Lycat and cloche at the switch between blood vessel growth and differentiation?

The formation of the cardiovascular system starts in the mouse embryo at approximately embryonic day (E)7.0 to E7.5. The first blood vessels in the extraembryonic membranes, the major intraembryonic vessels, and the heart form by vasculogenesis, the in situ differentiation of mesodermal cells that give rise to “blood-islands.” The latter are composed of hemangioblasts, the common precursors of endothelial and blood cells. Hemangioblasts situated in the lumen of the blood islands will further differentiate into hematocytoblasts, the precursors of all 3 lineages of blood cells. In contrast, hemangioblasts lining the walls of the blood islands will give rise to angioblasts that form endothelial cells.1 Migrating angioblasts from the proximal lateral mesoderm assemble symmetrically at the lateral sides of the embryo to establish 2 preendocardial tubes. They fuse to give rise to the primordial heart.2 While vasculogenesis is still proceeding, the uniform blood islands begin to remodel to a network of large and small vessels by the process of angiogenesis, preferentially intussusceptive microvascular growth.3,4 Gene expression and targeting studies have identified vascular endothelial growth factor and its 2 receptors, KDR/ flk-1 and flt-1, as critical for the formation and early remodeling of the blood islands. Vascular endothelial growth factor is produced by endodermal and mesodermal cells at the onset of hemangioblast formation, whereas its receptors are expressed in the future endothelial cells lining the blood islands.5 Flk-1 / embryos lack blood islands throughout the embryo and yolk sac.6 In flt-1 / embryos, blood islands do not properly remodel but form large blood channels.7 Inactivation of a single vascular endothelial growth factor allele caused multiple embryonic malformations including the heart, rudimentary dorsal aortae, and a reduced number of blood cells.8,9 All deletions were lethal between days E8.5 and E11 to E12. Angiopoietin (Ang)-1, expressed by mesodermal cells, and its corresponding tie-1 and tie-2 receptors, located on the endothelium, form another important endothelial specific regulatory system that is critical for the mechanisms of intussusceptive microvascular growth.10–12 Recently, the zebrafish mutation cloche has been characterized to affect blood vessel and blood cell formation at a very early stage.13 In this issue of Circulation Research, Xiong et al14 report the isolation of lysocardiolipin acyltransferase, lycat, from the deletion interval of cloche, in the attempt to determine the molecular components of the cloche gene in zebrafish. Morpholino-mediated lycat knockdown results in a strikingly similar phenotype as compared with the cloche mutation. The central embryonic vascular network is established; however, the intersegmental vessel loops are approximately one-third longer and reduced in number with largely increased intervascular spaces as compared with controls. These vessels, as well as the central large vessels, the dorsal aorta, and the axial vein, express flk-1 and tie-1 in a mosaic pattern, in the way that some endothelial cells exhibit normal levels, whereas others produce none or very low amounts. The lycat-deficient embryos lack cranial blood vessels and possess an extremely thin common endocardial– myocardial layer, comparable to the cloche mutation. Flk-1, sc1, gata1, etsrp, and fli1 act in hemangioblasts downstream of lycat, as was previously demonstrated for cloche. This establishes lycat as one of the earliest known regulators of hemangioblasts.14 What does the lycat phenotype suggest concerning its function in endothelial cells, as well as in blood vessel and heart morphogenesis? For the formation of the heart, it is essential that the endocardial tube and the surrounding mesoderm that differentiates to myocardium interact. This communication is likely disturbed in the lycat mutants, and a normal heart therefore cannot form. Mice deficient of Ang-1, or its tie-2 receptor, were unable to recruit mesodermal cells to the endocardium resulting in a pathological heart morphology with a thinned myocardium that remained distant to the endocardium.10,11 It is not known presently whether Ang-1/ tie-2 signaling is affected by the lycat knockdown. In their video file (available in the online data supplement to the article at http://circres.ahajournals.org), Xiong et al show impressively “endocardial” contractions that are, however, less effective, because the myocardium forms no separate thick layer.14 This suggests, that mesodermal cells are recruited and likely incorporated into the endocardium to give rise to a common 1-layered, enlarged endocardial–myocardial structure instead of forming the adjacent myocardium. Low expression levels of tie-1 could support this inclusion of mesodermal cells into the endocardium. Tie-1 was identified as a counter player of tie-2, and low tie-1 expression would thus support tie-2–mediated periendothelial recruitment. (Figure, A).10,11 Alternatively, it could be argued that the endocardium never forms. However, the supplemental video in the article by Xiong et al shows that the circulation is intact.14 This means that the “myocardium” is properly connected to the large vessels, indicating that the sinus venosus and The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Department of Anatomy and Cell Biology, State University of New York Downstate Medical Center, Brooklyn. Correspondence to Sybill Patan, Assistant Professor, Department of Anatomy and Cell Biology, SUNY Downstate Medical Center, 450 Clarkson Ave, Box 5, Brooklyn, NY 11203-2098. (Circ Res. 2008;102:1005-1007.) © 2008 American Heart Association, Inc.