|Home | About | Journals | Submit | Contact Us | Français|
Inheritable missense mutations in small molecular weight heat-shock proteins (HSP) with chaperone-like properties promote self-oligomerization, protein aggregation, and pathologic states such as hypertrophic cardiomyopathy in humans. We recently described that human mutant αB-crystallin (hR120GCryAB) overexpression that caused protein aggregation cardiomyopathy (PAC) was genetically linked to dysregulation of the antioxidant system and reductive stress (RS) in mice. However, the molecular mechanism that induces RS remains only partially understood. Here we define a critical role for the regulatory nuclear erythroid 2-related factor 2 (Nrf2)-Kelch-like ECH-associated protein (Keap1) pathway—the master transcriptional controller of antioxidants, in the pathogenesis of PAC and RS. In myopathic mice, increased reactive oxygen species signaling during compensatory hypertrophy (i.e., 3 months) was associated with upregulation of key antioxidants in a manner consistent with Nrf2/antioxidant response element (ARE)-dependent transactivation. In transcription factor assays, we further demonstrate increased binding of Nrf2 to ARE during the development of cardiomyopathy. Of interest, we show that the negative regulator Keap1 was predominantly sequestrated in protein aggregates (at 6 months), suggesting that sustained nuclear translocation of activated Nrf2 may be a contributing mechanism for RS. Our findings implicate a novel pathway for therapeutic targeting and abrogating RS linked to experimental cardiomyopathy in humans. Antioxid. Redox Signal. 14, 957–971.
Inheritable disorders linked to mutations in genes encoding molecular chaperones (e.g., CryAB or heat-shock protein binding factor-2 [HSPB2]) and cytoskeletal proteins contribute to the pathogenesis of heart failure (21–23, 29). Among such pathogenic mechanisms, redox signaling pathways involving glutathione (GSH), a major nonprotein thiol, possess antioxidant properties for detoxifying and scavenging toxic reactive oxygen species (ROS) generated in disease states (25, 26). Paradoxically, we have demonstrated recently that the development of cardiac hypertrophy and heart failure from human R120GαB-crystallin (hR120GCryAB) overexpression caused dysregulation of glutathione homeostasis, termed “reductive stress (RS),” in transgenic mice (TG) (29). Further, quenching of reducing power by inhibiting glucose 6 phosphate dehydrogenase (G6pd) levels/activity and thereby decreasing NADPH rescued the hR120GCryABTG mice from cardiac hypertrophy and cardiomyopathy/heart failure (29). However, the causal mechanism for “RS” remains obscure.
Redox-sensitive transcription factors are attractive candidates whose dysregulation of key molecular targets might contribute to the pathogenic changes in the intracellular redox milieu. Nuclear erythroid 2-related factor 2 (Nrf2; NF-E2-related factor 2) (18), a basic leucine zipper protein, is a redox-sensitive master transcriptional regulator of several cytoprotective genes including the major antioxidants (12, 18, 44). Under basal conditions, Nrf2 is sequestered in the cytoplasm by the Kelch-like ECH-associated protein (Keap1). Interaction with Keap1 promotes rapid ubiquitination of Nrf2 and subsequent proteasome degradation (18, 27). In response to oxidative or electrophilic stress, Keap1-dependent ubiquitin ligase activity is inhibited and the Nrf2 translocates to the nucleus to induce transactivation of the antioxidant response element (ARE)-containing genes (14, 49). Transcriptional pathways under the control of Nrf2 have been implicated in protection from oxidative stress of vital organs and various diseases such as acute respiratory distress syndrome, hyperoxic injury, pulmonary fibrosis, hepatocellular necrosis, hepatotoxicity, cancer, and neuronal dysfunction (7, 13–19). Using in vitro and in vivo Nrf2 knockout models, substantial evidence has been developed to support the regulatory role of Nrf2 in antioxidative and cytoprotective functions against pathological conditions, including cancer, pulmonary, inflammatory, and neurodegenerative diseases (4, 9, 18, 28, 32, 33). Thus, Keap1-Nrf2 interaction is an important regulatory nodal point for the overall response to redox stress. Both Nrf2 and Keap1 (a cysteine-rich protein) are redox sensitive proteins (6, 45) and are vulnerable to altered intracellular concentrations of redox couples (i.e., GSH:GSSH, NADPH:NADP, NADH:NAD, −SH:S-S, and cysteine:cystine) (20). Relevant to the cardiovascular system, a recent study has shown that Nrf2−/− mice are susceptible for mechanical pressure overload cardiac hypertrophy, which demonstrates a critical pathophysiological role for Nrf2 in the heart (19).
We identified constitutive activation of Nrf2 in the TG-mouse heart in which human mutant CryAB was overexpressed. Such an activation of Nrf2 is associated with RS and mutant protein aggregation cardiomyopathy (MPAC) in mice. Collectively, these observations suggest that the Nrf2-Keap1 pathway acts as key player in protecting against the mutant protein-induced stress, but eventually contributing to RS, which may be deleterious to the heart. In the current study, we hypothesized that the mutant protein-induced self-oligomerization and formation of insoluble protein aggregates could induce ROS, which can then trigger dissociation of the Nrf2-Keap1 complex, thereby activating Nrf2. We further hypothesized that interactions between the mutant protein aggregates and Keap1 play a key role in stabilization of Nrf2 and activating ARE-dependent cytoprotective mechanisms, which might also lead to RS in the hR120GCryAB TG mouse heart. Since the Nrf2-Keap1 system is crucial for regulating a battery of genes encoding antioxidants and cytoprotective proteins in various oxidative stress disorders, unraveling its direct and key role in heart tissue and cardiovascular pathogenesis will be of potential therapeutic importance. Using the TG mouse model for human heart disease, we have investigated the mechanisms for Nrf2-dependent transactivation of antioxidant genes and resultant RS.
Studies of mice were approved by the Institutional Animal Care and Use Committee at the University of Utah (IACUC No. 08-07006). Nontransgenic (NTG) and hR120GCryAB TG mice in C57/Bl-6 background were generated as previously described (29), and were housed under controlled conditions for temperature and humidity, using a 12-h light/dark cycle. At the indicated time points (3 and 6 months of age), mice were assessed for cardiac hypertrophy by echocardiography.
The following antibodies and reagents were used: Nrf2-ab (SC-772; Santa Cruz Bio), Keap1 (10503-2-AP; Proteintech), HO-1 (ab13248; Abcam, Inc.), NQO-1 (ab34173), phospho-CryAB (ab5577), glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (ab9485), lamin-B1 (ab16048), super oxide dismutase-1 (SOD-1) (ab13498), Catalase (219010; Calbiochem, Merck kGaA), γ-glutamyl cysteine synthase (γ-GCS) (RB-1697-P1; Labvision/Neomarkers), and mono-/polyubiquitin (BML-PW0150; Enzo Life Sciences). An anti-CryAB polyclonal antibody (Ab) that reacts with mouse and human proteins was raised against 164–175 amino acid residues of human CryAB. Secondary antibodies conjugated with horseradish peroxidase IgG Rabbit and Mouse (PI-1000 & PI-2000; Vector Labs). Radical detecting electron paramagnetic resonance (EPR) probes 1-hydroxy-3-methoxy-carbonyl-2, 2,5,5- tetramethyl pyrrolidine (CMH) (NOX-2.1), TEMPOL (705748; Sigma-Aldrich), EPR grade water (NOX-7.7.1), Krebs-4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid [HEPES] buffer (NOX-7.6.1), diethyl dithio carbomate (DETC) (NOX-10.1), defroxamine (NOX-9.1), glass capillary tubes (NOX-G.3.1), and critoseal (NOX-A. 3.1-VP) were purchased from the Noxygen Diagnostics. Fluorescent probes dihydroethidium (DHE)/dihydro difluro diacetate (H2DCFDA) for detecting ROS and superoxide were obtained from Molecular Probes. Bio-Rad Protein Assay (500-0006; Bio-Rad) was used to determine protein levels in the heart tissue extracts. All reagents and primers for RNA extraction and real-time RT-PCR quantification were purchased from Qiagen, Inc.
Glutathione redox (GSH/GSSG) state was determined in heart tissue extracts from NTG and TG mouse at 3 and 6 months. In brief, heart tissue extracts were prepared in MES buffer and centrifuged at 5000rpm for 5min at 4°C. A small aliquot of the supernatants were used for protein determination and the remaining samples were mixed with equal volumes of 10% metaphosphoric acid (Cat. No. 239275; Sigma) to precipitate proteins. A known volume of the metaphosphoric acid extracts was treated with triethanolamine reagent and 2-vinyl pyridine (only for GSSG analysis) and appropriate GSH and GSSG standards were treated similarly to prepare a standard graph. Kinetic GSH-reductase recycling assay was performed following the manufacturer's instructions (703002; Cayman Chemicals) using a plate reader (Bio-Tek; FLx-800) (29, 31).
To understand the electrophilic signals that are critical to dissociate the Nrf2-Keap1 complex and favor Nrf2 activation, we measured the tissue redox stress/ROS formation and assess the temporal effect of hR120GCryAB expression. Briefly, mice were injected (i.p.) with heparin (4U/g b.w.) and sacrificed by CO2 inhalation. Hearts were perfused in situ with HEPES buffer (20μM, pH 7.4) to remove residual blood, a portion of ventricle was cut into small pieces, placed in a clean 24-well tissue culture plate containing 500μL HEPES, washed twice, and immersed in 300μL of the buffer. About 10mg of processed ventricular tissues was incubated with 150μL of 500μM CMH, at 37°C for 30min. ROS released by tissues react with the CMH to form a stable nitroxide radical that can be measured using electron spin resonance (ESR) (15). Aliquots of the incubated probe media were taken in 50μL glass capillary tubes and ESR spectra measured using an EMX-ESR spectroscopy (Bruker Instruments) as reported previously (1).
ROS generation in the mouse myocardial tissue from 3- and 6-month-old NTG and TG mice was determined using DHE/H2DCFDA fluorescence in a manner similar to that reported previously (8, 43). Cytosolic DHE exhibits blue fluorescence; however, when this probe is oxidized to ethidium it intercalates with DNA, staining the nucleus a bright red fluorescence. In the presence of superoxide (O2−.), the cell-permeable nonfluorescent DHE is oxidized to fluorescent 2-hydroxyethidium, which is then detected by using the excitation/emission filters appropriate for rhodamine. H2DCFDA is oxidized by hydrogen peroxide to convert the nonfluorescent DCF to fluorescent DCF, which can be detected by a fluorescent microscope using appropriate excitation/emission filters for fluorescein as reported previously (43). Briefly, at the end of each protocol the hearts were snap-frozen using optimal cutting temperature medium and cut into 5-μm-thick transverse sections placed in clean glass slides. Sections were appropriately covered with the probe solution containing DHE/H2DCFDA (10μM) along with the nuclear stain DRAQ5 and were incubated in a light-protected chamber at 37°C for 30min and washed thoroughly with 1×TBS-T thrice, fixed, and mounted using Flouromount-G. Images of the tissue sections were obtained using a Zeiss 510 Meta laser scanning confocal microscope (Carl-Zeiss, Inc.). Fluorescence intensity, which positively correlates with the amount of O2−. generation, was determined in the myocardial tissue and quantified by automated image analysis using the ImageJ software (National Institutes of Health).
Harvested hearts from 3- and 6-month-old NTG and TG animals were initially flash frozen in liquid nitrogen. Cytosolic lysates from the tissues were prepared by homogenizing the tissue using cytosolic/homogenizing buffer (10mM HEPES, 10mM KCl, 0.1mM EDTA, 0.5mM MgCl2, with freshly prepared 1mM dithiothreitol and 0.1mM phenyl methylsulfonyl fluoride, and 1% Triton×100, pH 7.9), followed by centrifugation at 5200rpm for 5–6min. The nuclear pellet was washed with 4 volumes of homogenizing buffer to get rid of any cytosolic contaminants. Nuclear fractions were prepared in complete lysis buffer (20mM HEPES, 420mM NaCl, 0.1mM EDTA, 1.5mM MgCl2, 25% glycerol, 1mM dithiotheitol, and 0.5mM phenyl methylsulfonyl fluoride, pH 7.9). Samples were incubated on ice with mild shaking and centrifuged at 8200rpm for 8min. Later, proteins were determined using Biorad Bradford reagent and the samples for Western blots (WB) were prepared in 0.25 volumes of Laemmli buffer with 5% freshly added β-mercapthoethanol and boiled for 5min. About 30μg of both nuclear and cytosolic proteins was resolved separately on 10%–12% SDS-PAGE. They were probed using antibodies against CryAB, P-CryAB, Nrf2, HO-1, NQO-1, catalase, Keap1, ubiquitin, SOD-1, γ-GCS, GAPDH, and lamin-B1. Secondary antibodies conjugated with horseradish peroxidase IgG Rabbit and Mouse were used for chemiluminescence detection.
The level of Nrf2 activation and DNA binding efficacy at the RS state was evaluated in the NTG and TG mouse hearts using Trans AM Nrf2 Kit (50296; Active Motif). About 10μg of nuclear proteins was incubated with immobilized oligonucleotide containing ARE consensus binding site (5′-GTCACAGTACTCAGCAGAATCTG-3′) and the active form of Nrf2 that bound to the oligo was detected using anti-Nrf2 primary Ab recognizing epitope on Nrf2 protein after treating with HRP-conjugated secondary Ab. The resultant chromogen formed due to the specific activity of the transcription factor in the nuclear extracts was read in a plate reader. Absorbance at 450nm was expressed as the direct activity of Nrf2.
Since Keap1 regulates the fate of Nrf2 ubiquitination and stabilization in cytoplasm, we have analyzed the ubiquitin-conjugated Nrf2 by immunoprecipitation (IP) using ubiquitin Ab, followed by WB against anti-Nrf2 Ab. Briefly, 200μl of 0.5mg/ml cytosol proteins from NTG and TG mouse hearts was incubated with an anti-ubiquitin Ab (for 12h at 4°C with gentle rotation) and protein-A/G agarose beads (Invitrogen Corporation) and washed the precipitates thoroughly in wash buffer. IP proteins were analyzed by WB with an anti-Nrf2 Ab. Further to investigate the degree of Nrf2-Keap1 association, we performed IP from the cytosol extracts with anti-Keap1 Ab and probed with anti-Nrf2 Ab.
Hearts were harvested after in situ perfusion with 10ml of RNase free PBS and 10ml of RNA later reagent. To extract RNA, ~30mg of ventricular tissue was processed from 3- and 6-month-old NTG and TG mice (n=5) using RNA extraction kit (Qiagen; Cat. No. 74106) following supplier's instructions. RNA samples were then quantified by measuring the absorbance at 260nm. Reverse transcription reaction was performed using 2.5μg RNA using Qiagen Reverse Transcription Kit (Cat. No. 205311) as per manufacturer's instructions to synthesize cDNA using oligo(dT). For quantitative real-time RT-PCR analysis, 100ng of cDNA template, 10μL of SYBR green master mix (Qiagen; Cat. No. 204054), and respective Qiagen primer sets for HO-1 (QT00095270), NQO-1 (QT00094367), catalase (QT01058106), G6pd (QT00120750), glutamyl cysteine ligase (catalytic) (GCLC) (QT QT00130543), glutamyl cysteine ligase (modulatory) [GCLM] (QT00174300), Nrf2 (QT00095270), and Keap1 (QT00147371) were used and analyzed in a Light Cycler real-time thermocycler (Roche Bio). Copy numbers of cDNA targets were quantified using Ct values and the messenger RNA (mRNA) expression levels for all samples were normalized to the level of the housekeeping gene Arbp1 (QT00249375) or GAPDH (QT01658692).
Frozen sections at 5μm thickness were obtained from the 3- and 6-month-old NTG and TG mice hearts to localize Keap1/CryAB by immunostaining method as reported previously (42). The sections attached to the cover slips were fixed with 3.7% paraformaldehyde for 10min, permeabilized with 0.25% Triton X-100 in TBS containing 0.01% Tween-20 for 5min, and blocked for 30min with 1% BSA in 0.01% tris buffered saline-Tween-20 (TBST) and incubated with primary anti-CryAB/Keap1 at a dilution of 1:2000 and 1:500, respectively, in 0.01% TBST containing 1% BSA for 1h at room temperature. After treatment of sections with the chosen primary antibodies, they were incubated with secondary anti-rabbit Alexa Fluor 488–conjugated and secondary anti-mouse AlexaFluor 586–conjugated antibodies (1:1000 dilution), wherever necessary, for 1h at room temperature. After thorough washing with 1×TBS-T the sections were mounted with the antifade mounting medium (Fluoromount-G) and viewed with 510 Meta Zeiss Confocal (Carl-Zeiss, Inc.) microscope at a magnification of 60×for observation of specific localization and interactions between CryAB and Keap1 proteins. WB analysis was done on the cytosolic fraction of the respective mouse hearts to confirm the active Keap1 protein levels at 3 and 6 months of age. Further, we performed IP with anti-CryAB-ab followed by WB with anti-Keap1-ab to indentify and quantify the interactions between the Keap1 and mutant CryAB in supernatant (cytosolic) and pellet (cytoskeletal & nuclear) fractions of NTG and TG mouse hearts at 6 months of age.
Data are expressed as mean±standard deviation. Analysis of variance followed by the Student–Newman–Keul multiple comparison tests was used to determine the significant differences between the groups. p-Values <0.05 were considered significant.
We have recently reported that moderate overexpression of the missense human R120G mutant CryAB associated with protein aggregation causes cardiomyopathy and dysregulation of glutathione homeostasis, termed RS (29, 30). We have repeated these experiments to validate our previous findings in the hR120GCryAB TG overexpressors in 3- and 6-month-old mice that were subsequently used for Nrf2 studies. We determined heart-to-body weight ratios and found no differences between the NTG and TG at 3 months, but the heart-to-body weight ratio was significantly increased for mutant CryAB-TG compared with NTG mice at 6 months (Fig. 1A, B). Given that cardiac hypertrophy occurred in these hearts at 6 months, we next assessed the major intracellular thiols, reduced (GSH), and oxidized (GSSG) glutathione ratios, for evidence of redox imbalance linked to the pathogenic events. Interestingly, we found similar intracellular GSH levels between the NTG and TG at 3 months, but these values were significantly increased in the TG mouse hearts at 6 months along with profoundly increased GSH/GSSG ratios (Fig. 1C, D). Since the TG hypertrophied hearts exhibit enhanced reducing power, we tested the hypothesis that Nrf2, a master transcriptional regulator of numerous antioxidants/cytoprotective genes, played a role in pathogenic redox imbalance (18).
Apart from basal nuclear expression of Nrf2 under unstressed conditions (18), increased production of ROS, RNS, and other electrophile chemical species induces Nrf2 activation and its nuclear translocation (32, 34, 37, 41). To quantify the redox stress and ROS formation, we utilized EPR spectroscopy with redox-sensitive spin probes (Fig. 2). With the spin probe CMH that in the presence of ROS is oxidized to form its EPR detectable nitroxide radical, a prominent increase in radical formation is seen in the TG compared to NTG myocardium at 3 months, but these signals are not distinguishable at 6 months (Fig. 2A, B). We next determined the levels of ROS using two independent fluorescent probes (DCFH-DA/DHE) of NTG and TG hearts at 3 and 6 months, respectively (Fig. 2C–E). These results confirmed that increased ROS formation occurs in 3-month-old TG compared to NTG mice, suggesting the development of oxidative stress in the TG mice at 3 months. In contrast, at 6 months no increase was seen in the TG compared to the NTG myocardium, suggesting that reductive compensation may occur.
Together, these results are consistent with the notion that alterations of ROS might underlie early activation of signaling cascades, including Nrf2-Keap1 pathway and contribute to subsequent development of pathologic cardiac hypertrophy and GSH imbalance in vivo. Next, we have assessed whether this apparent excess ROS generation in the TG hearts had direct consequence on Nrf2 nuclear translocation and transactivation of key antioxidants.
Under physiological conditions, Nrf2 is tethered to its cytosolic partner, Keap1 and degraded by the ubiquitin/proteasome pathway (32, 37). In turn, the release of Nrf2 from its negative regulator translocates into the nucleus, binds to recognition sequences, and mediates the transcriptional activation of downstream target genes. We hypothesized that increase in ROS signaling by hR120GCryAB expression would induce Nrf2 nuclear translocation during compensatory hypertrophy. To determine the intracellular localization of Nrf2, we isolated cytosolic and nuclear fractions from NTG and TG myocardium and performed WB analysis using Nrf2-Ab. We found that Nrf2 expression in the nuclear fraction was eight- and sixfold higher in TG compared with NTG at 3 and 6 months, respectively (Fig. 3A, C, D, F). This prominent nuclear translocation of Nrf2 was coupled with nuclear translocation of mutant CryAB and its phosphorylated form in the TG mouse heart (Fig. 3A, B, D, E). These results suggest that hR120GCryAB expression increases translocation of Nrf2 into the nucleus of TG compared with NTG mice.
Next, to investigate the consequent function of Nrf2 in the hypertrophy hearts, we determined the Nrf2 activity using ARE-oligonucleotide-based transactivation assay in the nuclear extracts of NTG and TG mouse hearts at 6 months of age. The Nrf2 activity was significantly higher (>2.0-fold) in the TG mouse hearts, suggesting its activation along with increased antioxidant capacity (Fig. 4A). To further confirm these results, we have analyzed ubiquitination of Nrf2 in the NTG and TG mouse at 3 and 6 months. Of interest, IP/Co-IP experiments detected significantly lower ubiquitination of Nrf2 in the TG heart extracts (Fig. 4B) than the NTG, indicating that possible dissociation of Nrf2 from its cytosolic negative regulator, Keap1, in TG mouse is consistent with elevated nuclear translocation seen in Figure 3. IP of Keap1 and followed by WB with anti-Nrf2-Ab revealed poor interactions of these proteins in the cytosol of TG mouse when compared with NTG (Fig. 4C), suggesting dissociation of the Nrf2/Keap1 complex and thereby decreased the ubiquitination of Nrf2 as shown in Figure 4B. Taken together, these results indicate that significant dissociation of Nrf2/Keap1 complex prevented proteasomal degradation of Nrf2 and promoted its activation/nuclear translocation in the TG hearts.
We hypothesized that increased translocation and DNA-binding activities of Nrf2 might induce transcriptional activation of downstream targets, including critical antioxidative pathways associated with glutathione homeostasis. To test this hypothesis, we performed real-time RT-PCR using RNA isolated from the 3- and 6-month-old NTG and TG mouse hearts. We selected several potential target genes under the control of NRF2/ARE-dependent regulation, including the major antioxidant genes. Catalase (Fig. 5G), G6pd (29, 30), and NADH-quinone oxidase (Nqo1) were significantly upregulated in the TG compared with NTG mouse hearts at 3 months (Fig. 5A), suggesting that onset of compensatory mechanisms against mutant CryAB induced cardiac hypertrophy. Expression was elevated for Gclm, Gclc, and hemoxygenase-1 (Ho1) in the TG mouse compared with NTG, but these changes did not reach the level of significance (Fig. 5B–F). In 6 months' hearts, demonstrating cardiac hypertrophy, the antioxidant genes Nqo1, Ho1, G6pd, catalase, Gclm, and Gclc were significantly upregulated in TG compared to NTG hearts (Fig. 5I–O). Although mRNA levels for Nrf2 did not altered significantly at 3 months, there was an increasing trend (p<0.072) at 6 months in TG compared to age-matched NTG mice (Fig. 5M), indicating that there might be ARE-dependent transcriptional upregulation of Nrf2 gene. ARE-mediated auto-regulation of Nrf2 has been reported previously (17).
To characterize whether the Nrf2 activation, in part, was responsible for the increased GSH levels and RS in the TG mouse heart (29), we measured protein expression for the major targets of Nrf2 and enzymes that are involved in GSH metabolism. The WB analysis in either cytosol or nucleus showed dramatic increase in various antioxidants (Fig. 6). At 3 months of age, Nqo1 (>2.0-fold), Ho1 (2.0-fold), and catalase (~4.0-fold) were significantly increased and sod-1 was decreased (50%) in TG hearts related to NTG animals. Unexpectedly, increased catalase did not effectively quenched H2O2 (DCFH-DA), which could be due to superoxide-mediated inactivation of catalase in TG mice. It has been previously shown that superoxide radicals could inhibit catalase (16). Interestingly, at 6 months majority of the antioxidants (catalase, G6pd, glutathione peroxidase-1 [Gpx1], and glutathione reductase [Gsr]) (29, 30) and Nqo1 were increased significantly. These results are consistent with those of RT-PCR and indicate that the activation of Nrf2 is strongly associated with significant upregulation of GSH metabolic enzymes and proteins as compensatory mechanisms against pathogenesis of mutant CryAB-induced protein aggregation in the TG mice (Fig. 6A–D). However, no change in the protein levels of γ-GCS, the rate-limiting enzyme for GSH biosynthesis, was observed at either 3 or 6 months of age, indicating possible feed back inhibition on its production/catalytic function. It is conceivable that upon onset of protein aggregation and ROS generation in the TG mouse, there might be possible increased recruitment/utilization of antioxidants to compensate the stress induced by mutant protein overexpression at 3 months than 6 months of age.
Immunofluorescence analysis (Fig. 7) performed in heart cryosections indicated that Keap1 (red) is colocalized with mutant CryAB (green) aggregates (yellow). Although endogenous CryAB (green) and Keap1 (red) were observed, no prominent interactions (colocalization) of CryAB and Keap1 were evident in the Nrf2-KO mouse heart (Fig. 7A). At 3 months, minimal interactions between the Keap1 and mutant protein (anti-CryAB-Ab) were evident. Along with the increased ROS as determined by EPR, colocalization of Keap1 into the tiny protein aggregates might diminish its potential interactions with Nrf2 (Fig. 7C; inset arrows). Obviously, at 6 months the majority of the Keap1 was bound to the protein aggregates in the TG mouse heart (Fig. 7E; inset arrows), indicating its poor ability to interact with Nrf2, which might explain the sustained nuclear translocation and increased activity of this transcription factor. Inset boxes in each panel showing magnified view (100μm) of the Keap1-CryAB colocalization with the aggregates (yellow on merged images).
To determine the effects of mutant CryAB expression on the negative regulator Keap1, we obtained cardiac extracts and probed for the levels of Keap1 in cytosol from NTG and TG mouse at 3 and 6 months. Although protein levels of Keap1 were unchanged between NTG and TG at 3 months, we found a significant decrease (50%) of Keap1 protein in cytosolic fractions in the TG compared to NTG at 6 months (Fig. 8A, B). Further, our results from IP and WB analyses revealed prominent and strong interactions between the Keap1 and mutant CryAB in TG mouse when compared with NTG (Fig. 8C). These robust interactions were confirmed by immunohistochemistry (Fig. 7E; inset box) in which Keap1 appears sequestered into the mutant CryAB aggregates of hR120GCryAB hearts.
Our findings identify Nrf2-Keap1 pathway as a critical regulator of cardiac defense system and demonstrate the contributing mechanism for RS in the MPAC. An imbalance in the redox equilibrium toward an increase in ROS and oxidative stress has been implicated in the pathophysiology of multiple cardiovascular diseases, including cardiac hypertrophy and cardiomyopathy (5, 7, 31, 40). Surprisingly, we have discovered recently that the “RS” is associated with the molecular pathogenesis of MPAC in the hR120GCryAB TG mouse heart (29). Our observations are consistent with links between mutant-protein-induced ROS, and sustained Nrf2 nuclear translocation occurs at 3 months along with onset of cardiac hypertrophy. Further, such activation of Nrf2 is prominent at 6 months with cardiac hypertrophy and RS due to surpassing sequestration of Keap1 in the mutant-CryAB aggregates. Thus, the mechanisms for Nrf2 activation in the MPAC occur at two stages: initially due to ROS generation, and later due to Keap1 dysfunction through its sequestration into the mutant protein aggregates. These results strongly support our hypothesis that “sustained activation and nuclear translocation of Nrf2 in the hR120GCryAB mouse heart will lead to ceaseless transcriptional upregulation of antioxidants contributing to RS.”
Previous reports have authenticated that the ROS and electrophilic stress could be the primary trigger for the dissociation of Keap1-Nrf2 complex and consequent activation of Nrf2 (3, 13, 14). Our multiple observations reveal sustained activation/translocation of Nrf2 into the nucleus of MPAC mouse heart. A fundamental question addressed in this study is whether the mutant-CryAB aggregates could trigger ROS generation in the mouse heart. Previous studies have documented direct evidence that β-amyloid protein aggregates generate hydrogen peroxide and ROS in Alzheimer's disease (10, 38, 39). Here we show that increased ROS/superoxide were generated in the TG mouse heart at 3 months of age with no evidence of cardiac hypertrophy. These observations suggest that there could be an ROS-mediated dissociation of Keap1-Nrf2 complex in the TG mouse at 3 months age as previously demonstrated in other models of Nrf2-dependent protection (3, 14, 18, 19). An important interpretation of these results might be that the mutant-protein that forms self-oligomers is toxic to intracellular environment; in other words, such toxic events could generate ROS, which then would challenge the intracellular defense system (2, 29, 30, 38, 46). At 3 months of age, our results demonstrate that Nrf2 is activated and translocates into the nucleus, which may be attributed to increased ROS generation in the TG mice as evident from the EPR and fluorescent microscopy analysis.
While Nrf2 is activated before the onset of cardiac hypertrophy, the precise mechanism for its sustained nuclear translocation is of particular interest. To understand this aspect, we measured ROS formation in the hR120GCryAB mouse heart at 6 months of age using similar EPR analysis. Interestingly, we observed that there are no significant differences in ROS formation between the control (NTG) and TG mouse hearts, indicating a potential shift in the intracellular redox milieu toward a reductive state. In fact, our previous studies have documented that at 6 months, the intracellular reducing power (glutathione and NADPH) is significantly increased (termed as RS) along with diminished protein carbonylation/lipid peroxidation in the MPAC mouse (29). Although there is strong evidence for initial activation of Nrf2 via increased ROS generation at 3 months of age, unraveling the mechanism for its sustained activation at 6 months is important to understand the progression of RS.
A significant finding of this study is that misfolded proteins such as mutant CryAB play a key role in the activation of Nrf2/ARE pathways by sequestering the Keap1 into protein aggregates. Our immunofluorescence findings reveal prominent colocalization of Keap1 with the mutant CryAB in the TG mouse hearts at 6 months, suggesting potential sequestration of Keap1 favoring Nrf2 dissociation. Continued synthesis of mutant protein in the TG mouse heart challenges the intracellular environment and activates compensatory pathways, including heat-shock protein (Hsp) signaling (29, 35). Our data suggest that mutant protein stress is also able to activate and upregulate the Nrf2-dependent antioxidant response more substantially than the basal/physiological conditions. Further, our central finding authenticated sustained production and self-oligomerization of dominant negative mutant CryAB sequesters Keap1 (susceptible for conformational changes due to its cysteine rich and redox sensitive nature) and leading to its dysfunction that promotes prolonged dissociation of Nrf2-Keap1 complex; thereby, Nrf2 is freed from ubiquitination, leading to constitutive translocation into the nucleus of TG mouse, which might activate reductive compensation pathways to cause RS. Such chronic activation of master transcription factor Nrf2 resulted in sustained transactivation of ARE-dependent antioxidants in the MPAC contributing to RS. In summary, Keap1-Nrf2 segregation may be plausible contributing mechanism in hR120G CryAB-induced RS and cardiomyopathy (Fig. 9).
Usually, mutant proteins serve as bona-fide stimulus of stress response pathways to prevent protein aggregation and enhance quality control by degrading unwanted proteins (29, 36). Here we show evidence, besides induction of stress pathways that are involved either in chaperone function or in ubiquitin proteasome system, for the prolonged activation of major antioxidant pathways at transcriptional and translation levels. The importance of Nrf2 in affording protection during oxidative stress has been reported in several studies (18, 19, 24, 33, 37, 48, 50). In association with other proteins, such as small maf proteins, ARE binding protein-1 and other coactivators, Nrf2 binds to ARE, leading to transcriptional induction of target genes (11, 13, 24, 37, 47). Consistent with the nuclear accumulation of Nrf2 and increased binding to ARE in TG mouse hearts (as shown in Figs. 3 and and4),4), transcription of most of the ARE containing antioxidants was significantly upregulated during MPAC, along with elevated GSH levels, contributing to RS in the TG mouse (29, 30). Profound upregulation of the key antioxidants under the control of Nrf2 activation might be an adaptive mechanism to overcome the mutant protein aggregation expression in TG myopathic hearts.
In this report, we present evidence that sustained Nrf2 activation and its nuclear translocation as one of the causal mechanism for RS in the mouse heart overexpressed with mutant human R120GCryAB transgene. Based on the molecular and biochemical data obtained, we present novel and contributing mechanisms for the RS in this mouse heart failure model that simulates human heart disease.
Indeed, our previous findings have claimed that expression of the antioxidant genes early on (before the onset of cardiac hypertrophy) in the TG mouse could be a potential “Bio-Signature” for the prediction/prognosis of RS-dependent cardiomyopathy (30). The current results demonstrate that there is ROS generation and Keap1 sequestration into protein aggregates, which leads to dissociation of Nrf2-Keap1 complex and promotes Nrf2 activation to cause or contribute to RS/PAC. Thus, future experiments should test if abrogating Nrf2 (partial or complete knockdown) could prevent RS/protein aggregation and thereby rescues the TG mouse from cardiac hypertrophy/heart failure.
Beginning Grant-In Aid Award from the American Heart Association, Western States Affiliate (0865015F; July 2008–December 2010), and Pilot Grant on Aging (Center for Aging, University of Utah; July 2009–June 2011) to Namakkal Soorappan Rajasekaran, Ph.D., resources from Dr. Ivor Benjamin's laboratory and Dr. Jay L. Zweier's laboratory (NIH grants HL63744, HL65608, and HL38324) provided support for this work. The authors appreciate the helpful suggestions and comments from Dr. John R. Hoidal (Chair of Internal Medicine, University of Utah) and Dr. Dennis Winge (Professor of Medicine). Authors thank Dr. Goutam Karan for his expert help in performing whole-heart images/microscopy. Mrs. Vasanthi Rajasekaran provided extensive technical assistance in performing WBs, scanning and density analysis of images, and preparing graphs. Ms. Jennifer Schroff provided editorial assistance.
No competing financial interests exist.