From the Servicio de Neurocirugía, Hospital La Paz (J.M.P., F.C.,
J.M.R.), and Instituto de Investigaciones Biomédicas del C.S.I.C.
(S.C.), Madrid, Spain.
Correspondence to Dr Sebastian Cerdán, Instituto de Investigaciones Biomédicas del C.S.I.C., c/Arturo Duperier 4, E-28029 Madrid, Spain. E-mail scerdan{at}biomed.iib.uam.es
Methods—Right hemisphere infarct was induced in adult Long-Evans
rats by permanent occlusion of the right middle cerebral artery and
transitory occlusion of both common carotid arteries.
(1,2-13C2) acetate was infused for 60 minutes
in the right carotid artery immediately after carotid recirculation had
been re-established (1-hour group) or 23 hours later (24-hour group).
Extracts from both cerebral hemispheres were prepared and
analyzed separately by 13C nuclear magnetic
resonance and computer-assisted metabolic modeling.
Results—FCI decreased the oxidative metabolism of
glucose in the brain in a time-dependent manner. Reduced glucose
oxidation was compensated for by increased oxidations of
(13C) glutamate and (13C) GABA in the
astrocytes of the ipsilateral hemispheres of both groups. Increased
oxidative metabolism of (13C) glutamine in the
neurons was favored by increased activity of the neuronal pyruvate
recycling system in the 24–hour group.
Conclusions—Data were obtained consistent with
time-dependent changes in the utilization of glutamate and GABA or
glutamine as metabolic substrates for the glial or neuronal
compartments of rat brain after FCI.
The mammalian brain contains essentially two different
metabolic compartments: the glial and the neuronal. Classic
radiolabeling experiments and more recent 13C NMR evidence have shown
that these two compartments differ in their utilization of particular
substrates, their metabolic products, and the presence
of characteristic enzymes.7 8 9 10 The glial
compartment uses glucose or acetate as its main substrates, is
characterized by the presence of glutamine synthase, and produces
glutamine, which is transferred to the neuronal compartment and
metabolized. The neuronal compartment uses glucose and glial glutamine
as its main substrates, is characterized by the presence of high
glutaminase and GABA decarboxylase activities, and releases glutamate and GABA, which
are transferred to the glial compartment and
metabolized.11 In addition, glial lactate has
been proposed recently as an important substrate for the neuronal
compartment, at least in neural cell cultures of fetal or neonatal
origin during neuronal activation.12 The
metabolisms of neurons and glial cells interact closely,
competing for glucose as a primary substrate, and using the glutamate,
glutamine, and GABA produced in the other cell type as an alternative
substrate for oxidation in the tricarboxylic acid cycle. Thus, several
substrates are available to neurons and glial cells in situ, allowing
to compensate deficits in one substrate with increased utilization of
others.
The aim of this study was to investigate the use of glutamate,
glutamine, and GABA as alternative substrates to glucose by the
neuronal and glial compartments in both rat brain hemispheres after
partially reversible FCI. To this end we used a combination of
previously validated 13C NMR and mathematical
modelling techniques.13 14 15 The approach used
(1,2-13C2) acetate as a
substrate and computer-assisted analysis of extracts prepared
from the ipsilateral and contralateral hemispheres, 1 hour or 24 hours
after the ischemic insult. It has been shown that after FCI,
glutamate and GABA behave, in addition to their role as excitatory or
inhibitory transmitters, as true alternative substrates to
glucose in the glial compartment of the in situ brain, while glutamine
assumes a similar role in the neuronal compartment. These alterations
in substrate selection in both compartments are time dependent and
affect both the ipsilateral and contralateral hemispheres.
High Resolution 13C and 1H NMR Spectroscopy
of Brain Extracts
High-resolution 1H NMR spectra (360.1 MHz,
22°C) of the extracts used for 13C NMR
spectroscopy were acquired using the following:
90o pulses, 3968 Hz spectral width, 16K computer
memory (1.3-second acquisition time), 128 transients for each spectrum,
and 20-second relaxation delay. The residual water signal was
attenuated with a 3-second presaturating pulse applied with the
decoupler on the water resonance. Chemical shifts were referenced to
internal TSP. 1H and 13C
NMR assignments were made by comparison with published
values.13 23
Computer-Assisted Simulation of 13C NMR Spectra From
(1,2-13C2) Acetate Metabolism in
Both Hemispheres of Rat Brain After FCI
Under steady state conditions the METASIM program calculates the
probability of formation (or disappearance) of individual 13C isotopomers of the metabolites located in the network
nodes, using an algorithm based on the recursive definition of
input-output equations.24 Calculated 13C isotopomer populations are used to compute and display
the resulting 13C NMR spectrum as the weighted
sum of the individual 13C isotopomer
contributions, using predefined chemical shifts and 13C-13C coupling constants for each
carbon. The METASIM curve fits an experimental spectrum by
automatically optimizing the values of pool sizes and relative
metabolic fluxes converging on the nodes of the specified
network using an iterative, nonlinear least-squares procedure. Briefly,
the following procedure was followed for flux and pool size
optimization. Simulations of the 13C
multiplet structures of all the observable carbons of glutamate,
glutamine, and GABA in both cerebral hemispheres were performed with
METASIM, starting with the values for pool sizes and relative fluxes
previously validated in the normoxic brain.14 15
Progressive modifications of these starting values generated 13C NMR spectra that were virtually identical to the
experimental ones (c.f. Figure 5
Other Determinations
Materials
Effect of FCI on the 1H NMR Spectra of
Extracts From Both Hemispheres of Rat Brain After
(1,2-13C2) Acetate Infusion
Effects of FCI on the 13C NMR Spectra of Extracts From
Both Hemispheres of Rat Brain After (1,2-13C2)
Acetate Infusion
The multiplet structure of individual 13C
resonances is a direct consequence of the metabolic
pathways that produce and degrade the observed metabolite
carbon.13 14 15 26 The main pathways affecting the
relative intensities of the singlets, doublets, and multiplets observed
in the resonances of glutamate, glutamine, and GABA carbons are as
follows: the relative amounts of labeled and unlabeled acetyl-CoA
entering the neuronal and glial tricarboxylic acid cycles and the
relative exchanges of glutamate, glutamine, and GABA between the
neuronal and glial compartments (c.f. Figure 1
It is possible to obtain a more quantitative interpretation of the
changes observed in the multiplet structures of Figure 3
In the 1-hour group, the glial compartment of the ipsilateral
hemispheres showed decreased acetate oxidation (flux 3) and increased
oxidation of the (13C) glutamate (flux 6) and
(13C) GABA (flux 7) imported from the neurons.
The neuronal compartment showed increased oxidation of unlabeled
glucose (flux 8) and decreased contribution of the pyruvate recycling
system to the neuronal cycle (flux 10). The glial glutamate pool size
decreased. There was a trend for a decreased glial glutamine pool size
and increased neuronal glutamate pool size. These results are
consistent with astrocytic swelling. Twenty-four hours later,
acetate oxidation augmented (flux 1) in the glial compartment of the
ipsilateral hemisphere and increased (13C)
glutamate oxidation (flux 6) was maintained. The enhanced glucose
oxidation relative to pyruvate recycling seen in the ipsilateral
neuronal compartment of the 1-hour group was not seen at 24 hours.
Glutamine oxidation (flux 5) in the neuronal compartments of both
hemispheres increased as compared with the 1-hour group. It should be
mentioned that morphological studies indicate that the effects of focal
cerebral ischemia are heterogeneous in the
ipsilateral hemisphere. Heterogeneity includes the
presence of a zone of well-perfused healthy tissue and a necrotic core
surrounded by a penumbral zone with increasingly better
perfusion.18 19 20 Notably,
heterogeneity has also been reported in the
contralateral hemisphere.18 19 The relative flux
values and pool size measurements presented in Figure 5
Metabolic utilization of glutamate, glutamine, and GABA as
fuel alternatives or complementary to glucose is determined by their
relative availability in the extracellular fluid as well as by the
kinetics of their corresponding transport systems. The extracellular
concentration of these amino acids in normoxic and ischemic
brain tissues32 33 and the kinetics of their
uptake by cultures of cortical neurons or astrocytes have been
addressed previously.34 35 36 37 38 These studies allow
us to calculate values for the availability of different substrates for
the neuronal and glial metabolisms using in vitro
experiments (Table 2
Glucose is the main substrate for the adult brain under
physiological conditions. The rates of glucose
utilization as measured by the deoxyglucose method are approximately,
3.7 and 2.1 nmol glucose · min ·
mg protein–1 in cultivated astrocytes and
cortical neurons, respectively.39 Glutamate, GABA
and glutamine are known to be transported to astrocytes and neurons by
cell specific transporters.34 35 36 37 38 Given
physiological extracellular concentrations of
glutamate, GABA and glutamine of approximately 1, 3 and 94
µmol/L respectively,32 33 the amounts of
glutamate and GABA transported to the astrocyte can be calculated as
0.07 and 0.0042 nmol · min–1 · mg
protein–1, respectively, and the amount
calculated for glutamine transported to the neuron is 0.85 nmol
· min–1 · mg
protein–1. On the other hand, tricarboxylic acid
cycle rates of approximately 13 and 3 nmol
min–1 · mg
protein–1 have been proposed for the neuronal
and glial compartments in vivo.10 Thus, under
normoxic conditions complete oxidation of glutamate in the
tricarboxylic acid cycle of astrocytes would represent at most
2% of their tricarboxylic cycle flux, whereas complete oxidation of
glutamine would contribute at most 6% of the tricarboxylic acid cycle
flux in neurons. These results match well with those obtained in the
present study, indicating that the contribution of glutamate
oxidation to the energetics of the astrocytes (or glutamine to that of
neurons) is minor under physiological conditions.
This situation is observed in the contralateral hemisphere of the
1-hour group.
Under ischemic conditions the availability of extracellular
substrates for neurons and astrocytes undergoes a drastic change.
Glucose deprivation and hypoxia cause the
extracellular levels of glutamate, glutamine, and GABA to rise to
approximately 200, 250, and 20 µmol/L,
respectively.32 33 These concentrations saturate
the corresponding amino acid transport systems of neurons or
astrocytes, transiently increasing the metabolic
availability of glutamine or glutamate and GABA, in the respective
intracellular environments (Table 2
Notably, an increased contribution of the previously described pyruvate
recycling system13 is observed in ipsilateral
hemisphere of the 24-hour group as compared with the 1-hour group.
Neuroprotective actions of this pathway may include providing the
neurons with (1) an increased NADPH generation through mitochondrial
and synaptosomal malic enzymes40 and (2) a
mechanism to generate acetyl-CoA from glial glutamine under conditions
of reduced glucose oxidation.14 Indeed, our
current results show that the relative contribution of the pyruvate
recycling system to the neuronal cycle increases in the ipsilateral
hemisphere during the late phases of ischemic insult when
glucose oxidation is reduced and neuronal glutamine oxidation is
increased. Interestingly, recently bicarbonate has been shown to
modulate glutamine synthesis in cultivated
astrocytes.41 Thus, a reduction in tissue
bicarbonate caused by ischemic acidosis could contribute to the
reduced neuronal glutamine oxidation and pyruvate recycling observed in
the 1-hour group as compared with the 24-hour group. An alternative
mechanism in which an astrocytic pyruvate recycling would provide the
neurons with recycled lactate, rather than with glutamine, has been
proposed, using astrocyte cultures prepared from neonatal brain
tissues.42 43 Recently, enzymes of this pathway
have been shown to experience important changes in activity and
cellular localization during ontogenic development, and malic enzyme
and phosphoenolpyruvatecarboxykinase activities have been reported in
neuronal mitochondria and synaptosomes prepared from adult brain
tissues.40 44 It should also be mentioned here
that the use of lower doses of acetate could decrease the reported
contributions of recycling.
In summary, our current results are consistent with important
time-dependent changes in substrate selection during the development of
ischemic brain damage. Glutamate and GABA or glutamine appear
to be efficiently oxidized as alternative substrates to glucose by the
glial or neuronal compartments of the adult brain in situ.
Investigation of the effects of ischemic damage in the
different perfusion zones surrounding the necrotic foci constitutes an
attractive topic for future research in this field.
Received October 12, 1997;
revision received February 5, 1998;
accepted February 19, 1998.
2.
Huang R, Hertz L. Effect of anoxia on glutamate
formation from glutamine in cultured neurons: dependence on neuronal
subtype. Brain Res. 1994;660:129–137.[Medline]
[Order article via Infotrieve]
3.
Huang R, Shuaib A, Hertz L. Glutamate uptake and
glutamate content in primary cultures of mouse astrocytes during
anoxia, substrate deprivation and simulated
ischemia under normothermic and hypothermic
conditions. Brain Res. 1993;618:346–351.[Medline]
[Order article via Infotrieve]
4.
Kimmelberg HK, Rutledge E, Goderie S, Charniga C.
Astrocytic swelling due to hypotonic or high K+
medium causes inhibition of glutamate and aspartate uptake and
increases their release. J Cereb Blood Flow Metab. 1995;15:409–416.[Medline]
[Order article via Infotrieve]
5.
Badar-Goffer RS, Bachelard HS, Morris PG.
Neuronal-glial metabolism under depolarizing conditions:
a 13C NMR study. Biochem J. 1992;282:225–230.
6.
Cox DWG, Morris PG, Bachelard HS. Kinetic
analysis of the cerebral creatine kinase reaction under hypoxic
and hypoglycemic conditions in vitro. Biochem J. 1988;255:523–527.[Medline]
[Order article via Infotrieve]
7.
Bachelard H, Badar-Goffer R. NMR spectroscopy in
neurochemistry. J Neurochem. 1993;61:412–429.[Medline]
[Order article via Infotrieve]
8.
Berl S, Clarke DD. The metabolic
compartmentation concept. In: Hertz L, Kvame E, Mc Geer G, Schousboe A,
eds. Glutamate, Glutamine and GABA in the Central Nervous
System. New York, NY: Alan R. Liss Inc; 1983:205–217.
9.
Van den Berg CJ, Krzalic LJ, Mela P, Waelch H.
Compartmentation of glutamate in brain: evidence for the existence of
two different tricarboxylic acid cycles in brain. Biochem J. 1969;113:281–289.[Medline]
[Order article via Infotrieve]
10.
Van den Berg CJ, Garfinkel D. A simulation study of
brain compartments: metabolism of glutamate and related
substances in mouse brain. Biochem J. 1971;123:211–218.[Medline]
[Order article via Infotrieve]
11.
Erecinska M, Silver I. Metabolism and role
of glutamate in mammalian brain. Prog Neurobiol. 1990;35:245–296.[Medline]
[Order article via Infotrieve]
12.
Tsacopoulos M, Magistretti PJ. Metabolic
coupling between glia and neurons. J Neurosci. 1996;16:877–885.
13.
Cerdán S, Künnecke B, Seelig J.
Cerebral metabolism of
(1,2-13C2) acetate as
detected by in vivo and in vitro 13C NMR.
J Biol Chem. 1990;265:12916–12926.
14.
Künnecke B, Cerdan S, Seelig J. Cerebral
metabolism of
(1,2-13C2) glucose and
(U-13C4)
ß-hydroxybutyrate as detected by in vivo and in
vitro 13C NMR. NMR in Biomed.. 1993;6:264–277.
15.
Chapa F, Künnecke B, Calvo R, Escobar del Rey F,
Morreale de Escobar G, Cerdán S Adult onset hypothyroidism and
the cerebral metabolism of
(1,2-13C2) acetate as
detected by. 13C NMR. Endocrinology. 1995;136:296–305.[Abstract]
16.
Chen ST, Hsu CY, Hogan EL, Maricq MD, Balentine JD. A
model of focal ischemic stroke in the rat: reproducible
extensive cortical infarction. Stroke. 1986;17:738–743.
17.
Liu TH, Beckman JS, Freeman BA, Hogan EL, Hsu CY.
Polyethylene glycol-conjugated superoxide dismutase and catalase reduce
ischemic brain injury. Am J Physiol. 1989;256:H589–H593.
18.
Avendaño C, Roda JM, Carceller F,
Díez-Tejedor E. Morphometric study of focal cerebral
ischemia in rats: a stereological evaluation. Brain
Res. 1995;673:83–92.[Medline]
[Order article via Infotrieve]
19.
Roda JM, Carceller F, Díez-Tejedor E,
Avendaño C. Reduction of infarct size by
intra-arterial nimodipine administration at reperfusion in
a rat model of partially reversible focal ischemia.
Stroke. 1995;26:1888–1892.
20.
Roda JM, Carceller F, Pajares RL, Díez-Tejedor
E. Prevention of cerebral ischaemic reperfusion by
intraarterial administration of superoxide dismutase in the
rat. Neurol Res. 1991;13:160–163.[Medline]
[Order article via Infotrieve]
21.
Carceller F, Roda JM, Gutiérrez-Molina M,
López-Pajares R, Albariño AR, Díez-Tejedor E. Can
nimodipine prevent ischemic reperfusion in the rat brain?
Neurol Res. 1993;15:395–400.[Medline]
[Order article via Infotrieve]
22.
Preece NE, Cerdan S. Metabolic precursors
and compartmentation of cerebral GABA in vigabratin treated rats.
J Neurochem. 1996;67:1718–1725.[Medline]
[Order article via Infotrieve]
23.
Cerdán S, Parrilla R, Santoro J, Rico M.
1H NMR detection of cerebral myo-inositol.
FEBS Lett. 1985;187:167–172.[Medline]
[Order article via Infotrieve]
24.
Malloy CR, Sherry AD, Jeffrey FMH. Analysis of
the tricarboxylic acid cycle of the heart using 13C isotope isomers. Am J Physiol. 1990;259:H987–H995.
25.
Künnecke B. Computer assisted simulation and
interpretation of metabolic data derived from 13C NMR spin coupling patterns. PhD Thesis. University of
Basel, Switzerland, 1991:155–327.
26.
Künnecke B. Application of 13C NMR spectroscopy to metabolic studies in
animals. In: Beckman N, ed. Carbon-13 NMR Spectroscopy of
Biological Systems. New York, NY: Academic Press; 1995:159–267.
27.
Künst A, Draeger B, Ziegenhorn J. D-glucose:UV
methods with hexokinase and Glucose 6-phosphate dehydrogenase. In:
Bergmeyer HU, Bergmeyer J, Grassl M, eds. Methods of Enzymatic
Analysis. Weinheim, Germany: Verlag-Chemie;
1984;6:163–172.
28.
Chang JY, Knecht R, Brown DG. Amino acid
analysis in the picomole range by precolumn derivatization and
HPLC. Methods Enzymol. 1981;91:41–48.
29.
Mies G, Paschen W, Hossman KA. Cerebral blood flow,
glucose utilization, regional glucose and ATP content during the
maturation period of delayed ischemic injury in gerbil brain.
J Cereb Blood Flow Metab. 1990;10:638–645.[Medline]
[Order article via Infotrieve]
30.
Gadian DG, Williams SR, Bates TE, Kauppinnen RA. NMR
spectroscopy: current status and future possibilities. Acta
Neurochir. 1993;57:1–8.
31.
McKenna MC, Sonnewald U, Huang X, Stevenson J, Zielke
HR. Exogenous glutamate concentration regulates the
metabolic fate of glutamate in astrocytes. J
Neurochem. 1996;66:386–393.[Medline]
[Order article via Infotrieve]
32.
Hagberg H, Lehmenn A, Sandberg M, Nyström B,
Jacobson I, Hamberger A. Ischemia induced shift of
inhibitory and excitatory amino acids from intra to
extracellular compartments. J Cereb Blood Flow Metab. 1985;5:413–419.[Medline]
[Order article via Infotrieve]
33.
Takagi K, Ginsberg M, Globus MYT., Dietrich D, Martinez
E, Kraydieh S, Busto R. Changes in amino acid neurotransmitters and
cerebral blood flow in the ischemic penumbral region following
middle cerebral artery occlusion in the rat: correlation with
histopathology. J Cereb Blood Flow Metab. 1993;13:575–585.[Medline]
[Order article via Infotrieve]
34.
Yu ACH, Hertz, L. Uptake of glutamate, GABA and
glutamine into a predominantly GABAergic and a predominantly
glutamatergic nerve cell population. J Neurosci Res. 1982;7:23–35.[Medline]
[Order article via Infotrieve]
35.
Garlin AB, Sinor A, Sinor JD, Jee SH, Grinspan JB,
Robinson MB. Pharmacology of sodium dependent high affinity
L-[3H]glutamate transport in glial cultures.
J Neurochem. 1995;64:2572–2580.[Medline]
[Order article via Infotrieve]
36.
Schousboe A, Svenneby G, Hertz L Uptake and
metabolism of glutamate in astrocytes cultured from
dissociated mouse brain hemispheres. J Neurochem. 1977;29:999–1005.[Medline]
[Order article via Infotrieve]
37.
Lewin L, Mattsson MO, Rassin DK, Sellstrom A. On the
activity of
38.
Nilsson M, Hansson E, Rönnback L. Interactions
between valproate, glutamate, aspartate and GABA with respect to uptake
in astroglial primary cultures. Neurochem Res. 1992;17:327–332.[Medline]
[Order article via Infotrieve]
39.
Peng L, Zhang X, Hertz L. High extracellular potassium
concentrations stimulate oxidative metabolism in a
glutamatergic neuronal culture and glycolysis in cultured astrocytes
but have no stimulatory effect in GABAergic neuronal culture.
Brain Res.. 1994;663:168–172.[Medline]
[Order article via Infotrieve]
40.
Cruz F, Scott R, Barroso I, Santisteban P, Cerdán
S. Ontogeny and cellular localization of the pyruvate recycling system
in rat brain. J Neurochem. 1998 (in press).
41.
Gamberino WC, Berkich DA, Lynch CJ, Xu B, La Noue K.
Role of pyruvate carboxylase in facilitation of synthesis of glutamate
and glutamine in cultured astrocytes. J. Neurochem.. 1997;69:2312–2325.[Medline]
[Order article via Infotrieve]
42.
Sonnewald U, Westegard N, Jones P, Taylor A, Bachelard
H, Schousboe A. Metabolism of (U-13C)
glutamine in cultured astrocytes studied by NMR spectroscopy: first
evidence of astrocytic pyruvate recycling. J Neurochem. 1996;67:2566–2572.[Medline]
[Order article via Infotrieve]
43.
Hassel B, Sonnewald U. Glial formation of pyruvate and
lactate from TCA cycle intermediates: implications for the inactivation
of transmitter amino acids. J Neurochem. 1995;65:2227–2234.[Medline]
[Order article via Infotrieve]
44.
Vogel R, Hamprecht B. Enzymatic and immunocytochemical
studies for localizing mitochondrial malic enzyme in brain.
J Neurochem. 1997;69(suppl):S170B. Abstract.
Anesthesiology/Critical
Care Medicine The Johns Hopkins
University Baltimore, Maryland
In the study of Pascual et al, the authors use an innovative approach
of carbon-13 NMR spectroscopy to track changes of carbon-13
incorporation into key energy substrates and to allow computer
simulation of metabolic pathways that can predict changes
in glutamate, glutamine, and GABA oxidation during focal
ischemia in the rat. Ischemia was induced by permanent
middle cerebral artery occlusion plus bilateral carotid occlusion for
90 minutes. The critical feature of the experiment is the use of
carbon-13 in both the one and two carbon positions of acetate. When
these two adjacent carbons are incorporated into the TCA cycle, they
provide both singlet and multiplet signatures on the NMR spectra for
glutamate, glutamine, GABA, and lactate carbons. Labeled carbon
positions depend on the number of turns of the TCA cycle and on a
pyruvate cycling pathway. By assuming two metabolic
compartment pools based in neurons and glia, the authors conclude (1)
that oxidation of neuronally derived glutamate and GABA increases in
the glial compartment after 90 minutes of focal ischemia and
(2) that oxidation of glial-derived glutamine increases in the neuronal
compartment in conjunction with an increase in activity of the pyruvate
recycling pathway after 24 hours of focal ischemia. The latter
effect of increased glutamine oxidation also occurred in cortex
contralateral to the middle cerebral artery occlusion on the day after
bilateral carotid occlusion. These results are consistent with
the notions (1) that glia increase their utilization of carbon
skeletons derived from amino acids for energy metabolism
during the early period of focal ischemia and (2) that neurons
outside the ischemic core and surviving the moderate
ischemia associated with bilateral carotid occlusion increase
their utilization of carbon skeletons derived from amino acids for
energy metabolism 1 day after the insult.
In performing the computer simulation of the metabolic
fluxes, the authors used the most simplified model consistent
with the information content of their data. Reducing the model to the
simplest elements is often necessary for computer iterations to
converge on a solution. However, when applying these models to the in
vivo situation with multiple neuronal and glial subtypes and
subcellular compartments, some of which will be selectively altered by
the ischemic insult, one needs to critically assess the
simplifying assumptions of the model.
The main assumptions of the model are that (1) all of the labeled
acetate is metabolized in the glial compartment, (2) all of the
glutamine is synthesized in the glial compartment, and (3) all of the
pyruvate recycling occurs in the neuronal compartment. These
assumptions, which are based on reports from the literature and
previous work from this laboratory, are reasonable first
approximations. However, future studies need to be directed at
validating these assumptions during ischemia and at further
refining the model to include additional pathways.
For example, blood-brain barrier injury during ischemia may
increase intracellular availability of
13C-acetate relative to glucose, particularly in
neurons if 13C-acetate can more readily diffuse
past astrocyte foot processes. Because 13C-NMR is
relatively insensitive, a 1.5 millimolar plasma concentration of
labeled acetate was used rather than a tracer concentration. Thus, it
is possible that substantial changes in intracellular acetate
availability during ischemia could occur and influence the rate
of glycolysis. The increased signal intensity across all peaks of the
24-hour spectra might be due to increased acetate transport across the
barrier.
Another consideration is the gliosis and increased glial fibrillary
acidic protein–staining typically seen 1 day after ischemia.
Such phenotypic changes in astrocytes may be accompanied by an altered
expression of metabolic enzymes. Thus, pyruvate recycling
enzymes present in neonatal glial cultures could become prominent
in postischemic astrocytes in vivo and modify the
assumption of selective neuronal localization. Because the observed
changes in the ratio of singlet-to-doublet resonances in C4 glutamate
are presumed to reflect changes in pyruvate recycling activity, changes
in cell localization of this pathway during ischemia will have
a considerable impact on the data interpretation. Moreover, an
increased fraction of the glutamine synthesized in glia may be
metabolized by glutaminase in glial mitochondria rather than in
neuronal mitochondria when neurons are selectively injured. Finally,
the model does not consider the potential transport of carbon between
astrocytes and neurons in the form of lactate, which is known to
increase during ischemia.
This study breaks new ground in demonstrating increased use of
alternative energy substrates during ischemia. However, further
work is needed to validate the assumptions of the model in
postischemic tissue, to expand the number of
metabolic compartments and pathways, and to use other
carbon-labeled substrates such as 13C-glucose to
complement the information derived from
13C2-acetate.
Received October 12, 1997;
revision received February 5, 1998;
accepted February 19, 1998.
© 1998 American Heart Association, Inc.
Original Contributions
Glutamate, Glutamine, and GABA as Substrates for the Neuronal and Glial Compartments After Focal Cerebral Ischemia in Rats
Abstract
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Background and Purpose—Even though
the utilization of substrates alternative to glucose may play an
important role in the survival of brain cells under ischemic
conditions, evidence on changes in substrate selection by the adult
brain in vivo during ischemic episodes remains very limited.
This study investigates the utilization of glutamate, glutamine, and
GABA as fuel by the neuronal and glial tricarboxylic acid cycles of
both cerebral hemispheres after partially reversible focal cerebral
ischemia (FCI).
Key Words: cerebral ischemia, focal • cerebral metabolism • middle cerebral artery occlusion • neuronal damage • stroke, experimental
Introduction
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Some of the
metabolic mechanisms underlying the development of
ischemic brain damage have been investigated using highly
enriched cultures of neural cells1 2 3 4 or cerebral
cortex slices.5 6 However, the
metabolic events occurring in vivo in the neuronal and
glial compartments of the adult brain during the development of
ischemic damage remain less understood.
Materials and Methods
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Animals, Surgical Procedures, and Experimental Design
The experimental protocols used in this study were approved by
appropriate institutional review committees, and meet the guidelines of
the responsible governmental agency. Long-Evans female rats (250±25 g,
n=10), receiving rat chow and water ad libitum, were subjected to focal
cerebral ischemia after an overnight fasting period.
Anesthesia was induced with 2.5 mL/kg body weight of a
mixture of ketamine hydrochloride (25 mg/mL), diazepam (2
mg/mL), and atropine (0.1 mg/mL) injected
intraperitoneally, and prolonged when necessary
with one third of the initial dose. Animals were allowed to breathe
unassisted. The surgical procedure was essentially that described by
Chen et al in 198616 and Liu et al in
1989,17 with the modifications previously
reported.18 19 Briefly, under a surgical
microscope, a small craniectomy was made over the main trunk of the
right middle cerebral artery and above the rhinal fissure. The right
middle cerebral artery was permanently ligated just before its
bifurcation between the frontal and parietal branches. During this
occlusion period, the right external carotid artery was cannulated
retrogradely, and the catheter was placed near the origin of the
internal carotid artery.20 21 Both common carotid
arteries were then exposed and clamped for 90 minutes. Restoration of
carotid blood flow was directly observed in all cases, after removal of
the clamps. Then, the animals were infused with
(1,2-13C2) acetate solution
(32 µmol · min–1 · 100 g
body wt–1, pH 7.2) for 60 minutes, either
immediately after carotid blood flow was reestablished (1-hour group,
n=5) or 23 hours later (24-hour group, n=5). The right femoral artery
was cannulated during surgery and subsequent infusion, allowing
continuous monitoring of arterial pressure. At the end of
the infusion, 0.3-mL blood samples were drained from both jugular
veins and from the femoral artery to determine arterial and
venous blood gas values, pH, and glucose. Body temperature was
maintained at 37±0.5°C during the experiment with a servo-controlled
rectal probe heating pad. An additional thermistor probe was placed in
the region of the middle cerebral artery under the temporal muscle and
over the cerebral cortex, while brain temperature was maintained at
36±0.5°C with a tungsten lamp located over the head. At the end of
the infusion, the head of the animals was funnel-frozen with liquid
nitrogen. The ipsilateral and contralateral hemispheres were dissected
separately while still at low temperature, powdered under liquid
nitrogen, and extracted separately with methanol/HCl/perchloric
acid.22 Extracts were lyophilized to dryness and
resuspended in D2O (99.9%D) before high
resolution 13C and 1H
NMR.
High resolution 13C NMR spectra (90.56
MHz, 22°C, pH 7.4) of neutralized perchloric acid extracts of
individual hemispheres, were obtained at 8.4 T with a Bruker AM-360 NMR
spectrometer (Bruker Analytik GMBH). Acquisition conditions were as
follows: 60o pulses, 18.5-KHz sweep width, 64K
word computer memory (1.769-second acquisition time), total recycle
time of 5.7-seconds, and approximately 22 000 scans. Broad band
de-coupling (1 watt average forward power) was applied only during the
acquisition. Chemical shifts were calibrated with the resonance of a
10% (vol/vol) dioxane solution (67.4 ppm) placed in a coaxial
capillary.
Computer-assisted interpretation of the 13C NMR spectra from extracts of each cerebral hemisphere,
in both time groups, was performed using the METASIM program version
2.1.14 METASIM simulates the 13C NMR spectrum produced by the metabolism of
a given 13C-labeled substrate in a user-defined
metabolic network consisting of metabolite pool sizes and
interconnecting fluxes (Figure 1
). The
network consists of two main cerebral compartments, namely neuronal and
glial, with extracellular spaces for plasma and
interstitial fluid. The neuronal compartment groups both
glutamatergic and GABAergic neurons, whereas the glial compartment
contains both astroglia and oligodendroglia.
View larger version (94K):
[in a new window]
Figure 1. Metabolic network used to simulate
cerebral metabolism of (1,2-13C2)
acetate in both hemispheres of rat brain after FCI. The network
consists of fluxes (circled numbers) and metabolite nodes (abbreviated
names). Two metabolic compartments associated with the
glial (left) or neuronal (right) environments exist in the brain.
Infused (1,2-13C2) acetate (flux 1) is
activated to (1,2-13C2) acetyl-CoA only
in the glial compartment, where it is diluted with unlabeled acetyl-CoA
from unlabeled glucose (flux 2), originating the glial acetyl-CoA pool
(Node AcCoAg). AcCoAg enters the glial cycle
(flux 3) labeling the small glutamate pool (Node Glug) and
the glutamine pool (Glng). The glial compartment also
metabolizes glutamate (flux 6) and GABA (flux 7) produced in the
neurons (Nodes Glug and SUCCg, respectively).
The neuronal tricarboxylic acid cycle uses unlabeled glucose (flux 8)
and some glial glutamine (flux 5) as substrates (Nodes
AcCoAn and Glun, respectively).
AcCoAn enters the neuronal cycle (flux 9) labeling the
large neuronal glutamate pool (Node Glun), which is the
main precursor of GABA. Only the glial compartment contains glutamine
synthase activity.11 The neuronal compartment contains the
glutamate decarboxylase and most of the glutaminase activity as well as
the pyruvate recycling system (flux 10). The METASIM program was used
for the simulations.14 15 ). Finally, the set of refined
parameter values was automatically optimized by curve
fitting the 13C multiplets of the experimental
spectra to those of the simulated ones. More detailed descriptions of
METASIM have been provided.14 25 26
View larger version (35K):
[in a new window]
Figure 5. Pool sizes and relative flux values obtained from
computer-assisted interpretations of 13C NMR spectra from
extracts of both hemispheres of rat brain infused with
(1,2-13C2) acetate, 1 hour and 24 hours after
FCI. Values were optimized iteratively using METASIM as indicated in
"Materials and Methods" (c.f. References 14 and 15) and illustrated
in Figure 4 
. Results represent the mean±SD of the fittings
performed on four different 13C NMR spectra of extracts
from the ipsilateral (dark bars) and contralateral (light bars)
hemispheres of the 1-hour and 24-hour groups, respectively. Circled
numbers and node names refer to the nomenclature of Figure 1.
Subscripts "g" or "n" refer to the glial or neuronal glutamate
pools. Relative flux refers to the probability (Pi) of
occurrence of a flux (0 < Pi < 1), as compared with
the added probability of occurrence of all fluxes converging on the
same node taken arbitrarily as one (
Pi=1).
*P<0.05, §P<0.01,
#P<0.001.
MABP was recorded in the femoral artery during the infusion
using a pressure transducer (Schiller). Once the acetate solution had
been infused, PO2,
PCO2, and pH were determined in
samples from arterial and venous blood taken from the
femoral artery and jugular veins, respectively, using a clinical
automatic analyzer (Nova Biomedical Analyzer). Glucose
was determined spectrophotometrically with an NADP-coupled
assay.27 Total amino acid concentrations were
measured by reverse-phase HPLC.28 The Student's
t test for unpaired values was used to assess statistical
significance (P<0.05, P<0.01, and
P<0.001) between the groups.
(1,2-13C2) acetate
(99.9% 13C) and D2O
(99.9% D) were purchased from ISOTEC Inc. The rest of the reagents
were of the highest purity available from Sigma Chemical Co.
Results
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Physiological Parameters of
Arterial and Venous Blood After FCI
Table 1 shows
physiological parameters measured in
arterial and venous blood in the course of FCI.
Measurements performed in the right and left jugular veins gave very
similar results, and the mean value was taken as
representative of venous blood in every animal. MABP
remained within normal values before and during the carotid occlusion
period, but decreased slightly in the post-occlusion period. After the
infusion of (1,2-13C2)
acetate, arteriovenous differences in glucose were larger in the 1-hour
group (1.2±0.12 mmol/L) than in the 24-hour group (0.7±0.07
mmol/L). Venous pH was lower in the 1-hour group than in the 24-hour
group. Similar metabolic changes have been reported
previously.29
View this table:
[in a new window]
Table 1. Physiological
Parameters in Arterial and Venous Blood of Rats
in the Course of FCI Followed by Infusion of
(1,2-13C2) Acetate
Figure 2 shows
representative 1H NMR spectra
obtained from extracts of the contralateral (left) and ipsilateral
(right) cerebral hemispheres from rats infused with
(1,2-13C2) acetate in the
1-hour (top) and 24-hour (bottom) groups. Spectra obtained from the
ipsilateral hemispheres showed higher Lac/NAA ratios (1.32±0.12 and
0.78±0.02) than the contralateral hemispheres (0.27±0.055 and
0.20±0.03). The latter Lac/NAA ratios are similar to those previously
reported for normoxic brains.30 1H NMR spectra
from the ipsilateral hemispheres also showed more intense signals for
the acetate methyl group (resonance 2) than those from the
contralateral hemispheres, a situation that was specially apparent in
the 24-hour group. In this case, 13C satellites
corresponding to the methyl signals of
(1,2-13C2) acetate were
clearly visible (arrows). This observation indicates that
(1,2-13C2) acetate
accumulates in the ipsilateral hemispheres, probably because of reduced
oxidation in the tricarboxylic acid cycle. No other significant
alterations could be detected in the remaining metabolites observable
by 1H NMR. Amino acid analysis revealed
no significant difference in the total concentrations of glutamate,
glutamine, and GABA between the ipsilateral and contralateral
hemispheres of the 1-hour and 24-hour groups (not shown).
View larger version (23K):
[in a new window]
Figure 2. Representative 1H NMR
spectra (360.13 MHz) of extracts from both hemispheres of rat brain
infused with (1,2-13C2) acetate, 1 hour or 24
hours after FCI. Permanent occlusion of the middle cerebral artery was
performed in the right (ipsilateral) hemisphere. Only the aliphatic
portion of the spectra is shown. 1 indicates lactate H3; 2, Ac H2
(12C bonded); 3, NAA H6; 4, Glu-Gln H3, H3'; 5, Glu H4,
H4'; 7, NAA H3, H3'; 8, Cr (PCr) methyl; 9, trimethyl ammonium groups
of choline derivatives; 10, Ino H1, H3; 11, Cr (PCr) methylene; and 12,
Ino H2. Arrows indicate 13C satellites of
(1,2-13C2) acetate H2.
Figure 3 depicts the aliphatic
portion of representative proton-decoupled 13C NMR spectra of extracts from both cerebral hemispheres
of rats infused with
(1,2-13C2) acetate, either
1 hour after FCI (top) or 24 hours later (bottom). Resonances from the
C2, C3, and C4 carbons of glutamate and glutamine and the C2 carbon of
GABA were clearly detected in all spectra (see Figure legend for
assignments). Even though infusion conditions were identical,
intensities of the 13C signals were lower in the
1-hour group than in the 24-hour group. In most cases, these 13C resonances depicted an apparent triplet structure
produced by the superposition of a singlet and a doublet. Singlets are
derived from metabolites containing the observed 13C bonded to12C neighbors, while
doublets, and doublets of doublets, are derived from molecules
containing the observed 13C bonded to one or
two 13C neighbors,
respectively.13 14
View larger version (23K):
[in a new window]
Figure 3. Representative proton
decoupled 13C NMR spectra (90.55 MHz) of extracts from both
hemispheres of rat brain infused with
(1,2-13C2) acetate, 1 hour and 24 hours after
FCI. Permanent occlusion of the middle cerebral artery was performed in
the right (ipsilateral) hemisphere. Only the aliphatic portion of the
spectra is shown. 14 indicates lactate C3; 15, Gln C3; 16, Glu C3; 17,
Gln C4; 18, Glu C4; 19, GABA C2; 20, Gln C2; and 21, Glu C2. ). A comparison of
the 13C NMR spectra shown in Figure 3 reveals
that the relative proportions of singlets, doublets, or multiplets
corresponding to the same carbon resonance are different in the
ipsilateral and contralateral hemispheres, irrespective of the time
elapsed after FCI. These results reveal differences in substrate
selection between the two hemispheres. Of particular interest is the
ratio of singlets to doublets in the glutamate C4 resonance (number 18
in Figure 3). This ratio provides an estimate of the relative flux
through the pyruvate recycling system, with larger values of the ratio
corresponding to higher relative fluxes through the recycling
pathway.13 14 in terms of
metabolic fluxes using the METASIM program. Figure 4 illustrates this procedure with a
comparison of representative experimental spectra (top)
from the ipsilateral (right panels) and contralateral (left panels)
hemispheres, with the METASIM simulations (bottom) obtained using the
parameters indicated in the legend. Similar simulations
were performed with five animals of the 1-hour group and five animals
of the 24-hour group. Values of pool size and relative
metabolic flux that best fit the model described in Figure 1 in these animals are summarized in Figure 5.
View larger version (27K):
[in a new window]
Figure 4. Representative experimental (top)
and simulated (bottom) 13C NMR spectra from extracts of
each cerebral hemisphere 24 hours after FCI. Simulations were performed
using the METASIM program configured to calculate the 13C
NMR spectra resulting from (1,2-13C2) acetate
metabolism in the network described in Figure 1 . Fifteen
iterations were used with an artificial line broadening of 1 Hz and a
simulated signal/noise ratio of 35/1. Pool size (µmol/g wet weight)
and relative flux values used for the depicted simulations of
ipsilateral (contralateral) hemispheres are Glug 0.66
(0.16); Glun 5.0 (5.5); Glutamine 7.5 (6.1); GABA 1.2
(1.2); flux 1 0.4 (0.45); flux 2 0.29 (0.53); flux 3 0.69 (0.98); flux
4 0.67 (0.69); flux 5 0.11(0.36); flux 6 0.30 (0.02); flux 7 0.33
(0.30); flux 8 0.24 (0.45); flux 9 0.89 (0.66); and flux 10 0.65
(0.19).
represent the weighted averages of the value of each
parameter over the complete ipsilateral or contralateral
hemispheres.
Discussion
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Cerebral ischemia is the consequence of the reduction in
blood flow below a critical threshold and the concomitant limitation in
the supplies of primary cerebral substrates like glucose and oxygen.
Under these limiting circumstances the ability of neurons and glial
cells to obtain energy from substrates other than glucose contributes
importantly to cellular survival. The amino acids glutamate, GABA, and
glutamine are available as substrates in the extracellular fluid, and
may be used as alternative fuels by astrocytes or neurons under
conditions of glucose deprivation. Although it is known
that these amino acids are used as substrates by enriched cultures of
neurons or glial cells,31 evidence on the
physiological competence of this protective
mechanism during glucose deprivation has been missing in
the adult brain in vivo. ) and to compare the
results with those found in the present study for the in vivo brain
(Figure 6).
View this table:
[in a new window]
Table 2. Calculated Availability of Glutamate, Glutamine, and
GABA as Oxidative Substrates in Cultivated Neurons and Glial Cells
Under Normoxic or Ischemic Conditions
View larger version (114K):
[in a new window]
Figure 6. Summary of neuronal-glial interactions in both
hemispheres of rat brain infused with
(1,2-13C2) acetate, 1 and 24 hours
after FCI. Primary cerebral substrates and metabolic end
products in plasma are shown schematically in the blood vessels
located outside the hemispheres. Exchanges of glutamate, glutamine, and
GABA between the tricarboxylic acid cycles of neurons and glial cells
are shown by arrows inside the hemispheres. The relative size of the
arrows indicates approximately the relative contributions of the
fluxes. Black arrows indicate ipsilateral fluxes; white arrows,
contralateral fluxes. Numbers indicate the contribution (%) of the
corresponding flux, taken from Figure 5 .). The calculated rates of amino
acid transport under ischemic conditions represent an
important increase in the intracellular metabolic
availability of glutamate and GABA or of glutamine for oxidation. The
results obtained in the present study (Figure 6) confirm this
prediction showing increased oxidations of (13C)1
glutamate and (13C) GABA in the glial
tricarboxylic acid cycle and (13C) glutamine in
the neuronal tricarboxylic acid cycle of the adult brain in vivo,
respectively. The detection of increased glutamate and GABA oxidation
in the glial compartment agrees with previous observations in astrocyte
cultures30 and shows the important role of glial
metabolism in regulating the extracellular concentrations
of these amino acids in the ischemic brain tissue in
situ.
Selected Abbreviations and Acronyms
Ac
=
acetate
AcCoA
=
acetyl-CoA
Cr
=
creatine
PCr
=
phosphocreatine
FCI
=
partially reversible focal cerebral ischemia
GABA
=
-aminobutyryc acid
Gln
=
glutamine
Glc
=
glucose
Glu
=
glutamate
HPLC
=
high-performance liquid chromatography
Ino
=
myo-inositol
Lac
=
lactate
MABP
=
mean arterial blood pressure
NAA
=
N-acetyl-aspartic acid
NMR
=
nuclear magnetic resonance
OAA
=
oxalacetate
Succ
=
succinate
TSP
=
2,2'-3,3' tetradeutero trimethyl-silyl-propionate
Acknowledgments
This work was supported by grants PB-94011, PB96–0864-C02, and
AC-105/96 to Dr Cerdan and grant FISss 1177/96 to Dr Carceller. The
authors wish to thank Dr. B. Künnecke for providing a copy of the
METASIM v2.1 program, Dr O. Herreras for helpful comments, Mr. J.
Pérez for assistance with the illustrations, and Mrs. C.F. Warren
for editorial processing of the manuscript.
Footnotes
1 We use the term (13C) X to denote the total 13C accumulated in metabolite X during the infusion of a 13C-labeled substrate. 
References
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
1.
Sochocka E, Juurlink BHJ, Code WE, Hertz V, Peng
L, Hertz L. Cell death in primary cultures of mouse neurons and
astrocytes during exposure to and recovery from hypoxia,
substrate deprivation and simulated ischemia.
Brain Res. 1994;638:21–24.[Medline]
[Order article via Infotrieve]-aminobutyric acid and glutamate transporters in
chick embryonic neurons and rat synaptosomes. Neurochem Res. 1992;17:333–337.[Medline]
[Order article via Infotrieve]
Editorial Comment
Introduction
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
In the mature brain, glucose is the predominant energy
substrate with small contributions from plasma-derived lactate,
ß-hydroxybutyrate, and acetoacetate. In addition, components of the
amino acid pool can exchange with intermediates of the tricarboxylic
acid cycle (TCA) and thereby provide alternative energy substrates when
glycolysis is reduced. During prolonged focal cerebral
ischemia, glucose availability and acidosis may limit the
activity of glycolytic and other energy-associated enzymes. Whether
amino acids such as glutamate and glutamine, which are normally thought
to cycle between neuronal and astrocytic compartments, undergo
increased oxidation as alternative energy substrates during
ischemia in vivo is not well-established.
Selected Abbreviations and Acronyms
Ac
=
acetate
AcCoA
=
acetyl-CoA
Cr
=
creatine
PCr
=
phosphocreatine
FCI
=
partially reversible focal cerebral ischemia
GABA
=
-aminobutyryc acid
Gln
=
glutamine
Glc
=
glucose
Glu
=
glutamate
HPLC
=
high-performance liquid chromatography
Ino
=
myo-inositol
Lac
=
lactate
MABP
=
mean arterial blood pressure
NAA
=
N-acetyl-aspartic acid
NMR
=
nuclear magnetic resonance
OAA
=
oxalacetate
Succ
=
succinate
TSP
=
2,2'-3,3' tetradeutero trimethyl-silyl-propionate
This article has been cited by other articles:
 |
|
 |
|
  J. P. van der Zijden, P. van Eijsden, R. A. de Graaf, and R. M. Dijkhuizen 1H/13C MR spectroscopic imaging of regionally specific metabolic alterations after experimental stroke Brain, August 1, 2008; 131(8): 2209 - 2219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Zoghbi, H. U. Shetty, M. Ichise, M. Fujita, M. Imaizumi, J.-S. Liow, J. Shah, J. L. Musachio, V. W. Pike, and R. B. Innis PET Imaging of the Dopamine Transporter with 18F-FECNT: A Polar Radiometabolite Confounds Brain Radioligand Measurements J. Nucl. Med., March 1, 2006; 47(3): 520 - 527. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Behar and D. L. Rothman In Vivo Nuclear Magnetic Resonance Studies of Glutamate-{gamma}-Aminobutyric Acid-Glutamine Cycling in Rodent and Human Cortex: the Central Role of Glutamine J. Nutr., September 1, 2001; 131(9): 2498S - 2504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Hertz, S. R. Robinson;, J. L. Griffin;, P. J. Magistretti, L. Pellerin, R. G. Shulman, and D. L. Rothman; Energy for Neurotransmission Science, July 30, 1999; 285(5428): 639a - 639. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||