Long-term potentiation promotes proliferation/survival and neuronal differentiation of neural stem/progenitor cells

Abstract

Neural stem cell (NSC) replacement therapy is considered a promising cell replacement therapy for various neurodegenerative diseases. However, the low rate of NSC survival and neurogenesis currently limits its clinical potential. Here, we examined if hippocampal long-term potentiation (LTP), one of the most well characterized forms of synaptic plasticity, promotes neurogenesis by facilitating proliferation/survival and neuronal differentiation of NSCs. We found that the induction of hippocampal LTP significantly facilitates proliferation/survival and neuronal differentiation of both endogenous neural progenitor cells (NPCs) and exogenously transplanted NSCs in the hippocampus in rats. These effects were eliminated by preventing LTP induction by pharmacological blockade of the N-methyl-D-aspartate glutamate receptor (NMDAR) via systemic application of the receptor antagonist, 3-[(R)-2-carboxypiperazin-4-yl]-propyl-1-phosphonic acid (CPP). Moreover, using a NPC-neuron co-culture system, we were able to demonstrate that the LTP-promoted NPC neurogenesis is at least in part mediated by a LTP-increased neuronal release of brain-derived neurotrophic factor (BDNF) and its consequent activation of tropomysosin receptor kinase B (TrkB) receptors on NSCs. Our results indicate that LTP promotes the neurogenesis of both endogenous and exogenously transplanted NSCs in the brain. The study suggests that pre-conditioning of the host brain receiving area with a LTP-inducing deep brain stimulation protocol prior to NSC transplantation may increase the likelihood of success of using NSC transplantation as an effective cell therapy for various neurodegenerative diseases.

Figure 1. NMDAR-dependent LTP enhances proliferation of NPCs in the DG.

(A) The induction of LTP in the hippocampal DG requires activation of NMDARs. A stable basal level of fEPSPs was obtained by electrical stimulation of MPP (0.05 Hz) from the DG of anesthetized rats and LTP was induced by application of sTPS through the same stimulation electrode. Representative traces of fEPSP on the left were averages of all individual recordings before and after the establishment of LTP. Systemic application of the competitive NMDAR antagonist CPP (10 mg/kg) prevented the induction of LTP (CPP+sTPS), without affecting basal level of fEPSP (0.05 Hz+CPP). (B) LTP increases the total numbers of PCNA-positive immature neurons in the DG. Representative images from coronal sections from non-LTP (0.05 Hz+ saline) and LTP-induced (LTP) rats co-stained with PCNA (red) and DAPI (blue) in the DG area. Examples of doubled-stained cells in boxed areas were shown in higher magnification on the right panels. Bar graph on the right summarizes data from various groups, suggesting enhancement of proliferation by LTP induction. Note CPP pretreatment alone (0.05 Hz+CPP) has no effect on intrinsic proliferation, but prevented LTP-promoted NPC proliferation (CPP+sTPS). **p<0.01, n = 8 or 9. Statistical analysis was performed with One Way ANOVA. Post hoc testing revealed a significant difference between the LTP group and other three groups. Data are mean ± SEM. significantly different from each other.

Figure 2. NMDAR-dependent LTP enhances neuronal differentiation and maturation of NPCs in the DG.

(A) Schematic representation of the experimental design. (B) The induction of LTP increases the total number of GFP-labeled NPCs, and DCX and GFP double-labeled immature neurons differentiated from NPCs in the DG. Left: Representative images from coronal sections double-stained with antibodies against an immature neuronal marker DCX (red) and GFP (green) in the DG. Examples of double-stained immature neurons in boxed areas are shown in higher magnification in the panels on the right. Right: Bar graphs summarizing effects of LTP induction on the total number of GFP-positive cells (top panel), and DCX and GFP double-labeled cells (bottom panel) in the DG. *p<0.05 and **p<0.01; n = 8 or 9 in each group. (C) The induction of LTP increases the total number of GFP-labeled NPCs, and NeuN and GFP double-labeled differentiated mature neurons in the DG. Left: Representative images from coronal sections double-stained with antibodies against a mature neuronal marker NeuN (red) and GFP (green) in the DG. Examples of double-stained mature neurons in boxed areas are shown in higher magnification in the panels on the right. Right: Bar graphs summarizing effects of LTP induction on the total number of GFP-positive cells (top panel), and NeuN and GFP double-labeled cells (bottom panel) in the DG. *p<0.05 and **p<0.01; n = 8 or 9 in each group. Statistical analyses were performed with One Way ANOVA. Post hoc testing revealed a significant difference between the LTP group and other three groups. Data are mean ± SEM.

Figure 3. The induction of LTP increases neuronal differentiation of exogenous NSCs transplanted into the CA1 region.

(A) Schematic representation of the experimental design. (B) NMDAR-dependent LTP in the hippocampal CA1 region was reliably induced by HFS of the Schaffer collateral inputs in anesthetized rats. Left: Representative traces of fEPSP were averages of individual recordings taken before and after the establishment of LTP. Right: Systemic application of the competitive NMDAR antagonist CPP prevented the induction of LTP (CPP+HFS), without affecting basal level of fEPSP (0.05 Hz+CPP). (C) Representative images from coronal sections show LTP (LTP) enhanced the total GFP (green)-positive cells in the CA1 compared with the control group (0.05 Hz+ saline). Bar graph below summarizes data from each group of rats. LTP increased the total numbers of transplanted NSCs. (D) Representative images from coronal sections double-stained for GFP and NeuN showing LTP (LTP) increased numbers of NeuN-positive (red) NSCs (green) in the CA1 region, compared with the control group (0.05 Hz+ saline). Bar graph below summarizing data of LTP enhancement of neuronal differentiation. *p<0.05, n = 7, 7, 10, and 8. Statistical analyses were performed with One Way ANOVA. Post hoc testing revealed a significant difference between the pre-conditioning LTP group and other three groups. Data are mean ± SEM.

Figure 4. Isolation and characterization of NSCs in cultures.

(A) Immunostaining characterization of neurosphere and dissociated NSCs. Top two panels show NSCs in neurospheres isolated and characterized from E14 telencephalon. The representative phase contrast image on the left showing the formation of neurospheres and representative immunostaining image on the right showing that majority of cells in the neurosphere express nestin, a NPC marker. The remaining 4 panels are immunostained images of individual NSCs dissociated from neurospheres doubled stained with a nuclear marker DAPI (blue) and one of various cell type markers (green): nestin, vimentin (NPCs), MAP2 (mature neuron) or GFAP (radial glia and/or progenitors). (B) and (C) Biochemical characterization of NSCs. Western blots of lysates of the adult rat brain, or NSCs grown in media were sequentially probed with various cell type-specific antibodies against nestin, vimentin, MAP2 and GFAP (B), and various ionotropic neurotransmitter receptor GluN1, GluA1, GluA2 and GABA_AR-α1 subunits (C). (D) The LTP-inducing treatment does not affect either proliferation or neuronal differentiation of NSCs in the absence of neurons in the cultures. Representative images of GFP-labeled NSCs (green) immunostained BrdU or MAP2 (red). Bar graphs of the grouping data on the right (D) showing that treatment of pure NSCs in cultures with the cLTP-inducing protocol did not alter either BrdU (upper panel) or MAP2 (lower panel). *p<0.05; n = 5 in each group. Statistical analyses were performed with Student's t-test. Data are mean ± SEM.

Figure 5. Chemical LTP enhances proliferation/survival and neuronal differentiation in NSC-neuron co-cultures.

(A) Schematic representation of the experimental design. GFP-labeled NSCs were co-cultured with dissociated hippocampal neurons as described in the methods. (B) The cLTP-inducing protocol reliably induces LTP in co-cultures. Western blotting of biotinylated plasma membrane proteins revealed that in comparison with control (PBS), the cLTP-inducing treatment (cLTP) resulted in a specific increase in the level of GluA1 and GluA2 subunits of AMPARs, but not β-LRP1 (as a plasma membrane protein control), expressed on the plasma membranes 10 min after the treatment, and the increase was prevented by 50 µM NMDAR blocker APV (cLTP+APV), confirming the successful induction of NMDAR-dependent LTP. ***p<0.001, n = 7. (C) Representative Immunostaining images showing enhanced BrdU (top panel, red) and MAP2-positive (bottom panel, red) GFP-labeled NSCs following cLTP induction in co-cultures. (D) Bar graph summarizing data from multiple experiments shown in C. n = 5 and *p<0.05 for left panel; and n = 10 and ***p<0.001 for right panel. Statistical analyses were performed with One Way ANOVA. Post hoc testing revealed a significant difference between the cLTP group and other two groups (B) or three groups (D). Data are mean ± SEM.

Figure 6. Conditioned media from cLTP treated neuronal cultures increases neurogenesis of NSCs.

(A) Schematic representation of the experimental design. Conditioned media from non-, PBS- or cLTP-stimulated protocol-treated hippocampal cultures were collected and filtered before their use to treat NSCs in cultures. (B) Representative images of NSCs doubled stained with a nuclear marker DAPI (blue) and neuronal marker MAP2 (red) showing increased numbers of MAP2 (red) staining in cultured NSCs alone compared with a control group (non- and PBS-stimulated media). Data from multiple experiments were summarized in the bar graph on the right. *p<0.05, n = 6. A statistical analysis was performed with One Way ANOVA. Post hoc tests revealed a significant difference between conditioned-media stimulated group and control group. Data are mean ± SEM.

Figure 7. LTP promotes neurogenesis of NSCs in cultures at least in part through BDNF-TrkB system.

(A) cLTP increases the BDNF production in hippocampal cultures. Conditioned media was collected at 0, 10, 30, 60 min, and 1 day following cLTP induction. ELISA assays reveal a significant increase in the level of BDNF, but not NGF or NT-3, at 30 min and 1 hour following cLTP induction. *p<0.05, **p<0.01, n = 5 for each group. (B) cLTP-induced conditioned media activates membrane surface TrkB receptors in cultured NSC alone. Western blotting of surface biotinylated TrkB in NSCs sequentially probed with anti-TrkB tyrosine phosphorylation (p-TrkB) and anti-TrkB (TrkB) antibodies showing the increased level of receptor tyrosine phosphorylation by conditioned media from hippocampal cultures treated with cLTP protocol (cLTP), but not these treated with PBS (Control) and the increased phosphorylation was prevented in the presence of TrkB receptor inhibitor K252a (cLTP+200 nM). Data from 4 individual experiments were summarized in the bar graph at the bottom. **p<0.01, n = 4. (C) and (D) cLTP enhanced neurogenesis of NSCs in NSC-neuron co-cultures requires activation of TrkB receptors. Representative images (C) of GFP-labeled NSCs immunostained with MAP2 (red) illustrate cLTP increased, in a TrkB dependent fashion, both the total number of NSCs (green; summarized in the left Bar in D) and MAP2-positive NSCs (yellow; summarized in the right bar graph in D). **p<0.01, n = 6. Statistical analyses were performed with One Way ANOVA. Post hoc tests revealed a significant difference between conditioned media and other two group, and (D) the LTP group and two other groups. Data are mean ± SEM.