Apoptosis is an important process in the regulation of morphogenesis and has been recognized to occur in many embryonic and fetal tissues. Apoptosis can regulate morphogenesis in two ways, in tissue remodeling, or in tissue destruction by massive cell death. The latter event, although more rare, is associated with the production of reactive oxygen species (ROS), which can be demonstrated in situ by using specific histochemical reactions.

Fetal heart is known to be subjected to apoptosis. During heart morphogenesis in homeotherms, apoptosis plays a key role in cardiac tissue remodeling. Pexieder (1975) was the first to map the sites of apoptosis in a number of heart locations using standard haematoxylin-eosin staining. More recently, the TUNEL reaction is frequently used to unambiguously demonstrate sites of cell death, both in the muscular wall and in septal structures.

Septal structures are the target of important and complex morphogenetic changes during embryonal development. Their cellular components are mesenchymal cells which originate early in development in transient structures called endocardial cushions, which are primarily involved in heart septation. A number of studies have shown the importance of tissues such as the endocardium, neural crest cells, and epicardium in the production of cardiac mesenchymal cells.

Although the origin of these cells has been investigated in detail, the ultimate fate of mesenchymal cells is not yet clear, in part because of the lack of suitable biochemical and/or developmental markers of mesenchymal cells. In the past years, it has been hypothesized that mesenchymal cells can differentiate into cardiac myofibroblasts.

However, morphological analysis of cardiac malformations in human embryos led to the suggestion that the endocardial cushions are transitory structures which do not directly contribute to valvar or septal structures. Recently, Oosthoek et al. (1998) demonstrated the relevance of cushion tissue as origin of the tension apparatus in valve leaflets.

We have recently shown that an extracellular glycoprotein that is recognized by the PNA lectin is synthesized only by mesenchymal cells when they populate endocardial cushions and may thus represent a useful marker of endocardial cushion-derived tissues. Because PNA positivity was found to decrease with fetal age, which coincided with the histiogenesis of mature septal and valvular tissues, we have performed experiments to investigate whether apoptosis of mesenchymal cells is related to the loss of PNA positivity and to the differentiation of mature cardiac tissues, including the appearence of novel markers such as metallothioneins.

Swiss albino mice were used for all the experiments. Females were placed in a cage with a male from 6 pm to 6 am, the following day. At this time point, the presence of the vaginal plug in females was determined and it was taken as 0 dpc. On a specific day during pregnancy, females were anesthetized with chloroform and killed by cervical dislocation. All experiments were carried out in agreement with the legislation regulating the use of experimental animals.

To detect apoptosis in situ, we used the TdT (terminal deoxynucleotidyl transferase)-mediated dUTP nick end-labeling technique (TUNEL) for in situ detection of apoptosis using the In situ Cell Death Detection Kit (Roche SpA, Milan, Italy). Deparaffinized sections were rehydrated and treated with proteinase K. Labeled nucleotides were incorporated at the ends of nuclear DNA using TdT and were detected by alkaline phosphatase activity in the presence of nitro blue tetrazolium (NBT) and 5-bromo-4-chloro-3?-indolylphosphate p-toluidine salt (BCIP; both from Roche).

Lectin from Arachis hypogea (peanut agglutinin, PNA), recognizing the disaccharide residue ?-D-galactosyl (1-3)-N-acetylgalactosamine, was obtained from Vector Laboratories (Burlingame, CA, USA). Embryos were washed with phosphate-buffered saline (PBS), and fixed in ice-cold 4% paraformaldehyde in phosphate-buffered saline (PBS) for 2 h at 4 °C, then washed for the same time in PBS, dehydrated in ethanol series and embedded in paraffin.

Staging of embryos for descriptive purposes was carried out according to Kaufman (1992). Lectin binding was performed on thin paraffin sections (4–6-?m thick). Sections were incubated in the presence of horseradish peroxidase-labeled lectin diluted in PBS at a final concentration of 20 ?g/ml for 30–60 min at room temperature. In control experiments, sections were incubated with the lectin in the presence of the corresponding inhibitory sugar (D-galactose) in order to test for specificity of the binding. At the end of the incubation period, sections were carefully washed and peroxidase activity was stained using 0.5 mg/ml diaminobenzidine in 50 mM Tris–HCl buffer (pH 7.0) in the presence of 0.5% hydrogen peroxide. Sections were thoroughly washed, dehydrated and mounted for light microscopy without counterstaining.

Serial sections were stained with hematoxylin-eosin for histological analysis.

Sections were processed using the indirect immunoperoxidase method for demonstrating metallothioneins (MTs). Briefly, sections were deparaffinized and rehydrated. After blocking endogenous peroxidase with 0.3% hydrogen peroxide in PBS, sections were incubated overnight at 4 °C using a 1:100 dilution of the primary antibody E9 produced against a conserved epitope shared by both MT-1 and MT-2 isoforms (Dako, Copenhagen, Denmark). Sections were washed in PBS and incubated for 1 h with a peroxidase-conjugated goat-anti-mouse IgG (diluition 1:100; Sigma, St. Louis, MO, USA). Peroxidase activity was localized as described above.

Specificity of immunostaining was verified by replacing the primary antibody by nonimmune goat serum or PBS.

In mouse fetuses at stage 14–15 dpc, centripetal growth of the aortico-pulmonary septum (formerly called spiral septum) allowed it to reach both the atrioventricular and the interventricular septum and to fuse to them; separation of aorta from the pulmonary artery was completed, and semilunar valve leaflets were being formed.

At this stage (14.5 dpc), the region in the middle of the septum showed a large number of apoptotic cells as demonstrated with the TUNEL reaction. Microscopical analysis of the areas of cell death showed wide fenestrations, due to the confluence of intercellular spaces caused by cell depletion; remnants of apoptotic cells were often found at their periphery.