Sex-specific regulation of the cardiac transcriptome by the protein phosphatase 2A regulatory subunit B55α

Figures

• 

  Fig. 1. Homozygous knockout of B55α causes embryonic lethality without overt cardiac defects. a Schematic of the Ppp2r2a ‘knockout-first’ allele used to produce B55α heterozygous (Het) and homozygous (Hom) knockout mice (Model 1: Global B55a KO). Floxed Ppp2r2a mice were produced by removing the lacZ/neo cassette with Flp recombinase. Cardiomyocyte-specific B55α knockout mice (Model 2) were produced by removing exon 4 with Cre recombinase under control of the αMHC promoter. b Whole mount images of wildtype (Wt) and Hom littermates at 12.5 dpc, imaged using an AxioCamMRc5 camera with a 1x objective. Scale bar 2 mm. c Images of Wt and Hom embryos at 12.5 dpc, stained with haematoxylin & eosin and imaged using a Pannoramic Scan II slide scanner fitted with a 20x/NA 0.8 objective and a CCD camera. Left panel shows head (scale bar 1 mm), middle panel shows hindlimb (scale bar 0.5 mm), right panel shows skin layer (scale bar 0.05 mm). d Images of Wt and Hom hearts at 12.5 dpc, stained with haematoxylin & eosin and imaged using a Pannoramic Scan II slide scanner fitted with a 20x/NA 0.8 objective and a CCD camera. Left panel shows heart cross-section (scale bar 0.5 mm; asterisks indicate AV…

• 

  Fig. 2. Heterozygous expression of B55α leads to left ventricular wall thinning in male mice at 12 months of age. a Representative Western blot and b) quantitation of B55α protein expression. Ventricular tissue from 12-month old male and female Ppp2r2a wildtype (Wt) and heterozygous (Het) mice at 12 months of age. α-tubulin was used as a loading control. Data are expressed as fold difference vs Wt. Bars show median +/− IQR. Mann-Whitney U tests. n = 6–7 per group. c, d Interventricular septum (IVS) thickness, left ventricular posterior wall (LVPW) thickness, LV internal dimension (LVID) and fractional shortening, assessed by M-mode echocardiography in anaesthetised mice at 12 months of age. Bars show median +/− IQR. Mann-Whitney U tests. n = 6–8/group. e, f Heart weight (HW) and atria weight (AW) normalised to tibia length (TL) of mice at 12 months of age. n = 6-8/group. Bars show median +/− IQR. Mann-Whitney U tests.

• 

  Fig. 3. Effects of cardiomyocyte-specific deletion of Ppp2r2a on PP2A subunit expression and HDAC5 phosphorylation. a, c Representative Western blots and b, d quantitation of B55α, PP2A catalytic (PP2A-C) and scaffolding (PP2A-A) subunits in ventricular tissue from 10–12 week-old cardiomyocyte-specific B55α knockout (cKO) mice and floxed controls (FC). Vinculin was used as a loading control. Data are expressed as fold difference vs FC. Bars show median +/− IQR. Mann-Whitney U tests. n = 6 per group for all analyses except PP2A-C (n = 5 for female FC). Vertical lines indicate where images of Western blots have been spliced so the relevant genotypes could be displayed side by side. e, f Quantitative PCR analysis of PP2A B55 subunits Ppp2r2a (encoding B55α), Ppp2r2c (encoding B55γ) and Ppp2r2d (encoding B55δ). Data are expressed as fold difference relative to the FC group. Bars show mean +/− SEM. Unpaired t-tests. n = 4–6/group. g, h Representative Western blots and quantitation of HDAC5 phosphorylation at Ser259 relative to total HDAC5. Data are expressed as fold difference vs FC. Bars show median +/− IQR. Mann-Whitney U tests. n = 4–7 per group.

• 

  Fig. 4. Left ventricular structure and function was normal in mice with cardiomyocyte-specific knockout of B55α at 10–12 weeks of age. a, b Interventricular septum (IVS) thickness, left ventricular posterior wall (LVPW) thickness, LV internal dimension (LVID) and fractional shortening in 10–12 week old cardiomyocyte-specific B55α knockout (cKO) mice and floxed controls (FC), quantified from M-mode echocardiograms. Bars show median +/- IQR. Mann-Whitney U tests. n = 13/group. c, d Heart weight (HW) and atria weight (AW) normalised to tibia length (TL) in FC and cKO mice at 10–12 weeks of age. Bars show median +/- IQR. Mann-Whitney U tests. n = 12/group.

• 

  Fig. 5. Loss of cardiomyocyte B55α alters the expression of myosin heavy chain isoforms and pro-fibrotic genes. a, b Quantitative PCR analysis of genes that are typically dysregulated in settings of pathological cardiac hypertrophy and fibrosis. Ventricular tissue from male and female cardiomyocyte-specific B55α knockout (cKO) and floxed control (FC) mice. Data are expressed as fold difference relative to the FC group. Bars show mean +/− SEM. Unpaired t-tests. n = 6–8 per group for all analyses except Ctgf (n = 5 for female cKO).

• 

  Fig. 6. Loss of cardiomyocyte B55α had a greater impact on cardiac gene expression in males compared with females. a, b Volcano plots showing genes that were differentially expressed between cKO and FC hearts. RNA-seq analysis of ventricular tissue from male and female cardiomyocyte-specific B55α knockout (cKO) and floxed control (FC) mice. FDR < 0.05, n = 6-7/group. c, d Pathway enrichment analysis showing Gene Ontology terms that were over-represented in the set of genes that were significantly upregulated or downregulated in male cKO hearts vs FC. n = 6–7/group. e Representative LV sections stained with picrosirius red and quantitation of collagen (expressed as a percentage of left ventricular area). Scale bar: 50 μm. Two-way ANOVA followed by Sidak’s post hoc tests. n = 6-7/group. f–i Relative mRNA abundance (in counts per million, cpm) of collagen isoforms in FC (circles) and cKO (triangles) hearts. Two-way ANOVA followed by Tukey’s posthoc tests. n = 6–7/group.