Fig 1.
Expression of CD150 in human central nervous system (CNS) tissues.
(A) Immunohistochemical analysis of CD150 expression in human primary CNS tumors. Tumor paraffin sections of glioblastoma, anaplastic astrocytoma, diffuse astrocytoma, anaplastic ependymoma and anaplastic oligodendroglioma were stained with antibodies to CD150 (upper panel), glial fibrillary acidic protein, GFAP, astrocyte marker (middle panel) and nestin, neural stem/progenitor cell marker (lower panel). CD150 expression was detected in all presented histological variants of CNS tumors (upper panel). (B) Expression of CD150 in different regions of human normal brain tissues: cerebellum, thalamus, subventricular zone ependyma. The expression of CD150 was not found in human normal brain tissues. DAB staining shows specific reactions in brown. Cell nuclei are lightly counterstained with haematoxylin. Microscopic magnification of 400× was used for all images.
Table 1.
CD150 expression in CNS tumors.
Fig 2.
CD150 expression in serial sections of tumor and reactive brain tissues.
(A) Diffuse astrocytoma with extensive microcyst formation. (B) Zone of infiltration with tumor cells of perifocal brain tissue. Scattered tumor cells are shown with arrows. Activated microglia and reactive astrocytes are CD150 negative. (C) Infiltration of subventricular zone with tumor cells (arrows), negative reaction in ependymal cells (Ep). DAB staining shows specific reaction in brown. Cell nuclei are counterstained with haematoxylin. Microscopic magnification 400×.
Fig 3.
Expression of CD150 in human glioma cell lines.
(A) Flow cytometry study of CD150 surface expression on U87, U343 and A172 cells. One of four experiments. (B) Immunofluorescent analysis of CD150 expression in the cytoplasm of U87, U343 and A172 glioma cell lines. The cells were fixed, stained with anti-CD150 antibodies (IPO-3) followed by secondary antibody labelled with Alexa 488 (upper panel). Nuclei were visualized by staining with DAPI (middle panel). Results are representative of more than five experiments. Magnification: 630×. (C) Western blot analysis of CD150 expression in glioma cell lines using rabbit monoclonal anti-CD150 antibodies (Sino Biologicals Inc., China). Pre-B cell line REH and B-LCL MP-1 (in two different dilutions) were used as negative and positive controls respectively. One of five experiments is presented. The flow cytometry analysis of live cells does not show any CD150 expression in all used glioma cell lines, however fluorescent studies of fixed permeabilized cells and western blot detected CD150 expression in U87, U343, U251, TE671, NCH84, and NCH92 glioma cell lines.
Fig 4.
Measles virus (MV) does not infect human glioma cell lines.
(A) Measles virus cytopathic effect in glioma and control cell lines. U87 glioma cell line and Vero-SLAM cells (positive control) were infected with wild type MV strain G954 and laboratory MV strain Edmonston, at MOI of 1. Images show syncytia formation (white arrows) at 24 h (for Vero-SLAM cells) or 96 h (for U87 cells) post-infection. Microscopic magnification of 400× was used for all images. One of five experiments. (B-C) Wild type (B) and laboratory strain (C) MV production in glioma and control cell lines, determined daily by plaque assay on Vero-SLAM cells, 24 h (1), 48 h (2), 72 h (3) and 96 h (4) post-infection. Vero-SLAM and Vero cells were used as positive and negative control respectively. One of three representative experiments. All studied glioma cell lines were not sensitive to the infection with wild type measles virus (MV), which uses CD150 receptor for its entry. At the same time all cell lines were infected and produced high titers of laboratory adapted Edmonston MV strain that also uses ubiquitously expressed CD46 as entry receptor.
Fig 5.
Colocalization of CD150 with the markers of endoplasmic reticulum and Golgi apparatus.
U87 or MP-1 cells were stained for CD150 (green) and the markers of endoplasmic reticulum (A) or Golgi apparatus (B) (red). Markers for endoplasmic reticulum—kinectin-1 and GRP78, for Golgi—furin. Nuclei were visualized by staining with DAPI (4’,6-diamidino-2-phenylindole). Colocalization coefficients were determined using the Manders algorithm (which ranges from 0 to 1, where 0 is defined as no colocalization and 1 as perfect colocalization) and are indicated within the panels. Confocal microscopy shows the similar high colocalisation of CD150 and ER markers in both types of cells, but significantly lower colocalisation of CD150 and Golgi marker in glioma cell line U87 comparing to B cell line MP-1. Microscopic magnification of 630× was used for all pictures. Digital magnification of 3150× was made for the insertions. The data are presented as mean ± SD (n = 7).
Fig 6.
Expression of different domains of CD150 in glioma cell lines.
Specific primers to the extracellular (Extr CD150), cytoplasmic (Cyt-m CD150) and transmembrane plus cytoplasmic (TM CD150) parts of CD150 were used for RT-PCR analysis of CD150 expression. LCL and MP-1 cells were taken for positive control. The quality and quantity of cDNA was monitored by GAPDH expression. CD150 extracellular and transmembrane domains on mRNA level were detected in all studied glioma cell lines, while cytoplasmic domain was found only in U87 and A172 cell lines. One of more than five representative experiments.
Fig 7.
The expression of a novel CD150 splice isoform (nCD150).
(A) Schematic representation of the CD150/SLAM gene structure and alternative splicing of mRNA. The exon designated Cyt-new is flanked with canonical splice sites (AG/GT marked in grey in the above genomic sequence). Exons are shown by filled rectangles, noncoding sequences—by solid line. Abbreviations: LS—leader sequence, V and C2—extracellular domains, TM—transmembrane domain, Cyt—cytoplasmic tail. (B) Expression of a nCD150 splice isoform was found in TE671, A172, U343, NCH92, NCH89 and U87 glioma cell lines, human tonsillar B cells (TBC) and cell lines of B cell origin, human acute monocytic leukemia cell line THP-1 and dendritic cells (DC), T cells (CD3+), monocytes (Mon) and macrophages (Mac), but it was not detected in human primary monocytes (Mon). The quality and quantity of cDNA was monitored by GAPDH expression. One of three representative experiments.
Fig 8.
Real-time-PCR analysis of CD150 splice isoforms expression in glioma cell lines and primary tumors.
We used 6 glioblastoma cell lines, four glioma tumor samples (GB1, GB2—glioblastoma, AODG—anaplastic oligodendroglioma, DA—diffuse astrocytoma), human tonsillar CD38+ B cells and lymphoblastoid cell line T5–1 for the analysis. Expression level of mRNA coding for each CD150 isoform was calculated using (ddCt) method, normalized to TBP and then expressed relative to respective isoform in CD38+ B cells, the value for which was set at 1. The results, presented as mean of triplicates (±SEM), are from one of three independent experiments. In glioma cell lines and glioma primary tumors the novel CD150 isoform is expressed at the high level, while the isoforms with the conventional cytoplasmic tail are absent or detected at the low level.
Fig 9.
Cell lines transfected with mCD150 and nCD150 are susceptible to wild type measles virus infection.
(A) Surface expression of mCD150 and nCD150 isoforms in HEK293T and U87 cell lines after transfection with respective plasmids measured by flow cytometry, after staining with anti-CD150 mAb IPO3. (B) Wild type MV (G954 strain) production in HEK293T cells transfected with mCD150 and nCD150, determined daily by plaque assay on Vero-SLAM cells, 24 h (1), 48 h (2), 72 h (3), 96h (4) and 120 h (5) post-infection. Vero-SLAM and non-transfected HEK293T cells were used as positive and negative control respectively. One of three independent experiments. The expression of both mCD150 and nCD150 isoforms on the surface of transfected cells allows the entry of wt measles virus to the cell.