OUP user menu

★ editor's choice ★

Proteasome functional insufficiency activates the calcineurin–NFAT pathway in cardiomyocytes and promotes maladaptive remodelling of stressed mouse hearts

Mingxin Tang, Jie Li, Wei Huang, Huabo Su, Qiangrong Liang, Zongwen Tian, Kathleen M. Horak, Jeffery D. Molkentin, Xuejun Wang
DOI: http://dx.doi.org/10.1093/cvr/cvq217 424-433 First published online: 2 July 2010


Aims Proteasome functional insufficiency (PFI) may play an important role in the progression of congestive heart failure but the underlying molecular mechanism is poorly understood. Calcineurin and nuclear factor of activated T-cells (NFAT) are degraded by the proteasome, and the calcineurin–NFAT pathway mediates cardiac remodelling. The present study examined the hypothesis that PFI activates the calcineurin–NFAT pathway and promotes maladaptive remodelling of the heart.

Methods and results Using a reporter gene assay, we found that pharmacological inhibition of 20S proteasomes stimulated NFAT transactivation in both mouse hearts and cultured adult mouse cardiomyocytes. Proteasome inhibition stimulated NFAT nuclear translocation in a calcineurin-dependent manner and led to a maladaptive cell shape change in cultured neonatal rat ventricular myocytes. Proteasome inhibition facilitated left ventricular dilatation and functional decompensation and increased fatality in mice with aortic constriction while causing cardiac hypertrophy in the sham surgery group. It was further revealed that both calcineurin protein levels and NFAT transactivation were markedly increased in the mouse hearts with desmin-related cardiomyopathy and severe PFI. Expression of an aggregation-prone mutant desmin also directly increased calcineurin protein levels in cultured cardiomyocytes.

Conclusions The calcineurin–NFAT pathway in the heart can be activated by proteasome inhibition and is activated in the heart of a mouse model of desmin-related cardiomyopathy that is characterized by severe PFI. The interplay between PFI and the calcineurin–NFAT pathway may contribute to the pathological remodelling of cardiomyocytes characteristic of congestive heart failure.

  • Proteasome
  • Calcineurin
  • Nuclear factors of activated T-cells
  • Cardiac remodelling
  • Desmin-related cardiomyopathy

1. Introduction

Ubiquitin-proteasome system (UPS)-mediated proteolysis plays a central role in protein quality control (PQC) through removing terminally misfolded/damaged proteins in the cell.13 The UPS also degrades most normal cellular proteins that are no longer needed, thereby playing a critical role in regulating many other cellular processes including cell signalling.4 Proteasome functional insufficiency (PFI) has been observed in animal models of several forms of cardiomyopathy,1,3 especially cardiac proteinopathies.1,5 Proteasome functional insufficiency has also been implicated in failing human hearts as reflected by remarkable increases in ubiquitinated proteins in virtually all heart diseases,1 including ischaemic heart disease, idiopathic dilated cardiomyopathy, hypertrophic cardiomyopathy, pressure-overload cardiomyopathy,6 diabetic cardiomyopathy, and desmin-related cardiomyopathy (DRC).5,7 Therefore, we postulated that PFI might play an important role in cardiac remodelling and congestive heart failure (CHF).1,8 Supporting this postulate, cardiac malfunction and even CHF have been reported to occur in cancer patients receiving proteasome inhibition (PSMI) as a primary measure of chemotherapy.911 However, very little has been reported on the molecular mechanism by which PFI contributes to cardiac remodelling and CHF. Nevertheless, inadequate PQC is known to permit intracellular aberrant protein aggregation and the latter further impairs proteasome function.12 In heart disease, such sequelae are best illustrated in DRC.5,7,13,14

Desmin-related cardiomyopathy is pathologically featured by the presence of intrasarcoplasmic aberrant protein aggregates containing desmin, αB-crystallin (CryAB), ubiquitin, and amyloid precursor proteins.15,16 Genetic studies have demonstrated that mutations in desmin and CryAB genes can cause DRC.13,14,16,17 These mutant proteins are unable to fold correctly in the cell and tend to form aggregates. Cardiac hypertrophy, a leading predictor of progressive heart disease that leads to CHF, is an important pathogenic process in DRC. The molecular mechanism underlying cardiac hypertrophy and failure in DRC is, however, unknown. Interestingly, marked increases in intrasarcoplasmic amyloid-like oligomers were observed in most human hearts with end-stage CHF resulting from idiopathic dilated or hypertrophic cardiomyopathies,18 illustrating that aberrant protein aggregation is likely a common pathological phenomenon in the progression of major heart diseases to CHF. More recently, abnormal protein aggregates were observed in pressure overloaded mouse hearts.19 It is important to note that aberrant protein aggregation not only serves as the hallmark of PQC inadequacy but also further exacerbates PQC inadequacy by impairing UPS proteolytic function in the cell, including cardiomyocytes.12 Using a UPS reporter system, we unveiled that proteasome function is severely impaired in mouse models of DRC.5,7,12 The impairment is, at least in part, caused by aberrant protein aggregation and occurs before cardiac hypertrophy and dysfunction become discernible in mutant CryAB-based DRC mice,5,12 suggesting that the PFI plays a causative role in DRC progression.

The nuclear factor of activated T-cells (NFAT) regulates the transcription of many important genes in diseased hearts.20,21 The NFAT nuclear translocation is stimulated by calcineurin-mediated de-phosphorylation and represents a critical step of NFAT activation. Phosphatase calcineurin is normally activated by an increase in intracellular calcium.21 Calpain-mediated cleavage of the autoinhibitory domain in the carboxyl terminus of calcineurin A (CnA) can also activate CnA and lead to its nuclear translocation.22 Activation of the calcineurin–NFAT pathway mediates pathological cardiac growth in many cardiac disorders.21,22 Calcineurin can be ubiquitinated by the Atrogin-1 ubiquitin ligase complex and degraded by the proteasome.23 More recently, the stability and activity of NFATc4 was found to be reduced by its ubiquitination and proteasome-mediated degradation.24 Hence, we have examined the hypothesis that PFI stimulates the calcineurin–NFAT signalling pathway in cardiomyocytes, thereby participating in maladaptive cardiac remodelling. Our experimental data strongly support this hypothesis and also reveal that the calcineurin–NFAT pathway is activated in DRC mouse hearts.

2. Methods

Additional description of methods can be found in the online supplement.

2.1 Animals

This investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). It was also approved by the Institutional Animal Care and Use Committee of the University of South Dakota, Vermillion, SD, USA. Transgenic (TG) mice carrying an NFAT binding site-dependent luciferase (NFAT-Luc) reporter have been previously described.25 The creation of a DRC mouse model by cardiomyocyte-restricted overexpression of a 7-amino-acid deletion (R172 through E178), mutation (D7-des), and a TG control with a similar level of overexpression of the wild-type desmin (WT-des) were previously reported.14

2.2 Luciferase reporter assay

The NFAT-Luc reporter activities in myocardial protein extracts or cell lysates were measured using a Luciferase Assay Kit (Roche), according to the manufacturer's instructions.

2.3 Neonatal rat ventricular myocyte culture and adenoviral gene delivery

Neonatal rat ventricular myocyte (NRVM) isolation and culture and adenoviral infection were performed as previously described.5 The adenoviruses harbouring NFATc1 with C-terminal fusion of an enhanced green fluorescence protein (GFP) (Ad-NFAT-GFP), wild-type desmin (Ad-WT-des), a DRC-linked mutant desmin (Ad-MT-des), or a constitutively active calcineurin (Ad-ΔCnA) were respectively created as previously described.12,20,26 All viruses were used at ∼50 MOI.

2.4 Cytoplasmic and nuclear fractionation of NRVMs

The cell pellets were obtained by centrifugation and resuspended gently with Lysis Buffer A (10 mM HEPES pH 7.9, 0.1% NP40, 0.1 mM EDTA, 0.1 mM EGTA, 10 mM KCl, 1 mM DTT, 0.5 mM PMSF, and the protease inhibitor cocktail one tablet/50 mL) and placed on ice for 10 min. The supernatant as the cytoplasmic fraction was collected after centrifugation at 1500 g for 5 min. The pellets were resuspended with 50 µL of Nuclear Extraction Buffer (20 mM HEPES pH 7.9, 400 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 10 mM NaF, 10% glycerol, 0.5% NP40, 0.5 mM PMSF, and the protease inhibitor cocktail one tablet/50 mL) and placed on ice for 10min, and then sonicated for 10 s. The supernatants containing nuclear proteins were obtained by centrifugation at 12000 g for 10 min (4°C).

2.5 Morphometry of cultured NRVMs

Fixed NRVMs were stained for F-actin with Alexa Fluor-568-conjugated phalloidin to identify cardiomyocytes. Individual cells in the digitalized images were traced and the profile area, major and minor axes, and Roundness of each cell were measured using the Image-Pro Plus image analysis system (Media Cybernetics, Silver Springs, MD).27 (Roundness = [Perimeter]2/4π[Area]). Roundness of a perfect circle is 1 and that of any other shape is >1. Therefore, a thread-like shape results in a Roundness value approaching infinity.

2.6 Isolation and culture of adult cardiomyocytes

The isolation of adult mouse cardiomyocytes was based on established procedures.28 The cardiomyocytes were plated and incubated in minimum essential medium at 37°C, 2% CO2 for 24 h before being treated with MG262 (1 µM) or vehicle control [dimethyl sulfoxide (DMSO)].

2.7 Western blot analysis and proteasome chymotrypsin-like activity assay

These were performed as we have previously described.5

2.8 Immunofluorescence microscopic analysis of NFAT nuclear translocation

The anti-GFP primary antibody (Santa Cruz) and the Alexa-Fluor 488 conjugated secondary antibody were used to label NFAT-GFP in fixed NRVMs. F-actin was stained red for identifying cardiomyocytes. The image of the cells was captured and digitalized using confocal microscopy.

The criteria for a cardiomyocyte with NFAT nuclear translocation include that: (a) it has very abundant F-actin and is GFP positive; (b) its GFP fluorescence intensity in the nuclear area is higher than or equal to that in the cytoplasm. The number of cardiomyocytes, with nuclear NFAT in a total of 150 cardiomyocytes expressing NFAT-GFP in 10 fields in each treatment group, was counted.

2.9 Transverse aortic constriction and administration of bortezomib

Transverse aortic constriction (TAC) was performed on 10-week-old wild-type FVB/N mice as previously described with modifications.27 The modifications include using a 7–0 silk suture and a 30-gauge needle (as mode) for ligature creation.

Bortezomib (BZM) dissolved in 1% DMSO in phosphate buffered saline (PBS) was administered via intraperitoneal injection (1 mg/kg, once every 2 days), with the first dose given immediately before the surgery. The BZM regime was determined on the basis of the reported regime used on mice.29,30 Assessment of the proteasome activity in blood cells collected at 24 h after the final injection of BZM shows that the proteasome activities in the BZM-treated sham and TAC groups remained at ∼30–40% of the untreated controls (see Supplementary material online, Table S1), indicating that the PSMI induced here was less severe than what was reported for BZM-treated human patients whose remaining proteasome activities in the blood cells at the same time point were often below 30% of the normal level.31

2.10 Anatomic determination of left ventricular volume

On the fifth day post-surgery, the mouse was anaesthetized, echocardiography was performed as previously described,27 and the abdominal aorta was cannulated for retrograde perfusion. The heart was arrested at the end of diastole through brief perfusion of KCl (100 mM) and was immediately fixed by switching the perfusion solution to 10% neutral formalin. The right atrium was cut open upon starting the perfusion. The fixed heart was removed from the chest and weighed. The ventricular weight was measured after removing atria. Finally, the left ventricular (LV) volume was measured by determining the exact amount of PBS required for filling up the LV.

2.11 Statistical analysis

All quantitative data are presented as mean ± SD. Differences between experimental groups were evaluated for significance using Student's t-test for unpaired data or one-way ANOVA when appropriate. The Holm–Sidak test was used for post hoc comparison.

3. Results

3.1 Pharmacological inhibition of the proteasome activates the NFAT signalling in mouse hearts and cultured cardiomyocytes

First, we tested the acute effects of PSMI on NFAT transactivation activity in both intact mice and cultured adult mouse cardiomyocytes.

NFAT-Luc TG mice were intravenously treated with MG262 (5 µmol/kg), a cell-permeable proteasome inhibitor. At 24 h after the treatment, the high molecular weight ubiquitinated proteins [HMWUP (>60 kD)] in ventricular myocardium were modestly but statistically significantly increased (P < 0.05; Figure 1A) and the proteasome chymotrypsin-like activities of both ventricular myocardium (Figure 1B) and red blood cells (data not shown) were decreased by ∼70% (P < 0.01). Compared with the control group treated with DMSO, myocardial calcineurin protein levels were not increased (data not shown) but NFAT-luciferase activities were increased by 60% in MG262 treated NFAT-Luc mouse hearts (Figure 1C), indicating that PSMI is sufficient to activate NFAT signalling in mouse hearts. Perhaps due to the short-term nature of the treatment, echocardiography detected no significant morphometric and functional changes in the heart of the MG262 treated mice at this time point (Figure 1D–F).

Figure 1

Proteasomal inhibition activates NFAT in mouse hearts. NFAT-Luc mice were treated with MG262 (5 µmol/kg) or DMSO. Twenty-four hours after treatment, total protein extracts from ventricular myocardium were used for western blot analysis of ubiquitinated proteins (A). Densitometry analysis shows that the high molecular weight ubiquitinated proteins (HMWUP, >60 kDa) are increased by 35% in MG262 treated hearts, compared with the DMSO control group (P < 0.05). The soluble crude protein extracts were used for measuring the chymotrypsin-like activity (B) and the NFAT-Luc activity (C). Echocardiography was performed in a separate cohort at 24 h after treatment. Representative M-mode images (D) and a summary of key parameters (E, F) were presented. HR, heart rate; LVPWd and LVIDd, left ventricle posterior wall thickness and internal diameter at the end of diastole, respectively; FS, fractional shortening; EF, ejection fraction; *P < 0.05, **P < 0.01, vs. the DMSO group.

To demonstrate PSMI is sufficient to activate NFAT signalling at the cardiomyocyte level, ventricular myocytes from adult NFAT-Luc mouse hearts were cultured and treated with MG262 (1 µM) or DMSO. As expected, ubiquitin conjugates of a higher weight (>210 kD) were increased (Figure 2A, B) and the proteasome chymotrypsin-like activity was severely depressed (Figure 2C) after 10h of MG262 treatment. More importantly, the NFAT-luciferase activity of MG262 treated cells was ∼50% higher than that of DMSO treated cells (Figure 2D).

Figure 2

Proteasomal inhibition increases NFAT transactivation in cultured adult cardiomyocytes. Ventricular myocytes enzymatically isolated from adult NFAT-Luc mice were cultured under a serum-free condition. Cell lysates for western blot analyses of ubiquitinated proteins (A, B), proteasome chymotrypsin-like activities (C), and the NFAT-luciferase (NFAT-Luc) activity assay (D) were obtained from the cells that had been treated with MG-262 (1 µM) or DMSO (vehicle control) for 10 h. A quantitative comparison of the relative density of ubiquitinated proteins of a higher (>210 kDa; a) or a lower molecular weight (<210 kDa; b) between MG262- and DMSO-treated cells is summarized in panel (B). *P < 0.05, **P < 0.01, vs. the DMSO group.

3.2 MG262 enhances NFAT nuclear translocation in a calcineurin-dependent manner

Nuclear factor of activated T-cells translocation from cytoplasm to the nucleus is an important event in the activation of the calcineurin–NFAT pathway.21 To determine if PSMI causes NFAT nuclear translocation and to determine whether the activation of NFAT by PSMI is calcineurin-dependent, NFAT-GFP was expressed in cultured NRVMS via infection of Ad-NFATc1-GFP and the NFAT-GFP subcellular distribution was monitored via microscopic analysis. At 24 h after the viral infection, cells were treated, respectively, with MG262 (1 µM), calcineurin inhibitor—cyclosporine A (CsA, 1 µg/mL), and DMSO as a vehicle control in serum-free culture medium. Cells with NFAT-GFP nuclear translocation were significantly increased by MG262 but this increase was nearly abolished by CsA co-treatment (Figure 3). These results show that PSMI triggers calcineurin-dependent NFAT nuclear translocation.

Figure 3

Microscopic analysis of the effect of proteasomal inhibition on NFAT nuclear translocation. Neonatal rat ventricular myocytes grown in chamber slides were infected with Ad-NFAT-GFP 24 h before being treated with DMSO (vehicle control), cyclosporine A (CsA, 1 µg/mL), MG262 (1 µM), or both MG262 (1 µM) and CsA (1 µg/mL). After 6 h of treatment, the cells were fixed by 4% paraformaldehyde and immunolabelled for GFP (green). F-actin was stained red with Alexa Flour 568 conjugated phalloidin. Representative fluorescence micrographs are presented in panel (A). The percentage of NRVMs with nuclear NFAT-GFP in the indicated treatment groups (B) was measured from 10 representative fields of three experimental repeats. *P < 0.01 vs. DMSO; #P < 0.05 vs. MG262. Scale bar = 50 µm.

The findings from the microscopic analysis were confirmed by subcellular fractionation followed by quantitative western blot analysis of NFAT-GFP (Figure 4).

Figure 4

Western blot analysis of proteasomal inhibition induced NFAT-GFP nuclear translocation in cultured cardiomyocytes. Neonatal rat ventricular myocytes grown on Petri dishes were infected with Ad-NFAT-GFP and treated with indicated agents as described in Figure 3. Groups overexpressing a constitutive CnA (ΔCnA) with (ΔCnA + CsA) or without CsA treatment were included as positive controls. Cytosolic and nuclear protein extracts were, respectively, quantitatively analysed with western blots for GFP (NFAT-GFP). A set of representative western blot images are shown in panel (A). The nuclear to cytosolic (N/C) NFAT-GFP ratios derived from the densitometry of four independent experimental repeats are summarized and compared in panel (B). The average N/C ratio of DMSO treated cells is arbitrarily set as 1 and used to normalize other treatment groups. *P < 0.05, **P < 0.01, vs. DMSO; #P < 0.05 vs. MG; $P < 0.05 vs. ΔCnA.

3.3 PSMI mitigates the growth of cultured cardiomyocytes and promotes LV dilatation in TAC mice

We next tested whether PSMI affects baseline and pro-hypertrophic agonist-stimulated cardiomyocyte growth in cultured NRVMs. Morphometry showed that norepinephrine (NE)-induced increases in the profile area of culture NRVMs were nearly abolished by MG262 but MG262 significantly increased the major axis and the Roundness of the cell while tending to decrease the minor axis at baseline. NE treatment alone significantly increased both the major and the minor axis but did not alter the Roundness of the cell. However, co-treatment of MG262 further enhanced NE-induced increases in the major axis but prevented NE from increasing the minor axis of the cell, thereby rendering the cell to a thread-like shape, as indicated by a marked increase in the Roundness parameter of the cell in the NE+MG262 group, compared with both the vehicle control (CTL) and the NE-treated groups (P < 0.01; Figure 5A, B). Since increased growth in the length relative to width is the most recognized anatomic feature of cardiomyocytes underlying chamber dilatation in CHF,32 these in vitro findings raise a possibility that PFI may promote chamber dilatation and functional decompensation of hearts undergoing hypertrophy.

Figure 5

Effects of PSMI on the growth of cultured cardiomyocytes and LV remodelling. Neonatal rat ventricular myocytes were cultured in serum-free media for 12h before being treated with the indicated agents. Three days after the treatment, cells were harvested for total RNA extraction or fixed with 3.8% paraformaldehyde. (A, B) Morphometric characterization of cell profile area and cell shape. The fixed cells were stained for F-actin (red), imaged, and morphometrically analysed. A representative set of images are shown in panel (A) (scale bar = 25µm). The indicated computer-derived parameters of 90 cells/group from three experimental repeats are summarized in panel (B). *P < 0.05, **P < 0.01 vs. CTL; #P < 0.05, ##P < 0.01 vs. NE. (CF) Effects of PSMI on post-TAC LV remodelling in intact mice. Wild-type mice were treated with BZM (i.p., 1mg/kg/2 days) or vehicle, starting immediately before TAC or sham surgery. Two-dimensional-guided M-mode echocardiograph was recorded on the fifth day after surgery. Representative M-mode images (C) of the indicated groups and indicated key parameters (D) are presented. LVIDd, left ventricular end-diastolic internal dimension; LVPWd, left ventricular posterior wall thickness; FS, fractional shortening (%). On the fifth day after surgery, mouse hearts arrested at the end diastole were perfusion-fixed for gravimetric and anatomic analyses as described in the section Methods. Representative images of the whole heart (E, scale bar = 5 mm) and the heart weight to tibial length ratio (Hw/TL), the ventricular weight to tibial length ratio (Vw/TL), and LV chamber volume (LVv) (F) are shown. *P < 0.05 vs. Sham; #P < 0.05 vs. Sham + BZM; $P < 0.05 vs. TAC.

To test this possibility in vivo, we treated non-transgenic (NTG) mice with proteasome inhibitor BZM (1 mg/kg, once every 2 days) or vehicle control and subjected them to TAC or sham surgery. Compared with vehicle-treated mice, BZM-treated mice showed a significantly reduced tolerance and survival rate to TAC during the first 5 days after TAC (see Supplementary material online, Figure S1). Those surviving the TAC surgery displayed a significantly enlarged LV as revealed by both echocardiography and post-mortem LV anatomic examination. Interestingly, consistent with the ability of PSMI to activate the calcineurin–NFAT pathway, the BZM-treated sham group developed significant cardiac hypertrophy that is comparable with the vehicle-treated TAC group. In contrast to the abrogation of NE-induced increases in the cardiomyocyte profile area by PSMI in cultured NRVMs (Figure 5A, B), a longer term of systemic PSMI induced by BZM failed to alter TAC-induced increases in ventricular mass at 5 days after TAC. Echocardiography also revealed a significant decrease in LV fractional shortening (FS) in BZM-treated TAC mice, compared with the vehicle-treated TAC mice (see Figure 5C–F, Supplementary material online, Tables S2 and S3). These in vivo data indicate that chronic PSMI is sufficient to cause cardiac hypertrophy and induces stressed hearts to fail.

3.4 The calcineurin–NFAT pathway is activated in DRC mouse hearts

D7-des TG mouse hearts display aberrant protein aggregation, pathological hypertrophy, early diastolic malfunction, and late chamber dilatation, recapitulating human DRC.14 Recent studies have revealed that both calcineurin and NFAT are degraded through the UPS.23,24 Ubiquitin-proteasome system proteolytic function is severely impaired in D7-des mouse hearts.7 These studies led to a hypothesis that UPS malfunction in D7-des hearts may slow down the degradation of calcineurin and NFAT, thereby enhancing and/or activating the calcineurin–NFAT pathway. To test this hypothesis, our investigation revealed that full-length calcineurin A (CnA, ∼59 kD) protein levels in ventricular myocardium were markedly increased in D7-des TG mice but not in WT-des TG mice, compared with NTG littermate controls (Figure 6). An increase in a lower molecular weight band (∼48 kD, indicated by an arrow in Figure 6) that is CnA antibody reactive is also evident. This lower molecular weight band disappeared when the antibody against the carboxyl terminal end of CnA (Santa Cruz) was used for the western blot analysis (data not shown). Therefore, this shorter form of CnA is likely the carboxyl terminal cleaved form and an active form of CnA.22

Figure 6

Increases in calcineurin A (CnA) protein levels in desminopathic mouse hearts. CnA levels in left ventricular total protein extracts from 3-month old D7-des TG, WT-des TG, and NTG mice were analysed by quantitative western blots. Representative western blot images (A) and a summary of densitometry data (B) are presented. A brain tissue protein extract (Br) was included in the western blot to help identify the main band of CnA. AU, arbitrary units; **P < 0.01 vs. WT-des TG and NTG.

Activated calcineurin can de-phosphorylate NFAT, resulting in NFAT translocation to the nucleus where NFAT interacts with GATA4 and other transcription factors to initiate hypertrophic gene transcription.33 Therefore, NFAT transactivation activities reflect the activity of the calcineurin–NFAT pathway. To probe the downstream events of increased calcineurin in DRC hearts, D7-des mice were cross-bred with NFAT-Luc mice. The resultant double TG (DTG) mice displayed significant increases in the heart weight to body weight ratio, compared with NFAT-Luc TG or NTG mice at 3 months. There is no significant difference in the heart weight to body weight ratio between NFAT-Luc TG and NTG mice (Figure 7A). The cardiac hypertrophy observed in the DTG mice is similar to what was previously observed in D7-des single TG mice.14 More importantly, ventricular NFAT-luciferase reporter activities of DTG mice were significantly higher than that in NFAT-Luc single TG mice (Figure 7B), indicating that NFAT is activated in D7-des TG hearts.

Figure 7

Increased transactivation of NFAT in desminopathic mouse hearts. NFAT-luciferase (NFAT-Luc) reporter mice were cross-bred with D7-des TG mice and the resultant offspring littermates were examined at 3 months of age. The ratio of the heart weight to the body weight (HW/BW) and the reporter luciferase (Luc) activities (relative light units, RLU/µg; n = 4 for each group) are, respectively, presented in panels (A) and (B). DTG, D7-des-NFAT-Luc double transgenic; **P < 0.01 vs. NTG and NFAT-Luc TG.

3.5 MT-des increases calcineurin protein levels in cultured cardiomyocytes

To test a causal relationship between MT-des expression and the change in calcineurin signalling at the cellular level, we examined calcineurin protein expression in the cultured NRVMs infected with Ad-MT-des, Ad-WT-des, or Ad-β-Gal for 5 days. A modest level of MT-des overexpression increased significantly the calcineurin protein level but a comparable level of WT-des overexpression did not, compared with the Ad-β-Gal group (Figure 8). Interestingly, the western blot images show a marked increase in the intensity of a lower molecular weight band (∼48 kD, indicated by an arrow in Figure 8A) detected by the CnA antibody against the N-terminal CnA [CnA(N)] in the MT-des expressing cells. Similar to what was observed in D7-des TG hearts, this band could not be detected with the antibody against the carboxyl terminus of CnA [CnA(C), Figure 8A]. Based upon the molecular weight, this band is likely the calpain-cleaved and an activated form of CnA described previously by Burard et al.22 These data suggest that in the absence of changes in the mechanical function of cardiomyocytes, overexpression of MT-des is sufficient to increase calcineurin protein expression in cardiomyocytes.

Figure 8

Expression of MT-des increases calcineurin protein levels in cultured NRVMs. Cells infected with Ad-WT-des, Ad-MT-des, and Ad-β-gal were cultured for 5 days in a serum-free condition and the total proteins were extracted and analysed with quantitative western blots. (A) Representative images of western blots for desmin, CnA, and α-actinin (serving as loading controls). CnA was probed with an antibody against its N-terminal [CnA(N)] and an antibody against its C-terminal [CnA(C)] on two parallel gels. (B) Comparison of CnA(N) densitometry data among the three groups. Note that the intensity of a CnA antibody-reactive lower molecular weight band (indicated by arrow) is also substantially increased in MT-des expression cells. *P < 0.05, compared with the Ad-WT-des and the Ad-β-gal group.

4. Discussion

Cardiac accumulation of pre-amyloid oligomers in a large subset of human CHF patients and the increases of myocardial ubiquitin conjugates in nearly all heart diseases indicate that PFI is likely a common pathological process in cardiac dysfunction. Some cancer patients treated with proteasome inhibitors develop overt heart failure, suggesting that PFI is detrimental to the heart. Here we have demonstrated for the first time that PSMI is sufficient to activate cardiac NFAT signalling in both intact mice and cultured cardiomyocytes. Our in vitro studies further reveal that PSMI, albeit suppressing the pro-hypertrophic agonist-stimulated cell profile area increases, causes cardiomyocyte remodelling, giving rise to a maladaptive cell shape change. Confirming these in vitro findings, BZM-induced chronic PSMI promoted LV chamber dilatation and functional decompensation in pressure-overloaded mouse hearts while causing cardiac hypertrophy in the sham surgery group. Supporting a pathogenic role of PFI-induced NFAT activation, we have further found activation of the calcineurin–NFAT pathway in a DRC mouse model with severe PFI. These data suggest that PFI alone is sufficient to activate calcineurin–NFAT signalling and cause cardiac hypertrophy. Additionally, PFI promotes cardiomyocyte lengthening and chamber dilatation in a stressed heart.

4.1 PQC, calcineurin–NFAT signalling, and cardiac hypertrophy

Calcineurin–NFAT signalling plays diverse regulatory roles in multiple vertebrate cell types. In the heart, calcineurin has been implicated as a pivotal regulator of cardiomyocyte hypertrophy.21 An increase in calcineurin protein levels is often associated with the activation of calcineurin in the hearts.34,35 The increase can obviously be caused by an increase in calcineurin synthesis. Indeed, increases in calcineurin synthesis were demonstrated in cardiac hypertrophy caused by various pathological stimuli.34,35 Moreover, activation of the calcineurin–NFAT pathway has been shown to activate the promoter of CnAβ in vitro and in vivo, forming a positive feedback.33

Recent studies also suggest that calcineurin expression can be altered by a post-translational mechanism such as ubiquitination.23,36 The present study provides evidence that PSMI suffices to activate the calcineurin–NFAT pathway both in vitro and in vivo. In many cases, the activation of a normal protein molecule also sends a signal to the UPS for its degradation.8 The degradation of calcineurin by the UPS may well be the case. Notably, not only the full-length CnA but also the carboxyl terminal cleaved form of CnA were markedly increased in both D7-des TG hearts and cultured NRMVs expressing DRM-linked MT-des (Figure 6A and 8A, arrows). This raises the possibility that CnA may first be cleaved by a protease (e.g. calpain) prior to being degraded by the UPS so that both the full-length and cleaved forms of CnA increased when UPS proteolytic function is severely impaired. Hence, both increased synthesis and decreased proteasomal degradation may be responsible for the increase in CnA protein levels and activities in MT-des expressing cardiomyocytes and hearts. Increased CnA activities then result in NFAT activation, which has been shown by the majority of studies to participate in pathological remodelling of the heart. Indeed, we found that sustained systemic PSMI led to cardiac hypertrophy in mildly stressed mice and promoted LV dilatation and functional decompensation in mice with TAC (Figure 5C–F).

Protein quality control in the cell is carried out by an elaborate collaboration between two major players: molecular chaperones and targeted proteolysis. The latter is primarily performed by the UPS, with perhaps some supplementary role from macroautophagy.1,19,37 As the most abundant small heat shock protein in cardiomyocytes, CryAB is a major cytoplasmic chaperone for cardiomyocytes. We have previously shown that the loss of CryAB is sufficient to activate the calcineurin–NFAT pathway and causes cardiac remodelling and malfunction, while CryAB overexpression suppresses hypertrophic stimuli-induced NFAT activation in mouse hearts.27 Taken together, it appears that PQC inadequacy induced either by the loss of an important molecular chaperone or by PFI can both activate the calcineurin–NFAT pathway and facilitate maladaptive cardiac remodelling.

4.2 Activation of the calcineurin–NFAT pathway in DRC mouse hearts

Pathological hypertrophy and CHF are among the major clinical manifestations of DRC.16,17 The molecular pathway by which cardiac hypertrophy and malfunction occur in DRC was not understood. Here we have shown that the calcineurin–NFAT pathway is activated in the D7-des mouse hearts which were previously shown to have severe PFI. Because the D7-des mice develop cardiac hypertrophy and malfunction shortly after birth, it is difficult to tell from the in vivo data whether the activation is secondary to cardiac malfunction or not (Figures 6 and 7). Nevertheless, we found that expression of DRC-linked MT-des increased the protein levels of both the native and cleaved forms of CnA (Figure 8). This provides strong evidence to support that cardiac malfunction is not essential for the activation of the calcineurin–NFAT pathway in DRC hearts. Therefore, a pathogenic role of NFAT activation and a likely causative relationship between PFI and the NFAT activation in DRC are strongly implicated.

4.3 PSMI promotes maladaptive cardiomyocyte remodelling

Consistent with previous reports,38 administration of MG262 blocked NE-induced increases of the cell profile area. However, PSMI caused a remarkable cell shape change as evidenced by a dramatic increase in the Roundness. This occurred in PSMI alone and became more pronounced when coupled with NE stimulation (Figure 5A, B). Although not explicitly described, this characteristic myocyte shape change is evident in the previously reported images of cardiomyocytes subjected to PSMI.38,39 At the individual cardiomyocyte level, abnormal increases in the cell length relative to cell width (i.e., increases in the parameter Roundness) are the most consistent anatomic change seen in the ventricular dilatation characteristic of CHF.32 Indeed, we were able to demonstrate in intact mice that sustained systemic PSMI facilitates LV dilatation and functional decompensation and significantly reduces survival after TAC (see Figure 5C–F, Supplementary material online, Figure S1). In agreement with these findings, chamber dilatation and CHF occur in DRC mouse hearts that show severe PFI.13,14 A significant increase in cell length was also evident in cardiomyocytes isolated from mouse hearts with TG overexpression of an activated calcineurin.40

Overall, the findings of the present study dispute several recent reports reviewed by Hedhli and Depre.41 These reports demonstrated that systemic pharmacological PSMI prevents or even reverses pressure overload cardiac hypertrophy without affecting the function of the heart in mice, suggesting that PSMI is of therapeutic value to pressure overload cardiac hypertrophy. A definitive explanation for this dispute is currently lacking, but the severity and duration of PSMI as well as the interplay or cross-talk between PFI and other compounding pathological processes may all contribute to the outcome. For example, autophagy was activated by, and compensated for PFI in some pathological conditions.42,43 Nevertheless, the findings of the present study are well supported by recent clinical observations which showed that a regime with the proteasome inhibitor BZM can cause severe heart failure in cancer patients, especially when coupled with a pre-existing cardiac condition.911

It appears that in the presence of PFI, the growth signals from the activation of the calcineurin–NFAT pathway and perhaps other pathways by stress, neurohumoral factors, or PSMI drive cardiomyocytes to remodel toward the maladaptive direction. Additional studies, especially in vivo studies, are warranted to test further whether activation of calcineurin–NFAT signalling is essential to PSMI-induced maladaptive remodelling of cardiomyocytes, given that PFI is likely a very common phenomenon in a large subset of CHF patients.1


This study was supported by NIH grants R01HL072166, R01HL085629, and R01HL068936 and by AHA 0740025N (X.W.).


X.W. is an Established Investigator of American Heart Association (AHA). We thank Ms Emily J. McDowell for her excellent assistance in the preparation of this manuscript.

Conflict of interest: none declared.


View Abstract