Functional characterization of SHOX2 variants in vivo and in vitro. (A) Cardiac-specific overexpression of SHOX2 mutants compared with SHOX2 WT (wild-type) leads to pericardial edema (arrow) for p.G77D and p.H283Q but not for p.P33R and p.A293= in zebrafish 72hpf. (B) The heart rate of zebrafish embryos was significantly reduced upon overexpression of p.G77D and p.H283Q but normal for p.P33R and p.A293= 72hpf compared with empty vector-injected embryos. SHOX2 WT overexpression showed no effect. Data are expressed as mean ± SD of three to six independent experiments. All P-values were determined by one-way ANOVA with Tukey’s multiple comparison test (*P < 0.05; ***P < 0.001; ns: not significant). (C) Luciferase activity of BMP4 reporter construct co-expressed with SHOX2 wild-type (WT) or SHOX2 mutants (p.H283Q, p.G77D, p.P33R) in HEK293T cells 24 h after transfection. All values are normalized to the empty pGL3 basic vector co-transfected with the respective expression constructs. Data are expressed as mean ± SEM of four independent experiments performed in triplicate. All P-values were determined by one-way ANOVA followed by uncorrected Fisher’s least significant difference test (*P < 0.05). (D) nCounter analysis revealed downregulation of Bmp4 expression in zebrafish hearts (72 hpf) overexpressing the mutant p.P33R compared with wild-type SHOX2 (n= 40 hearts per condition from two independent experiments). Statistical differences were determined by a ratio paired t-test (**P < 0.01).

Fluorescence imaging of cardiac-specific overexpression of SHOX2 mutants in zebrafish. Bright field (upper image) and corresponding fluorescence imaging (lower image) of cardiac-specific overexpression of SHOX2 mutants compared to SHOX2 WT (wildtype) and empty vector control show pericardial edema (red arrow) for p.G77D and p.H283Q but not for p.P33R and p.A293= 72hpf.