|Home | About | Journals | Submit | Contact Us | Français|
There has been an increasing recognition that mitochondrial perturbations play a central role in human heart failure. Discovery of mitochondrial networks, whose function is to maintain the regulation of mitochondrial biogenesis, autophagy (‘mitophagy’) and mitochondrial fusion/fission, are new potential therapeutic targets. Yet our understanding of how the molecular underpinning of these processes is just emerging. We recently identified a role of the SWI/SNF ATP-dependent chromatin remodeling complexes in the metabolic homeostasis of the adult cardiomyocyte using cardiomyocyte-specific and inducible deletion of the SWI/SNF ATPases BRG1 and BRM in adult mice (Brg1/Brm double mutant mice). To build upon these observations in early alterated metabolism, the present study looks at the subsequent alterations in mitochondrial quality control mechanisms in the impaired adult cardiomyocyte. We identified that Brg1/Brm double-mutant mice exhibited an increased mitochondrial biogenesis, increases in ‘mitophagy’, and alterations in mitochondrial fission and fusion that led to small, fragmented mitochondria. Mechanistically, increases in the autophagy and mitophagy-regulated proteins Beclin1 and Bnip3 were identified, paralleling changes seen in human heart failure. Cardiac mitochondrial dynamics were perturbed including decreased mitochondria size, reduced number, and altered expression of genes regulating fusion (Mfn1, Opa1) and fission (Drp1). We also identified cardiac protein amyloid accumulation (aggregated fibrils) during disease progression along with an increase in pre-amyloid oligomers and an upregulated unfolded protein response including increased GRP78, CHOP, and IRE-1 signaling. Together, these findings described a role for BRG1 and BRM in mitochondrial quality control, by regulating mitochondrial number, mitophagy, and mitochondrial dynamics not previously recognized in the adult cardiomyocyte. As epigenetic mechanisms are critical to the pathogenesis of heart failure, these novel pathways identified indicate that SWI/SNF chromatin remodeling functions are closely linked to mitochondrial quality control mechanisms.
As heart failure treatment has evolved, increasing recognition mitochondria’s role in the pathogenesis of disease has emerged1. New techniques allow the discovery of mitochondrial networks, whose function is maintained by three processes, including 1) mitochondrial biogenesis (increase in mitochondrial number); 2) mitophagy; and 3) continuous mitochondrial fission (division) and fusion2, 3. Impaired mitochondrial biogenesis is a feature of myocardial hypertrophy and end-stage ischemic heart failure in humans4, while autophagy has been central to defects in human heart failure5, 6. Evidence of altered mitochondrial dynamics (fission and fusion) in human heart failure, specifically in idiopathic dilated cardiomyopathy, has recently been reported7.
Changes in the mitochondrial network influence the production of ROS and ATP production, most notably offset by activating mitochondrial biogenesis and mitophagy in exercise and other therapeutic modalities. Complementing mitochondrial biogenesis is the ongoing process of “mitophagy” (mitochondrial autophagy), where damaged/dysfunctional mitochondria are targeted for destruction via lysosomal degradation. Here mitochondrial proteins such as BNIP3L and BNIP3 are abundant in the heart and act as targeting molecules that attract autophagosome to mitochondria and are critical to maintaining cardiac function3. Lastly, regulation of mitochondrial dynamics (fission and fusion) involve the redistribution of mitochondrial content8. With the enormous continuous ATP demand of the cardiomyocyte, the importance of maintaining mitochondrial quality control is of particular significance in cardiac disease, including heart failure, ischemic heart disease/myocardial infarction, and diabetes9–11.
Epigenetic regulation of heart failure occurs by several key mechanisms, including ATP-dependent chromatin remodeling12. One ATP-dependent chromatin remodeling complex in cardiomyocytes are the SWI/SNF (SWItch/Sucrose Non-Fermentable) complexes, comprised of 9–12 subunits, including one of two ATPases critical to their function 1) BRG1 (Brahma-related gene 1) or 2) BRM (Brahma)13, 14. Recent studies demonstrate a role of BRG1 and BRM in maintaining homeostasis in the adult cardiomyocyte. These studies identified that cardiomyocyte-specific and inducible deletion of BRG1 and BRM in adult mice (Brg1/Brm double mutant mice), alterations in glucose-6-phosphate, fructose-6-phosphate, and myoinositol at day 10 post-induction occurred before severe cardiomyopathy is seen15. The identification of these early metabolic changes led to the current study to identify the underlying mechanisms by which BRG1 and BRM deletion affect mitochondrial quality control mechanisms in the impaired adult cardiomyocyte.
In the present study, we build upon our findings that Brg1/Brm double mutant hearts have early metabolic changes at day ten (10) by investigating the changes occurring 15 days after inducing Brg1 deletion, just before the development of a severe cardiomyopathy. We identified alterations in mitochondrial biogenesis, mitochondrial autophagy (mitophagy), and mitochondrial fission and fusion just before the death that occurred in all mice by day 22. We demonstrate a previously undescribed link between the chromatin remodeling SWI/SNF components BRG1 and BRM with mitochondrial homeostasis. On a more practical note, the present study should give those develop BRG1 and BRM inhibitors for cancer pause16.
All mouse experiments were approved by the Institutional Animal Care and Use Committees (IACUC) review board at the University of North Carolina at Chapel Hill. The αMHC-Cre-ERT mice were obtained from The Jackson Laboratory (#005657, Bar Harbor, ME); the Brg1 conditional mutant mouse line and Brm constitutive mutant mouse line have been described previously17–19. Genotyping of the αMHC-MerCreMer, Brg1 floxed, and Δfloxed alleles and the Brm mutation were genotyped by PCR as previously described17–20.
Immunohistochemistry (IHC) was performed on fixed histological section using an anti-BRG1 rabbit polyclonal antibody (Millipore #07-478, Temecula, CA), anti-GRP78 mouse monoclonal antibody21, anti-CHOP (GADD153), or phospho-eIF2, as previously described22, 23. Hearts analyzed by TEM were viewed on a LEO EM910 transmission electron microscope operating at 80 kV (LEO Electron Microscopy, Thornwood, NY). An average of 2,500 mitochondria were analyzed from 30 fields from multiple levels from three hearts per mouse cohort.
Autophagic flux was determined in vivo from the titration of bafilomycin A1 as previously described24. Antibodies for LC3B isoforms (Sigma-Aldrich #L7543) and beclin were used for western immunoblot analysis of markers of autophagy (Cell Signaling #3495P, Beverly, MA).
Protein aggregation was assessed using an Enzo Proteostat Protein Aggregation Assay (#ENZ-51023, Farmingdale, NY) in conjunction with their pre-formulated aggregation standards (#ENZ-51039) and pre-amyloid oligomer staining was performed as described in detail25.
Total RNA was isolated from cardiac tissue homogenized with a TissueLyser LT (Qiagen N.V. #69980, Venlo, The Netherlands) using
Trizol (Life Technologies #15596-026, Carlsbad, CA) according to the manufacturer’s instructions. TaqMan gene expression assays (Life Technologies) were performed using universal TaqMan master mix (Life Technologies #4304437) commercial probes, as described previously24.
Mitochondrial number was quantified by qPCR on DNA prepared from whole-heart homogenates using the DNAeasy Blood and Tissue Kit (Qiagen #69506). SYBR Green and primers for the mitochondrial DNA included mtNd1, mtCO1, and mtCytb1; H19 was run in parallel as a nuclear DNA control as previously detailed26.
SigmaPlot (Systat Software, Inc., San Jose, CA) was used to determine significant statistical difference by One-way ANOVA followed by post-hoc analysis using the Holm–Sidak method or a Student’s t-test, as indicated. A p value < 0.05 was considered significant.
Additional materials and methods detail may be found in the online supplement.
To investigate the combined role of BRG1 and BRM in adult cardiomyocytes, we crossed mice carrying an inducible, cardiomyocyte-specific αMHC-Cre-ERT transgene to Brg1fl/fl; Brm−/−mice. Adult Brg1fl/fl; αMHC-Cre-ERT+/0; Brm−/− mice were then fed tamoxifen-fortified chow for 7 days to induce the Brg1 floxed-to-Δfloxed excision event, which was confirmed by both PCR and IHC (Supplemental Fig. 1A, B). These mice (herein referred to as Brg1/Brm double mutants) were monitored by echocardiography until they died, which occurred within 22 days of initiating the tamoxifen diet (Supplemental Fig. 1C). A progressive heart failure occurred prior to death in Brg1/Brm double-mutant but not in control mice (Table 1). Six groups of control mice were analyzed in parallel (listed in Supplemental Fig. 1C) to ensure that we were observing a Brg1/Brm double-mutant phenotype rather than an artifact, such as tamoxifen itself or tamoxifen-induced Cre dysfunction, which has been previously described27.
Considering the severity of the cardiac phenotype and the rapid lethality of mice lacking Brg1 and Brm in their cardiomyocytes, the histopathology was surprisingly mild. Analysis of H&E-and Mason’s Trichrome-stained sections revealed that the Brg1/Brm double-mutant phenotype ranged from relatively normal to moderate vacuolization (Fig. 1A) with no evidence of lipid accumulation by oil red O staining (Supplemental Fig. 2) and with fibrosis increased only in rare individuals (Fig. 1B, Supplemental Fig. 3). Next, we assessed fetal cardiomyocyte gene expression because it is a typical pathological hypertrophic response that occurs during heart failure. Skeletal muscle actin was significantly decreased at both time points in the Brg1/Brm double-mutant hearts compared with controls (Fig. 1C). βMHC fetal gene expression was significantly elevated at day 10 post-tamoxifen induction (early time point) but not at 1-day pre-mortem (late time point) (Fig. 1C and 1D). In contrast, Bnp and Anf mRNA were not changed significantly at either time point. These data are not representative of a typical pathologic hypertrophy/heart failure response.
Ultrastructural analysis of the Brg1/Brm double-mutant heart by TEM was distinct by the presence of alterations limited to the interfibrillar areas (mitochondrial compartment). In all of the Brg1/Brm double-mutant hearts, we identified mitochondria degeneration (Fig. 2A and 2B) and an increase in double-membrane bound vacuoles with mitochondrial remnants within (Fig. 2B) that were absent in all of the parallel control hearts (Fig. 2C). Ultrastructurally, these vacuoles appeared similar to those found to be increased in the process of autophagy, specifically autophagosomes. Autophagy is a lysosomal degradation pathway for cytoplasmic material and organelles, starting at the endoplasmic reticulum-mitochondria interface28. The conserved autophagic process shuttles cytoplasmic components to fuse with lysosomes. This stepwise engulfment of cytoplasmic material by the phagophore, maturing into a double-membrane-bound vesicle, forms the autophagosome. To confirm that Brg1/Brm double-mutant hearts had quantitative alterations in autophagy, we next analyzed autophagic flux in vivo.
The role of autophagy in cardiovascular biology has proven to be context dependent, widely characterized in cardiomyocytes, and critical to the maintenance of cardiovascular homeostasis and function29. Depending on context, there is a window of optimal autophagic activity, whereby increases or decreases may be protective depending upon the cardiac stress disease state30, 31. Brg1/Brm double-mutant hearts exhibited a significant increase in autophagic flux, as illustrated by the LC3II:LC3I ratio by western immunoblot after bafilomycin A1 treatment, compared with control mice run in parallel (Fig. 3A). In parallel with this evidence of increased autaphagic flux, we identified increased Beclin 1 by western immunoblot analysis (Fig. 3B). With evidence of increased autophagy, we next investigated mechanisms responsible for upregulating autophagy.
Central to the regulation of autophagy in heart disease, the Beclin 1 protein functions as a scaffolding protein assembling Beclin 1 interactome to regulate Class II PI3K/VPS34 activity, which tightly controls autophagy at multiple stages32. The Brg1/Brm double-mutant mice had significantly increased Beclin 1 protein expression compared with control group mice (Fig. 3B). Autophagy is also regulated at the transcriptional level, with increased Atg5, Atg7, Atg12, Bnip3, and Vps34 mRNA to enhance autophagy33. RT-qPCR analysis of these genes demonstrated the Brg1/Brm double-mutant hearts had a significant increase in Bnip3 mRNA levels, while Atg12 and Vps34 were significantly decreased compared with control hearts at day 15 (Fig. 3C). At the earlier day 9 time point, Brg1/Brm double-mutant hearts had increased Vps34 mRNA and decreased Bnip3 mRNA (Supplemental Fig. 4), illustrating the dynamic expression of autophagy regulators. The transcriptional and epigenetic network regulating autophagy is emerging (as recently reviewed33) that underscore the diverse ways in which autophagy can be regulated depending on the stimulus. To test whether BRG1 and BRM occupy the Bnip3 promoter, we performed quantitative ChIP assays. Significant enrichment of BRG1 and BRM was detected in cardiac tissue (Fig. 3D), which strongly suggests that BRG1- and BRM-catalyzed SWI/SNF complexes directly regulate Bnip3. BNIP3, primarily located in the mitochondria as an integrated protein34, perturbs outer membrane integrity35, induces permeabilization of the inner mitochondrial membrane36, and can stimulate mitophagy directly by triggering depolarization and Parkin recruitment37. Taken together, these findings illustrate a novel role for BRG1 and BRM in the maintenance of cardiomyocyte autophagic flux, specifically related to the mitophagy as illustrated by the TEM analysis and the direct regulation of Bnip3.
BNIP3 alters cellular mitochondrial dynamics, a continuous process by which mitochondria fuse together (fusion) and subsequently divide (fission) to maintain their functional balance38. Specifically, increasing BNIP3 expression can induce mitochondrial fragmentation (fission) in cells34, 39. Understanding this link led us to evaluate further the Brg1/Brm double-mutant heart mitochondrial phenotype in our TEM studies. Here TEM revealed an apparent increase in Brg1/Brm double-mutant heart mitochondrial fragmentation compared with controls (Fig. 4A). Analysis of mitochondria over multiple LV regions revealed that the number of small mitochondria identified in the Brg1/Brm double-mutant hearts was skewed with a marked increase in small mitochondria with significantly smaller areas (Fig. 4B and 4C). To put in context the apparent increase in small mitochondria, we determined mitochondrial number by qPCR of three mitochondrial encoded genes normalized to the nuclear-encoded H19 and found a significant decrease in the number of mitochondria (Fig. 5A). The fusion and fission of mitochondria is a dynamic process that cells use for responding to mitophagy and maintaining protein quality. The regulation of mitochondrial dynamics by fusion and fission is regulated by four dynamin-related GTPases, including DRP1 and FIS1 (fission)40, 41 and mitofusins (e.g. MFN1) in the outer membrane and OPA1 in the inner membrane (fusion)42, 43. RT-qPCR analysis of these GTPases in the Brg1/Brm double-mutant hearts demonstrated significant decreases in Mfn1, Opa1, and Drp1 in late disease (Fig. 5B and 5C) but not at an earlier stage (Supplemental Fig. 5). These findings demonstrate the interesting juxtaposition of increased mitophagy and mitochondrial fission, which has been shown to allow the selective elimination of damaged mitochondria by autophagy44, 45
Histological analysis of the Brg1/Brm double-mutant hearts stained with H&E revealed stereotypical pink accumulation inside of the cardiomyocytes, resembling the unfolded proteins that accumulate in cardiac amyloidosis (Supplemental Fig. 3E). Since the accumulation of unfolded proteins in mitochondria has been shown to induce mitophagy in neurons46, we next investigated the presence of unfolded proteins using a colorimetric assay that has been used to identify amyloid and other unfolded proteins in experimental disease processes47–51. At an early time point, Brg1/Brm double-mutant hearts did not have an increase in unfolded proteins (Fig. 6A, left). However, at later time points when mitophagy and altered fission and fusion were present (Figs. 2–5), a significant increase in unfolded proteins was present with a ~3-fold increase compared with control hearts (Fig. 6A, right). The pathogenic effects of the accumulation of amyloid fibrils, independent of soluble precursors, have been described in the heart52. Recent studies have identified a critical role of pre-amyloid oligomers (PAOs) that form before amyloid/insoluble fibrils in mediating heart failure directly, paralleling the pathogenesis of common neurodegenerative diseases, including Alzheimer’s disease, Parkinson’s disease, and Huntington’s disease53.
Using the A11 antibody, we investigated the presence of PAOs in early and late Brg1/Brm double-mutant heart samples (Fig. 6B). The detection of low levels of PAO by the A11 antibody (green) present in cardiomyocytes (red) detected by an antibody against the heavy chain myosin II using immunofluorescence indicates that soluble PAOs were increasing at day 15 in Brg1/Brm double-mutant hearts (Fig. 6B). While not reaching significant levels, these findings suggest that the protein aggregates detected in the heart at day 15 (Fig. 6A) go through a PAO state very quickly to accumulate amyloid in significant amounts when Brg1 and Brm have been ablated in adult cardiomyocyte in vivo.
In response to stress, such as that induced by oxidized proteins, the heart induces the unfolded protein response (UPR), which includes the endoplasmic reticulum (ER) chaperone protein GRP7854. GRP78 is found in the endoplasmic/sarcoplasmic reticulum bound to PERK, IRE1, and ATF6 receptors to prevent their activation of the UPR55. In the presence of unfolded proteins, GRP78 preferentially shifts to these proteins to stabilize their confirmation, leaving PERK, IRE1, and ATF6 to signal survival and apoptotic signaling downstream55. Having established the presence of increased amyloid in these hearts, we next investigated cardiac expression of GRP78 by IHC and identified that Brg1/Brm double-mutant mice had markedly increased GRP78 expression compared with controls (Fig. 7A) at the late time point where PAOs and amyloid were increased.
The link between the UPR and cardiovascular disease is strengthening56, but UPRs are diverse and include signaling by IRE-1, PERK, and ATF657, 58. Identification of the spliced Xbp-1 by RT-qPCR is one way in which IRE-1 activity is measured58. We identified that Brg1/Brm double-mutant hearts had a significant increase in the spliced Xbp-1 mRNA compared with control hearts (Fig. 7B). Downstream of the IRE1-XBP-1 pathway, XBP1 binds the promoter of the C/EBPα (Cebpa) gene to increase its expression59. Brg1/Brm double-mutant mice had significantly more Cebpa mRNA expression compared with controls (Fig. 7B). The significant increase in Brg1/Brm double-mutant heart Xbp-1 splice variant mRNA and Cebpa mRNA demonstrate the increased activity via IRE-1 signaling. The relationship between XBP1 and C/EBPβ (Cebpb) is a bit more complex, with evidence that C/EBPβ induces expression of XBP1 protein59. Brg1/Brm double-mutant hearts had diminished Cebpb mRNA levels that were not significant (Fig. 7B).
The stress caused in the ER by the presence of unfolded proteins can also cause signaling through PERK, whereby PERK proteins dimerize and undergo autophosphorylation60. This autophosphorylation leads to downstream phosphorylation of eIF2, which we investigated by IHC (Fig. 7A, column 2). Here we found evidence that the Brg1/Brm double-mutant mice had a higher cardiac eIF2 expression histologically, and next looked at downstream ATF4 and CHOP expression by RT qPCR55. Significant increases in both Chop and Atf3 mRNA were identified by RT qPCR in the Brg1/Brm double-mutant mice (Fig. 7B), with increases in CHOP protein also seen by IHC in the Brg1/Brm double-mutant mice (Fig. 7A, column 3). Activation of ATF4 increases target gene expression, including the transcription factor CHOP (CCAAT/-enhancer-binding protein homologous protein)61, which itself induces the expression of Atf3 mRNA62. While the lack of reliable reagents prevented us from directly measuring ATF6 signaling61, we determined the mRNA levels of other reported genes involved in the UPR. We found that Brg1/Brm double-mutant hearts had either significant down-regulation of Ire-1a, Atf6a, and Grp78 reflecting compensatory responses to the clear GRP78 and IRE-1 signaling present, or no change in genes not as clearly related to the UPR in cardiomyocytes61 (Supplemental Fig. 6).
This study identifies a novel role of the SWI/SNF components BRG1 and BRM in the homeostasis of the adult cardiomyocyte regulating mitochondrial biogenesis, mitophagy, and mitochondrial fission and fusion. While there have been suggestions that SWI/SNF complexes may be related to mitochondrial biogenesis63 and autophagy64 in mammalian cells, the present study is the first to link them additionally to mitochondrial fission and fusion. While these findings parallel changes seen in heart failure phenotypes, BRG1 and BRM have previously been shown to have a role in the development of pathological cardiac hypertrophy. BRG1 interactions with the DNA-binding protein PARP1 have been implicated in the prototypical shift in cardiac MHC isoform expression in pathological cardiac hypertrophy65, and may explain the decreases (instead of increases) of βMHC seen with the heart failure in the present study (Fig. 1B). In the present study, we describe the redundant role of BRG1 and BRM to maintain the adult cardiomyocyte in vivo. Notably, multiple distinct alterations have been identified in this model, with early metabolomics changes previously identified15, accompanying the upregulation of autophagy/mitophagy, altered mitochondrial dynamics, accumulation of PAOs/aggregated protein, and activation of the UPR that culminated in severe bradycardia and death (reported here).
The increased autophagic flux present in Brg1/Brm double-mutant hearts was characterized morphologically by increases in mitochondria in what look like autophagosomes (Fig. 2) and by increased Beclin 1 protein and Bnip3 mRNA (Fig. 3). While the process of autophagy has largely been described in terms of transcriptional regulation, the epigenetic control of autophagic flux is gaining ground33. Multiple transcription factors and epigenetic regulators have been shown to be involved33. We demonstrate that BRG1 and BRM occupy the Bnip3 promoter and regulate its expression as well as autophagic flux. The transcriptional activation of Bnip3, as seen in the Brg1/Brm double-mutant hearts, has been reported to be by C/EBPβ, E2F1, FOXO1/3, HIF1, and STAT333. The enhanced Beclin 1, also seen in the Brg1/Brm double-mutant hearts, has been reported to be regulated by FOXO1/3, NF-κB, p63, and Stat33. BNIP3 is a receptor, like PINK1 and Parkin, regulates mitochondrial autophagy (also known as mitophagy)66, 67, consistent with the morphological findings in Brg1/Brm double-mutant hearts. Increases in BNIP3 expression occur after myocardial infarction and in response to chronic pressure overload hypertrophy and may be a mechanism by which mitochondrial pruning is mediated to get rid of damaged mitochondria during increased stress68. BRG1 and BRM may be linked to the regulation of HIF1, as the SWI/SNF complex regulates HIF-1-dependent mitophagy (regulated through BNIP3 and Beclin1) as a metabolic adaptation to stress in cancer cells69, 70.
The regulation of mitochondrial autophagy and mitochondrial dynamics is logically linked to mitochondrial quality control. Mitochondria are damaged by many environmental stressors, and repair is dependent upon the mitochondrial fission and fusion (broadly known as mitochondrial dynamics) and autophagy71. In fact, mitochondrial fusion and fission are processes essential for the preservation of normal mitochondrial function. In heart failure, OPA1 is important for maintaining normal cristae structure and function, while preserving the inner membrane structure and for protection against apoptosis. Confocal and TEM analysis have demonstrated that failing hearts have small, fragmented mitochondria, consistent with decreased fusion. Reducing OPA1 or MFN1 in cardiomyocytes results in increased mitochondrial fragmentation72, 73. The transcriptional regulation of genes involved with mitochondrial fusion in cardiomyocytes (i.e. OPA1 and MFN1) as well as fission (e.g. DRP1) has been recently shown to be under the control of PGC-1α by coactivating the estrogen-related receptor α74. While the SWI/SNF complex component BAF60a supports PGC-1α co-activation63, the regulation of mitochondrial fission by Brg1/Brm demonstrated here has not been described previously.
Supplemental Fig. 1. Tamoxifen-induced ablation of Brg1 in cardiomyocytes. A. PCR detection of the Brg1 floxed (fl, top panel) and Δfloxed (Δfl, bottom panel) alleles in genomic DNA prepared from cardiac tissue of Brg1fl/fl; αMHC-Cre-ERT mice that were untreated (−) or treated (+) with tamoxifen (TAM). The floxed PCR product is diminished but still detected in TAM-treated mice because the Cre-mediated excision event does not occur in cell types other than cardiomyocytes, such as vascular endothelial cells. NTC, no template control. B (left). IHC showing BRG1 nuclear staining in cardiomyocytes (encircled region) and vascular endothelial cells (arrow) of cardiac tissue sections from control mice (Group 2: Brg1fl/fl; no αMHC-Cre-ERT transgene but treated with TAM). B (right). IHC showing lack of BRG1 staining in cardiomyocytes (encircled region) in cardiac tissue sections from Brg1/Brm double-mutant mice. Presence of BRG1 staining in vascular endothelial cells (arrow) at levels comparable to the control serves as an internal positive control. C. Kaplan-Meier survival curve of mice after administration of tamoxifen (+ TAM) on days 1 through 7. The number of mice for each genotype and ± TAM treatment is listed below.
Supplemental Fig. 2. Oil red O staining of fresh frozen cardiac section to identify neutral lipids in tissue. Sections were counter-stained with hematoxylin using standard techniques to detect nuclei.
Supplemental Fig. 3. Histological analysis of H&E stained Brg1/Brm double-mutant and control hearts. A–C: Representative Brg1/Brm double-mutant cardiac sections. D. Representative control hearts. E: Representative Brg1/Brm double-mutant with vacuole formation (black arrows) and waxy accumulation of materials intracellularly (white arrows).
Supplemental Fig. 4. RT-qPCR analysis of autophagy genes (A, Atg5; B, Bnip3; C, Vps34) in control and double-mutant hearts at day 10 post-tamoxifen induction. Data are normalized to Gapdh and presented as means ± SEM based on 5 independent experiments with significant differences indicated (*, p < 0.05). See Fig. 3C for analysis at later (day 15) time point.
Supplemental Fig. 5. RT-qPCR analysis of mitochondrial fusion and fission genes (A, Drp1; B, Fis1; C, Mfn1; D, Opa1) in control and double-mutant hearts at day 10 post-tamoxifen induction. Data are normalized to Gapdh or 18S and presented as means ± SEM based on 5 independent experiments with no significant differences. See Fig. 6B and 6C for analysis at later (day 15) time point.
Supplemental Fig. 6. RT-qPCR analysis of genes involved in the unfolded protein response normalized to Gapdh in day 15 hearts from control and double-mutant mouse hearts. A. Down-regulated genes (Ire-1a, Atf6a, Grp78). B. Genes not significantly changed (Atf4, Fabp4, Pparg, Pref1). Data are presented as means ± SEM based on 5 independent experiments with significant differences indicated (*, p < 0.05).
We would like to thank Dr. Marco Sandri for his ideas and input in interpreting the relationship between autophagy and mitochondrial dynamics and Dr. Joe Hill for his guidance in measuring autophagic flux. We thank Janice Weaver in the UNC Animal Histopathology Laboratory for assistance in preparing histological specimens and Vicky Madden at the UNC Microscopy Services Laboratory for assistance with transmission electron microscopy.
Sources of Funding
This work was supported by funding from the National Institutes of Health (RO1HL104129 to M.W. and CA125237 to S.B.), a Jefferson-Pilot Corporation (fellowship to M.W.), and the Leducq Foundation (to M.W.).
The authors declare no conflicts of interest.
Compliance with Ethical Standards
All applicable international, national, and/or institutional guidelines for the care and use of animals were followed.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.