Research ArticlePHYSIOLOGY

FUNDC1 interacts with FBXL2 to govern mitochondrial integrity and cardiac function through an IP3R3-dependent manner in obesity

See allHide authors and affiliations

Science Advances  16 Sep 2020:
Vol. 6, no. 38, eabc8561
DOI: 10.1126/sciadv.abc8561
  • Fig. 1 HF intake–affected pathways and genes in hearts using RNA-seq and effect of obesity on mitophagy markers in the heart.

    (A to C) Rats were fed HF (60% fat) diet for 20 weeks, and cardiac tissues were collected and subjected to RNA-seq. (A) Signal transduction pathway of KEGG. (B) Cellular processes of KEGG. (C) Heatmap displaying relative expression of autophagy and Ca2+ signaling pathway–related genes in the heart (n = 3). (D) Mitophagy protein profiles (FUNDC1, Parkin, and BNIP3) in adult ob/ob mouse hearts. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. (E) Mitophagy protein profiles in adult db/db mouse hearts. (F) Mitophagy protein profiles in adult rat hearts following 20 weeks of HF intake. (G) Mitophagy protein profiles in lean and obese human heart samples. Mean ± SEM; n = 4 to 6 group, *P < 0.05 between indicated groups.

  • Fig. 2 HF diet intake–induced body weight gain, glucose intolerance, plasma profile, and cardiac remodeling in WT and FUNDC1−/− mice.

    (A) Body weight gain over 20-week feeding period. (B) Intraperitoneal glucose tolerance test (IPGTT). (C) Area underneath IPGTT curve (AUC). (D) Food intake (by weight). (E) Food intake (by calorie). (F) Plasma insulin. (G) HOMA-IR index. (H) Fasting blood glucose. (I) Serum triglycerides. (J) Gross images, lectin, or Masson’s trichrome staining in hearts from LF or HF diet–fed mice. (K) Pool data of cardiomyocyte cross-sectional area. (L) Quantitative analysis of interstitial fibrotic area (as a percentage of entire cardiac region). (M) ANP mRNA levels. (N) BNP mRNA levels. (O) GATA4 levels. Mean ± SEM; n = 7 to 10 mice per group, *P < 0.05 between indicated groups. Photo credits: Taken by Jun Ren, Zhongshan Hospital, Fudan University.

  • Fig. 3 HF diet intake–induced changes in echocardiographic indices in WT and FUNDC1−/− mice.

    (A) Representative echocardiographic images from all four experimental groups. (B) LV wall thickness. (C) Septal thickness. (D) LV ESD. (E) LV EDD. (F) Fractional shortening. (G) Ejection fraction. (H) LV mass. (I) Heart weight. (J) Heart rate. bpm, beats per minute. Mean ± SEM; n = 7 to 9 mice per group, *P < 0.05 between indicated groups.

  • Fig. 4 Effect of HF intake on myocardial ultrastructure, mitochondrial integrity, mitochondrial respiration, cell death, and mitochondrial O2 production in WT and FUNDC1−/− mice.

    (A) Mitochondria and sarcomere ultrastructure using transmission electron microscopy. (B) Mitochondrial number by image area. (C) Mitochondrial area (% total area). (D) Mitochondrial area (per mitochondrion). (E) UCP2 levels. (F to H) Mitochondrial respiration (complex II/III, complex I/III, and complex IV). (I) Representative MitoSOX Red fluorescence images. (J) Pooled data of mitochondrial O2 levels using MitoSOX Red. (K) Receptor-interacting protein (RIP)–associated cell death marker RIPK1. (L) RIP cell death marker RIPK3. (M) 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay in cardiomyocytes. (N) Plasma lactate dehydrogenase (LDH) levels. (O) Mitochondrial aconitase activity. Insets: Representative immunoblots depicting UCP2, RIPK1, and RIPK3 using respective antibodies. GAPDH was used as the loading control. Mean ± SEM, n = 5 to 6 images or visual fields per group, *P < 0.05 between indicated groups.

  • Fig. 5 Effect of HF intake on apoptosis, inflammation, mitochondrial apoptosis, and ER proteins including IP3Rs, SERCA2a, and RyR2 levels in hearts from WT and FUNDC1−/− mice.

    (A) Representative 4′,6-diamidino-2-phenylindole (DAPI) (nucleus) and TUNEL (apoptosis) staining in four experimental groups (arrowheads denote apoptosis). (B) Pooled data of TUNEL apoptosis. (C) Mitochondrial fraction of Bax. (D) Cytosolic fraction of Bax. (E) Immunofluorescence detection of IL-1β. (F) Mitochondrial Bcl2. (G) IP3R1. (H) IP3R2. (I) IP3R3. (J) SERCA2a [sarco(endo)plasmic reticulum Ca2+-ATPase 2a]. (K) Pan and phosphorylated RyR2. Insets: Representative immunoblots depicting levels of Bax, Bcl2, IP3Rs, SERCA2a, and pan and phosphorylated RyR2 using respective antibodies. GAPDH was used as the loading control. Mean ± SEM, n = 8 to 9 mice per group, *P < 0.05 between indicated groups.

  • Fig. 6 Effect of HF intake on autophagy and mitophagy protein markers in WT and FUNDC1−/− mice and evidence of FUNDC1-FBXL2 interaction.

    (A) LC3BI/II, (B) Beclin1, (C) Atg7, (D) Atg5, (E) p62, (F) Pink1, (G) Parkin, (H) FUNDC1, and (I) BNIP3. Insets: Representative immunoblots (IB) depicting levels of LC3BI/II, Beclin1, Atg7, Atg5, p62, Pink1, Parkin, FUNDC1, and BNIP3 using respective antibodies. GAPDH or COXIV (mitochondria) was used as the loading control. (J) Levels of FBXL2. (K) IP analysis for FBXL2-IP3R3 interaction. Bar graph depicts ratio of immunoprecipitated IP3R3 and FBXL2. (L) Co-IP mass spectrometry identified FBXL2 as an interacting protein of FUNDC1. (M) IP analysis of FBXL2 tagged with hemagglutinin (HA) (FBXL2-HA) and FUNDC1 tagged with Flag (FUNDC1-Flag) in H9c2 cells. IgG, immunoglobulin G. (N) IP analysis of FBXL2 domain mutations (Delta-F-box or Delta LRR1, HA-tagged) using FUNDC1-Flag in H9c2 cells. Mean ± SEM, n = 9 mice per group, *P < 0.05 between indicated groups.

  • Fig. 7 FBXL2 binding domain and bioinformatics prediction for FUNDC1-FBXL2 binding modality.

    (A) Structure-based protein interaction interface analysis between FUNDC1 and FBXL2. Cartoon represents the predicted FBXL2-FUNDC1 complex structure, where the interaction hotspot residues are labeled. (B) Schematic diagram of domain deletion of FBXL2-HA for co-IP experiments, and the predicted interaction sites on FBXL2 with FUNDC1. One interaction hotspot (ARG-75) is settled on the LRR1 domain of FBXL2, while the others (CYS-33, GLN-37, LYS-40, ASN-43, and LEU-47) are all from the F-box domain. Boxes indicate exons; lines indicate deleted domains. (C) IP analysis of FBXL2 and FUNDC1 in heart tissues from LF- and HF-fed WT mouse groups. (D) Representative images displaying mitochondrial Ca2+ fluorescence in neonatal cardiomyocytes from WT and FUNDC1−/− mice challenged with palmitic acid (PA, 0.5 mM for 8 hours). (E) Pooled data of mitochondrial Ca2+ levels. Mean ± SEM, n = 7 to 10 cell batches per group, *P < 0.05 versus between indicated groups.

  • Fig. 8 Effect of FUNDC1 transfection and/or FBXL2 stimulation on palmitic acid–induced mitochondrial structure in neonatal mouse cardiomyocytes in the presence or absence of FBXL2 localization inhibitor GGTi-2418.

    Neonatal cardiomyocytes were transfected with FUNDC1 overnight before incubation with palmitic acid (0.5 mM) for another 8 hours in the presence or absence of FBXL2 activator BC-1258 (10 μg/ml), the IP3R3 inhibitor 2-APB (30 μM), or the FBXL2 colocalization inhibitor GGTi-2418 (15 μM) before assessment of mitochondrial structure using the confocal microscopy. (A) Representative mitochondrial immunofluorescence. (B) Quantification of mitochondrial length. Mean ± SEM, n = 8 cells per group. *P < 0.05 between indicated groups.

  • Fig. 9 Effect of FUNDC1 or FBXL2 transfection on palmitic acid–induced changes in mitochondrial apoptosis, ΔΨm, and IP3Rs in neonatal cardiomyocytes.

    Neonatal cardiomyocytes from WT or FUNDC1−/− mice were transfected with FUNDC1 or FBXL2 overnight before the incubation with palmitic acid (0.5 mM) for an additional 8 hours. (A) Representative fluorescence probing of ΔΨm. (B) Mitochondrial Bax levels. (C) Cytosolic Bax levels. (D) Bcl2 levels. (E) IP3R1 levels. (F) IP3R2 levels. (G) IP3R3 levels. Insets: Representative immunoblots depicting levels of Bax, Bcl2, IP3R1, IP3R2, and IP3R3 using specific antibodies. GAPDH or COXIV (for mitochondria) was used as the loading control. Mean ± SEM, n = 8 cells per group, *P < 0.05 between indicated groups.

  • Fig. 10 Determination of stabilization of FBXL2 and IP3R3 using pulse-chase assay in neonatal cardiomyocytes from WT and FUNDC1−/− mice transfected with FUNDC1 or FBXL2 viral vector.

    (A) Western blots analysis of palmitic acid (0.5 mM)–induced time-dependent FBXL2 degradation (0 to 24 hours) in neonatal cardiomyocytes from WT or FUNDC1−/− mice with or without FUNDC1 transfection. A cohort of cardiomyocytes was incubated with the proteasomal inhibitor MG132 throughout the course of study. (B) Western blots analysis of palmitic acid (0.5 mM)–induced time-dependent IP3R3 degradation (0 to 24 hours) in neonatal cardiomyocytes from WT or FUNDC1−/− mice with or without FUNDC1 or FBXL2 transfection. A cohort of cardiomyocytes was incubated with MG132 throughout the study. (C) Pooled data of FBXL2 degradation in the absence of MG132. (D) Pooled data of IP3R3 degradation in the absence of MG132. (E) Diagram depicting proposed mechanism for the role of FUNDC1 in HF diet–induced changes in cardiac remodeling and contractile defects. HF intake suppressed FUNDC1–mediated mitophagy, leading to disturbed interaction between FUNDC1 and FBXL2, thus prompting destabilization of FBXL2. FBXL2 loss facilitates IP3R3-mediated Ca2+ mobilization into mitochondria (mitochondrial Ca2+ overload) and RIP-mediated cell death, resulting in mitochondrial injury and cardiac geometric and functional anomalies.

Supplementary Materials

  • Supplementary Materials

    FUNDC1 interacts with FBXL2 to govern mitochondrial integrity and cardiac function through an IP3R3-dependent manner in obesity

    Jun Ren, Mingming Sun, Hao Zhou, Amir Ajoolabady, Yuan Zhou, Jun Tao, James R. Sowers, Yingmei Zhang

    Download Supplement

    This PDF file includes:

    • Table S1
    • Figs. S1 to S3

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article