Getting to the heart of cardiac morphogenesis.

One of the most difficult concepts in heart development is to convert a looped tube with the presumptive right and left chambers in series to a 4-chambered heart with blood flowing from right atrium to right ventricle and left atrium to left ventricle. Try to explain how it happens to a room full of typical first-year medical students (“just tell me what I need to know”) if you are into self-laceration. The key to shifting from serially connected presumptive chambers to chambers connected in parallel is the atrioventricular (AV) canal. Because this little canal originally connects only the part of the tube that will become the atria with the part that becomes the left ventricle, it is critical that it moves or expands to gain access to the more distal part of the tube that will become the right ventricle. It is only possible because the tube is looped. It may be obvious that the myocardium at the inner part of the loop must be remodeled along with a change in position or expansion of the AV canal. My awakening to the complexity of this region occurred several years ago with the publication of a study by Webb et al1 in Circulation Research. Until then, my rather naive view of the AV canal was fairly simplistic and involved some shifting and fusion of the AV cushions. Webb et al presented a coherent exposition of the central mesenchymal mass (septum intermedium) in developing mouse embryos. This central mass comprises the AV endocardial cushions, the valves of the sinus venosus and the sinus septum, and mesenchyme of the right pulmonary ridge. This ridge is additionally complicated because it has a core (called the spina vestibuli) originating from body mesenchyme, which is covered by mesenchyme that is continuous with the leading edge of the primary atrial septum. If you have trouble visualizing how these structures get together to form a central mass of mesenchyme, this study can help. During later development, according to the authors, some areas of this central mass become fibrous, but the larger part of the mass is muscularized to form the inferior edge of the fossa ovalis (oval fossa). After delamination of the septal leaflet of the tricuspid valve, the central mass becomes the AV component of the membranous septum and the fibrous skeleton of the heart. Webb et al did not address what happened to the myocardium of the AV canal, which is really a separate issue. The fate of the AV myocardium is very important, because it is actively involved in changing the connection of atria to ventricles from being serial to parallel. Another important result of this remodeling is to restrict the electrical continuity of atria and ventricles to the AV node, which is specialized for electrical conduction. If nonspecialized (working) myocardium remains at the AV junction, propagation of the electrical current through the heart is not normal and results in premature ventricular contractions, as seen in WolffParkinson-White syndrome.2–5 In this issue of Circulation Research, Kim et al6 present a detailed study using molecular markers to show the reorganization of the AV myocardium in formation of the AV junction in developing human heart. They have been able to follow certain portions of the AV canal myocardium using the spatial expression patterns of 2 markers, which this group has studied in heart development for many years. The GlN2 antibody was used to track parts of a ring of positive myocardium at the midpoint of the looped tubular heart that separates the presumptive left and right ventricles (called the interventricular foramen). This antibody was developed against a carbohydrate epitope in nodose ganglion by Barbu et al7 in Dr Nicole Le Douarin’s laboratory. Because so many neural antigens are expressed by myocardial cells that ultimately become specialized conduction myocardium, it should come as no surprise that this ring of tissue is thought to become the AV bundle and bundle branches.8 The second marker used in this study was creatine kinase M, which is expressed by the ventricular myocardium but not the AV canal myocardium. So the story of AV myocardial fate is very cleverly based on the region where GlN2 expression overlaps with absent expression of creatine kinase M. Using the dynamic patterns of positive and negative staining and the areas where they overlap, Kim et al6 confirm the results of previous studies by their group that the myocardium of the AV canal does not contribute to ventricular myocardium but is mostly incorporated into the smooth walls of the atria just above the AV valves. Although this was known earlier,9,10 the exact location of the AV myocardium in the definitive atria was not determined. The present study shows definitively that the continuity of the right side of the AV canal with the ascending (right ventricular) portion of the cardiac loop forms by local expansion of the inner curvature of the heart and, finally, that the AV node that connects the atria and ventricles electrically develops as a persisting portion of this myocardium. At issue was how the AV canal repositions and remodels itself to allow blood to flow from an incipient right atrium through the right side of the canal into the ascending (right ventricular) portion of the cardiac loop. It has been acknowledged for years that the AV canal must expand to the right.11 Several studies proposed that the AV endocardial cushions, The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Institute of Molecular Medicine and Genetics, Medical College of Georgia, Augusta, Ga. Correspondence to Margaret L. Kirby, PhD, Institute of Molecular Medicine and Genetics, Medical College of Georgia, Augusta, GA 30912-2640. E-mail [email protected] (Circ Res. 2001;88:370-372.) © 2001 American Heart Association, Inc.