Abstract

   Temozolomide (TMZ) serves as the principal chemotherapeutic agent for
   glioma; nonetheless, its therapeutic efficacy is compromised by the
   rapid emergence of drug resistance, the inadequate targeting of glioma
   cells, and significant systemic toxicity. ARV‐825 may play a role in
   modulating drug resistance by degrading the BRD4 protein, thereby
   exerting anti‐glioma effects. Therefore, to surmount TMZ resistance and
   achieve efficient and specific drug delivery, a dual‐targeted
   glutathione (GSH)‐responsive nanoparticle system (T+A@Glu‐NP) is
   designed and synthesized for the co‐delivery of ARV‐825 and TMZ. As
   anticipated, T+A@Glu‐NPs significantly enhanced penetration of the
   blood‐brain barrier (BBB), facilitated drug uptake by glioma cells, and
   exhibited efficient accumulation in brain tissue. Additionally,
   T+A@Glu‐NPs exhibited augmented efficacy against glioma both in vitro
   and in vivo through the induction of apoptosis, inhibition of
   proliferation, and cell cycle arrest. Furthermore, mechanistic
   exploration revealed that T+A@Glu‐NPs degraded the BRD4 protein,
   leading to the downregulation of Notch1 gene transcription and the
   inhibition of the Notch1 signaling pathway, thereby augmenting the
   therapeutic efficacy of glioma chemotherapy. Taken together, the
   findings suggest that T+A@Glu‐NPs represents a novel and promising
   therapeutic strategy for glioma chemotherapy.

   Keywords: ARV‐825; glioma chemotherapy, notch1; target delivery;
   temozolomide
     __________________________________________________________________

   In this study, a Glu‐modified and GSH‐responsive nanoparticle is
   constructed to enhanced drug delivery and targeting capabilities. Sugar
   modification helps nanoparticles penetrate the BBB and target gliomas,
   eventually responsively releasing the drug in the high GSH environment
   within the glioma cells. Encapsulating TMZ and ARV‐825 within
   nanoparticles can enhance the chemotherapeutic efficacy of TMZ in
   gliomas.

   graphic file with name ADVS-11-2409753-g010.jpg

1. Introduction

   The infiltrative growth pattern of gliomas frequently precludes
   complete surgical resection of the tumor, making chemotherapy as an
   indispensable component in the treatment regimen for gliomas.^[ [42]^1
   , [43]^2 ^] Temozolomide (TMZ), as a clinical first‐line
   chemotherapeutic agent for gliomas, is capable of inducing DNA damages
   of glioma cells to further kill residual tumor tissue.^[ [44]^3 ,
   [45]^4 ^] Nonetheless, the median survival for the majority of glioma
   patients remains less than two years, potentially due to the rapid
   development of drug resistance, inadequate targeting of glioma cells,
   and systemic toxicity.^[ [46]^5 , [47]^6 , [48]^7 , [49]^8 , [50]^9 ^]
   Thus, it is urgent to explore novel drug combinations and delivery
   strategies to overcome these obstacles.

   The development of resistance to TMZ poses unprecedented challenges in
   the chemotherapy of gliomas. However, the mechanisms underlying TMZ
   resistance in gliomas remain insufficiently explored.^[ [51]^10 ,
   [52]^11 , [53]^12 ^] Accumulating researches show that
   bromodomain‐containing protein 4 (BRD4) is upregulated in various solid
   tumors including head and neck, breast and brain tumors,^[ [54]^13 ,
   [55]^14 , [56]^15 ^] in which BRD4 protein could regulate certain
   oncogenes expression to participate in tumorigenesis.^[ [57]^16 ,
   [58]^17 ^] Furthermore, some studies have demonstrated that BRD4
   protein would upregulate the O6‐methylguanine‐DNA methyltransferase
   (MGMT)/DNA repair pathway to repair the DNA damage caused by
   chemotherapy drugs.^[ [59]^18 , [60]^19 , [61]^20 ^] In addition,
   numerous bromodomain and extra‐terminal domain (BET) inhibitors have
   been employed in tumor chemotherapy, exhibiting some efficacy and
   conferring a survival benefit for glioma patients.^[ [62]^21 , [63]^22
   , [64]^23 ^] All of the evidences showed that the BRD4 protein, as an
   epigenetic regulation factor, may be a potential and novel target to
   enhance the efficacy of TMZ in glioma treatment. However, most
   small‐molecule inhibitors of the BRD4 protein are more susceptible to
   drug resistance and have severe side effects due to their high dosage
   requirements, low selectivity, and transient effects. To overcome these
   challenges, the novel proteolysis‐targeting chimera (PROTAC) strategy
   has been developed, which offers greater selectivity and can induce
   efficient degradation with minimal quantities.^[ [65]^24 , [66]^25 ^]
   Notably, ARV‐825, a BRD4 PROTAC, has been shown to significantly
   inhibit glioma growth at nanomolar concentrations by inducing apoptosis
   and inhibiting cell proliferation, thereby further corroborating this
   notion.^[ [67]^26 ^] Consequently, we endeavored to incorporate ARV‐825
   as an adjuvant in TMZ‐based glioma chemotherapy.

   Although novel drug combinations may sensitize glioma chemotherapy, low
   blood‐brain barrier (BBB) penetration and poor glioma targeting can
   impair the efficacy of the drugs.^[ [68]^27 , [69]^28 , [70]^29 ^] To
   overcome these hindrances, large number of researches had focused on
   the development of advanced BBB crossing strategies, such as
   receptor‐mediated transcytosis, the use of neurotropic viruses,
   nanoparticles and exosomes.^[ [71]^30 , [72]^31 , [73]^32 , [74]^33 ^]
   Compared to other strategies, receptor‐mediated transcytosis is more
   sensitive and specific for drug delivery, as numerous receptors are
   highly expressed on the cerebrovascular endothelium and can facilitate
   drug delivery systems with corresponding ligand modifications to cross
   the BBB, such as transferrin receptor (TfR), insulin and insulin‐like
   growth factor receptor (IGFR), low density lipoprotein receptor (LDLR),
   and glucose transporter (GLUT).^[ [75]^34 , [76]^35 , [77]^36 ^] Among
   these, the GLUTs are highly expressed on both the cerebrovascular
   endothelium and the cell membrane surface of glioma cells, which can
   promote the penetration of glucose through the BBB and its subsequent
   uptake by glioma cells. Therefore, a number of Glu‐modified drug
   delivery nanoparticles have been verified to indeed facilitate
   blood‐brain barrier penetration and rapid uptake by gliomas, thereby
   enhancing the efficacy of drugs.^[ [78]^37 , [79]^38 ^]

   Inspired by previous explorations, a Glu‐modified nanoparticle
   (T+A@Glu‐NPs) was designed and constructed for combined delivery of
   ARV‐825 and TMZ for the glioma treatment in this study (Scheme [80]1 ).
   The T+A@Glu‐NPs consist of Glu‐modified nano micelles, Glu‐PEG‐PCL, and
   glutathione (GSH)‐responsive nano micelles, PEG‐SS‐PCL, which form the
   shell of the nanoparticles. ARV‐825 and TMZ were encapsulated in the
   hydrophobic core. Elaborate experiments were designed to characterize
   and validate the function of T+A@Glu‐NPs. It is revealed that glucose
   modification significantly augmented blood‐brain barrier penetration
   and facilitates drug uptake by glioma, and T+A@Glu‐NPs rapidly
   disintegrated to release the drugs in the high GSH environment within
   glioma, resulting in effective intracranial accumulation of the drug.
   In vitro anti‐glioma experiments were conducted to analyze the effects
   of the combined drugs on cell proliferation, cell cycle, cell apoptosis
   and other tumor biological behaviors. In addition, the therapeutic
   effects of the combined drug groups were evaluated in three murine
   models of glioma. Further exploration of the underlying mechanisms
   disclosed that BRD4 protein binds to the Notch1 promoter region,
   regulates its transcriptional expression and leads to TZM treatment
   sensitization. In conclusion, our studies demonstrated that the
   Glu‐modified and GSH‐responsive nanoparticles enhanced drug delivery
   and targeting capabilities. Therefore, the T+A@Glu‐NPs represents a
   novel and promising drug combination strategy, offering new insights
   into the clinical treatment of glioma.

Scheme 1.

   Scheme 1
   [81]Open in a new tab

   Schematic illustration of the brain‐targeting and therapeutic mechanism
   of T+A@Glu‐NPs against glioma. Glu‐PEG‐PCL, PEG‐SS‐PCL, TMZ, and
   ARV‐825 were used to prepare T+A@Glu‐NPs nanoparticles via a
   self‐assembly method. Administered intravenously, T+A@Glu‐NPs leverage
   the glucose transporters expressed by the cerebral microvascular
   endothelium to penetrate the blood‐brain barrier and reach the tumor
   site. They then bind to the transporters highly expressed on the
   surface of glioma and are efficiently internalized into the cells.
   After lysosomal escape and response to glutathione (GSH) reduction,
   T+A@Glu‐NPs disintegrate and rapidly release TMZ and ARV‐825. TMZ acts
   on DNA to exert its cytotoxic effect, while ARV‐825 induces the
   sustained degradation of BRD4 protein through the ubiquitin‐proteasome
   pathway, thereby downregulating Notch1 and its associated signaling
   pathways. This further enhances the apoptotic and cell cycle arrest
   effects induced by TMZ in glioma cells, achieving a more potent
   glioma‐killing effect.

2. Results

2.1. Preparation and Characterization of T+A@Glu‐NPs

   Glu‐modified nano micelles (Glu‐PEG‐PCL) and glutathione
   (GSH)‐responsive nano micelles (PEG‐SS‐PCL) were used to encapsulate
   ARV‐825 and TMZ, to increase the water solubility and stability of the
   drugs, and to enhance the targeting ability of this nanoparticle and
   its ability to penetrate the BBB. To further identify the interaction
   process of Glu‐PEG‐PCL, PEG‐SS‐PCL, ARV‐825, and TMZ, molecular
   dynamics simulation was performed to investigate their interactions in
   normal physiological environment (pH 7.4, GSH 0.1mM) and
   tumor‐mimicking high GSH environment (GSH 10mM). In normal
   physiological environments, ARV‐825 and TMZ presented constant change
   in position and conformations while interacting with Glu‐PEG‐PCL and
   PEG‐SS‐PCL within 20 ns, and gradual formation of tight interactions
   between drugs and nano micelles, ultimately forming stable drug‐loaded
   nanoparticles (Figure [82]1A; Figure [83]S1A, Supporting Information).
   And in tumor‐mimicking high GSH environment, after shedding PEG in
   response to the breakage of PEG‐SS‐PCL disulfide bond, the drug and
   nano micelles detached from each other to promote the release of the
   drug within 20 ns (Figure [84]1B; Figure [85]S1B, Supporting
   Information). Then, we synthesized the PEG‐SS‐PCL and modified glucose
   to MAL‐PEG‐PCL to form targeted Glu‐PEG‐PCL (Figure [86]S2A–C,
   Supporting Information), whose structure was confirmed via infrared
   spectroscopy analysis (Figure [87]S3, Supporting Information).
   Subsequently, the drug loaded nanoparticles, termed as T+A@Glu‐NPs, was
   fabricated via self‐assembly of PEG‐SS‐PCL, Glu‐PEG‐PCL, ARV‐825 and
   TMZ. Detection of dynamic light scattering (DLS) showed that
   T+A@Glu‐NPs had a small particle size (41.50 ± 0.44 nm) and
   polydispersity index (0.120 ± 0.027), and zeta potential of ‐9.32±0.480
   mV (Figure [88]1C,D). T+A@Glu‐NPs presented a uniformly distributed
   spherical structure as observed by transmission electron microscope
   (TEM) (Figure [89]1E). Furthermore, stability and GSH‐responsive
   release of T+A@Glu‐NPs were determined. As shown in Figure [90]1F–H and
   Figure [91]S4 (Supporting Information), the nanoparticles were well
   stable in the simulated physiological environment and released rapidly
   and responsively in a 10 mM GSH environment. Finally, we used a
   single‐layer BBB model to initially assess the effect of T+A@Glu‐NPs
   penetration rate of the blood‐brain barrier compared with T+A@NPs in
   vitro, and found that after glucose modification, nanoparticles BBB
   penetration ratio increased by ≈15% (Figure [92]1I,J).

Figure 1.

   Figure 1
   [93]Open in a new tab

   Characterizations of T+A@Glu‐NPs. Molecular dynamics simulation
   analysis A) the self‐assembly of TMZ, ARV‐825 and Glu‐PEG‐PCL /
   PEG‐SS‐PCL in the normal physiology environment (PH 7.4, GSH 0.1 mM).
   B) the GSH responsive and drug release process of T+A@Glu‐NPs in high
   GSH environment (GSH 10 mM). Conformations (I), (II), (III), (IV), (V),
   and (VI) represented the snapshots of the interaction between TMZ,
   ARV‐825 and Glu‐PEG‐PCL/PEG‐SS‐PCL at 0, 4, 8, 12, 16 and 20 ns,
   respectively. C) Particle size distribution of T+A@Glu‐NPs. D) Zeta
   potential of T+A@Glu‐NPs. E) TEM image of T+A@Glu‐NPs (scale bar = 50
   nm). F) The stability experiment of T+A@Glu‐NPs. G) The glutathione
   reductive responsive release of TMZ and ARV‐825 from T+A@Glu‐NPs. H)
   TEM image of the reductive response morphological changes of
   T+A@Glu‐NPs. (I) Single‐layer BBB (Blood‐Brain Barrier) model
   penetration experiment of T+A@Glu‐NPs. (J) Statistical analysis of the
   BBB penetration rate. Data are presented as mean ± SD. No significant
   difference is marked with ns. *P < 0.05, **P < 0.01, ***P < 0.001 and
   ****P < 0.0001.

2.2. Validation of T+A@Glu‐NPs Targeting In Vitro and In Vivo

   Chlorin e6 (Ce6) is auto‐fluorescent and can be used to probe the
   metabolic kinetics of drug‐carrying nanomaterials in vivo and the
   synthesis process of Ce6, Ce6@NPs, and Ce6@Glu‐NPs is presented in
   Supporting information. We first carried out cell uptake studies in
   vitro. As shown in Figures [94]2A–C and [95]S5 (Supporting
   Information), well‐grown glioma cells were treated with the same
   concentration of Ce6, Ce6@NPs, and Ce6@Glu‐NPs. We subsequently
   determined the uptake ratio of Ce6 by flow analysis. We found that the
   uptake ratio of Ce6@Glu‐NPs reached higher than 90% at 2 h, and it did
   not decrease significantly at 6 h. Whereas the free Ce6 group had a low
   level of uptake, the Ce6@NPs showed an intermediate rate of uptake but
   a faster decline. Thus, Glu‐modified nanoparticles both increased
   cellular uptake and prolonged the intracellular action time of the
   drugs. Besides, we found that the proportion of Glu‐PEG‐PCL in the
   nanoparticle shell affects the efficiency of drug uptake by the cells,
   as showed in Figure [96]S6 (Supporting Information), the Ce6@Glu‐NPs L
   group took up more drug than the Ce6@Glu‐NPs H group. Second, we
   further examined nanoparticles travelling in glioma cells by confocal
   imaging and found that the concentration of Ce6@Glu‐NPs was the highest
   among the three groups in the cell at the same time point. Furthermore,
   we found that the Ce6@Glu‐NPs moved from the cytosol to lysosomes and
   then escaped into the nucleus (Figure [97]2D,E; Figure [98]S7,
   Supporting Information). Finally, we constructed a glioma orthotopic
   model to verify the ability of Ce6@Glu‐NPs to target tumors in vivo. As
   shown in Figure [99]2F,G and Figure [100]S8 (Supporting Information),
   the Ce6@Glu‐NPs accumulated in the brain tissue faster and more at the
   same time point and lasted for a longer time. Besides, the Ce6@Glu‐NPs
   concentrations remained highest in the isolated brain tissue at 24 h
   (Figure [101]2H,I; Figure [102]S9, Supporting Information). Further, we
   separated normal brain tissue from glioma tissue and conducted
   fluorescence intensity measurements. We found that the average
   fluorescence intensity of Ce6 in normal brain tissue was essentially
   consistent, while in the glioma tissue, the fluorescence intensity of
   the Ce6@Glu‐NPs group was the highest. This further proves the
   dual‐targeting functionality of Ce6@Glu‐NPs (Figure [103]S10,
   Supporting Information). Finally, we found that Ce6 was predominantly
   distributed in the liver in isolated organs, suggesting that it is
   metabolized through the liver (Figures [104]S11 and [105]S12,
   Supporting Information). All the results indicated that the
   Glu‐modified nanoparticles have superior targeting ability.

Figure 2.

   Figure 2
   [106]Open in a new tab

   Enhanced targeting ability of T+A@Glu‐NPs in vitro and in vivo. A) FCM
   of Ce6 fluorescence intensity after LN229 incubated with Ce6, Ce6@NPs,
   Ce6@Glu‐NPs for 2 h. B) Statistical analysis of cell uptake rate after
   LN229 incubated with Ce6, Ce6@NPs, Ce6@Glu‐NPs for 2 h. C) Cellular
   uptake rate curves of different treatment groups. The untreated group
   was used as the control. D) The confocal images of the cells after
   LN229 incubated with Ce6, Ce6@NPs, Ce6@Glu‐NPs for 1 h and 4 h. E) The
   distribution of Ce6, Hochest33442 and LysoTracker Green DND‐26 in LN229
   cells was examined using confocal imaging after incubated with
   Ce6@Glu‐NPs for 1h and 4 h. F) The distribution and changes in the Ce6
   average radiant efficiency in the in vivo images of mice after tail
   vein injection of Ce6, Ce6@NPs, Ce6@Glu‐NPs at 3 h, 6 h, 9 h, and 12 h.
   G) Statistical analysis of the Ce6 average radiant efficiency in mice
   at 12 h in different treatment groups. H) The initial tumor size in
   vivo images of mice in each group and the distribution images of Ce6
   average radiant efficiency in mouse brain tissues isolated at 24 h. I)
   Statistical analysis of Ce6 average radiant efficiency in mouse brain
   tissues isolated at 24 h. Data are presented as mean ± SD. No
   significant difference is marked with ns. *P < 0.05, **P < 0.01, ***P <
   0.001 and ****P < 0.0001.

2.3. T+A@Glu‐NPs Induced Cells Proliferation Inhibition and Cells Apoptosis
In Vitro

   We then assessed the cytotoxicity of T+A@Glu‐NPs against glioma cells
   through Cell Count Kit‐8 (CCK‐8) assay. The results showed the
   viability of cells treated with T+A@Glu‐NPs decreased more than the
   cells treated with T@Glu‐NPs only (Figure [107]3A; Figures
   [108]S13–S15, Supporting Information). Meanwhile, the inhibitory
   phenomenon was observed in colony formation assay. The clone numbers of
   cells incubated with T+A@Glu‐NPs were less than those of other groups,
   which were treated by T@Glu‐NPs, A@Glu‐NPs and NS respectively
   (Figure [109]3B, C; Figures [110]S16 and [111]S17, Supporting
   Information). We further evaluated the influence of the T+A@Glu‐NPs on
   the cell cycle in different glioma cell lines. We found that
   T+A@Glu‐NPs would significantly block the cell cycle of GL261 cells and
   U251 cells in G2/M phase, but there was a difference between U87cells
   and LN229 cells (Figure [112]3D,E; Figures [113]S18–S20, Supporting
   Information). The results may be caused by the glioma heterogeneity.
   Western blot analysis was carried out to investigate the molecular
   mechanism of the T+A@Glu‐NPs for glioma suppression. As showed in
   Figure [114]3H, Figures [115]S21 and [116]S22 (Supporting Information),
   despite T@Glu‐NPs showing opposite trends in CDK4 expression in GL261
   and LN229 cell lines, which could be due to the inherent differences
   between the two cells and the complex regulatory mechanisms of cyclins.
   The expression of cell cycle‐related proteins was downregulated
   obviously after glioma cells being incubated with T+A@Glu‐NPs for 48h,
   including cyclin D1 and CDK4. The results further confirmed the
   T+A@Glu‐NPs could block the cell cycle of glioma. Besides, we found the
   expression level of p‐STAT3 and STAT3 was also decreased, which were
   related to the cell proliferation in glioma.

Figure 3.

   Figure 3
   [117]Open in a new tab

   T+A@Glu‐NPs induced cells proliferation inhibition and cells apoptosis
   in vitro. A) Cell viabilities of GL261 cells treated with A@Glu‐NPs,
   T@Glu‐NPs and T+A@Glu‐NPs at different concentrations for 48 h (n = 3).
   The untreated group was used as a control. B) Representative images of
   colony formation of GL261 cells incubated with 100 ng mL^−1 A@Glu‐NPs,
   25 µM T@Glu‐NPs and T+A@Glu‐NPs. C) Quantification of colony formation
   (n = 3). D) GL261 cells cycle distribution determined by flow cytometry
   after treated with 100 ng mL^−1 A@Glu‐NPs, 100 µM T@Glu‐NPs and
   T+A@Glu‐NPs for 48 h. E) Quantification of cell cycle distribution (n =
   3). F) Flow cytometry detection of GL261 cells apoptosis after GL261
   cells treated with 200 ng mL^−1 A@Glu‐NPs, 75 µM T@Glu‐NPs and
   T+A@Glu‐NPs for 48 h. G) Statistical analysis of apoptosis cells (n =
   3). H) Representative Western blot analysis of P‐STAT3, STAT3, Cyclin
   D1, CDK4 and GAPDH in GL261 cells treated with 100 ng mL^−1 A@Glu‐NPs,
   100 µM T@Glu‐NPs and T+A@Glu‐NPs for 48 h. GAPDH served as a loading
   control. I) Expressions of apoptosis related proteins (pro‐caspase 3,
   cleaved caspase 3, PARP) and the loading control GAPDH were determined
   by western blot analysis after GL261 cells treated with 100 ng mL^−1
   A@Glu‐NPs, 100 µM T@Glu‐NPs and T+A@Glu‐NPs for 48 h. J) Quantitative
   analysis of all the western blot results. Data are presented as mean ±
   SD. No significant difference is marked with ns. *P < 0.05, **P < 0.01,
   ***P < 0.001 and ****P < 0.0001.

   To evaluate the effect of the chemotherapeutics on glioma cell
   apoptosis, FCM analysis and western blot assay were performed in the
   study. The T+A@Glu‐NPs showed strongest capacity to induce cell
   apoptosis in GL261 cells and U87 cells (Figure [118]3F,G; Figure
   [119]S23, Supporting Information). Meanwhile, Western blot analysis
   results demonstrated that the T+A@Glu‐NPs induced cleavage of caspase 3
   and PARP to activate the apoptosis pathway (Figure [120]3I; Figures
   [121]S21 and [122]S22, Supporting Information). All results showed that
   T+A@Glu‐NPs produce anti‐glioma efficacy by increasing cell apoptosis,
   blocking cell cycle and inhibiting cell proliferation.

2.4. T+A@Glu‐NPs Enhanced the Antitumor Efficacy of TMZ In Vivo

   To evaluate the anti‐glioma effect of T+A@Glu‐NPs in vivo, we first
   established a GL261 subcutaneous tumor model. According to the schedule
   shown in Figure [123]4A, the bearing tumor mice were randomly allocated
   into NC, Blank‐NPs, A@Glu‐NPs, T@Glu‐NPs or T+A@Glu‐NPs treatment group
   after the tumor volume grow up to ≈62.5 mm^3, which were intravenously
   injected with different drugs once the other day for five times. During
   the treatment, the tumor volume and the mouse body weight before each
   injection were monitored. At the endpoint, both the T@Glu‐NPs and
   T+A@Glu‐NPs exhibited superior glioma inhibition capacity compared with
   the NC, Blank‐NPs and A@Glu‐NPs groups, and the T+A@Glu‐NPs attained a
   tumor suppression rate of 90%, while it was merely ≈50% in the
   T@Glu‐NPs group (Figure [124]4C–E). Moreover, the results of
   immunohistochemistry and immunofluorescence showed more significantly
   decrease in Ki67 index and increase in the mean fluorescence density of
   TUNEL apoptosis in the T+A@Glu‐NPs group compared to the T@Glu‐NPs
   group, suggesting that the T+A@Glu‐NPs exerted anti‐glioma effects by
   inhibiting tumor cell proliferation and inducing cells apoptosis
   (Figure [125]4F–I). GL261 glioma orthotopic mouse model was
   subsequently established to further assess the anti‐glioma efficacy of
   T+A@Glu‐NPs in vivo. The tumor‐bearing mice were randomly divided into
   five groups, including NC, Blank‐NPs, A@Glu‐NPs, T@Glu‐NPs, and
   T+A@Glu‐NPs, which received the corresponding drugs administered
   through the tail vein every two days, and in vivo monitoring of the
   tumor growth and records of the body weight changes of the mice were
   conducted (Figure [126]5B). As shown in Figure [127]5A,C, there was no
   significant difference in tumor growth among NC, Blank‐NPs and
   A@Glu‐NPs groups, while tumor growth was inhibited in the T@Glu‐NPs
   group and the T+A@Glu‐NPs group, and the tumor suppression rate of
   T+A@Glu‐NPs group was ≈90%. After treatment, half of the mice were
   randomly selected for eye blood sampling in each group, and performed
   safety assessment and HE staining after stripping the mouse brain and
   other important organs. The results showed that the tumor volume in the
   brain tissue of the T+A@Glu‐NPs group was significantly reduced
   compared with that of A@Glu‐NPs and T@Glu‐NPs groups (Figure [128]5H),
   and the results of Ki67 and TUNEL staining were consistent with the
   subcutaneous glioma model (Figure [129]5F, G, I, and J). Moreover, the
   survival analysis showed that overall survival of the T+A@Glu‐NPs group
   was extended to 36 days and the overall survival of the T@Glu‐NPs group
   was only 28 days, which further verified the anti‐glioma efficacy of
   the T+A@Glu‐NPs was superior to that of T@Glu‐NPs (Figure [130]5E).
   Moreover, an orthotopic mouse model of human glioma LN229 in nude mice
   was established to further evaluate the anti‐glioma efficacy of
   T+A@Glu‐NPs in vivo. The experimental mice were randomized into 5
   treatment groups, the dosing method and dosing concentration remained
   unchanged, but the dosing interval was changed to once every 3 days for
   a total of 8 doses (Figure [131]6A). The in‐vivo imaging results showed
   that the T+A@Glu‐NPs group significantly inhibited tumor growth
   compared with the other groups, while the overall survival of mice in
   the T+A@Glu‐NPs was extended to 60 days from 49 days in the T@Glu‐NPs
   group (Figure [132]6C–E). Besides, there was no significant weight
   loss, organ damages and systemic toxicity observed in the experimental
   mice of the three glioma animal models, which suggested a reliable
   safety profile for the T+A@Glu‐NPs (Figures [133]4B, 5D, and 6B;
   Figures [134]S24‐S27, Supporting Information). In summary, we could
   propose a conclusion that the T+A@Glu‐NPs has stronger anti‐glioma
   efficacy and is a promising and safe drug combination regimen.

Figure 4.

   Figure 4
   [135]Open in a new tab

   Antitumor effect of T+A@Glu‐NPs in GL261 glioma subcutaneous model. A)
   Schedule of experimental design in GL261 mouse subcutaneous model. B)
   Changes of body weight of mice in different group during treatment. C)
   The tumor group curve of the mouse subcutaneous model. D) The image of
   tumors from different groups. E) Quantification of tumors weight from
   different groups. F,G) TUNEL immunofluorescence staining images and
   statistical analysis of the GL261 mouse subcutaneous tumor. H,I) Ki67
   staining images and statistical analysis of the GL261 mouse
   subcutaneous tumor. J,K) BRD4 immunohistochemical staining images and
   statistical analysis of the GL261 mouse subcutaneous tumor. Data are
   presented as mean ± SD. No significant difference is marked with ns. *P
   < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001.

Figure 5.

   Figure 5
   [136]Open in a new tab

   Antitumor effect of T+A@Glu‐NPs in GL261 glioma orthotopic model. A) In
   vivo imaging of GL261‐Luc tumor‐bearing mice from different groups at
   4, 8 and 12 days after injection of tumor cells. B) Schedule of
   experimental design in GL261 mouse orthotopic model. C) The curve of
   relative luminescence in different treatment groups. D) Changes of body
   weight of mice in different group during treatment (n = 5). E) The
   overall survival curve of the GL261 mouse orthotopic model and the
   humane endpoint is the persistent discomfort of mouse, such as severe
   hunched posture, reduced activity, apathy, leg dragging, or weight loss
   of more than 20%. F) Statistical analysis of TUNEL immunofluorescence
   staining. G) statistical analysis of Ki67 staining. H) H&E staining
   images of the GL261 orthotopic tumor on day 13 after the inoculation of
   GL261 cells. I) TUNEL immunofluorescence staining images of the GL261
   mouse orthotopic tumor. J) Ki67 staining images of the GL261 mouse
   subcutaneous tumor. Data are presented as mean ± SD. No significant
   difference is marked with ns. *P < 0.05, **P < 0.01, ***P < 0.001 and
   ****P < 0.0001.

Figure 6.

   Figure 6
   [137]Open in a new tab

   Antitumor effect of T+A@Glu‐NPs in LN229 glioma orthotopic model. A)
   Schedule of experimental design in mouse orthotopic model. B) Changes
   of body weight of mice in different group during treatment. C) In vivo
   imaging of LN229‐Luc tumor‐bearing mice from different groups at 5, 11,
   17, 23, and 30 days after injection of tumor cells. D) The curve of
   relative luminescence in different treatment groups. E) The overall
   survival curve of the LN229 mouse orthotopic model. Data are presented
   as mean ± SD. No significant difference is marked with ns. *P < 0.05,
   **P < 0.01, ***P < 0.001 and ****P < 0.0001.

2.5. Anti‐Glioma Mechanisms of T+A@NPs

   With the aim to disclose the key molecules and pathways involved in the
   enhanced anti‐tumor effects of T+A@Glu‐NPs, the transcriptome
   sequencing analysis was performed to analyze changes in the expression
   levels of relevant genes after drug treatment of GL261 glioma cells for
   48 h, and differential gene analysis and pathway enrichment analysis
   were sequentially implemented to find possible key genes and pathways.
   As shown in Figure [138]7A–D, Notch1 gene and related pathways were
   upregulated in T@Glu‐NPs group, while were downregulated significantly
   in the T+A@Glu‐NPs group, which indicated Notch1 gene and related
   pathways maybe play a critical role in glioma TMZ resistance. To
   further verify the hypothesis, the expression levels of the Notch1 gene
   and related pathway genes were detected by real time qPCR. The results
   demonstrated that Notch1 and notch pathway related genes were
   up‐regulated after T@Glu‐NPs treatment and could be down‐regulated
   after A@Glu‐NPs and T+A@Glu‐NPs treatment, including Maml2, Ccn3, Fat4,
   Ptp4a3 and Cdh6 (Figure [139]7E). Based on the above results, it is
   speculated that Notch1 gene might play a key role in glioma TMZ
   resistance, while ARV‐825 could degrade BRD4 protein to reverse glioma
   TMZ resistance by modulating Notch1 pathway related genes expression.
   Thereupon, western blot analysis was further performed to confer the
   correlation between BRD4 and Notch1 protein expression after glioma
   cells were treated for 48 h. As shown in Figure [140]7F,G, the
   expression of BRD4 and Notch1 was consistent in the same therapy group
   and was slightly upregulated in the T@Glu‐NPs group, while was
   significantly downregulated in the the A@Glu‐NPs and T+A@Glu‐NPs
   groups. The results were basically consistent with the previous
   transcriptome sequencing analysis and real time qPCR analysis results,
   substantiating the positive correlation between Notch1 gene and BRD4
   protein in glioma. To further determine this point, the protein
   expression levels in tumor samples from animal models were probed by
   immunohistochemical techniques and found that the proportions of BRD4‐
   and Notch1‐positive cells were slightly increased in T@Glu‐NPs and
   significantly decreased in tumor samples from both A@Glu‐NPs and
   T+A@Glu‐NPs (Figures [141]4J,K, and [142]7H,I). Therefore, we suggested
   that Notch1 may mediate glioma TMZ resistance, and T+A@Glu‐NPs can
   downregulate Notch1 by degrading BRD4 protein, thereby reversing glioma
   TMZ resistance.

Figure 7.

   Figure 7
   [143]Open in a new tab

   The mechanism of T+A@Glu‐NPs to reverse the TMZ resistance in glioma.
   A), B), and C) The RNA‐seq analysis and differential genes expression
   analysis of GL261 cells treated with 100 ng mL^−1 A@Glu‐NPs, 100 µM
   T@Glu‐NPs, and T+A@Glu‐NPs for 48 h. The untreated group was as2333 a
   control. D) The enrichment analysis based on the RNA‐seq. E) Real‐time
   quantitative PCR analysis of genes related to the Notch1 signaling
   pathway. F,G) Western blot analysis of BRD4 and NOTCH1 after the GL261
   cells treated with 100 ng mL^−1 A@Glu‐NPs, 100 µM T@Glu‐NPs, and
   T+A@Glu‐NPs for 48 h. The untreated group was as a control. H,I) Notch1
   immunohistochemical staining images and statistical analysis of the
   GL261 mouse subcutaneous tumor. Data are presented as mean ± SD. No
   significant difference is marked with ns. *P < 0.05, **P < 0.01, ***P <
   0.001, and ****P < 0.0001.

2.6. Knocking out Notch1 Gene Enhanced the Anti‐Glioma Efficacy of TMZ In
Vitro and In Vivo

   Many studies have shown that Notch1 is involved in the tumor growth,
   metastasis, drug resistance and recurrence of a variety of tumors,
   especially its close relationship with glioma stem cells.^[ [144]^39 ,
   [145]^40 , [146]^41 , [147]^42 , [148]^43 ^] In this study, we
   speculated that Notch1 could be a key molecule in the TMZ resistance of
   gliomas and subsequently constructed Notch1 knockout GL261 cell lines.
   As shown in Figure [149]8A,B, there were two Notch1 knockout GL261 cell
   lines, KO1‐1 and KO1‐2, which had been verified by quantitative PCR and
   Western blot analysis. After that, CCK‐8 was performed and it was found
   the GL261/KO1‐1 and GL261/KO1‐2 were more sensitive to TMZ after Notch1
   was knockout compared to the GL261/NC cell line (Figure [150]8C,D).
   Besides, the colony formation assay results showed that knockout of
   Notch1 would enhance proliferation inhibition of gliomas by TMZ
   (Figure [151]8E,F). Moreover, flow cycle and apoptosis analysis were
   further used to explore whether knockout of the Notch1 gene enhanced
   cycle block and increased cell apoptosis of glioma by TMZ. We found
   that knockout of Notch1 did enhance the TMZ‐induced cycle block in G2/M
   phase and induced an increase in apoptosis by TMZ (Figure [152]8G–J).
   Furthermore, NOTCH1 knockout human‐derived glioma cell lines were also
   performed. As shown in Figures [153]S28 and [154]S29 (Supporting
   Information), both the LN229 NOTCH1 gene knockout cell lines and the
   U251 NOTCH1 gene knockout cell lines were again sensitive to TMZ
   compared the normal LN229 and U251 cell lines. In short, knockdown of
   Notch1 gene would make glioma sensitized to TMZ again in vitro.

Figure 8.

   Figure 8
   [155]Open in a new tab

   Knocking out Notch1 gene enhanced the anti‐tumor efficacy of TMZ in
   vitro. A) The qPCR of two Notch1 gene knocked out clones. (B) Western
   blot analysis of Notch1. GAPDH served as a loading control. C) Cell
   viabilities of clone 1‐1 in GL261‐KO cells treated with T@Glu‐NPs at
   different concentrations for 48 h. D) Cell viabilities of clone 1–2 in
   GL261‐KO cells treated with T@Glu‐NPs at different concentrations for
   48 h. E,F) Representative images of colony formation of GL261(NC/KO)
   cells incubated with 25 µM TMZ/NPs and quantification of colony
   formation. G,H) Flow cytometry detection of GL261(NC/KO) cells
   apoptosis after different treatments for 48 h and statistical analysis
   of apoptosis cells. I,J) GL261 (NC/KO) cells cycle distribution
   determined by flow cytometry after treatment with 100 µM T@Glu‐NPs for
   48 h and quantification of cell cycle distribution. Data are presented
   as mean ± SD. No significant difference is marked with ns. *P < 0.05,
   **P < 0.01, ***P < 0.001 and ****P < 0.0001.

   Considering the complex effects of the internal environment and tumor
   microenvironment, the GL261/KO1‐2 and GL261/NC cell lines were
   inoculated into mice subcutaneously to construct both NOTCH1 KO and
   NOTCH1 NC mice models of subcutaneous tumor, respectively. As shown in
   Figure [156]9A, mice were randomly divided into four groups, NC‐GS,
   NC‐TMZ, KO‐GS, and KO‐TMZ, and were given the corresponding drug
   treatments. When the subcutaneous tumor volume of mice in the NC‐GS
   group grew to ≈1500 mm^3, the experiment was stopped and the
   subcutaneous tumor was peeled off. We found that there was no
   significant difference in the growth curves of gliomas, the weight and
   volume of isolated tumors between the NC‐GS and KO‐GS groups, while
   tumor growth was significantly slower in the KO‐TMZ group than that in
   the NC‐TMZ group (Figure [157]9C–E). And the knockout of Notch1 gene
   increased TMZ‐induced tumor proliferation inhibition and tumor cell
   apoptosis. Moreover, the results of Ki67 and TUNEL staining were
   consistent with previous results (Figure [158]9F–K). Subsequently, the
   GL261/KO1‐2 and GL261/NC cell lines were transfected with plasmids
   stably expressing luciferase to construct the new cell lines,
   GL261/KO1‐2‐Luc and GL261/NC‐Luc. Then the cells were inoculated into
   mouse brain to construct two kinds of orthotopic glioma mice, NC‐Luc
   and KO‐Luc (Figure [159]9N). The mice were randomly divided into NC‐GS,
   NC‐TMZ, KO‐GS and KO‐TMZ groups according to the results of the first
   live imaging, and were injected appropriate drugs (Figure [160]9N).
   During the treatment, small animal live imaging was performed every
   four days and all the death of the mice in the KO‐TMZ group as the
   endpoint of the survival observation. The results at the end of the
   experiment showed that there was no significant difference in the
   change of in situ tumor fluorescence intensity curves between the NC‐GS
   and KO‐GS groups, and tumor growth was significantly slower in the
   KO‐TMZ mice than that in the NC‐TMZ group (Figure [161]9L,M), and the
   overall survival of KO‐TMZ mice was extended to 35 days, while it was
   only 28 days in the NC‐TMZ group (Figure [162]9P). Besides, there was
   no weight loss in all groups (Figure [163]9B,O). All in all, we
   concluded that knockout of the Notch1 gene enhances the glioma
   therapeutic effectiveness of TMZ and further confirmed that the Notch1
   gene is a key molecule in the reversal of TMZ resistance in glioma.

Figure 9.

   Figure 9
   [164]Open in a new tab

   Knocking out Notch1 gene enhanced the anti‐tumor efficacy of TMZ in
   GL261 subcutaneous and orthotopic mouse model. A) Schedule of
   experimental design in GL26 (NC/KO) mouse subcutaneous model. B)
   Changes of body weight of mice in different group during treatment. C)
   The tumor group curve of the mouse subcutaneous model. D) The image of
   tumors from different groups. E) Quantification of tumors weight from
   different groups. F,G) BRD4 immunohistochemical staining images and
   statistical analysis of the GL261 (NC/KO) mouse subcutaneous tumor.
   H,I) Ki67 staining images and statistical analysis of the GL261 (NC/KO)
   mouse subcutaneous tumor. J,K) TUNEL immunofluorescence staining images
   and statistical analysis of the GL261 (NC/KO) mouse subcutaneous tumor.
   L) In vivo imaging of GL261‐Luc tumor‐bearing mice from different
   groups at 4, 8, and 12 days after injection of tumor cells. M) The
   curve of relative luminescence in different treatment groups. N)
   Schedule of experimental design in GL261‐Luc (NC/KO) mouse orthotopic
   model. O) Changes of body weight of mice in different group during
   treatment. P) The overall survival curve of the GL261‐Luc (NC/KO) mouse
   orthotopic model. Data are presented as mean ± SD. No significant
   difference is marked with ns. *P < 0.05, **P < 0.01, ***P < 0.001, and
   ****P < 0.0001.

2.7. Mechanisms of T+A@Glu‐NPs to Modulate Notch1 Expression to Enhance
Anti‐Glioma Efficacy of TMZ

   As an important epigenetic regulation factor, BRD4 acts as a
   transcriptional bridging platform and is able to regulate the
   transcription of a variety of genes. Besides, previous studies have
   shown that Notch1 expression is positively correlated with BRD4
   expression.^[ [165]^44 , [166]^45 ^] Therefore, we first performed a
   dual luciferase reporter gene assay to explore whether BRD4 was
   involved in the transcriptional regulation of NOTCH1 genes and whether
   T+A@Glu‐NPs could degrade BRD4 protein to downregulate Notch1
   expression. As shown in Figure [167]10A,B, the intensity of firefly
   luciferase in T+A@Glu‐NPs treated cells was significantly reduced
   compared to that in the control group, indicating a reduction in the
   transcription of the Notch1 gene, which verified that T+A@Glu‐NPs might
   reduce the activation of the Notch1 promoter by degrading BRD4 protein,
   thereby reducing luciferase expression. So, we could conclude that BRD4
   protein did regulate NOTCH1 gene expression. To further identify the
   promoter binding region of the BRD4 protein on the Notch1 gene, we
   designed 15 pairs of quantitative PCR primers for the NOTCH1 gene
   promoter sequence and set up a control group and an ARV‐825
   drug‐treated group. After 48 h, the DNA bound to the BRD4 protein in
   each group was precipitated and purified according to the Chromatin
   Immunoprecipitation Kit, after which quantitative PCR was performed to
   verify the results. As shown in Figure [168]10C, BRD4 protein bound to
   the promoter region of the Notch1 gene and was involved in the
   transcriptional regulation of this gene. In another word, ARV‐825 could
   degrade BRD4 protein thereby inhibiting the transcription of the Notch1
   gene.

Figure 10.

   Figure 10
   [169]Open in a new tab

   Mechanisms of T+A@Glu‐NPs modulate Notch1 expression to enhanced
   anti‐glioma efficacy of TMZ. A) The diagram of BRD4 acting on the
   Notch1 gene promoter and ARV‐825 reducing luciferase gene transcription
   initiated by the Notch1 promoter through inhibiting BRD4. B) Relative
   Notch1 luciferase reporter activity in GL261 cells treated with 100
   ng mL^−1 A@Glu‐NPs, 100 µM T@Glu‐NPs and T+A@Glu‐NPs for 48 h. The
   untreated group was as a control. C) ChIP‐qPCR analysis of Notch1
   promoter in DMSO or 100 ng mL^−1 A@Glu‐NPs treated GL261 cells for 48 h
   using IgG or anti‐BRD4. Data are presented as mean ± SD. No significant
   difference is marked with ns. *P < 0.05, **P < 0.01, ***P < 0.001, and
   ****P < 0.0001.

3. Discussion

   Currently, TMZ resistance in gliomas remains a significant challenge in
   clinical glioma chemotherapy, so there is an increased urgency for new
   drug combination therapies. A multitude of small molecule inhibitors of
   the BET family proteins have been employed to enhance the synergy of
   tumor chemotherapy, and these studies have demonstrated that the
   inhibition of BET proteins can indeed sensitize tumors to chemotherapy
   and extend survival in mice.^[ [170]^46 , [171]^47 , [172]^48 ^]
   However, these small molecule inhibitors are more prone to off‐target
   effects and systemic toxicity due to the necessity for higher doses. In
   contrast, the PROTAC technology ubiquitinates target proteins by
   linking the target protein ligand to a ubiquitin E3 ligase, and then
   efficiently and specifically degrades the target protein with the
   assistance of the ubiquitin proteasome system, achieving sustained
   degradation at very low concentrations.^[ [173]^49 , [174]^50 ^]
   Consequently, various PROTAC formulations have been applied in tumor
   chemotherapy, yielding improved efficacy. And the ARV‐825, as a BRD4
   PROTAC, requires minimal doses to efficiently degrade target proteins
   and exhibits reduced susceptibility to drug resistance in single‐agent
   glioma treatments. So, in this study, we propose for the first time the
   use of ARV‐825 as a therapeutic adjuvant for TMZ in gliomas.

   Furthermore, to overcome the challenges of poor systemic drug targeting
   and low blood‐brain barrier (BBB) permeability, we designed
   glucose‐modified nano micelles, Glu‐PEG‐PCL, to construct a drug
   delivery platform. This platform ingeniously exploits the glucose
   transporter, which is highly expressed in both the cerebral vascular
   endothelium and glioma cell membranes, to deliver drugs in a targeted
   and efficient manner. Additionally, to ensure effective drug release
   within glioma cells, we incorporated a disulfide bond into the
   nanomaterials, which can be responsively cleaved in the high GSH
   environment of the glioma periplasm, enabling targeted drug release.^[
   [175]^51 , [176]^52 , [177]^53 ^] The results indicated that the
   modified glucose molecules facilitated dual targeting of T+A@Glu‐NPs to
   the BBB and gliomas in a “two birds with one stone” style, and the
   GSH‐responsive crosslinker containing disulfide bonds ensured
   “directional blasting” cleavage of the nanoparticles to release TMZ and
   ARV‐825 in the high GSH environment of glioma cells. Moreover, the
   safety evaluation in mice revealed no significant toxic side effects in
   terms of blood biochemistry and vital organs, as well as no abnormal
   weight loss in mice during treatment, demonstrating the reliable safety
   of this delivery system as a drug delivery platform. All results
   suggest that T+A@Glu‐NP is an efficient and promising drug delivery
   platform for glioma chemotherapy.

   Ultimately, T+A@Glu‐NPs significantly inhibited the glioma progression
   and prolonged the overall survival of the tumor bearing mice through
   inducing cell apoptosis, inhibiting cell proliferation and blocking
   cell cycle. More importantly, we found that the expression levels of
   Notch1 were positively correlated with BRD4 proteins in Western blot
   analysis and immunohistochemistry results of in vivo tumor samples.
   Furthermore, T+A@Glu‐NPs were able to significantly downregulate the
   expression of Notch1, as revealed by transcriptome sequencing.
   Subsequent dual‐luciferase reporter assays and ChIP‐PCR experiments
   conclusively verified that BRD4 protein could bind to the Notch1
   promoter to regulate its transcription. Meanwhile, Notch1
   pathway‐related genes, including Maml2, Ccn3, Fat4, Cdh6, and Ptp4a3,
   exhibited similar expression trends and may be intimately linked to
   glioma development. Additionally, knockout of the Notch1 gene enhanced
   the efficacy of TMZ in gliomas and the survival benefit of mice. We
   therefore conclude that ARV‐825 can degrade BRD4 protein to
   downregulate the transcription of the Notch1 gene, reversing TMZ
   resistance in gliomas. It is somewhat regrettable that we did not
   explore the downstream targets of the Notch pathway, which would be a
   direction for future research.

   In brief, the T+A@Glu‐NPs is a novel and potential combined strategy
   for glioma chemotherapy and the sensitization of glioma chemotherapy by
   T+A@Glu‐NPs through modulation of Notch1 is a novel and main mechanism,
   offering a new direction for the development of new glioma
   chemotherapeutic regimens and providing more options for clinical
   glioma treatment.

4. Conclusion

   In summary, the Glu‐modified nanoparticles we constructed effectively
   penetrated the blood‐brain barrier and targeted gliomas, enabling
   precise drug delivery. Concurrently, T+A@Glu‐NPs have demonstrated
   strong efficacy against gliomas in both in vivo and in vitro studies.
   Furthermore, we have elucidated that T+A@Glu‐NPs exert their effects by
   modulating Notch1 expression. In conclusion, the T+A@Glu‐NPs we
   developed represent a novel and promising strategy for glioma
   chemotherapy. The elucidation of their mechanism of action could
   broaden the scope of drug combination regimens and offer valuable
   insights for clinical practice.

5. Experimental Section

Preparation and Characterization of T+A@Glu‐NPs

   The T+A@Glu‐NPs were prepared using a nano‐precipitation method. 2 mg
   TMZ, 2 mg ARV825, 30 mg PEG‐SS‐PCL, and 10 mg Glu‐PEG‐PCL were
   dissolved in 500 µL DMSO. The solution was then quickly added to ten
   times the volume of water and ultrasonicated for an additional 10–15
   min. The organic phase and free drug were removed by ultrafiltration
   centrifugation to obtain T+A@Glu‐NPs. Unmodified nanoparticles were
   prepared by the same method.

   Dynamic Light Scattering (Brookhaven Instruments, US) was initially
   employed to determine the size and polydispersity index (PDI) of
   T+A@Glu‐NPs. Malvern particle sizer (Malvern Nano ZS90 ZEN3690) was
   used for measuring the zeta potential of T+A@Glu‐NPs. Subsequently,
   Transmission Electron Microscopy (TEM, JEM‐2100Plus, Japan) was
   utilized to observe the dimensional and morphological characteristics
   of T+A@Glu‐NPs. In order to assess the stability of T+A@Glu‐NPs, the
   nanoparticles were diluted in pure water at 4°C and in 10% FBS at 37°C
   to simulate storage conditions and conditions in plasma, respectively.
   The particle size and PDI of T+A@Glu‐NPs were measured at intervals of
   0, 1, 2, 4, 6, 8, 24, 48, 96, and 144 h, with each measurement
   replicated three times under the same conditions. Moreover, to evaluate
   the reduction responsiveness of the nanoparticles, T+A@Glu‐NPs were
   diluted in three different PBS buffers at pH 7.4, each containing 10 mM
   GSH, 0.1 mM GSH, and 0 mM GSH. The samples were then incubated at 37°C
   on a shaker at 100 rpm to ensure consistent concentration and
   distribution of the drug and GSH. The size and PDI of T+A@Glu‐NPs were
   measured at 0, 1, 2, 4, 6, 8, and 12 h post‐incubation. Meanwhile, TEM
   was utilized again to observe the changes on dimensional and
   morphological characteristics of T+A@Glu‐NPs after incubated in 10 mM
   GSH buffer for 8 h.

Cell Uptake Assay

   GL261 or LN229 cells were digested with trypsin and inoculated into
   12‐well plates at a certain density (2 × 10^4 per well for GL261 and 5×
   10^4 per well for LN229), and were incubated overnight at 37°C in an
   incubator. The next day, the GL261 or LN229 cells were incubated with
   Ce6, Ce6@NPs and Ce6@Glu‐NPs at a concentration of 2 µg mL^−1.
   Afterwards, samples were collected for flow cytometry to determine
   intracellular Ce6 fluorescence intensity at 1 h, 2 h, and 4 h time
   points, respectively.

Confocal Imaging

   GL261 or LN229 cells were digested with trypsin and planted into
   confocal dishes at a certain density (1× 10^4 per well for GL261 and
   1.2× 10^4 per well for LN229), and were incubated overnight at 37°C in
   an incubator. The next day, the GL261 or LN229 cells were incubated
   with Ce6, Ce6@NPs and Ce6@Glu‐NPs at a concentration of 20 µg mL^−1. At
   1 h and 4 h timepoints, samples were taken to terminate the incubation,
   and confocal imaging was performed after Hoechst 33342 and LysoTracker
   Green DND‐26 staining.

Cell Viability Assay

   GL261, LN229, U251, and U87 cells were digested with trypsin and
   inoculated into 96‐well plates at a certain density (2 × 10^3 per well
   for GL261, 4 × 10^3 per well for LN229, 8 × 10^3 per well for U251 and
   U87), and were incubated overnight at 37°C in an incubator. The next
   day, the GL261 or LN229 cells were incubated with different
   concentrations of A@Glu‐NPs, T@Glu‐NPs and T+A@Glu‐NPs for 48 h. The
   untreated group served as a negative control. Sequentially, 20 µL of
   CCK‐8 was added to each well and incubated for another 1–2 h. The
   absorbance value (OD:450 nm) was determined by enzyme marker.

FCM Detection of Cells Apoptosis

   GL261 or LN229 cells were digested with trypsin and inoculated into
   6‐well plates at a density of 1×10^5 per well, and were incubated
   overnight in the incubator. The next day, GL261 or LN229 cells were
   incubated with certain concentrations of A@Glu‐NPs, T@Glu‐NPs and
   T+A@Glu‐NPs for 48 h. The untreated group served as a negative control.
   The supernatants were recovered and washed once with pre‐cooled PBS at
   4°C, then 500 µL EDTA‐free trypsin digestion was added to each well.
   The digested glioma cells were collected into flow‐through tubes,
   centrifuged, resuspended, and washed twice with pre‐cooled PBS.
   Finally, 100 µL prepared staining solution (containing 5 µL
   annexin‐FITC and 5 µL PI per 100 µL) was added to each sample tube, and
   incubated at room temperature and protected from light for 15 min. 400
   µL 1× Binding buffer was added to each tube and gently resuspended
   before flow apoptosis assay was performed using a flow cytometer.

FCM Analysis of Cell Cycle

   GL261 or LN229 cells were digested with trypsin, inoculated into 6‐well
   plates at a density of 1×10^5 per well, and incubated overnight in the
   incubator. The double‐free medium was changed 12 h before drug
   addition. After that, GL261 or LN229 cells were incubated with certain
   concentrations of A@Glu‐NPs, T@Glu‐NPs and T+A@Glu‐NPs for 48 h. The
   untreated group served as a negative control. The cells were washed
   once with pre‐cooled PBS at 4°C, and 500 µL trypsin was added to each
   well for digestion. The digested glioma cells were collected into flow
   tubes, centrifuged, resuspended, and washed twice with pre‐cooled PBS.
   Finally, the cells were resuspended in 2 mL of pre‐cooled 75% ethanol
   and fixed at 4°C for 24 h. The fixed samples were added with the
   configured staining solution per tube and incubated at room temperature
   away from light for 30 min. Sequentially, the flow cycle analysis was
   performed at a low speed.

Colony Formation Assay

   GL261 or LN229 cells were digested with trypsin and inoculated into
   6‐well plates at a density of 500–1000 per well, and were incubated
   overnight in the incubator. After that, GL261 or LN229 cells were
   incubated with certain concentrations of A@Glu‐NPs, T@Glu‐NPs and
   T+A@Glu‐NPs for 72 h. The untreated group served as a negative control.
   The drug‐containing medium was changed every 3 days until the cell
   number of single clone formation reached ≈50–60. Removed the
   supernatant, washed with PBS once, and fixed the cells with 1mL of 4%
   paraformaldehyde tissue fixative per well for 30–60 min. Then removed
   the fixative, washed with PBS again, and stain with 500 µL of 0.1%
   crystal violet staining solution per well for 10–20 min. Recycled the
   staining solution, washed the plates with PBS for 3 times or more, air
   dry, and take pictures.

shRNA Knockdown Experiment

   First, Notch1 shRNA primer sequences were first designed and then
   recombined with PLKO.1 plasmid to construct a new knockout plasmid. The
   empty vector as negative control. Then, viral supernatants containing
   large amounts of the target plasmid were obtained after plasmid
   transformation, single clone picking, sequencing and identification,
   and viral packaging. Sequentially, well‐grown GL261 cells were
   inoculated into 6‐well plates at 2×10^4 per well, cultured overnight,
   and added the virus supernatant for infection. After 48 h of infection,
   puromycin at concentration of 1 µg mL^−1 was added to screen cells for
   48–72 h. Finally, the screened GL261 cells were re‐laid in plates for
   subsequent WB and qPCR experiments to verify the knockdown efficiency.

Dual‐Luciferase Reporter Assay

   GL261 cells were digested with trypsin, inoculated into 12‐well plates
   at a density of 5×10^4 per well and cultured overnight. Double‐free
   medium was changed before transfection for 2–3 h. The prepared system
   containing PRL‐SV40 and Notch1‐Luc was added to the well plates
   according to the amount of 1 µg PRL‐SV40+1 µg Notch1‐Luc+2 µL
   Lipofectamine 2000 per well. After 4–6 h, the complete medium
   containing the drugs was switched, and the incubator was continued to
   incubate for 48 h. Finally, according to the Meilunbio Firefly &
   Renilla Luciferase Reporter Assay Kit to determine luciferase activity.

Chromatin Immunoprecipitation (ChIP) Assay

   ChIP assays were performed via the ChIP Assay Kit (Cell Signaling
   Technology, USA) according to the manufacturer's instructions. In
   brief, GL261 cells were incubated with 100 ng mL^−1 A@Glu‐NPs and equal
   volume of DMSO respectively for 48 h, followed by fixing with
   formaldehyde to crosslink targeted proteins and DNA to form chromatin
   complexes. After breaking cells with ultrasonic, the chromatin
   complexes were immunoprecipitated via anti‐BRD4 antibody (Cell
   Signaling Technology, USA) or negative control IgG, and subsequently
   the protein G magnetic beads were used to capture the
   antibody‐chromatin complexes formed in front, followed by elution and
   decrosslinking. Finally, the purified DNA samples were detected by
   qRT‐PCR. The primers are listed in Supporting Information Table.

Subcutaneous Xenograft Model

   GL261‐WT and GL261‐KO cells were subcutaneously inoculated into C57BL/6
   female mice at 5×10^6 cells each mouse. When the tumors grew to 5 mm ×
   5 mm, the mice were randomly divided into four groups: (A) NC‐GS, (B)
   NC‐TMZ, (C) KO‐GS, and (D) KO‐TMZ. Each mouse was injected with 200 µL
   of the drug solution in the tail vein, and the dose of T@Glu‐NPs was 10
   mg kg^−1 body weight. The drug was administered every two days for a
   total of four times, and the changes in tumor volume and body weight of
   the mice were recorded at the same time. When the tumor volume reached
   15 mm×15 mm, the mice were sacrificed and the tumors were weighted and
   photographed, and then the tumor tissues were sent for
   immunohistochemistry‐related sections.

Glioma Orthotopic Xenograft Model

   GL261‐luc or LN229‐luc cells were digested and washed twice with
   double‐free medium. C57BL/6 mice or nude mice were anesthetized and the
   hair on the head of the mice was removed. Then, a certain concentration
   cells (2×10^5 cells for GL261‐luc, 5×10^5 cells for LN229‐luc) were
   injected into mice brain. 4–5 days after inoculation, the mice were
   randomly divided into five groups according to fluorescence intensity:
   (1) NC; (2) Blank‐NPs; (3) A@Glu‐NPs; (4) T@Glu‐NPs; and (5)
   T+A@Glu‐NPs. The drugs were administered every 2 days for a total of 5
   times in GL261‐luc orthotopic model and every 3 days for a total of 8
   times in LN229‐luc orthotopic model. During treatment, tumor growth was
   monitored by in vivo imaging techniques and the body weight of the mice
   were recorded. At the end, blood samples serums were collected and
   prepared for the blood biochemistry test, and the whole brain and
   important organs were separated, fixed in 4% paraformaldehyde and the
   brain tumor tissues and vital organs were sent for immunohistochemistry
   and immunofluorescence staining. All animal experiments were approved
   by the Animal Experimentation Ethics Committee of the State Key
   Laboratory of Biotherapeutics (SKLB), Sichuan University, and were
   conducted in accordance with the Guidelines for the Care and Use of
   Institutional Animals.

Statistical Analysis

   In this study, data were presented as mean ± standard deviation of
   measurement (SD). The statistical differences of data were analyzed
   with the unpaired two‐tailed Student t test for two groups comparison
   and ordinary one‐way analysis of variance for multiple groups
   comparison in GraphPad Prism 9.0 software. The P value of more than
   0.05 indicated no significant difference and was marked with ns. P
   value of less than 0.05 was considered statistically significant and
   were presented as *P < 0.05, **P < 0.01, ***P < 0.001 and ****P <
   0.0001.

Conflict of Interest

   The authors declare no conflict of interest.

Supporting information

   Supporting Information
   [178]ADVS-11-2409753-s001.docx^ (5.6MB, docx)

Acknowledgements