Biology subjects

Desgrange, A.

Publications and source records attributed to Desgrange, A..

5 recordsLinked to original sources

bioRxiv2024

Deep learning-based detection of murine congenital heart defects from μCT scans

Congenital heart defects (CHD) result in high morbidity and mortality rates, but their origins are poorly understood. Mouse models of heart morphogenesis are required to study the pathological mechanisms of heart development compared to normal. In mouse fetuses, CHD can be observed and detected in 3D images obtained by thoracic micro-computed tomography (CT). However, diagnosis of CHD from CT scans is a time-consuming process that requires the experience of senior experts. An automated alternative would thus save time, empower less experienced investigators and could broaden analysis to larger numbers of samples. Here, we describe and validate an approach based on deep learning to automatically segment the heart and screen normal from malformed hearts in mouse CT scans. In an initial cohort, we collected 139 CT scans from thorax and abdomen of control and mutant perinatal mice. We trained a self-configurating neural network (nnU-Net) to segment hearts from body CT scans and validated its performance on expert segmentations, achieving a Dice coefficient of 96%. To identify malformed hearts, we developed and trained a 3D convolutional neural network (CNN) that uses segmented CT scans as inputs. Despite the relatively small training data size, our diagnosis model achieved a sensitivity, specificity (for a 0.5 threshold), and area under the curve (AUC) of 92%, 96%, and 97% respectively, as determined by 5-fold cross-validation. As further validation, we analyzed two additional cohorts that were collected after the model was trained: a prospective cohort, using the same experimental protocol as the initial cohort, and containing a subset of its genotypes, and a divergent cohort in which mice were subjected to a different treatment for heart arrest (cardioplegia) and that contained a new mouse line. Performance on the prospective cohort was excellent, with a sensitivity of 92%, a specificity of 100%, and an AUC of 100%. Performance on the divergent cohort was moderate (sensitivity: 69%, specificity: 80% and AUC: 81%), but was much improved when the model was finetuned on (a subset of) the cohort (sensitivity: 79%, specificity: 88% and AUC: 91%). These results showcase our models robustness and adaptability to technical and biological differences in the data, highlighting its usefulness for practical applications. In order to facilitate the adoption, adaptation and further improvement of these methods, we built a user-friendly Napari plugin (available at napari-hub.org/plugins/mousechd-napari) that allows users without programming skills to utilize the segmentation and diagnosis models and re-train the latter on their own data and resources. The plugin also highlights the cardiac regions used for the diagnosis. Our automatic and retrainable pipeline, which can be employed in high-throughput genetic screening, will accelerate diagnosis of heart anomalies in mice and facilitate studies of the mechanisms of CHD.

developmental biology↗

bioRxiv2024

Notch3 is a genetic modifier of NODAL signalling for patterning asymmetry during mouse heart looping

The TGF{beta} secreted factor NODAL is a major left determinant required for the asymmetric morphogenesis of visceral organs, including the heart. Yet, when this signalling is absent, shape asymmetry, for example of the embryonic heart loop, is not fully abrogated, indicating that there are other factors regulating left-right patterning. Here, we used a tailored transcriptomic approach to screen for genes asymmetrically expressed in the field of heart progenitors. We thus identify Notch3 as a novel left-enriched gene and validate, by quantitative in situ hybridization, its transient asymmetry in the lateral plate mesoderm and node crown, overlapping with Nodal. In mutant embryos, we analysed the regulatory hierarchy and demonstrate that Nodal in the lateral plate mesoderm amplifies Notch3 asymmetric expression. The function of Notch3 was uncovered in an allelic series of mutants. In single neonate mutants, we observe that Notch3 is required with partial penetrance for the development of ventricles, in addition to its known role in coronary arteries. In compound mutants, we reveal that Notch3 acts as a genetic modifier of Nodal, able to modulate heart looping direction and the curvature of the outflow tract. Whereas Notch3 was previously associated with the CADASIL syndrome, its contribution to asymmetric organogenesis is now relevant to severe laterality defects such as the heterotaxy syndrome.

developmental biology↗

bioRxiv2023

Torsion of the heart tube by shortage of progenitor cells : identification of Greb1l as a genetic determinant of criss-cross heart in mice

Despite their burden and impact, most congenital defects remain poorly understood by lack of knowledge of the embryological mechanisms. Here, we identify Greb1l mutants as the first mouse model of criss-cross heart. Based on 3D quantifications of shape changes, we demonstrate that torsion of the atrioventricular canal occurs together with supero-inferior ventricles at E10.5, after heart looping. Mutants phenocopy specific features of partial deficiency in retinoic acid signalling, suggesting that GREB1L is a novel modulator of this signalling. Spatio-temporal gene mapping and cross-correlated transcriptomic analyses further reveal the role of Greb1l in maintaining a pool of precursor cells during heart tube elongation, by controlling ribosome biogenesis and cell differentiation. Growth arrest and malposition of the outflow tract are predictive of abnormal tube remodelling in mutants. Our work on a rare cardiac malformation opens novel perspectives on the origin of a broader spectrum of congenital defects associated with GREB1L in humans. HighlightsO_LIGreb1l inactivation is the first model of criss-cross heart C_LIO_LIGrowth arrest of the outflow tract and reduced pole distance are predictive of the torsion of the atrioventricular canal, and also account for associated defects of supero-inferior ventricles and malposition of the great vessels C_LIO_LIVentricle position needs to be maintained after heart looping C_LIO_LIGREB1L, which is associated in humans with a spectrum of congenital defects, is required to maintain precursor cells, by promoting ribosome biogenesis and restricting cell differentiation. C_LIO_LIGREB1L is a novel factor involved in retinoic acid signalling. C_LI In BriefGREB1L is associated with a spectrum of congenital defects in humans. Bernheim et al now uncover its function in maintaining a reservoir of precursor cells. Inactivation of Greb1l in the mouse impairs the elongation of the heart tube leading to criss-cross heart with supero-inferior ventricles.

developmental biology↗

bioRxiv2019

Transient Nodal signalling in left precursors coordinates opposed asymmetries shaping the heart loop

The secreted factor Nodal has been shown to be a major left determinant. Although it is associated with severe congenital heart defects, its role in heart morphogenesis has remained poorly understood. Here, we report that Nodal is transiently active in precursors of the mouse heart tube poles, before the morphological changes of heart looping. In conditional mutants, we show that Nodal is not required to initiate asymmetric morphogenesis. We provide evidence of a heart-specific random generator of asymmetry that is independent of Nodal. Using 3D quantifications and simulations, we demonstrate that Nodal functions as a bias of this mechanism: it is required to amplify and coordinate opposed left-right asymmetries at the heart tube poles, thus generating a robust helical shape. We identify downstream effectors of Nodal signalling, regulating asymmetries in cell proliferation, cell differentiation and extra-cellular matrix composition. Our work provides novel insight into how Nodal regulates asymmetric organogenesis.

developmental biology↗

bioRxiv2019

Standardised imaging pipeline for phenotyping mouse laterality defects and associated heart malformations, at multiple scales and multiple stages

Laterality defects are developmental disorders resulting from aberrant left/right patterning. In the most severe cases, such as in heterotaxy, they are associated with complex malformations of the heart. Advances in understanding the underlying physiopathological mechanisms have been hindered by the lack of a standardised and exhaustive procedure in mouse models, for phenotyping left/right asymmetries of all visceral organs. Here, we have developed a multimodality imaging pipeline, which combines non-invasive micro-ultrasound imaging, micro-CT and HREM, to acquire 3D images at multiple stages of development and at multiple scales. Based on the position in the uterine horns, we track, in a single individual, the progression of organ asymmetry, the situs of all visceral organs in their thoracic or abdominal environment, together with fine anatomical left/right asymmetries of cardiac segments. We provide reference anatomical images and organ reconstructions in the mouse, and discuss differences with humans. This standardised pipeline, which we validated in a mouse model of heterotaxy, offers a fast and easy-to-implement framework. The extensive 3D phenotyping of organ asymmetry in the mouse uses the clinical nomenclature for direct comparison with patient phenotypes. It is compatible with automated and quantitative image analyses, which is essential to compare mutant phenotypes with incomplete penetrance and gain mechanistic insight into laterality defects.\n\nSummary statementLaterality defects, which combine anomalies in several visceral organs, are challenging to phenotype. We have now developed a standardised approach for multimodality 3D imaging in mice, generating quantifiable phenotypes.

developmental biology↗