Hyperglycemia impairs proteasome function by methylglyoxal

Abstract

Objective: The ubiquitin-proteasome system is the main degradation machinery for intracellularly altered proteins. Hyperglycemia has been shown to increase intracellular levels of the reactive dicarbonyl methylglyoxal (MGO) in cells damaged by diabetes, resulting in modification of proteins and alterations of their function. In this study, the influence of MGO-derived advanced glycation end product (AGE) formation on the activity of the proteasome was investigated in vitro and in vivo.

Research design and methods: MGO-derived AGE modification of proteasome subunits was analyzed by mass spectrometry, immunoprecipitation, and Western blots. Proteasome activity was analyzed using proteasome-specific fluorogenic substrates. Experimental models included bovine retinal endothelial cells, diabetic Ins2(Akita) mice, glyoxalase 1 (GLO1) knockdown mice, and streptozotocin (STZ)-injected diabetic mice.

Results: In vitro incubation with MGO caused adduct formation on several 20S proteasomal subunit proteins. In cultured endothelial cells, the expression level of the catalytic 20S proteasome subunit was not altered but proteasomal chymotrypsin-like activity was significantly reduced. In contrast, levels of regulatory 19S proteasomal proteins were decreased. In diabetic Ins2(Akita), STZ diabetic, and nondiabetic and diabetic G101 knockdown mice, chymotrypsin-like activity was also reduced and MGO modification of the 20S-beta2 subunit was increased.

Conclusions: Hyperglycemia-induced formation of MGO covalently modifies the 20S proteasome, decreasing its activity in the diabetic kidney and reducing the polyubiquitin receptor 19S-S5a. The results indicate a new link between hyperglycemia and impairment of cell functions.

FIG. 1.

Proteasome activity in endothelial cells under chronic hyperglycemia. Proteasome activity and 20S proteasome expression levels were analyzed in immortalized human endothelial Eahy.926 cells and in primary bovine retinal endothelial cells (BRECs) following incubation with low- and high-glucose medium for 14 days. Proteasomal trypsin-like (A) and caspase-like (B) activities following incubation with high-glucose medium were not significantly changed in comparison with those following low-glucose medium, whereas the proteasomal chymotrypsin-like activity was significantly reduced (C). The level of the catalytic 20S proteasome remained constant, while the 19S regulatory complex expression was decreased, as shown by Western blot (D). Data represent means ± SD from at least five separate experiments. *P ≤ 0.05. RFU, relative fluorescence unit.

FIG. 2.

Resistance of glycated albumin against proteasomal degradation in vitro. AGE-BSA (10 μg) was incubated with isolated 20S proteasome to examine the influence on proteasomal activity. The degree in carboxy-methyl-lysine modification of BSA (as demonstrated by Western blotting) did not significantly affect the chymotrypsin-like activity of the 20S proteasome. Ponceau S staining is shown as a loading control. Data represent means ± SD from at least three separate experiments. RFU, relative fluorescence unit.

FIG. 3.

Inhibition of proteasomal chymotrypsin-like activity in vitro by reactive carbonyl compounds. Reactive carbonyl compounds were incubated with the 20S proteasome and the enzymatic activities determined. Trypsin-like (A) and caspase-like (B) activities were not significantly changed, whereas the chymotrypsin-like activity was significantly reduced (C). Data represent means ± SD from at least three separate experiments. *P ≤ 0.01. GA, glyceraldehyde; GO, glyoxal; RFU, relative fluorescence unit.

FIG. 4.

Inhibition of proteasomal chymotrypsin-like activity in endothelial cells and accumulation of ubiquitinated proteins. Immortalized human endothelial Eahy.926 cells and primary bovine retinal endothelial cells were each incubated for 12 h with 500 μmol/l reactive carbonyl compounds. Trypsin-like (A) and caspase-like (B) activities were not changed, whereas the chymotrypsin-like activity was significantly reduced after incubation with each of the three reactive carbonyl compounds (C). Incubation with reactive carbonyl compounds exhibited accumulation of ubiquitinated proteins in a dose-dependent manner (D). Data represent means ± SD from at least three separate experiments. *P ≤ 0.05. GA, glyceraldehyde; GO, glyoxal; RFU, relative fluorescence unit.

FIG. 5.

Decreased proteasomal activity in organs from diabetic Ins2^Akita mice. Proteasomal activity and proteasome expression was determined in kidney, heart, and aorta from diabetic Ins2^Akita mice (n = 6) and control wild-type mice (n = 6). The three proteolytic activities of the proteasome were measured with fluorogenic substrates. Trypsin-like (A) and caspase-like (B) activities were significantly reduced in kidneys from diabetic mice. The chymotrypsin-like activity (C) was significantly reduced in kidney and aorta, whereas in the heart no significant changes were seen. The 20S proteasome protein levels were not significantly changed in any organs, as shown by Western blotting (D and E). Data represent means ± SD from at least three separate experiments. *P ≤ 0.05.

FIG. 6.

Increased MGO- and ubiquitin-modified proteins with impaired proteasomal activity in GLO1 knockdown and diabetic mice. Kidney lysate protein (40 μg) from wild-type (WT), GLO1-knockdown (GLO1-KD), and STZ diabetic and GLO1-knockdown diabetic (GLO1-KD-STZ) mice were analyzed by Western blotting. GLO1-knockdown mice exhibited a significant increase in MGO- and ubiquitin-modified proteins in kidney homogenates compared with levels in wild-type mice (A and B). GLO1-knockdown and STZ-injected mice exhibited significantly reduced proteasomal chymotrypsin-like activity in kidney homogenates compared with that in wild-type mice (C). The protein band densities were scanned and quantified using the Odyssey imaging system. Data are expressed as means ± SD. *P < 0.05 vs. wild type; n = 4. RFU, relative fluorescence unit. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 7.

Decreased 19S-S5a protein levels and MGO-modified 20S-β2 proteasome subunit in GLO1-knockdown (GL01-KD) and STZ diabetic mice. Kidney lysate protein (40 μg) from indicated mice were analyzed on NUPAGE 4–12% Bis-Tris gels, transferred to PVDF membranes, and blotted by 19S-S5a antibody (A). The lysates from indicated mice kidney tissues were immunoprecipitated with 20S-β2 antibody and analyzed by Western blotting, and 10% whole lysates were used for β-actin detection as input. MGO modification of the 20S proteasome β2-subunit was detected in kidneys from GLO1-knockdown and STZ diabetic mice (B). The protein band densities were scanned and quantified using the Odyssey imaging system. Data are expressed as means ± SD. *P < 0.05 vs. wild type; n = 4. RFU, relative fluorescence unit. (A high-quality digital representation of this figure is available in the online issue.)