Allosteric regulation of glutamate dehydrogenase deamination activity

Glutamate dehydrogenase (GDH) is a key enzyme interlinking carbon and nitrogen metabolism. Recent discoveries of the GDH specific role in breast cancer, hyperinsulinism/hyperammonemia (HI/HA) syndrome, and neurodegenerative diseases have reinvigorated interest on GDH regulation, which remains poorly understood despite extensive and long standing studies. Notwithstanding the growing evidence of the complexity of allosteric network behind GDH regulation, identifications of allosteric factors and associated mechanisms are paramount to deepen our understanding of the complex dynamics that regulate GDH enzymatic activity. Combining structural analyses of cryo-electron microscopy data with molecular dynamic simulations, here we show that the cofactor NADH is a key player in the GDH regulation process. Our structural analysis indicates that, binding to the regulatory sites in proximity of the antenna region, NADH acts as a positive allosteric modulator by enhancing both the affinity of the inhibitor GTP binding and inhibition of GDH catalytic activity. We further show that the binding of GTP to the NADH-bound GDH activates a triangular allosteric network, interlinking the inhibitor with regulatory and catalytic sites. This allostery produces a local conformational rearrangement that triggers an anticlockwise rotational motion of interlinked alpha-helices with specific tilted helical extension. This structural transition is a fundamental switch in the GDH enzymatic activity. It introduces a torsional stress, and the associated rotational shift in the Rossmann fold closes the catalytic cleft with consequent inhibition of the deamination process. In silico mutagenesis examinations further underpin the molecular basis of HI/HA dominant mutations and consequent over-activity of GDH through alteration of this allosteric communication network. These results shed new light on GDH regulation and may lay new foundation in the design of allosteric agents.

helices. Moreover, there is a lack of similarity in the NBD (α6 -α11) and Antenna regions.

Notes: Deviation of Cα atoms:
Cα deviation movement is measured using the following formula: where ( ) are the position of the Cα atoms of every residue within GDH (pdb ids: 3jd3 and 3jd4). The deviation is measured in all the cases including the apo GDH structure (pdb id: 3jcz). Then the deviations are compaired and visualized with the different conformation of GDH using Python coding. Helix numbers are also included with this comparison to indicate the region of changes in Cα deviation.    The above table summarizes the interfaces of open form structure (pdb id: 3jd3) and closed form structure (pdb id: 3jd4). N at, N res indicate the number of atoms and the number of residues respectively involved in the interface area. N HB indicates number of hydrogen bonds present at the interfaces and ΔG (Kcal/mol) is the solvation energy for folding. All information was extracted from PDBePISA (https://www.ebi.ac.uk/pdbe/pisa/).    This table shows the location and frequency of hyperinsulinism-hyperammonemia syndrome (HI/HA) associated mutations in the GDH. Out of the total 84 cases, 66% is sporadic and 34% familial 4,5 . The table also indicates the three mutational hotspot regions within the protein: GTP binding sites, Antenna, and the Pivotal helices region; however, the highest number of mutations in HI/HA cases has been observed at Leu441 whose position is in the junction of pivotal and Antenna helices. Another position of mutation is His265 whose frequency is also significantly higher in the HI/HA cases.

Table S6. Rigid cluster and flexibility analysis using ProFlex/FIRST
Hydrogen bond energy strength is considered from 0 to -2.0, as the communication is started after hydrogen bond cut off 0.0 and has been stopped before the cut-off of -1.5 ( Figure S5). The table also indicates the position and percentage of helices that are involved within the rigid cluster at the different hydrogen bond energy cut-off.

Fig. S6. Standard tripeptide chain geometry
Amino acids in proteins (or polypeptides) are joined together by peptide bonds. R 1 , R 2 and R 3 group represent the side chain of each amino acid.    2). All data were generated by SHIFTX2 (http://www.shiftx2.ca/) using experimental pH and temperature(K) available in PDB data 8 .

Fig.S10.
AlloSigMA webserver is used for predicting allosteric communications between allosteric sites GTP, NADH reg , and the catalytic sites triggered by GTP-binding for the GDH open conformer. It estimates per-residue allosteric free energies resulting from GTP and Fig.S11. AlloSigMA webserver is used for predicting allosteric communications between allosteric sites GTP, NADH reg , and the catalytic sites triggered by GTP-binding for the GDH closed conformer. It estimates per-residue allosteric free energies resulting from GTP and NADH reg binding. Positive sign shows the effects of local destabililization, whereas negative sign indicates effects of local stabilization.  Residues with same sign move together in the same direction and it is predicted to form a dynamically coupled regions. It also predicts the global hinge sites located between the sequences segments that undergo opposite direction movements along the slowest mode. Across all the six monomers, it shows domain separation of NBD and regulatory sites by the allosteric regions lie within residues ~200-400. Table S7. Frequency of amino acids in the high-energy transition region. ALA-194  12  Regulatory  ALA-341  31  coenzyme binding domain  ALA-375  22  Catalytic&376  ALA-443  91  Regulatory  ASN-254  20  Catalytic  ASP-119  17  --------GLU-25  42  --------GLU-328  49  Catalytic  GLU-36  19  --------GLY-243  68  coenzyme binding domain  GLY-350  53  Catalytic  GLY-376  197  Catalytic  GLY-377  19  Catalytic  HIS-189  24  Regulatory  HIS-209  31  Regulatory  LEU-371  14  coenzyme binding domain and alpha11  LYS-329  28  Catalytic  MET-169  63  Catalytic  PHE-122  84  ---------PHE-304  10  coenzyme binding domain  PRO-165  39  Catalytic  PRO-167  40  Catalytic  PRO-202  21  Regulatory  PRO-240  32  coenzyme binding domain  PRO-288  16  coenzyme binding domain  PRO-354  19  coenzyme binding domain  PRO-432  82  Antenna  PRO-7 Table listed all the amino acid residues with their frequency higher than 10 that fall into the region -35 0 < φ < 35 0 along the MD trajectory. Cyan colour indicates the frequency of Gly376 and its neighbouring residues. With Gly376Asp mutation, table listed the amino acid residues with the frequency more than 10 that lie in the high energy passes demarcated by -35 0 < φ < 35 0 along the computed MD trajectory. Gly376 and its neighbourhood residues were highlighted in cyan colour showing less frequency compared to the previous table.