Ids Barriers CNS (2016) 13:Page 3 ofwith Mayer’s hematoxylin (Wako, Tokyo, Japan). Immunofluorescence staining of the rehydrated sections was performed using the protocol outlined above for cryostat sections, but without the antigen retrieval. For immunofluorescence PubMed ID:https://www.ncbi.nlm.nih.gov/pubmed/27486068 staining of fresh frozen sections, sections was fixed in methanol (-20 , 30 min) and acetone (4 , 10 min), incubated with blocking buffer, primary, and fluorescently-labelled secondary antibodies. The following primary antibodies were used in this study: anti-GLUT9 (NBP1-05054, Novus Biologicals, 1/500 or 1/1000 dilution for immunofluorescence staining, 1/5000 dilution for immunohistochemistry); anti-acetylated-tubulin (6-11B-1, Sigma, 1:500); anti-NeuN (A60, Millipore, 1/500); anti-P Glycoprotein (C219, GeneTax, 1/200); anti-BCRP/ABCG2 (BXP-9, Abcam, 1/100). For antigen absorption experiment, GLUT9 peptide (NBP105054PEP, Novus Biologicals) was used at a concentration of 100 g/ml. In situ hybridization was performed using ViewRNATM ISH Tissue 2-Plex assay (Affymetrix, Tokyo, Japan) according to the manufacturer’s protocol. Paraffin sections were deparaffinized with Hemo-De, boiled in pretreatment solution for 10 min, and digested with protease for 10 min. Sections were then hybridized with ViewRNA probes (Mouse Slc22a12, Type 6; Mouse Slc2a9, Type 1: Mouse Abcg2, Type 1). Bound probes were amplified by hybridization with preamplifier and amplifier mix solutions. Sections were incubated with Label Probe 6-alkaline phosphatase (AP) solution and Fast Blue substrate. Subsequently, sections were incubated with Label Probe 1-AP Type 1 solution and Fast Red substrate. Finally, the sections were counterstained with Gill’s hematoxylin and cover-slipped with Ultramount Permanent Mounting Medium (Dako Japan, Tokyo, Japan).In situ hybridization Image acquisitionusing a rabbit polyclonal anti-GLUT9 antibody on PFAfixed murine brain frozen sections. Distinct GLUT9 immunoreactivity in the ependymal cells of the dorsal third ventricle, and weaker staining in the brain parenchyma region were detected (Fig. 1a). The specificity of anti-GLUT9 antibody was verified by antigen absorption test (Fig. 1b). Higher magnification imaging revealed that GLUT9 was not localized to the nucleus or the cilia, visualized using anti-acetylated–tubulin, suggesting an intracellular and plasma membrane localization (Fig. 1c, d). To identify the cell type of GLUT9-positive cells in the brain parenchyma, we performed double immunostaining for GLUT9 and NeuN (neuronal SB 202190 supplier marker; Fig. 1e ). All GLUT9-positive cells in the parenchyma co-localized with NeuN, suggesting that GLUT9 is also present in neurons.GLUT9 immunoreactivity is detected in ependymal cells of all ventriclesNext, we observed the distribution of GLUT9 in other ventricles using PFA-fixed, frozen coronal sections at different levels. In addition to the ependymal cells of dorsal third ventricle (Figs. 1a, 2c), GLUT9 immunoreactivity was detected in ependymal cells of all ventricles including the lateral, ventral third, and fourth ventricles, and the aqueduct (Fig. 2a, b, d ). GLUT9 was also prominent in tanycytes, which are specialized ependymal cells that line the floor of the ventral third ventricle, and in its long processes that extend into the parenchyma (Fig. 2d). No immunostaining was detected in the choroid plexus (Fig. 2b, c, f).Immunohistochemistry and immunofluorescence staining of GLUT9 in a methacarnfixed murine brainDAB-stained secti.
