Graphene-Derived Materials Interfacing the Spinal Cord: Outstanding in Vitro and in Vivo Findings

Abstract

The attractiveness of graphene-derived materials (GDMs) for neural applications has fueled their exploration as components of biomaterial interfaces contacting the brain and the spinal cord. In the last years, an increasing body of work has been published on the ability of these materials to create biocompatible and biofunctional substrates able to promote the growth and activity of neural cells in vitro and positively interact with neural tissues when implanted in vivo. Encouraging results in the central nervous tissue might impulse the study of GDMs towards preclinical arena. In this mini-review article, we revise the most relevant literature on the interaction of GDMs with the spinal cord. Studies involving the implantation of these materials in vivo in the injured spinal cord are first discussed, followed by models with spinal cord slides ex vivo and a final description of selected results with neural cells in vitro. A closing debate of the major conclusions of these results is presented to boost the investigation of GDMs in the field.

Keywords: graphene; in vivo models; neurons; scaffold; spinal cord injuries; toxicity.

Figure 1

(A) Mature blood vessels were detected inside reduced graphene oxide (RGO) scaffolds at 30 days post-injury by labeling RECA-1 (green) and laminin from basal membranes (LAM, red). (B) Axons were specifically detected inside RGO scaffolds at 30 days post-injury by expression of β-III tubulin (TUB, green) and neurofilaments (NF, white). Laminin from basal membranes was labeled in red (LAM). Arrow heads indicate areas with coexistence of TUB and NF (red; MERGE 1) and TUB with LAM (white; MERGE 2). Cell nuclei were labeled with Hoechst (HOE). Bright field images (BF) are included to visualize the scaffold structure. Scale bar: 100 μm. Adapted from López-Dolado et al. (2016), Copyright (2016), with permission from Elsevier.

Figure 2

Exposure to small graphene oxide nanosheets (s-GO) at high concentration impaired excitatory synapses. In (A), sample tracings of mPSCs recorded in control and s-GO-treated cultures (left panel) and plots reporting mPSC amplitude and frequency values (right panel) (***p < 0.001, Student’s-t-test). In (B), confocal reconstruction of control and s-GO-treated neurons immunolabeled for the vesicular glutamate transporter 1 (VGLUT1; green) and counterstained for cytoskeletal component β tubulin III (red; nuclei are visualized by DAPI labeling in blue; scale bar 10 μm) (***p < 0.001, Student’s-t-test). In (C), fluorescence images following staining with FM1–43, control and s-GO-treated (scale bar: 50 μm) (top). The areas in the boxes are higher magnifications to highlight the difference in vesicular staining between the two conditions (scale bar: 10 μm). The plot (top right) reproduces the representative (control and s-GO) traces of FM1–43 de-staining (each trace has been normalized to the maximum fluorescence detected). Bottom: the left plot summarizes the initial raw fluorescent intensities of hippocampal terminals from control and s-GO-treated cultures (**p < 0.01, Mann-Whitney test); the right plot summarizes the decay time constant τ of FM1–43 de-staining in the two conditions (***p < 0.001, Mann-Whitney test). Reprinted with permission from Rauti et al. (2016). Copyright (2016) American Chemical Society.