Limitation of damage after ischemia and reperfusion injury to the myocardium remains an elusive clinical goal. Previous studies have suggested that molecular chaperones, which include members of the heat shock protein (Hsp) family, may have cardioprotective effects, although the protective role of endogenous chaperones has not been well documented. CHIP (carboxyl terminus of Hsp70-interacting protein) is a cochaperone/ubiquitin ligase that integrates the response to stress at multiple levels. We tested the response of CHIP(−/−) mice to in vivo ischemia and reperfusion injury induced by left anterior descending coronary artery ligation. Compared with wild-type littermates, CHIP(−/−) mice had decreased survival and increased incidence of arrhythmias during reperfusion. The size of myocardial infarction, as assessed by the ratio of infarct area to area at risk, was 50% greater in CHIP(−/−) mice. Increased infarct size was accompanied by impaired upregulation of the chaperone Hsp70 after ischemia-reperfusion injury. In situ analysis also indicated that hearts of CHIP(−/−) mice were more prone to develop apoptosis in cardiomyocytes and especially endothelial cells of intramural vessels. Previous studies have found that CHIP plays a central role in maintaining protein quality control and coordinating the response to stress. The present data indicate that these functions of CHIP provide a critical cardioprotective effect in the setting of ischemia-reperfusion injury due in part to increased apoptosis in cardiac cells. Quality control mechanisms therefore may be underappreciated clinical targets for maximizing myocardial protection after injury.
- heat shock protein
ischemic heart disease is still the most prevalent cause of death in developed countries, yet improvements in access to care, pharmacological therapy, and acute interventional therapies have reduced the per capita incidence of death due to acute coronary syndromes. These advances in diagnosis and therapy allow rapid return of blood flow to jeopardized myocardium at early stages and improve survival, yet myocardial injury is rarely avoided totally and cardiomyocytes face the additional peril of reperfusion injury. Despite (or perhaps because of) these trends, the incidence of death due to congestive heart failure has increased by 60% over the past two decades (20a), reflecting the transition of ischemic cardiovascular disease and its sequelae from an acute to a chronic disease. Although strategies exist to target many points in the ischemic cascade, methods to provide cardioprotection directly are lacking, and reperfusion injury remains an elusive clinical target.
Among putative cardioprotective pathways, several lines of inquiry point to a role for molecular chaperones (many of which are heat shock proteins). Heat shock proteins rescue and refold damaged proteins and also have direct antiapoptotic activities (21). Heat shock proteins are abundantly expressed within myocardial cells, and the inducible heat shock protein Hsp70 is upregulated after ischemic injury to the heart (4). Overexpression of Hsp70 improves functional recovery and reduces infarct size after ischemia in hearts of transgenic mice (12, 22, 25), and the small heat shock proteins αB-crystallin and Hsp27 exert similar effects in cultured myocytes (19). Nevertheless, a role for the endogenous chaperone system in cardioprotection remains to be clearly defined, in part because of redundancy among proteins in this family and a paucity of careful physiological analyses in appropriate animal model systems.
Investigators in our laboratory (3) have recently identified a dual-function cochaperone/ubiquitin ligase, CHIP (carboxyl terminus of Hsp70-interacting protein), that is highly expressed in the heart and that regulates chaperone activity and protein quality control at multiple levels. CHIP interacts with Hsp70 and enhances refolding of stress-damaged proteins in vivo (15). CHIP also has ubiquitin ligase activity and triggers proteasome-dependent degradation of irreversibly damaged proteins to prevent cellular toxicity (5, 9, 13). CHIP is a physiological regulator of stress-dependent apoptosis, in part through inhibitory interactions with proapoptotic signaling pathways (Huang J and Patterson C, unpublished observations). Finally, and most surprisingly, CHIP induces trimerization and transcriptional activation of heat shock factor 1 (the major transcriptional regulator of the heat shock response), and CHIP is therefore required for maximal Hsp70 induction in all tissues, including the heart (7). Because CHIP exerts a coordinating and central role in protein quality control, mice deficient in CHIP provide an excellent model to test the extent to which the endogenous molecular chaperone and protein quality control machinery confer cardioprotection in the setting of a clinically relevant myocardial insult.
MATERIALS AND METHODS
Myocardial ischemia and reperfusion.
All animal procedures were performed with approval from the University of North Carolina Institutional Animal Care and Use Committee. The generation of CHIP(−/−) mice, which are maintained on a mixed C57Bl/6 × 129 SvEv background, was previously described (7). CHIP(−/−) mice and their wild-type littermates were used at 8–12 wk of age. Mice were subjected to a myocardial ischemia-reperfusion model as described previously (20). Anesthetized mice were placed in a supine position and mechanically ventilated. Ischemia was achieved by ligating the anterior descending branch of the left coronary artery (LAD) with an 8-0 prolene suture, using a 1-mm section of polyethylene PE-10 tubing placed on top of the LAD, 1 to 3 mm from the tip of the normally positioned left atrium. Regional ischemia was confirmed by myocardial blanching and electrocardiographic ST-segment elevation. After occlusion for 30 min, blood flow was restored by removing the PE-10 tubing. The chest wall was then closed with a 6-0 silk suture, with one layer through the chest wall and muscle and a second layer through the skin and subcutaneous layer. Sham-operated mice underwent an identical procedure with placement of the ligature but did not undergo coronary artery occlusion. Hearts were harvested after 24 h of reperfusion. Electrocardiograms were recorded on lead II with an Axon data acquisition system (Axon Instruments). ECG was recorded from the start of tracheotomy to 120 min of reperfusion. All data were analyzed using AxoScope 8.1 software. Infarct size was determined in a blinded fashion, as described previously (20). In brief, the aortas were cannulated with a 22-gauge tubing adapter and sequentially perfused with 3–5 ml of PBS. The LAD was reoccluded, and a solution containing 0.25% fluorescent polymer microspheres was perfused into the aorta and coronary arteries with distribution throughout the ventricular wall proximal to the coronary artery ligature (26). Hearts were frozen for 15 min and cut into four transverse sections. Sections were incubated in 1.0% triphenyltetrazolium chloride for 10 min. After triphenyltetrazolium chloride staining, viable myocardium stains brick red and the infarct appears pale white. Sections were then photographed with a Nikon digital camera and weighed after the right ventricular free wall was removed. The images were quantified with the use of image analysis software (ImageJ 1.30). The fractions of both area at risk (AAR) to total slice size and infarct size to total slice size were calculated and multiplied by the weight of the slice to determine AAR and infarct weight per slice. Infarct size was expressed as a percentage of left ventricular (LV) mass and of the AAR. Detection of cells undergoing apoptosis from six mice of each genotype was evaluated using the ApopTag peroxidase in situ oligo ligation kit (Intergen) according to the supplier's instructions.
Two-dimensional imaging and M-mode imaging were performed using a VisualSonics Vevo660 ultrasound biomicroscopy system. LV end-diastolic dimension (LVEDD), LV end-systolic dimension (LVESD), interventricular septum thickness, and posterior wall thickness were obtained, and all LV dimension data were presented as the averages of two measurements. The percentage of fractional shortening (%FS) was calculated from the equation %FS = [(LVEDD − LVESD)/LVEDD] × 100.
Immunohistochemistry for Hsp70.
Sections (5 μm thick) were deparaffinized, rehydrated, and blocked with 1.5% normal goat serum. Hsp70 expression was detected using a mouse anti-Hsp70 monoclonal antibody (5 μg/ml; Stressgen) and biotinylated secondary antibody and avidin-biotin horseradish peroxidase (Santa Cruz Biotechnology). The sections were visualized under bright-field microscopy, and images were collected with a Qimaging Retica 1300 digital camera using Qcapture software.
Western blot analysis for Hsp70 and Hsp25.
Total protein was extracted from hearts, and Western blot analysis was performed with a mouse anti-Hsp70 monoclonal antibody (diluted 1:1,000; Stressgen) and a rabbit anti-Hsp25 polyclonal antibody (diluted 1:10,000; Stressgen). The immunoblot bands were measured by densitometric analysis of the film using Labworks software.
Data are presented as means ± SE. Significance was evaluated using Student's t-test and χ2 test. A value of P < 0.05 was considered statistically significant.
Baseline assessment of cardiac function in CHIP(−/−) mice.
As a first step to understand the role of CHIP in the cardiac response to stress, we measured baseline cardiac function in vivo in CHIP(−/−) mice and their littermates. Transthoracic echocardiography was performed in conscious 2-mo-old mice. LV function was determined by calculating %FS. The LV %FS was not significantly different between CHIP(−/−) mice and wild-type mice. In addition, no significant differences in LV end-diastolic and end-systolic chamber diameters or wall-thickness values were detected between the groups (Table 1). Ultrastructural analysis disclosed no obvious abnormalities in hearts of CHIP(−/−) mice (data not shown).
Physiological response of CHIP(−/−) mice to ischemia-reperfusion injury.
We employed a modification of a previously described protocol to induce myocardial infarction by LAD occlusion and reperfusion in vivo (20). Eight- to twelve-week-old age- and sex-matched littermate mice were subjected to 30 min of ischemia followed by 24 h of reperfusion. There were no significant differences in the basal hemodynamic parameters, including heart rate, between CHIP(−/−) mice and wild-type mice (Table 1). Under the conditions used in our study, wild-type mice were totally resistant to ventricular arrhythmias (6 or more ventricular premature beats/min) during reperfusion. In contrast, the incidence of ventricular arrhythmias observed during reperfusion in CHIP(−/−) mice was 42.9% (P < 0.01 compared with wild-type mice; Fig. 1). The frequency of ventricular premature beats in these mice varied from 84 to 174 beats/min during the early phase of reperfusion. These findings indicate that CHIP(−/−) mice have increased arrhythmogenic susceptibility during reperfusion. In addition, CHIP(−/−) mice were more prone to sudden death after induction of myocardial infarction. Under the conditions of our study, all wild-type mice survived the transient LAD ligation protocol, an observation similar to that of other groups (8). In contrast, only 78% of CHIP(−/−) mice were alive after 24 h of reperfusion (Fig. 2). The deaths in CHIP(−/−) mice occurred during the early phases of reperfusion and, in instances when heart rhythms were captured, were uniformly preceded by asystolic arrest without evidence of cardiac rupture upon postmortem analysis. No significant gender differences were noted in the susceptibility to ischemia-reperfusion injury in these experiments.
Increased susceptibility to myocardial infarction in CHIP(−/−) mice.
Left coronary artery occlusion as performed in this study consistently produced an ischemic AAR and myocardial infarction in every mouse studied (Fig. 3A). The AAR-to-LV ratio in CHIP(−/−) mice (59.1 ± 2.5%) did not differ from that in wild-type mice (58.8 ± 1.9%). The infarct-to-AAR and the infarct-to-LV ratios in wild-type mice were 25.5 ± 1.8% and 14.9 ± 1.0%, respectively, which are consistent with infarct sizes reported by other groups under similar conditions (11, 20, 28). In contrast, infarct sizes were 50% larger in CHIP(−/−) mice (infarct-to-AAR and infarct-to-LV ratios of 37.0 ± 2.9% and 21.4 ± 1.3%; Fig. 3B), indicating that endogenous CHIP plays a powerful checkpoint role in protection against ischemia-reperfusion injury.
CHIP is required for maximal Hsp70 induction.
Our previous studies indicate that CHIP is required for maximal induction of Hsp70 in response to thermal challenge, given the ability of CHIP to activate heat shock factor 1 (7). Previous studies have shown that Hsp70 is also induced by myocardial ischemia (16), but whether this effect is heat shock factor dependent has not been determined, and we do not know whether CHIP is required for Hsp70 expression in conditions other than thermal challenge. In our studies, Western blot analysis indicated that Hsp70 levels were very low or undetectable in hearts of sham-treated mice, regardless of their genotype (Fig. 4, A and B). However, after myocardial ischemia and reperfusion injury, induction of Hsp70 was consistently lower in CHIP(−/−) mice than in wild-type mice. Hsp25, another abundant heat shock protein in the heart (17), was also induced, albeit less strongly, after ischemia-reperfusion injury, and this induction was modestly but nonsignificantly reduced in CHIP(−/−) mice (Fig. 4, A and C). We also characterized expression of the cochaperone Bcl-2-associated athanogene-1 (BAG-1), given recent evidence that BAG-1 also may have cardioprotective effects (24). Consistent with previous reports, BAG-1 protein levels were induced by ischemia-reperfusion injury and, remarkably, BAG-1 levels were elevated in the hearts of CHIP(−/−) mice both before and after injury, perhaps as a partial compensatory response (Fig. 4D). In contrast, CHIP total protein levels were unaffected by ischemia-reperfusion injury, which is consistent with previous data supporting regulation of CHIP activity at the posttranslational level (7). Immunohistochemical examination demonstrated high expression of Hsp70 in the cytoplasm of cardiomyocytes in the left ventricles of wild-type mice after ischemia-reperfusion injury in a pattern that reflected the ischemic AAR (Fig. 4E). In contrast, Hsp70 expression was barely detectible in the same areas of CHIP(−/−) mice. Using desmin as a myocyte marker, we showed that the majority of cells expressing Hsp70 in wild-type hearts after ischemia were in the cardiomyocyte compartment (Fig. 4F).
CHIP provides protection against apoptosis in the setting of ischemia-reperfusion injury.
To define the cellular mechanism leading to increased infarct size in CHIP(−/−) mice, we measured biochemical parameters of apoptosis after ischemia-reperfusion injury, given our previous observations that CHIP is required for protection against thermal stress-induced apoptosis in vivo (7). An ischemic period of 30 min followed by 24 h of reperfusion resulted in both cardiomyocyte and endothelial cell apoptosis of intramural blood vessels within the AAR in CHIP(−/−) mice that was much more prominent than in cells in similar sections from hearts of wild-type mice (Fig. 5A). Quantitatively, significant increases in apoptotic cells were detected in both endothelial cells and cardiomyocytes of CHIP(−/−) mice (Fig. 5B). As a separate measure, we analyzed activation of caspase 3 as an additional marker of apoptosis. Precaspase 3 was efficiently cleaved after ischemia-reperfusion injury in both wild-type and CHIP(−/−) mice (Fig. 5C) at levels that were quantitatively indistinguishable (Fig. 5D). In contrast, activated caspase 3 levels were increased by ∼50% in CHIP(−/−) mice (Fig. 5E), which is proportional to the increases observed with the use of the TdT-mediated dUTP nick-end labeling (TUNEL) method.
In this report, we characterize the cardiovascular phenotypes of mice deficient in the cochaperone/ubiquitin ligase CHIP. CHIP is highly expressed in the heart compared with non-muscle tissues (3), yet hearts develop normally in these mice, and baseline function determined by echocardiography in CHIP(−/−) mice is comparable to that in wild-type mice. In contrast, CHIP(−/−) mice tolerate LAD ligation much more poorly than their wild-type littermates, with higher mortality rates after infarction and more frequent reperfusion arrhythmias. This is accompanied by infarct areas 50% greater than in wild-type mice and by evidence of increased vascular and cardiomyocyte apoptosis.
These studies are among the first thorough tests of a pathophysiologically relevant function for a molecular cochaperone and provide the most stringent evidence to date demonstrating a necessary role for the chaperone system in regulating the response to cardiac ischemia-reperfusion injury. Previous studies have frequently, but not invariably, demonstrated cardioprotective effects of increasing Hsp70 levels. A recent report (11) indicated that mice lacking two major Hsp70 isoforms, Hsp70.1 and Hsp70.3, have an unaffected response to ischemia-reperfusion compared with wild-type mice, yet they do have a diminished response to late-phase preconditioning. The more severe phenotype in mice lacking CHIP may reflect the fact that still other Hsp70 family members exist that may compensate in the absence of Hsp70.1 and Hsp70.3. Alternatively, the extreme phenotype of mice lacking CHIP may indicate that a perturbation of chaperone function at an upstream, more global level (by affecting heat shock factor 1 activation and also by impairing degradation of misfolded proteins) may have more drastic consequences to the myocardium. Because protein quality control pathways are redundant and self-regulating, animal models in which critical upstream and integrative proteins such as CHIP are targeted may provide a superior method for testing the function of the molecular chaperone system than models in which individual chaperone members are affected.
It is of interest that another cochaperone, BAG-1, is also implicated in protection against ischemia-reperfusion injury in cardiomyocytes (24). BAG-1 isoforms are induced after ischemia-reperfusion injury in intact hearts, and overexpression of BAG-1 proteins that bind Hsc70 protect cultured cardiomyocytes against simulated ischemia-reperfusion injury. Although in vivo investigation of a protective role for BAG-1 has not been undertaken in the heart and loss-of-function phenotypes have not been reported, these observations point to an important cardioprotective role for BAG-1. CHIP and BAG-1 are known to function cooperatively to regulate Hsp70 function and, in particular, chaperone-dependent protein degradation (1, 9), so the present studies suggest that these proteins may similarly cooperate to regulate a cardioprotective mechanism involving protein quality control mechanisms.
We noticed markedly increased frequencies of reperfusion arrhythmias in CHIP(−/−) mice that were never noted in wild-type mice during either the ischemic or reperfusion phases. It is possible that this is simply a reflection of the larger infarct size in CHIP(−/−) mice, although the discordance in ventricular premature beats seems disproportional to the differences in infarcted myocardium. Previous studies have demonstrated that induction of Hsp70 by thermal preconditioning inhibits reperfusion arrhythmias (14), and at least one cardiac potassium channel component, the α-subunit of the IKr cardiac potassium current, requires the chaperones Hsp70 and Hsp90 for normal maturation and trafficking (10). Thus we cannot exclude the possibility that CHIP also exerts a direct effect to regulate the arrhythmogenic potential in the setting of ischemia and reperfusion injury, and further studies are indicated to test this possibility.
A number of events that contribute to myocardial infarction occur during reperfusion following ischemia. Certainly, direct damage to proteins that must then either be refolded or degraded to protect the cell is a major but difficult-to-measure burden induced in this situation, and CHIP is ideally suited to coordinate the triage of damaged proteins so that they can be refolded or degraded when damaged (18). Mitochondrial dysfunction and generation of oxidative species are also linked to myocardial damage (2), and several lines of evidence suggest that chaperones are needed for oxidative protection and normal mitochondrial function. In particular, heat shock factor 1 is required to protect mitochondrial proteins from oxidative damage and dysfunction, presumably through induction of heat shock proteins (27); we can speculate that CHIP is equally important in mitochondrial protection, given its central role in regulation of heat shock factor 1. It is also apparent that apoptosis plays a contributory role in the face of cardiac reperfusion injury, and CHIP exerts a checkpoint role in protection against stress-induced apoptosis. In particular, we noted that both endothelial cells and cardiomyocytes are more prone to apoptosis after ischemia and reperfusion injury in CHIP(−/−) mice (Fig. 5). The endothelial component of this apoptotic injury is particularly relevant, given recent observations indicating that endothelial cell injury may precede and account for a significant proportion of myocardial injury in ischemic conditions (23).
The present studies provide direct evidence that the molecular chaperone machinery plays a necessary role in protection against myocardial infarction after ischemia-reperfusion injury. Recent data reveal that CHIP plays an unexpectedly central role in integration of the stress response and protection against cell death after thermal challenge (7). The present studies indicate that CHIP also regulates the response to injury in the heart, where it is preferentially expressed, under pathophysiologically relevant conditions. Analysis of mice lacking CHIP may be an especially useful tool to unravel how protective systems defend against myocardial damage, because CHIP coordinates a critical tier of the response to stress and protein damage in injured cells. Reconsideration of the importance of protein quality control mechanisms in protection against cardiac injury is warranted in the effort to find ways to rescue myocardium after interventional strategies in acute coronary syndromes.
This work was supported by National Institutes of Health Grants GM-61728 and HL-65619 (to C. Patterson) and HL-42550 (to L. H. Michael). C. Patterson is an Established Investigator of the American Heart Association and a Burroughs Wellcome Fund Clinical Scientist in Translational Research.
We thank Gang Lu for assistance with ECG recording.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 by the American Physiological Society