Folic acid‐decorated astrocytes‐derived exosomes enhanced the effect of temozolomide against glioma

A direct strategy to achieve specific treatments and reduce side effects is through cell type‐specific drug delivery. Exosomes (Exos) can be modified with folic acid (FA) to prepare drug delivery systems targeting tumor cells that highly express FA receptors. This study aimed to produce an exo drug delivery system with FA decoration and temozolomide (TMZ) loading to improve the sustained TMZ release and targeting. We used DSPE‐PEG2000‐FA to modify exos derived from astrocyte U‐87 to prepare FA‐modified exos (Astro‐exo‐FA). TMZ was encapsulated into Astro‐exo‐FA or Astro‐exo through electroporation to produce TMZ@Astro‐exo and TMZ@Astro‐exo‐FA. In vitro drug release was examined using the dialysis bag method. Through cell experiments in vitro and mouse glioma models in vivo, the effect of TMZ@Astro‐exo‐FA on U‐87 cells was determined. Exo properties were not affected by FA modification and TMZ loading. The drug release rate of TMZ@Astro‐exo‐FA was slower. TMZ@Astro‐exo‐FA uptake by U‐87 cells was higher compared to TMZ@Astro‐exo, indicating that TMZ@Astro‐exo‐FA has a stronger targeting toward U‐87 cells. TMZ@Astro‐exo‐FA remarkably reduced U‐87 cell proliferation, migration, and invasion compared with TMZ@Astro‐exo and free TMZ. Treatment with TMZ@Astro‐exo‐FA reduced the side effects of TMZ (minimal change in body weight), prolonged survival, and inhibited tumor growth in mouse glioma models, and its efficacy was stronger than that of TMZ@Astro‐exo and free TMZ. TMZ@Astro‐exo‐FA could enhance the effect of TMZ against glioma, providing novel ideas for drug targeting delivery and exploring exos as drug carriers against glioma.


| INTRODUCTION
Glioma, a malignant intracranial tumor that occurs in the neuroectoderm, accounts for approximately 40%-50% of craniocerebral tumors. 1 Presently, surgical resection, radiotherapy, and chemotherapy or in combination are the main treatment strategies for glioma. 2 The characteristic infiltrative growth of glioma blurs the boundary between normal brain tissue and tumor tissue, leading to difficulty in complete tissue removal through surgical treatment, rendering a poor treatment effect with a high risk of recurrence. 3,4mozolomide (TMZ) is a first-line drug for the clinical treatment of brain tumors, with excellent properties such as lipophilicity, acid stability, less toxic side effects, and ability to penetrate the bloodbrain barrier (BBB). 5,6TMZ is hydrolyzed at physiological pH ≥7.4,releasing a methyldiazonium cation to methylate purine bases (O6-guanine, N7-guanine, and N3-adenine), stimulating anti-tumor activity. 7However, TMZ does not possess specific targeting to glioma cells and is unstable, making it prone to degradation under human environmental conditions; thus, clinically, frequent administration is required to maintain an effective TMZ concentration. 7,8Importantly, frequent administration not only leads to low drug utilization, but also causes serious damage to normal cells. 9,10Hence, the development of TMZ carriers with specific targeting has become a new research hotspot.
Exosomes (exos) are membranous small vesicles released into the extracellular matrix by different types of cells, which are widely present in cell culture supernatants and various body fluids, with a diameter of approximately 30-150 nm. 11Biologically active substances can be transported from donor cells to recipient cells through exos, allowing communication between cells. 12Exo-mediated delivery can circumvent the efflux system of P-glycoprotein drugs, thereby reducing drug resistance. 13Furthermore, the circulating stability of exos can impede the phagocytosis of macrophages and prolong the half-life of chemical drugs. 14,15Concurrently, exos can also undergo genetic engineering processing, and by modifying the surface proteins of exos, they can specifically bind to receptors on the target cell surface, thereby making exo-mediated transportation with cell and tissue specificity. 16Additionally, in clinical applications, the high biocompatibility and low toxicity of naturally derived exos render them superior to synthetic drug carriers. 17Thus, in the treatment of glioma, exos as effective anti-cancer drug carriers have attracted widespread attention.Folic acid (FA) is a small-molecular-weight vitamin.FA receptors are membrane glycoproteins that are highly expressed on some tumor cell surface. 18Therefore, FA can be used to modify the exos carrying the drug, causing engineered exos to have specific targeting properties, exerting a killing effect on tumor cells and reducing normal tissue damage. 19,20FA-functionalized exos co-loaded with resveratrol and celastrol demonstrated powerful anti-inflammatory and immunosuppressive activities against LPSstimulated macrophages in vitro. 21An exo-based drug delivery system expressing PH20 and modifying FA could enhance the tumor treatment efficiency and reduce the side effects of hyaluronidase treatment. 22Therefore, taking advantage of FA receptor overexpression in glioma cells, the FA-decorated exos loaded with TMZ can actively target TMZ-loaded exos to glioma cells with FA receptors as targets, thereby improving the distribution of TMZ-loaded exos in tumor tissues and achieving the aim of the targeted therapy.We mainly constructed FA-modified TMZloaded astrocyte-derived exos (TMZ@Astro-exo-FA) and evaluated the effect of TMZ@Astro-exo-FA on astrocytes through a series of cell experiments in vitro and glioma mouse models in vivo, providing a direction for the development of new drug carriers that can target glioma.

| Isolation of U-87 cell-derived exos
Ultracentrifugation is the most frequently employed method for the isolation of exos. 23For exo isolation, the cell culture supernatants containing exos were harvested after a 48-h culture.The obtained supernatant was first centrifuged (300g, 10 min) (L-80XP ultracentrifuge, Beckman Coulter, Brea, CA, USA) to eliminate dead cells, followed by re-centrifugation (2000 g, 10 min) to remove cellular debris.
Centrifugation was again performed at 10,000 g for 30 min to remove large vesicles.Subsequently, centrifugation (120,000 g) was conducted for 90 min to aggregate the pellet.The obtained pellets were then resuspended in cold PBS, followed by filtration via a 0.22-μm filter.Finally, centrifugation (120,000 g) was performed for 90 min to obtain exos.The quantification and analysis of the obtained exo fraction were performed through quantitative protein estimation using the BCA protein assay kit (Cat#P0011, Beyotime, Shanghai, China).
Typical exo markers CD63, CD81, and Tsg101 were analyzed using western blotting.The diluted exos were then applied to a copper grid for 1 min, followed by staining with 1% uranyl nitrate for 30 s.The Tecnai G2 Spirit transmission electron microscope (TEM) was used to examine the dried grids, operating at 120 kV.The size distribution and particle concentration were analyzed and recorded using the Nano-Sight NS300 system (Malvern, UK) and Nanoparticle Tracking Analysis software (NTA, version 2.3).
The mixture was electroporated at 500 V with one discharge for 1 ms.The suspension was incubated at 4 C for 15 min after electroporation, followed by a continued incubation for 1 h at 37 C to promote drug diffusion and pore closure.Finally, the mixture was centrifuged twice (4000g, 30 min) in 100-KDa ultrafiltration tubes to remove extra-exosomal TMZ.Consequently, the characterization of TMZ@Astro-exo and TMZ@Astro-exo-FA was analyzed using NTA, TEM, and western blotting.

| Western blotting
The exosomal markers CD63, CD81, and Tsg101 were confirmed through western blotting.In brief, radioimmunoprecipitation assay buffer (Cat#R0020, Solarbio, Beijing, China) supplemented with PMSF was used to lyse U-87 cells, TMZ@Astro-exo, and TMZ@Astro-exo-FA.Protein samples were electrophoresed with 10% sodium dodecyl-sulfate polyacrylamide gel electrophoresis and then were transferred onto polyvinylidene difluoride membranes (Sigma).After blocking with 5% non-fat dry milk, the membrane was incubated with primary antibody against CD63, CD81, Tsg101 or β-actin at 4 C overnight.All information on all antibodies is presented in Table S1.The PVDF membranes were visualized with a gel imaging system after incubating with HRP-conjugated secondary antibody (Millipore, Bedford, MA, USA).

| Analysis of drug release
The release rate of free TMZ, TMZ@Astro-exo, and TMZ@Astro-exo-FA was evaluated using PBS (pH 7.4) as the release medium.The release medium (25 mL) was used for the immersion of a sealed dialysis bag with TMZ@Astro-exo and TMZ@Astro-exo-FA (5 μL).Equal amounts of fresh release medium were added after collecting the release medium (500 μL).Each sample was measured at 254 nm through high-performance liquid chromatography (Shimadzu LC-2030, Japan).
The images were photographed under a fluorescence microscope.The number of EdU-positive cells was calculated as the percentage of total cells per random field.

| Clonogenic assay
The clonogenic assay is an in vitro survival assay based on the ability of a single cell to grow into a colony.After harvesting with 0.05% trypsin, 1 Â 10 3 cells were plated in six-well plates with 0.05% DMSO (control), TMZ, TMZ@Astro-exo, or TMZ@Astro-exo-FA.To allow colony formation, cells were incubated for 15 days after the treatment.Colonies were fixed with methanol and stained with 1% crystal violet.

| Wound-healing assay
Briefly, the U-87 cells were seeded in a complete medium of 6-well plates until a more than 80% confluency is attained.A wound was made in each monolayer using a 20-μL pipette tip.The cells were incubated for 48 h after removing the floating cells.Images were captured using an inverted microscope at 0 and 48 h and quantified using TScratch software.

| Transwell invasion assay
Briefly, the U-87 cells were added into 24-well plates with an 8-μmpore polycarbonate membrane (Cat#3422, Corning, NY, USA) coated with 20 mg of Matrigel (Cat#354230, BD Biosciences).Approximately 2 Â 10 4 cells were cultured in the upper chamber in serum-free medium.Furthermore, a medium containing 10% FBS was added to the lower chamber.After a 24-h incubation, the invasive cells were fixed using 4% PFA, stained using 0.1% crystal violet (Cat#C0121 Beyotime) for 30 min, and photographed using an inverted microscope.Three independent experiments were conducted.(1 μL) and implanted into mice brainstem as previously described operation steps. 25After a 6-day cellular implantation, mice were randomly assigned into four groups (n = 6) and treated with PBS, TMZ, TMZ@Astro-exo, or TMZ@Astro-exo-FA (5 mg/kg) to evaluate the therapeutic efficacy via the tail vein.The treatment was repeated every 3 days for a total of four doses.The body weight of mice was recorded every 3 days during therapy.On day 21, one mouse in each group was randomly euthanized to remove glioma for hematoxylineosin staining (H&E).The remaining mice were retained until they died for survival analysis.Blood was collected from each mouse for serum chemistry analysis.

| Statistical analysis
GraphPad Prism v8.0 (GraphPad Inc., La Jolla, CA, USA) was used for all statistical analyses.Data were evaluated for normality using the Kolmogorov-Smirnov test.Experimental data are presented as the mean ± standard deviation.A comparison of the two groups was executed using a two-sided Student's t-test.Multiple comparisons were conducted using one-way analysis of variance plus a two-sided Tukey's test.Survival curves were plotted using the Kaplan-Meier method and compared using log-rank tests.All tests were two-tailed and statistical significance was set at p < 0.05.

| Preparation of TMZ@Astro-exo-FA
The TMZ@Astro-exo-FA was prepared by following the procedures as illustrated in Figure 1.Exos were isolated from astrocyte U-87 through ultracentrifugation and then coated with the membraneinsertable FA derivative of DSPE-PEG 2000 -FA through its hydrophobic lipid tail.Loading of TMZ into Astro-exo-FA was achieved through an electroporation approach, and the resultant formulation was termed TMZ@Astro-exo-FA for brevity.

| TMZ@Astro-exo-FA morphology and characterization
TMZ@Astro-exo and TMZ@Astro-exo-FA morphology was further characterized using TEM.We observed that both TMZ@Astro-exo and TMZ@Astro-exo-FA were round and cup-shaped vesicles with a bilayer membrane structure (Figure 2A).Western blotting demonstrated that both TMZ@Astro-exo and TMZ@Astro-exo-FA expressed exo marker proteins CD63, CD81, and Tsg101 (Figure 2B).NTA results revealed that the main peak of the TMZ@Astro-exo diameter was 100 nm, while the main peak of the TMZ@Astro-exo-FA diameter was 110 nm.Although the particle size of TMZ@Astro-exo-FA had a minimal increase, it still conformed to the characteristics of exos, suggesting that the properties of exos were not affected by FA decoration and drug loading (Figure 2C).The in vitro release of TMZ, TMZ@Astro-exo, and TMZ@Astro-exo-FA was explored through a F I G U R E 1 Schematic diagram representing the generation of engineered exos TMZ@Astro-exo-FA.dialysis bag method.The free TMZ showed a faster release rate, with a cumulative release rate of up to 80% at 4 h.Differently, TMZ in TMZ@Astro-exo and TMZ@Astro-exo-FA was wrapped by exos, making TMZ release slower and smoother, with a cumulative release increasing to approximately 60% and 50% from 0 to 10 h, respectively (Figure 2D).These results manifested that TMZ@Astro-exo-FA did not alter exo properties and prolonged drug release in vitro.

| TMZ@Astro-exo-FA enhanced TMZ toxicity to astrocytes
TMZ@Astro-exo and TMZ@Astro-exo-FA were first labeled with the membrane dye PKH26 (red) and then co-cultured with U-87 cells to demonstrate that FA modification can improve the binding capability of Astro-exo to astrocytes.After incubation for 1 and 2 h, two types of exos (TMZ@Astro-exo and TMZ@Astro-exo-FA) were successfully detected in the recipient U-87 cells, and U-87 cells in the TMZ@ Astro-exo-FA group had a considerably higher PKH26 fluorescence signal than the TMZ@Astro-exo group (Figure 3A).Next, the toxicity of TMZ@Astro-exo-FA to U-87 cells was analyzed.Astrocytes were treated with solvent DMSO, free TMZ, TMZ@Astro-exo, and TMZ@Astro-exo-FA for different hours.We noted that TMZ demonstrated light cytotoxicity on astrocytes versus DMSO at 24, 48, and 72 h, while TMZ@Astro-exo had a stronger cytotoxicity versus free TMZ at 72 h.Notably, TMZ@Astro-exo-FA showed a significantly stronger cytotoxicity than TMZ@Astro at 24, 48, and 72 h (Figure 3B), confirming that the TMZ@Astro-exo-FA had a significant cytotoxicity to astrocytes in vitro.In line with the abovementioned results, TMZ@Astroexo-FA resulted in a two-to threefold increase in cell death as indicated by changes in the IC50 values compared to free TMZ in three types of glioma cell lines (Supplementary Table S2).To investigate the specificity and sensitivity of TMZ@Astro-exo-FA to glioma, the potential of TMZ@Astroexo-FA to kill various types of non-glioma cancers in vitro was evaluated.However, no significant difference was found between the efficiency of TMZ and TMZ@Astro-exo-FA in non-glioma cancers (Supplementary Table S2).These results indicate that the potential of Astro-exo-FA to enhance anti-tumor effect of TMZ might be limited to glioma.
3.4 | TMZ@Astro-exo-FA strengthened TMZ-mediated inhibitory effects on astrocyte proliferation, migration, and invasion he effects of TMZ@Astro-exo-FA on astrocyte proliferation, migration, and invasion were further evaluated.After 24 h of incubation, TMZ resulted in a significant decrease in the percentage of EdU-positive U-87 cells (to 40%) versus DMSO, while TMZ@Astro-exo-FA caused an overt reduction in the percentage of EdU-positive U-87 cells (to 20%) compared to TMZ@Astro-exo, as validated by the EdU assays (Figure 4A).
The clonogenic assay conducted in U-87 cells showed that TMZ repressed the colony formation ability of U-87 cells.However, TMZ@ Astro-exo exhibited a stronger inhibitory effect on the colony formation ability of U-87 cells than free TMZ, but weaker than TMZ@Astro-exo-FA.Similarly, after TMZ treatment, the migrating and invading capacities of U-87 cells were impaired.However, the inhibitory effect of TMZ@ Astro-exo on the U-87 cell behaviors was greater than that of free TMZ, but less than TMZ@Astro-exo-FA.Altogether, these findings indicated that TMZ@Astro-exo-FA strengthened TMZ-mediated inhibitory effects on astrocyte proliferation, migration, and invasion.

| TMZ@Astro-exo-FA improved the efficacy of TMZ on glioma in mouse models
We injected U-87 cells into the brains of nude mice to establish an intracranial orthotopic xenograft model to clarify the targeting ability of TMZ@Astro-exo-FA in glioma in vivo.Mice were randomly assigned to four groups, followed by PBS, TMZ, TMZ@Astro-exo, and TMZ@Astro-exo-FA administration (TMZ administered at a concentration of 5 mg/kg) through the tail vein, with injections every 3 days for four consecutive doses.Body weight changes in mice treated with different drugs were recorded.Only mice in the TMZ@Astro-exo-FA group did not have a significant decline in body weight, while mice in the remaining groups had a significant decrease in body weight with the following ranking: TMZ@Astro-exo-FA > TMZ@Astro-exo > TMZ > PBS (Figure 5A).Additionally, mice in the PBS group had the shortest survival time, which is attributed to the rapid growth of gliomas in mice in this group.In other groups, tumor growth was inhibited to varying degrees secondary to drug treatment, resulting in the prolonged survival time of mice, and the survival time ranking was TMZ@Astro-exo-FA > TMZ@Astro-exo > TMZ > PBS.Most importantly, mice in the TMZ@Astro-exo-FA group had the longest survival time, which highlighted the excellent anti-tumor therapeutic effect of TMZ@Astro-exo-FA (Figure 5B).H&E staining revealed that brain tissues derived from mice in the TMZ@ Astro-exo-FA group had the smallest tumor volume, and the ranking of brain tumor volumes in mice in four groups was as follows: TMZ@Astroexo-FA < TMZ@Astro-exo < TMZ < PBS (Figure 5C).These results indicate that TMZ@Astro-exo-FA could effectively improve the effect of TMZ in mouse models.Finally, to investigate whether Astro-exo-FA can reduce the toxicity of TMZ on the liver and kidneys of mouse model, we collected blood samples after finishing treatment and examined related parameters.Liver enzyme elevation (alanine transaminase and aspartate aminotransferase) and blood urea nitrogen were evident after treatment with free TMZ, but not with TMZ@Astro-exo-FA and TMZ@Astro-exo (Supplementary Table S3).These results show that TMZ encapsulation in the Astro-exo-FA has the potential to reduce the incidence of liver and kidney injury induced by TMZ.
LIU and ZHANG limited owing to its low lipophilicity and water solubility, including a short half-life in vivo.In recent years, the application of nanotechnology in the preparation of sustained-release targeted formulations of various drugs has developed rapidly. 27Researchers have utilized nanodrug delivery systems to heighten TMZ solubility, achieve targeted delivery, prolong circulation time in vivo, and reduce adverse drug reactions by encapsulating TMZ in liposomes, polymer micelles, or nanoparticles. 28For example, Zong et al. prepared angiopep 2-modified lipid-polymer nanoparticles for TMZ delivery to achieve synergistic effects against glioma. 29A delivery system of polyhedral oligosesesquitoxanes modified by FA and iRGD peptides shuttled TMZ to glioma cells, effectively enhancing the therapeutic effect of chemotherapy. 30Eudragit ® RS100 and chitosan-coated TMZ-loaded selenium nanoparticles improved the ability to target the delivery of TMZ to C6 cells and reduced protein levels associated with TMZ resistance. 31However, synthetic nanomaterials are difficult to degrade and easy to remove using the reticuloendothelial system, which are a disadvantage.Moreover, the TMZ nanodrug delivery systems need to overcome barriers such as blood, tumor cells, and intracellular transport after administration, which limits the efficacy of targeted delivery.
Exo is a natural carrier produced by endogenous cells, with excellent properties such as good biocompatibility, avoidance of phagocytosis by macrophages, blood vessel penetration, and capacity for crossing the BBB. 32The disadvantage of insufficient targeting of natural exos can be overcome through reasonable engineering modifications. 16A report revealed that glioma cell-derived exos carrying TMZ and dihydrotanshinone have a variety of advantages, including optimal BBB penetration, tumor-homing accumulation, and immune response stimulation. 33Liang et al. prepared a delivery system by which angiopep-2 and CD133 RNA aptamers-decorated exos loaded with O 6 -benzylguanine and TMZ exhibited efficient uptake by U-87 cells and excellent extension of the survival time of U-87-bearing mice. 34However, the efficacy of such delivery system not only hinges on their ability to target tumors efficiently, they must be able to avoid effects of systemic drug toxicity and adverse offtarget tissue effects.Surface modification of nanodelivery system with PEG can increase circulation time, allowing for tumor site accumulation, and minimizing adverse drug toxicities, which is observed with the PEGylated liposomal formulation of doxorubicin that significantly reduces doxorubicin cardiotoxicity. 35DSPE-PEG 2000 -FA is an amphiphilic linear polymer, with one end of the DSPE chain possessing lipophilicity and the other end possessing hydrophilicity. 36In the process of DSPE-PEG 2000 -FA modifying exos, the DSPE end can bind to phospholipid membranes and expose FA as a target tip to the surface of exos, which not only increases the in vivo circulation time of exos, but also targets lesions with high FA receptor expression. 37Here, F I G U R E 5 TMZ@Astro-exo-FA enhanced the effect of TMZ against glioma in mouse orthotopic xenograft models.(A) Body weight changes of mouse glioma models in PBS, TMZ, TMZ@Astro-exo, and TMZ@Astro-exo-FA groups (n = 6).
Mice were housed at 22 C in a 12-h light/12-h dark cycle in a specific pathogen-free animal house.All animal experiments were conducted in accordance with the guidelines provided by Zibo Vocational Institute.The orthotopic U-87 mouse model was established in female NCG (NOD/ShiLtJGpt-Prkdc em26 Il2rg em26 /Gpt) mice (Jicuiyaokang, Jiangsu, China).U-87 cells (1 Â 10 6 cells) were suspended in PBS (B) Kaplan-Meier survival curves of mouse glioma models in different treatment groups.(C) Hematoxylin-eosin staining images of brain tissues derived from different treatment groups.*p < 0.05, **p < 0.01, and ***p < 0.001.