Superviseur : Dr. Piet Van Vliet Collaborateurs: Dr. Alexandre Dubrac, Dr. Martin Smith / Résumé INTRODUCTION : Les maladies valvulaires du cœur surviennent dans 2% de la population, impliquant souvent un reflux sanguin dû au rétrécissement de la valve. Nous avons récemment identifié deux familles non apparentées étant atteintes par une sténose aortique. Le séquençage exomique des familles a révélé une mutation entrainant une perte de fonction homozygote pour le gène ADAMTS19. La relation entre la perturbation du gène ADAMTS19 et la sténose a été reproduite et donc confirmée grâce à une souris ADAMTS19-LACZ KO/KO. Cette souris montre également que ADAMTS19 est spécifiquement exprimé dans les cellules valvulaires interstitielles (VICs) dans les valves. Le rôle d’ADAMTS19 durant le développement des valves reste inconnu. Pour analyser le patron d’expression d’ADAMTS19 pendant le développement du cœur, nous avons obtenu un modèle de souris transgénique contenant une CRE-tamoxifen inductible (Cre-ERT2) qui est exprimé sous l’influence du promoteur humain ADAMTS19. HYPOTHESE/OBJECTIF : Nous émettons l’hypothèse que le promoteur humain d’ADAMTS19 inséré dans la souris ADAMTS19-Cre-ERT2 contient toutes les séquences régulatrices permettant d’exprimer le gène ADAMTS19 et que ADAMTS19 est principalement exprimé au niveau des cellules valvulaires interstitielles dans les valves. L’objectif est de caractériser le patron d’expression d’ADAMTS19 en analysant sa distribution durant le développement grâce à une souris reportrice tdTomato. Comme ADAMTS19 est spécifiquement exprimé dans les VICs, cet outil transgénique permettrait d’étudier ces cellules durant le développement. METHODE/RESULTAT : Suite à une étude in silico le promoteur ADAMTS19 est apparu comme extrêmement conservé. Par conséquent, pour analyser l’expression d’ADAMTS19 nous avons obtenu une souris BAC ADAMTS19-Cre-ERT2 contenant la séquence conservée que nous avons croisé avec une souris reportrice tdTomato. Le Tamoxifen est administré aux femelles gestantes par gavage aux jours embryonnaires E9,5, E11,5 ainsi que E13,5, et les cœurs sont extrait a E16,5. Des coupes de cœurs embryonnaires vont permettre d’identifier la localisation et la morphologie des cellules marquées. L’expression d’ADMATS19 dans les cellules valvulaires interstitielles est consistant avec le fait qu’ADAMTS19 est connu pour affecter les valves durant le développent et dans le cas de maladie valvulaire. Cependant, le patron des valves n’est pas reproductible au travers des générations. De plus, nous observons qu’ADAMTS19 est marqué dans des cellules des oreillettes et ventricule dans une lignée et dans une sous population de cellules de l’artère pulmonaire dans une autre. CONCLUSION : L’analyse des séquences de chaque lignée par séquençage permettre d’investiguer la raison de ses différents patrons et de mettre en évidence des régulateurs spécifiques. / BACKGROUND: Valvular heart disease (VHD) occurs in ~2% of the general population, often resulting in reduced or disturbed blood flow. We recently identified two unrelated families with recessive inheritance patterns of progressive polyvalvular heart disease in absence of any clear syndromic phenotype. Exome sequencing revealed homozygous, rare, loss of function (LOF) alleles in both families for the gene ADAMTS19. The relation between ADAMTS19 mutation and aortic stenosis were confirm via an ADAMTS19-LacZ KO/KO mouse model. This model also shows that ADAMTS19 is specific of VICs during valve development. The ADAMTS protein family includes 19 proteases that are involved in matrix remodeling, and tissue homeostasis in development and disease. However, the role of ADAMTS19 specifically during valve development remains unknown. We aim to characterize ADAMTS19 expression using a BAC transgenic ADMTS19-CRERT2 mouse. HYPOTHESE/OBJECTIVE We hypothesize that the BAC used to make the ADMTS19-CRERT2 mouse contains all the regulatory elements to express it and also that its expression will be specific to the VICs. The objective is to establish the ADAMTS19- CREERT2 and therefore to create a new tool to study VICs in vivo. METHODS/RESULTS: In silico analysis of the human and mouse ADAMTS19 genomic regions showed a high level of conservation. Thus, to analyze ADAMTS19 expression patterns during mouse development, we obtained a BAC transgenic mouse model containing a tamoxifen inducible Cre (CreERT2) that is expressed under the influence of the human ADAMTS19 promoter and surrounding genomic region. We crossed males from several lines created in parallel with Rosa-tdTomato reporter females to generate offspring in which expression of the fluorescent tdTomato reporter is activated in ADAMTS19-expressing cells upon tamoxifen administration. Surprisingly, whole mount imaging of embryos induced at E13.5 and isolated at E16.5 revealed strong, but distinct labelling patterns in offspring from different ADAMTS19CreERT2 sublines. Whereas one line exclusively labelled VICs, consistent with ADAMTS19 in situ RNA expression data from the Eurexpress database, another line specifically labelled cells in atrial and ventricular, but not VICs. A third line seems to label only a subset of cells in the pulmonary artery. Labeling of ADAMTS19-positive VICs is consistent with ADAMTS19 affecting valve development and VHD. In addition, we observed exclusive ADAMTS19-dependent labelling in atrial and ventricular cells or in a subset of pulmonary artery cells in two different sublines. CONCLUSION: The distinct expression patterns in offspring from different ADAMTS19-Cre-ERT2 lines indicates that although regulation of ADAMTS19 is conserved between human and mouse, expression in VICs versus other cells may be dependent on mutually exclusive regulatory mechanisms.

Genome engineering, the deliberate alteration of an organism's genetic material, has revolutionized biotechnology and biomedical research, enabling precise modifications to DNA sequences. Among the tools developed for this purpose, site-specific recombinases (SSRs) stand out for their ability to catalyze targeted DNA rearrangements between defined target sites. The Cre/loxP system, in particular, has been widely used for conditional gene inactivation and recombinase-mediated cassette exchange, facilitating targeted DNA excision, inversion, or integration through the recognition and recombination of loxP target sites. While the inherent specificity of Cre towards the loxP target sequence has been invaluable, it also limits its application to other genomic loci of therapeutic interest. Understanding the factors that govern the enzyme’s DNA specificity opens the possibility to engineer and retarget the complex to non-native sequences, significantly broadening the range of targetable genomic loci. To address this challenge, I describe the development of a high-throughput method to quantify Cre recombination efficiency across a library of loxP-like spacer variants. This method systematically analyzes the impact of spacer sequence alterations to reveal DNA specificity determinants. Through comprehensive screening, the study identified spacer sequences that exhibit inefficient recombination by Cre, despite both full lox sites having matching spacer sequences. Directed evolution was used to enhance Cre activity on these previously 'inert' spacer sequences, generating variants with altered spacer specificity. Detailed molecular analyses, including mutational studies and molecular dynamics simulations, elucidated the structural basis for altered spacer selectivity in evolved Cre variants. The study provides mechanistic insights into the role of specific amino acid residues in determining spacer specificity and highlights the potential for the rational design of recombinases with tailored spacer preferences. Building upon this foundation, I describe the engineering of heterospecific Cre-type SSRs capable of recombining asymmetric DNA target sites. By combining two evolved Cre variants with unique half-site specificities, a functional heterotetrameric complex forms, capable of excising DNA fragments flanked by asymmetric target sequences naturally occurring in the human genome. This approach expands the applicability of SSRs and holds promise for correcting chromosomal inversions underlying genetic disorders, as demonstrated in the correction of the int1h inversion associated with hemophilia A. However, harnessing the full potential of heterospecific SSRs presents challenges, particularly concerning off-target effects resulting from the formation of undesired functional homotetrameric complexes. To mitigate these risks, I investigated strategies to render SSR monomers functionally active in heterotetrameric, but not homotetrameric complexes. Through substrate-linked directed evolution, I identified mutations that confer obligate heterospecificity, leading to safer and more precise genome engineering applications. Together, these studies highlight the transformative potential of engineered SSRs in genome editing and underscore the importance of ongoing research efforts to enhance their specificity, efficacy, and safety for therapeutic interventions and biotechnological applications. By manipulating the highly specific Cre/loxP complex to retarget different lox sequences and analyzing evolved or naturally occurring recombinase recombination specificity, we can better understand how these enzymes can be optimized for therapeutic applications. Furthermore, the ability to confer obligate heterospecificity increases the overall safety of these engineered SSRs, expanding their potential applications in genome engineering, particularly for therapeutic targets that require editing asymmetric (non-palindromic) target sites.