• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jul 25, 2006; 103(30): 11270–11275.
Published online Jul 12, 2006. doi:  10.1073/pnas.0601319103
PMCID: PMC1544077
Applied Physical Sciences, Medical Sciences

Real-time metabolic imaging

Abstract

The endogenous substance pyruvate is of major importance to maintain energy homeostasis in the cells and provides a window to several important metabolic processes essential to cell survival. Cell viability is therefore reflected in the metabolism of pyruvate. NMR spectroscopy has until now been the only noninvasive method to gain insight into the fate of pyruvate in the body, but the low NMR sensitivity even at high field strength has only allowed information about steady-state conditions. The medically relevant information about the distribution, localization, and metabolic rate of the substance during the first minute after the injection has not been obtainable. Use of a hyperpolarization technique has enabled 10–15% polarization of 13C1 in up to a 0.3 M pyruvate solution. i.v. injection of the solution into rats and pigs allows imaging of the distribution of pyruvate and mapping of its major metabolites lactate and alanine within a time frame of ≈10 s. Real-time molecular imaging with MRI has become a reality.

Keywords: 13C, dynamic nuclear polarization, hyperpolarized, MRI, spectroscopy

The technique for increase in signal-to-noise ratio of >10,000 times in liquid-state NMR and the use of this technique for molecular imaging with endogenous substances generated through the process of dynamic nuclear polarization (DNP) has been reported (1, 2). In both these studies, 13C-enriched urea was used as an example of an endogenous substance that could be polarized to a high degree (37% for 13C and 8% for 15N) and used for high-resolution imaging of the cardiovascular system in rats.

It was suggested that the signal enhancement could be used not only for visualizing the cardiovascular system but also for improved perfusion measurements and that it may allow real-time metabolic mapping of other endogenous substances such as alanine, glutamine, and acetate. Such studies should be possible if the relaxation time of the 13C-labeled site in the hyperpolarized molecule is long enough and the metabolic products retain sufficient fraction of the nonequilibrium polarization. The possibilities for doing perfusion studies using 13C labeled hyperpolarized substances have recently been reviewed by Månsson et al. (3), but the application of visualizing metabolic processes by using hyperpolarized substances have not yet been described.

To reveal information about the metabolic status of the tissue, magnetic resonance (MR) spectroscopy has been used, employing nuclei like 1H, 13C, 31P, and 19F (4, 5). The main application areas have been brain, muscle, and prostate tissue. Information on fluxes through metabolic pathways is less straightforward to obtain though. Traditionally, 13C NMR spectroscopy in combination with 13C-labeled (enriched) substrates has been used to visualize the label applied, its metabolic intermediates, and/or its end products during steady-state conditions. In certain cases the metabolic rates can be indirectly estimated by using mathematical modeling (6), for example, in determining the flux through the tricarboxylic acid cycle in vivo (710).

As reviewed by Ross et al. (11) injection of 13C-labeled glucose and acetate now can uncover hitherto unknown disorders of N-acetyl aspartate (NAA) synthesis, tricarboxylic acid cycle, and glycolysis by using clinical 13C NMR (1.5 T) spectroscopy in man. The inherent low intrinsic sensitivity of 13C NMR in combination with low in vivo concentrations for relevant metabolites resulted in a need to sample for several minutes, i.e., 35 min (12). If information about flux rates is wanted, one needs to follow the fate of 13C-labeled substrate up to hours after injection (9, 10).

The increased signal amplitude available by the hyperpolarization method should allow the sampling times to be reduced to seconds. This reduction would open up a completely different perspective on metabolic mapping using NMR: detection of newly formed 13C compounds within minutes after injection of the 13C-hyperpolarized label with high image quality and no background interference. To examine the potential of the hyperpolarization technique for visualization of in vivo metabolic processes, we have chosen the substance pyruvate-13C enriched in the C1 position. Pyruvate is an intermediate common to three major metabolic and catabolic pathways in the mammalian cells (Fig. 1). Depending on the intracellular energy status of the tissue, pyruvate will be converted to a different degree into alanine, lactate, or carbon dioxide.

Fig. 1.
Simplified overview of the main metabolic pathways of pyruvate.

After an i.v. injection, pyruvate is rapidly distributed in the body and absorbed by the cells of most organs. It is known that only insignificant amounts of the injected pyruvate leaves the body via the normal excretory pathways, the bile and urine (13, 14). Consequently, pyruvate is completely metabolized within a short time after its injection.

The relative amount of metabolites produced from the injected pyruvate will depend on the actual condition of the cells and a number of basic cell viability parameters such as pO2, pH, and need for protein synthesis. If one therefore could determine not only pyruvate metabolite levels but also metabolic rate in vivo, it would be of great medical diagnostic importance for organ function and disease quantification. It is the aim of this work to evaluate whether the metabolic fate of injected pyruvate can be mapped and followed within the medically relevant time frame of seconds.

It is important to realize that hyperpolarized samples are characterized by a strong nonthermal polarization and that the hyperpolarization is created ex situ to the examined object. The nonthermal polarization condition cannot be regenerated after the NMR investigation. Characteristic for hyperpolarized samples is that during and after dissolution of the sample into the liquid space, the nuclear polarization will decrease according to the longitudinal relaxation time T1 (3). In addition, the distribution in the body will lead to a decrease of concentration of the agent depending on body size. Therefore, we conducted the experiments in both rats and pigs where the pharmacokinetic parameters widely differ. It will be shown that in both species it is possible to map pyruvate and some of its metabolic products in resting skeletal muscle within a clinically useful time frame of ≈10 s.

Results and Discussion

Most applications in in vivo NMR spectroscopy are performed from NMR signals of carbon-bound “nonexchangeable” protons in small molecules (Mw < 500). In general, several minutes are used for signal averaging to obtain sufficient signal-to-noise ratio to obtain an interpretable NMR spectrum. The spectra therefore informs in most applications about the steady-state situation only. Endogenous pyruvate concentrations in plasma are between 0.1 and 0.2 mM and thus currently challenging to examine with conventional 1H NMR spectroscopic techniques. Especially because of the high H-background signal, it is impossible to study the metabolism of pyruvate and its metabolic rate within a clinically interesting time frame using 1H NMR spectroscopy. The 13C spectral region is ≈10 times that of proton, but because of the low gyromagnetic ratio (1/4 of 1H) the intrinsic 13C NMR sensitivity is nearly 2 orders of magnitude lower than that of 1H.

To examine the capability of monitoring the metabolic fate of pyruvate in vivo in a noninvasive manner, 13C1-enriched hyperpolarized pyruvate (0.79 mmol/kg) was injected, and the 13C NMR spectra were acquired from the time of injection for 50 s. By injecting the hyperpolarized 13C1-enriched pyruvate in a rat, we obtained a NMR spectrum (Fig. 2) showing the fate of pyruvate in the lower half of the rat during the first 50 s after the injection. The dynamic of the production of its main metabolic products lactate and alanine can be followed as well as the rate of the 13C NMR signal disappearance. When working with hyperpolarized substances, the signal decay caused by the excitation pulses and the respective T1 of the formed products must be considered. An estimation of the in vivo T1 value of the 13C1 of pyruvate, lactate, and alanine shows that they are all within 15 ± 5 s in the rat and 20 ± 5 s in the pig. In addition to the metabolites, the pyruvate hydrate [CH3C(OH)CO213COO] also can be seen in the NMR spectra. Pyruvate hydrate, which is not metabolically active, is formed (<8%) under the conditions of pH 7.5–8.2 in the injection solution.

Fig. 2.
Metabolic production of lactate and alanine after the injection of 13C1-pyruvate. (A and B) The spectra (B) are acquired with a time interval of 3 s from the lower part of the animal as indicated by the proton MR image (A). (C) The formation of bicarbonate ...

The third main metabolite formed from 13C1-pyruvate is 13CO2, which is in rapid equilibrium with H13CO3. We could detect the H13CO3 signal only after all of the spectra acquired during the 50-s time window (Fig. 2) were added. The relative low amounts of detectable H13CO3 could be because of a short T1 of this compound, although experiments indicate that the T1 of bicarbonate should be close to the T1 of pyruvate (K.G. and R.i.t.Z., unpublished results). It is more likely that the skeletal muscle tissue that constitutes most of the volume covered by the radiofrequency (rf)-coil does not metabolize pyruvate to a large extent through the tricarboxylic acid cycle in the resting muscular tissue of the anesthetized animal under the experimental conditions (15, 16). Rates for pyruvate oxidation and formation of lactate and alanine have been calculated by using 13C-glucose as substrate (17). This kind of data has been obtained under steady-state conditions by using physiological levels of substrate. It is possible that a different metabolic pattern can be found when using high levels of 13C-pyruvate to visualize metabolism under non-steady-state conditions with a short time window. The ratio between CO2 and HCO3 in tissue is directly related to the pH, and a low concentration of HCO3 therefore also may be explained by a low pH in the examined area.

The spectra presented in Fig. 2 are affected by the fact that they cover most of the rat and do not inform about the relative contribution to the metabolism from the different organs in the body.

Spatial information can be obtained by using, for example, a chemical shift imaging (CSI) sequence (18, 19). Every excitation rf-pulse irreversibly destroys a certain fraction of the longitudinal magnetization; a 90° pulse, for example, uses all of the magnetization available within the excited slice in one single rf-pulse. For this reason a compromise should be made between the rf-pulse angle and the matrix size, e.g., the amount of rf-pulses needed to acquire a single CSI. It is feasible to acquire five chemical shift images with a modest matrix size of 12 × 12 and at the same time increase the rf-pulse angle with every subsequent measurement to compensate for the loss of signal over time.

CSI experiments in rats injected with hyperpolarized 13C1-enriched pyruvate were carried out to examine the differences in the metabolic transformation of pyruvate to the various oxidative, reductive, or transaminated products. The result of such an approach is shown in Fig. 3 where the CSI allows us to follow the fate of the injected pyruvate. (i) About 13 s after the injection, most of the signal originates from pyruvate located in the central vascular structures. (ii) At 21 s pyruvate has distributed into the whole body and already been metabolized in significant amounts to alanine and lactate. (iii) At 37 s even more has been metabolized, and it can be seen that the lumbar muscles are the most active areas for metabolism.

Fig. 3.
The time course of the build up of lactate and alanine in the imaged slice of the rat. (A) The location of the image slab in the rat and the corresponding transversal 1H NMR image are illustrated. (B) The NMR signal obtained simultaneously from pyruvate, ...

The strength of the NMR modality using the CSI sequence is that the chemical shift allows us to identify the metabolic product and at the same time localize them within the body organs. The results in Fig. 3B where the metabolite concentration changes with time indicate that metabolic rate measurements are possible, provided that the relaxation rates of all metabolites involved are known. It also shows that it is possible to simultaneously localize 13C1-pyruvate and the production of its metabolites in the rat with a spatial resolution of 7.5 × 7.5 × 32 mm3 and a time resolution of 8 s at a 1.5 T clinical NMR machine. The color scale used in Fig. 3B might suggest the lack of pyruvate in the animal in the 37-s image. To illustrate the presence of pyruvate in the whole image (t = 37 s), the pyruvate intensity has been rescaled with a factor of 1 → 20 in Fig. 3C. Even after 37 s the amplitude of pyruvate is higher than that of the metabolites in any voxel.

If one acquires the entire signal during the period 30–43 s, a higher spatial resolution (5 × 5 × 10 mm3) can be obtained (Fig. 4) at the expense of the metabolic rate information. These results demonstrate the previously undescribed feasibility to monitor the metabolic transformation of pyruvate, a key species involved in cellular energetics, noninvasively, in the region of interest, within a clinically interesting time frame, and coregistration of such maps with the anatomy.

Fig. 4.
Acquisition of a single CSI image at a higher matrix size of 16 × 16 and a field of view of 80 × 80 × 10 mm. The location of the image slab in the rat and the corresponding transversal 1H NMR image are illustrated. The NMR signal ...

To show that the method is also applicable in a larger animal, we decided to perform a nearly similar study as performed in rats in pigs. Such a large animal would be better suited in a clinical NMR system. It would provide important information as to whether the 13C-pyruvate dynamics also could be visualized in a more clinically relevant setting. The examination area chosen in the pig covers mainly muscular tissue in the legs as depicted from the standard proton-based image (Fig. 5A). If one measures the NMR signal amplitude generated at the different frequencies characteristic for the 13C1 in pyruvate, alanine, lactate, and pyruvate hydrate, it can be seen (Fig. 5B) that the time window of 30–45 s after the injection is an optimum time for imaging the metabolites. Consequently, a single CSI was collected during this time window. Fig. 5C shows the localization and indicates the relative concentration of pyruvate, alanine, and lactate in the chosen part of the legs of the pig. We thus can noninvasively get localized information about the metabolic status of the muscular tissue of interest. The low or missing signal in the upper part of the right leg is due to the fact that a femoral arterial blood sampling catheter limits the delivery of pyruvate in this area.

Fig. 5.
Pyruvate and its metabolism in the hind leg of a pig. (A) Unlocalized 13C NMR spectra were acquired from the lower legs of the pig as illustrated in the proton projection image. (B) The signal amplitudes of pyruvate, pyruvate hydrate, alanine, and lactate ...

Pyruvate is a molecule central to delivering energy to the body cells, and the metabolism is under control of a range of different physiological conditions (2022). Despite this control, it is possible to create metabolic images of 13C-lacate and 13C-alanine within a minute after injection of supraphysiological levels of 13C-pyruvate. General changes in organ function will affect the energy consumption (viability) of the cells and consequently the metabolism of pyruvate. These changes will be reflected in the capacity of the cells to label the existing lactate and alanine levels with 13C and the production of these metabolites. Ability to localize quantity and determine metabolic rate of pyruvate may be of importance for diagnosis and treatment of medical diseases. In this study, we did not pay attention to quantification of metabolite concentration but focused on the aim to prove that real molecular imaging with MRI within a clinically interesting time frame has become a reality.

Materials and Methods

Endogenous Substance.

The 13C1-labeled pyruvate was polarized in its acid form with the electron paramagnetic agent Tris(8-carboxy-2,2,6,6,-tetra(methoxyethyl)benzo[1,2-d:4,5d′]bis(1, 3)dithiole-4-yl)methyl sodium salt present in a 15 mM concentration in the neat acid. The polarization and subsequent dissolution of the substance were performed in a manner similar to the process described in ref. 2. After the polarization process (60 min of microwave irradiation at 1.2 K), the polarized sample of 40 mg (31.5 μl) labeled pyruvic acid for rat and 500 mg labeled pyruvic acid for pig was thawed, dissolved, and neutralized within 2 s by using 5.7 ml of a heated aqueous buffer for dissolving the rat sample and 18.7 ml of a heated aqueous buffer for dissolving the pig sample. The final injection solution for rat contained 78.8 mM pyruvate, 68.8 mM sodium ion, 20 mM Tris buffer, 0.27 mM Na2EDTA, and 83 μM paramagnetic agent. For pig the final injection solution contained 300 mM pyruvate, 100 mM Tris buffer, 0.27 mM Na2EDTA, 250 mM sodium ion, and 0.32 mM paramagnetic agent. The injectant temperature was ≈30°C, and pH was 7.5–8.2. The polarization measured immediately after the dissolution was 15–20% and was measured in a homebuilt polarimeter. The T1 of the 13C pyruvate was 55 s, and the transfer time to the imaging magnet was 15–20 s, resulting in a polarization of 10–15% at the moment of injection.

Dosing.

Hyperpolarized 13C-pyruvate was injected at a dose of 0.79 mmol/kg for rats, whereas for pigs the dose was 0.2 mmol/kg. The injection volume was 3 ml for the rat, and the injection rate was 0.25 ml/s, whereas in the pig 16 ml of the solution was injected at an injection rate of 1.25 ml/s. This technique results in a total injection time of 12 s in both species.

Animal Handling.

Male Wistar rats (300–350 g) were anesthetized by using isoflurane (2–3%) in 97% oxygen. A catheter was introduced into the tail vein, and another catheter was inserted into A. carotis communis sinistra. They were placed on a home-built pad that was heated to ≈37°C by means of circulating FC-104 Fluorinert to avoid background signals in the MR experiments. Body temperature was constantly kept at 37°C. Anesthesia was continued by means of the isoflurane mixture. The arterial catheter was connected via a T-tube to a pressure recorder and a pump delivering saline (0.15 ml·min−1) to prevent catheter clotting. After the examination, the animals were killed by a lethal injection of pentobarbital.

Pig Experiment.

A Swedish land pig (25 kg) was premedicated with 10 ml of Ketalar (50 mg/ml; Warner–Lambert, Ann Arbor, MI) and 1.5 ml of Midazolan (5 mg/ml; Pharma Hameln, Hameln, Germany) intramuscularly. The pig was tracheally intubated; two catheters were inserted i.v., one in the hind leg for administration of anesthesia and one in the front leg for administration of Ringer-acetate solution (150 ml/h) for hydration and injection of the 13C-pyruvate. Full anesthesia was induced by using an injection of 0.5 ml/kg Pentothal natrium (25 mg/ml; Abbott). The pig was connected to a volume-controlled respirator (PV 301A; Breas Medical AB, Mölnlycke, Sweden; 6–8 ml/kg; 20 breaths per min). Blood pressure and heart rate were continuously recorded through a catheter in A. carotis, and body temperature was measured through a rectal probe. Anesthesia during the examination was maintained by using a mixture containing isotonic NaCl (26 vol%), Ketalar (50 mg/ml; 42 vol%) (Pfizer AB, Sollentuna, Sweden), Norcuron (10 mg + 5 ml sterile water; 21 vol%) (Organon), and Midazolam (5 mg/ml; 11% vol) (Pharma Hameln) administered using an infusion pump at a rate of 0.6 ml/min. Injection of the hyperpolarized 13C-pyruvate was performed by a manual injection. After the examination, the animals were killed by a lethal injection of pentobarbital.

All animal experiments were approved by the local ethical committee.

MRI Equipment.

The spectroscopic and imaging experiments were performed on a 1.5-T clinical MRI (Sonata; Siemens) by using a 1H-13C Tx/Rx birdcage coil for the rat experiments (diameter 8.3 cm, length 10 cm; Rapid Biomedical, Rimpor, Germany) operating at 63.67 and 16.00 MHz, respectively. Pigs were positioned in a pig 13C-Tx/Rx NMR coil (diameter 26 cm, length 35 cm; Rapid Biomedical) and imaged with the proton body coil by using a standard proton NMR imaging sequence to ensure reproducible positioning of the hind legs of the pig.

NMR Protocol.

All automatic in line adjustments of the MR scanner were disabled to avoid the use of unwanted rf-pulses that would destroy the hyperpolarization signal. The 90° reference rf pulse was calibrated by using the natural abundance 13C-lipid signal. Based on the proton frequency as determined by the NMR system, the NMR frequency for 13C1-alanine was calculated according to frequency 13C1-alanine = 0.25144 × [(system frequency proton × 1.00021) − 0.000397708]. The frequency calculated will position the NMR signal arising from 13C1-alanine in the middle of the 13C-NMR spectrum with 13C1-lactate on the left and 13C1-pyruvate resonating on the right of 13C1-alanine. An unlocalized NMR spectroscopy sequence was run to check that the 13C-NMR coil and the system NMR frequency is setup correctly (bandwidth 10 KHz; 2K complex points; 90° rf-pulse; TR = 800 ms; 128 averages).

Unlocalized Free Induction Decay Experiments.

Pigs and rats were injected with 13C-pyruvate, and a series of 25 low-flip-angle (10°) unlocalized NMR spectra were acquired with an interval of 3 s, where the first NMR spectrum was acquired just before the injection of the 13C-pyruvate started. This finding gives insight in the metabolic rate of the 13C-labeled pyruvate and relative concentration of the metabolites in the whole animal. The 13C NMR signal obtained with the rat coil used covers the area from the kidneys down to the tail, whereas in case of experiments with pigs the 13C coil covers the lower legs of the animal. The 13C NMR spectra obtained therefore should be considered as a reflection of the average metabolism of pyruvate over the field of view.

Chemical-Shift Imaging Experiments.

For the chemical-shift imaging a standard sequence was used (Siemens V21B) except that centric K-space acquisition was added, and the possibility was created to be able to change the repetition time to be as short as possible as the timing of the sequence would allow.

Rat.

Chemical shift images were acquired with the 13C-image location positioned to cover the region of interest (80 × 80 × 32 mm3) with a matrix containing 12 × 12 elements. In the reconstruction phase, the matrix was zero-filled to 32 × 32. The injection of the pyruvate was started simultaneously with the start of the CSI sequence. In one version of the experiment, a total of five chemical shift images were acquired with a rf pulse angle of 2, 3, 4, 5, and 10°, respectively, with a time resolution of one chemical shift image of 8 s (Tr = 90 ms). In a second experiment, a single chemical shift image was acquired starting 30 s after start of the injection with a matrix size of 16 × 16 and a field of view of 80 × 80 × 10 mm3. The rf pulse angle used was 10°, and the total scan time 13.9 s (Tr = 90 ms).

Pig.

Chemical shift images were acquired with the same field of view as for the results of the unlocalized 13C NMR spectra experiments (250 × 250 × 80 mm3). NMR imaging resulted in a matrix containing 16 × 16 elements in which each element or voxel/pixel contains a 13C NMR spectrum. In the reconstruction phase, the matrix was zero-filled to 32 × 32. Thirty seconds after the start of the injection, i.e., 18 s after finishing the injection, the chemical shift 13C-NMR sequence was started (Tr = 90 ms; total acquisition time 13.9 s).

Analysis.

The series of 25 unlocalized 13C NMR spectra were analyzed by using the algorithm AMARES (23) as implemented in jmrui 2.2 (24, 25) with the following prior knowledge: Lorentzian line shape fitting; assumption of equal linewidth for alanine, lactate, pyruvate, and pyruvate hydrate; relative phase zero with respect to main phase; zero-order phase fixed, e.g., −30.1 for pig and −49.6 for rat free induction decay experiments; first-order phase fixed at 0.5 ms (delay between rf pulse and acquisition), 256 points in analysis; and to neglect the first 2 points of free induction decay in analysis. All graphs were made with graphpad prism 4.03 (GraphPad, San Diego).

The metabolic images were calculated from the CSI data set by using a home written program coded in matlab 6.5.1 (MathWorks, Natick, MA). This program incorporates time domain fitting algorithms (23). After manual phasing of the spectra, the amplitudes were estimated assuming constant phase; identical linewidth for alanine, lactate and pyruvate; and a fixed frequency shift between lactate and alanine (106 Hz) and pyruvate and alanine (−92 Hz). The amplitudes estimated for the metabolites pyruvate, alanine, and lactate are shown as color maps.

Acknowledgments

We thank Fredrik Ellner, Andreas Gram, Britt-Marie Lilja, Birgit Persson, and Kerstin Thyberg for excellent technical assistance and Mathilde Lerche for helpful discussions.

Abbreviations

CSI
chemical shift imaging
MR
magnetic resonance.

Footnotes

Conflict of interest statement: K.G., R.i.t.Z., and M.T. are employees of Amersham Health R&D AB Malmö, which is now part of GE Healthcare.

References

1. Ardenkjaer-Larsen J. H., Fridlund B., Gram A., Hansson G., Hansson L., Lerche M. H., Servin R., Thaning M., Golman K. Proc. Natl. Acad. Sci. USA. 2003;100:10158–10163. [PMC free article] [PubMed]
2. Golman K., Ardenkjaer-Larsen J. H., Petersson J. S., Mansson S., Leunbach I. Proc. Natl. Acad. Sci. USA. 2003;100:10435–10439. [PMC free article] [PubMed]
3. Månsson S., Johansson E., Magnusson P., Chai C. M., Hansson G., Petersson J. S., Stahlberg F., Golman K. Eur. Radiol. 2005;16:57–67. [PubMed]
4. Gadian D. G. NMR and Its Applications to Living Systems. Oxford: Oxford Univ. Press; 1995.
5. de Graaf R. A. In Vivo NMR Spectroscopy: Principles and Techniques. London: Wiley; 1998.
6. Wiechert W. Genet. Eng. (N.Y.) 2002;24:215–238. [PubMed]
7. Oz G., Berkich D. A., Henry P. G., Xu Y., LaNoue K., Hutson S. M., Gruetter R. J. Neurosci. 2004;24:11273–11279. [PubMed]
8. Sibson N. R., Dhankhar A., Mason G. F., Behar K. L., Rothman D. L., Shulman R. G. Proc. Natl. Acad. Sci. USA. 1997;94:2699–2704. [PMC free article] [PubMed]
9. Morris P., Bachelard H. NMR Biomed. 2003;16:303–312. [PubMed]
10. Bluml S., Moreno-Torres A., Shic F., Nguy C. H., Ross B. D. NMR Biomed. 2002;15:1–5. [PubMed]
11. Ross B., Lin A., Harris K., Bhattacharya P., Schweinsburg B. NMR Biomed. 2003;16:358–369. [PubMed]
12. Morikawa S., Inubushi T. J. Magn. Reson. Imaging. 2001;13:787–791. [PubMed]
13. Baverel G., Bonnard M., Pellet M. FEBS Lett. 1979;101:282–286. [PubMed]
14. Balagura-Baruch S., Burich R. L., King V. F. Am. J. Physiol. 1973;225:389–392. [PubMed]
15. Schadewaldt P., Munch U., Prengel M., Staib W. Biochem. Biophys. Res. Commun. 1983;116:456–461. [PubMed]
16. Schadewaldt P., Munch U., Staib W. Biochem. J. 1983;216:761–764. [PMC free article] [PubMed]
17. Jucker B. M., Rennings A. J., Cline G. W., Petersen K. F., Shulman G. I. Am. J. Physiol. 1997;273:E139–E148. [PubMed]
18. Maudsley A. A., Hilal A. K., Perman W. H., Simon H. E. J. Magn. Reson. 1983;51:147–152.
19. Brown T. R., Kincaid B. M., Ugurbil K. Proc. Natl. Acad. Sci. USA. 1982;79:3523–3526. [PMC free article] [PubMed]
20. Mallet R. T., Sun J., Knott E. M., Sharma A. B., Olivencia-Yurvati A. H. Exp. Biol. Med. (Maywood) 2005;230:435–443. [PubMed]
21. Sharma A. B., Knott E. M., Bi J., Martinez R. R., Sun J., Mallet R. T. Resuscitation. 2005;66:71–81. [PubMed]
22. Mongan P. D., Capacchione J., Fontana J. L., West S., Bunger R. Am. J. Physiol. 2001;281:H854–H864. [PubMed]
23. Vanhamme L., Sundin T., Hecke P. V., Huffel S. V. NMR Biomed. 2001;14:233–246. [PubMed]
24. Naressi A., Couturier C., Castang I., de Beer R., Graveron-Demilly D. Comput. Biol. Med. 2001;31:269–286. [PubMed]
25. Naressi A., Couturier C., Devos J. M., Janssen M., Mangeat C., de Beer R., Graveron-Demilly D. Magma. 2001;12:141–152. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • Compound
    Compound
    PubChem Compound links
  • PubMed
    PubMed
    PubMed citations for these articles
  • Substance
    Substance
    PubChem Substance links