Diphtheria Toxin-induced Autophagic Cardiomyocyte Death Plays a Pathogenic Role in Mouse Model of Heart Failure*

It is still not clear whether loss of cardiomyocytes through programmed cell death causes heart failure. To clarify the role of cell death in heart failure, we generated transgenic mice (TG) that express human diphtheria toxin receptor in the hearts. A mosaic expression pattern of the transgene was observed, and the trans-gene-expressing cardiomyocytes (17.3% of the total cardiomyocytes) were diffusely scattered throughout the ventricles. Intramuscular injection of diphtheria toxin induced complete elimination of the transgene-express-ing cardiomyocytes within 7 days, and (cid:1) 80% of TG showed pathophysiological features characteristic of heart failure and were dead within 14 days. Degenerated cardiomyocytes of the TG heart showed characteristic features indicative of autophagic cell death such as up-regulated lysosomal markers and abundant injection revealed an increase in the cross-sectional area of cardiomyocytes, indicating hypertrophic compensation of myocardial cells without transgene expression.

It is still not clear whether loss of cardiomyocytes through programmed cell death causes heart failure. To clarify the role of cell death in heart failure, we generated transgenic mice (TG) that express human diphtheria toxin receptor in the hearts. A mosaic expression pattern of the transgene was observed, and the transgene-expressing cardiomyocytes (17.3% of the total cardiomyocytes) were diffusely scattered throughout the ventricles. Intramuscular injection of diphtheria toxin induced complete elimination of the transgene-expressing cardiomyocytes within 7 days, and ϳ80% of TG showed pathophysiological features characteristic of heart failure and were dead within 14 days. Degenerated cardiomyocytes of the TG heart showed characteristic features indicative of autophagic cell death such as up-regulated lysosomal markers and abundant autophagosomes containing cytosolic organelles like cardiomyocytes of human dilated cardiomyopathy. The heart failure-inducible TG are a useful model for dilated cardiomyopathy, and provided evidence indicating that myocardial cell loss through autophagic cell death plays a causal role in the pathogenesis of heart failure.
Cardiomyocyte death is observed in a number of pathological conditions such as ischemic or dilated cardiomyopathy, hypertensive heart disease, and aging (1). Oncosis has been recognized to be a principal mechanism of myocardial cell death, but during the last decade, much emphasis has been put on the importance of apoptosis on the basis of detectable apoptotic cardiomyocytes in animal and human models of heart failure (2). Recently, autophagic cell death (ACD) 1 has been demonstrated as another type of myocardial cell death in human failing hearts (3)(4)(5)(6)(7). Although a decline in pumping capacity initiated by cardiomyocyte loss is supposed to induce ventricular remodeling and finally results in symptomatic heart failure (8), there still remains a controversy over the pathogenic role of cell death in progression of heart failure (9,10). An intractable problem that hampers mechanistic insights is the low occurrence of myocardial cell death in failing hearts, although it differs strikingly according to the models examined and technical specificity (2,8). Furthermore, in human hearts, most samples were obtained from patients with end-stage heart failure who underwent heart transplantation and thus it remains unknown whether myocardial cell death occurs persistently from an early stage and is causative to progression of heart failure (8). To circumvent these obstacles, we established an inducible heart failure mouse model utilizing diphtheria toxin (DT)-mediated cell ablation, in which a given number of cardiomyocytes are arbitrarily and synchronously ablated, and prospective and serial analysis is available.
DT is a two-peptide protein consisting of fragments A (DT-A) and B (DT-B) produced by Corynebacterium diphtheriae (11). DT binds to the DT receptor on the cell surface through DT-B and is internalized into acidic endocytic vesicles, which allows release of catalytic DT-A into the cytoplasm (12). DT-A exerts its cytotoxicity by ADP-ribosylating elongation factor 2 and thereby inhibiting protein synthesis in infected cells (13). DT receptor has been identified as a precursor of heparin-binding EGF-like growth factor (pro-HB-EGF) (14,15). Intriguingly, DT cannot bind to rodent pro-HB-EGF because of substitution of amino acids required for DT binding, whereas primate pro-HB-EGF acts as a functional DT receptor (16). Therefore, specific cells in mice are ideally sensitized to DT by forced expression of human pro-HB-EGF (17).
To enable cardiac-specific cell ablation, we generated transgenic mice (TG) expressing human pro-HB-EGF in the hearts under the control of ␣-myosin heavy chain promoter. Administration of DT induced ablation of transgene-expressing cardiomyocytes, and subsequently symptomatic heart failure. In this mouse model of heart failure, autophagy but not apoptosis was the mechanism of cardiomyocyte death. Autophagy is a dynamic process and intracellular constituents are sequestered by membranes and subsequently degradated or recycled in lysosome or vacuole (18 -20). In this sense, autophagy is involved in maintaining cellular homeostasis and turnover in physiological conditions. However, a growing body of evidence suggests that autophagy is implicated in execution of pro-grammed cell death and is closely linked to several pathological conditions (21)(22)(23). Our model of experimentally induced heart failure provided direct evidence that myocardial cell loss through ACD causes heart failure, and will be useful to dissect the molecular mechanisms underlying structural and functional changes in heart failure.

EXPERIMENTAL PROCEDURES
Generation of Transgenic Mice-Human pro-HB-EGF cDNA (gift from A. Ullrich, Max-Planck-Institute of Biochemistry, Martinsried, Germany) was subcloned into the ␣-myosin heavy chain promotercontaining expression vector (gift from J. Robbins, Children's Hospital, Cincinnati, OH). The 6.9-kb DNA fragment was microinjected as a transgene into pronuclei of eggs from BDF1 mice, and the eggs were transferred into the oviducts of pseudopregnant ICR mice. The transgene was identified by Southern blot and PCR analysis. All protocols using mice were approved by the Institutional Animal Care and Use Committee of Chiba University.
Administration of Diphtheria Toxin-Diphtheria toxin (Sigma) was reconstituted in 10 mM sodium phosphate buffer (pH 7.4) containing 5% lactose, and was administered by intramuscular injection.
Western Blot Analysis-Protein samples were fractionated by SDS-PAGE, and immunoblot analysis was performed as described previously (26,28).
Transthoracic Echocardiography-Mice were anesthetized by intraperitoneal injection of a mixture of ketamine (100 mg/kg) and xylazine (5 mg/kg). Cardiac function was evaluated with echocardiography (SO-NOS 4500, Philips, Eindhoven, the Netherlands) using a 12-MHz transducer as described previously (26).
Histological Analysis and Immunohistochemistry-Hearts were fixed in 10% neutralized formalin and embedded in paraffin. Serial sections at 5 m were routinely stained with hematoxylin-eosin for morphological analysis, and with Masson's trichrome for detection of fibrosis. For measurement of the myocyte cross-sectional area, semithin sections with silver staining were analyzed. Suitable cross-sections were defined as having round-to-oval cardiomyocyte sections and nearly round-shaped capillaries that perfused in the region. For immunohistochemistry, Vectastain ABC kit (Vector Laboratories, Burlingame, CA) was used to detect the primary antibodies. The sections were counterstained with hematoxylin.
Evaluation of DNA Fragmentation-TUNEL assay using paraffin sections was performed with an in situ apoptosis detection kit (Takara Biomedicals, Otsu, Japan). For agarose gel electrophoresis for DNA fragmentation, genome DNA (10 g) was electrophoretically fractionated on a 1.5% agarose gel and stained with ethidium bromide as described previously (29). To induce apoptosis in spleens as positive controls, we injected lipopolysaccharide (40 mg/kg) (Sigma) in phosphate-buffered saline intraperitoneally into age-matched mice. Mice were sacrificed 12 h after lipopolysaccharide injection and spleens were excised (30).
Electronmicroscopy-Hearts were fixed in 3% paraformaldehyde, 2.5% glutaraldehyde, and 0.1 M cacodylate buffer (pH 7.4). After washing with the buffer solution and post-fixation in 1% OsO 4 and 0.1 M cacodylate buffer (pH 7.4), they were washed with the buffer solution, dehydrated using alcohol and acetone, and embedded in epoxy resin. Ultrathin sections were examined under the electron microscope (31). Statistical Analysis-All values are expressed as mean Ϯ S.E. Comparisons were made by Student's t test or one-way analysis of variance as appropriate. Values of p Ͻ 0.05 were considered statistically significant.

Inducible Myocardial Cell Ablation in TG Expressing
Human DT Receptor in the Hearts-To confer DT sensitivity to cardiomyocytes in mice, we generated TG expressing human DT receptor, pro-HB-EGF, under the control of the ␣-myosin heavy chain promoter (Fig. 1A). Of two independent founder lines with successful germline transmission, one line was chosen for further analysis on the basis of transgene expression levels. By immunoblot analysis using an antibody specific for human pro-HB-EGF, we confirmed cardiac-specific expression of the transgene (Fig. 1B). In situ hybridization analysis using a specific riboprobe for human pro-HB-EGF further revealed a mosaic expression pattern of the transgene in the hearts (Fig.  1C). Expression of the transgene was scattered throughout the TG hearts, and the number of transgene-expressing cardiomyocytes was 17.3 Ϯ 6.0% out of total cardiomyocytes.
To induce DT-mediated myocardial cell ablation, we administered DT by intramuscular injection to TG and wild-type mice (WT) at 10 weeks of age. When 5 mg/kg DT was administered, TG became lethargic and ϳ80% of TG died within 10 days after injection of DT, although WT appeared normal (Fig. 1D). Immunoblot analysis in combination with in situ hybridization analysis revealed that expression of human pro-HB-EGF in the TG hearts was significantly decreased on the next day of DT injection, and almost disappeared on the following day ( Fig.  1E). After 7 days, transgene-expressing cardiomyocytes were undetected in the TG hearts, suggesting that they were completely ablated through DT-mediated cytotoxicity.
DT-mediated Myocardial Cell Loss Caused Heart Failure in Mice-We next examined the geometric, functional, and histological changes in the hearts caused by DT-mediated myocardial cell damages. Gross inspections of the TG hearts 7 days after DT injection showed global chamber dilatation with marked wall thinning and atrial thrombus ( Fig. 2A), and the heart to body weight ratios were ϳ1.3-fold increased (Fig. 2B), whereas the mock-treated TG hearts and DT-or mock-treated WT hearts showed no geometric change (Fig. 2).
To evaluate cardiac function, we performed transthoracic echocardiographic examination. Seven days after injection, a ϳ1.3-fold increase in the left ventricular end-diastolic dimension and a 1.5-1.7-fold decrease in left ventricular wall thickness were observed in DT-treated TG, whereas these parameters were unchanged in mock-treated TG (Fig. 2, C and D) and DT-or mock-treated WT. Echocardiographic examination also demonstrated a 2.4-fold reduction in % FS in DT-treated TG. These results suggest that injection of DT induced deterioration of LV systolic function with chamber dilatation and ventricular wall thinning in TG. In contrast, no discernible phenotype was observed in TG in the absence of DT, and DT had no harmful effect on WT.
Hematoxylin-eosin staining of the histological sections of TG hearts 7 days after DT injection revealed degenerated cardiomyocytes surrounded by inflammatory cells (Fig. 3A). These pathological findings were not observed in WT with or without DT injection (data not shown). The infiltrating inflammatory cells were identified as macrophages by immunohistochemical analysis using anti-Mac-3 antibody (Fig. 3B). Histological sections with Masson's trichrome staining showed interstitial fibrosis in DT-treated TG hearts (Fig. 3A). Silver staining of the sections of TG hearts on 14 days after DT injection revealed a 1.9-fold increase in cross-sectional areas of cardiomyocytes (Fig. 3C), indicating that the cardiomyocytes, which did not express the transgene and survived DT administration, underwent hypertrophic cell growth.
Alterations of Gene Expression in TG Presenting DT-induced Heart Failure-To characterize the molecular basis of heart failure caused by DT-induced myocardial cell ablation, we examined expression levels of several molecular markers. Expression of BNP was up-regulated 1 day after DT injection, and persistently elevated thereafter (Fig. 4). Increased expression of skeletal ␣-actin and decreased expression of SERCA2 were evident 4 days after DT injection (Fig. 4). Up-regulation of natriuretic peptide genes and fetal cardiac genes including skeletal ␣-actin is one of the characteristic cellular responses observed during cardiac hypertrophy (32,33). Especially, ventricular expression of BNP is induced promptly in response to volume expansion and pressure overload, and plasma BNP concentrations have proven to be valuable for diagnostic and prognostic assessment in patients with heart failure (34). In addition, down-regulation of SERCA2 has been reported to be a sensitive marker for heart failure (35). Therefore, these patterns of cardiac gene expression indicated that DT-induced myocardial cell ablation burdened hemodynamic overload and progressed overt heart failure concomitantly with cardiac hypertrophy.
Consistent with the histological finding of infiltration by macrophages, an increase in expression of MCP-1 was observed 1 day after DT injection, and expression levels of MCP-1 were further increased until 4 days and declined on 7 days (Fig. 4). The expression levels of TNF-␣, encoding an inflammatory cytokine produced by macrophages, changed in parallel with that of MCP-1 (Fig. 4). Insomuch as symptomatic heart failure was evident 7 days after DT injection, stressed myocardium could be another source of TNF-␣ production at this period. Following up-regulation of inflammatory markers, expression levels of the collagen genes (Col1a2 and Col3a1) were increased at 4 days after DT injection (Fig. 4). These temporal profiles of Expression of human pro-HB-EGF in TG was observed specifically in the hearts (right). B, brain; H, heart; Li, liver; K, kidney; Sp, spleen; Sk, skeletal muscle; Te, testis. C, in situ hybridization analysis using a riboprobe specific for human pro-HB-EGF. Transgene was expressed in a mosaic pattern, and cardiomyocytes expressing the transgene were 17.3 Ϯ 6.0% of the total cardiomyocytes within TG hearts. D, Kaplan-Meier survival curves of control mice (WT treated with mock or DT and TG treated with mock, n ϭ 21, respectively) and TG (n ϭ 21) treated with intramuscular injection of DT. E, complete ablation of transgene-expressing cardiomyocytes following DT injection revealed by immunoblot (left) and in situ hybridization analysis. Expression of pro-HB-EGF was remarkably diminished on day 2 (D2) and was completely undetected on day 7 (D7). gene expressions suggest that mobilization of macrophages are induced by up-regulated MCP-1 after the myocardial cell ablation, leading to cardiac fibrosis by enhanced production of inflammatory cytokines, and that inflammatory cytokines and cardiac remodeling might promote left ventricular dysfunction evoked by myocardial cell loss.
Myocardial Cell Death Induced by DT Is Not Primarily because of Apoptosis-To investigate the mechanisms of myocardial cell death in DT-treated TG hearts, we first performed a TUNEL assay. In the hearts of DT-treated TG, we could not detect any TUNEL-positive cardiomyocytes or inflammatory cells, whereas a marked increase in TUNELpositive cells was detected in the spleen of mice treated with intraperitoneal administration of lipopolysaccharide as positive controls (Fig. 5A). Likewise, analysis of genomic DNA by agarose gel provided no evidence of DNA laddering in DTtreated TG hearts (Fig. 5B). We further examined activation of caspase 3 (Fig. 5C), changes in expression of proapoptotic and antiapoptotic Bcl2 family proteins (Fig. 5D), and cytochrome c release from mitochondria (Fig. 5E), but biochemi- cal changes leading to typical apoptosis were not observed in DT-treated TG hearts.
Autophagy Is the Mechanism of Myocardial Cell Death in DT-induced Failing Hearts-ACD is a regulated process of caspase-independent programmed cell death, in which intracellular components are degradated by lysosomal or proteasomal proteases (21)(22)(23). Immunohistochemical analysis revealed positive staining for lysosomal protease cathepsin D, LAMP-1, and ubiquitin in cardiomyocytes of DT-treated TG hearts (Fig. 6A). It is notable that cathepsin D showed a diffuse cytosolic distribution, whereas LAMP-1 showed a granular localization. These results suggest an increase in formation of lysosomes and leakage of activated lysosomal enzymes into the cytosol. Recently, it has been reported that, in human failing hearts, ubiquitin accumulation in cardiomyocytes may be associated with up-regulation of ubiquitin-conjugating enzyme E2 (UBC2) and down-regulation of deubiquitinating enzymes such as UFD1 and isopeptidase T (6). However, Western blot analysis revealed that the amounts of the ubiquitin-activating enzyme E1, UBC2, ubiquitin-ligating enzyme E3 (E6-AP), and UFD1 were not unchanged in DT-treated TG hearts when compared with control hearts (Fig. 6B).
Electron microscopic analysis revealed abundant cytosolic vacuoles and segmented configuration of nuclei with lumpy con- densation of chromatin in degenerated cardiomyocytes in TG hearts 3 days after DT injection (Fig. 7, A and B). However, nuclear fragmentation and crescent-shaped chromatin condensation at the nuclear periphery typical of apoptosis were not observed. In higher magnifications, cytosolic vacuoles containing lipid droplets with myelin figures and degenerated mitochondria (Fig. 7, C and D), suggested that these vacuoles are typical autophagosomes. In terminally degenerated cardiomyocytes, lysis of myofibrills and other intracellular organelles were prominent (Fig. 7E). These findings suggest that myocardial cell loss in DT-treated TG hearts is primarily because of ACD.
Autophagic degeneration has been implicated in human failing heart failure (3-7). We also found cardiomyocytes undergoing autophagic cell death in a biopsied specimen from a 43year-old patient suffering from dilated cardiomyopathy. Similar to the electron micrographic findings in degenerated cardiomyocytes in our mouse model of heart failure, myofibrillar degeneration in association with cytosolic vacuoles and lipid droplets were observed in this biopsied specimen (Fig. 7F). The vacuoles were autophagosomes containing digested organelles. These findings are illustrative of the previously reported features characteristic of ACD in human heart failure. Therefore, degenerated cardiomyocytes in human dilated cardiomyopathy patients showed ultrastructural alterations similar to those in our mouse model of heart failure. DISCUSSION In this study, we generated a novel mouse model of heart failure, where cardiomyocyte loss through ACD is arbitrarily and specifically induced by intramuscular injection of DT. Recent technical progress in genetic manipulation and physiological measurements enabled us to produce several mouse models of heart failure (36). These models have improved our understanding of pathophysiology of heart failure and have been of great help for establishment and evaluation of new therapeutic approaches. Genetically engineered mice, in particular, expanded the list of gene products that are involved in generation or progression of heart failure, but they have specific limitations as animal models. For example, mice homozygous for muscle LIM protein (MLP) develop cardiomyopathy and heart failure, but the clinical courses of individual mice are divergent because the penetrance of phenotype is influenced by genetic backgrounds (37). About half of the MLP-deficient mice suffer from severe congestive heart failure and die during the second postnatal week, but the rest of the MLP-deficient mice survive to adulthood and are viable. Development of cardiomyopathy is also identified in tropomodulin-overexpressing transgenic mice (38). In this model, severe signs of heart failure are observed between 2 and 4 weeks after birth and most symptomatic mice die within a few days. The phenotypes of these model mice are primarily genotype-dependent, but are susceptible to the effects of genetic backgrounds. In addition, a difficulty in morphological and biochemical approach in studying a role of cardiomyocyte death in heart failure arises from the low occurrence of cell death in these models (2). In our mouse model of heart failure, cardiomyocytes expressing the DT receptor are selectively and simultaneously damaged by administration of DT, and this advantageous feature not only makes it possible to induce symptomatic heart failure arbitrarily but also provides insights into the roles of cell death in heart failure.
Our model appears conceptually similar to the one reported in the earlier work (39), in which DT-A expression is regulated by a tetracycline-responsive promoter. In that model, induction of DT-A in the hearts resulted in congestive heart failure as well. However, a leaky induction is occasionally observed in this tetracycline-inducible system, and subtle expression of DT-A might have nonspecific and undesirable effects, insomuch as the toxicity of DT-A is extremely high (17). In our model, the DT receptor, not injurious in the absence of DT, was expressed in the hearts, and myocardial cell ablation was specifically and ideally achieved.
Temporal histological analysis and profiling of gene expression revealed a series of cellular events that finally evoked heart failure (Figs. 3 and 4). Transgene expression was dramatically reduced in a few days after DT injection (Fig. 1E), suggesting that cardiomyocyte death occurs during that period. Following cardiomyocyte death, inflammatory cells infiltrated and produced inflammatory cytokines. Around 7 days, hemodynamic deterioration with apparent cardiac remodeling induced symptomatic heart failure. These findings strongly suggest that myocardial cell death causes symptomatic heart failure. In addition, our model allowed quantitative analysis of cardiomyocyte death. In situ hybridization analysis revealed that expression of the transgene was scattered diffusely and observed in 17.3 Ϯ 6.0% of cardiomyocytes in TG hearts. Because there was no cardiomyocyte expressing the transgene on day 7 after DT injection, all of the transgene-expressing cells (ϳ17% of cardiomyocytes) might be dead. These results suggest that loss of this population is sufficient to produce symptomatic heart failure, and this estimation is consistent with the notion that a diffuse loss of 10 -20% of cardiomyocytes accounts for cardiac failure, whereas equivalent cardiac failure is produced by a segmental loss of 40 -50% of cardiomyocytes after coronary artery occlusion (40).
Electron microscopic analysis revealed that, in our mouse model, damaged cardiomyocytes showed abundant cytoplasmic autophagosomes and chromatin condensation with more complex and lumpier shapes than in apoptosis, both of which are characteristic of ACD (Fig. 7). ACD is defined as a regulated pathway of cytoplasmic degradation executed by lysosomal and proteasomal proteases (21)(22)(23)41). Consistently, elevated expressions of cathepsin D and ubiquitin were recognized in DT-treated TG cardiomyocytes (Fig. 6A). In contrast, biochemical signals inducing apoptosis were not activated. ACD often occurs in large, cytoplasmic-rich, and post-mitotic cells, and has been implicated in several human diseases (19,(21)(22)(23)41). For example, ACD is associated with neurodegenerative diseases such as Parkinson's disease (42), Huntington's disease (43), and Alzheimer's disease (44). Degenerated cardiomyocytes displaying morphological features characteristic of ACD have recently been reported to exist in the failing hearts of patients with dilated cardiomyopathy, valvular heart diseases, congenital heart diseases, and hypertensive heart diseases ( Fig. 7F) (3-7). A recent paper demonstrated that cardiomyocyte apoptosis of rare occurrence comparable with that observed in human failing hearts were sufficient to cause lethal cardiomyopathy in transgenic mice with a ligand-induced caspase-8 activation in the hearts (45). It is noteworthy that autophagic cardiomyocyte death has been reported to be detectable more frequently than apoptosis or oncosis in failing or hemodynamically overloaded human hearts (3,6,7), although the frequency of the each form of cell death may be influenced by disease stages and backgrounds (e.g. age, etiology, clinical feature, and treatment) of the examined samples (2).
Recent studies have indicated a significant role of lysosomal cysteine and aspartic protease cathepsins in execution of ACD (21,46). PC12 cells cultured in a serum-deprived condition showed morphological features characteristic of ACD with an increase in proteolytic activity of cathepsin D (46), and forced expression of cathepsin D in PC12-induced rapid cell death, indicating a regulatory role of cathepsin D in ACD. Expression of cathepsin D was increased in DT-treated TG hearts (Fig. 6A). In damaged cardiomyocytes, translocation of cathepsin D from lysosomes to the cytosol was evident, which is indicative of enhanced proteolytic activity (47). Cathepsin D has been reported to be activated in human failing hearts (5) and also to be a positive mediator of apoptotic cell death (22,47). Further experiments are required to clarify whether cathepsin D is involved in execution of myocardial ACD in the pathogenesis of heart failure and how cathepsin D regulates these modes of cell death differentially.
According to the recent studies, ubiquitin-dependent protein degradation is linked to autophagy (48). Accumulation of ubiquitin in cardiomyocytes was observed in DT-treated TG hearts (Fig. 6A) as well as in human failing hearts (4, 6, 7). Kostin et al. (6) demonstrated a functional defect in the ubiquitin/proteasome pathway together with ubiquitin accumulation in human failing hearts, and speculated that an excess of ubiquitinated proteins might activate autophagic protein degradation. The precise molecular mechanisms of how ubiquitin accumulation enhances autophagy remain unknown, although ubiquitination is postulated to be required for maturation of autophagosomes (48). Unlike human failing hearts (6), the amounts of UBC2 and UFD1 were unchanged in the hearts of our model mice (Fig. 6B). Protein ubiquitination and deubiquitination are mediated by a large number of enzymes (49), and it is an important issue to be FIG. 7. Electron microscopic analysis of the hearts after DT injection. Electron microscopic analysis before (A) and after DT injection (B-E). In cardiomyocytes of DT-treated TG hearts, nuclear morphological changes associated with lumpy chromatin condensation were observed (arrow in B). Abundant vacuoles of various sizes in the cytoplasm (B) were typical autophagosomes containing degenerating lipid droplets and mitochondria (C and D). Arrows (in C and D) indicate myelin figures. Myofibrillar lysis (arrows) was observed in cardiomyocytes undergoing autophagic cell death (E). Electron microscopic analysis of biopsied myocardium from a patient with dilated cardiomyopathy (F) revealed cardiomyocytes containing many autophagosomes with degenerating mitochondria (arrows). addressed in the future how ubiquitin is accumulated in cardiomyocytes undergoing autophagic cell death.
Diphtheria is a communicable disease affecting the upper respiratory tract and occasionally the skin as primary infection (50). However, remote organs such as heart or peripheral nerves are often damaged when DT is absorbed into the systemic circulation. Although diphtheria infection has been rarely encountered in developed countries because of the high rates of vaccination after the mid 1960s, several sporadic outbreaks occurred, for example, in the former Union of Soviet Socialist Republics in 1990s (50). Myocardial involvement is a major complication that determines the prognosis, but the pathophysiology associated with diphtheritic cardiomyopathy remains largely unknown. Histological analysis of the postmortal heart of a diphtheria patient revealed hyaline degeneration of myocardium and infiltration of mononuclear cells, and cytosolic lipid droplets and clumped chromatin granules were observed by electron microscopic examination (51). Although vacuoles of autophagosomes were not described in this paper, these findings were similar to those observed in our mouse model. Interestingly, cardiomyocyte loss through ACD associated with a decrease in protein synthesis was also observed in anthracycline-induced cardiomyopathy (52). Therefore, autophagic cardiomyocyte death evoked by a decrease in protein synthesis might not be confined to diphtheritic cardiomyopathy, but a more generalized phenomenon that occurs during progression of heart failure arising from miscellaneous etiologies. In this regard, our model of experimentally induced heart failure will be useful to elucidate molecular mechanisms underlying structural and functional changes associated with ACD in heart failure.