Myocardial ischaemia inhibits mitochondrial metabolism of 4-hydroxy-trans-2-nonenal

Abstract

Myocardial ischaemia is associated with the generation of lipid peroxidation products such as HNE (4-hydroxy-trans-2-nonenal); however, the processes that predispose the ischaemic heart to toxicity by HNE and related species are not well understood. In the present study, we examined HNE metabolism in isolated aerobic and ischaemic rat hearts. In aerobic hearts, the reagent [(3)H]HNE was glutathiolated, oxidized to [(3)H]4-hydroxynonenoic acid, and reduced to [(3)H]1,4-dihydroxynonene. In ischaemic hearts, [(3)H]4-hydroxynonenoic acid formation was inhibited and higher levels of [(3)H]1,4-dihydroxynonene and [(3)H]GS-HNE (glutathione conjugate of HNE) were generated. Metabolism of [(3)H]HNE to [(3)H]4-hydroxynonenoic acid was restored upon reperfusion. Reperfused hearts were more efficient at metabolizing HNE than non-ischaemic hearts. Ischaemia increased the myocardial levels of endogenous HNE and 1,4-dihydroxynonene, but not 4-hydroxynonenoic acid. Isolated cardiac mitochondria metabolized [(3)H]HNE primarily to [(3)H]4-hydroxynonenoic acid and minimally to [(3)H]1,4-dihydroxynonene and [(3)H]GS-HNE. Moreover, [(3)H]4-hydroxynonenoic acid was extruded from mitochondria, whereas other [(3)H]HNE metabolites were retained in the matrix. Mitochondria isolated from ischaemic hearts were found to contain 2-fold higher levels of protein-bound HNE than the cytosol, as well as increased [(3)H]GS-HNE and [(3)H]1,4-dihydroxynonene, but not [(3)H]4-hydroxynonenoic acid. Mitochondrial HNE oxidation was inhibited at an NAD(+)/NADH ratio of 0.4 (equivalent to the ischaemic heart) and restored at an NAD(+)/NADH ratio of 8.6 (equivalent to the reperfused heart). These results suggest that HNE metabolism is inhibited during myocardial ischaemia owing to NAD(+) depletion. This decrease in mitochondrial metabolism of lipid peroxidation products and the inability of the mitochondria to extrude HNE metabolites could contribute to myocardial ischaemia/reperfusion injury.

Fig 1

HPLC separation of HNE metabolites generated in the heart after a bolus injection of [³H]HNE. In each experiment, isolated adult rat hearts were equilibrated with aerobic buffer for 10 min and then 50 nmol of [³H]HNE in 0.1 ml of KH buffer was injected into the side port of the aortic cannula. The hearts were perfused with aerobic KH buffer (A), were subjected to 30 min of ischaemia (B), or were subjected to 30 min of ischaemia followed by 10 min of reperfusion (C). Peaks were assigned to GS–HNE/GS–DHN (peak I), DHN (peak II) and HNA (peak III), and unmetabolized HNE (peak IV). The identity of peak V is unknown. Inset (i) shows the ESI/mass spectrum of peak I. Ions with m/z 464 and 446 were assigned to MH⁺ and [MH-18]⁺ ions of GS–HNE. The peak with m/z 466.3 corresponds to the MH⁺ ion of GS–DHN. Inset (ii) shows the GC/mass spectrum of HPLC peak III after derivatization with BSTFA [bis-(trimethylsilyl)trifluoroacetamide]. The ion with a t_R of 9.55 min, corresponding to the t_R of HNA, displayed a fragmentation pattern (inset iii) similar to HNA. ESI–MS analysis of peak I from ischaemic heart tissue (B, inset iv) showed only a single peak with m/z 466.3 corresponding to GS–DHN. GC/MS analysis of peak II from the ischaemic heart showed an ion with a t_R of 8.08 min (inset v) and a fragmentation pattern (inset vi) similar to reagent DHN. ESI–MS analysis of peak I obtained from perfusates obtained within the first 3 min of reperfusion showed a single ion (m/z=466.2) corresponding to GS–DHN (C, inset vii). Profiles are representative of 4–8 rats per group (see Table 1).

Fig 2

Isolated rat hearts were perfused ex vivo for 40 min with aerobic buffer or were subjected to 30 min of ischaemia alone or were made ischaemic for 30 min followed by 5 min of reperfusion. After completion of the protocol, hearts were removed, homogenized in the presence of D₁₁-HNE, D₁₁-DHN and D₁₁-HNA. The samples were derivatized with PFBHA [O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine] and BSTFA. (Ai) shows the gas chromatogram of HNE and D₁₁-HNE in chemical ionization mode. (Aii) shows the separation of D₁₁-DHN and D₁₁-HNA in electron-impact ionization mode. DHN and HNA co-eluted with their D₁₁ analogues (results not shown). To determine the relative metabolite levels, the abundance of the major ions was quantified by single ion monitoring. For quantification, ions in which C₅H₁₁/C₅D₁₁ fragments were intact [M-HF for HNE (m/z=403) and D₁₁-HNE (m/z=414); M-CH₃ for DHN (m/z=287) and D₁₁-DHN (m/z=298); and M-CH₃ for HNA (m/z=301) and D₁₁-HNA (m/z=312)] were used. (B–D) show representative spectra illustrating the abundance of selected ions for quantification of HNE, DHN and HNA in hearts subjected to perfusion alone (P in B), ischaemia (I in C) or reperfusion after ischaemia (IR in D). (E) Group data showing myocardial abundance of HNA, DHN and HNE in perfused hearts and hearts subjected to ischaemia/reperfusion. Values are means±S.E.M. (n=four to six rats per group). *P<0.05 compared with hearts perfused with the aerobic buffer alone. P, perfused; I, ischaemia; IR, reperfusion after ischaemia.

Fig 3

Following 10 min of KH buffer perfusion without or with the ALDH inhibitor, benomyl (20 μM), a bolus of [³H]HNE (50 nmol in 0.1 ml KH buffer) was injected into the aorta. The coronary perfusate was collected for 3 min and tritiated metabolites recovered in the perfusate were resolved by HPLC. Results are shown as the percentage of total HNE metabolites recovered in the perfusate. Values are means±S.E.M. (n=six rats per group). *P<0.05 compared with hearts perfused without benomyl.

Fig 4

Isolated rat heart mitochondria were incubated with 15 μM [³H]HNE in 1.0 ml of respiration buffer for 30 min at 37 °C. Mitochondrial pellets (A and C) and the incubation medium (B and D) were separated by centrifugation, and radioactivity retained inside the mitochondrial pellet and extruded in the incubation medium was resolved by HPLC. Peak identification, as indicated, was based on the t_R of GS–HNE (I), DHN (II), HNA (III) and HNE (IV). Inset (i) is the ESI/mass spectrum of the HPLC peak eluting at the t_R identical with GS–HNE. The ions with m/z 464 and 446 were assigned to MH⁺ and [MH-18]⁺ ions of GS–HNE. The chemical identity of the major HPLC peak in (B) was established by GC/MS after derivatization with BSTFA. The metabolite eluted as a strong peak at 9.52 min (inset ii), similar to HNA. The fragmentation pattern (inset iii) of this peak was identical with HNA. To probe the role of ALDH in HNE oxidation, mitochondria were incubated with the ALDH inhibitor, benomyl (20 μM), for 30 min at 37 °C. The mitochondria were then incubated with 15 μM [³H]HNE, and metabolites in the pellet (C) and in the medium (D) were separated by HPLC. Expression, as shown by Western blotting (10 μg of protein/well), of ALDH2 in whole mitochondria (WM), the mitochondrial matrix (Mat) and the mitochondrial membrane (Mem) fractions of the rat heart, is shown in inset (iv). Cytochrome c oxidase subunit 1 (COX IV-1) was used as a mitochondrial membrane marker. The percentage change in HNE metabolite distribution in control and benomyl-treated mitochondria is shown in inset (v). Values are means±S.E.M. (n=four to six individual experiments). *P<0.05 compared with the control.

Fig 5

Isolated rat hearts were equilibrated with aerobic buffer for 10 min, and 50 nmol of [³H]HNE in 0.1 ml of KH buffer was injected into the side port of the aortic cannula, followed by 30 min of ischaemia. After the ischaemic protocol, the mitochondria were isolated as described in the Experimental section in the presence of 2 μM cyclosporin A. Peaks were assigned to GS-HNE/GS-DHN (peak I), DHN (peak II), HNA (peak III) and unmetabolized HNE (peak IV). The identity of peak V is unknown. HNE metabolites from the cytosol (A, inset i) and mitochondria (B, inset ii) were quantified and are shown as the percentage of total recovered HNE metabolites. Inset (iii) shows the protein-bound [³H]HNE in cytosolic and mitochondrial fractions. Values represent the means±S.E.M. (n=four rat hearts).

Fig 6

HPLC profiles obtained from mitochondrial lysates incubated with [³H]HNE in the presence of the indicated pyridine nucleotides. Isolated rat heart mitochondria were solubilized in respiration buffer, containing 1% Nonidet P40, and 2 mg of mitochondrial protein was incubated with 15 μM [³H]HNE in the presence of (A) 1.0 mM NAD⁺ or (B) 2 mM NADP⁺ for 30 min at 37 °C. Protein was then precipitated with trichloroacetic acid and metabolites in the supernatant were separated by HPLC and quantified by scintillation counting.

Fig 7

(A) Hearts were subjected to aerobic perfusion for 40 min, global ischaemia for 30 min, or global ischaemia for 30 min followed by 5 or 60 min reperfusion, and their NAD⁺ and NADH contents were measured. Mean values of the ratio are shown below the histograms. The inset shows Western blots from perfused (P), ischaemic (I) and reperfused (I/R) heart lysates developed using an anti-ALDH2 antibody. Values are means±S.E.M. (n=three individual rat hearts). *P<0.05 compared with the 40 min perfusion control (40′P); #P<0.05 compared with the 30 min ischaemia sample (30′I). (B) Myocardial lysates were incubated with the indicated ratio of NAD⁺/NADH and [³H]HNE. The [³H]HNA generated was separated by HPLC and quantified by scintillation counting. Values are means±S.E.M. (n=three hearts per group). *P<0.01 compared with 11.2 NAD⁺/NADH ratio; #P<0.01 compared with 0.4 NAD⁺/NADH ratio.

Fig 8

HPLC profiles of mitochondria isolated from (A) perfused and (B) ischaemic hearts. After isolation, mitochondria were resuspended in aerobic respiration buffer (2.5 mg/ml) and exposed to 15 mM [³H]HNE for 30 min at 37 °C. After incubation, mitochondria were sedimented by centrifugation, and metabolites in the supernatant were separated by HPLC and quantified by scintillation counting. Inset (i) shows HNA/HNE ratios (means±S.E.M.) of perfused and ischaemic rat hearts (n=3).