PTEN regulates AMPA receptor-mediated cell viability in iPS-derived motor neurons

Abstract

Excitatory transmission in the brain is commonly mediated by the α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptors. In amyotrophic lateral sclerosis (ALS), AMPA receptors allow cytotoxic levels of calcium into neurons, contributing to motor neuron injury. We have previously shown that oculomotor neurons resistant to the disease process in ALS show reduced AMPA-mediated inward calcium currents compared with vulnerable spinal motor neurons. We have also shown that PTEN (phosphatase and tensin homolog deleted on chromosome 10) knockdown via siRNA promotes motor neuron survival in models of spinal muscular atrophy (SMA) and ALS. It has been reported that inhibition of PTEN attenuates the death of hippocampal neurons post injury by decreasing the effective translocation of the GluR2 subunit into the membrane. In addition, leptin can regulate AMPA receptor trafficking via PTEN inhibition. Thus, we speculate that manipulation of AMPA receptors by PTEN may represent a potential therapeutic strategy for neuroprotective intervention in ALS and other neurodegenerative disorders. To this end, the first step is to establish a fibroblast-iPS-motor neuron in vitro cell model to study AMPA receptor manipulation. Here we report that iPS-derived motor neurons from human fibroblasts express AMPA receptors. PTEN depletion decreases AMPA receptor expression and AMPA-mediated whole-cell currents, resulting in inhibition of AMPA-induced neuronal death in primary cultured and iPS-derived motor neurons. Taken together, our results imply that PTEN depletion may protect motor neurons by inhibition of excitatory transmission that represents a therapeutic strategy of potential benefit for the amelioration of excitotoxicity in ALS and other neurodegenerative disorders.

Figure 1

Characterization of human dermal fibroblasts used for hiPSC induction. (a) Culture of human dermal fibroblasts (bar=50 μm). (b) TE-7 immunostaining confirmed that the resulting cultures were of fibroblast identity (bar=100 μm). (c) Phase contrast image of human dermal fibroblasts transfected with GFP retroviruses (bar=100 μm). (d) The hESC-like hiPS colony (bar=50 μm). (e and f) These hESC-like colonies showed high alkaline phosphatase expression (bar=50 μm). (g) Immunocytofluorescence of the hiPSCs showing the expression of multiple hESC markers (bar=100 μm). (h) RT-PCR analysis of hESC marker genes in hiPSCs. (i) No karyotypic abnormalities were observed in the hiPSCs. (j) Open circles indicate unmethylated CpG dinucleotides, whereas closed circles indicate methylated CpGs in the promotor of Oct4. After transplantation into the hind limb muscle of NOD/SCID mice, hiPSCs generated teratomas showing (k) neural tissue (ectoderm; bar=100 μm), (l) cartilage (mesoderm; bar=100 μm), (m) muscle (mesoderm; bar=100 μm) and (n) primitive gut (endoderm; bar=100 μm)

Figure 2

Generation and characterization of spinal motor neurons from hiPSCs. (a) Schematic drawing of the experimental setup and strategy to derive induced spinal motor neurons. (b) When differentiated for 12 days, the hiPSC-derived primitive neuroepithelial cells are positive for OTX2 (red) and PAX6 (green), bar=25 μm. (c) When differentiated for 35 days, the hiPSC-derived motor neurons are positive for Islet 1/2 (red), bar=50 μm. (d) When differentiated for 45 days, hiPSC-derived motor neurons become mature, and are positive for ChAT (green), bar=50 μm. DAPI shown as blue denotes DAPI-stained nuclei

Figure 3

PTEN knockdown (b and f) increases phosphorylation of AKT (c and g) and Bad (d and h) in primary cultured motor neurons and iPS-derived motor neurons. Primary (a) or iPS-derived (e) motor neurons were immunostained with anti-tubulin (green), Hoechst (nuclear staining in blue) and PTEN (red). Scale bar, 20 μm. Western blotting on homogenates from primary cultured (b–d) and iPS-derived motor neurons. PTEN is normalized by GAPDH (b and f) and pAKT is normalized by AKT (c and g). pBAD is normalized by BAD (d and h). Results represent the mean+S.E.M. from four independent experiments. **P<0.01, tested by one-way ANOVA

Figure 4

PTEN knockdown decreases AMPA receptor expression in primary cultured (a–f) and iPS-derived (g–l) motor neurons. Motor neurons were immunostained with anti-tubulin (green), Hoechst (nuclear staining in blue) and AMPA receptor (GluR1, red). Scale bar, 20 μm. (b–f and h–l) Western blotting. PTEN knockdown decreases GluR1 (c and i) and GluR2 (d and j) but not GluR4 expression in both primary cultured and iPS-derived MNs. GluR3 decreases significantly in iPS-derived MNs (k) but not in primary cultured MNs (e). Results represent the mean+S.E.M. from four independent experiments. **P<0.01, tested by one-way ANOVA

Figure 5

PTEN knockdown decreases AMPA-induced whole-cell currents in primary cultured (a–d) and iPS-derived (e–h) motor neurons. Primary cultured (a) and iPS-derived (e) motor neurons were immunostained with anti-tubulin (green), Hoechst (nuclear staining in blue) and ChAT (red). Scale bar, 20 μm. (b and f) Representative current traces elicited by AMPA. (c and g) Dose response of AMPA-induced whole-cell currents were recorded in 20 mM extracellular Na+ at −60 mV, in response to AMPA concentrations ranging from 5 μM to 5 mM. Each point represents the mean±S.E.M. from three cells (*P<0.05, **P<0.01, tested by one-way ANOVA). EC₅₀ values estimated from fits to pooled data were 106 μM for primary cultured motor neurons and 113 μM for iPS-derived neurons. (d and h) Average peak AMPA-induced whole-cell currents were recorded in Na+-free extracellular solution containing 50 mM Ca²⁺ at −60 mV in primary cultured and iPS-derived motor neurons, evoked by AMPA 100 μM (n=8). Columns represent the peak amplitudes of agonist-induced whole-cell currents. Results represent the mean+S.E.M. from four independent experiments, with 8 motor neurons of each (*P<0.05, **P<0.01, tested by one-way ANOVA)

Figure 6

PTEN knockdown decreases AMPA-induced cell death in primary cultured and iPS-derived motor neurons. Primary cultured (a) and iPS-derived (c) motor neurons were immunostained with anti-tubulin (green), Hoechst (nuclear staining in blue) and TUNEL staining (red). Scale bar, 20 μm. (b and d) Results represent the mean+S.E.M. from four independent experiments. **P<0.01, tested by one-way ANOVA