Cardiac mitochondrial NADP+-isocitrate dehydrogenase is inactivated through 4-hydroxynonenal adduct formation: an event that precedes hypertrophy development.

Mitochondrial NADP+-isocitrate dehydrogenase activity is crucial for cardiomyocyte energy and redox status, but much remains to be learned about its role and regulation. We obtained data in spontaneously hypertensive rat hearts that indicated a partial inactivation of this enzyme before hypertrophy development. We tested the hypothesis that cardiac mitochondrial NADP+-isocitrate dehydrogenase is a target for modification by the lipid peroxidation product 4-hydroxynonenal, an aldehyde that reacts readily with protein sulfhydryl and amino groups. This hypothesis is supported by the following in vitro and in vivo evidence. In isolated rat heart mitochondria, enzyme inactivation occurred within a few minutes upon incubation with 4-hydroxynonenal and was paralleled by 4-hydroxynonenal/NADP+-isocitrate dehydrogenase adduct formation. Enzyme inactivation was prevented by the addition of its substrate isocitrate or a thiol, cysteine or glutathione, suggesting that 4-hydroxynonenal binds to a cysteine residue near the substrate's binding site. Using an immunoprecipitation approach, we demonstrated the formation of 4-hydroxynonenal/NADP+-isocitrate dehydrogenase adducts in the heart and their increased level (210%) in 7-week-old spontaneously hypertensive rats compared with control Wistar Kyoto rats. To the best of our knowledge, this is the first study to demonstrate that mitochondrial NADP+-isocitrate dehydrogenase is a target for inactivation by 4-hydroxynonenal binding. Furthermore, the pathophysiological significance of our finding is supported by in vivo evidence. Taken altogether, our results have implications that extend beyond mitochondrial NADP+-isocitrate dehydrogenase. Indeed, they emphasize the implication of post-translational modifications of mitochondrial metabolic enzymes by 4-hydroxynonenal in the early oxidative stress-related pathophysiological events linked to cardiac hypertrophy development.

Mitochondrial NADP ؉ -isocitrate dehydrogenase activity is crucial for cardiomyocyte energy and redox status, but much remains to be learned about its role and regulation. We obtained data in spontaneously hypertensive rat hearts that indicated a partial inactivation of this enzyme before hypertrophy development. We tested the hypothesis that cardiac mitochondrial NADP ؉ -isocitrate dehydrogenase is a target for modification by the lipid peroxidation product 4-hydroxynonenal, an aldehyde that reacts readily with protein sulfhydryl and amino groups. This hypothesis is supported by the following in vitro and in vivo evidence. In isolated rat heart mitochondria, enzyme inactivation occurred within a few minutes upon incubation with 4-hydroxynonenal and was paralleled by 4-hydroxynonenal/ NADP ؉ -isocitrate dehydrogenase adduct formation. Enzyme inactivation was prevented by the addition of its substrate isocitrate or a thiol, cysteine or glutathione, suggesting that 4-hydroxynonenal binds to a cysteine residue near the substrate's binding site. Using an immunoprecipitation approach, we demonstrated the formation of 4-hydroxynonenal/NADP ؉ -isocitrate dehydrogenase adducts in the heart and their increased level (210%) in 7-week-old spontaneously hypertensive rats compared with control Wistar Kyoto rats. To the best of our knowledge, this is the first study to demonstrate that mitochondrial NADP ؉ -isocitrate dehydrogenase is a target for inactivation by 4-hydroxynonenal binding. Furthermore, the pathophysiological significance of our finding is supported by in vivo evidence. Taken altogether, our results have implications that extend beyond mitochondrial NADP ؉ -isocitrate dehydrogenase. Indeed, they emphasize the implication of post-translational modifications of mitochondrial metabolic enzymes by 4-hydroxynonenal in the early oxidative stress-related pathophysiological events linked to cardiac hypertrophy development.
Mitochondrial dysfunction is considered to play a key role in the pathogenesis of cardiac hypertrophy and failure. Chronic alterations of fuel metabolism and oxidative stress status are factors that could impair the capacity of the mitochondria to fulfil their crucial role in energy production (1) and thereby contribute to the activation of signaling pathways governing cell death by apoptosis and/or necrosis (2,3). Although the role of mitochondrial energy fuel deficits or of oxidative stress are often investigated separately, accumulating evidence indicates that these two factors are linked. For example, several metabolic enzymes can be inactivated through post-translational modifications by oxidative stress-related molecular components (4 -11). The latter molecules include oxygen-and nitrogen-derived reactive species or aldehydes produced from lipid peroxidation. Of specific interest to this study is 4-hydroxynonenal (HNE), 1 the major ␣-␤-unsaturated aldehyde formed from peroxidation of both -3 and -6 polyunsaturated fatty acids, whose formation is enhanced in hypertrophied and ischemic/ reperfused hearts (12)(13)(14). Whether aldehydes are "second toxic messengers," as first hypothesized by Esterbauer et al. (15) or merely represent markers of tissue damage (16), has long been a controversial subject. However, the situation has changed recently (17) as growing evidence indicates that the formation of HNE-protein adducts is a key event in many free radical-related effects in the heart. These effects include (i) disturbances of myocardial contraction and rhythm (4,8), (ii) modulation of signal transduction leading to subsequent adaptive responses or apoptosis (13, 18 -20), and (iii) alterations of mitochondrial energy metabolism (7,21).
Although research on mitochondrial enzyme alterations by free radicals and aldehydes often focuses on electron transport chain components (4 -6), the detrimental consequences of inactivation of citric acid cycle (CAC)-related enzymes for NAD(P)H-linked respiration cannot be overlooked (6 -11, 21). Among these enzymes is NADP ϩ -dependent isocitrate dehydrogenase (NADP ϩ -ICDH), whose susceptibility to inactivation by nitric oxide has been demonstrated only recently (22). Mammalian tissues contain two classes of NADP ϩ -ICDH isoenzymes, mitochondrial NADP ϩ -ICDH (mNADP ϩ -ICDH) and cytosolic NADP ϩ -ICDH, which are encoded by two distinct genes (23,24). Both isoenzymes catalyze the reversible interconversion between isocitrate and ␣-ketoglutarate and have no known allosteric effector. It is unlike the bacterial enzyme, which is regulated by phosphorylation/dephosphorylation (25). The mNADP ϩ -ICDH isoform predominates (Ͼ95%) in the heart, where it is confined to cardiomyocytes (23,26,27). mNADP ϩ -ICDH exists as a homodimer of 413 amino acids per subunit with a molecular mass of 47 kDa and a pI of 9.0 (cf. Swiss-Prot accession no. P33198). It shows a high degree of homology in protein structure and cDNA sequences between species (Ͼ90% between mice, pigs, bovines, and humans (23,24)).
However, much remains to be learned about the role and regulation of mNADP ϩ -ICDH in mammalian cells, and especially in the heart where it presents its highest activity and expression (23,24). In 1994, Sazanov and Jackson (28) proposed that reverse flux through mNADP ϩ -ICDH, which generates isocitrate and NADP ϩ , would be part of a substrate cycle to allow CAC flux to adapt more precisely to changes in energy demand. This cycle also includes the participation of the CAC enzyme NAD ϩ -ICDH and H ϩ -transhydrogenases to regenerate ␣-ketoglutarate and NADPH, respectively. By using 13 C methods, we obtained data in the intact perfused rat heart that supported the Sazanov and Jackson hypothesis (29). This interpretation is challenged by the data of Jo et al. (26) in NIH3T3 cells, which support an antioxidant role of mNADP ϩ -ICDH through NADPH regeneration. To reconcile all these data, we proposed (29) that in the heart, a tissue where the mitochondrial NADPH/NADP ϩ ratio is high (Ͼ50 compared with Ͻ1 in NIH3T3) and which could inhibit CAC flux (crucial for contraction), the metabolic role of mNADP ϩ -ICDH could differ from that of other cell types.
While investigating the potential redox regulation of mNADP ϩ -ICDH in the heart, we obtained data indicating its inactivation before hypertrophy development. Our study model was the spontaneously hypertensive rat (SHR); Wistar-Kyoto (WKY) rats served as controls. Prompted by the potential pathophysiological significance of mNADP ϩ -ICDH inactivation, we tested the hypothesis that mNADP ϩ -ICDH is a target for binding by the lipid peroxidation product HNE.

MATERIALS AND METHODS
Animals-Animal experimentation was approved by the local ethics committee in compliance with the guidelines of the Canadian Council on Animal Care. Male SHR, WKY, and Wistar rats were purchased from Charles River Laboratories (St-Constant, QC, Canada). Prior to the day of the experiments, the rats were housed for Ն7 days in a 12-h light/12-h dark cycle facility with unlimited access to water and standard chow. SHR and control WKY rats were killed at 7 weeks of age. Body weights (n ϭ 6) at the time of death were, respectively, 210 Ϯ 10 g for SHR and 215 Ϯ 15 g for WKY rats. The hearts were isolated under sodium pentobarbital anesthesia (65 mg ϫ kg Ϫ1 , intraperitoneal; MTC Pharmaceuticals), cannulated rapidly, flushed with a cold saline solution, freeze-clamped, and stored in liquid nitrogen until further analyses. Wistar rats (320 -340 g) served for the isolation of heart mitochondria, as described below.
Enzyme Activity-Total mNADP ϩ -ICDH activity was assessed with a commercial kit (Sigma Diagnostics) in 100 mg of powdered tissues that were homogenized on ice in 1 ml of buffer containing 180 mM KCl, 5 mM MOPS, and 2 mM EDTA, pH 7 and centrifuged for 10 min at 800 ϫ g at 4°C. Supernatants were used for enzyme assays after 10-min centrifugation at 6,000 ϫ g at 4°C. Protein levels were measured with a kit (Bio-Rad) with bovine serum albumin (BSA) (Sigma) serving as the standard. Activities are expressed in units/mg of proteins, where 1 unit is defined as the amount of enzyme catalyzing the conversion of 1 mol substrate/min at 37°C.
Western Blotting-Proteins from roughly 100 mg of powdered heart tissues were extracted (as described above) with cold extraction buffer containing 1 mM phenylmethylsulphonyl fluoride, 10 M each of aprotinin, leupeptin, and pepstatin, and 1 mM orthovanadate. After protein determination, 20 g of tissue protein extract were subjected to discontinuous 4 -12% SDS-PAGE under reducing conditions and transferred electrophoretically onto nitrocellulose membrane (Bio-Rad), according to the method of Towbin et al. (30). Membranes were immersed overnight at 4°C in blocking buffer (Roche Applied Science), diluted 10-fold in Tris-buffered saline (TBS: 20 mM Tris HCl, 150 mM NaCl, pH 7.5), and incubated again overnight at 4°C in blocking buffer diluted 10-fold in 0.05% Tween-TBS containing rabbit anti-mouse mNADP ϩ -ICDH (1:1,000; a generous gift of Dr. T. L. Huh, Kyungpook National University, Taegu, South Korea). Then, the membranes were washed with Tween-TBS and incubated for 1 h with goat anti-rabbit IgG-horseradish peroxidase conjugate (1:25,000; The Jackson Laboratory). Immunoreactive proteins were visualized with BM chemiluminescence blotting substrate (Roche Applied Science) diluted 1:100 in luminescence substrate solution according to the manufacturer's specifications. The membranes were exposed to Kodak X-AR5 film (Eastman Kodak), and the autoradiographs were scanned with the ChemiImager 4000 system (Alpha Innotech). Quantification was achieved by optical densitometry and expressed as a percentage of control values.
Semi-quantitative Reverse Transcription-PCR (RT-PCR)-Total RNAs were extracted from frozen powdered heart tissues with TRIzol (Invitrogen) according to the manufacturer's specifications. Total RNAs were then treated with RNase-free DNase I following a standard protocol (Invitrogen). The integrity and quantity of the purified RNAs were controlled by formaldehyde denaturing agarose gel electrophoresis and by measurement of the A 260 /A 280 ratio. RT-PCR assays were carried out with a programmable thermal controller (PTC-100, MJ Research Inc.). Two g of total RNAs were reverse-transcribed into complementary DNA (cDNA) by incubation at 42°C for 1 h with a 20-l first-strand cDNA synthesis mixture containing 10 mM dithiothreitol, 1 g of random hexamers, 0.5 mM deoxynucleoside triphosphate, 10 units/l reverse transcriptase (Moloney murine leukemia virus), and 2 units/l of RNase inhibitor. cDNAs encoding mNADP ϩ -ICDH, atrial natriuretic factor (ANF), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were amplified as follows. 5 l of first-strand cDNA were added to the PCR mixture containing 50 mM KCl, 10 mM Tris, pH 9, 1.5 mM MgCl 2 , 0.2 mM deoxynucleoside triphosphate, 5 units of TaqDNA polymerase, 10 Ci [␣-35 S]dCTP, and 100 ng of each of the sense and anti-sense-specific primers: mouse mNADP ϩ -ICDH sense, 5Ј-GGCTG-TGAGCTCGCTCTGCAGAGC-3Ј, and antisense, 5Ј-TCTTGGTGCTCA-AGTAGAGCGGC-3Ј; rat ANF sense, 5Ј-ATCTGATGGATTTCAAGAA-CC-3Ј, and antisense, 5Ј-CTCCAATCCTGTCAATCCTACC-3Ј; and rat GAPDH sense, 5Ј-AGTGGACATTGTTGCCATCAACGACC-3Ј, and antisense, 5Ј-GTCATGAGCCCTTCCACGATGCCAA-3Ј. Briefly, cDNA was amplified for 20 or 25 cycles by incubation at 95°C for 1 min, 60°C for 30 s, and 72°C for 1 min, with a final incubation at 72°C for 5 min. The number of cycles for amplification was selected on the basis of linearity of the PCR product (data not shown). Negative controls for RT-PCR were also run in parallel (data not shown). PCR products (10 l/100 l) were separated on 1.8% agarose gel and transferred electrophoretically onto nylon membranes (Hybond-Nϩ, Amersham Biosciences) in 10 mM sodium acetate buffer, pH 7.8, 20 mM Tris, and 0.5 mM EDTA overnight at 4°C. The membranes were then exposed to ultraviolet light to cross-link labeled cDNA, dried, and exposed again to Kodak films. The autoradiographs were scanned by ChemiImager. cDNA values for mNADP ϩ -ICDH and ANF were normalized to that of GAPDH and expressed as a percentage of control values.
Isolation and Incubation of Mitochondria-Intact rat heart mitochondria were used to evaluate the effect of the lipid peroxidation product HNE on mNADP ϩ -ICDH activity. Mitochondria were isolated from Wistar rats as described by Nulton-Persson and Szweda (10). Briefly, isolated hearts were minced in ice-cold buffer (20 ml/g of tissue) containing 180 mM KCl, 5.0 mM MOPS, and 2.0 mM EGTA at pH 7.4 and homogenized in a Dounce glass homogenizer. Unbroken cells and nuclei were removed by 10-min centrifugation at 800 ϫ g. The supernatant was centrifuged at 10,000 ϫ g for 20 min. The pellet containing the mitochondria was re-suspended in 20 ml of the same buffer and centrifuged again at 10,000 ϫ g for 20 min before re-suspension in 200 l of buffer. The mitochondrial suspension was kept at 4°C and used within 1 h. Protein levels were measured by kit (Bio-Rad). For all studies, total mNADP ϩ -ICDH activity was assessed as described above after incubation of mitochondria under the following conditions. (i) For dose-response studies, a volume of mitochondrial suspension equivalent to 40 g of proteins was incubated for 10 min at 37°C in the presence of increasing concentrations (0 -100 M) of HNE (Cayman Chemical). (ii) For time-course studies, mitochondria were incubated for increasing periods of time (5-20 min) in the presence of 20 M HNE. (iii) To test the protective effect of antioxidants against mNADP ϩ -ICDH inhibition by HNE, we chose a thiol compound, either 0.5 mM cysteine or reduced glutathione (GSH), because of the known high reactivity of HNE to protein thiol groups. Cysteine or GSH was added to the mitochondrial incubation mixture 2 min prior to the addition of HNE. (iv) To assess the possibility of cofactor and substrate site modifications by HNE, mitochondria were incubated for 2 min with 1 mM NADP ϩ and/or 1 mM isocitrate before the addition of 20 M HNE for 10 min. Under all conditions, the incubation was stopped by the addition of 0.5 mM cysteine.
Immunoprecipitation-100 g of tissue protein extract, incubated in the presence of 10, 20, or 50 M HNE for 10 min, were pre-incubated with protein G resin (1:10 of a 50% slurry in water; Santa Cruz Biotechnology, Inc.) at 4°C with constant gentle shaking and then subjected to immunoprecipitation with rabbit anti-mouse mNADP ϩ -ICDH antibody (1:1,000) in radioimmune precipitation assay buffer (150 mM NaCl, 1 mM EDTA, 40 mM Tris (pH 7.6), 1% Triton X-100, 1 mM phenylmethylsulphonyl fluoride, 10 M each of aprotinin, leupeptin, pepstatin, and 1 mM orthovanadate). For the in vivo detection of HNE/ mNADP ϩ -ICDH adducts, 500 g of tissue protein extract from 7-weekold SHR and WKY rat hearts were subjected to immunoprecipitation, as described above, by using radioimmune precipitation assay buffer containing 10% glycerol. After overnight incubation at 4°C with constant gentle shaking, protein G resin was added, and the mixture was incubated for 2 h at room temperature. After centrifugation, the resin was washed three times with 1 ml of radioimmune precipitation assay buffer. Proteins were removed from the resin by the addition of 100 l of undiluted SDS-PAGE loading buffer. A 20-l aliquot of the immunoprecipitated proteins was heated at 95°C for 3 min prior to Western blot analysis, using rabbit anti-HNE antibody (1:1,000 dilution; Calbiochem) as the primary antibody.
Tissue Levels of HNE-protein Adducts-Total tissue levels of HNEprotein adducts were assessed by enzyme-linked immunosorbent assay, as described by Toyokuni et al. (31) with some modifications. To prepare the HNE-BSA standard, BSA (Sigma) was dissolved in 50 mM phosphate-buffered saline (PBS; 5 mg BSA/ml), pH 7.2, and incubated at 37°C for 1 h in the presence of 0.1 mM HNE. The HNE-BSA solution was then filtered and washed four times with PBS using 10,000 MWCO centrifugal filters (Millipore). Immulon II microtiter plates (VWR International) were precoated overnight at 4°C with 100 l of HNE-BSA standard or sample (tissue protein extracts concentrated 5-fold), washed four times with PBS, and then incubated with 200 l of blocking buffer (5% BSA and 5% fetal bovine serum) for 4 h at room temperature. The blocking buffer was discarded, and the plates were incubated overnight at 4°C with 200 l of primary antibody rabbit anti-HNE (1:1,000 dilution in 0.05% Tween-PBS). After washing four times with Tween-PBS, 200 l of the goat anti-rabbit IgG-horseradish peroxidase conjugate (1:50,000; The Jackson Laboratory) were added, and the plates were incubated for 2 h at room temperature. Detection was carried out with a Sigma Fast TM POD kit (Sigma). After a 10-to 20-min incubation at room temperature, the reaction was stopped with 50 l of 2N H 2 SO 4 , and absorbance at 450 nm was measured with a microplate reader. A linear response in the range up to 5 g/ml of HNE-modified BSA standard was observed (slope ϭ 0.30 Ϯ 0.03; coefficient of correlation ϭ 0.99; p Ͻ 0.0001).
Statistical Analyses-Data are expressed as means Ϯ S.E. Statistical significance at p Ͻ 0.05 of differences between mean values was assessed by unpaired t test or one-way analysis of variance followed by a Bonferroni multiple-comparison post-test.

Myocardial mNADP ϩ -ICDH Activity Is Decreased in 7-
weekold SHR-Cardiac hypertrophy development in SHR occurs between 9 and 12 weeks of age (32). Accordingly, compared with control age-matched WKY rats, 7-week-old SHR did not show any increase in myocardial mRNA levels of ANF, a marker of hypertrophy (data not presented). As depicted in Fig.  1, myocardial mNADP ϩ -ICDH activity was 25% lower in 7-week-old SHR hearts than in WKY rats, but protein (Fig. 1B) and mRNA (Fig. 1C) levels were comparable in both rat species. Taken together, these results indicated the existence of a posttranslational mechanism decreasing cardiac mNADP ϩ -ICDH activity before hypertrophy development. mNADP ϩ -ICDH Is a Target for HNE Binding-We tested the hypothesis of potential mNADP ϩ -ICDH inactivation by HNE, a product of free radical-induced lipid peroxidation. Our choice of HNE was based on the following reasoning. (i) The progression of cardiac disease is associated with increased oxidative stress and lipid peroxidation (for recent reviews, see Refs. 33 and 34). (ii) Because of its high chemical reactivity, HNE reacts readily with the amino acid residues of lysine and histidine, sulfhydryl groups of cysteine, and lipoate of proteins to form stable adducts (13, 15, 18 -20). (iii) mNADP ϩ -ICDH contains six cysteine residues and is particularly enriched in histidine and lysine residues, as evidenced by its pI of 9.0 (22,35).
In Vitro Evidence-To evaluate the effect of HNE on mNADP ϩ -ICDH activity, we conducted dose-response and time-course studies in isolated heart mitochondria. This would exclude any potential contamination by the low level of cytosolic NADP ϩ -ICDH. As shown in Fig. 2A, mNADP ϩ -ICDH inhibition by HNE was dose-dependent: there was a 58% decrease in activity with 20 M HNE (n ϭ 3 experiments) and a 90% decrease with 100 M HNE, which was the highest concentration tested. At 20 M, the inhibitory effect of HNE was already apparent after 5 min (20%; one-way analysis of variance, p Ͻ 0.001), although it was more pronounced after 20 min (64%; Fig. 2B). The addition of 0.5 mM cysteine, a thiol antioxidant, markedly reduced the magnitude of the effect of HNE on mNADP ϩ -ICDH activity (Fig. 2, A and B). Similar results were obtained with another thiol antioxidant, GSH (0.5 mM, data not shown). Finally, although incubation of mitochondria with 50 M HNE alone inhibited mNADP ϩ -ICDH activity by 85%, the addition of 1 mM NADP ϩ and/or isocitrate prior to HNE partially protected mNADP ϩ -ICDH from inactivation (Fig. 2C).

FIG. 1. Cardiac mNADP ؉ -ICDH activity (A), protein (B), and mRNA expression (C) in 7-week-old SHR and WKY rats.
Total proteins and RNA were extracted from SHR and WKY rat hearts and processed as described under "Materials and Methods." Western blotting was also performed on commercially purified pig heart mNADP ϩ -ICDH (std ϳ 46 kDa). Quantitation of protein and mRNA levels in arbitrary units was achieved by densitometric analysis of the autoradiograph bands. mRNA levels were normalized to those of GAPDH mRNA and expressed as a percentage of control WKY rats. Data are means Ϯ S.E. of six (activity) and four (protein and mRNA) rat hearts. Student's unpaired t test. SHR versus WKY; **, p Ͻ 0.01. The greater protective effect of isocitrate compared with NADP ϩ suggests that HNE reacts more readily with isocitrate than with the NADP ϩ -binding site.
To confirm that the inactivation of cardiac mNADP ϩ -ICDH was caused by the formation of HNE/mNADP ϩ -ICDH adducts, we performed Western blotting of mitochondrial samples incubated for 10 min with increasing concentrations of HNE. After immunoprecipitation with rabbit anti-mouse mNADP ϩ -ICDH antibody, protein complexes were separated by SDS-PAGE. Western blot analysis using anti-HNE antibody revealed a signal corresponding to HNE/mNADP ϩ -ICDH conjugates, whose intensity was proportional to HNE concentration (Fig.  2D). Samples from untreated mitochondria divulged also a low but detectable signal.
In Vivo Evidence-To support the pathophysiological significance of our finding, direct evidence for enhanced formation of HNE/mNADP ϩ -ICDH adducts in SHR hearts was sought by an immunoprecipitation approach with rabbit anti-mouse mNADP ϩ -ICDH antibody followed by Western blotting with rabbit anti-HNE antibody. As shown in Fig. 3A, there was a 2.1-fold increase in the relative tissue level of HNE/mNADP ϩ -ICDH adducts in SHR hearts compared with control WKY rat hearts which paralleled that of total HNE-protein adducts (Fig.  3B). DISCUSSION This study aimed at clarifying the regulation of cardiac mNADP ϩ -ICDH, an enzyme whose activity is a determinant of mitochondrial energy and oxidative stress status. One major finding of this study was a decline in myocardial mNADP ϩ -ICDH activity without change in its expression at the protein and mRNA levels in 7-week-old SHR compared with agematched control WKY rats. The SHR is a well established model of genetic hypertension, which also enables the investigation of sub-cellular mechanisms linked to cardiac hypertrophy development. This highly clinically relevant condition is associated with mitochondrial dysfunction because of energy deficits and increased oxidative stress (1,33,34,36). In agreement with the notion that cardiac hypertrophy in SHR develops between 9 and 12 weeks of age (32), we found no increase in the tissue level of ANF mRNA in 7-week-old SHR hearts compared with age-matched WKY rats.
In view of its potential pathophysiological significance, we conducted additional work to explain our finding of a decrease in mNADP ϩ -ICDH in 7-week-old SHR hearts. One potential candidate for inactivation of this enzyme is nitric oxide (22). In this study, we obtained the following evidence both in vitro and in vivo, indicating that the lipid peroxidation product (HNE) is another candidate for myocardial mNADP ϩ -ICDH inactivation. (i) Using isolated rat heart mitochondria, we showed that mNADP ϩ -ICDH activity was decreased progressively after incubation with increasing HNE concentrations ( Fig. 2A). Furthermore, taken together, the enzymatic and immunoprecipitation data obtained in isolated mitochondria provided evidence, for the first time, that mNADP ϩ -ICDH inactivation is associated with the formation of HNE/mNADP ϩ -ICDH adducts (Fig. 2, A and D). There was also a direct relationship mNADP ϩ -ICDH Inactivation by HNE Binding between the extent of mNADP ϩ -ICDH inactivation and the extent of HNE/mNADP ϩ -ICDH adduct accumulation with HNE concentrations. (ii) We provide direct immunological evidence for the in vivo occurrence of HNE/mNADP ϩ -ICDH adducts in the heart and for its increased level in 7-week-old SHR hearts, which was paralleled by a similar increase of total tissue HNE/protein adducts.
Other data provide information about the nature of the amino acid candidate(s) targeted by HNE. Because in vitro mNADP ϩ -ICDH inactivation was prevented by the addition of thiol agents (Fig. 2, A and B) or its substrate isocitrate (Fig.  2C), we conclude that HNE-binding sites are likely to include a cysteine residue near the catalytic site. These results concur with the known chemical reactivity of HNE and available structural information on mNADP ϩ -ICDH. (i) The ␣,␤-double bond of HNE reacts spontaneously, via Michael addition, to the sulfhydryl group of cysteine, the ⑀-amino group of lysine, and the imidazole function of histidine, but the sulfhydryl group shows the highest reactivity at neutral pH (15). Furthermore, although HNE binding to proteins via cysteine residues can potentially be reversed by thiol agents through thiol-disulfide exchange reactions (37), binding to histidine or lysine, via Michael addition or Schiff base formation, is expected to be irreversible. (ii) Although there is little information available on the rat enzyme, porcine mNADP ϩ -ICDH has seven cysteine residues (Swiss-Prot accession no. P33198), six of which have been shown to react with thiol reagents (35). Cysteine residues 305 and/or 387, which appear to be essential for its maximal catalytic activity, are particularly susceptible to modifications (22). However, additional HNE binding to histidine and lysine residues cannot be excluded. The rabbit anti-HNE antibody that we used in our study recognizes, but does not distinguish, HNE bound to proteins via histidine, lysine, or cysteine residues. mNADP ϩ -ICDH is particularly enriched in these basic amino acid residues, as evidenced from its pI value of 9.0 (Refs. 22 and 35). The extent of HNE binding to histidine and lysine residues as well as formation of Schiff base cross-links is expected to depend on intracellular HNE concentrations. The latter is determined by both the magnitude of the increase in free radicals, which will subsequently promote lipid peroxidation, and the intrinsic capacity of the cells to metabolize HNE, free or bound, to non-toxic products through oxidation, reduction, and conjugation with GSH (14, 38 -40).
Our finding of enhanced accumulation of HNE/mNADP ϩ -ICDH adducts in 7-week-old SHR hearts indicates that oxidative stress-related mitochondrial metabolic alterations are an early event in disease progression. Mitochondria have received considerable attention as a target and principal source of free radicals because of electron leakage from the respiratory chain to oxygen to form superoxide anions. However, their involvement in the pathogenesis of cardiac disease is predominantly considered in the context of the ischemic or failing heart (1,4,8,9,33,41). In SHR, enhanced production of superoxide anion has been reported as early as 4 weeks of age but was attributed mainly to dysfunctional constitutive nitric oxide synthase and/or angiotensin II-stimulated NAD(P)H oxidase activity (42)(43)(44)(45). Free radical-induced lipid peroxidation and HNE production could occur at these extra-mitochondrial sites. Indeed, given its relatively longer half-life compared with free radicals, HNE can diffuse within the cell to propagate the damage to mitochondria. Thus, the contribution of the various site(s) of HNE formation remains to be clarified.
Irrespective of the site(s) of HNE formation, our finding of enhanced accumulation of HNE/mNADP ϩ -ICDH adducts and total HNE/protein adducts in 7-week-old SHR hearts concurs also with increasing evidence in the literature indicating that lipid peroxidation products, specifically HNE, contribute to the pathophysiological oxidative stress-related events associated with cardiac disease (4,8,13,18,46). It is noteworthy that many signaling molecules and transcription factors modulated by HNE (e.g. mitogen-activated protein kinases, nuclear factor-B and activator protein-1) are also involved in processes implicated in cardiac hypertrophy and remodelling, such as fibrosis, inflammation, cell proliferation, and apoptosis (17)(18)(19)38). In this context, our results emphasize the importance of including the effects of HNE at the post-translational level on metabolic enzymes determining mitochondrial energy and NAD(P)H redox status in the network of signaling pathways and transcription factors that are triggered by HNE in the context of cardiac hypertrophy development.
Additional work is needed, however, to evaluate the specific significance of inactivation of mNADP ϩ -ICDH through HNE adduct formation and, in turn, cardiac hypertrophy development. We speculate that in the heart, where net substrate flux through mNADP ϩ -ICDH favors the generation of isocitrate and NADP ϩ (29) (thus supporting a predominant role of this enzyme acting in concert with H ϩ -transhydrogenases in the fine regulation of CAC flux; Ref. 28), decreased mNADP ϩ -ICDH activity should compromise mitochondrial energy metabolism. However, at the same time, this may spare NADPH for GSH regeneration by GSH reductase and, thus, increase mitochondrial capacity for HNE detoxification through GSH conjugation and export (38). Provided that HNE binding to mNADP ϩ -ICDH via cysteine residues is reversible, one could consider the existence of mechanisms for the acute control of its activity through post-translational modifications. Such a mechanism was proposed by Szweda and co-workers (47) for the thiol-containing CAC enzyme ␣-ketoglutarate dehydrogenase, which shows a susceptibility to inactivation by HNE that is similar to or less than that of mNADP ϩ -ICDH (50% inhibition at 40 versus ϳ20 M HNE, respectively; Ref. 7 and this study). Interestingly, these authors hypothesized the existence of a coordinated mitochondrial metabolic response to oxidative stress that involves GSH and glutathionylation of cysteine FIG. 3. Cardiac levels of HNE/mNADP ؉ -ICDH (A) and total HNE-protein (B) adducts in 7-week-old SHR and WKY rats. Total proteins from SHR and WKY rat heart extracts were immunoprecipitated and subjected to Western blotting (A) or concentrated five-fold, coated on microtiter plates, and subjected to enzyme-linked immunosorbent assay (B), as described under "Materials and Methods." Data are means Ϯ S.E. of three (A) or six (B) rat hearts. Student's unpaired t test. SHR versus WKY; #, p Ͻ 0.001. mNADP ϩ -ICDH Inactivation by HNE Binding residues of the active site of NAD(P)H-linked enzymes (47). In view of its metabolic role and the susceptibility of its cysteine residues to modification, mNADP ϩ -ICDH appears to fulfil criteria for participation in such a coordinated mitochondrial response. Clearly, in the short term, these changes may be considered partly adaptive, but in the long term, they would be maladaptive, leading to mitochondrial energy deficits.
In conclusion, the results of this study demonstrate the inactivation of cardiac mNADP ϩ -ICDH activity because of its post-translational modification in SHR before hypertrophy development. Specifically, mNADP ϩ -ICDH inactivation results from binding of the lipid peroxidation product HNE, probably to cysteine residues near the catalytic site. To the best of our knowledge, mNADP ϩ -ICDH had not yet been considered as an HNE target. The significance of our data extends, however, beyond mNADP ϩ -ICDH. Altogether, our results emphasize the implication of post-translational modifications of mitochondrial metabolic enzymes by HNE in the early pathophysiological events linked to cardiac hypertrophy development. Provided that HNE binding to mNADP ϩ -ICDH occurs at cysteine residues, without further formation of Schiff base cross-links, it could potentially be reversed by aldehyde-sequestering drugs (48,49). In fact, the reversal of HNE binding to proteins could be part of the mechanism by which N-acetylcysteine prevented cardiac hypertrophy in rats infused with angiotensin II, as recently demonstrated by Nakagami et al. (50). However, additional work is needed to clarify specifically whether a decline in mNADP ϩ -ICDH activity caused by HNE binding is part of the metabolic events that contribute to or is associated with cardiac hypertrophy development.