NUCLEAR MAGNETIC RESONANCE (View Printable Version)

Introduction

Nuclear magnetic resonance (NMR) a theoretical project for physicists and an analytical tool for chemists entered the field of biochemistry in 1974 – 1975, when phosphorus (³¹P) spectra of intact rat muscle (Hoult et al., 1974), and intact frog muscle (Bárány et al., 1975) were recorded. The noninvasive nature of NMR makes it unique among biochemical methods, i.e. the information is obtained without destroying the tissue. Muscle was selected for the first NMR studies because animal tissues contain about 40% muscle which is located near the surface of the body and, therefore, it is easily accessible. ³¹P-NMR was selected, because muscle is rich in phosphate metabolites which play the key role in the mechanochemistry of muscle contraction. Subsequently, methods for ¹H-NMR and ¹³C-NMR were developed and magnetic resonance spectroscopy (MRS) was applied for several tissues. Thus, MRS became a popular tool in biochemistry, because it could analyze the composition of the tissues or the dynamics of their metabolism at the level of the live animal or even of a human being.

 Basis of NMR:  Many atomic nuclei possess an intrinsic angular momentum or spin. When the nuclei are placed in a strong magnetic field of an NMR spectrometer, the nuclei will align with the magnetic field (like tiny bar magnets) so that the nuclei are pointing either “toward” or “against” the direction of the magnetic field, corresponding to two energy levels of the nuclei. Then the nuclei are irradiated with such electromagnetic frequency that the photons (quanta) have energy exactly equal to the separation of the two levels of energy. If the nucleus is in the lower level of energy state it can be promoted to the upper level and vice versa, if it is in the higher level of energy state it can be stimulated by the irradiation to emit a photon and fall to the lower state. Quantum mechanics indicates that if a particle is moving with periodic motion (oscillating), then the particle can absorb electromagnetic radiation if the frequency of that motion exactly matches the frequency of the radiation. This is nuclear magnetic resonance that can be recorded by the NMR spectrometer as a peak in the spectrum.

Machines and Accessories

 
History

Figure NMR 1 depicts the scheme of the NMR spectrometer that we have used in 1974 to record the ³¹P spectra of a frog gastrocnemius muscle, freshly dissected.  The muscle was transferred into an NMR tube and placed between the north (N) and south (S) poles of a permanent magnet; it was spun at a low speed so that we could see the tube (CELL) by eye. The CELL, placed inside the radiofrequency coil of wire, was mounted centrally in the magnet. A radiofrequency transmitter (SOURCE) operating at ³¹P frequency of 36.43 MHz  supplied pulses of radiofrequency (rf) current in the radiofrequency coil, thus generating rf pulses which were experienced  by the phosphorus atoms in the gastrocnemius muscle. The coil was also connected to a radiofrequency receiver (DETECTOR) tuned to 36.43 MHz, which picked up the NMR signal. The detector was supposed to amplify the signal and passing it to a recorder for display. In our case it took 90 min to get a readable ³¹P spectrum.

Fig. NMR 1. NMR spectrometer in 1974.

Figure NMR 2 illustrates the ³¹P spectrum of a frog gastrocnemius muscle recorded in 1974. There is a major decrease in phosphocreatine (PCr), at 3.2 ppm and a major increase in inorganic phosphate (P_i), at -1.7 ppm. At -3.7 ppm the sugar phosphate (SP) peak appears. The three phosphate groups of ATP, γ at 5.2, α at 10.4, and β at 19.2 ppm are well separated from each other. On the whole, this spectrum demonstrated that NMR spectroscopy is applicable to intact muscle.

Fig. NMR 2.   31P spectrum of a frog gastrocnemius muscle recorded in 1974. Note, negative chemical shifts are downfield and positive shifts are upfield.

Superconducting NMR spectrometer

³¹P spectra of intact muscle initiated research in the magnet industry and soon superconducting magnets with a vertical bore of 8 – 20 mm appeared on the market

(Fig. NMR 3). This type of spectrometer incorporates a deuterium lock for shimming, various modules for multiple resonance experiments, a pulse programmer, amplifiers, a sophisticated instrument computer and a software package. Both the magnet and the probe (³¹P, ¹H, or ¹³C) have to be tuned to obtain high resolution spectra.

Fig. NMR 3. Superconducting NMR spectrometer.

At the end of the 1970s these magnets had a field strength of 4.2 Tesla (T), corresponding to 180 MHz ¹H frequency. Since the NMR spectral resolution increases with higher magnetic field, a steady upgrading of the NMR spectrometers, 400, 500, and 600 MHz took place in the coming decades and at the time of this writing (2006) 900 MHz spectrometers are available. Cooling the magnets in liquid nitrogen or liquid helium   increases the homogeneity of the magnetic field and this is further increased by adjustable, “shimming” coils which optimize the uniformity of the magnet.

Superconducting NMR spectrometers revealed the diversity of  ³¹P spectra in various muscles (Burt et al., 1975, 1976); they were also used to record ¹³C (Doyle and Bárány, 1982)  and ¹H spectra (Arús et al., 1984) of muscles.

With the advance in technology, muscles in the high resolution spectrometers, could be perfused with physiological salt solutions, gassed with a mixture of O₂ and CO₂. The perfusion kept the muscles alive for a longer time, allowing metabolite studies, or functional studies, e.g. change in muscle phosphates during contraction (Dawson et al., 1977). Furthermore, dissected rat or rabbit hearts could be used to test the effect of drugs on the metabolism of a beating heart.

Horizontal bore superconducting NMR spectrometer

The next level of magnet technology produced the horizontal magnets with 20 - 40 cm bore; at 2 T (Fig. NMR 4). The 4.7 T (200 MHz for ¹H) machine came much later.

Fig. NMR 4. Two Tesla horizontal bore superconducting NMR spectrometer.

With the availability of wide bore superconducting magnets, the surface coil was developed (Ackerman et al., 1980). Anaesthetized frogs, rats, or rabbits were moved into the magnet and with aid of surface coils, spectra were recorded from the leg muscles, resting or contracting. Naturally, these instruments were also used to record spectra from various organs of the animals.

The 2 T horizontal bore spectrometer readily accommodated the human forearm and as such pioneered in recording the ³¹P spectra of muscle in live human beings. The muscles usually examined belonged to the flexor compartment of the forearm. Good ³¹P-spectra could be obtained in 5 min.  

In addition, ³¹P-NMR allowed the measurement of the intracellular pH of the muscle, resting or fatigued, through the shift of the frequency of the P_i peak (see Equation 2).

One meter horizontal bore magnets

In the middle of the 1980s, large superconducting magnets with a horizontal bore of 1 m in diameter became available, at 1.5 T, 63.9 MHz for ¹H. These machines were really built for MR imaging of patients but were also equipped for human spectroscopy. Figure NMR 5 illustrates such a magnet. The human subject lays on a flat bed and the technician programs the machine to move the subject into the interior of the magnet.

Fig. NMR 5. Front view of the 1 meter bore magnetic resonance scanner.

Figure NMR 6 illustrates the set up. Under the gastrocnemius muscle of the left leg there is a surface coil, used for the spectroscopy. The right leg is elevated and, thus, it is out of the range of the surface coil.

Fig. NMR 6. Set up for spectroscopy of human leg.

With time, the 1 m bore magnets were updated from 1.5 T, to 4 T, (170 MHz for ¹H) then to 7 T, (300 MHz), and at the time of this writing (2006), 9.4 T (400 MHz) magnets can be found in a few laboratories, although they are not in routine use.

³¹P-Spectroscopy

 ³¹P is the naturally occurring isotope of phosphorus and as such it is 100% abundant in muscle and other tissues. The major phosphate metabolites of muscle are: ATP, PCr, P_i, and (SP). ATP and PCr occur at high concentrations in normal resting muscle, whereas the appearance of P_i indicates fatigue, and SP in the muscle suggests disease.  In spite of the high concentration of the muscle phosphates, as many as 256 – 512 scans are needed to produce a good signal to noise ratio in the ³¹P-spectra. To economize the time of the spectrometer, the pulses are less than optimal (90^o) and, therefore, the spectra obtained are qualitative. Nevertheless the short pulse spectroscopy satisfies the need of the physiologist or the clinician.  Quantitative spectroscopy for estimation of the concentration of muscle phosphate metabolites has been described (Bárány-Glonek, 1982, Venkatasubramanian et al., 1988, Canioni and Quistorff, 1994); this is mainly needed for kinetic studies in biochemistry.

Characteristic Parameters of  ³¹P-NMR spectra (several of these parameters also apply for ¹H and ¹³C spectra)

(1)

For ³¹P-spectra, chemical shifts are reported relative to 85% inorganic orthophosphoric acid to be 0.0 ppm. According to the old recommendation of the International Union of Pure and Applied Chemistry, negative chemical shifts were assigned downfield and positive shifts were assigned upfield. Later this recommendation was reversed; upfield became negative and downfield positive. In muscle, PCr is the tallest peak in the center of the ³¹P-spectrum, therefore PCr may be selected as the 0.0 ppm reference (see Figure NMR 12). We have also used methylenediphosphonate as a reference compound that resonates at -16.3 ppm, far away from the center of the spectrum (Figure NMR 8).

(2)

Identification of resonances    Peak assignments in intact muscle are essentially based on accurate measurements of the chemical shift of a resonance. The NMR Chemical Shift Table (see below) lists the chemical shifts of phosphates known to occur in muscle, measured both in ¹H-coupled and decoupled spectra. The resonances of SP, P_i, GPC, SEP, PCr, γ-, α -, and β-phosphates of ATP are well separated in muscle, thus the identification of these metabolites present no problem. Within the SP group only glucose 1-phosphate can be resolved from the other hexosephosphates and from the triosephosphates. A better resolution of SP requires 145.8 MHz for ³¹P and proton decoupling. In muscle, it is very difficult to separate SP from AMP or IMP, and it is difficult to separate AMP and IMP from each other. Differentiation of arginine phosphate from PCr is relatively easy, and NADH-NAD⁺ appear as an individual peak at 72.88 MHz, although they are incorporated into the α-phosphate peak of ATP at 36.43 MHz. The chemical shift caused by complexing of the β-P of ATP is easily determined at 36.43 MHz.

Chemical shifts of biologically relevant ³¹P metabolites are described in Gillies (1994)

Coupling constants: A magnetically active nucleus associated with a given resonance can interact through the electronic structure with other magnetic active nuclei in the molecule. The effect of this is to split the resonance into smaller multiple peaks that sum to give the total intensity. This phenomenon is called spin-spin coupling. For example in ATP, the α and the γ end group phosphates each spin-couple to the β middle group to produce the characteristic doublet, doublet, triple pattern (see Figure NMR 19).

By proper irradiation of the sample, the spin coupling can be decoupled, resulting in single resonances instead of a doublet or triplet.

Fig. NMR 7. Line width in normal human muscle water at 63.86 MHz for 1H.

³¹P Spectral Profiles of Intact Muscles

³¹P-NMR discovered phosphate compounds, not detected by biochemical techniques that resonate in the center of the ³¹P-NMR spectrum and thereby could be identified as phosphodiesters (PDE).

Figure NMR8 compares the spectra of various intact muscles. PCr is the major peak in the gastrocnemius muscle of the North American winter frog, and a considerable amount of P_i is also present. Between the PCr and P_i peaks are two phosphodiester resonances at -0.4 ppm and +0.2 ppm. These resonances are found in large amount in the toad gastrocnemius but they are missing from the abalone muscle. Abalone contains P-arginine instead of PCr. The human gastrocnemius muscle, from a gangrenous leg, degraded most of its organic phosphates into P_i.

Fig. NMR 8. Comparison of 31P NMR spectra of various muscles. Peak assignments in the muscle spectra from left to right: the external methylenediphosphonate reference compound, -16.3 ppm; SP, -3.7 ppm; Pi, -1.7 ppm; PCr, 3.2 ppm; the γ, 5.6 ppm;  α, 10.7 ppm; and  β, 19.1 ppm, phosphate groups of ATP, ORTHO denotes the orthophosphate compounds, ENDS and  MIDDLES denote the end and middle phosphate groups of ATP. Ho, denotes the magnetic field.

Fig. NMR 9. Comparison of 31P NMR spectra of normal and dystrophic chicken breast muscle. For peak assignments see the legend to Fig. NMR 8.

The comparison of normal versus dystrophic muscle shows no difference in the resonances of SP, P_i, PCr, and ATP.  However, a new signal (see arrow), between 0.4 -0.5 ppm appears in the dystrophic muscle.

Fig. NMR 10.  Phosphodiesters in various intact muscles. For peak assignments see the legend to Fig. NMR 8. The arrow denotes the phosphodiester resonances.

In Figure NMR 10 the indicated resonances in the – 0.5 to 0.5 ppm area of the spectrum are shown in frog gastrocnemius, rabbit soleus, rabbit heart and dystrophic chicken pectoralis.  However, we observed these resonances in several other muscles as well. The compounds which gave rise to these resonances were extracted from their muscles with perchloric acid, and purified by barium and alcoholic fractionation and column chromatography. The 0.13 ppm resonance was isolated (Burt et al., 1976) as a homogeneous compound and identified as sn-glycerol 3-phosphoryl choline (GPC). (see Equation 3).

(3)

GPC is present in heart and slow muscles, but not in all skeletal muscles. For instance, the gastrocnemius muscle of Northern frog contains GPC but it is missing from the same muscle of the Southern frog. Similarly human leg muscles are rich in GPC, in contrast human arm muscles contain only minute amounts of GPC (Bárány et al., 1982).

The other pure phosphodiester compound that resonated at -0.4 ppm was isolated from dystrophic chicken muscle (Chalovich et al., 1977) and identified as L-serine ethanolamine phosphate (SEP, see Equation 4).

(4)

Like GPC, SEP is also present in slow muscle, such as the anterior latissimus dorsi of chicken, or turtle muscle (see Figure NMR 11).

Fig. NMR 11. Proton coupled 72.88 MHz 31P NMR spectrum of intact turtle muscle. The spectrum was obtained on a muscle 5 hr after dissection from the animal, which explains the high level of Pi and SP.  SEP is present in high concentration in turtle skeletal muscle. Note, this spectrum was obtained in 3.3 min.

Changes in  ³¹P Spectra of Stimulated Muscles

Figure NMR 12 shows the effect of a 35 sec tetanus on muscle ³¹P spectra. PCr decreases to about half of its resting value, P_i increases greatly, SP are increasing slightly, and ATP remains constant (Dawson et al., 1977). The same authors also followed the recovery of tetanus: both PCr and P_i recovered with a half time of about 10 min.

Fig. NMR 12. The effect of a 35 sec tetanus on the 31P spectrum (129.2 MHz) of toad gastrocnemius muscles.  Spectrum A: Resting muscle in oxygenated Ringer’s solution at 4oC; B: Stimulated muscle under the same conditions. Each spectrum was obtained over a period of 7 min and represents an accumulation of 200 scans at 2 sec intervals. PCr is the 0 ppm reference (Dawson et al., 1977).

Frog muscles can also be stimulated by chemical means; caffeine is a drug which causes the muscle to develop tension and shorten. If the muscle is not allowed to shorten, considerable isometric tension will be produced by the muscle. Figure NMR 13 (Panel A) shows the ³¹P profile of one of the paired gastrocnemius muscles shortly after it was mounted in the NMR tube. The muscle is rich in PCr, the sharp resonance at 3.2 ppm, and the three resonance bands from ATP are visible. The SP (-3.7 ppm) and the P_i (-1.7 ppm) concentrations are low. After 45 min of caffeine stimulated isometric contracture (Panel B) the spectrum shows the complete exhaustion of PCr and barely detectable levels of ATP. On the other hand SP and P_i are increased markedly.

Fig. NMR 13. Effect of isometric caffeine contracture on the 31P phosphate profile of frog gastrocnemius muscle at 31oC. (A) Spectrum of the freshly dissected muscle in the first five minutes. After the spectrum was taken, 20 mM caffeine in Ringer’s solution was added and the muscle underwent isometric contracture for 45 min. After this time spectrum (B) was obtained under the same spectrometer conditions as spectrum (A). After this analysis, the 31P phosphate profile from the other gastrocnemius muscle of the same frog was determined, again under the same spectrometer conditions as spectrum (A). Spectral conditions: 256 scans at 2 sec interval, at 36.47 MHz for 31P. The noisy spectrum is due to the low magnetic field used and avoiding baseline correction.

The Central Role of PCr in Analyzing ³¹P Spectra

Phosphocreatine is the predominant peak in the freshly dissected muscle. Only a small peak is detected in the resonance position of P_i and no peak is detected further downfield where SP would resonate (Figure NMR 14).

Fig. NMR 14.  Ptoton-coupled 31P-NMR spectrum of intact chicken pectoralis muscle. A Brucker CXP-180 spectrometer operating at 72.88 MHz for 31P was used. The muscle, chilled in ice, was placed into a 20 mm tube and the spectrum was recorded at 24oC for 10 min. The spectrum shows the signal average of 600 scans. Chemical shifts are relative to the PCr resonance in the muscle at -3.2 ppm.

Similar spectra were recorded with resting toad muscles at 4^oC (Dawson et al., 1977).   In contrast, a muscle that has been injured during dissection or an aged muscle shows a reduced PCr peak and increased P­_i and SP peaks. However, the ATP-phosphate resonances do not change until the PCr level is maintained above 50%.

In general, in normal muscle only the ratio of P_i/PCr is changing upon exercise; the muscle uses PCr to regenerate the ATP hydrolyzed (through the creatine kinase equilibrium reaction) resulting in reduction of PCr, whereas the P_i liberated from ATP increases, as a result of exercise. In other words, a low P_i/PCr ratio is characteristic for resting muscle, whereas a high ratio is characteristic for functioning muscle. Chance et al., (1986) pioneered the use of the P_i/PCr ratio as an indicator of ADP levels and thus an indicator of muscle activation. Thus, the P_i/PCr ratio indicates the relationship between the transduction of chemical energy and muscle work. The P_i/PCr ratio demonstrates differences between athletes, normal subjects and patients with various diseases. For example, ³¹P spectroscopy was used to characterize muscle metabolism and force production in sprint trained runners and endurance trained runners (Johansen and Quistorff, 2003).

Fig. NMR 15. Changes in the 31P spectrum during muscular exercise.  For details see the text.

In human patients, this type of  experiment combined with estimation of the intracellular pH (from the chemical shift of the P_i resonance) was used, for diagnosis of the absence of glycogen phosphorylase (Ross et al., 1981) and the deficiency of phosphofructokinase (Edwards et al., 1982).

Kinetics of PCr breakdown:  The ability of  NMR to follow changes in the concentration of phosphate metabolites allows kinetic measurements in intact muscle. Of the various phosphate metabolites, PCr is the phosphate to be monitored, because (a) it appears as a sharp resonance and, therefore, changes in the peak area can be measured accurately; (b) its concentration in the resting skeletal muscle is about 25 mM, more than the sum of other phosphates, and, therefore, changes in PCr concentration are large; (c) it is in rapid equilibrium with ATP, and, therefore, PCr participates in all energy-requiring reactions of the muscle cell.

Figure NMR 16 compares the logarithm of the percentage of PCr peak area relative to the sum peak areas of all the phosphates, as a function of the incubation time of normal and dystrophic chicken pectoralis muscles under virtually anaerobic conditions. In both cases the plots are straight lines, indicating that the breakdown of PCr is first order. From the slopes, the rate constants were calculated to be 0.011 and 0.024 min^-1 for normal and dystrophic muscles, respectively, thus showing about a twofold greater PCr utilization for the dystrophic muscle relative to the normal muscle. Similar experiments may be carried out with other muscle types. There is no need to know the absolute PCr concentration in muscle for such a plot, only the relative peak areas are required.

Fig. NMR 16.  31P NMR time course for the utilization of PCr by intact normal (empty circles) and dystrophic (full circles) chicken pectoralis muscle. The ordinate represent the log of the percentage of phosphorus present as PCr. Proton-coupled 31P spectroscopy (72.88 MHz) was carried out as described in the legend of Figure NMR14.

Myosin ATPase activity: Under anaerobic conditions the rate of breakdown of PCr may be used to estimate the ATPase activity of myosin in muscle.  This is based on Equations 5,6 and 7.

(5)   (6)     (7)

Since the equilibrium in reaction (6) is shifted far to the right under anaerobic conditions and before the onset of glycolysis (which uses P_i for the formation of glucose 1-phosphate from glycogen), the initial breakdown of PCr is equal to the breakdown of ATP. The activity of myosin ATPase in the resting muscle is described in Bárány et al., (1982).

³¹P-Spectroscopy of Heart Muscle

Fig. NMR 17.  31P spectrum of beating rat heart. Courtesy of Dr. E. Douglas Lewandowski

During ischemia, a rapid decrease in PCr, a slower decrease in ATP and a marked increase in the P_i resonance takes place. There is also an upfield shift of the P_i signal. Upon reperfusion, there is a rapid resynthesis of PCr and a reverse in the shift of P_i signal, whereas the recovery of ATP and reduction of P_i are relatively slow (reviewed in  

Bárány and Glonek, 1982).

In vivo spectroscopic studies of the heart require spatially localized spectroscopy measurements. From et al., (1994) developed techniques which allow acquisition of ³¹P spectra from several discrete layers from the canine left ventricular (LV) wall using a surface coil attached to the epicardial surface of the ventricle. They were able to record spectra from five layers, spanning the LV wall and marginally penetrating the LV chamber; the spectra were representing the subendocardium, the midwall, and subepicardium without overlap. Spectra with this spatial resolution could be obtained with a high signal to noise ratio in about 10 min. The intermost voxel showed the signal from 2,3-diphosphoglycerate (2,3-DPG) a typical compound of erythrocytes in the left ventricular chamber.

These studies showed that P_i was undetectable in any myocardial layer in short time experiments in contrast to studies with the  in vitro perfused heart. Evidently, in the live animal, the mitochondrial oxidative phosphorylation is fast enough to keep ATP levels at maximum. However, if such “transmural” spectra are accumulated over a longer period of time, P_i becomes observable. In general, the NMR spatial localization techniques has made possible detailed analyses of the interactions between metabolism, blood flow, and contractile function taking into account the intramural heterogeneities of these variables in both normal and abnormal hearts. These measurements gave a better understanding of the myocardial energy expenditure, than did the in vitro perfused heart experiments.

The ³¹P spectra of human hearts in live persons have been recorded (for summary see From et al., 1994). At 34.48 MHz for ³¹P in normal subjects the ATP- βP/PCr ratio varied from 0.48 – 0.91. The large variation is caused from the large voxels used which likely included blood in the heart chamber or chest wall. At 68.96 MHz for ³¹P, the ratio was ~0.56 which was in agreement with the values obtained with epicardial surface coils in dogs. In ischemic human myocardium the ATP- βP/PCr ratio increased markedly similar to that obtained in dog heart. Most studies have found a increase of the ATP- βP/PCr ratio in cardiomyopathy and vascular heart diseases.

³¹P spectroscopy was also used for characterization of human donor hearts before transplantation (Kober et al., 2002).     

³¹P-Spectroscopy of Smooth Muscle

The review of Dillon (1996) describes the ³¹P spectroscopy of smooth muscle in great detail.

³¹P-Spectroscopy of Normal and Diseased Human Muscle

This is a popular topic in biochemistry, physiology and medicine. Because of the low magnetic field of the clinical MR scanners, it is necessary to optimize the tissue homogeneity by shimming on the muscle water. Figure NMR7 shows a 0.1 ppm line width in healthy human leg muscle; in diseased muscle the line width is considerably greater.

Figure NMR 18 shows a relatively homogeneous axial image from the muscle of a healthy person.  Figure NMR 19 shows a high resolution ³¹P spectrum from the leg muscle of a healthy person, even the couplings of the ATP phosphate resonances are resolved. PCr (at – 3.2 ppm) is the prominent peak, whereas P_i (at 1.7 ppm) is a small peak; the P_i/PCr ratio is 0.09. Note the presence of GPC (at -0.23 ppm), and absence of SP.  The coupled γ-, α-, and β-phosphates of ATP resonate between -5.22 – 5.90, -10.35 – 10.93, and -18.49 – 19.80, respectively. The ratio of ATP-βP/PCr is 0.16.

Fig. NMR 18.  Axial image of leg muscle from a healthy person.

Fig. NMR 19. 31P spectrum of leg muscle from a healthy person. The spectrum was recorded in the proton-coupled mode at 25.86 MHz for 31P. The 0 ppm corresponds to 85% inorganic orthophosphoric acid. For peak assignments see the text.

In the following we will show the morphology of diseased muscles through axial images, and then compare  the ³¹P spectra of  diseased muscles with those of normal muscles:

Fig. NMR 20. Axial image of leg muscle from a patient with Werdnig-Hoffman Disease.

Fig. NMR 21. Comparison of 31P spectra of leg muscles from a patient with Werdnig-Hoffman disease (top) with that of a healthy person (bottom). For figure legend see that of Fig. NMR 19.

In myotonic dystrophy the axial image of the leg muscle (Figure NMR 22) shows much less alteration than that seen in the Werdnig Hoffman muscle, and the ³¹P spectrum of the myotonic muscle (Figure NMR23) is also less severe than that of the Werdnig-Hoffman muscle: Thus, the ratio of the ATP-βP peak (-19.5 ppm) to that of the PCr peak (-3.2 ppm) is 0.38. The ratio of the P_i peak (1.9 ppm) to that of the PCr peak (-3.2 ppm) is 0.48. The myotonic muscle contains a small amount of GPC (-0.03 ppm) and a small amount of SP (3.3 ppm).

Fig. NMR 22. Axial image of leg muscle from a patient with myotonic dystrophy.

Fig. NMR 23. Comparison of 31P spectra of leg muscles from a patient with myotonic dystrophy (top) with that of a healthy person (bottom). For figure legend see that of Fig. NMR 19.

Duchenne dystrophy, a severe disease, is compared with pedal dystonia, a mild muscle disease. The axial image (Fig. NMR 24) illustrates the diseased state of the Duchenne muscle. From the comparison of ³¹P spectra, (Fig. NMR 25) it appears that the ratio of ATP-βP/PCr is 0.43 in the Duchenne muscle vs 0.29 in the dystonic muscle. There is no P_i in the Duchenne muscle whereas the ratio of P_i/PCr is 0.22 in the dystonic muscle. The Duchenne muscle contains a significant GPC peak (-0.4 ppm) and a large quantity of SP (at 2.32 and 3.13 ppm).

Fig. NMR 24. Axial image of leg muscle from a patient with Duchenne muscular dystrophy.

Fig. NMR 25. Comparison of 31P spectra of leg muscles with a patient with Duchenne dystrophy (top) with that of a patient with pedal dystonia (bottom). For figure legend see that of Fig. NMR 19.

The values for the ratios of ATP-βP/PCr and P_i/PCr, as well as the appearance of SP in the spectra indicate a relationship between extent of muscle disease and characteristics of ³¹P spectra. This is documented by Bárány et al., (1989a).

The review of Cozzone and Bendahan (1994) describes the pathological applications of human ³¹P MRS.

¹H-Spectroscopy

¹H, a single proton, is the principal isotope for hydrogen. Its isotopic abundance is almost 100% and animal tissues contain 80% water. ¹H possesses the largest magnetic moment of the biologically important nuclei. Since at constant magnetic field, the NMR frequency of nuclei depend only on their magnetic moment, the ¹H frequency is the highest within the same spectrometer. For instance, in a 360 MHz spectrometer for ¹H, the frequency for ³¹P is 145.76 MHz, and for ¹³C it is about 90 MHz.  Since proton is the most sensitive nucleus, it takes only a few minutes to record ¹H spectra in conventional 1.5 T MR scanners at 63.85 MHz. Furthermore, much less tissue volume is required for ¹H spectroscopy than for ³¹P or ¹³C spectroscopy. For instance, the entire leg is required to get a good ³¹P or ¹³C spectrum of the muscle, but a 27 cm³ leg voxel is enough to get a good ¹H spectrum.

Assignment of ¹H resonances: Arús et al., (1985) were the first to identify the ¹H resonances in spectra of excised rat brain, using TSP (sodium 3-trimethylsilyl[2,2,3,3-²H₄]propionate) as an external standard at 0.00 ppm. In frog, rat and human skeletal muscle, the same ¹H resonances were found as in the rat brain. Furthermore, the =N-CH₃ resonance of creatine at 3.02 ppm could be used as internal reference for the chemical shifts. (Arús and Bárány, 1986).

Fig. NMR 26. Triple voxels in human leg.  Upper voxel: bone marrow of tibia; middle voxel: gastrocnemius muscle; bottom voxel: subcutaneous fat.

Fig. NMR 27. Comparison of 1H spectra in triple voxels of human leg.  The same voxels were used which are indicated in the axial images of Fig. NMR 26.

In the next example, ¹H spectra were recorded from voxels of normal and Duchenne dystrophic muscles. The image voxels (Figures NMR 28 and NMR 29) show uniform structure in the normal muscle and strong fatty infiltration in the muscle of the advanced Duchenne dystrophic muscle (equal size voxels were selected from both types of muscles). The comparative ¹H spectra (Figure NMR30) show virtually only water (4.7 ppm) in the normal muscle and virtually only fat (1.3 ppm) in the dystrophic muscle. Based on this observation, we have used the fat/water ratio for characterization of normal and diseased muscles Bárány et al., (1989b).

Fig. NMR 28. Voxel in normal leg muscle.

Fig. NMR 29. Voxel in advanced Duchenne dystrophic muscle.

Fig. NMR 30. Comparison of volume selective 1H spectra from normal (bottom) and advanced Duchenne dystrophic (top) muscle.

Water suppressed ¹H spectroscopy:   A strong water signal masks weaker signals from metabolites in localized proton spectra of live tissues and, therefore, to obtain metabolite spectra the water signal has to be suppressed (Frahm et al., 1987b).  Figure NMR 31 illustrates this effect. After significant water suppression (top) the characteristic muscle metabolites, creatine (3.0 ppm) and choline (3.2 ppm) become visible in the spectrum. Furthermore, the fat resonances are just discernible in the original spectrum at 1.4 and 1.6 ppm (bottom), after water suppression  they  become major peaks (top).

Fig. NMR 31. STEAM spectrum of normal human muscle with (top) and without water suppression (bottom).

¹H spectroscopy in the 21^st century:  By the end of the 20^th century vertical magnets with bore sizes from 12 to 30 cm, or horizontal magnets with a bore size of 12 cm became available with an ¹H frequency of 300 – 500 MHz. These spectrometers were equipped with shielded gradient systems, quadrature driven birdcage coils, and the point resolved spectroscopy (PRESS) single-voxel technique, allowing researchers to record ¹H spectra of various organs of live mouse and rat of a high quality, not seen before.

Schneider and collaborators (2004) worked out the in vivo cardiac ¹H-MRS in the mouse at 500 MHz. Stability and reproducibility were achieved by dedicated cardiac and respiratory gating. Water-suppressed and unsuppressed cardiac spectra from a 2 μl voxel positioned in the septum were acquired in diastole. Ten resonances were resolved in the spectrum, including creatine, taurine, carnitine, and intramyocardial lipids. Moreover, the cardiac metabolites were quantified relative to the total water content in the very small    2 μl.voxel. Thus, these authors increased the sensitivity of volume selective ¹H spectroscopy about 1000-fold.

Renema et al., (2003) used a 300 MHz instrument to carry out ¹H spectroscopy in muscle of guanidinoacetate methyltransferase (GAMT)-deficient mice. Several metabolic abnormalities were observed in the muscle of these animals by the in vivo spectroscopy; most important was the reduced creatine (Cr) content. Overall the in vivo ¹H spectroscopy of the muscle of the GAMT -/- mice was similar to that found in human GAMT deficiency. This opens up the way for study of Cr synthesis and transport and for diagnostic and therapeutic aspects of Cr deficiencies in humans. Subsequently, Nakae et al, (2004) found significantly reduced creatine content in human patients with chronic heart failure.

Neuman-Haefelin et al., (2003), also working at 300 MHz, studied the determinants of intramyocellular lipid (IMCL) concentrations in rat soleus and tibialis anterior muscles.

Interestingly, they have found that in addition to insulin resistance several other factors influence IMCL levels, such as age, gender, muscle type, and rat strain. Previously,

Jagannathan and Wadhwa (2002) reported the absence of IMCL in severely paralyzed postpolio human patients.

Walter et al., (2005) used ¹H-MRS for noninvasive monitoring of  gene correction in dystrophic mouse muscle.

¹³C-Spectroscopy

¹³C is the carbon isotope that can be used for NMR. However, its natural abundance is only 1.1%. Therefore, recording ¹³C spectra is time consuming. On the other hand, the ¹³C spectrum is as wide as 200 ppm, which allows easy identification of tissue metabolites.

The sensitivity of ¹³C spectroscopy can be increased by proton-observed   carbon-edited spectroscopy (Novotny et al., 1990; Henry et al., 2006).

Assignment of ¹³C resonances:  In proton-decoupled natural abundance ¹³C NMR spectrum of excised rat brain 55 resonances were resolved at 90.8 MHz for ¹³C (Bárány et al., 1985). The spectrum was accumulated with 29,472 transients and it was plotted without any line broadening. The resonances were identified by the correlation of intact rat brain ¹³C resonances with those of pure brain metabolites and the localization of brain resonances in various subfractions of brain.

Several of the brains resonances such as the amino acids taurine, glutamate, phospholipid-degradation products, such as inositol, GPC, and energy yielding products, such as creatine, lactic acid were also found in skeletal and heart muscles. Notable are the methyl carbons of creatine and creatine phosphate at 54.66 ppm, which can be used as internal reference. For external reference we use dioxane at 67.4 ppm                           (see Figure NMR 32).

Lactic acid  is an indicator for the extent of glycolysis and can be identified through its methyl carbon resonance at 20.95 ppm and alcoholic carbon resonance at 69.24 ppm   (see Figure NMR 32).

For detailed assignment  of  ¹³C resonances see Appendix  2  in Gillies (1994).

Comparative spectra of normal and diseased muscle:   Figure NMR 32 compares the spectra of diseased muscle (upper trace) with a normal muscle (lower trace). The resonances observed in the diseased muscle are mainly from neutral fats: at 172.6 ppm the esterified carboxyl carbon (OCO); at 130.3 and 128.7 ppm the unsaturated carbons ((CH₂)_n-CH= and =CH-CH₂-CH=, respectively); at 62.6 ppm the 1 and 3 carbons of glycerol (CH₂OH); at 30.4 ppm the repeating methylene carbons ((CH₂)_n); at 28.0 ppm the methylene carbon adjacent to an allylic carbon (CH₂-CH=); at 25.6 ppm the β-methylene carbon from the carboxyl terminal (OCOCH₂CH₂); at 23.5 ppm the methylene carbon adjacent to the methyl terminal (CH₂-CH₃); and at 14.8 ppm the terminal methyl carbon (CH₃). In addition, the spectrum of the diseased muscle shows the resonance from the methyl carbon of creatine, at 54.8 ppm.

In the spectrum of normal muscle, in the lower trace of the Figure, the resonances of α- and β-glucose carbons are visible at: 96.8, 93.0, 76.7, 75.1, 72.4, 70.5 and 61.6 ppm. This muscle also produced a considerable amount of lactic acid as evidenced by the appearance of resonances from the lactic acid methyl carbon (21.0 ppm), alcoholic carbon (69.3 ppm), and carboxylic carbon (183.3 ppm). Normal muscle contains more creatine than diseased muscle, as may be concluded by comparing the peak heights at 54.7-54.8 ppm. As a result of the increased creatine content, the methylene carbon (37.8 ppm) and the guanidino carbon (158.0 ppm) of creatine are detectable in normal muscle. The peak height of the 30.4-30.6 ppm methylene carbon of fatty acyl chains is greatly reduced in normal muscle as compared to that of the diseased muscle. Furthermore, the ratio of 130.4 to 128.7 ppm peak heights is about 1 in normal muscle but about 4 in diseased muscle.

Fig. NMR 32. Proton-decoupled natural abundance 13C spectra of diseased human muscle vs normal human muscle. The spectra were recorded at 118.2 MHz for 13C, and about 50,000 transients were collected for each muscle Upper trace; Cerebral palsy muscle, Lower trace, normal muscle. The signal at 67.4 ppm is from dioxane in a sealed capillary. For peak assignments see the text.

Changes in the natural abundance ¹³C NMR spectra of intact frog muscle:  The carbons of phospholipids have limited mobility in fresh, resting frog muscle but gain considerable mobility upon caffeine contracture. This is illustrated in Figure NMR32.  The strongest resonance in the caffeine treated (upper trace) muscle occurs at 31.0 ppm from the methylene carbons, repeating units of fatty acyl chains of phospholipids, whereas the same resonance in the untreated muscle reaches a height only equal to that of other resonances in the aliphatic envelope of the spectrum.  In contrast, the height of the 54.5 ppm major resonance from the monomethyl carbon of creatine and the trimethyl carbons of choline is the same in the caffeine-treated and untreated muscles. This indicates that the caffeine effect is specific for the membrane phospholipids.

Caffeine contracture depletes the high energy phosphate stores of the muscle and subsequently initiates glycolysis. The end-product of glycolysis is lactic acid, and we see the predominant resonances from the methyl carbon of lactic acid at 20.9 ppm, and from the alcoholic carbon at 69.2 ppm. These resonances are absent in the untreated muscle. This experiment demonstrates the power of high resolution ¹³C-NMR to follow changes in muscle membrane structure and energetics.

Fig. NMR 33. Proton-decoupled natural abundance 13C spectra of intact frog gastrocnemius muscles. One of the paired muscles was soaked in frog Ringer’s solution with and without 10 mM caffeine (upper and lower traces, respectively) at room temperature, then the spectra were recorded at 90.5 MHz for 13C for 10 min. The signal at 67.4 ppm is from dioxane.

¹³C spectroscopy with ¹³C-labeled substrates:  If tissues are perfused with ¹³C-enriched compounds the intensity of the resonances is due to the ¹³C-enrichment and not caused by the 1.1% natural abundant ¹³C. Thus, ¹³C-labeled substrates are powerful tools to get information about the intermediary metabolism in living tissues. It should be mentioned that the cost of enriched substrates is not negligible, but a lot of information can be obtained from a single experiment.

The perfused rat heart is the most commonly used organ for such studies. Figure NMR34, ¹³C-NMR spectra acquired from intact beating rat heart perfused with ¹³C-labeled palmitate, shows a selected set of dynamic data. Palmitate is oxidized and the ¹³C-label is incorporated into the 2^nd, 3^rd, and 4^th carbon of glutamate (GLU) and triglyceride (TRIG) pools. The reaction is completed in about 35 min. Similar experiments were performed to follow the pathway of the metabolites in the Krebs citric acid cycle (Sherry and Malloy, 1994). Furthermore, these authors also infused ¹³C-labeled compounds into live rats and studied the ¹³C spectra of extracts of frozen tissue samples. The frequently used ¹³C-enriched substrates are glucose, lactate, acetate, fatty acids, or ketone bodies.

Fig. NMR 34. 13C-NMR spectra acquired from intact beating rat heart perfused with 13C-labeled palmitate. Courtesy of Dr. E. Douglas Lewandowski

In general, ¹³C spectroscopy has been given new directions for studying the intermediary metabolism in muscle and other tissues in physiology. Labeled substrates, generally glucose, have enabled the fluxes to be determined in vivo, whereas natural abundance ¹³C has enabled concentrations to be measured (reviewed by Shulman and Rothman, 2001). Indeed, natural abundance ¹³C spectroscopy evaluated the muscle glycogen content in patients suffering from adult-onset acid maltase deficiency (Wary et al., 2003). This is a new approach to follow the success of therapeutic trials.

References

Ackerman, J.H., Groove, T.H., Wong, G.,  Gadian, D.G., and Radda, G.K. (1980). Mapping of metabolites in whole animals by ³¹P NMR using surface coil. Nature, 283, 167-170.

Arús, C., Bárány, M., Westler, W.M., and Markley, J.L. (1984). ¹H NMR of intact muscle at 11. FEBS Letters, 165, 231-237.

Arús, C., Chang, Y-C., and Bárány, M. (1985). Proton nuclear magnetic resonance spectra of excised rat brain assignment of resonances. Physiol. Chem. Physics and Medical NMR, 17, 23-33.

Arús, C., and Bárány, M. (1986). Application of high-field ¹H-NMR spectroscopy for the study of perfused amphibian and excised mammalian muscles. Biochim. Biophys. Acta, 886, 411-424.

Bárány, M., Bárány, K., Burt, C.T., Glonek, T., and Myers, T.C. (1975). Structural changes in myosin during contraction and the state of ATP in the intact frog muscle. J. Supramol. Struct, 3, 125 – 140.

Bárány, M., and Glonek T. (1982). Phosphorus-31 nuclear magnetic resonance of contractile systems. In Methods Enzymol. vol. 85 (L.W. Cunningham and , D. Frederikson eds.), pp.  624 – 676, Academic Press, New York.

Bárány, M., Chalovich, J. M., Burt, C.T., and Gloneck, T. (1982). Nuclear magnetic resonance studies of muscle. In Disorders of the Motor Unit (D. Schotland, ed), pp. 697-714, John Wiley & Sons, Inc. Hoboken, N.J.

Bárány, M., Arús, C., Chang, Y-C. (1985). Natural-abundance ¹³C NMR of brain. Magn. Reson. Med., 2, 289-295.

Bárány, M., Siegel, I.M., Venkatasubramanian, P.N., Mok, E., Wilbur, A.C. (1989a). Human leg neuromuscular diseases: P-31 NMR spectroscopy. Radiology, 172, 503-508.

Bárány, M., Venkatasubramanian, P.N., Mok, E., Siegel, I.M., Abraham, E., Wycliffe, N.D., and Mafee, M.F. (1989b). Quantitative and qualitative fat analysis in human leg muscle of neuromuscular diseases by ¹H MR spectroscopy in Vivo. Magn. Reson. Med., 10, 210-226.

Burt, C.T., Glonek, T., and Bárány, M. (1975). Analysis of phosphate metabolites, the intracellular pH, and the state of adenosine triphosphate in intact muscle by phosphorus nuclear magnetic resonance. J. Biol. Chem., 251, 2584-2591.

Burt, C.T., Glonek, T., and Bárány, M. (1976). Phosphorus-31 nuclear magnetic resonance detection of unexpected phosphodiesters in muscle. Biochemistry, 15, 4850-4853.

Burt, C.T., Cohen, S.M., and Bárány, M. (1979). Analysis of intact tissue with ³¹P NMR.

Ann. Rev. Biophys. Bioeng., 8, 1-25.

Canioni, P. and Quistorff, B. (1994). Liver physiology and metabolism.. In NMR in Physiology and Biomedicine (R.J. Gillies, ed), pp, 373- 388. Academic Press, San Diego

Chalovich, J. M., Burt, C.T., Cohen, S.M., Glonek, T., and Bárány, M (1977). Identification of an unknown ³¹P nuclear magnetic resonance from dystrophic chicken as L-serine ethanolamine phosphodiester.  Arch. Biochem. Biophys. 182, 683 – 689.

Cozzone, P.J. and Bendahan, D. (1994). ³¹P NMR spectroscopy of metabolic changes associated with muscle exercise: physiopathological applications, In NMR in Physiology and Biomedicine, (R.J. Gillies, ed), pp 389 – 403, Academic Press, San Diego

Dawson, M. J., Gadian, D.G., and Wilkie, D.R. (1977). Contraction and recovery of living muscles studies by 31P nuclear magnetic resonance. J. Physiol., 267, 703.- 735.

Dillon, P.F. (1996). ³¹P nuclear magnetic resonance spectroscopy. In Biochemistry of Smooth Muscle Contraction (M. Bárány, ed.), pp 393 – 404, Academic Press, San Diego.

Doyle, D.D., and Bárány, M. (1982). Quantification of lactic acid in caffeine-contracted and resting frog muscle by high resolution natural abundance ¹³C NMR. FEBS Letters, 140, 237-240.

Edwards, R.H.T., Wilkie, D.R., Dawson, M.J., Gordon, R.E., and Shaw, D. (1982). Clinical use of nuclear magnetic resonance in the investigation of myopathy.  Lancet i, 725-731.

Frahm, J., Merboldt, K., and Haenicke, W. (1987a). Localized proton spectroscopy using stimulated echoes, J. Magn. Reson., 72, 502-508.

Frahm, J., Merboldt, K., Haenicke, W., and Willringer, A. (1987b). Water suppressed localized proton spectroscopy using stimulated echoes. Abstracts 6^th Ann Meeting Soc. Magn. Reson. Med., New York USA, p. 137.

From, A.H.L., Bache, R.J., Zhang, J., and Uğurbil, K  (1994). Nuclear magnetic resonance studies in normal and abnormal myocardium. In NMR in Physiology and Biomedicine., (R.J. Gillies, ed), pp 413 – 437, Academic Press, San Diego

Gadian, D.G., Radda, G.K., Richards, R.E. , and  Seeley, P.J. (1979) In  Biological Applications of Magnetic Resonance, (R.G. Shulman ed.), p 463, Academic Press, New York

Gillies, R. (1994).  In NMR in Physiology and Biomedicine (R. Gillies ed.), Academic Press, San Diego.

Henry, P-G., Marjanska, M., Walls, J.D., Valette, J., Gruetter, R., and Uğurbil, K. (2006). Proton-observed carbon-edited NMR spectroscopy in strongly coupled second-order spin systems. Magn. Res. Med., 55, 250 - 257.

Hoult, D.I., Busby, S.J.W., Gadian, D.G., Radda, G.K., Richards, R.E. and Seeley, P.J. (1974). Observation of tissue metabolites using ³¹P nuclear magnetic resonance.  Nature, 252, 285 – 287.

Jagannathan, N. R., and Wadhwa, S. (2002). In vivo proton magnetic resonance spectroscopy (MRS) study of post polio residual paralysis (PPRP) patients. Magn. Reson. Imaging, 20, 113-117.

Johansen, L., and Quistorff, B. (2003). 31P-MRS characterization of sprint and endurance trained athletes. Int. J. Sports Med., 24, 183-189.

Kober, F., Caus, T., Riberi, A., Confort-Gouny, S., LeFur, Y., Izquierdo, M., Ranjeva, J. P., Viout, P., Mesana, T., Metras, D., Cozzone, P. J., and Bernard, M. (2002). Objective and noninvasive metabolic characterization of donor hearts by phosphorus-31 magnetic resonance spectroscopy. Transplantation, 74, 1752-1756.

Lundberg, J., Lindgard, A., Elander, A., and Soussi, B. (2002). Improved energetic recovery of skeletal muscle in response to ischemia and reperfusion injury followed by in vivo 31-P magnetic resonance spectroscopy. Microsurgery, 22, 158-164.

McCully, K.K., Vandenborne, K., Posner, J.D., and Chance, B. (1994). Magnetic resonance spectroscopy of muscle bioenergetcs. In NMR in Physiology and Biomedicine,

(R.J. Gillies, ed), pp 405 – 412, Academic Press, San Diego.

Nakae, I., Mitsunami, K., Matsuo, S., Matsumoto, T., Morikawa, S., Inubushi, T., Koh, T., and Horie, M. (2004). Assessment of myocardial creatine concentration in dysfunctional human heart by proton magnetic resonance spectroscopy. Magn. Reson. Med., Sci., 3, 19-25.

Neumann-Haefelin, C., Kuhlmann, J., Belz, U., Kalisch, J., Quint, M., Gerl, M., Juretschke, H-P., and Herling, A.W.(2003) Determinants of intramyocellular lipid concentrations in rat hindleg muscle. Magn. Reson. Med., 50, 242-248.

Novotny, E.J., Ogino, T., Rothman, D.L., Petroff, O.R.C., Richard, J.W., Shulman, R.C. (1990). Direct carbon versus proton heteronuclear editing of 2-¹³C ethanol in rabbit brain in vivo; a sensitivity comparison. Magn. Reson. Med., 16, 431 – 443.

Renema, W.K.J., Schmidt, A., van Asten, J.J.A., Oerlemans, F., Ullrich, K., Wieringa, B., Isbrandt, D., and Heerschap, A. (2003). NMR spectroscopy of muscle and brain in guanidinoacetate methyltransferase (GAMT)-deficient mice: validation of an animal model to study creatine deficiency. Magn. Reson. Med., 50, 936-943..

Ross, B.D., Radda, G.K., Gadian, D.G., Rocker, G., Esiri, M., and Falconer-Smith, J. (1981). Examination of a case of suspected McArdle's syndrome by 31P nuclear magnetic resonance.  N. Engl. J. Med., 304, 1338-1342.

Schneider, J.E., Tyler, D.J., ten Hove, M., Sang, A.E., Cassidi, P.J., Fischer, A., Wallis, J., Sebag-Montefiore, L.M., Watkins, H., Isbrandt, D., Clarke, K., and , Neubauer, S. (2005). In vivo cardiac ¹H-MRS in the mouse. Magn. Reson. Med., 52, 1029-1035.

Sherry, A. D. and  Malloy, C.R. (1994)  Cardiac metabolism. In NMR in Physiology and Biomedicine, (R.J. Gillies, ed), pp 439 – 449, Academic Press, San Diego

Shulman, R. G., and Rothman, D. L. (2001). 13C NMR of intermediary metabolism: implications for systemic physiology. Annu. Rev. Physiol., 63, 15-48.

Venkatasubramanian, P.N.,Mafee, M.F.,  and  Bárány, M (1988). Quantification of phosphate metabolites in human leg in vivo.  Magn. Reson. Med., 6, 359 -363.

Walter, G., Cordier, L., Bloy, D., and Sweeney, H.L. (2005) Noninvasive monitoring of gene correction in dystrophic muscle. Magn. Reson. Med., 54, 1369-1376.

Wary, C., Laforet, P., Eymard, B., Fardeau, M., Leroy-Willig, A., Bassez, G., Leroy, J. P., Caillaud, C., Poenaru, L., and Carlier, P.G. (2003). Evaluation of muscle glycogen content by 13C NMR spectroscopy in adult-onset acid maltase deficiency. Neuromuscul. Disord., 13, 545-553.

Recommended Readings

Chapters to read: Andrew E. R. (1994). Introduction to Nuclear Magnetic Resonance.

                            Chapter 1 In: NMR in Physiology and Biomedicine (R. Gillies ed.),

                            Academic Press, San Diego.

                            Martin, P., Gibson, H., Edwards, R. (1996). MRS of Muscle  In: MR

                            spectroscopy – clinical applications and techniques. (Young I, and

                            Charles, H. eds), Martin Dunitz. London

Book to read:      Lambert, J. B., and Mazzola, E. P. (2003). Nuclear Magnetic Resonance

                            Spectroscopy: An Introduction to Principles, Applications, and

                            Experimental Methods.  Prentice Hall, N.J.

_______________________________________________________________________

Back to the Home

Back to the Beginning of the Chapter

Go to the Previous Chapter