Molecular defects of the glycine 41 variants of alanine glyoxylate aminotransferase associated with primary hyperoxaluria type I

Abstract

G41 is an interfacial residue located within the alpha-helix 34-42 of alanine:glyoxylate aminotransferase (AGT). Its mutations on the major (AGT-Ma) or the minor (AGT-Mi) allele give rise to the variants G41R-Ma, G41R-Mi, and G41V-Ma causing hyperoxaluria type 1. Impairment of dimerization in these variants has been suggested to be responsible for immunoreactivity deficiency, intraperoxisomal aggregation, and sensitivity to proteasomal degradation. However, no experimental evidence supports this view. Here we report that G41 mutations, besides increasing the dimer-monomer equilibrium dissociation constant, affect the protein conformation and stability, and perturb its active site. As compared to AGT-Ma or AGT-Mi, G41 variants display different near-UV CD and intrinsic emission fluorescence spectra, larger exposure of hydrophobic surfaces, sensitivity to Met53-Tyr54 peptide bond cleavage by proteinase K, decreased thermostability, reduced coenzyme binding affinity, and catalytic efficiency. Additionally, unlike AGT-Ma and AGT-Mi, G41 variants under physiological conditions form insoluble inactive high-order aggregates (approximately 5,000 nm) through intermolecular electrostatic interactions. A comparative molecular dynamics study of the putative structures of AGT-Mi and G41R-Mi predicts that G41 --> R mutation causes a partial unwinding of the 34-42 alpha-helix and a displacement of the first 44 N-terminal residues including the active site loop 24-32. These simulations help us to envisage the possible structural basis of AGT dysfunction associated with G41 mutations. The detailed insight into how G41 mutations act on the structure-function of AGT may contribute to achieve the ultimate goal of correcting the effects of these mutations.

Fig. 1.

Effect of proteinase K on holoG41R-Mi. HoloG41R-Mi (15 μM) was incubated at 25 °C in 100 mM potassium phosphate buffer, pH 7.4, at a 1,000/1 (wt/wt) mutant/proteinase K ratio. At times indicated, aliquots were removed, treated (see Methods), and subjected to 12% SDS-PAGE. Plus and minus signs indicate presence or absence of proteinase K.

Fig. 2.

Time dependence of turbidity of AGT-Ma, AGT-Mi, and G41 variants. Absorbance at 600 nm as a function of time of 4 μM apo forms of AGT-Ma (▪), AGT-Mi (▾), G41R-Ma (▴), G41V-Ma (♦), and G41R-Mi (•) in potassium phosphate buffer, pH 7.4, I = 150 mM at 37 °C. Corresponding holoenzymes, open symbols.

Fig. 3.

Time-dependence of total count rate (measured as kilo counts per second) of AGT-Ma, AGT-Mi, and G41 variants in the holo (A) and apo (B) forms. Measurements performed at 4 μM enzyme concentration, 37 °C, I = 150 mM, pH 7.4. Color code: black, AGT-Ma; red, AGT-Mi; blue, G41R-Ma; green, G41V-Ma; fuchsia, G41R-Mi.

Fig. 4.

Time dependence of the apparent diameters of AGT-Ma, AGT-Mi, and G41 variants. Experimental conditions (see legend to Fig. 3). (A) holo and apo AGT-Ma. (B) holo and apo AGT-Mi. (C) holo and apo G41R-Ma. (D) holo and apo G41V-Ma. (E) holo and apo G41R-Mi. Color code: black, holo dimer; green, apo dimer; red, holo small aggregates; blue, apo small aggregates; cyan, holo high aggregates; fuchsia, apo high aggregates.

Fig. 5.

Comparison of the initial 3D model of G41R-Mi (dark gray) with the averaged structure obtained from MD simulation (light gray). Residues 1–46, roughly corresponding to the N-terminal arm of G41R-Mi, are highlighted in orange for the initial 3D model and in red for the averaged structure. (Inset) Detail of the α-helix in which the G41R substitution takes place (residues 34–42).

Fig. 6.

Electrostatic potential surface maps of AGT-Mi and its truncated form. Electrostatic gradient and map (kT/e) of AGT-Mi (A), and 1–44 truncated form of AGT-Mi (B).