Increased Neuronal Differentiation of Neural Progenitor Cells Derived from Phosphovimentin-Deficient Mice

Abstract

Vimentin is an intermediate filament (also known as nanofilament) protein expressed in several cell types of the central nervous system, including astrocytes and neural stem/progenitor cells. Mutation of the vimentin serine sites that are phosphorylated during mitosis (VIM ^SA/SA ) leads to cytokinetic failures in fibroblasts and lens epithelial cells, resulting in chromosomal instability and increased expression of cell senescence markers. In this study, we investigated morphology, proliferative capacity, and motility of VIM ^SA/SA astrocytes, and their effect on the differentiation of neural stem/progenitor cells. VIM ^SA/SA astrocytes expressed less vimentin and more GFAP but showed a well-developed intermediate filament network, exhibited normal cell morphology, proliferation, and motility in an in vitro wound closing assay. Interestingly, we found a two- to fourfold increased neuronal differentiation of VIM ^SA/SA neurosphere cells, both in a standard 2D and in Bioactive3D cell culture systems, and determined that this effect was neurosphere cell autonomous and not dependent on cocultured astrocytes. Using BrdU in vivo labeling to assess neural stem/progenitor cell proliferation and differentiation in the hippocampus of adult mice, one of the two major adult neurogenic regions, we found a modest increase (by 8%) in the fraction of newly born and surviving neurons. Thus, mutation of the serine sites phosphorylated in vimentin during mitosis alters intermediate filament protein expression but has no effect on astrocyte morphology or proliferation, and leads to increased neuronal differentiation of neural progenitor cells.

Keywords: Astrocytes; Bioactive3D culture system; GFAP; Intermediate filaments; Nanofilaments; Neural stem/progenitor cells; Vimentin.

Fig. 2

VIM ^SA/SA neurosphere cells show increased neuronal differentiation. a VIM ^SA/SA and VIM ^WT/WT brain cells showed comparable neurosphere forming capacity (assessed for primary, secondary, and quaternary neurospheres) and VIM ^SA/SA and VIM ^WT/WT neurosphere cells showed comparable proliferation. The data show the total number of neurospheres formed from 50,000 plated cells (n = 4 per genotype) and the number of neurosphere cells generated from primary dissociated neurospheres (n = 4 per genotype). b VIM ^SA/SA and VIM ^WT/WT neurosphere cells showed comparable vimentin and nestin immunoreactivity. Western blot analysis showed comparable expression of nestin and SOX-2 but lower expression of vimentin in VIM ^SA/SA neurosphere cells compared with VIM ^WT/WT (n = 4 per genotype). c–e To assess the neuronal differentiation, dissociated neurosphere cells were allowed to differentiate, and double-immunolabeled with neuronal makers βIII-tubulin or MAP2, and astrocyte marker GFAP. VIM ^SA/SA neurosphere cells showed highly increased relative (left) and absolute (right) neuronal differentiation compared with VIM ^WT/WT neurosphere cells in both 2D and Bioactive3D culture system (n = 5 per genotype in 2D system, n = 4 per genotype in Bioactive3D system; c, e, respectively). VIM ^SA/SA neurosphere cells showed also twofold increase of neuronal differentiation in 2D culture system when the neurons were immunolabeled with antibody against MAP2 (d). f Western blot indicates higher expression level of βIII-tubulin in differentiated VIM ^SA/SA neurosphere cells than in differentiated VIM ^WT/WT neurosphere cells, in both 2D and 3D culture systems (n = 1 per genotype, pooled lysates from four samples). Data are presented as a mean ± SEM. *p < 0.05; **p < 0.01. Scale bar in (a) and (d), 50 μm. Scale bar in (b), 20 μm

Fig. 3

The pro-neurogenic properties of VIM ^SA/SA neurosphere cells are not affected by the astrocyte environment. a, b Dissociated VIM ^WT/WT or VIM ^SA/SA neurosphere cells pre-labeled with BrdU cocultured with VIM ^WT/WT or VIM ^SA/SA astrocytes in 2D (a) and Bioactive3D (b) culture systems. The graphs show the absolute number of immunolabeled neurons (βIII-tubulin^pos/GFAP^neg/BrdU^pos) and astrocytes (GFAP^pos/βIII-tubulin^neg/BrdU^pos) per square millimeter (n = 5 per genotype in 2D cultures, n = 4 per genotype in Bioactive3D cultures; a, b, respectively). Data are presented as a mean ± SEM. *p < 0.05. NS, neurosphere cells; A, astrocytes

Fig. 4

VIM ^SA/SA mice exhibit a modest increase in the fraction of newly born and surviving neurons in the hippocampal dentate gyrus. a Image of dentate gyrus immunostained with antibodies against doublecortin (DCX) and BrdU; the nuclei were visualized with DAPI. Arrow, DCX^posBrdU^pos cell; arrowhead, DCX^negBrdU^pos cell. VIM ^SA/SA mice had higher percentage of DCX^posBrdU^pos neuroblasts among the proliferation cells (BrdU^pos) in the dentate gyrus, but the absolute numbers of BrdU^pos cells and DCX^posBrdU^pos neuroblasts were comparable in VIM ^WT/WT and VIM ^SA/SA mice. Ten serial coronal brain sections with a 180-μm inter-section distance analyzed per mouse, n = 5 per genotype. b Image of dentate gyrus immunostained with antibodies against NeuN and BrdU. The nuclei were visualized with DAPI. Arrow, NeuN^posBrdU^pos cell; arrowhead, NeuN^negBrdU^pos cell. VIM ^SA/SA mice showed higher percentage of neurons (NeuN^posBrdU^pos) among surviving newly born cells (BrdU^pos), but the absolute number of surviving newly born neurons was comparable between the genotypes. VIM ^SA/SA mice showed decreased absolute number of surviving newly born cells (BrdU^pos). Ten serial coronal brain sections with a 180 μm inter-section distance analyzed per mouse, n = 10 per genotype. Data are presented as a mean ± SEM. *p < 0.05; **p < 0.01. The inlets in (a) and (b) show individual fluorescence channels. Scale bar in (a) and (b), 20 μm