Bmi-1 over-expression in neural stem/progenitor cells increases proliferation and neurogenesis in culture but has little effect on these functions in vivo

Abstract

The polycomb gene Bmi-1 is required for the self-renewal of stem cells from diverse tissues, including the central nervous system (CNS). Bmi-1 expression is elevated in most human gliomas, irrespective of grade, raising the question of whether Bmi-1 over-expression is sufficient to promote self-renewal or tumorigenesis by CNS stem/progenitor cells. To test this we generated Nestin-Bmi-1-GFP transgenic mice. Analysis of two independent lines with expression in the fetal and adult CNS demonstrated that transgenic neural stem cells formed larger colonies, more self-renewing divisions, and more neurons in culture. However, in vivo, Bmi-1 over-expression had little effect on CNS stem cell frequency, subventricular zone proliferation, olfactory bulb neurogenesis, or neurogenesis/gliogenesis during development. Bmi-1 transgenic mice were born with enlarged lateral ventricles and a minority developed idiopathic hydrocephalus as adults, but none of the transgenic mice formed detectable CNS tumors, even when aged. The more pronounced effects of Bmi-1 over-expression in culture were largely attributable to the attenuated induction of p16(Ink4a) and p19(Arf) in culture, proteins that are generally not expressed by neural stem/progenitor cells in young mice in vivo. Bmi-1 over-expression therefore has more pronounced effects in culture and does not appear to be sufficient to induce tumorigenesis in vivo.

Figure 1. Transgene expression in Nestin-Bmi-1-GFP transgenic mice

A). Schematic representation of the Nestin-Bmi-1-GFP transgene. B). GFP epifluorescence shows transgene expression throughout the CNS in an E14.5 transgenic embryo but not in non-neural tissues or in a wild type control embryo (to the left of the transgenic embryo). C) The transgene was still widely expressed in most regions of the P0 transgenic brain, with nerve fiber bundles (ac, cc) and the wall of the ventricular system (aq) expressing the highest levels of GFP (left, whole brain overview; right, sagittal view from the midline). D-F). GFP expression became more restricted in the adult transgenic mouse brain (D) and concentrated in regions enriched for neural stem/progenitor cells, such as the SVZ of the lateral ventricle (E) and the subgranular layer of the dentate gyrus (F). All photos are from line C transgenic mice, but are also representative of line B mice. Flow-cytometry of cells from line B mice confirmed the presence of GFP⁺ cells in the ventricular zone, the cortex, and the cerebellum of neonatal transgenic mice (green histograms; G) but not in control littermates (unshaded histograms; G), as well as in the adult SVZ (H). I-J). The vast majority of neurospheres formed by neonatal VZ, cerebral cortex, and cerebellum (I) as well as adult SVZ (J) derived from the GFP+ fraction of cells. Data represent mean±s.d. from 3 independent experiments using line B transgenic mice. K). Many Bmi-1 transgenic neurospheres were capable of undergoing multilineage differentiation into astrocytes (GFAP⁺, green), neurons (Tuj-1⁺, red) and oligodendrocytes (O4⁺, black). ac, anterior commissure; aq, aqueduct; cbl, cerebellum; cc, corpus callosum; ctx, cerebral cortex; DG, dentate gyrus; LV, lateral ventricle; ob, olfactory bulb; str, striaum.

Figure 2. Bmi-1 expression was elevated in transgenic neural stem/progenitor cells

A). qRT-PCR showed that Bmi-1 transcript expression was markedly increased in uncultured transgenic P0 VZ and adult SVZ cells as well as in neurospheres cultured from adult transgenic SVZ cells as compared to littermate control cells. Fold change is presented (mean±s.d. from 3-5 independent experiments) relative to control cells, which were set to 1. Total cDNA levels in each pair of samples were normalized based on ß-actin expression. B) Western blotting showed the expression of HA tag, as well as increased Bmi-1 and decreased p16^Ink4a and p19^Arf in transgenic as compared to control neurospheres from lines B and C. Western blotting showed the expression of HA tag and increased Bmi-1 levels in E14.5 telecephalon (C) and P0 brain (D) from transgenic as compared to control mice. Note that p16^Ink4a and p19^Arf were not detected in E14.5 telecephalon (C) or P0 brain (D) from either transgenic or control mice, as expected. E). Representative immunohistochemical staining of Bmi-1 on coronal sections of E15 wild type and line B transgenic telencephalon. Bmi-1 staining was nuclear, LV is lateral ventricle.

Figure 3. Bmi-1 over-expression increased the self-renewal and neuronal differentiation of neural stem/progenitor cells in culture

A) A significantly higher percentage of freshly isolated SVZ cells from 2-3 month old transgenic mice gave rise to multipotent neurospheres than littermate control cells. (*, p<0.05); however, the total number of multipotent neurospheres formed per transgenic SVZ (from both brain hemispheres) was not significantly different from control SVZ (B, p=0.31 for line B and p=0.55 for line C). Transgenic neurospheres were significantly larger than control neurospheres (C; *, p<0.01) and gave rise to significantly more multipotent secondary neurospheres upon subcloning (D; *, p<0.05). E, F) Single CD15⁺CD24-SVZ cells from P5 wild type and transgenic mice were sorted per well of 96 well plates. The transgenic single cell-derived neurospheres were larger than controls (E; *, p<0.01) and generated significantly more multipotent secondary neurospheres upon subcloning (F; *, p<0.01). G) Adherent Bmi-1 transgenic neural stem cell colonies showed a significantly higher frequency of BrdU+ cells as compared to control colonies after a 1 hour pulse with BrdU (*, p<0.01). H, I) Transgenic neurospheres were capable of multilineage differentiation and gave rise to larger numbers of neurons as compared to control colonies. Significantly more transgenic colonies contained large numbers of neurons (>100) as compared to control colonies (p<0.01 for line B and p<0.05 for line C). Data represent mean±s.d. from 3-5 independent experiments.

Figure 4. Bmi-1 over-expression had little effect on fetal or adult neural stem/progenitor cell proliferation in vivo

A) The rate of BrdU incorporation by SVZ cells was comparable between transgenic and control mice after a 2 hour pulse of BrdU (p=0.31 for line B and p=0.08 for line C; data represent mean±s.d. from 4 independent experiments). B) Representative confocal images showing BrdU⁺ cell distribution in wild type and transgenic E14.5 telencephalon after a 15 minute pulse of BrdU. No differences in the gross distribution or frequency of BrdU⁺ cells were noted between wild type and transgenic embryos. C). No differences in the thickness of the cerebral cortex or VZ (as indicated in B) were found between transgenic and control embryos. Numbers in parenthesis indicate VZ thickness as a fraction of total cortex thickness.

Figure 5. Bmi-1 over-expression had little effect on fetal or adult neurogenesis

A) One month old transgenic and control mice were pulsed with BrdU for one week. Four weeks later, the frequency of BrdU⁺NeuN⁺ neurons in the olfactory bulb was significantly reduced among line B transgenics (p=0.02) but was not affected in line C (p=0.74; data represent mean±s.d. from 3-4 independent experiments). B-F) Representative images from line C transgenic mice show normal cortical neurogenesis in Bmi-1 transgenic mice. B) At E12, the frequency and localization of Tuj1⁺ neurons in the cerebral cortex was similar between transgenic and control littermates. C, D) Cortical layer organization appeared to be normal in the postnatal transgenic mouse brain. Reelin labels Cajal-Retzius neurons in layer I of the neonatal cortex (C) and FoxP2 labels layer VI neurons (D). E, F) Bmi-1 transgenics and littermate controls were pulsed with a single injection of CldU at E17 and analyzed at P20: no difference was observed in the frequency or localization of CldU⁺ neurons. Cux-1 labels upper layer (II-IV) neurons. G) High magnification confocal images of the same tissue sections as shown in (E,F). Cux-1 staining is shown in red while CldU staining is shown in green (double stained cells are yellow). The frequency of CldU⁺Cux-1⁺ neurons as compared to total CldU⁺ cells (H) and as compared to total Cux-1⁺ neurons (I) was normal in transgenic mice.

Figure 6. Bmi-1 over-expression had little effect on gliogenesis in vivo

A-E) We did not detect any effect of Bmi-1 over-expression on postnatal gliogenesis in line C transgenic mice, as indicated by the normal number and distribution of MBP⁺ oligodendrocytes (A) and GFAP⁺ astrocytes (B) in the adult transgenic brain as well as the normal timing of early postnatal gliogenesis (C-E). F,G) While no difference was observed in the frequency of GFAP⁺ astrocytes in P4 line B wild type and transgenic mouse brains (F); the frequency of MBP⁺ oligodendrocytes was transiently reduced as compared to littermate controls at P4 (G). As with line C, adult line B transgenic mice exhibited no difference relative to control mice in terms of the frequency of GFAP+ astrocytes or MBP+ oligodendrocytes (data not shown).

Figure 7. Bmi-1 over-expression increased proliferation, self-renewal, and neurogenesis by wild-type but not Ink4a-Arf deficient neural stem/progenitor cells in culture

Wild type and Ink4a-Arf deficient neural stem cells were infected with either a Bmi1-encoding retrovirus (MSCV-Bmi1-IRES-GFP) or control retrovirus (MSCV-GFP). A) qRT-PCR confirmed a 4 to 5.5-fold increase in Bmi-1 expression in MSCV-Bmi-1 infected neurospheres as compared to neurospheres infected with control virus. B) Wild type neurospheres infected with MSCV-Bmi1 were significantly larger than those infected with control virus or uninfected neurospheres (p<0.01). In contrast, MSCV-Bmi1 infection did not affect the size of Ink4a-Arf deficient neurospheres (p=0.57). Data represent mean±s.d. from 3-4 independent experiments. C) Wild type neurospheres infected with MSCV-Bmi1 virus exhibited significantly increased self-renewal as compared to neurospheres infected with control virus or uninfected neurospheres (p<0.05). In contrast, MSCV-Bmi1 infection did not affect the self-renewal of Ink4a-Arf deficient neurospheres (p=0.45). Data represent mean±s.d. for 3-5 independent experiments. D). MSCVBmi1 infection increased neuronal differentiation within wild-type (*, p<0.05 relative to WT no virus) but not Ink4a-Arf deficient neural stem cell colonies. Data represent mean±s.d. from 3-4 independent experiments. E-I) Bmi-1 transgenic mice (Bmi-1^TG+) were also mated with Ink4a-Arf deficient mice. E) Freshly dissociated SVZ cells showed that Bmi-1^TG+Ink4a-Arf^−/− mice expressed the transgene in vivo (green filled histogram), while Ink4a-Arf^−/− mice (black open histogram) did not. F). qRT-PCR analysis showed that Bmi-1 expression was increased in the SVZ of Bmi-1^TG+Ink4a-Arf^−/− mice. G). The frequency of multipotent neurospheres that arose in culture from Bmi-1^TG+Ink4a-Arf^−/− SVZ cells was not significantly different from Ink4a-Arf^−/− SVZ. H,I). Bmi-1^TG+Ink4a-Arf^−/− neurospheres did not show increased diameter or self-renewal potential as compared with Ink4a-Arf^−/− neurospheres.

Figure 8. A minority of Bmi-1 transgenic mice developed idiopathic hydrocephalus but showed no sign of brain tumors

A) Representative nissl stains of P5 line C and littermate control brain sections showed enlarged lateral ventricles (LV) in the transgenic brain but a normal laminar organization of the cortex. B) MRI images showed hydrocephalus in adult line C mice: lateral ventricles were enlarged, while the third and fourth ventricles as well as the sylvius aqueduct were relatively normal. C) No physical block was detected between the lateral and third ventricles in H&E stained serial brain sections. D) The ependymal cell layer (S100ß⁺ cells) in the transgenic mouse brain was abnormally folded in places. E) Bmi-1 over-expressing mice did not develop brain tumors, even when aged for approximately two years. Brains in the “gross examination” column are in addition to those examined by histology. Mice that did not develop hydrocephalus showed no sign of premature death relative to control mice.