Fig 1.
Loss of FlnA and Fmn2 causes a midline thoracoabdominal schisis, delayed bone formation and shortened gut.
A. Gross anatomical photographs of FlnA+Fmn2 double knockout E15.5 embryos show a thoracoabdominal schisis that is characterized by midline extrusion of the heart and lungs in the ventral thorax, and the liver and intestines in the ventral abdomen. The overlying skin is also translucent and thinned. Noticeably, the ribcage/sternum (see green arrows) are localized to the lateral sides of the thorax in the null FlnA+Fmn2 mouse but was present in the ventral midline of WT or null FlnA mice (see arrows). B. Gross anatomical photographs of E15.5 mice following Alcian blue (cartilage) and Alizarin red (bone) staining shows a delay in development (ossification) of the ribcage in null FlnA+Fmn2 embryos, compared to null FlnA and WT age matched littermates. Midline closure of the sternum and ribs (green arrows) is completely absent in the double knockout mice, slightly delayed in the null FlnA mice and completed in the WT mice. The green arrows and dashed red lines in the null FlnA+Fmn2 mice indicate the contour of the sternum/ribs as they progress toward the midline. C. The intestines of null FlnA and null FlnA+Fmn2 embryos are shorter than that seen in WT embryos (WT vs null FlnA * = p <0.015, WT vs null FlnA+Fmn2 ** = p <0.004, null FlnA vs null FlnA+Fmn2 * = p <0.047). D. Findings of thoracoabdominal schisis formation in the null FlnA+Fmn2 mice during embryonic development are summarized in the table including the numbers of embryos examined. We also examined tens of null Fmn2 or null FlnA embryos from single gene-deleted mother. The results showed that loss of Fmn2 alone was insufficient to cause midline closure defects, and loss of FmnA caused a delay in midline closure as displayed in Fig 1B, but could lead to the formation of thoracoabdominal schisis. In contrast, approximately half of the heterozygous FlnA and Fmn2 embryos (3/6) or null FlnA and heterozygous Fmn2 embryos (3/5) actually developed thoracoabdominal schisis with extrusion of the organs through the midline, suggesting the additional effect of FlnA and Fmn2 functions on formation of thoracoabdominal schisis. Finally, all the null FlnA and Fmn2 embryos examined (6/6) developed thoracoabdominal schisis, contrast to only a midline closure delay for null FlnA embryos, indicating a significant difference in development of phenotypes between null FlnA alone and null FlnA and Fmn2 embryos. In addition, measurement of the schisis at the chest midline showed greater defects in midline closure in the double FlnA+Fmn2 null vs single FlnA null vs WT mice (graphically summarized, WT vs null FlnA: ** = p< 0.0017, null FlnA vs null FlnA+Fmn2: ** = p < 0.0025, WT vs null FlnA+Fmn2: *** = p < 0.0011).
Fig 2.
Loss of FlnA and Fmn2 leads to smaller body size, thinned muscle, and skeletal wall, as well as defects in cardiac valves.
A. Brightfield photomicrographs of sagittal sections of E13.5 mouse embryos stained with hematoxylin and eosin (H&E) demonstrate the smaller body size and thinned or unformed ventral body wall (arrows) in the null FlnA and null FlnA+Fmn2 embryos. B. H&E staining of transverse whole body sections corresponding to the dashed lines in (A) shows severe developmental defects in the muscle/ribs along the lateral thoracic and ventral body wall. C. Higher magnification photomicrographs corresponding to boxed area (a) in panel B capture the thinned ventral body wall (bracket) in the double knockout mice. In the age-matched WT mice, the ventral wall is considerably thicker and the rib/sternum (arrow) has migrated near the midline. Higher magnification photomicrographs corresponding to the boxed area (b) in panel B display a severe defect in thoracic muscle formation in null FlnA, and null FlnA+Fmn2 embryos (arrows marked muscles). Scale bar = 500 μm.
Fig 3.
Loss of FlnA and Fmn2 leads to a defect in cell proliferation along the ventral thoracic wall.
A. Fluorescent photomicrographs of immunostaining performed on transverse sections of E13.5 mouse thorax for proliferation markers. Ki-67 (a marker for cells in the cell cycle) and BrdU (a marker labeling cells entering into S phase) revealed a significant decrease in the number of proliferating (Ki-67+ or BrdU+) cells per unit area in the ribcage and ventral midline of the null FlnA and null FlnA+Fmn2 embryos compared to control WT embryo. Brackets mark the thickness of the ribcage, whereas the rectangular box delineates the thickness of the ventral midline. B. The number of BrdU or Ki67 positive cells in the ventral midline of null FlnA and null FlnA+Fmn2 embryos was decreased by more than 50% in comparison with those in WT embryo (WT vs null FlnA *** = p <0.0002, WT vs null FlnA+Fmn2 *** = p <0.0008). C. Immunostaining of E13.5 mouse ribcage and sternum for desmin, a mesenchymal and muscle cell marker reveals a reduction in the numbers of mesenchymal-derived cells in null FlnA and null FlnA+Fmn2 embryos (arrows). Nuclei are counterstained with DAPI (blue color) in all the images. Three homotopic sections were examined from n>3 littermate mice per experimental variable. Scale bars = 50 μm.
Fig 4.
Programmed cell death does not cause the thoracoabdominal schisis seen in the FlnA+Fmn2 double null mice.
Fluorescent photomicrographs show decreased rates of programmed cell death (fluorescein) between WT vs null FlnA, and WT vs null FlnA+Fmn2 cortex within the mouse sternum (WT vs null FlnA ** = p < 0.00012, Wt vs null FlnA+Fmn2 ** = p < 0.0004, null FlnA vs null FlnA+Fmn2 * = p < 0.05). Immunostaining was performed on E13.5 ventral walls of the thorax (lower panels) for caspase 3, an apoptotic marker. Nuclei are counterstained with DAPI (blue color). Three homotopic sections were examined from n>3 littermate mice per experimental variable. Scale bars = 50 μm.