Letter by Anderson et al Regarding Article, “Abnormal Conduction and Morphology in the Atrioventricular Node of Mice With Atrioventricular Canal–Targeted Deletion of Alk3/Bmpr1a Receptor”

To the Editor:

The embryonic origin of the cardiac conduction system and the transcription factors involved are an important and emerging field of research.^1,2 We read with interest, therefore, the article by Stroud and colleagues, in which they describe important findings on the role of bone morphogenetic proteins in the formation of the atrioventricular conduction axis.³ Much of their argument rests on the histology of the atrioventricular conduction axis. In our opinion, their identification of the various nodal structures is incorrect. Just over 100 years ago, Sunao Tawara described in exquisite detail⁴ how the atrioventricular node is the atrial component of an axis of histologically specialized tissue providing the conduction pathway from the atrial to the ventricular myocardium. In subsequent equally important articles, Aschoff⁵ and Mönkeberg⁶ showed how, to be recognized histologically, the cells of the conduction tissues needed to be traced from section to section and shown to be histologically distinct from the adjacent working myocardium. This means that, for the atrioventricular node, the conduction axis needs to be traced dorsally from the penetrating bundle of His. For this reason, in frontal sections, as shown by Stroud and colleagues,³ it should not be possible to identify the bundle of His and the atrioventricular node in the same section. Thus, in our opinion, the tissues identified as “AVN” in Figures 3 and 6 of the article by Stroud et al³ cannot be atrioventricular node because the bundle of His is clearly present in the sections. Instead, the tissues identified as “AVN” have the histological appearance of blocks of working atrial myocardium. It would have been helpful if Stroud and colleagues had confirmed the “specialized” nature of these myocardial areas by staining for specific markers, such as connexin43 or HCN4 (as examples, see Yoo et al⁷ and Dobrzynski et al⁸). The use of such markers permits unequivocal identification of the cells making up the atrioventricular conduction axis. Furthermore, if Stroud and colleagues had traced the atrioventricular conduction axis dorsally in the series of sections shown in Figures 3 and 6, they would have discovered that it is a normal finding for the atrioventricular node to divide into 2 inferior extensions that become the ring tissues encircling the tricuspid and mitral valves. The images shown in Figure 8C through 8F of “twin or split AV nodes” in the article by Stroud et al³ could, therefore, be a normal finding, whereas the structure purportedly identified as the atrioventricular node in Figure 8A and 8B is again likely to be a block of normal working atrial myocardium. Stroud and colleagues are examining an important topic of study, but further work is required to unravel the role of bone morphogenetic proteins in the formation of the atrioventricular conduction axis.