The atrial septum divides the right and left atria. For fetal circulation, however, open communication between the two atria through the foramen ovale is essential for survival allowing right-to-left shunting of oxygenated blood supplied from the maternal circulation and thus by passing the fetal pulmonary circulation. Shortly after birth as the lungs become operational and the pulmonary circulation is established, the communication between atria is normally functionally closed. The structure of the atrial septum is complex and its development to maturation consists of multiple precursors in a series of events during fetal development (reviewed in Anderson and Brown, 1996; Wessels et al., 2000; Anderson et al., 2002; Wessels and Sedmera, 2003; Snarr et al., 2008).
Open to closed interatrial communication is directly achieved by the fossa ovalis, a remnant of the fetal foramen ovale, and the flap valve that seals it. They are composed of several specialized tissues including the septum primum and secundum. The fossa ovalis is bounded by the septum secundum at its superior margin, and by the muscular base of the septum primum at the floor (or antero-inferior rim). The muscular base of the primary atrial septum is formed by the myocardialized dorsal mesenchymal protrusion, to which the flap valve anchors. Superior margin of the fossa ovalis is formed by an infolding of the atrial roof (not true septum) in humans; however in mice, there is a small but distinct septum protrusion (true septum secundum) (Anderson and Brown, 1996; Wessels et al., 2000; Anderson et al., 2002; Wessels and Sedmera, 2003; Snarr et al., 2008).
When the fossa ovalis remains unsealed, this condition is termed patent foramen ovale (PFO) (Hagen et al., 1984; Kenny et al., 2008; Calvert et al., 2011). PFO is found in ∼27% of the general population, in which 98% of the PFO are small in size (1–10 mm, mean 4.9 mm) (Hagen et al., 1984). PFO is functionally closed most of the time as a result of a higher left atrial pressure than that of the right atrium under normal circumstances. Under conditions in which right atrial pressure exceeds that of the left atrium (e.g., coughing and sneezing), the foramen ovale can open, through which any materials including thrombi and air could pass from the venous to the arterial circulation. PFO has been linked to several conditions, including cryptogenic stroke, migraine with aura, decompression illness, and systemic arterial embolism. In humans, diagnosis of right-to-left shunt through the PFO is made by contrast echocardiogram with Valsalva maneuver, which transiently opens PFO (Kenny et al., 2008; Calvert et al., 2011).
When the two atria have a persistent open communication with blood flow shunting from left-to-right, this condition is called atrial septal defect (ASD), which accounts for about 7% of congenital heart disease (McMahon et al., 2002; Balzer, 2004). Sustained blood shunting through ASD increases right ventricular cardiac output and sometimes leads to pulmonary hypertension and heart failure in adults, which eventually leads to right-to-left shunts (Kirklin and Barratt-Boyes, 1993; Braunwald et al., 2001).
A growing number of genetically engineered mouse models have been generated to elucidate the mechanisms of cardiac anomalies. In fact, several studies have attempted to determine a link between PFO or the size of the foramen ovale and specific genetic backgrounds or genetic loci (Biben et al., 2000; Kirk et al., 2006; Williams et al., 2008). Because equivalent diagnostic criteria of PFO using contrast echocardiogram with Valsalva maneuver remain to be established in mice, cardiac anomalies are mostly analyzed by morphology.
Normal newborn mice have open interatrial communication, therefore newborn mice are not suitable to evaluate the presence of ASD (Williams et al., 2008). However, no studies have demonstrated when this communication normally closes in mice. To address this, we quantified the size of open interatrial communication in neonatal mice of the FVB/N strain on postnatal Days 2, 3, 4, and 7. Notably, the FVB/N mouse strain is most frequently used for generation of transgenic mice and a previous study showed that adult FVB/N mice do not exhibit PFO (Biben et al., 2000). We found that the size of open interatrial communication progressively decreases in developing neonatal hearts, and it closes by postnatal Day 7 (P7).
MATERIALS AND METHODS
Neonatal and adult mice of the FVB/N background starin were utilized with the date of birth defined as postnatal Day 1 (P1). All procedures were performed in compliance with relevant laws and institutional guidelines with approval from the University of Florida Institutional Animal Care and Use Committee.
Histological Analysis and Three-Dimensional Reconstruction
Paraffin-embedded heart tissue sections (5 or 7 μm thickness) from mice at P2, P3, P4, P7, 4 weeks and 3 months of age were stained with hematoxylin-eosin and analyzed. All the sections from a single heart have either 5 or 7 μm thickness. Three-dimensional (3D) reconstruction was performed as described previously (Soufan et al., 2003; Terada et al., 2011). The following antibodies were utilized for immunostaining: anti-Troponin T antibody (Sigma T6277), Palloidin-TRIC (Fluka 77418), and DAPI.
Values among groups were compared using ANOVA and Fisher PLSD post-hoc test (StatView version 5.01). P < 0.05 was considered significant.
RESULTS AND DISCUSSION
To define morphological changes associated with functional changes after establishment of the pulmonary circulation, we examined serial transverse tissue sections of neonatal hearts obtained from P2 to P7 FVB/N mice (a total of 15 hearts) with a reference to the mouse histology database (Petiet et al., 2008; Savolainen et al., 2009). The representative serial tissue sections are shown in Fig. 1A–J, selected from P3 heart separated by 161 μm (23 serial sections, 7 μm thickness each) from the posterior/dorsal to the anterior/frontal foramen ovale/fossa ovalis. Fossa ovalis was not yet sealed by the flap valve with the distance from the opening to the closure being 119 μm (17 serial sections, marked with*). Immunostaining visualized the troponin T-positive muscular structure in the septum primum (flap valve and its base, marked with arrowheads) and septum secundum forming superior margin of the fossa ovalis (Fig. 1K).
The size of the open interatrial communication (antero-posterior axis) calculated by the number of serial sections was progressively decreased from P2 hearts (151 ± 11 μm, N = 5) to P3 (111 ± 27, N = 3) and P4 (38 ± 12, N = 3). None of the P7 hearts (N = 4) exhibited open interatrial communication (Fig. 1L). Reconstructed 3D images of the tissue sections further clarified open interatrial communication in P3 heart (Fig. 1M, marked with *), which was no longer observed in P7 heart (Fig. 1N). However, in P7 hearts, a small gap between the septum secundum and flap valve was present (Fig. 2A–F, arrows), suggesting that process of fusion is under way. That was similar in 4-week-old hearts (N = 5, sections from two animals are shown in Fig. 2G,H). At 3 months of age (N = 3), the flap valve was positioned close to the septum secundum without an apparent gap between two septa (sections from two animals are shown in Fig. 2I,J).
Overall, our results indicate that interatrial communication is absent in P7 hearts followed by anatomical/structural fusion between two septa to be completed by several months.
In human, closure of interatrial communication occurs during the first year. Approximately 27% of the general population demonstrates PFO, yet it is functionally closed most of the time (Hagen et al., 1984; Kenny et al., 2008; Calvert et al., 2011). In mice, prevalence of PFO as well as ASD is considerably affected by genetic background (Biben et al., 2000). Thus, inclusion of control litters with the same genetic backgrounds is necessary in analysis of cardiac morphologies in genetically engineered mice. In addition, our study demonstrates the presence of open interatrial communication in P4 but not P7 hearts in wild-type FVB/N mice, suggesting that it is appropriate to determine abnormal interatrial communication after P7 in analysis of genetically engineered mice.
The authors greatly appreciate K. Fortin, E.O. Weinberg, L.E. Briggs, and S.A. Warren for technical help and valuable suggestions.