Junctophilin-2 Regulates Mitochondrial Metabolism

Right ventricular dysfunction (RVD) is a risk factor for mortality in multiple cardiovascular diseases, but approaches to combat RVD are lacking¹. Therapies used for left heart failure are largely ineffective in RVD, and thus the identification of molecules that augment RV function could improve outcomes in a wide-array of cardiac conditions. Junctophilin-2 (JPH2) is an essential protein that has important roles in cardiomyocytes, including calcium handling/maintenance of t-tubule structure and gene transcription². Additionally, JPH2 may regulate mitochondrial function as Jph2 knockout mice exhibit cardiomyocyte mitochondrial swelling and cristae derangements³. Moreover, JPH2 knockdown in embryonic stem cell-derived cardiomyocytes induces downregulation of the mitochondrial protein mitofusin-2 (MFN2), which disrupts mitochondrial cristae structure and transmembrane potential⁴. Impaired mitochondrial metabolism drives RVD¹, but JPH2’s potential interaction with MFN2 and its impact on RV mitochondrial activity are not well-defined.

Cellular fractionation, co-immunoprecipitation, and pulldown assays of recombinantly expressed and purified JPH2 and MFN2 constructs defined a direct JPH2-mitochondrial link. JPH2 was synthesized and cloned into pET151/dTOPO (Thermo) with deletions generated by Q5 mutagenesis PCR (NEB). His-tagged proteins were purified and tested for interaction by co-immunoprecipitation with MFN2 (Abcam), JPH2 (Invitrogen), isotype IgG, and V5 (VWR) antibodies. Human induced pluripotent stem cells (iPSC) (Allen Institute) were treated with CRISPR-Cas9 to knockout JPH2 (Synthego) and to express green fluorescent protein (GFP), JPH2_FL, or JPH2_Ct (Addgene 52343). Cell proliferation, nucleofection, and differentiation of iPSC into cardiomyocytes (iPSC-CM) were performed as recommended (Allen Institute). iPSC-CM mitochondria were stained with MitoTracker Orange (ThermoFisher) and network morphology was quantified with MiNA (https://github.com/StuartLab). Agilent XFp/XF96 Seahorse examined iPSC-CM mitochondrial respiration. In vitro lipid sensitivity was probed by incubating iPSC-CM with 16 μM oleate, linoleate, and palmitate (Sigma) overnight. Lipid droplets were visualized with LipidTOX Red (FisherScientific). Confocal micrographs were collected on a Zeiss LSM 900 Airyscan 2.0 microscope.

Adult male Sprague-Dawley rats (200-250g) were randomly allocated into three groups: injected with phosphate buffered saline, treated with an intraperitoneal injection of 1x10¹¹ vector genomes of adeno-associated virus serotype 9 encoding GFP (AAV-GFP) one week after subcutaneous monocrotaline (MCT) (60 mg/kg) injection, and MCT rats treated with 1x10¹¹ vector genomes of AAV9 encoding JPH2 (AAV-JPH2) one week post-MCT injection. The cardiac-specific TNT4 promoter directed expression for both viruses. RV mitochondrial enrichments were processed for quantitative proteomics using TMT10-plex labeling. Lipidomic profiling of RV specimens was performed by Metabolon, Inc. Electron micrographs of RV mitochondria were collected at the University of Minnesota (UMN) Imaging Center. Blinded image analysis was performed by SZP, FK, or MB. Echocardiography and closed-chest pressure-volume loops defined RV function and pulmonary hypertension severity. Animal studies were approved by the UMN Institutional Animal Care and Use Committee. Statistical analyses were performed on GraphPad Prism 9.5 and MetaboAnalyst software (https://www.metaboanalyst.ca/). All data and materials have been made publicly available at the publicly available Figshare database and can be accessed at doi: 10.6084/m9.figshare.24877764 and doi:10.6084/m9.figshare.24993054.

Super resolution microscopy revealed discrete areas of JPH2 and MFN2 co-localization in RV cardiomyocytes (Figure A). Both JPH2 and MFN2 were detected in mitochondrial fractions, and co-immunoprecipitation experiments showed JPH2 and MFN2 interacted in RV extracts (Figure A). Pulldown studies demonstrated JPH2 directly bound MFN2, and the binding domain was mapped to the amino-terminal third of JPH2 (Figure A).

Figure: JPH2 binds MFN2 and regulates mitochondrial metabolism and RV function.

(A) Confocal micrographs demonstrated JPH2 (green) and MFN2 (purple) co-localized in isolated RV cardiomyocytes. Coomassie brilliant blue (CBB) of SDS-PAGE of cytoplasmic and mitochondrial extracts. Western blot analysis of cellular fractionations showed both JPH2 and MFN2 were enriched in the mitochondrial fraction. Co-immunoprecipitation studies demonstrated a biochemical interaction between JPH2 and MFN2. Immunoblots of pulldown experiments of recombinantly expressed and purified JPH2 and MFN2 constructs revealed the two proteins directly bound each other, and the interaction site was localized to the amino-terminal third of JPH2. (B) Representative confocal micrographs revealed disruption of mitochondrial network morphology (p-values determined by Mann-Whitney U-test) in JPH2 KO iPSC-CM. Western blot analysis showed JPH2 expression was abolished in JPH2 KO iPSC-CM (above) without significantly altering mitochondrial fission/fusion protein expression (p-values determined by unpaired t-test). CBB gel showed equivalent loading (below). (C) Seahorse tracings and quantification of reduced oxygen consumption rates (OCR) in JPH2 KO iPSC-CM (p-values determined by unpaired t-test). Confocal micrographs of lipid droplets (left) and the area of cells occupied by lipid droplets in control and JPH2 KO iPSC-CM. (D) Re-expression of full-length JPH2 (JPH2_FL) augmented mitochondrial oxygen consumption as compared to C-terminal JPH2 (JPH2_Ct) overexpression (p-values determined by one-way ANOVA with Tukey’s multiple comparisons test or by Kruskal-Wallis test with Dunn’s multiple comparisons test). (E) AAV-JPH2 restored JPH2 protein abundance and did not significantly impact MFN2 levels (p-values determined by one-way ANOVA with Tukey’s multiple comparisons test) and improved peri-t-tubular (red arrow) mitochondrial cristae structure (p-values determined by Kruskal-Wallis test with Dunn’s multiple comparisons test). (F) KEGG, Reactome, and WikiPathway analysis of differentially expressed proteins in mitochondrial enrichments. Red boxes highlight FAO related pathways. Hierarchical cluster analysis demonstrated AAV-JPH2 increased mitochondrial FAO protein abundance in RV mitochondrial extracts. (G) Hierarchical cluster and random forest analyses suggested AAV-JPH2 treatment restructured RV lipid homeostasis. (H) AAV-JPH2 treatment enhanced RV function as assessed by echocardiography and pressure-volume loop analysis without significantly altering RV afterload. p-values determined by one-way ANOVA with Tukey’s multiple comparisons test for TAPSE, RV free wall thickness change, RV-pulmonary artery coupling (E_es/E_a), and effective arterial elastance (E_a). p-values for RVSP determined by Brown-Forsythe ANOVA with Dunnett’s T3 multiple comparisons test.

Next, we evaluated how JPH2 ablation impacted mitochondrial morphology and function in iPSC-CM. JPH2 knockout mitochondrial networks were fragmented despite no differences in fission/fusion protein abundance (Figure B). JPH2 knockout suppressed mitochondrial oxidative capacity and heightened lipid droplet accumulation (Figure C). Re-expression of JPH2_FL but not overexpression of JPH2_Ct in JPH2 KO iPSC-CM increased oxygen consumption (Figure D).

Then, we determined how JPH2 overexpression via AAV9 impacted mitochondrial morphology, mitochondrial protein regulation, and lipid levels in the RV of MCT rats. AAV-JPH2 treatment restored JPH2 levels in the RV and rescued defects in peri-t-tubule mitochondrial cristae morphology (Figure E). KEGG, Reactome, and WikiPathway identified fatty acid oxidation (FAO) as an altered pathway in mitochondrial proteomics (Figure F). Hierarchical cluster analysis of fatty acid handling/metabolizing proteins showed AAV-GFP rats exhibited downregulation of multiple FAO enzymes, which AAV-JPH2 mitigated (Figure F). Consistent with a FAO enhancing effect, RV lipidomic profiling demonstrated AAV-JPH2 combatted dysregulation of several lipid species (Figure G).

Finally, we quantified the effects of AAV-JPH2 on RV function. Echocardiographic and closed-chest pressure-volume loop analyses revealed AAV-JPH2 augmented RV function without significantly altering pulmonary hypertension severity (Figure H).

In conclusion, we show JPH2 directly interacts with mitochondria via MFN2. Ablation of JPH2 in human iPSC-CM induces mitochondrial fragmentation, suppresses oxidative capacity, and impairs lipid handling. In MCT rats, AAV-JPH2 restores cristae morphology, increases expression of FAO proteins, and restructures the RV lipidomic signature. These morphological and molecular changes lead to improvements in RV contractility. Interestingly, a proteomics study found reduced RV JPH2 levels in pulmonary arterial hypertension patients with RVD⁵, which provides potential human-disease relevance to our findings. In summary, our data suggest JPH2 may be an important therapeutic target for RVD due to both its crucial mitochondrial and non-mitochondrial functions.

Non-standard Abbreviations and Acronyms

AAV9

Adeno-associated virus serotype 9

CRISPR

Clustered regularly interspaced short palindromic repeats

Cas9

Clustered regularly interspaced short palindromic repeats associated protein 9

GFP

Green fluorescent protein

iPSC

induced pluripotent stem cell

iPSC-CM

induced pluripotent stem cell derived cardiomyocyte

FAO

Fatty acid oxidation

JPH2

Junctophilin-2

KEGG

Kyoto Encyclopedia of Genes and Genomes

MCT

Monocrotaline

MFN2

Mitofusin-2

PCR

Polymerase chain reaction

RVD

Right ventricular dysfunction

RV

Right ventricle

TAPSE

Tricuspid annular plane systolic excursion

UMN

University of Minnesota