Mitophagy Facilitates Cytosolic Proteostasis to Preserve Cardiac Function

Figures

• 

  Figure 1.. Loss of TRAF2 results in mis-localization of DESMIN in the cytosol and increased protein aggregation in the cytosol and mitochondria. A Representative images showing immunohistochemical staining for DESMIN (white) and p62 (green) in hearts from Traf2 floxed mice with the Myh6MerCreMer transgene (TRAF2-icKO) or Traf2 flox/flox (TRAF2 fl/fl) control mice. Adult mice were administered tamoxifen in chow for one week and sacrificed 3 weeks later. White arrows indicate DESMIN co-localized with p62-positive protein aggregates in TRAF2-icKO hearts. B, C Immunoblotting for protein aggregation markers p62 and polyubiquitin in the cytosolic and mitochondrial biochemical fractions from TRAF2-icKO and TRAF2 fl/fl control mouse hearts treated as in A. Immunoblotting for VDAC and GAPDH is shown to demonstrate separation of mitochondrial and cytosolic fractions, respectively. Ponceau S staining is shown to assess total protein loading. D, E Quantitation of the cytosolic (D) and mitochondrial (E) levels of polyubiquitin and p62 from panels B and C respectively. Polyubiquitin and p62 levels were normalized to total protein as assessed by Ponceau S Staining and expressed as a fold change relative…

• 

  Figure 2.. Cardiac-myocyte specific ablation of TRAF2 induces increased protein aggregation in mitochondria. A Representative super-resolution imaging of TRAF2-icKO and TRAF2 floxed control hearts following immunohistochemical staining for polyubiquitin (red) and p62 (green) to identify protein aggregates, and COXIV (blue) to identify the inner mitochondrial membrane. B, C Line scan analysis was performed along the yellow lines shown in merged images in panel A to assess the localization of p62 and polyubiquitin relative to COXIV. TRAF2 floxed hearts demonstrate p62 and polyubiquitin outside the COXIV-positive region (arrows in panel B), whereas TRAF2-icKO hearts demonstrate p62 and ubiquitin inside the region delineated with COXIV (arrows in panel C), indicating localization within the mitochondria. D, E Representative images showing immunohistochemical staining for polyubiquitin (green), p62 (red), and COXIV (blue) in postnatal day 1 TRAF2 floxed neonatal mouse cardiac myocytes (NMCMs) treated with adenoviral Cre (Ad-Cre) or adenoviral LacZ (Ad-LacZ) control virus, 3 days after adenoviral treatment. Arrows in magnified inset (E) from Ad-Cre treated cells (marked by dotted line in D) point…

• 

  Figure 3.. R120G mutant CRYAB protein localizes to mitochondria. A Representative images of immunohistochemical staining for mitochondrial marker COXIV (red) and GFP (green) in HEK293 cells transfected with GFP-tagged wild-type CRYAB (GFP-CRYAB-WT) or GFP-tagged CRYAB with the aggregation-prone R120G mutation (GFP-CRYAB-R120G). White arrows point to co-localization of COXIV with mutant GFP-CRYAB-R120G. B Representative images of induced pluripotent stem cell (iPSC)-derived cardiac myocytes from iPSC lines homozygous for knock-in of R120G mutation and isogenic controls expressing mKate-tagged α-actinin and stained with antibody against polyubiquitin. White arrows point to co-localization of α-actinin with polyubiquitin in aggregates. C Representative immunoblot demonstrating expression of CRYAB in mitochondria rich and cytosolic sub-cellular fractions from iPSC-derived cardiac myocytes homozygous for knock-in of R120G mutation (R120G) and isogenic controls (WT). GAPDH and COXIV are shown as markers for cytosol and mitochondrial fractions respectively, and Ponceau S staining is shown for protein loading control. D, E Quantitative evaluation of CRYAB abundance in mitochondrial (D) and…

• 

  Figure 4.. Aggregate-prone mutant proteins accumulate in the mitochondria. A Schematic depicting experiments for mitochondrial uptake of exogenously expressed proteins in HEK293 cells followed by mitochondrial protection assay. See methods for details. B, C GFP-CRYAB-WT or GFP-CRYAB-R120G were expressed in HEK293 cells and mitochondria-free cytosolic fractions were generated (as shown in input lanes). Cytosolic fractions were then incubated with isolated mitochondria from untransfected cells. Half of the mitochondrial pellet was treated with digitonin, shown as ‘+’ (and the other half with diluent shown as ‘-’) to disrupt the outer mitochondrial membrane and assessed by immunoblotting to determine the import of GFP-CRYAB into the mitochondria. GFP-CRYAB-R120G demonstrated increased association with mitochondria in comparison to GFP-CRYAB-WT, with less depletion by digitonin treatment (quantitated in C and expressed as fold change over control), indicating increased mitochondrial import of CRYAB-R120G and its retention within mitoplasts as compared with CRYAB-WT. P values shown are by one-way ANOVA followed by Tukey’s post-hoc testing. D, E GFP-BAG3-WT and mutant GFP-BAG3-P209L were…

• 

  Figure 5.. Transgenic CRYAB-R120G mice exhibit mitochondrial CRYAB localization with upregulation of TRAF2 and induction of mitophagy. A-D Representative immunoblot (A) demonstrating expression of CRYAB, TRAF2 and mitochondrial proteins, namely VDAC and COXIV, in crude whole heart extracts from 20–24 week-old CRYAB-R120G mice (R120G) or littermate controls (WT). R120G hearts exhibited increased TRAF2, reduced VDAC, and reduced COXIV (quantitated in B, C, and D, respectively) versus wild-type. For quantitation, samples were normalized to GAPDH to control for loading. P values shown are by t-test. E-H Representative immunoblot (E) on cytosolic and mitochondrial fractions from heart tissue from 20–24 week-old CRYAB-R120G mice (R120G) or littermate controls (WT), demonstrating expression of CRYAB, TRAF2, p62, and polyubiquitinated proteins. GADPH was used to demonstrate separation of the cytosolic fraction, and VDAC and COXIV were used as markers of the mitochondrial fraction. Ponceau S staining is shown as loading control. Mitochondrial fractions from R120G hearts demonstrated increased levels of the protein aggregation markers polyubiquitin and p62, as well as TRAF2 (quantitated in F, G, and…

• 

  Figure 6.. Reduction in cardiac myocyte TRAF2 exacerbates CRYAB-R120G-induced cardiomyopathy. A Experimental approach to deplete TRAF2 in cardiac myocytes. One Traf2 floxed allele and the cardiac myocyte-specific MerCreMer tamoxifen-inducible Cre allele were introduced into CRYAB-R120G transgenic line to generate CRYAB-R120G mice with cardiac myocyte-specific inducible deletion of one Traf2 allele (termed R120G TRAF2-icHET). These mice were compared to R120G TRAF2 fl/+ mice (without the MerCreMer transgene) and to TRAF2 fl/+ and TRAF2-icHET control mice without CRYAB-R120G. Mice were treated with tamoxifen chow for one week starting at 8 weeks of age. Echocardiography was done at 20 weeks of age, and survival was monitored to 60 weeks. B Representative M-mode echocardiograms for 20-week-old mice in indicated groups. White arrows indicate left ventricular internal diastolic diameter, red arrows indicate left ventricular internal systolic diameter. C Assessment of left ventricular systolic function by fractional shortening in 20-week-old mice. Fractional shortening was reduced in R120G TRAF2-icHET mice versus R120G TRAF2 fl/+ controls. P values shown are by one-way ANOVA followed by Tukey’s…

• 

  Figure 7.. Reduction in cardiac myocyte TRAF2 increased mitochondrial protein aggregates in R120G transgenic myocardium. A Representative immunoblot depicting expression of TRAF2, polyubiquitinated proteins and p62 in cytosolic and mitochondrial fractions from 20-week-old CRYAB-R120G mice with cardiac myocyte-specific inducible deletion of one Traf2 allele at 8 weeks of age (termed R120G TRAF2-icHET), and similarly treated R120G TRAF2 fl/+ mice, TRAF2-icHET and TRAF2 fl/+ as control as modeled in Figure 6A. GADPH was used to demonstrate separation of the cytosolic fraction, and LONP1, VDAC and COXIV were used as markers of the mitochondrial fraction. B Quantitation of TRAF2 levels in the cytosolic fractions from TRAF2-icHET and TRAF2 fl/+ mice treated as in A. P value shown is by Mann-Whitney test. C Quantitation of polyubiquitinated proteins in the mitochondrial fractions from R120G TRAF2-icHET and R120G TRAF2 fl/+ mice treated as in A. P value shown is by Mann-Whitney test.

• 

  Figure 8.. TRAF2 overexpression rescues CRYAB-R120G-induced cardiomyopathy. A Experimental approach to overexpress TRAF2 in cardiac myocytes. Wild-type (WT) and transgenic CRYAB-R120G mice were injected with AAV9 expressing TRAF2 (AAV9-TRAF2) or control AAV9 expressing GFP (AAV9-GFP). Cardiac troponin T promoter was employed in both viruses to specifically target cardiac myocytes. AAVs were injected via tail-vein at a dose of 7.0 X 10^11 viral genomes/mouse at 8 weeks of age. Echocardiography was performed at 20 weeks, and survival was monitored to 100 weeks. B Representative M-mode echocardiograms for 20-week-old mice in indicated groups modeled as in A. White arrows indicate left ventricular internal diastolic diameter; red arrows indicate left ventricular internal systolic diameter. C Assessment of systolic function by fractional shortening in 20-week-old mice. Left ventricular fractional shortening is improved in CRYAB-R120G mice treated with AAV9-TRAF2 versus AAV9-GFP. P values are by one-way ANOVA followed by Tukey’s test for multiple comparison testing between groups. D Survival of WT and CRYAB-R120G mice injected with AAV9-GFP and AAV9-TRAF2 viruses. CRYAB-R120G AAV9-TRAF2 mice…

• 

  Figure 9.. TRAF2 overexpression improves mitochondrial protein aggregation and normalizes DESMIN localization in CRYAB-R120G transgenic hearts. A-E Representative immunoblots (A) of cytosolic and mitochondrial fractions from 20-week-old wild-type (WT) AAV9-GFP, WT AAV9-TRAF2, CRYAB-R120G AAV9-GFP, and CRYAB-R120G AAV9-TRAF2 hearts. Mice were injected with AAV virus at 8 weeks, as previously described in Figure 8A. Quantitation of cytosolic TRAF2, mitochondrial TRAF2, mitochondrial polyubiquitin, and mitochondrial p62 levels are shown in B, C, D, and E respectively, with normalization to total protein level as assessed by Ponceau S staining. Values shown as fold change over control. P values are by t-test for B, D and E; and by Mann-Whitney test for panel C. F Representative images depicting immunohistochemical analysis to assess DESMIN localization in cardiomyocytes. CRYAB-R120G AAV9-GFP hearts exhibited disruption of normal striated DESMIN pattern seen in WT AAV-GFP and WT AAV9-TRAF2 mice, with localization of DESMIN to aggregates. DESMIN localization (pseudo colored in white) was partially rescued with reduced aggregates in CRYAB-R120G AAV9-TRAF2 mice. G Quantitative assessment of protein…