Neural Stem Cells Adopt Tumorigenic Properties by Constitutively Activated NF-κB and Subsequent VEGF Up-Regulation

Abstract

One of the challenges in stem cell research is to avoid transformation during cultivation. We studied high passage subventricular zone derived neural stem cells (NSCs) cultures of adult rats in the absence of growth factors epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF). We termed this culture exogenous growth factor independent neural stem cells (GiNSCs). GiNSCs expressed stemness markers, displayed a high constitutive NF-κB activity and an increased, aberrant, polyploid DNA content. GiNSCs showed a tumorigenic phenotype and formed colonies in a soft agar assay. Microarray analysis showed the up-regulation of the NF-κB target gene vascular endothelial growth factor (VEGF). In contrast, proneuronal genes were down-regulated. Under neuronal differentiation conditions GiNSCs adopted a glioma-like phenotype, with nuclear p53, preserving high amounts of Nestin positive cells and prolonged proliferation. Neutralization of VEGF strongly inhibited proliferation and induced differentiation. In a gain of function approach, the transfection of NSCs with constitutively active upstream kinase IKK-2 led to constitutively activated NF-κB, proliferation in absence of growth factors and augmented VEGF secretion.

In a rescue experiment a reduction of NF-κB activity by overexpression of IκB-AA1 was able to shift the morphology toward an elongated cell form, increased cell death, and decreased proliferation. Thus GiNSCs may provide a potent tool in cancer research, as their exogenous cytokine independent proliferation and their constitutively high NF-κB expression presumes cancerous properties observed in gliomas. In addition, this study might add a novel mechanism for detecting oncogenic transformation in therapeutic stem cell cultures.

Introduction

S tem cells are immature cells bearing a prolonged self-renewal capacity and, with regard to their origin, the ability to differentiate into multiple or even all cell types of an organism. Within the mammalian central nervous system (CNS) distinct neurons and glia arise from seemingly indistinguishable populations of neural stem cells (NSCs) to make up the diversity of cell types of the mature nervous system.

Such NSCs have been isolated from several regions of the mammalian brain, including the dentate gyrus [1], subcortical white matter [2], and the subventricular zone (SVZ) with the latter one considered being the largest of these germinal centers [3].

Characterization of NSCs in vitro is assessed by the expression of the intermediate filament Nestin [4], the RNA-binding Protein Musashi-1 [3,5], and the transcription factors Sox1 [6] and Sox2 [7].

Many experimental studies have implicated the NSCs residing in the SVZ as a source of oncogenicity and tumor formation [8,9] in vivo. In particular there is compelling evidence that gliomas arise from NSCs that have undergone malignant transformation [10].

Cell proliferation and cell numbers within the developing and adult CNS are tightly controlled under physiological conditions [11,12].

In a previous study, we were able to show that tumor necrosis factor (TNF) not only triggers the aggregation of 3D neurophere NSC cultures but also the proliferation of NSCs in vitro [13,14]. The latter one proved the known transduction of TNF to the IKK-α/β-complex [15] to apply to NSCs as well. In both approaches we found TNF signaling via the NF-κB pathway to have huge impacts.

NF-κB proteins in mammals are dimers assembled from a set of at least five different subunits termed p50, p52, p65/RelA, RelB, and c-Rel with p65/p50 dimers being the most frequent ones within the CNS [16]. Normally, NF-κB proteins are retained in the cytoplasm by a group of inhibitory proteins known as IκBs keeping them in an inactive state [17]. This trimeric NFκB–IκB complex can be activated by various stimuli such as cytokines, neurotransmitters, mitogens, and growth factors [17,18].

NF-κB transcription factors are crucially important in a wide variety of biological processes such as immune and inflammatory responses [19], neuroprotection and degeneration [17], and pathological tumor malignancies [15,20]. Inflammatory responses in turn are a prerequisite for immune defense and angiogenetic processes. As cancer frequently is associated with maladaptive chronic inflammation, it is interesting to note that TNF is described to strongly activate NF-κB in a row of experimental cancer models [15]. Hence, the role of NF-κB in cancer development and progression has been discussed very extensively [20 –23].

NSCs are hoped to be capable of replacing lost cells within the CNS because of their therapeutic potential in neurodegenerative disorders [24,25] and cancers [26]. Yet, as one of the crucial features of NSCs is the ability to proliferate, they eventually can accumulate genetic aberrations leading to tumorigenesis.

Materials and Methods

Isolation and culture of adult NSCs

Tissue was extracted from Wistar rats according to local guidelines (Bezirksregierung Detmold). NSCs from the lateral SVZ of 3–5 adult rats/preparation were isolated and cultivated as described in reference [13].

Isolation of GiNSCs

Subcultured secondary NSCs, primordially isolated from the SVZ of adult rats, were maintained in culture for 92 passages. After dissociation with Accutase cells were seeded on poly-d-lysine coated plates at a density of 5.0 × 10⁵ cells/3.8 cm². Incubation in medium without epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) was to initiate neuronal differentiation for 5 days. A few cells did not adhere to the provided substrate and instead remained in suspension. Separately cultured without EGF and bFGF they began to form neurospheres. The so obtained growth factor independent neural stem cells (GiNSCs) were cultured for at least 20 passages. To avoid effects of clonally derived cell lines, at least three heterogeneous populations were tested in each assay.

Semiquantitative RT-PCR

A total of 4.0 × 10⁶ cells of NSCs or GiNSCs, 50 μg of cortex or liver tissue were harvested per sample and total RNA was isolated with RNeasy (Qiagen, Valencia, CA) according to the manufacturer's instructions. cDNA was subsequently synthesized using SuperScript III First-Strand Synthesis System for reverse transcription-polymerase chain reaction (RT-PCR; Invitrogen, Carlsbad, CA) following manufacturer's guidelines.

Microarrays

Microarray analysis, data processing and handling were performed as described by Schwamborn et al.[27].

Immunocytochemistry

Neurospheres were harvested, fixed, and stained as described in reference [13]. The following antibodies were used: Nestin 1:100 (Milipore, Billerica, MA), Musashi 1:100 (Chemicon, Temecula, CA), Sox2 1:150 (Chemicon), Notch2 1:100 (Santa Cruz, Santa Cruz, CA), PSA-NCAM 1:100 (Chemicon) GFAP 1:100 (Chemicon), β-III-tubulin 1:100 (Promega, Madison, WI), Gad67 1:100 (Chemicon), Neurofilament 1:100 (Chemicon), PCNA 1:100 (Chemicon), p53 1:100 (Santa Cruz), and p65 1:100 (Chemicon).

Proliferation assays

The determination of proliferation of NSCs and GiNSCs spheres was performed as described in reference [13]. NSCs and GiNSCs conducted to neuronal or glial differentiation were seeded in an initial density of 1.0 × 10⁵ cells and cell numbers were determined by a trypan blue exclusion assay daily.

Alternatively, NSCs and GiNSCs induced to neuronal or glial differentiation for 5 or 4 days, respectively, were incubated with a bromodeoxyuridine (BrdU) solution of the BrdU Labeling and Detection Kit (Boehringer Mannheim, Mannheim, Germany) for 2 h, fixed and stained for BrdU incorporation according to the manufacturer's instructions. Differences in proliferation between NSC and GiNSC were assessed by 2-way analysis of variance (ANOVA) followed by post hoc t-test with Bonferroni correction. P ≤ 0.05 was considered as significant.

Cell differentiation

To analyze the neuronal and glial differentiation potential of NSCs and GiNSCs, cells were differentiated as described earlier [13].

Differences in marker expressions between differentiated NSC and GiNSC were determined by 2-way ANOVA followed by post hoc t-test with Bonferroni correction. P ≤ 0.05 was considered significant.

Reporter gene assay and rescue approach (transfection and analysis)

Neurospheres were dissociated as described above and transfected using a Rat NSC Nucleofector Kit (Lonza, Cologne). To detect NF-κB activity in NSCs, a Dual-Luciferase Reporter Assay System (Promega, Madison, WI) and κB-luc Reporter (BD Clonetech, Mountain View, CA) were used as described in reference [13]. In addition the IκB-AA1 [28] or caIKK2 (a constitutively active IKK2), cloned into the pLenti4/V5-DEST expression vector (Invitrogen, Carlsbad, CA) were cotransfected. A representative experiment is shown.

For a phenotype rescue approach GiNSCs were conducted to neuronal differentiation. Upon differentiation, the cells were either transfected with 2 μg of the IκB-AA1 or the mock vector pMETalpha. All samples were additionally transfected with 1 μg of the pmaxGFP (Lonza). After 5 days of growth, the cultures were processed for immunocy-tochemistry and proliferation assays. Differences in marker expressions and proliferation between NSC and GiNSC were determined by 2-way ANOVA followed by post hoc t-test with Bonferroni correction. P ≤ 0.05 was considered significant.

Cell cycle analysis using FACS

Nuclei were isolated by hypotonic lysis. 5 × 10⁵ cells were harvested and plasma membranes lysed for 30 min by a hypotonic lysis solution (0.1 % w/v Sodium citrate, 0.1% v/v Triton X-100). Nuclei were stained with 0.1% v/v propidium iodide and cell aggregates were removed by filtration through a 40 μm mesh. Nuclear staining was analyzed on a FACSCalibur flow cytometer (Becton Dickinson, Germany) and data processed with the WinMDI 2.8 software (J. Trotter, USA). Differences in DNA content between control NSCs and GiNSCs were assessed by 2-way ANOVA followed by post hoc t-test with Bonferroni correction. P ≤ 0.05 was considered significant.

Preparation of rat metaphase chromosomes

A number of each 1 × 10⁷ cells of NSCs and GiNSCs grown in tissue flasks were incubated for 2 h at 37°C in medium containing colcemide (0.05 μm/mL) to arrest cells in mitotic metaphase. After treatment with prewarmed hypotonic solution (0.075M KCl) for 15 min at 37°C the cells were progressively fixed three times in methanol/acetate (3:1). The suspension was dropped from a height of 40 cm onto cleaned wet coverslips and air dried. Chromosomes were stained with DAPI (50 ng/mL) and mounted with moviol (Hoechst).

Western blot analysis

GiNSCs, NSCs, and NSCs stably transfected with caIKK2 were cultured for 48 h (for transfected NSCs) or 72 h without media change. After centrifugation at 200g the supernatants (3 mL) were collected and concentrated to 100 μL using Amicon 10 K centrifuge spin column (Millipore, Temecula, CA). The blotting was performed according to Widera et al. [13] using polyclonal antibody against vascular endothelial growth factor (VEGF; R&D Systems, 0.2 μg/mL) or FLAG-tag (Sigma-Aldrich, St. Louis, MO; 1:5000) with gentle agitation.

Neutralization of VEGF and determination of proliferation

GiNSCs were cultivated as described above with supplementation of IgG-control or 0.1,0.5, and 1 μg/mL of neutralizing α-rat VEGF antibody (R&D Systems, Minneapolis, MN). For BrdU incorporation analysis 10 μM BrdU was added every 24 h. BrdU incorporation was analyzed by immunocytochemistry. Antibody staining was visualized using confocal laser scanning microscopy (LSM 510l; Zeiss, Germany) and analyzed using ZEN software (Zeiss, Germany).

Measurement of tumorigenicity using soft agar assay

Anchorage independent cell proliferation was investigated by a soft agar assay. Top agarose was prepared from Agarose (Biozym) in Dulbecco's modified Eagle's medium (DMEM)/F12 substituted with B27 supplement. Cells were seeded at a density of 3 × 10³ cells per cell culture well in 0.35% top agarose and cultured for 14 days on 0.5% base agar (Difco Agar Noble; Becton Dickinson, Franklin Lakes, NJ) at 37°C. GiNSCs and NSCs were cultured in absence of EGF and bFGF. As control NSCs with EGF/bFGF were used. The medium was changed every 3–4 days. When present, EGF and bFGF was added to both the soft agar and the medium at 20 ng/mL each. After 14 days of culture, cells were stained with 0.05% crystal violet for 2 h at 4°C. Colonies were counted in the entire well per condition under a dissecting microscope.

Generation of GiNSCs via stable transfection with caIKK2

NSCs (passage 14) were transfected using Amaxa (see above) with vector carrying caIKK2 [29], puromycin resistance and FLAG-tag or with a vector containing puromycin resistance and FLAG-tag alone as control. Transfected cells were cultured in medium with EGF and bFGF for 24 h followed by treatment with 1.5 μg/mL puromycin. Puromycin containing medium was changed every 24 h. After 72 h of culture in presence of puromycin, the medium was replaced by standard EGF/bFGF containing medium followed by culture for 72 h at 37°C. Thereafter the growth factors were deprived and the cells (1 × 10⁶/condition) were cultivated for additional 48 h followed by analysis of cell morphology, cell number, western blot or transfection for reporter gene assay (see above). In a control approach EGF and bFGF were left in the medium. In all approaches successful transfection was verified via western blot against the FLAG-tag.

Results

Characterization of exogenous Growth factor independent Neural Stem Cells (GiNSCs)

SVZ derived NSCs were kept as in vitro culture for 92 passages. These NSCs were exposed to differentiation promoting conditions in the absence of EGF and bFGF. We noticed a few cells remaining in suspension. Interestingly, these cells did not adhere to the substrate as the majority of cells and they appeared vital. To analyze this population of differentiation nonresponding cells, we harvested them and kept them in culture in the absence of growth factors (Fig. 1A). To our surprise, the cells were vital and, moreover, within 14 days began to proliferate and to form neurospheres. As these cells were able to form neurosphere like clusters in the absence of EGF/bFGF supplementation we termed them GiNSCs. GiNSCs were cultured for at least 20 passages.

FIG. 1.

A neural stem cell (NSC) population independent of growth factor supplementation (GiNSCs). (A) Subcultured NSCs at passage 92 were prompted to undergo neuronal differentiation. A subset of cells did not undergo differentiation, but survived remaining in the supernatant. These cells were harvested and cultured without growth factors. Within the following 10 days the cells began to form neurospheres. (B) Semiquantitative RT-PCR analysis of gene expression patterns. GiNSCs show an expression pattern similar to NSCs. Stemness marker transcripts were expressed in GiNSCs and NSCs at similar levels. (C) Immunocytochemical analysis expression of NSC markers Nestin (a), Musashi (b), Sox2 (c), and Notch2 (d). (D) Quantification of stemness marker expression yielded a pattern normally observed in NSCs in vitro. Bar, 100 μm (E) GiNSCs are able to proliferate without exogenously added growth factors. Proliferation capacity of GiNSCs was screened by a trypan blue exclusion assay over a period of 4 days. Initially seeded 1 × 10⁴ GiNSC cells per sample had a doubling time of about 28 h. Additional supplementation of incubation medium with growth-factors epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF) did not lead to increased cell numbers. ***P < 0.001.

Expression pattern of GiNSCs was studied by semiquantitative RT-PCR (Fig. 1B). NSC markers Sox-2, Musashi, Nestin [3], and Notch [30] were amplified. To verify PCR-analysis immunocytochemistry was performed (Fig. 1C). Most of the GiNSCs were Nestin immunoreactive, and featured high levels of Musashi, essential for neurosphere formation and proliferation [3]. The NSC-specific transcription factor SOX2 [31] and Notch2, a marker present in proliferating neural cells [30] were detected as well (Fig. 1C and 1D). Moreover, we found expression of PSA-NCAM, a marker for migrating neuronal precursor cells. The differentiation markers GFAP and β-III-tubulin (data not shown) were not expressed (Fig. 1C).

GiNSCs proliferate without growth factors

To analyze, if a growth factor supplementation with EGF/bFGF still had an effect on proliferation, samples in presence or absence of growth factors were prepared (Fig. 1E). Initial cell number of 1 × 10⁴ increased to over 1 × 10⁵ within 96 h (average doubling time of 28.9 h). In comparison, cultivation of GiNSCs in presence of EGF and bFGF, though yielding a more considerable increase during the first 24 h, did not lead to a higher total cell number after 96 h. Without EGF and bFGF addition population doubling time within the first 24 h was 18.16 h.

GiNSCs display a glioma phenotype upon differentiation

In neuronal differentiation attempts morphology of GiNSCs differed notable from that of the control NSCs (Fig. 2A). After 5 days GiNSCs appeared compact, with glial-cell like protrusions positive for Nestin, in contrast to the typical neuronal morphology. Moreover, expression patterns differed significantly between control NSCs and GiNSCs, as the latter ones were highly positive for Nestin (Fig. 2B). Moreover, GiNSCs displayed a higher percentage of GFAP positive cells and strikingly less numbers of β-III-tubulin expressing cells. Although NF-M [32,33] expression was at similar levels, GiNSCs had slightly less GAD67 [34] positive cells than NSCs. In addition, a tremendous morphological difference to NSC cells was apparent (Fig. 2A, a–i).

FIG. 2.

Growth factor independent neural stem cells (GiNSCs) feature an altered phenotype when prompted to differentiate. (A) GiNSCs and neural stem cells (NSCs) prompted to neuronal differentiation. After 5 days detection of the NSC marker Nestin (a, f), glial marker GFAP (b, g), neuronal markers β-III-tubulin (c, h), NF-M (d, i), and Gad67 (e, j) was screened. DNA: green. Bar, 50 μm. (B) Quantified percentage of marker positive cells of NSCs (black bars) and GiNSCs (gray bars) prompted to neuronal differentiation. GiNSCs show a significantly higher percentage of cells positive for Nestin, and significantly less cells positive for β-III-tubulin and Gad67. (C) GiNSCs and NSCs were kept in 10% fetal calf serum (FCS) supplemented medium for 4 days to prime glial differentiation and stained for the expression of stemness and differentiation markers. Antibodies were to visualize the NSC marker Nestin (a, d), glial marker GFAP (b, e) and neuronal marker β-III-tubulin (c, f). DNA: green. Bar, 50 μm. (D) Quantified percentage of marker positive cells of NSCs (black bars) and GiNSCs (gray bars) prompted to glial differentiation. Again, note the significant higher percentage of Nestin positive and a concomitantly lower percentage of β-III-tubulin positive GiNSC derived cells. *P < 0.05, ***P < 0.001.

Conducted to glial differentiation GiNSCs appeared to be very compact in shape and possessed enlarged nuclei (Fig. 2C). The glial like phenotype differed from the appearance of the NSCs controls. GiNSCs displayed the same drift in Nestin expression as they did in the neuronal differentiation (Fig. 2D).

GiNSCs continue proliferating upon differentiation

During neuronal and glial differentiations we observed the cell density of GiNSCs to increasingly shift in comparison to NSC controls. As cells for all samples were initially seeded in the same density and this phenomenon appeared in all independent differentiation approaches we assumed such a variation to be due to proliferation of GiNSCs.

After 5 days of neuronal differentiation cells were incubated with BrdU, followed by immunocytochemical detection of BrdU incorporation (Fig. 3A). While NSCs controls had a rather discrete to no presence of BrdU, nuclei of GiNSCs showed to take up marked amounts of BrdU. Quantification proved nearly 70% of GiNSCs to be highly BrdU positive, which was in significant contrast to <5% BrdU positive NSCs (Fig. 3A). In contribution a trypan-blue exclusion assay displayed that GiNSC cell numbers doubled within 72 h whereas NSC numbers remained invariable (Fig. 3C).

FIG. 3.

Differentiated growth factor independent neural stem cells (GiNSCs) continue to proliferate. (A) GiNSCs and neural stem cells (NSCs) were prompted to neuronal differentiation as described in Figure 3. After 5 days incorporation of bromodeoxyuridin (BrdU) into newly synthesized DNA was determined (a, b). The corresponding nuclei were counterstained by Sytox green (c, d) and images merged for NSCs (e) and GiNSCs (f) to screen BrdU positive cells. Bar, 50μm. Quantification of BrdU positive cells in NSCs (black bar) and GiNSCs (gray bar) with the latter ones displaying a marked higher rate. (B) Proliferation rate of NSCs and GiNSCs after induction of glial differentiation by detection of BrdU incorporation into NCS (a) and GiNSCs (b) analogous to neuronal differentiation. Sytox green counterstained nuclei (c, d) identify double positive NSCs (e), and GiNSCs (f) in merged images. Bar, 50 μm. Quantification of the rate of BrdU positive cells in NSCs (black bar) and GiNSCs (gray bar) shows a similar higher rate of proliferating GiNSC as in neuronal differentiation. (C) Alternatively, proliferation rate of GiNSCs and NSCs was assessed by increase in cell numbers within the 5 days of neuronal differentiation via trypan blue exclusion assay. 96 hours after differentiation induction GiNSCs had increased their numbers 4-fold. Black bars depict NSCs and gray bars depict GiNSCs. (D) Increase in cell number within the 4 days of glial differentiation as assessed by trypan blue exclusion assay. Black bars depict NSCs and gray bars depict GiNSCs. *P < 0.05, ***P < 0.001.

Analogous to neuronal differentiation, GiNSCs induced to glial differentiation incorporated BrdU as well (Fig. 3B). Similarly a trypan-blue exclusion assay revealed a tripled GiNSC cell number within 72 h, hence being significantly more cells compared to NSCs (Fig. 3D).

GiNSCs form colonies in soft agar assay

Anchorage independent proliferation was tested using soft agar assay (Fig. 4). After 14 days of culture GiNSCs formed 214 ± 37 colonies in medium without EGF/bFGF. In contrast NSCs control cultures in presence of growth factors formed no colonies at all, although single cells remained viable and showed signs of sprouting (see Fig. 4). In a control approach with NSCs cultivated without EGF and bFGF no viable cells could be detected.

FIG. 4.

Measurement of tumorigenicity using soft agar assay. Cells were seeded at a density of 3 × 10³ cells per cell culture well in 0.35% soft agar and cultivated for 14 days at 37°C. Growth factor independent neural stem cells (GiNSCs) and neural stem cells (NSCs) were cultured in absence of epidermal growth factor (EGF) and basic fibroblast growth factor (bFGF). As control NSCs with EGF/bFGF were used. If present, EGF and bFGF were added to both the soft agar and the medium at 20 ng/mL each. After staining with crystal violet, the number of colonies has been determined. GiNSCs formed 214+ colonies, whereas NSCs with growth factors show no colony formation but sprouting of single cells (see blow-up in the middle panel). In a control approach using NSCs without growth factors, no viable cells have been detected.

GiNSCs accumulate nuclear p53 and have a constitutively high NF-κB activity

A very common feature among tumor cell lines as well as tissues obtained from primary cancers is the nuclear accumulation of the tumor suppressor protein p53 accompanied or caused by its loss of function. For this reason we analyzed GiNSCs for nuclear accumulation of p53 during neuronal differentiation (Fig. 5A). In GiNSCs p53 displayed huge nuclear accumulation, whereas the NSC control had rather low amounts of p53 (Fig. 5A, a and b), predominantly located in the perinuclear space, as assessed by nuclear counterstaining (Fig. 5A, c–f). Nuclear p53 was found in almost 45% of all GiNSC cells, contrasted by <5% in NSC controls (Fig. 5B).

FIG. 5.

Growth factor independent neural stem cells (GiNSCs) accumulate nuclear p53 and exert a high constitutive NF-κB activity. (A) Neuronal differentiation of neural stem cells (NSCs) and GiNSCs was conducted as stated before and cells fixed 5 days postadhesion. Immunocytochemical detection of tumor suppressor protein p53 expression in NSCs (a) is predominantly observed within the perinuclear space, while GiNSCs (b) feature nuclear localization of p53. Nuclei were counterstained with Sytox green (c, d) and images merged to verify putative nuclear localization in NSCs (e) and GiNSCS (f). p53/Sytox (DNA) double positive cells are marked by arrows. Bar, 50 μm. (B) Statistical analysis of the percentage of p53 positive nuclei in NSCs (black bar) and GiNSCs (gray bar) depicts a vast nuclear accumulation in the latter ones. (C) GiNSCs possess a constitutively high NF-κB activity as measured by reporter gene activity. To verify measured NF-κB dependent Luciferase activity, NSCs and GiNSCs were either cotransfected with a transdominant negative NF-κB super repressor termed IκB-AA1 or a constitutively active kinase caIKK2. Basal NF-κB activities were analyzed with a mock control. Note the significantly higher basal NF-κB activity in GiNSCs compared to sNSCs. *P < 0.05, **P < 0.001, ***P < 0.001.

When eliciting regulation downstream of p53, it is noteworthy, that activated p53 was reported to lead to activated NF-κB in Saos-2 tumor cells [35]. Therefore, we wondered whether the independence of GiNSCs from growth factor supplementation might correlate with an altered NF-κB expression.

Investigation of NF-κB signal transduction was conducted by reporter gene analysis. Cotransfection of the NF-κB reporter plasmid and IκB-AA1 significantly decreased the activity measured, whereas the constitutively NF-κB activating caIKK2 significantly increased the κB-dependent bioluminescence. Notably, basal NF-κB activity in mock-transfected control cultures was significantly higher in GiNSCs than in mock transfected NSCs, indicating a high amount of constitutive NF-κB activity in GiNSCs.

GiNSCs harbor an aberrant DNA content

We analyzed the cell cycle of GiNSCs via FACS to quantify major differences to NSCs in the degree of the polyploidy. Normally, a stem cell population such as NSCs displays 3 subpopulations (Fig. 6A): A peak of cells with a single chromosome set (1n) that either are arrested in G0 state or have undergone cytokinesis currently being in replication state (S-phase) and another peak of cells that already have doubled their DNA (G2/M). In contrast to this GiNSCs additionally had a dispersed population of polyploid cells. In detail GiNSCs had a significantly higher percentage of cells being in G2/M phase and a significant number of cells possessing more than 2 chromosome sets (polyploidy, Fig. 6B).

FIG. 6.

(A) Growth factor independent neural stem cells (GiNSCs) harbor an aberrant DNA amount. For cell cycle analysis via flow cytometry 5 × 10⁵ undifferentiated cells per sample were harvested, plasma membranes lysed by hypotonic solution and cell aggregates and clumps were removed by filtration through a 40 μm mesh. Nuclei were stained with 0.1% propidium iodide. The cell cycle profiles obtained by fluorescence-activated cell sorting (FACS) analysis reveal a higher amount of GiNSCs than neural stem cells (NSCs) being in mitosis phase and the presence of polyploid subpopulations. G1-Phase (G1), S-Phase, Mitosis (G2/M). (B) Statistical analysis of the cell cycle assay significantly prove more GiNSCs to be in mitosis and to bear a polyploid DNA content. Black bars depict sNSCs and gray bars depict GiNSCs. (C) For classical karyotyping of mitotic sNSC and GiNSC metaphases chromosomes were spread on glass slides. Inverted DAPI-stained images of sNSC metaphase (a) compared with GiNSC metaphase (b) verify that GiNSCs harbor a dramatically increased chromosome number. Prospective karyotype of sNSC metaphase (c) and prospective, representative karyotype of GiNSC (d) depict 72 chromosomes in GiNSCs compared to 42 chromosomes in sNSCs. Please note the 4 chromosomes No. 9. *P < 0.05, ***P < 0.001. (D) GiNSCs feature a shift in gene-expression as visualized by microarray analysis. Total-RNA of sNSCs and GiNSCs was processed to cDNA, coupled wit fluorescence dyes Cy3 and Cy5 and hybridized to a cDNA microarray containing 7,200 genes. The array was produced according to the Stanford method and the fluorescence was determined using a GenePix 4000A scanner. For statistical analysis sNSCs gene expression values of candidate genes was set as 100% with GiNSC values set in relation to this baseline. Note the marked down regulation of neuronal genes such as MAP2 and up regulation of the oncogene vascular endothelial growth factor (VEGF) in GiNSCs located at the chromosome 9. (E) Supernatants of cultivated NSCs and GiNSCs were collected, concentrated using Amicon columns and processed to western blot using α-VEGF antibody. Equal loading of cell supernatants is shown by Coomassie staining. Please note the higher amount of VEGF in the supernatant of GiNSCs compared to NSCs.

To further analyze to what extend the DNA amount of GiNSCs had grown, we investigated their chromosome sets via karyotyping metaphases (Fig. 6C). Inverted DAPI-stained images of a NSCs metaphase (Fig. 6C, a) showed the expected 20 chromosome pairs plus 2 gonadosomes, typically of rat cells. Quite strikingly GiNSC metaphases (Fig. 6C, b) verified that GiNSCs harbor a dramatically increased chromosome number (Fig. 6C, a and b): 72 chromosomes and 2 gonadosomes, many of them constituting for more then 2 individual chromosomes each pair, and 4 pleiomorphic chromosomes that were not sortable to any of the 20 chromosome types (Fig. 6C, c and d).

Microarray analysis showed upregulation of VEGF and IRAK2

We visualized shifts in gene-expression patterns by microarray analysis. In particular, we were interested whether multiplication of chromosome pairs could be correlated with multiplications in gene expression.

Microarrays were processed with total RNA of GiNSCs and control NSCs. Targets, involved in a range of cellular tasks such as cytoskeleton assembly, neuronal differentiation and tumor suppression were selected with regard to their corresponding chromosomes, shown to be amplified in GiNSCs. Among those the angiogenesis factor VEGF [36] from chromosome 9 (9q12), with 4 individual chromosomes in GiNSC karyotyping (Fig. 6C, d), was increased in GiNSCs 3-fold. Interestingly VEGF is a NF-κB target gene [37]. Similarly, the downregulation of notch 1 inhibited cancer cell invasion by decreasing NF-κB DNA binding activity and subsequent VEGF downregulation [38]. IRAK2, involved in NF-κB dependent survival signaling [39], though its corresponding chromosome 4 was not found to be amplified, had 3 times higher expression levels than NSCs as well. Other targets, such as the DNA repair protein XRCC5 (9q34) [40], the neural cytoskeleton component MAP2 (9q32) [41] or the cation channel subunit ACCN2 (7q36) [42] were drastically downregulated (Fig. 6D). Candidate gene expression values of NSCs were set as 100% with GiNSC values set in relation to this relative baseline. A number of neuronal genes such as CamK2B, MAP2, ACCN2, and Fyn were found to be downregulated. Since patient derived gliomas rely on VEGF for growth we focused on the analysis of VEGF protein expression. Hence, we screened the VEGF expression in supernatants from cultured GiNSCs in comparison to control NSCs. Western blotting revealed a high amount of secreted VEGF in supernatants prepared from GiNSC and corroborated the microarray results (Fig. 6E).

Stable transfection of NSCs with caIKK2 leads to GiNSC-like phenotype in NSCs

In order to investigate the potential transformation of NSCs by constitutively activated NF-κB, NSCs (passage 14) were stably transfected with a constitutively active form of IKK-2 (caIKK2) carrying vector with additional puromycin resistance and FLAG-tag or puromycin resistance and FLAG-tag alone (Fig. 7A). IKK-2 is the upstream kinase activating NF-κB. The success of the transfection has been verified via western blot against FLAG-tag (data not shown). After 48 h of culture in absence of growth factors no cell proliferation has been observed in mock-transfected cells (Fig. 7C). In addition the morphology of the cells changed to more differentiated phenotype than caIKK2-transfected cells (Fig. 7B). In contrast, caIKK2-transfected cells proliferated moderately after growth factor deprivation. Furthermore, an increased NF-κB activity has been observed in caIKK2 transfected cells compared to mock transfected cells in presence or absence of growth factors (Fig. 7D). Finally, NSCs stably transfected with caIKK2 showed elevated level of the secreted NF-κB target gene VEGF compared to mock-transfected NSCs (Fig. 7E). Taken together, a constitutive activation of NF-κB using caIKK2 led to a GiNSC-like phenotype in NSCs.

FIG. 7.

Stable transfection of neural stem cells (NSCs) with caIKK2 leads to growth factor independent neural stem cell (GiNSC)-like phenotype in NSCs. (A) Schematic diagram of the experimental procedure. (B) Passage 14 NSCs has been stably transfected either with caIKK2 expressing vector or with a mock vector. CaIKK2-trasfected cells were able to growth as neurospheres after growth factor deprivation, whereas mock-transfected NSCs showed signs of differentiation (blow-up). (C) NSCs transfected with caIKK2 proliferate after growth factor deprivation. Control NSCs showed no proliferation. In contrast caIKK2 transfected cells proliferated without epidermal growth factor (EGF)/basic fibroblast growth factor (bFGF). (D) Bioluminescence of a reporter plasmid has been measured and displayed as relative luminescence in relation to mock-transfected NSCs in presence of EGF/bFGF. CaIKK2 transfected cells show significantly elevated NF-κB activity compared to mock transfected cells with and without growth factor supplementation. *P < 0.05. (E) CaIKK2 transfected cells show elevated levels of vascular endothelial growth factor (VEGF) secretion compared to mock transfected cells as demonstrated via western blot analysis. Equal loading of cell supernatants is shown by Coomassie staining.

Neutralization of VEGF results in decreased proliferation of GiNSCs

Cultures of GiNSCs were treated with a neutralizing antibody directed against rat VEGF for 48 and 72 h (Fig. 8A). Differences in neurosphere size and number were evident after 48 h and in addition after 72 h some stem cells started to differentiate (Fig. 8A). Control IgG did not interfere with proliferation (upper panel in Fig. 8A). Cell proliferation decreased in GiNSCs treated with anti-VEGF antibody (Fig. 8B). High power magnification showed nuclear BrdU incorporation under control conditions, which was nearly invisible after anti-VEGF treatment.

FIG. 8.

Neutralization of vascular endothelial growth factor (VEGF) leads to decreased proliferation and induces differentiation of growth factor independent neural stem cells (GiNSCs). (A) Cultivated GiNSCs show signs of differentiation if cultivated in presence of neutralizing antibody against rat VEGF (lower panel) compared to IgG-control treated cells (upper panel). Please note the differentiated cells shown in the blow-up section (right) of the lower panel. (B) GiNSCs incorporate significantly lower amounts of bromodeoxyuridine (BrdU) in presence of a neutralizing antibody against VEGF compared to IgG-control treated GiNSCs indicating decrease in proliferation. (C) Quantification of BrdU-incorporation assay.

Reduced NF-κB activity reverses the tumorigenic phenotype in differentiating GiNSCs

We hypothesized that a reduction of NF-κB activity in GiNSCs might reduce the proliferation rate and induces cell death.

As a sort of rescue approach we introduced IκB-AA1 into GiNSCs previous to induction of neuronal differentiation. An additional approach with a mock vector served as control. 5 days after adhesion of the cells to the provided substrate marker expression was analyzed. Mock transfected GiNSCs had the same compacted cell shape with glial-cell like protrusions as previously observed among untransfected GiNSCs (Fig. 9A, a–c). In contrast IκB-AA1 transfected GiNSCs differed remarkably and instead had an elongated shape (Fig. 9A, d–f). Differentiated cells were screened for their expression profiles of Nestin, GFAP, and β-III-tubulin (Fig. 9A, a3–f3). To assess the influence of NF-κB activity GFP reporter positive cells were analyzed exclusively (Fig. 9A, a2–f2). Approximately 55% of the GFP positive cells of the IκB-AA1 transfected approaches showed an elongated phenotype compared to <10% in Mock transfected approaches (data not shown). IκB-AA1 transfected cells still were positive for Nestin, but to a lesser percentage (Fig. 9B). Quantifications of GFAP showed a reduction of positive cells and a slight increase of neuronal marker β-III-tubulin in IκB-AA1 transfected GiNSCs. Immunodetection of activated NF-κB subunit p65 verified that IκB-AA1 transfected GiNSCs had less activated p65 (Fig. 9C). Quantification of p65 immunodetection revealed IκB-AA1 transfection to half the amount of p65/GFP double positive nuclei (Fig. 9D).

FIG. 9.

A reduction of constitutively high NF-κB activity in growth factor independent neural stem cells (GiNSCs) reverses the tumorigenic phenotype. (A) GiNSCs transfected with the NF-κB super repressor IκB-AA1 or a Mock control vector were exposed to neuronal differentiation promoting conditions. Immunocytochemical staining of Mock (Panels a–c) and IκB-AA1 (Panels d–f) transfected cells for the expression of stemness and differentiation markers was conducted 5 days after differentiation. Detection of the neural stem cell (NSC) marker Nestin (a3, d3), glial marker GFAP (b3, e3), and neuronal marker β-III-tubulin (c3, f3). Cells were cotransfected with a GFP reporter to trace expression of the Mock (a2, b2, c2)-and IκB-AA1 (d2, e2, f2)-constructs. Nuclei were counterstained with DAPI solution (a1–f1). In order to analyze marker expression exclusively in transfected cells, images were merged for Nestin (a4, d4), GFAP (b4, e4), and β-III-tubulin (c4, f4). Bars, 34 μm. (B) Quantified percentage of marker positive cells of Mock (light gray bars)-and IκB-AA1 (dark gray bars)-transfected GiNSCs. Note the decreased percentage of Nestin and GFAP positive cells and the higher amount of β-III-tubulin positive cells in IκB-AA1 super repressor transfected GiNSCs. *P < 0.05. (C) 5 days upon neuronal differentiation immunocytochemical staining for the expression of activated p65 in Mock (e)—and IκB-AA1 (f)—transfected cells was performed. In IκB-AA1 transfected samples a decrease of activated nuclear p65 was observed (f). Detection of p65 expression was to be analyzed only in GFP positive cells of Mock (c)—and IκB-AA1 (d)—approaches. Nuclei were counterstained with DAPI solution (a, b) and images merged to verify putative nuclear localization in Mock (g)—and IκB-AA1 (h)—transfected cells. Bar, 34 μm. (D) Statistical analysis of the percentage of GFP/p65 double-positive nuclei in Mock (light gray bar) and IκB-AA1 (dark gray bar) approaches. Reduced expression of activated NF-κB subunit p65 in IκB-AA1 transfected GiNSCs after induction of neuronal differentiation is displayed. (E) Proliferation rate of IκB-AA1 transfected GiNSCs after induction of neuronal differentiation was assessed by immunocytochemical detection of bromodeoxyuridine (BrdU) incorporation into Mock (a)-and IκB-AA1 (b)-transfected GiNSCs as explained previously. The corresponding DAPI counterstained nuclei (c, d) are displayed and merged images for Mock (e)-and IκB-AA1 (f)-approach show BrdU positive nuclei. Bar, 34 μm. (F) Quantification of the rate of BrdU positive cells in Mock (light gray bar) -and IκB-AA1 (dark gray bar)-transfected cells reveals a significant decrease in latter ones. (G) Increase in cell number within the 5 days of neuronal differentiation as assessed by trypan blue exclusion assay. Light gray bars depict Mock- and dark gray bars depict IκB-AA1-approach. **P < 0.01, ***P < 0.001.

We quantified proliferation by changes among cell numbers via a trypan blue exclusion assay over a period of 5 days and assessed BrdU incorporation via immunocytochemistry 5 days after differentiation (Fig. 9E). BrdU detection proved nearly 70% of Mock transfected GiNSCs to be highly BrdU positive which was in significant contrast to <50% BrdU positive IκB-AA1 transfected GiNSCs (Fig. 9F). During the neuronal differentiation Mock cell numbers doubled within 96 h whereas IκB-AA1 numbers remained more or less invariable, besides an initial fluctuation (Fig. 9F).

Trypan blue assay was used to determine the amount of living cells (Fig. 9G). When mock transfected GiNSCs proliferated rapidly up to 5 × 10,000 cells within 96 h after seeding of 10,000 cells. In contrast GiNSCs transfected with IκB-AA1 did not proliferate.

Discussion

In our present study we introduce a novel population of NSCs derived from SVZ that are independent of EGF and bFGF supplementation. In the absence of these growth factors these GiNSCs grow as neurospheres. Expression of stemness markers was remarkably similar to control NSCs. They showed to be highly proliferative and had a high constitutive NF-κB activity.

The morphological appearance differed remarkable between GiNSCs and NSCs. In particular, they had enlarged and seemingly decompacted nuclei, pointing at an increased DNA content and a higher transcriptional activity, in which they accumulated the tumor suppressor protein p53. Nuclear p53 without inducing cell death was reported to account for a loss of function and emerging tumorigenesis [43]. Using clinical samples it was shown that a nuclear localization of p53 in 10%–100% of analyzed cells was associated with p53 mutations and tumor progression in bladder cancer. [44] Here we suppose that nuclear p53 in GiNSCs is a sign of nonfunctional mutated p53. In this line a recent study showed that loss of p53 function (p53 –/–) is a major factor for the induction of high-grade malignant gliomas [45]. Additionally, it is very interesting that these gliomas express high levels of VEGF as shown here in GiNSCs [45]. Furthermore conditional expression of mutant p53 in mice resulted in an accumulation of p53 in nuclei of NSCs, similar to GiNSCs, and in the formation of malignant gliomas [46].

GiNSCs even continued to proliferate under differentiation conditions. After inducing neuronal differentiation a markedly persistance of glioma markers Nestin and GFAP was detected. In addition, GiNSCs have an aberrant DNA content, strikingly reduced gene expression levels of neuronal genes and a constitutive NF-κB expression. This type of unregulated NF-κB activity was already reported for many tumors (prostate, breast, lung, lymphomas, see ref. [47]) but is novel for NSC lineages. NF-κB was found to have a driving impact on GiNSC proliferation, as a reduction of its transcriptional activity by the superrepressor IκB-AA1 strongly decreased the proliferation and induced differentiation of the cells. Expression levels of Nestin and GFAP thereby were remarkably reduced and proliferation, measured by independent experimental approaches, decreased as well. The impact of NF-κB on this rescue approach was verified by reduced amounts of activated NF-κB. One of the most stringent state-of-art in vitro assays for detecting malignant transformation of cells is the soft agar assay [48]. It has been demonstrated that the degree of cell growth in semisolid medium consistently correlates with neoplastic growth of cells in nude mice [48]. Here we showed that GiNSCs were able to form colonies in soft agar after growth factor deprivation providing a strong evidence for their potential tumorigenicity. In contrast, nontransformed NSCs formed no colonies.

Cancerous alterations leading to tumors are thought to require initial mutations in proliferating cells. As proliferation is proceeding, NSCs would have the opportunity to accumulate genetic mutations leading to cancerous transformation [49]. In this regard ex vivo expansion of NSCs to obtain sufficient amounts of cells for therapeutic aims, bears the thread of becoming a germ centre for cancerous alterations itself. The here characterized GiNSCs derived from a long-term NSC culture underline this idea, which might add a cautious note for those interested in in vitro propagation of stem cells for therapy. Importantly, long-term culture of stem cells is not only transforming stem cells derived from neural origin as shown here but also leading to spontaneous malignant transformations of bone-marrow derived human mesenchymal stem cells (hMSCs) [50].

The ambivalent role of stem cells in tumorigenesis is underlined by a series of studies supporting the hypothesis of cancerous alteration preferentially occurring in stem cells. A study by Hemmati and colleagues described stem cell properties featuring tumorigenic cells from pediatric brain tumors, which formed neurosphere clusters yielding multipotent and self-renewing cells with the ability to differentiate into neuronal and glial lineages [51].

Taken together optimization of ex vivo culture paradigms is an urgent need for the development of clinical cell transplantation protocols.

Here we provide evidence, that long-term culture has genetically transformed NSCs in a “glioma-like” phenotype. These growth factor-independent GiNSCs have the following cancerous properties: (1) Exogenous growth factor independence; (2) Stemness marker expression as found in human gliomas: Nestin, Musashi, Sox-2; (3) Polyploidity; (4) Nuclear localization of the anti-oncogene p53, without resulting cell death; (5) Transcriptional changes directed by activated NF-κB including upregulation of its target gene VEGF; and (6) Proliferation, dependent on autocrinously produced VEGF.

A recent study linked the genesis of glioma development with aneuploidity and p53 inactivation. They report a clear difference between gliomas and colorectal carcinoma which are diploid or near-diploid [52]. To our knowledge this is the first study reporting on NSC population with genome instability, featuring a constitutively high NF-κB activity, and aberrant differentiation and continuing proliferation.

A line of experimental evidence has assessed the role of NF-κB dependent signaling in different types of cancers. In fact, NF-κB is constitutively active in most tumor specimens described to date and not only blocks apoptosis but also mediates tumor cell proliferation and induces resistance to chemotherapeutic agents [20 –23,53]. In this regard the constitutively high NF-κB activity in GiNSCs and its impact on a glial phenotype and excessive proliferation ignoring differentiation signals might provide a link between NF-κB, cancer and stem cells. Experimental approaches targeting a reduction of NF-κB activity in gliomas have proven promising reduction of proliferation and tumor size [54,55]. The cell of origin of a tumor however and the molecular mechanism, still remains largely unknown. Here we suggest that gliomas might arise from a NSC with genome instability. In accordance with this hypothesis some recent reports provide evidences that NSCs are the cellular origin of high grade gliomas [56,57]. Malignant gliomas are the most aggressive tumors of the brain, which can be clinically treated with drugs like Avastatin or Bevacizumab [58], recombinant humanized monoclonal antibodies against the main proangiogenic factor VEGF [59]. Interestingly, in GiNSC cultures as well, neutralization of VEGF strongly inhibited proliferation and induced differentiation. Interestingly, a constitutive activation of the NF-κB pathway in NSCs using caIKK2 [29] led to increased cell proliferation, growth factor independence and up-regulated VEGF (Fig. 7).

More recently Samuel Weiss and colleagues demonstrated that subpopulations of human glioblastoma stem cells are able to proliferate independently of exogenous mitogens in part through EGFR signaling [60]. Importantly it has been shown that EGF–EGFR interaction is mediated via activation of NF-κB in breast cancer cells [61] (reviewed in ref. [62]).

Thus, we investigated the influence of the NF-κB pathway in U373 human astroglioma cells cultivated as tumorispheres (Supplementary Results). Here we showed that IκB-AA1 transfected glioma cells formed significantly smaller tumorispheres than mock-transfected cells (Supplementary Fig. 1; Supplementary materials are available online at www.liebertonline.com/scd). In addition the blockade of NF-κB resulted in enhanced cell death and signs of differentiation (Supplementary Fig. 1A). Taken together the blockade of NF-κB in giNSCs and human U373 cells resulted in decrease in proliferation and a phenotypic switch.

These data suggest that glioma therapy might be feasible by inhibiting the NF-κB pathway.

Footnotes

Acknowledgments

This study was supported in part by the Deutsche Forschungsgemeinschaft (DFG). The Anti-PSA antibody was obtained from Dr. Christoph Piechaczek, Milteny Biotec, Bergisch Gladbach, Germany. The excellent technical help of Angela Krahlemann-Koehler is gratefully acknowledged.

Author Disclosure Statement

No competing financial interests exist.

References

Eriksson

, Perfilieva

, Björk-Eriksson

, Alborn

, Nordborg

, Peterson

, Gage

. 1998. Neurogenesis in the adult human hippocampus. Nat Med, 4:1313–1317.

Nunes

, Roy

, Keyoung

, Goodman

, McKhann

II , Jiang

, Kang

, Nedergaard

, Goldman

. 2003. Identification and isolation of multipotential neural progenitor cells from the subcortical white matter of the adult human brain. Nat Med, 9:439–447.

Sakakibara

, Nakamura

, Yoshida

, Shibata

, Koike

, Takano

, Ueda

, Uchiyama

, Noda

, Okano

. 2002. RNA-binding protein Musashi family: roles for CNS stem cells and a subpopulation of ependymal cells revealed by targeted disruption and antisense ablation. Proc Natl Acad Sci USA, 99:15194–15199.

Lendahl

, Zimmerman

, McKay

. 1990. CNS stem cells express a new class of intermediate filament protein. Cell, 60:585–595.

Sakakibara

, Imai

, Hamaguchi

, Okabe

, Aruga

, Nakajima

, Yasutomi

, Nagata

, Kurihara

, Uesugi

, Miyata

, Ogawa

, Mikoshiba

, Okano

. 1996. Mouse-Musashi-1, a neural RNA-binding protein highly enriched in the mammalian CNS stem cell. Dev Biol, 176:230–242.

Pevny

, Sockanathan

, Placzek

, Lovell-Badge

. 1998. A role for SOX1 in neural determination. Development, 125:1967–1978.

Brazel

, Limke

, Osborne

, Miura

, Cai

, Pevny

, Rao

. 2005. Sox2 expression defines a heterogeneous population of neurosphere-forming cells in the adult murine brain. Aging Cell, 4:197–207.

Sanai

, Alvarez-Buylla

, Berger

. 2005. Neural stem cells and the origin of gliomas. N Engl J Med, 353:811–822.

Uchida

, Mukai

, Okano

, Kawase

. 2004. Possible oncogenicity of subventricular zone neural stem cells: case report. Neurosurgery, 55:977–978.

10.

Barami

. 2007. Biology of the subventricular zone in relation to gliomagenesis. J Clin Neurosci, 14:1143–1149.

11.

Widera

, Kaus

, Kaltschmidt

. 2008. Neural stem cells, inflammation and NF-kappaB: basic principle of maintenance and repair or origin of brain tumours? J Cell Mol Med, 12:459–470.

12.

Zhong

. 2003. Diversifying neural cells through order of birth and asymmetry of division. Neuron, 37:11–14.

13.

Widera

, Mikenberg

, Elvers

, Kaltschmidt

. 2006. Tumor necrosis factor alpha triggers proliferation of adult neural stem cells via IKK/NF-kappaB signaling. BMC Neurosci, 7:64.

14.

Widera

, Mikenberg

, Kaus

, Kaltschmidt

. 2006. Nuclear Factor-kappaB controls the reaggregation of 3D neurosphere cultures in vitro. Eur Cell Mater, 11:76–84discussion 85.

15.

Karin

, Yamamoto

, Wang

. 2004. The IKK NF-kappa B system: a treasure trove for drug development. Nat Rev Drug Discov, 3:17–26.

16.

Widera

, Mikenberg

, Kaltschmidt

. 2006. Potential role of NF-kappaB in adult neural stem cells: the underrated steersman? Int J Dev Neurosci, 24:91–102.

17.

Kaltschmidt

, Widera

, Kaltschmidt

. 2005. Signaling via NF-kappaB in the nervous system. Biochim Biophys Acta, 1745:287–299.

18.

Hayden

, Ghosh

. 2008. Shared principles in NF-kappaB signaling. Cell, 132:344–362.

19.

Kumar

, Takada

, Boriek

, Aggarwal

. 2004. Nuclear factor-kappaB: its role in health and disease. J Mol Med, 82:434–448.

20.

Kim

, Hawke

, Baldwin

. 2006. NF-kappaB and IKK as therapeutic targets in cancer. Cell Death Differ, 13:738–747.

21.

Aggarwal

. 2004. Nuclear factor-kappaB: the enemy within. Cancer Cell, 6:203–208.

22.

Nakanishi

, Toi

. 2005. Nuclear factor-kappaB inhibitors as sensitizers to anticancer drugs. Nat Rev Cancer, 5:297–309.

23.

Schwartz

, Hernandez

, Mark Evers

. 1999. The role of NF-kappaB/IkappaB proteins in cancer: implications for novel treatment strategies. Surg Oncol, 8:143–153.

24.

Gage

. 2000. Mammalian neural stem cells. Science, 287:1433–1438.

25.

Lie

, Song

, Colamarino

, Ming

, Gage

. 2004. Neurogenesis in the adult brain: new strategies for central nervous system diseases. Annu Rev Pharmacol Toxicol, 44:399–421.

26.

Shah

, Hsich

, Breakefield

. 2004. Neural precursor cells and their role in neuro-oncology. Dev Neurosci, 26:118–130.

27.

Schwamborn

, Lindecke

, Elvers

, Horejschi

, Kerick

, Rafigh

, Pfeiffer

, Prüllage

, Kaltschmidt

. 2003. Microarray analysis of tumor necrosis factor alpha induced gene expression in U373 human glioblastoma cells. BMC Genomics, 4:46.

28.

Fridmacher

, Kaltschmidt

, Goudeau

, Ndiaye

, Rossi

, Pfeiffer

, Kaltschmidt

, Israël

, Mémet

. 2003. Forebrain-specific neuronal inhibition of nuclear factor-kappaB activity leads to loss of neuroprotection. J Neurosci, 23:9403–9408.

29.

Sasaki

, Derudder

, Hobeika

, Pelanda

, Reth

, Rajewsky

, Schmidt-Supprian

. 2006. Canonical NF-kappaB activity, dispensable for B cell development, replaces BAFF-receptor signals and promotes B cell proliferation upon activation. Immunity, 24:729–739.

30.

Jin

, Zhu

, Sun

, Mao

, Xie

, Greenberg

. 2002. Vascular endothelial growth factor (VEGF) stimulates neurogenesis in vitro and in vivo. Proc Natl Acad Sci USA, 99:11946–11950.

31.

Graham

, Khudyakov

, Ellis

, Pevny

. 2003. SOX2 functions to maintain neural progenitor identity. Neuron, 39:749–765.

32.

Perrin

, Boisset

, Docquier

, Schaad

, Descombes

, Kato

. 2005. No widespread induction of cell death genes occurs in pure motoneurons in an amyotrophic lateral sclerosis mouse model. Hum Mol Genet, 14:3309–3320.

33.

Shea

, Beermann

. 1999. Neuronal intermediate filament protein alpha-internexin facilitates axonal neurite elongation in neuroblastoma cells. Cell Motil Cytoskeleton, 43:322–333.

34.

Lindefors

. 1993. Dopaminergic regulation of glutamic acid decarboxylase mRNA expression and GABA release in the striatum: a review. Prog Neuropsychopharmacol Biol Psychiatry, 17:887–903.

35.

Ryan

, Ernst

, Rice

, Vousden

. 2000. Role of NF-kappaB in p53-mediated programmed cell death. Nature, 404:892–897.

36.

Maurer

, Tripps

, Feldmann

Jr , Kuschinsky

. 2003. Expression of vascular endothelial growth factor and its receptors in rat neural stem cells. Neurosci Lett, 344:165–168.

37.

Chilov

, Kukk

, Taira

, Jeltsch

, Kaukonen

, Palotie

, Joukov

, Alitalo

. 1997. Genomic organization of human and mouse genes for vascular endothelial growth factor C. J Biol Chem, 272:25176–25183.

38.

Wang

, Banerjee

, Li

, Rahman

, Zhang

, Sarkar

. 2006. Down-regulation of notch-1 inhibits invasion by inactivation of nuclear factor-kappaB, vascular endothelial growth factor, and matrix metalloproteinase-9 in pancreatic cancer cells. Cancer Res, 66:2778–2784.

39.

Gottipati

, Rao

, Fung-Leung

. 2008. IRAK1:a critical signaling mediator of innate immunity. Cell Signal, 20:269–276.

40.

Thacker

, Zdzienicka

. 2004. The XRCC genes: expanding roles in DNA double-strand break repair. DNA Repair (Amst), 3:1081–1090.

41.

Dehmelt

, Halpain

. 2005. The MAP2/Tau family of microtubule-associated proteins. Genome Biol, 6:204.

42.

Priest

, Kaczorowski

, Garcia

. 2006. Ion channel modulators: new targets and new indications for old targets. Curr Opin Drug Discov Devel, 9:587–594.

43.

Hait

, Yang

. 2006. The individualization of cancer therapy: the unexpected role of p53. Trans Am Clin Climatol Assoc, 117:85–101discussion 101.

44.

Barrett

, Horowitz

, Pollock

, Zhang

, Bortin

, Buchanan

, Camitta

, Ochs

, Graham-Pole

, Rowlings

. 1994. Bone marrow transplants from HLA-identical siblings as compared with chemotherapy for children with acute lymphoblastic leukemia in a second remission. N Engl J Med, 331:1253–1258.

45.

Zheng

, Ying

, Yan

, Kimmelman

, Hiller

, Chen

, Perry

, Tonon

, Chu

, Ding

, Stommel

, Dunn

, Wiedemeyer

, You

, Brennan

, Wang

, Ligon

, Wong

, Chin

, DePinho

. 2008. p53 and Pten control neural and glioma stem/progenitor cell renewal and differentiation. Nature, 455:1129–1133.

46.

Wang

, Yang

, Zheng

, Tomasek

, Zhang

, McKeever

, Lee

, Zhu

. 2009. Expression of mutant p53 proteins implicates a lineage relationship between neural stem cells and malignant astrocytic glioma in a murine model. Cancer Cell, 15:514–526.

47.

Bassères

, Baldwin

. 2006. Nuclear factor-kappaB and inhibitor of kappaB kinase pathways in oncogenic initiation and progression. Oncogene, 25:6817–6830.

48.

Freedman

, Shin

. 1974. Cellular tumorigenicity in nude mice: correlation with cell growth in semi-solid medium. Cell, 3:355–359.

49.

Singh

, Clarke

, Hide

, Dirks

. 2004. Cancer stem cells in nervous system tumors. Oncogene, 23:7267–7273.

50.

Røsland

, Svendsen

, Torsvik

, Sobala

, McCormack

, Immervoll

, Mysliwietz

, Tonn

, Goldbrunner

, Lønning

, Bjerkvig

, Schichor

. 2009. Long-term cultures of bone marrow-derived human mesenchymal stem cells frequently undergo spontaneous malignant transformation. Cancer Res, 69:5331–5339.

51.

Hemmati

, Nakano

, Lazareff

, Masterman-Smith

, Geschwind

, Bronner-Fraser

, Kornblum

. 2003. Cancerous stem cells can arise from pediatric brain tumors. Proc Natl Acad Sci USA, 100:15178–15183.

52.

Leung

, Yuen

, Chan

, Ho

, Kwan

, Fan

, Hung

, Chung

, Wyllie

. 2000. Chromosomal instability and p53 inactivation are required for genesis of glioblastoma but not for colorectal cancer in patients with germline mismatch repair gene mutation. Oncogene, 19:4079–4083.

53.

Shishodia

, Aggarwal

. 2004. Nuclear factor-kappaB: a friend or a foe in cancer? Biochem Pharmacol, 68:1071–1080.

54.

Cuevas

, Díaz-González

, Dujovny

. 2003. Glioma cell-associated sustained activation of the transcription factor, nuclear factor-kappa B, was inhibited by neomycin. Neurol Res, 25:271–274.

55.

Hui

, Zhang

, Chen

, Xi

, Purow

, Friedman

, Fine

. 2004. Agents with selective estrogen receptor (ER) modulator activity induce apoptosis in vitro and in vivo in ER-negative glioma cells. Cancer Res, 64:9115–9123.

56.

Walton

, Snyder

, Park

, Kobeissy

, Scheffler

, Steindler

. 2009. Gliotypic neural stem cells transiently adopt tumorigenic properties during normal differentiation. Stem Cells, 27:280–289.

57.

Siebzehnrubl

, Jeske

, Muller

, Buslei

, Coras

, Hahnen

, Huttner

, Corbeil

, Kaesbauer

, Appl

, von Horsten

, Blumcke

. 2009. Spontaneous in vitro transformation of adult neural precursors into stem-like cancer cells. Brain Pathol, 19:399–408.

58.

Norden

, Drappatz

, Wen

. 2008. Novel anti-angiogenic therapies for malignant gliomas. Lancet Neurol, 7:1152–1160.

59.

Ranieri

, Patruno

, Ruggieri

, Montemurro

, Valerio

, Ribatti

. 2006. Vascular endothelial growth factor (VEGF) as a target of bevacizumab in cancer: from the biology to the clinic. Curr Med Chem, 13:1845–1857.

60.

Kelly

, Stechishin

, Chojnacki

, Lun

, Sun

, Senger

, Forsyth

, Auer

, Dunn

, Cairncross

, Parney

, Weiss

. 2009. Proliferation of human glioblastoma stem cells occurs independently of exogenous mitogens. Stem Cells, 27:1722–1733.

61.

Biswas

, Cruz

, Gansberger

, Pardee

. 2000. Epidermal growth factor-induced nuclear factor kappa B activation: A major pathway of cell-cycle progression in estrogen-receptor negative breast cancer cells. Proc Natl Acad Sci USA, 97:8542–8547.

62.

Biswas

, Iglehart

. 2006. Linkage between EGFR family receptors and nuclear factor kappaB (NF-kappaB) signaling in breast cancer. J Cell Physiol, 209:645–652.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

2.57 MB