PEGylated squalenoyl-gemcitabine nanoparticles for the treatment of glioblastoma

Abstract

New treatments for glioblastoma multiforme (GBM) are desperately needed, as GBM prognosis remains poor, mainly due to treatment resistance, poor distribution of therapeutics in the tumor tissue, and fast metabolism of chemotherapeutic drugs in the brain extracellular space. Convection-enhanced delivery (CED) of nanoparticles (NPs) has been shown to improve the delivery of chemotherapeutic drugs to the tumor bed, providing sustained release, and enhancing survival of animals with intracranial tumors. Here we administered gemcitabine, a nucleoside analog used as a first line treatment for a wide variety of extracranial solid tumors, within squalene-based NPs using CED, to overcome the above-mentioned challenges of GBM treatment. Small percentages of poly(ethylene) glycol (PEG) dramatically enhanced the distribution of squalene-gemcitabine nanoparticles (SQ-Gem NPs) in healthy animals and tumor-bearing animals after administration by CED. When tested in an orthotopic model of GBM, SQ-Gem-PEG NPs demonstrated significantly improved therapeutic efficacy compared to free gemcitabine, both as a chemotherapeutic drug and as a radiosensitizer. Furthermore, MR contrast agents were incorporated into the SQ-Gem-PEG NP formulation, providing a way to non-invasively track the NPs during infusion.

Keywords: Convection-enhanced delivery; Gemcitabine; Glioblastoma; Nanoparticles; Squalene.

Fig. 1. Physico-chemical characterization of SQ-Gem NPs and SQ-Gem-PEG NPs

(A) The hydrodynamic diameter of the different formulations was measured using dynamic light scattering (DLS). All formulations presented a monodisperse population (PdI < 0.2), with a diameter below 150 nm, while the addition of SQ-PEG decreased the diameter of the particles. (B) The surface charge (zeta potential) was measured in 10 mM NaCl. The addition of SQ-PEG rendered the formulations nearly neutral (ZP ≈ − 1 mV). (C) Size stability was measured by DLS in aCSF at 37°C for 5 h. The diameter of the unmodified SQ-Gem NPs immediately increased upon dilution in aCSF, and kept dramatically increasing over time, while all formulations containing SQ-PEG were stable.

Fig. 2. Distribution of SQ-Gem NPs and SQ-Gem-PEG NPs after administration to healthy rats by CED

(A) Volumes of distribution (V_d) were calculated using fluorescent based volumetric reconstruction. All formulations incorporating SQ-PEG presented a V_d superior to 50 mm³. (B–G) Representative images of the distribution of the different NPs formulations after administration by CED. Unmodified SQ-Gem NPs (B) aggregated at the injection site, while all SQ-Gem-PEG NPs formulations (C − 5%, D − 10%, E − 30%, F − 50%, G − 70%) distributed uniformly throughout the healthy striatum.

Fig. 3. USPIO loaded SQ-Gem-PEG NPs for particle tracking during CED

(A) The hydrodynamic diameter of USPIO loaded SQ-Gem-PEG 70% NPs was measured using DLS. The formulation presented a monodisperse population (PdI < 0.2), with similar diameter than SQ-Gem-PEG 5% NPs and SQ-Gem-PEG 70% NPs. (B) The surface charge (zeta potential) was measured in NaCl 10 mM. The addition of USPIO slightly decreased the surface charge compared to USPIO unloaded formulations. (C) Size stability was measured by DLS in aCSF at 37°C for 5 h, and it was observed that the incorpor ation of USPIO did not compromise the stability of the formulation. (D) V_d was calculated using fluorescent-based volumetric reconstruction following CED of fluorescently labeled, USPIO loaded NPs. The incorporation of USPIO did not prevent the NPs from distributing widely throughout the brain, yielding a V_d comparable to USPIO unloaded PEGylated formulations. (E–H) Representative images of the distribution of the different formulations after administration by CED. Unmodified SQ-Gem NPs (E) aggregated at the injection site, while all the SQ-Gem-PEG NPs formulations (F − 5%, G − 70%), including those incorporating USPIO (H), distributed uniformly throughout the healthy striatum.

Fig. 4. In vitro uptake and cytotoxicity of SQ-Gem and SQ-Gem-PEG NPs

(A) NP uptake by U87 cells was measured by flow cytometry for up to 24 h. After 24 h of incubation, internalization of unmodified SQ-Gem NPs (B) and SQ-Gem-PEG 70% (C) was visualized by confocal microscopy, confirming reduced uptake of PEGylated NPs (scale bar = 20 µm). (D–E) Cytotoxicity of SQ-Gem NPs and SQ-Gem-PEG 70% NPs was evaluated in U87 (D) and RG2 (E) cells using an MTT assay, demonstrating that PEGylated NPs were as potent as unmodified NPs, and that they were both as cytotoxic as the free drug.

Fig. 5. Distribution of SQ-Gem NPs and SQ-Gem-PEG NPs after administration to tumor bearing rats by CED

(A) Volumes of distribution (V_d) after CED into brains with RG2 tumors of around 4 mm³ were calculated using fluorescent-based volumetric reconstruction. SQ-Gem-PEG 5% NPs yielded a significantly higher V_d compared to unmodified SQ-Gem NPs. (B) Percentages of tumor coverage in small tumors were calculated for both formulations. (C–D) Representative images of the co-localization of the NPs with the tumor mass (C – unmodified SQ-Gem NPs, D – SQ-Gem-PEG 5% NPs). The tumor appears in green, the NPs in red, and the colocalized area in yellow.

Fig. 6. Anticancer activity of SQ-Gem-PEG NPs in rats bearing RG2 glioma

Anticancer activity of SQ-Gem-PEG 5% NPs when administered by CED 4 days after RG2 tumor implantation was evaluated without (A) and with (B) additional RT (two doses of 5 Gy, 24 h and 48 h after CED). (C–D) Survival of the animals, when treated without (C) and with (D) additional RT. In both settings, SQ-Gem-PEG 5% NPs were more potent than free Gem as they significantly increased the mean survival time.