In order to confirm that lpr NPCs lack Fas, we stained NPCs with a Fas-specific phycoerythrin (PE)-conjugated antibody (or isotype control antibody). Flow cytometric analysis demonstrates high levels of Fas surface expression in wt NPCs, and no Fas expression in lpr NPCs (Figure 1).

Proliferation in response to mitogens and the ability to self-renew are defining properties of NPCs. All proliferation experiments were performed in the presence of epidermal growth factor (EGF) and fibroblast growth factor (FGF). A bromodeoxyuridine (BrDU) assay was used to determine the effects of Fas deficency on NPC proliferation. Lpr NPCs showed significantly less BrDU incorporation over 16vhours in complete media compared to wt NPCs. Lpr NPCs had 18% ± 1.6 less proliferation compared to wt (Figure 2(a)).

Additionally, neurosphere growth in lpr NPCs was abrogated compared to wt NPCs (Figure 2(b)). We performed the neurosphere assay to quantify this difference. In this assay, size of the neurosphere is an indicator of mitogenic or proliferative capacity, while neurosphere number is an estimate of the absolute total number of putative stem cells [10] . Over three subsequent passages, lpr NPCs developed significantly smaller neurospheres than wt (Figure 2(c)). While, the total number of neurospheres remained the same (Figure 2(d)).

To be certain the abrogated proliferative capacity of lpr NPCs was not due to altered apoptosis profiles, we also performed UV Live-Dead dye (labeling dead cells) and annexin (labeling early apoptotic cells) double-staining on NPCs growing in the presence of growth factors (EGF and FGF, similar to proliferation experiments). Four days post-cell plating, there was no difference in the amount of dead and dying cells in lpr and wt cultures (Figures 2(e) and (f)). Quantification of the data showed no significant difference between the percentage of live/

healthy cells in lpr (97% ± 2.3) and wt (95% ± 2.6) cells. Therefore, the decrease in proliferation cannot be attributed to increased cell death.

In summary, Fas deficient NPCs are deficient in proliferative properties, evidenced by decreased BrDU incorporation and smaller sphere size, but Fas expression has no impact on the total number of stem cells (number of neurospheres).

The ability of NPCs to survive in the absence of growth factors (EGF, FGF) is a fundamental issue for transplantation models. If NPCs are unable to survive after transplantation due to inadequate levels of growth factors, their beneficial effects are negated. Normally, upon growth factor withdrawal in vitro, a large percentage of NPCs die [7]. We observed significantly higher survival in lpr NPCs (42%) compared to wt NPCs (25%) after 48 hours in minimal media (no EGF or FGF) using UV/Annexin staining (described above). Therefore, the percentage of live/healthy cells in lpr cultures after growth factor withdrawal is nearly doubled when compared to wt NPCs

(Figures 3(a) and (b)).

One distinguishing feature of NPCs is their capacity to differentiate into the three neuroectodermal lineages. We compared lpr and wt NPC lineage marker RNA expression for astrocytes (glial fibrillary acidic protein (GFAP), glutamate aspartate transporter (GLAST)), OPCs (platelet-derived growth factor receptor-α (PDGFRα)) and neurons (βIII-tubulin, Doublecortin (Dcx)) using RTqPCR. After two days of differentiation, lpr NPCs expressed significantly higher levels of OPC and neuronal markers compared to wt NPCs (Figure 4(a)). This trend maintained statistical significance after seven days in differentiating media (Figure 4(b)).

We confirmed the enhanced OPC phenotype in lpr NPCs by staining for A2B5, a putative OPC cell surface marker, after two days of differentiation. Flow cytometric analysis revealed 60% of lpr NPCs were positive for A2B5, while only 20% of wt NPCs were A2B5-positive (Figures 5(a) and (b)).

NPC cell morphology was also confirmed at the protein level using immunocytochemistry. Staining for βIIItubulin and PDGFRα corroborated mRNA expression levels, as the percentage of neuronal cells (βIII-tubulin+) and OPCs (PDGFRα+) in lpr cultures was double that found in wt (Figures 5(c) and (d)).

Altogether, these data indicate that the absence of Fas enhances NPC differentiation to the neuronal and OPC lineages at both the genomic and protein level.

In search of possible mechanisms and pathways controlling the lpr phenotype, we ran a “Neurogenesis and Neural Stem Cell” (SA Biosciences^TM) PCR array using RNA isolated from wt or lpr NPCs after 7 days of differentiation. Of the >80 genes tested, four were downregulated (apolipoprotein E (Apoe), bone morphogenetic protein 2 (BMP2), FGF2, S100 calcium binding protein A6 (S100a6)) and 6 were up-regulated (α-filamin (Flna), hairy and enhancer of split 1 (Hes1), Neurod1, Noggin, neuronal pentraxin 1 (NptX1), Shh) in lpr compared to wt (Figure 6(a)). Neurod1, Noggin, and Nptx1 are all genes known to be involved in regulating neuronal fate specification. Shh is a major modulator of both oligodendrocyte and neuronal differentiation pathways. The Shh antagonist BMP2 is down-regulated in lpr NPCs (–3.87 fold, p < 0.0012); further indicating Shh pathway alterations as a probable mechanism for the altered lpr phenotype.

Upregulation of Hes1, Noggin, and Shh was confirmed using RT-qPCR (Figure 6(b)). To further evaluate the involvement of Shh in the lpr phenotype, we also tested expression levels of Gli1, a major regulator and target of Shh signaling. Gli1 was up-regulated 3.86-fold (±0.55) in lpr NPCs compared to wt.

NPCs were isolated from whole brains dissected from 4-day postnatal wild-type C57/BL6 or lpr mutant mice (C57/BL6 background). Brains were processed using NeuroCult^® Enzymatic Dissociation Kit for CNS Tissue (Stem Cell Technologies). Resulting single-cell suspension was plated in EGF and FGF (both 10ng/mL) containing complete media (Gibco^®Neural Basal Media supplemented with B27 and 1:1:2). Neurospheres were chemically dissociated (Kit, Stem Cell Technologies) into a single-cell suspension and plated on poly-D Lysine and laminin (both Sigma) coated Nunc^® flasks or plates for passaging and experimentation. Cells were plated at a density of 16,000 cells/mL media and grown to confluency (3 - 4 days) before experimentation or switching to minimal media. All experiments were performed on passage-matched wt or lpr NPC lines and only cells with passage numbers between 3 and 6 were used. Proliferation studies were performed in EGF and FGF containing media, cell death studies and early differentiation timepoints were performed in minimal media (Neural Basal Media with only B27 and 1:1:2 [penicillin/streptomycin: L-glutamine: glucose]), and extended differentiation experiments were performed in minimal media containing 0.2% FCS (to prevent extensive cell death in the absence of growth factors). All cells incubated in 37˚C/5% CO₂ conditions.

Cells were lifted using 0.05% trypsin/EDTA (Gibco) at 37˚C. Trypsin enzyme was stopped using 20% FCS/ neural basal media after 7 mins. Cells were placed in Falcon^® tubes (#352058) and then rinsed with 1XPBS and spun at 1400 rpm before proceeding with staining. All stains were added to a 100 µL volume cell suspension containing approximately 1 × 10⁶ cells.

Fas receptor stain: Samples blocked with hamster IgG (50 ug/mL). Samples stained with either anti-Fas or isotype control (Hamster IgG2, λ1) phycoerythrin (PE)-conjugated antibodies (1:30, BD Biosciences). All rinses and dilutions were done with 1%BSA/1XPBS. All incubations performed at 4˚C (15 min for block, 30 min for PE-conjugated antibodies).

A2B5 stain: biotynlated anti-A2B5 (50 uL, 1:100, BD Pharm) added to each sample. Primary antibody incubated for 30 min at 4˚C. After rinsing with PBS, Strepavidin-Phycoerythrin (100 uL, undiluted, BD Pharm) added to each sample and incubated for 30 min at 4˚C. Samples rinsed twice before analyzing on flow. Controls included: no primary and a sample of NPCs growing in the presence of EGF and FGF (to serve as a low percent differentiation negative control).

Death & Apoptosis Assay (UV/Annexin stain): UV Live/Dead Dye^® and Annexin Alexa Fluor-647 both used according to manufacturer instructions (Invitrogen). UV dye labels dead cells because of disruptions in the cell membrane. Viable cells to not stain positive for UV dye because of an intact cell membrane. Annexin binds strongly to phosphatidylserine residues that have flipped to the exterior surface of the cell, thus labeling early apoptotic cells. Briefly, 200 µL of diluted UV dye (2 µL per mL 1 × PBS) was added to cell suspension containing 10,000 cells/µL. Cells were incubated on ice, in the dark, for 30 min. After cells with 1 × PBS and aspirating the supernatant, 100 uL of diluted Annexin (1 µL per 400 µL HEPES buffer) was added to cells. Cells grown in complete media (growth factors present) were used as a negative control because >90% of cells in these cultures are live/healthy. Cells briefly treated with 3% paraformaldehyde were used as positive controls because >90% of cells are dead/dying.

Cell Proliferation: BrDU assay performed according to manufacturer instructions (FITC Detection BrdU Flow Cytometry Assay Kit, BD Biosciences). Briefly, cells were first pulsed overnight (16 hours) with BrDU and subsequently fixed and stained with a FITC-conjugated anti-BrDU antibody for flow analysis. Cells grown in minimal media (no growth factors) were used as a negative control because <10% of cells grown in these conditions will stain positive for BrDU incorporation.

Analysis: flow cytometry analysis was performed by a BD LSRII flow cytometer equipped with 3 lasers (488, helium neon, and UV) and data were analyzed with FlowJo software, respectively.

Single-cell suspensions were plated in uncoated 24-well plates (seeding density of 100,000 cells/mL) using NPC media containing EGF and FGF. After 7 days, 2-3 pictures of random frames from each well (3 wells per cell type) were taken at 20X using a phase-contrast microscope. After taking pictures, neurospheres were dissociated using Neurocult^® Chemical Dissociation Kit and then filtered through a 40 µm nylon cell strainer (BD Falcon^TM). Cells were then counted, plated as described above, and grown for 7 days before the next passage time point was recorded (this same process completed four total times over four weeks). Exact cell circumferences were measured using Neurolucida software.

Cells were plated in 24-well plates with poly-D/Laminin coated coverslips and grown to confluency. At this point, cells were rinsed and allowed to differentiate 1 week in minimal media containing 0.2% FCS. For all differentiation markers, 3 coverslips were scored per cell line and repeated for at least 3 biological replicates.

Stains performed as follows: Cells fixed with Zamboni’s fixative and blocked (10%HS/0.1% Triton/0.02% Azide in 1 × PBS) for 30 min. Block for PDGFRα stain did not include triton. Primaries incubated overnight at 4˚C, while secondary antibodies incubated for 1.5 hr at RT. Cells rinsed 3X between primary and secondary. All cells counterstained with Hoechst nuclear stain (1:2000).

The following antibodies were used: rabbit anti-BIII Tubulin (Sigma, 1:100), rabbit anti-PDGFRα (Santa Cruz, 1:50), rabbit anti-GFAP (Dako, 1:1000), goat anti-rabbit Cy3 (1:500, Jackson Immunoresearch).

All confocal images were acquired using the “Tile” feature on a Zeiss LSM 510 META confocal laser microscope. Cells were visualized on the left side of each cover slip at 20x magnification, and then the microscope acquired 16 consecutive field-of-view images in the form of a 4 × 4 square (referred to here as a tile). All cell scoring was performed using ImageJ software. Each tile was scored for the number of nuclei present (hoecsht positive cells) and the number of cells staining positive for the respective differentiation marker (a cell was considered positive if the cell body and projections were labeled). At least 9 tiles were acquired and tallied per slip, resulting in the total acquisition of >75 individual fields-of-view per coverslip. Percent positive was calculated by dividing the number of cells positive for a lineage marker (βIII-tubulin or PDGFRα) by the total number cells (hoecsht staining nuclei).

Performed as previously described [7]. Briefly, RNA was extracted from cells using RNeasy^® Mini Kit (Qiagen) according to manufacturer protocol. For RT-qPCR, cDNA was prepared using Superscript^® III First-Strand Synthesis Supermix for qRT-PCR. RT-qPCR completed using Taqman^® Master Mix and applied Biosystems 7500 Fast Software. Primers were purchased from Applied Biosystems (Taqman Assay on Demand).

Neurogenesis and Neural Stem Cell (PAMM-404) PCR arrays purchased from SA Biosciences^TM and RNA samples were processed by UVM Cancer Center DNA Facility.

Graphs were created and statistical analyses were run using GraphPad Prism software. Student’s t-test or ANOVA with Tukey’s post-test were used to determine significance (p-values).