mechanism of the enzyme must also be considered when interpreting the flux measurements. LDH has an ordered ternary complex mechanism, in which the coenzymes NAD+ and NADH bind to the enzyme before lactate and pyruvate, respectively.

  • equation image

This means that label exchange between pyruvate and lactate can be much faster than that between the coenzymes and much faster also than net chemical flux through the reaction (73). There has also been some speculation that the hyperpolarized signal may decay more rapidly in an enzyme–substrate complex, with consequences for the measured label flux. Both issues have been considered previously with magnetization transfer measurements of flux in enzyme-catalyzed reactions (74). Even if there is faster relaxation in these complexes, they turn over so rapidly and their lifetimes are so short that it is unlikely there will be significant loss of polarization.

Magnetization transfer measurements in a metabolic steady state and measurements of hyperpolarized 13C label exchange in a system that rapidly achieves near-chemical equilibrium are formally equivalent, and the same equations may be used to analyze the kinetics of polarization exchange. The most commonly used approach for analyzing pyruvate exchange data has been to fit it to a simple two-site exchange model

  • equation image

where ρP, ρL are the longitudinal relaxation rates for pyruvate and lactate, respectively, and kP and kL are the pyruvate [RIGHTWARDS ARROW] lactate (or “forward”) and lactate [RIGHTWARDS ARROW] pyruvate (“back”) exchange rate constants, respectively. A similar modeling approach can be used for alanine. It is worth noting that this simple model does not take into account the effect of radiofrequency (RF) pulsing. The flip angles used in dynamic DNP studies are usually low, and the effect of moderate pulse repetition rates is limited mainly to a slight overestimation of relaxation rates while the calculated forward exchange rates are relatively unaffected (19). In cases where flip angles are larger or variable flip angles are used (75), the effects of RF pulsing can also be included in the model provided that the flip angles can be measured accurately, a task that is not straightforward, especially in experiments using surface coils. Ideally, all four rate constants are fitted from the data. In practice, the fitting process can easily lead to unrealistic results when all four rate constants are fitted simultaneously, and the fitting is often limited to three rate constants. For example, Day et al. (19) assumed that the relaxation rates for both metabolites are similar and fitted two exchange rate constants and one relaxation rate constant. Lactate labeling can also be fitted to a two-site model, where the reverse flux from lactate to pyruvate is ignored (22,50,67). However, this does not mean that this flux is truly zero and that there is no exchange, only that this flux is poorly determined by the experiment with hyperpolarized [1-13C]pyruvate, because during the lifetime of the polarization the fractional labeling of lactate will be very low compared to that of pyruvate. Importantly, the forward exchange rate obtained using either approach is in most cases nearly identical (40). More complex kinetic models may be used to analyze the exchange in vivo, which also take into account the rate of pyruvate delivery (50, 76). In such approaches, an independent read-out of delivery, such as that given by coinjection of hyperpolarized 13C-urea, could be beneficial.

The demonstration that LDH catalyzes an exchange reaction means that the injected hyperpolarized [1-13C]pyruvate effectively labels endogenous lactate pools, such as we would expect to find in tumors or other hypoxic tissues. The reverse experiment with hyperpolarized [1-13C]lactate (35) works less well as the endogenous pyruvate pool is much lower in concentration. Compared to 1H magnetic resonance spectroscopy imaging of the endogenous lactate pool, the hyperpolarized [1-13C]pyruvate experiment allows very sensitive detection of endogenous lactate, with no background signals. More importantly, it also allows the kinetics of lactate labeling to be determined, which can give important information regarding tumor cell energy status, for example, a decreased rate of lactate labeling following induction of cell death (19). Similarly, the reaction catalyzed by alanine transaminase is also likely to be near-to-equilibrium in the cell and therefore observable alanine labeling from hyperpolarized [1-13C]pyruvate is to be expected in those tissues with a large endogenous alanine pool. However, in reactions such as those catalyzed by pyruvate dehydrogenase, which are clearly irreversible in the cell, the experiment measures net flux (77). The most appropriate method for analysis of hyperpolarized label exchange will be determined, therefore, by the specific metabolic pathway under investigation.


  1. Top of page
  2. Abstract
  8. Acknowledgements

The large increase in signal intensity afforded by hyperpolarization allows imaging of the spatial distribution of metabolism in real time. Depending on the molecule(s) studied, one may be interested in either acquiring a single high-resolution metabolic image (for example, a pH map) or a dynamic series of images in which metabolic changes are followed (for example, lactate labeling). Because of the transient nature of the hyperpolarized signal, fast imaging approaches are usually required. A wide array of fast spectroscopic imaging sequences has been developed for proton imaging and in principle all of them could be used for hyperpolarized 13C imaging. In contrast to proton, hyperpolarized 13C spectra usually contain only few resonances, and the lower gyromagnetic ratio leads to fewer artifacts from off-resonance effects. However, the lower gyromagnetic ratio of 13C means that the imaging gradient amplitudes need to be four times higher than for proton to achieve a similar spatial resolution. Furthermore, the chemical shift range of 13C metabolites is relatively broad. For example, the difference between [1-13C]lactate and [13C]bicarbonate resonances is 20 ppm, and the 13CO2 resonance is a further 40 ppm away from lactate. Therefore, wide spectral widths may be required, especially at higher magnetic fields. As each RF excitation irreversibly destroys some of the polarization, the number of RF excitation pulses used in image formation must be limited. The total imaging time is restricted to within two to three times the T1 of the metabolite, which is usually less than 2 min in most cases. Despite these limitations, a number of spectroscopic imaging methods have been developed to capture images of hyperpolarized metabolites (78–81). Importantly, a large part of the pulse sequence development has been performed at 3 T on clinical systems, which is likely to speed up their availability in the clinic.

Several studies have used a simple spectroscopic imaging sequence (19,26,40,43), with phase-encoding gradients in two or three directions. The sequence is widely available in a clinical setting, has relatively low-gradient amplitude requirements, and detects all the resonances simultaneously with high spectral quality. However, the inefficient use of the polarization, the slow imaging speed, and the large number of excitation pulses severely limit the size of the data matrices (usually 16 × 16 or 32 × 32) and therefore the spatial resolution. Low flip angle pulses are required to preserve the polarization over the whole imaging period, and the continuous loss of signal may lead to anomalies in the resulting image, although this can be minimized using variable flip angle schemes (75, 82) that allow more efficient use of the available polarization. Other limitations of the sequence can be addressed by acquiring reduced data sets, parallel imaging (83), or by using non-Fourier-based spatial encoding (84). Despite these improvements, the number of images that can be acquired is still very limited and so the sequence usually only allows a snapshot image of metabolic status at a single time point.

Fast imaging techniques can speed up imaging times to less than 1 s, a fraction of the time required for spectroscopic imaging, with only small losses in signal-to-noise (85). This allows real-time monitoring of label transfer between the hyperpolarized substrate and its products (22,86,87) (Fig. 5). The majority of the studies have used pulse sequences based on gradient refocusing, such as echo-planar (spectroscopic) imaging EP(S)I (78, 82), spiral chemical shift imaging (79,85,86), and rosette chemical shift imaging (80). Spatial information in one (EPSI) or two (spiral and rosette) dimensions is encoded with varying imaging gradients during data acquisition, whereas the spectroscopic information is obtained by analysis of signal behavior at different echo times. Typical of fast imaging sequences, the long data acquisition time makes them more vulnerable to distortions in the magnetic field homogeneity, and careful shimming is needed, although the lower gyromagnetic of ratio 13C somewhat alleviates the situation. The echo times used are dictated by the spectral bandwidth required and the separation of peaks and therefore become shorter at higher fields. This leads to increasing demands on gradient hardware due to the rapid gradient switching that is required. The problem can be partially resolved by narrowing the spectral bandwidth, while ensuring that the folded peaks do not overlap (79,82,88). The chemical shift-dependent phase variations of the metabolite resonances can be converted into spectral information using either direct fast Fourier transformation or by using least-squares separation techniques (88–90). The latter approach usually requires fewer input images to achieve good spectral separation and is therefore not as sensitive to T2* relaxation with long echo trains. It could also be used in multishot approaches where the initial echo time is adjusted between excitations and only a single image is acquired (91). Although spiral and rosette spectroscopic imaging require more complex data reconstruction than EPSI, they potentially allow single-shot imaging with short echo times and therefore use a minimal number of excitation pulses. However, in most cases, multiple shots are still required because of limited gradient performance (85). The problem of signal loss due to T2* relaxation can also be addressed by using RF refocusing pulses and multiple echo trains, which exploit the relatively long 13C T2s (81, 92). Another interesting approach to improve EPSI acquisition applies compressed sensing theory to either decrease the acquisition time of full 3D data sets to a few seconds (93) or to increase spatial resolution (94). Further development in the fast imaging methods and their combination with approaches such as parallel imaging (83) are likely to improve the current 1–5 mm2 in-plane spatial resolution.

thumbnail image

Figure 5. Real-time metabolic imaging using spiral chemical shift imaging. Six frames of a time series of metabolic maps of Pyr, Lac, and Ala (superimposed on the corresponding proton MR image) acquired from a 10-mm slice through the kidneys of a mouse. Each metabolic time series is individually scaled (threshold at 8% for Pyr and 20% for Lac and Ala). The time stamps are relative to the time of injection. The complete movie is in online at From Ref.86. Data courtesy of Dr. Dirk Mayer et al.

Download figure to PowerPoint

The relative simplicity of 13C spectra may be exploited by using spectrally selective excitation pulses in combination with less-demanding single-shot fast imaging sequences (EPI and single-shot spiral imaging) (95–97) (Fig. 3). Each resonance can be excited sequentially with different flip angles if necessary to optimize signal-to-noise and imaged with minimal disturbance from the other resonances provided that the chemical shift difference between the resonances is sufficient and provided the flux of 13C label between them is slow compared with the imaging timescale. It is feasible that this approach could provide higher spatial resolution than single-shot methods where all the signals are imaged simultaneously. For spatially and spectrally selective excitation, spatial-spectral RF pulses (95,98,99) are needed. The pulses consist of a series of RF subpulses and oscillating gradients and can be designed using the Shinnar–LeRoux approach (98, 100), taking into account the limitations of gradient and RF systems. At moderate field strengths the approach is very appealing, allowing image acquisition with fewer RF excitations than an EPSI sequence and with only a small penalty in terms of RF efficiency compared to single-shot spectroscopic spiral imaging (96). However, the approach may be more sensitive to B0 imperfections. Because these effects cannot be corrected by postprocessing, careful shimming is essential. Because of imperfections in the designed spatial-spectral RF pulses, only the first few side lobes are usually used for spectral selection. The approach therefore becomes challenging at high field strengths where increased frequency dispersion requires short high-power RF pulses and rapid gradient switching. Nevertheless, the approach has been implemented successfully at fields of up to 14 T (101). It is worth noting that spatial-spectral RF pulses can also be designed to give different excitation flip angles to different metabolites in a single pulse. This may be useful in spectroscopic imaging to preserve signal of the substrate while retaining the high signal-to-noise of the metabolites produced from it (93).


  1. Top of page
  2. Abstract
  8. Acknowledgements

The signal lifetime places a fundamental limit on the imaging window available with hyperpolarized substrates. As a result, carbonyl groups such as the C1 position in pyruvate are usually chosen for 13C labeling because of the lack of attached protons, giving T1 values on the order of tens of seconds and an imaging window of a few minutes. It has been shown previously that the quantum mechanical properties of two spin (or more) systems may be used to extend the hyperpolarization lifetime by exploiting the symmetry properties of singlet states in order to avoid the dominant NMR relaxation mechanisms, such as dipolar coupling (102). Transfer of hyperpolarization to a directly bonded proton or to solvent water may also provide an avenue to indirectly probe the hyperpolarized signal (103–106). Increasing the imaging lifetime of hyperpolarized 13C-labeled molecules could greatly increase the utility of DNP as well as the number of metabolic substrates that could potentially be imaged. Recent developments have allowed the conversion of magnetization into singlet order, even in weak and inhomogeneous magnetic fields, raising the prospect of magnetization generated by dissolution DNP being stored in singlet states before in vivo imaging.

Singlet States

Quantum Mechanical Description

Any quantum state of a pair of spins-1/2 in a magnetic field B0 can be expressed as a superposition of the four Zeeman product states |αα〉, |αβ〉, |βα〉, and |ββ〉, where α and β denote spin “up” and spin “down,” respectively. Furthermore, these Zeeman states can be combined to construct one singlet and three triplet states according to:

  • equation image

where the subscript refers to the total nuclear spin of the superposition. The spectra acquired from both magnetically equivalent and inequivalent (weakly coupled) spin pairs are shown in Fig. 6. In the special case where there is no chemical shift difference between spin pair (magnetic equivalence), the singlet and triplet states are eigenstates (energy levels) of the spin system. The singlet state |S0〉 is antisymmetric under exchange of spins and behaves like a single, nonmagnetic particle of total spin I = 0 and so does not give an NMR signal. It is the antisymmetric property of the singlet state that confers its longer lifetime, TS, which has been reported to be up to 10-fold longer than the corresponding T1 (107, 108). The strongest relaxation mechanisms, including dipole–dipole relaxation, are exchange symmetric, so the singlet state is immune to these processes unlike the symmetric triplet states. In the general case of a magnetically inequivalent spin pair, there is a chemical shift difference between them, and this interaction breaks the spin exchange symmetry allowing transitions between all states. Manipulating the magnetic equivalence of a spin system in the course of an experiment, therefore, allows the hyperpolarized signal to be “trapped” in the singlet state of the equivalent system and later returned to the measurable Zeeman state of the inequivalent case.

thumbnail image

Figure 6. Demonstration of the NMR spectra acquired from magnetically equivalent and inequivalent spins. α and β denote the spin “up” (angular momentum equation image along the magnetic field direction) and “down” (angular momentum equation image)states, respectively. In the equivalent case, the singlet state contributes no NMR signal, so the only NMR peak arises from transitions between triplet states. For inequivalent spins, a spin–spin (J) coupling exists between them as well as a chemical shift difference. Inversion of any one of these peaks will exchange the populations of the two Zeeman states that produce it.

Download figure to PowerPoint

Experimental Demonstrations of Singlet States

A number of methods are available to manipulate magnetic equivalence in a system that intrinsically lacks such a property, and these have been reviewed in detail elsewhere (102,109–112). The general principle of all such experiments involves preparation of a nonequilibrium population in the |αβ〉 and/or |βα〉 spin states (see Fig. 6), followed by conversion to a magnetically equivalent system where the perturbed population is distributed between the I = 0 singlet and triplet states. The polarization is then returned to the inequivalent state at some later time, when the triplet states have equilibrated and the singlet is unchanged, in order to use the stored polarization.

The chemical shift difference of an inequivalent system is suppressed at low magnetic field so any excess population can be preserved in a low-field singlet state using adiabatic “field cycling” (109). This first demonstration of lifetime extension using singlet states was performed using the two protons in 2,3-dibromothiophene (107): an RF pulse was applied to a sample at 9.4 T to create nonequilibrium spin-state populations, which were then converted to a long-lived singlet state by transporting the sample to the low field. Singlet-state population was converted back into a visible NMR signal when the sample was returned to high field (102). Recent advances have shown that the field cycling method may also be applied to other nuclei, such as the coupled nitrogen nuclei in 15N2-nitrous oxide (113), and importantly, that the whole process can be performed outside the magnet (108), although the method has yet to be applied to 13C-labeled hyperpolarized substrates.

The use of a chemical reaction to modify the symmetry of a hyperpolarized 13C substrate has also been demonstrated to extend the hyperpolarization lifetime by populating a singlet state. This experiment exploited the change in magnetic equivalence that arises from the chemical hydration of [2,3-13C2]diacetyl, whose carbonyl carbons are equivalent, to the monohydrate form, which has a 110 ppm difference between the 13C resonances and is the majority species in an aqueous solution (114). Following dissolution of hyperpolarized [2,3-13C2]diacetyl, populations of the |αα〉 and |αα〉 states were exchanged in the monohydrate form, and signal was preserved in the singlet state by shifting the equilibrium toward neat [2,3-13C2]diacetyl with acetone. A subsequent injection of water shifted the equilibrium back in favor of the monohydrate, and the singlet-state population was released as an observable NMR signal. For this to be exploited in vivo, the hyperpolarized precursor molecule must be inequivalent and the singlet state populated in an external system; the injected magnetically equivalent substrate must then be used by a relevant biological pathway that will break the symmetry of the 13C spin pair. A constraint on the application on this technique is that the scalar coupling between the 13C spin pair must be greater than their couplings to any other nuclei (JC[BOND]C >> JC[BOND]H) for the singlet population to be conserved, meaning perdeuterated molecules are likely to be the best candidates. Even with these considerations, a number of molecules have been suggested for in vivo application, including L-DOPA, which has previously been hyperpolarized (115) and it is anticipated that molecules of relevance in cancer imaging could also be identified for preparation in this manner.

Polarization Transfer to Other Nuclei

In addition to storing polarization in the singlet state of a 13C spin pair, it is also possible to extend the lifetime of the hyperpolarization and expand the number of metabolic processes that can be followed by transferring the polarization between nuclei in a hyperpolarized system. This may be achieved by the application of appropriate pulse sequences to transfer the polarization or exploiting the inherent cross relaxation between the different spin species. For example, in studies of slow enzymatic processes, where the hyperpolarized nucleus is not at a relevant metabolic site or exhibits only a small chemical shift change following metabolism, it may be more practical to detect signal changes via J-coupled protons. Furthermore, although proton imaging is routine on all clinical MR machines, few systems have the capability to detect 13C signal; indirect detection of hyperpolarized carbon using proton MRI could make DNP more widely applicable for routine patient imaging.

As an example, abnormal choline phospholipid metabolism is a common feature of many types of cancer and is frequently measured by 1H NMR (2–5). Sarkar et al. used a reverse INEPT sequence to transfer signal from long-lived hyperpolarized 15N in choline to hydroxyl protons (25) to follow phosphorylation by choline kinase using 1H NMR. The potential limitation of the approach is that although the 15N signal is retained for long periods of time, standard polarization transfer sequences commonly use 90° pulses and the majority of the hyperpolarized signal may therefore be lost in the first transfer. The experiment can be improved by using an INEPT-based sequence with spatial encoding techniques that allowed repeated transfers of small fractions of the 15N polarization to the hydroxyl protons (103). The maximal enhancement of the proton signal they observed was >100-fold.

Manipulation of the hyperpolarized signal into singlet states offers an interesting avenue for future DNP research, and with the present interest in exploiting such methods, the application of extending the hyperpolarized lifetime in an in vivo setting may be demonstrated in the near future.


  1. Top of page
  2. Abstract
  8. Acknowledgements

The limit of anatomical resolution that can be achieved in patients using conventional imaging techniques (such as MRI and CT) is rapidly being approached. There are a number of areas where functional and molecular imaging techniques could enhance patient management as they offer potentially powerful tools to aid diagnosis, identify disease heterogeneity, predict outcome, target biopsies, and determine treatment response noninvasively (51). DNP of 13C-labeled cell substrates has the potential to help achieve these targets.

Given the preclinical results obtained so far, the initial trials with DNP-hyperpolarized substrates are likely to center on early response markers for treatment, particularly in oncology. Early imaging biomarkers of chemotherapy response that serve as indicators of later definitive response as well as overall patient survival are important for the development of new drugs and for improving prognosis; patients may be changed from an ineffective therapy to an effective one very rapidly if the imaging test suggests that the patient is not responding to treatment, and this is particulary important in diseases where there is rapid progression (51). Furthermore, there is some evidence that hyperpolarization could aid in the assessment of tumor grade. The use of noninvasive cancer imaging to accurately diagnose disease with few (or no) invasive biopsies is an attractive prospect for the management of patients, facilitating repeated assessment over time and quantifiable sampling of a large volume of tissue.

Finally, the ability to detect residual or recurring disease is a major diagnostic challenge in oncology. Residual diseased tissue may be present, which is too small or diffuse to detect using conventional imaging techniques, and may lie dormant for many years before disease recurrence (116). Because hyperpolarized MRI does not require a radiation dose for the patient, it could be used for serial monitoring to detect recurrence at an early stage, for example, through increased lactate labeling. As with other forms of functional and molecular imaging, the challenge for hyperpolarized substrates will be to identify altered metabolism in a small volume of tissue when compared with the background metabolism.

Further work is needed to develop and standardize data analysis methods, design high-resolution fast imaging sequences, and to find methods that could extend the relatively short polarization lifetime; significant progress has been made already in each of these areas, suggesting that hyperpolarized magnetic resonance spectroscopic imaging could have a significant impact on the imaging of cancer patients in the future. The initial results from the first clinical trial, with hyperpolarized [1-13C]pyruvate in prostate cancer, show great promise (, accessed November 2010).


  1. Top of page
  2. Abstract
  8. Acknowledgements
  • 1
    Gadian DG, Radda GK. NMR studies of tissue metabolism. Annu Rev Biochem 1981; 50: 6983.
  • 2
    Kurhanewicz J, Vigneron DB. Advances in MR spectroscopy of the prostate. Magn Reson Imaging Clin N Am 2008; 16: 697710, ix–x.
  • 3
    Nelson SJ, Graves E, Pirzkall A, Li X, Antiniw CA, Vigneron DB, McKnight TR. In vivo molecular imaging for planning radiation therapy of gliomas: an application of 1H MRSI. J Magn Reson Imaging 2002; 16: 464476.
  • 4
    Sardanelli F, Fausto A, Podo F. MR spectroscopy of the breast. Radiol Med 2008; 113: 5664.
  • 5
    van der Graaf M. In vivo magnetic resonance spectroscopy: basic methodology and clinical applications. Eur Biophys J 2010; 39: 527540.
  • 6
    Henry PG, Adriany G, Deelchand D, Gruetter R, Marjanska M, Oz G, Seaquist ER, Shestov A, Ugurbil K. In vivo 13C NMR spectroscopy and metabolic modeling in the brain: a practical perspective. Magn Reson Imaging 2006; 24: 527539.
  • 7
    Jeffrey FM, Rajagopal A, Malloy CR, Sherry AD. 13C-NMR: a simple yet comprehensive method for analysis of intermediary metabolism. Trends Biochem Sci 1991; 16: 510.
  • 8
    Shulman RG, Rothman DL. 13C NMR of intermediary metabolism: implications for systemic physiology. Annu Rev Physiol 2001; 63: 1548.
  • 9
    Schafer JR, Fell DA, Rothman D, Shulman RG. Protein phosphorylation can regulate metabolite concentrations rather than control flux: the example of glycogen synthase. Proc Natl Acad Sci USA 2004; 101: 14851490.
  • 10
    Ardenkjaer-Larsen JH, Fridlund B, Gram A, Hansson G, Hansson L, Lerche MH, Servin R, Thaning M, Golman K. Increase in signal-to-noise ratio of > 10,000 times in liquid-state NMR. Proc Natl Acad Sci USA 2003; 100: 1015810163.
  • 11
    Golman K, Ardenkjaer-Larsen JH, Petersson JS, Mansson S, Leunbach I. Molecular imaging with endogenous substances. Proc Natl Acad Sci USA 2003; 100: 1043510439.
  • 12
    Ardenkjaer-Larsen JH, Golman K, Brindle KM. Hyperpolarized 13C magnetic resonance imaging—principles and applications. In: WeisslederR, GambhirSS, RossBD, RehemtullaA, editors. Molecular imaging: principles and practice. New York: McGraw-Hill; 2010. pp 377388.
  • 13
    Gallagher FA, Kettunen MI, Brindle KM. Biomedical applications of hyperpolarized 13C magnetic resonance imaging. Prog NMR Spectrosc 2009; 55: 285295.
  • 14
    Golman K, Petersson JS. Metabolic imaging and other applications of hyperpolarized 13C1. Acad Radiol 2006; 13: 932942.
  • 15
    Ross BD, Bhattacharya P, Wagner S, Tran T, Sailasuta N. Hyperpolarized MR imaging: neurologic applications of hyperpolarized metabolism. AJNR Am J Neuroradiol 2010; 31: 2433.
  • 16
    Rowland IJ, Peterson ET, Gordon JW, Fain SB. Hyperpolarized 13carbon MR. Curr Pharm Biotechnol 2010; 11: 709719.
  • 17
    Viale A, Aime S. Current concepts on hyperpolarized molecules in MRI. Curr Opin Chem Biol 2010; 14: 9096.
  • 18
    Chen AP, Albers MJ, Cunningham CH, Kohler SJ, Yen YF, Hurd RE, Tropp J, Bok R, Pauly JM, Nelson SJ, Kurhanewicz J, Vigneron DB. Hyperpolarized C-13 spectroscopic imaging of the TRAMP mouse at 3T-initial experience. Magn Reson Med 2007; 58: 10991106.
  • 19
    Day SE, Kettunen MI, Gallagher FA, Hu DE, Lerche M, Wolber J, Golman K, Ardenkjaer-Larsen JH, Brindle KM. Detecting tumor response to treatment using hyperpolarized 13C magnetic resonance imaging and spectroscopy. Nat Med 2007; 13: 13821387.
  • 20
    Golman K, in 't Zandt R, Thaning M. Real-time metabolic imaging. Proc Natl Acad Sci USA 2006; 103: 1127011275.
  • 21
    Schroeder MA, Swietach P, Atherton HJ, Gallagher FA, Lee P, Radda GK, Clarke K, Tyler DJ. Measuring intracellular pH in the heart using hyperpolarized carbon dioxide and bicarbonate: a 13C and 31P magnetic resonance spectroscopy study. Cardiovasc Res 2010; 86: 8291.
  • 22
    Hurd RE, Yen YF, Tropp J, Pfefferbaum A, Spielman DM, Mayer D. Cerebral dynamics and metabolism of hyperpolarized [1-(13)C]pyruvate using time-resolved MR spectroscopic imaging. J Cereb Blood Flow Metab 2010; 30: 17341741.
  • 23
    Halestrap AP, Price NT. The proton-linked monocarboxylate transporter (MCT) family: structure, function and regulation. Biochem J 1999; 343 ( pt 2): 281299.
  • 24
    Gabellieri C, Reynolds S, Lavie A, Payne GS, Leach MO, Eykyn TR. Therapeutic target metabolism observed using hyperpolarized 15N choline. J Am Chem Soc 2008; 130: 45984599.
  • 25
    Sarkar R, Comment A, Vasos PR, Jannin S, Gruetter R, Bodenhausen G, Hall H, Kirik D, Denisov VP. Proton NMR of (15)N-choline metabolites enhanced by dynamic nuclear polarization. J Am Chem Soc 2009; 131: 1601416015.
  • 26
    Golman K, Zandt RI, Lerche M, Pehrson R, Ardenkjaer-Larsen JH. Metabolic imaging by hyperpolarized 13C magnetic resonance imaging for in vivo tumor diagnosis. Cancer Res 2006; 66: 1085510860.
  • 27
    Park I, Larson PE, Zierhut ML, Hu S, Bok R, Ozawa T, Kurhanewicz J, Vigneron DB, Vandenberg SR, James CD, Nelson SJ. Hyperpolarized 13C magnetic resonance metabolic imaging: application to brain tumors. Neuro-oncology 2010; 12: 133144.
  • 28
    Albers MJ, Bok R, Chen AP, Cunningham CH, Zierhut ML, Zhang VY, Kohler SJ, Tropp J, Hurd RE, Yen YF, Nelson SJ, Vigneron DB, Kurhanewicz J. Hyperpolarized 13C lactate, pyruvate, and alanine: noninvasive biomarkers for prostate cancer detection and grading. Cancer Res 2008; 68: 86078615.
  • 29
    Yen YF, Le Roux P, Mayer D, King R, Spielman D, Tropp J, Butts Pauly K, Pfefferbaum A, Vasanawala S, Hurd R. T(2) relaxation times of (13)C metabolites in a rat hepatocellular carcinoma model measured in vivo using (13)C-MRS of hyperpolarized [1-(13)C]pyruvate. NMR Biomed 2010; 23: 414423.
  • 30
    Dafni H, Larson PE, Hu S, Yoshihara HA, Ward CS, Venkatesh HS, Wang C, Zhang X, Vigneron DB, Ronen SM. Hyperpolarized 13C spectroscopic imaging informs on hypoxia-inducible factor-1 and myc activity downstream of platelet-derived growth factor receptor. Cancer Res 2010; 70: 74007410.
  • 31
    Ward CS, Venkatesh HS, Chaumeil MM, Brandes AH, Vancriekinge M, Dafni H, Sukumar S, Nelson SJ, Vigneron DB, Kurhanewicz J, James CD, Haas-Kogan DA, Ronen SM. Noninvasive detection of target modulation following phosphatidylinositol 3-kinase inhibition using hyperpolarized 13C magnetic resonance spectroscopy. Cancer Res 2010; 70: 12961305.
  • 32
    Day SE, Kettunen MI, Krishna MC, Mitchell JB, Lizak MJ, Morris HD, Matsumoto S, Koretsky AP, Brindle KM. Detecting response of rat C6 glioma tumors to radiotherapy using hyperpolarized [1-13C]pyruvate and 13C magnetic resonance spectroscopic imaging. Magn Reson Med 2011; 65: 557563.
  • 33
    Johannesson H, Macholl S, Ardenkjaer-Larsen JH. Dynamic nuclear polarization of [1–13C]pyruvic acid at 4.6 tesla. J Magn Reson 2009; 197: 167175.
  • 34
    Hurd RE, Yen YF, Mayer D, Chen A, Wilson D, Kohler S, Bok R, Vigneron D, Kurhanewicz J, Tropp J, Spielman D, Pfefferbaum A. Metabolic imaging in the anesthetized rat brain using hyperpolarized [1–13C] pyruvate and [1–13C] ethyl pyruvate. Magn Reson Med 2010; 63: 11371143.
  • 35
    Chen AP, Kurhanewicz J, Bok R, Xu D, Joun D, Zhang V, Nelson SJ, Hurd RE, Vigneron DB. Feasibility of using hyperpolarized [1–13C]lactate as a substrate for in vivo metabolic 13C MRSI studies. Magn Reson Imaging 2008; 26: 721726.
  • 36
    Gallagher FA, Kettunen MI, Day SE, Hu DE, Ardenkjaer-Larsen JH, Zandt R, Jensen PR, Karlsson M, Golman K, Lerche MH, Brindle KM. Magnetic resonance imaging of pH in vivo using hyperpolarized 13C-labelled bicarbonate. Nature 2008; 453: 940943.
  • 37
    Wilson DM, Keshari KR, Larson PE, Chen AP, Hu S, Van Criekinge M, Bok R, Nelson SJ, Macdonald JM, Vigneron DB, Kurhanewicz J. Multi-compound polarization by DNP allows simultaneous assessment of multiple enzymatic activities in vivo. J Magn Reson 2010; 205: 141147.
  • 38
    Witney TH, Kettunen MI, Hu DE, Gallagher FA, Bohndiek SE, Napolitano R, Brindle KM. Detecting treatment response in a model of human breast adenocarcinoma using hyperpolarised [1-(13)C]pyruvate and [1,4-(13)C(2)]fumarate. Br J Cancer 2010; 103: 14001406.
  • 39
    Bohndiek SE, Kettunen MI, Hu DE, Witney TH, Kennedy BWC, Gallagher FA, Brindle KM. Detecting tumor response to a vascular disrupting agent using hyperpolarized 13C magnetic resonance spectroscopy. Mol Cancer Ther 2010; 9: 32783288.
  • 40
    Gallagher FA, Kettunen MI, Hu DE, Jensen PR, Zandt RI, Karlsson M, Gisselsson A, Nelson SK, Witney TH, Bohndiek SE, Hansson G, Peitersen T, Lerche MH, Brindle KM. Production of hyperpolarized [1,4–13C2]malate from [1,4–13C2]fumarate is a marker of cell necrosis and treatment response in tumors. Proc Natl Acad Sci USA 2009; 106: 1980119806.
  • 41
    Jensen PR, Peitersen T, Karlsson M, in 't Zandt R, Gisselsson A, Hansson G, Meier S, Lerche MH. Tissue-specific short chain fatty acid metabolism and slow metabolic recovery after ischemia from hyperpolarized NMR in vivo. J Biol Chem 2009; 284: 3607736082.
  • 42
    Keshari KR, Wilson DM, Chen AP, Bok R, Larson PE, Hu S, Van Criekinge M, Macdonald JM, Vigneron DB, Kurhanewicz J. Hyperpolarized [2–13C]-fructose: a hemiketal DNP substrate for in vivo metabolic imaging. J Am Chem Soc 2009; 131: 1759117596.
  • 43
    Karlsson M, Jensen PR, in 't Zandt R, Gisselsson A, Hansson G, Duus JO, Meier S, Lerche MH. Imaging of branched chain amino acid metabolism in tumors with hyperpolarized 13C ketoisocaproate. Int J Cancer 2010; 127: 729736.
  • 44
    Gallagher FA, Kettunen MI, Day SE, Lerche M, Brindle KM. 13C Magnetic resonance spectroscopy measurements of glutaminase activity in human hepatocellular carcinoma cells using hyperpolarized 13C-labeled glutamine. Magn Reson Med 2008; 60: 253257.
  • 45
    Gallagher FA, Kettunen MI, Day SE, Hu DE, Karlsson M, Gisselsson A, Lerche MH, Brindle KM. Detection of tumor glutamate metabolism in vivo using 13C magnetic resonance spectroscopy and hyperpolarized [1–13C]glutamate. Magn Reson Med 2011 [Epub ahead of print].
  • 46
    von Morze C, Larson PE, Hu S, Keshari K, Wilson DM, Ardenkjaer-Larsen JH, Kurhanewicz J, Vigneron DB. Imaging of blood flow using hyperpolarized 13C-urea in preclinical murine models. Proc Int Soc Magn Reson Med 2010; 18: 567.
  • 47
    Thaning M, Servin R. Method of dynamic nuclear polarisation (DNP) using a trityl radical and a paramagnetic metal ion. US Patent PCT/NO2008/000449, 2007.
  • 48
    Harrison CE, DeBerardinis RJ, Jindal AK, Yang C, Sherry AD, Malloy CR. Analysis of mitochondrial metabolism in cancer cells by combining hyperpolarization and isotopomer analysis. Proc Int Soc Magn Reson Med 2010; 18: 569.
  • 49
    Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, Dancey J, Arbuck S, Gwyther S, Mooney M, Rubinstein L, Shankar L, Dodd L, Kaplan R, Lacombe D, Verweij J. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45: 228247.
  • 50
    Zierhut ML, Yen YF, Chen AP, Bok R, Albers MJ, Zhang V, Tropp J, Park I, Vigneron DB, Kurhanewicz J, Hurd RE, Nelson SJ. Kinetic modeling of hyperpolarized 13C1-pyruvate metabolism in normal rats and TRAMP mice. J Magn Reson 2010; 202: 8592.
  • 51
    Brindle K. New approaches for imaging tumour responses to treatment. Nat Rev Cancer 2008; 8: 94107.
  • 52
    Hamstra DA, Rehemtulla A, Ross BD. Diffusion magnetic resonance imaging: a biomarker for treatment response in oncology. J Clin Oncol 2007; 25: 41044109.
  • 53
    Lyng H, Haraldseth O, Rofstad EK. Measurement of cell density and necrotic fraction in human melanoma xenografts by diffusion weighted magnetic resonance imaging. Magn Reson Med 2000; 43: 828836.
  • 54
    Gillies RJ, Raghunand N, Karczmar GS, Bhujwalla ZM. MRI of the tumor microenvironment. J Magn Reson Imaging 2002; 16: 430450.
  • 55
    Cardone RA, Casavola V, Reshkin SJ. The role of disturbed pH dynamics and the Na+/H+ exchanger in metastasis. Nat Rev Cancer 2005; 5: 786795.
  • 56
    Gillies RJ, Raghunand N, Garcia-Martin ML, Gatenby RA. pH imaging. A review of pH measurement methods and applications in cancers. IEEE Eng Med Biol Mag 2004; 23: 5764.
  • 57
    Joint Formulary Committee, National Health Service, U.K. British National Formulary, Vol. 60; 2010. Available at:
  • 58
    in 't Zandt R. Metabolic reactions studied with 13C-DNP-MR in vitro and in vivo. Proc Int Soc Magn Reson Med 2008; 16: 827.
  • 59
    Douard V, Ferraris RP. Regulation of the fructose transporter GLUT5 in health and disease. Am J Physiol Endocrinol Metab 2008; 295: E227-E237.
  • 60
    Tong X, Zhao F, Thompson CB. The molecular determinants of de novo nucleotide biosynthesis in cancer cells. Curr Opin Genet Dev 2009; 19: 3237.
  • 61
    Weggen S, Preuss U, Pietsch T, Hilger N, Klawitz I, Scheidtmann KH, Wiestler OD, Bayer TA. Identification of amplified genes from SV40 large T antigen-induced rat PNET cell lines by subtractive cDNA analysis and radiation hybrid mapping. Oncogene 2001; 20: 20232031.
  • 62
    Bode BP, Souba WW. Modulation of cellular proliferation alters glutamine transport and metabolism in human hepatoma cells. Ann Surg 1994; 220: 411422; discussion 422–414.
  • 63
    Wilson DM, Hurd RE, Keshari K, Van Criekinge M, Chen AP, Nelson SJ, Vigneron DB, Kurhanewicz J. Generation of hyperpolarized substrates by secondary labeling with [1,1–13C] acetic anhydride. Proc Natl Acad Sci USA 2009; 106: 55035507.
  • 64
    Veech RL, Lawson JW, Cornell NW, Krebs HA. Cytosolic phosphorylation potential. J Biol Chem 1979; 254: 65386547.
  • 65
    Williamson DH, Lund P, Krebs HA. The redox state of free nicotinamide-adenine dinucleotide in the cytoplasm and mitochondria of rat liver. Biochem J 1967; 103: 514527.
  • 66
    Kettunen MI, Hu DE, Witney TH, McLaughlin R, Gallagher FA, Bohndiek SE, Day SE, Brindle KM. Magnetization transfer measurements of exchange between hyperpolarized [1–13C]pyruvate and [1–13C]lactate in a murine lymphoma. Magn Reson Med 2010; 63: 872880.
  • 67
    Harris T, Eliyahu G, Frydman L, Degani H. Kinetics of hyperpolarized 13C1-pyruvate transport and metabolism in living human breast cancer cells. Proc Natl Acad Sci USA 2009; 106: 1813118136.
  • 68
    Brindle KM, Campbell ID, Simpson RJ. A 1H n.m.r. study of the kinetic properties expressed by glyceraldehyde phosphate dehydrogenase in the intact human erythrocyte. Biochem J 1982; 208: 583592.
  • 69
    Larson PE, Hurd R, Kerr AB, Bok R, Kurhanewicz J, Vigneron D. Stimulated-echo contrast with hyperpolarized [1–13C]-pyruvate. Proc Int Soc Magn Reson Med 2010; 375.
  • 70
    Xu T, Mayer D, Gu M, Yen Y-F, Johansson E, Tropp J, Hurd R, Spielman DM. Measuring Michaelis-Menten kinetics of hyperpolarized 13C1-pyruvate metabolism in a single bolus injection. Proc Int Soc Magn Reson Med 2009; 17: 58.
  • 71
    Simpson RJ, Brindle KM, Brown FF, Campbell ID, Foxall DL. Studies of lactate dehydrogenase in the purified state and in intact erythrocytes. Biochem J 1982; 202: 581587.
  • 72
    Simpson RJ, Brindle KM, Brown FF, Campbell ID, Foxall DL. A p.m.r. isotope-exchange method for studying the kinetic properties of dehydrogenases in intact cells. Biochem J 1982; 202: 573579.
  • 73
    Silverstein E, Boyer PD. Equilibrium reaction rates and the mechanisms of bovine heart and rabbit muscle lactate dehydrogenases. J Biol Chem 1964; 239: 39013907.
  • 74
    Brindle KM. NMR methods for measuring enzyme-kinetics in vivo. Prog NMR Spectrosc 1988; 20: 257293.
  • 75
    Nagashima K. Optimum pulse flip angles for multi-scan acquisition of hyperpolarized NMR and MRI. J Magn Reson 2008; 190: 183188.
  • 76
    Keshari KR, Kurhanewicz J, Jeffries RE, Wilson DM, Dewar BJ, Van Criekinge M, Zierhut M, Vigneron DB, Macdonald JM. Hyperpolarized (13)C spectroscopy and an NMR-compatible bioreactor system for the investigation of real-time cellular metabolism. Magn Reson Med 2010; 63: 322329.
  • 77
    Merritt ME, Harrison C, Storey C, Jeffrey FM, Sherry AD, Malloy CR. Hyperpolarized 13C allows a direct measure of flux through a single enzyme-catalyzed step by NMR. Proc Natl Acad Sci USA 2007; 104: 1977319777.
  • 78
    Cunningham CH, Chen AP, Albers MJ, Kurhanewicz J, Hurd RE, Yen YF, Pauly JM, Nelson SJ, Vigneron DB. Double spin-echo sequence for rapid spectroscopic imaging of hyperpolarized 13C. J Magn Reson 2007; 187: 357362.
  • 79
    Mayer D, Levin YS, Hurd RE, Glover GH, Spielman DM. Fast metabolic imaging of systems with sparse spectra: application for hyperpolarized 13C imaging. Magn Reson Med 2006; 56: 932937.
  • 80
    Schulte RF, Wiesinger F, Fish KM, Whitt D, Hancu I. Efficient hyperpolarised 13C metabolic imaging with rosette spectroscopic imaging. Proc Int Soc Magn Reson Med 2009; 17: 2433.
  • 81
    Leupold J, Mansson S, Petersson JS, Hennig J, Wieben O. Fast multiecho balanced SSFP metabolite mapping of (1)H and hyperpolarized (13)C compounds. MAGMA 2009; 22: 251256.
  • 82
    Yen YF, Kohler SJ, Chen AP, Tropp J, Bok R, Wolber J, Albers MJ, Gram KA, Zierhut ML, Park I, Zhang V, Hu S, Nelson SJ, Vigneron DB, Kurhanewicz J, Dirven HA, Hurd RE. Imaging considerations for in vivo 13C metabolic mapping using hyperpolarized 13C-pyruvate. Magn Reson Med 2009; 62: 110.
  • 83
    Arunachalam A, Whitt D, Fish K, Giaquinto R, Piel J, Watkins R, Hancu I. Accelerated spectroscopic imaging of hyperpolarized C-13 pyruvate using SENSE parallel imaging. NMR Biomed 2009; 22: 867873.
  • 84
    Chen AP, Hurd RE, Cunningham CH. Non-Fourier spatial encoding for improved point-spread-functions in hyperpolarized 13C CSI acquisitions. Proc Int Soc Magn Reson Med 2010; 18: 3262.
  • 85
    Mayer D, Yen YF, Levin YS, Tropp J, Pfefferbaum A, Hurd RE, Spielman DM. In vivo application of sub-second spiral chemical shift imaging (CSI) to hyperpolarized 13C metabolic imaging: comparison with phase-encoded CSI. J Magn Reson 2010; 204: 340345.
  • 86
    Mayer D, Yen YF, Tropp J, Pfefferbaum A, Hurd RE, Spielman DM. Application of subsecond spiral chemical shift imaging to real-time multislice metabolic imaging of the rat in vivo after injection of hyperpolarized 13C1-pyruvate. Magn Reson Med 2009; 62: 557564.
  • 87
    Larson PE, Bok R, Kerr AB, Lustig M, Hu S, Chen AP, Nelson SJ, Pauly JM, Kurhanewicz J, Vigneron DB. Investigation of tumor hyperpolarized [1–13C]-pyruvate dynamics using time-resolved multiband RF excitation echo-planar MRSI. Magn Reson Med 2010; 63: 582591.
  • 88
    Levin YS, Mayer D, Yen YF, Hurd RE, Spielman DM. Optimization of fast spiral chemical shift imaging using least squares reconstruction: application for hyperpolarized (13)C metabolic imaging. Magn Reson Med 2007; 58: 245252.
  • 89
    Reeder SB, Brittain JH, Grist TM, Yen YF. Least-squares chemical shift separation for (13)C metabolic imaging. J Magn Reson Imaging 2007; 26: 11451152.
  • 90
    Brodsky EK, Holmes JH, Yu H, Reeder SB. Generalized k-space decomposition with chemical shift correction for non-Cartesian water-fat imaging. Magn Reson Med 2008; 59: 11511164.
  • 91
    Wiesinger F, Menzel MI, Weidl E, Janich M, Schwaiger M, Schulte RF. Minimum-norm IDEAL spiral CSI for efficient hyperpolarized 13C metabolic imaging. Proc Int Soc Magn Reson Med 2010; 18: 3290.
  • 92
    Perman WH, Bhattacharya P, Leupold J, Lin AP, Harris KC, Norton VA, Hovener JB, Ross BD. Fast volumetric spatial-spectral MR imaging of hyperpolarized 13C-labeled compounds using multiple echo 3D bSSFP. Magn Reson Imaging 2010; 28: 459465.
  • 93
    Larson PE, Hu S, Lustig M, Kerr AB, Nelson SJ, Kurhanewicz J, Pauly JM, Vigneron DB. Fast dynamic 3D MR spectroscopic imaging with compressed sensing and multiband excitation pulses for hyperpolarized (13)C studies. Magn Reson Med 2011; 65: 610619.
  • 94
    Hu S, Lustig M, Balakrishnan A, Larson PE, Bok R, Kurhanewicz J, Nelson SJ, Goga A, Pauly JM, Vigneron DB. 3D compressed sensing for highly accelerated hyperpolarized (13)C MRSI with in vivo applications to transgenic mouse models of cancer. Magn Reson Med 2010; 63: 312321.
  • 95
    Cunningham CH, Chen AP, Lustig M, Hargreaves BA, Lupo J, Xu D, Kurhanewicz J, Hurd RE, Pauly JM, Nelson SJ, Vigneron DB. Pulse sequence for dynamic volumetric imaging of hyperpolarized metabolic products. J Magn Reson 2008; 193: 139146.
  • 96
    Lau AZ, Chen AP, Ghugre NR, Ramanan V, Lam WW, Connelly KA, Wright GA, Cunningham CH. Rapid multislice imaging of hyperpolarized (13)C pyruvate and bicarbonate in the heart. Magn Reson Med 2010; 64: 13231331.
  • 97
    Lupo JM, Chen AP, Zierhut ML, Bok RA, Cunningham CH, Kurhanewicz J, Vigneron DB, Nelson SJ. Analysis of hyperpolarized dynamic 13C lactate imaging in a transgenic mouse model of prostate cancer. Magn Reson Imaging 2010; 28: 153162.
  • 98
    Pauly J, Le Roux P, Nishimura D, Macovski A. Parameter relations for the Shinnar-Le Roux selective excitation pulse design algorithm [NMR imaging]. IEEE Trans Med Imaging 1991; 10: 5365.
  • 99
    Meyer CH, Pauly JM, Macovski A, Nishimura DG. Simultaneous spatial and spectral selective excitation. Magn Reson Med 1990; 15: 287304.
  • 100
    Lee KJ. General parameter relations for the Shinnar-Le Roux pulse design algorithm. J Magn Reson 2007; 186: 252258.
  • 101
    Sukumar S, Larson PEZ, Keshari KR, Kurhanewicz J, Vigneron DB. Single shot, chemical shift specific imaging methods for hyperpolarized carbon-13 studies at 14T. Proc Int Soc Magn Reson Med 2010; 18: 3268.
  • 102
    Levitt MH. Singlet and other states with extended lifetimes. In: HarrisRK, WasylishenRE, editors. Encyclopedia in nuclear magnetic resonance. Wiley; 2010.
  • 103
    Harris T, Giraudeau P, Frydman L. Indirect detection of enzymatic processes by hyperpolarized NMR: temporal information, enhanced spectral resolution and slow spin relaxation. Proc Int Soc Magn Reson Med 2010; 571.
  • 104
    Navon G, Song Y-Q, Room T, Appelt S, Taylor RE, Pines A. Enhancement of solution NMR and MRI with laser-polarized xenon. Science 1996; 271: 18481851.
  • 105
    Merritt ME, Harrison C, Mander W, Malloy CR, Sherry AD. Dipolar cross-relaxation modulates signal amplitudes in the (1)H NMR spectrum of hyperpolarized [(13)C]formate. J Magn Reson 2007; 189: 280285.
  • 106
    Vasos PR, Comment A, Sarkar R, Ahuja P, Jannin S, Ansermet JP, Konter JA, Hautle P, van den Brandt B, Bodenhausen G. Long-lived states to sustain hyperpolarized magnetization. Proc Natl Acad Sci USA 2009; 106: 1846918473.
  • 107
    Carravetta M, Johannessen OG, Levitt MH. Beyond the T1 limit: singlet nuclear spin states in low magnetic fields. Phys Rev Lett 2004; 92: 153003.
  • 108
    Pileio G, Carravetta M, Levitt MH. Storage of nuclear magnetization as long-lived singlet order in low magnetic field. Proc Natl Acad Sci USA 2010; 107: 1713517139.
  • 109
    Carravetta M, Levitt MH. Theory of long-lived nuclear spin states in solution nuclear magnetic resonance. I. Singlet states in low magnetic field. J Chem Phys 2005; 122: 214505.
  • 110
    Pileio G, Levitt MH. Theory of long-lived nuclear spin states in solution nuclear magnetic resonance. II. Singlet spin locking. J Chem Phys 2009; 130: 214501.
  • 111
    Sarkar R, Ahuja P, Vasos PR, Bodenhausen G. Measurement of slow diffusion coefficients of molecules with arbitrary scalar couplings via long-lived spin states. Chemphyschem 2008; 9: 24142419.
  • 112
    Sarkar R, Vasos PR, Bodenhausen G. Singlet-state exchange NMR spectroscopy for the study of very slow dynamic processes. J Am Chem Soc 2007; 129: 328334.
  • 113
    Pileio G, Carravetta M, Hughes E, Levitt MH. The long-lived nuclear singlet state of 15N-nitrous oxide in solution. J Am Chem Soc 2008; 130: 1258212583.
  • 114
    Warren WS, Jenista E, Branca RT, Chen X. Increasing hyperpolarized spin lifetimes through true singlet eigenstates. Science 2009; 323: 17111714.
  • 115
    Jenista E, Branca R, Bhattacharyya D, Chen X, Warren W. Extending the utility of hyperpolarized compounds by storing polarization in the singlet state. Proc Int Soc Magn Reson Med 2009; 57.
  • 116
    Harbeck N. Never too late: reducing late breast cancer relapse risk. Curr Med Res Opin 2008; 24: 32953305.