Role of ERK1/2 signaling in congenital valve malformations in Noonan syndrome

Abstract

Noonan syndrome (NS) is the most common nonchromosomal genetic disorder associated with cardiovascular malformations. The most prominent cardiac defects in NS are pulmonary valve stenosis and hypertrophic cardiomyopathy. Gain-of-function mutations in the protein tyrosine phosphatase Shp2 have been identified in 50% of NS families. We created a NS mouse model with selective overexpression of mutant Shp2 (Q79R-Shp2) in the developing endocardial cushions. In our model, Cre recombinase driven by the Tie2 promoter irreversibly activates transgenic Q79R-Shp2 expression in the endothelial-derived cell lineage. Q79R-Shp2 expression resulted in embryonic lethality by embryonic day 14.5. Importantly, mutant embryos showed significantly enlarged endocardial cushions in the atrioventricular canal and in the outflow tract. In contrast, overexpression of wild-type Shp2 protein at comparable levels did not enhance endocardial cushion growth or alter the morphology of the mature adult valves. Expression of Q79R-Shp2 was accompanied by increased ERK1/2 activation in a subset of cells within the cushion mesenchyme, suggesting that hyperactivation of this signaling pathway may play a pathogenic role. To test this hypothesis in vivo, Q79R-Shp2-expressing mice were crossed with mice carrying either a homozygous ERK1 or a heterozygous ERK2 deletion. Deletion of ERK1 completely rescued the endocardial cushion phenotype, whereas ERK2 protein reduction did not affect endocardial cushion size. Constitutive hyperactivation of ERK1/2 signaling alone with a transgenic approach resulted in a phenocopy of the valvular phenotype. The data demonstrate both necessity and sufficiency of increased ERK activation downstream of Shp2 in mediating abnormal valve development in a NS mouse model.

Fig. 1.

Endothelial lineage-specific expression of Q79R-Shp2 in the embryonic mouse heart. (A–F) Immunostaining for total Shp2 in the septal leaflets of the atrioventricular valves (A, B, E, and F) and in the pulmonary valve leaflets (C and D). (G) Quantification and comparison of Shp2 staining in 1 Q79R-Shp2 line (red) and 4 independent WT-Shp2 lines (gray) vs. endogenous levels (control, black) (n = 2–3 hearts with 2–4 sections each per group); *, P < 0.05 vs. nontransgenic control. (H) Percentage of cells positive for GFP in the endocardial cushion mesenchyme of the AVC (green) and OFT (blue); *, P < 0.05 vs. AVC group. (I) Endocardial cushion volumes in control (white), Q79R-DTG (red), control (yellow), and WT-DTG (green) hearts at E13.5 (n = 3 per group; *, P < 0.05 vs. control). (J–Q) Representative histological sections through AVC and right ventricular OFT leaflets in control, Q79R-, and WT-DTG hearts.

Fig. 2.

Ultrastructure, differentiation markers, and mitotic indices in endocardial cushion cells from control and Q79R-DTG embryos at E13.5. (A and B) Endocardial surface of atrioventricular cushion. (C and D) Mesenchymal cells within AVC cushions. (E and F) Immunostaining for NFATc1 (green) together with TO-PRO nuclear stain (blue). (G and H) Immunostaining for cartilage link protein in septal leaflet of AVC cushions. (I) Percentage of BrdU-positive nuclei in the endothelial cell (EC), mesenchymal cell (MC), and cardiomyocyte (CM) population at E13.5 (control, white; Q79R-DTG, black; n = 3 hearts per group; *, P < 0.05 vs. control).

Fig. 3.

ERK hyperactivation in endocardial cushions. (A–F) Immunostaining for Shp2 (blue) and phospho-ERK1/2 (green) in control and Q79R-DTG E13.5 AVC cushions. (G–J) Representative histological sections through AVC and right ventricular OFT leaflets in control and caMEK1-DTG hearts. (K) Endocardial cushion volumes in control (white) and caMEK1-DTG (blue) hearts at E13.5 (n = 3–4 per group; *, P < 0.05 vs. control). (L and M) Immunostaining for MEK1/2 (green) with TO-PRO nuclear stain (blue) of septal atrioventricular leaflet at E13.5 in control and caMEK1-DTG.

Fig. 4.

Deletion of ERK1 but not reduction of ERK2 protein rescues the valve phenotype in Q79R-DTGs. (A–H) Representative histological sections through AVC and right ventricular OFT leaflets in ERK1^−/−, Q79R-DTG;ERK1^−/−, ERK2^+/−, and Q79R-DTG;ERK2^+/− hearts. (I) Endocardial cushion volumes in ERK1^−/− (white), Q79R-DTG;ERK1^−/− (green), ERK1^−/+ (gray), and Q79R-DTG;ERK1^−/+ (blue) hearts at E13.5 (n = 3–5 per group; *, P < 0.05 vs. respective control). (J) Endocardial cushion volumes in ERK2^+/− (white) and Q79R-DTG;ERK2^+/− (orange) hearts at E13.5 (n = 3 per group; *, P < 0.05 vs. ERK2^+/− control). (K and L) Percentage of BrdU-positive nuclei in the endothelial cell (EC), mesenchymal cell (MC), and cardiomyocyte (CM) population at E13.5 (ERK1^−/− and ERK2^+/− controls, white; Q79R-DTG;ERK1^−/−, green; Q79R-DTG;ERK2^+/−, orange; n = 3–4 hearts per group; *, P < 0.05 vs. respective control). (M) Percentage of TUNEL-positive nuclei in AVC cushions at E13.5 (control, white; Q79R-DTG, black; ERK1^−/−, light gray; Q79R-DTG;ERK1^−/−, green; ERK2^+/−, dark gray; Q79R-DTG;ERK2^+/−, orange; n = 3–5 hearts per group; *, P < 0.05 vs. respective control).