Electroconductive nanoscale topography for enhanced neuronal differentiation and electrophysiological maturation of human neural stem cells

Background

Nano- and microscale topography can guide stem cell differentiation through contact guidance, integrin binding and focal adhesion signalling, but the authors note that simple topographical cues alone are often insufficient to drive the functional maturation of neurons derived from stem cells. Electrical cues, delivered through conductive polymers, carbon nanotubes or metal deposits, also promote neurite outgrowth and neuronal differentiation. The study asks whether a single substrate that provides defined nanotopography and conductivity, and can carry an applied current, promotes neuronal differentiation and electrophysiological maturation of hNSCs without neurotrophic supplements.

Design and approach

Ridge/groove nanopatterns were moulded in PUA by capillary force lithography (CFL) and coated with 30 nm of Ti by electron-beam evaporation, chosen for its conductivity and biocompatibility. The resulting Ti-coated nanopatterned substrate (TNS) presents grooves for contact guidance and a conductive surface; for active stimulation, wires fixed at both ends of the TNS deliver pulsed current parallel to the grating axis. The substrates were coated with fibronectin (FN) before cell seeding.

How it was evaluated

hNSCs isolated from human fetal telencephalic tissue at 13 weeks of gestation were cultured without basic fibroblast growth factor (bFGF) and leukemia inhibitory factor (LIF) on flat and nanopatterned PUA (FS, NS), Ti-coated flat substrates (TFS) and TNS. Neuronal and astrocytic differentiation was assessed by immunostaining and quantitative PCR (qPCR) for Tuj1, MAP2, NeuN, GFAP and neuronal subtype markers at 5 or 12 days, mechanotransduction by vinculin and F-actin staining, focal adhesion kinase (FAK) phosphorylation and inhibitors of β1 integrin, myosin II, Rho-associated protein kinase (ROCK) and MEK1/2, and function by whole-cell patch clamp after 5 days. Pulsed stimulation (1 Hz, 30 min twice a day, with current and voltage limited to 3 µA and 25 V) started 1 day after seeding on TFS and TNS (ETFS, ETNS). All experiments were in vitro.

Key results

Atomic force and scanning electron microscopy showed ridge/groove widths of 150, 200, 250 and 300 nm with a pattern height of 300 nm, and the Ti layer covered the grooves without significantly altering the topography; the abstract instead lists a groove thickness of 150 μm. The conductivity of the 150 nm TNS was 1250 ± 68 S cm−1 bare and 916 ± 23 S cm−1 after FN coating, whereas uncoated PUA had zero conductivity.

Without applied current, TNS with 150 and 200 nm grooves increased the number of Tuj1-positive cells and decreased GFAP-positive cells relative to TFS. After 12 days, NeuN-positive cells made up 86.2 ± 1.5% on the 150 nm TNS compared with 72.6 ± 3.1% on TFS, and Tuj1 expression was 4-fold higher on TNS than on the uncoated nanopattern of the same geometry, with no significant difference in GFAP. In the results section, TNS shifted differentiation towards GABAergic and dopaminergic rather than glutamatergic lineages, with higher GAD67 and TH and lower GluT expression after 12 days, whereas the abstract names dopaminergic and glutamatergic neurons. Neuron-like cells on TNS produced sodium currents and action potentials that disappeared with tetrodotoxin (TTX); 19.6% of patched cells generated sodium currents on TNS compared with 13.9% on TFS, and the difference in average current amplitude was not statistically significant.

Vinculin and FAK expression was higher on NS and TNS than on FS and TFS, which the authors take to indicate that nanotopography may be more critical than conductivity for focal adhesion formation, and FAK phosphorylation was higher on TNS than on TFS. Blocking β1 integrin inhibited focal adhesion assembly on TNS, ROCK inhibition with Y27632 and myosin II inhibition with blebbistatin dissipated F-actin alignment and reduced differentiation markers, and MEK-ERK inhibition with U0126 completely reversed the effect of TNS on neuronal differentiation. The differentiation readout of these inhibitor experiments (Tuj1 expression) was taken at about 1 day; at 5 days, Y27632 increased Tuj1 and MAP2 expression on TFS.

After 5 days, MAP2-positive cells were 24.6 ± 2.0% on FS, 37.1 ± 2.1% on NS, 40.6 ± 0.7% on TFS, 48.4 ± 2.1% on TNS, 48.9 ± 2.6% on ETFS and 56.2 ± 1.3% on ETNS, and ETNS increased Tuj1 and MAP2 expression by 2 to 2.5 fold compared with FS. Stimulation did not increase GFAP or change Nestin, Oct4 and Nanog, and its effect was larger on TFS than on TNS, with no significant difference in Tuj1 and MAP2 expression between ETFS and ETNS. On ETNS, expression of the voltage-gated channels CACNA1C and SCN1α increased, calcium imaging showed calcium influx and depolarization during stimulation, and ERK1/2 phosphorylation rose; the authors speculate that stimulation acts through voltage-gated ion channels and ERK1/2 signalling.

Open questions

The authors did not examine how the depth of the grating grooves affects differentiation, and they note that the inhibitor experiments were read at an early time point. They state that the long-term effects of electrical stimulation and the mechanisms underlying the enhanced differentiation need further study, and that the system must be scaled up, with better nanofabrication and reproducibility, before it could supply neuron-like cells in larger amounts.

How it connects to other research

The related graphene study examines hNSC adhesion and neuronal differentiation on a conductive two-dimensional carbon material, whereas this work combines defined nanogratings with a thin Ti coating and passes pulsed current through that coating to separate topographical, conductive and electrical cues.