[Frontiers in Bioscience, d701-718, July 20, 1998]
Reprints
PubMed
CAVEAT LECTOR




Table of Conents
 Previous Section   Next Section

EXCITATORY AMINO ACID NEUROTRANSMISSION. PATHWAYS FOR METABOLISM, STORAGE AND REUPTAKE OF GLUTAMATE IN BRAIN

Monica Palmada and Josep Joan Centelles

Departament de Bioquímica i Biologia Molecular, Facultat de Química, Universitat de Barcelona, Martí i Franquès, 1, 08028-Barcelona, Spain

Received 5/7/98 Accepted 5/23/98

5. METABOLIC INTERACTIONS BETWEEN NEURONS AND ASTROCYTES

If one tries to envisage the metabolic capabilities required to maintain glutamatergic, aspartatergic, and GABA-ergic transmission (Glu, Asp, and GABA) in a simple cell type residing behind a blood-brain barrier through which there is rapid exchange of CO2, H2O, and NH3 and a rapid net influx of glucose, but no net accumulation of either glutamate, aspartate, GABA, or glutamine (73), one would end up with a cell like that sketched in figure 4. Neurons are able to synthesize both glutamate and GABA from glutamine, and astrocytes form and release glutamine (which has no transmitter activity) after accumulation of either glutamate or GABA.

Figure 4. Schematic diagram of glutamate/glutamine and GABA/glutamine cycles. In neurons, glutamate and g -aminobutyric acid (GABA) are synthesized from glucose or glutamine. Neurotransmitter released in synaptic cleft can then interact with receptor sites. To terminate this effect, transport to astroglial cells is performed with higher affinity than to neuron cells. In astroglial cells there is a higher glutamine synthase activity and glutamate and GABA are metabolized to glutamine, which has no neurotransmitter effects and can be recycled to neurons to form glutamate or GABA.

Extracellular glutamate is, to a larger extent, accumulated into astrocytes both in the intact brain (74) and in cultured cells (75,76), but much of the accumulated glutamate (how much probably depends upon the experimental conditions) is degraded as a metabolic fuel to CO2 and H2O in astrocytes and thus not converted to glutamine (75,77). New glutamate and GABA precursor molecules will, therefore, have to be synthesized from glucose. Since carboxylation of pyruvate to oxaloacetate, an intermediate of the TCA cycle, occurs in astrocytes, net synthesis of a -ketoglutarate in the TCA cycle can also take place in these cells.

The rate of glutamate metabolism to CO2 is high in astrocytes but not in the two neuronal types, especially not in the glutamatergic cerebellar granule cells. The rapid CO2 formation in astrocytes from glutamate appears to represent mainly a net utilization of glutamate (78), not just an isotope exchange between glutamate and a -ketoglutarate. Utilization of glutamate or glutamine as a metabolic substrate is not restricted to cultured cells but has also been observed in brain slices and dissociated cell preparations (79,80). As long as the glutamate utilized as a metabolic fuel originally is produced from glucose behind the blood-brain barrier, this is not a violation of the well-established fact that the entire adult brain in vivo under normal conditions almost exclusively utilizes glucose as its substrate for energy metabolism (81).

Exposure to very high potassium concentrations (> 20 mM) causes depolarization and a depolarization-induced increase in free intracellular calcium concentration in synaptosomes (82), neurons, and astrocytes (83) and enhances metabolic interactions between these two cell types by facilitating glutamate (and GABA) release from neurons and CO2 fixation in astrocytes, which in turn promotes astrocytic formation of transmitter precursors for neurons. During exposure to slightly elevated potassium. In the following section, we describe some properties of the enzymes involved in the glutamic acid metabolism (figure 5).

Figure 5. Glutamate metabolism. Glutamate from mammalian brain can be obtained from glutamine (glutaminase) or from a -ketoglutarate (glutamate dehydrogenase and glutamate oxalacetate transaminase). Degradation of glutamate can generate a -ketoglutarate by reversible reactions, GABA (glutamate decarboxylase) or glutamine (glutamine synthetase).

5.1. Glutamine synthetase (E.C. 6.3.1.2.)

Glutamine synthetase catalyzes several reactions (84), although the main reaction is the following one:

Glu + NH3 + ATP <-----> Gln + ADP + Pi

The enzyme has two important functions: assimilation of ammonia and biosynthesis of glutamine. The enzyme from brain has been studied in rat, ox, sheep, pig

and human. It has been isolated from a variety of sources and the proteins vary greatly in their ability to catalyze the reverse reaction. With the mammalian enzyme, however, the forward rate relative to the reverse rate is about 10 to 1 (85).

The purification of the enzyme usually proceeds through four steps consisting on an acetone powder extract, precipitation by acid, hydroxylapatite and DEAE-cellulose column chromatography (86); with a yield of 15-30% and a purification of about 200-fold.

Glutamine synthetase is composed of eight subunits which are identical (44,000-50,000 daltons) showing rather similar amino acid composition (84). The enzyme has a molecular weight of 400,000 daltons, distributed in two tetramers (87). Distribution of glutamine synthetase and glutaminase are uneven in central nervous system. Berl (88) determined the distribution of glutamine synthetase in 16 brain areas of the adult rat. The site of glutamine synthesis, from glutamate amidation via glutamine synthase (GS), is in glial cells (89); and with a cytosolic location (9). Additionally, glia cells contain carbonic anhydrase, which catalyzes the CO2 hydration reaction and HCO3- formation. Therefore, glia cells could be important in acid-base regulation and related amino acid metabolism in CNS. Glutamine formed by GS can move into adjacent nerve endings where it is utilized either in metabolic pathways unrelated or not directly related to neurotransmitter glutamate and GABA synthesis or in formation of neurotransmitter glutamate and GABA via glutaminase and glutamate decarboxylase, respectively.

The enzyme is irreversibly inhibited by methionine sulfoxamine (MSO) (90). ATP and magnesium are necessary for the binding of glutamate to the enzyme, whereby it becomes activated. Tate et al. (84) calculated in rat liver Km for ATP = 2.3 mM; and Deul et al. (86) Km for ammonia = 0.3 mM. Certain anions, particularly bicarbonate and chloride, activate the enzyme when nonsaturating concentrations of L-glutamate are used. Although liver glutamine synthase is activated by 2-oxoglutarate, brain enzyme is less affected by this compound (84). The mammalian enzymes are inhibited by inorganic phosphate and carbamyl phosphate (91). This effect can be due to the reaction of ATP synthesis from ADP and carbamyl phosphate, catabolized also by glutamine synthetase.

5.2. Glutaminase (E.C. 3.5.1.2.)

The reverse reaction of glutamine synthase is catalyzed by the ubiquitous enzyme glutaminase, which is present in both neurons and astrocytes (92). Nevertheless, glutaminase is predominantly a neuronal enzyme and it has been localized in mitochondria (93,94). Pig brain mitochondria have been shown to contain two major forms of glutaminase, one soluble located in the matrix and one membrane-bound enzyme located in the inner membrane and both activated by phosphate (95).

Soluble glutaminase has been purified from pig brain (96,97) and some other sources (98-100). The soluble enzyme is present as a dimer (101) with a relatively low specific activity and a highly aggregated form of the enzyme with a considerably greater specific activity. By an incubation with phosphate, the soluble enzyme is reversibly and slowly aggregated (101-103). Because of this complicated regulatory behavior, some evidences have been suggested that the membrane-bound and soluble enzyme may have different function in the brain (95).

Kinetic studies of the two enzymes have been performed by Nimmo and Tipton (101) and they show Km values for glutamine of about 0.8-1.4 mM and 3.4-9.7 mM. One of the products, glutamate, inhibits the enzyme strongly, whereas the other product ammonia has only a slight inhibitory action on the enzyme. Glutamate inhibition is mixed (Kislope= 1.6 mM and Kiintercept= 3.3 mM).

In astrocytes exposed to 1.2 mM valproate, glutaminase activity increased 80% in primary culture by day 2 and remained elevated by day 4; glutamine synthetase activity was decreased 30% (104).

5.3. Glutamate dehydrogenase (E.C. 1.4.1.2.-4.)

Glutamate dehydrogenases catalyze the following reversible reaction:

L-glutamate + H20 + NAD(P) <-----> a -ketoglutarate + NH3 + NAD(P)H

although it appears that the reaction velocity is higher when the formation of glutamate is studied.

Three different enzymes are considered depending on the use of NAD/NADH (E.C. 1.4.1.2.), NADP/NADPH (E.C. 1.4.1.4.) or both (E.C. 1.4.1.3.) as coenzymes. Neural glutamate dehydrogenase (E.C. 1.4.1.3.) can use both coenzymes, although it has been shown that NAD is used more effectively than NADP (105). The direction of neural glutamic dehydrogenase (E.C. 1.4.1.3.) activity appears to be regulated in part by the tissue NAD(P)/NAD(P)H concentration ratio (105, 106). In rat brain GDH activity exists in two distinct forms differing in solubility, kinetic parameters, resistence to heat inactivation and allosteric properties (107). These forms have been designed soluble and particulate GDH (107,108).

When the ratio is high, e.g., in the absence of glucose, oxidative deamination of glutamate occurs. In the presence of glucose, when the ratio falls, and a -ketoglutarate is not rate limiting, reductive amination of a -ketoglutarate is favored. Kinetic parameters have been studied for both directions of the reaction (105).

GDH’s from various organs of the same species are similar, if not identical. Complete cross-reaction occurs between antibodies induced by bovine liver GDH and extracts of bovine spleen, brain and heart. Most animal GDH are inhibited by GTP and activated by ADP (109). GDH from bovine brain has been partially purified by Grisolia et al. (106) and shows a specificity either for NAD or NADP. This enzyme has a molecular weight of 332,000 daltons, as judged from the amino acid sequence of the six identical subunits (110).

As compared with other enzymes, the Vmax of GDH in synaptic mitochondria from rat brain is 20-40% lower than Vmax of aspartate aminotransferase, but 4-5fold higher than Vmax of phosphate dependent glutaminase (111).

5.4. Transaminases

Transaminases play an important role in the aminoacid metabolism, as they are able to catabolize a reversible transference of an amino group from an aminoacid to a ketoacid acceptor. Those enzymes use pyridoxal phosphate as a coenzyme, which will perform the transference of the amino group. Among those transaminases, aspartate aminotransferase or glutamate oxalacetate transaminase (E.C. 2.6.1.1.) catabolizes the reversible transference of amino group of Asp to a -ketoglutarate and generates oxalacetate and glutamate. There have been found two isoenzymes of aspartate aminotransferase in animals: a citoplasmic and a mitochondrial. Both are dimeric proteins of 45,000 daltons and 2,000 daltons subunits (112).

Aspartate amino transferase (AAT) activity seems to be higher than glutamate dehydrogenase (GDH). The ratio AAT/GDH is between 10 and 20 in rat brain (113) or in squid giant nerve (114). Transamination of glutamate by using aspartate amino transferase generates aspartate, which is also a neurotransmitter; whereas glutamate dehydrogenase yields ammonia but not another neurotransmitter. The higher activity of AAT can be therefore due to prevent the toxic action of ammonia.

Another transaminase is alanine aminotransferase (E.C. 2.6.1.2.), an enzyme which transfers amino from alanine to a -ketoglutarate to yield pyruvate and glutamate. This enzyme presents also a cytoplasmic and a mitochondrial isoenzymes.

5.5. Glutamic acid decarboxylase (E.C. 4.1.1.15.)

The enzyme removes the a -carboxyl group of glutamate to produce a g -carboxyl amino acid called g -amino butyric acid (GABA). This decarboxylation of glutamate to GABA is not very different from decarboxylation of L-DOPA or tryptophan to dopamine and serotonin. Like those enzymes, glutamic acid decarboxylase requires the cofactor pyridoxal phosphate (vitamin B6).

The enzyme is highly substrate specific, although Homola and Dekker (115) showed that some glutamate analogs can also be decarboxylated.