F‐Box and Leucine‐Rich Repeat Protein 4 (FBXL4) Maintains Sarcomere Integrity and Cardiac Function by Enhancing K48‐Linked Ubiquitinated Degradation of Profilin‐1 (PFN1)

Figures

• 

  FIGURE 1. FBXL4 expression is down‐regulated in cardiac tissues of heart failure. (A) Uniform Manifold Approximation and Projection (UMAP) plot showing single cells isolated from the hearts of control and HF mice (from the GSE271946 database). (B) Heatmap showing the differentially regulated genes between control and HF. (C) The transcriptional level of FBXL4 in human control (n = 8) and HF (n = 60) from the GSE271946 database. (D) The transcriptional level of FBXL4 in mouse control (n = 4) and HF (n = 7) from the GSE36074 database. (E) The transcriptional level of FBXL4 in human control (n = 136) and HF (n = 176) from the GSE57345 database. (F) Immunofluorescence staining of adult cardiomyocytes from sham‐operated and TAC treated‐mice demonstrated cytoplasmic localization of FBXL4 with decreased expression post‐TAC surgery, stained with DAPI (blue) α‐actinin (green) and FBXL4 (red), n = 6, scale bar = 100 µm. Zoomed images show a higher magnification. Scale bar = 20 µm. (G, H) Western blot and qRT‐PCR analyses of FBXL4 expression in left ventricular tissues of human non‐HF subjects (n = 3) and patients with HF (HF, n = 4). (I) Protein levels of ANP, β‐MHC, and FBXL4 in left ventricular…

• 

  FIGURE 2. Inducible cardiomyocyte‐specific deletion of FBXL4 in an adult heart develops HF. (A) Schematic of tamoxifen injection and the echocardiography experiments in adult mouse hearts. Echo, echocardiography. (B, C) Representative echocardiographic images of M‐mode and mitral valve pulse‐wave Doppler in CTR vs. iCKO mice at baseline, day 28, and day 56. (D) Echo parameters of CTR vs. iCKO mice at baseline (BL), day 5, day 7, day 14, day 28, and day 56 were calculated (n = 6–8/group). (E) Representative 4‐chamber H&E images on day 56. Scale bar = 2 mm. (F) Heart weight normalized body weight (HW/BW) and heart weight normalized to tibia length (HW/TL) of CTR vs. iCKO on day 56 (n = 8/group). (G) Survival curve of Myh6‐Cre vs. FBXL4‐iCKO mice (n = 19 for Myh6‐Cre, n = 21 for iCKO). H, I) Representative WGA staining images and quantitation of cardiomyocyte size on days 28 and 56 (n = 300 cells/group). Scale bar = 100 µm. (J, K) Representative trichrome staining images and quantitation of cardiac interstitial fibrosis on days 28 and 56 (n = 5/group). Scale bar = 100 µm. (L, M) Representative trichrome staining images and quantitation of cardiac perivascular fibrosis on days 28 and 56 (n =…

• 

  FIGURE 3. Cardiac‐specific FBXL4 overexpression restore FBXL4‐iCKO induced HF. (A) Schematic diagram of tail intravenous AAV9‐cTnT‐FBXL4/Vector and tamoxifen intraperitoneal injection. (B–D) Representative M‐mode and doppler echocardiography of the left ventricle and statistical data of EF%, FS%, LVIDd, LVIDs, and E/A ratio. n = 8–11. (E, F) Gross morphology and HE staining of the hearts (scale bar = 2 mm). G, H) Heart weight (HW)/body weight (BW) and HW/tibia length (TL) ratios in the different groups, n = 8–12/group. (I, J) Representative images and statistics of WGA staining (2–3 sections/mouse, Scale bar = 100 µm, n = 4). (K, L) Masson's trichrome staining (3 sections/mouse, Scale bar = 100 µm, n = 4 mice). (M) Representative IF images and statistics of Z‐disc organization and F‐actin architecture in AAV9‐Vector or AAV9‐FBXL4 treated with FBXL4‐iCKO mice, stained with DAPI (blue), ACTN2 (red), and α‐actinin (green), 3 sections/mouse, n = 4 mice. Scale bar = 100 µm. Scale bars: upper, 100 µm; lower, 20 µm. (N) Representative TEM analysis of sarcomere structure in AAV9‐Vector or AAV9‐FBXL4 treated with FBXL4‐iCKO mice. The red and yellow arrows indicate Z‐disc and M‐band, respectively.…

• 

  FIGURE 4. Cardiomyocytes‐specific FBXL4 overexpression alleviates cardiac hypertrophy and dysfunction. (A–C) Representative M‐mode echocardiography and doppler echocardiography of the left ventricle and statistical data of EF%, FS%, LVIDd, LVIDs and E/A ratio n = 8–10/group. (D, E) Gross morphology and HE staining of the hearts (scale bar = 2 mm). (F, G) Heart weight (HW)/body weight (BW) and HW/tibia length (TL) ratios in the different groups, n = 8–12/group. (H, I) Representative images and statistics of WGA staining (2‐3 sections/mouse, Scale bar = 100 µm, n = 4). (J, K) Masson's trichrome staining (3 sections/mouse, Scale bar = 100 µm, n = 4 mice). L) Representative immunofluorescence images and quantitation of Z‐disc organization and F‐actin architecture in FBXL4‐cKI mice following TAC‐induced hypertrophy, stained with DAPI (blue), ACTN2 (red), and α‐actinin (green), 3 sections/mouse, n = 4 mice. Scale bars: upper, 100 µm; lower, 20 µm. M) Representative TEM analysis of sarcomere structure in FBXL4‐cKI mice following TAC‐induced hypertrophy. The red and yellow arrows indicate Z‐disc and M‐band, respectively. Scale bars, 100 µm and 20 µm. n = 3/group. (N) Representative western blot…

• 

  FIGURE 5. FBXL4 physically interacts with PFN1 to promote its degradation. (A) Bar plot showing Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of FBXL4‐binding genes. (B) The volcano plot highlights the top 13 genes exhibiting the most significant correlations with FBXL4. (C) The Venn diagram illustrates the overlap between 837 proteins identified by mass spectrometry and candidates prioritized by Spearman's correlation analysis, defining the putative substrate of FBXL4. (D) Scatter plots illustrating the correlation between FBXL4 and PFN1 in the single‐cell dataset (GSE120064): A two‐tailed p‐value < 0.05 denotes statistical significance, and an absolute Spearman correlation coefficient |R| ≥ 0.30 indicates a meaningful association. (E) IF staining showing the cellular distribution of FBXL4 and PFN1 in CMs. FBXL4 stained in green, PFN1 stained in red, and the nucleus stained in blue with DAPI. Scale bars: upper, 100 µm; lower, 10 µm. (F, G) Lysates from HEK293T cells transfected with Flag‐FBXL4 and HA‐PFN1 plasmids were immunoprecipitated with anti‐Flag/anti‐HA followed by WB with anti‐Flag (FBXL4) and anti‐HA (PFN1) (n = 3). (H, I) Endogenous protein interactions were examined…

• 

  FIGURE 6. K70 of PFN1 is important for its K48‐linked ubiquitination mediated by FBXL4. (A) HEK293T cells overexpressing FBXL4 were transfected with the indicated plasmid combinations to measure ubiquitination of PFN1. (B) HEK293T cells overexpressing FBXL4 were transfected with the indicated plasmid combinations to measure K6/K11/K27/K29/K33/K48/K63‐linked ubiquitination of PFN1. (C–E) HEK293T cells overexpressing FBXL4 were transfected with the indicated plasmid combinations to measure K29, K48, and K63‐linked ubiquitination of PFN1. (F) The ubiquitination sites of PFN1 were predicted by BDM‐PUB and UBPRED. (G) HEK293T cells were transfected with the indicated plasmid combinations to measure K48‐linked ubiquitination of PFN1. (H) WebLogo output for amino acid sequences of PFN1 from Mouse, Homo, Rat, Cuniculus, Chimpanzee, and Pig. (I) HEK293T cells overexpressing Flag‐FBXL4 and Myc‐K48 were transfected with the indicated plasmid combinations to measure the ubiquitination of HA‐PFN1/PFN1‐K70R. (J, K) Representative immunoblot and quantification showing PFN1 protein levels in NMCMs following transduction with K70R or WT PFN1, n = 5. (L) IF staining showing the cellular distribution of PFN1…

• 

  FIGURE 7. Knockdown of PFN1 partially rescues the HF in FBXL4‐iCKO Mice. (A) Schematic diagram of tail intravenous AAV9‐cTnT‐SiPFN1/Vector and tamoxifen intraperitoneal injection. (B–D) Representative M‐mode echocardiography of the left ventricle and statistical data of EF%, FS%, LVIDd, LVIDs, and E/A ratio. n = 8–10/group. (E, F) Gross morphology and HE staining of the hearts (scale bar = 2 mm). (G, H) Heart weight (HW)/body weight (BW) and HW/tibia length (TL) ratios in the different groups, n = 10/group. (I, J) Representative images and statistics of WGA staining (2–3 sections/mouse, Scale bar = 100 µm, n = 4). (K, L) Masson's trichrome staining (3 sections/mouse, Scale bar = 100 µm, n = 4 mice). (M) Representative IF images of Z‐disc organization and F‐actin architecture in AAV9‐cTnT‐Vector or AAV9‐cTnT‐SiPFN1 treated with Myh6‐Cre and FBXL4‐iCKO mice, stained with DAPI (blue), ACTN2 (red), and α‐actinin (green). Scale bars: upper, 100 µm; lower, 20 µm, 3 sections/mouse, n = 4 mice. (N) Representative TEM image of sarcomere structure in AAV9‐cTnT‐Vector or AAV9‐cTnT‐SiPFN1 treated with Myh6‐Cre and FBXL4‐iCKO mice. The red and yellow arrows indicate Z‐disc and M‐band, respectively.…

• 

  FIGURE 8. FBXL4 deficiency induces hypertrophic phenotypes in human induced pluripotent stem cell‐derived cardiomyocytes (iPSC‐CMs). (A) Verification of the efficacy of human FBXL4 siRNA (siFBXL4) in silencing FBXL4 expression at the protein levels (n = 4) in hiPSC‐CMs. NC is a negative control siRNA. (B, C) Representative images and quantification of myofibrillar disarray in hiPSC‐CMs by staining with ACTN2 (red), α‐actinin (green), and DAPI (blue). Scale bar, 50 µm. Zoomed images show a higher magnification. Scale bar, 1 µm. At least 100 cells were counted in each group from 6 independent experiments. (D, E) Representative immunoblotting analysis and quantification of β‐MHC, PFN1, pERK1/2, and ERK1/2 protein levels after transfection of siFBXL4 in hiPSC‐CMs. (n = 4). (F) Verification of the efficacy of the human FBXL4 plasmid in FBXL4 overexpression at the protein levels (n = 4). GFP is a negative control FBXL4. (G, H) Representative images and quantification of myofibrillar disarray in hiPSC‐CMs by staining with ACTN2 (red), α‐actinin (green), and DAPI (blue). Scale bar, 50 µm. Zoomed images show a higher magnification. Scale bar, 1 µm. At least 100 cells were counted in each group from…