PKN2 deficiency leads both to prenatal ‘congenital’ cardiomyopathy and defective angiotensin II stress responses

Figure 1.. PKN2 knockout in cardiomyocytes causes defective embryonic heart development leading to failure prior to adulthood. (a) H&E and Gomori's Trichrome stained histological cardiac short-axis sections of the longest surviving XMLC2Cre+/− Pkn2fl/fl genotype mouse (left; male), and two littermates (middle & right; male XMLC2Cre+/− Pkn2fl/+ & female XMLC2Cre+/+ Pkn2fl/+, respectively) all culled aged 5 weeks. Scale bars are 1 mm. In lower boxed regions, various sections are also shown at 20× zoom relative to whole-heart cross-sections. (b) H&E sections of lungs of the same XMLC2Cre+/− Pkn2fl/fl genotype mouse (left), and its XMLC2Cre+/− Pkn2fl/+ littermate (right; both male). Scale bars are 200 µm. (c) Images of High-Resolution Episcopic Microscopy (HREM) reconstructions of E14.5 embryo hearts for the genotypes indicated, showing surface and four chamber views. Boxed regions of sections are shown at 4× zoom; a diverticulum is indicated by the yellow arrowhead.

Figure 2.. SM22aCre+/− Pkn2fl/+ mice suffer cardiac failure on aging. (a and b) Short-axis mid-ventricle sections of hearts of 39 week male littermates of the genotypes indicated, stained with H&E (a), or Gomori's Trichrome (b). (c) H&E stained sections of lungs, littermates as in (a and b). (d) Time course of symptom onset for mice of genotype SM22αCre+/− Pkn2fl/fl(n = 23). (e and f) Montage of time-series through the cardiac cycle recorded by cine-cardiac-MRI through the short-axis plane at mid-ventricle from SM22αCre+/+ Pkn2fl/fl (e) and SM22αCre+/− Pkn2fl/fl (f) 32 week old female littermates. (g) micro-CT of the thorax of a 24 week SM22αCre+/− Pkn2fl/fl female shown as a projected image and slice images in three planes. Scale-bars in histopathology images are 1 mm in whole heart sections (panels a and b) and 200 µm in zoomed sections (panels b and c). Scale-bars in MRI montages are 2 mm (panels e and f).

Figure 3.. The heart-specific defects in SM22αCre+/−Pkn2fl/fl mice. (a) H&E stained sections of (E18.5) lungs of littermates with the indicated genotypes. Lower boxed panels are 5× zoomed; scale bars indicate 200 µm in upper panels and 40 µm in lower zoomed panels. (b) Reconstructions of E14.5 hearts imaged by HREM are shown in surface (top images) and slice (lower images) views. The genotypes are as indicated. (c) H&E stained E14.5 heart sections featuring an outer ventricular surface nodule, identified as a diverticulum. Scale bars are 100 µm. (d) Reconstructions of E18.5 hearts imaged by HREM shown as per panel (b). Genotypes are as indicated.

Figure 4.. PKN2 is associated with human heart failure and required for cardiac adaptation to hypertension in mice. (a), Expression of PKN1, PKN2 and PKN3 in human hearts. Data were mined from an RNASeq database of patients with dilated cardiomyopathy (DCM; n = 97) and normal controls (Con; n = 108). Data for individual samples are shown with false discovery rates (FDR). (b and c) Immunoblotting of phospho-PKN1/2, total PKN1, total PKN2 and GAPDH in hearts from wild-type (WT) or Pkn2Het mice treated for 7 days with vehicle (Veh, V) or 0.8 mg/kg/d angiotensin II (AngII, A). Immunoblots are shown on the left with densitometric analysis on the right (normalised to the mean of vehicle-treated controls). Individual data points are provided with means ± SEM. Analysis used two-way ANOVA with Holm–Sidak's post-test. (d–h), Echocardiography of hearts from WT and Pkn2Het mice at baseline (d and e) or treated with vehicle or AngII for 7 days (f–h). Representative short-axis M-mode images used for assessment of cardiac dimensions are shown at baseline (d) and after treatment (f) (the same animals are shown). (g), Representative long-axis B-mode images used for speckle-tracking and strain analysis to…

Figure 5.. Pkn2 haploinsufficiency reduces cardiac adaptation to hypertension, affecting cardiomyocyte size, perivascular fibrosis and gene expression. WT or Pkn2Het mice were treated with vehicle or AngII (0.8 mg/kg/d) for 7 days. (a–c), Sections of WT and Pkn2Het mouse hearts were stained with haematoxylin and eosin (H&E), picrosirius red (PSR), or Masson's Trichrome. Representative images are shown with quantification provided below. Individual data points are provided with means ± SEM. Analysis used two-way ANOVA with Holm–Sidak's post-test. (d), RNASeq analysis of hearts from WT and Pkn2Het mice. (i) Summary of differentially-expressed genes (P < 0.01 resulting from AngII treatment. (ii)–(viii) Clusters of DEGs according to function. Results are the mean normalised count values with 95% CI for the n values indicated. Analysis used one-way ANOVA with Holm–Sidak's post-test. (e) Comparison of Tagln (SM22α) mRNA expression assessed by RNASeq and qPCR. Individual data points are shown with means ± SEM. Analysis used two-way ANOVA with Holm–Sidak's post-test.

Figure 6.. Effects of aging on aortic and pulmonary blood flow in WT and Pkn2Het mice. Echocardiograms were taken for Pkn2Het and WT littermates with an average age of 12 or 42 weeks. (a), End diastolic left ventricular mass (EDLVM) was measured from long-axis B-mode images using speckle-tracking strain analysis. (b), The width of the aorta was measured from B-mode images from the widest diameter taken at cardiac systole (upper panel) and assessing the ratio of this to the narrowest diameter taken after the aortic valve closed (L/S). (c–f), Pulsed wave Döppler was used to assess blood flow as it leaves the heart into the pulmonary artery (c and d) and the aorta (e and f). Additional data are provided in Supplementary Table S1. Representative images are shown (c and e) and with the analysis (d and f). Individual data points are shown with means ± SEM. Analysis used two-way ANOVA with Holm–Sidak's post-test.