Convection-enhanced delivery of liposomal drugs for effective treatment of glioblastoma multiforme

Abstract

The blood-brain barrier (BBB) impedes the efficient delivery of systemically administered drugs to brain tumors, thus reducing the therapeutic efficacy. To overcome the limitations of intravascular delivery, convention-enhanced delivery (CED) was introduced to infuse drugs directly into the brain tumor using a catheter with a continuous positive pressure. However, tissue distribution and retention of the infused drugs are significantly hindered by microenvironmental factors of the tumor such as the extracellular matrix and lymphatic drainage system in the brain. Here, we leveraged a liposomal formulation to simultaneously improve tissue distribution and retention of drugs infused in the brain tumor via the CED method. Various liposomal formulations with different surface charge, PEGylation, and transition temperature (Tm) were prepared to test the cellular uptake in vitro, and the tissue distribution and retention in the brain. In in vitro studies, PEGylated liposomal formulations with a positive surface charge and high Tm showed the most efficient cellular uptake among the tested formulations. In in vivo studies, the liposomal formulations were infused directly into the brain via the CED method. PEGylated liposomal formulations with a positive surface charge and high Tm showed more efficient distribution and retention in both normal and tumor tissues while only-PEGylated formulations displayed rapid clearance from the tissues to cervical lymph nodes. Furthermore, we demonstrated that the CED of liposomal everolimus prepared with the PEGylated formulation with a positive surface charge and high Tm resulted in superior therapeutic effects for glioblastoma treatment compared to other formulations.

Graphical abstract

Introduction

Glioblastoma multiforme (GBM) originates from the astrocyte, which accounts for the majority of cellular composition in the brain. The World Health Organization (WHO) classifies GBM as a grade IV astrocytoma because it is the most aggressive malignant brain tumor. Malignant gliomas account for ~ 40% of primary brain tumors and more than 15,000 new cases are diagnosed in the USA each year [1]. Although the incidence rate of GBM is low, the median survival with standard-of-care therapy is only 14.6 months and 12.1 months with radiotherapy alone [1, 2]. The standard treatment procedure for GBM is maximal surgical resection, followed by radiation and chemotherapy with temozolomide (TMZ) [3]. Since TMZ is administered orally to patients, the delivery efficiency to the brain tumors is limited by the blood-brain barrier (BBB) [4]. Although previous studies reported that the BBB is disrupted in the glioma [5, 6], the delivery of TMZ to the brain tumors was still inefficient [7].

Convection-enhanced delivery (CED) is a delivery technique first introduced in the 1990s that infuses drugs or therapeutic agents directly to the brain parenchyma [8]. Because the infusates are injected via a continuous positive pressure, the bulk flow of fluids can penetrate the tissues more effectively than simple diffusion (e.g., carmustine wafer) where the drug molecule is moved via a concentration gradient. CED is not clinically approved yet, but phase studies have tried to achieve meaningful results [9,10,11,12]. In one clinical trial, transferrin-conjugated diphtheria toxin was infused to the brain as therapeutic agents via CED for the treatment of glioma [13]. In the preclinical studies, adeno-associated virus (AAV)-based vectors [14, 15] and siRNAs [16] were injected using CED to transfer genes to the brain. Other therapeutic agents such as epidermal growth factor [17] and anti-cancer drugs [10] were also injected using CED to improve their therapeutic effects in the brain tissues. Recently, various nanoparticle formulations including magnetic nanoparticles [18], lipid nanoparticles [19, 20], polymeric nanoparticles [1, 21], and nanodiamonds [22] have been used for CED to introduce additional functions and further improve the delivery efficiency in the brain tissues. Because of the small size (~ 64 nm) of pores in the brain parenchyma [23], nanoparticles smaller than this size are preferred to distribute the drugs in the brain tissues via CED. Nanoparticles with a highly positive surface charge are generally avoided for efficient tissue distribution because they bind strongly to the cells and the extracellular matrix (ECM) via electrostatic interactions [24]. Previous studies reported that the nanoparticles coated densely with the polyethylene glycol (PEG) can penetrate deep in the brain parenchyma [21, 25]. However, it is also known that a PEG coating impedes cellular uptake of nanoparticles [26], which may lower the therapeutic efficacy of drugs targeting intracellular molecules. Furthermore, the nanoparticles that penetrate the brain tissues without interacting with cells and ECM may be washed rapidly out of the brain tissues presumably via the clearance mechanism of brain wastes [27, 28]. Thus, the therapeutic nanoparticles injected into the brain tissues via CED should be precisely engineered to achieve both efficient distribution and subsequent internalization into the target cells.

In this study, we engineer liposomal formulations to improve both tissue distribution and cellular uptake of the drugs infused in the brain via the CED method. We prepare various liposomal formulations with different surface charge, PEGylation, and transition temperature (Tm), and test their cellular uptake and cytotoxicity in vitro and tissue distribution and retention in vivo. Additionally, we investigate the clearance pathway of the liposomal formulations infused into the brain tissues via CED. Lastly, in a mouse model of GBM, we examine the therapeutic efficacy of CED of liposomal everolimus.

Materials and methods

Preparation and characterization of liposomal formulations

1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (ammonium salt) (DSPE-mPEG 2000), L-α-phosphatidylcholine, hydrogenated (soy) (HSPC), 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), 1,2-dioleoyl-3-trimethylammonium-propane (chloride salt) (DOTAP), and cholesterol (Chol, ovine wool, > 98%) were purchased from Avanti Polar Lipids, Inc. (AL, USA). Lipophilic fluorescence dyes, 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (DiI) and 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide (DiR) were purchased from Invitrogen. Everolimus powder was purchased from Sigma. For preparation of liposomal formulations, DMPC, HSPC, DSPE-PEG 2000, DOTAP, and/or Chol were used at the specific molar ratio as shown in Table 1. The lipid stock solution was prepared in chloroform at a concentration of 10 mg/ml. To embed fluorescence dye DiI or DiR to the liposomal membrane, DiI or DiR stock solution was prepared with 1 mg/ml in ethanol. The lipid solution was mixed with 1% of DiI or DiR solution and then dried in fume hood using desiccator. The completely dried lipid cake was hydrated with a phosphate-buffered saline (PBS) and then stirred for 30 min at 250 rpm. The hydrated lipid solution was extruded on a 50-nm polycarbonate membrane to control the size of liposomal formulation. The temperature during extrusion was 45 °C and 65 °C for DMPC- and HSPC-based liposomal formulations, respectively. The hydrodynamic size, surface charge, and polydispersity of liposomal formulations were obtained using a Zetasizer ZS90 machine operating in intensity mode using a 640-nm HeNe laser (Malvern Instruments, Worcestershire, UK). To investigate stability of liposomal formulations in the normal brain tissue, each liposomal formulation (total lipid mole 1.40526 × 10−6) was incubated in 1 ml of PBS containing 1% hyaluronic acid solution (normal brain tissue-mimicking solution) at 37 °C for 48 h. To investigate stability of liposomal formulations in the brain tumor and lysosome, HSPC-based liposomal formulations were incubated in the brain tumor-mimicking solution (PBS with 1% hyaluronic acid solution at pH 6.7) [29] and lysosome-mimicking solution (PBS at pH 5) [30] 37 °C for 24 h. Their hydrodynamic size, PDI, and surface charge were measured by Zetasizer ZS90. For preparation of everolimus-loaded liposomal formulations, everolimus was dissolved in ethanol at 1 mg/ml concentration. The lipid solution was mixed with 2% of everolimus solution instead of DiI or DiR solution in the abovementioned procedure. Free everolimus was removed using a dialysis method with molecular weight cutoff 100-kDa tube. Physical size and morphology of everolimus-loaded liposomes were observed by transmission electron microscopy (JEM-2100F, JEOL). The loading stability of everolimus was examined by measuring the release of everolimus from the liposomal formulation in the brain tumor (pH 6.7) and lysosomal conditions (pH 5). Briefly, a dialysis membrane tube containing 1 ml of liposome solution was incubated in 50 ml of solution with different pH (6.7 or 5). At specific time points, 20 μl of liposome solution was collected from the dialysis membrane tube and mixed with 80 μl of ethanol to break the liposome structure. The amount of everolimus in the mixed solution was measured using a standard curve of everolimus at 280 nm.

Table 1 Lipid compositions and physiochemical properties of liposomal formulations used in this study

Cell culture

GL261 murine glioblastoma cells were incubated in Dulbecco’s modified Eagle’s medium (DMEM, Hyclone, UT, USA) supplemented with 10% fetal bovine serum (FBS, Hyclone, UT, USA) and 1% penicillin/streptomycin (Hyclone, UT, USA) at 37 °C in 5% CO2 condition.

In vitro cellular experiments

To examine cellular uptake of liposomal formulations, GL261 cells were seeded in a 96-well plate with 5000 cells/well density for microplate fluorescence reader analysis and in a 6-well plate with 250,000 cells/well density for confocal microscopy. After 1 day, the cells were incubated with DiI-loaded liposomal formulations (lipid mole 2.10789 × 10−7) for 30 min, 2 h, or 5 h. After washing three times with PBS, cells were imaged with confocal microscopy (Nikon). To quantify the cellular uptake of liposomal formulations, the cells were incubated with a RIPA buffer (200 μl/well) for 30 min to break the cell structure and fluorescence of DiI was measured using microplate fluorescence reader (ex 550 nm/em 565 nm). The cellular uptake of liposomal formulations was also analyzed via flow cytometry (Supplementary Methods). To investigate cellular internalization pathway of liposomal formulations, cells were incubated with DiI-loaded liposomal formulations for 2 h, stained with LysoTracker™ Green DND-26 (200 nM, ThermoFisher) to label lysosomes, and imaged with confocal microscopy. To examine cytotoxicity of everolimus-loaded liposomal formulations, GL261 cells were seeded in a 96-well plate with 5000 cells/well density. After 1 day, the cells were treated with everolimus-loaded liposomal formulations in the everolimus concentration range from 0.01 to 10 μM for 24, 48, and 72 h. The cells were washed with PBS three times and the cell viability was evaluated using an MTT assay. Cell viability was also analyzed using flow cytometry (Supplementary Methods).

Animal model

Seven-week-old female BALB/c and C57BL/6 mice weighing 17–22 g were purchased from Koatech, Korea. A mouse model of GBM was prepared with conditions modified from previous studies [31,32,33,34,35]. GL261 cells were suspended in antibiotics and serum-free media with 2 × 107 cells/ml density and 10 μl of the cell solution was loaded to the Hamilton syringe with a 26-G needle. The cells were infused into the caudate putamen of mouse using an injection pump. The coordinate is 0 mm anterior-posterior, 2.5 mm lateral, and 3 mm deep from the bregma, and the infusion rate was 1 μl/min. All mice were housed in standard facilities and provided free access to feed and water. All animal experiments were performed with approval from the KAIST Institutional Animal Care and Use Committee (IACUC). Zoletil 50 (25 mg/kg) and Rompun (1:1) were administered during surgery via i.p. injection.

Convection-enhanced delivery to brain

A cranial window with 2 mm × 2 mm available imaging area was made on the right side of skull by using dental drill. The center of cranial window was 2.5 mm lateral from the bregma. A Hamilton syringe with a 33-G needle was loaded with 3 μl of DiR-loaded liposomal formulations and fixed to the stereotaxic frame. The needle was put to the caudate putamen of mouse brain, which coordinate is 0 mm anterior-posterior, 2.5 mm lateral, and 3 mm deep from the bregma. The liposomal formulation solution was infused with 0.1 μl/min rate using an injection pump. To prevent any backflow of solution or blood, the needle was maintained for 5 min before and after the infusion. To observe the tissue distribution of liposomal formulations in the brain, brains were collected 30 min, 24 h, and 48 h after infusion and sectioned using cryotome (Leica) with 10 μm of thickness. The brain sections were then analyzed using Li-cor odyssey NIR imaging system. To observe the tumor distribution of liposomal formulations in the brain, 3 μl of DiR-loaded liposomal formulations was infused to the brain via CED method after 1 week of tumor implantation. To examine biodistribution of liposomal formulations, liver, spleen, kidneys, heart, lungs, and cervical lymph nodes were collected 30 min and 48 h after infusion of DiR-loaded liposomal formulations. All organs were washed three times with PBS and imaged using Li-cor odyssey NIR imaging system. To examine therapeutic effects of everolimus-loaded liposomal formulations, GL261 cells were infused into caudate putamen of C57BL/6 mouse using injection pump according to the abovementioned protocol. Considering our preliminary data on survival curve of mice after inoculation of GL261 cells and the treatment conditions used in the previous studies [31,32,33,34,35], mice were injected with 3 μl of everolimus-loaded liposomal formulations or free everolimus at an everolimus concentration of 10 μM into the same region of tumor site with 0.1 μl/min rate 1 week after inoculation of GL262 cells and sacrificed to collect brains for H&E staining 3 weeks after sample injection.

Results

Preparation and characterization of liposomal formulations

Various liposomal formulations were prepared with different compositions of lipids and cholesterol at specific molar ratios (Table 1). DMPC and HSPC were particularly chosen as base lipids because their different phase transition temperatures (Tm) can affect the rigidity of the liposomal formulations; Tm of DMPC and HSPC is lower and higher than body temperature, respectively. The average size of the resulting liposomal formulations was 80 nm ± 10 nm. As the molar ratio of cationic lipid DOTAP increased, the surface charge of the formulations was generally increased. The PEGylation reduced the surface charge in all liposomal formulations while the existence of cholesterol showed no effect. The colloidal stability of liposomal formulations was examined in a PBS containing 1% hyaluronic acid, which mimics the brain tissue [36,37,38]. None of the PEGylated formulations altered their size over a 48-h period while the cationic formulations without a PEG coating showed a significant size increment (Supplementary fig. 1).

In vitro cellular uptake of liposomal formulations

We first investigated how the surface charge, PEGylation, and Tm of liposomal formulations affect their cellular uptake in vitro. GL261 murine glioblastoma cells were incubated with DiI-labeled liposomal formulations for 2 h and then imaged with confocal microscopy. Among the same PC-based formulations, more positively charged formulations showed superior cellular uptake compared to less positively charged formulations regardless of PEGylation (Fig. 1). Liposomal formulations with only-PEG (DP0, HP0, and HP0C) showed the lowest cellular uptake, while the only-positives (D20 and H20) were predominantly internalized into the cells. Between different base lipids, the cellular uptake of the HSPC-based formulations was greater than that of the DMPC-based formulations. Among all formulations, H20 showed the highest cellular uptake while HP0 displayed the lowest. This trend was also observed in the cells incubated with liposomal formulations for different incubation periods (30 min and 5 h) (Fig. 1b, Supplementary fig. 2). Next, to elucidate the cellular internalization pathway of liposomal formulations, cells were incubated with DiI-labeled liposomal formulations for 2 h, stained with Lysotracker, and imaged with confocal microscopy. HP20 and H20 showed strong co-localization with Lysotracker signals (Supplementary fig. 3), indicating that they enter the cells via the endo-lysosomal pathway. In addition, to investigate the binding effect of liposomal formulations, cells were incubated with DiI-labeled liposomal formulations for 30 min at 37 °C and 4 °C and analyzed via flow cytometry (Supplementary figs. 4 and 5). DiI-positive cells were observed only with D20 and H20 treatments at 4 °C while they were found after treatments of positively charged liposomal formulations at 37 °C, indicating cellular binding ability of highly positive liposomal formulations. Considering the time-dependent aggregation of H20 in the brain tissue-mimicking fluid (Supplementary fig. 1), the HP20 formulation would show the most efficient cellular uptake to the glioblastoma cells in the brain tissues. These results indicate that the positively charged formulations can easily enter the cells while the PEGylation hinders the cellular uptake. Furthermore, we also found that the cellular uptake of PEGylated formulations with high colloidal stability can be enhanced with incorporation of cationic lipids at the proper molar ratio.

Fig. 1
figure1

In vitro cellular uptake of liposomal formulations. a Representative confocal microscopic images of GL261 cells 2 h after incubation with DiI-loaded liposomal formulations (red). Nuclei were stained with Hoechst 33342 (blue). Scale bar indicates 50 μm. b Quantification of DiI-loaded liposomal formulations taken up by the cells after different time of incubation. To avoid the unwanted quenching effect, DiI fluorescence was measured after the cells were lysed with RIPA buffer. Data are means ± s.e.m. (n = 6; *p < 0.05, ***p < 0.001, two-way ANOVA and Bonferroni post-test)

CED of liposomal formulations in normal brains

We next investigated the tissue distribution of liposomal formulations in normal brains after CED. For in vivo experiments, only HSPC-based liposomal formulations were tested because DMPC-based formulations generally showed inefficient cellular uptake. HP0C represented the FDA-approved formulation used for cancer therapy (e.g., Doxil®) while HP10, HP20, and H20 were the formulations tested to find the optimal distribution and retention in the brain tissue after CED. These liposomal formulations were labeled with NIR fluorescence dye DiR and infused to normal mouse brains via the CED method. The brain tissue sections were obtained 30 min, 24 h, and 48 h after infusion and then imaged using a NIR fluorescence imaging system. NIR fluorescence images of brain tissue sections revealed that HP0C spread out over a large area of the brain immediately after CED while HP10, HP20, and H20 were relatively localized in the specific areas (Fig. 2a–c). These observations indicate that HP0C penetrated the brain tissue, presumably without interacting with the surrounding cells during CED, as observed in the in vitro experiments (Fig. 1). Among cationic formulations that can interact efficiently with the cells, the tissue distributions of HP10 and HP20 were much broader than that of H20. This indicates that the PEG coating is likely to reduce their aggregation during tissue penetration as observed in the colloidal stability study (Supplementary fig. 1). Furthermore, the fluorescence signal of HP20 in the brain tissue was not altered significantly from 24 to 48 h after infusion while those of HP10 and H20 were somewhat diminished over this period, indicating that HP20 remained in the brain tissue longer than HP10 and H20 (Fig. 2a, d). The increase of fluorescence signals in all formulations from 30 min to 24 h after infusion is likely due to the dequenching effect that occurred during their diffusion through the brain tissue after CED. The fluorescence signal of HP0C in the brain tissue was much weaker at all time points compared to those of other formulations, showing that a significant amount of HP0C appears to be cleared out of the brain tissue after CED. Collectively, these results suggest that, among the tested formulations, HP20 is an optimal liposomal formulation that can spread over the brain tissue via CED and remain there for a lengthy period of time.

Fig. 2
figure2

Convention-enhanced delivery of liposomal formulations in normal brains. a Representative NIR fluorescence images of brain sections obtained at specific time points after CED of DiR-loaded liposomal formulations. White arrow indicates the injection site. Scale bar indicates 2 mm. b Quantification of the mean areas showing the fluorescence in a. c Quantification of the mean distance of liposome penetration in a. d Quantification of the mean fluorescence intensity measured in the brain section in a. Data are means ± s.e.m. (n = 3; *p < 0.05, ***p < 0.001, two-way ANOVA and Bonferroni post-test)

Biodistribution of liposomal formulations after CED

Having observed that liposomal formulations could be cleared out of the brain tissue after CED, we investigated their time-dependent accumulation in other major organs. DiR-labeled liposomal formulations were infused to normal mouse brains via CED. At 30 min, 24 h, and 48 h post-infusion, heart, lungs, liver, spleen, kidneys, and cervical lymph nodes were harvested as the major organs and imaged using the NIR fluorescence imaging system. The cervical lymph nodes were particularly included as the organ where the cerebrospinal fluid is drained from the brain parenchyma [27, 28, 39,40,41]. NIR imaging of major organs revealed that a significant amount of HP0C was found in the cervical lymph nodes 30 min after CED and the amounts of HP0C accumulated in cervical lymph nodes, liver, and spleen gradually increased over time after CED (Fig. 3, Supplementary fig. 6). These observations show that the HP0C was drained to the cervical lymph nodes immediately after CED, entered the systemic circulation, and finally accumulated in the liver and spleen, thereby leaving a minimal amount in the brain tissues. By contrast, the accumulation of H20 in all organs was negligible over the 48-h period after CED, indicating that the majority of H20 remained in the brain tissues. Comparing the formulations composed of both PEG and cationic lipids, a significant amount of HP20 was found in the cervical lymph nodes from 48 h after CED while the HP10 was in the cervical lymph nodes from 24 h and in the liver and spleen from 48 h after CED. These results indicate that increasing the positive charge on the PEGylated surface of liposomal formulations improved their tissue retention in the brain after CED. Collectively, these results suggest that the liposomal formulation infused in the brain tissue via CED can either remain by interacting with the surrounding cells and extracellular matrix or be eliminated from the brain parenchyma to the cervical lymph nodes and further to the liver and spleen, and that this biodistribution pattern can be modulated by controlling the ratio of PEG and cationic moieties on the liposomal surface.

Fig. 3
figure3

Biodistribution of liposomal formulations after CED to normal brains. a Representative NIR fluorescence images of major organs obtained at specific time points after CED of DiR-loaded liposomal formulations. H, Lu, CLN, Li, S, and K indicate heart, lung, cervical lymph nodes, liver, spleen, and kidneys, respectively. bd Quantification of the mean fluorescence intensity of cervical lymph nodes (b), liver (c), and spleen (d) in a. Data are means ± s.e.m. (n = 3; *p < 0.05, *p < 0.01, ***p < 0.001, two-way ANOVA and Bonferroni post-test)

CED of liposomal formulations in brain tumors

Since biophysical characteristics are different between cancerous and normal tissues in the brain [42] and this difference could affect the tissue distribution of liposomal formulations, we next investigated tissue and organ distributions of liposomes after direct infusion into the brain tumor via CED. To prepare a mouse model of glioblastoma, GL261 cells were implanted to the caudate putamen of the mouse brain. After 1 week, liposomal formulations were infused to the same site for tumor implantation via the CED method. At 30 min and 48 h of post-infusion, the organs were harvested and imaged using the NIR fluorescence imaging system. NIR fluorescence images of brain tissue sections at 30 min post-infusion revealed that HP0C was well distributed over a large area of the brain while the distributions of other formulations were relatively localized (Fig. 4a–d), which was similar to the distribution observed in the normal brain (Fig. 2). However, at 48 h post-infusion, the majority of HP0C was cleared out of the brain tumor while a significant amount of HP20 remained in a relatively large area. Quantification of NIR fluorescence in the cervical lymph nodes also showed significant drainage of HP0C to the cervical lymph nodes 48 h after CED to the brain tumor (Fig. 4e), as observed in the normal brain (Fig. 3, Supplementary fig. 6). These results demonstrate that the tissue distribution profiles of liposomal formulations infused directly to the brain tumor via CED were similar to those infused to the normal brain while the clearance of liposomal formulations in the cancerous tissues was slightly faster than that in the normal tissues. Considering the results of both tumor and normal tissues in the brain, the HP20 formulation showed the tissue distribution and clearance profiles that were most suitable for the treatment of glioblastoma.

Fig. 4
figure4

CED of liposomal formulations in brain tumors. a Representative NIR fluorescence images of brain sections obtained 30 min and 48 h after CED of DiR-loaded liposomal formulations. White arrow indicates the injection site. Scale bar indicates 2 mm. b Quantification of the mean areas showing the fluorescence in a. c Quantification of the mean distance of liposome penetration in a. d Quantification of the mean fluorescence intensity measured in the brain section in a. e Quantification of the mean fluorescence intensity measured in the cervical lymph nodes obtained 30 min and 48 h after CED of DiR-loaded liposomal formulations. Data are means ± s.e.m. (n = 3 for HP0C and HP10, n = 4 for HP20 and H20; *p < 0.05, **p < 0.01, ***p < 0.001, two-way ANOVA and Bonferroni post-test)

CED of liposomal everolimus for the treatment of brain tumors

Having found the optimal liposomal formulation for CED, we prepared and characterized liposomal drugs for the treatment of brain tumors. HP0C and HP20 formulations were particularly included because H20 aggregated in the focused area and HP10 showed less penetration and retention in the brain compared to HP20. The stability of HSPC-based liposomal formulations was further examined in the brain tumor-mimicking (pH 6.7) [29] and lysosomal conditions (pH 5) [30]. HP0C and HP20 formulations showed that their size and PDI were slightly increased in the lysosomal condition (Supplementary fig. 7), indicating that they can be destabilized in the lysosome for drug release. The effects of drug-loaded liposomes for in vivo treatment of brain tumors were examined. Everolimus was used as an anti-cancer drug because it is one of the FDA-approved drugs for the treatment of brain tumors, especially for subependymal giant cell astrocytoma [43], and can be loaded easily in the liposomal membrane due to the high hydrophobicity. Everolimus-loaded HP0C and HP20 (HP0C-Eve and HP20-Eve) showed similar sizes (~ 80 nm) (Fig. 5a, b) and drug release profiles at both brain tumor and lysosomal conditions (Fig. 5c, d). Both formulations released less than 20% of the drug in the brain tumor-mimicking condition for 24 h while they released around 50% in the lysosomal condition, implying that the liposomes would efficiently release drugs inside the cell. The cytotoxicity of HP20-Eve in the tested drug concentrations against GL261 cells was higher than that of HP0C-Eve regardless of incubation time (Fig. 5e, Supplementary fig. 8). The Annexin V/PI cell viability assay via flow cytometry showed that the percentage of early and late apoptotic cells in the HP20-Eve treatment was significantly higher than that in the HP0C-Eve treatment (Supplementary fig. 9). Collectively, these results suggest that the HP20 formulation enhances the cytotoxicity against GL261 cells via efficient cellular uptake and subsequent drug release in lysosomes.

Fig. 5
figure5

Preparation and characterization of everolimus-loaded liposomal formulations. a, b Hydrodynamic sizes (a) and TEM images (b) of everolimus-loaded liposomal formulations. Scale bars indicate 50 nm. c, d Drug release profiles of everolimus-loaded liposomal formulations at pH 6.7 (brain tumor environment) (c) and at pH 5 (lysosome) (d) at 37 °C. e Cytotoxicity of free everolimus and Eve-loaded liposomal formulations against GL261 cells for 48 h of incubation. Data are means ± s.e.m. (n = 6; *p < 0.05, **p < 0.01, ***p < 0.001, two-way ANOVA and Bonferroni post-test)

Lastly, we tested the therapeutic potential of CED of liposomal drugs for the treatment of glioblastoma. Free everolimus (Eve), HP0C-Eve, or HP20-Eve was infused to the tumor site via the CED method 1 week after tumor implantation. At 3 weeks post-treatment, the brain sections were prepared, stained with H&E, and visualized with optical microscopy. The H&E staining of brain tissue sections revealed that the tumor areas after CED of free Eve and HP0C-Eve were similar to the area of untreated tumors (Fig. 6a, b). However, the tumor area after CED of HP20-Eve was significantly smaller than those of untreated, free Eve-treated, and HP0C-Eve-treated tumors. More importantly, the HP20-Eve treatment delayed the loss of mouse weight while free Eve-treated and HP0C-Eve-treated mice showed a gradual decrease similar to untreated mice (Fig. 6c), indicating that the CED of HP20-Eve effectively reduced the tumor size. Both HP0C-Eve and HP20-Eve treatments did not show significant nanotoxicity (Supplementary fig. 10). Collectively, these results suggest that the liposomal formulations to distribute the drugs over a large area of the brain tumor and internalize them efficiently into the tumor cells can be used to improve the treatment of glioblastoma via the CED method.

Fig. 6
figure6

CED of liposomal drugs for the treatment of brain tumors. a Representative H&E images of brain tissue sections 3 weeks after CED. Yellow dotted lines indicate the tumor regions. Scale bar indicates 2 mm. b Quantification of the tumor areas measured in a. c Monitoring of mouse weights over the 1-week period before and the 3-week period after CED. Arrows indicate the day when CED was performed. Data are means ± s.e.m. [n = 4; *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA and Tukey’s multiple comparison test for b and repeated measures ANOVA and Tukey’s multiple comparison test for c]

Discussion

Numerous delivery strategies have been developed to efficiently localize therapeutic agents to the brain tumors. However, most of the approaches including intravenous injection could not overcome the blood-brain barrier, which is a highly selective barrier separating the circulating blood from the brain tissues, thus significantly reducing the delivery efficiency of therapeutic agents to the brain tumors. CED is a delivery strategy that was developed to localize therapeutic agents efficiently to the targeted area of the brain regardless of the blood-brain barrier. During the CED, the bulk flow transports the therapeutic agents through the extracellular space via a pressure gradient. However, the CED method cannot control the tissue retention of therapeutic agents after distributing them over the targeted area of the brain, thus limiting the therapeutic efficacy. Furthermore, most anti-cancer drugs induce their therapeutic effects by binding their intracellular targets, but do not enter the cell alone. Thus, precise engineering of pharmaceutical formulations is required to achieve a broad distribution, prolonged retention, and efficient cellular uptake of therapeutic agents in the brain tissues after CED for effective treatment of brain tumors.

In this study, we employed and engineered liposomal formulations to improve the therapeutic efficacy of CED for the treatment of brain tumors. It is well known that the positive surface charge enhances the cellular uptake of nanoparticles via electrostatic interactions while the PEGylated surface prevents the nanoparticles from being taken up by cells [44, 45]. In addition, it was previously reported that the dense PEG coating enhances the penetration of nanoparticles within the brain parenchyma [25]. Thus, PEGylated and cationic moieties on the nanoparticles should be balanced to achieve both tissue distribution and subsequent cellular uptake after CED. Based on this hypothesis, we prepared various liposomal formulations by precisely controlling the ratio of PEGylated and cationic lipids and incorporating the base lipids with different Tm, and tested their cellular uptake in vitro, and their distribution and retention in the brain and organ distribution after CED. First, we found that the liposomal formulations composed of base lipids with higher Tm showed superior cellular uptake in mouse glioblastoma cells compared to those with lower Tm, presumably due to the more rigid structure. Secondly, we found that PEGylated liposomal formulations spread over a large area of the brain via CED while their cellular uptake was poor. However, PEGylated liposomal formulations incorporated with cationic lipids at the proper molar ratio significantly enhanced the cellular uptake and tissue retention while being distributed over a relatively large area of the brain via CED, which was similar to a recent finding involving polymeric nanoparticles with bioadhesive moieties [46]. Lastly, we observed that the liposomal formulations cleared out of the brain after CED accumulated in the cervical lymph nodes. The brain has a drainage pathway to clear out the waste molecules from the brain parenchyma to the cervical lymph nodes [27, 28, 39,40,41]. Based on the clearance mechanism, the liposomal formulations that do not interact with the cells and extracellular matrix are first drained to the cervical lymph nodes.

The extracellular matrix (ECM) of the brain is mainly composed of proteoglycans and hyaluronic acids [47, 48]. These ECM components play important roles in cushioning the cells and providing the strength and resilience in the brain. In particular, it is known that the production of highly negatively charged heparin sulfate proteoglycans is upregulated in the brain tumors [49]. The leaky vasculature is also developed in the peripheral regions of the brain tumor [50, 51], which facilitates a transport of biomolecules between blood circulation and cancerous tissues. Unfortunately, it is unknown how the lymphatic drainage system is altered in the brain tumor microenvironment. Based on the previous findings, it is speculated that HP20, PEGylated liposomal formulations incorporated with cationic lipids at the proper molar ratio, can spread over a large area of the brain tumor via CED and remain there for a relatively long period of time by either entering the cells or interacting with the negatively charged ECM. The liposomal formulations coated only with PEG do not interact with cells and ECM both during and after CED, thus reducing their tissue retention in the brain tumors. By contrast, the liposomal formulation with a highly positively charged surface tends to aggregate and interact strongly with ECM during the CED, not allowing their efficient distribution in the cancerous tissues. Most importantly, efficient tissue distribution and retention of HP20-based liposomal drugs after CED led to more effective treatment of brain tumors compared to free and HP0C-based liposomal drugs. However, one limitation of this study is that we could not observe the development of the brain tumor mass and its response to the CED of liposomal drugs longitudinally.

In summary, we precisely engineered liposomal formulations to improve the therapeutic efficacy of CED for the treatment of glioblastoma. We found that PEGylated liposomal formulations incorporated with cationic lipids at the proper molar ratio exhibited specific cellular uptake in the glioblastoma cells in vitro, and efficient tissue distribution and retention in the brain tumor after CED. We believe that liposomal drugs engineered based on the biophysical properties of disease microenvironments will significantly improve the therapeutic efficacy of CED for the treatment of brain diseases.

References

  1. 1.

    Zhou J, Patel TR, Sirianni RW, Strohbehn G, Zheng MQ, Duong N, et al. Highly penetrative, drug-loaded nanocarriers improve treatment of glioblastoma. Proc Natl Acad Sci U S A. 2013;110:11751–6.

    CAS  PubMed Central  Google Scholar 

  2. 2.

    Allard E, Passirani C, Benoit JP. Convection-enhanced delivery of nanocarriers for the treatment of brain tumors. Biomaterials. 2009;30:2302–18.

    CAS  Google Scholar 

  3. 3.

    Stupp R, Mason WP, van den Bent M, Weller M, Fisher B, Taphoorn MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352:987–96.

    CAS  Google Scholar 

  4. 4.

    Dong X. Current strategies for brain drug delivery. Theranostics. 2018;8:1481–93.

    CAS  PubMed Central  Google Scholar 

  5. 5.

    Dhermain FG, Hau P, Lanfermann H, Jacobs AH, van den Bent MJ. Advanced MRI and PET imaging for assessment of treatment response in patients with gliomas. Lancet Neurol. 2010;9:906–20.

    Google Scholar 

  6. 6.

    van Tellingen O, Yetkin-Arik B, de Gooijer MC, Wesseling P, Wurdinger T, de Vries HE. Overcoming the blood-brain tumor barrier for effective glioblastoma treatment. Drug Resist Updat. 2015;19:1–12.

    Google Scholar 

  7. 7.

    Wohlfart S, Gelperina S, Kreuter J. Transport of drugs across the blood-brain barrier by nanoparticles. J Control Release. 2012;161:264–73.

    CAS  Google Scholar 

  8. 8.

    Bobo RH, Laske DW, Akbasak A, Morrison PF, Dedrick RL, Oldfield EH. Convection-enhanced delivery of macromolecules in the brain. Proc Natl Acad Sci U S A. 1994;91:2076–80.

    CAS  PubMed Central  Google Scholar 

  9. 9.

    Kunwar S, Chang S, Westphal M, Vogelbaum M, Sampson J, Barnett G, et al. Phase III randomized trial of CED of IL13-PE38QQR vs Gliadel wafers for recurrent glioblastoma. Neuro-oncology. 2010;12:871–81.

    CAS  PubMed Central  Google Scholar 

  10. 10.

    Lidar Z, Mardor Y, Jonas T, Pfeffer R, Faibel M, Nass D, et al. Convection-enhanced delivery of paclitaxel for the treatment of recurrent malignant glioma: a phase I/II clinical study. J Neurosurg. 2004;100:472–9.

    CAS  Google Scholar 

  11. 11.

    Mittermeyer G, Christine CW, Rosenbluth KH, Baker SL, Starr P, Larson P, et al. Long-term evaluation of a phase 1 study of AADC gene therapy for Parkinson’s disease. Hum Gene Ther. 2012;23:377–81.

    CAS  PubMed Central  Google Scholar 

  12. 12.

    Souweidane MM, Kramer K, Pandit-Taskar N, Zhou Z, Haque S, Zanzonico P, et al. Convection-enhanced delivery for diffuse intrinsic pontine glioma: a single-centre, dose-escalation, phase 1 trial. Lancet Neurol. 2018;19:1040–50.

    Google Scholar 

  13. 13.

    Laske DW, Youle RJ, Oldfield EH. Tumor regression with regional distribution of the targeted toxin TF-CRM107 in patients with malignant brain tumors. Nat Med. 1997;3:1362–8.

    CAS  Google Scholar 

  14. 14.

    Bankiewicz KS, Eberling JL, Kohutnicka M, Jagust W, Pivirotto P, Bringas J, et al. Convection-enhanced delivery of AAV vector in parkinsonian monkeys; in vivo detection of gene expression and restoration of dopaminergic function using pro-drug approach. Exp Neurol. 2000;164:2–14.

    CAS  Google Scholar 

  15. 15.

    Nguyen JB, Sanchez-Pernaute R, Cunningham J, Bankiewicz KS. Convection-enhanced delivery of AAV-2 combined with heparin increases TK gene transfer in the rat brain. Neuroreport. 2001;12:1961–4.

    CAS  Google Scholar 

  16. 16.

    Stiles DK, Zhang Z, Ge P, Nelson B, Grondin R, Ai Y, et al. Widespread suppression of huntingtin with convection-enhanced delivery of siRNA. Exp Neurol. 2012;233:463–71.

    CAS  Google Scholar 

  17. 17.

    Yang W, et al. Convection-enhanced delivery of boronated epidermal growth factor for molecular targeting of EGF receptor-positive gliomas. Cancer Res. 2002;62:6552–8.

    CAS  Google Scholar 

  18. 18.

    Hadjipanayis CG, Machaidze R, Kaluzova M, Wang L, Schuette AJ, Chen H, et al. EGFRvIII antibody-conjugated iron oxide nanoparticles for magnetic resonance imaging-guided convection-enhanced delivery and targeted therapy of glioblastoma. Cancer Res. 2010;70:6303–12.

    CAS  PubMed Central  Google Scholar 

  19. 19.

    Mamot C, Nguyen JB, Pourdehnad M, Hadaczek P, Saito R, Bringas JR, et al. Extensive distribution of liposomes in rodent brains and brain tumors following convection-enhanced delivery. J Neuro-Oncol. 2004;68:1–9.

    Google Scholar 

  20. 20.

    Saito R, Bringas JR, McKnight TR, Wendland MF, Mamot C, Drummond DC, et al. Distribution of liposomes into brain and rat brain tumor models by convection-enhanced delivery monitored with magnetic resonance imaging. Cancer Res. 2004;64:2572–9.

    CAS  Google Scholar 

  21. 21.

    Nance E, Zhang C, Shih TY, Xu Q, Schuster BS, Hanes J. Brain-penetrating nanoparticles improve paclitaxel efficacy in malignant glioma following local administration. ACS Nano. 2014;8:10655–64.

    CAS  PubMed Central  Google Scholar 

  22. 22.

    Xi G, Robinson E, Mania-Farnell B, Vanin EF, Shim KW, Takao T, et al. Convection-enhanced delivery of nanodiamond drug delivery platforms for intracranial tumor treatment. Nanomedicine. 2014;10:381–91.

    CAS  Google Scholar 

  23. 23.

    Thorne RG, Nicholson C. In vivo diffusion analysis with quantum dots and dextrans predicts the width of brain extracellular space. Proc Natl Acad Sci U S A. 2006;103:5567–72.

    CAS  PubMed Central  Google Scholar 

  24. 24.

    Frohlich E. The role of surface charge in cellular uptake and cytotoxicity of medical nanoparticles. Int J Nanomedicine. 2012;7:5577–91.

    PubMed Central  Google Scholar 

  25. 25.

    Nance EA, et al. A dense poly(ethylene glycol) coating improves penetration of large polymeric nanoparticles within brain tissue. Sci Transl Med. 2012;4:149ra119.

    PubMed Central  Google Scholar 

  26. 26.

    Mishra S, Webster P, Davis ME. PEGylation significantly affects cellular uptake and intracellular trafficking of non-viral gene delivery particles. Eur J Cell Biol. 2004;83:97–111.

    CAS  Google Scholar 

  27. 27.

    Weller RO, Djuanda E, Yow HY, Carare RO. Lymphatic drainage of the brain and the pathophysiology of neurological disease. Acta Neuropathol. 2009;117:1–14.

    CAS  Google Scholar 

  28. 28.

    Weller RO, Subash M, Preston SD, Mazanti I, Carare RO. Perivascular drainage of amyloid-beta peptides from the brain and its failure in cerebral amyloid angiopathy and Alzheimer’s disease. Brain Pathol. 2008;18:253–66.

    CAS  Google Scholar 

  29. 29.

    Rottenberg DA, Ginos JZ, Kearfott KG, Junck L, Dhawan V, Jarden JO. In vivo measurement of brain tumor pH using [11C]DMO and positron emission tomography. Ann Neurol. 1985;17:70–9.

    CAS  Google Scholar 

  30. 30.

    Mindell JA. Lysosomal acidification mechanisms. Annu Rev Physiol. 2012;74:69–86.

    CAS  Google Scholar 

  31. 31.

    Ramachandran M, Yu D, Dyczynski M, Baskaran S, Zhang L, Lulla A, et al. Safe and effective treatment of experimental neuroblastoma and glioblastoma using systemically delivered triple microRNA-detargeted oncolytic Semliki Forest virus. Clin Cancer Res. 2017;23:1519–30.

    CAS  Google Scholar 

  32. 32.

    Wei J, Marisetty A, Schrand B, Gabrusiewicz K, Hashimoto Y, Ott M, et al. Osteopontin mediates glioblastoma-associated macrophage infiltration and is a potential therapeutic target. J Clin Invest. 2019;129:137–49.

    Google Scholar 

  33. 33.

    Belmans J, van Woensel M, Creyns B, Dejaegher J, Bullens DM, van Gool SW. Immunotherapy with subcutaneous immunogenic autologous tumor lysate increases murine glioblastoma survival. Sci Rep. 2017;7:13902.

    PubMed Central  Google Scholar 

  34. 34.

    Enríquez Pérez J, Kopecky J, Visse E, Darabi A, Siesjö P. Convection-enhanced delivery of temozolomide and whole cell tumor immunizations in GL261 and KR158 experimental mouse gliomas. BMC Cancer. 2020;20:7.

    PubMed Central  Google Scholar 

  35. 35.

    IYENGAR S, SCHWARTZ D. Failure of inositol trispyrophosphate to enhance highly effective radiotherapy of GL261 glioblastoma in mice. Anticancer Res. 2017;37:1121–5.

    CAS  Google Scholar 

  36. 36.

    Zhang T, Yan Y, Wang X, Xiong Z, Lin F, Wu R, et al. Three-dimensional gelatin and gelatin/hyaluronan hydrogel structures for traumatic brain injury. J Bioact Compat Polym. 2007;22:19–29.

    Google Scholar 

  37. 37.

    Wu S, Xu R, Duan B, Jiang P. Three-dimensional hyaluronic acid hydrogel-based models for in vitro human iPSC-derived NPC culture and differentiation. J Mater Chem B. 2017;5:3870–8.

    CAS  PubMed Central  Google Scholar 

  38. 38.

    Xiao W, Zhang R, Sohrabi A, Ehsanipour A, Sun S, Liang J, et al. Brain-mimetic 3D culture platforms allow investigation of cooperative effects of extracellular matrix features on therapeutic resistance in glioblastoma. Cancer Res. 2018;78:1358–70.

    CAS  Google Scholar 

  39. 39.

    Engelhardt B, Carare RO, Bechmann I, Flügel A, Laman JD, Weller RO. Vascular, glial, and lymphatic immune gateways of the central nervous system. Acta Neuropathol. 2016;132:317–38.

    CAS  PubMed Central  Google Scholar 

  40. 40.

    Louveau A, Plog BA, Antila S, Alitalo K, Nedergaard M, Kipnis J. Understanding the functions and relationships of the glymphatic system and meningeal lymphatics. J Clin Invest. 2017;127:3210–9.

    PubMed Central  Google Scholar 

  41. 41.

    Ahn JH, Cho H, Kim JH, Kim SH, Ham JS, Park I, et al. Meningeal lymphatic vessels at the skull base drain cerebrospinal fluid. Nature. 2019;572:62–6.

    CAS  Google Scholar 

  42. 42.

    Quail DF, Joyce JA. The microenvironmental landscape of brain tumors. Cancer Cell. 2017;31:326–41.

    CAS  PubMed Central  Google Scholar 

  43. 43.

    Krueger DA, Care MM, Holland K, Agricola K, Tudor C, Mangeshkar P, et al. Everolimus for subependymal giant-cell astrocytomas in tuberous sclerosis. N Engl J Med. 2010;363:1801–11.

    CAS  Google Scholar 

  44. 44.

    Hu Y, Xie J, Tong YW, Wang CH. Effect of PEG conformation and particle size on the cellular uptake efficiency of nanoparticles with the HepG2 cells. J Control Release. 2007;118:7–17.

    CAS  Google Scholar 

  45. 45.

    Oyewumi MO, Yokel RA, Jay M, Coakley T, Mumper RJ. Comparison of cell uptake, biodistribution and tumor retention of folate-coated and PEG-coated gadolinium nanoparticles in tumor-bearing mice. J Control Release. 2004;95:613–26.

    CAS  Google Scholar 

  46. 46.

    Song E, Gaudin A, King AR, Seo YE, Suh HW, Deng Y, et al. Surface chemistry governs cellular tropism of nanoparticles in the brain. Nat Commun. 2017;8:15322.

    CAS  PubMed Central  Google Scholar 

  47. 47.

    Ferrer VP, Moura Neto V, Mentlein R. Glioma infiltration and extracellular matrix: key players and modulators. Glia. 2018;66:1542–65.

    Google Scholar 

  48. 48.

    Oohashi T, Edamatsu M, Bekku Y, Carulli D. The hyaluronan and proteoglycan link proteins: organizers of the brain extracellular matrix and key molecules for neuronal function and plasticity. Exp Neurol. 2015;274:134–44.

    CAS  Google Scholar 

  49. 49.

    Wade A, Robinson AE, Engler JR, Petritsch C, James CD, Phillips JJ. Proteoglycans and their roles in brain cancer. FEBS J. 2013;280:2399–417.

    CAS  PubMed Central  Google Scholar 

  50. 50.

    Kim WY, Lee HY. Brain angiogenesis in developmental and pathological processes: mechanism and therapeutic intervention in brain tumors. FEBS J. 2009;276:4653–64.

    CAS  PubMed Central  Google Scholar 

  51. 51.

    Liebner S, Dijkhuizen RM, Reiss Y, Plate KH, Agalliu D, Constantin G. Functional morphology of the blood–brain barrier in health and disease. Acta Neuropathol. 2018;135:311–36.

    CAS  PubMed Central  Google Scholar 

Download references

Funding

This work was supported by the Basic Science Research Program (Grant No. NRF-2017R1E1A1A01074847) through the National Research Foundation funded by the Ministry of Science and ICT, Republic of Korea.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Ji-Ho Park.

Ethics declarations

All animal experiments were performed with approval from the KAIST Institutional Animal Care and Use Committee (IACUC).

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

ESM 1

(PDF 13.2 mb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Han, Y., Park, JH. Convection-enhanced delivery of liposomal drugs for effective treatment of glioblastoma multiforme. Drug Deliv. and Transl. Res. 10, 1876–1887 (2020). https://doi.org/10.1007/s13346-020-00773-w

Download citation

Keywords

  • Cervical lymph node
  • Chemotherapy
  • Convection-enhanced delivery
  • Glioblastoma
  • Liposome