Breaking Barriers in Glioblastoma Targeting through Advanced Nanoparticle Cell Membrane Coating

Abstract

Glioblastoma multiforme (GBM) is the most common and aggressive malignant brain tumor characterized by poor prognosis and limited treatment options. Despite current therapies combining surgery, radiotherapy, and chemotherapy, GBM remains highly resistant to treatment, largely due to the challenges of drug delivery across the blood-brain barrier (BBB). Nanoparticles (NPs) have shown promise as drug carriers, but their clinical translation is hindered by limited brain accumulation and rapid clearance by the immune system. In this study, we explored the potential of GBM cell membrane (CM)-coated NPs (G-NPs) as a strategy to improve GBM targeting and, therefore, efficient treatments. We optimized the CM isolation protocol using U87-MG human GBM cells and identified the Heidolph homogenizer as the most effective technique for producing pure, enriched CM fractions, proposing it as a standard method due to its high scalability. G-NPs were extensively characterized, demonstrating excellent colloidal stability under biological conditions. Flow cytometry revealed the enhanced uptake of G-NPs by U87-MG cells compared to non-coated NPs. Notably, the specific homotargeting capability of G-NPs toward human glioblastoma cells was ultimately confirmed by demonstrating a marked specificity of the glioblastoma CM coating when compared to human fibroblast CM-coated NPs, highlighting selective tumor cell-type targeting. Additionally, the coating of NPs with GBM CMs not only did not impede the physiological passage of NPs across the human in vitro BBB, but interestingly, increased the BBB permeability to G-NPs. These findings highlight that biomimetic coating of NPs with GBM cells is a potential strategy to create platforms for the targeted chemotherapy of GBM.

Keywords: blood-brain barrier; cell membrane; coating; glioblastoma; nanoparticles.

1

Cell membrane isolation protocols. (a) Western Blotting of CM P17 and P100 fractions obtained from the different CM isolation protocols. The molecular weight marker (MW) is shown as well as the control sample (total cell lysate before centrifugation). The 100 kDa band (within the red square) indicates the cell membrane marker (Anti-Na⁺/K⁺ ATPase antibody), the 75 kDa band shows the endoplasmic reticulum marker (Anti-GRP78 antibody), the 55 kDa band shows the mitochondrial membrane marker (Anti-ATP5a antibody), the 37 kDa band shows the cytosolic marker (Anti-GAPDH antibody), and the 17 kDa band shows the nucleus marker (Anti-Histone H3 (dimethyl K9) antibody). (b) Quantitative measurement of 100 kDa bands (cell membranes, highlighted in a) from different CM isolation protocols represented as the mean values of 3 different experiments ± SEM. The control density value was normalized to 1.

2

Characterization of NP coating. (a) Size (nm) of CMs, PS-NPs, and G-NPs by DLS in Milli-Q water at RT. (b) Z-potential (mV) values of CMs, PS-NPs, and G-NPs in water at RT. Data are represented as the mean average of 4 independent replicates ± SEM. (c) TEM micrographs of PS-NPs and G-NPs showing the CM coating layers onto the G-NPs surface.

3

Colloidal stability of PS-NPs and G-NPs. (a) Z-potential and (b) size of PS-NPs and G-NPs across a pH range from 4 to 9. (c) Size of PS-NPs and G-NPs in cDMEM, SF-DMEM, and PBS after 24 and (d) 72 h of incubation. The standard deviation ± is displayed (3 replicate measurements).

4

Assessment of biomolecular corona formation. (a) The Z-potential and the (b) size of PS-NPs are shown along a pH range, comparing bare PS-NPs, soft corona PS-NPs, and hard corona PS-NPs. (c) The Z-potential and the (d) size of G-NPs are shown along a pH range, again comparing bare PS-NPs, soft corona PS-NPs, and hard corona PS-NPs. The standard deviation is displayed (3 replicate measurements).

5

PS-NPs and G-NPs bionanointeractions with U87-MG cells measured by flow cytometry and confocal microscopy. (a) Flow cytometry. Statically significant differences (T-student mean comparison test p < 0.05) are highlighted with “*”. The SEM is displayed (4 independent replicates). (b) (c) PS-NPs and G-NPs bionanointeractions at 1 and 24 h (d, e) and their orthogonal views (XZ and YZ) under confocal microscopy. Confocal microscopy samples were stained with Hoechst (blue nucleus), phalloidin (red cytoskeleton), and FITC- (green NPs).

6

Transport Study of PS-NPs and G-NPs across the in vitro BBB model. (a) Schematic representation of the passing assay across the in vitro BBB model (created in BioRender.com). (b) Normalized Transported Mass and Apparent Permeability of PS-NPs and G-NPs through the in vitro BBB model at 5h. Mean ± standard deviation is displayed (9 replicate measurements). Statically significant differences (T-student mean comparison test p < 0,05) are highlighted with “*”.