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

https://doi.org/10.1016/j.mri.2006.01.003Get rights and content

Abstract

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.

Introduction

Carbon-13 NMR spectroscopy is a powerful tool to investigate intermediary metabolism. The high chemical specificity of 13C NMR, which can distinguish 13C isotope incorporation not only into different molecules, but also into specific carbon positions within the same molecule (13C isotopomers), allows one to follow the fate of 13C label through multiple metabolic pathways. However, the interpretation of 13C NMR data to derive quantitative metabolic fluxes requires analysis with a metabolic model. This metabolic modeling is particularly complex in the brain due to the highly organized interaction between different cell types corresponding to different metabolic compartments.

The potential of 13C NMR spectroscopy to study metabolic pathways was demonstrated using suspensions of microorganisms [1], [2]. The first attempt to model the flow of 13C label into the TCA cycle was made in 1983 by Chance et al. [3] in the heart, at a time when such analysis required some of the fastest computers available. Initial studies in the brain in vivo were reported in the late 1980s and early 1990s [4], [5], [6], [7]. Since then, the steady increase in magnetic fields available for in vivo studies and progress in NMR methodology have allowed detection of 13C labeling time courses in localized regions of the brain with constantly improving sensitivity. In parallel, methods for metabolic modeling have evolved from relatively simple models into complex two-compartment models. Together, these improvements have allowed 13C metabolic modeling studies to make important contributions to our understanding of brain metabolism and compartmentation, showing for example that the glutamate–glutamine cycle is a major metabolic pathway in the brain [8], [9], that the neuronal TCA cycle rate increases with neuronal activity [10], [11], [12], that glial TCA cycle is significant in the brain [13] and that anaplerotic pyruvate carboxylase activity is significant in the brain [13] and increases with neuronal activity [14].

The aim of this review is to provide the reader with an overview of the experimental design for a 13C metabolic modeling study. This is not intended as a comprehensive review of the field of 13C metabolic modeling, but rather as a practical guide for readers (not only NMR spectroscopists, but also neuroscientists in general) interested in understanding the experimental details of such studies. The emphasis has been placed on metabolic modeling rather than NMR spectroscopy, and NMR spectroscopy methodology is described only when it is relevant to metabolic modeling. We refer the reader to recent reviews for further details on NMR methodology [15], [16].

We have chosen to focus in this review on metabolic modeling which uses a one-compartment model and [1-13C]glucose or [1,6-13C2]glucose as a metabolic substrate. Other tracers and more complex metabolic models can be used, such as two-compartment (neuron–astrocyte) models. However the one-compartment model is relatively simple and is well suited for explaining the principles of metabolic modeling. We have also chosen to focus on “dynamic” metabolic modeling, meaning modeling of 13C labeling time courses. Therefore other analysis methods such as isotopomer analysis will not be considered in this review.

13C metabolic studies commonly involve four steps (Fig. 1):

  • (1)

    choice of 13C-labeled substrate and infusion protocol

  • (2)

    detection of 13C label incorporation into brain metabolites during infusion of 13C labeled substrate

  • (3)

    quantitation of 13C spectra to obtain 13C turnover curves

  • (4)

    metabolic modeling of 13C turnover curves to obtain quantitative fluxes through specific biochemical pathways

In the following sections, each of these steps will be examined separately, with a focus on the relevance of each aspect to metabolic modeling.

Section snippets

Choice of 13C-labeled substrate

The first step in the design of a 13C metabolic study is the choice of a metabolic substrate. Since glucose is the main fuel for the brain, 13C-labeled glucose has been the preferred substrate for metabolic studies in the brain. Most in vivo metabolic studies have been performed using [1-13C]glucose or [1,6-13C2]glucose. Both substrates lead to the formation of [3-13C]pyruvate, with [1-13C]glucose generating one unlabeled pyruvate and one labeled pyruvate per molecule of glucose, while [1,6-13C2

Detection and quantification of 13C label. What can be measured from NMR spectra in vivo?

The next step in the design of a 13C metabolic study is the detection of 13C label incorporation into brain metabolites during infusion of the 13C-labeled substrates and the quantification of the resulting NMR spectra. Since most TCA cycle intermediates are not concentrated enough to be detected by NMR in vivo, one relies on detection of 13C label in the larger pools of brain amino acids such as glutamate, glutamine, aspartate and (if sensitivity is sufficient) GABA.

A fundamental choice in the

Metabolic modeling

The fourth and final step in a 13C metabolic study is the analysis of 13C labeling time courses with a metabolic model to derive quantitative metabolic fluxes. Time courses of 13C labeling obtained after measurement and quantitation of 13C NMR spectra offer only limited metabolic information unless they are further analyzed using a metabolic model. For example, the rate of labeling of glutamate C4 depends not only on the TCA cycle rate (VTCA), but also on the exchange rate between

Conclusion

13C metabolic modeling is still an evolving field. Although this review has focused on the one-compartment model to illustrate the key points in metabolic modeling, much progress can be expected in the next few years as two-compartment models are being refined and validated. We expect Monte Carlo simulations to play a key role in the assessment of the robustness of two-compartment models in a variety of experimental conditions.

With the availability of very high magnetic fields for human studies

Acknowledgments

We would like to thank Dee Koski for technical assistance. This work was supported in part by NIH grants P41-RR08079, R01-NS38672, the Keck Foundation and the MIND Institute. The General Clinical Research Center was supported by NIH grant M01-RR00400.

References (49)

  • R. Gruetter et al.

    Localized 13C NMR spectroscopy in the human brain of amino acid labeling from d-[1-13C]glucose

    J Neurochem

    (1994)
  • N.R. Sibson et al.

    In vivo 13C NMR measurements of cerebral glutamine synthesis as evidence for glutamate–glutamine cycling

    Proc Natl Acad Sci U S A

    (1997)
  • F. Hyder et al.

    Oxidative glucose metabolism in rat brain during single forepaw stimulation: a spatially localized 1H[13C] nuclear magnetic resonance study

    J Cereb Blood Flow Metab

    (1997)
  • W. Chen et al.

    Study of tricarboxylic acid cycle flux changes in human visual cortex during hemifield visual stimulation using 1H-{13C} MRS and fMRI

    Magn Reson Med

    (2001)
  • N. Chhina et al.

    Measurement of human tricarboxylic acid cycle rates during visual activation by 13C magnetic resonance spectroscopy

    J Neurosci Res

    (2001)
  • R. Gruetter et al.

    A mathematical model of compartmentalized neurotransmitter metabolism in the human brain

    Am J Physiol

    (2001)
  • G. Oz et al.

    Neuroglial metabolism in the awake rat brain: CO2 fixation increases with brain activity

    J Neurosci

    (2004)
  • R. Gruetter et al.

    Localized in vivo 13C NMR spectroscopy of the brain

    NMR Biomed

    (2003)
  • R.A. de Graaf et al.

    In vivo 1H-[13C]-NMR spectroscopy of cerebral metabolism

    NMR Biomed

    (2003)
  • N.R. Sibson et al.

    In vivo 13C NMR measurement of neurotransmitter glutamate cycling, anaplerosis and TCA cycle flux in rat brain during [2-13C]glucose infusion

    J Neurochem

    (2001)
  • V. Lebon et al.

    Astroglial contribution to brain energy metabolism in humans revealed by 13C nuclear magnetic resonance spectroscopy: elucidation of the dominant pathway for neurotransmitter glutamate repletion and measurement of astrocytic oxidative metabolism

    J Neurosci

    (2002)
  • J.W. Pan et al.

    [2,4-13C2 ]-beta-Hydroxybutyrate metabolism in human brain

    J Cereb Blood Flow Metab

    (2002)
  • B. Ross et al.

    Clinical experience with 13C MRS in vivo

    NMR Biomed

    (2003)
  • R.B. Duckrow et al.

    Regional cerebral glucose utilization during hyperglycemia

    J Neurochem

    (1987)
  • Cited by (88)

    • INCA 2.0: A tool for integrated, dynamic modeling of NMR- and MS-based isotopomer measurements and rigorous metabolic flux analysis

      2022, Metabolic Engineering
      Citation Excerpt :

      Furthermore, parallel tracer experiments significantly increase the number of biological replicates required for flux assessment (Crown et al., 2016). Because of the capability to acquire NMR time-course data non-invasively (Cherix et al., 2021; Henry et al., 2006; Jeffrey et al., 2013; Judge et al., 2019), INST-MFA of dynamic 13C labeling studies offers an alternative approach to precisely characterize metabolic fluxes, while minimizing the number of parallel labeling experiments and biological replicates. Currently, the primary challenge in analyzing dynamic labeling studies is the lack of computational software available to model such datasets.

    • Biomolecular MRI reporters: Evolution of new mechanisms

      2017, Progress in Nuclear Magnetic Resonance Spectroscopy
    • Progress towards in vivo brain <sup>13</sup>C-MRS in mice: Metabolic flux analysis in small tissue volumes

      2017, Analytical Biochemistry
      Citation Excerpt :

      Metabolic steady state is usually assumed, which means that total concentrations of the metabolic pools as well as metabolic fluxes are constant throughout the infusion experiment. Studies tracking the time evolution of the pool sizes, as possible with 1H-[13C] MRS, have shown that the assumption of constant metabolic pool sizes is admissible for studies with infusion of 13C-glucose [3,48], while it might need refinement for other substrates, such as for [2–13C] acetate, where an increase in glutamate was measured [99]. The following paragraph will illustrate how differential equations can describe metabolite kinetics in a single-pool and more complex models.

    View all citing articles on Scopus
    View full text