Structural features and regulatory properties of the brain glutamate decarboxylases

doi:10.1016/S0197-0186(00)00014-0

Neurochemistry International

Volume 37, Issues 2–3, 1 August 2000, Pages 111-119

https://doi.org/10.1016/S0197-0186(00)00014-0 Get rights and content

Abstract

It is widely recognized that the two major forms of GAD present in adult vertebrate brains are each composed of two major sequence domains that differ in size and degree of similarity. The amino-terminal domain is smaller and shows little sequence identity between the two forms. This domain is thought to mediate the subcellular targeting of the two GADs. Substantial parts of the amino-terminal domain appear to be exposed and flexible, as shown by proteolysis experiments and the locations of posttranslational modifications. The carboxyl-terminal sequence domain contains the catalytic site and shows substantial sequence similarity between the forms. The interaction of GAD with its cofactor, pyridoxal-5′ phosphate (pyridoxal-P), plays a key role in the regulation of GAD activity. Although GAD₆₅ and GAD₆₇ interact differently with pyridoxal-P, their cofactor-binding sites contain the same set of nine putative cofactor-binding residues and have the same basic structural fold. Thus the cofactor-binding differences cannot be attributed to fundamental structural differences between the GADs but must result from subtle modifications of the basic cofactor-binding fold. The presence of another conserved motif suggests that the carboxyl-terminal domain is composed of two functional domains: the cofactor-binding domain and a small domain that closes when the substrate binds. Finally, GAD is a dimeric enzyme and conserved features of GADs superfamily of pyridoxal-P proteins indicate the dimer-forming interactions are mediated mainly by the carboxyl-terminal domain.

Introduction

GABA and glutamate decarboxylase (GAD) were discovered in brain almost 50 years ago (Awapara et al., 1950, Roberts and Frankel, 1950a, Roberts and Frankel, 1950b, Udenfriend, 1950, Wingo and Awapara, 1950), at a time when the prolonged debate over the very existence of chemical synaptic transmission had just concluded (Bacq, 1975). At that time, of course, the function of brain GABA was yet to be established. GABA is now known best as the predominant inhibitory transmitter in brain, but very substantial fractions of GABA and GAD are found far from synaptic endings in cell bodies and dendrites, and the idea that GABA has important nonsynaptic functions cannot be dismissed (reviewed by Martin and Rimvall, 1993, Soghomonian and Martin, 1998). The discovery that vertebrates have two genes for GAD and express two major forms of the enzyme (Bu et al., 1992, Erlander et al., 1991) opened the possibility that each form is specialized to synthesize a specific pool of GABA that serves a specific function (Martin and Barke, 1998). In this paper we will summarize the functional differences and similarities between the two forms and examine aspects of the structures of the two forms that help us to understand their functions.

The two forms of GAD, which are called GAD₆₅ and GAD₆₇, are the products of two independently regulated genes (Bu et al., 1992, Erlander et al., 1991). The two forms are widely but not equally expressed among cell types in different brain regions (Esclapez et al., 1993, 1994; Feldblum et al., 1993, Mercugliano et al., 1992). GAD₆₅ and GAD₆₇ are almost equally abundant in several regions of mouse brain but GAD₆₅ is much more abundant than GAD₆₇ in most regions of rat brain (Sheikh et al., 1999).

Immunocytochemical and subcellular fractionation studies have shown that the two GADs are distributed differently within neurons (Erlander et al., 1991, Esclapez et al., 1994, Hendrickson et al., 1994, Henry and Tappaz, 1991, Reetz et al., 1991, Rimvall and Martin, 1994, Wood et al., 1976). GAD₆₅ is more concentrated within terminals and appears to be associated with synaptic vesicles while GAD₆₇ is more uniformly distributed throughout the neuron, being easily detected in cell bodies and dendrites. In brain, GAD₆₅ and GAD₆₇ are present both as soluble enzymes and associated with membranes. In addition, brain contains GAD₆₅–GAD₆₇ heteromers (Sheikh and Martin, 1996). The difference in subcellular targeting of the two enzymes, the membrane association of the enzymes and the formation of heteromers are important matters for which clues may be sought in the structures of the enzymes.

Both forms of GAD require the cofactor pyridoxal-P for activity, and the interaction of GAD with pyridoxal-P plays a central role in the regulation of GAD activity. At least half of brain GAD is present as apoenzyme (GAD without bound cofactor) which provides a reserve of inactive enzyme that can be activated when additional GABA synthesis is required (Itoh and Uchimura, 1981, Miller et al., 1977). Active-site labeling studies indicate that the high proportion of apoGAD in brain is unusual, if not unique, among pyridoxal-P dependent enzymes (Martin et al., 1991). The reserve function of apoGAD appears to operate in tissues, as depolarizing synaptosomes leads to the conversion of apoGAD to holoGAD and stimulates GABA synthesis (Gold et al., 1978, Gold and Roth, 1979, Miller et al., 1980). Finally, the interconversion of apo- and holoGAD is a complex, regulated process that involves a large conformational change in the enzyme (Chen et al., 1998, Martin and Rimvall, 1993).

GAD₆₅ and GAD₆₇ appear to differ in their interactions with pyridoxal-P, although these differences have not yet been studied in detail. Experiments with the individual forms of recombinant enzyme and with GAD₆₅ and GAD₆₇, prepared by immunoprecipitation from cerebellum, indicate that GAD₆₅ is more responsive to added pyridoxal-P than is GAD₆₇ (Erlander et al., 1991, Kaufman et al., 1991). In addition, GAD₆₅ accounts for most of the apoGAD in synaptosomes and several regions of rat brain (Martin et al., 1991). At first glance the latter result seems consistent with the idea that GAD₆₅ is more highly regulated by the cofactor than is GAD₆₇. However, GAD₆₅ is much more abundant than GAD₆₇ in synaptosomes and most rat-brain regions, and the greater abundance of apoGAD₆₅ appears to result in large part from the greater abundance of GAD₆₅ (Rimvall and Martin, 1994, Sheikh et al., 1999).

The unusual and complex interaction of GAD with pyridoxal-P and the related differences between the two GADs suggest that these enzymes have interesting and unusual structural features.

Section snippets

Materials

Chemicals were obtained from Sigma (St Louis, MO, USA). Sequencing grade trypsin and endoproteinase Asp-N were obtained from Sigma and Boehringer Mannheim (Indianapolis, IN). Q-Sepharose was from Pharmacia Biotech (Piscataway, NJ, USA). Immobilon-P was from Millipore (Bedford, MA, USA). AntiGAD sera W887, W883, and W624 were prepared in this laboratory (Sheikh and Martin, 1996). W887 was prepared against a peptide corresponding to residues 570–585 of GAD₆₅ and 578–593 of GAD₆₇. This shared

Sequence domains of the GADs

GAD₆₅ and GAD₆₇ are each composed of two major sequence domains as revealed by pairwise sequence alignment (Erlander et al., 1991). One of these domains is composed of residues 1–95 in human GAD₆₅ and 1–101 in GAD₆₇ and is strikingly different in the two GADs, as it has only 23% sequence identity. This domain, called the N-terminal domain appears, to be responsible for subcellular targeting and the formation of GAD₆₅–GAD₆₇ heteromers (Dirkx et al., 1995, Sheikh and Martin, 1996, Shi et al., 1994

Acknowledgements

Fluorimaging, mass spectrometry and protein sequencing were conducted in the Wadsworth Center’s Molecular Immunology, Biological Mass Spectroscopy, and Protein Chemistry Core Facilities, respectively. We thank Charles R. Hauer for his invaluable advice and assistance with mass spectrometry. This work was supported by grant MH35664 from the National Institute of Mental Health, DHHS.

References (41)

J. Awapara et al.
Free γ-aminobutyric acid in brain
J. Biol. Chem.
(1950)
C.-H. Chen et al.
Structural characteristics of brain glutamate decarboxylase in relation to its interaction and activation
Arch. Biochem. Biophys.
(1998)
R. Dirkx et al.
Targeting of the 67-kDa isoform of glutamic acid decarboxylase to intracellular organelles is mediated by its interaction with the NH₂-terminal region of the 65-kDa isoform of glutamic acid decarboxylase
J. Biol. Chem.
(1995)
M.G. Erlander et al.
Two genes encode distinct glutamate decarboxylases
Neuron
(1991)
S. Henry et al.
Relative enrichment of the lighter 59 kDa form of glutamic acid decarboxylase in nerve endings: an immunoblotting study in the pituitary neurointermediate lobe
Neurosci. Letts
(1991)
C.A. McPhalen et al.
Domain closure in mitochondrial aspartate aminotransferase
Journal of Molecular Biology
(1992)
L.P. Miller et al.
Glutamate holodecarboxylase levels and the regulation of GABA synthesis
Brain Res. Bull.
(1980)
C. Momany et al.
Crystallographic structure of a PLP-dependent ornithine decarboxylase from Lactobacillus 30a to 3.0 A resolution
J. Mol. Biol.
(1995)
M. Namchuk et al.
Phosphorylation of serine residues 3, 6, 10, and 13 distinguishes membrane anchored from soluble glutamic acid decarboxylase 65 and is restricted to glutamic acid decarboxylase 65alpha
J. Biol. Chem.
(1997)
W.H. Oertel et al.
Production of a specific antiserum to rat brain glutamic acid decarboxylase by injection of an antigen-antibody complex
Neurosci.
(1981)

E. Roberts et al.

γ-Aminobutyric acid in brain: its formation from glutamic acid

J. Biol. Chem.

(1950)

J.-J. Soghomonian et al.

Two isoforms of glutamate decarboxylase: why?

Trends Pharmacol. Sci.

(1998)

S. Udenfriend

Identification of γ-aminobutyric acid in brain by the isotope derivative method

J. Biol. Chem.

(1950)

W.J. Wingo et al.

Decarboxylation of l-glutamic acid by brain

J. Biol. Chem.

(1950)

Z.M. Bacq

D.-F. Bu et al.

Two human glutamate decarboxylases, 65-kDa GAD and 67-kDa GAD, are each encoded by a single gene

Proc. Natl. Acad. Sci. USA

(1992)

S. Christgau et al.

Membrane anchoring of the autoantigen GAD₆₅ to microvesicles in pancreatic β-cells by palmitoylation in the NH₂-terminal domain

J. Cell Biol.

(1992)

M. Esclapez et al.

Comparative localization of two forms of glutamic acid decarboxylase and their mRNAs in rat brain supports the concept of functional differences between the forms

J. Neurosci.

(1994)

M. Esclapez et al.

Comparative localization of mRNAs encoding two forms of glutamic acid decarboxylase with nonradioactive in situ hybridization methods

J. Comp. Neurol.

(1993)

S. Feldblum et al.

Different distributions of GAD65 and GAD67 mRNAs suggest that the two glutamate decarboxylases play different functional roles

J. Neurosci. Res.

(1993)

Cited by (31)

Structural determinants of GAD antigenicity
2009, Molecular Immunology
Our aim was to ascertain structural determinants of autoantigenicity based on the model of the diabetes autoantigen glutamic acid decarboxylase 65 kDa isoform (GAD65) in comparison with that of the non-autoantigenic isoform GAD67. This difference exists despite the two isoforms having the same fold and high sequence identity. Autoantibodies to GAD65 precede the development of type 1 diabetes and are clinical markers of this and certain neural autoimmune diseases. To date, epitope mapping has been based on particular amino acid differences between the two isoforms, and there is no explanation as to why autoantibodies that react with GAD65 only infrequently cross-react with GAD67. To characterize each isoform of the enzyme and gain insights into their contrasting autoantigenic properties, we have used the recently determined crystal structures of GAD65 and GAD67 to compare their structure, hydrophobicity, electrostatics, flexibility and physiochemical properties. The results revealed striking differences which appear almost exclusively at the C-terminal domain of the isoforms. Whereas GAD65 displayed a highly charged and flexible C-terminal domain containing numerous patches of high electrostatic and solvation energies, these characteristics were absent in the GAD67 molecule. Additionally, analysis indicated potential N-terminal and PLP domain binding sites surrounding the C-terminal domain of GAD65, a major region of autoantigenic activity, but not of GAD67. These features agree with good accuracy with published epitope-mapping data. Our analysis suggests that the high flexibility and charge of GAD65 in the C-terminal domain is coupled with the mobility of its catalytic loop, a property that is absolutely required for its enzymatic function. Thus, the structural features that distinguish GAD65 from GAD67 as a B cell autoantigen are related to functional requirements for its enzymatic mechanism. This could well apply to the various other enzyme autoantigens and, if so, these features could be used as the basis of future predictive strategies.
Time course of changes in pyridoxal 5′-phosphate (vitamin B<inf>6</inf> active form) and its neuroprotection in experimental ischemic damage
2007, Experimental Neurology
In the present study, we investigated ischemia-induced changes of pyridoxal 5′-phosphate synthesizing enzyme and degrading enzyme and neuroprotective effects and roles of pyridoxal 5′-phosphate against ischemic damage in the gerbil hippocampal CA1 region. Pyridoxal 5′-phosphate oxidase and pyridoxal phosphate phosphatase immunoreactivities were changed in neurons up to 2 days after ischemia, while 4 days after ischemia their immunoreactivities were expressed in astrocytes. Pyridoxal 5′-phosphate oxidase immunoreactivity and its protein level were highest 12 h after ischemia, while those in pyridoxal phosphate phosphatase were highest 2 days after ischemia. Total activities of these enzymes were changed after ischemia, but specific activities of the enzymes were not altered. Treatment with pyridoxal 5′-phosphate into brains (4 μg/5 μl, i.c.v.) at 30 min before transient ischemia protected about 80% of CA1 pyramidal cells 4 days after ischemia and induced elevation of glutamic acid decarboxylase 67 immunoreactivity in the CA1 region. However, pyridoxal 5′-phosphate treatment into ischemic brains decreased GABA transaminase immunoreactivity in the CA1 region after ischemia. These results indicate that pyridoxal 5′-phosphate may be associated with the inhibitory discharge of GABA in the hippocampal CA1 neurons, and the increased level of GABA may protect hippocampal CA1 pyramidal cells from ischemic damage.
Expression of recombinant goldfish glutamic acid decarboxylase 65 and evidence for differential pH and PLP responsiveness compared to the human enzyme
2006, Comparative Biochemistry and Physiology - B Biochemistry and Molecular Biology
Glutamic acid decarboxylase (GAD) catalyzes the conversion of glutamate to γ-aminobutyric acid (GABA) that acts as an important inhibitory neurotransmitter in the vertebrate brain, as well as in the regulation of neuroendocrine function. GAD65 and GAD67 are the two main isoforms that exist in vertebrates. The biochemical properties of recombinant forms of goldfish and human GAD65 were examined. The recombinant goldfish GAD65 (gfGAD65) was expressed at high levels using a maltose binding protein fusion system for biochemical characterization. The human GAD65 (hGAD65) was expressed as a GST fusion and was also purified. The recombinant goldfish GAD65 protein has properties that are different from the human counterpart. In particular, the gfGAD65 is less active at acidic pH compared to hGAD65, which is moderately active over a wider range of acidic and basic pH. Interestingly, however, gfGAD65 is less dependent on a cofactor pyridoxal-5′-l-phosphate (PLP) for activity. In the absence of added PLP, cleaved recombinant gfGAD65 showed approximately 20% of maximal activity whereas hGAD65 showed no detectable activity. The physiological and evolutionary significance of these findings is discussed in light of the conserved function of GAD in two vertebrate species that are separated in evolutionary time by more than 200 million years.
The blood supply of the cat's visual cortex and its postnatal development
2004, Brain Research
Citation Excerpt :
Barone et al. [3] have noted that alkaline phosphatase is highly expressed in regions of neuropil where glutamic acid decarboxylase (GAD, the rate-limiting enzyme in the synthesis of GABA) is also highly expressed, for example, layer 4 of visual cortex [4,26]. They argue that alkaline phosphatase plays a critical role at GABAergic terminals, dephosphorylating pyridoxal phosphate, thereby allowing pyridoxal to be transported into the terminal, where it is rephosphorylated and acts as a co-factor for GAD65 [61]. This suggestion is consistent with earlier observations that alkaline phosphatase knockout mice exhibit seizures [65] that resemble those in mice fed diets deficient in vitamin B-6 and can be mitigated by administering pyridoxal HCL [66].
We examined the blood supply of the cat's visual cortex using alkaline phosphatase histochemistry to demonstrate the capillary endothelial cells. In the adult, layer 4 is marked by a band that is of obviously greater density, extends throughout areas 17 and 18, and ends abruptly at the 18/19 border. We quantified blood vessel density in area 17, observing a 23% greater density in layer 4 than in supragranular and infragranular layers. This difference reflects a laminar difference in metabolic rate. In three animals studied using the metabolic marker 2-deoxyglucose, layer 4 was 25% denser than the other layers. The band of greater density in layer 4 is not present in newborn kittens, but becomes apparent at about 5 weeks of age. Early in development, the endothelial cells form filopodia as the capillaries grow and branch. The density of blood vessels decreases slightly during the first week of postnatal life, but increases between 1 and 6 weeks of age, so that by 6 weeks, the blood supply of the visual cortex resembles that seen in the adult. This pattern resembles that of cortical metabolism seen with 2-deoxyglucose [J. Cereb. Blood Flow Metab. 11 (1991) 35], but the increase in vascular density precedes that in glucose metabolism.
Changes of pyridoxal kinase expression and activity in the gerbil hippocampus following transient forebrain ischemia
2004, Neuroscience
In the previous study, we observed chronological alterations of glutamic acid decarboxylase (GAD), which is the enzyme converting glutamate into GABA. GAD isoforms (GAD65 and GAD67) differ substantially in their interactions with cofactor pyridoxal 5′-phosphate, which is catalyzed by pyridoxal kinase (PLK). In the present study, we examined the chronological changes of PLK expression and activity in the hippocampus after 5 min transient forebrain ischemia in gerbils. PLK immunoreactivity in the sham-operated group was detected weakly in the hippocampus. Ischemia-related change of PLK immunoreactivity in the hippocampus was significant in the hippocampal cornu ammonis (CA1)region, not in the hippocampal CA2/3 region and dentate gyrus. PLK immunoreactivity was observed in non-pyramidal GABAergic neurons at 30 min to 3 h after ischemic insult. At 12 h after ischemic insult, PLK immunoreactivity was shown in many CA1 pyramidal cells as well as some non-pyramidal cells. At this time point, PLK immunoreactivity and protein content was highest after ischemia. Thereafter, PLK immunoreactivity and protein content is decreased time-dependently by 4 days after ischemic insult. Four days after ischemia, some astrocytes expressed PLK in the CA1 region. The specific PLK activity was not altered following ischemic insult up to 2 days after ischemic insult. Thereafter, the specific PLK activity decreased time-dependently. However, total activity of PLK was significantly increased 12–24 h after ischemic insult, and thereafter total activity of PLK decreased. Therefore, we suggest that the over-expression of PLK in the CA1 pyramidal cells at 12 h after ischemia may induce increase of GAD in the CA1 pyramidal cells, which plays an important role in delayed neuronal death via the increase of GABA or enhancement of GABA shunt pathway.
Post-mortem degradation of brain glutamate decarboxylase
2003, Neurochemistry International
The post-mortem stability of the GABA synthesizing enzyme glutamate decarboxylase (GAD) was studied by using SDS–PAGE and quantitative immunoblotting to measure the rates of degradation of GAD in the cerebral cortex, hippocampus, and cerebellum of rats and mice as a function of time after death. The intact 65- and 67-kDa isoforms of GAD (GAD₆₅ and GAD₆₇) disappeared gradually over a 24-h period. In both rats and mice, the degraded GAD appeared as a band with an apparent molecular mass of 55–57 kDa; no significant amounts of smaller forms were observed. The 55–57 kDa band reacted with antiserum W887, which recognizes a shared epitope at the carboxyl-terminal end of both GADs, indicating that GAD was cleaved near the amino-terminal end of the molecule. GAD₆₇ was cleaved at a site between the amino-terminus and the epitope for antiserum W883 (located within residues 79–93 of GAD₆₇), as antiserum W883 stained a 56-kDa band on the blots. The appearance of degraded GAD paralleled the loss of total GAD (GAD₆₅+GAD₆₇), and after 24 h the 55–57 kDa band accounted for 97, 88, and 59% of the intact GAD lost from rat cerebellum, cerebral cortex and hippocampus. On a percentage basis, GAD₆₇ was degraded more rapidly than was GAD₆₅ in all brain regions studied. The loss of GAD activity was greater in rat than mouse brain, even though the percent loss of intact GAD protein was similar.

View all citing articles on Scopus

View full text

Published by Elsevier Ltd.

Structural features and regulatory properties of the brain glutamate decarboxylases

Abstract

Introduction

Section snippets

Materials

Sequence domains of the GADs

Acknowledgements

J. Biol. Chem.

Arch. Biochem. Biophys.

J. Biol. Chem.

Neuron

Neurosci. Letts

Journal of Molecular Biology

Brain Res. Bull.

J. Mol. Biol.

J. Biol. Chem.

Neurosci.

J. Biol. Chem.

Trends Pharmacol. Sci.

J. Biol. Chem.

J. Biol. Chem.

Two human glutamate decarboxylases, 65-kDa GAD and 67-kDa GAD, are each encoded by a single gene

Proc. Natl. Acad. Sci. USA

Membrane anchoring of the autoantigen GAD65 to microvesicles in pancreatic β-cells by palmitoylation in the NH2-terminal domain

J. Cell Biol.

Comparative localization of two forms of glutamic acid decarboxylase and their mRNAs in rat brain supports the concept of functional differences between the forms

J. Neurosci.

Comparative localization of mRNAs encoding two forms of glutamic acid decarboxylase with nonradioactive in situ hybridization methods

J. Comp. Neurol.

Different distributions of GAD65 and GAD67 mRNAs suggest that the two glutamate decarboxylases play different functional roles

J. Neurosci. Res.

Membrane anchoring of the autoantigen GAD₆₅ to microvesicles in pancreatic β-cells by palmitoylation in the NH₂-terminal domain