GAD65 (glutamic acid decarboxylase 65 kDa), the product of the GAD2 gene, is one of two isoforms of the pyridoxal phosphate (PLP)-dependent enzyme that catalyzes the decarboxylation of L-glutamate to γ-aminobutyric acid (GABA), the principal inhibitory neurotransmitter in the vertebrate nervous system. In contrast to its constitutively active paralog GAD67, GAD65 exists largely as a dormant apoenzyme that is recruited to membrane compartments and activated in response to demand for synaptic GABA. This review focuses on the molecular architecture of GAD65, the structural basis of its catalytic mechanism and activity-dependent autoinactivation, its regulation by phosphorylation, palmitoylation, and calcium-dependent proteolysis, and the downstream GABA receptor signaling pathways—including GABAA ligand-gated chloride channels and GABAB metabotropic receptors—through which GAD65-derived GABA shapes inhibitory neurotransmission and non-neuronal signaling such as pancreatic islet function. To support neuroscience and signaling-pathway research, we provide multiplexed analysis services using protein array technology, meeting diverse needs in phosphoprotein profiling, neurotransmitter pathway activity mapping, and signaling network discovery.
γ-Aminobutyric acid (GABA) is the most abundant inhibitory neurotransmitter in the adult vertebrate brain and also acts as a signaling and trophic molecule in several non-neural tissues, including the pancreas. The committed, rate-limiting step in GABA biosynthesis is the α-decarboxylation of L-glutamate, a reaction catalyzed by glutamic acid decarboxylase (GAD; EC 4.1.1.15). In mammals, GAD exists as two closely related isoforms, GAD65 and GAD67, which are encoded by separate, independently regulated genes and differ markedly in their regulation, subcellular localization, and mode of activation. GAD65, the product of GAD2 on human chromosome 10p11.23, is the inducible, activity-dependent isoform: it is synthesized as a largely cytosolic, PLP-free apoenzyme that is recruited to membrane compartments and activated when the demand for synaptic GABA rises. GAD67, by contrast, is constitutively active and sustains the basal cytoplasmic GABA pool. The molecular properties that make GAD65 a tunable, signal-responsive enzyme—rather than a housekeeping catalyst—are the central subject of this review.
At the signaling level, GAD65 sits at the top of the GABAergic pathway. The GABA it produces is packaged into synaptic vesicles (in neurons) or released from cytosolic and granule pools (in pancreatic beta cells), after which it engages ionotropic GABAA receptors and metabotropic GABAB receptors on neighboring cells. Through these receptors, GAD65-derived GABA controls chloride permeability, potassium conductance, calcium influx, and adenylyl cyclase activity, thereby shaping inhibitory tone in the brain and paracrine communication in the islet. The remainder of this review examines, in turn, the isoform pair, the catalytic machinery of GAD65, its post-translational control, its vesicular targeting, and the receptor systems that transduce its product into cellular signals.
Figure 1. Open and closed dimers of GAD65 and the mechanism of dimer opening in holo → apo conversion. (Source: Kass I, et al. 2014)
Two Isoforms of Glutamate Decarboxylase
The existence of two GAD isoforms was established by the cloning of distinct GAD genes from brain cDNA. Erlander and colleagues showed that the brain contains two forms of GAD, GAD65 and GAD67, which derive from two separate genes and differ in molecular size, sequence, intracellular location, and interaction with the PLP cofactor. Bu and colleagues subsequently isolated and sequenced the human cDNAs, demonstrating that human GAD65 encodes a 585-residue, ~65 kDa polypeptide and human GAD67 a 594-residue, ~67 kDa polypeptide, each the product of a single gene; fluorescence in situ hybridization mapped GAD2 to chromosome 10p11.23 and GAD1 to chromosome 2q31. The two isoforms share roughly 65% overall sequence identity but diverge sharply in their N-terminal ~95 residues, a region that carries the isoform-specific targeting and modification signals.
The functional significance of maintaining two isoforms was clarified by Soghomonian and Martin, who reviewed the division of labor between them. GAD65 is preferentially targeted to membranes and nerve endings, whereas GAD67 is more widely distributed through the cytosol. Both forms can synthesize transmitter GABA, but GAD67 appears to maintain the cytoplasmic basal pool, while GAD65 is poised to synthesize GABA for vesicular release and to respond to acute increases in neurotransmitter demand. This segregation—basal supply versus activity-dependent supply—is reflected at every level of GAD65 biology, from its dormant apoenzyme state to its calcium-triggered activation, and it is what distinguishes GAD65 from its paralog as a signaling-responsive enzyme.
Structure and Catalytic Mechanism
GAD65 is a PLP-dependent decarboxylase. The cofactor pyridoxal 5'-phosphate forms a Schiff base with a conserved active-site lysine, and the resulting internal aldimine acts as an electron sink that stabilizes the carbanion intermediate of glutamate decarboxylation, yielding GABA and carbon dioxide. The three-dimensional basis for this chemistry was revealed by Fenalti and colleagues, who determined the crystal structures of N-terminal truncations of both GAD isoforms. The GAD fold comprises a large and a small α/β domain that sandwich the PLP cofactor at the dimer interface; the active site is formed between subunits of the homodimer.
The structures exposed a striking difference between the isoforms in the conformation of the catalytic loop that borders the active site. In GAD67, this loop is tethered over the active site, providing a protected catalytic chamber that sustains continuous GABA production. In GAD65, the same loop is inherently mobile. Kinetic analyses indicate that this mobility permits a side reaction in which the PLP cofactor is released, converting the active holoenzyme into the inactive apoenzyme—a process termed autoinactivation. GAD65 therefore cycles between an active holo form (with PLP bound) and an inactive apo form (without PLP), and the equilibrium between these states is a key determinant of how much GABA the enzyme can supply on demand. This structural insight explains why GAD65 behaves as a reserve rather than a constitutive catalyst and why its activity is exquisitely sensitive to cofactor availability, local conformation, and the post-translational modifications discussed below.
Regulation by Post-Translational Modification
Because GAD65 functions as a signal-responsive reserve, its activity is tightly controlled by several post-translational modifications. Namchuk and colleagues identified phosphorylation of GAD65α at serine residues 3, 6, 10, and 13 in the N-terminus, a modification that distinguishes the membrane-anchored from the soluble form of the enzyme and is mediated by a membrane-bound kinase. Although this phosphorylation does not substantially alter the Km or Vmax for glutamate, it marks GAD65 after membrane anchoring and is consistent with a model in which only one subunit of the native GAD65 dimer is phosphorylated, perhaps to tune enzyme function at the synaptic vesicle membrane. More recently, phosphorylation at a threonine residue in the catalytic domain has been reported to activate GAD65, linking kinase signaling at the synapse to GABA synthetic capacity.
Calcium entry during neurotransmission provides a second, direct route to GAD65 activation. Wei and colleagues showed that sustained neuronal stimulation triggers calcium-dependent cleavage of GAD65 by the cysteine protease calpain, releasing a truncated form (GAD65 lacking residues 1–69) that is more active than the full-length enzyme. Calcium-chelating reagents block the cleavage, and calpain inhibitors prevent the conversion, establishing calpain as the effector that couples elevated intracellular Ca2+ to a more active GAD65 pool. Together with the PLP-dependent holo/apo equilibrium, phosphorylation and calpain cleavage form a layer of activity-dependent regulation that allows GABA synthesis to track the firing state of the neuron. Oxidation of the PLP cofactor and changes in local pH further modulate GAD65, reinforcing its role as a dynamically controlled node rather than a constant-rate enzyme.
Figure 2. A model depicting the role of protein phosphorylation of GAD65 on GABA neurotransmission. (Source: Chou CC, et al. 2017)
Trafficking and Vesicular Targeting
GAD65 is synthesized as a soluble, hydrophilic protein but rapidly acquires hydrophobic character and becomes anchored to the cytosolic face of endoplasmic reticulum and Golgi membranes, after which it is forwarded to post-Golgi vesicles and ultimately to the cytosolic face of synaptic vesicles. The itinerary is governed by a palmitoylation cycle. Kanaani and colleagues demonstrated that reversible S-palmitoylation dynamically partitions GAD65 between ER-Golgi and post-Golgi membranes: palmitoylation clears the enzyme from the ER-Golgi, directs it through the trans-Golgi network, and sorts it into a post-Golgi vesicular pathway, while depalmitoylation returns it to the Golgi by a non-vesicular route. This cycle can redistribute GAD65 between cytosolic and membrane-anchored pools far faster than vesicular trafficking alone, providing a mechanism to tune where GABA is made.
At the nerve terminal, membrane-anchored GAD65 is positioned to load newly synthesized GABA into synaptic vesicles. Jin and colleagues provided direct evidence for functional coupling between GABA synthesis and vesicular GABA transport, showing that GAD65 at the synaptic vesicle membrane, together with chaperone proteins such as HSC70 and cysteine-string protein, facilitates the efficient transfer of GABA into vesicles via the vesicular GABA transporter (VGAT). In pancreatic beta cells, GAD65 localizes to Golgi membranes and to cytosolic vesicles that are distinct from insulin-containing granules, reflecting a comparable logic in which the enzyme is placed at the site of GABA production for release. The convergence of palmitoylation-dependent trafficking and VGAT coupling ensures that GAD65-derived GABA is delivered into the same compartments from which it is secreted.
Frequently Asked Questions
GAD65 is regulated by phosphorylation, palmitoylation, and calcium-dependent proteolysis. Phosphorylation of N-terminal serines distinguishes the membrane-anchored from the soluble pool, and phosphorylation in the catalytic domain can activate the enzyme. A reversible palmitoylation cycle controls its partitioning between ER-Golgi and post-Golgi membranes. During neuronal activity, calcium influx activates calpain, which cleaves N-terminal residues from GAD65 to generate a more active truncated form. Together these modifications tune GABA synthesis to physiological demand.
GABA acts through two main receptor classes. GABAA receptors are pentameric ligand-gated chloride channels with extensive subunit diversity; synaptic receptors mediate fast phasic inhibition and extrasynaptic receptors mediate tonic inhibition, the direction depending on the intracellular chloride set point. GABAB receptors are GB1/GB2 heterodimeric GPCRs that couple to Gi/o, inhibiting adenylyl cyclase, opening GIRK potassium channels, and closing calcium channels to produce slow inhibition. Both classes transduce GAD65-derived GABA into changes in membrane excitability and second-messenger signaling.
Yes. Protein array platforms enable the parallel, quantitative measurement of dozens of phosphoproteins, receptor subunits, and signaling intermediates from a single small-volume sample. Combined with readouts of GAD65 modification state (phosphorylation, palmitoylation, truncated forms) and GABAA/GABAB pathway activity (chloride flux, CREB phosphorylation, GIRK activity), these platforms can provide a multidimensional picture of GABAergic and islet signaling in basic and translational neuroscience research.
References
- Kass I, et al. Cofactor-dependent conformational heterogeneity of GAD65 and its role in autoimmunity and neurotransmitter homeostasis. Proc Natl Acad Sci U S A. 2014 Jun 24;111(25):E2524-9.
- Chou CC, et al. Activation of Brain L-glutamate Decarboxylase 65 Isoform (GAD65) by Phosphorylation at Threonine 95 (T95). Mol Neurobiol. 2017 Mar;54(2):866-873.
- Fenalti G, et al. GABA production by glutamic acid decarboxylase is regulated by a dynamic catalytic loop. Nat Struct Mol Biol. 2007 Apr;14(4):280-6.
- Kanaani J, et al. A palmitoylation cycle dynamically regulates partitioning of the GABA-synthesizing enzyme GAD65 between ER-Golgi and post-Golgi membranes. J Cell Sci. 2008 Feb 15;121(Pt 4):437-49.
- Hagan DW, et al. The role of GABA in islet function. Front Endocrinol (Lausanne). 2022 Sep 29;13:972115.