Localized 13C NMR spectroscopy of myo-inositol in the human brain in vivo

Mapping oxidative metabolism in the human brain with calibrated fMRI in health and disease

Conventional functional MRI (fMRI) with blood-oxygenation level dependent (BOLD) contrast is an important tool for mapping human brain activity non-invasively. Recent interest in quantitative fMRI has renewed the importance of oxidative neuroenergetics as reflected by cerebral metabolic rate of oxygen consumption (CMR_O2) to support brain function. Dynamic CMR_O2 mapping by calibrated fMRI require multi-modal measurements of BOLD signal along with cerebral blood flow (CBF) and/or volume (CBV). In human subjects this "calibration" is typically performed using a gas mixture containing small amounts of carbon dioxide and/or oxygen-enriched medical air, which are thought to produce changes in CBF (and CBV) and BOLD signal with minimal or no CMR_O2 changes. However non-human studies have demonstrated that the "calibration" can also be achieved without gases, revealing good agreement between CMR_O2 changes and underlying neuronal activity (e.g., multi-unit activity and local field potential). Given the simpler set-up of gas-free calibrated fMRI, there is evidence of recent clinical applications for this less intrusive direction. This up-to-date review emphasizes technological advances for such translational gas-free calibrated fMRI experiments, also covering historical progression of the calibrated fMRI field that is impacting neurological and neurodegenerative investigations of the human brain.

Comparison of direct 13C and indirect 1H-[13C] MR detection methods for the study of dynamic metabolic turnover in the human brain

A wide range of direct ¹³C and indirect ¹H-[¹³C] MR detection methods exist to probe dynamic metabolic pathways in the human brain. Choosing an optimal detection method is difficult as sequence-specific features regarding spatial localization, broadband decoupling, spectral resolution, power requirements and sensitivity complicate a straightforward comparison. Here we combine density matrix simulations with experimentally determined values for intrinsic ¹H and ¹³C sensitivity, T₁ and T₂ relaxation and transmit efficiency to allow selection of an optimal ¹³C MR detection method for a given application and magnetic field. The indirect proton-observed, carbon-edited (POCE) detection method provides the highest accuracy at reasonable RF power deposition both at 4T and 7T. The various polarization transfer methods all have comparable performances, but may become infeasible at 7T due to the high RF power deposition. 2D MR methods have limited value for the metabolites considered (primarily glutamate, glutamine and γ-amino butyric acid (GABA)), but may prove valuable when additional information can be extracted, such as isotopomers or lipid composition. While providing the lowest accuracy, the detection of non-protonated carbons is the simplest to implement with the lowest RF power deposition. The magnetic field homogeneity is one of the most important parameters affecting the detection accuracy for all metabolites and all acquisition methods.

Progress towards in vivo brain 13C-MRS in mice: Metabolic flux analysis in small tissue volumes

The combination of dynamic ¹³C MRS data under infusion of ¹³C-labelled substrates and compartmental models of cerebral metabolism enabled in vivo measurement of metabolic fluxes with a quantitative and distinct determination of cellular-specific activities. The non-invasive nature and the chemical specificity of the ¹³C dynamic data obtained in those tracer experiments makes it an attractive approach offering unique insights into cerebral metabolism. Genetically engineered mice present a wealth of disease models particularly interesting for the neuroscience community. Nevertheless, in vivo¹³C NMR studies of the mouse brain are only recently appearing in the field due to the numerous challenges linked to the small mouse brain volume and the difficulty to follow the mouse physiological parameters within the NMR system during the infusion experiment. This review will present the progresses in the quest for a higher in vivo¹³C signal-to-noise ratio up to the present state of the art techniques, which made it feasible to assess glucose metabolism in different regions of the mouse brain. We describe how experimental results were integrated into suitable compartmental models and how a deep understanding of cerebral metabolism depends on the reliable detection of ¹³C in the different molecules and carbon positions.

Quantitative fMRI and oxidative neuroenergetics

The discovery of functional magnetic resonance imaging (fMRI) has greatly impacted neuroscience. The blood oxygenation level-dependent (BOLD) signal, using deoxyhemoglobin as an endogenous paramagnetic contrast agent, exposes regions of interest in task-based and resting-state paradigms. However the BOLD contrast is at best a partial measure of neuronal activity, because the functional maps obtained by differencing or correlations ignore the total neuronal activity in the baseline state. Here we describe how studies of brain energy metabolism at Yale, especially with (13)C magnetic resonance spectroscopy and related techniques, contributed to development of quantitative functional brain imaging with fMRI by providing a reliable measurement of baseline energy. This narrative takes us on a journey, from molecules to mind, with illuminating insights about neuronal-glial activities in relation to energy demand of synaptic activity. These results, along with key contributions from laboratories worldwide, comprise the energetic basis for quantitative interpretation of fMRI data.

A two-compartment mathematical model of neuroglial metabolism using [1-(11)C] acetate

The purpose of this study was to develop a two-compartment metabolic model of brain metabolism to assess oxidative metabolism from [1-(11)C] acetate radiotracer experiments, using an approach previously applied in (13)C magnetic resonance spectroscopy (MRS), and compared with an one-tissue compartment model previously used in brain [1-(11)C] acetate studies. Compared with (13)C MRS studies, (11)C radiotracer measurements provide a single uptake curve representing the sum of all labeled metabolites, without chemical differentiation, but with higher temporal resolution. The reliability of the adjusted metabolic fluxes was analyzed with Monte-Carlo simulations using synthetic (11)C uptake curves, based on a typical arterial input function and previously published values of the neuroglial fluxes V(tca)(g), V(x), V(nt), and V(tca)(n) measured in dynamic (13)C MRS experiments. Assuming V(x)(g)=10 × V(tca)(g) and V(x)(n)=V(tca)(n), it was possible to assess the composite glial tricarboxylic acid (TCA) cycle flux V(gt)(g) (V(gt)(g)=V(x)(g) × V(tca)(g)/(V(x)(g)+V(tca)(g))) and the neurotransmission flux V(nt) from (11)C tissue-activity curves obtained within 30 minutes in the rat cortex with a beta-probe after a bolus infusion of [1-(11)C] acetate (n=9), resulting in V(gt)(g)=0.136±0.042 and V(nt)=0.170±0.103 μmol/g per minute (mean±s.d. of the group), in good agreement with (13)C MRS measurements.

Polarization transfer for sensitivity-enhanced MRS using a single radio frequency transmit channel

Polarization transfer techniques are used to enhance sensitivity and improve localization in multinuclear MRS, by transferring polarization from highly polarized or even hyperpolarized nuclei to less sensitive spin systems. Clinical MR scanners are in general not equipped with a second radio frequency (RF) transmit channel, making the conventional implementation of polarization transfer techniques such as distortionless enhanced polarization transfer (DEPT) impossible. Here we present a DEPT sequence using pulses sequentially that can be used on a single RF transmit channel (SC-DEPT). Theoretical simulations, phantom measurements, and in vivo results from human brain at 3 T show that the SC-DEPT method performs as well as the conventional DEPT method. The results indicate that an independent second RF transmit channel for simultaneous pulsing at different nuclear frequencies is not needed for polarization transfer, facilitating the use of these methods with common clinical systems with minor modifications in the RF architecture.

In vivo 13C NMR spectroscopy and metabolic modeling in the brain: a practical perspective

In vivo 13C NMR spectroscopy has the unique capability to measure metabolic fluxes noninvasively in the brain. Quantitative measurements of metabolic fluxes require analysis of the 13C labeling time courses obtained experimentally with a metabolic model. The present work reviews the ingredients necessary for a dynamic metabolic modeling study, with particular emphasis on practical issues.

1H-localized broadband 13C NMR spectroscopy of the rat brain in vivo at 9.4 T

Localized (13)C NMR spectra were obtained from the rat brain in vivo over a broad spectral range (15-100 ppm) with minimal chemical-shift displacement error (<10%) using semi-adiabatic distortionless enhancement by polarization transfer (DEPT) combined with (1)H localization. A new gradient dephasing scheme was employed to eliminate unwanted coherences generated by DEPT when using surface coils with highly inhomogeneous B(1) fields. Excellent sensitivity was evident from the simultaneous detection of natural abundance signals for N-acetylaspartate, myo-inositol, and glutamate in the rat brain in vivo at 9.4 T. After infusion of (13)C-labeled glucose, up to 18 (13)C resonances were simultaneously measured in the rat brain, including glutamate C2, C3, C4, glutamine C2, C3, C4, aspartate C2, C3, glucose C1, C6, N-acetyl-aspartate C2, C3, C6, as well as GABA C2, lactate C3, and alanine C3. (13)C-(13)C multiplets corresponding to multiply labeled compounds were clearly observed, suggesting that extensive isotopomer analysis is possible in vivo. This unprecedented amount of information will be useful for metabolic modeling studies aimed at understanding brain energy metabolism and neurotransmission in the rodent brain.

In vivo 13C NMR studies of compartmentalized cerebral carbohydrate metabolism

Localized 13C nuclear magnetic resonance (NMR) spectroscopy provides a unique window for studying cerebral carbohydrate metabolism through, e.g. the completely non-invasive measurement of cerebral glucose and glycogen metabolism. In addition, label incorporation into amino acid neurotransmitters such as glutamate (Glu), GABA and aspartate can be measured providing information on Krebs cycle flux and oxidative metabolism. Given the compartmentation of key enzymes such as pyruvate carboxylase and glutamine synthetase, the detection of label incorporation into glutamine indicated that neuronal and glial metabolism can be measured in vivo. The purpose of this paper is to provide a critical overview of these recent advances into measuring compartmentation of brain energy metabolism using localized in vivo 13C NMR spectroscopy. The studies reviewed herein showed that anaplerosis is significant in brain, as is oxidative ATP generation in glia and the rate of glial glutamine synthesis attributed to the replenishment of the neuronal Glu pool and that brain glycogen metabolism is slow under resting conditions. This new modality promises to provide a new investigative tool to study aspects of normal and diseased brain hitherto unaccessible, such as the interplay between glutamatergic action, glucose and glycogen metabolism during brain activation, and the derangements thereof in patients with hepatic encephalopathy, neurodegenerative diseases and diabetes.

Galactose metabolism in normal human lymphoblasts studied by (1)H, (13)C and (31)P NMR spectroscopy of extracts

The development of tools to follow and quantitate the fate of galactose in mammalian cells is crucial to the study and understanding of the inherited disorders of galactose metabolism. In this study we incubated normal human lymphoblasts with 1- or 2-(13)C galactose for 2.5 or 5 h and prepared TCA extracts of the cells. The various galactose metabolites were identified and quantified using a combination of proton, carbon and phosphorus NMR spectra. Galactose-1-phosphate (gal-1P), uridine diphosphogalactose, uridine diphosphoglucose and galactitol were present in the extracts. Average levels for gal-1P were around 10 nmol/mg protein and for uridine diphosphoglucose, uridine diphosphogalactose and galactitol in the range of 0.5-2 nmol/mg protein. Galactonate was never found in any conditions. Percentage labeling could be estimated for gal-1P and for the ribose carbons of AMP. The labeling agrees with a conversion of galactose to glucose through the Leloir pathway.

Single-shot, three-dimensional "non-echo" localization method for in vivo NMR spectroscopy

Metabolite signals with short T(1) or T(2) are difficult to localize with full sensitivity. This limitation was overcome with the development and implementation of a single-shot, complete three-dimensional "non-echo" localization method with reduced sensitivity to spatial B(1) variation, which is suitable for measuring signals with very short T(1) or T(2), e.g., the (13)C NMR signals of glycogen. The proposed method is based on a T(1)-optimized outer volume suppression scheme using pulses of the hyperbolic secant type applied at different power levels, which is robust over a fivefold range of T(1). Strong lipid, muscle glycogen, and glucose signals originating outside the rat brain were suppressed. Signals of glycogen, aspartate, glutathione, GABA C4, N-acetyl aspartate as well as the C3 and C4 signals of glutamate and glutamine with resolved homonuclear (13)C-(13)C coupling were fully resolved in vivo at 9.4 Tesla using higher-order shimming. The method can be extended to other nuclei and to localized MRS of humans.

Novel peak assignments of in vivo (13)C MRS in human brain at 1.5 T

(13)C MRS studies at natural abundance and after intravenous 1-(13)C glucose infusion were performed on a 1.5-T clinical scanner in four subjects. Localization to the occipital cortex was achieved by a surface coil. In natural abundance spectra glucose C(3beta,5beta), myo-inositol, glutamate C(1,2,5), glutamine C(1,2,5), N-acetyl-aspartate C(1-4,C=O), creatine CH(2), CH(3), and C(C=N), taurine C(2,3), bicarbonate HCO(-)(3) were identified. After glucose infusion (13)C enrichment of glucose C(1alpha,1beta), glutamate C(1-4), glutamine C(1-4), aspartate C(2,3), N-acetyl-aspartate C(2,3), lactate C(3), alanine C(3), and HCO(-)(3) were observed. The observation of (13)C enrichment of resonances resonating at >150 ppm is an extension of previously published studies and will provide a more precise determination of metabolic rates and substrate decarboxylation in human brain.

Chronic treatment of human astrocytoma cells with lithium, carbamazepine or valproic acid decreases inositol uptake at high inositol concentrations but increases it at low inositol concentrations

Inositol uptake was measured at concentrations of 25, 40 and 50 microM in human astrocytoma cell cultures treated for 1-3 weeks with pharmacologically relevant concentrations of LiCl, valproic acid or carbamazepine as well as in control cultures that had not been treated with any drug. After at least 2 weeks of treatment, each of these 3 conventional anti-bipolar drugs increased the uptake significantly at 25 microM inositol, had no effect at 40 microM, and decreased it at 50 microM inositol. Reduction of the drug concentrations by 50% abolished the stimulation of uptake at 25 microM inositol by lithium and valproic acid and reduced that by carbamazepine. These findings may contribute to an understanding of the mechanisms of action for anti-bipolar medication, and explain the controversy in the literature whether or not brain inositol is reduced after chronic administration of lithium.

Molecular aspects of magnetic resonance imaging and spectroscopy

Magnetic resonance imaging (MRI) is a well known diagnostic tool in radiology that produces unsurpassed images of the human body, in particular of soft tissue. However, the medical community is often not aware that MRI is an important yet limited segment of magnetic resonance (MR) or nuclear magnetic resonance (NMR) as this method is called in basic science. The tremendous morphological information of MR images sometimes conceal the fact that MR signals in general contain much more information, especially on processes on the molecular level. NMR is successfully used in physics, chemistry, and biology to explore and characterize chemical reactions, molecular conformations, biochemical pathways, solid state material, and many other applications that elucidate invisible characteristics of matter and tissue. In medical applications, knowledge of the molecular background of MRI and in particular MR spectroscopy (MRS) is an inevitable basis to understand molecular phenomenon leading to macroscopic effects visible in diagnostic images or spectra. This review shall provide the necessary background to comprehend molecular aspects of magnetic resonance applications in medicine. An introduction into the physical basics aims at an understanding of some of the molecular mechanisms without extended mathematical treatment. The MR typical terminology is explained such that reading of original MR publications could be facilitated for non-MR experts. Applications in MRI and MRS are intended to illustrate the consequences of molecular effects on images and spectra.

Localized 15N NMR spectroscopy of rat brain by ISIS

Three-dimensional image-selected in vivo spectroscopy (ISIS), combined with proton-decoupled nuclear-Overhauser-enhanced 15N nuclear magnetic resonance (NMR), was used to localize [15N]metabolites, observed by a head coil, to the brain in rats. In spontaneously breathing anesthetized rats given intravenous [15N]ammonium acetate infusion, brain [5-15N]glutamine was observed in the localized spectrum with a v1/2 of 5 Hz in 19-28 min at 4.7 T, while the signal from blood [15N]urea was eliminated by the localization process. In rats given [15N]leucine infusion, the peak representing predominantly (89%) brain [15N]glutamate was observed, with elimination of the signal from muscle [15N]alanine. In vivo peak areas of the brain [15N]metabolites in the localized spectra were proportional to their concentrations. The advantages and limitations of localization by ISIS using a volume coil with a homogeneous B1 field are compared with those of localization by a surface coil for in vivo 15N NMR study of neurotransmitters in the brain.

In vivo quantitation of cerebral metabolite concentrations using natural abundance 13C MRS at 1.5 T

A method for the quantitation of cerebral metabolites on a clinical MR scanner by natural abundance 13C MRS in vivo is described. Proton-decoupled spectra were acquired with a power deposition within FDA guidelines using a novel coil design. myo-Inositol, quantified by a separate proton MRS, and readily detectable in 13C MRS, was used as an internal reference. Normal concentrations, measured in four control subjects, age 7 months to 12 years, were glutamate 9.9 +/- 0.7, glutamine 5.6 +/- 1.0, and NAA 8.8 +/- 2.8 mmol/kg. In a patient diagnosed with Canavan disease, examined four times, glutamate was reduced to 46% of normal, 4.6 +/- 0.5 mmol/kg. NAA was increased by 50% to 13.2 +/- 1.6 mmol/kg in 13C MRS, consistent with the 41% increase to 12.3 +/- 1.1 from control 8.7 +/- 1.1 mmol/kg assayed by 1H MRS. Limited concentration of glutamate may impact on glutamatergic neurons and excitatory neurotransmission in Canavan disease. Quantitation of cerebral glutamate in human brain may have clinical value in human neuropathologies in which glutamate is believed to play a central role.

Heteronuclear cross polarization for enhanced sensitivity of in vivo 13C MR spectroscopy on a clinical 1.5 T MR system

The potential of heteronuclear ¿1H-13C¿ cross polarization was studied for optimization of the signal-to-noise ratio in in vivo 13C MR spectroscopy at the clinical field strength of 1.5 T. Experiments on the human calf showed a significant chemical-shift selective signal enhancement on triglyceride signals of 3.9 by heteronuclear cross polarization, compared to a standard pulse-acquire sequence. Studies on a neonatal piglet brain showed an enhancement by cross polarization of 2.2 for the detection of 13C-1-glucose. This enhancement allowed a fourfold improvement in time resolution in dynamic 13C MR of 13C-1-glucose inflow in piglet brain. Phantom experiments demonstrated the efficiency of this technique for interleaved detection of two spectral regions. Tests with a volume coil showed the feasibility of signal enhancement by cross polarization over a large volume of interest.

Resolution improvements in in vivo 1H NMR spectra with increased magnetic field strength

The measurement of cerebral metabolites using highly homologous localization techniques and similar shimming methods was performed in the human brain at 1.5 and 4 T as well as in the dog and rat brain at 9.4 T. In rat brain, improved resolution was achieved by shimming all first- and second-order shim coils using a fully adiabatic FASTMAP sequence. The spectra showed a clear improvement in spectral resolution for all metabolite resonances with increased field strength. Changes in cerebral glutamine content were clearly observed at 4 T compared to 1.5 T in patients with hepatic encephalopathy. At 9.4 T, glutamine H4 at 2.46 ppm was fully resolved from glutamate H4 at 2.37 ppm, as was the potential resonance from gamma-amino-butyric acid at 2.30 ppm and N-acetyl-aspartyl-glutamate at 2.05 ppm. Singlet linewidths were found to be as low as 6 Hz (0.015 ppm) at 9.4 T, indicating a substantial decrease in ppm linewidth with field strength. Furthermore, the methylene peak of creatine was partially resolved from phosphocreatine, indicating a close to 1:1 relationship in gray matter. We conclude that increasing the magnetic field strength increases spectral resolution also for 1H NMR, which can lead to more than linear sensitivity gains.

Identification of a high concentration of scyllo-inositol in the brain of a healthy human subject using 1H- and 13C-NMR

The peak at 3.35 ppm in the 1H-NMR spectrum characteristic for scyllo-inositol may be a marker for cerebral pathology, although it has a well-known constant concentration relative to myo-inositol. Such a peak was observed with an intensity at least 300% above normal in the brain of a healthy volunteer. The scyllo-inositol signal was assigned based on the detection of a corresponding peak at 74.5 ppm in the 13C-NMR spectrum and on the demonstration of singlet characteristics of the proton signal. The presence of substantial brain concentrations of scyllo-inositol suggests that scyllo-inositol metabolism may be regulated independently from myo-inositol and that such concentrations are compatible with normal health.

Broadband decoupled, 1H-localized 13C MRS of the human brain at 4 Tesla

Broadband proton decoupling of the entire 13C spectrum was possible within power absorption guidelines and resulted in the detection of narrow (as low as 2-3 Hz), natural abundance signals from metabolites such as myo-inositol, glutamate, N-acetyl-aspartate, and glutamine from 72 cm3 volumes in the human brain. To overcome the chemical shift displacement error, three-dimensional localization on the 1H z magnetization was combined with polarization transfer. Efficiency of the heteronuclear localization method was demonstrated by the elimination of all scalp lipid resonances. A signal-to-noise ratio of 5:1 for 0.07 mM [13C] was achieved in 12 min, which is approximately a fivefold improvement over the sensitivity reported at 2.1 Tesla.

Image directed proton spectroscopy of gerbil brain at 7 tesla

Image directed localized 1H NMR spectra were obtained at 7 T (300 MHz) from cubic volumes of less than 40 microL in the gerbil brain. Signals from cerebral metabolites commonly detected in other rodent species were easily assigned, and high resolution spectroscopy (360 MHz) of aqueous brain extracts assisted the tentative identification of partially overlapping resonances from lower concentration compounds like alanine, lysine, gamma-aminobutyrate, valine, leucine and isoleucine. Weak coupling at 7 T was manifest in the resolution of signals from the gamma-CH2 groups of glutamine and glutamate. Down-field of water, signals assigned to purine nucleotides were conspicuous in the extract spectra, but localized spectra acquired routinely in vivo, using selective excitation and gradient crushing (SUBMERGE) for water suppression, exhibited little or no signal from purines. When localized in vivo spectra were acquired without water suppression, however, or using a low power binomial excitation sequence rather than SUBMERGE, a broad signal appeared at the resonant frequency of purine aromatic protons. NMR experiments on the nucleotide adenosine 5'-monophosphate (AMP) in 90% glycerol/10% D2O solution demonstrated that pre-irradiation of the water signal even for less than 100 ms attenuated the nucleotide signal appreciably. This implies that the soft pulses required for selective excitation of water in sequences such as SUBMERGE induce spin-diffusion which eliminates or diminishes the signal from nucleotides in vivo.

Simultaneous determination of the rates of the TCA cycle, glucose utilization, alpha-ketoglutarate/glutamate exchange, and glutamine synthesis in human brain by NMR

13C isotopic tracer data previously obtained by 13C nuclear magnetic resonance in the human brain in vivo were analyzed using a mathematical model to determine metabolic rates in a region of the human neocortex. The tricarboxylic acid (TCA) cycle rate was 0.73 +/- 0.19 mumol min-1 g-1 (mean +/- SD; n = 4). The standard deviation reflects primarily intersubject variation, since individual uncertainties were low. The rate of alpha-ketoglutarate/glutamate exchange was 57 +/- 26 mumol min-1 g-1 (n = 3), which is much greater than the TCA cycle rate; the high rate indicates that alpha-ketoglutarate and glutamate are in rapid exchange and can be treated as a single combined kinetic pool. The rate of synthesis of glutamine from glutamate was 0.47 mumol min-1 g-1 (n = 4), with 95% confidence limits of 0.139 and 3.094 mumol min-1 g-1; individual uncertainties were biased heavily toward high synthesis rates. From the TCA cycle rate the brain oxygen consumption was estimated to be 2.14 +/- 0.48 mumol min-1 g-1 (5.07 +/- 1.14 ml 100 g-1 min-1; n = 4), and the rate of brain glucose consumption was calculated to be 0.37 +/- 0.08 mumol min-1 g-1 (n = 4). The sensitivity of the model to the assumptions made was evaluated, and the calculated values were found to be unchanged as long as the assumptions remained near reported physiological values.

Symbiosis between in vivo and in vitro NMR spectroscopy: the creatine, N-acetylaspartate, glutamate, and GABA content of the epileptic human brain

High resolution 1H NMR spectroscopy was used to analyze temporal lobe biopsies obtained from patients with epilepsy. Heat-stabilized cerebrum, dialyzed cytosolic macromolecules, and perchloric acid extracts were studied using one- and two-dimensional spectroscopy. Anterior temporal lobe neocortex was enriched in GABA, glutamate, alanine, N-acetylaspartate, and creatine. Subjacent white matter was enriched in aspartate, glutamine, and inositol. The N-acetylaspartate/creatine mole ratio was lower in anterior temporal neocortex with mesial (0.66) than neocortical (0.80) temporal lobe epilepsy. Human brain biopsy samples were separated into crude and refined synaptosomes, neuronal cell bodies, and glia using density gradient centrifugation. Neuronal fractions were enriched in glutamate and N-acetylaspartate. Glial cell fractions were enriched in lactate, glutamine, and inositol. The creatine content was the same in biopsied epileptic cortex (8.8-8.9 mmol/kg) and normal in vivo occipital lobe (8.9 mmol/kg). Glutamate content was higher in epileptic cortex at biopsy (10.1-10.5 mmol/kg) than normal in vivo occipital lobe (8.8 mmol/kg). GABA content was higher in biopsies of epileptic cortex (2.3-2.2 mmol/kg) than in normal in vivo occipital lobe (1.2 mmol/kg). N-acetylaspartate content was lower in biopsied epileptic temporal cortex (5.8-6.8 mmol/kg) than normal in vivo occipital lobe (8.9 mmol/kg). Paired in vivo and ex vivo measurements are critical for a firm understanding of the changes seen in the 1H-spectra from patients with epilepsy.

Analysis of magnetic resonance spectra by mole percent: comparison to absolute units

A variety of metabolites present in perchloric acid extracts of brain tissue were measured by 1H and 31P magnetic resonance spectroscopy (MRS) and HPLC in the same tissue sample and the MRS results were expressed both in terms of mole % and mumole/g based on an internal standard. The levels of 16 metabolites were compared by linear regression analysis and the mole % results were found to correlate very well with the results expressed as mumole/g. To compare the two units under typical experimental conditions, the percent change in metabolites in a group of Alzheimer's disease brains was compared to a control group using both units. The results were essentially identical for the mole % and mumole/g methods. We conclude that the use of the mole % method of expressing MRS data yields results which are equivalent to those expressed in absolute units and suggest that, for in vivo MRS studies, use of the mole % method is preferable because fewer artifacts, such as partial volume effects, are introduced.

Decrease in cerebral inositols in rats and humans

Rats with portacaval shunts and humans with hepatic encephalopathy show severe myo-inositol depletion in the brain. The portacaval-shunted rat may therefore be a useful model for the investigation of neurochemical pathways containing myoinositol, which are modulated not only in hepatic encephalopathy but also in diabetes mellitus and Alzheimer's disease.

Automatic, localized in vivo adjustment of all first- and second-order shim coils

The implementation of a "fast, automatic shimming technique by mapping along projections" (FASTMAP) on a 2.1-Tesla whole-body system is described. The method provides a localized adjustment of all first- (X, Y, Z) and second-order (Z2, ZX, ZY, X2-Y2, 2XY) coils in 2 min. The time savings are achieved by mapping the magnetic field along six projections rather than from whole-imaging data sets. An analysis of noise error propagation suggests that in 64-ml volumes the residual linewidths obtained with the method are negligible when the signal-to-rms-noise ratio is above 30. The initial application of the method to localized 13C, 31P, and 1H spectroscopy of the human brain resulted in linewidths of approximately 2 Hz for 13C (144-ml volume), approximately 2.5 Hz for 31P (36-ml volume) and approximately 4 Hz for 1H (36-ml volume) with symmetric and reproducible lineshapes.

Applications of magnetic resonance spectroscopy and diffusion-weighted imaging to the study of brain biochemistry and pathology

The first practical demonstration that nuclear magnetic resonance (NMR) spectroscopy could be applied to the study of brain biochemistry in vivo came in 1980, with the studies of the rat brain using a surface coil. Since then the technique has been rapidly and extensively developed into a versatile, non-invasive tool for the investigation of various aspects of brain biochemistry, physiology and disease. NMR is non-destructive and can be used to examine a wide variety of samples, ranging from localized regions within the whole brain in humans or animals, through tissue preparations (perfused organ, tissue slices and homogenates), to isolated cells and aqueous solutions, such as tissue extracts. 31P and 1H NMR spectra deriving from endogenous compounds of the brain in situ allow assessment of tissue metabolites and provide information about high-energy phosphates, lactate, certain amino acids, intracellular pH and ionic concentrations. Exogenous substrates or probes labelled with stable isotopes can also be introduced into the brain and used to monitor metabolism. Animal models of brain diseases have given some impetus to rapid progress in clinical NMR spectroscopy and also magnetic imaging techniques. The purpose of this article is to highlight the type of information available from these NMR techniques, and to present this in a neuroscience context, emphasizing the biochemical, physiological and pathological information that can be obtained using these methods.