Glycerol_3_phosphate_Dehydrogenase

Glycerol-3-phosphate dehydrogenase (PDB ID: 2QCU) from Escherichia coli                                                     Created by: Luke Cavanah

            Sn-glycerol-3-phosphate dehydrogenase (GlpD; PDB ID: 2QCU) is a wild-type monotopic membrane enzyme found in Escherichia coli (E. coli). GlpD is an oxidoreductase within the inner membrane of E. coli and has implications for respiration, glycolysis, and phospholipid biosynthesis (1-2). The biological significance of GlpD is supported by its role in the three aforestated vital metabolic processes, multi-level regulation, and high number of homologs found across nearly all organisms (2-3).

GlpD is one of the main dehydrogenases of the electron transport chain (ETC), where it catalyzes the oxidation of glycerol-3-phosphate (G3P) to dihydroxyacetone phosphate (DHAP) and the reduction of flavin adenine dinucleotide (FAD) to FADH­­2 (2). Subsequently, GlpD passes electrons on to ubiquinone (UQ) in the ETC (2,4). Together, ETC and chemiosmosis generate the most ATP during cellular respiration, reinforcing GlpD’s biological importance (4).

GlpD was expressed in E. coli (1-2). The protein was crystallized through vapor diffusion sitting drop and hanging drop methods (2). β-octylglucoside (βOG) was used as detergent in expression, purification, and the detergent reconstitution process of GlpD (2). Structural data was obtained through x-ray diffraction. Phosphate ion, ethylene glycol, tris(hydroxyethyl)aminomethane, and imidazole are ligands present in the crystal structure that were used merely to induce crystallization, and serve no biological function in this biomolecule. βOG is another ligand present in the crystal structure, which has no biological function in GlpD (1-2).

 GlpD has a molecular weight of 113483.07 Da, and an isoelectric point of 7.03 (5). GlpD has two identical subunits (1-2). The primary structure of GlpD consists of 1002 residues between its two subunits. Though the subunits consist of both hydrophobic and hydrophilic amino acids, hydrophobic residues greatly predominate (1-2). The high hydrophobic content suits its location in the highly nonpolar cell membrane (2).

The secondary structure of each subunit is approximately 23% β-sheet (26 strands, 119 residues), 36% α-helix (25 helices, 184 residues), with the remaining percentage being random coils and 3/10 helices (1). Both parallel and anti-parallel β-sheet are present. Hydrogen bonding between carbonyl oxygen atoms and amide nitrogen atoms in the peptide backbone provide significant stabilization to GlpD’s structure and neutralization of the polar N-H and C=O functions of the peptide backbone (2).

GlpD contains two important domains: a soluble extramembranous C-terminal domain and a N-terminal FAD-binding site (2). All of GlpD is represented in the published structure, and the entire structure was crystallized (1-2). Residues 1-388 make up the N-terminal domain, and residues 399-501 make up the C-terminal domain; no residues are unresolved (2).

In regard to tertiary structure  of GlpD as described by secondary-structure elements, the β-sheets tend to predominate toward the periphery of the protein, and the α-helices are prevalent throughout all parts of the protein (1). The base of GlpD that is located in the plasma membrane is distinctly positive. Electrostatic attraction between the positively charged regions of GlpD and the negatively charged phospholipid head groups serves to anchor GlpD in the fluid membrane. The cap domain is distinctly negative, which is hypothesized to be the UQ-binding surfaces (2-3).

Due to GlpD’s homodimeric nature, GlpD possesses quaternary structure (1-2). Dimer formation is structurally important because it reduces the surface-area-to-volume ratio, thus enabling burying of the hydrophobic residues; this hydrophobic collapse is especially important because it is of GlpD’s high hydrophobic content. Furthermore, burying of the hydrophobic residues is entropically favorable, providing significant structural stability. The dimer structure also allows protection of the active site because the monomeric units enclose the site where dehydrogenation occurs. The apolar microenvironment created by the dimer structure is required for catalysis (2).

Phosphoenolypyruvate (PEP), glyceric acid 2-phosphate (2-PGA), and glyceraldehyde-3-phosphate (GAP) are three substrates that complex to GlpD, and DHAP is a product that complexes to GlpD (1-2). Hydrogen-bonding interactions between PEP, 2-PGA, GAP, and DHAP define the active site. Ser-46, Leu-48, His-50, Leu-355, and Thr-356 interact with FAD, so substrates only bind at one face of the isoalloxazine ring. Arg-316 mediates the binding of the phosphate of the substrate analogues. Binding of the phosphate of the substrate aligns the substrate for dehydrogenation. Arg-54, Tyr-55, Thr-270, Thr-271, Asp-272, Arg-317, and Arg-332 mediate binding of substrates. Arg-317 likely catalyzes the initial deprotonation of the hydroxyl group bonded to C2 of G3P due to its proximity. The hydrogen bonding between the guanidino nitrogen atoms of Arg-317 to the oxygen atoms of the phosphate Asp-272 likely increase the basicity of Arg-317, thus facilitating the initial deprotonation of G3P. Following dehydrogenation, hydride transfer occurs from C2 of G3P to N5 of FAD. Leu-355 and especially Lys-354 stabilize the reduced flavin formed after the hydride transfer occurs. His-233 likely acts as a general base catalyst. FAD is the only prosthetic group present in GlpD. No associated metal ions appear to be required for GlpD activity (2).

            Position-specific-iterated Basic Local Alignment Search Tool (PSI-BLAST) is a program that identifies proteins with similar primary structure as the sequence queried. PSI-BLAST calculates an E value for the proteins identified with similar sequence to the query by examining the total sequence similarity and assigning gaps. Sequence homology is inversely related to the E value, and gaps are directly related to the E value. An E value of less than .05 suggests the protein has a similar sequence to that of the query (6). Flavoenzyme PA-4991 (PDB: 5EZ7), typically derived from Pseudomonas aeruginosa, and GlpD have a comparative E value of 3•10-20, indicating highly similar primary structures (6-7).

            The Dali server compares tertiary structures of proteins and calculates the differences in intramolecular distances by a statistical method known as the “sums-of-pairs” method. The Dali server then assigns a Z-score for the proteins identified to have similar tertiary structures to the query. A Z-score greater than 2 suggests the protein has a similar tertiary structure to that of the query. Flavoenzyme PA-4991 and GlpD have a comparative Z-score of 23.2, indicating highly similar tertiary structures (8).

            Both GlpD and flavoenzyme PA-4991 are part of the family of FAD-dependent oxidoreductases, and thus they both contain the FAD-binding domain and a FAD ligand (2, 3, 9). Additionally, both GlpD and flavoenzyme PA-4991 contain three of the same active-site residues: His-53, Arg-316, and Lys-345. Both proteins also share the structural feature of an isoalloxazine ring (2, 9).

Despite the structural similarities between GlpD and flavoenzyme PA-4991 above, there are many structural and functional differences as well. In terms of primary structure, flavoenzyme PA-4991 contains only one subunit, which has 392 residues (7, 9). The secondary structure of flavoenzyme-4991 contains a significantly lower proportion of α-helices, and a slightly higher proportion of β-sheets: flavoenzyme is 26% helical and 27% β-sheet (7). One major difference in the tertiary structures of GlpD and flavoenzyme PA-4991 is the structure of their active sites: the re face of the isoalloxazine ring is accessible from the solvent in flavoenzyme PA-4991, unlike in GlpD (2, 9). Flavoenzyme PA-4991 only has one chain, and GlpD has two; thus flavoenzyme PA-4991 lacks quaternary structure. Another notable structural difference between the two proteins is flavoenzyme PA-4991 has mercury (II) ion ligate to the structure, when GlpD does not (1, 2, 7, 9). The differences between GlpD and flavoenzyme PA-4991 manifest as functional changes, too, as illustrated by the lack of enzymatic activity of flavoenzyme PA-4991 when DHAP is the substrate, and their difference in substrate specificity (9).

            Ultimately, the biological significance of GlpD in metabolism is due to its involvement in the dehydrogenation of G3P (2). GlpD’s unique primary, secondary, tertiary, and quaternary levels of structure enable it to be a successful monotopic-membrane oxidoreductase (2). The transfer of electrons from GlpD to UQ is not well-understood, particularly UQ’s docking site. Future directions include investigating UQ’s docking site on GlpD.