Lipid nanomedicine simultaneously inhibits BRD4/PI3K and MDM2/XIAP signaling pathways for effective treatment of Medulloblastoma

Abstract

Medulloblastoma (MB) is the most common childhood brain tumor arising from the cerebellum. PI3K and BRD4 signaling pathways are known to induce MB cell growth, cancer stem cell (CSC) proliferation, and tumor resistance. Further, the tumor suppressor gene TP53 is found to be inactivated in MB due to overexpression of its negative regulator MDM2. In this study, we synthesized MDP5, a potent BRD4/PI3K dual inhibitor, and JW475A, a potent dual MDM2 and XIAP inhibitor. The combination of these two drugs significantly decreased the colony formation capacity compared to individual drugs. Given the challenge of inefficient drug transport across the blood-brain barrier (BBB), we prepared rabies virus glycoprotein (RVG) peptide decorated lipid nanoparticles (LNPs), which showed 4.9 ± 0.1 and 4.8 ± 0.1 % loading for MDP5 and JW475A, respectively. In vivo studies in mice showed that Cy5.5 labeled RVG-LNPs were detected in the brain after systemic administration. Combination drug-loaded RVG-LNPs significantly decreased the MB growth in orthotopic mouse model of MB compared to free drug combination and non-targeted LNPs. This study indicates that MDP5 and JW475A -loaded RVG-LNPs are a promising drug brain delivery system worth exploring further in clinical settings for MB therapy.

1. Introduction

Medulloblastoma (MB) is a malignant pediatric brain tumor, accounting for about 63.7 % of all embryonal tumors in children and adolescents [1,2]. Due to its complexity and heterogeneity, the traditional classification of MB was based on four molecular subgroups: wingless (WNT), sonic hedgehog (SHH), Group 3, and Group 4, with WNT group having the best while Group 3 MB having the worst clinical prognosis [3]. The revised 2016 WHO classification divides tumors into WNT-activated, SHH-activated-TP53 wild type, SHH-activated-TP53 mutant and non-WNT-SHH, where Group 3 and Group 4 MB are provisional entities to form the basis for the next generation of clinical trials and targeted therapy [4]. Tremendous improvements in therapy using a combination of surgery, craniospinal radiotherapy and chemotherapy with vincristine, cisplatin and cyclophosphamide have yielded an overall survival of 70–80 %; however, radio-polychemotherapy increases the risk of significant toxicities including neurocognitive impairment, endocrinopathy and neurologic deficits [5]. In addition, many patients need to receive craniospinal irradiation to protect against developing metastasis, which places patients at the risk for significant long-term neurocognitive side-effects and sequelae that may impair their quality of life [6]. Targeted treatment is hence strongly desired based on the aggressive nature of the tumor and treatment-related side effects [7].

MYC and BRD4 signaling pathways are persistently activated in MB and could serve a crucial role in its tumorigenesis [8]. Once activated, they may contribute to uncontrolled proliferation of cancer cells by promoting cell cycle progression, the inhibition of apoptosis and tumor metastasis [9]. Excessive MYC activation is associated with drug resistance and relapse. BRD4 activation has been found in many types of solid tumors including MB [10]. Although BRD4 inhibition works well in preclinical models, these treatments are not curative in patient and show resistance due to upregulation of compensatory pathways such as PI3K/AKT. We have previously detected high expression of MYC and BRD4 in human MB samples and cell lines (HD-MB03 and DAOY cells). PI3K signaling also cooperates with BRD4 in tumor growth and compensates the signaling upon latter’s inhibition. Based on these observations, we designed a BRD4 and PI3K dual inhibitor MDP5. In preclinical studies, MDP5 showed tumor inhibition in orthotopic MB-bearing mice by modulating MYC [11].

Tumor protein 53 (TP53) mutations greatly depend on the molecular subgroup of MB, with TP53 mutant SHH-MB forming an extremely high-risk group of patients [12]. Therefore, there is now a significant drive to learn how TP53 contributes to MB’s inherent resistance to treatment. In majority of MB cases, TP53 inhibition by murine double minute 2 (MDM2); rather than p53 mutations is responsible for its loss of activity [13]. The increased MDM2 expression in MB relative to normal cerebella, suggests that this is a promising therapeutic strategy in non-TP53-mutated MB [14]. The tumor-promoting p53-MDM2 interaction can be inhibited by the small molecule, nutlin-3, restoring p53 function [15].

X-linked inhibitor of apoptosis protein (XIAP) is an important member of the inhibitor of apoptosis. MDM2 protein contains a C-terminal RING domain that can bind to XIAP internal ribosome entry site (IRES) mRNA, enhance its translation, and result in resistance to anti-cancer treatment [16]. Further, targeting IAP gene shows reduced MB cell proliferation through disruption of the cell cycle [17]. The plausibility of targeting IAPs was first shown using a SMAC mimetic peptide. However, cannot be used clinically for brain cancer treatment. We discovered a new compound JW475A which showed dual MDM2 and XIAP inhibitor activity; hence is more tractable for clinical translation. Accordingly, we found that targeting MDM2 by JW475A showed synergistic anti-MB effects when combined with MDP5 treatment. MDP5 inhibits BRD4 and PI3K, disrupting MYC signaling, cell proliferation, and survival pathways like PI3K/AKT. On the other hand, JW475A inhibits MDM2 and XIAP, two key anti-apoptotic and tumor survival proteins: MDM2 inhibits p53, a critical tumor suppressor. XIAP blocks caspases, preventing apoptosis. By combining the two, MDP5 reduces pro-survival signaling (MYC, PI3K/AKT) and JW475A restores p53 activity and releases the brake on apoptosis by inhibiting XIAP. This dual blockade converges on both growth arrest and enhanced cell death, amplifying therapeutic efficacy.

In addition, inefficient drug transport across the BBB, and drug-induced neurotoxicity reduces benefits of chemotherapy. Often, preclinical studies do not take these barriers into account (e.g., cell culture studies, murine flank xenografts), therefore, many anti-cancer drugs fail once they are challenged with an intact brain/central nervous system (CNS) barrier system [18]. Together, these challenges highlight the need for innovative approaches to develop new effective drug delivery systems for pediatric CNS tumors. Nanoparticles (NPs) could be customized for functionalities to enhance drug transport across the BBB for carrying diverse payloads. Using receptor-mediated transcytosis (RMT), engaging receptors/proteins expressed at the BBB is used to the nonspecific interaction decrease of NPs with the body components and enhance their brain-specific uptake. However, designing NPs decorated with ligands targeting the BBB receptor has been reported with modest success due to the lack of brain-specific target protein expression. Therefore, selection of target protein is crucial. RVG peptide has the sequence of rabies virus glycoprotein that recognizes the nicotinic acetylcholine receptor (nAchR) at neuromuscular junctions, crosses the BBB and enters brain cancer cells, therefore represent a potential strategy [19,20].

In this study, we applied the strategy of simultaneous inhibition of PI3K/AKT, BRD4, MDM2, and XIAP to enhance the effectiveness of current MB therapies. We prepared RVG peptide decorated MDP5 and JW475A loaded LNPs to enhance the drug transport across the BBB to tumor cells to treat MB while decreasing the systemic toxicity.

2. Materials and methods

2.1. Cell culture and reagents

Human MB cell lines (DAOY and HD-MB03) and normal human astrocytes were purchased from the ATCC. Cells were maintained in Eagle’s minimum essential medium (EMEM) (ATCC) and Dulbecco’s Modified Eagle Medium (DMEM) (Corning) supplemented with 10 % fetal bovine serum (FBS) (Atlanta Biologicals, Fort Collins, CO, and 1 % Penicillin/Streptomycin (Sigma-Aldrich, St. Luis, MO) incubated with 5 % CO₂ at 37 °C.

All the reagents were purchased as follows: dimethyl sulfoxide and acetonitrile (Sigma-Aldrich, St. Louis, MO), 1, 2-Distearoyl-sn-glycero-3-phosphoethanolamine-poly(ethylene glycol) (DSPE-PEG₂₀₀₀), 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), and cholesterol from Avanti Lipids, Inc. (Alabaster, AL).

2.2. Synthesis of MDP5 and JW475A

Synthesis scheme of MDP5 was described previously [11]. For the synthesis of JW475A, the following method was used: Briefly, 4-amino-N-(3-(1-hydroxyethyl)phenyl)benzene sulfonamide (1 g, 3.42 mmol), 4-isopropyl benzaldehyde (0.51 g, 3.42 mmol), Sc(OTf)₃ (0.34 g, 0.68 mmol) and 4 Å molecular sieves (1 g) were mixed together in anhydrous CH₃CN (20 mL) at room temperature under argon (Fig. 1A). The reaction mixture was stirred for 1 h at room temperature. 2,3-Dihydrofuran (0.36 g, 5.13 mmol) was added via a syringe. The resulted mixture was stirred at room temperature under argon overnight. The reaction was quenched by addition of 50 mL of water at room temperature and neutralized by adding NaHCO₃. The solution was extracted with ethyl acetate (3 × 50 mL). The extracts were dried over anhydrous MgSO4, filtered, and concentrated under reduced pressure to dryness. The solid residue was purified by column chromatography (silica gel, CH₂Cl₂/acetone = 9/1 v/v) to give a white solid product (1.04 g, 61.5 % yield). We characterized JW475A with ¹H NMR and confirmed its molecular weight with mass spectroscopy.Further, we have designed the chiral synthesis scheme of the two pairs of stereoisomers in JW475A.

Fig. 1. Synthesis scheme and characterization of JW475A.

A) The two chiral centers in the 5-member rings defined, and there main two chiral centers are labeled with*. Thus, there are a total of four stereoisomers. B) ¹H NMR spectroscopy showing the characteristic protons and inset; Time Of Flight mass spectrometry determining the exact mass and corresponding molecular formula of the drug molecule. C) Chiral syntheses of the two pair of stereoisomers in JW475A. The JW475A-S-OH pair contains two isomers; and the JW475A-R-OH contains the other two isomers. Each of these two isomers can be separated by established chiral HPLC methods.

2.3. Molecular docking

The four JW475A diastereomers were docked against individual domains of the E3 ubiquitin-protein ligases XIAP and MDM2 (accession numbers: P98170, Q00987) using Glide software (Schrödinger, LLC.). Existing X-ray structures served as reference models for docking the XIAP BIR1 (residues 16–100), BIR2 (residues 154–238), BIR3 (residues 238–358), and RING domain, and the MDM2 N-terminal domain, which correspond to Protein Data Bank (PDB) PDB IDs: 4MTZ (chains A/C), 4KJU, 4HY0, 4IC2, and 1T4E (chain A), respectively. Using PyMOL, waters and excess chains were removed from the reference pdb files and each model, represented as a biological assembly, was imported into Maestro (Glide interface) and prepared for docking using the protein preparation workflow tool to perform energy minimization and optimize protein geometry and hydrogen-bond networks. Receptor grids were defined by the presence of existing ligands in each reference model, except for the XIAP RING domain, which was defined as the centroid of selected residues (A: 496 Met 1, A: 468 Leu 1, B: 496 Met 3, B: 497 Ser 3, B: 444 Leu 3). 3D structures depicting each of the four possible JW475A diastereomers were generated using SMILES strings and prepared using the LigPrep tool in Glide. The parameters for docking were set as the following: Ligand scaling factor: 0.8, partial charge cutoff: 0.15; General settings: extra precision (XP), flexible ligand sampling (sample nitrogen inversions and ring conformations, including input ring conformation), biased sampling of torsions for all predefined functional groups, and Epik state penalties applied.

2.4. MTT cell viability assay

MB cells (HD-MB03 and DAOY) were seeded in 96-well plates at a density of 3000 cells per well. The next day, different concentrations of MDP5 and JW475A inhibitors and their combination were added. The cells were incubated at 37 °C for a period of 48 h. Then, 30 μL of 3-(4, 5-dimethylthiazolyl)-2, 5-diphenyltetrazolium bromide (MTT) was added to each sample and incubated for 4 h. After that, media was removed and 200 μL of dimethyl sulfoxide solubilization solvent were added to each well. The absorbance was noted, and half-maximal inhibitory concentrations (IC₅₀) were determined by GraphPad Prism software7.0. Combination index (CI) was determined using CompuSyn software [21]. CI values indicate a synergistic effect when <1, an antagonistic effect when >1, and an additive effect when equal to 1.

2.5. Colony-formation assay

Clonogenic capacity was evaluated according to the protocol of Kumar et al., [21] where 1000 cells were treated with MDP5 or JW475A alone or in combination at their IC₅₀ values. After 48 h of treatment, a drug-free medium was added and incubated for an additional one to two weeks. Colonies were washed with PBS and fixed with formalin for 20 min and stained with 0.5 % crystal violet dye for 15 min. After staining, the plates were washed with distilled water to get rid of excessive crystal violet and dried. To determine the relative number of colonies, 10 % acetic acid was used to dissolve stained colonies and the absorbance was detected at 590 nm wavelength light in a spectrophotometer.

2.6. Cell migration assay

For Transwell migration assay, cells were trypsinized and approximately 1 × 10⁵ cells were suspended in serum free media and added into apical chamber of insert while media containing FBS was placed in the basolateral chamber of a 24 well plate. After 24 h, cells from apical side were washed with PBS thrice and removed with a cotton tipped applicator. Next, migrated cells on the dorsal side of the insert membrane were fixed and stained for 30 min with formalin and 0.5 % crystal violet, respectively. The images were captured using a light microscope at 20× magnification. After that, 10 % acetic acid solution was used to dissolve stained cells and optical density (OD) was measured at 590 nm.

2.7. Western blot assay

Treated and untreated MB cells were washed twice with cold PBS after 24 h of treatment. Cell flasks were kept on ice and lysed the cells in cell lysis buffer (Cell Signaling Technology, USA) with the cocktail of protease and phosphatase inhibitors, followed by scrapping cells with a sterile scraper. The lysates were collected in Eppendorf tubes and placed on a shaker for 30 min on ice. Then, Eppendorf tubes were centrifuged at 13,000 rpm for 20 min at 4 °C to collect the supernatant fractions. Cell lysates were separated by 4–15 % sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and subjected to Western blot analysis using a 1:1000 or 1:2000 dilution of primary antibodies according to the protocols and a 1:10000 dilution of horseradish peroxidase-conjugated secondary antibodies. The following antibodies were used for Western blot analysis: N-MYC, phosphorylated AKT (Ser473), XIAP, MDM2, Cyclin D1, p53, and β-actin (all from Cell Signaling Technology, USA). Membranes were analyzed using enhanced chemiluminescence plus reagents and scanned with the iBright Imager (ThermoFisher Scientific., San Diego, CA). The relative protein levels were quantified by densitometry with ImageJ software (National Institutes of Health, Bethesda, MD) according to the manufacturer’s instructions.

2.8. Cell cycle analysis and apoptosis assay

Cell cycle distribution was determined in proliferating MB cells. Cells (0.2 × 10⁶) were seeded in a 6 well plate, and growth inhibition was induced by adding drug solutions at the concentration of 1–5.0 μM for 48 h. Cells were harvested with 0.25 % EDTA-free trypsin (Gibco, USA) and washed twice in ice-cold PBS. Control and treated cells (0.1 × 10⁶) were stained with FxCycle PI/RNase Staining Solution (ThermoFisher Scientific) and analyzed for DNA content via flow cytometry. For apoptosis assay, the cells were collected as above and stained in the dark for 5 min with 5 μL Annexin V-FITC and 10 μL propidium iodide (PI) in 500 μL of 1 × binding buffer. Flow cytometry (FACS Calibur cytometer, BD Biosciences) was performed to quantify and calculate the cell apoptosis rate.

2.9. Tumor spheroid assay

For tumor spheroids assay, 1500 and 500 DAOY and HD-MB03 cells were suspended in 100 μL media and seeded in ultra-low attachment (ULA) 96-well plate for a week and 50 μL of media was replaced by fresh media every third day and spheroid size was tracked under fluorescent microscope. When tumor spheroid size reached to 200 μm, then spheroids were treated with MDP5, JW475A, or their combination at different concentrations. After 7 days of treatment period, spheroids were stained with Calcein^AM (2 μM) and ethidium homodimer1 (3 μM) (LIVE/DEAD^™; L3224) to check the live/dead cells. Images were taken by confocal laser microscope (Zeiss LSM 710) with 10× magnification at 40 μm depth.

2.10. Conjugation of RVG peptide with DSPE-PEG2000-NHS and characterization

RVG-PEG-DSPE was synthesized by the click chemistry between azide and alkyne group. Briefly, 1:1.5 ratio of N3-RVG and DSPE-PEG-DBCO in anhydrous DMF was subjected for stirring at 37 °C for 3 h. Product was kept at 4 °C for 24–48 h and unreacted RVG was removed by dialysis against deionized water (MW cut off 3500 Da) for 48 h at room temperature. Finally, RVG-PEG-DSPE was obtained by freeze-drying. The product was confirmed by NMR (Bruke) and MALDI-TOF-MS (Shimadzu, Japan).

2.11. Lipid nanoparticle formulation and characterization

LNPs were prepared by film hydration as described previously [22]. Briefly, millimolar ratios of lipids such as cholesterol (0.5), phosphatidyl choline (PC) lipid 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) (0.9 mol), and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino(polyethyleneglycol)-2000] (ammonium salt) (DSPE-PEG₂₀₀₀) (0.1 mol) were dissolved in chloroform and formed thin layer using a rotary evaporator to remove organic phase. Next, the round bottom flask was kept in a desiccator for overnight to dry residual organic solvents. The membrane was rehydrated with warm PBS at 60 °C, vortexed for 5 min, followed by water bath sonication for 10 min. and probe sonication with 20 % amplitude, 10 s per pulse on and off on ice-bath for 10 min.

For targeted LNPs, RVG-decorated LNPs, RVG-PEG2000-DSPE was incorporated at 5 mol% of the total lipid composition, replacing an equivalent amount of unmodified DSPE-PEG2000. The particle size and polydispersity index (PDI) of LNPs were measured using dynamic laser scattering (DLS) using a Malvern Zetasizer. The detection angle was set to 90° with 633 nm wavelength at 25 °C. The morphology of LNPs was observed by transmission electron microscopy (TEM). For that one drop of drug-loaded NPs solution was placed on a copper grid and dried at room temperature and observed under the TEM. To evaluate the encapsulation efficiency (EE) and loading efficiency (LE), drug loaded LNPs were dissolved in acetonitrile (ACN) and vortex for 5 min. Then samples were run on HPLC to get the concentration of drugs.

2.12. Establishment of xenograft and orthotopic MB tumor

Xenograft tumors were generated by injecting approximately 1 × 10⁶ MB cells subcutaneously (S.C.) on the right flank. The orthotopic MB tumors were generated by injecting 1 × 10⁵ HD-MB03 cells stereotaxically in the NSG mouse cerebellum using our established protocol [23]. Bioluminescence imaging using IVIS was done to monitor the tumor growth weekly after implantation of MB cells. When the bioluminescence reached 1 × 10⁶ p/s/cm²/sr, mice were divided into four groups (n = 5/4): i) Control group with PBS, ii) free MDP5 and JW475A (15 mg/kg), iii) non-targeted-LNPs loaded with MDP5 and JW475A (15 mg/kg), iv) RVG-LNPs loaded with MDP5 and JW475A (15 mg/kg) treatment will be given every alternative day for up to 2 weeks. At the end of the study, the mice were euthanized, and the brain tissues were isolated. We used the tissues for histopathology grading, immunohistochemistry.

2.13. Statistical analysis

In vitro experiments were performed in triplicates. The student’s t-test was used to compare two different groups, and one-way ANOVA was performed for three or more groups. The data obtained were shown as the mean ± S.D., and p < 0.05 was considered statistically significant.

3. Results

3.1. Synthesis and docking study for JW475A with MDM2 and XIAP

We have illustrated the synthesis and binding mechanism of MDP5 to BRD4 and PI3K in our previous study [11]. Here we synthesized a series of MDM2 and XIAP dual inhibitors, termed as JW series, which are structural analogs of MX-69 (Fig. S1). The synthesis and ¹H NMR characterization of the most potent compound in the series, JW475A, were detailed in Fig. 1A and B. Chiral isomers were also successfully separated by established chiral HPLC methods (Fig. 1C). To begin characterizing possible mechanisms of JW475A inhibitory activity, we performed docking studies of the four different JW475A diastereomers with the structurally characterized domains of human XIAP and MDM2 with druggable sites. These domains include the MDM2 N-terminal domain and four domains in XIAP: BIR1, BIR2, BIR3, and the C-terminal RING domain. Docking scores for each of the 20 possible complexes are in Table S1. Only two of the four diastereomers, the (S, R) and (R, R) structures, formed convincing complexes with any of the tested domains (Fig. 2 and Fig. S2). For clarity, the top scores include the diastereomers with either the S or R configuration on the ethylalcohol stereocenter, but only the R configuration at the stereocenter ortho to the piperidine nitrogen. Also note that the stereocenters on the tetrahydrofuran moiety (S, S) are the same in all four molecules.

Fig. 2. Potential complexes formed by JW475A.

In each panel, nitrogen, oxygen, and sulfur atoms are blue, red, and yellow, respectively. Carbon atoms of XIAP are cyan or light pink and differential subunits of the homodimer. Predicted polar interactions are indicated by dashed bonds with indicated distances. The molecular surfaces are shown when helpful to highlight the binding pocket. A) The pose showing the (S,R)-JW475A docked with MDM2. Carbon atoms of MDM2 are gray. B) The pose for (R,R)-JW475A docked at the dimer interface formed by two subunits of the XIAP N-terminal domain. C) The pose showing the (S,R)-JW475A diastereomer docked at the dimer interface formed by two subunits of the XIAP N-terminal domain. D) The pose for the (S,R)-JW475A diastereomer has the molecule bound in a deep hydrophobic pocket formed at the interface of the XIAP C-terminal domain homodimer. F495, known to impact MDM2 substrate binding, is positioned at the entry of pocket. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The seven rotatable bonds in JW475A, although possibly limiting potency, affords significant conformational variation, which is observed in the docking results. The best docking result with the MDM2 N-terminal domain (Glide score: −4.7) shows a compact S,R)-JW475A conformation forming two hydrogen bonding interactions with the backbone carbonyl oxygens of residues Q72 and V93 but suggests limited hydrophobic interactions necessary to support tight binding (Fig. S1). Binding to XIAP seems to be limited to the BIR1 domain and the C-terminal RING domain. The XIAP-BIR1 dimer interface can accommodate both the (R,R) and (S,R) configurations in extended conformations but in slightly different but overlapping locations giving scores of −5.1 and −6.1, respectively. While the R,R diastereomer forms four hydrogen bonds with XIAP-BIR1, there are again limited hydrophobic interactions to drive binding (Figs. 2). In contrast, the better scoring S,R diastereomer forms two hydrogen bonds with the S87 side chain hydroxyls of each of the two subunits in the XIAP-BIR1 dimer. Also, the central benzyl ring of JW475A forms a cation-pi interaction with the side chain of H67 (Table S1). If H67 lack a charge at physiological pH, a proton-pi interaction could still form.

The XIAP-RING domain possesses a deep hydrophobic pocket at the homodimer interface that was used as the target (Glide score −5.1). The S,R diastereomer forms hydrogen bonds with backbone carbonyl oxygen atoms of E447 and I458 with the JW475A sulfonamide and hydroxyl moieties, respectively. This is the only complex that exhibits a hydrogen bond with the tetrahydrofuran moiety. While the predicted 2.6 Å distance between the tetrahydrofuran oxygen and the side chain of K451, lysine side chains are quite flexible, so that hydrogen bond length may be an overestimate. Most interesting is that F495 is known to interact with XIAP substrate binding, and binding of JW475A in this mode could clearly hinder XIAP-substrate binding.

3.2. MDP5 and JW475A combination therapy inhibits MB cell proliferation more effectively compared to individual drugs

MDP5, JW475A and their combination were tested for anti-proliferative activities using SHH-driven DAOY and MYC-driven HD-MB03 cells. As shown in Fig. 3, the combination of MDP5 and JW475A inhibited the proliferation of both these cell lines more efficiently than single drugs. Both MDP5 and JW475A inhibited cell proliferation in dose-dependent manner, For MDP5, the IC₅₀ values in DAOY and HD-MB03 cells were 11.35 μM and 4.51 μM., whereas the IC₅₀ values of JW475A were 30.53 μM and 16.57 μM in DAOY and HD-MB03 cells, respectively. The cell proliferation was significantly inhibited in combination group, the IC₅₀ values were 6.58 μM in DAOY cells and 1.02 μM in HD-MB03 cells (Fig. 3). The combination of both drugs at lower concentrations showed synergistic effect calculated by Compusyn.

Fig. 3. MDP5 and JW475A combination inhibits MB cell proliferation more effectively than these individual drugs.

IC₅₀ values were calculated by using cell viability assay on DAOY and HD-MB03 MB cells. Both cell lines were treated with different concentration of A) MDP5 and B) JW475A alone and C) their combination for 48 h. The combination of lower drug concentration showed the synergistic effect in both MB cell lines.

3.3. MDP5 and JW475A combination inhibits colony formation and cell migration

Incubation of DAOY and HD-MB03 cell lines with either MDP5 or JW475A showed fewer colonies compared to the control group. However, incubation of DAOY and HD-MB03 cells with the combination of these two drugs showed the lowest number of colonies compared to the cells treated with individual drugs (Fig. 4A and 3B). Further, combination treatment significantly decreased the mobility of MB cells to cross the transwell insert membrane more efficiently than the control and single drug treated cells (Fig. 4C).

Fig. 4. MDP5 and JW475A was effective to inhibit the colony formation as well as cell migration.

A) Colony formation assay showed fewer colonies in combination group compared to alone treated and untreated groups. B) For quantitation of stained colonies, 10 % acetic acid solution was used to measure the absorbance at 590 nm. C) Transwell migration assay showed fewer number of cells migrated in combination group compared to other groups. D) Histogram showed the optical density (O⋅D) of migrated cells measured through ImageJ software (* indicates p < 0.05 compared to control untreated group, ** indicates p < 0.01, *** indicates p < 0.001, **** indicates p < 0.0001).

3.4. MDP5 and JW475A combination inhibits cell cycle progression, tumor spheroids, and promotes apoptosis in MB cells

We determined target protein expression by Western blot by reacting extracted total proteins with the following antibodies: p-Akt, N-MYC, p-PI3K, cyclin D1, MDM2, XIAP, and p53. When DAOY and HD-MB03 cells were incubated with MDP5, JW475 A or their combination for 24 h, the combination treatment showed decrease in the following protein expression more efficiently than individual drug treatment. XIAP and MDM2 are direct substrate for JW475 A, and we observe that they get decreased with combination of both drugs. Similarly, we observed that MDP5 is able to downregulate p-PI3K and p-AKT. N-MYC is the major regulator of MB, and its protein expression was downregulated by treatment of combination of MDP5 and JW475 A (Fig. 5A).

Fig. 5. Effect of MDP5 and JW475A combination treatment on target genes after incubation of DAOY and HD-MB03 cells.

A) Western blot analysis, B) apoptosis assay, C) cell cycle analysis, and D) 3D tumor spheroid assay. (* indicates p < 0.05 compared to control untreated group, ** indicates p < 0.01, *** indicates p < 0.001, **** indicates p < 0.0001).

The percentages of apoptotic DAOY cells were found to be 31.63 % and 26.4 % in MDP5 and JW475A treated group, respectively whereas the combination group showed higher apoptotic cells (46.4 %) compared to the control group (21.35 %) (Fig. 5B). MDP5 and JW475A arrested DAOY and HD-MB03 cell growth in G1 phase and their combination increased cell arrest in G1 phase to 80 % (Fig. 5C). Consistently, we observed decrease in cyclin D1 expression in the combination of MDP5 and JW475A (Fig. 5A). To further confirm the apoptosis in tumor microenvironment, we generated 3D tumor spheroids and treated with these drugs alone and their combination for 7 days. Tumor spheroids were then stained for live and dead cells. We observed reduced tumor spheroid size in all the treated groups compared to the control group. However, apoptotic cells (red) were significantly higher in the combination-treated tumor spheroids (Fig. 5D).

3.5. Spherical shaped LNPs depicted sustained release of MDP5 and JW475A

MDP5 and JW475A were encapsulated in LNPs using different lipids in millimolar ratio by thin film hydration (Fig. 6A). The average particle size for MDP5 and JW475A encapsulated LNPs were 75.3 ± 1.5 nm and 79.2 ± 1.5 nm, respectively with 0.27 ± 0.01 PDI measured by DLS. Drug-loaded LNPs showed an average 98 % of encapsulation efficiency (EE) and 4.8 % of loading efficiency (LE) (w/w%) for both drugs loaded LNPs (Fig. 6B). Surface morphology of MDP5 and JW475A co-loaded LNPs was confirmed through TEM microscope and size were found to be very similar to our DLS measurements (Fig. 6C). Further, in-vitro drug release studies showed no burst release for both MDP5 and JW475A (Fig. 6D.)

Fig. 6. MDP5 and JW475A loaded LNP formulation development and characterization.

A) Different formulation components and their concentrations used for preparing lipid nanoparticles (LNPs). B) The average particle size of drug loaded LNPs. C) Surface morphology of particles depicted by transmission (TEM) image. D) Drug loading capacity and encapsulation efficiency of LNPs. E) In vitro drug release studies. Results are presented as the mean ± S.D., n = 3.

3.6. RVG peptide conjugated LNPs effectively deliver MDP5 and JW475A to orthotopic MB tumor in mice

RVG-PEG₂₀₀₀-DSPE was successfully conjugated by strain-promoted alkyne azide reaction. PEG₂₀₀₀-DSPE-DBCO reaction with RVG-N3 in the DMF solvent followed by dialysis for 48 h. The conjugation was confirmed by MALDI-TOF-MS (Fig. S3). We did not observe a significant change in the size and entrapment efficiency of RVG peptide-conjugated LNPs. Further, we checked the biodistribution of Cy5.5 labeled non-targeted and RVG targeted LNPs in orthotopic MB bearing NSG mice. We observed a higher Cy5.5 fluorescence in vivo in the brain after 5 h of administration of RVG peptide-conjugated LNPs as compared to non-targeted LNPs when administered through the IV route. Also, ex vivo study showed higher epifluorescence in brain of RVG-LNPs treated MB mice as compared to the non-targeted LNPs treated MB mice, which confirms RVG peptide increasing brain biodistribution (Fig. S4). The data show a statistically significant increase in the brain fluorescence in the RVG-targeted LNP group compared to the non-targeted LNPs, confirming enhanced brain targeting. Additionally, we observed a significant increase in fluorescence in the spleen. This is likely due to the uptake of LNPs by the mononuclear phagocyte system, regardless of surface modifications. The presence of signal in the kidneys may be due to their role in the clearance of fluorescent dye which was conjugated to one of the lipids used for LNPs preparation rather than conjugated to the RVG peptide. However, for other organs including the heart, liver and lungs no statistically significant differences in fluorescence intensity were observed between the RVG-targeted and non-targeted groups (Fig. S4B,C).

3.7. In vivo efficacy of MDP5 and JW475A in subcutaneous and orthotopic MB tumor

We first checked the efficacy of MDP5 and JW475A in subcutaneous DAOY and HD-MB03 NSG mice. We observed a significant reduction in tumor volume and weight with MDP5 and JW475A combination treated group as compared to the control group at a dose of 15 mg/Kg (Fig. 7). After confirming the in vivo efficacy of MDP5 and JW475A, we generated orthotopic MB tumor-bearing NSG mice using stably luciferase expressing HD-MB03 cells. These tumor-bearing mice received systemic administration of RVG-targeted and non-targeted LNPs loaded with a combination of MDP5 and JW475A at the dose of 15 mg/kg every alternate day for 2 weeks. Mice were imaged by IVIS to monitor tumor growth. As shown in Fig. 8A and B, the mice treated with RVG-LNPs loaded with MDP5 and JW475A exhibited significant inhibition in tumor growth evidenced by decrease in bioluminescence signals compared to the mice treated with non-targeted LNPs and non-treated mice. We observed that body weights (Fig. 8C) and physical activities of mice were normal in both the treatment group with more weight loss in control group.

Fig. 7. Combination of BRD4 and PI3K dual inhibitor MDP5 and MDM2 and XIAP dual inhibitor JW475A as free drugs inhibits tumor grow more effectively than these individual drugs when administered systemically at the dose of 15 mg/kg into DAOY & HD-MB03 generated subcutaneous MB-bearing mice.

A&B) Tumor volume change during therapy. C&D) Photos of tumors harvested from the mice at the termination of therapy. E&F) Tumor tissue mass at the termination of therapy. Results are presented as the mean ± SD, n = 3 or 5.

Fig. 8. Combination therapy of BRD4 and PI3K dual inhibitor MDP5 and MDM2 and XIAP dual inhibitor JW475A inhibits tumor grow more effectively than these individual drugs when administered systemically at the dose of 15 mg/kg into orthotopic MB-bearing NSG mice.

A) Measurement of bioluminescence at different time points. B) Average radiance signal quantified through IVIS machine. C) Body weight of mice during therapy. Results are presented as the mean ± SD, n = 5/4. T-MDP5 + JW475A stands for targeted MDP5 + JW475A and NT-MDP5 + JW475A stands for non-targeted. MDP5 + JW475A.

Brain tissue sections from different treatment groups were examined by hematoxylin and eosin (H&E) and immunohistochemical (IHC) staining. H&E images confirmed the contained dense of MB cells in the control group. There was a significant decrease in the number of cell proliferation marker Ki-67 positive cells but a significant increase in apoptotic cell marker cleaved caspase 3 positive cells in the brain after treatment with MDP5 and JW475A loaded RVG-LNPs compared to control, free drug and non-targeted LNPs groups, indicating the most efficient tumor growth inhibition after treatment with targeted LNPs (Fig. 9).

Fig. 9. Hematoxylin and eosin (H&E) staining along with immunohistochemical staining of isolated brain tissues after systemic administration of MDP5, JW475A and their combination into DAOY cell generated orthotopic MB bearing NSG mice.

MB tissues were stained with Ki67 and Cleaved Caspase 3. Scale bar 200um, inset image 20×. T-MDP5 + JW475A stands for targeted MDP5 + JW475A and NT-MDP5 + JW475A stands for non-targeted. MDP5 + JW475A.

4. Discussion

MB is the most common childhood brain tumor arising from the cerebellum. PI3K and BRD4 signaling also play key roles in MB cell growth, CSC proliferation, and tumorigenesis. Further, MB treatment is challenging due to the development of chemoresistance, inefficient drug transport across the BBB and drug induced neurotoxicity. Hh inhibitors are effective initially to treat SHH-MB, but their repeated use develops chemoresistance due to mutations in smoothened (SMO) but can be overcome by modulating GLI, which is downstream of SMO. We previously synthesized and demonstrated that our novel BRD4 and PI3K dual inhibitor MDP5 binds strongly to BD1 and BD2 and inhibits the MYC pathway to kill MB tumor cells [11]. As monotherapy usually results in resistance and ineffective response in most cancers, combination therapies of drugs are preferred. MDM2 and XIAP are cell-survival proteins in tumor cells. MDM2 acts as an oncoprotein, promoting cancer progression by inhibiting the tumor suppressor p53, while the anti-apoptotic protein XIAP inhibits apoptosis. Simultaneously target MDM2 and XIAP expression are expected to promote cancer cell apoptosis and death. Therefore, we investigated whether MDP5 and JW475A, which is a dual MDM2 and XIAP inhibitor, effectively treat MB in mice. We selected DAOY and HD-MB03 cell lines to represent the two major molecular subtypes of MB. DAOY cells are a well-characterized model of SHH-driven MB, known for TP53 wild-type status and moderate aggressiveness. In contrast, HD-MB03 cells are derived from a Group 3 MB, characterized by high MYC amplification, aggressive tumor behavior, and poor prognosis. Using these two cell lines allowed us to evaluate the therapeutic efficacy of our dual inhibitor combination across distinct molecular backgrounds, including one that mimics treatment-resistant, high-risk MB (Group 3) and one representing a moderate-risk subtype (SHH). This selection provides a more clinically relevant assessment of our therapeutic strategy.

Amplification of c-MYC, N-MYC, and TP53 mutations are typical molecular changes driving DNA damage-induced apoptotic resistance and are commonly enriched during the recurrence of MB. While c-MYC amplification is more common in Group 3 MB and is linked with a worse prognosis, it also occurs in other subgroups [24]. Recent breakthroughs in MB of heterogeneity research have further revealed the involvement of TP53 mutations/inactivation [12]. TP53 mutations are substantially dependent on the molecular subgroup of MB, with TP53 mutant SHH-MB patients constituting an exceedingly high-risk group. Therefore, therapies that can inhibit the MYC pathway and TP53 are expected to reduce the recurrence of MB and improve the overall survival of the MB patients.

Overcoming anti-apoptotic mechanisms is an effective therapeutic strategy in highly resistant brain tumors, including MB. Overexpression of both human MDM2 and XIAP is detected in different tumor cell types including MB cells. In this study, we have explored MDP5 combination with a potent MDM2 and XIAP dual inhibitor JW475A, which was designed based on the tetrahydroquinoline scaffold from the previously reported lead compound JW-2–107 [25]. MDM2 and XIAP are crucial proteins for the survival of tumor cells and control TP53 protein expression. MDM2 protein contains a C-terminal RING domain that can bind to XIAP internal ribosome entry site (IRES) mRNA, enhance its translation, and result in resistance to anti-cancer treatment. Wu et al. reported JW475 A to be 25-fold more potent than nutlin-3 (MX-69) against leukemia EU-1 cells [26]. Similarly, we observed JW475 A to be more potent in D-341, HD-MB03, and DAOY cells than MX-69, JW373, and JW-3151 (Fig. S1). Our compound, JW475A, is an MDM2 and XIAP dual inhibitor; hence is more suitable for clinical translation. Accordingly, we found that targeting MDM2 by JW475A showed synergistic anti-MB effects when combined with MDP5 treatment. Moreover, we observed JW475A and MDP5 combination could synergistically inhibit MB cell proliferation by downregulating both p53 and MYC protein as compared to their monotherapies (Figs. 3 and 5).

Also, colony formation and cell migration ability of MB tumor cells were significantly reduced by MDP5 and JW475A combination (Fig. 4). The protein expression of key genes such as N-MYC, p-AKT, p-PI3K, MDM2, XIAP, and p53 was downregulated after incubation with JW475A and MDP5 combination (Fig. 5A), indicating the combination therapy of these two drugs could synergistically regulate the oncogenic pathways and increase apoptosis.

MDP5 and JW475A arrested the cells in G1 phase in DAOY and HD-MB03 cells (Fig. 5C). We further checked the ability of drugs to penetrate inside the 3-D tumor spheroids. We observed that the combination of both drugs was able to disrupt the 3-D spheroid and an increase in the red signal depicting dead cells (Fig. 5D). Further, we confirmed the ability of both drugs to demonstrate anti-tumor activity in-vivo in subcutaneous MB-bearing mice. We observed that MDP5 and JW475A combination was able to decrease the tumor volume by eight-fold compared to the control group in DAOY cells-generated subcutaneous tumors at a dose of 15 mg/Kg in 21 days (Fig. 7A, C and E). Also, we observed significant tumor volume reduction in HD-MB03 cells-generated subcutaneous tumors, indicating that MDP5 and JW475A combination has an effective anti-tumor response in SHH-driven MB as well as MYC-driven MB (Fig. 7B, D and F).

Herein, we utilized lipid nanocarriers to encapsulate MDP5 and JW475A. LNPs are increasingly reported to improve tumor penetration and biocompatibility, improve drug solubility, pharmacokinetics, and reduce off-target toxicity. We optimized the ratio of mPEG-DSPE, cholesterol, and DOPC to produce spherical-shaped nanoparticles with a hydrodynamic size of ~80 nm (Fig. 6). However, BBB is the major hurdle for treating brain disease, which has impeded the translation of a gamut of therapies irrespective of their high anti-tumor efficacy in vitro as well as in subcutaneous mouse models. Peptide-targeted LNPs have demonstrated great potential to circumvent BBB and increase the drug biodistribution to the brain [27]. In our previous studies, we used ApoE mimetic peptide as a targeting agent which binds to low density lipoprotein receptors (LDLR) present on the BBB [23,27]. We observed that our ApoE-targeted nanocarriers circumvented BBB and demonstrated greater uptake in tumor cells than non-targeted nanocarriers [29]. Despite an increase in drug delivery to tumor cells, we also observed an increase in drug accumulation in the liver due to the presence of LDL receptors in the liver. Therefore, in this study, we conjugated RVG peptide to DSPE-PEG₂₀₀₀ and utilized it as a moiety to circumvent the BBB. We observed that RVG-LNPs depicted higher Cy5.5 fluorescence in the brain ex-vivo as well as in-vivo after 5 h of systemic administration (Fig. S3), demonstrating better ability to circumvent BBB as compared to the non-targeted nanoparticles.

Despite of great efficacy of MDP5 and JW475A free drug combinations in the subcutaneous MB-mouse model (Fig. 7), we did not observe a significant effect of the free drug combination on the orthotopic mouse model at a dose of 15 mg/Kg IV administration compared to the control group. The possible explanation for the sub-optimal efficacy of free drugs is their inability to cross the tight junctions of BBB. We observed that our LNPs showed a better anti-tumor effect with a more prominent decrease in tumor bioluminescence. RVG-targeted LNPs showed an eight-fold reduction in tumor bioluminescence compared to the control group and a five-fold reduction in the free drug combination group with marginal weight loss (Fig. 8).

In conclusion, the key hurdles to MB therapy are the BBB, chemoresistance, and safety. Systemic administration of MDP5 and JW475A-loaded RVG-LNPs effectively delivered these drugs to MB tumor sites and significantly inhibited MB progression compared to free drug physical mixtures.

Abbreviation:

BBB MB

Blood-Brain Barrier

BRD4

Bromodomain-containing Protein 4

CSC

Cancer Stem Cell

DLS

Dynamic Light Scattering

DMEM

Dulbecco’s Modified Eagle Medium

DOPC

Dioleoyl phosphatidyl choline

DSPE

Distearoyl phosphatidyl ethanolamine

EMEM

Eagle’s Minimum Essential Medium

FACS

Fluorescence-Activated Cell Sorting

FBS

Fetal Bovine Serum

FITC

Fluorescein Isothiocyanate

GLI

Glioma-associated Oncogene Homolog

HPLC

High Performance Liquid Chromatography

IC₅₀

Half-maximal Inhibitory Concentration

IRES

Internal Ribosome Entry Site

IV

Intravenous

IVIS

In Vivo Imaging System

LNPs

Lipid Nanoparticles

MALDI-TOF-MS

Matrix-Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry

MDM2

Murine Double Minute 2

MTT

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

nAChR

Nicotinic Acetylcholine Receptor

NMR

Nuclear Magnetic Resonance

NPs

Nanoparticles

NSG

NOD scid gamma (mice strain)

OD

Optical Density

PBS

Phosphate-Buffered Saline

PC

Phosphatidylcholine

PEG

Poly(Ethylene Glycol)

PI

Propidium Iodide

PI3K

Phosphoinositide 3-Kinase

RMT

Receptor-Mediated Transcytosis

RVG

Rabies Virus Glycoprotein

SDS-PAGE

Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis

SHH

Sonic Hedgehog

SMAC

Second Mitochondria-derived Activator of Caspase

SMO

Smoothened

TEM

Transmission Electron Microscopy

TP53

Tumor Protein p53

WNT

Wingless-related Integration Site (signaling pathway)

XIAP

X-linked Inhibitor of Apoptosis Protein.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jconrel.2025.114266.

Data availability

The paper and/or the Supplementary Materials contain all the information required to assess the conclusions made in the paper. You may ask the authors for more information about this paper.