Ranule neurons expressed CD44 (Fig. 8D and 8G). At P7, about 40 of CD44-positive cells isolated by FACS expressed NeuN (Fig. 8M). As it is possible to isolate the cells expressing CD44 only 374913-63-0 web weakly from the CD44negative cells by FACS, there is a possibility that neurons expressing CD44 only weakly were detected by FACS but not by immunostaining. This datum also supported that a part of immature granule neurons might express CD44 weakly. In the adult, most of the granule neurons in the GL expressed CD44 (Fig. 8I ). Interestingly, whereas immature granule neurons less or weakly expressed CD44 (Fig. 8D ), mature neurons strongly expressed CD44 (Fig. 8I ), suggesting that CD44 might be related to maturation and/or functions of these neurons.DiscussionIn this study, we revealed dynamic changes of CD44 expression in the mouse cerebellum during development. At E12, CD44 was observed in the ventricular zone of the IVth ventricle, but not in the rhombic lip. The expression of CD44 expanded into the entire cerebellum 3-Amino-1-propanesulfonic acid site between E14 and P3 (Fig. 2), after which expression became restricted to specific regions: strong expression of CD44 was observed in the PCL and WM at P7, and in the WM only at P14 (Fig. 3). Interestingly, three types of astrocytes showed different CD44 expression patterns (Fig. 6). GFAP-positive astrocytes appearing in the GL around P7 are protoplasmic astrocytes, and these astrocytes did not express CD44. CD44 expression was observed in immature Bergmann glia at P3 25837696 and P7. However, the expression of CD44 in Bergmann glia was never observed after P14. Fibrous astrocytes in the WM expressed CD44 from P3 to P14. The expression of CD44 in the WM (Fig. 6) decreased at P14, and it was lost by the adult stage. Thus, CD44 expression in astrocytes was transiently restricted to immature Bergmann glia and fibrous astrocytes during development. We previously isolated astrocyte precursor cells from P3 cerebellum by selecting cells that express high levels of CD44 [9]. The characteristics of CD44high cells, however, could not be discerned from those of CD44low cells sorted by FACS (Fig. 1) and in the sections from developing cerebellum (Fig. 6), and a part ofCD44 Expression in Developing CerebellumFigure 4. Representative FACS plots of CD44 positive cells from P3 mouse cerebellum. A : Unstained sample. D : CD44-stained sample. The CD44-positive population and CD44-negative population overlapped in the histogram of cell number vs. fluorescence intensity from CD44 antibody staining (CD44-PE) (A and C). To separate CD44-positive cells from CD44-negative cells clearly, dot plots depicted SSC vs. CD44-PE (B and D), and gates to separate the CD44-negative population from the CD44-positive population were determined. The gate information was reflected again by histogram (C and F). The CD44-positive cell population shown in E was isolated by FACS. doi:10.1371/journal.pone.0053109.gCD44high cells yield neurospheres (Fig. 1), raising the question whether CD44 expression was restricted to only astrocytic lineages. Therefore, we examined CD44 expression 
in other cell types. Remarkably, many CD44-positive cells co-expressed Sox2 and nestin, markers for neural 
stem cells and progenitor cells (Fig. 5). The majority of CD44-positive cells were thought to be neural progenitor cells at P3, and the percentage of CD44-positive cells that were identified as neural progenitor cells decreased during development. Consistent with the expression of CD44 by progenitors,.Ranule neurons expressed CD44 (Fig. 8D and 8G). At P7, about 40 of CD44-positive cells isolated by FACS expressed NeuN (Fig. 8M). As it is possible to isolate the cells expressing CD44 only weakly from the CD44negative cells by FACS, there is a possibility that neurons expressing CD44 only weakly were detected by FACS but not by immunostaining. This datum also supported that a part of immature granule neurons might express CD44 weakly. In the adult, most of the granule neurons in the GL expressed CD44 (Fig. 8I ). Interestingly, whereas immature granule neurons less or weakly expressed CD44 (Fig. 8D ), mature neurons strongly expressed CD44 (Fig. 8I ), suggesting that CD44 might be related to maturation and/or functions of these neurons.DiscussionIn this study, we revealed dynamic changes of CD44 expression in the mouse cerebellum during development. At E12, CD44 was observed in the ventricular zone of the IVth ventricle, but not in the rhombic lip. The expression of CD44 expanded into the entire cerebellum between E14 and P3 (Fig. 2), after which expression became restricted to specific regions: strong expression of CD44 was observed in the PCL and WM at P7, and in the WM only at P14 (Fig. 3). Interestingly, three types of astrocytes showed different CD44 expression patterns (Fig. 6). GFAP-positive astrocytes appearing in the GL around P7 are protoplasmic astrocytes, and these astrocytes did not express CD44. CD44 expression was observed in immature Bergmann glia at P3 25837696 and P7. However, the expression of CD44 in Bergmann glia was never observed after P14. Fibrous astrocytes in the WM expressed CD44 from P3 to P14. The expression of CD44 in the WM (Fig. 6) decreased at P14, and it was lost by the adult stage. Thus, CD44 expression in astrocytes was transiently restricted to immature Bergmann glia and fibrous astrocytes during development. We previously isolated astrocyte precursor cells from P3 cerebellum by selecting cells that express high levels of CD44 [9]. The characteristics of CD44high cells, however, could not be discerned from those of CD44low cells sorted by FACS (Fig. 1) and in the sections from developing cerebellum (Fig. 6), and a part ofCD44 Expression in Developing CerebellumFigure 4. Representative FACS plots of CD44 positive cells from P3 mouse cerebellum. A : Unstained sample. D : CD44-stained sample. The CD44-positive population and CD44-negative population overlapped in the histogram of cell number vs. fluorescence intensity from CD44 antibody staining (CD44-PE) (A and C). To separate CD44-positive cells from CD44-negative cells clearly, dot plots depicted SSC vs. CD44-PE (B and D), and gates to separate the CD44-negative population from the CD44-positive population were determined. The gate information was reflected again by histogram (C and F). The CD44-positive cell population shown in E was isolated by FACS. doi:10.1371/journal.pone.0053109.gCD44high cells yield neurospheres (Fig. 1), raising the question whether CD44 expression was restricted to only astrocytic lineages. Therefore, we examined CD44 expression in other cell types. Remarkably, many CD44-positive cells co-expressed Sox2 and nestin, markers for neural stem cells and progenitor cells (Fig. 5). The majority of CD44-positive cells were thought to be neural progenitor cells at P3, and the percentage of CD44-positive cells that were identified as neural progenitor cells decreased during development. Consistent with the expression of CD44 by progenitors,.
