Functionalized carbon nanotubes in the brain: cellular internalization and neuroinflammatory responses

Abstract

The potential use of functionalized carbon nanotubes (f-CNTs) for drug and gene delivery to the central nervous system (CNS) and as neural substrates makes the understanding of their in vivo interactions with the neural tissue essential. The aim of this study was to investigate the interactions between chemically functionalized multi-walled carbon nanotubes (f-MWNTs) and the neural tissue following cortical stereotactic administration. Two different f-MWNT constructs were used in these studies: shortened (by oxidation) amino-functionalized MWNT (oxMWNT-NH3(+)) and amino-functionalized MWNT (MWNT-NH3(+)). Parenchymal distribution of the stereotactically injected f-MWNTs was assessed by histological examination. Both f-MWNT were uptaken by different types of neural tissue cells (microglia, astrocytes and neurons), however different patterns of cellular internalization were observed between the nanotubes. Furthermore, immunohistochemical staining for specific markers of glial cell activation (GFAP and CD11b) was performed and secretion of inflammatory cytokines was investigated using real-time PCR (qRT-PCR). Injections of both f-MWNT constructs led to a local and transient induction of inflammatory cytokines at early time points. Oxidation of nanotubes seemed to induce significant levels of GFAP and CD11b over-expression in areas peripheral to the f-MWNT injection site. These results highlight the importance of nanotube functionalization on their interaction with brain tissue that is deemed critical for the development nanotube-based vector systems for CNS applications.

Figure 1. Structural features of f-MWNTs.

A) Chemical structures of f-MWNTs. B) TEM images of f-MWNTs dispersed in 5% dextrose at 250 µg/ml final concentration (scale bar 100 nm).

Figure 2. Parenchymal distribution of MWNT-NH₃ ⁺and oxMWNT-NH₃ ⁺ after stereotactic administration in the motor cortex of C57BL/6 mice by light microscopy of H&E stained coronal sections, two weeks after injection with nanotubes.

A) MWNT-NH₃ ⁺; B) oxMWNT-NH₃ ⁺; C) High magnification images of the MWNT-NH₃ ⁺ injected tissue. Cortex (CX), ependymal layer (EL), and ventricles (VT) are noted, respectively. The distance from the Bregma line for each section is indicated above each image. Black arrows indicate the presence of f-MWNTs in the sections.

Figure 3. Uptake of f-MWNTs by neural tissue cells in vivo.

Transmission electron micrographs (TEM) of brain parenchyma close to the injection site: A) low magnification TEM following injection with MWNT-NH₃ ⁺. It was also possible to observe individualized MWNT-NH₃ ⁺ at the extracellular spaces. B) low magnification TEM following injection with oxMWNT-NH₃ ⁺ mainly localized intracellularly in clusters and within vesicles. C) and D) show TEM images of increasing magnification of a neural tissue cells containing MWNT-NH₃ ⁺ and oxMWNT-NH₃ ⁺ intracellularly. i (C & D) low magnification TEM of a neural cell containing intracellular CNT. ii and iii (C&D) shows higher magnification of the cytoplasm of the cell imaged on pannel i (C&D). iv shows higher magnification of cytoplasm of cell imaged on panel i, being able to identify the presence of more individualized CNT (iv C) or clusters and within vesicle CNT (iv D).

Figure 4. Neuroinflammatory response following intra-cortical injection of f-MWNTs.

A) Panel of cytokine expression profiles in C57BL/6 mouse cortex following stereotactic injection of 5% dextrose solution, LPS, MWNT-NH₃ ⁺ and oxMWNT-NH₃ ⁺ (n = 4 for each group). mRNA expression levels were assessed using quantitative rtPCR. Relative gene expression was calculated using the 2-ΔΔCT method. Relative gene expression levels (plotted as fold increases) were calculated by first normalizing the values for each primer transcript against the house-keeping gene β-actin, and then further normalized against the 5% dextrose control group, giving relative gene expression values (fold-change (mean ± SD)). Both f-MWNTs caused a transient increase in the pro-inflammatory cytokine expression levels, peaking at 16 hr. oxMWNT-NH₃ ⁺ induced significantly higher expression levels of TNF-α and IL-1β at 16 hr compared to MWNT-NH₃ ⁺ (p<0.001). B) Glia activation studies 30 days after stereotactic injection with 5% dextrose saline solution, MWNT-NH₃ ⁺, oxMWNT-amide-NH₃ ⁺ and oxMWNT-NH₃ ⁺ (n = 3 for each group). Histograms of green (GFAP) or red (CD11b) positive pixel were analysed. The positive pixels at lower intensity (5% of the total pixels) were excluded in order to avoid background noise.The intensity was measured and quantified in an area 1×0.5 mm surrounding the injection site.