Myocardial Notch-Rbpj deletion does not affect heart development or function

During vertebrate cardiac development NOTCH signaling activity in the endocardium is essential for the crosstalk between endocardium and myocardium that initiates ventricular trabeculation and valve primordium formation. This crosstalk leads later to the maturation and compaction of the ventricular chambers and the morphogenesis of the cardiac valves, and its alteration may lead to disease. Although endocardial NOTCH signaling has been shown to be crucial for heart development, its physiological role in the myocardium has not been clearly established. Here we have used a genetic strategy to evaluate the role of NOTCH in myocardial development. We have inactivated the unique and ubiquitous NOTCH effector RBPJ in the early cardiomyocytes progenitors, and examined its consequences in cardiac development and function. Our results demonstrate that mice with cTnT-Cre-mediated myocardial-specific deletion of Rbpj develop to term, with homozygous mutant animals showing normal expression of cardiac development markers, and normal adult heart function. Similar observations have been obtained after Notch1 deletion with cTnT-Cre. We have also deleted Rbpj in both myocardial and endocardial progenitor cells, using the Nkx2.5-Cre driver, resulting in ventricular septal defect (VSD), double outlet right ventricle (DORV), and bicuspid aortic valve (BAV), due to NOTCH signaling abrogation in the endocardium of cardiac valves territories. Our data demonstrate that NOTCH-RBPJ inactivation in the myocardium does not affect heart development or adult cardiac function.


95
We first compared expression of the NOTCH effector RBPJ to the pattern of NOTCH activation in the 96 E12.5 heart. RBPJ was widely expressed in the nucleus of endocardial, myocardial and epicardial cells (Fig 1a-97 a'), while NOTCH1 activity was restricted to the endocardium (Fig 1b-b'). Expression of the NOTCH transgenic 98 reporter CBF:H2B-Venus in endocardial cells indicated that NOTCH activity was restricted to the endocardium 99 (Fig 1c-c').

128
Genetic manipulation of NOTCH elements leading to signal inactivation in the endocardium, disrupts 129 myocardial patterning and chamber maturation (17,27). We analyzed if ventricular patterning was affected after

164
Genotyping of neonatal and adult litters showed that all genotypes appeared at the expected Mendelian 165 proportions ( Table 1), indicating that RBPJ loss in the myocardium did not compromise postnatal viability.

166
Morphological analysis of 6-month old Rbpj flox ;cTnT-Cre adults revealed normal heart structure compared to 167 control animals (Fig 4a,b). In order to detect potential physiological impairments in the heart, we analyzed cardiac 168 function by echocardiography (Fig 4e). Ejection fraction (EF%) and fractional shortening (FS%) were similar in 169 wild type and mutant mice (Fig 4e). The diastolic function, indicated by the E/A ratio (ratio of early diastolic 170 velocity to atrial velocity) (54), was also normal. Physiological measurements indicate that ventricular volumetric 171 and mass parameters were normal compared to control mice. Overall, the echocardiography study indicates that 172 myocardial Rbpj inactivation does not affect postnatal heart growth and adult myocardial function. We confirmed 173 these results by inactivating Notch1 with the cTnT-Cre driver. Six-month old Notch1 flox ;cTnT-Cre adult hearts

198
For ECG, n = 6 control mice and n = 6 mutant mice. P<0.05 by Student's t-test; n.s., not significant; *P<0.05; significant differences neither in the main intervals PR and QT nor in the QRS complex duration compared to 206 control mice, suggesting that the conduction system is fully functional in Rbpj flox ;cTnT-Cre adult mice (Fig 4g).

207
To further confirm that myocardial RBPJ is dispensable for cardiac development and function, we used a   (Fig 5e-f). Seven mutants (54%) had bicuspid aortic valve (BAV), characterized by either right 214 to non-coronary (75% of cases) or right to left (25% of cases) morphology and resulting in a two-leaflet valve 215 instead of the normal three-leaflet valve (Fig 5c-d). Rbpj flox ;Nkx2.5-Cre mutant embryos developed a normal 216 compact and trabecular myocardium layers, with a thickness similar to controls (Fig 5g). Rbpj flox ;Nkx2.5-Cre 217 mutants showed perinatal lethality and died around postnatal day 0 (P0; Table 2).

239
To determine the precise contribution of Nkx2.5-expressing cells to the developing heart, we took 240 advantage of the mTmG system in which following CRE-mediated excision, the mTomato transgene is removed 241 so that the CAG promoter drives the expression of membrane localized EGFP (56). Lineage tracing analysis of 242 mTmG;Nkx2.5-Cre mice revealed both myocardial and endocardial contribution of CRE-expressing cells (Fig   243   5h-j), including partial recombination in the E9.5 outflow track (OFT) endocardium (Fig 5i) and in the mitral 244 valve endocardium at E16.5 (Fig 5j). RBPJ immunostaining in E16.5 control and Rbpj flox ;Nkx2.5-Cre embryos 245 showed efficient RBPJ abrogation throughout the ventricular myocardium (99.98 ± 0.02 of cardiomyocytes 246 recombined), while RBPJ was preserved in the majority of ventricular endocardial cells (29.06 ± 9.67% of 247 endocardial cells recombined; Fig. 5k-l''). In contrast, in endocardial cells overlying the valves, RBPJ depletion 248 was significantly more efficient (76.91 ± 6.48 %; P <0.01 by Student's t-test) (Fig. 5m-n'). Our results are in 249 agreement with previous reports showing that NOTCH signaling abrogation by deletion of Notch1 or Jagged1, Notch signaling for valve morphogenesis, and suggesting that the lethality observed in Rbpj flox ;Nkx2.5-Cre mutant 252 mice was very likely due to Rbpj inactivation in valve endocardium.

253
These results demonstrate that myocardial inactivation of Rbpj in cTnT-Cre;Rbpj flox mice does not affect 254 heart development and structure, nor does impair adult heart function, as it occurs with NOTCH signaling 255 inactivation in the endocardium (4, 17, 18, 26, 30, 57). In contrast, Rbpj deletion driven by the Nkx2.5-Cre driver 256 leads to VSD, DORV and BAV, phenotypes due to Nkx2.5-mediated CRE activity in valve endocardial cells in 257 which RBPJ mediates NOTCH signaling, with VSD being the likely cause of perinatal lethality of these mutants.

306
Olympus BX51 microscope and analyzed with ImageJ software. Using an inverted gray scale, pixel intensity was 307 measured throughout the myocardium and a mean pixel intensity was obtained for every heart sections. 7 sections 308 of each heart at different levels were analyzed and a mean pixel intensity was obtained for each heart.

310
Left ventricle (LV) function and mass were analyzed by transthoracic echocardiography in 6 months of age mice.

311
Mice were mildly anaesthetized by inhalation of isoflurane/oxygen (1-2%/98.75%) adjusted to obtain a target 312 heart rate of 450±50 beats/min and examined using a 30MHz transthoracic echocardiography probe. Images were 313 obtained with Vevo 2100 (VisualSonics). From these images, cardiac output, stroke volume and LV mass were 314 calculated. These measurements were normalized by the tibial length of each mice. Ventricular systolic function 315 was assessed by estimating LV shortening fraction and the ejection fraction. Diastolic function was assessed by 316 the E/A ratio. We performed a second echocardiography analysis two weeks after the first one and calculate the 317 mean for each parameter and each mouse.

319
Electrocardiograms were recorded with and MP36 system and analyzed using the Acknowledge 4 software. 6 320 months old mice were anesthetized by inhalation of isoflurane/oxygen (1-2%/98.75%) adjusted to obtain a target Statistical analysis was carried out using Prism 7 (GraphPad). All statistical test were performed using a two-324 sided, unpaired Student's t-test. Data are represented as mean ±s.e.m. All experiments were carried out with at 325 least three biological replicates. In the case of adult image analysis by echo and electrocardiogram analysis, the 326 experimental groups were balanced in terms of age and sex. Animals were genotyped before the experiment and