Mammalian glutamate dehydrogenase (GDH) is a mitochondrial enzyme that catalyzes the reversible oxidative deamination of L-glutamate to 2-oxoglutarate using NAD(P)+ as coenzyme.  The enzyme is a homohexamer that is tightly regulated by a large number of positive and negative allosteric effectors as well as by cooperative interactions between subunits.
Although the allosteric regulation of GDH has been intensively studied, its role in human physiology has remained unclear until recent demonstration of the involvement of GDH in a newly discovered form of congenital hypoglycemia; hyperinsulinism and hyperammonemia (HI / HA) syndrome (Weinzimer, Stanley et al. 1997) Stanley, Lieu et al. 1998).  Congenital hyperinsulinism is the most common cause of hypoglycemia in early infancy and can result in seizures or coma (Stanley and Baker 1976).  The majority of these cases appear to be due to genetic defects in the regulation of insulin secretion by pancreatic beta cells (Stanley and Baker  1999).  In some children, however, hyperinsulinism is also associated with hyperammonemia.  In these children, plasma ammonium concentrations are 100-200 µM (3-8 times normal).  The hyperammonemia is not associated with symptoms and is unaffected by changes in blood glucose levels.  In a recent report, eight patients were found to be heterozygous for mutant forms of GDH that were unresponsive to the inhibitor, GTP.  If GDH were to mainly operate in the oxidative deamination reaction in the pancreas and liver, then decreased levels of L-glutamate will inhibit urea production since high concentrations of glutamate are needed for N-acetylglutamate production (Stewart and Walser 1980).  In the beta cells of the pancreas, the increased glutamate to a-ketoglutarate conversion will increase respiration rate via the Krebs cycle, increase the ATP/ADP ratio, leading to inhibition of K+ channel activity, and result in excessive release of insulin.  This may also explain why these patients are particularly sensitive to protein-induced hypoglycemia (Hsu, Kharlip et al. 1998; Kelly, Ferry et al. 1999).  Leucine is an activator of GDH (Sener and Malaisse 1980) and therefore ingestion of protein could exacerbate the already hyperactive state of GDH.
The observations in children with the HI/HA syndrome implicate GDH as the key sensor for protein mediated insulin secretion, analogous to the role of glucokinase as the sensor for glucose regulated insulin secretion (Matschinsky and Sweet 1996).  GDH also appears to be an important site for controlling hepatic amino acid oxidation and ureagenesis in response to protein ingestion.  GDH is also highly expressed in the kidney where it might have a similar function in linking renal function to amino acid and protein metabolism.  In addition to our new evidence that GDH plays a major role in regulating hepatic ureagenesis and insulin release, GDH is likely to be important for normal function of the brain, since activity in brain is also very high. Defects in GDH activity have been reported in patients with cerebellar degeneration (Plaitakis, Flessas et al. 1993).   The absence of symptoms related to hyperammonemia in HI/HA syndrome children suggests that increased GDH activity may protect against cerebral ammonium toxicity, perhaps by reducing the levels of glutamate, an important excitatory neurotransmitter.  Thus, the detailed information about how GDH activity is regulated that will be provided by the proposed studies is highly relevant to a wide range of normal and abnormal human physiology, including diabetes, protein catabolism, and the control of brain neurotransmitters. Our work will provide a major advance in our understanding of the biochemistry of GDH and, through our mutagenesis studies, the tertiary structural details of the complex allosteric regulation of this key enzyme. These results, combined with binding and kinetic analysis of the HI/HA mutants, will allow us to fully understand the basis of the HA/HI syndrome.  Furthermore, these analyses will allow us to resolve the controversy surrounding the role of mammalian GDH in-vivo. This detailed information about GDH regulation is highly relevant to a wide range of normal and abnormal human physiology, including diabetes, protein catabolism, and the control of brain neurotransmitters. The Figure above is a ribbon diagram of the entire boGDH hexamer.  In this diagram, each of the identical subunits are represented by different colors.  The N-terminal GLU binding domain forms all of the interactions across the non-crystallographic 2-fold axes.  The NAD+ binding domain rests on top of the GLU binding domain giving the subunit a ‘clam-like’ appearance with the active site lying in the cleft between the two domains.  The NAD+ binding domain makes very little contact with the other subunits in the hexamer.  In contrast, the antenna region, extending from the top of the NAD+ binding domain, forms extensive interactions with other subunits in the trimers.  The long helix is quite amphipathic with a large number of leucines and valines contacting the other helices that form this right-handed helical bundle.  This large, 48-residue antenna region is not found in non-mammalian GDH and may be intimately involved in allosteric regulation.
	During initial model building, the active site glutamate and NADH molecules were clearly visible and added to the atomic model.  As observed in C. symbiosum GDH, glutamate binds to the deeper recess of the active site and NADH binds towards the opening (Stillman, Baker et al. 1993).  The B-face of the nicotinamide ring lies immediately above the glutamate Ca atom.  The active site structure in contact with the bound glutamate and the nicotinamide portion of NADH is remarkably similar to that observed in csGDH (Stillman, Baker et al. 1993).  However, the residues in contact with the adenine ribose moiety are significantly different - perhaps leading to differences in coenzyme usage (i.e. NAD(H) vs. NADP(H)) between bacterial and mammalian GDH.
Shown here is a single subunit of bovine glutamate dehydrogenase.  Shown in purple is the NAD binding domain, yellow is the glutamate binding domain, and orange is the allosteric 'antenna'.  The core of the hexamer is at the lower right side, the nearest 2-fold related subunit is directly beneath this subunit.  Shown in dark mauve is the NADH molecule that resides in the active site.  The red molecule is the substrate, glutamate.  The light mauve molecule is the NADH molecule that, when bound, inhibits the reaction.  We propose that this inhibition is due to a 'gumming up' of the ability of the subunit to move and release product after the reaction.  There are two molecules of GTP bound in this model.  The major, inhibitory, molecule is shown in white.  A second molecule, whose function is not clear, is shown in light and dark green.  These latter two molecules are equivalent - one is just from the adjacent subunit.  It is important to note that the children with hyperinsulinism/hyperammonemia have mutations about the first (white) GTP molecule.  We propose that GTP acts in a manner analogous to the second NADH site - it blocks movement of the NAD binding domain that is needed to release product.
Shown here are the locations of the hyperinsulinism/ hyperammonemia mutations with respect to the GTP molecule bound to the first site.  The GTP molecule is shown as a ball&stick figure, the C-alpha backbone is shown as a stick model (with color ramped from red->blue as you go from the N to C terminus), and the sidechains that are mutated in this disorder are shown as stick figures extending from the backbone.  All of these mutations decrease the sensitivity of the enzyme to inhibition by GTP.  Since most mutations do not contact the bound GTP, it seems likely that much of the mutational effect is on how GTP acts rather than merely how it binds.  As reviewed above, these mutations probably make the children's GDH too active, make too much 2-oxoglutarate that feeds the Krebbs cycle.  This increases the ATP/ADP ratio and causes too much insulin to be released from the pancreas.

[Return to top]