Soluble amyloid precursor protein: a novel proliferation factor of adult progenitor cells of ectodermal and mesodermal origin

Primary neurosphere culture

Two-month-old C57/Bl6 wild-type mice were euthanized and their brains were removed and placed into sterile Dulbecco's modified Eagle's medium/F12 (DMEM/F12). A coronal slice (approximately 1 mm) was dissected starting 1 to 2 mm posterior to the olfactory bulb. The region occupying the lateral wall and anterior horn of the lateral ventricles was removed with the aid of a dissecting microscope and diced with a sterile scalpel. Neurosphere culture was prepared as previously described [20]. Briefly, tissue pieces were collected in a mixture of Papain and DNase in Earl's balanced salt solution and incubated at 37°C for 40 minutes. Then, tissue pieces were pelleted by centrifugation and dissociated to a single-cell suspension, and cells were plated in complete medium-water, DMEM/F12 (Gibco, now part of Invitrogen Corporation, Carlsbad, CA, USA), glucose (Sigma-Aldrich, St. Louis, MO, USA), NaHCO₃ (Sigma-Aldrich), HEPES (Sigma-Aldrich), L-glutamine (Invitrogen Corporation), penicillin/streptomycin (Invitrogen Corporation), putrescine (9.6 μg/mL; Sigma-Aldrich), apotransferrin (0.1 mg/mL; Sigma-Aldrich), insulin (0.025 mg/mL; Roche, Indianapolis, IN, USA), selenium (5.2 ng/mL; Sigma-Aldrich), progesterone (6.3 ng/mL; Sigma-Aldrich), bovine serum albumin (BSA) (2 mg/mL; Sigma-Aldrich), heparin (4 μg/mL; Sigma-Aldrich), EGF (20 ng/mL; PeproTech, Rocky Hill, NJ, USA), and bFGF (10 ng/mL; PeproTech)-and passaged after 10 days.

Isolation of mesenchymal stem cells

After euthanasia, the bone marrow contents of the femurs and tibia of donor Balb/C mice were flushed through a 40-μm filter (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) into a 50-mL tube (Corning, Corning, NY, USA) containing MSC media: 40% alpha-modified Eagle's medium (Invitrogen Corporation, Rockville, MD, USA), 40% F-12 nutrient mixture (Invitrogen Corporation), 10% fetal bovine serum (Valley Biomedical, Winchester, VA, USA), and 1% antibiotic-antimycotic solution (Invitrogen Corporation). Bone marrow cells were plated at a density of 20 × 10⁶ per 9.6 cm² in MSC media at 37°C in 5% CO₂ as previously described [21]. The non-adherent population was removed after 72 hours, and the adherent cells were washed with fresh media and cultured for 7 days. The resulting adherent cells were harvested by incubating with 0.25% trypsin (Invitrogen Corporation) followed by gentle scraping. By means of negative selection via immunomagnetic column (Miltenyi Biotec, Auburn, CA, USA), cells negative for CD11b (eBioscience, San Diego, CA, USA) and CD45 (eBioscience) were placed back into culture. A homogenous cell phenotype was confirmed on the basis of the expression of CD29, CD44, and Sca1 and the absence of hematopoietic (CD45, CD14, and CD11b) markers. Prior to use, cells had been passaged from one to four times.

Human decidua parietalis placenta stem cells

All studies were approved by the Institutional Review Board of the University of Illinois. hdpPSCs were isolated from the decidua parietalis dissected from placental membranes after normal vaginal delivery at term, as previously described in detail [22]. Human placenta tissue was obtained from the Human Female Reproductive Tissue bank in the Center for Women's Health and Reproduction at the University of Illinois at Chicago. Cells were cultured in RPMI-1640 medium supplemented with 10% heat-inactivated and charcoal-stripped fetal bovine serum, 0.1 mM sodium pyruvate, and 1% penicillin/streptomycin. At confluence, cells were trypsinized, propagated, and used for experiments in passage numbers three to five.

Recombinant sAPP

sAPPα (Sigma-Aldrich) was used at 10 nM concentrations unless otherwise indicated (dissolved in phosphate-buffered saline, or PBS).

Conditioned media

Neurosphere media was conditioned by plating 3 × 10⁵ NPCs in each well of a 12-well plate in 500 μL of complete media. After 1 hour, media was removed and spun at 1,000g for 10 minutes to remove any cells or debris. For depletion of sAPP, conditioned media was precleared with protein A-agarose beads (Pierce, Rockford, IL, USA) and then incubated overnight at 4°C with 22C11 antibodies against the N-terminus of APP or IgG antibodies (Millipore Corporation, Billerica, MA, USA) as a control. Protein A-agarose beads were added for 30 minutes, the mixture was spun at 4,000 revolutions per minute (rpm) for 3 minutes, and the supernatant was used as depleted media. Regular conditioned media was subjected to the same process without antibody incubation as a control. All media was filtered through a 0.22-mm filter prior to addition.

Detection of sAPP

For the detection of soluble APP from brain lysates, protein was extracted in immunoprecipitation buffer containing 150 mM NaCl, 50 mM Tris-Cl, 5 mM ethylenediaminetetraacetic acid (EDTA), 1% Triton-X 100, 0.5% sodium deoxycholate, protease inhibitor cocktail, and 250 μM phenylmethylsulfonyl fluoride (PMSF). To remove full-length APP, protein samples were immunodepleted by using 369 antibodies against the C-terminus of APP (a gift from Sangram S Sisodia, The University of Chicago). Briefly, samples were precleared with 50 μL of immobilized protein A-agarose beads (Pierce) at 4°C for 30 minutes. Samples were centrifuged at 4,000 rpm for 3 minutes, and the pellet was discarded. To the supernatant, 5 μL of 369 antibody was added and incubated overnight at 4°C. The next morning, 50 μL of immobilized protein A was again added for 30 minutes at 4°C and spun at 4,000 rpm for 3 minutes. The pellet contains the full-length APP-369 antibody complex, and the supernatant was probed for sAPP by using 22C11 antibodies raised against the N-terminus of APP (Millipore Corporation).

Neurosphere formation (clonogenic) assay

Briefly, neurospheres were singly dissociated by mechanical dissociation and plated at 1,000 cells per well onto 96-well plates. For matrix-metalloproteinase (MMP) inhibitor experiments, cells were then treated with the indicated molar concentration of GM6001 or GM6001 negative control (Millipore Corporation) and the indicated molar concentrations of sAPP or conditioned media. If not otherwise indicated, 1 μM GM6001 and negative control inactive inhibitor (NC) were used. Cells were treated every 72 hours for 10 days. After 10 days in culture, neurospheres were counted under an inverted light microscope, and the average neurosphere diameter was calculated from 25 randomly assigned squares of the grid by using a Zeiss AX10 microscope (Carl Zeiss Ltd., Hertfordshire, UK) and StereoInvestigator software (MBF Bioscience, Williston, VT, USA). After sphere size determination, cells were singly dissociated with a p200 pipette and counted with a hemocytometer. The remaining cells were placed onto Matrigel-coated chamber slides for 30 minutes and then fixed in 4% paraformaldehyde for 30 minutes for immunocytochemistry. Briefly, cells were washed four times in Tris-buffered saline (TBS) and then placed into blocking solution (5% normal donkey serum, 0.25% Triton-X 100 in TBS) for 30 minutes at room temperature. Next, cells were incubated in primary antibodies-mouse anti-nestin (1:100; Millipore Corporation) and goat anti-Sox2 (1:200; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA)-in TBS containing 0.25% Triton-X 100 for 1 hour at room temperature. After primary antibodies, cells were again incubated in blocking solution for 30 minutes at room temperature before secondary antibody incubation-anti-goat cy5 (1:250; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA) and anti-mouse cy3 (1:500; Jackson ImmunoResearch Laboratories, Inc.)-in TBS with 0.25% Triton-X 100 for 30 minutes at room temperature in the dark. Cells were then washed four times in TBS and incubated for 5 minutes with DAPI (4'-6-diamidino-2-phenylindole) (1:50,000; Invitrogen Corporation) at room temperature in the dark. Cells were then washed three times and mounted with polyvinyl alcohol-DABCO mounting solution. Cell counts were made by means of StereoInvestigator software version 8 (MBF Bioscience).

Mesenchymal and placental cell proliferation experiments

MSC or hdPSC were trypsinized for 5 to 10 minutes with 0.05% trypsin, collected after trypsin inactivation, and spun at 500g for 5 minutes. Pellets were then dissociated and plated at 1,000 cells per well in 96-well plates. Cells were then treated with the indicated molar concentration of GM6001 or GM6001 negative control (Millipore Corporation) and the indicated molar concentrations of sAPP. After 3 days in culture, cells were trypsinized for 5 to 10 minutes, spun, and counted with a hemocytometer.

Brain tissue for biochemistry and immunohistochemistry

Experiments using animals were performed according to guidelines of the National Institutes of Health and the University of Illinois at Chicago Institutional Animal Care and Use Committee. The APP knockout [APP(-/-)] model has been described previously [23, 24]. Briefly, the authors generated the mice through homologous recombination in embryonic stem cells. Mice heterozygous for APP expression were cross-mated and APP(+/+) (APP wild-type), APP(+/-), and APP(-/-) resulted from this breeding. Our colony is maintained via group housing (fewer than five mice per cage) in a barrier facility under a 14:10 light/dark cycle with free access to food and water. Animal care and procedures were conducted according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals[25].

Brain tissue processing

For in vivo immunohistochemical staining, male APP(+/+) mice from 3 to 5 months old were used. All mice were anesthetized with a mixture of ketamine and xylazine and transcardially perfused with 100 mL of ice-cold PBS. The brains were then removed and halved in the sagittal plane. The left half was immediately placed into 4% paraformaldehyde on ice. From the right half of the brain, the following regions were dissected for biochemical analysis and immediately placed into Eppendorf tubes on dry ice: SVZ, hippocampus, olfactory bulb, frontal cortex, and cerebellum.

Immunohistochemistry

Left hemibrains from PBS-perfused mice were post-fixed in 4% paraformaldehyde for 3 days and stored in 30% sucrose at 4°C. Hemibrains were sectioned sagittally at 50 μm by using a microtome and placed into cryopreservent (47.6% PBS, 28.57% ethylene glycol, and 25% glycerin vol/vol). Sections were blocked by using a solution containing 0.25% vol/vol Triton-X 100 (Sigma-Aldrich) and 5% vol/vol Normal Donkey Serum (Jackson ImmunoResearch Laboratories, Inc.) in TBS. The following antibodies were used: Dlx-2 (1:200; Millipore Corporation), nestin (1:100; Millipore Corporation), Sox2 (1:100; Santa Cruz Biotechnology, Inc.), ADAM10 (1:200; Millipore Corporation), and APP (22C11; Millipore Corporation and A8717; Sigma-Aldrich). Floating sections were incubated in primary antibodies for 72 hours at 4°C before continuing with blocking, biotin conjugation (Jackson ImmunoResearch Laboratories, Inc.), and secondary antibody incubation (cy2 Streptavidin, anti-mouse cy3, anti-goat cy5, and anti-rabbit cy5; Jackson ImmunoResearch Laboratories, Inc.).

Western blotting

Protein extraction from brain tissue was performed in lysis buffer containing 1X TNE, 50 mM Tris, 150 mM NaCl, 5 mM EDTA, protease inhibitor cocktail (Sigma-Aldrich), and 100 mM PMSF. Quantification of protein was performed by using the bicinchoninic acid (BCA) method (Pierce), and equal amounts of protein were subjected to direct immunoblotting. For the extraction of protein from neurosphere, mesenchymal, and Human decidua parietalis placenta stem cells cultures, a lysis buffer containing 150 mM NaCl, 50 mM Tris-Cl, 5 mM EDTA, 1% Triton-X 100, 0.5% sodium deoxycholate, protease inhibitor cocktail, and 250 μM PMSF was used. For quantification, at least three cultures were used.

Erk and Akt signaling

To assay phosphorylation of Erk and Akt, neurospheres were singly dissociated and plated at 5 × 10⁵ cells per well in a six-well plate in Earle's balanced salt solution (Sigma-Aldrich) and treated immediately with 1 μM GM6001 or GM6001 negative control. After 1-hour incubation at 37°C, one of the GM6001-treated groups was subsequently treated with 10 nM recombinant sAPPα (Sigma-Aldrich) for 15 minutes before all groups were lysed in ROLB buffer: 10 mM HEPES, pH 7.4, 0.5% Triton X-100, 80 mM β-glycerophosphate, 50 mM sodium fluoride, 2 mM sodium orthovanadate, 100 nM staurosporine, 100 nM K252a, 50 nM okadaic acid, 50 nM microcystin, mammalian protease inhibitor cocktail (Sigma-Aldrich), and phosphatase inhibitor cocktail II (Calbiochem, now part of EMD Biosciences, Inc., San Diego, CA, USA) in water. After lysis, protein quantification was performed by using the BCA method (Pierce), and equal amounts of protein were run on Tris-glycine gels and transferred to nitrocellulose membranes. For blocking and antibodies, we employed a solution of 0.05% vol/vol Tween, 10% wt/vol milk, and 0.1% wt/vol BSA (Sigma-Aldrich) in TBS. The following antibodies were used: pErk (1:500; Santa Cruz Biotechnology, Inc.), Erk (1:1,000; Santa Cruz Biotechnology, Inc.), pAkt (1:1,000; Cell Signaling Technology, Inc., Danvers, MA, USA), and Akt (1:500; Cell Signaling Technology, Inc.). (N = 3 for Erk and Akt Western blot quantification.)

sAPPα rescues proliferation of neural progenitor cells after treatment with a matrix-metalloproteinase and ADAM inhibitor

To examine the role of sAPPα in NPC proliferation and the significance of α-secretase cleavage of APP in this process, NPCs isolated from the SVZ of adult mice were treated with GM6001, a potent broad-spectrum hydroxamic acid-based inhibitor of MMPs and ADAMs which has been shown to decrease levels of sAPP and sAPLP2 [11]. First, using Western blot analysis, we confirmed that levels of sAPP are reduced in the conditioned media of NPCs treated with GM6001 (Figure 1a, b). Next, we singly dissociated NPCs and treated them with either inactive inhibitor (NC) or GM6001 or with GM6001 and sAPPα in a neurosphere formation assay. We observed that treatment of neurosphere culture with GM6001 significantly reduced proliferation of NPCs in a dose-dependent manner (Figure 1c). Inhibition of NPC proliferation could be rescued by the addition of recombinant sAPPα (Figure 1c), suggesting that sAPPα is sufficient for the induction of NPC proliferation. To further characterize the effect of sAPPα on NPC proliferation, we repeated the clonogenic assay, treated singly dissociated NPCs with NC or GM6001 or with GM6001 and sAPPα, and determined the number of neurospheres generated, their diameter, and the total number of NPCs (Figure 1d-f). We observed that, after treatment with GM6001, there were significant reductions in the diameter of neurospheres (Figure 1e) and the total number of NPCs observed in the cultures (Figure 1d) but not a significant reduction in the number of clones (Figure 1f), suggesting that treatment with GM6001 affects NPC proliferation without affecting clone formation capability. Likewise, treatment of neurosphere cultures with GM6001 and sAPPα significantly rescued impaired neurosphere diameter and total number of NPCs without affecting the number of neurospheres (Figure 1d-f). To determine whether recovery of NPC proliferation by sAPPα occurs in a dose-dependent manner, sAPPα was added to neurosphere culture at a concentration range of 100 pM to 100 nM. sAPPα recovered NPC proliferation after GM6001 treatment in a concentration as low as 100 pM. Rescue of NPC proliferation increased as a function of sAPPα dose and was maximal at 10 nM (Figure 1g). Taken together, these data suggest that α-secretase cleavage of APP is an important event in NPC cell cycle control and that sAPPα regulates NPC proliferation.

To determine the subpopulation of NPCs affected by GM6001 inhibition and recovered by sAPPα, singly dissociated NPCs treated with GM6001 or GM6001 + sAPPα were allowed to form neurospheres in a clonogenic assay as before and then immunolabeled with antibodies raised against nestin, an intermediate filament protein expressed in stem and progenitor cells [26], and Sox2, a transcription factor expressed in neural stem cells (NSCs) and NPCs of the adult brain [27]. We show that the number of nestin⁺Sox2⁺ cells was significantly reduced after treatment with GM6001, and their number was fully recovered after addition of sAPPα (Figure 2a-c, 2e, f), suggesting that sAPPα affects the proliferation of both NSCs and NPCs. Thus, we proceeded to examine whether sAPPα affects self-renewal of NSCs (Figure 2d). For this purpose, secondary neurospheres were singly dissociated and cultured at one cell per well in 96-well plates, and the number of cells that generated tertiary neurospheres was quantified. We show that treatment of cells with GM6001 significantly reduced the number of generated neurospheres but that the addition of sAPPα significantly enhanced the number of neurspheres generated (Figure 2d), suggesting that sAPPα regulates both NSC self-renewal and NPC proliferation.

sAPPα regulates neural progenitor cell proliferation in an EGF- and bFGF-independent manner

To address whether sAPPα is a stand-alone proliferation factor or its proliferative effect depends on other growth factors, we first compared the ability of NPC-conditioned media containing sAPPα to rescue NPC proliferation with that of conditioned media depleted of sAPPα. We observed that conditioned media containing sAPPα successfully rescued NPC proliferation but that conditioned media depleted of sAPPα failed to do so (Figure 3a). Next, we excluded proliferation factors (namely, EGF and bFGF) from the proliferation medium of NPCs and examined the rate of proliferation of cells by measuring clone formation and size (Figure 3b). As expected, neurosphere proliferation was reduced without EGF and bFGF. Interestingly, treatment of these neurospheres with GM6001 further reduced the extent of proliferation, suggesting that factors aside from EGF and bFGF regulate neurosphere proliferation (Figure 3b). Importantly, the addition of soluble APPα recovered GM6001-induced deficits in proliferation even without EGF and bFGF, suggesting that sAPPα operates independently of EGF and bFGF (Figure 3b). To determine whether sAPPα operates synergistically to these growth factors in regulation of neurosphere proliferation, we quantified the effect of sAPPα in medium devoid of EGF/bFGF and supplemented with GM6001 as well as the effect of EGF/bFGF in a medium supplemented with GM6001 and compared the effect on neurosphere diameter with the effect of EGF/bFGF on neurosphere diameter in medium supplemented with the negative control for GM6001. We observed that the additive effect of sAPPα in medium devoid of EGF/bFGF and supplemented with GM6001 and the effect of EGF/bFGF in a medium supplemented with GM is greater than the effect of EGF/bFGF in medium supplemented with the negative control for GM, suggesting a synergistic effect of sAPPα, EGF, and bFGF (Figure 3b).

sAPPα regulates the proliferation of adult stem cells of non-neural origin

As APP is ubiquitously expressed and was shown to act upon fibroblasts and thyroid epithelial cells [14, 15], we examined whether it regulates adult stem cells of non-neural origin. Thus, we first tested whether reduced proliferation after inhibition of MMPs and ADAMs was cell-autonomous to NPCs or applied to other stem cell types. For this purpose, we examined the extent of proliferation of MSCs after GM6001 treatment. As in NPCs, GM6001 treatment reduced secreted levels of sAPPα (Figure 4a, b) and significantly reduced proliferation of MSCs (Figure 4c) in a dose-dependent manner, suggesting that MMP family members may play an important role in the proliferation of MSCs. The addition of recombinant sAPPα significantly recovered MSC proliferation (Figure 4c, d). Like in the NPCs, 100 pM of sAPPα was sufficient to significantly rescue MSC proliferation and enhance their proliferation in a dose-dependent manner (Figure 4e). Taken together, these results suggest that APP is expressed in MSCs and processed by α-secretase activity to yield sAPPα that in turn regulates the proliferation of MSCs. Next, we investigated the effect of α-secretase inhibition and sAPPα on the proliferative activity of hdpPSCs. The effect of GM6001 on hdpPSCs was similar to the one observed with NPCs and MSCs (Figure 5). Treatment with GM6001 significantly reduced levels of secreted sAPPα concomitantly with hdpPSC proliferation (Figure 5a-c), and this deficit was ameliorated by the addition of recombinant sAPPα in a dose-dependent manner (Figure 5c-e). Taken together, these results suggest that sAPPα is an essential proliferation factor of a variety of adult stem cells of different origins and provides a functional significance for the abundance of APP.

High levels of sAPP are expressed in the subventricular zone of adult mice

To establish that sAPPα plays a role in proliferation of NPCs in vivo, we examined expression levels of full-length APP and sAPPα in the SVZ, SGL, and a non-neurogenic region (cerebellum) of APP(-/-) and APP(+/+) mice. For this purpose, we used antibodies recognizing the N◯-terminus of APP (22C11), thus recognizing both full-length APP and sAPPα. We show that levels of 22C11 are comparable in the SGL and cerebellum with a trend of increased expression in the SVZ (Figure 6a). To examine whether this trend is a result of increased levels of sAPPα specifically, we immunodepleted full-length APP from the protein extract by using antibodies recognizing the carboxyl-terminus of APP, thus not capturing sAPPα. We observed that levels of sAPPα were particularly high in the SVZ of both APP(+/+) and APP(+/-) (Figure 6b-d). These results support our notion that sAPPα plays a major role in the SVZ.

To determine whether the NSC and NPC populations that express APP resemble those observed in vitro in neurosphere culture, we double-stained brain sections of adult mice with antibodies raised against APP, Sox2, nestin, and Dlx-2, the last of which is a homeobox protein expressed in transit-amplifying C cells of the adult SVZ [28]. APP expression colocalized with all three markers, suggesting that APP is expressed in NSCs and transit-amplifying NPCs in the adult SVZ (Figure 6e).

ADAM10 is expressed in the subventricular zone and hippocampus of adult mice

To examine whether high levels of sAPPα in the SVZ reflect levels of α-secretase in this region, we examined expression of ADAM10, recently shown to be a critical α-secretase during brain development [13]. We show that levels of ADAM10 are significantly higher in the SVZ compared with the SGL (Figure 7a, b). In support of a correlation between levels of sAPPα and ADAM10 is the observation that ADAM10 levels are lower in the SVZ of APP(-/-) compared with APP(+/+) (Figure 7a, c). To examine whether ADAM10 is expressed in the same neurogenic populations as APP, we immunostained brain sections with antibodies raised against ADAM10, nestin, Sox2, and BrdU 24 hours after a single dose. We observed that ADAM10 colocalizes with all three markers, suggesting that, like APP, ADAM10 is expressed in NSCs and transit-amplifying NPCs in the adult SVZ (Figure 7d, e).

sAPP regulates extracellular signal-regulated kinase activity

The mitogen-activated protein (MAP) kinase pathway plays an important role in the proliferation of NPCs derived from the adult SVZ [29]. Treatment with sAPP has been shown to stimulate ERK signaling in PC-12 cells [30], and SORLA-deficient mice showing enhanced APP cleavage to sAPP and Aβ display increased neurogenesis and ERK signaling in hippocampal neurons [19]. Therefore, we tested whether inhibition of MMP activity, decreasing sAPPα production, would result in deficient ERK signaling. For this purpose, NPCs were treated with GM6001 as before and placed into Hank's balanced salt solution for 1 hour to remove any presence of exogenous growth factor signaling. Western blot analysis reveals that pERK is reduced approximately 25% in NPCs treated with GM6001 (Figure 8a, b). To examine whether this reduction can be recovered by addition of sAPPα, we added recombinant sAPPα prior to lysing the cells. Recombinant sAPPα was able to reverse the deficits in ERK phosphorylation caused by GM6001 treatment (Figure 8a, b). To test whether this effect was specific to ERK, we examined another integral proliferation pathway known to respond to sAPPα [31], protein kinase B (Akt). Phosphorylation of Akt did not change significantly from negative control in any of the groups tested (Figure 8c, d), suggesting that sAPPα is associated specifically with ERK signaling.