High glucose induces mitochondrial dysfunction independently of protein O-GlcNAcylation

Abstract

Diabetes is characterized by hyperglycaemia and perturbations in intermediary metabolism. In particular, diabetes can augment flux through accessory pathways of glucose metabolism, such as the hexosamine biosynthetic pathway (HBP), which produces the sugar donor for the β-O-linked-N-acetylglucosamine (O-GlcNAc) post-translational modification of proteins. Diabetes also promotes mitochondrial dysfunction. Nevertheless, the relationships among diabetes, hyperglycaemia, mitochondrial dysfunction and O-GlcNAc modifications remain unclear. In the present study, we tested whether high-glucose-induced increases in O-GlcNAc modifications directly regulate mitochondrial function in isolated cardiomyocytes. Augmentation of O-GlcNAcylation with high glucose (33 mM) was associated with diminished basal and maximal cardiomyocyte respiration, a decreased mitochondrial reserve capacity and lower Complex II-dependent respiration (P<0.05); however, pharmacological or genetic modulation of O-GlcNAc modifications under normal or high glucose conditions showed few significant effects on mitochondrial respiration, suggesting that O-GlcNAc does not play a major role in regulating cardiomyocyte mitochondrial function. Furthermore, an osmotic control recapitulated high-glucose-induced changes to mitochondrial metabolism (P<0.05) without increasing O-GlcNAcylation. Thus, increased O-GlcNAcylation is neither sufficient nor necessary for high-glucose-induced suppression of mitochondrial metabolism in isolated cardiomyocytes.

Figure 1. High glucose depresses the bioenergetic reserve capacity of cardiomyocytes

(A) Immunoblot of whole-cell protein O-GlcNAcylation following 0, 5, 10, 20 and 33 mM glucose treatment. An osmotic control (Osm, 5 mM glucose + 28 mM mannitol) was also used. (B) Densitometric measurement of O-GlcNAcylation in relation to 5 mM control demonstrated a significant induction of protein O-GlcNAcylation following 33 mM glucose. (C) Immunoblotting for mitochondrial protein O-GlcNAcylation. NRCMs were fractionated to isolate mitochondrial protein following treatment with 5 mM glucose, 33 mM glucose and the osmotic control. (D) Representative XF assay and diagram of how mitochondrial measurements were calculated. (E) Mitochondrial function assay following 48 h of high glucose treatment. From (E) parameters of mitochondrial function were assessed (as described in D): (F) basal respiration, (G) maximal respiration, (H) reserve capacity, (I) ATP-linked respiration, and (J) proton leak. As indicated in bars, n = 7 independent experiments; ^*P < 0.05, compared with 5 mM.

Figure 2. Overexpression of OGT or OGA does not affect bioenergetic reserve

NRCMs were subjected to adenoviral overexpression of OGT and OGA. (A) Immunoblot for OGT and OGA protein expression following adenoviral treatment. (B) Immunoblot for protein O-GlcNAcylation demonstrated an induction of O-GlcNAc in response to Ad-OGT and a reduction in response to Ad-OGA. (C) Mitochondrial function assay of NRCMs 48 h post-transfection. Parameters of mitochondrial function were calculated from (C), including: (D) basal respiration, (E) maximal respiration, (F) reserve capacity, (G) ATP-linked respiration, and (H) proton leak. As indicated in bars, n = 5 independent experiments; ^*P < 0.05, compared with Ad-Null.

Figure 3. Inhibition of OGA has negligible effects on mitochondrial function

NRCMs were treated with TMG (1 µM) for 48 h to induce protein O-GlcNAcylation. (A) Immunoblot for whole-cell protein O-GlcNAcylation. (B) Densitometric measurement of O-GlcNAcylation. (C) Immunoblot for mitochondrial protein O-GlcNAcylation following treatment with 1 µM TMG. (D) Mitochondrial function assay of NRCMs following 48 h treatment with TMG. Parameters of mitochondrial function were measured from (D), including: (E) basal respiration, (F) maximal respiration, (G) reserve capacity, (H) ATP-linked respiration, and (I) proton leak. As indicated in bars, n = 6 independent experiments; ^*P < 0.05.

Figure 4. High glucose depresses Complex II-dependent State 3 and 4_o respiration

NRCMs were treated for 48 h with 5 or 33 mM glucose or an osmotic control. During XF analysis, NRCMs were permeabilized and provided with succinate (to support Complex II-dependent respiration) as well as Rot (to inhibit Complex I activity). (A) XF assay of Complex II-dependent respiration: following three baseline OCR measurements in MAS buffer, the permeabilization agent, saponin, and succinate + Rot were injected. After two measurements, oligomycin (Oligo), then AA + Rot (AA–Rot) were injected sequentially, with measurements recorded after each injection. (B) State 3 OCR: The AA + Rot rate was subtracted from the succinate-stimulated rate to determine the State 3 rates. (C) State 4_o OCR: The AA + Rot rate was subtracted from the oligomycin rate to obtain State 4_o rates. (D) RCR: State 3/State 4_o. As indicated in bars, n = 6 independent experiments; ^*P < 0.05, compared with 5 mM glucose.

Figure 5. Overexpression of OGA does not rescue high-glucose-induced suppression of bioenergetic reserve

NRCMs were transduced with virus to overexpress OGA prior to treatment with hyperglycaemia. Ad-Null was used as a vector control. (A) Immunoblot for whole-cell OGA protein expression following viral transduction. (B) Whole-cell protein O-GlcNAcylation levels following adenoviral transduction. (C) Immunoblot for mitochondrial OGA following adenoviral overexpression. (D) Immunoblot for mitochondrial protein O-GlcNAcylation following adenoviral overexpression of OGA. (E) Mitochondrial function assay following 48 h of hyperglycaemic treatment. From assay in (E) parameters of mitochondrial function were measured: (F) basal respiration, (G) maximal respiration, (H) reserve capacity, (I) ATP-linked respiration, and (J) proton leak. As indicated in bars, n = 5 independent experiments; ^*P < 0.05, compared with 5 mM glucose + Ad-Null.

Figure 6. Inhibition of OGA increases Complex II-dependent State 3 respiration

NRCMs were treated for 24 h with 1 µM TMG to induce protein O-GlcNAcylation. During XF analysis NRCMs were permeabilized and provided with succinate (to support Complex II-dependent respiration), as well as Rot (to inhibit Complex I activity). (A) XF assay of Complex II-dependent respiration: following three baseline OCR measurements in MAS buffer, the permeabilization agent, saponin, and succinate + Rot were injected. After two measurements, oligomycin (Oligo) and AA + Rot (AA–Rot) were injected sequentially, with two measurements recorded after each injection. (B) State 3 OCR: the AA + Rot rate was subtracted from the succinate-stimulated rate to determine the State 3 rates. (C) State 4_o OCR: the AA + Rot rate was subtracted from the oligomycin rate to obtain State 4_o rates. (D) RCR: State 3/State 4_o. As indicated in bars, n = 7 independent experiments; ^*P < 0.05.