Contribution of breakdown in the regulation of mouse liver glutathione S-transferase subunits during protein depletion and re-feeding

Glutathione transferases in the bioactivation of azathioprine

The prodrug azathioprine is primarily used for maintaining remission in inflammatory bowel disease, but approximately 30% of the patients suffer adverse side effects. The prodrug is activated by glutathione conjugation and release of 6-mercaptopurine, a reaction most efficiently catalyzed by glutathione transferase (GST) A2-2. Among five genotypes of GST A2-2, the variant A2*E has threefold-fourfold higher catalytic efficiency with azathioprine, suggesting that the expression of A2*E could boost 6-mercaptopurine release and adverse side effects in treated patients. Structure-activity studies of the GST A2-2 variants and homologous alpha class GSTs were made to delineate the determinants of high catalytic efficiency compared to other alpha class GSTs. Engineered chimeras identified GST peptide segments of importance, and replacing the corresponding regions in low-activity GSTs by these short segments produced chimeras with higher azathioprine activity. By contrast, H-site mutagenesis led to decreased azathioprine activity when active-site positions 208 and 213 in these favored segments were mutagenized. Alternative substitutions indicated that hydrophobic residues were favored. A pertinent question is whether variant A2*E represents the highest azathioprine activity achievable within the GST structural framework. This issue was addressed by mutagenesis of H-site residues assumed to interact with the substrate based on molecular modeling. The mutants with notably enhanced activities had small or polar residues in the mutated positions. The most active mutant L107G/L108D/F222H displayed a 70-fold enhanced catalytic efficiency with azathioprine. The determination of its structure by X-ray crystallography showed an expanded H-site, suggesting improved accommodation of the transition state for catalysis.

Differences among allelic variants of human glutathione transferase A2-2 in the activation of azathioprine

Azathioprine has been clinically used for decades in connection with organ transplantation, autoimmune disease, and treatment of cancer. Toxic side-reactions are common and have been linked to the liberation of excessively high concentrations of 6-mercaptopurine and corresponding toxic metabolites. An allelic variant of thiopurine methyltransferase with low activity is associated with elevated concentrations of 6-mercaptopurine. However, other genetic markers remain to be identified in order to fully account for adverse reactions and efficacy failure. In the present study, we studied the five known allelic variants of human glutathione transferase A2-2 (GST A2-2) (EC 2.5.1.18), abundantly expressed in liver and efficiently catalyzing the bioactivation of azathioprine to release 6-mercaptopurine. All five variants exhibited high activity with azathioprine, but allelic variant E of GST A2-2 displayed a 3-4-fold elevated catalytic efficiency compared to the other variants. High GST activity can lead to overproduction of 6-mercaptopurine, and the nature of the multiple forms of GSTs in a patient will obviously affect the metabolism of azathioprine. In addition to GST A2-2, the polymorphic GST M1-1 is also highly active with azathioprine. Considering our findings, it appears that the genotypic and phenotypic variations in the GST complement may influence the presentation of adverse reactions in patients treated with azathioprine. Clinical trials will be required to clarify the impact of the GST expression in comparison with the established biomarker thiopurine methyltransferase as predictors of adverse reactions.

The mouse liver content of carbonic anhydrase III and glutathione S-tranferases A3 and P1 depend on dietary supply of methionine and cysteine

The contents of glutathione S-transferase (GST) subunits, carbonic anhydrase III (CAIII), glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and a 230 kDa protein are affected by protein deprivation in mouse liver. In order to know if particular amino acids control these contents, the effects of feeding for 5 days with diets containing different amino acids were examined. After an exploration using SDS-PAGE analysis, the action of selected diets was further examined by distinct techniques. The 230 kDa protein was identified as fatty acid synthase (FAS) by both mass spectrometry and amino acid sequence analyses. Dietary tests showed that: (1) a protein-free diet (PFD) increased the content of glutathione S-transferases P1 and M1, and glyceraldehyde-3-phosphate dehydrogenase, while the content of glutathione S-transferase A3, fatty acid synthase and carbonic anhydrase III decreased; (2) a protein-free diet having either methionine or cysteine preserved the normal contents of glutathione S-transferases P1, A3, M1 and carbonic anydrase III; (3) a protein-free diet having threonine preserved partially the normal contents of glutathione S-transferases P1, A3, M1 and carbonic anhydrase III; (4) a protein-free diet having methionine, threonine and cysteine prevented in part the loss of fatty acid synthase; and (5) the glyceraldehyde-3-phosphate dehydrogenase content was controlled by increased carbohydrate level and/or by lower amino acid content of diets, but not by any specific amino acid. These data indicate that methionine and cysteine exert a main role on the control of liver glutathione S-transferases A3 and P1, and carbonic anhydrase III. Thus, they emerge necessary to prevent unsafe alterations of liver metabolism caused by protein deprivation.

Two dimensional non equilibrium pH gel electrophoresis mapping of cytosolic protein changes caused by dietary protein depletion in mouse liver

Two-dimensional non-equilibrium pH gel electrophoresis (2D-NEPHGE) analysis was used to evaluate the effects of dietary protein depletion on the protein composition of mouse liver cytosol. Analysing the cytosol from both normal and protein depleted liver, the position in gels of more than three hundred protein spots was determined. After 5 days of protein depletion, about 20% of the spots either increased or decreased more than 2 fold. Five spots of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were recognised by specific antibodies. The glutathione S-transferase (GSTs) subunits Ybl, Yc and Yf were identified by the simultaneous analysis of both glutathione-binding cytosolic proteins and the corresponding standards. As estimated by internal optical density (IOD) of spots, the changes caused by protein depletion in GAPDH and GST subunit contents were similar to those obtained by other methods. By means of mass spectrometric analysis of tryptic peptides generated from spots and/or comparison of two-dimensional gel electrophoretic patterns, carbonic anhydrase III (CAIII), Cu, Zn superoxide dismutase (CuZnSOD) and a cytochrome P450 cytosolic protein (cyt P450) were identified. These three proteins, as well as GSTs, are related with intracellular detoxification and free radical scavenging systems. Their contents were regulated by dietary protein restriction in a manner indicative of diminished liver defence against oxidising agents.

The effect of cysteine on the altered expression of class alpha and mu glutathione S-transferase genes in the rat liver during protein-calorie malnutrition

Protein-calorie malnutrition (PCM) represents a global health problem. The breakdown rate of glutathione S-transferase (GST) subunits determines their differential contents during protein depletion. Hepatic GST expression and the underlying mechanistic basis were investigated in PCM rats. PCM caused no change in rGSTA1/2 subunit. In contrast, rGSTA3/5 subunit was 2.4-fold induced during PCM, while the levels for rGSTM1 and M2 subunits were 30% and 70% suppressed. Increased GSTA3/5 expression was significantly prevented by cysteine or methionine treatment, although such treatment failed to restore the rGSTM2 level. In contrast to differential GST protein expression, PCM caused a 5-10-fold increase in rGSTA2/A3/A5 and M1 mRNAs, whereas rGSTM2 mRNA was 70% decreased. The elevations in rGSTA2/A3/A5 and M1 mRNAs were completely abolished by cysteine or methionine treatment during PCM, although the rGSTM2 mRNA level was not restored. PCM induced oxidative stress in the liver, as evidenced by protein carbonylation. Antioxidant response element (ARE)-binding activity of nuclear extracts from PCM rats was increased, which was immunodepleted with anti-Nrf-1/2 antibodies. Activation of nuclear ARE-binding proteins was inhibited by cysteine. Data showed that hepatic GSTs were differentially expressed during PCM, that certain GST mRNAs were increased with the ARE activation, and that cysteine was active in preventing increases in GST mRNAs and ARE activation.