Transient muscarinic and glutamatergic stimulation of neural stem cells triggers acute and persistent changes in differentiation

Abstract

While aberrant cell proliferation and differentiation may contribute to epileptogenesis, the mechanisms linking an initial epileptic insult to subsequent changes in cell fate remain elusive. Using both mouse and human iPSC-derived neural progenitor/stem cells (NPSCs), we found that a combined transient muscarinic and mGluR1 stimulation inhibited overall neurogenesis but enhanced NPSC differentiation into immature GABAergic cells. If treated NPSCs were further passaged, they retained a nearly identical phenotype upon differentiation. A similar profusion of immature GABAergic cells was seen in rats with pilocarpine-induced chronic epilepsy. Furthermore, live cell imaging revealed abnormal de-synchrony of Ca(++) transients and altered gap junction intercellular communication following combined muscarinic/glutamatergic stimulation, which was associated with either acute site-specific dephosphorylation of connexin 43 or a long-term enhancement of its degradation. Therefore, epileptogenic stimuli can trigger acute and persistent changes in cell fate by altering distinct mechanisms that function to maintain appropriate intercellular communication between coupled NPSCs.

Keywords: Connexin; Epileptogenesis; Glutamate receptor; Neural stem cell; Pilocarpine.

Fig 1. Increased GABAergic differentiation and decreased neurogenesis in Pilo+Glut exposed NPSCs

P2 NPSCs were subjected to Phosphate Buffer Saline (PBS; Veh), 4μM pilocarpine (Pilo), 30μM glutamate (Glut) or dual pilocarpine and glutamate (Pilo+Glut) treatment. Cells were incubated for 22 hours with Pilo followed by 2 hours of Glut. P2 NPSCs were then either immediately subjected to 10 days of differentiation (P2) or passaged an additional 3 times and then differentiated (P5-memory). (a–b) Mean cell counts +SEM from differentiated NPSCs expressing Tuj-1 (a) or GAD67 (b); cells treated as indicated while proliferating. (n=3 independent experiments, counts are an average obtained from four random fields per image; 1-way ANOVA, p=0.0032 for Tuj-1; P=0.0019 for GAD67; post hoc Tukey’s Multiple Comparison Test, *= P<0.05). (c) Representative IIF image of differentiated P2 NPSCs. (d) qRT-PCR analysis of GAD67 expression in differentiated P2 NPSCs (n=4; error bars=SEM, 1-way ANOVA, P<0.0001; post hoc Tukey’s Multiple Comparison Test, *=P<0.05). (e–f) Mean cell counts +SEM from P5-memory differentiated NPSCs expressing Tuj-1 (e) or GAD67 (f); cells treated as indicated while proliferating at P2. (n=3; 1-way ANOVA, P=0.0260 for Tuj-1; P=0.0280 for GAD67; post hoc Tukey’s Multiple Comparison Test, *=P<0.05). (g) Representative IIF images of differentiated P5-memory NPSCs. (h) qRT-PCR analysis of GAD67 expression in differentiated P5 memory NPSCs (n=4; error bars=SEM; 1-way ANOVA, P=0.0025; post hoc Tukey’s Multiple Comparison Test, *= P<0.05). (i–j) Mean cell counts +SEM from P5-memory differentiated NPSCs expressing Tuj-1 (i) or GAD67 (j); cells treated as indicated while proliferating at P2 (i.e. 3 μM AIDA or 10μM MK801 30 minutes prior to Glut). (n=4; error bars = SEM, 1-way ANOVA, P<0.0001; post hoc Tukey’s Multiple Comparison Test, *=P<0.05). (k) Representative IIF images of differentiated P5-memory NPSCs.

Fig 2. Increased immature GABAergic phenotype in the hippocampus of chronic TLE rats and in differentiated Pilo+Glut stimulated NPSCs

(a) SOX2/GAD67 co-localization, as measured by the Pearson’s Correlation Coefficient, from IIF images (b) of hippocampi of young adult rats (3 weeks old) injected IP with Veh or Pilo (Chronic TLE) (n=6; student’s t-test; *=p<0.05). (c) Mean cell counts +SEM from differentiated P2 NPSCs expressing Sox-2; cells treated as indicated while proliferating (n=3; 1-way ANOVA, P=0.0005; post hoc Tukey’s Multiple Comparison Test, *=P<0.05). (d) Representative IIF images of differentiated P2 NPSCs. (e) Mean cell counts +SEM from differentiated P5-memory NPSCs expressing Sox-2; cells treated as indicated at P2 while proliferating (n=3; 1-way ANOVA, P=0.0073; post hoc Tukey’s Multiple Comparison Test, *= P<0.05). (f) Representative IIF images of differentiated P5-memory NPSCs. (g) SOX2/GAD67 co-localization in P2 and P5-memory NPSCs treated as indicated, as measured by the Pearson’s Correlation Coefficient (inset, representative image from differentiated Pilo+Glut treated P2 and P5-memory NPSCs). (h) qRT-PCR analysis of NKCC1 and KCC2 expression in differentiated P2 and P5-memory NPSCs treated with Pilo and Glut (n=3). (i) IIF staining of differentiated P2 NPSCs transiently transfected while proliferating with either an siRNA directed against mGluR1 (si mGluR) or a scrambled siRNA (Scr si) and also treated as indicated while proliferating. Identical results were obtained in a separate biological replicate transfection.

Fig 3. mGluR1 dependent calcium desynchrony in Pilo+Glut stimulated NPSCs

Representative plots (a,b) of live cell intracellular Ca²⁺ flux stimulated by treatment with glutamate where indicated (red arrow), measured using the ratiometric calcium indicator Fura-2 AM. Each line in the graphs represents a single NPSC within a neurosphere. Ca²⁺ traces from P2 NPSCs treated with Veh (a) or Pilo+Glut (b). (c) Pearson’s Correlation Coefficient +SEM from P2 NPSCs treated with Veh, Pilo, Glut or PIlo+Glut (n=3; 1-way ANOVA, P<0.0001; post hoc Tukey’s Multiple Comparison Test, *=P<0.05). (d) Pearson’s Correlation Coefficient +SEM from P5-memory NPSCs treated at P2 with Veh, Pilo, Glut or Pilo+Glut (n=3; 1-way ANOVA, P<0.0001; post hoc Tukey’s Multiple Comparison Test, *=P<0.05; **=P<0.01; ***=P<0.005). Pearson’s Correlation Coefficient +SEM from P2 NPSCs (e) or P5-memory NPSCs treated at P2 (f) with Pilo+Glut, MK801+Pilo+Glut or AIDA+Pilo+Glut (n=3; 1-way ANOVA, P<0.0001; post hoc Tukey’s Multiple Comparison Test, *=P<0.05). Pearson’s Correlation Coefficient of live cell calcium synchrony between individual cells in P2 NPSCs (g) or P5-memory NPSCs (h) transfected with mGluR1 siRNA (SimGluR1) or scrambled siRNA (Scr) at P2 and treated with Veh or Pilo+Glut at P2 (n=3; 1-way ANOVA, P<0.0001; Tukey’s Multiple Comparison Test, *P=<0.05).

Fig 4. Decreased Cx43 phosphorylation in Pilo+Glut stimulated P2 NPSCs is associated with enhanced PP2B activity and increased GJIC

Representative Western blots (a,c) and densitometric scans (b,d) of total Cx43 (a,b) and pCx43s368 in P2 NPSCs treated as indicated (d; n=3 independent experiments; 1-way ANOVA, P=0.0358; post hoc Tukey’s Multiple Comparison Test, *=P<0.05). (e) PP2B activity as assessed by a phosphate release assay in P2 NPSCs treated with vehicle or Pilo+Glut (n=3; student’s t-test; p=0.0409). Representative FRAP curves (f) from Calcein AM loaded P2 NPSCs treated as indicated with Box plots for t1/2 +/− SEM (g,h) shown of various treatment groups (n=5; g. Kruskal Wallis test p=0.0129; post hoc Dunn’s Multiple Comparison Test *=P<0.05, h. Kruskal Wallis test p=0.0057; post hoc Dunn’s Multiple Comparison Test *=P<0.05).

Fig 5. Pilo+Glut and Glut only P5-Memory NPSCs have lower total Cx43 protein levels and decreased GJIC

Representative Western blots (a,b) and densitometric scans (c,d) of total Cx43 in hippocampi of young adult rats 6 months following IP injection with vehicle or Pilo (c; n=3; student’s t-test; *=p<0.01) or P5-memory NPSCs treated as indicated during P2 (d; n=3 independent experiments; 1-way ANOVA, P=0.0015; post hoc Tukey’s Multiple Comparison Test, P<0.05). Representative Western blot of Cx43 protein half-life (e) or densitometric scans of Western blots (g) of total Cx43 from lysates of P5-memory NPSCs treated with Veh or Pilo+Glut at P2 and treated at P5 for 1, and 3hrs with the protein synthesis inhibitor cycloheximide (g; n=3, student’s t-test; *=p<0.05). P5-memory NPSCs treated at P2 as indicated were preloaded with Calcein AM and subjected to FRAP analysis (f). Box plots for t1/2 +/− SEM (n=6; Kruskal Wallis test p=0.0029; post hoc Dunn’s Multiple Comparison Test *=P<0.05).

Fig 6. Pilo+Glut treated human IPSC derived NPSCs demonstrate decreased neurogenesis and an enhanced immature GABAergic phenotype

(a) Representative IIF images of differentiated NPSCs generated from human IPSCs, subject to treatment with Veh, Pilo+Glut, or AIDA+Pilo+Glut. Mean cell counts +SEM of Tuj-1 (b) and GAD67 (c) staining following IIF from differentiated human NPSCs treated as indicated (n=3; 1-way ANOVA, p=0.001 for Tuj-1; P=0.0001 for GAD67; post hoc Tukey’s Multiple Comparison Test, **= P<0.01; ***=P<0.005).