Hemodynamic forces regulate developmental patterning of atrial conduction

Abstract

Anomalous action potential conduction through the atrial chambers of the heart can lead to severe cardiac arrhythmia. To date, however, little is known regarding the mechanisms that pattern proper atrial conduction during development. Here we demonstrate that atrial muscle functionally diversifies into at least two heterogeneous subtypes, thin-walled myocardium and rapidly conducting muscle bundles, during a developmental window just following cardiac looping. During this process, atrial muscle bundles become enriched for the fast conduction markers Cx40 and Nav1.5, similar to the precursors of the fast conduction Purkinje fiber network located within the trabeculae of the ventricles. In contrast to the ventricular trabeculae, however, atrial muscle bundles display an increased proliferation rate when compared to the surrounding myocardium. Interestingly, mechanical loading of the embryonic atrial muscle resulted in an induction of Cx40, Nav1.5 and the cell cycle marker Cyclin D1, while decreasing atrial pressure via in vivo ligation of the vitelline blood vessels results in decreased atrial conduction velocity. Taken together, these data establish a novel model for atrial conduction patterning, whereby hemodynamic stretch coordinately induces proliferation and fast conduction marker expression, which in turn promotes the formation of large diameter muscle bundles to serve as preferential routes of conduction.

Figure 1. Developmental changes in atrial conduction patterning.

A) Isochronal map (0.3 ms/div) depicting action potential propagation across the atria of a HH Stage 18 heart. The action potential initiates in the newly differentiated pacemaker cells (red) and propagates towards the left side of the atria (purple). B) Bright field image of the heart from (A), with the area of the atria and inflow region of the heart indicated by dashed lines. C) As in (A), for a HH Stage 24 atria. Note the emergence of finger-like projections in the isochronal map depicting regions of higher conduction velocity. D) Bright field image of heart from (C), with the area of the atria and inflow outlined by white dashed lines. E) As in (A), for a HH Stage 29 heart. Note the prominent finger-like extensions in the isochronal map demonstrating areas of rapid propagation. F) Bright field image of heart from (E), showing the location of the muscle bundles extending from Bachman's Bundle (BB) along the roof of the atria. G) Optical mapping data was superimposed on bright field images and the areas of the prominent muscle bundles were outlined (white dashed area) to demonstrate that rapid conduction followed the muscle bundles. Green signal  =  dV/dT max from a single frame of optical mapping recording. H) Quantification of average time for the right atria to activate at HH Stage 18, HH Stage 24, and HH Stage 29 (n = 8 for each). I) Quantification of right atrial conduction velocity from hearts in (H). P-values indicated significance relative to HH Stage 18 (***p<0.001, *p<0.05). v – ventral, l – left, d – dorsal, r – right, A - atria, LA – left atria, RA – right atria, BB – bundle of Bachman.

Figure 2. Molecular diversification of atrial muscle bundles.

A,B) Whole mount in situ hybridization for Cx40 and Nav1.5 at HH Stage 18. Viewed from a superior angle of the atrial roof. C,D) Whole mount in situ hybridization for Cx40 and Nav1.5 at HH Stage 29. Note deep staining along the muscle bundles. E,F) In situ hybridization for Cx40 and Nav1.5 performed on transverse sections through the roof of the atria. Note that Cx40 and Nav1.5 are enriched in the prominent pectinate muscle bundles (red arrows). G,H) In situ hybridization for Cx40 and Nav1.5 performed on transverse sections inferior to the atrial roof. Note deeper staining in the muscle bindles than in the thin-walled myocardium. I,J) Plots of staining intensity from sections in (G,H). Note that the relative intensity of the thin-walled myocardium (yellow/green) is lower than the myocardium at the tips of the muscle bundles (blue/purple).

Figure 3. Stretching induces conduction marker expression in the atria.

Whole embryonic hearts were isolated at HH Stage 18 and placed in culture for 8 hrs. A) Image of control heart (left lateral view) following 8 hrs of culture with vehicle (Tyrode's solution). B) Image of heart in which a bead of silicone oil was microinjected into the atria following 8 hrs. of culture with vehicle. C) Image of silicone injected atria cultured for 8 hrs with 5 uM GsMTx4. D) Quantification of gene expression levels in atria following whole heart culture. P-values indicated significance relative to control (***p<0.001, *p<0.05). A – atria, AVJ – atrioventricular junction, V – ventricle.

Figure 4. Atrial muscle bundles display increased proliferation rate relative to thin-walled myocardium.

A) Immunohistochemical staining in HH Stage 29 atrial muscle pulsed for 12 hrs with BrdU using antibodies against sarcomeric myosin (MF20-red) and BrdU (green). B) Section from (A) showing overlap of BrdU (green) and the nuclear marker Dapi (grey), with the MF20 positive area from (A) outlined (red line). C,D) As in (A,B) for the ventricular trabeculae. E) Quantification of BrdU incorporation within atrial muscle bundles (thicker then 4 nuclei) and thin-walled myocardium (4 nuclei or less). *p<0.05. F) Quantification of BrdU incorporation within ventricular trabeculae (thicker than 4 nuclei) and compact myocardium (4 nuclei or less).

Figure 5. Right vitelline vessel ligation decreases atrial conduction velocity.

A) Bright field image of sham operated embryo at HH Stage 18. Suture was placed under the vitelline artery and vein but was not tied. B) Bright field image of vitelline vessel ligation in a HH Stage 18 embryo. C) Isochronal map (0.3 ms/div) of HH Stage 29 sham operated embryo viewed from a superior angle to capture the roof of the right atria. Action potentials propagated from the sinoatrial junction (red) to ventral extent of the right atria (blue). D) Isochronal map of action potential propagation in an embryo at HH Stage 29 following vitelline vein ligation. Note that the distance between adjacent isochronal divisions is decreased, demonstrating slower conduction velocity. E) Quantification of right atrial activation time in sham operated (black diamonds) vs. ligated embryos (white boxes). n = 8, *p<0.05. F) Quantification of conduction velocity in sham operated (black bar) vs. ligated embryos (white bar). n = 8, **p<0.01.