Hyperglycemia-induced cardiomyocyte death is mediated by lysosomal membrane injury and aberrant expression of Cathepsin D

Abstract

Hyperglycemia is an independent risk factor for diabetic heart failure. However, the mechanisms that mediate hyperglycemia-induced cardiac damage remain poorly understood. Previous studies have shown an association between lysosomal dysfunction and diabetic heart injury. The present study examined if mimicking hyperglycemia in cultured cardiomyocytes could induce lysosomal membrane permeabilization (LMP), leading to the release of lysosome enzymes and subsequent cell death. High glucose (HG) reduced the number of lysosomes with acidic pH as shown by a fluorescent pH indicator. Also, HG induced lysosomal membrane injury as shown by an accumulation of Galectin3-RFP puncta, which was accompanied by the leakage of cathepsin D (CTSD), an aspartic protease that normally resides within the lysosomal lumen. Furthermore, CTSD expression was increased in HG-cultured cardiomyocytes and in the hearts of 2 mouse models of type 1 diabetes. Either CTSD knockdown with siRNA or inhibition of CTSD activity by pepstatin A markedly diminished HG-induced cardiomyocyte death, while CTSD overexpression exaggerated HG-induced cell death. Together, these results suggested that HG increased CTSD expression, induced LMP and triggered CTSD release from the lysosomes, which collectively contributed to HG-induced cardiomyocyte injury.

Graphical Abstract

1. Introduction

Hyperglycemia is an independent risk factor for diabetic cardiovascular complications. Indeed, high glucose increases reactive oxygen species, which has been suggested as a major causative factor in the pathogenesis of diabetic cardiomyopathy [1]. This hypothesis is strongly supported by the ability of various antioxidants to reduce diabetic heart injury in animal studies[2]. However, clinical trials have shown very limited effects of antioxidant therapy on cardiovascular outcomes including heart failure[3], suggesting a need to more thoroughly understand the cellular and molecular mechanisms that mediate diabetic cardiomyopathy and heart failure.

The lysosome is a cellular organelle that contains proteolytic enzymes to break down large molecules and cellular debris. The autophagy-lysosomal degradation pathway is a cytoplasmic quality control system which is extremely important for maintaining cardiac homeostasis [4]. Altered autophagy has been observed in diabetic mouse hearts[5], which is often associated with lysosomal dysfunction [6]. The contribution of diminished lysosomal function to cardiac hypertrophy [7] and myocardial aging[8] has been well documented. However, it remains unclear whether lysosomal dysfunction is causatively involved in the pathogenesis of diabetic cardiomyopathy.

A variety of cellular stresses can cause lysosomal damage leading to different degrees of lysosomal membrane permeabilization (LMP)[8]. Severe LMP can lead to the release of lysosomal proteases including cathepsins from the lysosomal lumen into cytosolic compartment, triggering cell death[9]. This has been strongly supported by the ability of lysosome inhibitors to prevent cell death[9]. Cathepsins include 11 cysteine proteases, 2 serine proteases and 2 aspartate proteases. Cathepsin D (CTSD) is the only lysosomal aspartic protease that is ubiquitously expressed, suggesting that it may have some unique functions that cannot be performed by any other cathepsins. Most lysosomal enzymes easily lose their activities in non-acidic pH environments such as the cytosol. However, CTSD is very unique in that it is still functional in the cytosol with multiple physiological substrates[10]. It is thus not surprising that CTSD appears to be the major lysosomal protease that mediates cell injury[11].

CTSD activity was increased in serum from patients with diabetes [12], but it was decreased in diabetic mouse hearts [13, 14]. However, the protein expression levels of CTSD and its subcellular localization in the diabetic heart have not been investigated. It remains unclear whether CTSD plays a role in diabetic cardiac injury. Accordingly, our present study determined CTSD expression levels and its distribution in diabetic mouse hearts and in cardiomyocytes cultured under high glucose conditions. We also determined the functional significance of altered CTSD in high glucose-induced cardiomyocyte injury.

2. Materials and methods

2.1. Neonatal Rat Ventricular Cardiomyocyte Culture and High Glucose Treatment

Neonatal rat ventricular cardiomyocytes (NRVC) were isolated from 0-2 day old Harlan Sprague-Dawley rats. The heart tissue was digested with trypsin (Sigma T4799). The resulting cells were plated onto an uncoated 100mm dish for fibroblasts to attach. Cardiomyocytes were lifted, re-plated onto gelatin-coated 60mm dishes, and cultured overnight in Dulbecco’s modified essential medium (DMEM; Corning Cellgro, 10-017CV) with 10% fetal bovine serum and 5’-bromo 2’-deoxy-uridine (BrdU; Sigma, B5002). The cells were washed the following day with phosphate-buffered saline and changed to serum-free DMEM. For high glucose studies, cardiomyocytes were cultured for 72h in glucose-free DMEM (GIBCO, 11966) supplemented with 5.5, 17, or 30 mmol/liter of glucose (Sigma, G7021). The osmolarities of all media were made equal to 30mM by adding different amounts of mannitol (Sigma, M9647).

2.2. Antibodies and Reagents

The antibodies against PARP (#9542/AB_2160739), cleaved Caspase 3 (#9664/AB_2070042) β-Actin (#4967/AB_330288) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, #5147/AB_10622025) were purchased from Cell Signaling Technology. The antibodies against Cathepsin D (sc6486/AB_637896, sc10725/AB_2292414) and LAMP-1 (sc8098/AB_2134494) were obtained from Santa Cruz Biotechnology. Alexa-Fluor™ Antibodies (A11006/ AB_141373, A11008/AB_143165, A21431/AB_2535852) were from Thermo Fisher Scientific. LysoSensor™ Green DND-189 (L7535) and Lysotracker™ Blue DND-22 (L7525) were purchased from Thermo Fisher Scientific and were dissolved in absolute ethanol (Fisher Scientific, AC61508). Pepstatin A was from RPI (P30100) and dissolved in dimethyl sulfoxide (DMSO; Sigma, 472301).

2.3. Western Blot Analysis

Protein extracts from cultured cardiomyocytes were prepared as described previously[15], except that 1% of protease inhibitor cocktail (Sigma, P8340) was added to the extraction buffer. Protein samples were subjected to SDS-polyacrylamide gel electrophoresis, transferred to polyvinylidene difluoride membrane (Amersham Pharmacia, RPN3031), and then blocked in 5% milk prepared with 0.1% Tween-20 Tris-buffered saline (TBST) for 1 h.. The membrane was incubated at 4 °C in 2.5% milk with Primary antibodies overnight and an appropriate horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotech, sc-2004, sc-2005, sc-2020, and sc-2438) for 1 h and processed for chemiluminescent detection using Lumigen ECL Ultra (TMA-6 Lumigen, MI, USA) and Amersham Imager 600.

2.4. Replication-deficient Adenoviruses

Adenovirus encoding human Cathepsin D was created as previously described[15]. The human CTSD cDNA was amplified by PCR from an ORF clone (OriGene, SC118936). The tfGalectin3 reporter (tfGal3) plasmid was kindly provided by Dr. Tamotsu Yoshimori[16]. Each insert was sub-cloned into pShuttle-CMV vector at NotI-XhoI sites. Recombinant adenoviruses were then generated using the AdEasy Adenoviral Vector System (Stratagene, 240009). Cardiomyocytes were infected with adenovirus at a multiplicity of infection of 100 plaque forming unit for 18h before glucose treatments.

2.5. Assessment of Lysosomal pH with Fluorescent Microscopy

LysoSensor™ Green is a fluorescent pH indicator that exhibits a pH-dependent increase in fluorescence intensity upon acidification; it is almost nonfluorescent except when inside acidic compartments. LysoSensor dye was added directly to the culture media at 2μM. After incubation in a CO₂ chamber for 30 minutes, the cells were washed with DMEM and the live cell images were captured with a fluorescent microscope (OLYMUS-IX71) under green fluorescence filter using 40x objective lens.

2.6. RNA isolation and RT-PCR

The reverse transcriptase-PCR (RT-PCR) was carried out using the TaqMan™ Reverse Transcription Reagent (Cat. # N8080234, Thermo Fisher Scientific). Briefly, 1 μg of total RNA isolated with TRIzol Reagents (Cat. # 15596026, Thermo Fisher Scientific) from NRVCs was used for first-strand cDNA synthesis using Random Hexamers as the primer. Then 2 μl of the reverse transcription reaction was used as cDNA template for semiquantitive PCR amplification in a volume of 50 μl containing gene-specific primers for CTSD, CTSL, or CTSB. Aliquots of PCR reaction were taken at different cycles for agarose gel analysis to determine the linear range of amplification. All reactions were run on a 1.5% agarose gel. The sequences of the primers are as follows: CTSD (Accession number NM134334.2) forward, 5’-GGCAACCTGGAGGAGAACTAA-3′, reverse, 5′-TTGGCAAAGCCGACCCTAT-3’; CTSL (NM013156.1) forward, 5’-GGCTATGGTTATGAAGGAACA-3’, reverse, 5’-TTCGGATGTAGTGTCCGTCA-3’; CTSB (NM022597.2) forward, 5’-CTGGGGTGATAATGGTTTCTT-3’, reverse, 5’-AGGTAGGGTGGCTCTGATG-3′.

2.7. Knockdown of Cathepsin D (CTSD)

The siRNAs targeting CTSD (NP599161) or control siRNA were transfected into NRVCs at 50nM using Lipofectamine RNAiMAX Transfection Reagent (ThermoFisher Scientific). Culture media were replaced 24hrs later by serum-free DMEM with the indicated glucose conditions. The Silencer® Pre-Designed siRNAs targeting CTSD (Cat. # AM16708 IDs: s191000, 191001, 191002) and the Silencer™ Select Negative Control siRNA (Cat. #4390846) were obtained from Thermo Fisher Scientific.

2.8. Propidium iodide staining

Cardiomyocyte death was determined by staining with propidium iodide (PI, 1 μg/ml, Sigma 81845). PI was added directly to the culture medium 10 minutes prior to imaging, and cardiomyocytes were examined and photographed with a fluorescent microscope (OLYMUS-IX71) under both phase contrast and fluorescent conditions. The images were captured at 10x or 20x magnification. PI positives cells were manually counted and calculated as the percentage of the total cells examined. The experiments were repeated at least three times in triplicate culture dishes.

2.9. Immunofluorescence staining

Cardiomyocytes were fixed with 4% paraformaldehyde in PBS for 10 minutes and permeabilized with 1% Triton-X100 for 2 minutes. The cells were blocked with 2% bovine serum albumin (A7070 Sigma-Aldrich) for 1 hour and incubated with primary antibodies against CTSD and LAMP1 overnight. After washing, the cells were incubated with the respective Alexa Fluor™-conjugated secondary antibodies for one hour and examined with a confocal microscope.

2.10. Confocal Microscopy

Cardiomyocytes cultured in 12-well plate with gelatin coated cover slip were infected with AdtfGalectin3 for 18 hours, and cultured in DMEM for 72h. Cells were fixed with 4% paraformaldehyde for 10 minutes. The coverslips were mounted on microscope slide (Cat. 22-230-900 Thermo Fisher Scientific, MA) with fluoromount G (Cat. 17984-25 Electron Microscopy Sciences, PA) and observed with a laser scanning confocal microscope (Nikon C2). The confocal images were captured at 60x magnification. The numbers of RFP puncta were counted manually from at least three independent experiments in triplicate dishes as described[15].

2.11. Cathepsin D enzymatic activity assay

The proteolytic activities of CTSD were determined by using a fluorescence-based assay kit (Cat. ab65302 Abcam, MA). Briefly, 1x10⁶ cells were collected and lysed with 50uL of Cell Lysis Buffer followed by washing with cold PBS. The lysates were mixed with 52uL Reaction Buffer containing the Substrate and incubated in dark at 37 °C for one hour. The activity was measured by using a fluorescence microplate reader at Ex/Em = 328/460 nm. The relative fluorescence units (RFU) were normalized with total protein amount of each sample. The CTSD activities in the culture media were also determined as an indicator of CTSD leakage from the cells.

2.12. Mouse models of Type 1 Diabetes

Two-month old C57BL/6 mice were treated with three doses of streptozotocin (STZ, 50 mg/kg/d for 3 days) to induce diabetes. Fasting blood glucose content of 15 mM or greater was considered diabetic, and vehicle-treated mice were used as non-diabetic controls. Another type 1 diabetic model is the OVE26 mice which are an inbred diabetic line with early onset hyperglycemia and insulinopenia resulting from calmodulin overexpression in pancreatic beta cells[17].

2.13. Statistical Analysis

Quantitative data were presented as the Means ± SD. Differences between experimental groups were examined by t-test and one-way or two-way analysis of variance (ANOVA) followed by the Bonferroni post-test using Prism software (GraphPad). For all analysis, p < 0.05 were considered statistically significant.

3. Results

3.1. High glucose induced lysosomal membrane permeabilization (LMP) in cardiomyocytes

Our previous results suggested that high glucose impaired the autophagy-lysosome system in cardiomyocytes [15]. Lysosomal membrane permeabilization (LMP) is a prominent feature of lysosomal dysfunction which can lead to cellular damage under certain conditions [15, 17, 18]. LMP is characterized by the perturbation of membrane integrity and the consequent leakage of the lysosomal contents such as ions and proteins. We used the LysoSensor™ Green and Galectin-3-RFP to assess the lysosomal acidity and membrane integrity of NRVCs. The LysoSensor™ Green is a fluorescent pH indicator that exhibits a pH-dependent increase in fluorescence intensity upon acidification. Thus, LysoSensor labeled the functional acidic lysosomes as bright puncta on the fluorescent images of NRVC (Fig.1A). Compared with 5.5 mM glucose, high glucose (HG, 30 mM) markedly reduced the number and the fluorescent intensity of the puncta, suggesting that HG decreased the lysosomal acidification (Fig.1A). Galectin-3 is a cytosolic carbohydrate-binding lectin that is recruited to damaged lysosomes and has been used as a marker of lysosomal membrane injury [19]. As shown in Fig. 1B, Galectin-3-RFP was diffusely distributed with very few puncta at 5.5 mM glucose. However, high glucose led to an increased accumulation of Galectin-3-RFP puncta, indicating increased lysosomal membrane injury. Galectin-3-RFP puncta colocalized with lysosomes stained by Lysotracker blue (Fig. 1C). These results suggested that high glucose treatment induced LMP and lysosomal damage.

Figure 1. High glucose (HG) induced LMP and triggered CTSD release in cardiomyocytes.

Neonatal Rat Ventricle Cardiomyocytes (NRVC) were cultured in DMEM media containing 5.5 mM or 30 mM glucose for 72hrs. (A) LysoSensor™Green (2 uM) was added to the culture dish and the cells were imaged 30 min later under a fluorescent microscope. (B) NRVC were infected with AdtfGalectin3, cultured for 72 hrs and imaged with the Nikon C2 confocal microscope. (C) Galectin-3-RFP puncta colocalized with Lysotracker blue-stained lysosomes. NRVC were infected with AdtfGalectin3 and cultured for 72 hrs. The cells were incubated with Lysotracker™ Blue dye at 100nM for 30 minutes and imaged using a fluorescent microscope. (D) NRVCs were cultured in DMEM for 72hrs, co-immunostained with antibodies against CTSD and LAMP1 and observed under the Nikon C2 confocal microscope. (E) NRVCs were cultured in DMEM for 72h and stained with CTSD antibodies. CTSD stained area was measured by ImageJ software and expressed as a percentage of the total cell area. (F) The CTSD enzymatic activity in culture media was determined. Both the LysoSensor™Green puncta and the Galectin3-RFP puncta were manually counted in at least 5 myocytes. All above experiments were repeated three times in triplicate culture dishes. All data were expressed as mean ± S.D, n≥3, *p<0.05. Scale bar represents 20um.

3.2. High glucose (HG) induced CTSD release from the lysosome

LMP initially allows passage of small molecules such as ions, but eventually leads to the release of lysosomal proteases from the lysosomal lumen into the cytosolic compartments[9]. We examined the subcellular localization of CTSD using co-immunostaining with lysosomal marker protein (Fig.1D). Most CTSD (red) was colocalized with lysosomal membrane protein LAMP1 (green) in NRVC cultured at 5.5 mM glucose, indicating that CTSD was normally restricted in the lysosome. However, HG treatment induced the leakage of CTSD into the cytosol as shown by its relocation from granular localization to a more diffused distribution in many areas which was no longer limited to the lysosomal lumen. As shown in Figure 1E, the anti-CTSD antibody-stained puncta in NRVC appeared mostly sharp-edged at 5.5 mM glucose, but the puncta looked swollen with diffused edges at HG. The total area stained by CTSD was markedly increased by HG as shown in the bar graph. Strikingly, HG was able to induce CTSD release from the lysosome not only into the cytosol but also into the culture media as shown by the increased CTSD enzymatic activity in the media (Fig. 1F). Together, these results suggest that HG induced lysosomal dysfunction in NRVC, which was characterized by altered pH, occurrence of LMP and CTSD leakage. These changes may contribute to HG-induced myocyte death as suggested before in other cell types.

3.3. CTSD expression is increased in high glucose (HG)-treated NRVCs and the diabetic heart

Except for inducing CTSD leakage, HG may affect the expression levels of CTSD to cause cardiomyocyte injury. CTSD is initially made as an inactive 43 kDa preprocathepsin D that is cleaved and glycosylated to form a 46 kDa procathepsin D and then further cleaved to produce mature and enzymatically active single chain forms (34 kDa and 15 kDa). As shown in Fig. 2, HG (17 or 30 mM) dramatically increased the protein levels of the 34 kDa mature CTSD (p<0.01). HG also elevated the mRNA expression levels of CTSD, but it did not affect the expression of cathepsin B (CTSB) or L (CTSL). The increased CTSD expression correlated with enhanced CTSD enzyme activity (Fig. 2C), indicating that high glucose induced CTSD expression and activity in cardiomyocytes. Cathepsins have been implicated in diabetic cardiac injury [12-14]. We determined the CTSD expression levels in the hearts of two mouse models of types 1 diabetes. As shown in Fig.2D, the mature forms of CTSD were dramatically increased in the diabetic hearts from either STZ-injected mice or OVE26 mice, consistent with the results from cardiomyocytes cultured under HG.

Figure 2. HG increased the expression of cathepsin D in cardiomyocytes and in diabetic mouse heart.

NRVC were cultured in DMEM for 72 hours. (A) CTSD protein levels were determined by Western blot analysis and quantified by ImageJ. The results were normalized to GAPDH, expressed as the mean ± SD, and analyzed by one-way ANOVA. N=6, *p<0.01 vs 5.5mM Glucose. (B) The mRNA levels of CTSD, CTSL and CTSB were assessed by semi-quantitative reverse transcriptase-PCR (RT-PCR). Shown is an agarose gel image stained with ethidium bromide. (C) The proteolytic activity of CTSD in NRVC was determined and expressed as the mean ± SD and analyzed by student t-test. N=3, *p<0.01 vs 5.5mM Glucose. (D) The heart tissues were taken from STZ or OVE26 diabetic mice and compared with their corresponding non-diabetic mice. Western blot analyses of CTSD were quantified by densitometry and results are presented as means ± SD. Values were normalized with βActin and analyzed by student t-test. n=4, *P < 0.05 as compared with non-diabetic mice.

3.4. CTSD mediated high glucose-induced cardiomyocyte injury.

If the increased expression and aberrant distribution of CTSD were causatively involved in high glucose toxicity, reducing CTSD levels or inhibiting its activity would protect cardiomyocytes. Indeed, siRNA-mediated CTSD knockdown was able to reduce HG-induced cardiomyocyte death as determined by the cleavage of Poly (ADP-ribose) polymerase (PARP) and caspase 3 (Fig. 3A), and Propidium iodide staining (Fig. 3B). Similarly, the aspartic protease inhibitor pepstatin A markedly attenuated HG-induced cardiomyocyte death (Suppl Fig. 1), suggesting a necessary role for CTSD in HG cardiotoxicity. In addition, as shown in Fig.4, overexpression of CTSD by adenovirus significantly increased HG-induced PI positive cardiomyocytes and tended to elevate HG-induced apoptosis as assessed by cleaved caspase 3 and PARP (p=0.10). Collectively, these results suggest that HG induced LMP, allowing CTSD to leak out from the lysosomes, ultimately leading to cardiomyocyte injury and death.

Figure 3. siRNA-mediated Knockdown of CTSD alleviated HG-induced Cardiomyocyte Death.

NRVC were transfected with siRNAs targeting CTSD or control siRNA for 24 hrs, and then cultured for 72 hrs. (A) Western blot analyses of CTSD, cleaved caspase 3 and PARP. Results are presented as means ± SD. Values were normalized sequentially with GAPDH and control siRNA-transfected samples at 5.5 mM glucose, and analyzed by one-way ANOVA. n=4, *P < 0.05 as compared with control siRNA-transfected samples at 30 mM glucose. (B) Propidium iodide (PI) was added to the culture media for 10 minutes to stain the dead cells. The images were captured under both phase contrast and fluorescent conditions. PI positives cells were manually counted and calculated as the percentage of the total number of cells examined. The experiments were repeated multiple times in triplicate culture dishes. The results were analyzed by one-way ANOVA and expressed as mean ± SD. N=6, P< 0.01 as compared with control siRNA-transfected samples at 30 mM glucose.

Figure 4. Overexpression of CTSD exaggerated cardiomyocyte death.

NRVC were infected with adenovirus encoding CTSD or beta-galactosidase (βgal) and cultured for 72 hrs. (A) Western blot analyses of CTSD, cleaved caspase 3 and PARP. Results are means ± SD. Values were normalized sequentially with Ponceau S staining and Adβgal-infected cells at 5.5 mM glucose, and analyzed by one-way ANOVA. n=4, *P < 0.05 as compared with Adβgal-infected cells at 5.5 mM glucose. (B) Propidium iodide (PI) was added to the culture media to stain the dead cells. The images were captured under both phase contrast and fluorescent conditions. PI positives cells were manually counted, calculated as the percentage of the total number of cells (mean ± SD) and analyzed by one-way ANOVA. N=6, P< 0.01 as compared with Adβgal-infected cells at 5.5 or 30 mM glucose.

4. Discussion

Diabetic cardiomyopathy contributes to increased heart failure and mortality in diabetic patients. However, the mechanisms underlying diabetic cardiac injury remain poorly understood, and strategies to prevent diabetic heart failure are currently unavailable. The lysosome is an essential organelle which turns over proteins to maintain cellular homeostasis. Previous studies have shown an association between lysosomal dysfunction and cardiac maladaptation in diabetes [9, 18, 20]. However, it remains unknown whether lysosomal dysfunction can directly induce diabetic cardiac injury. In this study, we found that CTSD was increased in the hearts of two different mouse models of type 1 diabetes and in high glucose (HG)-treated cardiomyocytes. HG also induced CTSD leakage from the lysosomal lumen to the cytosol and into the culture media. Importantly, CTSD knockdown or inhibition attenuated HG-induced cardiomyocyte death. These results suggested that the increased lysosomal membrane permeability and the ensuing CTSD release led to the aberrant accumulation of CTSD in the cytosol which contributed to HG-induced cardiomyocyte injury.

In the present study, HG-induced lysosomal dysfunction was demonstrated by the reduced LysoSensor puncta that indicate reduced number of lysosomes with acidic pH (Fig.1A). The ones not stained by LysoSensor were injured lysosomes with increased galectin-3 accumulation (Fig.1B) and cathepsin D (CTSD) release (Fig. 1DEF), which triggered cardiomyocyte death (Figs.3, 4 and Suppl Fig. 1). The importance of CTSD upregulation and its release from injured lysosomes was highlighted by the ability of both siRNA and pepstatin A to reduce HG-induced cardiomyocyte death (Fig. 3 and Suppl Fig. 1). This observation suggested that the CTSD relocation or aberrant distribution following the lysosomal membrane permeabilization was necessary for HG to induce cardiomyocyte death. Although the underlying mechanisms that mediates CTSD-induced cell death remain poorly understood, the pro-apoptotic protein Bid is activated by cytosolic CTSD[21], which may be a downstream effector that triggers apoptosis.

Dr. Wang’s group previously reported that myocardial infarction induced-upregulation of CTSD was an adaptive response that protected against cardiac injury [22], suggesting that the increased expression of CTSD per se in HG-treated cardiomyocytes may improve rather than compromise lysosomal function if CTSD were retained within the lysosomal lumen. Thus, preventing CTSD leakage may be effective in reducing HG cardiotoxicity. This may be achieved by strategies that can stabilize lysosomal membrane, reduce lysosomal membrane permeability or repair lysosome damage[23]. Conceivably, an even more effective approach to alleviating hyperglycemic cardiotoxicity is to eliminate the injured lysosomes altogether before they have any chance to release CTSD into the cytosol. This could be carried out by lysophagy, a process in which injured lysosomes are engulfed by autophagosomes and degraded by healthy lysosomes[16]. Enhancing lysophagy would be expected to remove dysfunctional lysosomes and maintain a pool of healthy lysosomes. Restoration of lysosomal function is also important for degrading other harmful cellular components including dysfunctional mitochondria that are considered a major mechanism for diabetic cardiomyopathy[24].

In summary, our results strongly suggested that lysosomal membrane permeabilization (LMP) and the subsequent protease Cathepsin D release are important cues that trigger HG-induced cardiomyocyte death. Preventing Cathepsin D leakage by eliminating damaged lysosomes may be an effective approach to reducing HG cardiotoxicity. Future studies are warranted to test this hypothesis in both cell culture and whole animal studies.

Highlights.

1. Hyperglycemia induced Cathepsin D in mouse heart and in cultured cardiomyocytes

2. High glucose increased lysosomal membrane permeability in cultured cardiomyocytes

3. High glucose triggered Cathepsin D leakage from the lysosome in cardiomyocytes

4. Cathepsin D knockdown or inhibition reduced high glucose cardiotoxicity

5. Cathepsin D overexpression exaggerated high glucose-induced cardiomyocyte death

Abbreviations:

cCasp3

cleaved Caspase3

CTSB

Cathepsin B

CTSD

Cathepsin D

CTSL

Cathepsin L

DAPI

4,6-diamidino-2-phenylindole

DMEM

Dulbecco’s modified essential medium

DMSO

dimethyl sulfoxide

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

GFP

green fluorescence protein

LAMP-1

lysosomal-associated membrane protein 1

mRFP

monomeric red fluorescence protein

PepA

Pepstatin A

PI

Propidium iodide

PARP

Poly (ADP-ribose) polymerase

siRNA

short interferance RNA

β-gal

β-galactosidase