PARK7 Catalyzes Stereospecific Detoxification of Methylglyoxal Consistent with Glyoxalase and Not Deglycase Function

Abstract

The protein PARK7 (also known as DJ-1) has been implicated in several diseases, with the most notable being Parkinson's disease. While several molecular and cellular roles have been ascribed to DJ-1, there is no real consensus on what its true cellular functions are and how the loss of DJ-1 function may contribute to the pathogenesis of Parkinson's disease. Recent reports have implicated DJ-1 in the detoxification of several reactive metabolites that are produced during glycolytic metabolism, with the most notable being the α-oxoaldehyde species methylglyoxal. While it is generally agreed that DJ-1 is able to metabolize methylglyoxal to lactate, the mechanism by which it does so is hotly debated with potential implications for cellular function. In this work, we provide definitive evidence that recombinant DJ-1 produced in human cells prevents the stable glycation of other proteins through the conversion of methylglyoxal or a related alkynyl dicarbonyl probe to their corresponding α-hydroxy carboxylic acid products. This protective action of DJ-1 does not require a physical interaction with a target protein, providing direct evidence for a glutathione-free glyoxalase and not a deglycase mechanism of methylglyoxal detoxification. Stereospecific liquid chromatography-mass spectrometry (LC-MS) measurements further uncovered the existence of nonenzymatic production of racemic lactate from MGO under physiological buffer conditions, whereas incubation with DJ-1 predominantly produces l-lactate. Collectively, these studies provide direct support for the stereospecific conversion of MGO to l-lactate by DJ-1 in solution with negligible or no contribution of direct protein deglycation.

Figure 1

Purified DJ-1 actively detoxifies methylglyoxal. (A) Schematic depicting the formation and cellular fates of methylglyoxal. (B) Representative Coomassie gel and anti-FLAG Western blot of purified FLAG-DJ-1 isolated from HEK293T cells stably overexpressing Flag-DJ-1. (C) MGO quantification in recombinant assays containing the indicated MGO concentration and protein condition following incubation for 24 h at 37 °C. Data plotted in (C) are mean ± SEM from n = 6 independent biological replicates. Statistical analyses are by ordinary one-way analysis of variance (ANOVA). *p < 0.05; **p < 0.01; ***p < 0.001; and ****p < 0.0001.

Figure 2

DJ-1 detoxifies an alkynylated analogue of MGO and protects proteins from modification. (A, B) Structure (A) and extracted ion chromatogram (B) of the predicted lactoyl-alkyne metabolic product of MG-alkyne treated with equal amounts of DJ-1 or BSA for 24 h at 37 °C. (C) Quantification of remaining MGO from the reaction where 1 mM MGO was treated with equal amounts of recombinant DJ-1 or BSA or with the vehicle for 24 h at 37 °C. (D) Quantification of remaining MG-alkyne from the reaction where 1 mM MG-alkyne was treated with equal amounts of DJ-1 or BSA or with the vehicle for 24 h at 37 °C. (E) Representative rhodamine gel of 2 mg/mL BSA treated with 300 μM MG-alkyne as well as recombinant DJ-1 or vehicle for 24 h at 37 °C. (F) Quantification of labeling of 2 mg/mL BSA treated with 100 or 300 μM MG-alkyne as well as recombinant DJ-1 or vehicle for 24 h at 37 °C. Data plotted in (C–F) are mean ± SEM from n = 6 (C), 4 (D), or 3 (F) independent biological replicates. Statistical analyses are by ordinary one-way analysis of variance (ANOVA). *p < 0.05; **p < 0.01; ***p < 0.001; and ****p < 0.0001.

Figure 3

DJ-1 protects proteins from glycation with or without a physical interaction with target proteins. (A) Schematic depicting experiments where BSA and recombinant DJ-1 are incubated with MG-alkyne together or separated by a dialysis membrane alongside predicted interpretations. (B, C) Rhodamine gel (B) and quantification of labeling (C) of BSA incubated with 100 μM MG-alkyne in the presence or absence of recombinant DJ-1 with and without separation by a dialysis membrane. Data plotted in (C) are mean ± SEM from n = 3 independent biological replicates. Statistical analyses are by ordinary one-way analysis of variance (ANOVA). ****p < 0.0001.

Figure 4

Lactate product stereochemistry is consistent with a GSH-free glyoxalase mechanism. (A, B) Previously proposed DJ-1 glyoxalase (A) and deglycase (B) mechanisms and predicted product lactate stereochemistry. (C, D) LC-MS separation of DATAN-derivatized lactate standards (C) and DJ-1-catalyzed products (D). Both chromatograms show the extracted ion chromatogram (EIC) for the DATAN-derivatized product. (E–G) Quantification of the integrated peak area of l- and d-lactate formed by DJ-1, BSA, or PBS treated with MGO for 0–8 h at 37 °C. (H) Representative chromatograms of derivatized l- and d-lactate formed from PBS treated with 1 mM MGO for 0–8 h at 37 °C along with l- and d- lactate synthetic standards. Data plotted in (E–G) are mean ± SEM from n = 4 independent biological replicates normalized across all conditions.

Figure 5

Schematic depicting the stereospecific glyoxalase activity of DJ-1 supported by this study.