On average, galectin 3 was positive in 10% of the OLCs. Olig2 was diffusely positive with a positive rate of 88%. On the other hand, NeuN-positive OLCs were rare, exhibiting a positive rate of only 0.7%. To further characterize OLCs and floating neurons, we performed

double fluorescent immunohistochemistry (Fig. 6). For this procedure, we first confirmed that galectin 3 colocalized with GFAP in the cytoplasm and the processes of astrocytes (figures not shown). Galectin 3 also labeled the nuclei of astrocytes. While galectin 3 and Olig2 were Selisistat colocalized in the nuclei of the OLCs, both NeuN and Olig2 were mutually exclusive. In general, the number of NeuN-positive cells was greater than that of floating neurons, with NeuN-positive nuclei being found to be much larger than Olig2-positive nuclei. Sections cut perpendicular to the cortex were selected for evaluation. In such sections, the specific glioneuronal elements were embedded within the surface of the cortex and the NeuN-positive cells appeared to be sparser in the center compared to that AUY-922 seen in the periphery of the lesion. In addition, the NeuN-positive cells possessed a continuous laminar arrangement that was continuous with the adjacent cortex (Fig. 7). In contrast, a specific glioneuronal element

within the white matter contained no NeuN-positive cells (Fig. 8). For the quantitative analysis, we measured the density of the NeuN-positive cells in the specific glioneuronal elements within the cortex and those within the white matter (Table 3). As a control, we also measured the cells

in the adjacent cortex. The density of the NeuN-positive cells in the specific glioneuronal elements in the cortical area was 35% compared to the density of the NeuN-positive cells found in the adjacent normal cortex. In contrast, the density Diflunisal of the NeuN-positive cells in the specific glioneuronal elements in the white matter was only 2.6%. These differences were statistically significant. In order to confirm that the floating neurons are NeuN-positive, we decolorized representative sections with HE and then performed NeuN immunohistochemistry on the same section (Fig. 9). All of floating neurons were NeuN-positive and some OLCs were also positive for NeuN. We next manually traced the captured images of the nuclei of the NeuN-positive cells and then converted the traces into binary images (Fig. 10), which were analyzed using an image analysis system. The mean value and standard deviation of the area of the NeuN-positive nuclei in these elements were identical to those of the nuclei in the adjacent cortex (Table 4). However, the perimeters of the nuclei were significantly shorter in the areas in the elements. In addition, the circulatory factor, which represents the roundness of nuclei, was significantly larger in these elements. Next, we performed morphometry on the nuclear areas of the Olig2-positive cells.