Hmox-1 Constitutes an Adaptive Response to Effect Antioxidant Cardioprotection
A Study With Transgenic Mice Heterozygous for Targeted Disruption of the Heme Oxygenase-1 Gene
Background—Heme oxygenase-1 (Hmox-1) has been implicated in protection of cells against ischemia/reperfusion injury.
Methods and Results—To examine the physiological role of Hmox-1, a line of heterozygous Hmox-1-knockout mice was developed by targeted disruption of the mouse Hmox-1 gene. Transgene integration was confirmed and characterized at the protein level. A 40% reduction of Hmox-1 protein occurred in the hearts of Hmox-1+/− mice compared with those of wild-type mice. Isolated mouse hearts from Hmox-1+/− mice and wild-type controls perfused via the Langendorff mode were subjected to 30 minutes of ischemia followed by 120 minutes of reperfusion. The Hmox-1+/− hearts displayed reduced ventricular recovery, increased creatine kinase release, and increased infarct size compared with those of wild-type controls, indicating that these Hmox-1+/− hearts were more susceptible to ischemia/reperfusion injury than wild-type controls. These results also suggest that Hmox-1+/− hearts are subjected to increased amounts of oxidative stress. Treatment with 2 different antioxidants, Trolox or N-acetylcysteine, only partially rescued the Hmox-1+/− hearts from ischemia/reperfusion injury. Preconditioning, which renders the heart tolerant to subsequent lethal ischemia/reperfusion, failed to adapt the hearts of the Hmox-1+/− mice compared with wild-type hearts.
Conclusions—These results demonstrate that Hmox-1 plays a crucial role in ischemia/reperfusion injury not only by functioning as an intracellular antioxidant but also by inducing its own expression under stressful conditions such as preconditioning.
Heme oxygenase (HO) catalyzes the reaction for heme metabolism, yielding equimolar quantities of CO, iron, and biliverdin.1 The HO system consists of 2 isoforms: oxidative stress–inducible HO-1 (Hmox-1), also known as heat-shock protein (HSP) 32, and the constitutive isozyme HO-2.2 In addition to oxidants, a variety of other environmental stresses, including heat stress, hypoxia, metals, endotoxin, and certain hormones, can induce Hmox-1.3 4
Recent evidence suggests that Hmox-1 induction plays a role in cellular protection against injury caused by the reactive oxygen species (ROS).5 Enhancement of HO activity by endotoxin attenuated renal failure, and this protection was lost when HO was inhibited with protoporphyrin.6 In another study, Hmox-1 mediated an adaptive response to oxidative stress in skin fibroblasts.7 ROS produced in the ischemic/reperfused myocardium has been shown to induce Hmox-1 gene expression.8
The physiological significance of Hmox-1 induction during myocardial ischemia remains unknown. It seems reasonable to speculate that Hmox-1 induction during ischemia/reperfusion is the heart’s own stress signal for survival against oxidative stress. Although a role of Hmox-1 in myocardial protection has been speculated upon, a definitive cardioprotective role of Hmox-1 is lacking. To fill this gap, we developed an Hmox-1+/− mouse by targeted disruption of the mouse Hmox-1 gene. The isolated hearts with 1 functional copy of the Hmox-1 gene and those of matched wild-type mice were subjected to ischemia/reperfusion. To examine whether Hmox-1 plays a role in stress adaptation, the hearts were subjected to preconditioning by cyclic episodes of ischemia/reperfusion. The results demonstrated impaired ventricular recovery and increased infarct size for the Hmox-1+/− hearts compared with those from wild-type controls, suggesting that Hmox-1+/− hearts are vulnerable to ischemia/reperfusion injury. Improvement of cardiac function was less apparent for the Hmox-1+/− mice after preconditioning.
Targeted Disruption of the Mouse Hmox-1 Gene
As shown in Figure 1⇓, the XhoI fragment of the mouse Hmox-1 gene9 spanning from part of exon 1 to intron 2 was cloned into the XhoI site of the targeting vector pPNT.10 The 3′ XbaI-EcoRI fragment of the gene-containing sequence of intron 3 to 3′ flanking region of the gene was subcloned into the corresponding XbaI and EcoRI sites in pPNT. The targeting vector was linearized with NotI digestion and then transfected into R1 embryonic stem (ES) cells11 by electroporation. Five homologous recombinant ES clones were identified from 224 clones by DNA blot analysis using the 5′ external probe encompassing sequences from nucleotide −133 to +73. Male chimeric mice, generated by microinjecting C57BL/6 blastocytes with 3 independent clones, were used in breeding with C57BL/6 female mice. Germ-line transmission of 129/Sv chromosomes was achieved by chimeric mice derived from each of the homologous recombinant ES clones. Line 183 Hmox-1+/− mice were used in this study.
Protein Blot Analysis of Hmox-1 in Hearts
Hearts were homogenized in 50 mmol/L phosphate buffer, pH 7.8, containing 0.1% Triton X-100 with a Polytron homogenizer. The homogenates were centrifuged at 20 000g and stored at −70°C. Protein contents were determined with an assay kit (Pierce).
For protein blot analysis, total cellular protein was separated on a 12% SDS-polyacrylamide gel and electrophoretically transferred onto a nitrocellulose paper. The protein blot filter paper was incubated in 1× HBSS plus a 1:2000 dilution of a monoclonal anti-human Hmox-1 antibody (StressGen) for 2 hours. The filter paper was washed with HBSS containing 1% nonfat milk, incubated with a 1:3000 dilution of horseradish peroxidase–conjugated goat anti-mouse IgG (Bio-Rad, Hercules) in HBSS plus 5% nonfat milk for 1 hour, washed again with HBSS, and then subjected to autoradiography in the presence of an enhanced chemiluminescent substrate (Pierce). Western blot data were quantified by densitometric scanning.
Measurement of Contractile Function
Twenty-four mice (10 to 12 weeks old) were divided into 2 groups: Hmox-1+/− (n=12) and wild-type (n=12). They were anesthetized with pentobarbital (200 mg/kg). The heart was excised, the aorta cannulated, and the heart perfused with Krebs-Henseleit buffer (KHB) by the Langendorff method.12 The effluent was collected before ischemia and during reperfusion and stored at −20°C for subsequent measurement of creatine kinase (CK) and malondialdehyde (MDA). A 4-0 silk suture on a round-bodied needle was passed through the apex of the heart and attached to the apex, which in turn was attached to a force transducer. The heart rate, force developed by the heart (DF), and first derivative of developed force (dF/dt) were recorded.12 The data on myocardial contractile function were recorded and analyzed in real time with the Cordat II data acquisition and presentation system (Triton Technologies). The hearts were subjected to 30 minutes of ischemia by clamping of the aortic cannula, followed by 2 hours of reperfusion.
To examine whether antioxidants could protect the Hmox-1+/− hearts from ischemia/reperfusion injury, 2 different protocols were used. (1) For in vivo studies, both wild-type and Hmox-1+/− mice were injected once every 24 hours with either Trolox (10 mg/kg IV), an analogue of vitamin E, or N-acetylcysteine (NAC, 1.5 mg/kg IV), a precursor of glutathione synthesis and an antioxidant. Control hearts were injected with saline only. (2) In vitro experiments were performed by preperfusion of the isolated wild-type and Hmox-1+/− hearts for 15 minutes with KHB buffer in the presence or absence of Trolox (10 mmol/L) or NAC (10 mmol/L). All hearts were subjected to 30 minutes of ischemia followed by 2 hours of reperfusion. The cardioprotective abilities of the antioxidants were examined by study of ventricular function, infarct size, and CK release.
Another group of hearts from Hmox-1+/− and wild-type mice were preconditioned by subjection to 3 cycles of 5 minutes of ischemia each, followed by another period of 5 minutes of reperfusion.13 14 This preconditioning protocol has been found to render the heart tolerant to the subsequent ischemic stress. All hearts were subjected to 30 minutes of ischemia followed by 2 hours of reperfusion.
Estimation of CK Release
CK release from the heart was estimated in the perfusate collected from the heart by use of a CK assay kit (Sigma). The results are expressed as IU/L of total CK release from the heart.
Estimation of MDA
MDA was assayed in the coronary perfusate as described previously15 to monitor the development of oxidative stress. The MDA was derivatized with DNPH.15 Aliquots of 25 μL of derivatized MDA in acetonitrile were injected onto a Beckman Ultrasphere C18 (3-mm) column in a Waters high-performance liquid chromatograph (Waters Corp). The products were eluted isocratically and detected at 307, 325, and 356 nm. The amount of MDA was quantified by use of a Maxima software program (Waters).
Measurement of Infarct Size
At the end of reperfusion, a 10% (wt/vol) solution of triphenyl tetrazolium in phosphate buffer was infused into the aortic cannula.16 The hearts were excised and stored at −70°C. Sections (0.8 mm) of frozen heart were fixed in 2% paraformaldehyde, placed between 2 coverslips, and digitally imaged with a Microtek ScanMaker 600z. To quantify the areas of interest in pixels, an NIH Image 5.1 (a public-domain software package) was used. The infarct size (transmural) was quantified in pixels.
For statistical analysis, a 2-way ANOVA followed by Scheffé’s test was first carried out to test for any differences between groups. If differences were established, the values were compared by Student’s t test for paired data. The values were expressed as mean±SEM. The results were considered significant at a level of P<0.05.
Generation and Characterization of Hmox-1+/− Mice
As shown in Figure 1a⇑, the mouse Hmox-1 gene was inactivated by replacement of exon 2 with a neomycin resistance (neo) cassette. Insertion of the neo gene creates a new PstI restriction site in the Hmox-1 gene, resulting in a shorter PstI genomic fragment from the targeted allele than that from the wild-type allele (Figure 1b⇑). Heterozygous Hmox-1+/− mice were apparently healthy and showed no pathological or phenotypic differences from the wild-type littermates. No homozygous knockout mice could be detected in the progeny from heterozygote intercross, however, indicating that embryonic lethality occurs in the homozygous mice. A 40% reduction of Hmox-1 was evident in the heart of the heterozygous mouse compared with that of the wild-type mouse (Figure 1c⇑). We reasoned that a decrease in expression of Hmox-1 protein due to the targeted mutation in Hmox-1+/− mice might render the mice more susceptible to an increased oxidative stress than controls.
All hearts recovered their beats spontaneously after 30 minutes of ischemia. The heart rates remained slightly but not significantly lowered, however, compared with baseline (results not shown). No significant difference was found in heart rate between groups throughout the experiment. Coronary flow progressively lowered during the reperfusion, but differences were not significant (Figure 2⇓). The DF during reperfusion was lower for the Hmox-1+/− mouse hearts than wild-type hearts (Figure 3⇓, top). In control hearts, DF recovered to near baseline levels after 90 minutes of reperfusion. Conversely, DF did not recover beyond 87% of baseline for the Hmox-1+/− hearts. At all points, DF showed significantly lower recovery for Hmox-1+/− hearts than wild-type hearts after 30 minutes of reperfusion (P<0.05). A similar trend was seen in dF/dtmax (Figure 3⇓, bottom). A significantly lower recovery of dF/dtmax occurred in Hmox-1+/− hearts after 30 minutes of reperfusion (P<0.05). These results demonstrate that Hmox-1+/− mouse hearts showed significantly lower contractile recovery after 30 minutes of ischemia.
Preconditioning rendered the hearts of the wild-type mice tolerant to ischemia/reperfusion injury, as evidenced by improved DF and dF/dt compared with wild-type hearts not subjected to preconditioning. In contrast, the preconditioned hearts of the Hmox-1 mice did not exhibit similar improvement in functional recovery, and there were no significant differences of DF and dF/dt between preconditioned and nonpreconditioned hearts.
Both in vivo and in vitro, Trolox and NAC treatments partially improved the contractile function. For example, postischemic DF and dF/dt were improved significantly but only partially compared with those of Hmox-1+/− hearts (Table⇓⇓). Interestingly, both antioxidants (in vivo or in vitro) improved DF and dF/dt to the same extent. Baseline values of the antioxidant-treated (in vivo or in vitro) and untreated groups were identical.
CK Release From Heart
Total CK release from the heart (Figure 4⇓) was negligible for all the groups before ischemia. After ischemia, CK release increased in both groups, but the amount of release was much higher for Hmox-1+/− hearts. At 30 minutes of reperfusion, CK release was 70±9.7 IU/mL for Hmox-1+/− mice versus 50±11.5 IU/mL (P<0.05) for nontransgenic controls. At 60 minutes of reperfusion, CK release was 120±8.7 IU/mL for Hmox-1+/− hearts versus 82±9.4 IU/mL (P<0.05) for nontransgenic controls. Similarly, after 120 minutes of reperfusion, CK release from the knockout mouse hearts amounted to 239±8.1 IU/mL compared with 135±7.4 IU/mL for controls.
Preconditioning significantly reduced the amount of CK in wild-type hearts compared with the nonpreconditioned group. The amount of CK release was also less for the preconditioned Hmox-1+/− hearts than for nonpreconditioned hearts, but the differences did not reach significance except at 90 and 120 minutes of reperfusion.
In vivo and in vitro, Trolox and NAC treatments partially but significantly reduced the CK release compared with that from the Hmox-1+/− hearts (Table⇑⇑). Again, both antioxidants reduced the CK release to the same extent. Baseline values of the antioxidant-treated and untreated groups were identical.
The portion of the heart not stained by tetrazolium was measured and calculated as the infarct area. Mean value of infarct size in the knockout group was significantly higher than that in the wild-type group (48.5±3.2% versus 28.5±2.4% for wild-type, P<0.05) (Figure 5⇓). Our results indicated that Hmox-1+/− hearts had significantly higher myocardial necrosis.
A reduced amount of infarct size was apparent in the hearts subjected to preconditioning. Significantly less infarct was noticed, however, for the preconditioned wild-type hearts than nonpreconditioned hearts.
Both in vivo and in vitro, Trolox and NAC treatments partially but significantly reduced the infarct size compared with the Hmox-1+/− hearts (Table⇑⇑). Similar to the previous results, both antioxidants reduced the infarct size to the same extent.
MDA, measured as MDA-DNPH, increased progressively during reperfusion for the nonpreconditioned hearts (Figure 6⇓). The amount of MDA production, however, was significantly higher at all points in the Hmox-1+/− hearts than in control hearts, demonstrating that Hmox-1+/− hearts were subjected to increased amounts of oxidative stress. Preconditioning reduced the amount of MDA formation in both groups of hearts. A significantly higher amount of MDA was found in the Hmox-1 hearts than in the wild-type control hearts.
As mentioned earlier, Hmox-1+/− mouse hearts showed ≈40% reduction of Hmox-1 protein compared with Hmox-1+/+ hearts. The activities of other key antioxidant enzymes, including manganese superoxide dismutase (Mn-SOD), catalase, glutathione peroxidase, glutathione reductase, and glucose-6-phosphate-dehydrogenase, remained unaltered in the Hmox-1+/− hearts compared with wild-type controls (data not shown). The results thus truly reflect the effects of Hmox-1+/− in myocardial ischemia/reperfusion injury.
Hmox-1+/− hearts were more susceptible to ischemia/reperfusion injury, as evidenced by greater infarct size, increased CK release, and reduced ventricular recovery compared with wild-type controls. In addition, these Hmox-1+/− hearts could not be preconditioned like the hearts of the wild-type controls, suggesting that Hmox-1 plays a role in the adaptation to stress. This is further supported by the results with MDA, because MDA formation, a presumptive marker for oxidative stress, was higher for the preconditioned Hmox-1+/− mouse hearts than the wild-type hearts.
During the course of our studies, Poss and Tonegawa17 18 have reported generation of Hmox-1+/− mice in which exons 3, 4, and part of 5 were replaced by the neo cassette. Whereas they were able to obtain homozygous knockout mice from heterozygous breeding at a lower-than-expected frequency, an indication of embryonic lethality due to Hmox-1 deficiency, we could not detect any homozygous knockout mice in our breeding colony. The discrepancy in our breeding results is not understood. It may result from the fact that different ES cells were being used in each laboratory. The studies by Poss and Tonegawa using homozygous knockout mice have clearly established the antioxidant role of Hmox-1 against the toxicity resulting from treatment with endotoxin, heme, and hydrogen peroxide. We rationalized that a modest decrease in Hmox-1 gene expression in Hmox-1+/− hearts might also cripple the cellular antioxidant defense mechanism, which is protective in certain disease models whose pathogenesis is known to be associated with an overproduction of ROS. Similar to what was observed by Poss and Tonegawa,18 the Hmox-1+/− mice generated in our laboratory showed no increased susceptibility to endotoxin toxicity compared with that of wild-type mice (unpublished data). This report documents that heterozygous Hmox-1+/− mice with an inactivated Hmox-1 allele are more vulnerable to myocardial ischemia/reperfusion injury.
Heme oxygenase, which catalyzes the initial rate-limiting step of heme catabolism, is known to be induced by a wide variety of oxidative stresses, including H2O2, UV radiation, and reduced glutathione depletion.3 4 A study from our laboratory documented that ischemia/reperfusion could induce the expression of Hmox-1 mRNA in the heart.8 Induction of Hmox-1 gene expression increased as a function of duration of reperfusion, which could be inhibited with ROS scavengers and SOD, suggesting that oxidative stress developed during ischemia/reperfusion induces the HO-1 gene. Our results are consistent with these previous reports, indicating that Hmox-1 is induced by oxidative stress.
The role of HO-1 in oxidative stress adaptation was further supported by the observation that the hearts of the Hmox-1+/− mice could not be preconditioned to make them resistant to subsequent oxidative stress compared with wild-type hearts. Preconditioning by repeated brief episodes of ischemia/reperfusion is known to develop oxidative stress.19 The development of oxidative stress presumably leads to adaptive modification of the hearts, which is achieved through a chain of reactions including generation of intracellular mediators and modification of signal transduction processes and gene expression.20 Oxidative stress has been shown to function as a second messenger in this process.21 The results of our study suggest that induction of HO-1 by oxidative stress plays a crucial role in preconditioning-mediated adaptive modification of the heart.
HO, an oxidative stress–inducible protein, is also an HSP, HSP 32. HO-1 is evolutionarily conserved in the primary amino acid and nucleotide sequences, and its extent of homology between mouse, rat, and human is >80%. Hmox-1 is readily induced not only by oxidative stress but also with many other pathological and environmental stresses, such as hypoxia, hyperoxia, cellular transformations, and heat shock.22 This enzyme catalyzes the conversion of heme into biliverdin and bilirubin, which function as in vivo antioxidants.1 In addition, a messenger molecule, carbon monoxide (CO), is generated during the production of biliverdin. Previous studies demonstrated that production of cGMP from CO signaling is beneficial for the heart.23
To examine whether antioxidants could rescue the Hmox-1+/− hearts from increased ischemia/reperfusion injury, the Hmox-1+/− mice were treated with 2 different antioxidants, Trolox and NAC. Both of these antioxidants are cell-permeable. Trolox, an analogue of vitamin E, is a hydrophilic antioxidant that has been found to be cardioprotective. NAC is a precursor of glutathione and functions as an antioxidant. NAC also prevents intracellular oxidation and maintains glutathione levels by reducing cystine to cysteine. Both Trolox and NAC provided similar degrees of cardioprotection when used in vitro or in vivo. Neither of these antioxidants, however, could completely rescue the hearts from the ischemia/reperfusion injury. Nevertheless, they significantly reduced the amount of ischemia/reperfusion injury compared with control Hmox-1+/− hearts. Interestingly, the amount of cardioprotection afforded by these antioxidants was comparable to that obtained by preconditioning the Hmox-1+/− hearts.
Numerous reports in the literature indicate cardioprotective effects of preconditioning.24 25 Preconditioning potentiates a signal transduction mechanism coupled with both G-protein and tyrosine kinase receptors, leading to the activation of protein kinase C and mitogen-activated protein kinases and redox-sensitive transcription factors such as nuclear factor-κB and activating protein-1.26 Preconditioning-mediated cardioprotection is believed to be achieved through its ability to induce several cardioprotective genes and proteins, including HSP 27, HSP 70, HSP 89, SOD, and reduced glutathione peroxidase-1.27 These genes/proteins have been shown to constitute the defense system of the myocardium. Our results demonstrate that HO-1 is also an important member of the antioxidant defense system of the heart.
In summary, our study documents the crucial role of Hmox-1 in cardioprotection. A decrease in HO-1 in the heart makes it vulnerable to ischemia/reperfusion injury. In vivo or in vitro antioxidant therapy only partially rescued the Hmox-1+/− hearts from the ischemia/reperfusion injury, suggesting that HO-1 plays an important role in cardioprotection. Preconditioning, which provokes oxidative stress and induces a large variety of oxidative stress–inducible genes, was unable to rescue the hearts of heterozygous Hmox-1+/− mice from cellular injury, presumably because of the reduced level of Hmox-1. This paves the way for the potential of Hmox-1 gene therapy for cardioprotection in pathological conditions in which oxidative stress is implicated.
This study was supported by NIH grants HL-22559, HL-33889, and HL-34360 (Dr Das), a Grant-in-Aid from the American Heart Association and HL-56803 (Dr Maulik), HL-56421 and P30-ES-06639 (Dr Ho), and DK-43135 (Dr Alam).
- Received August 25, 2000.
- Revision received October 9, 2000.
- Accepted October 10, 2000.
- Copyright © 2001 by American Heart Association
Tenhunen R, Marver HS, Schmid R. The enzymatic conversion of heme to bilirubin by microsomal heme oxygenase. Proc Natl Acad Sci U S A. 1968;61:748–755.
Maines MD. Heme oxygenase: function, multiplicity, regulatory mechanisms, and clinical applications. FASEB J. 1988;2:2557–2568.
Vile GF, Tyrrell RM. Oxidative stress resulting from ultraviolet A irradiation of human skin fibroblasts leads to a heme oxygenase-dependent increase in ferritin. J Biol Chem. 1993;268:14678–14681.
Nath KA, Balla G, Vercellotti G. Induction of heme oxygenase is a rapid, protective response in rhabdomyolysis in the rat. J Clin Invest. 1992;90:267–270.
Vile GF, Basu-Modak S, Waltner C, et al. Heme oxygenase 1 mediates an adaptive response to oxidative stress in human skin fibroblasts. Proc Natl Acad Sci U S A. 1994;91:2607–2610.
Alam J, Cai J, Smith A. Isolation and characterization of the mouse heme oxygenase-1 gene. J Biol Chem. 1994;269:1001–1009.
Nagy A, Rossant J, Nagy R, et al. Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc Natl Acad Sci U S A. 1993;90:8424–8428.
Yoshida T, Maulik N, Engelman RM, et al. Glutathione peroxidase knockout mice are susceptible to myocardial ischemia reperfusion injury. Circulation. 1997;96(suppl II):II-216–II-220.
Miller DL, Van Winkle DM. Ischemic preconditioning limits infarct size following regional ischemia-reperfusion in in situ mouse hearts. Cardiovasc Res. 1999;42:680–684.
Das DK, Engelman RM, Kimura Y. Molecular adaptation of cellular defences following preconditioning of the heart by repeated ischemia. Cardiovasc Res. 1993;27:578–584.
Yoshida T, Maulik N, Engelman RM, et al. Targeted disruption of the mouse SodI gene makes the hearts vulnerable to ischemic reperfusion injury. Circ Res. 2000;86:264–269.
Poss KD, Tonegawa S. Heme oxygenase 1 is required for mammalian iron reutilization. Proc Natl Acad Sci U S A. 1997;94:10919–10924.
Poss KD, Tonegawa S. Reduced stress defense in heme oxygenase 1-deficient cells. Proc Natl Acad Sci U S A. 1997;94:10925–10930.
Maulik N, Das DK. Redox regulation of ischemic adaptation. In: Sen CK, Sies H, Baeuerle PA, eds. Antioxidant and Redox Regulation of Genes. New York, NY: Academic Press; 1999:491–516.
Maulik N, Engelman D, Watanabe M, et al. Nitric oxide/carbon monoxide: a free radical-dependent molecular switch for myocardial preservation during ischemic arrest. Circulation. 1996;94(suppl II):II-398–II-406.
Flack JE, Kimura Y, Engelman RM, et al. Preconditioning the heart by repeated stunning improves myocardial salvage. Circulation. 1991;84(suppl III):III-369–III-374.
Maulik N, Yoshida T, Zu Y-L, et al. Ischemic adaptation of heart triggers a tyrosine kinase regulated signaling pathway: a potential role for MAPKAP kinase 2. Am J Physiol. 1998;275:H1857–H1864.